Wari

Wari
Lords of the Ancient Andes

Susan E. Bergh
Introduction by Luis G. Lumbreras

With

Luis Jaime Castillo Butters

Anita G. Cook
Mary Glowacki
William H. Isbell and Margaret Young-Sánchez
Justin Jennings
Heidi King
Patricia J. Knobloch
Gordon F. McEwan and Patrick Ryan Williams
Donna Nash
Ann Pollard Rowe
Katharina Schreiber
Published on the occasion of the exhibition Wari: Lords of the Ancient
Andes, organized by the Cleveland Museum of Art.

The Cleveland Museum of Art

28 October 2012–6 January 2013

Museum of Art Fort Lauderdale

10 February–19 May 2013

Kimbell Art Museum, Fort Worth

16 June–8 September 2013
Continuing a tradition of strong support and
active involvement in the arts, Hahn Loeser
& Parks is proud to sponsor the first North
American major exhibition devoted to the
startlingly beautiful arts of the Wari. An
ancestor of the better-known Inca, the Wari
are believed to have forged one of the first
empires of the ancient Andes.
Hahn Loeser has a long history of supporting
the Cleveland Museum of Art. Together with
our clients and staff, we congratulate the
museum for offering stellar exhibitions such
as Wari: Lords of the Ancient Andes. Hahn
Loeser looks forward to a strong future for
the arts, thanks to the efforts of the Cleveland
Museum of Art.
Lawrence E. Oscar
Chief Executive Officer
Stephen J. Knerly Jr.
Partner
Hahn Loeser & Parks LLP
Wari: Lords of the Ancient Andes has been
made possible in part by the National
Endowment for the Humanities: Exploring the
human endeavor. This exhibition is supported
by an indemnity from the Federal Council
on the Arts and the Humanities. Support for
exhibition programming has been provided
in part by Georgia and Michael DeHavenon
and by the Ohio Humanities Council, a state
affiliate of the National Endowment for the
Humanities. Research for this exhibition was
supported by a Curatorial Research Fellowship
from the Getty Foundation.
This publication is made possible in part by
the Andrew W. Mellon Foundation.
The Cleveland Museum of Art is generously
funded by Cuyahoga County residents through
Cuyahoga Arts and Culture. The Ohio Arts
Council helped fund this exhibition with
state tax dollars to encourage economic
growth, educational excellence, and cultural
enrichment for all Ohioans.
 Contents

ix Lenders to the Exhibition

x Director’s Foreword

xi Acknowledgments

xiv Maps

1 Introduction
Luis G. Lumbreras

5 The History of Inquiry into the Wari and Their Arts

Susan E. Bergh and Justin Jennings

Interpreting a Cosmopolitan Andean Society

31 The Rise of an Andean Empire

Katharina Schreiber

47 Looking at the Wari Empire from the Outside In

Luis Jaime Castillo Butters

Transforming the World

65 The Wari Built Environment: Landscape and Architecture of Empire

Gordon F. McEwan and Patrick Ryan Williams

82 The Art of Feasting: Building an Empire with Food and Drink

Donna Nash

103 The Coming of the Staff Deity

Anita G. Cook

122 Archives in Clay: The Styles and Stories of Wari Ceramic Artists
Patricia J. Knobloch

145 Shattered Ceramics and Offerings
Mary Glowacki
 159 Tapestry-woven Tunics
Susan E. Bergh

193 Tie-dyed Tunics

Ann Pollard Rowe

207 Featherwork
Heidi King

217 Inlaid and Metal Ornaments

Susan E. Bergh

233 Figurines
Susan E. Bergh

243 Wood Containers and Cups

Susan E. Bergh

The Aftermath

251 Wari’s Andean Legacy

William H. Isbell and Margaret Young-Sánchez

268 Checklist of the Exhibition

279 Reference List

 Lenders to the Exhibition

Private Collectors Museo de Arte de Lima

I. Michael Kasser Collection
Private collections
Institutional Lenders
American Museum of Natural History,
New York City
The Art Institute of Chicago
Brooklyn Museum, New York City
The Cleveland Museum of Art
Dallas Museum of Art
The Dayton Art Institute, Ohio
Denver Art Museum
Deutsches Textilmuseum, Krefeld
Dumbarton Oaks Research Library and
Collection, Washington, DC
Ethnologisches Museum, Berlin
Fine Arts Museums of San Francisco
Fowler Museum at UCLA, Los Angeles
Fundación Museo Amano, Lima
Kimbell Art Museum, Fort Worth
Linden-Museum, Stuttgart
Los Angeles County Museum of Art
The Metropolitan Museum of Art,
New York City
Michael C. Carlos Museum, Emory
University, Atlanta
Milwaukee Public Museum
Museo de América, Madrid
Museo de Arqueología y Antropología,
Universidad Nacional Mayor de San Marcos,
Lima
Museo de Sitio Huaca Pucllana, Lima
Museo Histórico Regional “Hipólito Unanue,”
Ayacucho
Museo Larco, Lima
Museo Nacional de Arqueología,
Antropología e Historia del Perú, Lima
Museo Regional de Ica “Adolfo Bermúdez
Jenkins”
Museum der Kulturen, Basel
Museum of Art, Rhode Island School of
Design, Providence
Museum of Fine Arts, Boston
Museum of Fine Arts, Houston
Museum Rietberg, Zurich
Museum zu Allerheiligen, Schaffhausen
Niedersächsiches Landesmuseum, Hannover
Peabody Museum of Archaeology and
Ethnology, Harvard University, Cambridge
Pontificia Universidad Católica del Perú,
Lima
Princeton University Art Museum
Roemer- und Pelizaeus-Museum, Hildesheim
Royal Ontario Museum, Toronto
Staatliches Museum für Völkerkunde,
Munich
The Textile Museum, Washington, DC
University of Pennsylvania Museum of
Archaeology and Anthropology, Philadelphia

ix
© 2012 The Cleveland Museum of Art. All rights reserved. Library of Congress Control Number: 2012932588
No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or Casebound, Thames & Hudson Inc.:
mechanical, including photocopy, recording or any other ISBN 978-0-500-51656-0
information storage and retrieval system, without prior
Paperback, The Cleveland Museum of Art:
permission in writing from the Cleveland Museum of Art.
ISBN: 978-1-935294-07-8
The works of art themselves may also be protected by
copyright in the United States of America or abroad and
may not be reproduced in any form or medium without
First published in 2012 in hardcover in the United States
permission from the Cleveland Museum of Art.
of America by Thames & Hudson Inc., 500 Fifth Avenue,
Images of works of art in the exhibition were provided New York, New York 10110
by the lenders, unless noted otherwise. Objects in the
thamesandhudsonusa.com
collection of the Cleveland Museum of Art were photo-
graphed by museum photographers Howard Agriesti and
Gary Kirchenbauer. David Brichford and Bruce Shewitz
Produced by the Cleveland Museum of Art
assisted with preparation of the digital files, and Howard
Agriesti supervised the color management. All works Editing: Barbara J. Bradley and Kathleen Mills
of art themselves may be protected by copyright in the Proofreading: Amanda A. Mikolic and Andrea F. Vazquez
United States or abroad and may not be reproduced in Design: Thomas H. Barnard III
any form without permission from the copyright holders.
Prepress production: Howard Agriesti and David Brichford
Where provided, individual photographers and copyright
holders are acknowledged in figure captions. Printed and bound in Italy by Graphicom

The Cleveland Museum of Art

www.ClevelandArt.org

[103]. Frontispiece, detail of a tunic fragment with bird-

headed staff-bearing creature in profile. (For the complete
textile, see fig. 152.)
 Director’s Foreword
The central Andes is one of only four places in
the world where civilization emerged indepen-
dently; the complex societies that developed
there are unique in having done so in the
absence of an alphabet and writing. Among
them, the Wari in Peru, who held sway in the
region from about AD 600 to 1000, forged a
society of such unprecedented complexity that
many today interpret it as the region’s first
empire. In spite of this remarkable accomplish-
ment, the Wari remain poorly known outside
of specialists’ circles. The Cleveland Museum
of Art is pleased to bring this crucial but sur-
prisingly unfamiliar chapter in Amerindian
achievement to public attention.
The timing of Wari: Lords of the Ancient
Andes is felicitous; it precedes by less than
a year the opening of galleries for the mu-
seum’s well-regarded Pre-Columbian collec-
tion, which go on view in summer 2013 as our
renovation and construction project draws
to a close. This exhibition and these galleries
confirm our pledge to display the arts of Am-
erindian peoples, a commitment the museum’s
founders established when they resolved to
make “the first attempt of an American art
museum to show . . . the art of those who lived
here [in the Americas] before the white man
came.” We are pleased to renew and carry for-
ward this important aspect of their vision as
the museum approaches its 2016 centenary.
Initial thanks go to Susan E. Bergh, the
museum’s curator of the arts of the ancient
Americas, for bringing Wari to the attention of
audiences in the United States with the exhi-
bition and this catalogue. She began working
on the project in 2007, conducting research in
many collections internationally and gather-
ing together a team of expert advisors, both art
historians and archaeologists.
x
Most sincere gratitude is due the public
institutions in Europe, Peru, the U.S., and
Canada and the several private collectors who
allowed the splendid works of art in their care
to travel long distances. The exhibition would
never have taken place without their enthusi-
asm, generosity, and help, often provided at the
cost of precious time and effort, particularly in
the case of the Textile Museum. In Peru, we ex-
tend very special thanks to President Ollanta
Humala and to the Ministry of Culture under
the direction of Dr. Luis Peirano Falconí for
critical assistance in arranging loans.
Equally crucial is the support of several
funders, to whom we offer heartfelt apprecia-
tion. They include Hahn Loeser & Parks LLP,
the sponsor of the Cleveland presentation, and
the National Endowment for the Humanities.
The Federal Council on the Arts and the Hu-
manities indemnified the exhibition. Georgia
and Michael DeHavenon and the Ohio Human-
ities Council, a state affiliate of the National
Endowment for the Humanities, provided
funding for the exhibition’s public program-
ming. A Curatorial Research Fellowship grant
from the Getty Foundation supported early
research. Finally, this engaging and beautiful
catalogue is made possible by the Andrew W.
Mellon Foundation.
We are delighted that Wari: Lords of the
Ancient Andes will be on view at the Museum
of Art Fort Lauderdale and then at the Kimbell
Art Museum in Fort Worth and confident that
it will captivate audiences at all three venues.
David Franklin, President and CEO
Sarah S. and Alexander M. Cutler Director
The Cleveland Museum of Art
 Acknowledgments
These acknowledgments begin with the people
to whom this catalogue is dedicated: my
parents, Florence and Norman Bergh, who are
bedrock, and the scholars Dorothy Menzel and
Alan Sawyer, both pioneers in Wari studies
whose seminal, time-tested contributions are
recognized universally.
No exhibition comes into being without
the collegiality, friendship, and cooperation
of dozens of people and institutions. A
milestone in the inception of this project
was the meetings of advisors, both local
and international, who provided invaluable
advice: Luis Jaime Castillo Butters; Anita
G. Cook; Peter Dunham; Enrique González
Carré, who has been a friend to the exhibition
in many ways; Anne Helmreich, who
was both personally and professionally
important; William H. Isbell; Justin Jennings;
Miriam Levin; Luis G. Lumbreras; Gordon
F. McEwan; Christian Mesía Montenegro;
Donna Nash; Katharina Schreiber; Rebecca
Stone; John Topic; and Patrick Ryan Williams.
Professional facilitator Richard Buchanan
expertly guided discussion, assuring that no
time was wasted. I also deeply appreciate the
efforts of the authors whose thoughtful essays
make up this catalogue; it has been a privilege
to work with them. A special salute is owed
the funders whose generosity David Franklin
recognizes in his foreword. I add my gratitude
to his, and send personal thanks to Barbara
Bays of the NEH, the DeHavenons, and
Stephen J. Knerly Jr. for their kind assistance.
Crucial to any exhibition is the
willingness of institutions and individuals to
lend treasures from their collections and to
invest considerable time and effort in doing so.
For their generous help with loans and access
to collections for study, I am very grateful to
xi
Charles Spencer, Sumru Aricanli, and Judith
Levinson, American Museum of Natural
History, New York, where errant loan requests
were processed with great speed and courtesy;
Nancy Rosoff, Brooklyn Museum, who has
been a friend and project supporter; Carol
Robbins and Roslyn Walker, Dallas Museum
of Art; Sally Kurtz, the Dayton Art Institute;
Margaret Young-Sánchez, Denver Art Museum,
who contributed through both word and last-
minute deeds of generosity; Isa Fleischmann-
Heck and Petra Brachwitz, Deutsches
Textilmuseum, Krefeld; Juan Antonio
Murro, Dumbarton Oaks Research Library
and Collection, Washington, DC; Manuela
Fischer and Lena Bjerregaard, Ethnologisches
Museum, Berlin, who offered unstinting
hospitality and access to collections in storage;
Marla Berns and John Pohl, Fowler Museum
at UCLA, Los Angeles; Rosa Watanabe,
Fundación Museo Amano, Lima; Jennifer
Casler Price, Kimbell Art Museum, Fort Worth,
with whom it has been a pleasure to work
as a venue partner; Doris Kurella and Inés
de Castro, Linden-Museum, Stuttgart; Kaye
Spilker and the late, much missed Virginia
Fields, Los Angeles County Museum of Art;
Julie Jones, Heidi King, and Christine Giuntini,
the Metropolitan Museum of Art, New York,
who gave me the benefit of their professional
advice and personal friendship; Bonnie
Speed, Michael C. Carlos Museum, Emory
University, Atlanta; Ellen Censky and Dawn
Scher Thomae, Milwaukee Public Museum;
Werner Rutishauser, Museum zu Allerheiligen,
Schaffhausen; Concepción García Sáiz and
Ana Verde, Museo de América, Madrid; Rector
Pedro Cotillo Zegarra, Carlos Del Águilar
Chávez, and Christian Altamirano, Museo
de Arqueología y Antropología, Universidad
 Nacional Mayor de San Marcos, Lima; Kate
Irvin, Museum of Art, Rhode Island School of
Design, Providence; Natalia Majluf and Cecilia
Pardo Grau, Museo de Arte de Lima; Pamela
Parmal, Museum of Fine Arts, Boston; Frances
Marzio and Chelsea Dacus, Museum of Fine
Arts, Houston; Mario Teodoro Cueto Cárdenas
and Jorge Luis Soto Maguino, Museo Histórico
Regional “Hipólito Unanue,” Ayacucho;
Alexander Brust, Museum der Kulturen, Basel;
Andrés Alvarez-Calderón and Ulla Holmquist,
Museo Larco, Lima; the staff at the Museo
Nacional de Arqueología, Antropología e
Historia del Perú, Lima, including current
and former directors Carmen Teresa Carrasco
Cavero and Carmen Arellano Hoffman and
section heads Sonia Amparo Quiroz Calle,
Gabriela Schworbel, Carmen Thays Delgado,
and Milano Trejo Huayta, along with all of
the helpful personnel in their departments;
Peter Fux, Museum Rietberg, Zurich; Emma
Susana Arce Torres, Museo Regional de Ica
“Adolfo Bermúdez Jenkins”; Isabel Flores
Espinoza and Pedro Vargas Nalvarte, Museo
de Sitio Huaca Pucllana, Lima; Jutta Steffen-
Schrade, Niedersächsisches Landesmuseum,
Hannover; Jeffrey Quilter and Susan Haskell,
Peabody Museum of Archaeology and
Ethnology, Harvard University, Cambridge;
at Lima’s Pontificia Universidad Católica del
Perú, Rector Marcial Rubio Correa, Enrique
González Carré, and Luis Jaime Castillo
Butters and his dedicated students; Bryan
Just, Princeton University Art Museum; the
several private collectors, including I. Michael
Kasser, who kindly assumed the burden of
coordinating loans; Regine Schulz, Roemer-
und Pelizaeus-Museum, Hildesheim; Justin
Jennings and Anu Liivandi, Royal Ontario
Museum, Toronto; Christine Stelzig, Elke
x ii
Bujok, and Regina Stumbaum, Staatliches
Museum für Völkerkunde, Munich; Maryclaire
Ramsey, Esther Méthé, Ann Rowe, and Rachel
Shabica, the Textile Museum, Washington,
DC, whose efforts on behalf of the project were
extraordinary; and Lucy Fowler Williams
and William Wierzbowski, University of
Pennsylvania Museum of Archaeology and
Anthropology, Philadelphia. The staff at
many other museums also accommodated my
research requests and visits; only limitations
of space prevent me from thanking these
colleagues by name.
Others outside of the museum also
contributed in indispensable ways. Patricia
Díaz translated official letters and documents
into graceful, nuanced Spanish. Patricia
Knobloch selflessly provided hours of
consultation about the ins and outs of Wari
ceramic styles. In the project’s final phases,
fellow Andrea F. Vazquez arrived in the nick
of time and adroitly handled many tasks in
relation to the catalogue and other matters.
She followed in the footsteps of Wendy Earle,
who as an intern competently assisted with
earlier phases of research. At the cost of much
time and effort, Barbara Wolff of Ayni guided
our efforts to arrange a visit to Cleveland by
master weavers from Ayacucho. Fatma Wille
of Christie’s evaluated our loan list for federal
indemnity, and David Bernstein, Steve Burger,
and Stacy Goodman provided important help.
An especially heartfelt expression of gratitude
must be extended to Bertha Vargas Vargas, who
served as our knowledgeable, accomplished
impresario of the Peruvian loan process.
Neither the museum nor I could have managed
without her.
Recognition of individuals at the Cleveland
Museum of Art must start with the curatorial
 assistants who have formed the other half of
the Department of Pre-Columbian and Native
North American Art. Lisa Simmons deftly
got the project under way before she began
walking north on the Appalachian Trail.
Amanda Mikolic immediately stepped in,
and her masterful organization, efficiency,
and unruffled calm have helped to keep both
me and the project on track, even in times of
trouble. I thank them both affectionately.
I am particularly grateful to Director David
Franklin and Deputy Director C. Griffith
Mann for their courageous and unwavering
support. Several factors made the catalogue
a challenge to bring to fruition, but editor
Barbara J. Bradley met the challenge with
great professional expertise, intelligence,
and patience; her efforts were abetted by
the skillful copy editing of Kathleen Mills.
Thomas Barnard is responsible for the clarity
of many of the catalogue’s graphics and the
elegance of its design, which he accomplished
with characteristic flair and good humor.
Howard Agriesti and David Brichford worked
miracles with many of the photographs that
appear in the following pages. The daunting
task of arranging loans from nearly fifty
national and foreign lenders fell on the
strong shoulders of Mary Suzor’s Collections
Management staff, especially Kim Cook.
The conservators who, under Marcia Steele’s
leadership, brought their professionalism and
helpfulness to bear include Colleen Snyder,
Samantha Springer, and especially Robin
Hanson, whose efforts on the project’s behalf
have been extensive and exceptional. James
Englemann, under Jeffrey Strean’s direction,
created the exhibition’s poetic design, an
effort that our mount makers, graphic design
team, and label editors complemented. In the
x iii
Education Department, Caroline Goeser has
been a valued colleague and ally; her dedicated
team created imaginative programming to
improve the public’s understanding of Wari,
an effort in which the Textile Art Alliance
energetically participated. John Ewing and
Massoud Saidpour infused the past with
the present through inventive contemporary
music and film programming. Heidi Strean
and her staff in the Exhibition Office—first
Sheri Walter, then Sarah Otto—ably handled
myriad organizational details and responded
with aplomb to emergencies. Fundraising was
adeptly orchestrated by members of August
Napoli’s department, including Marianne
Bernadotte, Kathy Rowe, Achala Wali, and
Cindy Flores, who stayed late to beat the
deadlines. Librarians Betsy Lantz, Louis
Adrean, and Matthew Gengler were fast on
their feet in providing research materials and
assistance; they are consummate professionals.
In marketing and communications, Elizabeth
Bolander conducted audience research and
oversaw marketing outreach with typical
intelligence and focus. John Baburek went
to much effort to stock the gift shop with
culturally appropriate offerings.
Finally are the family and friends who
supported me during the demanding process
of transforming the idea into reality. To
all, including nieces who were especially
boisterous and refreshing, I am sincerely
grateful for encouragement, advice, patience,
and help both material and intangible.
Susan E. Bergh
Curator, Pre-Columbian and Native North
American Art
The Cleveland Museum of Art
 E C U A D O R
P I URA
MARAÑON

A N
N
Piura

O
AÑ

D E
AR
M

A N
I
AL
AY
UC

C
H
O
TA

M O
UE

N
EQ

O
AY
MB
LA

U N
San José de Moro Cajamarca

T A
JEQUETEPEQUE

MA

I N
ICA
CH
Huamachuco
E

S
CH
Chan Chan MO Viracochapampa B R A Z I L
Trujillo Moche
MA
RA
ÑO

N TA
N

SA

Honco Pampa
CASM
A Chavín de Huántar
Huaráz

UCA
El Castillo

YA L
HUAR MEY

I
A
LC
VI
TI
PA

Chimú Capac

P A C I F I C
P E R U
M
A
N
TA

Ancón
RO

O C E A N AC
UR

Lima RÍM
UB

Huaca Pucllana
ÍN Huancayo
AM

LU R
Pachacamac Wari Wilka
BA

A
AL
M

Huaca E Wari Espíritu

Malena ET SAN Pampa
Ñ Azángaro
CA MIGUEL
R PA

VALLEY
Jargampata
HUA

Ayacucho
Conchopata
Maymi CO
PIS PA M
PA S
ICA Cuzco
A Pikillacta
V ILC

N Huaro
D
ANO
DE

E Pomacanchi
AN

A
TA

Jincamocco
N
GR

IO
EN
ING
NASCA
Pataraya
Nasca M
Pacheco
AC A
RI O
U
N
A

Acarí
C

T
U

x iv Northern and C entral P eru

YA

Pucara
Corral A
Redondo I
N
LC
A S
A

CO
S
OÑ

JE
 ON
N TA
SA

Honco Pampa
CASM
A Chavín de Huántar
Huaráz

UCA
stillo

YA L
HUAR MEY

I
A

B R A Z I L
LC
VI
TI
PA

mú Capac

P E R U
M
A
N
TA

Ancón
RO

Colombia
AC Ecuador

UR
Lima RÍM

UB
Huaca Pucllana
ÍN Huancayo Brazil

AM
LU R Peru
Pachacamac Wari Wilka

BA
A
AL Bolivia
M

Huaca E Wari Espíritu

Malena ET SAN Pampa Chile
Ñ Azángaro
CA MIGUEL
R PA

VALLEY
Jargampata
HUA

Ayacucho
Conchopata
Maymi CO
PIS PA M
PA S
ICA Cuzco
A Pikillacta

V ILC
N Huaro
D

ANO
DE

E Pomacanchi
AN

A
TA
Jincamocco
N
GR

E NIO
ING
NASCA
Pataraya
Nasca M
Pacheco
AC A
RI O
U
N
A

Acarí
C

T
U
YA

Corral A Pucara B O L I V I A
Redondo I
N
LC
A S
A

CO
ES
OÑ

AJ
OC

LAKE
M

Puno
TI TI C AC A
Á
AN

Arequipa
OR
M

VIT
CA

La Paz
P A C I F I C Tiwanaku

Moquegua Cerro Baúl

A
U
EG
U

O C E A N
Q

TAC NA
O
M

Wari site
Site where Wari-style
materials were found
Ancient, non-Wari site San Pedro
Arica
N de Atacama
Modern city AZAP (500 kilometers)
A

200 kilometers
C H I L E

x v C entral and S outhern P eru

Luis G. Lumbreras
Introduction
When the Spaniards landed in Peru in 1532,
they found the Inca Empire, which encom-
passed a vast territory extending south from
southern Colombia through all of Ecuador,
Peru, and Bolivia, two-thirds of Chile, and the
northwestern wedge of Argentina. The Inca
referred to their domain as “Tawantinsuyu”
(The Land of Four Parts), and the image of
their empire soon predominated European
thinking about the ancient Andes, often
obscuring essential questions and affecting
Western thinking about the nature and origins
of civilization in the region. Indeed, the con-
quistadors tended to picture Tawantinsuyu as
a kind of replica of the Roman Empire.
In the centuries following the conquest,
understanding Inca origins was no easy task
since indigenous writing systems never devel-
oped in the Andes. The only reports available
are accounts the Spaniards created based on
their own eyewitness observations along with
testimony from the Inca and other natives who
had survived the conquest and the scourge of
diseases Europeans introduced. According to
the origin myths these natives recounted, the
Inca arrived in Cuzco, their capital in Peru’s
southern highlands, from the shores of Bo-
livia’s Lake Titicaca, near which the Tiwanaku
civilization had developed in pre-Inca times.
The obvious inference was that the Tiwa­naku
must have been the Incas’ ancestors. (The
creation myth also held that nearly all Inca
deities had sprung forth from the depths of
the lake, which natives therefore regarded as
a pacarina or “birthplace.”) This conceptual
model persisted from the time of the conquest
until the early twentieth century, when ar-
chaeological research began to reveal a much
longer period of occupation and provide a
better picture of the development of civiliza-
1
tion in the Andes. We now know that the Inca
were a brief, 100-year episode that culminated
a complex history of settled life reaching back
to about 3000 BC. We also now know that dur-
ing the period today known as the Middle Ho-
rizon (AD 600–1000), Tiwanaku was not alone
in the Andean cultural landscape. Rather, it
had a powerful contemporary—Wari (see pp.
5–27, “The History of Inquiry into the Wari
and Their Arts”).
Given Tiwanaku’s prominence in colonial
records as well as the renown of its fine stone
sculpture among early explorers and travelers
(see fig. 6a), it is not surprising that some of
the earliest modern archaeological research,
conducted in the late nineteenth and early
twentieth centuries, focused on the Tiwanaku
capital city. Important in this regard is the de-
scription that German archaeologist Max Uhle
published of Tiwanaku in 1892, based on the
research conducted by his fellow countryman
Alphons Stübel.1 Uhle went on to discover that
Tiwanaku-like influence had registered at sites
in the western Pacific coastal regions of Peru.
During excavations he conducted in 1896
at Pachacamac, the sprawling archaeologi-
cal site on the outskirts of modern-day Lima,
Uhle unearthed Tiwanaku-like ceramics and
textiles that stratigraphically predated any
Inca remains. He thus definitively established
the greater antiquity of the Tiwanaku-like
style and confirmed its spread to the central
coast of Peru.2 Some years later, Wendell Clark
Bennett conducted excavations at Tiwanaku,
again asserting that its development had pre-
ceded the Inca. Bennett formulated a cultural
timeline, including a “Classic” period that led
to a Tiwanaku expansion period and finally to
the Inca, as Uhle had originally surmised.3 Ti-
wanaku occupied the middle of the timeline,
 and all indications thus tended to confirm the
role of this culture as a source for the Incas.
But Bennett also excavated at a huge site
in Ayacucho, Peru, that is now recognized as
the Wari capital, there discovering ceramics
in a style then termed “Tiwanakoid” because
their iconography is similar to that found on
Tiwanaku artifacts. Based on this and other
evidence, the archaeologists Rafael Larco
Hoyle and Julio C. Tello hypothesized that the
Ayacucho site—not Tiwanaku—had served
as the source of “Tiwanakoid” incursions
along the coast and in other regions of Peru,
a pattern of far-flung distribution suggesting
the existence of an empire that predated the
Inca.4 This idea of empire, however, did not
gain major acceptance until the 1960s. The
ensuing decades have been filled with intense
archaeological research bolstered by advances
in scientific dating and the study of archaeo-
logical remains.
This research has revealed that an agricul-
tural people known as the Huarpa, after the
name of the region’s main river, emerged be-
tween the second and fourth centuries AD in
the Ayacucho Valley, the Wari heartland. By
the fifth century, the Huarpa had converged
along the forty-five-mile length of the valley
and established a unified realm between the
Mantaro and Pampas Rivers. The Wari culture
coalesced from these humble Huarpa roots.
As it did, its development focused on two
key settlements in the Ayacucho region: the
Wari capital, also known as Wari, and nearby
Conchopata.
By the middle of the sixth century, virtu-
ally all of present-day Peru was undergoing
a general crisis triggered by environmental
perturbations, including drought and floods.
These conditions affected all cultures in the
region and led to a period of great social up-
heaval as the Moche, Lima, and Nasca peoples
of the Peruvian coast as well as the highland-
ers of the Cajamarca, Huaylas, and Titicaca
regions migrated to other areas or struggled
and competed to increase their water and food
supplies. It was at this time that Wari began
its initial expansion, which apparently arose
from the need to obtain more land in the face
of the climate crisis. In my view, current infor-
mation about this period points to a long-term
2 Luis G. Lumbreras
Wari invasion of many areas of the Andes. Per-
haps initially peaceful, these campaigns soon
became hostile in nature; waves of Ayacucho
warriors may have swept across the region. On
the north coast, the Moche civilization had en-
tered a decline marked by tremendous change
that Wari influenced (see pp. 47–61, “Look-
ing at the Wari Empire from the Outside In”).
The central coast Lima culture, which had
undergone considerable earlier local develop-
ment, was also beset by strong Wari influence,
particularly at Pachacamac. In the Nasca
region on the south coast, an earlier period
marked by internal violence and upheaval
(between the Nasca III and Nasca V phases)
led to the collapse of the Nasca culture and
its final assimilation by the Wari in the early
stages of their existence. This process also oc-
curred in the highlands, from Cajamarca and
Huamachuco in the north to the Vilcanota and
Cuzco Valleys, the later domain of the Inca in
the southern highlands. Even farther south,
Wari’s incursion into Tiwanaku territory first
occurred in the Moquegua Valley. The Wari
may even have attempted to penetrate into
Tiwanaku’s highland territory on the Boliv-
ian altiplano; if so, they were turned back
and settled along the Chuquibamba and Colca
Rivers in the southern Peruvian sierra, on the
outskirts of Tiwanaku’s realm. There is also
evidence of significant interaction between
Wari and central Ecuadorian groups, as well
as between Wari and cultures of the Tacna and
Arica regions of Chile.
Wari’s direct contact with these cultures
allowed it to absorb, broaden, and refine a
range of manufacturing techniques from the
Nasca, the Lima, the Tiwanaku, and others.
Important among them were ceramic tech-
nologies, but also others. The Wari also likely
came into contact with new customs and
beliefs that furthered their development. The
process of expansion also led to the evolution
of more complex social structures among the
Wari. The result was an empire that extended
across virtually the entire territory we now
call Peru. This empire had many of the same
characteristics as the Inca Empire: a major
capital city in which power was concentrated
as it was in Cuzco, the Inca capital; adminis-
trative centers of various sizes throughout the
 region; the Wariñan (Wari road), a network of
roads like the Inca road system, which linked
important nodes throughout the empire and
facilitated transportation and communication;
a powerful army; an elaborate state-sponsored
religion. Like the later Inca, the Wari Empire
also seems to have maintained state-supervised
workshops for the production of textiles and
ceramics, employed a system of recording and
accounting based on the use of khipus (fiber
recording devices), and created symbols of
power in the form of architecture and the fine
works of art featured in this catalogue. Some
modern scholars also believe that the Wari
had an institution of compulsory rotating
labor service, known among the Inca as m’ita
(see pp. 251–67, “Wari’s Andean Legacy”).
This empire came into being about a mil-
lennium before the Inca; some suggest that it
may not have entered into a final decline until
the Inca began their rise to prominence in
the Cuzco area in the twelfth and thirteenth
centuries. Some interpreters believe it is pos-
sible that, between the eleventh and twelfth
notes 1. Stübel and Uhle 1892.
2. Uhle [1903] 1991.
3. Bennett 1934.
4. Larco Hoyle 1948; Tello [1931] 1970.
3 I ntroduction
centuries, the Wari transferred their seat of
government to Cuzco after the decline of their
Ayacucho capital metropolis. If so, perhaps
Manko Qhapaq and Zinchi Roq’a, the early,
legendary Inca rulers, witnessed the last ves-
tiges of the Wari, whom the Inca might have
conceived as their traditional enemies (chan-
cas). The Spanish chronicles tell us that it was
the defeat of the chancas that launched the
Inca Empire. Thus, the relationship between
the Wari and the Inca empires may be one of
continuity rather than simple antecedence.
Whether or not this was the case, most
researchers now agree that Wari was the cen-
ter of an archaic Peruvian state, the vestiges
of which are still being discovered and de-
scribed. Although much research remains to
be conducted, particularly at the Wari capital,
it seems clear that Wari was a precursor of
Tawantinsuyu, and perhaps its direct fore-
runner. The essays in this catalogue, which
accompanies a landmark exhibition, offer a
range of perspectives on this important period
in history.
Susan E. Bergh and
Justin Jennings
Figure 1 [16]. Urn with
staff deities, from Pacheco
(for details, see fig. 5);
ceramic and slip; 83.5 x
86 cm. Museo Nacional de
Arqueología, Antropología
e Historia, Lima, S/C.
Photo: Daniel Antonio
Giannoni Succar.
5
The History of Inquiry into the Wari
and Their Arts
The historian-soldier Pedro de Cieza de León
was about thirty years old when, in 1547, he
set out on a historic journey: a trek by foot
and horseback into the vast, rugged territory
of Peru, then a Spanish colony that spanned
much of an ancient Amerindian culture region
known today as the central Andes (see maps,
pp. xiv, xv). Traveling with fellow Spanish
troops, Cieza entered Peru from the northern
reaches of the Andes mountain range, skirting
misty precipices and passing over plunging
gorges on roads and bridges built by the Inca,
whose empire had shattered violently just a
few years earlier. He then descended to the
stark, dun-colored desert that blankets the Pa-
cific shoreline, fording the rivers that genera-
tions of Indians had used to coax the parched
land into bloom, before turning again into the
mountains. During his journeys, he gathered
irreplaceable eyewitness information about
the natives and their lands that he published
in his legendary Crónica del Perú, the first
part of which has been called a “Baedeker’s
guide” to sixteenth-century Peru.1
In about 1548, Cieza visited an abandoned,
ruined city some 2,800 meters (about 9,200
feet) above sea level in the Guamanga Valley,
today’s Ayacucho Valley, an arid region in
Peru’s central highlands where agriculture is
poor and Andean camelids (llamas, alpacas,
and others) graze on the slopes of snow-
capped peaks. Confronted with this mysteri-
ous ruin, Cieza used the instincts of an ama-
teur archaeologist, along with the testimony of
Quechua-speaking natives whose voices many
earlier Spanish chroniclers had neglected, to
draw insightful conclusions.
The greatest river of the region is called
Vinaque, where there are some large and
very ancient buildings, which certainly
because of their weathered and ruinous
condition must have lasted through many
ages. Questioning the neighboring Indians
as to who built that ancient place, they
answer that it was other bearded, white
people like us, who, a long time before the
rule of the Incas, are said to have come
to this region and made their residence
in it. And in regard to these and other
ancient buildings in this realm, it seems
to me that their plan is different from
those [structures] which the Incas built or
ordered built, for this building was square
and Inca buildings are long and narrow.
There is also a tale that writing was found
on a stone slab of this building . . . in my
opinion, people of great skill and reason
arrived here in ancient times, and they
built these things and others that we do
not see.2
Cieza’s is the first written note about the
great capital city of the Wari (also spelled
“Huari”), a term that today refers to a people
and a culture that flourished about AD 600–
1000, a period known as the Middle Horizon.
Specialists now recognize that the capital,
also known as Wari, is the center of one of the
largest archaeological zones in South Ameri-
ca.3 But astonishingly, despite being familiar
to Ayacucho natives, it disappeared from the
published record for nearly four centuries af-
ter Cieza’s brief notice. Not until the twentieth
century did pioneering archaeologists, first
from Peru and then also from other countries,
begin to describe the city and the “people of
great skill and reason” who, centuries before
the Inca, wrote an ambitious chapter in both
world history and the pre-European history of
the Western Hemisphere.

Figure 2. Vicuñas grazing
in the high, wet grasslands
(bofedales) of northern
Chile. Photo: Susan E.
Bergh.
6 S usan E . Bergh and J ustin J ennings
Although many strides have been made
in recuperating Wari, the culture remains
surprisingly little known outside Peru, where
the Pre-Columbian past is a part of daily life
through media reports of new archaeologi-
cal discoveries and a landscape littered with
ancient monuments. This catalogue and the
accompanying exhibition—the first in North
America and only the second in the world of
its kind4—are correctives, organized in the
belief that Wari and its unsung achievements
have interest far beyond Peru. For they tell
the story of a people of modest beginnings
whose charismatic leaders, without the aid
of a system of writing and far from the influ-
ence of the Old World, created a cosmopolitan
civilization that many now regard as one of
the Americas’ first empires. In some respects it
is a familiar story of human struggle, ingenu-
ity, and vision, the latter perhaps more a result
than the impetus of an effort that at times
must have been contingent and scrambling. In
other ways, however, Wari’s habits of thought
and action are foreign and provide intrigu-
ing, fresh perspectives on our own. This may
be especially so of the arts, which, as among
other pre-industrial cultures, were central
to Wari’s political system, religion, economy,
and technologies: they commanded enormous
resources and intellectual investment, served
as forms of wealth and power, promoted the
spread of religious and political ideas, and in
the absence of writing functioned as durable
forms of communication.5 The arts, then, were
not passive aesthetic by-products, as they
are too often regarded today, but active tools
that abetted human achievement. If, as some
suggest, the Wari succeeded in part by posi-
tioning themselves to outsiders as powerful
mediators of earthly and cosmic affairs,6 the
arts were crucial to giving reality to the claim.
As Cieza discovered, the territory in which
Wari developed is spectacularly diverse in
geography, climate, and natural resources.
The Andes—after the Himalayas, the highest
mountain range in the world—comprise two
parallel chains of peaks that slice from north
to south, dividing the land into three broad
zones. The extreme vertical topography of the
Peruvian sierra itself, where Wari’s capital and
provincial outposts lie, creates remarkable
environmental diversity, with distinct eco-
logical niches that vary according to altitude.
Between elevations of 4,000 and 5,000 meters
(13,000 and 16,000 feet) are high, cold grass-
lands almost deserted except by camelids,
including the wild vicuña and guanaco as
well as herds of domesticated llamas and
alpacas valued for their meat and their silky
hair, which in antiquity was woven into fine
textiles (fig. 2). Below, on land that plunges
into deep river valleys, more moderate climate
and the annual cycle of highland rain make
agriculture possible; here, both today and
in the past, quinoa, hundreds of varieties of
potatoes, and other tubers thrive at higher alti-
tudes, giving way at lower elevations to fields
for maize and various kinds of beans, squash,
chili peppers, and fruits (fig. 3). Major mar-
kets never developed in the Andes because of
the proximity of these diverse zones; ancient
settlement commonly occurred at elevations
that allowed easy access to all.
The western coast, where many Wari
objects have been found in graves and buried
deposits, is a nearly rainless desert where
life was sustained by the sea’s abundance
and crops irrigated in the oasis-like valleys
of freshwater rivers that flow down from the
 highlands (fig. 4). Here ancient farmers grew
maize, squash, gourds, and such fruits such
as the creamy chirimoya and caramel-flavored
lúcuma, along with peanuts and, importantly,
cotton, the second fiber crucial to textiles. So
productive were these desert valleys that the
coast was always relatively densely populated
in antiquity, and it remains so today.
The lush green montaña of the eastern
Andean slopes, which fall steeply to the Ama-
zonian rainforest, was more marginal to Wari
although trade with both regions brought such
exotics as the brilliant feathers of tropical
birds into reach. Another important product
of this region was coca, used for its medicinal
properties, as an important ceremonial mate-
rial, and sometimes as a medium of exchange.
In places it is possible to travel from the
Pacific Ocean to the Amazon jungle in less
than 200 kilometers (125 miles), and this
juxtaposition of regions led early on to long-
distance exchange of products among regions.
These links were especially important in
the Andes, one of the most environmentally
unstable regions in the world. Prone to earth-
quakes, volcanic eruptions, landslides, and,
more rarely, tidal waves, Peru is most fre-
Figure 3. Sierra landscape
with modern agricultural
fields, Colca Valley, Peru.
Photo: Susan E. Bergh.
7 T he H istory of I nquiry into the Wari and T heir A rts
quently destabilized by El Niño and La Niña,
climatic disturbances that devastate marine
life and play havoc with precipitation pat-
terns, causing torrential rainfall on the coast
and droughts in the highlands. With resource
zones scattered and prone to periodic failure,
long-distance relationships were critical for
daily subsistence and also provided a safety
net when disaster struck in one locale.
Wari was preceded in this dramatic,
turbulent, and varied landscape by many
complex societies, the first of which took root
on the coast between about 3000 and 1800 BC,
when the first ceremonial centers appeared on
the coast and in the highlands. (The history of
simpler human groups in the Americas goes
back at least 13,000 years, when hunters fol-
lowed big game out of Siberia.) Thereafter, ma-
jor developments alternated regularly between
the coast and the highlands, focusing first on
one and then the other. Only during periods
of highland dominance, known as “horizons,”
were the geographic regions drawn together in
some way, first by the Chavín people during
the Early Horizon (1000 BC–AD 1), then by the
Wari and their powerful neighbors the Tiwa-
naku (also spelled “Tiahuanaco”; 600–1000),
Figure 4. Dunes in the Ica
Valley on the southern
reaches of Peru’s western
coast. Photo: Susan E.
Bergh.
and finally by the Inca of the Late Horizon
(1400–1532) whose empire, which expanded
to control most of the Andean world, fell to
Spanish forces after 1532.
A History of Wari Scholarship
archaeological research . The history of
Wari investigation, both art historical and
archaeological, is tightly bound up with Ti-
wanaku, the contemporary state-level soci-
ety whose large capital, replete with finely
cut, monumental stone sculpture, is near the
shores of Lake Titicaca on the altiplano (high
plateau) of modern Bolivia.7 Wari and Tiwan-
aku art and architecture are very distinct in
most respects, but curiously the two shared
an iconography so crucial to the period that it
functions as a signature: an imposing frontal
figure, its head radiating a halo of append-
ages, who is often flanked by winged acolytes
shown in profile and with features ranging
from human to fusions of the human and ani-
mal. All clutch staffs, a potent Andean symbol
of authority (figs. 1, 5, 6). The appendages, the
animal features (which often include a fanged
mouth), and a divided eye seem to mark these
creatures as supernatural beings—a powerful
“staff deity” and its minions.8
8 S usan E . Bergh and J ustin J ennings
Early investigators noted the presence of
this imagery at a great many coastal sites in
Peru and, faced with the question of its source,
pinpointed the Tiwanaku capital. The site
had been famous since the Inca and Spanish
colonial periods in part because of its theatri-
cal sculpture, which prominently features the
divine imagery and was well known by the
latter half of the nineteenth century.9 Thus,
they applied the name “Coast Tiwanaku” to
the Peruvian style, although they recognized
that Peruvian and Bolivian manifestations of
the imagery differ in several important ways
and some speculated that an unknown, more
northern center had been involved in dis-
persing the Peruvian version of the style. In
1931, Julio C. Tello, the most famous Peruvian
archaeologist of his era, discovered that north-
ern source during a vacation reconnaissance
trip to Ayacucho. The then-grueling trip from
the coast into the mountains was unusual
(most archaeologists of the period focused
their attention on the coast), and it was stimu-
lated by Tello’s lifelong interest in demon-
strating the importance of the sierra and the
eastern jungle in ancient Peruvian history.10
Once he reached Ayacucho, locals directed
him to the Wari site where he found ceramics
Figure 5a. Staff deities on
the interior of an urn from
Pacheco (for complete
object, see fig. 1). Museo
Nacional de Arqueología,
Antropología e Historia,
Lima, S/C. Photo: Daniel
Antonio Giannoni Succar.

Figure 5b. Staff deities

on the interior of an urn
from Pacheco. American
Museum of Natural
History, New York,
41.0/5314. Drawing: cour-
tesy American Museum of
Natural History, Division
of Anthropology.

that closely matched the Coast Tiwanaku style
in association with some of the city’s most
impressive buildings. The question of how the
site fit within the grand scheme of Peruvian
antiquity preoccupied Tello for the next ten
years, and in 1942 he suggested that Wari was
both the source of the Coast Tiwanaku style
and the capital of a newly christened Wari
civilization.11 Unfortunately, Tello was unable
to publish the results of his Wari excavations
before his untimely death in 1947, but the
potential significance of his discovery was not
lost on his peers.12
In “considerable confusion” over Tello’s
few published notes about his breakthrough,
a trio of young North American archaeolo-
gists—John Rowe, Donald Collier, and Gordon
Willey—visited Wari during a jeep trip across
the Peruvian sierra in 1946. After spending
only an hour or so at the site, they came away

9 T he H istory of I nquiry into the Wari and T heir A rts

 convinced that Ayacucho was “one of the key
areas in the development of Andean Civiliza-
tion” and that the city, which sprawls over
more than 200 hectares (500 acres), would
take “months to explore in its entirety.”13
They were particularly struck by similari-
ties between the architecture in one of the
city’s sectors and that of Viracochapampa and
Pikillacta, massive centers at opposite ends
of the Peruvian sierra with high walls, few
entrances, and groups of small rooms with
raised doorways that they thought may have
served as storage facilities (see fig. 40).
Although their article on the site was
largely dedicated to a preliminary classifica-
tion of its ceramics and sculpture, it is tempt-
ing to imagine the excited conversations that
took place in the jeep as the three men specu-
lated on the city and its art style, colonies,
and relationship to Tiwanaku. If this city was
indeed the source of the “Coast Tiwanaku”
10 S usan E . Bergh and J ustin J ennings
style, what was the relationship between it
and all of those places in Peru where the style
had been found? In other words, how did Wari
relate to the spread of Wari art? If all of these
sites were built by the same people, what
purpose did they serve? Wari was the last big
piece to be found in the puzzle of ancient An-
dean history, and how it fit in was to become
one of the field’s most hotly debated topics.
With a source site finally defined, scholars
began to place Wari within a master Andean
chronology, a project abetted by Willard
Libby’s revolutionary work with radiocarbon
dating. In the 1950s, John Rowe began a series
of coastal excavations designed to link broad
cultural phases with radiocarbon dates; in
1962 he published a chronology that applied
the term “Middle Horizon” to the period of
widespread Wari influence. His student Doro-
thy Menzel soon followed with monumental
studies that sorted Wari and Wari-influenced
 ceramics into a number of regional styles
and assigned the styles to temporal phases or
epochs; Wari’s apogee occurred during the
first two of these epochs, known as Middle
Horizon Epochs 1 and 2, both of which Men-
zel further divided into A and B subphases.14
Rowe then linked Menzel’s stylistic phases
to a series of radiocarbon dates running from
about 600 to 1000.15 With minor revisions,16
Rowe’s and Menzel’s chronologies remain
widely used today.
This work on chronology was a crucial
step forward but scholars had done little to
relate the Wari capital to the spread of its art
style. Researchers over the last half-century
have attempted to explain the development
of the Wari civilization by focusing on three
main areas. The first is the Wari capital and
the surrounding Ayacucho Valley. In work
since the 1950s but particularly during the
1960s, archaeologists who worked at Wari
itself—Mario Benavides Calle, Wendell Ben-
nett, Enrique Bragayrac Dávila, William H.
Isbell, and Luis G. Lumbreras, among oth-
ers17—have confirmed the size and splendor
of the city, which boasted as many as 40,000
inhabitants and contained wealthy neighbor-
hoods, monumental ceremonial architecture,
Figure 6a. The Gateway
of the Sun at Tiwanaku.
Photo: Fernando Sánchez.
Figure 6b. Drawing of the
central portion of the lintel
of the Gateway of the Sun,
which is carved with the
image of the staff deity and
its profile winged atten-
dants. After Donnan 1992,
83, fig. 150.
11 T he H istory of I nquiry into the Wari and T heir A rts
and now-looted megalithic tombs that were
probably the resting places of Wari royalty.
The city abounded in exotic materials, such as
ritually charged Spondylus shell from Ecua-
dor, and many kinds of specialists (ceramists,
sculptors, weavers, metalworkers, and others)
transformed these materials into highly val-
ued works of art. Research in the surrounding
valleys demonstrated that as villagers moved
into the city the Wari converted the country-
side into a breadbasket: urban officials di-
rected the construction of elaborate irrigation
and field systems, and created administrative
compounds to organize the agricultural sur-
plus needed to feed the growing metropolis.18
The Sendero Luminoso (Shining Path)
guerrilla movement disrupted archaeological
work throughout Ayacucho from 1980 to 1992,
including at the Wari capital. When work re-
sumed, it concentrated on reconstructing the
lives of those who lived in the area during the
Middle Horizon, for example, the quotidian af-
fairs of farmers and the mortuary customs that
continued despite the great changes sweeping
the Ayacucho Valley.19 Most ambitious was
the project focused between 1999 and 2003 on
Conchopata, the largest Middle Horizon site in
Ayacucho aside from the Wari capital itself.20
Figure 7. Wari warriors
wearing tapestry-woven
tunics, from an urn recov-
ered at Conchopata. The
tunics are decorated with
profile-face and stepped-
fret motifs. See also fig.
10. After Seville 2001, 200,
203; drawings: José A.
Ochatoma Paravicino.
10 centimeters
Previous excavations had shown that it was an
important ceramic production center,21 but the
more recent excavations also found surpris-
ing wealth at the site in terms of architectural
elaboration and sumptuary goods.22 Excavators
were particularly struck by the militaristic
iconography found on some of the pottery (fig.
7), a theme underlined by the discovery of a
cache of trophy heads made from the skulls of
children born outside of Ayacucho.23
The second issue that has drawn archae-
ologists’ attention is provincial sites with Wari
style architecture of which more than two
dozen had been tentatively or definitely identi-
fied by the end of the 1990s.24 Among the latter
are Viracochapampa, a site in the northern
highlands that Wari abandoned unfinished,25
Figure 8. Site plans of two
reported Wari outposts that
have proved to be local,
non-Wari settlements.
Plans: Justin Jennings.
Numero 8, Sector C
Chuquibamba Valley N 10 meters
12 S usan E . Bergh and J ustin J ennings
10 centimeters
along with Pikillacta, Jincamocco, and Cerro
Baúl, all also in the highlands but to the south
of the capital. Extensive surveys and excava-
tions at these three southern sites have re-
vealed the footprint of Wari colonization and
prove that at least in some places Wari settlers
founded impressive centers from which they
administered the affairs of local populations.26
In other cases, however, suspected Wari af-
filiations have not proved true (fig. 8).27 Some,
for example, were occupied much later,28 and
excavations at others revealed that they had
been built and occupied not by the Wari but
by locals.29 There are also many places in Peru
where Wari or Wari-influenced artifacts are
plentiful but its provincial settlements are ab-
sent. Among them is the Chicha-Soras Valley
CLI
FF F
AC E
Achachiwa
Colca Valley N 100 meters
Figure 9. A long-haired
suri alpaca. Photo: Susan
E. Bergh.
in the central highlands, which the Wari may
have administered indirectly in order to gain
access to and maximize the area’s valuable
production of llama and alpaca fiber, some of
it no doubt put to use in the making of the fine
textiles in which the Wari specialized (figs.
2, 9).30 The same kind of indirect control has
been suggested for other regions.31 Wari arti-
facts also seem to have come into some places
as trade goods;32 in still others, they are rare or
absent altogether.33
How can this variable distribution of Wari
style architecture and artifacts be explained?
Early on, Menzel,34 Lumbreras,35 and others
embraced the notion of empire; this view was
corroborated as fieldwork in the 1970s and
1980s made it clear that the capital city had
both transformed the surrounding valleys and
established a number of settlements through-
out Peru. The empire model, however, was not
fully articulated until Katharina Schreiber
published Wari Imperialism in Middle Hori-
13 T he H istory of I nquiry into the Wari and T heir A rts
zon Peru in 1992. She suggested that the Wari
established a “mosaic of control”: they con-
quered much of Peru and built administrative
centers in some regions in order to establish
direct control, but elsewhere they manipu-
lated groups through local leaders36 or left
locales completely unmolested.37 Schreiber
noted that Wari’s varied impact through space
and time was in keeping with the ways in
which other empires adapted their strategies
to locally varying conditions.
The marked variability in Wari’s archaeo-
logical footprint has for many years fostered
resistance to the empire interpretation in some
scholarly quarters,38 but few viable alterna-
tive models have so far been offered. Richard
Schaedel, for example, proposed vague “vec-
tors” of cultural transformation to explain the
distribution of Middle Horizon styles,39 while
Ruth Shady Solís’s counter-model ignored
the evidence for provincial sites and sug-
gested that Wari was only one of many equally
Figure 10 [118]. Tunic with
face-fret motif; camelid
fiber and cotton; 110.5 x
118.1 cm. Museum of Art,
Rhode Island School of
Design, Providence, Mary
B. Jackson Fund and Edgar
J. Lownes Fund, 40.007.
Photo: Erik Gould, cour-
tesy Museum of Art, Rhode
Island School of Design.

Figure 11 [147]. Four-
cornered hat with mythical
creature; camelid fiber;
14 x 12 cm. Staatliches
Museum für Völkerkunde,
Munich, 58-20-233 (NM
223).
important cities that exchanged goods and
ideas during the period.40 More recent models
have attempted to downplay Wari’s dominance
while acknowledging the pivotal role that the
capital city had in the surging interregional
interactions of the Middle Horizon.41
The third topic that has captured research
attention remains one of the most vexing in
ancient Andean studies: the Wari-Tiwanaku
relationship. As mentioned above, the two
cultures shared essential iconography—the
staff deity and its winged acolytes, along with
tapestry-woven tunics and four-cornered hats
(figs. 10, 11). But they did not engage in other
types of exchange, even along their frontiers
in far southern Peru. For instance, in the
Moquegua Valley, the only region where their
occupations overlapped, there is little evi-
dence of interaction until the very end of the
Middle Horizon.42 Also, the Wari inhabitants
of Huaro and Pikillacta, two other frontier
towns near Cuzco, seem to have been disin-
terested in nearby Tiwanaku communities.43
Thus, although the two cultures’ chronologies
cannot be correlated in any tight way and the
origin of the staff deity iconography is still a
matter of debate, few experts doubt that Wari
and Tiwanaku were independent polities with
distinct territorial spheres of operation and
influence.
Certainly their approaches to architecture,
art, site planning, and public ceremonial life
were radically different. For example, Tiwa­
naku architects excelled in the use of exqui-
site, precisely cut stone masonry, which they
used to face massive mounds and sunken
courts, while the Wari built grand, high-
walled compounds with roughly dressed stone
and copious mortar (see fig. 39). In art the
Wari showed little interest in inscribing the
staff deity and its companions on monumen-
tal stone gateways and sculpture, a hallmark
of Tiwanaku artistic production (fig. 6), and
instead transferred it to small, finely made,

14 S usan E . Bergh and J ustin J ennings

Figure 12 [91]. Three
plumes with staff deity
head, from Pomacanchi;
silvered copper; 34.8 x 10
cm, 40 x 13.8 cm, 36.6 x 12
cm. Princeton University
Art Museum, Gift of
Leonard H. Bernheim Jr.,
Class of 1959, 1982-27,
1982-29, 1982-28. Photo:
Bruce M. White.
portable arts—textiles, ceramic vessels of
various kinds, personal ornaments of precious
materials, and sculptured objects that nestle
in the hand (figs. 12–16). Further, while the
spacious Tiwanaku capital seems designed to
accommodate crowds of pilgrims, the plan-
ning of Wari sites limits public gathering and
ritual movement to a remarkable degree. These
phenomena imply that Wari did not rely on a
centripetal strategy of drawing people into its
centers to witness great civic spectacles honor-
ing the staff deity and its winged attendants,
who perhaps stood as an idealized metaphor
for Wari rulers and society. Rather, the Wari
sent the image of the deity out from its high-
land capital into far-flung territories on objects
and textiles.44
art history’s neglected role . In numbers
of studies and the breadth of their collective
focus, the exploration of Wari arts, and of
ancient Andean arts in general, has lagged
15 T he H istory of I nquiry into the Wari and T heir A rts
behind archaeological investigation and today
remains nascent. The complex suite of reasons
for this disparity has its origins in the atti-
tudes and rationales of the colonial period.45
Among those factors that continue to have
impact today is the lack of indigenous writing.
Although Andean people developed the khipu,
a fiber mnemonic device used to record and
recall numerically enciphered information
(see fig. 180 and [155], p. 276), they are unique
in the world in having created great civiliza-
tions without the aid of true writing systems.46
But this remarkable fact seems to have been
prejudicial to study rather than encouraging
it, despite the implication that the arts played
an enhanced role in communication. The pau-
city of information about the arts in Spanish
colonial documents compounds the problem.
Also, the objects and media that the Wari and
many other ancient Andeans emphasized—
not paintings or monumental stone sculpture
but intricate garments of camelid fiber and
Figure 13a [165]. Cup with
staff-bearing creatures
in profile; wood; 11.4 x
6.4 cm. The Metropolitan
Museum of Art, New
York, The Michael C.
Rockefeller Memorial
Collection, Purchase,
Nelson A Rockefeller Gift,
1978.412.214. Image: ©
The Metropolitan Museum
of Art. Image source: Art
Resource, NY.
Figure 13b. The drawing
shows the two different
types of figures on the cup.
After Lapiner 1976, 246,
fig. 569.

cotton, useful clay or wood containers, pre-
cious personal ornaments today classified as
jewelry, and others (figs. 10–18)—lie outside
modern Western definitions of fine art, fall-
ing instead under the rubrics of decorative art
or craft, which hold inferior rank in Western
artistic hierarchies because we pejoratively
affiliate them with skilled hand labor and the
mastery of materials rather than with genius
and originality of intellectual conception.47 As
one contemporary art critic revealingly com-
mented in this regard, “hands cannot contem-
plate.”48 (The West has often associated craft
and the decorative with the racialized other,
such as Amerindians, as well as with feminin-
ity.49) The small scale of Andean objects has
also worked against them, particularly in the
modern museum settings that are now their
main venue but distance them behind glass in
spacious, high-ceilinged galleries that often
overwhelm them. Their subject matter too is
a barrier, for while their flashy beauty invites

16 S usan E . Bergh and J ustin J ennings

Figure 14 [56]. The head
of the staff deity appears
at the rounded ends and
winged attendants appear
on the sides of this vessel.
Container with staff deity
head and profile winged
creatures; ceramic and
slip; 16.6 x 16.9 x 21.8 cm.
The Cleveland Museum of
Art, Norman O. Stone and
Ella A. Stone Memorial
Fund 1999.2.
and beguiles, these objects present no obvious
heroic or dramatic narrative in the manner
of many other ancient arts, such as those of
Greece, Rome, Egypt, the Near East, and even
Mesoamerica.
Esther Pasztory has recently addressed
these and other factors—in James Clifford’s
words, the “powerful discriminations made
at particular moments”—that have affected
the reception of Andean arts in the modern
West.50 By taking the objects on their own
terms or at least on terms closer to their own,
she sees their potential to challenge our
expectations and broaden our ideas about art
and about ancient thought and capacity. For
in Andean hands the things that the West
regards as minor became major arts invested
with intellectual originality that informs not
only subject, style, color, and composition
but also manufacture and structure, which
ancient peoples seem to have endowed with
meanings that are unfamiliar in the West
and, in partial consequence, still faintly
understood.51 In other words, Andean arts
level the distinction between fine art and
craft, assimilating them into one and mak-
ing the categories irrelevant. Indeed, ancient
Andean languages seem to have had no words
17 T he H istory of I nquiry into the Wari and T heir A rts
for “art” and “craft,” which are conceptual
inventions of the Italian Renaissance.52 One
crucial implication of Pasztory’s approach is
that Wari arts must be imagined in the small-
scale contexts in which they were used: worn,
carried, manipulated, and touched; animated
by speeches, poetry, song, and conversation;
examined, admired, and surely critiqued in
lively ways. She believes that the insistence on
the personal and intimate reflects a cultural
ideal that emphasized face-to-face contact and
interaction,53 which allowed participants to
build networks of friendly or grudging al-
liance based on the mutual obligation and
reciprocity that were likely major elements of
Wari statecraft.
Of all the categories of Wari art produc-
tion, tapestry-woven tunics (shirts), the
raiment of Wari rulers and other elites, have
received the lion’s share of scholarly attention
(figs. 10, 15). Several factors likely account for
this focus, among them that, by the mid-1900s,
scholars recognized that the tunics were a
hallmark of the period and that fiber was a key
Andean artistic medium, a point John Murra
affirmed resoundingly in his milestone study
of the economic, political, social, and religious
importance of cloth among the Inca.54 Also,
Figure 15 [115]. Tunic
with sacrificer-related
creature; camelid fiber
and cotton; 105.4 x 114 cm.
Pre-Columbian Collection,
Dumbarton Oaks Research
Library and Collection,
Washington, DC, PC.B.496.
Image: © Dumbarton Oaks,
Pre-Columbian Collection,
Washington, DC.
18 S usan E . Bergh and J ustin J ennings
the tunics are by far the largest, most com-
positionally complex, and most technically
intricate objects that Wari artists created. As
collecting increased feverishly during the
twentieth century, tunics became plentifully
available for study.
Academic interest coincided with and in a
general way may have been encouraged by the
tide of enthusiasm for the tunics among col-
lectors and nonspecialists. This appreciation
stems in part from the parallels between a few
of the tunics’ features and Western modernist
painting, particularly a figural style that re-
calls Modernism’s geometric abstractions. For
instance, seeing in the tunics’ “abstraction”
the potential to “recall the esthetic problems
of our time,” René d’Harnoncourt singled the
tunics out for comment in his introduction to
the catalogue for the Museum of Modern Art’s
Ancient Arts of the Andes, a 1954 exhibition
Figure 16 [67]. Vessel with
staff deity head; ceramic
and slip; 21.4 x 16.6 x 11.2
cm. Fowler Museum at
UCLA, Los Angeles, Gift
of Mr. and Mrs. Herbert L.
Lucas Jr., X90.488. Photo:
Don Cole.
Figure 17 [133]. Overleaf,
Tunic; camelid fiber; 83.8 x
121.9 cm. The Metropolitan
Museum of Art, New York,
Gift of Arthur M. Bullowa,
1980, 1980.564.2. Image: ©
The Metropolitan Museum
of Art. Image source: Art
Resource, NY.
19 T he H istory of I nquiry into the Wari and T heir A rts
important for the fine-arts imprimatur that
it gave to material before regarded mainly as
the domain of anthropology or natural history
museums.55 The comparison to Modernism
may be promoted by the size and square shape
of the tunics, which resemble easel paint-
ings when mounted flat and within frames (a
method of display that many collectors today
prefer), as well as a compositional struc-
ture based on a grid, which has been termed
emblematic of the modernist ambition in the
visual arts.56
Against this background, it is fitting that
the first major art historical study devoted to
any Wari art form concerned the tunics and
their “abstraction.” Written in 1963 by Alan
Sawyer, then director of the Textile Museum
in Washington, DC, this splendid, article-
length analysis (which “mis-attributed” Wari
tunics to Tiwanaku during a period of shifting
understanding) briefly summarized the tunics’
highly standardized features. The article then
turned to the description of an equally rule-
bound convention for distorting form that is
based on the lateral manipulation of propor-
tions: the parts of each motif closest to the
garment’s center are elastically expanded or
stretched from side-to-side and the elements
proximate to the tunic’s sides are compressed
or narrowed. At its most extreme, this cerebral
practice transforms motifs into barely legible,
geometric stylizations that often require the
help of an expert guide to decipher (fig. 15).
Today, one such tunic is sometimes celebrated
as an abstract masterpiece (see fig. 174); that
the Wari might not agree is suggested by the
fact that it incorporates less than half of the
yarn—and the corresponding dyes, labor,
time, and talent—as other tapestry-woven
tunics (for example, see fig. 164).
Sawyer termed his study a broad “sketch,”
and he lamented the lack of more systematic
investigation. In the years following the publi-
cation of his article, researchers have focused
their attention on several basic aspects of this
task, none more fundamental than determin-
ing under whose auspices the tunics were
created since, aside from the staff deity and
its attendants, nearly identical tapestry-woven
tunics were one of the few other features that
Wari and Tiwanaku shared.57 It was not until
1986 that Amy Oakland Rodman, building
on the work of others, began to resolve the
dilemma by identifying several differences
in construction that distinguish Tiwanaku
tapestry-woven tunics, which survive in far
smaller numbers than their Wari counter-
parts.58 Since the 1980s a much larger sample
of Wari tunics has been published, providing
much-needed illustrations and descriptions
of both regularities and variations in artistic
and technical traits.59 During the same period,
a small corps of art historians and textile spe-
cialists—including Susan E. Bergh, the early
pioneer William Conklin, Mary Frame, Ann
Pollard Rowe, and Rebecca Stone60 —have of-
fered important studies focused on the tunics’
complex color and on distortion, a feature that
continues to fascinate, as well as analyses of
iconography and patterns of different kinds,
at least some perhaps rooted in mathematical
concerns.
In art historical terms Wari ceramics have
received less attention. As mentioned above
the most extensive studies, conducted in the
1960s by Lumbreras and especially Menzel,
applied themselves to the fundamental task
of synthesizing a broad chronology for Wari
Figure 18 [157]. Bag with
human face; alpaca or
llama hide, human hair,
pigment, cotton, coca leaf
contents; H. 26 cm (bag),
L. 64.7 cm (strap). The
Cleveland Museum of Art,
Leonard C. Hanna Jr. Fund
2011.35.

22 S usan E . Bergh and J ustin J ennings

 based on changes in ceramics, which are the
most ubiquitous, durable, and diagnostic of
Wari artifact types.61 In terms of artistic treat-
ment, ceramic manifestations of the staff deity
and its attendants (figs. 1, 14, 16) have been a
focus, especially from the perspective of their
origins in earlier art styles, such as Chavín
of the Peruvian highlands and Pucara of the
Lake Titicaca area, and from the viewpoint of
their synthesis by the Wari into a new ico-
nography of power that perhaps served as a
model for centralized hierarchy in the human
realm.62 The healthy thrust of many of these
studies has been to challenge the historical
tendency to view Tiwanaku as the source
of the divine imagery—a proposition that
awaits concrete proof. As a provocative and
hypothetical alternative, William Isbell and
Patricia J. Knobloch have proposed recently
that the imagery developed as a result of early
interaction across a wide sphere of the south-
ern Andes and appeared simultaneously at
Tiwanaku and Wari.63 Krzysztof Makowski
Hanula makes an equally unorthodox sugges-
tion: that depictions of the frontal deity por-
tray not a single supernatural being, as others
have usually supposed, but instead represent
many deities who, like the Christian saints,
share such major features as haloed heads but
vary in smaller attributes such as the orna-
mentation of belts or staffs.64 The idea is not
altogether new—in 1964, Menzel suggested
that male and female versions of the deity are
depicted artistically (fig. 5) 65—and at a mini-
mum it highlights how little we understand
iconographic variation and its meanings.
Wari images of the staff deity and its
companions are most prominently featured
on very large, finely made culinary vessels
recovered from deposits of deliberately shat-
tered ceramics, some of them enormous and
extremely complex in the variety of imagery
and vessel forms that they incorporate (figs.
1, 5). Some interpretation has been offered
for motifs unrelated to the staff deity from a
few of these deposits, although, because of
the obvious handicap of reconstruction, the
imagery and vessel forms of only one have
been analyzed comprehensively.66 The mean-
ing and purpose of the shattering, which may
vary among the deposits, also awaits analysis.
23 T he H istory of I nquiry into the Wari and T heir A rts
A final development worthy of note is a typol-
ogy of Wari human representations, the first of
its kind, which Patricia Knobloch continues to
develop based on ceramics and other media.67
With scattered notable exceptions, other
types of Wari arts have received no sustained,
in-depth analysis, including textiles of various
kinds such as tie-dyed cloth and featherwork
(fig. 17), headgear,68 ornaments made of gold or
silver or inlaid with colorful stone and shell
(fig. 12), wood containers, and stone sculpture,
both exquisitely carved figurines and a limited
number of larger-scale figures from the Wari
capital itself.69 In several of these categories,
even the project of basic description and quan-
tification has yet to be undertaken.
The Structure of This Volume
In the essays that follow, art historians and
archaeologists, many with long-standing
interests in the arts, present their insights,
informed conjectures, and still-unanswered
questions about Wari and the meanings and
functions of its arts. The pair of essays that
forms part one, “Interpreting a Cosmopolitan
Andean Society,” sketch essential archaeologi-
cal background for understanding the Wari
phenomenon as it occurred in different areas
of the Andes. Speaking from experience both
within and outside of the Wari heartland,
Katharina Schreiber outlines the case for the
interpretation of Wari as an empire, a view
that many of the authors in this volume en-
dorse. By “empire” she means a political state
that, starting in the mid-eighth century AD,
rapidly expanded beyond its regional borders
in Ayacucho to take control of a very large ter-
ritory that encompassed much of highland and
coastal Peru as well as many groups of people
of diverse ethnicities, cultures, languages, and
social organization. As noted above, Wari’s
archaeological footprint in these far-flung
areas is uneven, ranging from simple remains
of small-scale artifacts (ceramics and textiles)
to enormous architectural installations, and
she relates this variability to levels of control
that fluctuated from region to region, depend-
ing upon Wari’s goals and the local situations
to which it was responding.
Standing “outside Wari walls,” Luis Jaime
Castillo Butters provides a perspective on
 Wari from the Peruvian north coast, a vantage
point distant from the heartland where, dur-
ing early Wari times, the Moche people were
entering the final stages of their centuries-long
life span. As Castillo Butters argues, Wari
does not appear to have established an impe-
rial presence on the north coast, but its impact
nevertheless registered there, in part through
fine Wari objects found in ritual contexts. He
attributes the presence of these objects to the
agency not of the Wari but of Moche elites who
wanted to affiliate themselves with Wari’s
prestige and parlayed this affiliation in their
power strategies, and he asks whether analo-
gous situations might have pertained to other
places where Wari’s presence seems to have
been lighter, such as the central coast and the
northern highlands.
The set of essays in part two, “Transform-
ing the World,” describe and interpret the
transformative impact that Wari had in many
regions of the central Andes during its life-
time. Wari cannot be understood outside of its
monumental centers and architecture, which
it used to make its most public and over-
whelming proclamations of power. These cen-
ters have received more modern attention than
any other form of Wari material culture, and
Gordon F. McEwan and Patrick Ryan Williams
synthesize decades of research that of neces-
sity has focused on provincial installations
rather than the capital in Ayacucho, where po-
litical violence forced archaeological research
to a near halt during the 1980s. Next they turn
to the spread of agriculture and water infra-
structures (terraces, canals, aqueducts) that
accompanied Wari’s expansion and, in the
wake of a millennial drought that preceded its
rise, probably were key to its claim to medi-
ate nature’s unpredictable and devastating
whims. They conclude with an interpretation
of the messages that Wari may have intended
to convey through this built environment.
Donna Nash explores the human contexts
in which Wari strategies played out: the feasts
and beer-drinking parties at which Wari elites
certainly paraded many of the fine objects,
ornaments, and garments in this catalogue. As
Nash explains, these formal events were likely
a major aspect of Wari statecraft, providing
the social glue that bound the empire together
24 S usan E . Bergh and J ustin J ennings
by allowing Wari elites to forge essential con-
nections with local leaders in many different
areas and to negotiate official matters with
them. While doing so, they may have posi-
tioned themselves as a pivot of human affairs
through ritual hospitality and largesse. Anita
G. Cook addresses the focus of Wari’s state re-
ligion: the staff deity, who in the past has been
interpreted mainly as a symbol of power that
affirmed the political and religious authority
of Wari elites. While recognizing that the deity
did communicate such messages across the
ethnic and linguistic boundaries within Wari
territory, she advances another reading based
on indigenous cosmological beliefs, suggest-
ing that the staff deity embodies an Andean
concept of a cosmos sustained by balance,
renewal, reciprocity, and cooperation. Thus,
she implies, it is error to emphasize politi-
cal meanings at the expense of the view that
religious beliefs, evanescent and difficult to
reconstruct, were sincerely held and compel-
ling enough to find audiences in many differ-
ent areas of the Andes.
The arts abetted Wari’s efforts to transform
the world by communicating messages not
only about the encompassing power and the
promise of its supernatural beings but about
its leaders’ earthly might and wealth, encoded
in virtuosic and materials-intensive textiles,
the precious materials of inlaid ornaments,
and other beautifully made objects. In the fi-
nal section of part two, several authors discuss
the categories of Wari objects that have most
benefited from scholarly analysis, revealing
what is known of their function and symbolic
import.
Patricia J. Knobloch and Mary Glowacki
take on ceramics. Glowacki’s topic is the in-
triguing deposits of finely made vessels, many
probably used to serve or drink the native
corn beer chicha, that the ancients deliberately
shattered in diverse contexts, including buried
caches and the floors of important structures.
Such pottery smashes are widely documented
in Wari territory, and Glowacki believes they
relate to an ancient effort to keep the world in
balance through the offering of sacrifices and
chicha to the ancestors, who were guarantors
of water and fertility. Knobloch systematically
reviews the Wari or Wari-influenced ceramic
 styles, some included in the pottery smashes
but also many others, that came into being
over the four centuries of Wari’s life span.
Using the ceramics to reconstruct a narrative
of Wari’s development, she traces major ico-
nographies in the ceramic arts from the early
emergence of a collection of imaginative zoo-
morphic creatures that, she suggests, refer to
constellations, the roughly concurrent appear-
ance of the staff deity and distinctive human
persona who might represent elite individuals
or groups within the Wari hierarchy, and an
uptick in the production of feasting vessels
that she believes reflects increased feasting
activity in the centuries before Wari’s decline.
Susan Bergh’s discussion of tapestry-woven
tunics reviews the major characteristics of
these impressive garments: their iconography,
which is often connected to the staff deity,
along with their complex color, formats, and
the distortion of form mentioned above. In an
argument that relates to Anita Cook’s observa-
tions about the staff deity, she suggests that
the features of many tunics are motivated by
the Andean concept of dualism, the belief that
the world comes into being through the inter-
action of two principles that must be balanced
and harmonized. She further suggests that the
tunics evince an interest in number systems
and mathematics that may have given rise to
the distortion so widely admired today for the
figural abstraction that it produces.
Other authors provide overviews of im-
portant Wari object types that have not yet
received much systematic analysis. The most
surprising among these lacunae are exuber-
antly colored and geometrically patterned
tie-dyed garments, a well-known, important
type of elite wear that involved a construction
technique so demanding that it surprises even
contemporary textile specialists. Ann Rowe
describes the technique and, among other
things, provides a more thorough typology of
garment types, formats, and design principles
than has before been available, a contribu-
tion that will serve as a touchstone for future
investigation. With tie-dyed and tapestry-
woven cloth, fabric covered with the brilliant
feathers of tropical birds was likely one of the
most prized of Wari textile types. Analysis is
plagued by problems in cultural attribution
25 T he H istory of I nquiry into the Wari and T heir A rts
that owe to several factors, including the use
of geometric motifs that are not culturally spe-
cific. Heidi King discusses several feathered
works of confirmed Wari affiliation, making
intriguing, important observations based on
a fresh examination of the reported context
of the most spectacular group yet recovered.
Finally, Susan Bergh provides the first sum-
maries of fine Wari personal ornaments and
objects—ear ornaments, mirrors, pendants,
and figurines inlaid with brightly colored
mosaics of semi-precious materials along with
ornaments made of metal—and of small, fine
wood containers often carved in the image of a
fearsome supernatural sacrificer.
In the catalogue’s final section, “The
Aftermath,” archaeologist William Isbell and
art historian Margaret Young-Sánchez outline
Wari’s legacy in the ancient Andes. While
their focus is Wari, they broaden consideration
to include Tiwanaku, for the two cultures
developed together and shared a few aspects
of material culture, tapestry-woven tunics
among them, that are difficult to tease apart in
terms of later influence. According to Isbell,
one crucial element of Wari’s impact, achieved
in concert with Tiwanaku, was to shift the
Andean axis of political power and cul-
tural complexity, before centered in northern
coastal regions, to the southern highlands. In
Wari’s realm, this accomplishment rested on
improved agricultural and pastoral techniques
as well as on transformations in the nature of
governance. They then trace the potential arc
of Wari’s afterlife in specific features of later
Andean cultures, especially the Inca Empire.
In the arts, this task is complicated by the
turn of Inca textile and ceramic iconography
away from the figurative and toward a visual
vocabulary that focuses on spare, geometric
motifs. Nevertheless, Young-Sánchez presents
suggestive evidence of the importance of Wari
precedents to critical Inca imperial symbols,
both artistic and architectural.
The Past and the Present
Just sixty years ago, Wari re-emerged into
history from beneath the thick overburden
of topsoil, rubble, and prickly pear cactus
that in the millennium since its decline had
gradually obscured its achievement and its
 legacy. In the ensuing decades, a handful of
dedicated archaeologists and art historians,
several of whom have lived in Ayacucho, have
resurrected this once-unknown culture by
tracking its influence and ambitious expan-
sion over a challenging landscape, studying its
architecture and impressive provincial cen-
ters, recuperating its broad chronology, and
analyzing the corpus of startlingly beautiful
artworks that it left behind in buried depos-
its, offerings, and the tombs of the honored
dead. This work has succeeded in establish-
ing Wari as a powerful Andean polity in its
own right—in dialogue with but independent
of its contemporary, Tiwanaku, from whose
shadow it is stepping—and as a society that
has much to contribute to our understanding
of early human experiments with civilization.
This contribution comes not only from ar-
chaeology but also from the arts, which fluster
Western definitions of art and craft and thus
hold the potential to add needed dimension to
our ideas about the forms that human creative
genius assumes and the intriguing, often un-
familiar ways in which people invest things
with meaning.
In recent years, the number of scholars
with professional interests in Wari has in-
creased and their work has brought to light
exciting new archaeological discoveries that
have refined the understanding of Wari and of
26 S usan E . Bergh and J ustin J ennings
the Middle Horizon. But much investigation
remains to be undertaken in many parts of
Wari territory, not least at the Wari capital city
itself. The same is true, perhaps especially so,
of Wari’s arts, important categories of which
await systematic study and interpretation,
which the present project hopes to encourage.
The accuracy of future interpretation, both art
historical and archaeological, hinges on the
integrity of original archaeological context,
which has too often been devastated by com-
mercially motivated plundering, particularly
during the mid-twentieth century, as well as
by the increasing encroachment of modern de-
velopment into archaeological sites—airports
and housing projects in urban areas along
with rural agricultural expansion.
Wari: Lords of the Ancient Andes brings
together some 170 objects and textiles from
international collections that have harbored
and cared for them since as early as the late
nineteenth century. In all but a few instances,
original archaeological context has been lost.
This project is offered in the conviction that
there is a world of information embedded in
the objects themselves that with the help of
comparative data from scientific excavations
and the attention of sensitive analysts holds
promise to illuminate the past and its rele­
vance to the present.
notes 1. Cieza de León [1553] 1984; Cook
and Cook 1998; Pease 2008.
2. Cieza de León [1553] 1984, 249
[114]. This translation combines ele-
ments of those found in J. Rowe et al.
1950, 120; Schreiber 1992, 79.
3. Some suggest that, to avoid confu-
sion, the spelling “Wari” be used
to indicate the culture and “Huari”
to refer to the ancient capital (see
W. Isbell 2002, 457). This volume
employs a single spelling in the hope
that context and wording will make
the reference clear.
4. See Seville 2001 for the first exhi-
bition catalogue.
5. Pasztory 1984.
6. This idea surfaced during the
meeting of an NEH-funded advisory
committee for the exhibition in
October 2010.
7. For example, Young-Sánchez
2004d.
8. Lyon 1978, 96–97.
9. See Angrand 1866; D’Orbigny
1839; Squier 1877 for early illus-
trations of Tiwanaku sculpture;
DeHavenon 2009 provides a survey of
early literature on Tiwanaku and its
monuments.
10. Tello [1931] 1970, 520.
11. Tello 1942, 682–84.
12. Bennett 1946; Larco Hoyle 1948;
Schaedel 1948.
13. J. Rowe et al. 1950, 120, 122, 136.
14. Menzel 1977; Menzel 1968;
Menzel 1964.
15. J. Rowe 1967.
16. Knobloch 1976.
17. Benavides Calle 1991; Benavides
Calle 1984; Bennett 1953; Bragayrac
Dávila 1991; W. Isbell 2009; W. Isbell
1991; Lumbreras 1974a; Lumbreras
1960.
18. For instance, Anders 1991; W.
Isbell 1985; W. Isbell 1977b; W. Isbell
and Schreiber 1978.
19. Ochatoma Paravicino and Cabrera
Romero 2001; Valdez et al. 2006;
Valdez et al. 2002.
20. For instance, Cook and Benco
2001; W. Isbell 2002; W. Isbell and
Cook 2002; Ochatoma Paravicino
2007; Ochatoma Paravicino and
Cabrera Romero 2002; Ochatoma
Paravicino and Cabrera Romero 2001.
21. Pozzi-Escot 1991.
22. W. Isbell 2006.
23. Ochatoma Paravicino and Cabrera
Romero 2002; Tung and Knudson
2008.
24. See lists in Jennings and Craig
2001; Schreiber 1992.
27 T he H istory of I nquiry into the Wari and T heir A rts
25. J. Topic 1991.
26. McEwan 1996; McEwan 1991;
McEwan 1987; McEwan 1984a;
Schreiber 1992; Schreiber 1978.
27. Jennings 2006.
28. Jennings and Yépez Álvarez 2009.
29. Coleman-Goldstein 2010.
30. Meddens 1991; Meddens and
Branch 2010.
31. For instance, Coleman-Goldstein
2010; Tung and Owen 2006.
32. For instance, Castillo Butters
et al. 2008a; Lau 2006; Lau 2005;
Marcone Flores 2010.
33. For instance, Grossman 1983;
Hastorf 1993; Mackey 1982; Shimada
1985; Wilson 1988.
34. Menzel 1964.
35. Lumbreras 1974b.
36. See also W. Isbell and Schreiber
1978.
37. See also Schreiber 2005a;
Schreiber 2005b; Schreiber 2001;
Schreiber 1999.
38. See summaries in Glowacki 1996;
W. Isbell and McEwan 1991a.
39. Schaedel 1993.
40. Shady Solís 1988; Shady Solís
1982.
41. Jennings 2011; Jennings 2010a;
Owen 2010; T. Topic and J. Topic
2010a.
42. Moseley et al. 2005; Moseley et al.
1991; Williams 2001; Williams and
Isla Cuadrado 2002; Williams and
Nash 2002.
43. Glowacki 2002; Glowacki and
McEwan 2002.
44. Conklin 1991; Cook 1994; Cook
1983; W. Isbell and Vranich 2004;
Morris and von Hagen 1993, 122–23;
Schreiber 1992, 279–81.
45. See Braun 1993 for a more
detailed treatment of the post­
colonial reception of Pre-Columbian
arts in the West that focuses on
Mesoamerica.
46. For the khipu, see, for example,
Ascher and Ascher 1981; Urton
2003a.
47. For instance, Auther 2004;
Whittick 1984.
48. Mays 1985–86, 8.
49. Auther 2004, 341.
50. Clifford 1988; Pasztory 2010.
51. See, for example, Frame 1986 and
Lechtman 1979 for the meaning of
structure in ancient Andean metal-
work and textiles, respectively.
52. See Paternosto 1996, especially
5–8.
53. Pasztory 2010, especially chap. 5.
54. Murra 1962.
55. D’Harnoncourt 1954, 13.
D’Harnoncourt attributed Wari
tunics to Tiwanaku; subsequent
scholarship has corrected the
attribution.
56. Krauss 1979, 50.
57. The third is four-cornered hats
(e.g., fig. 11).
58. Oakland 1986a; Oakland 1986b,
31–41, 230–31. See also Rodman
and Cassman 1995; Rodman and
Fernández Lopez 2001. Contributors
to the effort were Bird and Skinner
(1974); Conklin (1983, 1970); A. Rowe
(1979).
59. Two dissertations pulled together
large samples based primarily on
museum collections (Bergh 1999
and Stone 1987) and other publica-
tions described collections either in
individual museums or from specific
archaeological sites (among others,
Angeles Falcón and Pozzi-Escot
2001; Conklin 1996; Eisleb and
Strehlow 1985; Eisleb and Strehlow
1980; Prümers 1990; Ramos and
Blasco 1977; Rodman and Fernández
Lopez 2005; A. Rowe 1986b).
60. Bergh forthcoming; Bergh 2009;
Bergh 1999; Conklin 2004a; Conklin
1986; Conklin 1970; Frame forth-
coming; Frame 2007; Frame 2005;
Frame 2001; A. Rowe 1979; Stone
1987; Stone 1986; Stone-Miller 1992b.
61. Lumbreras 1974a; Lumbreras
1960; Menzel 1977; Menzel 1968;
Menzel 1964. Recently, Matthew
Edwards and Katharina Schreiber
have put together a fully illustrated
beta version of Menzel’s underillus-
trated 1964 ceramic chronology.
62. Cook 2001b; Cook 1994; Cook
1984–85; Cook 1983; W. Isbell 2008;
W. Isbell 1988, 180–81; W. Isbell
1984–85; W. Isbell 1983; W. Isbell
and Cook 1987.
63. W. Isbell and Knobloch 2006; see
also W. Isbell and Knobloch 2009.
64. Makowski Hanula 2009;
Makowski Hanula 2002.
65. Menzel 1977, 54; Menzel 1964, 19,
26; see also Lyon 1978, 108–13; Bergh
1999, 70n86.
66. Cook 2001b; Cook 1984–85; W.
Isbell 2004b, 191; Knobloch 2000;
Ochatoma Paravicino and Cabrera
Romero 1999; Ochatoma Paravicino
and Cabrera Romero 2002.
67. Knobloch 2010; Knobloch 2002.
68. But see Frame 1990 on four-
cornered hats.
69. But see Cook 1992 on the
figurines.
Interpreting a Cosmopolitan
Andean Societ y
Katharina Schreiber
Figure 19 [86]. Warrior
plaque; silver; 25.7 x 19.7
x 2.5 cm. Museum of
Fine Arts, Houston, Gift
of Alfred C. Glassell, Jr.
2001.1174.
[129]. Overleaf, Bag;
camelid fiber and cotton;
18.7 x 16.5 cm. The Textile
Museum, Washington,
DC, Museum Purchase,
1959.10.1.
31
The Rise of an Andean Empire
The Wari Empire arose in the Huamanga or
Ayacucho region of the central highlands of
modern Peru. Its capital city, also known as
Wari, sits at an elevation of 2,800 meters (9,200
feet) above sea level in the relatively open
Ayacucho Basin. Wari emerged out of a period
of intense interaction during the sixth and
seventh centuries AD among a number of cul-
tures located in different parts of the central
Andean region. By the mid-eighth century the
Wari had embarked on a campaign of expan-
sion that was to bring most of Peru’s highlands
along with its southern and central coastal
regions under a single regime. The Wari ruled
an enormous territory and had sovereignty
over many diverse peoples, supporting their
empire through the control of economic
production and distribution, and legitimat-
ing their power through a state religion and
the symbols associated with that religion. By
the late eleventh century the empire had col-
lapsed, leaving only scattered, ruined traces
for today’s archaeologists to uncover.
This remarkable achievement was unprec-
edented. Wari is known to archaeologists as
a “pristine” empire. It came from a historical
tradition that included no prior empires. The
first of its kind in Andean South America,
it had no antecedents, no predecessors, no
examples from which to draw knowledge or
inspiration. Thus, the Wari people created a
form of governance that was completely new
in the region.
Rising within a physical context of great
diversity, the Wari located their capital city
within a maize-growing ecological zone. Resi-
dents would have had easy access to maize
fields around the city, potato fields at higher
elevations to the east, south, and west, and
lower, warmer lands to the north where they
could grow fruits, other vegetables, and chili
peppers. The empire expanded to include vast
areas of the high Andes, encompassing many
discrete valleys, each with its own system of
vertically arranged ecological zones. These
valleys were surrounded by high grassy
plains, where vast herds of llamas and alpacas
grazed. Further, the empire controlled large
portions of the desert coast, where many crops
including cotton could be grown in the irri-
gated fertile soils.
The Cultural Context of the Wari Empire
Humans have occupied the Andes for at least
12,000 years. By 2500 BC they were growing
cotton on the coast of Peru, and construction
of monumental platform mounds was under
way on the north central coast of Peru. Ar-
chaeologists divide the last four millennia of
Andean prehistory into six major segments:
the Initial Period, followed by three Horizons
(Early, Middle, and Late), which are separated
by Intermediate Periods (Early and Late). Pe-
riods were times during which developments
were regionally confined; during horizons,
however, a single well-defined cultural style
spread broadly through the Andes. The most
obvious example of the latter is the Late Ho-
rizon, when the expansion of the Inca Empire
distributed Inca-style material culture from
Ecuador to Argentina.
During the Initial Period (1800–1000 BC),
most people began to live in villages and
derive their subsistence from domesticated
plants and animals. On the central and north
coasts of Peru people built massive platform
mounds and pyramids, evidence of a rich
ceremonial life. During the Early Horizon
(1000 BC–AD 1), a great temple and impor-
tant pilgrimage center emerged at Chavín de
 Huántar.1 This was the first time that a single
cultural tradition transcended local bound-
aries and provided unifying elements for
multiple cultures. There is no indication that
Chavín influence had any political compo-
nent; rather, the site was the center of a shared
belief system.
Chavín influence ceased around 500 BC,
at which time several regional cultures began
to emerge. They coalesced in the Early Inter-
mediate Period (AD 1–600) and included the
north coast Moche, perhaps the first state-
level society in the Andes, the Lima culture
of the central coast, and the Nasca on the
south coast. The Nasca produced some of the
most spectacular ceramic and textile art of
the ancient world, and there were strong links
between them and the later Wari. Regional
highland cultures of this period include Caja-
marca in the far north of Peru, known for its
ceramics made of fine white clay, and the Re-
cuay culture, which occupied the deep valley
called the Callejón de Huaylas. The Ayacucho
Basin was replete with settlements of the
Huarpa culture, the direct ancestor of the
Wari. To the south in the Lake Titicaca Basin,
Pucara of the northern basin and Tiwanaku
of the southern basin were important religious
centers; the latter would grow into an impor-
tant political center as well.
The Middle Horizon (AD 600–1000) was
the period of the expansion of the Wari Em-
pire and the time during which Tiwanaku was
also the capital of a major polity to the south.
Wari and Tiwanaku shared distinctive ele-
ments of religious and political iconography,
and perhaps some common antecedents. But
the two were quite different. Tiwanaku was a
major pilgrimage center, open and welcoming
to large numbers of people; Wari and its pro-
vincial sites were built of high walls designed
to keep people out. Tiwanaku iconography
was seen by the people at the capital, carved
on immovable stone monuments; Wari’s
iconography of power was depicted on por-
table media such as textiles and ceramics that
could be taken to the people. Tiwanaku held
dominion over much of the Titicaca Basin
and established colonies in several far-flung
regions where crops could be grown that were
not available in the high wind-swept plains of
32 K atharina S chreiber
the altiplano; Wari had more imperial inten-
tions, establishing political and economic
control over much of the central Andes. After
both polities collapsed, distinctive regional
cultures appeared once again during the Late
Intermediate Period (1000–1400). All were
conquered by the Inca in the Late Horizon
(1400–1532). The indigenous Andean culture
sequence came to an end with the arrival of
Francisco Pizarro and his band of men, who
captured the Inca emperor and toppled the
Inca Empire.
For centuries during the Early Intermedi-
ate Period the distinctive regional cultures of
the Andes developed and grew with minimal
interaction. Times were good: populations
swelled, agricultural systems expanded, art-
ists created ever more beautiful objects. But
in the sixth and seventh centuries, during
the final generations of the period, something
changed. The climate deteriorated and periods
of prolonged severe drought in the mountains
affected irrigation systems on the desert coast,
which depended on rivers with headwaters
in the highlands.2 Warfare and violence seem
to have heightened. There was also sudden
increased contact among the different regional
groups. Moche potters included Recuay de-
signs on their ceramics; Nasca potters did like-
wise with Moche images; Nasca designs and
pigments began to appear on Huarpa pottery in
Ayacucho. More than ideas were moving from
one place to another; people were also moving,
as conditions forced the abandonment of some
places.
As the Early Intermediate Period drew to a
close most of the regional cultures collapsed or
were diminished in some way. Wari, however,
transformed. Out of the villages and towns of
the Huarpa culture the great capital city co-
alesced, and on it centered a state that would
expand its control into the territories of most
of the former regional cultures, and into the
lives of their diverse peoples.
What Is an Empire?
Before considering the specific case of Wari,
it is useful to define just what empires are
and how they work. Put simply, empires are
political states that expand beyond regional
borders—usually rapidly, using military force
 as well as diplomacy—to take control of very
large territories and many groups of people,
ranging from tribal-level societies to com-
plex (often competing) states. Thus, empires
are physically diverse, incorporating regions
that may be very different in environment
and ecology, and they are ethnically eclectic,
including people of many nations with differ-
ent cultural practices and languages. In order
to organize and maintain order in the face of
unprecedented size and diversity, empires
must develop centralized forms of control and
institutions, which must have been a major
challenge for a group such as the Wari, who
had no prior examples on which to draw.
The forms of imperial control can be
divided into three general categories: po-
litical, economic, and ideological. Political
control involves establishing a hierarchy that
is centralized but reaches outward and con-
nects all the disparate pieces of the empire,
binding them together into a single structure.
The capital city—usually established in the
polity’s heartland, as Rome was—forms the
capstone of the hierarchy, followed by impe-
rial cities or administrative centers, located at
strategic nodes throughout the empire. Within
each conquered province, the empire may
build administrative facilities, or it may rely
on existing facilities if the local authorities
are cooperative. Herein lies one of the greatest
challenges facing an empire: every situation
is different. Some newly conquered provinces
have large well-organized political systems,
while others are only loosely held together;
some are cooperative, others are hostile; some
are of great strategic importance or contain
crucial resources, others are of marginal
interest. In each case the empire must tailor
its strategies individually to incorporate each
new province. There is no “one size fits all”
approach. At each point in the hierarchy of
control are individuals who hold key offices:
the king or emperor in the capital, regional
governors in the major administrative centers,
provincial overseers at the next level, and so
on down the chain of command, which paral-
lels or in many cases is the same as the mili-
tary chain of command.
Empires support themselves through vari-
ous forms of economic control. Building and
33 T he R ise of an A ndean E mpire
maintaining all the things an empire needs is
expensive business, and an empire depends
absolutely on a constant and reliable source of
income. For the conquered this comes in the
form of taxation or in more general terms trib-
ute payments—money, produce, or labor. Non-
payment of tribute is punished harshly. It may
be necessary for the empire to organize chang-
es in the local economy to increase production
of the things it values and needs. Empires may
also establish control over key resources, such
as precious metals or other valued minerals,
wresting this control from conquered groups.
By establishing a monopoly on the produc-
tion of bronze, for example, artifacts of bronze
become associated with the empire, increase
in value, and enable the empire to manipulate
and use them to maintain power. Further-
more, the production of particular kinds of art
works, tightly controlled by the empire, can
have a similar result. An example of this is the
extremely fine cumbi cloth of the Inca, which
included exceptional tapestry-woven textiles.
Only the Inca emperor could wear cumbi gar-
ments, but he could make gifts of the cloth to
other individuals, usually to cement political
alliances (see pp. 159–91, “Tapestry-woven
Tunics”).
Finally, ideological control is crucial
although in some ways less tangible than
aspects of political or economic control. In one
form ideological control may be accomplished
by imposing a state religion that brings the
deities of the conquerors to the conquered and
helps legitimate the new regime. In another
form, which may cross-cut religion to a greater
or lesser degree, the state ideology may be
manifest in an iconography of power: symbols
such as individuals holding staffs of sover-
eignty that make clear to all who view them
that the new regime is right, powerful, and
justified (see figs. 1, 5, 6). In early empires re-
ligious and political symbols may be one and
the same. Political leaders take on deity-like
characteristics and may even become deities
upon their death. Other humans such as sol-
diers may take on nonhuman features such as
wings or animal traits that make them appear
to be supernatural beings. The blending of
the natural and supernatural, the earthly and
heavenly, provides the empire with divine jus-
Figure 20 [35]. Figure in
a litter; ceramic and slip;
26.3 x 21.6 x 24.8 cm. The
Cleveland Museum of Art,
Leonard C. Hanna Jr. Fund
1997.1.
tification and legitimization for its existence
(and activities) on earth.
Weaving together these forms of control
are a number of threads, some of them physi-
cal and material. Clearly, one of the first things
an empire must do after the initial conquest
of a new province is take stock of what it has:
territory, resources, and above all, people. It
needs to take a census. Such a survey requires
not only a numbering system but also a way to
keep records. Virtually all historically known
empires had systems of writing. Andean cul-
tures did not, but they did develop a record-
keeping system.
34 K atharina S chreiber
Political control and the resulting hier-
archy require an extensive built infrastruc-
ture. The capital city is large, complex, and,
aside from housing for thousands or tens of
thousands of residents, may contain “classic”
features of imperial art and architecture such
as government buildings, royal palaces, major
temples, and other types of monumental struc-
tures. Spread throughout the empire will be
administrative centers and other government
buildings, usually constructed in a style iden-
tified with the empire. Linking all these nodes
together are road systems, some of the most fa-
mous ancient examples of which are associat-
Figure 21. The Wari capital
as it currently appears, in
ruins and heavily over-
grown with cactus. Photo:
Katharina Schreiber.
Figure 22. View of the Wari
capital taken from a nearby
mountain, looking to the
southwest. The entire
plateau area, covered
with green vegetation, is
the architectural core of
the city. Photo: Katharina
Schreiber.
35 T he R ise of an A ndean E mpire
ed with empires. As “all roads lead to Rome,”
so do the roads of an empire converge on the
capital city. Finally, the art associated with an
empire plays a part, its iconography consist-
ing of imposing deities and powerful humans,
also rendered in recognizably imperial styles
on exquisitely made ceramics, textiles, and
sumptuary objects fashioned of noble metals
or hard-to-obtain shells and precious stones
(figs. 19, 20). All these things—the centers,
monumental architecture, vast networks of
roads, imperial styles, and fine art works made
of costly materials—are visible symbols of not
only the empire’s might but also its grandeur
and magnificence.
Typically scholars who study empires base
their work on written documents produced
by, or at least at the time of, the empire in
question. Wari presents major challenges to
archaeologists and art historians because it
left no written records to aid in understanding
its past.3 Nor are there eyewitness accounts
of the Wari written by others, in contrast to
the later Inca, who also had no writing system
but about whom the Spanish conquistadors
and those who followed them to the New
World left many accounts. Yet the Andes were
entirely unknown to the outside world in the
first millennium AD. The only sources of in-
formation about Wari lie in material remains,
ranging from city-sized archaeological sites to
textiles and fragments of pottery, which like
parts of a puzzle must be painstakingly reas-
sembled in order to reconstruct a portrait of
this early empire.
Evidence of political hierarchy can be
found in the existence of a capital city, admin-
istrative centers, and other government build-
ings constructed in the Wari imperial style,
along with the remains of a road system tying
these nodes together. We can glimpse imperial
interference in local production and control
of special resources. Imperial styles of ceram-
ics, textiles, and other artifacts that spread
throughout the Andes are strong testimony
to the presence of the empire. The imagery
depicted on those artifacts, the symbols of
religion and power, give us a sense of Wari
imperial ideology.
What Was the Original Name of the Wari
Empire?
What the Wari called themselves or their em-
pire we do not know—archaeologists and art
historians use the term “Wari” because that is
the modern name of the site that was its capi-
tal. At the time of the Spanish conquest, how-
ever, that ancient city was called “Vinaque.” 4
Could this have been the name of the capital,
the people, or the empire in Wari times? An-
other possibility exists. A legend recorded in
the sixteenth century about the Wari capital
stated that it was built by a people who were
bearded and white (see pp. 5–27, “The History
of Inquiry into the Wari and Their Arts”). A
similar legend was told about a Wari provin-
cial center, Jincamocco; in this version the
strangers were termed “Viracochas.”5 The
Inca gave the name Viracocha to their creator
deity, the god who brought civilization to the
Andes. Does this name perhaps harken back to
the earlier empire—Wari, which first brought
this form of civilization to the Andes? Or does
 Viracocha simply refer to any foreigner, as the 24). The city is not only large but diverse.
term does today? Sadly, we cannot answer The architectural core includes ceremonial
these questions with any degree of certainty. structures and perhaps some open plazas,
Today the Wari capital is in ruins and but most of the site is characterized by sub-
badly overgrown with cactus (fig. 21). The stantial, carefully planned architecture of a
remains of large stone-walled buildings cover particular and unique type: large rectangular
about 6 square kilometers (2.3 sq. miles)—the compounds, each two or even three stories
core of the site (fig. 22)—around which, in an tall, divided into square units, much like a
area nearly as extensive, the ground is covered checkerboard; the squares contain a central
with broken bits of Wari styles of pottery and open patio flanked by long narrow rooms,
other artifacts. Although sporadic excavations or galleries, that are sometimes two or three
at the site have taken place since the 1880s, deep (see pp. 65–81, “The Wari Built Environ-
very little has appeared in print about Wari.6 ment”).7 The ground floors of these structures
The settlement at Wari has a long history, are now covered by centuries of soil accumu-
even before the rise of the empire. Centuries lation, which gives the (mistaken) impression
earlier, a square sunken court of fine cut stone, that these structures have no doorways, but of
24 m (79 ft.) on a side, was built at the site (fig. course they do—at ground level, now deeply
23). This significant religious structure tells us buried. (When similar forms of architecture
that the site was an important place very early were first identified in the provinces, they
on. During the Early Intermediate Period, the were thought perhaps to be jails because it
Figure 23. Corner of the Huarpa people occupied Wari extensively, seemed so difficult to get into or out of them.)
early sunken court that leaving behind many typical ceramics that The great rectangular enclosures tend to stand
occurs beneath Middle have been exposed in deep excavations be- alone when found in the provinces, but at the
Horizon walls in the
Moraduchayoq sector of
neath Wari imperial levels. capital they have a more organic character.
the Wari capital. Some Most of what can be seen today at the site While there are many large subdivided com-
of the floor stones have represents the Middle Horizon occupation of pounds at Wari, they are more oriented to the
been removed, exposing
Wari: the period of imperial expansion (fig. landscape, and the spaces between them are
even earlier walls and
construction phases. Photo:
Katharina Schreiber.
Figure 24. Satellite image
of the Wari capital show-
ing the diversity of visible
architecture at the site. The
distance from east to west
is approximately that of the
National Mall between the
Washington Monument and
the U.S. Capitol building.
Modern roofs protect sev-
eral of the archaeological
monuments. The area of
the detail appears in Figure
25. Image: Google Earth.
© 2012 GeoEye; © 2012
Google; © 2012 Europa
Technologies.

36 K atharina S chreiber
 filled in with irregular forms that still attempt
to conform to the basic patio-with-galleries
plan. This style of architecture is ubiquitous
at Wari, and it seems likely that most of these
buildings were elite residences and state ad-
ministrative buildings.
One sector of the capital seems to have
served more ceremonial functions. Along the
western side of the site is a series of possible
temples, tombs, and other religious buildings
(fig. 25). Near the northwest corner is Capilla
Pata, a long, narrow, trapezoidal enclosure
that in size and shape is unlike other struc-
tures at the site. South of it are two large
D-shaped temples in the Vegachayoq Moqo
sector, one of which, the Templo Mayor (Great
Temple), has been excavated within the arms
of a U-shaped mound; a high stone wall sur-
rounds the complex (see fig. 46). The second
large D-shaped temple, visible only in outline

Figure 25. Detail of the

Wari capital. Capilla Pata is
at the upper left. Below (to
the south), is Vegachayoq Capilla Pata
Moqo’s Templo Mayor, a
large D-shaped structure,
between long areas with
modern roofing; a second
D-shaped structure occurs
to the upper right of that
complex. Just below is
Monjachayoq. Below
that and to the right is
Moraduchayoq, where the D-shaped temple
outlines of old excavations
can be discerned. The Vegachayoq
Cheqo Wasi sector is at the Moqo
lower right. Image: Google
Earth. © 2012 GeoEye;
© 2012 Google; © 2012
Europa Technologies.

D-shaped temple
Monjachayoq

Moraduchayoq

Cheqo Wasi

37 T he R ise of an A ndean E mpire


Figure 26. Stone slab on aerial views, lies just to the northeast. Next
structure in the Cheqo
in line are the subterranean tunnels and cut-
Wasi sector at the Wari
capital. Photo: Katharina stone galleries of the Monjachayoq complex,
Schreiber. whose function is unknown. But the care-
ful attention paid to layout and construction
indicates that they were very special places. A
smaller D-shaped temple is located just west
of Monjachayoq. A bit farther along to the
south-southeast is the Moraduchayoq com-
plex, location of the old sunken court, which
was buried and covered with other buildings
in the Middle Horizon, and immediately to the
south-southeast are the stone slab chambers
of Cheqo Wasi (fig. 26), thought to have been
tombs, perhaps of the Wari kings. This se-
quence of features would thus appear to form
the ceremonial sector of the site, as Enrique
González Carré and Enrique Bragayrac Dávila
suggest.8
The extensive distribution of artifacts and
isolated structures surrounding the capital’s
architectural core may simply be trash dis-
carded from the center, or it may represent the
remains of more perishable structures. If the
lower classes lived here, they did not inhabit
the large patio-and-gallery units but more
humble abodes that have not survived. Fu-
ture excavations may help solve some of these
mysteries.
There can be little doubt that the archaeo-
logical site of Wari was indeed the capital city
of the empire: there is simply no settlement
anything like it in terms of size and com-
38 K atharina S chreiber
plexity elsewhere in the realm. Its architec-
ture is identical to the building forms found
throughout the empire, with the distinction
that it is more varied. Its architecture seems
to have grown over time rather than being
built all at once, as Wari provincial centers
were, and it seems to have had large sectors or
barrios dedicated to different functions, such
as cere­monial rites and burials, elite or com-
moner residence, and so on. And the number
and density of Wari style artifacts is simply
astounding; at no other site in the empire does
one find pure Wari styles so completely domi-
nant (although provincial styles occasionally
turn up). In the provinces, however, Wari style
artifacts are always mixed with those of what-
ever local group had been conquered there.
The Wari Heartland and Beyond
Some very interesting things happened in the
capital’s immediate hinterland as Wari rose to
become an empire. Recalling that during the
Early Intermediate Period the region was home
to the Huarpa culture, archaeologists have
identified dozens of sites with Huarpa ceram-
ics in the region.9 Huarpa ceramics were gener-
ally painted with black geometric designs on a
white slip, with occasional use of red paint as
well.10 Toward the end of the Early Intermedi-
ate Period new colors and vessel shapes from
the south coast Nasca style appeared in Huar­
pa ceramics, indicating an upswing in com-
munication and perhaps even migrations of
people between the two groups.11 The Huarpa
also began to intensify agricultural production
both by building terraces to conserve scarce
rainfall and spring water (terraces prevent
run-off and encourage deeper soil formation)
and by constructing networks of irrigation
canals and reservoirs to distribute water. They
may have also developed ways to grow two or
three crops per year at the lowest elevations.12
This process of agricultural intensification
probably accelerated during Wari times in the
Middle Horizon when the population of the
heartland appears to have increased markedly.
At the beginning of the Middle Horizon
there was not only an abrupt transition in
ceramics from the mundane Huarpa style
to the beautifully made and decorated Wari
styles, but also an apparent major reorgani-
 zation of the people of the Ayacucho Basin.
In Huarpa times small clusters of villages
arranged around larger towns were spread
throughout the basin. In the early Middle
Horizon, Wari grew into a city and became the
focal point for local settlements. In the later
Middle Horizon the number of settlements in
the basin was sharply reduced and the popu-
lation of the hinterland dropped. What hap-
pened? Where did all the people go? It seems
likely that most moved to town. The emptying
of the rural hinterland around early cities is
a common pattern that archaeologists have
detected in other cases, such as Uruk in Meso-
potamia, or Teotihuacán in Mexico.
So far, data on the nature of agricultural
production in the heartland are limited. The
Wari built Azángaro in the northern, lower
portion of the basin, where warmer-climate
crops such as chili peppers could be grown
(see fig. 45).13 This great rectangular enclo-
sure included a large sector of structures that
probably served as storehouses, suggesting
state-controlled production and distribution
of crops. Another much smaller Wari site,
Jargampata, located just to the east of the
Ayacucho Basin, may have served as a point to
gather and store produce intended for the sup-
port of the capital city.14
Recent excavations at Conchopata, one of
the major Wari sites in the Ayacucho Basin,
have provided some interesting new data
about the people of the Wari heartland. Con-
chopata was a second-tier settlement in the ur-
ban hierarchy and the home to both elite and
commoner residents. The site includes domes-
tic structures, some of them formally planned
patio-and-gallery units and others of a more
irregular layout, as well as several D-shaped
temples (see fig. 70). Of the many tombs found
at the site, at least eight different types can
be distinguished that vary according to a
person’s social status, age, and/or gender.15 A
study of thirty-one “trophy” heads (human
crania removed from their bodies, specially
prepared, and displayed as trophies) found in
a D-shaped temple at Conchopata shows that
42 percent have evidence of physical trauma;
chemical analysis suggests that many of those
individuals who lost their heads were foreign-
ers.16 This evidence of a military component to
39 T he R ise of an A ndean E mpire
Wari expansionism is also found in iconogra-
phy at Conchopata and elsewhere (see fig. 7).
Some of the best evidence for the existence
of the Wari Empire comes from the remains of
imperial infrastructure discovered in regions
outside the Ayacucho Basin. Over time several
very large rectangular enclosures of the dis-
tinctive architectural style we now associate
with Wari were identified, and more are being
discovered as archaeologists explore more
remote regions of the Andes. These are most
common in the Peruvian Andes, from Caja-
marca in northern Peru to the Cuzco region in
the southeast and Moquegua in the far south
(see maps, pp. xiv, xv). Although provincial
sites are less common on the coast, sites have
been found in the Nasca region and elsewhere
on the central and south coasts of Peru.
If we use the distribution of provincial
centers to estimate the size of the empire,
Wari’s control of mountain territory reached
800 km (497 mi.) to the north of the capital,
525 km (326 mi.) to the south, and 275 km (171
mi.) to the east (to the region around Cuzco); it
also stretched down to the central and south
coast, 350 km (217 mi.) to the west and south-
west. Thus the empire extended more than
1,300 km (807 mi.) along its north-south axis;
its width varied from about 100 km (62 mi.)
in the north, where it encompassed only the
highlands, to some 400 km (248 mi.) in the
south, where it spanned both highlands and
coast. The total spatial extent of the empire
could have been as much as 320,000 sq. km
(124,000 sq. mi.).
The large rectangular enclosures that con-
stitute many provincial Wari sites are subdi-
vided into square or rectangular units accord-
ing to a rigid grid plan (e.g. Viracochapampa,
Pikillacta, Azángaro, Jincamocco). The basic
unit, as at Wari, is a central patio surrounded
by long narrow galleries. A few sites are made
up of free-standing patio-gallery units (e.g.,
Honco Pampa, Huaro). This architectural
style is unique and unmistakable. Many of the
provincial sites are associated with prehistoric
roads, some of which the Inca later incorpo-
rated into their system of royal highways.17
In fact, some previously unknown Wari sites
were discovered during investigation of the
Inca roads.18
Figures 27a, 27b. Two
views of Pataraya, a small
Wari installation in the
upper Nasca drainage on
Peru’s south coast. Satellite
image: Google Earth.
© 2012 GeoEye; © 2012
Google; © 2012 Europa
Technologies. Photo:
Matthew Edwards.
These sites vary greatly in size, from a
single patio unit (Jargampata) to huge rectan-
gular enclosures, up to 800 m (2,600 ft.) on a
side, as in the case of Pikillacta. The variation
depends on the diversity of activities carried
out in the center, the importance of the center
and the region in which it lies, and perhaps
the size of the region or population governed
from that center. Some very small sites may
have served very specific purposes, especially
those associated with one of the larger Wari
centers (see below). But most of the medium
and large sites—Pikillacta, Cerro Baúl, Jinca-
mocco, and others—served as regional admin-
istrative centers, nodes in the political hierar-
chy of the imperial ruling structure where a
40 K atharina S chreiber
range of political, economic, and ceremonial
activities occurred.
Excavations of the larger provincial Wari
sites have provided new lines of evidence
about the kinds of activities that took place
within them. These sites (or portions of them)
served as elite residences, where Wari admin-
istrators, regional governors, or other officials
lived. Some sectors were set aside for large-
scale food preparation, brewing beer, and
feasting, as at Cerro Baúl (see pp. 82–101, “The
Art of Feasting”). In some sites there is evi-
dence for craft production, including ceram-
ics and textiles. A few sites may have storage
sectors. Ceremonial areas are also apparent,
and both Viracochapampa and Pikillacta
have large niched halls that perhaps served as
places of ritual.19
Smaller special-purpose sites may be locat-
ed to control the movement of people into and
out of regions. Some are located along ancient
roads and may have functioned in part as way
stations, places to house travelers on official
state business. They also may have served
to organize the movement of resources from
one place to another. For example, Pataraya,
a recently excavated site in the upper Nasca
Valley, was a small, four-patio enclosure lo-
cated along a road that connected the coastal
region with the mountains (figs. 27a, 27b).20
It included two patio units devoted to food
preparation and domestic activities, one for
ceremonial activities, and one that probably
was the residence of the Wari administrator.
The remains of large quantities of cotton in
an external compound indicate that the fiber
was being shipped from lower coastal regions,
where it grows, to higher elevations, where it
could be used in textile production.
The remodeling of some provincial sites
reveals that Wari needs and strategies changed
over time; the Wari went into a region with
one plan, which they then modified. For
instance, Jincamocco expanded until it was
several times its original size, indicating that
the Wari ideas for this region had evolved and
that a larger facility was needed.21 One pos-
sibility is that more space was required first to
house laborers who were brought in to build
terraces and then to store the maize grown on
those terraces. Viracochapampa was neither

Figure 28. View of
Jincamocco, a Wari instal-
lation in the Sondondo
Valley near the Wari capi-
tal. The traces of the Wari
enclosure and rectangular
subdivisions within it can
be seen in the outlines of
modern fields in the center
of the photograph. Remains
of the Wari occupation
extend outward to the edge
of the agricultural ter-
racing and to the modern
town of Cabana, seen at
the upper right. Photo:
Katharina Schreiber.
41 T he R ise of an A ndean E mpire
finished nor occupied,22 perhaps because it
was no longer needed to fulfill the purpose for
which it had been planned. Some sectors of
Pikillacta were completed and occupied, but
others were still unfinished when the site was
abandoned.23
In the Sondondo Valley, 125 km (77 mi.)
south of the capital, the Wari built Jinca-
mocco, a medium-sized administrative center
consisting of a rectangular enclosure subdi-
vided into patio-gallery units, most of which
were probably used for domestic activities (fig.
28).24 One large patio served as an area for food
preparation and feasting, and a raised plat-
form, finely coated with white plaster and as-
sociated with finely made elite ceramics, had a
ceremonial function. Substantial evidence for
textile production at the site exists. The fact
that Wari invested heavily to build this center
suggests both that the region was important
(probably because it was the halfway point
on the journey to Nasca on the south coast)
and that it did not have a pre-existing politi-
cal authority and infrastructure that the Wari
could use.
In some areas in which Wari centers are
present, regional survey data complement ex-
cavation. These data tell us where people were
living before, during, and after the period of
imperial occupation, and also reveal much
about resources and their use at these times.
For example, when the Wari moved into the
Sondondo Valley to build Jincamocco, they
forced most of the local people to abandon
their villages and relocate to lower elevations.
(They moved from about 3,600 to about 3,300
m [12,000 to 11,000 ft.].) At the same time,
virtually all of the valley sides below 3,300
m were terraced, allowing the cultivation of
maize in these high regions. Thus, the move-
ment of people to lower elevations provided
the labor to increase maize production
An intensive archaeological survey of
the entire valley discovered four more, much
smaller, Wari imperial sites.25 One, located
along a major ancient road, probably served to
monitor access into the valley from the north
and also to exploit a small obsidian source
there. A second probably administered the
north end of the valley. A third may have been
a storage center as it was located adjacent to
some of the most productive agricultural land
in the valley. And the fourth was a ceremonial
site with stone slab tombs, like those at Wari,
and two small D-shaped temples. Several
major roads lead into and out of the valley;
portions of the Wari route leading directly
to Jincamocco were later incorporated in the
Inca system of royal roads.26 Thus, the Wari
occupation of a single region could be quite
complex, involving multiple centers that
played diverse roles.
But what about all the regions and valleys
where we do not find Wari provincial sites?
Were these also incorporated into the empire?
While the presence of imperial infrastructure
certainly indicates imperial control, the lack
of infrastructure does not necessarily mean
lack of control. Historians and archaeologists
sometimes distinguish between “direct” and
“indirect” imperial occupations,27 although
there are many forms of rule that fall between
these extremes.28 We can think of an empire
as a mosaic, with different territories needing
different levels of control, striking a balance
between the needs of the empire and the pre-
existing cultural situation in each region.
In more direct forms of control, the empire
builds its own infrastructure—administrative
centers, roads, and the like—because it needs
facilities that do not already exist. The pres-
 ence of imperial infrastructure tells us that a
particular region was important to the em-
pire but was probably not already politically
centralized: there was no controlling authority
that had established the infrastructure neces-
sary to rule the region. So the empire had to
step in and invest in constructing the facilities
it required such as at Pikillacta, Cerro Baúl,
Jincamocco, and others.
In a region with a centralized authority al-
ready in place—and with it an existing infra-
structure—an empire might need not to build
completely new facilities. It can use what is
already there. In these regions archaeologists
will not find major architectural complexes
in the style of the empire, instead encounter-
ing more subtle clues of imperial control such
as changes in settlement locations as people
were reorganized or moved around to serve
imperial purposes, shifts in the production
and distribution of local crops and resources,
and modifications in diet or patterns of vio-
lence, which indicate disruptions in daily life
caused by the presence of the empire. Or they
may find only very small imperial sites, but
not large administrative centers. This seems
to be the case in the Wari occupation of the
Nasca region; the local culture maintained a
relatively complex political organization, and
only a few very small Wari sites are found,
mostly in peripheral zones.
Wari Forms of Control
This evidence of infrastructure spread
throughout Peru provides a basic picture of
Wari’s political system: a capital city, major
administrative centers extending outward in
all directions, and smaller special-function
facilities placed where they were needed, all
connected to one another by a system of roads.
Such a scheme created the political hierar-
chy through which the empire controlled its
subjects and collected and transmitted infor-
mation from one part of the empire to another.
The roads also served to move armed soldiers
to wherever they were needed and allowed
caravans of llamas carrying their burdens of
goods to travel between the provinces and the
capital.
But what was the basis of economic sup-
port of the empire? How and in what form did
4 2 K atharina S chreiber
Wari collect tribute? We have only indirect ev-
idence to answer these questions, but can turn
to the later Inca Empire for guidance. The Inca
collected tribute in the form of labor, m’ita
(literally “rotation” or “taking one’s turn”). We
cannot know if the Wari also required subjects
to labor on behalf of the empire, but it is obvi-
ous that people supplied vast amounts of labor
to it. Thousands of workers would have been
required to build each of the major administra-
tive centers, the smaller centers, and the roads,
all of which required continued maintenance.
Wari economic control can also be seen in
the reorganization of local agricultural pro-
duction in many regions. In Moquegua, the
agricultural zones around Cerro Baúl were
modified and irrigation systems designed
to increase agricultural production. In the
Sondondo Valley most land below 3,300 m
in elevation was remodeled with stone-faced
terraces, and production of maize increased.
Not only did such efforts require enormous
amounts of labor, but planting and harvesting
these new fields also required labor. The new
crops produced, maize for the most part, were
probably destined for imperial coffers. Chemi-
cal analysis of bones shows that the consump-
tion of maize greatly increased during Wari
times and that even llamas were eating it.29
Perhaps, as in the case of the later Inca, the
Wari converted large quantities of maize into
a fermented beer-like beverage, chicha, which
was served to subjects during religious and
state festivals.
Wari may have controlled a number of ma-
jor sources of important raw materials, includ-
ing Quispisisa, the largest obsidian (volcanic
glass) quarry in the Andes. Obsidian was one
of the best materials available for making
sharp knives and points for spears and arrows.
In the Middle Horizon, the distribution of this
distinctive type of obsidian was suddenly
restricted to sites with Wari associations. The
sources of other minerals, precious metals, and
some imported shells may also have been un-
der Wari’s exclusive control. For example, Wari
sites contain bronze artifacts made of an alloy
of copper and arsenic, derived from smelting
an ore called enargite.30 There are a number of
enargite sources in Wari territory, and Wari
may have controlled one or more of them.
Figure 29. View of Cerro
Baúl, which has a Wari
site on the summit. Photo:
Katharina Schreiber.
This evidence of organization and exten-
sive political and economic control begs the
question: How did the Wari keep records?
Without a writing system, they must have had
some system for keeping numerical records:
census data, tribute payments, the contents of
imperial stores, and so forth. The Inca used
a device of knotted strings, the khipu, which
was quite effective in keeping track of such
things. While rare, khipu have been found in
Middle Horizon contexts, suggesting that the
Wari, too, used them for record keeping (see
fig. 180; see also [155], p. 276).31
Evidence of Wari ideological control is
apparent in the spread of its religious iconog-
raphy, depicted on the ceramics and textiles
that are found in all parts of the empire. More
subtle, however, is evidence that the Wari
also interfered with local ideologies as they
insinuated themselves and their new forms
of domination over conquered peoples. Not
only did the Wari remake local political and
economic landscapes, they also remade the lo-
cal sacred landscapes. Cerro Baúl, a towering,
sheer-sided mesa in the Moquegua Valley, is
an obvious example of an impressive landform
that even today is a sacred place; it was likely
a sacred mountain to the local people when
43 T he R ise of an A ndean E mpire
the Wari arrived on the scene (fig. 29). Despite
the profound inconvenience of establishing an
imperial occupation on its summit, they did
just that, in large measure to make their mark
on the local sacred landscape. Likewise in the
Sondondo Valley, the Wari created a site, com-
plete with D-shaped temples and stone slab
tombs, that blocked access to a local shrine
and sacred mountain. In this case they did not
co-opt the local shrine, but they did change
the way in which it was experienced.32
The Collapse of Wari
Why do empires collapse? There are prob-
ably more answers to this question than there
were empires. It is unlikely that a single fac-
tor explains Wari’s collapse, which seems to
have occurred late in the eleventh century.
Was Wari overextended? Was it trying to
administer more territories than its institu-
tions were able to handle? Like the shark that
must keep swimming in order to live, must an
empire keep expanding in order to survive?
Did Wari’s existence depend on the continual
availability of new regions, resources, and
tribute payers to support what it already had?
And when it ran out of places to conquer, did
it collapse? Perhaps.
 We can probably rule out external invasion
as a factor contributing to the collapse since
no other groups in the Andes were strong
enough to have posed a threat. But internal
rebellion may have played a role as the bene­
fits of incorporation into the empire wore
thin, or tribute demands grew too onerous.
An epidemic illness may have wiped out large
numbers of people, or a disease could have in-
fected one or more of the staple crops, causing
famine and widespread death. In either case
the amount of labor, or produce, available to
support the empire may have dropped below
the minimum needed to support the system.
Climate may have also played a role. A
period of severe and prolonged drought pre-
ceded the rise of the Wari Empire and another
period of climatic deterioration occurred in
the twelfth century. But this drought occurred
after Wari had begun to decline. Climate may
have contributed to the final demise, but other
fatal blows had probably already been struck.
While any one of these factors or a com-
bination of them, or something else entirely,
may be the answer, what we do know is that
when Wari collapsed, there was a major de-
mographic disruption throughout the Andes,
44 K atharina S chreiber
even in areas peripheral to Wari control.
Settlements were abandoned, entire regions
may have been deserted, warfare increased
dramatically, and people moved to new loca-
tions and established new settlements. Clearly
Wari’s collapse brought with it terrible times.
New evidence indicates that, whatever the
cause, the collapse was not a sudden event.
The provincial centers were abandoned gradu-
ally; sections were walled off and doorways
were closed. At Jincamocco, patios and galler-
ies that once had been the center of activities
were filled with trash. Doorways and cor-
ridors were sealed, blocking access to whole
sectors of the imperial enclosure. Likewise at
Pikillacta doorways were sealed, and in one
case an offering appears to have marked the
abandonment of the site.33 In the upper Nasca
Valley, Pataraya was ceremonially closed, and
a thin layer of river sand was spread on every
floor surface when the site was abandoned.34
These were not the actions of people fleeing
for their lives in the face of attack but the
planned, deliberate, and thoughtful responses
of those who recognized that the end was in
sight, and knew what they were giving up.
notes 1. Burger 1992, 192–95; Lumbreras 22. J. Topic 1991, 151–52; J. Topic and
1977. T. Topic 1983–85, 37–42.
2. Thompson et al. 1985, 791–93. 23. Glowacki 2005b, 123.
3. Schreiber 2001, 70–74. 24. Schreiber 1992, 256; Schreiber
4. Cieza de León [1553] 1984, 249. 1991b, 209.
5. Monzón [1586] 1965, 245. 25. Schreiber 1999, 163–65.
26. Schreiber 1991a, 249; Schreiber
6. Besides the initial description by
1984, 89–91.
the colonial traveler, Pedro Cieza
de León, the excavations by Julio 27. See D’Altroy 1992, 14–24;
C. Tello, and visit by American Sinopoli 1994.
archaeologists in 1948, mentioned 28. Schreiber 2001, 71–80; Schreiber
above, there have been a number of 1993, 112–16; Schreiber 1992, 17–27.
small excavations at the site over 29. Finucane et al. 2006, 1771–773.
the past half century or so by Yale
30. Lechtman 2005, 131–33.
University’s Wendell Bennett (1953),
who created the first map of the site; 31. Conklin 1982.
members of Richard MacNeish’s 32. Schreiber 2005b, 144–45.
Ayacucho Archaeological-Botanical 33. Glowacki 2005b, 123.
project (MacNeish et al. 1981);
34. Edwards 2010, 449–51.
William Isbell and his students
from Binghamton University,
who undertook excavations in the
Moraduchayoq sector (Brewster-
Wray 1989; W. Isbell et al. 1991;
Knobloch 1983; Wagner 1981);
Mario Benavides Calle (1979),
who exposed cut stone chambers
thought to be royal tombs in the
Cheqo Wasi sector; Peru’s National
Institute of Culture, under the direc-
tion of Enrique González Carré,
who explored the Templo Mayor
and associated structures in the
Vegachayoq Moqo sector as well as
the Monjachayoq sector (Bragayrac
Dávila 1991; González Carré and
Bragayrac Dávila 1996). Since the
political upheavals of the 1980s and
1990s, work by Peruvian archaeolo-
gists has begun to appear in limited
publications (e.g., González Carré
and Soto Maguino 2004; Pérez
Calderón 1999).
7. Schreiber 1978, 151–60.
8. González Carré and Bragayrac
Dávila 1996, 14–19.
9. Benavides Calle 1978; MacNeish
et al. 1981, figs. 8-12, 8-13.
10. Benavides Calle 1964; Knobloch
1976.
11. Menzel 1964, 8–10.
12. Lumbreras 1974a, 97–98.
13. Anders 1991, 166–67.
14. W. Isbell 1977b, 56.
15. W. Isbell and Cook 2002.
16. Tung and Knudson 2008.
17. Schreiber 1991a, 252; Schreiber
1984, 89–91.
18. Hyslop 1984, 271–73.
19. McEwan 2005b, 152–58; J. Topic
1991, 145–46.
20. Edwards 2010.
21. Schreiber 1992, 199.

45 T he R ise of an A ndean E mpire

Luis Jaime Castillo
Butters Looking at the Wari Empire from the
Outside In
Figure 30 [17]. Cup with After more than a century of research by
axe-bearing supernatural
hundreds of archaeologists at sites located in
being, from San José de
Moro; ceramic and slip; Peruvian coastal deserts and valleys, the high-
15 x 7.4 cm. Pontificia lands, and tropical forests, there is very little
Universidad Católica del doubt that Wari was one of the largest, most
Perú, Lima, M-U1242-C09.
Photo: Daniel Antonio complex cultural and political entities that de-
Giannoni Succar. veloped in the central Andean region. Ample
material evidence supports the claim that it
was bigger, wealthier, and more elaborate than
any Andean society that had existed before.
This evidence comes in the form of monu-
mental sites, such as the Wari capital city and
regional administrative centers including
Pikillacta and Cerro Baúl, as well as in objects
of exceptional beauty: textiles of many kinds,
polychrome ceramics, sculptures in stone and
wood, inlaid ornaments, and metal artifacts
crafted by some of the most gifted artists ever
to work in the Central Andes. Their creations
are the subject of this catalogue.
Only the Inca Empire, several centuries
along the line, achieved more complexity in
terms of organization and influence, or en-
compassed a larger territory and incorporated
more pre-existing societies. Some consider
the Wari to be the antecedent to the Inca, not
only because the two cultures occupied the
same general region but also because the Wari
may have laid the economic, administrative,
and perhaps linguistic foundations on which
the Inca developed an even larger empire in
a short period of time (see pp. 251–67, “Wari’s
Andean Legacy”). Furthermore, to most re-
searchers Wari is the first empire in ancient
South America, and thus the earliest such
form of political organization in the Southern
Hemisphere. In the previous essay Katharina
Schreiber provides a wonderful summary of
this point of view. Thus, Wari was the first
Andean society that went through the troubles
47
involved in creating a polity of a magnitude
never seen before, which it accomplished
without the advantage of foreign inspiration or
influence and with no knowledge that, centu-
ries before, the Egyptians, Assyrians, Chinese,
and Romans had grown to become empires
in their own worlds. Being the first had its
advantages: there was little competition, less
resistance, ample space for innovation, un-
tapped resources, and great opportunities for
reorganization of economics and society. But
climbing to the top of the ladder carried enor-
mous risks and challenges, such as exploring
uncharted territories and developmental tra-
jectories, interacting with unknown societies,
confronting enormously difficult organization-
al issues caused by heterogeneity, and testing
untested internal strengths and capacities, and
these issues are only what had to be tackled in
the early stages of building an empire.
Nevertheless, when the Wari phenomenon
is seen from the outside, from beyond Wari’s
borders, its might dims and its image blurs.
The ways in which Wari interacted with the
societies that existed on its periphery—the
Moche, Recuay, Nasca, and Cajamarca, to
name a few—are far more diverse, more
adapted to local circumstances and opportu-
nities, than the ways Wari exercised power
within its borders. My task here is to give
an alternative point of view, that of someone
standing “beyond Wari walls”1 and looking
in, over the centuries it took for this colossal
society to emerge, grow, decline, and col-
lapse. My particular point of view is that of
the Moche of the northern Peruvian coast, a
society that for much of Wari’s early history
coexisted with it and was influenced by it in
more ways that we usually want to accept.
The Moche, regarded as one of the first state-
 level societies to emerge in the central Andes,
had to confront some of the challenges the
Wari faced later on, but on a much smaller,
regional scale. The Moche never attempted to
go beyond their natural territory, nor did they
try to control peoples of dissimilar ethnici-
ties. The Moche seem to have been too busy
developing their own territories, making the
coastal desert valleys fertile by deploying one
of the largest irrigation systems in the Andes,
to engage in international adventures.2 What
stopped the Moche from conquering their
Recuay and Cajamarca neighbors in the high-
lands and thus becoming the first empire in
ancient Peru? Answering this question might
help us understand better the singularity of
the Wari and their legacy. My research and
fieldwork has focused during the last twenty
years on San José de Moro, one of the Moche
sites that interacted with and received more
influence from Wari than others. More Wari
artifacts, ceramics and obsidian in particular,
have been found at this site than in all the rest
of the north coast of Peru. San José de Moro,
its local populations, and what happened
there during the Middle Horizon (600–1000)
as the Wari expanded and collapsed may be a
barometer of how foreign and distant societies
perceived and interacted with the Wari.
Wari as an Empire
Before recounting the times when the Moche
of northern coastal Peru started to hear about
a mighty empire growing in the high Andean
mountains, assessing several basic ideas is
necessary. The first is whether Wari was an
empire. The critical issue here is whether
we look at Wari from the inside or from the
outside. Only those who work within Wari
walls—the heartland and the largest regional
administrative centers—can assess its true
nature and complexity; those outside are only
able to see parts and pieces, like the elephant
seen through a hole in the wall.
For most researchers, Wari was an em-
pire (that is, a political entity that began as a
regional state and then absorbed and incorpo-
rated neighboring societies under its adminis-
tration, resulting in a cosmopolitan conglom-
erate of ethnicities, languages, and religions).
Military action as well as economics and
48 Luis Jaime C astillo Butters
ideology all must have played important roles
in the construction and long-term mainte-
nance of this large political unit. In compari-
son to small regional states, an empire has to
be more efficient in its organization since it
must manage territories that are not familiar,
populations that are not necessarily loyal, and
resources that must both support the locals
and contribute to the central finances. Devel-
oping an empire requires a strong motivation,
be it commercial interests, political alliances,
or geopolitical, economic, or demographic
pressure; perhaps all these elements play a
part. Once a state starts to grow and evolve
into an empire, incorporating more and more
territory is the only alternative given the in-
creasing cost of running the empire.
Ayacucho was the heartland of the Wari
state, and its capital city, also known as Wari,
served throughout its history as the residence
of its rulers and most important elites, its reli-
gious center, and the headquarters of its suc-
cessful production and distribution network.
William H. Isbell extends the territory directly
under Wari control to the Cuzco region in
the southeast, the deserts of Moquegua in the
far south, and the coastal valleys of Nasca in
the west.3 A recent discovery of sumptuous
Wari tombs at Espíritu Pampa, due east of
Ayacucho, places Wari deep into the tropical
forest for most of its history. (Ceramics of both
early [Ocros and Chakipampa] and late [Vi-
ñaque] Wari styles, as well as outstanding sil-
ver and gold artifacts, were found in the stone-
lined funerary chambers.4) This vast territory,
which encompasses most of southern Peru, is
larger than any previous polity, many times
larger than the sum of all Moche territory, and
even larger than the Chimú Empire that devel-
oped on the north coast a few centuries after
the Wari had vanished. Most of the archaeolo-
gists who enthusiastically support the notion
of a Wari Empire work within this territory,
where the Wari built an impressive infrastruc-
ture of provincial centers that incorporate
distinctive architecture (multistory cellular
patio groups, D-shaped temples, niched halls,
and orthogonal enclosures) to administer their
territories, house their expatriates, store the
goods that they commercialized or collected
as taxes, and host religious ceremonies and
 feasts (see pp. 65–81, “The Wari Built Envi-
ronment,” and pp. 82–101, “The Art of Feast-
ing”). These archaeologists thus feel Wari’s
imposing presence, perceive the changes in
cultural trajectories after the Wari interacted
with a specific region, and recognize the pres-
ence of foreign Wari ceramics and textiles as
well as the changes in local styles and tech-
niques for the manufacturing of all kinds of
artifacts. Moreover, Wari’s arrival seems to be
connected with an increase in land use and an
intensification of agriculture in order to pro-
duce higher yields, which in turn may have
been conditioned by climatic events that gave
an advantage to the kind of organization Wari
represented.5
Only the northern frontiers of Wari terri-
tory are disputed at this point, particularly
the central and northern coasts of Peru and
the northern highland regions of Huamachuco
and Cajamarca. In these areas Wari’s presence
is sporadic and of a lower intensity, and the
interactions that the Wari had with local com-
munities is the subject of much debate. One
of the most intriguing cases of indirect Wari
influence can be found in Huamachuco, in the
northern highlands. There, the Wari began to
build Viracochapampa, which is tradition-
ally regarded as a Wari regional center. These
Wari style buildings, which John and Theresa
Topic interpret as facilities under control of
the local communities, were never completed
or used in spite of the immense labor invested
in them.6 For researchers working in these
areas, particularly those like me, Wari seems
a bit less centralized and coherent, less pow-
erful and forceful, even to the point that we
sometimes question whether it was an empire.
Maybe it was, but not in our neighborhood.
We are accustomed to seeing archaeological
regional states expanding, as the Moche did
on the north coast, incorporating adjacent
territories inhabited by societies similar to
the expanding one, and soon after transform-
ing all aspects of their material culture, from
funerary practices to construction techniques.
Influences occur in palaces and high temples,
fancy elite wares and ritual objects, but they
eventually affected the most humble domestic
settings, where even the household pottery
ends up imitating the wares produced in the
49 L ooking at the Wari E mpire F rom the Outside I n
state capital. But the polities that fomented
these changes before the Wari were not em-
pires. This intensity of cultural transforma-
tion, contradictorily, is not necessarily seen in
the case of the Wari even in its core territory,
for example, around Pikillacta, the largest
Wari provincial capital in the Lucre Valley,
close to Cuzco.7
Schreiber has correctly pointed out that if
we want to know what kind of political entity
Wari was, our questions should not focus on
a definition; otherwise we will end up trying
to fit Wari into a preconceived model. Rather,
we should ask what the Wari did and how
they acted in specific times and spaces, and
under specific circumstances. If we take this
approach, we must recognize that they had an
influence never seen before in the Andes. They
positioned themselves and their material cul-
ture at a level that elites throughout the cen-
tral Andes desired and ended up transforming
the lives of individuals and the trajectories of
societies, even the Moche.
Three issues make the task of understand-
ing Wari, or any historical episode, a very
complicated endeavor. First, the Wari phenom-
enon lasted some 400 years, and in that time
it no doubt changed many times, adapting to
circumstances and opportunities, weaken-
ing or temporarily collapsing, and regain-
ing strength. Second, the Wari responded to
extremely diverse ecosystems, topographies,
and climates: they interacted with socie­
ties in territories as radically different as the
coastal deserts of the Jequetepeque Valley and
the tropical forest of Espíritu Pampa. Finally,
and even more important, all societies are
multidimensional phenomena, composed of
factions and divisions, conflicting interests
and loyalties. So when Rafael Larco or Gordon
Willey8 long ago suggested that the Wari had
conquered the Moche we have to ask: which
Moche were conquered? And reciprocally,
which Wari did the conquering? All these vec-
tors—time, space, social actors, and dimen-
sions—generate incredibly complex processes.
So if we want to understand Wari, or for that
matter any ancient society, we must consider
when, where, and who.
With those considerations, it becomes
evident that Wari was different things in dif-
 ferent times and places. One situation would
have pertained to its relations with a weak
neighbor during its early stages of develop-
ment, when its power and might were in
high supply, but another must have applied
when, during an apparent period of temporary
decline at the onset of the Middle Horizon’s
second epoch, Wari interacted with the pow-
erful Moche states. Its attempts to intervene in
a territory with a totally alien ecology would
have contrasted with its efforts to annex a
neighbor living in an ecosystem similar to
its own heartland, where it could activate a
familiar set of blueprints for a redevelopment
program, including population relocation and
the deployment of a new, more efficient agri-
cultural infrastructure. Schreiber crafted the
concept of “mosaic of control” to describe this
checkered pattern. She was also trying to pro-
pose a flexible conception of Wari’s expansion
different from the more traditional, less subtle
notion that it imposed a “blanket of power,”
that at all times and under all circumstances
Wari acted to impose its will by means of co-
ercion and violence (see pp. 31–45, “The Rise
of an Andean Empire”). Over the last century,
this notion has dominated the way we have
thought of several ancient Andean societies:
researchers have envisioned the Moche mili-
tary marching across the north coast, taking
over one valley after another; a Chimú polity
built by force and run as a corporation; even a
Nasca state expanding in the southern deserts;
and, of course, the Wari Empire imposing its
will and control, subjugating populations,
extracting taxes, and spreading its religion
at will. As archaeological research has pro-
gressed, more moderate views have replaced
many of these conceptions.
The Wari and the Moche
The north coast of Peru, where the Moche and
Chimú developed, differs from the highlands
of Ayacucho, Wari’s heartland. In contrast
to Ayacucho’s green mountains and valleys,
the desert coast—bracketed by mountains
and sea—is a barren, narrow, flat continental
platform that grew from immense amounts of
sediment that washed into the ocean, particu-
larly when heavy glacial caps covering the
Andean mountains melted. Temperatures are
50 Luis Jaime C astillo Butters
mild, never below 15 degrees C (59 degrees
F) or above 30 degrees C (86 degrees F). The
only sources of fresh water are seasonal rivers
that flow down from the highlands during the
summer months. In contrast to the inhospi-
table land, the Pacific Ocean is bountiful, its
offshore waters containing one of the richest
concentrations of fish and mollusks in the
world. Early development on the coast of Peru
seems to have been dependent on the ability
to harvest the ocean, but as societies grew and
agriculture became indispensable, the combi-
nation of mild temperatures, a steady supply
of water, and rich soil made the coastal val-
leys the scenario for the development of the
first Andean states. Irrigation technologies
that transformed coastal valleys into prosper-
ous desert oases seem to have been key to the
development of these complex civilizations,
starting with such local formative societies as
the Cupisnique and Salinar but reaching full
maturity several hundred years later with the
Virú and Moche.
In the lack of written documents, only
archeological research allows us to recon-
struct these societies. This research started
with the pioneering work of Max Uhle, one of
the founders of Peruvian archaeology, who in
1899 conducted the first Moche excavations in
and around the Huaca del Sol and Huaca de
la Luna; these enormous temples, the largest
adobe brick structures in the New World, rest
on the southern banks of the Moche River,
close to the modern city of Trujillo.9 As Uhle
expected, his excavations produced numerous
burials loaded with beautiful Moche ceram-
ics as well as later objects of origin in the
Chimú and Lambayeque cultures. What was
not expected and surprised even Uhle were
several polychrome keros (ceremonial cups)
that postdated the abandonment of the Huacas.
The decoration of these keros reminded Uhle
of the figures on the Gateway of the Sun at
Tiwanaku, a place he knew well (see fig. 6a);
he concluded that a Tiwanacoid occupation
followed the Moche occupation of the site, and
that the former predated the construction of
the nearby Chimú capital, Chan Chan, as well
as the Inca conquest of the north coast. Thanks
to these observations, Uhle figured out four of
the five eras of the Peruvian pre-Columbian
 chronology, which fifty years later John Rowe
formalized as the sequence of Horizons and
Intermediate Periods.10 Rafael Larco,11 and
subsequently Gordon Willey and members of
the Virú Valley Project,12 used Uhle’s observa-
tions to characterize the end of the Moche era
as a confrontation between weakened coastal
societies and a powerful, charismatic, ex-
pansionist society from the highlands, today
known as Wari. These early findings heralded
a period in modern scholarship during which
Wari was thought to have had a long, fruit-
ful relationship with north coast societies,
a theory shored up by the expectation that
with further exploration Wari sites would be
found peppered across the coast and plenty of
Wari ceramics would be excavated, superim-
posed over Moche levels of occupation. In the
century since Uhle conducted his excavations
of the Huacas de Moche, however, no equally
important Wari artifacts have turned up and
no Wari site has ever been documented on the
north coast.13 The only exception is San José
de Moro in the northern Jequetepeque Valley,
where hundreds of Wari objects have surfaced
in their original contexts.
The Moche culture of northern coastal
Peru was never a centralized state with a
capital or a centralized administration, and it
was certainly not an empire. Rather, current
research reveals that between AD 200 and
850 many Moche polities coexisted on the
north coast in a manner somewhat analogous
to the dozens of Maya city-states that thrived
in Mesoamerica during roughly the same
period. The Moche polities fall into northern
and southern clusters and had different sizes
and configurations: on the northern north
coast they were small local states, while in the
southern reaches of Moche territory they seem
to have coalesced into a state centered around
the Huacas del Sol and de la Luna in the
Moche Valley that between AD 450 and 650
(during the Middle Moche phase) expanded
and incorporated neighboring territories to the
north and south. For a while after archaeolo-
gists abandoned the idea that there had been a
single centralized Moche state,14 they assumed
that each valley contained one polity. It seems
likely now that more than one polity devel-
oped within some of the larger valleys.15
51 L ooking at the Wari E mpire F rom the Outside I n
Economically, some Moche states special-
ized in shoreline resources, some were more
heavily agricultural, some controlled strate-
gic commercial routes, and some engaged in
intense “globalization processes,”16 including
long-distance trade and international relations
with foreign Andean societies. In the arts
certain Moche polities excelled in ceramic
production, while others crafted metal objects
in technologies and shapes never seen before.
Most seem to have been ruled by elites who
manipulated religious symbols to achieve
legitimacy and power, and among them some
seem to have been ruled by priestesses of a
cult connected with human sacrifice.17 To
make things even more complicated, it is
evident that over the 550 years of their exis-
tence the Moche states had plenty of time to
change and reinvent themselves, to emerge
from previous Virú and Cupisnique ancestors,
to be newly created thanks to land reclama-
tion programs, to confederate into larger units,
to make alliances among themselves or with
foreign powers, to dissolve, and to disband.
Thus, the map of Moche polities was drawn
and redrawn many times.
Of all the coastal valleys that the Moche
occupied, the Jequetepeque River Valley,
located at the southern periphery of northern
Moche territory, is singular in many regards
(fig. 31). It is one of the largest valleys on the
coast and one of the natural routes into the
northern highlands of Peru, leading to Caja-
marca, a region directly to the east, and to the
tropical jungles of the Marañon River. Even
before the Moche controlled it, the Jequete-
peque Valley was the natural route for season-
ally migrating herders, and along this route
occur Initial Period (1800–1000 BC) chiefdoms
in Kuntur Wasi, Tembladera, Lurífico, and Pu-
emape. Exchange of products—fish and mol-
lusks from the coast, coca leaves and animal
pelts from the highlands, and gold and feath-
ers from the tropical forests—was intense. The
Jequetepeque Valley routes were certainly the
backbone of interactions among the Moche,
Wari, and Cajamarca recounted below.
The process that triggered the emergence
of the Moche was the development and exten-
sion of irrigation systems and technologies. In
the Jequetepeque Valley, centuries of gradual,
Figure 31. Map of Moche
territory showing the
northern and southern
spheres. Drawing: Proyecto
Arqueológico San José de

A
Moro.

IR
H
C

A
N
Vicús

D
Loma Negra

E
N

A
R
IU

N
P

M
O
NORTHERN

U
MOCHE

N
HE

T
LEC
LA

A
L AM BAYE Q U E

I N
Pampa Grande

S
Sipán
ZA
ÑA

Cerro to Cajamarca
San José de Moro Chepén
JEQ
Pacatnamú UET
EPE
QUE
Dos Cabezas La Mina
P

Ascope CH
A

IC
AM
El Brujo A
C

E SOUTHERN
I F

Mocollope and Mayal CH

MO
Galindo
Huanchaco
MOCHE
I C

Huacas del Sol and de la Luna

RÚ
VI

Huaca de la Cruz O
O

A
H
Huancaco C
C
E

Tanguche
SA NTA
A

Castillo del Santa

N

Guadalupito

Pañamarca
ÑA
PE
Northern Moche Region NE

Southern Moche Region

CA SM A
Excavated archaeological site
mentioned in text S
RA
L EB
Excavated archaeological site CU

Unexcavated archaeological site

Y
ME
AR
HU

100 kilometers to Lima and

Central Coast

52 Luis Jaime C astillo Butters

 limited land reclamation projects led by for-
mative societies such as Cupisnique or Salinar
had expanded agricultural land to a maxi-
mum of 10,000 hectares (24,710 acres). Other
developments followed, particularly improve-
ments in plant husbandry, seed selection, and
crop rotation. The natural outcome of this
agricultural expansion was a more reliable
food supply and steady population growth.
Longer, deeper, better-designed canals allowed
communities of somewhat later Virú cultural
affiliation to penetrate farther into the deserts
and to farm vast tracts of land. By the second
century AD, when the Moche began to evolve
from the Virú, the increase in agricultural
land outpaced population growth, thus creat-
ing surpluses and a situation of sudden new
wealth. The technological advancements in
irrigation and agriculture, and the advantages
they brought about for the Moche offer inter-
esting parallels to Wari’s emergence, which
has also been linked to agricultural intensi-
fication. It is likely that the earlier scenario
perceivable on the coast was replicated in the
highland valleys of Ayacucho.18
Early, Middle, and Late Moche
Excavations in Early Moche (AD 200–450)
sites in the lower Jequetepeque Valley, particu-
larly those conducted by Christopher Donnan
at Mazanaca and Dos Cabezas, have docu-
mented the development of distinctive elites
who grew out of Virú roots.19 These elites are
generally known as Moche, and it is only in
Figure 32. A nearly life-
size metal mask (H. 27.5
cm) with gilded copper
sheet metal decorating
the forehead, brow, and
chin, and a gilded copper
ornament hanging from the
nose. The eyes are inlaid
with shell and stone. The
mask was found in Tomb
2 at Dos Cabezas, a Moche
site in the Jequetepeque
Valley. After Milan 2003,
306, fig. 311.
53 L ooking at the Wari E mpire F rom the Outside I n
their burials, temples, and dwellings that su-
perbly crafted ceramics and metal objects in a
spectacular new style, quite different from its
humble Virú predecessor, are found (fig. 32).
Early Moche artists used a wide array of new
techniques to produce black, red, and white
ceramic vessels modeled in the form of exqui-
site figures in a realistic style: humans, deities,
mythical animals, much of the fauna common
to coastal valleys, and a few exotic animals.20
But the artistic and technological revolutions
that characterized the emergence of the Moche
extended to many fields beyond ceramic pro-
duction. In Early Moche times, metalwork in
gold, silver, and copper saw its most remark-
able developments, with the introduction of
gilding and silver plating techniques that were
the most advanced in the central Andes. The
extension of agricultural fields and irrigation
systems peaked during the Middle Moche
(450–650); at the beginning of the Late Moche
phase (650–850), the Jequetepeque Valley had
more than 80,000 hectares (198,000 acres) of
agricultural land.21
In terms of interactions with foreign so-
cieties, the Early and Middle Moche phases
are silent. With the exception of north coast
traditions that coexisted with the Moche—in-
cluding Virú and Vicús, which continued for a
time after the Moche established themselves—
neither foreign artifacts nor influences can be
detected. Even though sites where the Moche
interacted with contemporary highland neigh-
bors have been found in the higher elevations
of Moche territories, these societies’ influence
on the Moche is minimal if not nonexistent.
This state of isolation from events occurring
in the rest of the central Andes seems to have
been an intentional policy, a possible outcome
of the antagonism between the Moche and
their highland neighbors, several of which—
Chachapoyas, Cajamarca, Huamachuco, and
Recuay—were much more regional phenom-
ena than the Moche.
Around 650 things started to change rap-
idly in the Moche world, particularly in the
Jequetepeque Valley. At about this time the
Jequetepeque ceramic sequence shifts from
Middle to Late Moche phases. This chrono-
logical change reflects transformations in
funerary practices, ceramic styles, construc-

Figure 33a. Stirrup- tion technologies, and iconographic themes
spouted ceramic bottle (H.
that eventually precipitated the collapse of
23 cm) painted in the Late
Moche Fineline style. The the Moche by around 850. Fortunately we
image represents a winged can follow the changes step by step thanks to
decapitator that holds a extensive stratigraphic excavations and a large
ceremonial knife (tumi)
and a severed human head. collection of burials dug at San José de Moro.
This vessel was found The transformations that characterize the
in a simple Late Moche Late Moche phase were mostly restricted to
pit burial (M-U 743) at
San José de Moro. Photo:
elite ceremonial and funerary practices, along
Archive of the Proyecto with the artifacts associated with them. Many
Arqueológico San José de other things remained the same, including
Moro.
diet and domestic wares. The three types of
Figure 33b. Stirrup- burial forms that existed in the Middle Moche
spouted ceramic bottle
from San José de Moro.
phase—pits, boot-shaped shaft tombs, and
Drawing: Archive of the chambers—continued on, as did many funer-
Proyecto Arqueológico San ary offerings, such as miniature ceramics
José de Moro.
(known locally as crisoles), large containers
for water and chicha (paicas), and faceneck
vessels. Pit tombs, the simple and poorest
burials belonging to the least privileged sec-
tors of society, show little change.22
In elite contexts, however, there is a sud-
den eruption of new ceramic styles and for-
eign artifacts. The most emblematic new style,
and the one that defines the phase, is Late
Moche Fineline,23 one of the most advanced
styles ever to be developed by Moche artists
in terms of the forms of the objects (predomi-
nantly bottles with triangularly shaped spouts
and conical necks), the technique used to
54 Luis Jaime C astillo Butters
decorate the artifacts (very fine lines used to
execute drawings in great detail), and com-
plex iconographic themes that focus on ritual
and mythical subjects (fig. 33). I have argued
elsewhere that the Late Moche Fineline style
did not develop locally because during the
Early and Middle Moche phases there was no
fineline decorative tradition in Jequetepeque
(the valley’s Middle Moche ceramics are plain
or very simply decorated) or in the entire
northern Moche realm.24 Northern Moche art-
ists selected metal, not ceramic, to represent
complex iconography. So, for example, in the
famous Middle Moche tombs of the lords of
Sipán, ceramics are plain and have little deco-
ration, while metal objects are fantastically
complex. Thus, an artistic style based entirely
on complex line drawings did not evolve in
Jequetepeque. Rather, it may have arrived dur-
ing a migration of artists from the Chicama
Valley in southern Moche territory, where
compelling parallels for the style exist. The
processes that triggered a sudden northward
migration of highly qualified potters from the
Chicama Valley to the Jequetepeque Valley is
mysterious and cannot be understood without
taking into account the role that Wari played
in the transformations of Late Moche society.
Wari in the Context of Late Moche Society
In the same funerary and other contexts that
Late Moche Fineline ceramics occurred, we
discovered direct evidence of interactions
with Wari, particularly in the form of arti-
facts in the Chakipampa style, which the
Wari produced in their heartland during the
first epoch of their existence (fig. 34).25 Even
though not many Chakipampa artifacts have
been found in early Late Moche burials at
San José de Moro, enough have shown up to
correlate unmistakably their appearance with
the first examples of Late Moche Fineline art,
making the arrival of these two styles part of
the same process. Their co-occurrence is not
likely a simple coincidence; instead, either
one generated the other in some way or they
shared a common cause. Perhaps Wari’s sud-
den appearance is related to the movement of
Moche artisans from the Chicama Valley to
the Jequetepeque Valley. It would help if we
could find evidence that the Wari had inter-

Figure 34. The form of this
vessel (H. 17.5 cm), a jar
with a funnel-like neck,
comes from Moche ceram-
ics but its painted motifs
are common to the early
Wari Chakipampa style.
The white clay (kaolin)
from which it is made is
typical of ceramics from
the northern highlands.
The vessel was found sur-
rounded by classic Moche
artifacts in a Late Moche
boot-shaped shaft tomb
(M-U 736) at San José de
Moro. Photo: Archive of
the Proyecto Arqueológico
San José de Moro.
Figure 35. This double-
spout-and-bridge head
effigy vessel (H. 18 cm) in
the Nievería style of the
central coast features the
effigy of an individual with
a moustache and double-
lobed ears, an attribute of
deities. This unique piece
was found in a niche in a
Late Moche chamber burial
(M-U 1022) at San José de
Moro. Photo: Archive of
the Proyecto Arqueológico
San José de Moro.
55 L ooking at the Wari E mpire F rom the Outside I n
acted with the Chicama Moche, where the
fineline style originated, but no such evidence
has yet been found. In any case, the explana-
tion is certainly not a Wari occupation of the
Chicama Valley that forced a refugee move-
ment northward; the Chicama Moche were not
like the sixteenth-century Byzantines, who
migrated out of Constantinople as the Turks
advanced into the city. While the number of
Late Moche Fineline and related ceramics at
San José de Moro implies production by artists
working locally, the same cannot be said of
Wari ceramics at the site, which seem to be
present as a result of exchange and commerce
rather than being produced locally.
Obsidian bifacial blades of typical Wari
form and Wari blackware keros were also
found in elite burials dating from the early
part of the Late Moche phase at San José de
Moro.26 Their materials and manufacturing
techniques suggest that these objects must
have arrived in Jequetepeque as finished items
from areas close to the Wari heartland in the
southern highlands of Peru. It is likely that
Wari textiles were also imported, but no sign
of them has been found, perhaps because pres-
ervation of organic materials is not particu-
larly good at the site. Likewise, Wari figurines,
made of crisocola (a blue to green mineral) or
copper and frequently found at Wari locales,
did not make their way to San José de Moro.
Thus, the number of Wari artifacts found in
early Late Moche contexts is small. They con-
sist of portable objects restricted to the most
complex chamber and boot-shaped burials,
including the tombs of some of the priestesses
of Moro. Wari objects were interred in the most
exclusive areas of these burials—close to the
main body or in the tomb’s niches—and in the
company of many Late Moche artifacts, includ-
ing fineline bottles. Evidently, Wari objects did
not have to climb the social prestige ladder;
as soon as they arrived they were regarded as
status markers. This privileged positioning im-
plies that Wari was perceived as a prestigious
society.
Aside from Wari artifacts, other ceramic
objects of foreign origin were found in the
same high-status tombs, including middle
period Cajamarca (ca. 700–1000) plates and
bowls, produced in the highlands immediately
adjacent to Moche territory. These artifacts,
made exclusively with kaolin clays, are rare in
Moche tombs, while Coastal Cajamarca wares,
a variant made with red clays and decorated
with a kaolin-rich slip, are more common.
(Coastal Cajamarca is a little-understood style
that seems to have originated in the middle
Jequetepeque Valley, midway along the routes
that connect highlands and lowlands.) The
number of both Cajamarca and Coastal Caja-
marca artifacts increase in the latter part of the
Late Moche phase and in the subsequent early
Transitional Period (850–1000), when they be-
come one of the predominant ceramic styles.
Even more unusual artifacts found in
conjunction with Wari (Chakipampa style)
and Cajamarca wares in the San José de Moro
tombs are vessels in the Nievería style, the
elite ceramic style of the Lima culture, a Late
Moche contemporary of the central coast
Rimac Valley, 700 kilometers (435 miles) to the
south (fig. 35). Nievería style artifacts of high
quality and intricate decoration are rare even
on the central coast. The few found in Moro
burials are exceptional in their quality and
preservation, and they attest both to the con-
temporaneity of these styles and to the fact that
these societies—Late Moche, Wari, Cajamarca,
and Lima—were somehow connected. Other
 ceramic styles have been reported from collec-
tions presumably found in San José de Moro,
although their specific context is unknown;
they include the Wari Atarco style from the
south coast Ica Valley and the Teatino style
from the valleys north of Lima. Again in these
cases, their co-occurrence could be a coinci-
dence, but it is much more likely that a single
process was affecting all these regions at the
same time. Whatever this process was, the
Moche did not motivate it, because their scope
was more limited, even though some Late
Moche objects have been found in the Rimac
Valley.27 Wari, which was interacting simulta-
neously with different societies, seems to be
the agent of these international correlations.
Finally, Moche potters also produced a
new style during the early days of the Late
Moche phase: Moche Polychrome,28 which
combined forms, iconography, and decorative
techniques from both Moche and Wari tradi-
tions (fig. 36). Typical forms in the Moche
Polychrome style are stirrup spout bottles and
faceneck jars from the earlier Moche tradition,
but vessel shapes more familiar on the south
coast, such as double spout and bridge bottles
and “lyre-shaped” bowls, are also frequent.
Decoration consists of geometric designs that
Figure 36. This Moche
Polychrome style vessel
(H. 34 cm) combines Wari
polychrome painting of
a diamond-shaped motif
with traditional Moche
ceramic production
techniques and painting.
Faceneck jars are common
both in Moche and Wari
traditions, and this one
has an additional Wari ele-
ment, vertical side lugs. It
was found in a late Moche
boot-shaped shaft tomb
(M-U 1512) at San José de
Moro. Photo: Archive of
the Proyecto Arqueológico
San José de Moro.
56 Luis Jaime C astillo Butters
Moche artists copied from Wari artifacts (the
Chakipampa and Ocros styles) along with tra-
ditional Moche themes, particularly winged
figures, birds, and complex representations
of priestesses in reed boats. Paradoxically,
Moche Polychrome artifacts are the most
frequent fine wares in Late Moche elite tombs;
they outnumber Late Moche Fineline bottles
and imported artifacts, which is strange for a
style that shows such strong foreign influenc-
es. Another surprising aspect of this style is
its very early development in the Late Moche
sequence. It appeared at almost the same time
as the Late Moche Fineline style at San José
de Moro, just as foreign styles also showed
up. I have interpreted the development of this
style as a response to the demand for poly-
chrome artifacts by Moche elites after the first
Wari polychrome ceramics made their way to
the north coast.29 Imported artifacts seem to
have been in short supply, so skilled Moche
artisans produced their own versions of Wari
ceramics, adapting them to the available
resources and the iconographic traditions of
the Moche. Other hybrid Wari styles appeared
throughout the central Andes in conjunc-
tion with the Wari expansion. At San José de
Moro, this hybrid style was not produced as
a response to the forceful imposition of Wari
traditions; rather, it was a local response to
the need to supply a large elite with objects
that identified them with the prestigious Wari
phenomenon.
By 850 all Moche states had collapsed: the
Moche abandoned most of their settlements,
stopped worshiping their gods and creating
their idiosyncratic arts, and their elites and
leaders lost legitimacy and power. The Moche
collapse has been explained in a number of
ways, most of which attribute the decline to
external forces—catastrophic rains and floods
brought on by El Niño events, the encroach-
ment of deserts into agricultural valleys,
droughts, or earthquakes. I am convinced
that the real reasons for the collapse lie in the
internal weakness of political and administra-
tive mechanisms. The Moche created intricate
political systems aimed at running small-scale
societies; these systems relied primarily on
power that derived from ideology and ritual
rather than on coercion. Moche elites created
Figure 37 [19]. Lyre cup true theatrical states and in doing so trans-
with supernatural head,
formed Moche society into a more advanced
from San José de Moro;
ceramic and slip; H. civilization—larger, wealthier, and more
9.5 x 8.4 cm. Pontificia complex. But their success ultimately doomed
Universidad Católica del them because in conditions of greater com-
Perú, Lima, M-U1242-C08.
Photo: Daniel Antonio plexity their organizational principles failed.
Giannoni Succar. Indeed, these principles began to exacerbate
social unrest and, under pressure, eventually
gave way to other forms of organization. Thus,
the demise of the Moche was followed by a
period of turmoil and change that took differ-
ent forms in each region and eventually led
to the creation of larger political entities: the
Chimú and Lambayeque states, both estab-
lished around the turn of the first millennium
of the common era.
In San José de Moro, the Moche collapse
initiated the Transitional Period on the north
coast, a time marked by great cultural eclecti-
cism and a lack of perceivable political leader-
ship. During this time, relations among the
descendants of the Moche, Wari, and Cajamar-
ca people seem to have intensified, perhaps
Figure 38 [18]. Head vessel, because, with the Moche gone, these societies
from San José de Moro;
could begin to exploit the fertile coastal val-
ceramic and slip; 16.1 x 9.2
cm. Pontificia Universidad leys. Thus, agency shifted out of north coast
Católica del Perú, Lima, hands. The Wari had gone through a profound
M-U1242-C06. Photo: transition and re-engineering at the onset of
Daniel Antonio Giannoni
Succar. the second epoch of the Middle Horizon. A
new ceramic style, Viñaque, replaced ear-
lier, venerable Chakipampa wares, and Wari
undertook much more aggressive interaction
with its neighbors. In San José de Moro these
tumultuous events are on display in a series of
early Transitional Period burials, particularly
three large chamber tombs, each of which
contained the remains of priestesses. For ex-
ample, burial 1241, a chamber 20 feet square,
contained the coffin of a priestess surrounded
by hundreds of ceramic artifacts, architectural
models, and metal objects; it tells the story of
continuity and change common to periods of
uncertainty. In the niches that surround the
coffin were clusters of artifacts organized by
regional style: Post-Moche in one, Proto-Lam-
bayeque in another, Cajamarca and Coastal
Cajamarca in a third, and one entire niche
reserved for three of the most remarkable
Wari objects ever excavated in the north coast
(figs. 30, 37, 38). These ceramic vessels include
a lyre cup with a simplified head that may

57 L ooking at the Wari E mpire F rom the Outside I n

 represent the Wari staff deity, a portrait head
that depicts a human who wears a polychrome
headband, and a kero painted with an extraor-
dinary and powerful image—a fierce super-
natural that seems to represent the staff deity,
although the staffs that the deity normally
grasps have been replaced by the sacrificer’s
axe and a severed human head.
Final Thoughts
A profusion of ceramics styles showed up all
at once at San José de Moro during the early
days of the Late Moche phase, which coin-
cided with Epoch 1 of the Middle Horizon
(600–850) and the first Wari expansion. This
and the contexts in which these artifacts oc-
cur provide insights into the transformations
and processes that were operating at the time.
They also allow us to glimpse one of the many
faces that Wari could have had in peripheral
territories.
Wari’s presence on the north coast cannot
be classified as an occupation force, for there
is very little proof of any kind of aggression,
Wari settlements are entirely lacking, and
all evidence of Wari interaction seems to be
confined to the northern Jequetepeque Valley,
specifically, to San José de Moro and its Late
Moche neighbor, Cerro Chepén. Instead, Wari’s
presence on the north coast seems to have
been essentially ideological in that almost all
the artifacts that signal its presence—authen-
tic Wari objects, artifacts associated with soci-
eties in Wari’s orbit, or even locally produced
Moche Polychrome style ceramics—are ritual
materials found in ritual contexts. Even the
obsidian bifacial blades were found in such
contexts and may have been used for human
sacrifices and bloodletting ceremonies rather
than as the weapons of war. This emphasis on
ideology is expected of early charismatic so-
cieties whose influence becomes widespread,
such as the earlier Chavín phenomenon,
which seems to have been religious rather
than political. But it comes as a surprise when
we think of the might of Wari.
Contrary to all expectations, the most ac-
tive agents in these interactions seem to have
been the Moche, not the Wari. This was a time
of profound social and cultural change, when
centuries of isolation suddenly ended and
58 Luis Jaime C astillo Butters
societies in many parts of Peru became open
to foreign influences. In Moche territory the
changes took place on an unprecedented scale
and seem to have been material expressions of
much more complex and widespread cultural
processes. Little or no previous evidence
of foreign influence can be found in Moche
archaeological contexts in the way of exotic
artifacts. Clearly, the Moche opened their
doors to all these artifacts at the beginning of
the Middle Horizon, lifting their restrictions,
accepting material symbols of a foreign ideol-
ogy, and initially placing these symbols in
association with the upper echelons of society.
The sudden arrival of Wari artifacts and the
preferential treatment they received surely
reflects the weakening of Moche elites in the
face of challenging social or meteorological
conditions. But it was not Wari’s exercise of
imperialistic policy that forced the Moche
to accept Ayacucho ideas and symbols. In-
stead, it seems that Moche elites at San José
de Moro and Cerro Chepén wanted to demon-
strate strong affiliation with the prestigious
Wari phenomenon. These elites seem to have
monopolized access to the Wari and to have
parlayed this exclusivity in their local and
interregional power strategies.
In some respects Late Moche elites from
Jequetepeque may have been suffering an
intense case of cultural alienation. It can be
suggested, following Jennings’s concept,30 that
the Moche had entered a phase of “globaliza-
tion,” multiplying their contacts with societies
across the central Andes—Nievería, Caja-
marca, Atarco, Teatino, and of course Wari.
But they may not have interacted with each
of these societies individually and indepen-
dently. Since Wari had an identifiable pres-
ence and vested interests in these territories,
it seems more logical that Wari congregated
them in its transactions with the Late Moche
of the Jequetepeque Valley. What we can-
not explain is the singularity of each Moche
context; for example, the first priestess burial
found at San José de Moro (burial M-U 41) fea-
tured two Nievería bottles and one Cajamarca
plate but no Wari artifacts.31 Could it be that
specific Moche individuals, once in the orbit
of the Wari conglomerate, chose to associate
themselves differentially with the societies
 involved? But why would the Wari promote
the distribution of artifacts from the central
coast or the northern highlands as part of its
interaction strategies with the Moche? What
would they gain?
These questions are not easily answered,
particularly because we do not understand
the kinds of relations that the Wari had with
these societies and interactions could have
been radically different from one to the other.
For example, until recently we had no idea
that the Wari had important strongholds in
Cajamarca, something that Shinya Wata-
nabe’s work in Huaquerones and El Palacio
de Miraflores is demonstrating,32 although
specifically connected with artifacts in Wari’s
Viñaque style, which developed during the
second epoch of the Middle Horizon. At these
sites Watanabe has found Wari architecture
in association with mostly local wares but
also a minute number of Wari ceramics. The
same kind of situation occurs at Cerro Chepén,
where the architecture is partially of Wari and
highland Huamachuco styles, while most of
the artifacts are of local origin.33 Could it be
that local leaders in Cajamarca and in Jequete-
peque employed Wari architects and masons
to build palaces and ceremonial places?
In my interpretation, the data signal that
the Wari and the local population in Cajamar-
ca had forged an alliance, and that the same
model was in the process of being extended to
the coast. This relationship would not im-
ply that Wari had territorial control of these
regions, as the imperial model would have
it, but that it formed strong coalitions with
northern cultures based on shared interests.
This interpretation is quite reasonable, given
the fact that Wari’s presence in the region was
limited to specific strategic locations, leaving
the vast majority of territory strictly in local
hands. The Cajamarca relationship seems to
have been essential to Wari’s overall strategy
since Wari moved through the highlands on
a route that likely connected Cajamarca and
Huamachuco with regions to the south and de-
scended to the coast via perpendicular roads.
As discussed earlier, on the entire north coast
the route to Jequetepeque seems to have been
preferred for highland-to-coast movement,
which explains why almost all evidence of
59 L ooking at the Wari E mpire F rom the Outside I n
Wari’s impact in northern Peru is found in
Cajamarca and San José de Moro.
In spite of Cajamarca’s great distance from
the Wari capital, it seems to have been one
of Wari’s strongest associates, hopping along
wherever the Wari went. Cajamarca artifacts
have been found at Wari’s capital metropolis
and even in the recently discovered elite Wari
tombs at Espíritu Pampa in the Department of
Cuzco.34 But the relationships between Wari
and Cajamarca seem to have been different
from other interactions the Wari developed.
In contrast to what happened in other regions,
there are no perceivable changes in the forms
and decorations of Cajamarca ceramics after
contact with Wari. There is no Cajamarca-Wari
style or Cajamarca Polychrome style, imply-
ing that the Cajamarca were not very heavily
influenced by the Wari in spite of their close
association. But why choose a best friend
who lives so far away? Cajamarca must have
had something very important to offer to the
Wari—although just what remains to be deter-
mined—and it is likely that Wari’s presence in
Cajamarca will prove to be much more intense
than archaeologists expected.
The relationship between the Wari and the
Lima society is more vague.35 Archaeologists
working on the central coast deny that Wari
had any territorial domination and reduce the
Wari phenomenon there to influence that had
ideological overtones rather than real geopolit-
ical control. But then, why do we get Nievería
ceramics at San José de Moro? This is a tough
question to answer, given the fact that to my
knowledge this style of artifact is found on the
north coast only at San José de Moro.
In this essay I have explored the relations
between two immensely complex phenom-
ena: Moche and Wari. To understand them
and their relationship we must keep in mind
a number of dimensions that influenced the
ways in which they interacted: time, space,
and the social actors involved. It is important
to emphasize that the Moche were never a
centralized society but a number of regional
states only one of which, centered at San José
de Moro, managed to control most Moche in-
teraction with Wari. It may have done so to the
exclusion of other Moche states and at least at
the beginning it was Moche elites who took
 the lead in the relationship, not Wari. Thus,
San José de Moro and the transformation that
occurred in the Late Moche phase are essen-
tial to understanding how the Wari confronted
the challenge of embracing the Moche.
Could this peculiar situation have been
the case with other Wari interactions? Wari
appears in the Late Moche archaeological
record in ways that are absolutely unexpected
and perplexing, with the strong mediation of
Cajamarca and a great cultural eclecticism
that included the Nievería, among others. But
Wari’s influence was rapidly transformed into
a local phenomenon, embodied by the Moche
Polychrome style, that seems to have escaped
Wari’s control. It is evident that in the Jequete-
peque Valley the Wari deployed a strategy that
was most appropriate to the conditions of the
local Moche state: they focused on the higher
levels of society and exercised influence
mostly based on the restricted distribution
of portable artifacts loaded with meaningful
iconography. In the process the Wari involved
some of the societies already under their influ-
ence, demonstrating a high degree of cosmo-
politanism in doing so and allowing their
closest associates, the Cajamarca, to assume a
60 Luis Jaime C astillo Butters
dominant role in interactions with the Moche.
Actually, in every territory outside its borders
the Wari behaved in a different way: building
characteristic architecture in some places,
influencing architectural styles in others,
transforming the production of ritual object in
still others. This eclecticism, or adaptability,
is one of the reasons Wari’s imperial character
has been doubted. Whatever the Wari were—
an occupational force, a political influence,
a state extracting tribute from dominated
territories, or all these things—one thing is
undeniable: they had an enormous impact on
the development of local communities. This
impact, the dimensions of which are only
now surfacing thanks to new archaeological
research, is perceivable in Moche territory
in the transformation of local ceramic styles
and the development of hybrid styles that
amalgamated Wari icons and techniques with
local tradition. Perhaps in regard to its foreign
policy Wari was in a state of metamorphosis, a
monster of many heads that rapidly adapted to
circumstances and opportunities, waiting for
the right moment to realize its true essence,
that of the first empire in the central Andes.
notes 1. Jennings 2010b.
2. Castillo Butters and Uceda
Castillo 2008.
3. W. Isbell 2010.
4. Fonseca et al. 2011.
5. Segura Llanos and Shimada 2010.
6. Marcone Flores 2010; T. Topic and
J. Topic 2010a; Watanabe 2002.
7. Belisle and Covey 2010.
8. Larco Hoyle 1945; Willey 1953.
9. Menzel 1977; Uhle 1915.
10. J. Rowe 1962b.
11. Larco Hoyle 1945.
12. Strong and Evans 1952; Willey
1953.
13. Mackey 1982.
14. Larco Hoyle 1945.
15. Castillo Butters 2010.
16. Jennings 2010a.
17. Castillo Butters and Uceda
Castillo 2008; Quilter 2002.
18. W. Isbell 2010.
19. Donnan 2007; Donnan 2006.
20. Donnan 2009.
21. Eling 1987.
22. Castillo Butters 2009b.
23. Castillo Butters 2009a;
McClelland et al. 2007.
24. Castillo Butters 2009a; Castillo
Butters 2001b.
25. Castillo Butters et al. 2008.
26. Castillo Butters 2001a.
27. Stumer 1958.
28. Castillo Butters 2001a.
29. Castillo Butters 2001a; Castillo
Butters 2001b.
30. Jennings 2011.
31. Castillo Butters 2005; Donnan
and Castillo Butters 1992.
32. Watanabe 2002.
33. Rosas Rintel 2007.
34. Fonseca et al. 2011.
35. Marcone Fores 2010; Segura
Llanos and Shimada 2010.

61 L ooking at the Wari E mpire F rom the Outside I n

Transforming the World
Gordon F. McEwan
and Patrick Ryan
Williams
Figure 39. View of walls at
Pikillacta. Photo: Susan E.
Bergh.
[55]. Overleaf, Super­
natural head vessel;
ceramic and slip; 17.6 x
17.1 x 14.7 cm. Milwaukee
Public Museum,
54569/20517.
65
The Wari Built Environment:
Landscape and Architecture of Empire
The Wari set out to change not only society but
the physical reality of their world. For the first
time, centralized planning on a massive scale
remade the Andes into something new. Domi-
nating central Peru for as many as four cen-
turies, Wari overlords left an indelible mark
on the landscape, building some of the largest
monuments ever seen in the region. They are
best represented by the well-studied pro-
vincial centers Pikillacta, Viracochapampa,
Cerro Baúl, Azángaro, and Jincamocco, and
by the less well known capital of the empire,
the city of Wari. These sites provide an excel-
lent sample of Wari architecture but are by no
means all of the Wari monuments. The scale
and cost of this imperial infrastructure was
not equaled until the rise of the Inca Empire
centuries later. The Wari achievement in-
cludes not only distinctive formal architecture
but equally impressive agricultural infrastruc-
ture in the form of terraces, aqueducts, and
canal systems that made a statement of power
still reverberating today.
The City of Wari
The Wari capital, occupied by perhaps 20,000
to 40,000 inhabitants, was a sprawling me-
tropolis covering several square kilometers of
rolling hills.1 It lacks the rigid organization
and grid plan that characterize the provincial
centers because it grew organically as the
empire grew, with new neighborhoods crop-
ping up and expanding the footprint of the
urban center. Wari was a cosmopolitan city,
with a variety of residents representative of
the far-flung reaches of the imperial realm. Di-
verse ceramic styles from both hinterland and
distant regions litter the surface, now covered
with rubble.
Wari is organized into discrete neighbor-
hoods, dissected by high-walled roads that
crisscross the city. Small cities unto them-
selves, these areas have modern names that
recall the Quechua heritage of today’s inhabit-
ants of the region: Moraduchayoq, Cheqo Wasi,
Vegachayoq Moqo, Robles Moqo, and Mon-
jachayoq, to name a few. A system of canals
and drains provided water to residents and
allowed wastewater to leave the city. Within
these neighborhoods—the building blocks of
the city—warren-like passageways connected
buildings. These alleyways were designed for
the residents who knew them; there was no
overt geometrical plan to guide navigation.
Structures in the neighborhoods varied in
construction technique and building style but
followed a central building canon. The most
common building type in the capital was a
courtyard house, known as a patio group. This
ubiquitous form was used not only in the capi-
tal, but in all the provincial centers as well.
Larger scale variants of the patio group served
as administrative buildings.
The capital also contained a number of
D-shaped temples, which are dispersed in
several neighborhoods. These temples, rare in
the provinces, were obviously of major im-
port to the peoples of the Wari capital, where
the largest example, in the Vegachayoq Moqo
neighborhood, measures 20 meters (66 feet)
in diameter. This was likely the principal D-
shaped temple in Wari’s realm, as it is twice as
grand as any other known example.2 D-shaped
temples in the Wari heartland’s second city,
Conchopata, contained evidence that may
relate them to a trophy-head cult practiced
by Wari’s ruling elite.3 Anita Cook has argued
that the D-shaped temples were also places
of ritual sacrifice, identifying a probable D-
Figure 40. Aerial photo-
graph of Pikillacta. After
W. Isbell and McEwan
1991b, 96, fig. 3; courtesy
Servicio Aerofotográfico
Nacional, Peru.
shaped temple in Wari iconography that is
depicted next to a warrior or priest holding a
severed human head.4
Cut stone construction has also been found
in the Wari capital, though it is not typical
in the Wari provinces, where plastered walls
of fieldstone and mortar are more common.
Tombs in the Cheqo Wasi neighborhood are
made of large cut slabs of dark volcanic stone,
fitted together with precision.5 William Isbell
has described a cut stone subterranean temple
in the deepest levels of Moraduchayoq remi-
niscent of earlier Formative Period (2000 BC–
AD 400) construction techniques at Tiwanaku,
Wari’s competitor of the Bolivian altiplano.6
Missing from the capital’s architectural
repertoire are the blocks of small, geometri-
cally organized, conjoined rooms of several
provincial centers. Likewise, the niched hall,
another important architectural form found in
the largest provincial centers, is absent from
the Wari capital.
The Provincial Centers
The Wari imperial architectural style is eas-
ily recognized in the provinces by its rigid
geometry based on rectangular ground plans.
It is widespread throughout a large portion of
6 6 G ordon F. M c E wan and Patrick Ryan W illiams
Peru and represents the signature of the Wari
imperial presence; thus the extent of the Wari
Empire can largely be seen in the distribution
of these distinctive architectural complexes,
which appeared abruptly in the Middle Hori-
zon (AD 600–1000) and spread as the empire
expanded and imposed its rule throughout the
Andes.7
Monumental construction in this archi-
tectural tradition often takes the form of vast
rectangular enclosure-compounds, which are
believed to have served as the administrative
centers and elite residences for those govern-
ing the empire.8 These enormous complexes
each contain numerous individual structures
of several distinct types that exhibit a high
degree of uniformity from one site to another.
This uniformity has led to the belief that the
provincial centers are the product of a central-
ized department of public works responsible
for planning and constructing architectural
monuments throughout the empire.
The compounds exhibit a number of
striking and peculiar characteristics. One
of the first things the observer notices is the
scale of the walls. Where architecture is well
preserved and complete ground plans are vis-
ible—such as at Pikillacta (fig. 40), the larg-
 est provincial site, located near Cuzco in the
southern highlands, and Viracochapampa (fig.
41), near Huamachuco in the northern high-
lands—walls still tower as high as 6 to 12 m
(20 to 40 ft.) and have an average thickness of
up to 1.5 m (5 ft.). Another notable attribute is
size: the best preserved provincial compounds
are enormous. Pikillacta measures 1.68 by 1.12
km (1.04 by 0.69 mi.) and contains more than
700 structures, many of which are as long as
50 m (165 ft.) on a side.9 Cuzco, the famous
Inca imperial capital city, is approximately
the same size. Viracochapampa measures 560
by 580 m (1,840 by 1,900 ft.);10 although the
site was never finished, the builders appear
to have intended to fill it with structures as
densely as Pikillacta.
The high surviving walls in some Wari
sites are often fitted with rows of projecting
stones, narrow shelves, or rows of small niches
interpreted as supports for upper stories (fig.
39). Excavations at Pikillacta revealed stair-
cases and collapsed upper floors, demonstrat-
ing that some Wari buildings were as many as
three stories in height.11 Most Pre-Columbian
Peruvian architecture, including that of the
Inca, is generally restricted to no more than
two stories. Despite their high walls, Wari
buildings have few or no windows. The inte-
riors of the rooms would have been very dark,
even in the noonday sun.
Given their monumental scale and density
of rooms, the interiors of these large com-
pounds are surprisingly inaccessible. Few
corridors or streets penetrate the dense blocks
of architecture, and those that exist tend to
give access to very few structures. Although
doorways are reasonably common between

Figure 41. Plan of Viraco­

chapampa. After W. Isbell
and McEwan 1991b, 143,
fig. 2; courtesy John Topic.

Plaza

Niched Halls Finished

Patio Groups Unfinished
N 200 meters Foundations

67 T he Wari Built E nvironment: L andscape and A rchitecture of E mpire


Figure 42. Three-
dimensional model of
patio groups at Pikillacta.
Drawing: Gordon F.
McEwan.
6 8 G ordon F. M c E wan and Patrick Ryan W illiams
the components of individual structures, such
as patio groups and their surrounding nar-
row chambers, few permit passage between
buildings and rooms. Likewise, entrances to
the entire site, which is often surrounded by
high perimeter walls, are rare and access to
these entrances is restricted. The whole effect
is a maze-like array of passages surrounded
by high walls. Access throughout is extremely
limited as are visibility and sight lines.
Although complex in overall design, the
large Wari provincial compounds comprise
a small group of constantly repeated mod-
ules—relatively simple room and building
types—that combine to form the larger whole.
The result is a complex geometric floor plan.
Studies of Wari provincial architectural com-
pounds, especially at Pikillacta because it is
the best preserved, but also at Viracochapam-
pa, Azángaro, Cerro Baúl, and Jincamocco,
suggest the following typology of structures to
define Wari provincial architecture.
Architectural Types and Functions
patio groups. The most commonly occur-
ring forms in Wari architecture, patio groups
are rectangular structures containing long,
narrow, roofed chambers surrounding a patio
that is open to the sky (fig. 42). Patio groups
may be further subdivided into symmetrical
and asymmetrical subtypes (sometimes the
same number of chambers does not occur on
all sides of the patio), although a functional
distinction between the subtypes has not yet
been discerned. Gordon McEwan and Nicole
Couture suggest that the differences in the
internal layout and number of long narrow
chambers around the open patio may be the
result of the structure expanding over time.12
Since patio groups are cellular in nature
and embedded with the surrounding struc-
tures, the only expansion possible is upward
through the addition of stories or inward
through the addition of parallel narrow cham-
bers. This architectural plan produces an
inward-looking arrangement and offers a great
deal of privacy because of the limited number
of access points, sight lines, and the extremely
high walls. Privacy and security seem to be
emphasized in these spaces.
Evidence for the function of these rooms
consists of the artifacts found within them. At
Pikillacta they consist principally of camelid
bones (llamas and alpacas) and pottery shards
but also a small number of bronze objects,
including sewing needles and shawl pins
(tupus) that were worn by women, which im-
plies their presence in these structures. A few
exotic and valuable artifacts made of imported
materials such as Spondylus princeps (a red or
orange spiky oyster shell), colored stone beads,
and obsidian suggest the presence of elite and
high-ranking persons.
Mary Glowacki’s study of the Pikillacta
ceramic collection indicates that the majority
of the vessels were used to prepare and serve
food and drink, not for storage.13 A number of
hearths were encountered in some of the patio
groups, which could represent locations where
food was prepared or warmed. Although a
majority of the chambers excavated within
the patio groups were empty, a few showed
evidence of having been used to store food.
These chambers do seem well suited to stor-
age, however, and if the contents were valu-
able (perishable nonfood stuffs such as fine
cloth or feathers, for example), they may have
been removed at the time of abandonment.
The cellular nature and small dimensions of
these chambers make their use as living quar-
ters seem unlikely although we cannot rule
out this function. There is also the problem
of illumination in these multistory buildings;
 they are dark and uncomfortable spaces to the
modern observer.
Drawing on the Inca analogy, many ar-
chaeologists suppose that the function of the
patio groups was feasting,14 which probably
incorporated both administrative ritual and
religious practice, perhaps simultaneously.
Inca administrative ritual similarly involved
bringing together a ruler or his representatives
with his subjects for the purposes of ceremo-
nial performance related to a formal request
for labor. Ceremonial feasting and drinking in
which great amounts of food and native corn
beer (chicha) were consumed characterized
these rituals. Conspicuous generosity was
also practiced at these feasts, and the numer-
Figure 43. Three-
dimensional model of a
niched hall at Pikillacta.
Drawing: Gordon F.
McEwan.
ous chambers in the patio group structures
perhaps were used to store goods to be distrib-
uted to feast participants. The archaeological
evidence for ritual feasting is the presence
of large numbers of ceremonial serving and
drinking vessels in the artifact collections
of a site, as at Pikillacta. Drinking chicha is
ritually essential yet also dangerous in that
it renders the participants drunk, insensible,
and therefore vulnerable. The patio groups
could have provided secure, private locations
for such performances. Variations in sizes of
patio groups perhaps reflect the size of the
administrative unit or kinship group involved
in the feast (see also pp. 82–101, “The Art of
Feasting”).
Patio groups at other Wari sites have occa-
sionally produced different kinds of evidence,
69 T he Wari Built E nvironment: L andscape and A rchitecture of E mpire
suggesting that the correspondence between
this architectural form and the activities
carried out within it are not as specific as
we might think. At Jincamocco, Katharina
Schreiber found evidence for cooking in patio
groups, which is consistent with feasting.15
At the Wari capital city, however, William H.
Isbell, Christina Brewster-Wray, and Lynda
Spickard found trash deposits, suggesting
domestic occupation in some of the patio
groups.16 Unfortunately whether this was an
original or secondary usage of these buildings
is not clear. The lack of primary archaeological
deposits in many patio groups makes a secure
diagnosis of structural function impossible. It
is also likely that a variety of functions took
place in these structures, ritual feasting being
among the most important for state purposes.
niched halls. These chambers appear to
have been completely roofed over without an
internal open-air court. They are found only
at Pikillacta and Viracochapampa, sometimes
embedded in the surrounding architecture and
at other times freestanding in the courtyards
of patio groups (fig. 43). As a class of build-
ings, niched halls have consistent diagnostic
characteristics but vary considerably in terms
of size and form. One defining characteristic
is large wall niches with a trapezoidal foot-
print that is narrower at the front and wider at
the back. The niches are found in a variety of
sizes, and there are a number of different niche
placement patterns within the walls. Some
buildings have them only in the short end
walls. In others, they are located on either side
of a corner. Examples from Viracocha­pampa
exhibit multiple rows of niches located in all
walls. The other major category of variation
among the niched halls is in size. The small-
est known example at Pikillacta is 3.5 by 4 m
(11.5 by 13 ft.), whereas the largest example is
15 by 42 m (49 by 138 ft.). The significance of
this disparity is, at present, unknown. These
buildings have internally rounded corners,
and offering pits are present in walls of the
corners, in the floor in front of each corner,
and beneath the thresholds.
In the case of niched halls the argument
for a single specific function can be made with
more confidence than for the patio groups.
 Niched halls seem to be the focal points of
activity at both Pikillacta and Viracocha-
pampa because of their association with the
largest open spaces within the site. Although
not as numerous as the patio groups at each of
these sites, substantial numbers are present:
eighteen at Pikillacta and nineteen at Viraco-
chapampa.
The excavations of niched halls at Piki­
llacta uncovered surprisingly few artifacts,
which were almost exclusively from offerings
in specially prepared pits located below floor
level in the corners or thresholds of the build-
ings. Almost all come from previously looted
contexts. (The few pottery fragments found
outside these offerings are not very illuminat-
ing.17) The reason for the scarcity of artifacts
is that the floors of niched halls were lined
with gypsum plaster; its hard surface does not
permit artifacts to be trod into the floor or ac-
cidentally deposited by other means. The care
taken in building these white gypsum floors
implies that they were kept clean and probably
frequently swept of any debris that might have
accumulated. Further, evidence for deliberate
and orderly abandonment implies that noth-
ing of value was left on floor surfaces. As a
result, the only artifacts remaining were those
deposited as offerings, including bronze ob-
jects, Spondylus princeps shells, and camelid
bones. These offerings provide evidence that
the function of these buildings was religious.
Originally these offerings probably contained
additional objects that directly link the func-
tion of the buildings with ancestor worship.
This suggestion is supported by two
elaborate offerings recovered at Pikillacta,
both consisting of turquoise-colored stone
figurines representing costumed humans
(see fig. 223). Looters found these offerings in
1927 in offering pits 3 m (10 ft.) deep in the
corners of a niched hall that McEwan re-
excavated in 1989. Each set contained forty
figurines dressed in distinctive costumes and
headgear.18 Cook has provided the results of a
thorough analysis of these objects that sheds
light on the function of Pikillacta as a whole.19
She concludes that these figurines “arguably
represent the legendary 40 founding ances-
tors of the Wari polity.”20 Further, she suggests
that “ancestor worship could be appropriated
70 G ordon F. M c E wan and Patrick Ryan W illiams
by the state to ensure rights of inheritance,
domination, and sacred legitimacy and in this
sense serve administrative ends.”21 A similar
use of ancestral figurines has been reported by
Katherine Julien for the Inca.22 She comments
that, according to the chronicle of the Spanish
conquistador Juan de Betanzos, miniature gold
figures representing the lineages descended
from Manko Qhapaq, the legendary founder of
the Inca ruling dynasty, were buried at the foot
of a stone representing the sun that was set up
in the main plaza of Cuzco.23
At Viracochapampa, the only other site
with niched halls, there is not much evidence
related to the halls’ function because the site
was never completed or occupied. Niched halls
at both sites, however, contained evidence
of secondary human burials. At Pikillacta,
this evidence consisted of a cache of human
skulls located under the floor in one corner of
one niched hall; looted pits in the corners of
other niched halls suggest that the walls had
originally held human remains. At Viracocha-
pampa secondary burials consisting of parts
of several bodies were found in a pit in the
corner of a niched hall.24 These results suggest
a similar function for these structures at both
sites.
Useful comparisons also can be made
with earlier and later examples of ceremo-
nial buildings from other Andean cultures.
The association of wall niches and internally
rounded corners with ritual buildings is very
ancient in the Andes. The most salient ex-
amples are found in highland temples of the
late Preceramic Period (2900–1800 BC), such
as La Galgada.25 Later, buildings with rounded
corners and wall niches occur at the Initial
Period (1800–1000 BC) site of Moxeke in the
Casma Valley on the north-central Peruvian
coast. Huaca A at Moxeke, a multichambered
structure made up of rooms with rounded cor-
ners and walls containing numerous niches,
is believed to have been used to store ritual
paraphernalia and foodstuffs.26
John Topic has proposed that niched
halls were introduced into Wari architecture
through contact with the earlier north high-
land Huamachuco culture, which commonly
built such halls.27 He proposes a sequence of
these structures starting with the Huama­
 chuco forms of the Early Intermediate Period
(AD 1–600) and continuing with the Wari
forms of the Middle Horizon. This sequence
culminates in the Inca kallanka, a very large
niched hall with numerous doorways that,
the Spaniards say, had ceremonial functions.
Topic believes the kallanka was inspired by
the niched halls the Inca saw at Pikillacta,
which is close to Cuzco, the Inca capital.
Another important analogous Inca struc-
ture known to have functioned as a ritual/
ceremonial building is the great temple of
Viracocha at Raqchi, southeast of Cuzco. This
temple and many of its adjacent structures
were created in the form of niched halls.
They are built of stone and adobe and exhibit
similar proportions to both Inca kallankas and
Wari niched halls. Further, excavations by ar-
chaeologist Bill Sillar28 are demonstrating that
a Wari occupation underlays the Inca temples
at this site.
Figure 44. Plan of
Pikillacta. After McEwan
1990, 102, fig. 3; plan:
Gordon F. McEwan.
71 T he Wari Built E nvironment: L andscape and A rchitecture of E mpire
Thus, multiple lines of evidence suggest
that Wari niched halls were possibly related
to ancestor worship. The prominence and
frequency of these buildings within the two
largest Wari provincial sites, Pikillacta and
Viracochapampa, speaks to their importance.
In addition, artifacts related to ancestor wor-
ship seem to have been associated with this
class of structures. In terms of form, internally
rounded corners and distinctive wall niches
are characteristics of ceremonial buildings
dating back as early as late Preceramic times,
and niched halls continued to serve similar
purposes into the Inca period. By analogy,
the presence of these characteristics in Wari
niched halls implies ceremonial function.
small conjoined rooms. These rooms are
built in rows and share end walls. Their size
varies but is always considerably smaller than
the other types. Those at Pikillacta (fig. 44) av-
N 100 meters
Figure 45. Plan of erage about 4 by 5 m (13 by 16 ft.) while those
Azángaro. After Anders
1991, 169, fig. 3. at Azángaro (fig. 45), a small Wari agricultural
facility near the capital, average about 2.5 by 9
m (8 by 29 ft.). They have internally rounded
corners and sometimes externally rounded
corners.29 These small structures, arranged

North
in groups of rows with carefully controlled
access, have occasioned much speculation in
regard to their function. A number of archae-
ologists have thought that they might be stor-
age structures like those used by the Inca.30
Others have suggested a ritual function, which
is plausible given their rounded corners. Exca-
vations at both Pikillacta and Azángaro could
not confirm any particular interpretation.31
Most of these structures at both sites appear
never to have been completed, making inter-
pretation of their intended function difficult.

d - shaped structures. As their name implies,

these structures have a floor plan in the shape
of a “D” (fig. 46). In both the provinces and the
Central

heartland, most measure approximately 10 m

(33 ft.) in diameter. They invariably looked out
onto an open plaza or courtyard. The entrance
to the structure was by way of a door in the
center of the straight wall; on the interior, a
pair of niches flanked either side of the door.
Likewise, a set of four niches was set in the
curved interior wall in each of the other three
directions. This pattern of sixteen niches
organized in four groups of four is repeated in
every known D-shaped structure.32
As mentioned above, D-shaped struc-
South

tures are relatively abundant at the capital

and select sites in the heartland, such as at
Conchopata and in the Chicha-Soras region.33
They are often found in pairs in the provin-
cial sites. South of the Wari heartland, they
are thus far best known from Cerro Baúl, the
southern frontier provincial center (fig. 47).34
In the northern realm, they are present at
Honco Pampa in the Callejón de Huaylas. They
Excavations are, however, not present everywhere, includ-
. . . Limits of irregular ing Pikillacta and Viracochapampa. The cult
buildings
practiced in the D-shaped structures was obvi-
ously diffuse, but not critical to Wari existence
50 meters N in every provincial setting.
Where D-shaped structures have been
excavated, they are in one of two patterns. At
Cerro Baúl, the interiors are completely clean

72 G ordon F. M c E wan and Patrick Ryan W illiams


Figure 46. View of a
D-shaped structure in the
Vegachayoq Moqo sector
of the Wari capital. Photo:
Susan E. Bergh.
as a result of either ongoing maintenance or
meticulous cleaning prior to abandonment.
Small offerings, buried in pits around the exte-
rior walls or beneath the floor in the center of
the room, include ceramic or gourd vessels or
metal foil cut in the shape of a llama.35 Else-
where, such as Conchopata, floors were strewn
with quantities of smashed ceramic vessels
used for brewing and serving chicha. Several
human trophy heads were also found in these
contexts.36
Construction, Engineering, and Costs
What was involved in building a Wari center,
how much did it cost, and how do we interpret
the cost? In studying a society without money
it is impossible to come up with a cash figure
for costs. Nevertheless we can obtain an idea
of how much human energy was involved in
these projects and assess costs in terms of
person-days of labor. McEwan has done this
for Pikillacta by systematically examining
the various steps involved in bringing a Wari
construction project to fruition.37
The building of the monumental Wari sites
throughout the realm was surprisingly labor
intensive in ways that are not at first obvi-
ous. It is apparent, however, that a tremen-
dous amount of stone had to be quarried and
brought to the construction site; thousands
of tons were transported by human energy
alone since the Wari lacked wheeled vehicles
and draft animals. But even before begin-
ning to acquire the basic building material,
the plan of the monument had to be recorded
and transmitted by engineers who did not use
writing or paper. We do not know how this
was done, but one possibility would have been
to record the information on a khipu, a device
made of strings on which numerical informa-

Figure 47. Plan of Cerro

Baúl. Plan: Patrick Ryan SECTOR E
Williams.
m SECTOR A
75
Ritual 2,5
platform

SECTOR B
D-shaped
temples

SECTOR C

m Palace
0
30

Brewery
m
5
57
2,

SECTOR D
Administration
Temple
and Storage
Annex

Excavated areas N 30 meters

73 T he Wari Built E nvironment: L andscape and A rchitecture of E mpire

 tion was enciphered via a complex system
of knots (see fig. 180; see also [155], p. 276).
Khipus record information in a highly orga-
nized and orderly hierarchy, and the repetitive
geometry of Wari architecture lends itself to
being recorded by such means.
Arriving at the site with the appropriate
plans, the engineer in charge of the project
was faced with the task of surveying the
ground and laying out the plan. The next step
was excavating the wall foundations, an enor-
mous earth-moving project for at some sites
the foundation trenches were as much as 4 m
(13 ft.) deep. While the excavated soil could
be saved to make adobe mortar for use in
constructing the walls, more soil would have
to be excavated to provide enough mortar.
Adobe mortar also requires large quantities of
grass, used as a temper, and crews were likely
devoted to finding, cutting, and transporting
sufficient amounts.
Many Wari architectural monuments were
apparently completely plastered with white
gypsum, which posed another set of problems.
Before it can be mixed with water to make
plaster of paris, gypsum must be mined and
then reduced to a powder by heating it in fire.
The quantities necessary would involve a con-
siderable labor force dedicated to producing
the plaster. Once prepared, it had to be moved
to the construction site and then applied to
the walls and floors of the buildings.
The wood beams supporting the floors
of the many two- or three-storied structures
presented a particularly difficult problem.
Building these floors required huge quantities
of wood—more than was commonly available
near many highland sites—which implies a
large labor force trekking to distant areas to
acquire enough wood and carry it back. Wood
was also necessary to fuel the fires used to re-
duce the gypsum to plaster as well as to build
the framework for the thatched roofs of the
buildings. Such usage increased the amount of
wood needed to tremendous amounts.
Once all the building materials were as-
sembled on-site, some of the not-so-obvious
costs began to appear. In order to make the
adobe mortar and the gypsum plaster, a large
quantity of water was required. The most ef-
fective way to move vast amounts of water is
74 G ordon F. M c E wan and Patrick Ryan W illiams
through canals and aqueducts, which seems to
be what the Wari did. Creating them, in turn,
was a separate major construction project.
Finally, one of the greatest costs was the ne-
cessity to feed the labor force, provide shelter,
and haul away human waste and garbage so as
to keep the work force healthy and not spoil
the environment. For a large monument such
as Pikillacta, the total cost of the project mea-
sured in labor would have come to almost 8
million person-days. That the Wari were able
to engage in such an undertaking is amazing.
What is even more impressive and provides a
true measure of their power is that they were
able to build multiple centers throughout the
empire at about the same time.
Wari Hydraulic Works
The Wari, master landscape engineers, were
the first to transform the highland Andean
landscapes of the Pacific watersheds through
terraced irrigation agriculture. Their water­
works were especially well developed in
the regions surrounding the most important
provincial centers: Pikillacta and Cerro Baúl,
located, respectively, in the Lucre Basin of the
Cuzco Valley and in the Moquegua Valley.
Valencia Zegarra has documented a total
of seven principal canals with a total length
of 48 km (30 mi.) in the Lucre Basin.38 The
largest of them, Canal A, is 16 km (10 mi.) long
and runs from the Chelque River southwest of
Pikillacta to the city itself. One of the largest
ever constructed in the area, Canal A had the
capacity to carry 850 to 1,700 liters (225 to 450
gal.) of water per minute. During its course, it
crosses two aqueducts (Cambayoq and Rumi-
colca) and passes through a tunnel 5 m (16.5
ft.) long cut from solid rock. The final 5 km (3
mi.) of its course before it reaches Pikillacta
averages a mere 0.07 percent grade, a true feat
of engineering skill. Most ancient earthen
contour canals maintain a grade of 0.5 to 1.5
percent. Achieving a slope close to zero over
long distances requires advanced engineering
capabilities lest part of the course run uphill.
The aqueducts of Cambayoq and Rumi-
colca (fig. 48; see also fig. 238) are impressive
monuments in their own right. Both have a
zigzag plan, with the former 87 m (285 ft.) long
and the latter 239 m (784 ft.) long. Both were

Figure 48. View of the
Rumicolca aqueduct at
Pikillacta. The aqueduct
flowed along the top of the
long wall shown in the
center of the photograph.
The Inca later made the
two ruptures in the center
of the wall in order to
create a gateway. Photo:
Gordon F. McEwan.
75 T he Wari Built E nvironment: L andscape and A rchitecture of E mpire
more than 7 m (23 ft.) in height, maintain-
ing water elevation across low points in the
landscape. The principal agricultural land
that Canal A fed is upstream of the aque-
ducts, indicating that the main function of the
aqueducts was to transport water for urban
consumption and impress visitors with their
massive structures.
The Wari also constructed agricultural ter-
races in great numbers throughout the farmed
area. This modification of the mountain slopes
into a series of platforms resembling giant
staircases created a microenvironment ideal
for crop growth. Each terrace was constructed
by building a long stone wall—often more
than 1 m (3 ft.) high and tens or hundreds of
meters long—behind which gravel was laid.
Soil from above was then pulled down to
create a series of flat platforms on what was
once a steep slope. Terracing provides several
benefits, including moisture retention without
inundation, prevention of soil erosion, and
adequate space to expand agricultural produc-
tion. Water is brought to the terrace surface by
canals fed from springs, rivers, or reservoirs.
Several reservoirs also graced the Wari
irrigation system in the Lucre Basin. The three
principal reservoirs associated with the larg-
est canals could store 225 to 230 cubic meters
(294 to 300 cubic yards) of water each, and one
had a stone-paved floor and an outlet measur-
ing 30 by 30 cm (12 by 12 in.). These large
man-made lakes were used to feed canals that
reached agricultural zones at great distances
from the water sources and to augment the
flow of Canal A to Pikillacta. In the Middle
Horizon, the reservoirs, in conjunction with
the canal systems, provided water for Piki­
llacta and the 572 hectares (1,413 acres) of
irrigable land in the basin.
The effect of Wari hydraulic works on the
Lucre Basin cannot be overestimated. Dem-
onstrating the Wari mastery over water, they
effectively transformed the landscape around
Pikillacta into lush gardens, man-made lakes,
and impressive stone monuments. The conver-
sion of the 105 sq. km (40.5 sq. mi.) basin from
dry hillsides into a garden of immense propor-
tions is on a scale of engineering prowess that
rivals the construction of the city itself.
While the waterworks around Pikillacta
are impressive, the changes to the landscape
the Wari initiated in the Moquegua Valley on
their southern frontier are even more remark-
able. The settlement there is centered on the
great mesa of Cerro Baúl, a mountain (cerro)
with sheer cliffs that towers 600 m (2,000
ft.) above the valley floor and has a flat top
1.5 km (0.6 mi.) long (fig. 49). The Wari sites
in Moquegua are located on the summit and
slopes of Cerro Baúl and Cerro Mejía, which
are adjacent mountains. The most sumptuous
monumental architecture and highest status
elite architecture were located on the peaks.
Positioned to control the sacred pinnacles,
these locales also provided a defensive loca-
tion on the Wari-Tiwanaku frontier. Situated
on mountain crests, the elite placed them-
selves closest to the gods. In this, the driest
desert environment of their realm, the Wari
were challenged to supply their citadel with
water and food. To do so, they constructed a
massive canal system that drew water from
the Torata River to the ridge upon which their
mountaintop cities were built (fig. 50).
The principal canal was at least 14 km
(8.7 mi.) long and had a discharge capacity of
400 liters (106 gal.) of water per second as it
reached the base of the urban settlements at
Cerro Baúl and Cerro Mejía. This same canal
irrigated up to 324 hectares (800 acres) of
agricultural land on the slopes of the colony’s
Figure 49. View of Cerro three principal hilltop settlements: sites on
Baúl, facing south. Photo:
Cerros Baúl, Mejía, and nearby Petroglifo.39
Patrick Ryan Williams.
Another branch of this canal reached 6 km
(3.7 mi.) on the slopes above the Quebrada
Cocotea (a quebrada is a ravine) to irrigate
up to 150 hectares (370 acres). Because both
canals followed the contour of the slope in
sections and flowed down the tops of ridges in
others, the Wari were forced to invent ingen­
ious ways of managing the force of the water.
They created broad, deep, stone-lined seg-
ments (which reduce the drag on water move-
ment) to keep large quantities of water flowing
along the low (1 percent) grade of the contour
canals, and they made mountain road-like
switchbacks and stone steps to slow the water
as it flowed down the steeper grade (5 percent)
ridgetop canals.
Wari engineers in Moquegua also built
at least one massive aqueduct, at El Paso, to
transport water across mountain passes. Sev-
eral smaller aqueducts were also constructed.
The El Paso aqueduct was destroyed by the
construction of a colonial road, later a high-
way, but footings on one side were preserved.
The aqueduct, 6 to 10 m (20 to 33 ft.) in height,
would have spanned some 50 m (165 ft.).
Wari terracing in Moquegua, though
heavily eroded from seismic activity and
76 G ordon F. M c E wan and Patrick Ryan W illiams
occasional wet season downpours,40 is still
preserved on the slopes of Cerros Baúl, Mejía,
and Petroglifo. Archaeological survey data
suggest that the abandonment of the settle-
ment in Moquegua also included the agricul-
tural fields around Cerros Baúl and Mejía and
on the slopes of Quebrada Cocotea. But the
upper part of the Wari agricultural system was
used by later peoples, including the Inca, and
is still in use today.
Wari waterworks in the Moquegua Val-
ley created the region’s first intervalley canal,
bringing water from the Torata River across
the intermontane divide and into the drain-
age basin of the Quebrada Cocotea and the
Tumilaca River. Unprecedented at its time, the
extent of this system would not be matched
until the Inca arrived in the Moquegua Valley
and transformed the agricultural landscape
once again.41 In fact, the scale of Wari water
management—which involved constructing
and maintaining canals capable of carrying
1,000 liters (265 gal.) per second—would not
be surpassed in the Moquegua Valley until
present-day national irrigation projects. Most
modern canal systems transmit 75 to 150 liters
(20 to 40 gal.) per second. The Wari operated at
a different scale, demonstrating a mastery over
water and landscape unequaled even today.
Figure 50. Map of the canal
system at Cerro Baúl. Map:
Patrick Ryan Williams. Ancient cultivation
Architecture
Cerro Petroglifo
Canal
100 m contour
Agricultural terrace
17.10° S
Cerro Mejía
Wari Canal
Cerro Baúl
17.12° S
70.87° W
Similar Wari waterworks and agricultural
transformations spread across the Peruvian
highlands. At Viracochapampa, John Topic
notes the existence of a canal entering the city
that connects to an earthen aqueduct at La
Cuchilla, 5 km (3 mi.) distant.42 This aqueduct
is 800 m (0.5 mi.) long with a height of 6 to
10 m. Topic suggests, given the steep slope of
4 percent between the aqueduct and Viraco-
chapampa, that a high-pressure water system
to feed the city’s canals may have been the
intention. He observes, though, that the site
was never finished and the canal was never
put into operation. Katharina Schreiber argues
that, at Jincamocco, agricultural terracing
was introduced during the first phase of Wari
occupation, around AD 700.43 This transfor-
mation profoundly altered the landscape,
creating an environment in which maize
77 T he Wari Built E nvironment: L andscape and A rchitecture of E mpire
El Paso
N 600 meters
70.85° W
agriculture, previously unsustainable at the
region’s altitude, could flourish. The story of
Wari skills as the masters of water manipula-
tion and as landscape sculptors is repeated
throughout the valleys of Peru’s highlands.
The Wari truly changed the world as they and
the peoples they conquered knew it.
Transformation of agrarian landscapes was
not all for show, of course. Wari agricultural
technology—high-elevation canals and irri-
gated, terraced mountainsides—proved to be
adaptive. One important benefit was to situate
fields, previously located at lower altitudes,
much closer to water sources. The empire
expanded in a time of climatic variability,
with rainfall amounts varying substantially
over decades. In fact, during the initial expan-
sion beyond the Wari heartland (AD 560–90),
a drought characterized by 30 percent less
 rainfall than the long-term norm apparently
ravaged the Andes.44 In Moquegua, agricul-
tural fields closer to the sources of water
running through mountain rivers were more
than twice as efficient in water use than lower
valley agriculture.45 Such efficiency occurs be-
cause high-elevation terraced agriculture loses
less water to evaporation and seepage since
the water does not have to travel tens or hun-
dreds of kilometers to the distant lower valley
fields. Likewise, the terracing technology
the Wari introduced conserved soil moisture
behind stone walls in a way that normal fields
do not. Wari hydraulic technology, involving
canal design and agricultural engineering,
transformed landscapes in both aesthetic and
functional terms.
Interpreting the Wari Built Environment
The Wari built environment also reflects the
empire’s ideology of expansion and conception
of its place in the cosmos. Wari elites strategi-
cally positioned themselves as intermediar-
ies with the supernatural by usurping sacred
places in the environment.46 They illustrated
their control over life-giving water, mediated
at least at Cerro Baúl through their interaction
with the ancestral mountain spirits (apu) of
local peoples, by means of canal construction
and irrigation. Their radical alteration of the
landscape they inhabited had major repercus-
sions for local populations.
Wari lords manipulated elevation and
architecture to mimic the cosmological move-
ment of water,47 orienting both their archi-
tecture and their rituals (performed at sacred
places in the landscape) toward mountain
peaks, the wellsprings of water and the homes
of the apu. The vestiges of these actions in-
clude platform complexes situated to provide
vistas of holy mountains at places like Cerro
Baúl,48 as well as traditional ritual architec-
ture built near huacas (landscape features
of spiritual importance).49 Thus, not only
did the Wari physically transform the world
before the perhaps incredulous eyes of their
conquered subjects, they also appear to have
actively co-opted local ideologies and sacred
places into imperial doctrine while introduc-
ing and implanting Wari belief systems. That
is, instead of attempting to replace subject
78 G ordon F. M c E wan and Patrick Ryan W illiams
peoples’ existing beliefs with their own, the
Wari may have incorporated such into a larger,
logical narrative that supported the legitimacy
of their conquest and naturalized their rule.
The Inca Empire is known to have used this
kind of incorporative technique, building on
the long-standing, widespread tradition of
reciprocity and mutual obligation—the social
sacrament by which Andean communities
literally lived and died—and using it to struc-
ture conquered communities’ relationships
with the empire. Among the cultures of the
ancient Andes, there was no sharp division
between religion and the state (the head of
state could also be the high priest), and thus
no clear distinction between religious ritual
and administration. Ritual was a technique
employed in the exercise of administrative
power.
Pikillacta, the best-preserved provincial
center, may reveal another way in which the
Wari exercised power. Among the most sacred
objects in the ancient Andes were the corpses
of the dead—the ancestors. Many groups seem
to have mummified their deceased leaders
or to have taken other steps to preserve their
bodies; some even made surrogate images.
These ancestors were responsible for the water
supply, land tenure rights, health, and fertil-
ity; they also provided oracular advice. Since
these deceased individuals had usually been
political leaders, their influence on the people
was understood in terms of political authority.
A priest spoke for each of the dead, who were
consulted on all important matters, and rites
were conducted to “wine, dine, praise, and
reassure the ancestors.”50 The political power
of a society could be severely damaged or even
destroyed by loss of ancestral remains, and
physical control of these remains allowed the
exercise of enormous social power.
Prestige and power could also be ac-
crued through alliances with important and
exalted lineages via marriage, adoption, or
“discovery” of more ancient links to common
ancestors. These kinship bonds engendered a
set of reciprocal obligations and legitimized
power relationships in a way that mere force
of arms could not. By inserting themselves
into the existing web of social relationships
at the highest level, the Wari validated their
 rule as a natural consequence of social obliga-
tion. Furthermore, the rulers would take upon
themselves the responsibility to sustain the
cycle of life by caring for both the living and
the dead.51
Perhaps Pikillacta and some of the other
large Wari provincial centers can be best un-
derstood as administrative devices for the gov-
ernance of the empire, places where the living
and the dead could interact safely. These de-
vices allowed the Wari to demonstrate to sub-
ject populations that their own familiar, local,
and ancient beliefs justified Wari control, from
which benefits must have flowed, whether to
local leaders or more broadly. This is a much
more efficient way to run an empire than by
direct coercion. With kinship-based reciprocal
obligations established, it became ideologi-
cally imperative for the conquered to cooper-
ate for the greater good of all. After inserting
themselves into the local governing lineages,
the Wari would also have direct control of
the ancestors’ remains. Thus there was the
veiled physical threat to the subject popula-
tions’ immediate ancestors and through them
to the continuing legitimacy of the locals’ land
tenure and water rights.
The actual operation of the Pikillacta
complex can be viewed as a transformational
device. The monument was composed of ar-
chitectural spaces well suited for the safe stor-
age of ancestors and as a theatrical setting for
experiencing administrative rituals and feasts.
Figure 51. Three-
dimensional reconstruc-
tion of the central sector
of Pikillacta, ca. AD 700.
Drawing: Gordon F.
McEwan.
79 T he Wari Built E nvironment: L andscape and A rchitecture of E mpire
Individuals went there to receive their instruc-
tions as cogs in the vast imperial mechanism.
They received their marching orders in a way
that had a profound psychological impact.
The experience included not only instructions
about what was to be done from the practical
point of view of imperial administration but
also a powerful psychological message of the
majesty of the empire and how the individual
and his or her corporate group fit into the
greater whole.
The psychological impact was carefully
calculated. With its white plaster coating,
Pikillacta was an enormous, shining beacon
on the landscape. As visitors approached and
entered, they passed through a long, narrow
avenue with high stone walls on either side.
Because it was impossible to see out of the
confines of the walled avenue, visitors im-
mediately became disoriented. Escorted by a
guide, they wound their way through a series
of maze-like passages that took them into the
heart of the monument. As they advanced, the
walls rose higher around them and only the
sky above and the path ahead were visible. The
intended effect was sensory deprivation and
the beginning of a transformation of mental
state. Visitors found themselves helpless in the
hands of their guide. After a long traverse of
narrow corridors, the passageways suddenly
delivered the visitor to the open court of a
patio group (fig. 51). Here, too, the walls and
floors were coated with white plaster, dazzling
Figure 52 [124]. Tunic
with stepped-cross and
interlocked U-shaped
motifs; camelid fiber and
cotton; 96.5 x 110.5 cm. Los
Angeles County Museum
of Art, M70.3.1. Digital
image: © 2012 Museum
Associates/LACMA.
Licensed by Art Resource,
NY.
80 G ordon F. M c E wan and Patrick Ryan W illiams
in the bright Andean sun. It was impossible to
see outside of the patio and all attention was
therefore focused within. On the white stage
provided by the plastered architecture, the
visitor confronted the other participants in
the event as well as the officiating Wari lords
dressed in brilliantly colored tapestry-woven
tunics and headgear (figs. 52, 53). Ancestral
mummies and images brought from the nearby
niched halls also participated in and oversaw
the festivities.
The ceremonies took place, gifts and re-
wards were bestowed, cosmological order was
reinforced, and instructions and reports were
given, followed by feasting and heavy drink-
ing. An altered state of consciousness was
achieved. The impression of power, majesty,
and mystery must have been overwhelming.
Later, visitors were led back out of the complex
by a guide and returned home profoundly im-
pressed with their place within the new order
and the power and majesty of the Wari lords.
Figure 53 [145]. Four-
cornered hat with geo­
metric motifs; camelid
fiber and cotton; 13 x 18
cm. The Metropolitan
Museum of Art, New York,
Gift of George D. Pratt,
1933, 33.149.101. Image: ©
The Metropolitan Museum
of Art. Image source: Art
Resource, NY.

notes 1. W. Isbell 2008, 750.
2. Bragayrac Dávila 1991.
3. Ochatoma Paravicino and Cabrera
Romero 2002.
4. Cook 2001a.
5. Benavides Calle 1991.
6. W. Isbell et al. 1991.
7. W. Isbell and McEwan 1991.
For the sake of consistency in this
volume, the editor has adopted
the dates AD 1–600 for the Early
Intermediate Period and 600–1000
for the Middle Horizon. McEwan
disagrees for reasons expressed in
his recent article (McEwan 2012). He
prefers the date ranges assigned by
John Rowe (1966).
8. Anders 1991; W. Isbell 1977b;
McEwan 2005c; Schreiber 1978.
9. McEwan 1991; McEwan 2005c.
10. J. Topic and T. Topic 1983.
11. McEwan 1991; McEwan and
Couture 2005, 21–27; J. Topic 1991,
149.
12. McEwan and Couture 2005, 25.
13. Glowacki 1996.
14. Cook and Glowacki 2003;
Glowacki 1996; Glowacki 2005b,
109–11; W. Isbell 1988; W. Isbell et
al. 1991; McEwan 2005c.
15. Schreiber 1992, 257.
16. W. Isbell et al. 1991.
17. McEwan 2005c.
18. The contents and circumstances
of this find were initially reported
in Valcárcel 1933 and again in
Trimborn and Vega 1935.
19. Cook 1992.
20. Ibid., 358.
21. Ibid., 360.
22. Julien 2000, 257.
23. Betanzos [1551] 1987, 50–53.
24. J. Topic and T. Topic 1983, 16.
25. Grieder et al. 1988.
26. Pozorski and Pozorski 1986.
27. J. Topic 1986.
28. Personal communication 2011.
29. Anders 1991, 170; McEwan
2005b, 158; McEwan 1991.
30. Harth-Terre 1959; Lanning 1967;
Sanders 1973.
31. Anders 1991; Glowacki 1996;
McEwan 2005c.
32. For tunics’ similar emphasis on
the number four, see pp. 159–191,
“Tapestry-woven Tunics,” in this
volume.
33. Cook 2001a; Schreiber 2005b.
34. Williams 2001.
35. Williams and Isla Cuadarado
2002.
36. Ochatoma Paravicino and
Cabrera Romero 2002.
37. McEwan 2005c.
38. Zegarra 2005.
39. Williams 2006.
40. Williams et al. 2005.
41. Williams 2006.
42. J. Topic 1991.
43. Schreiber 1992.
44. Shimada et al. 1991.
45. Williams 2003.
46. Glowacki and Malpass 2003;
Schreiber 2005b; Williams and Nash
2006.
47. Glowacki and Malpass 2003;
Williams and Nash 2006.
48. Williams and Nash 2006.
49. Schreiber 2005b.
50. Salomon 1995, 323.
51. McEwan 2005b, 149.

81 T he Wari Built E nvironment: L andscape and A rchitecture of E mpire

Donna Nash
Figure 54 [46]. The figure
painted on this vessel
wears a tunic with face-fret
motif. Faceneck vessel
with tapestry-woven tunic;
ceramic and slip; 15.6 x
10 cm. Museo Nacional de
Arqueología, Antropología
e Historia, Lima, C 54760.
Photo: Daniel Antonio
Giannoni Succar.
82
The Art of Feasting: Building an
Empire with Food and Drink
The earliest empire of the central Andes, Wari
(AD 600–1000) extended its influence over
hundreds of miles of the western watershed
of South America. In several regions, the Wari
built monumental administrative centers
in subject territories using the architectural
canons of the empire’s capital, located in Aya-
cucho (see pp. 65–81, “The Wari Built Envi-
ronment”). Since Wari had faded away before
European contact and no form of indigenous
Andean writing has been deciphered,1 under-
standing how the empire managed its spec-
tacular achievements is a challenge. Without
historical records such as we have for the Ro-
mans, Han China, or the Inca, scholars must
piece together evidence based on art, artifacts,
site features, and architecture. More durable
than textiles, metal, shell, or bone, pottery
vessels are the predominant kind of object on
which complex iconography occurs. These
types of vessels are not widely distributed and
most appear to have been used to serve or con-
sume food and drink. Decorated ceramic con-
tainers are often found smashed in dense pit
deposits, filling small rooms, or strewn across
patio floors.2 Although they seem to have been
shattered as offerings, analysis shows that
many were used multiple times at festive gath-
erings before their ritual destruction.
Wari vessels used to serve and drink chi-
cha (native corn beer) exhibit elaborate deco-
ration and have been found in palaces, plazas,
temples, and tombs. Evidence for the produc-
tion of fine pottery and the brewing of beer
exists in or adjacent to elite dwellings. There
are also indications that chicha was stored
and perhaps consumed in Wari temples. Taken
together, this evidence suggests that the ritual
consumption of beer and special foods us-
ing decorated pottery was an important and
frequent activity of Wari elites, including the
leaders of the empire. These feasting activi-
ties brought powerful people together to share
drink, food, and ritual in several settings;
thus, feasting may have played a role in more
than one of Wari’s fundamental institutions,
serving as the crucial social glue that bound
the empire together.
Understanding Feasting
Archaeologists and anthropologists refer
to parties or celebrations as feasts. Modern
societies mark many kinds of occasions with
feasts: personal affairs such as birthdays,
weddings, and funerals; events of national
importance, including Thanksgiving or In-
dependence Day; religious festivals, among
them Christmas, Hanukkah, and Eid al-Fitr.
All have specific customs. Such rich diversity
surely existed in the past. Despite this vari-
ety, all festive gatherings share a significant
feature: they bring people together and provide
settings to make friends, meet future spouses,
tell stories, discuss business, talk politics,
and have a wide variety of other social inter-
actions. This facet of feasts makes them an
important aspect of ancient societies, perhaps
dating back to Paleolithic times.3 Feasts can be
self-perpetuating mechanisms for maintaining
the relationships they establish; in many so-
cieties they create networks of mutual obliga-
tions between hosts and guests with far-reach-
ing implications.4 In other words, feasting can
be a significant institution that binds people
together, defines their relationships, and drives
economic production.
Not all feasts have the same significance.
The differences are clear in the materials that
each leaves behind for the archaeologist. For
instance, garbage from a White House state
 dinner might include broken pieces of crystal
wine glasses, porcelain plates marked with
the presidential seal, bottles of wine, lost jew-
elry, and the remains of high-priced meats or
small game birds. Unfortunately, plant waste
from the meal would disappear. The backyard
barbeque creates different traces: aluminum
cans, plastic plates, cups, and bottles, per-
haps the bones of low-cost cuts of meat. Using
many lines of evidence archaeologists can
often recognize these differences and even
distinguish between events related to personal
affairs versus those that were sponsored by
large institutions such as the Wari Empire. To
identify feasts, of course, a basic understand-
ing of daily meals is needed5 since by defini-
tion feasts go beyond the typical daily meal in
some way: the presence of more guests, lavish
foods, ritual activities, or special displays.6
Feasts are ideal vehicles to build prestige
and garner recognition. Feasts also can be
significant social mechanisms for asserting
power or effecting transformative change.
Thus, anthropologists have defined the roles
that feasts play in establishing distinctions
between people (status, rank), accomplishing
the goals of hosts, and indoctrinating guests
into institutions (religious or secular).
Feasting as a Social and Economic Institution
Festive gatherings bring people together and
provide opportunities for individuals, fami-
lies, factions, communities, organized reli-
gious cults, and even state-level societies to
further their social and political agendas. Eth-
nographies from many areas7 demonstrate that
some groups strategically plan feasts and their
hoped-for outcomes while others engage in the
“sincere fiction of disinterested exchange.”8 In
other words, some recognize feasting as a po-
litical and economic vehicle and overtly seek
to manipulate it, with all participants being
aware of their obligations and responsibilities.
But other societies, our own included, stress
attributes such as generosity and hospitality as
the primary motivation. (Gifting is similarly
regarded, with the attached obligations for a
return attributed to good manners.) Regardless
of the sentiment, feasting can be used to build
significant social, political, and economic rela-
tionships in all societies.
84 D onna N ash
Feasts can be venues to foment solidarity,
form or strengthen alliances, attract new fol-
lowers, show off (aggrandize or demonstrate
success), reaffirm the current hierarchy, or
challenge the pecking order. Feasts can also
garner loyalty and be viewed as favors and ac-
companied by gifts, which oblige repayment.
None of these social acts are mutually exclu-
sive and all can occur simultaneously.9
Feasts come in all sizes. Some are inti-
mate, others are large. In fact, state-sponsored
festivals (Independence Day, for example) can
be opportune times to promote state agendas,
demonstrate state success through pageantry,
and reinforce state ideologies through ceremo-
nies or performances. From a political perspec-
tive feasts are opportunities for people to assert
power over others by creating obligations,
winning the admiration of followers, or out-
competing rivals. Michael Dietler, who has
studied feasting in ancient Europe and modern
Africa, describes three types: empowering
feasts, patron-role feasts, and diacritical feasts.10
The broadest category is the empowering
feast because the relationship between partici-
pants is not fixed—hosts may become guests on
the next occasion. Hosts of empowering feasts
earn prestige and/or wealth. While both can be
fleeting or marginal, in societies where prestige
from hosting a feast confers decision-making
power, hosts become competitive. In contrast,
the participants of patron-role feasts have rela-
tively fixed relationships. Leaders always play
hosts and followers are always guests. Medi-
eval European courts offer good examples; the
royal couple always hosted their courtiers and
this relationship came with well-defined privi-
leges and responsibilities. Patron-role feasts are
overt statements of the patron’s power over fol-
lowers. The patron is generous and clients are
loyal although revolts do occur. Empowering
or patron-role feasts celebrated by an exclusive
group to distinguish itself by using distinctive
features—special foods, elaborate serving ves-
sels, expensive wardrobe, or fancy locales—are
termed diacritical feasts. Special credentials,
manners, or tastes are required to obtain an
invitation to such events, and the group may
consist either of people of the same high rank
competing for power or of a patron and high-
status followers.
Figure 55 [120]. Tunic with
face-fret motif; camelid
fiber and cotton; 110 x
110.5 cm. Ethnologisches
Museum, Berlin, VA
64374. Image: bpk, Berlin/
Ethnologisches Museum/
Art Resource, NY. Photo:
Dietrich Graf.
Archaeological remains do not always re-
veal the subtle relationships between partici-
pants, but it is likely that most of the materials
related to feasting in this catalogue were used
during diacritical feasts. These objects were
part of the symbol system that signaled the ex-
clusive status of elite participants. In Wari so-
ciety people wore their status and identity on
their bodies in the form of fine textiles and or-
naments, which were part of personal displays
(see pp. 159–91, “Tapestry-woven Tunics,”
and 217–31, “Inlaid and Metal Ornaments”).
The most elaborate imagery on pottery occurs
on figure effigies shown wearing tunics that
match examples found archaeologically (figs.
54, 55). Intricately decorated vessels were
85 T he A rt of F easting : Building an E mpire with F ood and Drink
special; archaeologically, they are restricted
to elite dwellings and monumental precincts,
where they are found with other rare, expen-
sive items. Even the buildings where feasting
took place were built by skilled masons and
finished with decorative plaster. The invest-
ments people made in their clothing, other or-
naments, special foods, fermented beverages,
elaborate dining halls, as well as the expense
of supporting staff and artisans who produced
these items must have been worthwhile or the
Wari elite would not have dedicated so many
resources to feasting.
In our prepackaged, catered world a host
can decide at noon to throw a dinner party
that evening. In traditional societies, however,
 hosts must accumulate resources over time,
sometimes years. The same is true for labor.
For a small intimate affair the members of the
household may manage, but large events re-
quire help from other relatives, close friends,
followers, or hired staff. Sponsors may call in
many personal favors to muster the cooking
labor and ingredients, or they may require fol-
lowers to make contributions. Planning is es-
sential, especially if alcohol is served because
traditional forms sour in just a few days.
In some societies hosts pull together a
few feasts over the course of their lives; other
leaders are expected to arrange feasts on a
regular basis.11 In these cases leaders usually
have permanent preparation facilities, insti-
tutionalize the contributions expected from
followers, and may resolve labor demands by
having multiple wives,12 servants, or both.
Feasting may seem frivolous, but many ex-
amples demonstrate that sponsoring a feast
can create power and generate wealth. Ethno-
graphic cases are so common and archaeologi-
cal evidence is so prevalent that some scholars
suggest that farming developed as a way of life
from the desire to create the surpluses needed
to reap the rewards of hosting a feast.13 which funded the ongoing feasting competi-
Feasts can pay in ways that range from
modest to extravagant. A few case studies
demonstrate the gains to hosts. On the modest
end of the scale are modern Andean labor ex-
changes between farming families, which are
required for annual planting and harvesting.14
These events obligate the hosts to contrib-
ute labor to those who show up to work, and
the meal serves as a thank you to engender
continued good relations between the partici-
pants, who are of a relatively equal status.15
In the Andes people openly recognize that
there are no free favors or meals. Obligations
are purposely created and recreated. These
investments pay off because they establish
a support network vital to the survival of all
participants.
Community resources are often shared but
the shares are not always equal. Among the
Akha of Southeast Asia, a council of elders
made all important village decisions and
distributed land to community members.16 The
elders formed and maintained factions, the
most powerful of which dominated the coun-
8 6 D onna N ash
cil, through feasts. To join a powerful group
an elder had to sponsor feasts and be a gener-
ous host to other faction members. Such feasts
signaled group success and built relationships
among prominent lineages. In this case, host-
ing feasts did not deliver solitary power but
created alliances and gave elders a voice to
advocate for their lineage.
In the Philippines, chiefly leaders and
up-and-coming rivals competed for followers
by throwing lavish feasts that created strong
ties among elites and indebted followers who
could be called to contribute labor or re-
sources for future agricultural work, warfare,
or yet another competitive feast.17 Participants
recognized the debts accrued by the host,
who required help from their network of elite
friends, kin, and followers to gather the feast’s
resources. Elders purposely sponsored young,
rising stars, wishing to obligate the newcom-
ers; these young stars also profited since they
needed the elders’ resources to compete. Thus,
feasting created a complex web of debts and
obligations by building alliances among com-
munities; when mobilized, these alliances
could amass large quantities of food and gifts,
tion.18 More important, the leader who at-
tracted the largest group of followers held the
advantage in military raids and could work
large tracts of lands to produce food surpluses,
both important sources of wealth before Span-
ish contact. Complex webs of debt relation are
common and notation systems such as tokens
or notched sticks are sometimes used to keep
track of who owes what to whom.19
Among several African groups, kings and
local leaders are expected to have special food
and drink on hand to entertain notable guests
and feed followers. In exchange, community
members owe labor (working the leader’s
lands), some form of tribute (raw or prepared
foods), or both.20 Leaders need contributions
to maintain their stores. Among the Méta of
Cameroon, one village leader fed people when
they worked his fields, but he was also expect-
ed to sponsor a community feast once a year to
maintain the reputation as a generous chief.21
Exchanging food for labor is not the ex-
clusive prerogative of leaders. Work feasts can
draw together large work parties, sometimes
 beyond one’s social network. Such events
convert perishable foodstuffs to more durable
goods such as religious buildings, wells, or
canals. These investments may increase the
economic distance between host and worker-
guests. Among the precolonial Samia of
Kenya, for example, a wealthy individual
staged a feast that put guests to work min-
ing iron-rich hematite.22 The host then paid
metalworkers to make iron hoes, a valuable
commodity in the region. In this case no long-
term obligation was established. The host and
the guests part ways having made an equal ex-
change. Unlike the Méta case, no annual party
created and renewed long-term obligations.
The guests owed no tribute and there was no
obligation on the host’s part to represent the
group. The Méta example is a good model for
understanding how leaders orchestrated labor
projects,23 whereas the Akha and Philippine
chiefdoms shed light on how alliances and
networks of cooperative elites form through
feasting. A brief review of Inca feasting will
show that both were important in the Andes.
Feasting among the Inca
The Wari Empire was smaller than the later
Inca Empire (1350–1532), which was the
largest pre-Hispanic state in the New World,
dominating nearly 2,500 miles of the Andes
and incorporating parts of Argentina, Bolivia,
Chile, Columbia, Ecuador, and Peru.24 Given
the difference in time, there are undoubtedly
important distinctions between the two, but
artifacts exhibit similarities that suggest some
continuities with earlier Wari practices.
Figure 56. Pair of matched
Inca wood keros (cups)
with incised geometric
motifs. Museo Inka, Cuzco.
Following a common pat-
tern, one member of the
pair is smaller than the
other; this may signify that
the relationship between
the two toasting parties
was not equal. Photo:
Daniel Antonio Giannoni
Succar.
87 T he A rt of F easting : Building an E mpire with F ood and Drink
Many Spanish colonial documents focus
on the activities of the Inca central govern-
ment in Cuzco, the capital. They record elite
Inca festivals in the annual religious cycle
as well as special celebrations such as the
ascensions or funerals of emperors, although
the Spaniards did not directly observe many
of these events. Exaggerations are likely and
interpretations were skewed by the European
observers’ feudal sensibilities. Historians
discuss Inca feasts in terms of reciprocity,
either between an Inca overlord and subordi-
nate nobility, or between local lords and their
followers.
In his analysis of colonial sources, Thomas
Cummins suggests that the Inca made a fun-
damental distinction between these two types
of reciprocal relations.25 The first took the form
of patron-role relations: the Inca feasted the
subjugated lords and gave them token gifts
in return for pledges of labor and resources.
This contrasts with the empowering relations
that local lords had with their followers in
which the former feasted the latter with food
and drink to obligate them to labor projects.
Cummins suggests that the Inca were above
reproach and became so powerful that they
could flout their followers’ expectations,
whereas local leaders always had to offer food
and drink to access the labor of their follow-
ers. This distinction likely represents the
difference between high and lesser nobles.
Nevertheless, both types of feasting created
unbalanced relationships that were essential
to maintaining the empire. That is, the hosts
(the Inca or the local lords) always received
far more than their guests. Feasting paid more
than it cost and kept followers indebted to
their patrons.
Scholars suggest that the Inca were build-
ing on long-standing Andean traditions. They
used feasting and gifting to obligate foreign
leaders, subjugated through conquest or
political maneuvering. Special cups (keros)
reportedly were used, often in matched pairs,
to make chicha toasts that sealed agreements
or alliances (fig. 56). For example, Pedro de
Cieza de León, a colonial Spaniard, states that
Cari, a leader of the Colla people of the Lake
Titicaca region, and Viracocha Inca, an Inca
emperor, celebrated their alliance by drink-
 ing from a goblet that was taken to a “temple
where such pledges between the Inca and
lords were kept.”26
Inca royal lineages (panacas) apparently
maintained solidarity with toasts of chicha
made using matched pairs of keros. Upon
arriving at the house of another panaca, they
offered chicha in one cup while drinking from
a matching cup, and they were met with the
same from their high-status host.27 Paired cups
were used in a symbolic way during rituals;28
these toasts mimicked the pledges between
people but called upon supernatural beings to
fulfill the needs of human supplicants. There
are at least three such examples: the emperor
toasted the sun, his deified ancestor, who
legitimized imperial Inca control (fig. 57);29
warriors charged with cleansing Cuzco dur-
ing an annual rite (the Situa ceremony) drank
from one cup and threw a second cup into a
river, perhaps as an offering to the river that
carried “sin” out of the city;30 mourners lined
up to honor a new ancestor with toasts during
funerals (fig. 58).31 Thus, toasts, an aspect of
feasting, were a ritualized way to create obli-
gations, and cups symbolized these relation-
ships in the Inca Empire.
The Inca sponsored the distribution of
food surpluses through regional administra-
tors to the general tax-paying public, which
obligated them to perform labor (the medium
for most tax payments). Perishable foods were
transformed into monumental architecture,
roads, prestige goods made by skilled art-
ists, and agricultural infrastructure, includ-
ing fields, terraces, and canals. This process
may have involved enormous work feasts, but
archaeological evidence for them has not been
found and some chroniclers say that the Inca
handed out raw foods during labor projects.32
Nevertheless the Inca economy was based on
this kind of redistribution, which has been
described as the “Inca mode of production.”33
Feasts accompanied many imperial events
and celebrations. Some religious festivals were
restricted to Inca of royal blood.34 Featuring
the finest foods and service wares, they were
held in special locations such as temples or
palaces.35 During other annual events, high-
ranking foreign nobles joined the Inca in
feasts but likely were served with vessels of
lower quality than their royal hosts. The state
provided the lion’s share of the resources, but
reportedly subordinates did make contribu-

Figure 57. The Inca emperor

toasting the sun, his ances-
tor. An elite woman (per-
haps the Inca queen) pours
chicha into two additional
cups, which she may use to
toast the sun in turn. After
Guaman Poma [1615] 2009,
190, figs. 246, 248.
Figure 58. The Inca emperor
toasting the mummy of
another Inca emperor. The
urn serves as a receptacle
for the chicha “consumed”
by the dead. Two Inca
queens participate in the
symbolic toasting ritual.
After Guaman Poma [1615]
2009, 225, figs. 287, 289.

88 D onna N ash

Figure 59 [33]. Cup-holding
figure in tie-dyed tunic
and four-cornered hat;
ceramic and slip; 19.3 x
19.5 cm. Museo Regional
de Ica “Adolfo Bermúdez
Jenkins,” MRI-00176-01.
Photo: Daniel Antonio
Giannoni Succar.
89 T he A rt of F easting : Building an E mpire with F ood and Drink
tions to fund feasting.36 Festive celebrations
must have been considered crucial to the in-
tertwined domains of religion and governance
as they were carried out both in the imperial
heartland and in all the empire’s provinces.37
The demands of the feasting economy
impacted every aspect of daily life: lands
were taken from commoners; labor taxes were
demanded from the general populace; arti-
sans were required to make items used during
feasting for the empire. In fact, some people’s
lives were seemingly devoted to the feasting
economy. When the Inca took control of a new
region they subdivided community lands,
allocating portions to the state and the sun
cult, and the remainder to commoners. The
emperor also had the prerogative to set aside
land as an imperial estate, which eventually
supported his lineage and the maintenance of
his mummy.38 The produce from state lands
went to state-sponsored labor projects, state of-
ficials, and any related feasting activity. Like-
wise, the lands devoted to the sun cult funded
feasting during religious festivals. Therefore,
in addition to losing lands, commoners also
worked harder because they were required
to contribute a turn of labor (m’ita), although
they were fed while performing this service.
Both state and temple celebrations used
the same type of specialized labor (mama-
cona), comprising females trained as young
girls (aqlla) to fulfill the duties of elite women.
Cuzco and provincial capitals had facilities
(aqllawasi) where this training occurred.
Some aqlla received instruction to serve in
temples, to sing the histories of Inca emperors,
or to play musical instruments; most learned
to weave fine textiles, to prepare food, and
to brew chichi; mamacona provided lifelong
service to the state with these skills.39 A few
aqllawasi have been identified and provide
reference points for distinguishing elite, state-
sponsored diacritical feasting.40 The variation
and irregular distribution of Inca artifacts
used during diacritical feasts41 exemplifies the
patchiness of prestige goods in a complex em-
pire. Wari artifacts also exhibit great variation
and uneven distribution.
Feasting among the Wari
A great deal of what we know about the Wari
can be linked to feasting or the production of
prestige goods that were either gifted or dis-
played during festive gatherings. Diacritical
feasting was practiced at most Wari sites, and
Wari feasting wares are widely distributed.
Many of these items are decorated in similar
ways and likely communicated messages that
tied feasting to ritual and the supernatural
world. Understanding feasting requires mul-
tiple lines of evidence.
the iconography of feasting. It is fair to
say that many fine Wari art works were on
display at feasts as either fine decorated table
and service wares42 or elaborate costumes; in
some, Wari elites may represent themselves as
hosts (fig. 59). Great effort went into making
feasting wares, perhaps because they were a
crucial aspect of display. The abstract icons
that appear on tableware—cups and bowls—
may have functioned in part as heraldry,
identifying family, occupation, rank, title, or
Figure 60 [26]. Cup with patron. Serving urns and faceneck jars carried
staff deity; ceramic and
elaborate state iconography43 surely designed
slip; 12.4 x 7.9 x 8.2 cm.
Ethnologisches Museum, to communicate particular messages to the
Berlin, VA 19167. Image: guests, perhaps including the empire’s hierar-
bpk, Berlin/Ethnologisches chical organization. For example, the impor-
Museum/Art Resource, NY.
Photo: Martin Franken. tant front-facing staff deity is shown on very
large Wari faceneck feasting jars and urns
Figure 61 [44]. Faceneck
vessel with mutilated from Conchopata (see figs. 75a–f, 102)—as it
nose; ceramic and slip; is on Tiwanaku’s Gateway of the Sun (see fig.
18.2 x 13.5 x 14.8 cm. 6a)—at the center of a composition completed
Ethnologisches Museum,
Berlin, VA 49450. Image: by a multitude of profile attendants, either
bpk, Berlin/Ethnologisches axe-wielding sacrificers or winged figures who
Museum/Art Resource, NY. sometimes take on animal features and always
Photo: Claudia Obrocki.
carry a staff (see pp. 103–121, “The Coming
of the Staff Deity”).44 In these scenes the
frontal deity is visually more important than
its companions in profile, and perhaps domi-
nant over them.45 In feasting contexts these
kinds of representations may have conveyed
the ideal relationship between Wari patrons
90 D onna N ash
and their followers, reminding participants of
that relationship and of bonds of reciprocity.
Staff deities also appear alone on jars, cups, or
bowls (fig. 60); at times only the head is pres-
ent, which may be a shorthand convention for
the whole or could represent a different type
of being.46 Such vessels may have reflected
the relative position of the user or the roles of
participants during toasting rituals.
It is also possible that the vessels them-
selves were at times regarded as participants.
Feasts were places where people interacted,
and faceneck jars, which represent people,
may have portrayed an absent sponsoring
leader or a remembered ancestor. Some face­
neck jars exhibit life-like facial features
whereas others are abstract and may have
been identified by the combination of icons
shown on their garments. If feasting vessels
were regarded as honored participants, the
Figure 62 [6]. Urn with wealth of the feast was literally dispensed
heads of mythical crea-
from their bodies.
tures, from Conchopata;
ceramic and slip; 34 x A number of other complex scenes and
64 cm. Museo Histórico themes are illustrated on urns, and these seem
Regional “Hipólito to portray mythical creatures or episodes, or
Unanue,” Ayacucho,
MHRA-834. Photo: Daniel legendary actions that perhaps conveyed the
Antonio Giannoni Succar. ideals of elite Wari society, including military
victories that legitimized royal power (see
fig. 103). Some vessels depict vegetation that
may have referenced a relationship between
feasting rituals and nature’s fertility, a natural
association. The variation in iconography sug-
gests that feasting may have played a role in
many different types of personal and seasonal
celebrations.
feasting ware . Wari feasting wares were pro-
duced in a range of qualities, suggesting status
differences among their owners, although
variations are often subtle and all were made
with great effort and skill. The most elaborate
probably pertain to high-ranking officials of
91 T he A rt of F easting : Building an E mpire with F ood and Drink
the central government (fig. 54) while less em-
bellished objects were likely used by the local
leaders of small groups (fig. 61). Evidence for
Wari feast activity comes from conspicuous
concentrations of decorated vessels,47 which
take a number of forms.48 Urns (fig. 62), large
faceneck jars (see, for example, fig. 198), and
perhaps specially modeled vessels (see figs.
133, 137) were likely used as serving decanters
for freshly fermented chicha, although some
may have held foods such as rich stews. These
dishes were not used behind the scenes but in
the feasting area, where their elaborate deco-
ration could be seen by guests. Tableware49
included cups (figs. 60, 63) and bowls (see fig.
108); such vessels are found in a variety of
contexts, including tombs, but, again, clusters
found together provide evidence of feasting. It
is possible that double-spouted vessels (fig. 64)
and modeled vessels with narrow spouts (figs.
65, 66), which come in a number of forms, also
played a role during some feasts, but these
Figure 63 [60]. Foot vessel;
ceramic and slip; 11.7 x
12.1 cm. Museo Larco,
Lima, ML018890. Photo:
Daniel Antonio Giannoni
Succar.

Figure 64 [51]. Double-

spouted feline-head vessel;
ceramic and slip; 18 x 16
x 13 cm. Niedersächsiches
Landesmuseum, Hannover,
I/10456. Image: ©
Landesmuseum Hannover.

92 D onna N ash
Figure 65 [65]. Vessel with
head of mythical creature;
ceramic and slip; 24.1 x 16.5
cm. Denver Art Museum
Collection, Gift of Olive
Bigelow by exchange,
1996.36. Photo: © Denver
Art Museum 2012. All rights
reserved.
Figure 66 [62]. Vessel
with bird-headed creature
(“Pachacamac griffin”), from
Pachacamac; ceramic and
slip; 18.3 x 17.3 x 15 cm.
University of Pennsylvania
Museum of Archaeology and
Anthropology, Philadelphia,
26709. Image: courtesy the
Penn Museum.
93 T he A rt of F easting : Building an E mpire with F ood and Drink
ceramics would be very difficult to clean and
thus may have been intended for a single use
or to contain a special offering. Larger-than-
normal cooking pots and specialized brewing
wares were used to prepare feasts, but usually
they were not decorated.
At any given event several serving vessels
may have been used or one may have been
refilled several times. Nevertheless, serving
vessel sizes suggest feasting took place at a
number of different scales. At the upper end
are the reconstructed urns from Pacheco,
which could hold more than 200 liters (53 gal-
lons) of beer (see figs. 1, 130). One faceneck jar
has a 130 liter (34 gal.) capacity (see fig. 198).
Smaller urns and faceneck jars in the exhibi-
tion would hold 80 liters (21 gal.) and 20 liters
(5 gal.), respectively (see figs. 134–36). At the
smallest scale is a modest faceneck jar (fig. 67),
which holds a mere 1.5 liters (6 cups), hardly
enough for a feast; it may represent a single
serving. If that is the case, the large Pacheco
urns held chicha for around 135 people and
the large faceneck jar, enough for 85.
Tableware also comes in a wide variety
of sizes, could be refilled, and may have been
passed from person to person, which is typical
of Andean drinking today. Like serving ves-
sels, much tableware is high quality in manu-
facture and decoration. Cups are less common
than bowls archaeologically, perhaps because
eating and drinking are separate phases of
Figure 67 [41]. Faceneck feasting in the Andes—food comes before
vessel with birds; ceramic
drink—and participants could use bowls for
and slip; 18.4 x 14.5 x
14.3 cm. Ethnologisches both.50 High-ranking individuals may have
Museum, Berlin, VA used different vessels at each stage. Since cups
49536. Image: bpk, Berlin/ are not present at all feasting locales, they
Ethnologisches Museum/
Art Resource, NY. may have been used for specific rituals rather
than all types of feasts. Like Inca keros, some
Wari vessels were made in pairs (see fig. 120)
or matched sets of four cups (fig. 68);51 it is
possible that sets held special significance and
were required for some events.
feasting spaces. Feasting took place primar-
ily in the homes of leaders, but some special-
ized areas adjacent to leaders’ homes have
also been identified.52 The latter appear to be
within elite precincts at Wari sites. Feasting
spaces, which vary in size and lavishness,
usually contain the remains of smashed serv-
ing wares, smashed tableware, or both and
94 D onna N ash
often are accompanied by elite ornaments or
other prestige goods.
Feasts were held in house patios or pla-
zas,53 which were designed with feasting in
mind.54 At the Wari capital, the partially ex-
cavated Moraduchayoq compound is a walled
complex containing several related elite fami-
lies, each of which had its own patio group
residence (fig. 69). The excavators concluded
that Moraduchayoq was a residence and work-
space for a group of mid-level administrators
who held feasts as part of their managerial
responsibilities.55 Remains of feasting wares
and the detritus of everyday life were found
in the patios of the dwelling units. The four
larger patios are about 100 sq. m (1,075 sq. ft.)
and may have been used to entertain some
fifty guests comfortably.56
More elaborate feasting wares have been
found at Conchopata, a site close to the capital,
Figure 68. A set of four
matched cups found
ritually smashed in the
chicha brewery at Cerro
Baúl. Photos: Patrick Ryan
Williams.
where feasting plazas are larger (fig. 70). One,
Plaza B, measures roughly 225 sq. m (2,400 sq.
ft.) and the other, Plaza E, is 294 sq. m (3,164
sq. ft.). These feasting spaces each had a large
urn embedded in the floor, presumably for
serving chicha. Taking into account entrances
and placement of the urns, a host could enter-
tain around seventy-eight guests in Plaza B or
eighty-nine in Plaza E. The urn found in Plaza
E (fig. 62) has a capacity of 80 liters (21 gal.);
we can infer that each guest drank approxi-
mately 0.92 liter (4 cups) of chicha or that the
urn was refilled one or more times during the
celebration.57 The latter seems the most likely.
Large jars with chicha residue were
recovered from one D-shaped temple at Con-
chopata,58 and fifteen similar jars, each of 95
Figure 69. Plan of the
Moraduchayoq compound
at the Wari capital. Based
on W. Isbell et al. 1991, 37,
fig. 19.
platform
N 20 meters
95 T he A rt of F easting : Building an E mpire with F ood and Drink
liter (25 gal.) capacity, may have been stored in
another such temple.59 It is unclear if drink-
ing took place inside; the temple may have
stored sacred chicha, or it may have been an
offering. D-shaped temples at most sites are
fairly small; thus, if feasting played a role in
the temple institution the participants were an
exclusive few.
Evidence of feasting at Wari provincial
sites is very similar to that from the Mora-
duchayoq compound: high concentrations
of feasting wares occur in the patio group
residences of elite leaders.60 In most instances,
smashing the feasting wares appears to have
been part of the festive ritual, but evidence of
the meal’s preparation or the brewing of chi-
cha is not commonly found. The rare excep-
tions are described below.
feasting fare . Chicha, an alcoholic corn
beer, was the most important element of a
Wari feast. The dregs—dense deposits of molle
seeds or pits—have been found at several Wari
sites.61 Schinus molle, the Peruvian pepper
tree, yields bright magenta berries (drupes)
with a hard central pit; the berries taste like
pepper and have pockets of sugary resin.62
Today in the Moquegua region of far southern
Peru, molle-flavored chicha is made by briefly
steeping the pepper seeds in hot water and
adding the seed-free liquid to a boiling corn
mash, which is later strained, cooled, and fer-
mented. Evidence from Cerro Baúl, described
below, suggests that the Wari used a very simi-
lar recipe—corn and molle—for their chicha.63
The Wari typically burned their garbage
to a fine ash, but molle has been found near
brewing areas and in a few cases feasting
remains were left in place, perhaps as some
form of ritual. At several sites camelid bones
 N 10 meters

D-Shaped Temples
Firing Ovens
Ceramic Ceramic
Offerings Concentrations

Plaza E

Pink
Plaza
Plaza B

White
Court
Y
2 04
205

GHWA
RN HI

walled street?
MODE

Slope Toward
Canyon

Figure 70. Plan of Concho­
pata. After W. Isbell 2009,
208, fig. 10.8.
accompany smashed feasting wares, making it
likely that feasters ate roasted llama or alpaca
meat (whole ribs are common). This meat
would have been a luxury to most people; data
from modest houses suggest that commoners
had very little access to meat.64
The remains of one feasting event, found
in a patio of a palace at Cerro Baúl, a Wari
administrative center located in the sierra of
the Moquegua region in far southern Peru (fig.
71),65 appears to correspond with the aban-
donment of the elaborate dwelling. Analysis
revealed that the festive meal consisted of
vizcacha (the Andean hare), deer, camelid
(alpaca or llama), river shrimp, and at least
nine types of fish from the Pacific Ocean.66
Chemical analysis showed that some of the
camelids ate a coastal diet,67 which may have
come from more than 70 km (40 mi.) away; the
ocean fish perhaps were a contribution from
one of the guests. Plants are less durable and
materials such as potatoes and other tubers
rarely leave a trace. Nevertheless, evidence
suggests that coca, quinoa, maca (a turnip-
like vegetable), mauka (a large tuber), beans,
peanuts, and squash featured in Cerro Baúl
feasting cuisine.68
preparing the feast. Several researchers
have noted the absence of evidence for feast
preparation near Wari feasting locales,69 but
no elite compound has been excavated in

9 6 D onna N ash
Figure 71. View of Cerro
Baúl, a Wari administra-
tive center located in
Moquegua, Peru, on Wari’s
southern frontier. The site
was built on top of a natu-
ral mesa; given the mesa’s
towering height and the
lack of a water source on its
summit, daily living was
itself a form of aggrandize-
ment. Photo: Donna Nash.
Figure 72. A tupu pin, used
by women to fasten cloth-
ing, from the chicha brew-
ery at Cerro Baúl. Photo:
Cerro Baúl Project.
97 T he A rt of F easting : Building an E mpire with F ood and Drink
its entirety. It is possible that large kitchens
were located away from living quarters but
still within these compounds.70 The possible
distance between feasting locales and prepara-
tion areas may imply that some women of high
status did not participate in preparing feast
meals, thus freeing them to engage in the feast
with their spouse as an equal cosponsor.71 It
also may imply the existence of brewing and
cooking specialists who had their own kitch-
ens and living quarters.
The best evidence for feast production
comes from the large-scale brewery (chicheria)
at Cerro Baúl, which had an estimated capac-
ity of 1,800 liters (475 gal.).72 Several tupus,
metal pins that elite women used to fasten
their clothing (fig. 72), and spindle whorls
used for spinning thread were found in the
brewery, suggesting that, if this facility was
run by specialists, they were women.
The brewery consists of a patio-group like
those used for elite dwellings: a trapezoidal
patio surrounded by several rooms, one of
them L-shaped (fig. 73). The long, western
room was used to soak corn for sprouting and
had five large slabs of volcanic stone (rhyolite)
to grind the corn once it dried. The northern
portion of the L-shaped room contained eight
or more deep pit hearths, each with a set of
stones to support large brewing jars with
conical bottoms. Molle was found near the
hearths. The rest of the L-shaped room, now
empty, may have been used for storage. The
specific purpose of the southern room remains
unclear. Fermentation occurred in the patio:
along one wall were vessels sunken into the
floor surface.
Smaller-scale chicha production has
been identified at Cerro Mejía, a large village
near Cerro Baúl. On the summit of the hill in
an elite patio-group residence (fig. 74), four
hearths for boiling chicha and three large
hearths for roasting meat were found in a sin-
gle room flanking the patio. Brewing vessels
are much smaller and there are no decorated
serving wares. In contrast to Cerro Baúl, a
major Wari provincial center, Cerro Mejía was
a secondary center with elites of lower rank.
This may explain the differences between the
two feasting facilities: Cerro Baúl had a dedi-
cated facility where a group of female special-
ists worked, while feasts on Cerro Mejía were
prepared by members of the lower-ranking
elite household.
It also appears that elite household mem-
bers prepared chicha at Conchopata. Although
Figure 73. Plan of the
chicha brewery in the elite
precinct of Cerro Baúl.
Illustration: Donna Nash
after Cerro Baúl Project.
Figure 74. Plan of the elite
patio-group residence
(Unit 145) on Cerro Mejía.
Illustration: Donna Nash.
Patio
9 8 D onna N ash
plaza bench
patio
bench
N 1 meter
no feast kitchen has been reported, there are
indications that elite women made beer; they
may have also made the large decorated ves-
sels.73 One burial of a woman in her forties,
placed under the floor in a residential room
(labeled 205 on the plan reproduced in fig.
70), was associated with offerings that seem
to span two small rooms (204 and 205). A
clay cap covered the grave and the area was
surrounded by pit offerings. The woman was
placed in a stone-lined cist in a seated, flexed
position with a decorated bowl inverted on her
head and two faceneck jars positioned around
her. The offerings in the pits and overlying the
floor included camelid remains and bowls per-
haps related to feasting, along with figurines,
Stairs
N 2 meters
pottery-making tools, pigments, and large jars
possibly used for the preparation, storage,
and service of chicha. The vessels and other
artifacts suggest the woman was of moderate
status. Since many women at Conchopata were
buried with men, investigators suggest the
“brewer woman” was a specialist who made
chicha for elite households.74 More research is
needed to understand elite feasting practices
and the personnel who made them possible,
especially because it appears that different
personnel took on this role, depending on the
host’s rank in the imperial hierarchy.
Interpreting Wari Feasts
A substantial portion of the Wari economy
was dedicated to feasting. Large quantities
of corn, molle, and other resources went into
making beer; special imported foods marked
some feasts as high class; herd animals were
dispatched to provide ample meat for many
guests. Much material, skilled labor, and
artistry went into making decorated serving
and drinking wares as well as large brewing
vats and fermentation jars. Also, gifting went
along with feasting, not to mention elaborate
personal displays in dress and ornamentation.
Feasting was costly but, based on the grand
achievements and long-lived success of the
empire, it must have paid off in many ways.
The feasts that have been recognized are
predominantly elite gatherings. Even though
we cannot describe the specific decision-
making powers of Wari officials or list privi-
leges of the elite, research shows that they
had access to more resources and enjoyed a
greater variety of foods in their diets.75 Their
houses are bigger and they apparently could
enlist people to help them build these elabo-
rate dwellings. There were several elite ranks,
scales of wealth, and sets of responsibilities,
but it appears that all Wari officials engaged
in feasting and many were called upon to host
these diacritical affairs.
Maintaining or improving the position of
the family and its status in state operations re-
quired the constant management of resources
both for feasts and gifts. Subordinates needed
to be looked after, equals were entertained,
and resources may have been requested by
superiors for the more elaborate events hosted
 in the royal palaces of the capital. Resources
and gifts that passed between individuals
may have been viewed through an ideology of
generosity but, based on sentiments held by
modern Andeans, such exchanges were likely
carefully tracked and overtly manipulated. It
is possible that khipus (see fig. 180; see also
[155], p. 276) or some other device was used to
record contributions of different kinds and to
monitor the obligations between parties.
That these festive interactions were pri-
marily set in elite dwellings demonstrates the
prominence of the palace in the Wari Empire.
Temples may have been another locale where
9 9 T he A rt of F easting : Building an E mpire with F ood and Drink
feasting or symbolic toasting took place. In
addition, at least at Conchopata, special areas
(Plazas B and E) were designed for feasts and
may have been dedicated to celebrations in
the empire’s annual religious cycle. All these
institutions brought elites together, fostering
cooperation and alliances that produced a
complex web of obligations among those in the
imperial hierarchy. States and empires, like
other types of societies, are made up of the
relationships among people. For the ancient
Wari Empire feasting appears to have been one
significant way crucial relationships were cre-
ated and maintained.
Notes 1. The recording traditions of the
Inca, both the khipu and histories
reportedly documented on painted
textiles (see Cobo [1653] 1979, 99),
cannot be deciphered beyond nota-
tions of numerical data (see Quilter
and Urton 2002; Urton 2003b) or
have not been identified and were
likely destroyed during the colonial
era. A few Wari era khipu have
been recovered from the coast (for
example, Conklin 1982), but how
they were used during Wari times
remains unknown.
2. Cook 2004; Cook 2001a; Cook
1984–85.
3. Conkey 1980.
4. See Bray 2003b; Dietler and
Hayden 2001a; Grignon 2001.
5. Nash 2010.
6. See Wiessner 2001.
7. For instance, the Philippines
(Claver 1985), the northwest coast of
North America (Perodie 2001), and
Africa (Rehfisch 1987).
8. Bourdieu 1990, 112, cited in
Dietler 2001, 76.
9. See Clarke 2001; Dietler 2001.
10. Dietler 2001; Dietler 1996.
11. Dillon 1990.
12. Dietler 2001; Junker and Niziolek
2010.
13. For instance, Bender 1978;
Hayden 1995.
14. Meyerson 1990.
15. Mayer 2002.
16. Clarke 2001.
17. Junker and Niziolek 2010.
18. See also Firth 1983.
19. Hayden 2001.
20. Dietler 2001; Dillon 1990.
21. Dillon 1990.
22. Dietler and Herbich 2001.
23. Nash 2010.
24. D’Altroy 2002.
25. Cummins 2002.
26. Cieza de León [1553] 1959, 220.
27. Betanzos [1576] 1996, 67.
28. Vega [1609] 1966, 363–65.
29. See Guaman Poma [1615] 2009,
190.
30. Betanzos [1576] 1996, 66–67.
31. See Guaman Poma [1615] 2009,
225.
32. See Cieza de León [1553] 1959,
163, also during military service, 61;
Nash 2010.
33. Godelier 1977b.
10 0 D onna N ash
34. For example, Cobo [1653] 1990,
151–53. These exclusive events
appear to be related to ancestor wor-
ship; Inca of royal blood, considered
“children of the sun,” worshiped
their ancestor, the sun.
35. See Morris 2004.
36. Betanzos [1576] 1996, 54–55.
37. D’Altroy 2002.
38. Inca queens also owned estates
to support their mummy cults,
and estates might be rewarded to
other members of the royal family
(Betanzos [1576] 1996).
39. Not all mamacona remained in
the aqllawasi. Many were matched
with elite men in marriage, some
joined the emperor’s retinue of
wives, others worked in the royal
estates (Betanzos [1576] 1996, 78),
and a few of royal blood served in
the temples of the sun around the
empire.
40. Morris and Thompson 1985.
The scale of feast preparation facili-
ties at Huánuco Pampa indicates
that massive groups were not fed
in the site’s public plaza. Instead
the aqllawasi likely supplied the
diacritical feasts taking place
between Inca officials and local
leaders (see Nash forthcoming).
41. Hyslop 1993.
42. See Cook and Glowacki 2003.
43. Compare the feasting urn area
BC-B at Conchopata, W. Isbell 2007.
44. Cook 1994; Cook 1983.
45. Ibid.
46. W. Isbell and Knobloch 2006.
47. W. Isbell 1977b.
48. Cook and Glowacki 2003.
49. As far as we know, the Wari
had no tables but likely sat on
benches or on the floor during feasts.
Nevertheless this term communi-
cates the appropriate type of vessel.
50. Cook and Glowacki 2003.
51. Williams et al. 2008.
52. For instance, Brewster-Wray
1983; Green and Goldstein 2010; W.
Isbell 2007; W. Isbell 2001; W. Isbell
et al. 1991; Moseley et al. 2005; Nash
2010.
53. Wari patios are spaces circum-
scribed by rooms, whereas plazas
can be walled spaces in monumental
complexes that are not directly
accessible from the surrounding
rooms. The term plaza can also be
used to describe a more open space
surrounded by buildings in a larger
community context. Both are typi-
cally open to the sky with no roof.
54. Nash 2010.
55. W. Isbell et al. 1991.
56. I calculated that during any
celebration people would not stand
for the entire affair but would sit on
a bench or the floor. Since a seated
person occupies a space of about 60
x 60 cm (24 x 24 in.), the maximum
number in a 100 sq. m (1,075 sq. ft.)
space would be around 165. But
this would leave no space to move.
If, however, the central space is left
open as a performance and serving
space, and guests lined the walls, up
to 50 could sit comfortably, allow-
ing 10 cm (4 in.) between people
to avoid “knocking knees” and
unimpeded access to the door. All
other estimates provided here are
based on the latter configuration.
Although other seating arrange-
ments may have been used, the
presence of benches along walls in
some patios and plazas supports the
one suggested here, which provides
a relative measure of the differences
in group size.
57. Some data from the Andes
suggest that women may not have
consumed as much chicha as men
(e.g. Hastorf 1991). But, given vessels
depicting females who may be host-
esses (see Moseley et al. 2005), it is
likely that elite women partook in
at least some drinking ceremonies.
Although all guests may not have
participated in the drinking aspects
of the feast, it remains probable
that the urn was filled several times
over the course of a festive gather-
ing, perhaps with great fanfare and
display.
58. Ochatoma Paravicino and
Cabrera Romero 2002.
59. W. Isbell 2007.
60. For example, Glowacki 2002
(Cotocotuyoq); Glowacki 2005c
(Pikillacta); Nash 2010 (Cerro Baúl).
61. Moseley et al. 2005 (Cerro Baúl);
Green and Goldstein 2010 (Cerro
Trapiche); Sayre and Whitehead
2003 (Conchopata); Tung and Owen
2006 (Beringa).
 62. Some bottles of mixed, colored
peppercorns sold in markets today
contain molle seeds.
63. Chicha-making experiments
were conducted in Moquegua with
recent Aymara-speaking migrants
from Carumas, a highland commu-
nity in the Department of Moquegua.
These women made several types
of chicha in ceramic vessels so that
the resulting fragments could be
used to run chemical comparisons
with materials from the brewery on
Cerro Baúl. This chemical analysis
is ongoing.
64. Moseley et al. 2005; Nash 2010.
65. Moseley et al. 2005; Nash 2010;
Nash and Williams 2005.
66. DeFrance forthcoming.
67. Thornton et al. 2011.
68. Williams et al. 2008.
69. For instance, W. Isbell et al. 1991
(Moraduchayoq); Nash 2010 (Cerro
Baúl).
70. See Morris 2004.
71. See Seville 2001, 401 for a rep-
resentation of an elite Wari woman
who holds a drinking cup. Her
posture may not be one of service
because the cup is not outstretched
as if being offered to another.
72. Moseley et al. 2005.
73. W. Isbell 2007.
74. Isbell and Groleau 2010.
75. Moseley et al. 2005; Nash 2010.

101 T he A rt of F easting : Building an E mpire with F ood and Drink

Anita G. Cook
The Coming of the Staff Deity
Figure 75c (detail) [2]. Wari, the first empire known in the Andes
Fragment of a faceneck
prior to the Inca, left its signature both on
vessel from Conchopata;
ceramic and slip; 43 x the landscape in the form of imperial archi-
39.5 cm. Museo Histórico tecture and on a multitude of artifacts that
Regional “Hipólito often display staff deity iconography, which
Unanue,” Ayacucho,
MHRA-1778. Photo: Daniel dominated politico-religious rituals for the
Antonio Giannoni Succar. latter half of the first millennium AD (figs.
75a–f). The main protagonist of this iconog-
raphy is a front-facing deity who grasps staffs
or other implements and sometimes appears
with companions that include a winged, staff-
bearing attendant and a sacrificer. In Wari
art, the most complex and complete configu-
rations of these figures, which I collectively
refer to as the staff deity complex, is preserved
on ceramics and textiles. But the figures also
appear on ornaments made of gold, silver, or
semi-precious inlay; carved stone, wood, and
bone objects; and musical instruments (see
pp. 159–91, “Tapestry-woven Tunics”; pp.
217–31, “Inlaid and Metal Ornaments”; and
pp. 233–41, “Figurine Offerings”). In the lack
of any known form of alphabetic writing, the
characters of the staff deity complex were part
of a communication system that also included
a khipu (recording device of knotted strings;
see fig. 180 and [155], p. 276) and many other
motifs, some of them geometric and abstract,
that were likely meaningful. Thus, these ele-
ments serve as a portal into the past and a way
to reconstruct and understand Wari’s cosmol-
ogy, religion, and ideology.
Before proceeding, it is useful to define
these terms: cosmology, religion, and ideol-
ogy. Cosmology, or worldview, explains how
the world came into being, how it functions,
and how people and cultures relate every-
day events to causality.1 Religious practice
involves particular rituals and beliefs; differ-
ent religions may share a worldview and yet
103
undertake distinct practices, as in the case
of Christians, Jews, and Muslims. Ideologies
legitimate the structure of human power by
invoking both cosmological and religious
principles. Cosmology, then, is a framework
for understanding the world that people with-
in a historical tradition, such as the Andes,
broadly share and put into practice at differ-
ent times, in different places, and in different
ways through religion and ideology.
Ideology is usually defined in political
terms as the set of views or ideas that one
social group or class attempts to impose on
another and, in this Marxist sense, includes
strategies of dominance whereby elites attempt
to mask reality for the masses. This notion has
shaped modern understanding of the staff de-
ity complex and has led to the interpretation
that the complex functioned first and foremost
as a symbol of human power that affirmed the
authority of Wari elites. This reading is based
on two facts: the staff deity complex was at
the core of Wari’s state religion, and it appears
on elite objects in compositions that seem to
convey hierarchy.
While recognizing that the complex, like a
logo or flag, communicated a politico-religious
message across the many linguistic boundar-
ies and ethnic groups of the Wari Empire,
this essay takes an alternative approach to
interpretation by relating the staff deity to
native cosmological concepts that have not
so far been extensively considered. These
concepts are recorded in colonial and eth-
nographic sources—respectively, documents
created shortly after the Spanish conquest or
studies of modern Andean native beliefs by
anthropologists. Although these sources date
from a time long after the empire had disap-
peared, they may allow us to draw closer to
10 4 A nita G. C ook
Four types of human faces
occur on faceneck vessels
from a shattered offering
found at Conchopata. The
vessels are human effigies
wearing tunics with the
image of the staff deity
with its winged staff-bear-
ing attendants in profile.
Museo Histórico Regional
“Hipólito Unanue,”
Ayacucho. Photos: Daniel
Antonio Giannoni Succar.
Figures 75a, 75b [3]. Two
views of a fragment of
a faceneck vessel from
Conchopata; ceramic and
slip; 44.5 x 48 cm. MHRA-
1779.
Figure 75c [2]. Fragment
of a faceneck vessel from
Conchopata; ceramic and
slip; 43 x 39.5 cm. MHRA-
1778.
Figure 75d [4]. Fragment
of a faceneck vessel from
Conchopata; ceramic and
slip; 53 x 48 cm. MHRA-
1784.
Figure 75e. Reconstruction
drawing of a faceneck
vessel from Conchopata.
Drawing: Jeffrey Splitstoser.
10 5 T he C oming of the S taff Deit y
Wari’s understanding of the world since they
reveal a native theory of knowledge, one that
honors the existence of otherworldly spirit
beings to which humans make offerings in
order to establish reciprocal relationships.
Social and biological reproduction is made
possible through these “payments” made in
exchange for a life-giving, generative land-
scape on which people depend for their sub-
sistence.2 For the Wari such offerings ranged
from shells, feathers, and plants to the blood
spilled during rituals of sacrifice that claimed
the lives of animals and, on certain occasions,
humans.
The Staff Deity Complex
As the introductory essay in this volume
explains, the staff deity complex was the most
important supernatural imagery depicted
artistically during the Middle Horizon (600–
1000), and it was central to religious experi-
ence both among the Wari and their great con-
temporaries, the Tiwanaku, whose homeland
was located on the Bolivian altiplano (high
plateau), far to the south of Wari territory. In
10 cm
Wari art, members of the staff deity complex
were generally depicted in one of three pos-
tures: full frontal (the staff deity), full profile
(the staff-bearing, winged attendants), and
partial profile (the sacrificer). All three can
also be represented as a disembodied head,
which appears to stand for the whole. These
postures, along with figure size, position, and
attire, have led to the interpretation that the
staff deity complex encodes a hierarchy of im-
portance, status, and authority, with the staff
deity at the pinnacle.
Certain decidedly non-naturalistic traits
seem to mark the characters of the staff deity
complex as supernatural, including vertically
divided eyes,3 traits drawn from the animal
world, heads or headdresses that emanate a
halo-like set of appendages, and streamers
emerging from mouths, belts, chins, and feet.
At least in tapestry-woven tunics and ceram-
ics, depictions of the members of the staff
deity complex display remarkable consistency
in how body parts are shaped and assembled
to create the final figure. In ceramics, this
treatment suggests the widespread use of de-
Figure 75f. Reconstruction
of a faceneck vessel from
Conchopata. Photo: Anita
G. Cook.
sign templates for constituent body parts that
could be neatly assembled to create the whole.
Yet the accouterments depicted—headdresses,
appendages, streamers, and other decorative
elements that dress and give identity to each
figure—are more unpredictable and suggest
that this aspect of subject matter was left in
the hands of specialists to paint without the
use of templates.
the staff deity ( principal huaca). When the
staff deity appears in multifigure compositions
in Wari art, it sometimes takes center stage, as
it does in Tiwanaku art (see figs. 6a, 6b). More
often, it is shown in isolation (see fig. 122). Its
generic traits, in addition to the full frontal
stance, are elaborate garments and ornaments
of various kinds, including leg and arm bands
that closely resemble silver or gold examples
found archaeologically. Its head is surrounded
or partially framed by a band, usually deco-
rated with interlocking frets, that radiates
appendages or streamers tipped with various
motifs. Facial features include a fanged mouth
perhaps based on a predatory feline’s, verti-
cally divided eyes, and a panel, often referred
to as a tear band, that falls from the eye onto
the cheek. To either side of its body it holds the
staffs after which it is named; the staffs some-
times seem to be replaced by other kinds of
implements, discussed below. A stepped ped-
estal that may represent a temple, a mountain,
or both is also associated with Wari representa-
tions of the staff deity, although only rarely.
In the past, the staff deity has been inter-
preted as a single divine entity and identified
as a precursor to such Inca deities as Thunder,
the Sun, or Viracocha, a creator god who was
the loftiest of Inca divine beings.4 As a matter
of convenience, this essay refers to the staff
deity in the singular, but this terminology
appears to be incorrect in several respects. In
1964, Dorothy Menzel identified two versions
of the staff deity on large urns from Pacheco on

Figure 76a. Fragment of a

large, shattered ceramic
urn from Conchopata. Here
the staff deity appears with
a sacrificer. Provenience:
EA2, Locus 903, Special
Find 35A. Photo: courtesy
Conchopata Archaeological
Project.

10 6 A nita G. C ook
Figure 76b. Reconstruction
drawing of an urn from
Conchopata. Drawing:
Anita G. Cook.
Figure 77a, 77b. The two
types of staff deities that
alternate on the exteriors
of urns from Conchopata.
Drawing: Jeffrey Splitstoser.
107 T he C oming of the S taff Deit y
the south coast, based in part on differences
in wardrobe: on the interiors of these vessels,
a deity who wears a belted garment alternates
with one clothed in an unbelted garment (see
figs. 5a, 5b). The same difference appears in
deity representations from Conchopata, a site
close to the Wari capital in the highlands (figs.
76, 77).5 Beyond these distinctions, the staff
deity appears in other iterations. For instance,
the deities who wear belted garments in the
Pacheco and Conchopata urns differ from one
another in some respects, including the forms
of their staffs, the appendages streaming
from their heads, and the ornaments hanging
from their belts. It is not clear whether these
versions refer to a single deity whose traits
changed somewhat through time and space,6
a single deity depicted in different aspects, or
multiple staff deities.7
Another caveat concerns the term “de-
ity,” which, although used in this essay and
throughout this catalogue, may not be an
appropriate term for these formidable, frontal
figures because it brings to mind a Greek-like
pantheon and conjures a worldview radically
different from what we know about Andean
cosmology and how it was put into practice
through religion. If there is an Andean corol-
lary, it would appear to be huaca, which in
Quechua—one of the major, ancient languages
of the Andes—refers to a multiplicity of sacred
forces, objects, or places: supernatural beings
as well as unusually shaped stones, bodies
of water, shrines, idols, images, and many
Figure 78a. Profile staff
bearer on a fragment of
an urn from Conchopata.
Photo: courtesy Concho­
pata Archaeological
Project.
other things. Imbued with camay, or life force,
huacas held the potential of serving as oracles
that spoke through intermediaries to living
audiences.8
profile staff bearers. When they appear
with the staff deity, profile staff bearers as-
sume flanking, ancillary positions and thus
have been interpreted as the deity’s attendants
(fig. 75). They also appear in isolated contexts,
such as on tapestry-woven tunics (see fig. 152)
and ceramics (fig. 78, see also fig. 105), and
as bodiless heads (see fig. 62). (On ceramics
from Conchopata, their bodiless heads appear
with a human figure shown either as a cap-
tive or wielding a knife, suggesting a narrative
that cannot now be reconstructed because the
Figure 78b. Profile staff
bearers on fragments of
urns from Conchopata.
Photo: courtesy Concho­
pata Archaeological
Project.
10 8 A nita G. C ook
ceramics were shattered and are fragmentary.)
As their descriptive name implies, profile
staff bearers carry a single staff, or sometimes
a spear-thrower, to the front of their bodies,
which are shown entirely in profile and with
two legs often bent as though kneeling or run-
ning. A hallmark of this creature is the bird
wing that sprouts from its back and a verti-
cally divided eye, often with a pendant tear
band or surrounded by an elaborate ornament.
Like the staff deity, the attendants usually
have ankle and wrist bands; apart from the
belts that often encircle their waists, however,
they do not appear to wear garments. Their
headdresses are decorated with appendage-
like elements tipped with a variety of motifs,
some of which also occur on the staff deity’s
head appendages. Profile staff bearers occur in
upright or horizontal positions and in many
different versions: 9 many have bird attributes
in addition to their wings, while others appear
to be based on animals, including camelids
and felines, and even humans.
sacrificers. The third main character in the
Wari supernatural universe is a sacrificer, the
only member of the trio to be shown in three-
dimensional as well as two-dimensional forms
(fig. 76).10 In two dimensions, this creature
 appears in partial profile: it has a frontal torso
from which two arms extend but a profile
head and legs, which sometimes appear to be
running. It, too, appears in different iterations,
typically holding a knife or axe in one hand
and a human trophy head or trophy body in
the other; sometimes the trophy is attached
to the end of a staff, as though animating it
and giving it power. Sacrificers wear elaborate
headdresses, belts, garments, and leg and arm
bands. On very large ceremonial ceramics they
appear with staff deities, but they are also de-
picted in isolation (see fig. 233). Many hybrids
of the sacrificer and profile staff bearer exist,
including in early, less well-known examples
of Wari art, which displays experimentation
in a wide variety of styles and imagery at the
local level. These figures eventually coalesce
and culminate in an artistic canon for a formal
religion sometime after 750.
Cosmology in the Andes
An important cosmological principle among
contemporary Quechua-speakers is the con-
cept of “camay,” which refers to the idea of
creation and implies the “energizing of extant
matter.”11 (It contrasts to the Western notion
that creation occurred ex nihilo, out of noth-
ing.) This energy flows through a multitiered
universe: Hanan Pacha, the upper world of
future time and divinities; Kay Pacha, the
present world of humans and other living be-
ings; Uku Pacha, the lower or inside world, the
place of the ancestors and the past.12 The flow
of camay (energy) among these worlds allows
for their reproduction.
The amount of this energizing force is
limited, and the ways in which it can be re-
cycled vary. A myth from Huarochirí (inland
from Lima) recorded in Quechua during the
colonial period illustrates this point. Humans
were originally immortal but they could only
have two children; one of the huacas ate one
of those children, leaving a single surviving
child for each immortal couple. Even with one
surviving child, the world became overpopu-
lated with humans, causing food to be scarce,
so they lived in misery and poverty. Eventu-
ally, people became mortal.
While the consequences of this change
are not spelled out in the Huarochirí myth,
10 9 T he C oming of the S taff Deit y
another myth from the central coast illustrates
one aspect of this transition to mortality.13 In
the central coast case, a poor woman with no
children and nothing to eat asked the Sun
to help her. Instead, or in response, he im-
pregnated her. When she gave birth to a son,
another huaca who was also a son of the Sun
and thus a half-brother of the infant, became
jealous. He murdered the infant and sowed
the body parts into the earth. From the teeth
grew maize, from the bones grew yucas (Mani-
hot esculenta), and from the flesh grew fruits.
Thus, the woman lost a son but had been given
the plants that sustain human life.
Andean peoples made sacrifices to main-
tain the reproductive capacity of their envi-
ronment; sometimes (though not always) these
sacrifices were of human life. Thus death is
intimately linked to fertility and the flow
of energy between worlds and between spe-
cies. Ethnographies often mention death as
movement between this world and another
world, from which the deceased can continue
actively to contribute to the well-being of the
living.14 The three-tiered Andean cosmology
forms a regenerative system in which time,
energizing forces, and the generations recycle.
The concept of transformation is key: energy
continuously flows between tiers, plants and
animals, and animate and inanimate objects.
Mallqui, the Quechua term for “ancestor,” also
means “sapling,” which underscores how the
once-living generate new life. Death is neces-
sary for renewal of the world. This concept is
referred to below as the life-death continuum.
The cycle of myths recorded in Huarochirí
describe numerous examples of rituals relat-
ing to the impounding and releasing of irriga-
tion water in gendered terms. These rites are
usually expressed as conflicts over water in
which women act as erotic sexual beings and
use their wiles to seduce male deities in order
to obtain water for their communities. Thus,
water huacas are male while earth huacas
are female—but the story does not end there.
Women have the power and ritual obligation
to impound water and avoid flooding, a meta-
phor for excessive male sexuality, and both a
man and a woman are involved in rituals to
release irrigation water. So gender complemen-
tarity and conflict are essential for renewal,

Figure 79. Stone tablet
depicting a deity holding
Strombus and Spondylus
shells, from Chavín de
Huántar. After J. Rowe
1962a, 12, fig. 11; drawing:
Fred D. Ayres.
110 A nita G. C ook
with female sexuality often acting as the
control on exuberant male sexuality.15 In the
Andes, gender complementarity can reach well
beyond the human body to include personified
landscapes that have one or more genders.16
Two Quechua words help in understand-
ing these ideas. Yanantin (to serve together)
refers to two complementary parts, or two
things that are both necessary. It can allude
to men and women but also a pair of gloves or
a pair of shoes; maleness is the complement
of femaleness and the left is the complement
of the right. In order to function properly,
both are needed. Tinku denotes where things
meet, join, or intersect to create a whole.
While yanantin distinguishes the two comple-
mentary elements, tinku describes how they
come together: two tributaries of a river, the
medial line of the human body at which the
two halves join, and the mingling of male
and female elements. Creation and renewal
take place where complementary forces come
together—in human intercourse; in the union
between mountain peaks, conceived as power-
ful masculine forces, and fertile flatter lands,
the realm of the earth mother Pachamama;
in the rain (semen) that falls from the sky to
fertilize the womb of the earth.17 All these
notions are expressions of an Andean concept
referred to as dualism.
In Andean relativistic thought, no person
is entirely male or entirely female.18 In fact,
there is a tension between the need to have
two complementary parts (yanantin) and
the urge for them to come together in union
(tinku). While the complementary parts are
conceived of as male and female, be they land-
forms or social groupings, one gender aspect
can be emphasized over the other in dynamic
interactions,19 and this is the necessary con-
stant in renewal. In other words, the gendered
elements are only effective when united.
The most general principles of Andean
cosmology, then, are complementarity between
male and female forces, a balance in the flow
of energy, and the need for reciprocal relation-
ships, whether between men and women or
people, plants, animals, and deities. These
principles have deep and widespread roots,
even in the Amazon, where contemporary
natives still maintain extraordinary cosmolo-
gies.20 Many believe that these basic principles
date at least as far back into the past as Chavín
de Huántar, an important temple site that
functioned between 1200 and 500 BC and is
known for its complex stone sculpture. One is
the Tello Obelisk, a slender, tapering, four-
sided monument carved with a male cayman
on one of its main faces and a female cayman
on another; plants sprout from parts of their
bodies.21 Another is a small stone plaque that
depicts a deity with upturned mouth—the
so-called Smiling God—who holds a Strombus
(conch) shell in one hand and a Spondylus
(thorny oyster) shell in the other; these shells
are associated, respectively, with male and
female principles (fig. 79). Finally, the Lanzón,
a blade-shaped monument set within a cruci-
form chamber in one of Chavín’s temples, is
often considered to represent an axis mundi,
a vertical axis that connects different levels
of the cosmos.22 These monuments provide
an example of how art works may illustrate
a complex cosmology involving the flow of
energy through a multilayered universe, facili-
tated by the joining and balancing of male and
female principles.
Cosmological Principles in the Wari Staff Deity
Complex
the life - death continuum . Andean iconog-
raphy very commonly features a disembodied
trophy head. For instance, during the pre-Wari
Early Intermediate Period (AD 1–600), the arts
of the Recuay culture of the north highlands,23
the north coast Moche, and the south coast
Figure 80. To the left of
the main figure’s head
on this bowl is a trophy
head with plants sprout-
ing from its mouth. Bowl
with sacrificial ritualist;
180 BC–AD 500; Nasca;
ceramic and pigment; 10.2
x 17.1 cm. The Art Institute
of Chicago, Kate S.
Buckingham Endowment,
1955.1934. Photo: The Art
Institute of Chicago.
Nasca all depict decapitated human heads. In
these cultures’ art severed heads and bodily
orifices as well as staffs sprout a variety of
things: among them, plants of various kinds;
streamers or appendages—perhaps abstract
representations of a vital force—that ter-
minate in new life forms, including human
beings; and serpents, including the amaru, a
hybrid serpent-feline that today symbolizes
the rainbow and water, and is closely associ-
ated with maize. These images seem to be
inspired by the concept of a continuum in
which life springs from death and particularly
from the head, uma in Quechua, which also
means “mountain top”—the sacred realm of
life-giving deities known today as apus or
wamani—and carries additional connotations
of seniority, leadership, and power. Given
these associations, it is no surprise that, in
many Andean ethnographic studies, the head
is identified as the most important part of the
body and the repository of a person’s essence.24
This may be at least part of the reason why
the members of the Wari staff deity complex
are often represented by their heads alone,
although whether they represent trophies is
unknown.
Donald Proulx addresses what he per-
ceives as a link in Nasca iconography between
the severed human head, concerned with
“death/decapitation/blood,” and the germinat-
ing plant forms that emerge from the head,
which embody “regeneration/rebirth/agricul-
111 T he C oming of the S taff Deit y
tural fertility” (fig. 80).25 In other words, heads
were the locus of power and could assure suc-
cessful harvests; indeed, Nasca trophy heads
that germinate plants suggest the spring plant-
ing season while plants decorated with trophy
head faces may refer to harvest. The Nasca and
the Wari used distinct artistic conventions to
express this life-death continuum,26 exempli-
fying how shared cosmology is expressed in
slightly different ways at different times. For
while Wari art also features severed heads, it
adds an ambiguous design that can be inter-
preted as either a plant or human organs (the
heart, lungs, and trachea), hereafter termed
the sacrifice motif.27 The ambiguity may be
intentional.
This artistic evidence of trophy heads
is supported by archaeological finds: actual
decapitated bodies and caches of trophy heads
have been recovered in excavations in Nasca
territory.28 In Ayacucho, the Wari heartland,
the ancients placed human trophy heads in
temple foundations as early as the first mil-
lennium BC.29 In Wari contexts, such heads
have been documented in ritually important
locations at Conchopata30 and in an elite
cemetery at Cotocotuyoq in the Huaro Valley,
near Cuzco.31 There is plentiful evidence that
the practice of taking human heads has even
deeper roots in the Andes.32 During the Wari
period, the interest in head taking is likely
explained not only by ritual practices but also
by emerging elites’ interest in empowering
themselves, potentially by impersonating or
imbuing themselves with camay, the energet-
ic, creative life force of the cosmos.
Trophy heads and trophy figures, often
portrayed as captives, are prominent in the
iconography of the staff deity complex, which,
not surprisingly, appears to be associated with
both life-giving and life-taking—the life-
death continuum that generates renewal—and
perhaps with the male-female gender comple-
mentarity essential to this renewal. Large
ceremonial urns used to serve beverages or
food are painted with iconography that links
the staff deity alternatively with agricultural
abundance and human sacrifice. The urns
that demonstrate the staff deity’s connection
to life-giving fecundity in the form of agricul-
tural plants come from Pacheco, the site on
 the south coast in the Nasca region; in these
vessels, motifs associated with death and hu-
man sacrifice are entirely absent.33 The Pa-
checo vessels (see figs. 1, 5) were part of a huge
(allegedly 3 ton) deposit of ceramics that were
shattered in antiquity and buried; a hand-
ful have been reconstructed. Two versions of
the staff deity alternate with one another on
the interiors of several urns. Both versions
are decorated with ears of maize, which tip
the appendages that flow hair-like from their
heads and additionally festoon the garment
and the staff of one of the figure types. The
two are distinguished in part by their gar-
ments and, based on this distinction in dress,
they have often been interpreted as male and
female, the male wearing a belted tunic and
the female an unbelted dress along with a
shoulder mantle or shawl. Although questions
still remain about these identifications,34 the
two may embody the gender complementar-
ity discussed above. Whatever the case, it
seems clear that they refer to Andean concepts
concerning dualism. On the exterior of the
Pacheco urns, the staff deity in the belted gar-
ment alternates with an image of the deity’s
head. (The alternation in the color of the back-
ground—red behind the full-bodied deity and
black behind the head—may encode an oppo-
sition that could refer to wet and dry seasons
or day and night.) Other ceramics found in
the Pacheco deposit also allude to abundance;
they include many large urns painted with
images of blossoming highland plants of dif-
ferent kinds (see figs. 130a, 130b) along with
effigies of domesticated animals, particularly
camelids, which were essential sources of
protein and wealth (see figs. 137–39). Thus, the
staff deity may also have had ties to animal
fertility.35
The staff deity is connected more fully
to both terms of the life-death continuum in
the iconography of urns from Conchopata,
the highland site; many of these vessels were
also intentionally broken and buried. The
urns frequently depict two different kinds of
staff deities who, again, wear either a belted
garment or an unbelted garment.36 On a few
urns, both versions of the deity co-occur with
imagery that refers to generative qualities, as
at Pacheco: maize, cultivated plants, and the
112 A nita G. C ook
amaru (water serpent). On others, the two dei-
ties appear in separate scenes with sacrificers
and carry a staff in each hand (figs. 76, 77; see
also fig. 102).37 The lower end of one staff, car-
ried in the proper left hand, has an elite hu-
man captive, hands bound behind its back, at
its tip; the other staff terminates in an animal
head with the sacrifice motif (the disembodied
human heart, lungs, and trachea) emerging
from its mouth. On these vessels, panels of
face-fret motifs common in tapestry-woven
tunics (see fig. 144) subdivide the figures into
scenes, each perhaps addressing different
stages or rituals of sacrifice. These urns, found
in contexts that were radiocarbon dated to
the tenth century, represent the final genera-
tions of the Wari empire,38 and their imagery
has been interpreted as reflecting conflict that
came about as Wari leaders tried to increase
political control in the face of instability.39
However, a perspective that links these repre-
sentations to principles of cosmology is vitally
important.
The sacrifice motif also appears on other
Wari objects. One is a wood container carved
in the shape of a winged figure that may wear
a feline or fox mask and pelt; the sacrifice
motif appears on its back (see fig. 234). A mag-
nificently carved wooden mirror also displays
a sacrificer, its belt formed by a bicephalic
serpent that may represent the amaru; the sac-
rifice motif emerges from the mouth of each
serpent and appears in more complete form
in the rib cage of the small figure that dangles
from one of the sacrificer’s staffs. From its
upper reaches, this staff sprouts motifs that
may represent foliage generated by the sac-
rifice (see fig. 206). A third example is seen
on a tapestry-woven tunic with decapitated
heads that emit butterflies; here the sacrifice
motif alternates with a butterfly in the narrow
bands along the tunic’s sides (fig. 81). An-
other example occurs on additional urns from
Conchopata; the motif appears in the beak of
a bird associated with the staff deity complex
(fig. 82).
Another indication of the staff deity’s role
as the transformer who both destroys and
gives life may come from the objects it holds
in its outstretched hand, often the staffs that
in ancient times were crucial symbols of both
Figure 81 [128]. Tunic with
heads, insects, and heart-
lung-trachea motifs; cotton
and camelid fiber; 101.6 x
105.4 cm. Private collec-
tion. Photo: Renée Comet
Photography.
113 T he C oming of the S taff Deit y
human and divine authority. So important to
indigenous symbol systems are staffs that they
remain strongly affiliated with leadership in
contemporary native communities. For in-
stance, today in some areas of the highlands,
staffs are considered to be animate and en-
dowed with special, sometimes terrible pow-
er.40 They are used in symbolic gestures that
mark the passing of power from one authority
figure to the next—a use that may be relevant
to Wari elites, their actions sanctioned and de-
fined by their relationship to the staff deity. In
some Wari representations, however, the staffs
may be replaced by either a spear-thrower or
an agricultural tool, the Andean foot plow
(chaquitalla). Both have a hook-like protrusion
from the side. In the spear-thrower, the hook
serves as a thumb rest; in the plow, a laborer
puts a foot on the hook to push the tool into
the earth (fig. 75c).
In these different contexts and ways, then,
the staff deity seems to be associated with

Figure 82. Reconstructed
urn showing a mythical
bird with the heart-
lung-trachea motif in its
beak, from Conchopata.
Provenience: EA79, Locus
1590, Special Find 1080.
Photo: courtesy José
Ochatoma Paravicino and
Martha Cabrera Romero.
Figure 83. The head of one
of the deities on urns from
Pacheco. Drawing: cour-
tesy American Museum of
Natural History, New York,
Division of Anthropology,
cat. 41.0/5314.
114 A nita G. C ook
both the destruction of life and nature’s fertil-
ity. The deity thus may express the transfor-
mative capacity of sacrifice—the life-death
continuum—and illustrate how life is nour-
ished by sacrifice. The taking of trophy heads
results in bloodletting that, in its turn, feeds
the earth and allows life to regenerate. The
head and internal organs, while indicative of
death, also enable life. More important, each
version of the staff deity embodies the trans-
formative power of camay, the life force itself.
other aspects of cosmology. The elements
at the tips of the appendages radiating from
the staff deity’s head may also relate to cosmo-
logical principles. These extremely interesting
terminus motifs vary somewhat from one rep-
resentation to the next but frequently include
the profile heads of felines (probably jaguars or
pumas), feline-camelid hybrids, foxes, birds of
prey, deer, a few other animal-like heads with
distinct noses (perhaps including the bear),
and a peculiar frontal head that may represent
the amaru, the serpent-like creature associ-
ated with water and maize. The heads often al-
ternate with a variety of unidentified designs
such as disks with a central dot or motifs that
resemble feathers. For instance, moving in the
clockwise direction from the center top of the
head of the Pacheco deity who wears a belted
garment (fig. 83; see also fig. 5), the first motif
represents a “feather,” the next is an ear of
maize, then a bird head, a feline head, maize
and a bird again, the head of a serpent, and
finally another bird and feline. The longer-
eared profile heads atop the staff may repre-
sent foxes; they are similar to fox heads shown
on whistles found at Conchopata.
Scholars have so far offered no interpre-
tation of these motifs, but information from
ethnographic sources is suggestive: in these
sources, felines—pumas and jaguars—fulfill
the role of mediator between the Hanan Pacha,
the upper world of divinities, and Uku Pacha,
the underworld where the ancestors reside.41
The fox occupies a middle position, always
subordinate to the puma, although foxes and
pumas are often considered to be brothers.
Raptors in modern lore are mediators between
Kay Pacha, the present world, and Hanan
Pacha.42 Deer are spirits of the mountain apus/
wamanis (deities); if killed, they need to be
put to very good use. All these creatures,
then, serve in mediating, communicator roles
among opposing concepts; they transform
and move generative forces among levels of
the cosmos. Their presence on the staff deity’s
head appendages and staffs may imply that the
deity fulfilled similar functions. It is possible
that the profile staff bearers, who serve as the
deity’s attendants, refer to the same concepts
since their heads and other traits vary among
birds, felines, camelids, perhaps other ani-
mals, and humans.43 Perhaps they were also
conceived as mediators between the staff deity
and the human community.
Other motifs that Wari artists employed
are more difficult to interpret. Particularly
common in staff deity imagery is a zigzag that
often appears on chin bars, belts, headdress
platforms, and staffs. The concept of camay
may provide a useful way to think about this
design, which potentially references move-
 ment, flow, and the cycling of life force (in-
cluding water) so central to biological and
social regeneration. Even more speculatively,
designs shown at specific points of physical
articulation or joints, such as the jaw, neck,
and waist, may refer to an animating force.
Antecedents and Sources of the Staff Deity
Complex
As mentioned above, the staff deity and its
profile attendants have traditionally been
interpreted as a symbol of human authority,
hierarchy, and power. The foregoing discus-
sion highlights the cosmological beliefs that
the staff deity complex may embody and that
many cultures in the Andes may have shared
over a period of some two millennia and that
Wari shaped into a state religion. Each of these
two very different perspectives, which are not
mutually exclusive, recognizes the importance
of earlier iconographic traditions in the evolu-
tion of the staff deity complex.
As noted above, the staff deity complex
appears in the art of both Wari and Tiwanaku,
the latter famous for its stone carvings that
include figural monuments as well as massive
stone portals with decorated architraves or
lintels. The Gateway of the Sun is among the
most extensively carved of such architectural
elements. It incorporates a large, centrally
positioned staff deity executed in high relief;
the figure stands on a stepped platform and is
flanked on either side by three rows of smaller
Figure 84. Yaya-Mama style
tablet from Copacabana,
Bolivia, 80–100 BC;
stone; 40 x 32 x 3.5 cm.
Ethnologisches Museum,
Berlin. Image: bpk, Berlin/
Ethnologisches Museum/
Art Resource, NY.
115 T he C oming of the S taff Deit y
winged figures shown in profile and executed
in lower relief. Beneath this ensemble is a
meander that encloses slightly different ver-
sions of the staff deity’s head flanked at either
end by a small human who wears a feline pelt
and, like a sacrificer, grasps a human trophy
head in one hand and a knife or trumpet in
the other.
At present, sufficient lines of evidence ex-
ist to dispel the old idea that the Wari version
of the complex derives in a direct line from
Tiwanaku.44 But early cultures in the south,
some in Tiwanaku’s neighborhood, have
remained the focus of attention in tracking
the development of staff deity imagery. These
iconographic traditions—recently dubbed the
“Southern Andean Iconographic Series”—de-
veloped over the course of a millennium in
the Lake Titicaca Basin, northern Chile, the
far south coast of Peru, and elsewhere.45 In the
Titicaca region, they include the imagery asso-
ciated with a widespread phenomenon known
as the Yaya-Mama Religious Tradition, a late,
refined manifestation of which occurred at Pu-
cara, now recognized as the place where a pre-
cocious synthesis of religious ideas emerged.
The tradition was also important to the north
Chilean people living in the desert oases at
San Pedro de Atacama.
the yaya- mama religious tradition. The
Yaya-Mama Religious Tradition, which existed
from about 800 BC to AD 200/300 in the Lake
Titicaca Basin, is characterized by ceremo-
nial sunken-court architecture, specific ritual
paraphernalia, and distinctive art.46 The term
“Yaya-Mama” combines the Quechua words
for father or man (yaya) and mother or woman
(mama), and the tradition is named after a
stone stela from Taraco, Peru, that features a
male on one of its principal faces and a female
on the other.47 These are some of the earliest
examples of gendered figures in the region.
Despite the highly variable nature of early
Yaya-Mama iconography, which includes
references to water and its denizens (fish,
serpents, and toads), it incorporates traits that
continue into later Pucara, Wari, and Tiwanaku
art, including full-bodied frontal and profile
figures (without staffs) as well as disembodied
heads with appendages or streamers (fig. 84),

Figure 85. Rectangular
stone vessel from Pucara,
Bolivia. After Chávez 2004,
fig. 3.23; drawing: Sergio
Chávez.
Figure 86. “Camelid
Woman” or “Woman
with Alpaca” depicted
on ceramics from Pucara.
After Chávez 2004, fig.
3.24a; drawing: Sergio
Chávez.
Figure 87. “Feline Man”
depicted on ceramics from
Pucara. After Chávez 2004,
fig. 3.24b; drawing: Sergio
Chávez.
116 A nita G. C ook
and a T-shaped eyebrow and nose. Yaya-Mama
ceramics may provide an early occurrence of
another important staff deity trait: the verti-
cally divided eye with a tear band that de-
scends from the eye onto the cheek.48
pucara . This important, mid-size polity of
the northern Lake Titicaca Basin flowered
between about 200 BC and AD 200/300. Its
ritual architecture consists of stepped pyram­
idal platforms with sunken courts on their
summits; its art, a late and extremely fine
expression of the Yaya-Mama artistic tradi-
tion, essentially revolutionized imagery in the
altiplano and introduced a few images that
are pivotal antecedents of Wari and Tiwana­
ku art. Among them are a three-dimensional
sculpture of a staff-carrying figure purport-
edly from the Pucara site itself;49 rectangular
carved stone vessels50 and ceramics decorated
with appendaged heads that resemble the
heads of the later Wari and Tiwanaku staff
deity (fig. 85);51 and a beautiful gold plume
from the Cuzco region that features a feline-
like figure with appendages emanating from
its framed head.52 Pucara head appendages
are often tipped with designs of various sorts,
including profile animal heads; perhaps they
have regenerative qualities. The vertically
divided eye continues in Pucara ceramics and
sculpture, as do disembodied heads.
Pucara was also the place where gendered
figures become fully developed, particularly in
ceramics.53 These figures include a front-facing
female who holds a staff in one hand and, in
the other, a tether attached to an alpaca (fig.
86). (The staff appears to represent a distaff, an
implement used when fiber is spun into yarn.)
For this reason, she is known as “Camelid
Woman” or “Woman with Alpaca.” She is also
associated with plants. These conjunctions
have prompted Sergio Chávez to identify her
with agriculture, pastoralism, and abundance
in general.54 As Chávez notes, in a general
way—her frontality, outstretched arms, and
staff—she seems to be a precedessor of the
later staff deity, with significant differences.
Another important Pucara persona is a
male depicted with head and running legs
in profile but torso in frontal view (fig. 87).
Known as “Feline Man” after his fangs and the
feline pelts that he sometimes wears, he, too,
wields a staff in one of his outstretched hands.
In the other, he clutches an axe or knife and a
human trophy head or decapitated body. Sev-
eral of Feline Man’s features—the partial pro-
file stance, trophy, and axe—are components
of the later Wari and Tiwanaku sacrificer.55
According to Chávez, the Feline Man is always
depicted in pairs that either confront or seem
to chase one another. These scenes may refer to
tinku battles, a practice that survives today as
 ritual battles that are underpinned by the idea
of the merging of complementary opposites
(the opponents) and a spilling of blood that
produces fruitfulness of the earth and general
fertility. Of course, as mentioned above, the
term “tinku” also connotes the place where
two complementary halves come together.
Thus, the Pucara Camelid Woman and Feline
Man may have relevance to the interpreta-
tion proposed here for later Middle Horizon
iconography. The use of a design template
for significant parts of the body in Pucara art
seems apparent and is later employed in Wari
iconography, as mentioned above.
northern chile . San Pedro de Atacama in
the Chilean desert was an important oasis
center whose contacts with the Lake Titicaca
area are evident in well-preserved burials that
contain artifacts with iconography related to
the staff deity complex that dates to both the
Pucara and Wari-Tiwanaku time periods.56
Here burials included a rich array of wood
snuff tablets, tubes, and spoons with incised
and carved handles that depict three main
figure types: staff deities, profile staff bearers,
and sacrificers (figs. 88a, 88b). These objects
were used in special medicinal and ritual
practices involving the inhalation of mind-
altering substances popular in the southern
Andean region but far less evident in the ar-
chaeological remains within the Wari sphere.
Figure 88a. Wood snuff
tablet from San Pedro de
Atacama, Chile. Photo:
Constantino Manuel
Torres.
Figure 88b. Image that
appears on the tablet from
San Pedro de Atacama,
Chile. After Sandweiss
1987, 217, fig. 1; courtesy
Donna P. Torres.
117 T he C oming of the S taff Deit y
early tiwanaku. Recently, a number of
remarkable tapestry-woven tunics have been
recovered on the far south coast and add to
a growing body of material that immediately
predates the Middle Horizon.57 These tunics
show a disembodied head with appendages
that is very similar to that of the Wari and
Tiwanaku staff deities. Radiocarbon dates and
studies of the iconographic elements suggest
that these textiles and other art works from
the far south coast are probably early Tiwa­na­
ku in style. Thus, they fill a gap in the se-
quence between Pucara and the later Wari and
Tiwanaku periods.58
other iconographic sources. It is evident
that Wari also interacted with many con-
temporary cultures outside of the southern
sphere; certainly the Wari capital was a nexus
that drew craft specialists who brought with
them knowledge of a wide array of art styles.
The Wari also must have been aware of tra-
ditions that, by the time of its fluorescence,
were part of an honored past. Some of these
past and contemporary traditions may have
influenced Wari’s decision to adopt staff deity
imagery, and many are known or suspected
to have contributed other elements to Wari art
and architecture. During the Early Horizon
(1000 BC–AD 1), for example, a front-facing
staff-bearing deity and profile winged atten-
dant figures appear in the monumental and
 portable art of Chavín de Huántar, the impor- include early Chavín-inspired staff deities
tant pre-Wari site in the northern Peruvian and profile figures and the so-called Oculate
highlands discussed above (fig. 89). A con- Being, a mythical creature associated with
temporary of Chavín, the Paracas culture of human sacrifice whose frontal head radiates
Peru’s south coast was also likely a significant appendages (fig. 90).
source of inspiration since the south coast is Several other groups developed during the
far closer to Wari’s heartland than any of the following Early Intermediate Period, which
other regions mentioned here. Some possible preceded Wari’s rise. One of particular impor-
Paracas antecedents to staff deity iconography tance was the south coast Nasca, descendant
of Paracas; it is known that the Nasca and
Figure 89. The Raimondi the Wari had very close relations for several
Stone from Chavín de
centuries around the dawn of the Middle Ho-
Huántar. The monument
is carved in relief with an rizon and, aside from focus on trophy heads,
image of a staff-bearing Nasca iconography includes mythical beings
supernatural; from its with streamers that generate plants, animals,
head flows an elaborate
configuration of fanged and human figures that may have informed
heads conceptually similar the later Middle Horizon staff deity and its
to the Middle Horizon staff companions (fig. 80). Recent research reveals
deity’s appendages. After
Burger 1992, 175, fig. 176.
that Wari precedents may also be found in
Recuay culture of the northern highlands;
these include compositions that feature
frontal figures whose heads sprout append-
ages and who are flanked by profile zoomor-
phic creatures.59 The north coast Moche, too,
probably made contributions since Moche art
had long featured a sacrificer who wields a
knife in one hand and a trophy head in the
other. Many scenes involving the life-death
continuum occur in Moche iconography as do
processions of various figures encountering a
central lord. Also, murals at the Huaca de la
Luna, one of the principal civic-ceremonial
buildings in the Moche sphere, display a head
with append­ages reminiscent of the Wari staff
deity.60 There are also other, underresearched
antecedents outside of the southern sphere,
but these few examples serve to indicate that
the staff deity complex seems to have emerged
from many widespread roots.
Most scholars agree that when the staff
deity complex appears in the Wari region, it
does so abruptly, in about 550–600. This early
dating is confirmed by a set of unusual Con-
chopata urn fragments with staff deity com-
plex iconography that, unlike anything else
found in the Wari heartland, display a num-
ber of Pucara traits61 previously known only
in the southern altiplano.62 These fragments
offer evidence that the staff deity complex
was fully developed in the Ayacucho region
by at least the very early Middle Horizon,

118 A nita G. C ook

Figure 90. The enormous
eyes give this creature its
name: the Oculate Being.
Attenuated serpent-head
appendages radiate from
the edges of the mask.
Oculate Being mask; 300
BC–AD 1; ceramic and
resin-based paint; 23.6
x 22.5 x 13.2 cm. The
Cleveland Museum of Art,
Purchase from the J. H.
Wade Fund 2003.39.
and perhaps even by the end of the preced-
ing Early Intermediate Period. The very large
Conchopata jars depicting humans who wear
tunics painted with staff deity iconography
also likely date to the first epoch of the Middle
Horizon, when Wari ritual practitioners and
artists were experimenting with, selecting,
and blending foreign religious concepts and
iconography. (This dating is based on stylistic
analysis rather than radiocarbon dates, which
are not available for the context in which the
jars were found.) In structure and layout, the
imagery shown on these vessels is the closest
known Wari parallel to the imagery on the
Tiwanaku Gateway of the Sun,63 evidence that
the two polities interacted. Aside from the
staff deity, the vessels feature a mix of Wari
and coastal Nasca designs, 64 including highly
stylized versions of Nasca figures grasping
trophy heads that hark back to Nasca trophy
heads,65 two types of humpback animals
executed in the Chakipampa style of the Wari
heartland, and a design that may represent an
imperial D-shaped temple (see pp. 65–81, “The
Wari Built Environment”).66 The fact that staff
deity imagery is accompanied by local and
coastal icons indicates a high degree of open-
ness and experimentation during the stages
when Wari religion was forming.
119 T he C oming of the S taff Deit y
Synthesis
Why did the Wari develop an interest in the
staff deities? Perhaps Wari leaders attempted
to validate their rule by affiliating themselves
with an honored past and with images and
ideas that were widespread in the Andes
at the time that they came to power. They
were also intent on creating a state religion
that would work at home and abroad. At the
center of this religion was an innovative form
of a staff deity who was tied to a concept of
a cosmos sustained by reciprocity, balance,
cooperation, and renewal that likely had been
prominent in the Andes for centuries, if not
millennia, before the Wari rose to power.
The staff deity and its companions served to
underscore relationships that allowed Andean
people to rely on one another and to exchange
goods and spouses among different ecologi-
cal zones and along the vertical gradient of
the highlands (see pp. 1–3, “Introduction”).
Centuries later, the Inca, perhaps following in
Wari’s footsteps, did the same.67
In synthesizing staff deity iconography,
the Wari took account of sources from many
different regions of the Andes. As the empire
incorporated new areas, it transformed this
iconography into a new visual paradigm. The
iconography and cosmological concepts most
likely took shape together, developing from an
incipient period of multiethnic regionalism
at the beginning of the Middle Horizon to an
interregional understanding based on personi-
fied huacas responsible for the regeneration of
life. Although the exact sequence in which im-
portant icons developed through time remains
under study, it is known that, early in Wari’s
development, the staff deity complex appeared
at Conchopata on effigy jars decorated with in-
dividuals who wear tunics that show the staff
deity and its attendants. The staff deity repre-
sents a foreign idiom in familiar terms. Over
time the staff deity continued to appear on ce-
ramics, both large ceremonial serving vessels
and many kinds of smaller vessels, many used
in official feasts68 at which Wari elites pursued
their agendas with the local leaders (see pp.
82–101, “The Art of Feasting” and pp. 145–57,
“Shattered Ceramics and Offerings”).
 The other major medium in which com-
plex-related figures appear is tapestry-woven
cloth, especially tunics worn by elites, most
likely gifts from the ruler and surely worn on
special state occasions. With a few exceptions,
the tunics’ iconographic repertoire is re-
stricted to profile staff bearers and sacrificers.
The absence of the staff deity from the tunics
is curious. It may be, as Susan Bergh suggests
elsewhere in the volume (see pp. 159–91,
“Tapestry-woven Tunics”), that by wearing the
tunics Wari elites cast themselves as surro-
gates for the staff deity in some way; dressed
120 A nita G. C ook
in the tunics, they assumed the position of the
staff deity flanked by the figures woven into
the cloth. Or these elites may have fulfilled the
roles of the winged staff bearers and sacrifi-
cers shown on their tunics and thus served as
intermediaries between human communities
and the staff deity, the powerful huaca suf-
fused with camay and the awesome capacity
to transform death into life. Eventually the
tunics were buried with their owners, who at
death became ancestors that, like the deity,
were responsible for their descendants’ future
prosperity.
Notes During the final stages of writing
this article, John Topic graciously
provided his intellectual acumen,
insight, editing advice, and support
in times of doubt, at the cost of
many hours. It has been an excit-
ing journey—thank you. I am also
deeply indebted to Susan Bergh,
who devoted much time to com-
menting and editing; our engage-
ment with Wari iconography has
only increased my respect for her
knowledge and attention to detail.
I also thank Tom Zuidema and his
progeny, who planted the seeds,
and the many colleagues who have
given me the benefit of their insights
over the years. I am responsible for
content and any errors.
1. Flannery and Marcus 1996; J.
Topic 1992; T. Topic and J. Topic
2009, 23–25.
2. For instance, Allen 2002; Delgado
Sumar 1980; Earls and Silverblatt
1978.
3. Although the divided eye has
been considered a supernatural trait,
its significance remains unresolved.
It often appears on human and hybid
figures and deserves more careful
study.
4. Demarest 1981; Menzel 1977;
Menzel 1968; J. Rowe 1960b; J. Rowe
1946; Valcárcel 1959.
5. Cook 1984–85; Menzel 1977, 54;
Menzel 1964, 19, 26.
6. W. Isbell and Knobloch 2009;
Menzel 1977; Menzel 1968; Menzel
1964.
7. Makowski Hanula 2009;
Makowski Hanula 2002.
8. For instance, Curatola Petrocchi
2008; Gose 1996; J. Rowe 1946;
Szeminski 1987; J. Topic 2008.
9. See Bergh 1999 for a compre-
hensive description of attendant
variants in tapestry-woven tunics.
No comparable classification is
available for other media.
10. For instance, Cook 2001a; Cook
1983; Valcárcel 1959.
11. Salomon 1991, 26.
12. Anders 1986, 911; Delgado Sumar
1980; Earls and Silverblatt 1978;
B. Isbell 1978, 207–14; Morisette
and Racine 1973; Roe 2008; Urbano
1981; Urbano 1980; Urbano 1978;
Valcárcel 1980, 78–79.
13. Calancha [1638] 1974–82.
14. Allen 2002; Glowacki and
Malpass 2003; Gose 2008; Gose 1994.
15. B. Isbell 1997.
121 T he C oming of the S taff Deit y
16. Bastien 1995; Classen 1993;
Douglas 1996; Gose 1994; Rösing
2003, 101–7; Urton 1996.
17. J. Topic 2008; J. Topic et al. 2002.
18. Allen 2002, 180.
19. B. Isbell 1997.
20. For instance, Guss 1989; Hugh-
Jones 1978; W. Isbell 1978; Roe 2008.
21. Burger 1992, fig. 141.
22. Ibid., figs. 126, 127, 140.
23. Lau 2011; Rodman and
Fernández Lopez 2001, fig. 34.
24. Arnold and Hastorf 2008.
25. Ibid., 63; Proulx 1999.
26. Arnold and Hastorf 2008;
Carmichael 1994; Franquemont et al.
1992; Proulx 1999.
27. Martha Cabrera Romero brought
the interpretation of the motif as
internal organs to my attention.
28. DeLeonardis 2000; Browne et al.
1993; Verano 1995.
29. Lumbreras 1981.
30. Andrushko and Bellifemine
2006; Tung et al. 2007; Tung and
Knudson 2008.
31. Glowacki 2007.
32. In some instances, coastal
mummy bundles encase only a
human arm or leg; while not techni-
cally representing trophies, they
may be substitutes for the whole
body.
33. Recent excavations have
revealed that Pacheco style (Robles
Moqo) pottery was also produced
at Conchopata; it is also known at
Wari.
34. Menzel (1977, 54; 1964, 19, 26)
and Lyon (1978) see gender distinc-
tions in wardrobe. I have some res-
ervations about this interpretation
(Cook 2001b; Cook 1992) but believe
that the pair embodies the Andean
principle of duality. See Bergh 2009
for a discussion of pairs in textile
imagery and Cook 1992 for a discus-
sion of pairs in caches of figurines
from Pikillacta.
35. Bergh, personal communication
2011. It may be worth noting that, on
rare early examples of the staff deity,
the band around its head is filled
with S motifs that today symbolize
fertility (B. Isbell 1978).
36. Knobloch fig. 102 in this volume
is a more recent and accurate recon-
struction drawing, based on more
complete information.
37. Cook 1984–85.
38. W. Isbell 2001, fig, 28; W. Isbell
and Cook 2002; W. Isbell and
Knobloch 2009.
39. Some suggest that a drought
helped bring down the Wari empire.
If so, Wari agriculture may have
been threatened and the direct
exploitation of adjacent areas may
have become necessary.
40. For instance, B. Isbell 1978; Skar
1982.
41. B. Isbell 1978; Salomon and
Urioste 1991; Urton 1985; Zuidema
1973.
42. B. Isbell 1978.
43. Bergh, personal communication,
2011.
44. Cook 1994; see overview by
Janusek 2008.
45. W. Isbell and Knobloch 2009.
46. Chávez 2004.
47. Ibid., 70–73; Janusek 2008, fig 3.8.
48. Chávez 2004, 90.
49. Young-Sánchez 2004d, figs.
3.9a,b.
50. Chávez 2004, fig. 3.23.
51. Chávez 2002, fig. 2.4g.
52. Young-Sánchez 2004b, 94–95; J.
Rowe and Brandel 1976, pl. IV, figs.
15–19. It shares a number of traits,
including posture, with figures from
Chavín de Huántar. This figure’s
face is reminiscent of south coast
pre-Paracas Puerto Nuevo ceramic
iconography (García 2010).
53. Chávez 2002.
54. Chávez 2004, 91.
55. Cook 1983.
56. For instance, Oakland 1992;
Berenguer Rodríguez 2000.
57. Haeberli 2002; Young-Sánchez
2004d, figs. 1.9, 2.22, 2.26.
58. See Haeberli 2008, Haeberli 2001,
and W. Isbell and Knobloch 2009 for
discussions that relate the textiles
to a provincial expression of the
Pucara style.
59. Isbell 1991; Lau 2011.
60. Menzel 1977, 60–66.
61. W. Isbell and Knobloch 2009,
fig. 27.
62. Cook 1994, pls. 50–53, fig. 20;
Cook 1983.
63. Cook 1984–85; W. Isbell and
Cook 1987.
64. Cook 2001a; Cook 1994; Cook
1984–85; W. Isbell and Cook 1987.
65. Proulx 2001; Proulx 1999.
66. Cook 2001.
67. Cummins 2002.
68. Cook and Glowacki 2003.
Patricia J. Knobloch
Figure 91 [64]. The
chamber of this Atarco
style vessel features both
a humpback animal and
a ventral animal. Vessel
with feline head; ceramic
and slip; 20.3 x 11.4 x 6.4
cm. The Metropolitan
Museum of Art, New
York, Purchase, Arthur
M. Bullowa Bequest
and Rogers Fund, 1996,
1996.290. Image: © The
Metropolitan Museum
of Art. Image source: Art
Resource, NY.
122
Archives in Clay: The Styles and
Stories of Wari Ceramic Artists
Like most New World cultures, the Wari
people never developed writing. Only a few
oral legends survived as myths when the
Spaniards arrived. They were curious about
the vast Wari capital near Ayacucho, Peru,
but never pursued stories of this lost culture.
The massive fieldstone walls of this ancient
city—its urban core covering nearly a square
mile,1 as London did when the Romans walled
it in—stood silent for centuries as looters and
stonecutters harvested its remains. At the turn
of the nineteenth century, Wari style pottery
captured the attention of scholars, who began
to resurrect these mysterious ancestors. This
pottery reflects artistic affiliations with other
ancient Andean societies from as far away as
the Peruvian north highlands, the Bolivian
altiplano (high plateau), and the desert oases
of northern Chile. Potters created images of
warriors, priests, and many high-ranking hu-
man figures, sometimes including well-attired
women, as well as cultivated plants and wild
animals (fig. 92). But overwhelmingly, these
artisans honored their deities in both the im-
agery and the function of their pottery, which
was used in celebrations of conquests and
bountiful harvests, among other things.
Creating and Recovering a Ceramic Legacy
Many Wari objects made of fragile organic
materials, including textiles, have perished
in the rainy highland climate, but well-fired
pots survived. Although tens of thousands
of Wari urbanites must have created an enor-
mous demand for cooking, eating, and stor-
age vessels, very few actual workshops with
kilns have been found. Yet the Wari capital is
located near several clay resources.2 Clay is
found in hard-packed layers and must first be
pulverized into powder that, with the addition
of water, makes a dough-like medium.3 Pot-
ters hand-formed vessel walls with flattened
slabs or rolled out coils of clay that were then
pressed into one another. The bodies of small
and at least some very large vessels were then
uniformly shaped using a tilla, which consists
of two stacked plates: the vessel rested on the
upper plate, which, like a potter’s wheel, was
rotated by hand atop the lower plate. To shape
and thin the walls of larger vessels, a stone
anvil, its form resembling a cookie jar lid,
was held against the vessel interior to support
the clay wall while the exterior surface was
pounded, either with a hand or perhaps a pad-
dle. Surfaces were smoothed with stones and
shards of ceramic with rounded, worn edges.
Potters also used molds, most often to produce
small figurines or the three-dimensional faces
that often appear on the necks of jars.4
After vessel surfaces were prepared
through polishing, they were decorated with
pigments that were usually mineral based and
applied with brushes of hair or plant fibers.
A common practice was to outline colorful
designs with fine black lines. Then polish-
ing pebbles were again vigorously rubbed
over the vessel’s surface to improve bonding
of the pigment, provide a lustrous finish, and
strengthen the surface by aligning the clay
molecules. Finally, the vessel was fired, which
fused the flat crystalline clay molecules into
strong laminated layers.5 On the base, some
vessels were marked with small, geometric
“potters marks.”6 This term seems to be a mis-
nomer, however, as such marks are too rare
and too simple to correspond to a potentially
large population of potters; the marks more
likely record the vessel’s use at a special event.
Perhaps because of the effort invested in
their production, at least some broken vessels
Figure 92 [34]. This figure
has tupu pins at her
shoulders. Female figure;
ceramic and slip; H. 28.4
cm. American Museum of
Natural History, New York,
41.2/8596. Image: cour-
tesy American Museum
of Natural History,
Anthropology. Photo: Craig
Chesek.
were cleverly repaired by drilling holes along
fracture edges; the holes were used to lace the
pieces back together.
Wari pottery is found in graves, ceremo-
nial offerings, and dumps of shards that may
document, respectively, a person’s lifetime,
a communal religious rite, or urban garbage
disposal. The fine vessels in this catalogue
derive from the first two contexts and many
appear as though they were made yesterday.
In a sense they were since many were locked
away below the ground. Their condition also
reflects circumstances of burial. Many are
from coastal graves where sand packed easily
around vessels; when discovered, the sand
can be carefully brushed away without harm
to the vessel. In the highlands, graves have
been found beneath the floors of structures
but they are rare.7 More often, Wari’s “urban”
dead were enshrined within built enclosures
such as megalithic stone boxes, stone-lined
cists with a large capping stone, catacombs,
12 4 Patricia J. K nobloch
and large niches in temple walls.8 Such tombs
often received later, additional burials that
damaged earlier grave goods, and they were
also obvious to looters, both in antiquity and
today. A millennium later, archaeologists be-
gan to retrieve what was left.
Reconstructing the History of Wari Pottery
The ceramic landscape in which Wari pot-
ters worked was complex. Andean geography
comprises multiple valley systems in both
inhospitable deserts and mountain ranges.
The terrain allowed numerous, independently
formed cultures and pottery styles to develop;
each of the latter blended with Wari art as the
empire expanded. Dating events during Wari’s
expansion starts with determining the se-
quence in which pottery styles were made and
then identifying the styles of pottery associ-
ated with each event. The sequencing process
is based on two assumptions about stylistic
change: first, potters continually either bor-
rowed designs or innovated new ones; second,
people discarded artifacts in the same order
in which they were created, ideally in deep,
multilayered deposits. Also, analysis requires
careful attention to the details of design
changes. For example, imagine reconstructing
the history of the VW Beetle based only on
the car itself: over decades the basic style may
appear unchanged, but the radios, lights, and
tires would assign the manufacture of some
cars to Germany in the 1930s and others to
Mexico in the 1970s.
Fortunately, in 1964 Dorothy Menzel ac-
complished the daunting task of organizing
Wari ceramics and the preceding decades
of research about them.9 Her major sources
included collections assembled by the German
archaeologist Max Uhle, the first to excavate
scientifically in Peru, at Pachacamac on the
central coast;10 Julio Tello, the first Peruvian
archaeologist, who discovered Wari ceremo-
nial pottery at Pacheco on the south coast and
at Conchopata in the highlands near the Wari
capital; the North American archaeologists
Alfred Kroeber, 11 John Rowe,12 and Lawrence
Dawson, who analyzed Uhle’s collections
and added their own surface collections from
south coast and highland Ayacucho sites,
as well as Wendell Bennett, who excavated
 extensively at the Wari capital;13 and, finally, vessel inventory is noteworthy for includ-
Luis Lumbreras, a Peruvian archaeologist ing spoons16 and whose ceramic designs are
whose extensive surveys and excavations predominately black and red geometric motifs
produced the most comprehensive study of on a white background (fig. 93). During the
ancient history in the Ayacucho region.14 sixth and seventh centuries,17 Huarpa and
Menzel divided the Wari past into four time Nasca cultures were in contact, possibly as
units known as epochs, the first two of which trade partners, though they did not exchange
she subdivided into two parts. ceramics. Complete Nasca icons never occur
in Huarpa art, but Huarpa potters did selec-
Wari Precursors tively adopt small, colorful Nasca designs
Important among coastal forerunners during edged with fine black outlining; among them
the Early Intermediate Period (AD 1–600) were are symmetrical medallions with a circled-dot
Figure 93. Huarpa style the north coast Moche and the south coast center and four appended elements shaped
tumbler with spirals that
Nasca, who had existed for centuries before like fleur-de-lis.18 By the end of the seventh
may have been inspired
by south coast Nasca style the Wari; late manifestations of their cultures century, several Huarpa communities had
pottery. Cup; ceramic and overlapped with Wari during the early years grown in density, especially on the mesa
slip; 17.1 x 13.3 cm. Fine of the Middle Horizon. Moche potters excelled where Wari’s capital would soon develop.
Arts Museums of San
Francisco, Gift of Bob and in realistic three-dimensional modeling and
Mary Maarshall, 1986.70.1. bichrome painting depicting many different middle horizon epoch 1a (600–700).19 At
themes; Moche ceramic art is the likely source the Wari capital, pottery shards have been
of Wari modeling techniques. Nasca potters excavated layer by layer to a depth of 4 meters
concentrated on two-dimensional representa- (13 feet) and each layer provides a successive
tions that are often more stylized and exe­ snapshot of artistic changes.20 As shards of
cuted in a range of clear, vibrant earth tones; Huarpa ceramics diminished, new pottery
modeling in Nasca ceramic art is limited and styles known as Ocros and Chakipampa ap-
includes effigy jars portraying humans with peared. Although these styles shared many
heads at the jar’s neck. Nasca ceramics were a motifs, a bright yellow-orange background
major source of inspiration for the Wari, who pigment and minor modeling distinguish
adopted Nasca pigmentation techniques along the Ocros style (fig. 94). With darker orange
with their habit of surrounding motifs with or white backgrounds, Chakipampa potters
fine black outlines. experimented with curvilinear motifs such
In the highland Ayacucho region was as wavy bands with curled tips known as
the contemporary and relatively less com- “recurved rays” to create elaborate octopus-
Figure 94. Undulating plex Huarpa culture (AD 1–650/700),15 whose like designs (fig. 95), and a banded rectangle
bands with recurved rays
appear on one side of this
Ocros style (Epoch 1A)
bowl from Conchopata.
On the other side are a
modeled face and painted
curving limbs. Universidad
Nacional de San Cristóbal
de Huamanga, Laboratorio
de Arqueología. Photo:
Patricia J. Knobloch.
Figure 95. This
Chakipampa style (Epoch
1A) bowl from Conchopata
has an octopus-like motif of
recurved rays. The interior
color is like Ocros style
pigmentation. Universidad
Nacional de San Cristóbal
de Huamanga, Laboratorio
de Arqueología. Photo:
Patricia J. Knobloch.

125 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists

 became standard on the interiors of straight-
sided, open bowls. Effigy jars of figures wear-
ing a headband of multicolored chevrons and
a version of the banded rectangle on the cheek
first appear at this time. The chevron band,
the meaning of which is unknown, became
the most diagnostic trait of Wari ceramic art;
the banded rectangle, wherever it occurs, may
refer to ethnic identity.
middle horizon epoch 1b (700–850).21 At
this time Chakipampa potters dramatically
and innovatively began to create a veritable
bestiary of zoomorphic symbols. They include
a lizard-like animal shown in profile with
horns and a zigzag band across its back;22 an
insect-like “ventral animal,” which rests on its
belly (ventral side) and has four or more limbs,
a triangular tail, and sometimes a pointed
“stinger” nose (fig. 96); 23 a single, S-shaped
band with a profile feline head at each end
that is an antecedent to a feline-like creature
with arching back and human-like hands
and feet known as the “humpback animal”
(fig. 97);24 and many more. Most notable is
the “Ayacucho serpent,” a legless, centipede-
like creature with a multilobed body, two
eyes, open mouth, and whiskers (fig. 98). This
imaginative imagery may represent an initial
conception of mythical entities and, perhaps,

Figure 96. Provincial

Chakipampa style (Epoch
1A) vessel with ventral
animal, from the south
coast. Phoebe A. Hearst
Museum of Anthropology,
University of California,
Berkeley, 4-9016.
Figure 97 [66]. This ves-
sel belongs to the Atarco
rather than Chakipampa
style. Vessel with hump-
back animal; ceramic and
slip; 26.5 x 21.6 x 11.4 cm.
Brooklyn Museum, New
York, Henry L. Batterman
Fund, 41.420.

126 Patricia J. K nobloch

Figure 98. Ayacucho document the initial Wari occupation at Cerro
serpent motif on a
Baúl’s mesa redoubt (see pp. 65–81, “The Wari
Chakipampa style (Epoch
1B) bowl from Concho­pata. Built Environment”). At this crucial time,
Photo: courtesy William H. when the Wari were on the road, one potter
Isbell. immortalized the grand visitation of a dig-
nitary carried on a litter by four porters (fig.
101). The potter painted the Chakipampa style
chevron band and medallion motif of re-
curved rays with a central dot. The absence of
weaponry indicates a peaceful procession.
During these forays into the south, the
a localized, autochthonous Wari cult based on Wari encountered a new cult of anthropomor-
a two-dimensional interpretation of the layout phic and zoomorphic deities depicted artisti-
of the constellations. Simultaneously, the Wari cally either as bodiless heads or as full-bodied
began to expand their presence and author- and holding staffs. The full-bodied figures are
ity by founding extensive settlements in the of two varieties: frontally posed staff deities
Ayacucho and Cuzco valleys. Human effigy who hold a staff in each hand, and supernatu-
jars in these distant areas depict figures with ral beings depicted in profile with a single
chevron headbands, cheek motifs, body deco- staff. This cult may have developed from a
ration of bicolored “octopus” motifs, and styl- society of shamans who lived at the northern
ized, rectangular hands (fig. 99).25 Throughout Chilean desert oases of San Pedro de Atacama.
the empire, these human images may depict There, hundreds of mummies were buried
a single individual or denote a more general with paraphernalia used to inhale hallucino-
ethnic identity. genic plant-based snuff; the paraphernalia,
Wari’s relationship with the central coast including snuff trays and tubes, is decorated
was less intrusive. The Nievería people, like with the frontal or a profile staff-bearing figure
their Moche neighbors to the north, excelled (see fig. 88). It is possible that states of ecstatic
in three-dimensional modeling, especially
effigy jars and figurine elements appliquéd to
vessels.26 But the Nievería pottery style also
features a distinctive, bright orange back-
ground color similar to Ocros pottery from Ay-
acucho along with renditions of the Ayacucho
serpent motif (fig. 100) and a ventral animal,
Figure 99. Chakipampa both seen in Wari’s Chakipampa style.27 Thus,
style (Epoch 1B) faceneck
the Nievería style most likely reflects contact
vessel with headband of
multicolored chevrons and with the Moche and their ceramic art prior to
banded rectangle on the the Middle Horizon; it seems to have existed
cheeks, from Conchopata. throughout Epoch 1 and to have taken inspi-
Photo: courtesy William H.
Isbell. ration from Wari styles as contact with the
highlanders set the stage for later Wari expan-
Figure 100. Nievería style
vessel with Ayacucho sion in this region.
serpent motif; ceramic and Wari leaders apparently focused on a
pigment; H. 17.1 cm, W. southward campaign with a settlement at
14.6 cm. The Art Institute
of Chicago, Kate S. Pacheco on the south coast (see maps, pp. xiv,
Buckingham Endowment, xv), where a provincial Chakipampa style in-
1955.2236. Photo: © The troduced the mythic animal icons, including
Art Institute of Chicago.
the humpback animal, the Ayacucho serpent,
and the ventral animal. Moving farther south
to the Moquegua Valley, small amounts of
Ocros and Chakipampa style pottery also

127 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists

 intoxication informed or even engendered
the iconography,28 which most likely spread
to far-flung areas by means of highly portable
textiles that were decorated with it.29 Its ap-
pearance in the Wari heartland, at the site of
Conchopata, is documented on twenty-three
large effigy jars of Wari elites, all wearing a
chevron headband, waist-length tunic, and
various kinds of facial decoration.30 The jars’
barrel-shaped bodies were used like billboards
to advertise the new cult: they are painted
with a frontal staff deity flanked by rows of
winged staff-bearing creatures in profile who
here serve as attendants (see figs. 75a–f). The
layout is similar to Tiwanaku’s Gateway of the
Sun (see fig. 6a), and the Wari staff deity is
Figure 101 [40]. Nievería
style? vessel with litter
group; ceramic and slip;
28 x 16 x 14 cm. The
Cleveland Museum of Art,
Leonard C. Hanna Jr. Fund
2011.36.
128 Patricia J. K nobloch
almost identical to one carved on a Tiwanaku
stone statue.31 On shards of similar Wari effigy
jars, the profile staff-bearing beings closely
resemble carved images on the San Pedro de
Atacama snuff paraphernalia.32 The shoulder
area of the twenty-three effigy jars was painted
with large and small versions of the humpback
animal, the latter similar to those on Nievería
style pottery and textiles. Remarkably, the
larger versions are more like those depicted on
Aguada style pottery of northwest Argentina.
These are just a few of the indications that
staff deity imagery developed in the context
of wide-ranging interactions; the Wari seem
to have been in touch with and open to many
different cultures and sources as the iconogra-
 phy, and the religion it represented, developed
into one of the Middle Horizon’s most impor-
tant signatures.
At Pacheco, the cult is depicted on ceremo-
nial pottery in the Robles Moqo style, another
of Wari’s many ceramic styles.33 These vessels
feature effigy jars depicting empty-handed
elites; modeled animals, especially camelids;
human hands and feet; and open, urn-like
vessels or huge tumblers that are painted with
plants, deity faces, and staff deities (see figs.
130–40). One spectacular effigy jar depicts a
Wari figure whose elite status is apparent not
so much in his simple, striped tunic as in the
beautifully depicted black jaguar pelt hanging
from his hat (see fig. 134). Jaguar habitation
once included all of South America, except
the western coast and southernmost regions.
Black jaguars are a rare, morphed peculiar-
ity; they represent only 6 percent of today’s
jaguar population.34 Thus, this Wari personage
may be someone of great distinction. Several
other Pacheco effigy jars show a figure with
an unusual complex facial design who appar-
ently wears a tie-dyed tunic (see figs. 135, 136;
see also pp. 193–205, “Tie-dyed Tunics”). This
human image occurs on more artifacts than
any other Wari personage and is here termed
the “paramount warrior” since, as shall be
seen, he may represent a heroic warrior or an
ethnic group who took up arms.35 The remain-
Figure 102. Composite ing vessels are crucial in understanding Wari’s
reconstruction drawing new religious advances. The staff deities are
based on urn fragments in
the most elaborate icons thus far of the new
the Conchopata style, from
Conchopata. Drawing: cult (see figs. 1, 5). The large urns and tum-
Patricia J. Knobloch. blers that bear these images could have held
129 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists
the chicha (native corn beer) that was served
to participants during feasting events (see fig.
133; see also pp. 82–101, “The Art of Feast-
ing”). Figurine cups, which seem awkward for
drinking, may have been used to make liba-
tions.36 The distinctive shapes of camelid ef-
figy vessels also may suggest ritual rather than
more routine use (see figs. 137, 138).
The cult continued to develop at Con-
chopata in the Wari heartland, but apparently
not as peacefully. Urns not unlike those from
Pacheco are variously painted with staff-bear-
ing figures, bodiless supernatural heads—in
profile or full-faced—as well as warriors and
captives.37 In the example illustrated in Figure
102, a human, perhaps a guardian, stands next
to a staff deity; a supernatural companion in
profile carries an axe and a staff tipped with
a human captive. The human also holds an
axe and is attired in a tasseled, four-cornered
hat, a necklace, and a belted, wrap-around
loincloth—all indicative of his elite status.
(The belt and the vertical registers between
figures are decorated with the face-fret design
commonly seen in tapestry-woven tunics [see
fig. 144]). The staff deity may represent a sun
god; its head is suggestively surrounded by a
corona of rays and the guardian holds a small,
circular object that might represent a mirror
used to reflect the sun’s rays. Such mirrors
are known archaeologically. Dangling from
the bottom of one of the deity’s two staffs is a
heart-and-lung motif that probably alludes to
the ritual killing of captives.
Interestingly, the composition is not like
those on the large Wari faceneck jars or the
Figure 103 [5]. Partial
recon­struction of an urn
from Conchopata in the
Conchopata style depicting
warriors carrying shields
and an axe or a bow and
arrows. Urn fragments
with warriors; ceramic
and slip; 40 x 85 cm.
Museo Histórico Regional
“Hipólito Unanue,”
Ayacucho, MHRA-1777.
Photo: Daniel Antonio
Giannoni Succar.

Tiwanaku Gateway of the Sun, on which many
profile staff-bearing figures obediently face
a central, staff deity. Rather, the figures are
represented in repeated pairs and perhaps in
opposition since the profile being turns away
from its frontally posed companion. It may be
that this layout, with each figure type holding
different human captives, records a critical
disruption within Wari’s cult that involved
Figure 104. Profile heads warfare among groups identified with either
on urn fragments in the the staff deity or the profile being.38
Conchopata style, from
Conchopata. Photos:
Further evidence of warfare exists in
William H. Isbell. representations of the paramount warrior,
who appears without weaponry in his in-
carnation at Pacheco on the south coast but,
at Conchopata, carries a shield and a bow
with arrows as he kneels in a reed boat (fig.
103).39 On another Conchopata urn, seven
profile heads appear; their lolling tongues may
indicate strangulation and their distinctive
headgear and facial ornamentation are shown
in some detail (fig. 104). One has a bicolored
face and wears a black cap with a brim and
perhaps a chin strap of square plaques, each
marked with two dots; another wears an ar-
row point tied to his headdress and bangles in
his pierced ears, nose, and lower lip. Although
interpretation of these heads varies, they may
represent enemies (or enemy groups) of the
Wari state who were tortured into submission
and thus offer further evidence of conflict as
an embattled society defended its new territo-
ry or sought to aggressively expand. Certainly
military imagery is common at Conchopata
and elsewhere in Wari territory during this
period, and, as Menzel observed, there is a
dramatic change in Wari’s local pottery after
the cult had been established, as though some
crisis—perhaps a revolt or epidemic—severely
disrupted politics in the Ayacucho Valley and
on the south coast.40

130 Patricia J. K nobloch

 middle horizon epoch 2 (850–1000). The
diversity of exquisite Epoch 2 artifacts in this
catalogue is a testament to the Wari invest-
ment in fine craftsmanship and demand for
affluent attire. The increase in cups, bowls,
and jars also attests the possible transition
from an exclusive cult functioning with a
few large urns to a full-blown religious move-
ment with obligatory feasting and ceremo-
nies of oblation.41 Also apparent is a marked
increase in the representation of zoomorphic
supernatural beings, including profile, staff-
bearing creatures with avian or feline heads.
Their depiction on Wari tapestry-woven tunics
and accessories may have symbolized the
wearer’s god-like authority and allegiance to
the empire (see pp. 159–91, “Tapestry-woven
Tunics,” and pp. 217–31, “Inlaid and Metal
Ornaments”), yet pottery was the most acces-
sible and expedient means of distributing the
symbols of this authority. A new generation of
Wari potters began to stylize and enhance the
Figure 105 [1]. A bowl in
the Viñaque style’s geo-
metric-on-light technique,
from Conchopata. Bowl
with bird-headed staff-
bearing creature in profile;
ceramic and slip; 12.7 x
33.4 cm. Museo Histórico
Regional “Hipólito
Unanue,” Ayacucho,
MHRA-925. Photo: Daniel
Antonio Giannoni Succar.
131 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists
cult imagery. The finest pottery comes from the
Viñaque style in the Wari heartland,42 the At-
arco style on the south coast,43 and the Pachac-
amac style on the central coast.44
Trademark Viñaque vessels include small
lyre cups, so-called because their shape resem-
bles Old World lyre harps with sides that curve
in slightly, as well as tall glass-shaped tum-
blers with slightly flared sides.45 The lyre cups
are painted with supernatural heads, leaving
little doubt about their ceremonial purpose.
One lyre cup found in an elite Moche grave
suggests that the Moche held Wari’s beliefs and
artifacts in high esteem (see fig. 37; see also pp.
47–61, “Looking at the Wari Empire from the
Outside In”).46 A profile staff bearer with an
avian head and wing on its back was painted
on the interior of a beautiful, open bowl in
the Viñaque geometric-on-light technique (fig.
105).47 Pachacamac potters shared this im-
age and developed a signature icon named
the “Pachacamac griffin,” which has an avian
Figure 106 [61].
Pachacamac style ves-
sel with bird-headed
creature (“Pachacamac
griffin”); ceramic and
slip; 26.8 x 20.5 x 20.9 cm.
Ethnologisches Museum,
Berlin, VA 19059. Image:
bpk, Berlin/Ethnologisches
Museum/Art Resource, NY.
Photo: Claudia Obrocki.
Figure 107 [58]. Pachaca­
mac style double-spouted
vessel with bird-headed
creature (“Pachacamac
griffin”); ceramic and slip;
16 x 16.5 x 14.5 cm. Fowler
Museum at UCLA, Los
Angeles, Gift of Mr. and
Mrs. Herbert L. Lucas Jr.,
X86-3702. Photo: Don Cole.
132 Patricia J. K nobloch
head and sometimes a staff and a feline body,
probably that of the jaguar (figs. 106, 107).48
(Of the eleven South American wild cats,
only the jaguar’s pelt has broken circlets with
central spots.49) A similar icon on a Viñaque
bowl shows a less elaborate and perhaps early
version of this mythical griffin (fig. 108). The
ventral animal version with a stinger nose
was also painted on a Viñaque geometric-on-
light jar with an interesting hourglass shape
(fig. 109).
Representations of the paramount warrior
continue into this period, and he is shown
wearing either a tie-dyed or a tapestry-woven
tunic with face-fret motifs along with an elite,
four-cornered hat (fig. 110; see also fig. 146). In
these effigies, the hands cup to form holes into
which other objects, perhaps made of perish-
able materials, may have been inserted—staffs
of authority, weapons, or agricultural tools are
among the possibilities.50 Curiously, unlike the
vast majority of Wari effigies in ceramic, both
objects are open at both top and bottom, and
thus did not serve as vessels; their function is
unknown. The figure with bangles in his ears,
nose, and lip, seen earlier at Conchopata, may
be memorialized in a Pachacamac double-
spout bottle (compare figs. 104 and 111) and in
another Moche burial as an effigy head jar.51
That jar’s association with Cajamarca style
pottery from the north highlands may suggest
that this personage is somehow tied to that
region, especially since both his facial pierc-
ings and Cajamarca ceramic techniques are
The animal on the bowl is
related to the Pachacamac
griffin; the creature on the
hourglass-shape vessel is
an Epoch 2B example of
the ventral animal. Both
vessels are in the Viñaque
style and come from a
grave at Wari Wilka, a site
in the highlands (Flores
Espinoza 1959).
Figure 108 [21]. Bowl
with mythical creature;
ceramic and slip; 8 x 12.6
cm. Museo Arqueología y
Antropología, Universidad
Nacional Mayor de San
Marcos, Lima, 3001 0041.
Photo: Daniel Antonio
Giannoni Succar.
Figure 109 [22]. Vessel with
ventral animal; ceramic
and slip; about 14 x 6 cm.
Museo Arqueología y
Antropología, Universidad
Nacional Mayor de San
Marcos, Lima, 3001 3453.
Photo: Daniel Antonio
Giannoni Succar.

Figure 110 [37]. Viñaque

style figure in tie-dyed
tunic and four-cornered
hat; ceramic and slip; 30 x
22 cm. Museum Rietberg,
Zurich, RPB 320. Photo:
Rainer Wolfsberger.

133 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists

Figure 111 [52].
Pachacamac style
double-spouted head
vessel; ceramic and slip;
15.8 x 15.3 cm. Museo
Nacional de Arqueología,
Antropología e Historia,
Lima, C 54789. Photo:
Daniel Antonio Giannoni
Succar.
dramatically different from Wari customs. The
figure with the bicolored face and a headdress
of square plaques also seems to be depicted
in various contexts, including a fine tapestry-
woven panel (fig. 112).52 A clue to interpreting
the plaques comes from recently discovered
burials at Espíritu Pampa; this figure’s image
occurs on ceramics recovered from these buri-
als, and one elite tomb also contained more
than one hundred square silver “sequins,”
each with two holes used to stitch the plaques
to a cloth.53 Remarkably, this elite burial also
included a U-shaped silver pectoral with
feline heads almost identical to one now in a
German collection (see fig. 219b). The latter
pectoral is accompanied by two nearly identi-
cal smaller versions that might be interpreted
as staff finials since they are similar to the
unusual U-shaped finials on staffs held by the
tapestry’s figures.
On the south coast, most Atarco style
pottery comes from elite burials that reveal
13 4 Patricia J. K nobloch
the practice of accompanying the dead with
prized items. Among those objects are inno-
vative bottles with two thin, tapering spouts
connected by an arching, solid bridge handle
(double-spouted bottles); liquids poured easily
from one spout while air flowed into the other
(fig. 113). One of the most impressive and rare
examples of Atarco ceramic art is a drum, its
animal-skin membrane still intact, with the
head of one of the mythical felines that were
the focus of Atarco’s religious imagery (figs.
114, 115; see also fig. 65). (Ceramic drums
were traditional musical instruments among
south coast cultures, originating with the
Paracas culture [900–200 BC] and continu-
ing with the Nasca.54) Feline images had also
been part of the earlier south coast Robles
Moqo tradition; for example, the impressive
human head shown in an effigy cup wears a
feline headdress that combines both artistic
traditions (fig. 116). The feline may represent
a jaguar, and the human head could be that of
Figure 112. Tapestry-
woven panel; camelid
fiber and cotton; 70.5 x 117
cm. Museum of Fine Arts,
Boston, Charles Potter
Kling Fund, 1996.50.

Figure 113 [53]. Atarco

style double-spouted skull
vessel; ceramic and slip;
14.5 x 17.8 x 12.6. The
Cleveland Museum of Art,
John L. Severance Fund
1996.292.

135 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists

Figure 114 [31]. Atarco
style drum; ceramic, slip,
cotton, and animal hide;
45 x 21 cm. Staatliches
Museum für Völkerkunde,
Munich, 89-311 922. Photo:
Staatliches Museum für
Völkerkunde, Munich.
Photo: Marietta Weidner.

Figure 115 [42]. Atarco

style faceneck vessel
with felines; ceramic
and slip; 17.8 x 12.7 x 12.7
cm. I. Michael Kasser
Collection, KP 246.
Figure 116 [54]. The model-
ing of this Atarco style ves-
sel comes from the Robles
Moqo style. Head vessel;
ceramic and slip; 18 x 12
cm. Museo Nacional de
Arqueología, Antropología
e Historia, Lima, C-54786.
Photo: Daniel Antonio
Giannoni Succar.

136 Patricia J. K nobloch

 a political leader or a shaman; if the latter, the
feline was likely a source of spiritual power,
as it is today among shamans of the Amazoni-
an lowlands. Another Atarco canteen-shaped
vessel depicts a handsome, partially mod-
eled feline that dominates two creatures who
survived from earlier Chakipampa times and
benefited from improved artistic treatments by
Atarco artists: the humpback animal appears
on each side of the vessel and, on the curved
surface beneath the feline’s mouth, there is a
ventral animal with an open, toothed mouth
(fig. 91). Continuation of highland/coastal
contact is also represented by an intricate
double-chambered vessel with deep, stacked
receptacles that are supported by a hollow,
modeled serpent. Its south coast provenience
and unique modeling suggest that an Atarco
potter employed Viñaque designs, which in
turn were directly derived from earlier Chaki-
pampa chevron bands, recurved rays, and an
angular rectangle known as the “three-fillet
band”(fig. 117).55
Death and warfare is prominently re-
corded in Wari style ceramics with images of
skulls (fig. 113) and warriors. One effigy bottle
depicts a warrior festooned with severed hu-
man heads, legs, and arms that may refer to
the gruesome aftermath of battle and the dis-
memberment of enemies (fig. 118). On another
vessel, a fierce warrior with axe in hand is
flanked by two profile heads whose identity is
unclear; they may represent his captives or his
allies or patrons (fig. 119).
Andean beliefs are also strongly under-
pinned by dualistic thought that manifests in
paired artifacts, such as two effigy jars that
may represent women, who only rarely appear
in Wari art (fig. 120). Although they lack the
tupus (shawl pins) normally shown on women,
the protuberances on their chests may refer to
breasts, beneath which a vagina-like motif ap-
pears. Hair braids are indicated with incisions
in the ceramic, and their bodies, or perhaps
garments, are ornamented with humpback
animals, the ventral animal, and birds. If
women, they may embody a generalized con-
cept of earth mother, although they lack overt
supernatural features that might identify them
with the female earth deity of later times,
Pachamama. Fertility is also emphasized in
other ways. For example, another Atarco effigy
jar shows an elite male holding a staff topped
with an ear of maize (see fig. 237), and a Vi-

Figure 117 [24]. This cup

reportedly was buried
with Atarco style vessels.
It could be of sierra origin
or locally rendered in the
Viñaque style. Cup on ser-
pent pedestal; ceramic and
slip; 14.6 x 7.9 cm. Denver
Art Museum Collection,
Gift of Olive Bigelow by
exchange, 1996.37. Photo: ©
Denver Art Museum 2012.
Figure 118 [39]. Viñaque
style warrior vessel;
ceramic and slip; 15 x 9.5
x 7 cm. Linden-Museum,
Stuttgart, 119016. Image: ©
Linden-Museum Stuttgart.
Photo: A. Dreyer.

137 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists

Figures 119a, 119b [68].
Atarco style vessel with
warrior (front and back);
ceramic and slip; 31 x
39.4 cm. Museo Regional
de Ica “Adolfo Bermúdez
Jenkins,” MRI-00178-01.
Photo: Daniel Antonio
Giannoni Succar.

138 Patricia J. K nobloch

Figure 120 [38]. These
Viñaque style vessels may
depict females. Two figural
vessels; ceramic and slip;
19.2 x 10.5 x 10.3 cm (each).
Dallas Museum of Art,
1976.W.216, 1976.W.217.
Figure 121. Viñaque style
tumbler with plants,
alpacas, and felines, from
Maymi. Photo: Martha
Anders.
Figure 122 [25]. Pachaca­
mac style cup with
staff deity; ceramic and
slip; 12.1 x 8 x 7.8 cm.
Ethnologisches Museum,
Berlin, VA 19174. Image:
bpk, Berlin/Ethnologisches
Museum/Art Resource, NY.
Photo: Martin Franken.
139 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists
ñaque tumbler delicately depicts in paint and
low relief an amazing “garden” of plants along
with alpacas and felines (fig. 121).56
Among the most impressive of Wari style
ceramic accomplishments are the very large
effigy vessels found at Corral Redondo on the
far south coast; these vessels are said to have
contained nearly one hundred feathered pan-
els (see fig. 198; see also pp. 207–15, “Feather-
work”). Thus, it cannot be assumed automati-
cally that all large vessels were used as chicha
containers. The Inca are also known to have
stored clothing in large jars.57
At Pachacamac on the central coast,
priests hosted massive feasts to meet ritual
demands. Potters complied by producing
tumblers with stylized, sometimes almost
cartoonish versions of the staff deity (fig.
122), whose disembodied head also appears
on jars (see fig. 16). Many small human effigy
vessels in the Pachacamac style may depict
the participants in these ceremonies (see fig.
67); they have individual facial features. The
designs painted on their bodies could refer to
fine textiles, although the characteristic verti-
Figure 123 [28]. One of the
chambers of this Pachaca­
mac style vessel is modeled
as a human who plays a
fox-head whistle and may
wear a tie-dyed tunic.
Double-chambered vessel
with human; ceramic and
slip; 14.9 x 9 x 20.7 cm.
Ethnologisches Museum,
Berlin, VA 49699. Image:
bpk, Berlin/Ethnologisches
Museum/Art Resource, NY.
Photo: Claudia Obrocki.
cal banding of highland Wari tapestry-woven
tunics is absent. The numerous examples of
these tumblers and effigy jars in museum col-
lections suggest that Pachacamac was a center
of religious power that may have rivaled the
Wari capital itself. Strikingly, warriors are
lacking in Pachacamac imagery, perhaps in ac-
cord with the emphasis on religious imagery;
this may indicate that the priests who ruled
Pachacamac maintained a degree of neutrality.
Among the many other types of vessels in the
Pachacamac style are musicians, one of whom
wears a tie-dyed tunic and plays a fox-head
whistle (fig. 123), and many representations
of marine creatures that may relate to Pacha-
camac’s location on the shores of the Pacific
Ocean: fish (fig. 124), sea cucumbers (fig. 125
and perhaps fig. 126), which are still caught
and dried for market in Peru, and snails (fig.
127).
middle horizon epochs 3 and 4 (1000–
1050). As Wari authority declined, regional
autonomies returned.58 The capital was
abandoned—the exact causes still require
14 0 Patricia J. K nobloch
study—but its religion survived and gradually
culminated on the coast, lasting longer among
Moche descendants on the north coast. There,
Wari icons blended with the Moche tradition
of press molding, which created imagery in
low relief, usually on unpainted blackware59
or in black-and-white designs painted on red
backgrounds. Two vessels that show the per-
sistence of Wari-derived figures into this pe-
riod include one that depicts a human flanked
by birds; a mythical figure with an avian head
appears in another (figs. 128, 129).60 But in
general, stylistic inheritance from Wari was
confined to simple design elements—white
bands, wavy dashes, crosses, and circles with
or without thin black lines—on pottery from
the far south coast Camaná61 and Majes val-
leys,62 to the north coast Huarmey Valley63 as
well as at Pachacamac (see also pp. 251–67,
“Wari’s Andean Legacy”). Walled within a
Vatican-like city of pyramids, cemeteries, and
elite compounds, Pachacamac became a reli-
gious stronghold that outlasted Wari’s high-
land authority. By Inca times, it had become
an extremely important oracle site. Its author-
[47] [47]
[49] [48]
Figures 124a, 124b [47].
Two Pachacamac style dou-
ble-spouted fish vessels;
ceramic and slip; 11.4 x 5.9
x 17.2 cm and 11.7 x 6.2 x
17.4 cm. Ethnologisches
Museum, Berlin, VA19128,
VA19129. Images: bpk,
Berlin/Ethnologisches
Museum/Art Resource, NY.
Photos: Claudia Obrocki.
Figure 125 [49]. Pachaca­
mac style double-spouted
sea cucumber vessel;
ceramic and slip; 9.6 x 5
x 18 cm. Ethnologisches
Museum, Berlin, VA19130.
Image: bpk, Berlin/
Ethnologisches Museum/
Art Resource, NY. Photo:
[50] Claudia Obrocki.
141 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists
Figure 126 [48]. Pachaca­
mac style double-spouted
sea creature vessel;
ceramic and slip; 17 x 7 x
17.5 cm. Ethnologisches
Museum, Berlin, VA19127.
Image: bpk, Berlin/
Ethnologisches Museum/
Art Resource, NY. Photo:
Claudia Obrocki.
Figure 127 [50]. Pachaca­
mac style double-spouted
snail vessel; ceramic and
slip; 12.9 x 15.1 x 10 cm.
Ethnologisches Museum,
Berlin, VA19149. Image:
bpk, Berlin/Ethnologisches
Museum/Art Resource, NY.
Photo: Claudia Obrocki.
Figure 128 [59]. Double-
spouted vessel with figure
and birds; ceramic and
slip; 18.6 x 19.4 x 10.5 cm.
Fowler Museum at UCLA,
Los Angeles, Gift of Dr. and
Mrs. Robert Kuhn, X71-417.
Photo: Don Cole.
Figure 129 [57]. Double-
spouted bird-headed
creature vessel; ceramic
and slip; 16.5 x 20.6 x 11.6
cm. Museo Larco, Lima,
ML010864. Photo: Daniel
Antonio Giannoni Succar.
14 2 Patricia J. K nobloch
ity also came to an abrupt end when Hernando
Pizarro stepped foot into the Pachacamac
temple in 1533. Legend has it that, at that very
moment, an earthquake shook the city.
Invasion and Resolution
Wari ceramics spread to many different places
in the Andes during the Middle Horizon.
While we do not know to what degree this
spread reflects migrations, trade relations, or
military conquest that forced an era of unifica-
tion, all three circumstances were certainly
involved. The memory of this great civiliza-
tion may be recorded in a remarkable tale of
war and peace recorded during the Spanish
colonial period. The tale, which comes from
Huarochirí in the Rimac Valley region of the
central coast, refers to a period long before
the Inca when a deity, Paria Caca, came down
from the highlands to save the Yunca coastal
people from a cruel cannibal god. Then
Paria Caca conquered the Yunca and cre-
ated “a cultic order in which both victors and
vanquished would participate.” The Yunca
eventually accepted Paria Caca into their
cosmology by bestowing upon him the status
of “brother” to the wife of Pacha Camac, the
supreme god of the central coast.64 This tale
of invasion by a highland god and resolution
with those living on the central coast65 may
refer allegorically to the Wari, who traveled
from their heartland into coastal territories
with a new cult that priests at Pachacamac ap-
parently embraced. Perhaps Paria Caca was a
Wari god of war and unification.
notes 1. W. Isbell et al. 1991, 24, where the
figure is stated as 250 hectares.
2. Arnold 1975, 191.
3. O’Neale 1977, 43.
4. Milliken 2006, 346–47, figs.
119–21; Pozzi-Escot 1991, 87.
5. Shepard 1956. For more on Wari
ceramic production techniques, see
Anders et al. 1998 and Pozzi-Escot et
al. 1998.
6. Bennett 1953, 66.
7. W. Isbell 2001, 28–36; Ochatoma
Paravicino and Cabrera Romero
2001, 474–77.
8. W. Isbell 1997, 184–88.
9. Menzel 1977; Menzel 1968;
Menzel 1964.
10. Uhle 1903.
11. Kroeber 1944.
12. J. Rowe 1960a.
13. Bennett 1953; see also Spielvogel
1955. Bennett accumulated over
50,000 ceramic fragments from
Wari, Conchopata, and Acuchimay,
south of Ayacucho; this collection
is now housed at Yale University’s
Peabody Museum.
14. Benavides Calle 1965; Lumbreras
1974a; Lumbreras 1960.
15. Knobloch 2003; Knobloch 1983;
Knobloch 1976. Collections of
Huarpa style pottery at Wari made in
1974 by William H. Isbell, Katharina
Schreiber, and the author included
a carbon sample that returned the
date BP 1713±120, cal. AD 255–536,
at 1 sigma (Stuiver and Reimer 1993;
Stuiver et al. 2005).
16. Sizes range from large ladles to
very small spoons that were possibly
used to feed infants.
17. Leoni 2004. Excavations of a
temple at Ñawinpukyo provided
radiocarbon dates of BP 1600±70,
cal. AD 430–576 and BP 1583±34,
cal. AD 441–593, both at 1 sigma
(Stuiver and Reimer 1993; Stuiver et
al. 2005).
18. This style is known as Cruz Pata
(Lumbreras 1974b, 137–38, fig. 147,
bottom).
19. This date range is not based on
calibrated carbon dates but on an
estimate of 14C calibrations based on
earlier Huarpa calibrated dates and
later materials dating to Epoch 1B.
20. Knobloch 1983.
143 A rchives in C lay: T he S t y les and S tories of Wari C eramic A rtists
21. This date range is based on a
suite of radiocarbon dates from
Conchopata that were rounded to
the nearest 50 years and at least 85%
of the 1 sigma range (W. Isbell and
Knobloch 2009, table 2; Ketteman
2002; see also Knobloch 2002, http://
www-rohan.sdsu.edu/~bharley/
WWWWHEN.html#Conchopata).
None of the carbon samples was
directly associated with pottery and
thus should not be used to sequence
Epoch 1B ceramics at the site.
22. Menzel 1968, 57, fig. 2b-6. This
creature may be based on a Basilisk
lizard that is hunted for food
and prepared like jerky for trade
(Holmberg 1957; Knobloch 1983,
296–98).
23. Menzel originally christened
this creature the “ventrally
extended animal,” a description
meant to imply that the extended
body is shown with belly (ventral
side) down and back (dorsal side)
exposed (Menzel 1964, 11; personal
communication, 2012). For an
interpretation of this icon, see
Knobloch 2002, http://www-rohan.
sdsu.edu/~bharley/WWWINSECTS.
html#KissingBug.
24. Knobloch 1983, pl. 58b; Menzel
1968, fig. 35.
25. Glowacki and McEwan 2002, fig.
14.
26. Castillo Butters 2001b, fig. 17.
27. Gayton 1927, pl. 97, fig. b.
28. Torres and Repke 2006. These
figures have animal heads and often
grasp a decapitated human head in
one hand, as though human sacri-
fice was part of the cult’s rituals.
However, shamans are described as
those who replace their heads with
those of animal spirits to partake of
supernatural knowledge. Thus, these
figures may represent shamans.
29. Bennett (1953, 117) suggested
that textiles were the optimal
medium for the diffusion of imagery,
reasoning that “ceramics were too
fragile to be taken on long journeys”
(Spielvogel 1955, 254n21). Thus,
potters may have shared concepts of
the motifs but developed different
stylistic expressions; this may help
to explain the difference in Middle
Horizon ceramics in Peru and
Bolivia (Spielvogel 1955, 9).
30. Cook 1984–85, figs. 1–4, 15–20;
Knobloch 2010.
31. W. Isbell and Knobloch 2009,
fig. 33.
32. Ibid., fig. 27.
33. Menzel 1964, 24.
34. See “Colour morphism” at http://
en.wikipedia.org/wiki/Jaguar.
35. My initial research of Wari
human representations began with
this image, formerly individual “A”
(Knobloch 1993), now “Agent 100” in
an online database (Knobloch 2002).
36. Menzel 1977, figs. 125, 135.
37. W. Isbell and Knobloch 2009,
figs. 2, 3, 25, 30, 31, 34; Ochatoma
Paravicino and Cabrera Romero
2001, figs. 7, 8, 10a–c.
38. Knobloch 2010.
39. Ochatoma Paravicino and
Cabrera Romero 2001, figs. 9, 10.
40. Menzel 1964, 69.
41. Menzel (1964) suggested that the
Wari cult became ecumenical.
42. Menzel 1964, 38–46.
43. Ibid., 46–53.
44. Ibid., 55–61.
45. These shapes avoid the need for
handles and are improvements over
straight-sided cups that can slip
one’s grasp.
46. Castillo Butters 2001b, fig. 15,
upper right.
47. Milliken 2006, 180–81, fig. 47.
48. Menzel 1964, 59–61.
49. See http://www.wotcat.com/
wildlife/Mammal/South%20
America.html.
50. Bergh 2004.
51. Castillo Butters et al. 2008a, fig.
47.
52. For example, Castillo Butters
2001b, fig. 15, bottom right.
53. Fonseca et al. 2011. Except for
some human teeth, all organic
material had rotted away, leaving
the sequins in a pile.
54. Proulx 2006, 2, 120–21.
55. Menzel 1964, 50n299. A fillet
is a band and here one sees two
border bands and an interior band
of dashes.
56. Anders 1990, 34.
57. J. Rowe 1946, 224.
58. Menzel 1964, 62–64, 73.
59. Menzel 1977, 32–33.
60. Donnan 1992, 86–87.
61. Owen 2010, 68, fig. 4.5, lower
right.
62. Owen 2007, fig. 11, top right; fig.
12, bottom; fig. 16, left.
63. Prümers 2001, fig. 8.
64. Salomon 1991, 6.
65. Ibid., 8–9.
Mary Glowacki
Figure 130a [15]. Urn
with plants, from Pacheco
(front; for side see fig.
130b, p. 147); ceramic and
slip; 56 x 86 cm. Museo
Nacional de Arqueología,
Antropología e Historia,
Lima, C-54798. Photo:
Daniel Antonio Giannoni
Succar.
Figure 131. A pottery smash
discovered at Conchopata.
Photo: William H. Isbell.
145
Shattered Ceramics and Offerings
Among early societies offerings served to
keep the world in balance as a way of pay-
ing back what had been granted by the gods
and other supernatural forces. In the ancient
Andes, offerings were seen as revitalizing
forces, with animal and human sacrifices be-
ing the most powerful, the latter reserved for
extreme circumstances. Early Andean peoples
also believed that objects were animated and
could substitute for living sacrifices.1 Among
the Wari, pottery vessels were the offering of
choice, typically used with libations, ritually
smashed, and then buried in the ground (fig.
131).
The ceremonial practice of smashing
ceramic vessels is much broader than Andean
Wari culture. Societies throughout time and
across continents have associated pottery ves-
sels and the human sustenance served in them
with offerings to the ancestors and the energy
of the afterlife. A third millennium BC soci-
ety of northern China, the Neolithic Minoan
Greeks, the ancient Mimbres culture of the
North American Southwest, and ancient Mis-
sissippian peoples of the southeastern United
States are but a few examples.2 What is espe-
cially interesting about these many and varied
ceramic offering traditions is that they require
the ritual “killing” of the pottery vessels,
breaking them to engage symbolically in an
experience with the supernatural realm. In the
specific case of the Wari, these ritual practices
seem to have been closely tied to surviving a
prolonged drought and beseeching the ances-
tors for assistance in this effort.
Our knowledge of Wari ritual behavior is
the result of many archaeological excavations
in the Wari heartland and its provinces. Start-
ing with the Wari capital city in Ayacucho,
archaeologists have documented large-scale
examples of this practice. Since the 1940s, two
offerings of numerous large vessels have been
found at Conchopata, near the Wari capital.3
The first contained urns, the second faceneck
jars (see figs. 75a–f); all were smashed and
buried in association with the interment of
five young women. Both types of vessels were
decorated with supernatural images, a number
of which represent the frontal staff deity, the
most important supernatural being that the
Wari depicted artistically (see pp. 103–21, “The
Coming of the Staff Deity”). In the case of the
urns, the exteriors bear this image while the
jars represent humans wearing tunics deco-
rated with staff deity iconography. (The jars are
known as facenecks because a human visage
appears on the neck of the vessel; the human’s
body is synonymous with that of the vessel.)
Of ancient origin, the frontal deity image dates
back as far as the Chavín culture of the Early
Horizon (1000 BC–AD 1) and can be traced
forward through time and different Andean
cultural traditions to as late as the supreme
deity of the Inca, Viracocha.4 Among the other
kinds of imagery painted on shattered ceram-
ics recovered at Conchopata are the disembod-
Figures 132a, 132b [16].
Urn with staff deities, from
Pacheco (front and side);
ceramic and slip; 83.5 x
86 cm. Museo Nacional de
Arqueología, Antropología
e Historia, Lima, S/C.
Figures 132c, 132d. Staff
deities on the interior of
the urn shown in Figures
132a and 132b.

ied heads of mythical creatures, perhaps the
staff deity’s attendants (see fig. 62).
Wari artists appear to have manufactured
a large portion of the Conchopata vessels ex-
pressly for the purpose of an important ritual
because they show no signs of long, continu-
ous use. Archaeologists have documented a
ceramic workshop at Conchopata, which likely
was reserved for the production of the pottery
offered at the site. The investigators posit that
the ritual activity associated with these ves-
sels involved the consumption of chicha (na-
tive corn beer), still drunk during ceremonies
today in traditional Andean communities. Af-
ter the presiding Wari leaders concluded their
ceremony with toasts,5 they smashed the ves-
sels and interred them with the young women
who also were sacrificed as part of this solemn
occasion. Excavations further revealed that a
channel ran from the tomb to the surface, and
chicha may have been poured into this chan-
nel at a later time to renew the offering.6
The ceramic offerings recorded at Con-
chopata are grand, but not unique. Others
involving large vessels are known from pro-
vincial sites, two located in the Nasca region
in southern coastal Peru. The first is Pacheco,
discovered in 1927 by Julio C. Tello and so far
the largest of the Wari offering sites: Tello re-
covered more than three tons of ceramic frag-
ments there, although the exact circumstances
of the find are unknown because he did not

14 6 M ary G lowacki
Figure 130b [15]. Urn with
plants, from Pacheco (side;
for front, see fig. 130a, p.
144); ceramic and slip; 56 x
86 cm. Museo Nacional de
Arqueología, Antropología
e Historia, Lima, C-54798.
Photo: Daniel Antonio
Giannoni Succar.

Figure 133 [11]. Cup with

supernatural head and
plants, from Pacheco;
ceramic and slip; about
59 x 55 cm. Museo de
América, Madrid, 8.315bis.

publish his notes about it and the site was
later used for agriculture.7 Nevertheless, some
vessels have been reconstructed and the major
forms include large urns similar to those from
Conchopata (figs. 130a, 130b, 132a, 132b; for
130a, see p. 144) along with huge cups or tum-
blers known as keros (fig. 133), faceneck jars
(figs. 134–36), large and small camelid (llama
or alpaca) effigies, their sex often clearly in-
dicated (figs. 137, 138), severed or skeletal ca-
melid heads (figs. 139, 140), and a few others.
Several of the urns depict two versions of the
staff deity that have been identified as male
and female based on their garments, the male
wearing a belted tunic and the female an un-
belted dress and a mantle over her shoulders
(figs. 132c, 132d; see also fig. 5b).8 Many other
urns are painted with highland agricultural
plants that also appear beneath the rims of the
large cups, which additionally bear the visage
of a supernatural being (figs. 130a, 130b, 133).
Based on this plant iconography, it is tempting
to think that Wari leaders facilitated the intro-
duction of certain crops to other regions, the
Nasca peoples being recipients. Many figures
depicted in Wari art, in turn, are derived from
Nasca art,9 suggesting that the Wari likewise
were beneficiaries of this relationship. The
plant urns display representations of archi-
tectural structures with dome-shaped roofs
that have been interpreted tentatively as either
Wari administrative buildings or niched halls,

147 S hattered C eramics and Offerings

 which are prominent in some Wari provin-
cial centers and seem to have had ceremonial
functions (see pp. 65–81, “The Wari Built
Environment”).10
Maymi, located in the Pisco Valley of
southern Peru, is the second major ceramic
Wari offering found in the south coast region.
Research at this site11 revealed an extraor-
dinary complex with one area devoted to
the production of ceramics. Close by were
a number of pits containing various kinds
of elaborate vessels, smashed and interred
as offerings. While a few vessels have been
reconstructed from the numerous excavated
Figure 134 [12]. Faceneck fragments, the collection remains largely
vessel, from Pacheco;
unanalyzed and unpublished. The recon-
ceramic and slip; 45.6
x 30.8 cm. Museo structed vessels include urns not unlike those
Nacional de Arqueología, from Conchopata and Pacheco, human and
Antropología e Historia, animal effigy vessels, and lyre-shaped cups
Lima, C-63067. Photo:
Daniel Antonio Giannoni and bowls. One particularly impressive piece
Succar. is a tall drinking vessel. Its exterior is decorat-
14 8 M ary G lowacki
ed with modeled and painted images of plants,
ears of corn, and felines and camelids (see
fig. 121).12 Equally intriguing are bowls with
human figures modeled as vessel supports (fig.
141).13 Additionally, investigators recovered
a cache of unfired human figurines wrapped
in different types of cloth. Some were associ-
ated with miniature unfired ceramic cups, a
jar, Spondylus princeps (thorny oyster) shell,
and ears of corn. While scholars await further
analysis of this material, it is clear that Maymi
demonstrates the importance of the ceramic
offering tradition in the Wari effort to extend
its reach into different regions.
The Wari provincial complex on Cerro
Baúl illustrates yet another example of the
ceramic offering tradition in southern Peru.
Located in Moquegua, a short distance from
the coast, this site stands out as an exceptional
embodiment of Wari ceremonialism. The Wari
built this city on a high mesa above the dry
Figure 135 [13]. Faceneck
vessel, from Pacheco;
ceramic and slip; 50 x 35.3
cm. Museo Nacional de
Arqueología, Antropología
e Historia, Lima, C-64075.
Photo: Daniel Antonio
Giannoni Succar.
Figure 136 [14]. Faceneck
vessel, from Pacheco;
ceramic and slip; 50 x 34.8
cm. Museo Nacional de
Arqueología, Antropología
e Historia, Lima, C-66969.
Photo: Daniel Antonio
Giannoni Succar.
149 S hattered C eramics and Offerings
desert valley floor. The site is difficult to ac-
cess but inspiring: it has an unparalleled vista
of the surrounding region and sierra, home to
mountain spirits, or apus. Even today offerings
are placed on its slopes to honor the moun-
tains. These offerings, or illas, are miniature
statues of animals and other forms associated
with farming and the procreation of the An-
dean landscape. Offerings made to the ancient
ones residing in the mountains and earth are
thought to be reciprocated with good fortune,
including plentiful crops and herds.14
The summit of the Cerro Baúl complex
consists of many structures organized around
plazas. Investigators identified a large elon-
gated building type that served as the place
of elite residence and administration. In one
of these halls the Wari brewed chicha and
sponsored ceremonies that revolved around
the consumption of this beverage (see pp.
82–101, “The Art of Feasting”). Excavation
of one of these structures revealed a burnt
deposit of ceramic drinking vessels and fine
necklaces made of lapis lazuli or chrysocolla
beads (fig. 142; see also fig. 68). Archaeologists
who investigated the site believe that the hall
was intentionally burned, possibly as part
of its ritual abandonment. This event was so
important that fine vessels, brought hundreds
of miles from the Wari heartland, were chosen
to be sacrificed. After toasting and drinking,
the Wari smashed these vessels and threw
Figure 137 [10]. Standing
camelid vessel, from
Pacheco; ceramic and slip;
74.5 x 51.5 x 32 cm. Museo
Nacional de Arqueología,
Antropología e Historia,
Lima, C-60592. Photo:
Daniel Antonio Giannoni
Succar.

150 M ary G lowacki

 them into the burning building as part of its
ceremonial sealing. After the fire was extin-
guished, the Wari made a second offering of
necklaces.15 One could speculate that Cerro
Baúl was established to mark the extent of
Wari expansion in the southern frontier, and
rituals associated with it were dedicated to the
ancestors who spiritually protected these new
regions.
Cerro Amaru offers another example of the
Wari ceramic offering tradition in the north.
Located in the highlands of Huamachuco, the
site was contemporaneous with and centrally
located to three larger sites found within a 4
kilometer (2.5 mile) radius. A team of archae-
ologists who have conducted considerable
research within the region argue that Cerro
Amaru functioned as an important shrine
tied to water rituals and ancestor worship not
only for the local community but for people
from other parts of Peru who came to make
offerings there.16 The shrine consisted of three
wells located on a man-made mound that
captured rain and served as a reservoir, along
with storage facilities and a mausoleum.
Figure 138 [9]. Reclining
camelid vessel, from
Pacheco; ceramic and slip;
17.5 x 24.8 x 80 cm. Museo
Nacional de Arqueología,
Antropología e Historia,
Lima, C-55041. Photo:
Daniel Antonio Giannoni
Succar.
151 S hattered C eramics and Offerings
A small cache of ceramics was discovered
behind the reservoir wall that included an
anthropomorphic figurine, a blackware lyre
cup, and a small decorated jar, all of which
were smashed. The investigators posit that the
Cerro Amaru ceramic deposit was the result
of a chicha offering, and that the libation was
kept in the storage buildings expressly for sup-
plying such ceremonies. After consuming the
chicha, the Wari smashed the vessels and laid
them next to the reservoir. This, however, was
not the only offering at the site. In 1900, one
of the Cerro Amaru wells was dredged. From
it thousands of dumortierite, turquoise, and
Spondylus shell beads along with large pieces
of Spondylus shell and metal objects were
recovered. Among early Andean societies,
these materials—symbols of the earth (stone
and metal) and water (shell)—were common
items offered to the ancestors.17 They were
interred in the ground or cast into bodies of
water, such as lakes, places of ancestral origin.
In turn, the ancestors reciprocated with fertil-
ity in the form of rain.18 In Inca times, bod-
ies of water, both natural and artificial, were
Figure 139 [7]. Camelid
head vessel, from Pacheco;
ceramic and slip; 15.6
x 17.8 x 12.6 cm. Museo
Nacional de Arqueología,
Antropología e Historia,
Lima, C-55032. Photo:
Daniel Antonio Giannoni
Succar.
considered portals to the underworld of the
ancestors,19 a view that is believed to date back
as early as Wari culture, if not earlier. What is
fascinating about Wari ceremonialism is that
no significant ritual seems to have been car-
ried out without a ceramic vessel offering, and
many were intimately tied to ancestor worship
with stone, metal, and Spondylus shell serving
as basic offering components.
The elite individuals interred at Cerro
Amaru may have been the very ancestors
around whom much of the ceremonial activ-
ity of the site revolved. They were buried with
sumptuous and exotic grave goods, including
many of the Wari style. It may even have been
the case that in life these personages were
believed to be able to predict the future. As
their wisdom and powers became renowned,
people came from far and wide to consult
them. After death, these ancestors may have
continued to communicate with the living
through the medium of an oracle. During
152 M ary G lowacki
this same time period, a major oracle center
existed at Pachacamac, near present-day Lima.
It had a significant Wari occupation and, along
with other sites including Cerro Amaru, may
have formed a network of ceremonial centers
that helped people harness the energies of the
ancestors.20
Three other examples of ceramic offering
smashes come from the southern highlands
Cuzco region, the most intensively occupied
provincial region of the Wari Empire. The
first is from the monumental site Pikillacta.21
Apart from the Wari capital, Pikillacta was
the largest of the Wari complexes, but it was
never completed or fully occupied, leaving
unanswered questions about its intended role.
Nonetheless, many activities can be docu-
mented at the site. One was the ritual use of
a niched hall, considered to be a ceremonial
building type not unlike the halls of Cerro
Baúl. When the Wari abandoned the site,
the hall was left unfinished in that it lacked
 a plastered floor. Before the Wari sealed it
closed, they made a very large ceramic offer-
ing by breaking hundreds of vessels in the
space, as if ceremonially to mark the event of
leaving the site; they then capped the room
with clay. The majority of these vessels are
simple single-serving bowls thought to have
been used for drinking chicha by the work-
ers at the site; elites, however, used keros. The
majority of ceramic fragments in this deposit
represent the Wamanga style, a late Wari
ceramic pottery style. This offering shows that
the ceramic smashing tradition continued into
the late phase of the Wari period.
A much earlier offering is recorded at
Huaro, a complex of Wari sites located 17 km
(10 mi.) southeast of Pikillacta. At Qoripata,
an administrative node of this settlement that
was likely the earliest Wari administrative
center in the Cuzco region,22 a large faceneck
jar representing a human was excavated from
a room that had been used for ceremonial
feasting and drinking. Although decorated
with unique iconography, it is very similar
in size and shape to other large faceneck jars
from Conchopata and Pacheco (see fig. 99).
Buried next to this vessel, which was struck
directly on the front of the body to “kill” it
Figure 140 [8]. Camelid
skull vessel, from Pacheco;
ceramic and slip; 17 x
22.8 x 11.8 cm. Museo
Nacional de Arqueología,
Antropología e Historia,
Lima, C-55035. Photo:
Daniel Antonio Giannoni
Succar.
153 S hattered C eramics and Offerings
ritually, were pairs of smashed drinking ves-
sels. The faceneck jar, which once stood 1.5
meters (5 feet) tall, is believed to have held
chicha, served as part of a ritual to secure this
important political space ceremonially. Much
like the hall at Cerro Baúl, the room was sub-
sequently closed off and then burned, perhaps
in anticipation of violence that threatened
the Wari representatives who controlled the
region. Alternatively, Qoripata may have been
ritually shut down because the Wari estab-
lished Pikillacta, a new administrative center
that never fully came to fruition.
Another example of the Wari ceramic of-
fering tradition comes from Muyu Roqo, south
and west of the Cuzco Valley in the Paruro
region.23 It is dominated by drinking vessels,
smashed and deposited in a single offering pit.
Curiously, most of the ceramic styles repre-
sented do not belong to Wari; they date to the
Middle Horizon (600–1000) but are local ver-
sions of what the Wari would have used. Local
peoples who were influenced or dominated by
the Wari created these pottery styles to prac-
tice a Wari ritual. The role of the Wari ceramic
offering tradition must have made an impact
on the local Cuzco population; not only was it
practiced in areas outside the sphere of direct
 Wari control, such as Paruro, but also with
pottery influenced by Wari styles.
In Cuzco and elsewhere in Peru there are
other examples of buried Wari offerings, some
of which include ceramics and some of which
do not. Among the latter are three offerings of
small human figurines, two recovered un-
scientifically, but nonetheless relatively well
recorded. These two each consisted of forty
greenstone figurines interred in a room as a
dedicatory offering (see fig. 223). Like other
Wari buried offerings, they also included
Spondylus shell and copper, materials with
symbolic ties to the ancestors. The third
cache was also dedicatory in nature but more
elaborate in composition (see figs. 225, 226).
Not only did these figurines include human,
supernatural, and animal representations
made of different types of materials, including
stone, Spondylus shell, and metal alloys, but
they were also interred with Spondylus shell
and other metal objects. Moreover, the statu-
ettes represent figures associated with war-
fare, such as warriors and prisoners,24 perhaps
implying that force was necessary to establish
the Wari southern province and that offerings
to the ancestors made in their likeness were
required.25
In assessing the Wari offering tradition,
consistent patterns can be observed. First
and foremost, fundamental to it are smashed
pottery vessels, many of which are drink-
ing or serving vessels and others made to
represent humans who wear tunics with
Figure 141. Vessel with
supports modeled as
human female figures,
from Maymi. Photo: Scott
Raymond.
Figure 142. One from
a set of four shattered
cups recovered from the
chicha brewery at Cerro
Baúl. Each is decorated
with the head of the staff
deity. Photo: Patrick Ryan
Williams.
15 4 M ary G lowacki
images of supernatural beings. These vessels
are sometimes archaeologically recovered in
association with stone, shell, and metal. The
Wari often buried them in the ground, a loca-
tion closely tied to the ancestors. As several
scholars have suggested,26 this ideology can be
linked to the Inca, who saw the ancestors as
residents of the underground waterways and
purveyors of water. In a region often threat-
ened by drought, peoples of the Andean sierra
were no doubt preoccupied by availability
of water to sustain crops and pastureland,
and frequently called upon the ancestors for
assistance in fulfilling this need by making
offerings to the earth and the ancestral world.
The typical offering included Spondylus shell,
fundamental for rain-making rites (fig. 143;
see also fig. 209). Other appropriate offerings
to the ancestors were stone objects and metal
(e.g., copper), elements of the earth. These
objects and imagery are consistently found to-
gether in offerings today known as pagos, pay-
ments to the earth and the spiritual entities
tied to it, including the ancestors, a predomi-
nant life force associated with water. From co-
lonial and modern ethnographic accounts we
find that the ancestors “thirsted” and that this
thirst needed to be quenched. From ancient
times to the present, such libation offerings to
the ancestors were made with chicha.
Although the vessel forms of Wari ceramic
offerings are variable, all were used in the
service and consumption of beverages, prob-
ably primarily chicha. Many of the vessels
Figure 143 [76]. Pendant
with figurine; Spondylus
shell, stone, and metal;
13.3 x 11.4 x 5.1 cm. Fowler
Museum at UCLA, Los
Angeles, Gift of Mr. and
Mrs. Herbert L. Lucas, X88-
255. Photo: Don Cole.
included in these ritual smashing acts are
decorated with imagery of supernatural beings
and human figures drawn from an old ico-
nography linked to monumental religious and
political centers of Andean culture. The offer-
ings are also often linked to the earth, which
connects them to the ancestors, the life-giving
source of water. Water is the underlying theme
of these ceremonies, which were surely given
urgency by the fact that the Wari period began
and ended with extended droughts.
Just as water is a source of fertility in
nature, so too is chicha in association with the
earth and the ancestors. In Inca times it was
offered to sacred places and to the ancestors
by being poured on the ground.27 One early
155 S hattered C eramics and Offerings
historic account of the role of chicha in Inca
religious practice reports that “the best and
most important part of the Indian sacrifices,
is chicha. By it and with it the festivals of the
huacas [sacred places or objects] begin, be-
cause of it they happen, and with it, they end.
It is everything.”28
This belief and practice exist even today
in traditional societies.29 In southern high-
land society, chicha is a fluid associated with
fertility.30 The substance itself has a frothy ap-
pearance, resembling semen, and rituals play
out this analogy. As early as Moche culture,
which predated and overlapped with Wari
culture on the north coast, evidence exists for
the consumption of chicha from vessels that
 explicitly link the beverage to sexuality and
fertility.31 Elsewhere in South America, such
as the Amazonian region, whose people share
related ideologies, a ceramic bowl contain-
ing chicha represents the female or wife, and
it is offered to the male, the husband. The
ceremony thus signifies procreation by the
two. Moreover, the ceramic vessel offering
of chicha is ceremonially and ritually tied to
the perpetuation of crops and livestock. As
mentioned above, according to various Ande-
an ethnographic accounts, both the spiritual
and earthly worlds (especially the ancestors)
“thirst” for this libation as a means of main-
taining their fertile powers.32
A drought that lasted for several decades
during the sixth century33 may have been
the impetus for the Wari expansion; it drove
the Wari to acquire arable lands outside the
heartland. The regions on which they imposed
the heaviest control were those that could
supply a regular flow of food and other key
resources to the Ayacucho heartland. Con-
sequently, the rituals the Wari performed in
these territories were aimed at legitimizing
their presence through the promise of their
156 M ary G lowacki
ancestors. By implanting their own ancestors,
perhaps represented by the co-opted ancient
frontal staff deity, as idols and effigies into the
new territories and offering them chicha to
satisfy their thirst, they hoped to draw water
to local farm and pasture lands. Other types of
offerings were also essential: items of copper,
Spondylus shell, and stone associated with
the earth, water, and the ancestors. A similar
technique was employed to obtain important
ritual items, such as Spondylus shell and cop-
per: ceramic vessels with key Wari imagery,
namely supernatural creatures, seem to have
been used to facilitate trade with north coast
societies for Spondylus shell (see fig. 30).
Throughout Peru archaeologists have re-
corded symbols of Wari power, one expression
being the Wari ceramic offering tradition. For
several hundred years the Wari were a pre-
dominant presence on the Andean landscape,
their ideology influencing many peoples.
While evidence of the power of Wari ances-
tors can never be fully understood, scholars
continue to gain insights through the broken
pieces of this rich and intriguing tradition.
notes 1. Cobo [1653] 1890–95, vol. 3, bk. 13,
chap. 21; Benson and Cook 2001, 1.
2. For northern China, see Kuen
Lee and Zhu 2002; for the Minoan
Greeks, see Tompkins 2009; for the
Mimbres culture, see Brody 1977;
for the Mississippian peoples, see
Pollack et al. 1987.
3. Cook 1985; Cook 1979; W. Isbell
and Cook 2002; W. Isbell and Cook
1987; Tello 1942.
4. Demarest 1981.
5. Figure 59 likely depicts this
activity.
6. W. Isbell and Cook 1987.
7. Menzel 1964, 23.
8. Ibid., 19, 26.
9. Ibid., 3–4.
10. W. Isbell 1977b, 231; McEwan
1998, 82.
11. El Proyecto Arqueológico Maymi
(Anders 1990), directed by Martha
Anders. See also Anders et al. 1998.
12. See Anders 1990, fig. 10.
13. See ibid., figs. 12a, b.
14. Moseley et al. 1991.
15. Williams et al. 2002, 69–73.
16. J. Topic and T. Topic 1992.
17. Glowacki and Malpass 2003,
441–43.
18. Ibid.
19. Cobo [1653] 1890–95, vol. 3,
bk. 13, chap. 47; Duviols 1978;
Sarmiento de Gambo [1572] 1942, 70;
see also Sherbondy 1982.
157 S hattered C eramics and Offerings
20. This is based on an analogy with
the site Pachacamac, near Lima. The
Spaniards say that, during the Inca
period (AD 1400–1532), Pachacamac
functioned as an oracle that had
“branch” or “satellite” temples in
various areas of the Andes (see, for
example, Vega [1609] 1987, 71, 78,
379–80, 384). Some suggest that
the same was true during the Wari
period (see Shea 1969, v–vi).
21. Glowacki 1996, 456–57; McEwan
1984b, 16.
22. Glowacki 2000. Julinho Zapata
and I recorded a single but signifi-
cant offering during 1996–97 field
investigations.
23. Bauer 1999, 64–66.
24. Arriola Tuni and Tesar 2011.
25. For instance, Cook 1992;
Glowacki and Malpass 2003; J. Topic
and T. Topic 1992.
26. For instance, Allen 1988, 153–54.
27. Arriaga [1621] 1968, 137; Cobo
[1653] 1890–95, vol. 3, bk. 13, chap.
21–22.
28. Arriaga [1621] 1968, 209.
29. Allen 1988, 149.
30. Weismantel, 273.
31. Bergh 1993.
32. Allen 1988, 47.
33. Thompson et al. 1985.
Susan E. Bergh
Tapestry-woven Tunics
Figure 144 [117]. Tunic Wari tapestry-woven tunics, versions of an
with face-fret motif;
ancient garment type known in the native lan-
camelid fiber and cotton;
102.2 x 102.2 cm. Dallas guages of the Andes as unku (Quechua) and
Museum of Art, The khawa (Aymara), belong to a distinguished
Eugene and Margaret tradition of ancient Andean tapestry weav-
McDermott Art Fund, Inc.,
in honor of Carol Robbins’ ing that culminated chronologically with the
40th anniversary with the Inca, the last completely indigenous culture
Dallas Museum of Art, to develop in the region before the Spanish
2004.55McD.
conquest (fig. 144).1 Tapestry, which refers to
cloth woven in a specific way rather than to
pictorial cloth in general, made its debut in
the Andes during the first millennium BC2
and quickly became a prestige fabric used not
for interior furnishings, its principal function
in the West, but for sumptuous garments such
as tunics (shirts), mantles (shoulder wraps),
and loincloths. By the time of the Inca Em-
pire, tapestry-woven textiles were classified
as cumbi (also spelled “qompi”), a category
of treasured, superior-quality cloth that Inca
royalty claimed as their exclusive privilege,
whether for personal wear or to bestow as
esteemed gifts to strengthen bonds of loyalty.3
In the early years following the conquest,
Spanish commentators shared this enthusi-
asm for cumbi, which they uniformly ranked
as finer than European cloth and admired for
its exquisite, silk-like softness and technical
refinement.4
The Spaniards’ eye for textiles is not sur-
prising since in pre-industrial Europe cloth
was highly valued because of the enormous
amount of labor and time that its creation de-
manded. As textile scholar Ann Pollard Rowe
remarks,5 it is no accident that the Industrial
Revolution focused first on streamlining the
production of this costly, essential commodity,
which was so expensive that in the late eigh-
teenth-century United States it was harder to
obtain than food and lodging.6 Rowe goes on
159
to say that, in areas of the world where cloth is
made entirely by hand as it was in the ancient
Andes, the process of creating it usually ranks
second only to food production in economic
and occupational importance—an astonishing
statement from a contemporary perspective.
But in the Andes cloth’s importance went far
beyond the economic. For instance, in one of
the most celebrated quotes in Andean stud-
ies, John Murra concluded that among the
Inca “no political, military, social, or religious
event was complete without textiles being
volunteered or bestowed, burned, exchanged,
or sacrificed.”7 Experts assume that cloth had
similar importance, if not identical uses, in
many earlier Andean cultures, which together
created one of the most aesthetically accom-
plished and technically innovative textile
legacies in the world.
Wari textiles are a crucial chapter in this
history, particularly tapestry-woven cloth.
Using the tapestry weave Wari weavers fabri-
cated several types of garments, among them
mantles and headbands (fig. 145; see also [131],
p. 274). Far more common, however, are tu-
nics, which likely served as partial inspiration
for the cumbi tapestry-woven tunics that Inca
rulers, nobles, and state functionaries wore
(see fig. 240).8 Except for a handful of stone
sculptures, the tunics are the largest of Wari
artifact types and certainly the most com-
plex. Their intricacy derives from the physi-
cal structure of the cloth—for the initiated,
a fascinating world into which the ancients
poured intellectual energy—along with more
visible systems of artistic composition, includ-
ing flamboyant color, format, imagery, and an
arcane, cerebral convention for distorting im-
agery that culminates in a geometric abstrac-
tion admired today for its “modern-ness.”
Figure 145 [130]. Head­ The Tapestry Weave and Tunic Construction
band; camelid fiber
A few features of the tapestry weave—a
and cotton; 67 x 12 cm.
The Textile Museum, simple structure from which the ancients
Washington, DC, Museum wrought great aesthetic complexity—contrib-
Purchase, 1965.32.1. uted to its choice as a preferred fabric type for
high-status garments. First, it lends itself to
the creation of intricate, mosaic-like patterns
made up of areas of pure, undiluted color that
are woven into the cloth rather than added to
a pre-existing fabric with needlework. This is
accomplished by passing the wefts—the mul-
ticolored yarns that the artist worked horizon-
tally during weaving—back and forth in areas
that range from tiny to large, and then packing
the wefts down so tightly that they completely
conceal the undyed, vertical warps.9 The
packing-down consumes extravagant amounts
of yarn and, of course, the labor and time the
extravagance implies.
The process begins with gathering and/
or growing the fiber (silky camelid hair for
the weft and either cotton or camelid fiber for
the warp) and continues with harvesting and
cleaning. Next are the very time-consuming
tasks of spinning and plying, and then dyeing,
often with precious colorants. Only then can
weaving commence, followed by garment con-
struction. In other words the tapestry weave is
resource-intensive at every stage of manufac-
ture, which is no doubt another reason both
the Inca and the Wari revered it as a noble
cloth. A shorthand way to state the human in-
vestment is to say that a Wari tapestry-woven
tunic of routine quality incorporates around
seven miles of carefully handmade yarn,
while the finest example so far documented
has an extraordinary eighteen miles.10 Based
on comparing yarn counts, an objective mea-
sure of quality, Wari tapestry weaving far out-
strips even the greatest tapestry weavings of
sixteenth-century northern Europe, which are
16 0 S usan E . Bergh
among the most famous examples of the tech-
nique in the world.11 In an added refinement
Wari tapestry weavers painstakingly finished
the cloth on both its faces, one of the essential
qualities of cumbi among the Inca; in contrast,
European counterparts have a distinct back or
wrong side marked by dangling yarns.
In creating a tapestry-woven tunic the
great majority of the effort went into producing
the cloth. Garment construction was straight-
forward since, like most Andean clothing,
Wari tunics are not elaborately tailored affairs
made of pieces cut from a larger whole. Rather,
the cloth was woven to shape on the loom
and while there completely finished on all but
one of its edges, a process that required weav-
ers to conceptualize every aspect of design
before work commenced. Wari tunics consist
of two such loom-shaped panels that most
often are simple rectangles, each about 50 by
200 centimeters (20 by 80 inches). The panels
were placed side by side and stitched together
along a seam that falls at the tunic’s center;
they were then folded in half to form the
shoulder line and seamed up the sides. Gaps in
the seams serve as openings for the neck and
arms. The resulting, roomy garment is roughly
100 cm (40 in.) on a side and on a person of
five-foot stature fell to the knees at the front
and back and well below the elbows at the
sides. Artistic representations suggest that
tunics were sometimes belted and were worn
without a lower body garment, or at least one
that was visible below the tunic’s lower edge.
Imagery and Wearer
In the Andes tunics were an essential article
of men’s attire; as the scholar R. Tom Zuidema
has observed, they cannot be understood with-
out imagining the presence of the lords who
wore them—the iconographic whole was the

Figure 146 [36]. Figure in
tapestry-woven tunic and
four-cornered hat; ceramic
and slip; 28.7 x 23 cm.
Fundación Museo Amano,
Lima, FMAC-000020.
Photo: Daniel Antonio
Giannoni Succar.
Figure 147 [144]. Four-
cornered hat with geomet-
ric motifs; camelid fiber
and cotton; 12.4 x 17.5 cm.
The Metropolitan Museum
of Art, New York, Gift of
Arthur M. Bullowa, 1983,
1983.497.6. Image: © The
Metropolitan Museum
of Art. Image source: Art
Resource, NY.
161 Tapestry-Woven T unics
lord, including tunic, ornaments, headgear,
and other paraphernalia.12 Based on artistic
representations, ornaments included large
ear spools that undoubtedly also signaled
high status, as they did among the Inca (see
pp. 217–31, “Inlaid and Metal Ornaments”).
Headgear ranged over several different types,
among which four-cornered hats were impor-
tant (figs. 146, 147).13 Other items certainly
sometimes included staffs, one of the period’s
most important symbols of human and divine
sovereignty (see fig. 1).
The sumptuousness and great standardiza-
tion in the size, format, color, construction,
and technical features of Wari tunics have
long suggested that, as among the Inca, they
were made under state auspices and worn by
those important to the administration of the
Wari polity: rulers, their representatives, and
probably valued allies, who received them
as prestigious gifts. This supposition raises
the possibility that the tunics’ imagery corre-
sponds to official functions, if only in a loose
way, since that imagery also is standardized to
encompass a narrow range of motifs, only one
of which usually repeats in different orienta-
tions and colors in any given tunic.14 Unfor-
tunately, little can now be said about these
functions as most representations of tunic-
wearing individuals provide few hints and the
vast majority of tunics come from unscientific
excavations, most probably of tombs that may
have held insignia related to the roles the de-
ceased played in life. There are one or two ex-
ceptions, however, and the tunics themselves
can be used to make broad generalizations.
Profile-face and Stepped-fret Tunics
One of the most common of Wari tunic types
features a profile-face and stepped-fret (face-
fret) motif (fig. 144).15 In the arts such tunics
appear on humans of unknown identity but
obvious power: one strikes a pose akin to that
of the staff deity, arms outstretched and hands
cupped to form holes into which implements,
perhaps staffs of authority or weapons, were
once likely inserted (fig. 146). Clear military
associations for this tunic type occur in ce-
ramics recently unearthed at Conchopata, an
important Wari site near the capital; on these
vessels fierce, axe-wielding warriors dressed
in face-fret tunics, probably military leaders,
parade in belligerent ceremonial display (see
fig. 7).16 The Wari also occasionally associated
the face-fret with supernatural sacrificers in
Figure 148 [121]. Views
of the front and back of
this tunic are arranged
as though the tunic is
unfolded at the shoulder,
marked by a white line. A
single profile face appears
in the third row from
the bottom. Tunic with
paired-fret motif; camelid
fiber and cotton; 98 x 106
cm. Staatliches Museum
für Völkerkunde, Munich,
57-20-245 (NM 245).

162 S usan E . Bergh

Figure 149 [123]. Tunic
with face-fret and inter-
locked U-shaped motifs;
camelid fiber; 106 x 94
cm. American Museum of
Natural History, New York,
41.2/8604. Image: cour-
tesy American Museum
of Natural History,
Anthropology. Photo: Craig
Chesek.

tapestry-woven tunics as well as ceramics
on which the staff deity also appears (see fig.
102).17 In at least some cases, then, the motif
seems to relate to conflict and death, some of it
cosmically sanctioned.
What the motif represents is still mysteri-
ous. The consistent pairing of the face and fret
implies that they have complementary and
reinforcing meanings, but the fret’s formal
simplicity and presumed abstraction have so
far resisted interpretation.18 The face too has
few identifying features beyond its eye orna-
ment, vertically divided eye, and the N-shaped
canines that it often bares, all generic traits
of the suprahuman in Wari art. Unconfirmed
suggestions about its identity, to which the
sacrificer should be added, include a trophy
head or the head of one of the staff deity’s
winged attendants.19
With a few exceptions, face-fret tunics are
only ordinary in quality (as measured by yarn
counts); thus their wearers, though distin-
guished, probably did not occupy the summit
of the Wari hierarchy.20 These tunics divide
into several subtypes that may correlate with
variations in time or place of manufacture;21
a small group of tunics combines the face-
fret with other geometric motifs, and the fret
sometimes appears on its own in composition-
ally related examples (figs. 148, 149).

163 Tapestry-Woven T unics

Figures 150a, 150b [114].
Views of the front and
back of this tunic are
arranged as though the
tunic is unfolded at the
shoulder, marked by a
white line. Perhaps in
antiquity one row of
figures was removed from
one edge (at the top of the
photograph). The detail
below shows two figures
from the lower left corner.
Tunic with sacrificer; cam-
elid fiber and cotton; 106.7
x 112 cm. The Cleveland
Museum of Art, John L.
Severance Fund 2007.179.

16 4 S usan E . Bergh
Figure 151 [116]. Tunic Winged Attendant and Sacrificer Tunics
with sacrificer-related
Another very large group of tapestry-woven
creature; camelid fiber
and cotton; 100 x 112 cm. tunics features the figures that in other con-
Museum der Kulturen, texts accompany the staff deity: sacrificers,
Basel, collected by Hans which appear in more than a half-dozen
Theodor Cron (1921–1964),
IVc23577. Photo: Markus iterations that always include a weapon and
Gruber, 2008. a human victim or its head (figs. 150, 151),
and, much more commonly, the more benign
winged attendants, which occur in bewilder-
ing variety (figs. 152, 153). The attendants are
always shown in profile, bent on one knee,
holding a staff of authority to the front of the
body, and wearing a complex headdress; an
elaborate wing sprouts from the back, usu-
ally over an appendage that streams from the
figure’s neck (fig. 154). But there the similar-
ity ends. Ornaments that festoon the figures
change kaleidoscopically and so do the
figures’ heads, which range from birds and
animals to humans and others whose heritage

16 5 Tapestry-Woven T unics
Figure 152 [103]. These
large fragments come from
a tunic that had sleeves.
Each represents the length
of the tunic from the
shoulder to the lower edge.
Together, they probably
formed a single panel that
has been divided along the
shoulder line. Tunic frag-
ments with bird-headed
staff-bearing creature in
profile; camelid fiber and
cotton; 90 x 53 cm and 89.4
x 53.5 cm. The Cleveland
Museum of Art, Purchase
from the J. H. Wade Fund
2005.53.a–b.
16 6 S usan E . Bergh
is unclear. In total, the tunics feature more
than twenty distinct versions of the winged
attendant.22
Most versions of this figure type ap-
pear in only a few tunics, but three occur in
many more examples. Of those three, two are
adorned with figures, both with heads raised
up, that are different and yet so similar they
raise suspicion of kinship.23 One is a bird-
headed attendant that may conflate the fea-
tures of several species, including the Andean
condor (one of the world’s largest birds of
flight) and other raptors such as a falcon or the
harpy eagle, as well as a parrot, perhaps one
whose brilliant feathers were a form of wealth
(figs. 152, 154 right). The second figure is an
attendant with an animal head of unclear deri-
vation and a persistently two-fingered “hand”
undoubtedly based on the cloven hooves
of a deer or, perhaps more likely, one of the
Andean camelids—llama, alpaca, vicuña, or
guanaco (figs. 153, 154 left). Although the two
figures’ heads and elaborate eye markings are
distinct, many of the remaining ornaments are
virtually identical, down to a unique combi-
nation of headdress trimmings that include
small heads with L-shaped mouths. (In a few
tunics with the bird-headed attendant, such
as the one illustrated, bird heads substitute
in the headdress.) None of the other tapestry-
woven attendants share as many features as
these two. In terms of quality, however, tunics
Figure 153 [104]. Tunic
with camelid- or deer-
headed staff-bearing crea-
ture in profile; camelid
fiber and cotton; 103.7
x 108.5 cm. Deutsches
Textilmuseum, Krefeld,
12299/2558.

Figure 154. Comparison

of the camelid- or deer-
headed creature (left) and
the bird-headed creature
(right). The camelid (or
deer) has somewhat differ-
ent side-to-side proportions
because of the effects of
distortion. Tracings: Susan
E. Bergh, based on a tunic
at the Textile Museum,
Washington, DC, 91.386
(left) and Taullard 1949, 56
(right).

167 Tapestry-Woven T unics

Figure 155 [106]. Tunic
with feline-headed
staff-bearing creature
in profile; camelid fiber
and cotton; 104.7 x 102.8
cm. The Textile Museum,
Washington, DC, Museum
Purchase, 1961.3.17.

Figure 156. Feline-headed with the bird-headed attendant are superior to
staff-bearing creatures
those with the camelid (or deer). They incor-
from a tunic very similar
to that shown in Figure porate many more figure repeats, an index of
155. Tracing: Susan E. weaving skill and investment, as well as far
Bergh, based on a tunic at more yarn, much of it a deep indigo-dyed blue,
the Staatliches Museum
für Völkerkunde, Munich, the most prestigious color that Wari weavers
34-50-6. employed.
That the bird and camelid (or deer) relate
to one another is further suggested by the
interest that most other winged attendant and
sacrificer tunics evince in paired figures.24
Rather than being segregated in different tu-
nics, however, these two figures—often closely
similar but sometimes very different—alter-
nate regularly with one another in the same
garment. In the third large group of winged at-
tendant tunics, for instance, the figures, whose
upright ears may refer to ancestry in the feline
world, are identical except for the decoration
of headdresses, neck appendages, and staffs,
here perhaps transformed into the hunter’s
(or warrior’s) spear-thrower by the side hook
that emerges from the upper portion (figs. 155,
156). The staffs are of particular interest since
they represent two types that the Wari used
very often to distinguish figural variants, at
least in the tunics: a wavy zigzag decorates the
shaft of one and a nested square the length of
the other. The same contrast occurs in sac-
rificer tunics, one very beautiful example of
which provides illustration (figs. 157, 158). To
one side of its body the figure holds a panpipe

16 8 S usan E . Bergh
Figure 157 [113]. Tunic
with sacrificer; camelid
fiber and cotton; 103.4
x 110.8 cm. The Textile
Museum, Washington,
DC, Museum Purchase,
1966.5.2.

Figure 158. The sacrificer

featured on the tunic
shown in Figure 157. On
the left is the entire figure,
its rear portions contracted
and nearly illegible. On the
right is the expanded rear
portion of another figure
repeat. Tracing: Susan E.
Bergh.

169 Tapestry-Woven T unics

 and to the other an axe with a haft patterned
by either the zigzag or the nested square.25
These two versions are otherwise established
only by consistent, subtle differences in the
orientation of the bird heads that dangle from
the sacrificer’s belt.
This insistence on paired figures finds its
likeliest explanation in dualism, a principle so
fundamental in the Andes that it has struc-
tured native thought and practice from ancient
times to the present day (see pp. 103–21, “The
Coming of the Staff Deity”).26 Generally stated,
dualism is founded on the conviction that the
world comes into being and continues to exist
through the dialectical, give-and-take balanc-
ing of two forces that are at once intrinsically
antagonistic but profoundly complementary
and indispensable to one another. The two
forces or principles are often gendered male
and female but also conceived as many other
natural dyads, such as left-right and upper-
lower. Today and in the past, Andeans activate
this pervasive way of thinking in two broad
ways that often interpenetrate, one reinforcing
and legitimizing the other, but do not neces-
sarily imply each other.27 One is in the realm
of symbolic thought, including beliefs about
the unseen structure and workings of the
universe. Given the tunics’ numinous imag-
ery—the attendants and sacrificers that are the
intimates of the all-important staff deity—it
seems safe to say that, at a minimum, the du-
alism upon which the tunics insist character-
ized important aspects of Wari cosmological
belief, religion, and perhaps ritual, although
how remains elusive. Other testimony in this
regard is offered by the two staff deities that
alternate with each other on the interiors of
Wari feasting vessels from the south coast,
one identified as male and the other as female
based on differences in wardrobe (see figs. 5a,
5b).28
It is harder to know whether the Wari
also put dualism to use in social and political
domains in the manner of many late pre-
Hispanic and contemporary Andean people.
For instance, at the time of the Spaniards’ ar-
rival, the Aymara-speaking Lupaqa of the Lake
Titicaca region divided their realm into two
parts (moieties) that they described as upper
(alasaa) and lower (masaa) and ranked hierar-
170 S usan E . Bergh
chically in relation to each other. Two para-
mount lords joined forces to govern the Lupaqa
polity as a whole, Qari of the upper part and
Qusi of the lower part, but Qari’s status and
wealth were greater, reflecting the ranking.
Each Lupaqa province was similarly organized
and ruled by subordinate paired lords who
were subject to Qari and Qusi.29 Inca soci-
ety adopted the same kind of ranked, binary
organization although there is no agreement
about whether Inca rulers governed in pairs in
which one partner held higher status.30 If the
Wari bird and camelid (or deer) tunics were in
simultaneous use—a still-open question since
the tunics’ chronology is unsettled—it may be
that they reflect such dual social organization
and its concomitant paired, ranked political
offices since the two tunic groups offer clear
evidence of ranking through the very marked
differences in their quality. Conceivably, those
who wore other tunics with paired figures
also carried out political functions in tan-
dem, their complementary roles symbolized
by their different staffs and accouterments.
But further investigation, both archaeological
and art historical, is needed before it can be
said unequivocally that dualism premised the
organization of Wari society in addition to the
symbolic domain.31
Other Tunic Types
In extant tunics other iconography occurs less
frequently than the face-fret motif, winged at-
tendants, and sacrificers. Among these scarcer
images is a profile creature whose head varies
in aspect between human and animal. Most of-
ten a tail-like angular scroll, sometimes tipped
with a bird or animal head, curves behind its
severely geometricized “body,” but a hand-
like motif sometimes replaces the scroll (fig.
159). The most complex versions include other
motifs, including plants (fig. 160).32 Typically,
this profile creature confronts a mirror image
of itself, and the two profile visages merge into
a single frontal face.33 What it connotes is un-
known. It appears on a tunic, rare in its small
size, that may have clothed an important and
treasured child or perhaps an object of some
kind (fig. 159). An unusual small tapestry-
woven panel that has the proportions of a man-
tle may be another child’s garment (fig. 161).34
Figure 159 [125]. This
small tunic may have been
made for a child or for an
object. Tunic; camelid fiber
and cotton; 55 x 53.7 cm.
Private collection. Photo:
Maury Ford.

171 Tapestry-Woven T unics

Figure 160 [126]. Tunic;
camelid fiber and cotton;
100 x 106 cm. The Textile
Museum, Washington,
DC, Museum Exchange,
1962.5.1.

172 S usan E . Bergh

Figure 161 [98]. Panel;
camelid fiber and cotton;
77 x 109.5 cm. Royal
Ontario Museum, Toronto,
931.11.1. Photo: with
permission of the Royal
Ontario Museum © ROM.

173 Tapestry-Woven T unics

 Figure 162 [127]. Tunic
with skulls; camelid fiber
and cotton; 220 x 115 cm.
Museo de Arte de Lima
Collection, Prado Family
Bequest, 2.1-1241-IV.
Conserved with the sup-
port of the Southern Peru
Copper Corporation 2001.
Photo: Daniel Antonio
Giannoni Succar.
Figure 163. The geometric
motif with profile bird
heads. After Sawyer
1963, type IIb; courtesy
the Textile Museum,
Washington, DC.

Another group of tunics features a mean-

der motif disposed in horizontal fields, a radi-
cal departure from the far more typical verti-
cal banding. Imagery is often superimposed
on the meander and much of it seems strongly
oriented toward death, usually skulls (fig. 162)
but also columns of vertebra-like motifs and,
in one instance, a stunning sacrificer that is
among the most complex and monumental
known.35 In technical terms tunics of the final
group, adorned with a sophisticated geometric
motif that incorporates profile bird heads, are
among the finest that Wari weavers created
(fig. 163). Their iconography again indicates
that the Wari associated birds with the high-
est status members of their society, who may
have carried the title “mallku” (condor) or

174 S usan E . Bergh

Figure 164 [107]. Tunic
with staff-bearing creature
in profile; camelid fiber
and cotton; 97 x 144.9 cm.
Brooklyn Museum, New
York, Gift of the Ernest
Erickson Foundation, Inc.,
86.244.109.
175 Tapestry-Woven T unics
“huamani” (falcon), as paramount rulers did
in later Andean times.36 Sadly, these tunics
often survive only as fragments.37 In addition
to these discrete groups of tunics are several
examples with imagery that is so far less usual
(see fig. 81).
Sleeved Tunics
In Wari art the staff deity consistently appears
in a garment that represents a tapestry-woven
tunic, to judge from the vertical banding that
stripes the chest. But this is no ordinary tunic.
Rather, it appears to correspond to a special
rare type that unlike all others has sleeves (fig.
164);38 in artistic representations the sleeves,
patterned with interlocked L-shaped motifs,
fit snugly around the god’s upper arms. There
can be no doubt that sleeved tunics were the
raiment of only the most exalted individuals—
in all likelihood, paramount rulers them-
selves—since they are among the most sub-
lime achievements of ancient Andean tapestry
weaving, standing at the apex of that long and
bannered tradition in its entirety.39 Several
tunics with the bird-headed partner of the
camelid (or deer) once were sleeved (fig. 152).
Most other sleeved tunics also feature winged
attendants or sacrificers, but a few feature
other imagery.
Three technical features provide the
gauge of these tunics’ quality. In comparison
to unsleeved examples, they routinely incor-
porate between four and eight more miles of
yarn and flaunt many more than twice the
Figure 165 [102]. This number of figure repeats, which generally
fragment is from a tunic’s
do not skimp on iconographic details even
side seam, a remnant of
which runs up the frag- though they are smaller because the weavers
ment’s center. The original fit them into a standard tunic size. Also, half-
tunic does not appear to figures appear along the side seams of some
have had sleeves, but the
fragment is one of the best sleeved tunics; when the seam was created,
extant examples of the it united a front half with a back half to form
matching of half-figures a nearly perfectly matched whole (fig. 165).
across seams. The tunic
shown in Figure 168 also
These figures demand very precise control of
had matched half-figures at spacing and proportions among many other
the side seams. Tunic frag- things and cause modern weavers to sigh in
ment with staff-bearing
admiration. The industrialization of textile
creatures in profile;
camelid fiber and cotton; production has dulled our sensitivity to such
54 x 15 cm. Ethnologisches refinements, but the original audience, deeply
Museum, Berlin, VA familiar with the hand-weaver’s art, would
66028. Image: bpk, Berlin/
Ethnologisches Museum/ have been alive to both them and the virtu-
Art Resource, NY. Photo: osity and status that they represent. It is not
Claudia Obrocki. surprising, then, that the sleeved tunic had a
special ceremonial charge signaled not only
by its association with the deity but also by
the fact that the only known miniatures of
Wari tapestry-woven tunics have sleeves (figs.
166, 167). These exquisite tunics, too small for
even a human infant, likely had devotional
purposes. Ceremony may also have motivated
the ancients to slice the sleeves from some tu-
nics, as though decommissioning them, an act
that may have occurred before the tunics were
pulled over mummy bundles and deposited in
tombs. One of the very few tunics to have been
scientifically recovered, from a grave at Ancón
on Peru’s central coast, has been so desleeved
(figs. 168, 169).40
Why did the ancients endow the sleeved
tunic with such high status? One possible rea-
son is that sleeves were adopted as prestigious
exotica from the Moche, among whom the
sleeved tunic seems to have been a tradition.41
The Moche dominated the north coast of Peru
in the years just before the Wari came to pow-
er. If true, the emulation is one of several hints
that at least some sleeved tunics were cre-
ated very early, at the threshold of the Middle
Horizon as elites scrambled to take advantage
of shifting conditions provoked by the decline
and transformation of earlier cultures and the
ascendance of a new order. In these circum-
stances the sublime quality of sleeved tunics
would have helped establish the prestige both
of the humans who wore them and of the
religion that the staff deity represents. That

176 S usan E . Bergh

178 S usan E . Bergh
Figure 168 [108]. This
large fragment, excavated
at Ancón, comes from a
tunic that had sleeves.
The shoulder line, toward
the top of the fragment,
is marked by a reversal
in the orientation of the
figures. Tunic fragment
with figures; camelid
fiber and cotton; 118.5 x
103.5 cm. Ethnologisches
Museum, Berlin, VA 7468
(16). Image: bpk, Berlin/
Ethnologisches Museum/
Art Resource, NY. Photo:
Dietrich Graf.
Figures 169a–c [110–112].
Mummy bundle from the
necropolis at Ancón, a site
on Peru’s central coast,
along with a rendering
of the tomb in which the
bundle was found. The
bundle is dressed in the
tunic shown in Figure 168;
the work basket shown
in Figure 170 was found
at the side of the bundle.
After Reiss and Stübel
1880–87, pls. 17, 16, and 10.

179 Tapestry-Woven T unics

Figure 170 [109]. Weaver’s
work basket and contents,
from Ancón; bone, camelid
fiber, cotton, reeds, and
wood; 20 x 26 x 18 cm.
Ethnologisches Museum,
Berlin, VA 5816a–t. Image:
bpk, Berlin/Ethnologisches
Museum/Art Resource, NY.
Photo: Claudia Obrocki.

Figure 171 [132]. Glove;

camelid fiber and cotton;
28.6 x 22.1 cm. Brooklyn
Museum, New York,
Charles Stewart Smith
Memorial Fund and
Museum Collection Fund,
58.204.

18 0 S usan E . Bergh
Figure 172 [119]. Tunic
with face-fret motif; cam-
elid fiber and cotton; 108.6
x 109.7 cm. The Textile
Museum, Washington, DC,
acquired by George Hewitt
Myers in 1941, 91.343.
is, rather than simply reflecting a grandeur
already achieved, sleeved tunics may have
played an active role in conveying the appeal
of the new cult and promoting its spread.42 A
unique glove-like tapestry testifies to continu-
ing cross-fertilization between Moche and
Wari weaving traditions later in the Middle
Horizon (fig. 171). On it a weapons-bearing
Moche warrior appears with small felines and
profile zoomorphic heads of Wari derivation.
Distortion
The most fascinating and peculiar of the
tunics’ features is a deliberate, systematic, and
rule-bound method of distorting form that
is uniquely Wari; it occurs at no other time
181 Tapestry-Woven T unics
in the Andes or even in any other Wari me-
dium.43 The concept is simple but the aesthetic
ramifications are complex: the portions of
each motif closest to the tunic’s center expand
from side to side and the parts closest to the
sides narrow and compress.44 The system’s
operation may be easiest to grasp in face-fret
tunics: Figure 144 illustrates a tunic with
relatively undistorted imagery; Figure 172
shows severe distortion of the same imagery.
But distortion also appears in most other tunic
groups; since Alan Sawyer first defined it in
1963, it has been most celebrated for its effects
on winged attendant and sacrificer imagery,
which according to some anticipate twentieth-
century abstract art (see pp. 5–27, “The Histo-
Figure 173. Distortion staff head headdress wing staff head headress wing staff head headress wing
in two tunics with bird-
headed staff-bearing crea-
tures in profile. Drawings:
Milton Sonday, with minor
modifications; courtesy
the Textile Museum,
Washington, DC.

foot leg and foot foot leg and foot foot leg and foot

Figure 174 [105]. Tunic

with camelid- or deer-
headed staff-bearing
creature in profile; camelid
fiber and cotton; 100 x 92
cm. Museo Nacional de
Arqueología, Antropología
e Historia, Lima, RT-1650.
Photo: Daniel Antonio
Giannoni Succar.

182 S usan E . Bergh

 ry of Inquiry into the Wari and Their Arts”).45
Figure 173 compares undistorted and mod-
erately distorted versions of the bird-headed
winged attendant discussed earlier. On the
left, the proportions of the attendant, which is
rendered in a somewhat geometricized style,
are normal except for an enlarged back foot.
On the right, one repeat of the same figure
appears on either side of the tunic’s unusually
striped center line. The staff of the rightmost
figure has widened considerably while the
headdress and three-feathered wing have nar-
rowed. The reverse occurs in the leftmost fig-
ure in accord with its changed relationship to
the tunic’s center and side: the wing and head-
dress expand but the staff contracts. All other
Figure 175. Distortion
parts of the figure, now very geometricized,
in the tunic shown in follow suit. In the most extreme application
Figure 174. Drawing: whole sections of the figure disappear and the
Milton Sonday, with minor
remainder is reduced to a collection of geomet-
modifications; courtesy
the Textile Museum, ric forms whose legibility is further compro-
Washington, DC. mised by color (figs. 174, 175). An added twist
staff head headdress wing
leg foot
staff head headdress wing
leg leg foot
183 Tapestry-Woven T unics
affects the width of the vertical bands, which
often narrow progressively toward the tunic’s
sides; in a few cases this narrowing creates
the illusion of cylindrical volume as the cen-
tral bands appear to advance and the sidemost
bands to recede.46
Why did they do it? There is no final an-
swer, but many believe that distortion is not
iconographic—that is, it holds no symbolic or
other meaning that can be interpreted. Rather,
it is sheerly an aesthetic contrivance perhaps
undertaken to relieve the tunics’ repetitive
simplicity,47 to disguise and mystify their
sacred imagery,48 or as a delightful intellectual
exercise with form that endows the tunics
with a pleasing rhythmic syncopation and,
by providing a glimpse into the workings of a
lively, playful intelligence, gives them a hu-
man approachability.49 The so-far unproven
implication of some of these views is that
distortion registers chronology, that its effects
became more profound through time as weav-
ers pushed the system to its extreme and logi-
cal conclusion.50 Although nothing is known
of the relationship that Wari artists had with
their state patrons, aesthetic motivations must
be given serious consideration since not do-
ing so risks denying these ancient artists the
creative genius accorded their counterparts in
other parts of the world, especially the West.51
If the wellspring was purely artistic invention,
however, it is much harder to say whether its
goal was abstraction, particularly in the sense
that it is understood today in the West.
But in view of the tunics’ presumed state
sponsorship it is possible that artists devel-
oped this aesthetic innovation in concert with
iconographic concerns—in other words, dis-
tortion has meaning, and this meaning does
nothing to diminish the ingenuity with which
artists chose to express it. It was explained
above that the great majority of winged atten-
dant and sacrificer tunics depict two figural
variants: bird and camelid (or deer) as well as
many others differentiated by the staffs they
carry and a range of other traits both subtle
and obvious. Distortion also results in figural
variants that are both the same and different;
as though in vacillating states of being, one
expands at the front but contracts at the rear,
while the other reverses its companion by nar-
 rowing at the front and widening at the back
(fig. 173).52 When these effects are combined
with figures’ directional orientations, the
number of visually discrete figural variants
doubles from two to four, the last compris-
ing two right-facing versions (one with front
expanded and the other with front collapsed)
and two left-facing versions of the same kind.
Reasons to believe that left-right directional-
ity may have had significance come from the
tunics themselves53 as well as from many late
pre-Hispanic and contemporary societies in
the Andes, which, in line with habits of dual-
istic thought and social organization, routinely
accord meaning to left and right by associat-
ing them with the members of complementary
but opposed dialectical pairs, including male
and female and the parts of dually partitioned
communities.54
If this logic concerning distortion and
motif orientation is applied to other kinds of
tunics—such as those with the face-fret, the
profile creature, or the geometric motif with
bird heads—the number of variants doubles
again to eight since the motifs in these groups
appear not only in left-right orientations but
also rightside-up and upside-down. The same
is true of the skull and vertebra-like imagery
that appears on tunics with the meander motif;
all are rendered in at least two, often four, and
sometimes eight distinct variants if the effects
of distortion and orientation are taken into ac-
count. Thus, distortion may be involved in the
exploration of a series of numbers that today
is important to many realms of mathematical
inquiry: a geometric progression (or geometric
sequence) generated by multiplying successive
terms by a fixed number, known as the com-
mon ratio. In the tunics’ case the common ratio
is two and the sequence is two, four, eight, and
on. The tunics’ format often shows a similar
preoccupation.
Format
Two number systems seem to guide the tu-
nics’ compositional format, which in the great
majority of cases is based on the alternation of
plain and patterned vertical bands and, across
the latter, the horizontal alignment of motifs in
rows. Most commonly, the numbers of verti-
cal bands, horizontal rows, and motifs are
18 4 S usan E . Bergh
founded on the geometric progression rooted
in the number two. The tunic in Figure 155
offers an easy illustration: the two main pat-
terned bands together contain a total of sixteen
repeats of a winged attendant, eight on each
side of the tunic, or two sets of four, one to
either side of the center seam. Face-fret tunics
extend the progression; for instance, the tunic
in Figure 144 has four main patterned bands
and eight rows that generate sets of sixteen,
thirty-two, and sixty-four motifs on each of
the tunic’s sides. The same is true of profile
creature, profile bird head, and even the oddly
formatted meander tunics, which lack vertical
bands but invariably repeat their additional
imagery—the skulls, vertebra-like motifs, and
others—in numbers that correspond to the
progression. Pairs and quartets also seem to
be emphasized in several other ways, such as
the four-part mirroring of imagery over cross-
like vertical and horizontal axes (figs. 159,
160, 163). The second most common progres-
sion embedded in the tunics’ format seems to
be based on the number five and doublings to
ten, twenty, and forty. The incidence of the
two types of progression varies among tunic
groups; the first is by far the most common
except in face-fret tunics, which split about
evenly between the two. Other format-based
progressions are rarer and often occur in
tunics that are idiosyncratic in other ways.
Color
The tunics’ complex color also often plays with
the predominant set of numbers that seems
to guide format and motif variation.55 In most
tunic groups the repeated imagery is woven in
several standard color blocks or combinations
that repeat in very regular sequences down the
length of the vertical bands; the overwhelming
majority of tunics feature a total of four blocks,
although other numbers also occasionally oc-
cur. For instance, in face-fret tunics, the frets
typically are either red with a gold surround
or two shades of gold, and the faces are either
gold on a brown “ground” or pink on tan (fig.
144); the bodies of winged attendants and sac-
rificers usually appear in red and three shades
of gold to which the colors of details add fur-
ther distinctive character (fig. 153). Blue and,
less commonly, green sometimes substitute for
 other colors, especially tan and related pale
shades; this enhancement is one of the indica-
tions that, as during the Italian Renaissance,
blue was a rare and prized exotic.56
When traced across the body of the tunic,
the color blocks generate large-scale geometric
patterns, most based on diagonals that con-
tinue from one vertical band of motifs to the
next, skipping over the intervening solid band
as though it does not exist.57 In four-block tu-
nics the most frequent pattern has two sets of
diagonals, each a pair, that oppose or reverse
one another in direction. One pair rises from
right to left (fig. 176a) and interleaves with
another pair that rises from left to right (fig.
176b). In some tunics, the colors of the diago-
nals traverse the seam without color change,
while in others they shift at center seam and
in doing so consistently work in pairs, as Mary
Frame has phrased it.58 This opposed diagonal
pattern is best documented among face-fret tu-
nics, one of which serves as illustration, but it
also occurs in all other major groups. In some
tunics, including several with the profile crea-
ture, the directions of the diagonals reverse
for one step (motif) in the narrow band at each
side of the tunic.59
Another common pattern again divides
the four blocks into two pairs, but disposes
the pairs in two horizontally extended check-
erboards. The profile creature tunic in Figure
177 provides an example. Within each check-

Figures 176a, 176b. The 2 3 2 3 2 3 2 3 2 3 2 3

opposed diagonal color 4 1 4 1 4 1 4 1 4 1 4 1
pattern in a tunic with 2 3 2 3 2 3 2 3 2 3 2 3
4 1 4 1 4 1 4 1 4 1 4 1
face-fret motif. The diago-
nals formed by the faces 1
3 2
4 1
3 2
4 1
3 2
4 1
3 2
4 1
3 2
4 1
3 2
4
are shown on the left; the
fret diagonals are on the 1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2
right. To improve clar-
2 3 2 3 2 3 2 3 2 3 2 3
ity, not all diagonals are 4 1 4 1 4 1 4 1 4 1 4 1
colored. Graphic: Susan
2 3 2 3 2 3 2 3 2 3 2 3
E. Bergh and Amanda 4 1 4 1 4 1 4 1 4 1 4 1
Mikolic, based on a tunic 1 4 1 4 1 4 1 4 1 4 1 4
at the Los Angeles County 3 2 3 2 3 2 3 2 3 2 3 2
Museum of Art, M77.70.3. 1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2
2 3 2 3 2 3 2 3 2 3 2 3
4 1 4 1 4 1 4 1 4 1 4 1

2 3 2 3 2 3 2 3 2 3 2 3
4 1 4 1 4 1 4 1 4 1 4 1
1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2

1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2
2 3 2 3 2 3 2 3 2 3 2 3
4 1 4 1 4 1 4 1 4 1 4 1

2 3 2 3 2 3 2 3 2 3 2 3
4 1 4 1 4 1 4 1 4 1 4 1
1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2

1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2
2 3 2 3 2 3 2 3 2 3 2 3
4 1 4 1 4 1 4 1 4 1 4 1

2 3 2 3 2 3 2 3 2 3 2 3
4 1 4 1 4 1 4 1 4 1 4 1
1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2

1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2
2 3 2 3 2 3 2 3 2 3 2 3
41 1 4 1 4 1 41 1 4 1 4 1

2 3 2 3 2 3 2 3 2 3 2 3
4 1 4 1 4 1 4 1 4 1 4 1
1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2

1 4 1 4 1 4 1 4 1 4 1 4
3 2 3 2 3 2 3 2 3 2 3 2

18 5 Tapestry-Woven T unics
Figure 177. The paired
checkerboard color
pattern in a tunic with
profile creatures similar
to that shown in Figure
159. To improve clarity,
not all checkerboards are
colored. Graphic: Susan
E. Bergh and Amanda
Mikolic, based on a tunic
at the Museo Nacional de
Antropología, Arqueología
e Historia del Perú, Lima,
03565.
Three additional patterns are frequent
1 2 1 4 3 4
enough among four-block tunics to warrant
3 4 3 2 1 2 mention. One consists of opposed chevrons in
which two sets of chevrons, each a pair, lay out
4 3 4 1 2 1 in interleaved directional reversals, one point-
ing up and the other down (figs. 179a, 179b); al-
2 1 2 3 4 3
most invariably each chevron is monochrome.
Two others are based either on single-direction
1 2 1 4 3 4
diagonals (figs. 149, 164, 172) or single-direc-
3 4 3 2 1 2 tion chevrons.60 In the former, the colors of
two of the diagonals usually exchange at the
4 3 4 1 2 1
center seam, while the other two pass over the
center seam without color change; in the latter,
2 1 2 3 4 3
both legs of each chevron are usually the same
1 2 1 4 3 4 color. All three of these patterns are best repre-
sented in tunics with the face-fret and related
3 4 3 2 1 2 motifs although the last two sometimes occur
in other tunics as well.61
4 3 4 1 2 1

2 1 2 3 4 3 3 1 3 1 3 1 3 1 3 1
2 4 2 4 2 4 2 4 2 4

1 2 1 4 3 4 3 2 3 2 3 1 4 1 4 1
1 4 1 4 1 3 2 3 2 3
3 4 3 2 1 2
4 2 4 2 4 2 4 2 4 2
1 3 1 3 1 3 1 3 1 3
4 3 4 1 2 1
4 1 4 1 4 2 3 2 3 2
2 3 2 3 2 4 1 4 1 4
2 1 2 3 4 3
3 1 3 1 3 1 3 1 3 1
2 4 2 4 2 4 2 4 2 4

Figure 178. The X-shaped erboard the diagonally aligned color blocks 3 2 3 2 3 1 4 1 4 1
cross color pattern of the 1 4 1 4 1 3 2 3 2 3
can be imagined to form two opposed chev-
tunic with face-fret motif
shown in Figure 144. To rons that interlock with each other (one /\ and 4 2 4 2 4 2 4 2 4 2
improve clarity, not all the other \/ ). Here, too, the colors sometimes 1 3 1 3 1 3 1 3 1 3
diagonals are colored. change at the center seam as they do in the 4 1 4 1 4 2 3 2 3 2
Graphic: Susan E. Bergh 1 3 2 3 2 4 1 4 1 4
and Amanda Mikolic. illustrated example, but in other cases they do
not. This pattern also occurs in winged atten- 4 1 4 1 4 2 3 2 3 2

dant and meander garments. 2 3 2 3 2 4 1 4 1 4

The idea of opposed diagonals plays out in 3 1 3 1 3 1 3 1 3 1

2 4 2 4 2 4 2 4 2 4
a different way in another distinctive pattern
seen almost exclusively in face-fret tunics. 3 2 3 2 3 1 4 1 4 1
1 4 1 4 1 3 2 3 2 3
Here the color diagonals divide the tunic
into quarters by reversing direction at both 4 2 4 2 4 2 4 2 4 2
1 3 1 3 1 3 1 3 1 3
the shoulder line and center seam; in doing
so they create two sets of upward-oriented 4 1 4 1 4 2 3 2 3 2
2 3 2 3 2 4 1 4 1 4
chevrons, one on either side of the tunic,
that merge to form an X-shaped cross when 3 1 3 1 3 1 3 1 3 1
2 4 2 4 2 4 2 4 2 4
the shoulder is unfolded (fig. 178). The col-
ors usually shift at both axes, again working 3 2 3 2 3 1 4 1 4 1
1 4 1 4 1 3 2 3 2 3
in pairs; in a few tunics, however, the same
color continues across the center seam to form 4 2 4 2 4 2 4 2 4 2
1 3 1 3 1 3 1 3 1 3
monochrome chevrons.

18 6 S usan E . Bergh
Figures 179a, 179b. The
1 4 1 4 1 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1
opposed chevron color
3 2 3 2 3 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3
pattern of the tunic with
face-fret motif shown in 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1
Figure 55. The chevrons 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3
formed by the faces are
2 3 2 3 2 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2
shown on the left; the fret
4 1 4 1 4 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4
chevrons are on the right.
To improve clarity, not 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2
all chevrons are colored. 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4
Graphic: Susan E. Bergh
and Amanda Mikolic. 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1
3 2 3 2 3 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3

1 4 1 4 1 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1
3 2 3 2 3 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3

2 3 2 3 2 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2
4 1 4 1 4 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4

2 3 2 3 2 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2
4 1 4 1 4 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4

1 4 1 4 1 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1
3 2 3 2 3 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3

1 4 1 4 1 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1
3 2 3 2 3 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3

2 3 2 3 2 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2
4 1 4 1 4 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4

2 3 2 3 2 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2
4 1 4 1 4 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4

1 4 1 4 1 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1
3 2 3 2 3 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3

1 4 1 4 1 1 4 1 4 1 1 4 1 4 1 1 4 1 4 1
3 2 3 2 3 3 2 3 2 3 3 2 3 2 3 3 2 3 2 3

2 3 2 3 2 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2
4 1 4 1 4 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4

2 3 2 3 2 2 3 2 3 2 2 3 2 3 2 2 3 2 3 2
4 1 4 1 4 4 1 4 1 4 4 1 4 1 4 4 1 4 1 4

Whether the patterns have meaning is,
again, an open question. It could be that
the color variations have strictly aesthetic
underpinnings. If so, they are akin to musi-
cal variations in which static, repeated, and
threateningly wearisome formal elements—
here, the diagonal, the color blocks, the motif,
and format—take on satisfying complexity and
beauty through improvisations with harmony,
rhythm, counterpoint, and orchestration.
Certainly several commentators have observed
that the tunics’ riotous color often dominates
first impressions, as if attempting to divert
attention from the imagery, which sometimes
coalesces only slowly, especially when under-
mined by distortion. The art historian Rebecca
Stone has pursued this argument in regard to
color with particular vigor though different
emphasis.62 Thus, as with distortion, to over-
look aesthetic motivations as the source of this
aspect of intricacy is to risk diminishing the
creators’ artistry and the sophistication with
which they enlivened the tunics by investing
them with visual puzzles.
Still, the tunics are the largest, most
complex portable objects that the Wari made
and those on which Wari elites may have
most depended to convey their message. Also,
although the tunics’ color may strike our eye
as riotous, we are a twenty-first century audi-
ence to whom pattern analysis, developed
through lifelong immersion in textiles and
the mathematically based process of weaving,
is foreign. If my experience with the tunics

187 Tapestry-Woven T unics

 and the visual skills of contemporary Andean
people are any guide,63 the ancients had little
trouble sorting out the tunics’ color and seeing
great order in it. To some, this order suggests
that the color patterns have content,64 although
interpreting that content is perforce risky
because of the figural austerity of the patterns
and the relative nascency of Wari studies,
among other things.
Nevertheless, the textile specialist Mary
Frame has offered a provocative opening salvo
in interpretation by suggesting that the color
patterns represent a systematized code that
reflects a form of cultural knowledge and that
the patterns may embody schema or formulas
that had potential application to a wide range
of experience.65 More specifically, she suggests
that emphasis on four-color blocks may reflect
an ideal of four-part division and that, collec-
tively, the patterns might constitute a kind of
catalogue of social geometries that structured
human interaction in different settings or re-
gions: as alternating, opposed dyads (opposed
diagonals), twinned, alternating pairs (check-
erboards), linear sequences (single-direction
diagonals), and so forth. It is true that similar
kinds of geometries structure many kinds of
activities in the Andes today and also in antiq-
uity. The Inca habit of dual social organization
comes to mind, made relevant by the fact that
they divided their bipartite society into four,
after which they named their empire (Tawan-
tinsuyu, “The Four Parts Together”), and that
they mapped both bipartite and quadripartite
divisions onto the plan of Cuzco, their capi-
tal. Quadripartite division also characterized
other conquest-period Andean societies, some
of which subdivided their populations or
lands into eight and sixteen parts in further
expression of dual organization.66
Of course, any suggestion of equivalent
social organization among the Wari remains
speculative and, if the color patterns do hold
meaning, it could well be anchored in other
realms. In this regard an interesting albeit
contemporary example comes from the folk
astronomy of natives who today live in the
environs of Cuzco. These Quechua-speakers
recognize several X-shaped crosses in the
Southern Hemisphere sky. These include the
Southern Cross, the annual motions of which
18 8 S usan E . Bergh
bracket the agricultural season, as well as two
others traced by the solstitial risings and set-
tings of the sun and by the movements of the
Milky Way, conceived as a river of stars that
streams through the night firmament. They
also identify other types of starry celestial for-
mations, some of them chevron-shaped.67
Past and the Present
It has been proposed elsewhere in this volume
that Wari’s success—gauged by its impressive
building projects, its transformative impact in
many areas of the Andes, and the cachet that it
and its artworks seem to have acquired—came
in part from the belief that Wari lords pos-
sessed the ability to mediate human and cos-
mic affairs, to act both as intercessors in cru-
cial matters that shaped the lives of men and
women and as masters of the intersections that
those matters had with the realm of unseen
forces that sway them profoundly. If so, many
of the tapestry-woven tunics that these men
wore surely helped them to instantiate these
powers, particularly those thronged by hosts
of winged attendants and sacrificers, the staff
deity’s numinous companions. These tunics
imply that to a very great degree Wari elites’
authority derived from trust in their privileged
access to the sacred realm and its denizens.
Indeed, by donning such tunics Wari lords
may have identified themselves with or even
transformed into these figures. Or, since the
deity itself seems to wear a tapestry-woven
tunic, perhaps these elites, so-clothed and
standing before audiences with a potent staff
of sovereignty in each hand, fulfilled the role
not of the faithful vassal but of the deity itself,
its acolytes swarming in ordered registers on
either side of the body. The divine group may
even have provided a model for a hierarchi-
cally ordered human society,68 its apex defined
by the deity’s human associates.69
Beyond this broad affiliation with the
supernatural may have lain more specific
assertions embedded in the pairs of figures
routinely depicted in the tunics, which likely
testify to a creed of dualism. If this creed has
echoes in the thought of contemporary Andean
people, it was based on the conviction that the
world achieves existence through the ongoing
dialectical balance and fusion of two compet-
 ing but complementary principles. In highland
Andean communities today, harmony between
the two is achieved through the give-and-take
of reciprocity and the bonds of mutual obliga-
tion that it generates. The concept structures
social relations, with both assistance and
injury calling forth measured repayment, as
well as interactions with the natural world
and with the deities of the mountains and the
earth, which through offerings are induced to
behave according to rules of reciprocity.70 In
the words of the anthropologist Catherine
Allen,71 reciprocity is the essence of modern
indigenous Andean life, and the same seems
to have been true among the Inca.72 The insis-
tent dualism of some tunics, then, may testify
to a belief in Wari lords’ power to bring con-
flicting forces, both cosmic and human, into
synthesis and harmony and thus to guarantee
health, prosperity, and a foothold in the fu-
ture, matters that would have been lent special
urgency and force by droughts that plagued
the Andes just before Wari rose to power.73
The tunics’ color patterns, which so strongly
emphasize the even, balanced distribution
of paired directional oppositions or paired
alternations, may have amplified this message
whether or not they carried other meanings.74
We do not know the rituals of reciprocity with
which Wari lords sought to influence the forc-
es of nature, but in the world of human affairs
Figure 180 [156]. Khipu; they focused on feasting, the exchange of food
cotton; L. 190 cm (primary and drink, a crucial Andean tradition that re-
cord), 36 cm (longest
secondary cord). Private minded participants of the mutual obligations
collection. that in later times allowed family, community,
189 Tapestry-Woven T unics
province, and polity to survive and thrive (see
pp. 82–101, “The Art of Feasting”).
The number systems that the tunics may
persistently explore—most prominently the
geometric progression moored on two and dou-
blings to four, eight, and beyond—also bespeak
an interest in mathematical concepts as do the
various symmetries of the plane that guide
motif distribution across the body of the tunic,
a fascinating aspect of composition that has
not been explored in these pages.75 Although
the state of research does not now allow the ar-
gument to be developed in detail, this concern
with numbers and their interrelations suggests
that, as others have observed,76 there is an
evolutionary continuum between textiles and
the khipu, a fiber recording device comprised
of cords that is the Andes’ closest approach
to writing. The khipu (fig. 180) is best docu-
mented among the Inca, who used it to record
many kinds of numerically based informa-
tion—statistics including censuses and tribute
accounts as well as such narratives as histories
and genealogies—via knots and distinctions of
cord color, construction, and attachment.77 The
code of the Inca khipu has been cracked inso-
far as its numbers and their hierarchies can
be read, but the memory of what the numbers
refer to died with the ancient khipucamayoqs
(khipu makers).
Examples of a distinctive kind of khipu,
its cords’ upper reaches wrapped with col-
ored yarns, have been radiocarbon dated to
the Middle Horizon (fig. 180; see also [155], p.
276).78 Although cultural attribution cannot
 be confirmed for those without archaeologi-
cal context, a few have been found with Wari
ceramics and may be Wari.79 In comparison to
Inca examples, much less is known about the
ways in which these earlier khipu encipher
information but it is likely that they are also
concerned with recording numbers. If in the
tunics distortion redundantly joins other sys-
tems of composition in expressing a concern
with numbers and their sequences, it holds
important implications for the understanding
of the indigenous roots of abstraction, which
has sometimes been explained in a way that
takes inspiration from one strand of twentieth-
century modernism: as an artist’s experiment
with form undertaken for its own sake. It may
be that the tunics’ great aesthetic complexity
instead has its source in the realm of math-
ematics—an interest in numbers and geometry
ingeniously translated into form and color by
fiber artists who are among the world’s most
inventive.

notes 1. Wari and its contemporary
Tiwanaku made remarkably
similar tapestry-woven tunics.
Amy Oakland Rodman identifies
a number of technical features
that distinguish the two (Oakland
1986a; Oakland 1986b, 31–41,
230–31; Rodman and Cassman
1995; Rodman and Fernández Lopez
2001). This dichotomy, today widely
accepted, has guided the selection
of tunics for this project, although
I will not be surprised if future
research forces some refinement of
current understanding. See Bergh
forthcoming; Bergh 1999, 6–7,
72–100.
2. Conklin 1978.
3. See Phipps (2004b, 21–25) and J.
Rowe (1979, 239–41) for cumbi, a
category that may have consisted
entirely of tapestry-woven cloth but
may also have encompassed other
kinds of fine fabric, such as feath-
ered cloth.
4. Murra 1962, 711.
5. A. Rowe 1996, 330.
6. Kidwell 1976, 28.
7. Murra 1962, 722.
8. See A. Rowe (1978) and J. Rowe
(1979) for Inca tunics. Tiwanaku
tapestry-woven tunics also form
part of the background of the Inca
successors; see Cummins (2002,
59–68) for the general importance of
Tiwanaku to the Inca.
9. See Emery 1966, 76–90, for more
about the tapestry weave.
10. The averaged thread counts on
which this rough, conservative cal-
culation is based are, for a routine
Wari tapestry-woven tunic, 50 wefts
and 12 warps per centimeter and, for
the finest example, today in a private
collection, 124 wefts and 24 warps
per centimeter.
11. Phipps 2004b, 24.
12. Zuidema 1992, 179.
13. Frame 1990.
14. Sawyer 1963, 3.
15. Menzel (1968; 1964) refers to the
face-fret motif as the split face.
16. Ochatoma Paravicino and
Cabrera Romero 2001, 202.
17. For these tunics, see Bergh
(1999, cat. 309, 310); the latter has
been radiocarbon dated to cal. AD
685–770 (Haeberli forthcoming).
18. But see Frame forthcoming;
Frame 2005, 9–11; Frame 2001,
120, 128–30; Posnansky, cited in
Goldstein 1989, 154; Stone-Miller
1995, 125, 132.
19. Conklin 1996, 383–89; Menzel
1968, 79.
20. See Cook 1996 for a contrasting
opinion.
21. Bergh 1999, 510–60.
22. Ibid., 117–44, 446–509. The
tunics also feature several hybrids
of these two types as well as a few
apparently human figures. See Bergh
1999.
23. Bergh 2009.
24. Bergh 1999, 117–44.
25. See A. Rowe 1979 for musician
iconography in Wari tapestry-woven
tunics.
26. The literature on dualism in
ancient and contemporary Andean
societies is very extensive. See
Bergh 1999, chap. 3, for a partial
review and bibliography.
27. For instance, Lévy-Strauss, cited
in Moore 1995, 176; Urton 1993.
28. Lyon 1978, 108–13; Menzel 1977,
54; Menzel 1964, 19, 26; see also A.
Rowe 1991, 116–18; A. Rowe 1979,
11.
29. Murra 1968.
30. Duviols 1979; Pärssinen 1992,
200–27; J. Rowe 1946, 202; Zuidema
1964.
31. But see Anders (1986, chap. 11)
for a speculative reconstruction of
Wari as a dual and quadripartite
empire and J. Topic and T. Topic
(1992; 2001) for their opinion that
dual organization was endemic to
Wari.
32. Bergh 1999, 561–84.

19 0 S usan E . Bergh
 33. See Conklin (1996, 398; 2004c)
and Sawyer (1963, 2) for their specu-
lations concerning the identity of
the motif and its components.
34. If the panel is a mantle, the ori-
entation of the imagery in its borders
indicates that it was not worn with
the upper edge folded down; this
contrasts to custom in later Andean
times. It is difficult to establish Wari
practice because bordered, tapestry-
woven constructions that could
have served as mantles are not very
common. In those that exist, the
imagery’s orientation would have
accommodated folding. The small
panel may well have served another
purpose.
35. Knobloch 2010, fig. 21; see also
Bergh 1999, fig. 86.
36. Gose 1993.
37. Bergh 1999, 606–12.
38. Rather than being stitched onto
the body of the tunic, the sleeves
were woven as extensions that are
continuous with it.
39. Bergh 1999, 585–605; Bergh
forthcoming.
40. See Young-Sánchez (2000) for
an analysis of the Ancón tomb. As
she points out, the tunic had been
recycled into this context.
41. Other routes of transmission are
possible, especially via the coastal
Nasca culture. See Bergh forthcom-
ing.
42. See Richard Burger (1988,
130–31) for his suggestion that the
same was true of the awe-inspiring
arts of the earlier Chavín culture.
43. However, distortion occurs in a
few Tiwanaku tapestries, perhaps
as a consequence of contact with
Wari (Rodman and Fernandéz Lopez
2001, 125).
44. Technically stated, the rules
of the system call for expansion of
those parts of the motif closest to
the long edge of the panel at which
weaving began; this edge normally
falls along the tunic’s center seam.
It follows that contraction affects
the elements closest to the panel’s
finishing edge, generally at the side
seams. A handful of tunics reverse
the positions of these edges and,
therefore, the direction of distortion.
See Bergh 1999, 47–48.
191 Tapestry-Woven T unics
45. Paternosto 1996, 227.
46. Sawyer 1963, 37.
47. Ibid.
48. Stone-Miller 1995, 148.
49. Pasztory 2010, 133–34.
50. A. Rowe 1996, 402; A. Rowe
1979, 18n28; Sawyer 1963, 8, 11.
Those who agree with a formalist
interpretation include Bird and
Skinner (1974, 11), Rodman and
Fernandéz Lopez (2001, 121), and
Tulchin (1997). For other interpreta-
tions, see Frame (2001, 130), Gayton
(1961, 127), and Conklin (2004c,
179), whose analysis W. Isbell (2002,
456) critiques.
51. See Pasztory 1990–91.
52. Stone-Miller (1995, 147) makes a
similar observation.
53. Bergh 1999, 144–48.
54. The literature in this regard is
very large. See Bergh 1999, chap. 3,
for a partial review.
55. Bergh 1999, 54–66.
56. Baxandall 1989, 11–15; Bergh
2009; Stone 1987; Stone-Miller 1995,
148; Stone-Miller 1992b.
57. Bergh 1999, 54–66; Frame forth-
coming. The following patterns can
be classified in more than one way.
58. Frame forthcoming.
59. See Bergh 1999, fig. 9.3.
60. Ibid., figs. 11, 12.
61. There are other rare color pat-
terns, including several found in
tunics with two-color blocks; see
Frame forthcoming for several of
them.
62. Stone 1987; Stone 1986; Stone-
Miller 1995; Stone-Miller 1992b.
63. For instance, Franquemont
et al. 1992; Franquemont and
Franquemont 1987.
64. Bergh 1999; Frame forthcoming.
65. Frame 2005, 9–12; Frame forth-
coming.
66. The literature on Inca and other
late pre-Hispanic socio-political
organization is extensive. See Bergh
(1999, 203–15) for a partial review
and bibliography, to which D’Altroy
(2002), Morris and von Hagen (2011),
and Pärssinen (1992) can be added.
67. Silverman-Proust 1988, 226;
Urton 1981. See Zuidema (1977)
and Martinez (1987) for calendrical
interpretations of a Wari tapestry-
woven mantle.
68. Cook 1984–85; W. Isbell 1984–85.
69. See Knobloch (2010) and
Makowski Hanula (2009) for dissent-
ing opinions.
70. Allen 1988, 93–94, 187; Earls and
Silverblatt 1978, 310; Mannheim
1991, 90–93.
71. Allen 1988, 91, 93, 208.
72. Cummins 2002.
73. Williams 2002, 365.
74. Bergh 1999.
75. See Frame (forthcoming) for
charts of some of these symmetries
and her ideas about how they relate
to fiber technologies.
76. See Bergh 1999, 3–4, 34; Conklin
1986, 126; Conklin 1982, fig. 3;
Frame 2005, 8–9; Frame 2001, 123;
Frame 1991, 145–47; Stone-Miller
1992a, 337.
77. See, for instance, Ascher and
Ascher 1981; Mackey et al. 1990;
Quilter and Urton 2002; Urton
2003a; Urton 2003b.
78. For instance, one khipu returned
the date AD 719–981 (fig. 180) and
another, cal. AD 690–900 ([155], p.
276). My thanks to Gary Urton for
sharing the dates of the khipu in
[155] and several other wrapped-
cord khipus at the American
Museum of Natural History, all of
which belong to the Middle Horizon.
79. William Conklin describes the
khipu that appears in Figure 180 in
detail, mentioning that Yoshitaro
Amano found similar wrapped-cord
khipu fragments in a tomb at Pampa
Blanca in the Nasca drainage,
with Wari ceramics of the Middle
Horizon’s second epoch (Conklin
1982). Another example was found
in architectural fill at the Huaca San
Marcos in Lima in association with
Pachacamac, Lima 9, and Nievería
style ceramics; the context suggests
that the khipu was manufactured
before or during the second epoch
of the Middle Horizon (Shady Solís
et al. 2000). Ruales (2001, 371–72)
reports the find of a khipu-like
object, apparently without wrapped
cords, in a Middle Horizon context
at Cerro de Oro in the Cañete Valley.
See Conklin 2011 for reflections on
wrapping.
Ann Pollard Rowe
Figure 181 [134]. This tie-
dyed tunic is one of the few
examples in original condi-
tion. Tunic with stepped
blocks; camelid fiber; 86.5
x 122 cm. The Textile
Museum, Washington, DC,
acquired by George Hewitt
Myers in 1941, 91.341.
193
Tie-dyed Tunics
Wari tunics with vivid tie-dyed colors have
an immediate visual appeal.1 While these
spectacular fabrics are common in collec-
tions, most are fragmentary and since compil-
ing pieces from one or more originals into a
new composition is easy (whether in ancient
or modern times) their original format and
composition have often been lost. The focus
here is therefore primarily on the more intact
examples, emphasizing those I have studied
firsthand (chiefly in the Textile Museum and
the Metropolitan Museum of Art) in order to
clarify the original format and design prin-
ciples to a greater extent than has previously
been possible. An examination of the colors
also provides new clues about the origin and
diffusion of these striking textiles, and de-
tailed consideration of archaeological associa-
tions highlights the textiles’ affiliation not just
with Wari but, more precisely, with an impor-
tant faction of men within the Wari hierarchy.
Weaving and Construction
The examples associated archaeologically
with Wari’s presence are consistent in struc-
ture and pattern. They are fascinating because
of their complex method of manufacture,
which involved weaving, disassembly into
individual parts, tie-dyeing the parts, and
then re-assembly to produce color patterns
with sharp edges not normally possible with
tie-dyeing. The fabric was apparently wo-
ven in strips or small sections using undyed
camelid-fiber (hair) yarns; each strip or sec-
tion consisted of several pattern units attached
to one another by temporary yarns that passed
alternately through the loops of warp along
the upper and lower edges of each pattern
unit, creating a dovetailed join.2 The tempo-
rary yarns were withdrawn after weaving,
which separated the many individual pattern
units. The units were then tie-dyed3 with the
fabric bunched and bound tightly with yarn in
order to prevent the dye from penetrating the
fibers under the bindings. Removing the bind-
ings revealed a pattern (typically of hollow
diamonds) in the original, lighter color of the
fabric. The pattern units were then arranged
in the final color alternation, and new yarns
made of undyed camelid fiber were inserted
along the horizontal edges (again dovetailed)
to hold the pieces in their new arrangement.
The vertical slits between color areas were
sewn closed with overcast stitches in dyed
camelid fiber yarn, and the strips of pattern
units were sewn together in the same manner.
The individual units are not always exactly
the same size, so a larger unit was gathered
slightly to be sewn to a smaller one. The tech-
nique lends itself to the manufacture of mul-
tiple garments at once, with each step conceiv-
ably done by a different person. In such a case,
pattern units linked during weaving might not
end up in the same tunic.
The weaving method results in cloth in
which both the warp and the weft are discon-
tinuous, which simply means that no yarn
runs continuously from top to bottom or from
side to side. Although a warp that runs the
full length of the loom is a major labor-saving
device, some of the most prestigious Peruvian
textiles have warp yarns that wind back and
forth only within a single color area. The last
2.5 centimeters (1 inch) of each pattern unit in
discontinuous-warp fabrics cannot be woven
in the usual way but must be darned with a
needle. Discontinuous weft yarns, turning at
the edges of each color area, are also used in
tapestry weaving, but tapestry is a denser, less
flexible fabric.

Figure 182. The side seams The structure is plain weave that is usu-
of this tie-dyed tunic
ally warp-predominant. The yarns (Z-spun
have been opened. Tunic
with stepped blocks; and 2-plied S) and weaving are not especially
camelid fiber; 187 x 114 fine: warp counts range from 9 to 20 per cm
cm. Museum of Fine Arts, (23 to 51 per in.) and weft counts from 7 to 12
Boston, Textile Fund and
Helen and Alice Colburn per cm (18 to 31 per in.).4 The end selvedges
Fund, 1983.252. Photo: © have a fringe consisting of groups of warp
2012 Museum of Fine Arts, loops (twisted Z), usually about 1 cm (3/8 in.)
Boston.
long.
Color Use
The overall color scheme in these discontin-
uous-warp-and-weft pieces is consistent—six
color combinations routinely appear—and
color pairing is also consistent (fig. 181). Thus,
a unit with white diamonds tie-dyed on a
red ground is paired with one having yellow
diamonds on a green ground. In a second com-
bination, white diamonds on a medium blue
ground appear with white and red diamonds
on a purple ground. The purple units were
made by first binding and dyeing the fabric
19 4 A nn P ollard Rowe
red (yielding the white diamonds) and then by
additional binding and overdyeing (yielding
the red diamonds).5 The third pairing consists
of white diamonds on a darker blue ground
and a unit that is red and yellow with larger-
scale patterning. Unfortunately, the dyes have
not yet been analyzed, although the red is
likely to have been cochineal, an insect dye
still cultivated commercially in the Ayacucho
area.6
Aesthetically the red-and-yellow pattern
units break up the repeating small diamonds
and add an explosive effect, a brilliant artis-
tic touch. In a few examples, the diamonds
are yellow and the ground is red, the normal
result of binding a yellow fabric and dyeing
it red,7 comparable to the green units. More
commonly, however, the opposite effect oc-
curs, with red diamonds on a yellow ground
(figs. 181, 182). In these cases, it appears that
the background was bound, not the diamonds,
and indeed there is telltale leakage of red dye
into the background in some examples. In oth-
er cases, the red diamonds are painted on (fig.
183),8 while in still others, they are absent (fig.
184), either not there to begin with or fugitive.9
In a few examples the pattern is so large that
the distinction between ground and pattern
is obscured (fig. 185). The idea of binding the
ground areas was evidently so unusual that
the technique was not obvious to some dyers,
who attempted to achieve the correct overall
effect by other techniques.
Because the blue pattern units are often
in poor condition, they have frequently been
patched or replaced (fig. 185). Presumably the
dye was indigo, which requires an alkaline
bath that tends to dissolve protein fibers like
camelid hair.10 Although it is possible to dye
protein fibers with indigo (if the alkali is not
too strong and the cloth is not dipped for too
long), and there are south coast textiles of
various periods with blue camelid-fiber yarns,
they are not common in the rest of the coast
or in other Wari textiles. Other Wari tie-dyed
tunics evidently employed a different, fugitive
blue dye.11 In the example shown (fig. 183),
the original purple has changed to dark red
(with lighter red and white diamonds), the
blue units have become khaki, and green units
have turned yellowish (with faint lighter yel-

Figure 183. Fugitive blue
dye was used in this strip,
said to be from Corralones
in the Acarí Valley. Strip
from a tunic; camelid fiber;
186 x 19 cm. The Textile
Museum, Washington,
DC, Anonymous Gift,
1966.7.164.
Figure 184. Tunic with
squares within squared
meanders; camelid fiber;
90.17 x 114.3 cm. Los
Angeles County Museum
of Art, M75.50.1. Digital
image: © 2012 Museum
Associates/LACMA.
Licensed by Art Resource,
NY.
19 5 T ie - dy ed T unics
low diamonds). The red-on-yellow diamonds
are always painted in this group.
The variations in both the red-and-yellow
units and the blue units indicate that some ex-
amples were made by people who did not fully
understand the original dyeing technology,
which probably came from the south high-
lands where indigo dyeing on camelid fiber
was common. Thus, the variations strongly
suggest that the textiles were made in several
different places, each with a different solution
for handling the unfamiliar techniques.
Garment Type and Patterns
The most complete of the discontinuous-warp-
and-weft tie-dyed fabrics appear to be men’s
tunics, a suggestion corroborated by ceramics
depicting men wearing such tunics. They are
slightly wider than high, a different shape
than Wari tapestry-woven tunics, which are
roughly square. The most intact tie-dyed tunic
in the sample measures 86.5 cm high by 122
cm wide (34 by 48 in.), which appears to be
typical (fig. 181).12 Its side seams are overcast,
like the other vertical seams, and there is no
special finish on the neck slit or armholes.
Although examples have sometimes been
identified as mantles (large, rectangular
shawl-like garments), no distinctive mantle
format or size can be identified in the avail-
able sample. It is, however, a common and un-
fortunate modern practice to remove the side
seams of tunics in order to allow the fabrics to
be displayed flat. Some of these flat pieces are
slightly narrower than those with surviving
side seams. It is conceivable that some of these
narrower examples are mantles, but it seems
more likely that they are tunics now missing
columns of pattern units, or are simply within
the range of tunic sizes.
The different shapes of pattern units
present different artistic options. Any other
meaning they may have had in antiquity is
unknown. Pairs of stepped blocks are the most
common pattern (figs. 181, 182). The opened-
out example has the remains of side seams
and the same number of rows (ten, counting
both sides) and columns (six) as the seamed
example, although it is slightly longer and nar-
rower.13 A similar substantially complete piece
(opened and with repairs in the neck area) is
also known.14
The diagonal rows of small diamonds in
each unit combine with those of neighboring
units to form larger diamonds, and the colors
repeat along diagonal axes. The color units
typically rotate from one diagonal row to an-
other, adding kaleidoscopic variety (fig. 181).
In some examples, rotation of the color units
occurs in the same diagonal, and some have
additional tie-dyed horizontal lines, adding to
the shimmering effect (fig. 182). The diagonals
reverse direction at the shoulder of an eight-
row example;15 a five-column piece is likely
incomplete.16
Another standard design consists of
square blocks within vertical meanders made
of solid-color strips (fig. 184).17 The width of
Figure 185. One row may
be missing from this tie-
dyed tunic, which is said
to have been found near
Palpa in the Nasca drain-
age. Tunic with squares;
camelid fiber; 75 x 116.5
cm. The Textile Museum,
Washington, DC, 91.308.
the garment accommodates three pairs of
meanders that change color along the shoulder
line. The small diamonds typically form an X
in each block; placed side by side, these blocks
coalesce into a larger diamond. Some of these
pieces are still sewn up the sides, creating
tunics of dimensions similar to those with
stepped blocks. Several complete or nearly
complete examples have been published.18
Other tunics feature simple, square blocks
arranged in diagonal color repeats. One
example is aberrant in having five and a half
rows and twelve columns (fig. 185), for a total
of eleven rows counting both sides of the tu-
nic.19 An uneven number of rows is uncharac-
teristic of pre-Hispanic tunics and the length
of this piece is also slightly short, suggesting
that a row might have been removed. But it is
unclear how this might have been done since
the diagonal repeat is consistent except in one
corner (not shown). The small diamonds form
an X in each block, and it is obvious that the
blue and purple blocks with diagonal rows of
diamonds were taken from another tunic and
added to replace the missing dark blue units,
probably in modern times, since they are not
dovetailed. Because some of these replacement
blocks occur on the side seams and armholes,
19 6 A nn P ollard Rowe
it appears that the present side seams and
armhole fringe are part of the same repair.
Armhole fringe is not usual in Middle Horizon
tunics, and the colors here do not match those
in the tie-dye. The repair is skilled, but it is
possible to ascertain that the original tunic
probably had six rows of blocks on each side
and diagonal color repeats.
The number of blocks in a column or row
varies in other examples, with the size of each
block adjusted to fit the general size of a tunic.
An opened-out example with fugitive blue
has twelve rows and six columns of squares
patterned with diagonal rows of diamonds.20 A
fragment with only one column remaining has
ten rows and accordingly only one shade of
(fugitive) blue (fig. 183).21 Another opened-out
example has fourteen rows and ten columns
of X-patterned rectangles and is slightly nar-
row.22 Its diagonal repeat changes direction in
the center.
The spectacular piece in Figure 186 has
clearly been altered from a tunic into a man-
tle.23 The red bands, now on the outer edges,
probably originally ran down the center, as
shown in some ceramics (see fig. 110). This
center band was cut in half up the middle and
moved to the outside edges; the raw cut edges
Figure 186 [135]. Tunic
with squares and solid
strips, converted to a
mantle in modern times;
camelid fiber; 182 x 112.5
cm. The Textile Museum,
Washington, DC, acquired
by George Hewitt Myers in
1931, 91.90.

197 T ie - dy ed T unics
Figure 187. Tunic fragment
with opposed hooks; cam-
elid fiber; 83.5 x 151 cm (as
mounted). Private collec-
tion. Photo: Renée Comet
Photography.
were overcast with a camelid-fiber yarn of
uneven color (as if unraveled from cloth dyed
after weaving) except in the center where the
neck slit was originally located. The neck slit
was made using discontinuous-weft yarns, so
it has selvedge edges. Neither the cutting nor
this type of overcasting is likely to be ancient.
The vertical seams are sewn with dark blue
camelid-fiber yarn, except for the new center
seam, which uses tan cotton yarn, an obvious
mismatch. The small diamonds make Xs in
the rectangles. Interestingly, natural golden-
tan camelid fiber has been used instead of a
yellow dye. A related more fragmentary piece
has stepped blocks alternating with plain
vertical strips, and one strip has part of a neck
slit made with discontinuous weft yarns.24
Another tunic combines square blocks
and stepped blocks, both with diagonal rows
of small diamonds (see fig. 17).25 The pattern
changes along the shoulder as well as along
the center seam. Although the rotation of the
stepped blocks is not completely regular and
the rows of small diamonds do not consis-
tently join to make larger ones, there is no sign
of tampering (the dovetail joins appear intact)
and the color repeat is consistent on the diago-
nal (except in one corner).
19 8 A nn P ollard Rowe
I have not examined any intact tunics
with hook patterns. One published example,
opened out, with six columns and eight rows,
has the same proportions as the tunics dis-
cussed above.26 The color repeat is regular
with hooks of the same color rotating in the
same diagonal, and the tie-dyed diamonds
align in horizontal rows. In another example,
hooks of each color are in the same orienta-
tion in each diagonal but rotate in adjacent
diagonals (similar to fig. 181), though with
some irregularities that might not be origi-
nal.27 A portion of the original diagonal repeat,
without rotation, is apparent in a fragmentary
piece mounted with other colors substituted
for damaged blue units (fig. 187).28 A few small
fragments have hooks with diagonal rows of
small diamonds.29
Only a few fragmentary examples have
opposed L-shaped units with diagonal rows
of small diamonds. One opened-out example
(fig. 188), nicely mounted so that both the
losses and the original format are obvious, has
six columns, eight rows, and a diagonal color
repeat similar to that in Figure 181.30 A half
tunic features a column of rectangular blocks
to either side of four columns of L-shaped
pairs; the color repeat includes rotation in the
Figure 188. The blue
units of this Ica Valley
fragmentary tie-dyed
tunic are almost entirely
deteriorated. Fragmentary
tunic with opposed L
shapes; camelid fiber; 155
x 116 cm. Ethnologisches
Museum, Berlin, VA
29101a–m.
Figure 189. Detail of frag-
ment with squares within
S shapes; camelid fiber;
W. 34 cm. Fowler Museum
at UCLA, Los Angeles,
X86.3956.

same diagonal.31 Two-thirds of a tunic with fu-
gitive blue and an end border of square blocks
has embroidery on the neck slit and surviv-
ing armhole, presumably done by the original
owner rather than the producer.32
Another pattern has squared S-shaped
units with inset square blocks, all tie-dyed,
with the small diamonds in horizontal rows.
This pattern is represented only by three re­
assembled fragments with some diagonal color
repeats (fig. 189).33
Other pieces combine tie-dyed and solid-
color pattern units, though none is complete or
in its original form. The most intact example,
although narrow and missing a fringed row at
each end, combines sections of tie-dyed hooks
and plain stepped blocks, each with four rows
and four columns, in a checkerboard arrange-
ment.34 Another fragmentary example com-
bines tie-dyed stepped blocks and solid-color
hooks (fig. 190).35 The longer-than-usual warp-
loop fringe is uniform, confirming that the

Figure 190. Fragment with

stepped blocks and solid-
color hooks; camelid fiber;
68 x 164 cm. The Textile
Museum, Washington, DC,
91.302.

19 9 T ie - dy ed T unics
Figure 191 [136]. Tunic stepped blocks and hooks are from the same
said to be from Chilca original, but the vertical seams may be mod-
on the central coast;
ern since they are sewn with an interlacing
camelid fiber; 87 x 124.8
cm. The Textile Museum, stitch rather than overcasting and the color
Washington, DC, Gift of repeat is irregular.
Leo Drimmer-Lichtemberg,
1965.40.43.
Documented Find-spots and Dating
Tie-dyed tunics are a coherent group techni-
cally and stylistically, and can be confidently
described as Wari in style. Fragments have
been found in all parts of the coast under Wari
influence, and several are from sites of known
Middle Horizon date. Radiocarbon dates of the
textiles themselves also verify a Wari associa-
tion. One example, from Acarí, is reported
as 720–90036 and another yielded a date of
20 0 A nn P ollard Rowe
671–862 (calibrated, 95.4 percent probabil-
ity; see fig. 187).37 The following summary of
find-spots, more comprehensive than previous
attempts, is organized geographically from
north to south. The greater number of finds
from the south coast may be attributed
primarily to more favorable preservation
conditions.
north coast. The best association is from
excavations at Huaca Cao in the Chicama
Valley of an unusual burned burial that also
contained a Wari tapestry-woven tunic frag-
ment, on top of which was a secondary burial
with ceramics of Middle Horizon Epoch 2B.38
Max Uhle also found several fragments at
 Chimú Capac in the Supe Valley, a site dating
to Epochs 2B and 3.39 Heiko Prümers made
surface collections that included a tie-dyed
fragment and a tapestry-woven imitation of a
tie-dyed pattern at the Middle Horizon site of
El Castillo in the Huarmey Valley.40
central coast. Uhle found two fragmentary
tie-dyed tunics of the Wari style in his Pacha-
camac excavations. One was on the same
mummy bundle as a Wari tapestry-woven tu-
nic from his Cemetery III, which also included
ceramics of Epoch 2B.41 He recorded no asso-
ciations for the other piece.42 Uhle also found
some coarse tie-dyed fragments of aberrant
pattern at Pachacamac, plausibly suggested to
be provincial Wari imitations of the standard
style.43 From these fragments, Ina VanStan was
able to reconstruct a tunic with rows of square
blocks in a diagonal repeat, with plain strips
at the center and on each side.44
A fragment of the usual kind was also
found without specified associations at Huaca
Malena in the Asia Valley, a site with a large
Middle Horizon component.45 Fragments have
also been reported from Chancay and Ancón.46
south coast. Heinrich Ubbelohde-Doering
found fragments at Usaca and Copara, both
on the Trancas River in the Nasca drainage.47
Giuseppe Orefici reports fragments from
Quemado, a site on the road from the high-
lands to Nasca.48 Uhle also found a fragment
in the Yauca Valley.49 Tiffiny Tung excavated
fragments from a disturbed context at the
provincial Wari site of Beringa in the Majes
Valley.50 Others have been reported from the
Ica valley, near Palpa in the Nasca drainage
(fig. 185), Santa Cruz in the Nasca drainage,
and in Acarí (fig. 183).51 A mummy exhibited
in the Museo Regional de Ica wears a dove-
tailed-warp tie-dyed tunic and a headcloth
embroidered in the south coast style of Middle
Horizon 2.
Confusion has, however, resulted from
Hans Disselhoff’s find of such a tunic at the
earlier Nasca 3 site of Cabezas Achatadas, near
Camaná in the Majes Valley.52 Unfortunately
he did not report the exact associations, mak-
ing it impossible to assess this anomaly ad-
equately, so it seems necessary to set it aside.
201 T ie - dy ed T unics
Other Tie-dyed Tunics
Wari tie-dyed tunics made with continuous
warp and weft have not often survived. One
is said to have been found with a striped Wari
tunic and a four-cornered pile hat at Chilca,
between the Mala and Lurín Valleys on the
central coast (fig. 191).53 Although warp-faced
and lacking fringe, it is similar in size and
proportions to the discontinuous-warp-and-
weft tunics but made by sewing two large
panels together with an interlacing stitch.
Uhle found fragments of two other tie-dyed
red camelid-fiber tunics in Middle Horizon
contexts on the central coast, one at Nievería54
and the other at Pachacamac.55 The latter has a
loom width of 43 cm (17 in.), which is narrow.
The tie-dyed diamonds are larger than those
on the discontinuous-warp-and-weft tunics.
One must be cautious about supposing a Wari
connection for plain tie-dyed tunics without
similar context, since tie-dyeing on camelid-
fiber tunics occurred at other times on the
south coast and south highlands, especially in
the Early Intermediate Period (AD 1–600).
Not all discontinuous-warp-and-weft
tie-dyed textiles are Wari either. Some differ
in structure and design from those described
above and they also vary among themselves.
None has any archaeological association
information. One group is made with rect-
angular pattern units that are usually sewn
rather than dovetailed together (fig. 192).56
The tie-dyed diamonds form either an X or
horizontal rows in each rectangle; there are
no partial diamonds, nor any of different scale
than the others. The ground colors are usu-
ally red, purple, and green or blue-green, not
yellow. Most examples have been mounted as
large rectangular panels, but one seemingly
complete tunic with ten columns and ten rows
measures 123 x 140 cm (48½ x 55 in.), larger
than the Wari pieces.57 None has any fringe.
Because this group has previously been
considered a variation of the Wari tie-dyed
style, we requested a radiocarbon analysis of
one example (fig. 192). The resulting date was
AD 414–575 (calibrated, 95.4 percent prob-
ability), which clearly pre-dates the Middle
Horizon. Possibly these pieces are from far
southern Peru where other large, predomi-
nantly camelid-fiber textiles have been found
Figure 192. Panel with
sewn tie-dyed rectangles,
probably from southern
Peru; camelid fiber; 232.5
x 146.5 cm. Private collec-
tion. Photo: Maury Ford.
in recent years. Some are described as Nasca
style, but they differ from those of the Nasca
drainage; a more accurate description might
be Southern Nasca style. It is tempting to sup-
pose that these pieces represent an antecedent
to Wari tie-dyed tunics, but more archaeologi-
cal information would be helpful. If they are,
then perhaps Wari tunics with X-patterned
blocks are more conservative (earlier) than
those with diagonal patterning.
Ceramic Representations of Tie-dyed Tunics
Some ceremonial ceramic fragments from
Conchopata that date to Epoch 1 depict a man
wearing a tunic with block patterns that might
represent discontinuous warp and weft; each
block contains a solid circle that could repre-
sent a tie-dyed pattern (see fig. 103). This fig-
ure, who may represent either an individual or
a group (such as an ethnic group), is defined
by his distinctive face paint and also wears a
hat with a diamond-patterned lower border.
He is among several others carrying a bow,
arrows, and a shield, and kneeling on what ap-
202 A nn P ollard Rowe
pears to be a boat made of totora, a kind of bul-
rush; the boat suggests a possible connection
to Lake Titicaca, in the area of Tiwanaku in-
fluence to the south, where such boats remain
in common use.58 This figure, with the same
face paint, hat, and a tunic more obviously
representing tie-dyed discontinuous warp and
weft, is prominent among the faceneck jars in
the Pacheco offering deposit of Epoch 1 (see
figs. 135, 136), and is also represented on an
impressive leather bag, probably dating to the
same period (see fig. 18). Several scholars have
identified him as a key figure in the formation
and spread of the Wari religion.59 The associa-
tion with the sacred seems to be underlined
by the supernatural being depicted on the
Pacheco cups, which appears to wear a similar
tunic (see fig. 133).
In Epoch 2 ceramics, this figure’s costume
changes to a four-cornered hat and a tunic
with vertical stripes. The patterned stripes
may have either a tie-dyed hook design (see
fig. 110), or the face-fret motif common in
tapestry-woven tunics of the period (see fig.
146).60 The Wari affiliation of the tie-dyed
tunics with hooks is verified by a ceramic
fragment with this design from excavations at
the Wari capital.61
Since the figure appears to be riding on a
totora boat, we may look for related tie-dyed
tunics in the art of Tiwanaku, especially since
some Tiwanaku and Wari tapestry-woven
tunics are very similar. Textile preservation
is less common in the Tiwanaku area, but two
tie-dyed fragments are known from just south
of Arica in northern Chile.62 One, from the site
Azapa-1, was associated with Cabuza style
pottery, which dates it to the Middle Horizon.
It has a discontinuous-warp-and-weft block
design in red, blue, and a faded color, with
diagonal rows of small tie-dyed diamonds
but no larger or partial diamonds; the twisted
fringe is similar to Wari examples.63 The
second fragment, said to be from the Azapa
Valley, appears identical to Wari examples,
although it is inadvisable to conclude that
identical tunics were made in both areas from
a single nonscientific find.64 There is schol-
arly debate about whether the tunics with
quartered format and large tie-dyed diamonds
found at San Pedro de Atacama to the south
Figure 193. Wari tapestry-
woven tunic fragment
that imitates a tie-dyed
design; camelid fiber and
cotton; 41.5 x 57 cm. Museo
Nacional de Antropología,
Arqueología e Historia,
Lima, T16142.
are of Tiwanaku affiliation or not.65 San Pedro
de Atacama does seem to have a different cul-
tural relationship to Tiwanaku than Arica.
Other representations of important Wari
figures sometimes wear tie-dyed tunics
though usually without discontinuous-warp-
and-weft patterning. One person who can be
identified in multiple versions during Epoch 2
wears a south-coast style headdress ornament
and often also a tie-dyed tunic (see fig. 123).66
The remaining representations are more dif-
ficult to classify, however.67
One tapestry-woven tunic fragment is
interesting because it combines a clear imi-
tation of discontinuous-warp-and-weft tie-
dyeing with a motif that typically occurs in
203 T ie - dy ed T unics
tapestry during Epoch 1 (fig. 193).68 Together
with the ceramic representations mentioned
above, it confirms that the discontinuous-
warp-and-weft tie-dyed tunics originated in
Epoch 1. The other archaeological associations
suggest a continuation through Epoch 2. Most
preserved Wari tie-dyed tunics, like tapestry-
woven ones, probably date to Epoch 2B. This
period was a relatively long one, from which
many objects are preserved, and designs that
had been rare and sacred earlier were now
produced in multiples for important men in all
regional areas. The large number of surviving
tie-dyed garments and fragments attests to the
enduring aesthetic impact of these textiles and
to the importance of the men who wore them.
notes I am very grateful to Susan E. Bergh
for inviting me to participate in this
exhibition catalogue. I also received
invaluable assistance from Patricia
Knobloch, Julie Jones and Christine
Giuntini (Metropolitan Museum of
Art), Lydia Fraser (Textile Museum
library), Bridget Gazzo (Dumbarton
Oaks library), Lucy Fowler-Williams
(University of Pennsylvania
Museum of Archaeology and
Anthropology), Mary Frame, Jeff
Splitstoser, and the private collec-
tors who provided access to their
pieces and radiocarbon dates.
1. I prefer to describe the tie-dyed
textiles as “Wari style” rather
than “Wari” since Wari is also the
name of an archaeological site. By
conforming to the convention used
in this catalogue, I do not mean to
imply that these textiles were all
made at the Wari site or by a single
ethnic group. In fact, one of the
points made here is that these tex-
tiles were produced in at least three
different places.
2. The entire process was replicated
by Rehl 2010, pl. 19a–e.
3. Rehl 2010, 213n92; Rehl 2001,
15n9.
4. These numbers are from Rehl
2003, 602–52, since she has counted
more than I have.
5. The second dye could have been
blue or an alkaline bath of cochineal
(Rehl 2010, 167).
6. It is also interesting to note that in
Salasaca, Ecuador, the mordanting
process for cochineal, which helps
prepare the fabric to accept the dye,
includes plants that dye the cloth
a strong yellow color (A. Rowe and
Miller 2007, 270–71, 274).
7. Frame 1999, 336; Orefici 1993, pl.
35, top right; another reassembled
fragment is at the Textile Museum
(91.469) (Rehl 2010, fig. 60; Rehl
2003, cat. 111).
8. See also Haberland 1965.
9. Lempertz 2010, 111, lot 146
(fugitive). Additional examples of
apparent absence are Anton (1992,
no. 29), Lommel (1977, abb. 94), Reid
(1986, pl. 40, fig. 58), and Strelow
(1996, 69), as well as most of those
with the same design as the tunic in
Figure 184.
10. I am grateful to Christine
Giuntini, textile conservator at the
Metropolitan Museum of Art, for
this suggestion. For the chemistry of
indigo, see Liles 1990, 54–55.
20 4 A nn P ollard Rowe
11. Larsen et al. 1976, 49; Moraga
2005, 48–49; Rodman and Fernández
Lopez 2001, fig. 7; Sinclaire Aguirre
1999, 19.
12. Previously published in
Crawford 1946, frontispiece;
Johnston and Kaufman 1967, 23;
Reeves 1949, 104; Rehl 2010, fig. 58;
Rehl 2003, cat. 106; A. Rowe 1977,
32, fig. 24.
13. Previously published in Cook
1996, 120; Frame 1999, 334; Rehl
2010, fig. 57; Rehl 2003, cat. 103;
Rehl 2001, fig. 1; Stone-Miller 1992c,
99–101. I have not examined this
piece; the seam information is from
Rehl 2003, 622.
14. Lavalle and González García
1993, 217 (detail); Lavalle and
González García 1989, 223 (detail);
Rehl 2003, cat. 104; Tsunoyama
1979, 23; Tsunoyama 1977, 23.
15. Medina Castro and Gheller Doig
2005, 81.
16. Lempertz 2010, 111, lot 146.
17. Previously published in Rehl
2010, fig. 55, and Rehl 2003, cat. 98.
18. Azoy 1985, 31; Esteban-
Johansson 2002, 226, fig. 3; Frame
1999, 336; Moraga 2005, 52–53;
Nakajima [1969], 24–25. The Pitt-
Rivers Museum in Oxford also
has a half tunic of this pattern
(1934.70.40).
19. Previously published in Rehl
2010, fig. 49; Rehl 2003, cat. 85.
20. Brugnoli and Hoces de la
Guardia Chellew 1989, 34–35; Hoces
de la Guardia Chellew et al. 2006, 17;
Sinclaire Aguirre 1999, 19.
21. Previously published in Rehl
2003, cat. 87.
22. Stuhr 2008, 166.
23. Previously published in King
1965, pl. 30; Rehl 2010, fig. 50; Rehl
2003, cat. 88.
24. Lommel 1977, 157, cat. 722; Rehl
2010, fig. 51; Rehl 2006, 17; Rehl
2003, cat. 89. Another fragmentary
example now has square blocks
flanked by solid-color strips (Anton
1992, no. 30). Both pieces have
irregular color repeats that probably
indicate modern tampering. The
Museum of International Folk Art
in Santa Fe has a folded and sewn
tunic with dark blue solid strips of
different widths separating three
hook-patterned strips (Rehl 2010,
pl. 47, fig. 56; Rehl 2003, cat. 99).
Asymmetrical arrangement of strips
is not a Wari stylistic feature, and
the width is on the narrow side
(100.5 cm [39½ in.]). Since I have
not examined the piece, I hesitate to
make any further comment.
25. Previously published in Lavalle
and González García 1993, 215
(detail); Lavalle and González
García 1989, 221 (detail); Rehl 2003,
cat. 109.
26. Reid 1986, pl. 40. The published
dimensions are too large (124 x 76
in.) and may be erroneous.
27. Benavides Calle 1999, 389;
Lavalle and González García 1993,
219; Lavalle and González García
1989, 225. The odd number of rows
(nine) and columns (seven) may also
not be original.
28. Previously published in Frame
1999, 337, pl. 23. I have examined
the piece and noted that some joins
were not original, but did not have
the opportunity to map them in
detail.
29. Rehl 2003, cats. 101, 102.
30. Previously published in Eisleb
and Strelow 1966, 295; Rehl 2003,
cat. 96; Strelow 1996, 69, 156. An
example not in its original form is in
the Peabody Museum, Harvard (Rehl
2003, cat. 94).
31. Anton 1992, no. 29 (color). See
also Doering 1936, pl. 58.; Lommel
1977, abb. 94; Rehl 2010, fig. 54; Rehl
2003, cat. 92.
32. Larsen et al. 1976, 49 (color). See
also Rehl 2010, fig. 54; Rehl 2003,
cat. 99; Stone-Miller 1992a, 337;
Tidball 1969, 30.
33. Previously published in Rehl
2010, fig. 62; Rehl 2003, cat. 125.
Another fragment is now part of a
modern composition (Kajitani 1982,
66–67). Rehl considers that the two
fragments may be from the same
original, which also appears pos-
sible to me. The Liverpool Museum
also has a small fragment (Feltham
1989, 12).
34. Schildkraut 1996, 6–7; Shiga
2004, no. 79. The piece now has
eight columns and fourteen rows,
and is 167.5 x 104 cm (66 x 41 in.).
Susan Bergh, who has seen the
piece, also reports that the seaming
yarn does not quite match the woven
blocks in color and that there is a
four-block-long gap in the center
seam where a neck slit might be
(personal communication, August
2011).
35. Previously published in Kelemen
1943, pl. 175; Rehl 2010, fig. 53;
Rehl 2003, cat. 92. A reassembled
fragment with an area of solid-color
blocks and of tie-dyed blocks is
in the National Museum of the
American Indian, Washington, DC
(23/9071).
 36. Esteban-Johansson 2002, 223.
37. Courtesy of the Historic Textile
Research Foundation.
38. Rodman and Fernández Lopez
2005, 123, fig. 10; Rodman and
Fernández Lopez 2001, fig. 30.
39. Rehl 2003, cats. 120–22, 126;
Rodman and Fernández Lopez 2001,
fig. 7; Rowe 1977, 33n14.
40. Prümers 2001, figs. 21–22.
41. Uhle [1903] 1991, 32, fig. 31, for
the textiles and 27, figs. 17, 20, for
ceramics. I am grateful to Patricia
Knobloch for her help in identifying
some of these ceramics. One small
bottle was possibly Middle Horizon
Epoch 3–4. Uhle unfortunately did
not record any specific grave asso-
ciations from Cemetery III except for
the two textiles mentioned.
42. University of Pennsylvania
Museum of Archaeology and
Anthropology (29779). Fragments
of another example said to be from
Pachacamac are in Berlin (Eisleb
and Strelow 1966, 300–06; Rehl
2003, cat. 114; Strelow 1996, 71–72).
43. Rehl 2010, pl. 49; Rehl 2003,
650–51, cat. 134; Strelow 1996, 70;
VanStan 1963; VanStan 1961, 36.
44. VanStan 1963, 167, fig. 1;
VanStan 1961, 36.
45. VanStan 1966, fig. 52.
46. Haberland 1965.
47. For Usaca, see Doering 1936, pl.
58, and other references in note 31;
for Copara, see Rehl 2010, fig. 63;
Rehl 2003, cats. 129–30.
48. Orefici 1993, 125, fig. 35.
49. Phoebe A. Hearst Museum of
Anthropology, Berkeley (4-8093c).
50. Tung 2007, 283; Tung and Owen
2006, 445.
51. For Ica, see Strelow 1996, 69; for
Santa Cruz, see Tsunoyama 1979,
202; for Acarí, see also Esteban-
Johansson 2002, 226.
52. Biermann 2006, 229; Sawyer
1997, 152–54. Disselhoff did not
publish an illustration of the piece
but shared a slide with Alan Sawyer
and probably others.
53. Previously published in A. Rowe
1986b, 161, 182.
54. O’Neale and Kroeber 1930, pl.
27a.
55. VanStan 1967, 28, figs. 20a, 21;
VanStan 1961, 35, upper left.
56. Rehl (2010, fig. 48; 2003, cat.
82) says that the example in the
Art Institute of Chicago is dove-
tailed. She reports the Museum of
20 5 T ie - dy ed T unics
International Folk Art example as
sewn (2010, fig. 47; 2003, cat. 79).
An example at the Metropolitan
Museum of Art is sewn (Rehl 2003,
cat. 83). I have seen the latter as
well as the example illustrated here.
Jiménez Díaz (2009, 134–36) reports
the examples in the Museo de
América as sewn.
57. Metropolitan Museum of Art,
New York (1980.564.3) (Rehl 2003,
cat. 83).
58. The spelling “Tiwanaku” is used
here in place of the older spelling
“Tiahuanaco” only in order to
conform to the style of the volume.
Since the new spelling reflects
modern pronunciation, it seems
reasonable in referring to the mod-
ern village, but I do not agree that
it is appropriate for the ancient site.
For example, “Tia” is the attested
pronunciation in early Spanish his-
torical records (Cieza de León [1553]
1984, chap. 104–5, pp. 281–85).
59. Cook 1996, 87, who calls him
“Figure A”; Knobloch 2010, 202–3,
who calls him “Agent 100.”
60. Other ceramics of Epoch 2
have what may be an abbreviated
representation of this person, with
a tie-dyed tunic: see Lapiner 1976,
fig. 557; Schmidt 1929, abb. 267;
Watanabe 2002, fig. 4.
61. Bennett 1953, pl. 6G.
62. I am grateful to Calogero Santoro
and Liliana Ulloa for providing me
with further information and photo-
graphs of these pieces.
63. Santoro and Ulloa 1985, 77; Ulloa
2001. The size of this fragment is 43
x 80 cm (17 x 31½ in.).
64. Sinclaire Aguirre 1999, 39; Ulloa
1985, 83, no. 250.
65. Cases and Agüero 2004, 123–25;
Rodman and Fernández Lopez 2001,
fig. 10.
66. See also Knobloch 2010, fig.
14; Menzel 1968, fig. 47; Schindler
2000, 147. Identified by Knobloch
as “Agent 104”; Knobloch 2011
describes her identifications in more
detail.
67. From Pacheco in Epoch 1, with-
out face paint or other regalia, see
Menzel 1977, fig. 128. From Epoch
2, with face paint and headgear, see
for example Disselhoff 1967, pl. 35;
Knobloch 2010, 203–4; Menzel 1968,
fig. 48; Seville 2001, 405.
68. Ubbelohde-Doering (1966, 172–
75) found a mummy with a tunic of
this tapestry design in a grave with
Epoch 1 ceramics.
20 6
Heidi King
Figure 194 [137]. Four-
cornered hat; feathers,
cotton, and reed; 17 x 14 x
14 cm. Brooklyn Museum,
New York, A. Augustus
Healy Fund, 41.228.
207
Featherwork
Textiles covered with brilliant feathers of rain
forest birds count among the most striking
works created by textile artists in Pre-Colum-
bian Peru. In the 1530s, the luxurious texture
and lustrous iridescence of native feathered
cloth filled the Spanish conquistadores with
such awe and admiration that one Spanish
missionary wrote: “The gloss, splendor, and
sheen of this feather cloth was of such excep-
tional beauty that it must be seen to be appre-
ciated.”1
From as early as the late third millen-
nium BC, Peruvian cultures used feathers in
ritual contexts as well as for elite clothing and
finery. Grand headdresses, elegant garments,
exquisite pectorals, and ear ornaments dense-
ly decorated with a mosaic of delicate feathers
bespeak the extraordinary skill and manual
dexterity of ancient Peruvian featherwork-
ers. They used different techniques to create
the colorful mosaic. Smaller objects such as
headgear (fig. 194) and various types of orna-
ments (fig. 195) were decorated by gluing tiny
trimmed feathers to the foundation.2 Larger
objects such as tabards (tunic-like garments
without side seams) (fig. 196) and panels (fig.
197) were made by sewing strings of feathers—
the feathers having been knotted individu-
ally onto the strings in a separate process—to
plain weave cotton cloth in overlapping hori-
zontal rows starting from the bottom.
Regrettably few of the featherworks that
survive were scientifically excavated, yet
many without provenance are called “Wari” or
“Nasca/Wari”; most of these attributions are,
however, unconfirmed or have proved incor-
rect. Based on their characteristic iconography
or scientific dating, only a small group can be
safely attributed to the Wari; among them are
an impressive large tabard with borders boldly
patterned with hook motifs in alternating red
and blue feathers surrounding a field of bright
yellow (the hook motif is also known from
woven textiles of the period); hats with a flat
square top; and diamond-shaped ornaments.
The distinctive square Wari hats with upright
peaks projecting from the corners are usually
made in a knotting technique with brightly
dyed yarns (see fig. 147). The unusual hat il-
lustrated here (fig. 194), its corner peaks now
missing, is constructed of a cane framework
covered with cotton cloth and feather mosaic.3
The top and four sides are divided into quar-
ters filled with geometric patterns—stepped
triangles and diamonds—and profile heads,
perhaps inspired by felines, with lozenge-
shaped eyes and tear bands, bared teeth, and
ornaments on their heads. The diamond-
shaped ornaments (fig. 195), originally joined
with threads, feature monkey-like faces with
split eyes and huge “smiling” mouths; their
original function is unknown, but the fine
mosaic of layered cut feathers in five colors
suggests use on a high-status or important
votive object.4
Large, spectacular panels decorated with
the feathers of blue-and-yellow macaws (fig.
197) are also among the featherworks of
secure Wari affiliation. These panels were
found inside several monumental faceneck
ceramic jars that had been buried together in
the Churunga Valley on Peru’s far south coast.
Although local people rather than professional
archaeologists found this offering, the circum-
stances of recovery were reported in several
Peruvian publications.5 In the past, three of
those publications do not seem to have been
considered in their entirety, but they provide
important insights into the potential function
of the site and the purpose of the offering.6
 In contrast to the English-language litera-
ture, which describes the offering simply as
a ceremonial deposit or cache,7 the Peruvian
authors consistently say that mummy bundles
were also found at the site. Thus the offering
may have commemorated either an elite Wari
burial, not many of which have been docu-
mented, or an important human sacrifice. In
the Peruvian reports, the panels are described
incorrectly as mantles (shawl-like shoulder
garments); instead, the format, size, and ties
at the upper corners indicate that the panels
likely served as hangings, perhaps to decorate
the walls of large compounds.
The first account of the find, published in
February 1943 in La Crónica, a Lima news-
paper, was based on interviews with Dr. Luis
Valcárcel, then director of Peru’s national
museum of archaeology, and Ernesto Tejada,
deputy prefect of the Condesuyos Province,
where the panels were unearthed.8 The report
states, “Last month a pre-Columbian burial
mound was accidentally discovered contain-
ing various artifacts, ceramics and garments
and among other interesting objects forty
mantles covered with Amazonian bird feath-
ers (probably of macaws and humming birds);
the mantles are of the most accomplished
manufacture and have decorative motifs in a
perfect and austere style.” The article goes on
to say that the discovery was made by work-
ers making adobe bricks in the hamlet of La
Figure 195 [138].
Victoria about 120 kilometers (75 miles) from
Ornaments; feathers and
cotton; 8 x 6.5 cm (each). the town of Chuquibamba on Peru’s southern
The Princeton University Pacific coast. They came upon a large, an-
Art Museum, Anonymous cient ceramic jar, which led to the discovery
gift 1996, 1996-228.1,
1996-228.2, 1996-228.4, of a total of “seven vessels, each one meter
1996-228.5. [about 3 feet] tall and two meters [6.5 ft.] in
20 8 H eidi K ing
circumference which contained forty rolled
up feathered mantles which represent an ac-
complished sample of feather arts made by
the Tiwanaku cultures.” (One of the vessels is
illustrated here in Figure 198; the cultural at-
tribution was based on the artistic style of the
jars, then incorrectly thought to be Tiwanaku
rather than Wari.) The burial mound consisted
of rough fieldstones and was surrounded by
two concentric circles, the larger measuring
52 m (170 ft.) in diameter. Very importantly,
“mummies in typical fetal position were also
found; they were immediately burnt by the
locals who—filled with cosmic terror—sought
to protect themselves from the wrath of their
ancestors.” Tejada “seized the excavated
objects in the name of the state,” but the local
excavators had already sold many of them. All
confiscated objects were taken to Lima except
three panels, which were given to local insti-
tutions in the Arequipa area.
The second report of the discovery, written
by Leonidas Bernedo Málaga, was published
in 1950 in the newspaper El Deber in Are­qui­
pa.9 This more detailed account reports that
on 12 January 1943 on the Hacienda Hispana
in the Churunga Valley, the native Plácido Coa
discovered a “pre-Columbian tomb which had
three walls built of stone and clay in concen-
tric circles; the first wall measured one meter
above ground, the other two were subterra-
nean.” In this version, the excavations are said
to have uncovered eight faceneck jars, each
containing twelve feathered mantles, between
the two outer walls; the mantles “are made of
the finest cotton cloth with a border of vicuña
wool and covered with blue and yellow, and
sometimes red feathers.”10 In the second circle,
Figure 196 [142]. Tabard;
feathers and cotton; 143.5 x
132 cm. Private collection.
Photo: Maury Ford.

20 9 F eatherwork
Figure 197 [139]. Panel,
probably a hanging, from
Corral Redondo; feath-
ers, cotton, and camelid
fiber; 63.5 x 208.9 cm. The
Metropolitan Museum
of Art, The Michael C.
Rockefeller Memorial
Collection, Bequest of
Nelson A. Rockefeller,
1979, 1979.206.471. Image:
© The Metropolitan
Museum of Art. Image
source: Art Resource, NY

210 H eidi K ing

211 F eatherwork
Figure 198 [45]. Faceneck
vessel with mythical
creatures, from Corral
Redondo; ceramic and
slip; 83.5 x 86 cm. Museo
Nacional de Arqueología,
Antropología e Historia,
Lima, C-64874. Photo:
Daniel Antonio Giannoni
Succar.

a number of objects in unmistakable Inca style
were found: “three small silver jars, one small
gold llama, two silver figurines, three deco-
rated wooden cups, three silver tupus [clothes
pins], a small mantle of alpaca wool with 42
silver plaques, a small poncho in different col-
ors, one aryballoid bottle and ceramic plates.”
The third, innermost circle was not excavated.
The article continues that the “feather
mantles taken from the ceramic vessels num-
bered 97 [sic] and were immediately divided
among the more than 60 persons who had par-
ticipated in the excavations during the night.”
These “textiles of rare beauty” were taken to
nearby mining centers and villages “to be sold
for ridiculous [very low] prices.” Several days
later, Jesús Rojas, son of the hacienda’s owner,
took one of the panels to Chuquibamba, and
on 26 January this “rare and most beautiful
archaeological garment” was exhibited in the

212 H eidi K ing

 office of the province’s notary. The deputy made. He writes that, according to Plácido
prefect of the province ordered all items exca- Coa’s niece, the find was “not so accidental”
vated from the hacienda (now in the hands of and that her uncle talked about it as early as
numerous people) be returned. The authori- 1941. She explained that wheat and barley
ties succeeded in ferreting out “almost all were grown at “Corral Redondo” (redondo
the Inca-style objects, but of the 96 Tiwanaku means “round,” a reference to the circular
[Wari] mantles they only found 44, which they walls at the site); when pounders were hit-
brought to Chuquibamba.” By order of the gov- ting the ground to separate the grain, a dull,
ernment ministry, “the entire treasure found hollow sound could be heard and people
on the Hacienda Hispana” was taken to Lima’s suspected they had found a “treasure.” One
national museum. day in December 1942, a pounder penetrated
In 1990 the renowned archaeologist Eloy a hollow and “when Plácido pulled it out, a
Linares Málaga published the first of two piece of cloth with feathers came with it; this
volumes dedicated to significant archaeologi- was the signal to prepare for the excavation of
cal discoveries in the department of Arequipa the treasure; it was on January 12, 1943 that
since 1851. While much of his information the culmination of the excavation occurred.”
about the find is based on Bernedo Málaga’s The whole village participated and the find
previous accounts,11 Linares Málaga added was distributed among the people.
important facts collected during visits to the Linares Málaga describes the context as
site in 1964 and 1981—including a plan of the follows: “As they excavated, they found three
site of Corral Redondo (fig. 199)—in inter- concentric circles consisting of walls built of
views with people present when the find was rough stones set in mud mortar and covered

Figure 199. Plan of Corral ı

Redondo, the archaeo- ı
ls
logical monument in ı al
ı w
the Churunga Valley, ı ne
Department of Arequipa, sto
ı
January 1981. After Linares ı
Málaga 1990, 120. ı
ı
tomb
ı
ı

evidence of
featherwork
ı

an ancient wall
ı

FIELDS
ı
ı
ı
ı

ı
ı

.
.6
Iqu

"W
LA

15
ipi

8'
T:

0
15

°
irri

73
°5

:
NG
g

6'

LO
atio

00
"S

d
n

roa
dit

Tombs
ch

ino
Necropolis ol
M
to

Featherwork
Al

altitude: 450 masl Churunga

surface area: 1,200 m 2 N Valley

213 F eatherwork
Figure 200. Five ceramic
vessels in the Robles Mojo
style shown in situ at
the Churunga cemetery,
Department of Arequipa.
The photo probably cap-
tures a staged reconstruc-
tion of the find, February
1943.
with stucco; in the center was the funerary
bundle on a base of carved stone and encircled
by the eight jars, finely decorated, and in each
were 12 mantles, 96 in total.” One ceramic jar
and one panel were exhibited in Arequipa in
1943, causing a sensation, especially among
intellectuals, students, and artists “who were
very much aware of the archaeological value
of these objects.” An official petition for the
return of objects from Lima to Arequipa was
unsuccessful; only four of the feathered pan-
els remained in the Department of Arequipa.
Aside from the three mentioned above, one
was rescued from the attic of a school and
placed in a local institution. Linares Málaga
attributes the panels to Wari, based on the
Robles Moqo style of the vessels in which they
were found.
In 1988, Linares Málaga visited Lima’s
national anthropology museum, where he
saw photographs taken of the “pre-Columbian
cemetery of Churunga,” which show five of
the faceneck jars placed in holes in the ground
(fig. 200; the photo has the appearance of a
staged reconstruction).12 Before his visit
Linares Málaga had been told only nineteen
feathered mantles were housed in the mu-
seum, but thirty-two were located during his
visit. He photographed three of the jars and
two storeroom trays with panels.13 Although
conservation issues prevented him from
examining the mantles, he published a basic
214 H eidi K ing
catalogue of all of them, based on museum re-
cords.14 In the last paragraph of his account he
states that “thirteen mantles were lost during
transit from Arequipa to Lima; since four are
in Arequipa, nine still are missing.”
These three Peruvian accounts of the
panels’ discovery shed new light on what the
original purpose of the panels might have
been. The fact that one states they were found
in “a pre-Columbian tomb” and two mention
that mummies were found at the site strongly
supports the suggestion that the Wari vessels
and panels were offerings that accompanied
either a very important deceased individual
or a human sacrifice made to appease cosmic
forces. The former may be supported by the
fact that the jars containing the panels had
not been deliberately broken, a pattern docu-
mented in other important Wari votive offer-
ings (see pp. 145–57, “Shattered Ceramics and
Offerings”). The Inca-style objects, however,
recall the sumptuous offerings found in buri-
als made in conjunction with the capacocha
ceremony,15 which involved human sacrifice.
Although destruction of the mummies at the
time of the find make it impossible to confirm
whether they were sacrifices or elite burials, it
seems clear that Corral Redondo was a huaca
(sacred site) that both Wari and Inca peoples
commemorated by performing ceremonies and
burying precious offerings.
notes 1. Cobo [1653] 1990, 226.
2. For a detailed discussion of
ancient Peruvian featherwork-
ing techniques, see Greene 1991;
Giuntini 2006, 1–13. For general
information on featherwork in
ancient Peruvian cultures, see King
et al. 2012.
3. Only two other examples are
known at this time; see Baessler
1902–3, pl. 147; Brinkerhoff 2000,
cat. no. 37.
4. A group of similar objects is in
the collection of the Staatliches
Museum für Völkerkunde in
Munich.
5. I would like to thank Matthew
Robb of the Saint Louis Art Museum
for sharing copies of the La Crónica
and El Deber articles with me and
Ryan Williams of the Field Museum
for bringing the Linares Málaga
article to my attention. I also thank
my colleague Patricia Llosa at the
Metropolitan Museum of Art and
Lucy Salazar of Yale University for
their generous help with the often
obscure wording of the Spanish
texts.
6. I hope to publish a more detailed
discussion of the Peruvian accounts
of the find in Ñawpa Pacha in the
near future.
7. Bird 1958, no. 3; Menzel 1964,
n196.
8. La Crónica 1943; translated by the
author, with help from Llosa and
Salazar.
9. Bernedo Málaga 1950; translated
by the author, with help from Llosa
and Salazar.
10. Only two panels with blue and
orange/red feathers are currently
known; one is in the collection of
the Museo Nacional de Arqueología,
Antropología e Historia del Perú in
Lima and one in the Textile Museum
215 F eatherwork
in Washington, DC. Two panels, one
in the collection of the Metropolitan
Museum of Art, New York, and
one in the Museum of Fine Arts in
Houston, are covered only with yel-
low feathers.
11. Bernedo Málaga also pub-
lished articles on the find in
1952 (“El Arte Plumario entre
los antiguos Peruanos,” Revista
Universitaria [Arequipa] 36, Año
XXIV, Segundo Semestre) and
1962 (“El Arte Plumario entre los
Antiguos Peruanos: reliquias de
este arte descubiertas en el Dpto. De
Arequipa,” Kontisuyo [Arequipa],
no. 1).
12. Some of the ceramic jars from
Corral Redondo were exhibited in
the Museo de la Nación in Lima,
probably between 1996 and 2002;
on the text label in one of the cases
were two black-and-white photo-
graphs of the site including the one
Linares reproduces (Linares Málaga
1990, 137). The other image is shown
here. Both photographs look staged.
Unfortunately, it was not possible
to find out when and by whom the
photographs were taken.
13. See Linares Málaga 1990 for
color illustrations of two jars (p.
143) and one of the trays (back cover
and fig. 3) and Linares Málaga 1993
for the third jar and second tray (p.
333).
14. Linares Málaga (1990, 152) sug-
gests that the last one on his list
is not part of the Churunga find
because it is significantly smaller,
sewn together from several frag-
ments with modern yarn, and has a
different design executed in several
colors.
15. For information on this impor-
tant Inca ritual see Reinhard and
Ceruti 2010.
Susan E. Bergh
Figure 201 [71]. Pair of
ear ornament frontals
with skulls; shell and
stone; 5.9 x 5.9 cm each.
Ethnologisches Museum,
Berlin, VA 41595a,b. Image:
bpk, Berlin/Ethnologisches
Museum/Art Resource, NY.
Photo: Claudia Obrocki.
217
Inlaid and Metal Ornaments
Among the theatrical arts of regalia, the Wari
seem to have placed greatest emphasis on
fine textiles, especially tapestry-woven tu-
nics, which survive in comparatively large
numbers. Less common are personal trap-
pings made of precious materials—necklaces
and ear ornaments, among them—that touted
the status of royals and lesser elites both in
life and as burial goods. This picture may be
skewed somewhat because the media from
which many Wari ornaments are made did not
survive well in the rainy highlands; indeed,
where find-spots of such materials are known
or reported, they are almost always coastal.
Nevertheless, the numbers of extant orna-
ments are few enough to suggest restricted use
and de-emphasis, particularly in comparison
to the Moche and Chimú, north coast cultures
that temporally bracketed Wari.
The most artistically ornate of Wari
personal ornaments are fashioned of noble
metals, both gold and silver, and also intri-
cate brightly colored mosaics inlaid with a
resinous, sometimes red-tinged adhesive on a
variety of materials, including wood.1 Among
the inlay materials are shell, such as purple
and red-orange Spondylus, purple mussel, and
iridescent mother-of-pearl; a range of blue and
green stones including lapis lazuli, serpen-
tine, and turquoise or its mineral look-alike,
chrysacolla; and metals such as gold, silver,
and pyrite (“fool’s gold”). Aesthetically, these
exquisite inlaid ornaments share much with
tapestry-woven, tied-dyed, or feathered cloth,
which are also based on colorful mosaics. Like
the tapestries, the ornaments’ iconography is
often devoted to the all-important staff deity
and its companions, from which elites seem to
have derived aspects of their authority.
Inlaid Ear Ornaments
From at least the first millennium BC, ear
ornaments made of precious materials were a
mark of distinction in the Andes; among the
Inca their sizes and materials correlated with
social rank.2 Although the Wari seem not to
have placed emphasis on ear ornaments—and
in this may have followed the earlier Nasca,
among whom such ornaments had limited
importance as status indicators3—elaborate
Wari examples exist and apparently both men
and women were privileged to wear them. Al-
though most lack archaeological context, a few
pairs with finely inlaid circular frontals have
been found in tombs of both sexes at Ancón,
a site on the central coast.4 Interestingly, one
of these tombs is said to have belonged to a
young girl, her mummy bundle decorated with
other valued materials in addition to the ear
spools (of the same general type as those in
figs. 201, 202).5 The Wari seem to have con-
fined the use of ear ornaments to the realm
of human pageantry; in artistic depictions,
supernatural beings wear them only rarely.
Wari artists created two main types of
inlaid ear ornaments. In the first group, each
was made from a single piece of lightweight
wood, and the carved frontals assume the
shape of small, appealing human heads of
unknown identity (fig. 203). While the facial
features are generic, the headdresses fall into
at least two categories distinguished by the
ornamentation of the band-like base and the
crest that rises from it. Shell tesserae impart
a life-like appearance to the eyes and are also
sometimes inset in the cheeks. The U-shaped
flange surrounding the face, its original shell
surface occasionally still present, may repre-
sent a collar or the headdress’s chin strap. In
Figure 202. Pair of ear
ornament frontals with
animal heads; shell, stone,
silver, and copper; Diam.
6.5 cm. The Art Institute of
Chicago, 1955.2543a,b.
Figure 203 [69]. Ear orna-
ment, from Pachacamac;
wood and shell; 3.8 x
6 x 2.4 cm. University
of Pennsylvania
Museum of Archaeology
and Anthropology,
Philadelphia, Max Uhle,
William Pepper Peruvian
Expedition, Funded by
Phoebe A. Hearst, 26720.
Image: courtesy the Penn
Museum.
218 S usan E . Bergh
a few the shafts bear tiny nail holes, implying
that they once were sheathed; if the remains
of a silver nail in one are any indication,6 the
sheathing material was thin metal sheet, ei-
ther silver or gold, which also may have been
attached with resinous adhesive to the fron-
tals in some cases.7
The second and most sumptuous type of
Wari ear ornament features complex inlays
covering the surface of rimmed, disk-shaped
frontals made of shell, some with an irides-
cent, nacreous surface (figs. 201, 202, 204). The
shafts often do not survive but, where they
do, they are fashioned of bone or wood and
fastened to the frontals with either ancient
adhesive, thin cords threaded through holes in
both frontal and shaft, or both; small nail holes
again sometimes pierce the wood examples,
implying sheathing.
A favored inlaid design comprises four re-
peats of a charmingly rendered profile animal
head with a button nose and, usually, a verti-
cally divided eye that marks the creature as su-
pranormal in some way. Although this head is
sometimes identified as feline, it may well rep-
resent another native beast, such as one of the
Andean camelids (fig. 202). If a camelid, then
the alternation of the head with that of a bird
in some examples recalls the bird and camelid
(or deer) tunics that may have functioned as
a pair and were worn by Wari’s most distin-
guished elites, perhaps its paramount rulers
(see pp. 159–91, “Tapestry-woven Tunics”).8
Additional representatives of this type
feature other symbols of authority: elaborate
profile heads that likely represent either sac-
rificers or the staff deity’s winged attendants,
skulls (fig. 201), the flowers of Anadenanthera
colubrina (a hallucinogenic plant that Wari
 elites seem to have used for religious purpos-
es), and a motif that may represent a truncated
staff of authority.9 Finally, still special in spite
of some loss of its fragile inlay, is an ornament
that survives without its mate and depicts a
single, full-figured version of one of the staff
deity’s winged attendants (fig. 204). Bent on
one knee with head looking skyward and a
four-feathered wing at its back, it clutches the
ghost of a staff; the shaft curves to conform to
the frontal’s edge and the lower terminus is
formed by a human head.10
Mirrors
In Mesoamerica, mirrors were used as cos-
tume ornaments, as cosmetic accessories,
and for divination; they also had complex
symbolic references.11 The uses and meanings
of ancient Andean mirrors—their reflective
surfaces made of either a single piece or a mo-
saic of obsidian (volcanic glass), anthracite (a
lustrous coal), pyrite, or metal—are less well
known although many are pierced and could
have been worn in a manner akin to Meso-
american practice.12 That they at least some-
times functioned in Wari times as insignia of
high status is suggested by one, made of pyrite
inlays set into a rectangular stone frame,
Figure 204 [70]. Ear orna-
ment frontal with staff-
bearing creature in profile;
shell and stone; 5.9 x 5.8
x 1 cm. Ethnologisches
Museum, Berlin, VA
41596. Image: bpk, Berlin/
Ethnologisches Museum/
Art Resource, NY. Photo:
Claudia Obrocki.
219 I nlaid and M etal Ornaments
excavated from an elite mausoleum containing
Wari ceramics at Cerro Amaru in the northern
highlands.13 The luxuriousness and imagery of
a few other Wari examples also establish elite
usage: the reflective materials are set within a
shallow recess on one side of a slightly trap-
ezoidal, handled wood frame, its reverse deco-
rated with the inlaid head of the staff deity or
a sacrificer carved in relief (figs. 205, 206).
Although their form resembles modern
hand-held mirrors,14 these Wari examples were
likely not used as looking glasses. Rather, as
archaeologist Anita Cook observes,15 their
shape is reminiscent of snuff trays, also often
trapezoidal and ornamented with carved,
inlaid images of the staff deity or sacrificer.
Such trays are better known in Tiwanaku’s
sphere to the south, where they were used
in conjunction with religious practices that
involved inhaling the powder of psycho-
tropic plants of the Anadenanthera genus.16
As mentioned, the Wari seem to have used
the same plant, although to date few snuff
trays have been discovered in Wari territory.
One important, large exception, which has a
handled shape like that of the mirrors, was
found at El Castillo, a funerary structure on
the north coast that contained objects in both
Figures 205a, 205b [73].
Mirror with staff deity
head (detail of handle
below); wood, stone, and
shell; 23.9 x 12 x 2 cm.
Pre-Columbian Collection,
Dumbarton Oaks Research
Library and Collection,
Washington, DC, PC.B.432.
Image: © Dumbarton Oaks,
Pre-Columbian Collection,
Washington, DC.

Wari and hybrid Moche-Wari styles (fig. 207).17
The report that one Wari mosaic mirror (fig.
205) was associated with a small copper spoon
perhaps like those used during snuff rites may
strengthen the connection to such practices.18
Tiwanaku snuff spoons have handles that, like
those of the mirrors, terminate in an elegantly
rendered animal head,19 as do a few larger
Wari spoons of unknown function found in
a woman’s tomb at San José de Moro on the
north coast (fig. 208).20
Wari mirrors, then, perhaps take inspira-
tion from snuff equipment used by the Tiwa-
naku, with whom the Wari shared religious
customs involving the staff deity and vision-
ary revelation facilitated by Anadenanthera. In
other words, these rare mirrors may be presti-
gious reinterpretations of exotic paraphernalia
related to elite Wari ritual.21 In contrast to the
smaller Tiwanaku trays, the recessed well of
the mirrors is not reserved for mixing snuff
with spoons prior to inhaling it through tubes.
Rather, it is inlaid with shiny, reflective mate-
rial that—by analogy to the snuff it replaces,
by the enhanced powers of vision to which
it refers, and by association with numinous
imagery—likely relates to an ancient quest for
spiritual illumination and insight.
The mosaic mirror illustrated above, its
surface entirely encrusted with shell and
stone tesserae, features an abbreviated version
of the staff deity’s head with four geometri-

220 S usan E . Bergh

Figures 206a, 206b [166].
Mirror with sacrificer
(front and back); wood
and pigment; 28.6 x 15.9 x
1.9 cm. Private collection.
Photo: Bruce Schwarz.
Figure 207. Drawing of
fragment of a wood snuff
tray with a handle, from
El Castillo in the Huarmey
Valley. After Prümers 2001,
fig. 25 (1).

5 cm

221 I nlaid and M etal Ornaments

Figure 208 [20]. Spoon,
from San José de Moro
(tomb M-U1512); ceramic;
9.8 x 3.5 x 1.8 cm.
Pontificia Universidad
Católica del Perú, Lima,
M-U1512-C16. Photo:
Daniel Antonio Giannoni
Succar.

cized appendages that are tipped with frontal
faces and disposed symmetrically around
the central visage. During burial, the mir-
ror was in contact with a textile that left its
impression on the surface; areas where the
impression is discontinuous likely correspond
to replacements of the original mosaic. An-
other mirror with traces of red pigment on its
surface (fig. 206) features a snarling figure that
appears to be a sacrificer: although it lacks a
weapon, its posture (two arms extended from
a frontal torso) is typical of sacrificers, and
its zigzagging staffs are tipped by subjugated
humans, one with a skeletonized chest. Four
human heads peer over the upper rim and a
register of animal and bird heads in profile
surround the now-empty recess for the mirror.
A feline (or perhaps feline-serpent) head oc-
curs on the handle of each mirror.22
Inlaid Tunic-wearing Figures
Wari inlaid objects also include exquisite hu-
man figurines that are either freestanding or
attached to one-half of a bivalve shell, usually
Spondylus princeps (fig. 209), a red-orange
oyster encrusted with thorny spines that here
have been removed, but also others (figs. 210,
211; see also fig. 143).23 The figures on these
objects—some pierced, perhaps for suspension
as necklace pendants—wear knee-length gar-
ments that probably represent tapestry-woven
tunics, given the vertical bands that alternate

Figure 209. Spondylus

princeps (thorny oys-
ter) shell. Photo: Gary
Kirchenbauer.

222 S usan E . Bergh


Figure 210. Pendant with between plain and patterned (see fig. 144);
figure; Spondylus shell,
tiny dots sometimes occur in the plain bands.
shell, silver, and cop-
per; 14.2 x 9.5 cm. The The tunics, headdresses, lunate collars often
Art Institute of Chicago, made of metal, and huge ear ornaments signal
Buckingham Fund, that these are elite males. (The ear ornaments
1955.2542.
are surely simplified representations of the
circular, inlaid variety.) The tunics’ decora-
tion features a schematic profile head reminis-
cent of the button-nosed creature on many ear
ornaments although details—the nose, mouth,
and often the eye—are omitted, perhaps for
reasons of scale (fig. 202). The extreme sche-
matization makes correlating the imagery of
the inlaid garments to that of actual tunics
guesswork; among the candidates are tunics
with the face-fret motif, winged attendants,
or sacrificers, all reduced to heads alone (see
figs. 144, 153, 157). It may be coincidence that
223 I nlaid and M etal Ornaments
in some instances the inlaid head visually
recomposes as a standing animal in profile
whose long neck and bulging chest recall an
unshorn camelid (fig. 211).
Although these small figurines cannot yet
be identified, the Spondylus upon which some
of them perch provides clues about their broad
associations, based in part on Spanish records
concerning the shell’s use among the Inca.
These records agree that Spondylus—imported
from Ecuador, where it flourishes in coastal
waters that are warmer than Peru’s—held a
value greater than gold or precious stones and,
therefore, that the shell was a form of wealth,
a badge of prestige, and a basis of power, espe-
cially for those who controlled its trade and
distribution networks.24 The figures-on-a-shell
support the suggestion that Wari elites, too,
legitimized their earthly authority with this
prized, exotic commodity, whose circulation
they may have sought to regulate and manipu-
late to their benefit.25
Spondylus also had very important ritual
functions among the Inca, who attributed to
it the power of fecundity and put it to use as
an essential ingredient of rain-making rites be-
cause it comes from the sea, conceived as the
ultimate source of water that cycles through
the cosmos.26 Thus, say the Spaniards, the Inca
offered prodigious amounts of the shell, some-
times pulverized into dust, to sacred springs
and other water sources in fervent efforts to
avert drought and promote bountiful crops.
That the Wari and other contemporary people
attributed similar capacities to Spondylus is
suggested by the widespread surge in the vo-
tive use of the shell, previously scarce among
Peruvian cultures, following a disastrous, late-
sixth-century drought (the longest and most
intense of the last 1,500 years) that coincided
with Wari’s rise.27 One place that registered
this heightened use was the Huamachuco
region in the northern highlands, including
Cerro Amaru. This shrine captured Wari’s
attention in the years following the drought,
perhaps in part because of its sacred water
wells and its location on the highland route
over which Spondylus traveled from Ecua-
dor.28 The figures-on-a-shell, then, may refer to
a belief that Wari tunic-wearing elites had the
power to intercede in the cosmic hydrological
Figures 211a, 211b [75].
Figure pendant (front and
back); wood, shell, stone,
and silver; 10.2 x 6.4 x 2.6
cm. Kimbell Art Museum,
Fort Worth, AP 2002.04.

22 4 S usan E . Bergh
Figure 212 [77]. Trumpet
with figure; Strombus
shell, shell, and stone; H.
19.1 cm. The Dayton Art
Institute, Museum pur-
chase, 1970.32.
cycle in order to blunt its destructive caprices
and assure the renewing, seasonal arrival of
rain upon which the future hinged—a claim
given reality by the water and agricultural
infrastructures that the Wari introduced in
several regions (see pp. 65–81, “The Wari Built
Environment”).29
Related meanings may lie behind a few
trumpets made of Strombus (conch shell also
native to Ecuadorian coastal waters) that are
inlaid with staff-bearing humans similarly
attired in tunics, ear ornaments, collars, and
headdresses (fig. 212). Such trumpets, fash-
ioned by removing the shell’s spire to form
225 I nlaid and M etal Ornaments
a mouthpiece, are still used today during
ceremonies in areas of the highlands.30 Among
the Inca, trumpets were men’s instruments
and the same may have been true among the
Wari, for the figures inlaid on the trumpets are
males and elsewhere another male is depicted
playing a trumpet of a different type.31
In addition to the categories of objects
discussed above, Wari inlaid ornaments and
objects include necklaces or necklace pen-
dants made of small trapezoidal plaques (fig.
213), the thumb rests of spear throwers (fig.
214), spoons or spatulas, and containers.
Figure 213 [74]. Ornament Metalwork
with figure; shell, stone,
Most known Wari works in precious metals—
and metal (silver?); 6.6
x 3.6 cm. The Cleveland gold and silver objects made from hammered
Museum of Art, In memory sheet and sometimes also with inlaid de-
of Mr. and Mrs. Henry tails—fall into a few major categories. Impor-
Humphreys, gift of their
daughter Helen 1944.291. tant among them are flashy plumes, their sur-
faces often ornamented with sacred imagery,
especially the head of the staff deity but also
lesser beings such as the deity’s profile atten-
dants.32 The tapering, pin-like shafts of these
ornaments suggest they were worn, shimmer-
ing and quivering, in clothing or perhaps more
likely as projecting additions to headdresses
or crowns. A large number of silvered copper
plumes are said to come from several buri-
als at Pomacanchi, southeast of Cuzco, which
contained a total of at least 141 metal orna-
ments, including many bells and wide undec-
orated cuffs for the wrists or ankles (see fig.
12).33 An example with an especially elegant
shape was excavated from a burial at Huaca
Pucllana on the central coast (fig. 215); several
others are said to derive from different sites in
the same region, including Pachacamac and
Ancón (figs. 216, 217).

Figures 214a, 214b [78].

Spear-thrower thumb rest
with bird (two views);
bone and stone; 8 x 4 cm.
Roemer- und Pelizaeus-
Museum, Hildesheim,
V. 5522.

226 S usan E . Bergh

Figure 215 [90]. Plume, The Wari also made silver or gold ear orna- bottom of the U to a fan of metal that may
from Huaca Pucllana; sil-
ments in sparing numbers. One especially mimic the shape of a blade. Now in Stuttgart
ver; 28.9 x 15.3 cm. Museo
de Sitio Huaca Pucllana, handsome and elaborate silver pair features and accompanied by two smaller versions of
Lima, MSHP-97-156 (ME). flaring, concave frontals attached to thick hol- unknown function (fig. 219a), this pectoral is
Photo: Daniel Antonio low shafts that contain pellets and rattle when remarkably similar to another example recent-
Giannoni Succar.
they move (fig. 218). The posts, held in place ly excavated under scientific conditions from
Figure 216 [88]. Plume;
when worn by cotton stoppers that remain a burial at Espíritu Pampa in the Vilcabamba
gold; 21.8 x 4.6 x 0.7 cm.
Ethnologisches Museum, intact, are cleverly punned as the scaly bodies region (Department of Cuzco).35 The burial,
Berlin, VA 31795. Image: of serpents with threatening toothy heads that perhaps that of a local leader who doubled as
bpk, Berlin/Ethnologisches seem to mingle serpent and feline features.34 a Wari governor, also contained a pair of gold
Museum/Art Resource, NY.
Only a few metal neck ornaments can be wrist cuffs and a silver mask. Such collec-
Figure 217 [89]. Plume;
attributed securely to the Wari. One of the tions as the one from Pomacanchi may come
gold; 27.5 x 6.9 x 0.7 cm.
Ethnologisches Museum, largest and most spectacular is a U-shaped from comparable tombs (see also pp. 207–15,
Berlin, VA 31797. Image: pectoral, its arms tipped by the heads of “Featherwork,” and pp. 251–67, “Wari’s An-
bpk, Berlin/Ethnologisches fanged mythical creatures (fig. 219b). The dean Legacy”). The same is likely true of the
Museum/Art Resource, NY.
Photo: Claudia Obrocki. steps along the outer edge give way at the Stuttgart pectoral.

227 I nlaid and M etal Ornaments

Figure 218 [81]. Pair of
ear ornaments; silver
and cotton; 9.5 cm (with
shafts) x 8.6 cm. The
Metropolitan Museum of
Art, New York, Bequest
of Jane Costello Goldberg,
from the Collection of
Arnold I. Goldberg, 1986,
1987.394.580–81. Image: ©
The Metropolitan Museum
of Art. Image source: Art
Resource, NY.

Figures 219a, 219b [84].
Three ornaments; silver;
17 x 17 cm, 17 x 16 cm, 55
x 44 cm. Linden-Museum,
Stuttgart, M31039. Image:
© Linden-Museum
Stuttgart. Photo: A. Dreyer.
Finally among Wari ornaments are a few
impressive gold or silver plaques. Probably
once mounted on a backing, such as a textile,
they often feature hook-beaked birds—perhaps
raptors such as the condor—with outspread
wings and tail (fig. 220). In one famous ex-
ample said to come from Pachacamac or its
vicinity, however, the bird’s head is replaced
by that of a snouted, fanged animal that may
be feline (fig. 221). These precious-metal ob-
jects are another indication that the Wari as-
sociated bird imagery with the highest status
members of their society (see also pp. 159–91,
“Tapestry-woven Tunics”). Another beautiful-
ly realized plaque takes the shape of a sumo-
like warrior, his menacing power established
as much by the axe and shield he carries at his
sides as by his looming bulk and high-relief
head, the eyes likely once inlaid. He wears a
four-cornered hat and a tunic covered with

228 S usan E . Bergh

229 I nlaid and M etal Ornaments
Figure 220 [85]. Bird interlocked hooks found in both tie-dyed and forehead. The face shares features with both
plaque; gold; 13.7 x 14.6
tapestry-woven examples; the band across the the staff deity, mainly its frontality, and sacri-
cm. Private collection.
Photo: Bruce Schwarz. waist may represent a belt (see fig. 19). The last ficers, a kinship established by the upturned,
Figure 221 [87]. Winged
example here is a large, arresting supernatu- fanged mouth lined with teeth.36 “Tear bands”
creature plaque; gold; 13 ral head made of silver (fig. 222). Although fall from the eyes, a feather-like motif rests
x 15.8 cm. Ethnologisches mask-like in appearance, it is more likely an on the forehead, and whisker-like elements
Museum, Berlin, VA
ornament that was attached to something else extend from the cheeks.
28787. Image: bpk, Berlin/
Ethnologisches Museum/ via the pairs of holes that pierce the chin and
Art Resource, NY. Photo:
Waltraut Schneider-
Schuetz.

Figure 222 [83]. Mask-like

ornament; silver; 20.5 x 18
x 5 cm. American Museum
of Natural History, New
York, B/9450. Image: cour-
tesy American Museum of
Natural History, Anthro­
pol­ogy. Photo: Craig
Chesek.

230 S usan E . Bergh

Notes 1. Cook (1996, 185) also notes this
colored resin.
2. D’Altroy 2002, 94; Morris and von
Hagen 2011, 63.
3. Proulx 2006, 194.
4. Ravines 1981, 90–95; Ravines
1977, 362–65.
5. Ravines (1981) dates this burial
to the third epoch of the Middle
Horizon.
6. The nail is preserved on an orna-
ment at the Museum of Fine Arts,
Houston (2010.328.1, 2010.328.2).
7. In 1896, the archaeologist Max
Uhle ([1903] 1991, 30, pl. 4, fig. 5)
excavated one of these ornaments at
Pachacamac from a cemetery at the
base of the Temple of Pachacamac,
now known as the Painted or
Polychrome Temple (Shimada
1991, xxxi), on the central coast
(fig. 203). Pachacamac is also the
reported origin of another pair
with the second headdress type
(Ethnologisches Museum, Berlin,
VA40267a, b) while another example
is said to come from Ocucaje on the
south coast (Staatliches Museum
für Völkerkunde, Munich, NM 329)
(Schindler 2000, 144).
8. An example with bird heads is in
the Museum of Fine Arts, Houston
(2011.1165.1, 2011.1165.2). In other
variations, geometric motifs rather
than bird heads alternate with the
animal heads (Lapiner 1976, 255,
fig. 593, now at the Metropolitan
Museum of Art, New York,
1978.412.215, 1978.412.216).
9. For the winged attendants, see
American Museum of Natural
History, New York, 41.2/8597. For
the skulls, see also Misugi 1985, fig.
104b, a pair of frontals whose cen-
ters are formed by a high-relief head
with skull-like features. For the A.
colubrina flowers, see Sotheby’s New
York, 16 May 1995, lot 11; Knobloch
2000. For the truncated staff motif,
see Misugi 1985, fig. 104a.
10. This ornament and one other
(Ethnologisches Museum, Berlin,
VA41595) are said to have been
found at Pachacamac on the central
coast. Lapiner (1976, 255 and note)
reports the find-spot of another
pair of this type as Cahuachi on
the south coast, but suggests this
provenience may be incorrect and
that the pair may instead come
from Pativilca on the central coast
(Metropolitan Museum of Art,
1978.412.215 and 1978.412.216).
231 I nlaid and M etal Ornaments
Finally is an ornament recovered 25. Jennings 2006, 277; Jennings and
on the north coast at La Libertad, Craig 2001, 481.
Hacienda Casa Grande, Chicama 26. Gose 1993, 501–2; Murra 1975.
Valley (Peabody Museum, Harvard
27. Cordy-Collins 1990, 396, 408;
University, 47-63-30.5683).
Shimada 1991, LI.
11. Miller and Taube 1993, 114.
28. J. Topic and T. Topic 1992, 172–
12. Bruhns and Kelker 2010, 150; 74; J. Topic and T. Topic 1983–85, 47;
Cook 1996, 184. T. Topic and J. Topic 2010b.
13. T. Topic and J. Topic 1984, 38–40. 29. Two of the figures-on-a-shell are
14. Bruhns and Kelker 2010, 151. said to come from Pachacamac on
15. Cook 1996, 184. the central coast (Schmidt 1929, pl.
VII). The larger of the two was lost
16. Llagostera et al. 1988, pl. 7, 8,
during World War II; the smaller
12; Torres et al. 1991. For more on
still survives at the Ethnologisches
the use of hallucinogens in ancient
Museum, Berlin (VA 41598). On
Andean ritual practice, see, for
Wari’s interest in water, see also
instance, Knobloch 2002 and Stone
Glowacki and Malpass 2003.
2011.
Glowacki (2005) suggests that the
17. Prümers 2001. ancients also used Spondylus to
18. Cook 1996, 184. Mary Glowacki induce visions.
(2002, 282) reports having found 30. For instance, Allen 1988, 116;
snuff spoons and perhaps a snuff Urton 1981, 28.
tube in the tombs of women at
31. A. Rowe 1979, 6, 11. A few other
Qoripata, a Wari site near Pikillacta
Wari shell trumpets feature registers
in the Cuzco region; Gordon
of inlaid, profile heads like those
McEwan (2005a, 34; personal com-
that decorate inlaid tunics. One
munication, 2011) also found two
is at the Ohara Gallery of Art in
small spoons in Units 16B-D and 40
Kobe, Japan, which also owns an
at Pikillacta.
example with a tunic-wearing figure
19. For example, Young-Sánchez (Misugi 1985, 60), and another is
2004a, 65. at the Museum zu Allerheiligen in
20. Rengifo Chunga et al. 2008, Switzerland (Schaffhausen 1999,
129–31. 251). The trumpet illustrated may
21. Knobloch (2000, fig. 10) repro- have been found in the Huacho-
duces a drawing of a Wari indi- Pativilca region on the central coast
vidual who holds a small circular (Lapiner 1976, 255).
object that may be a mirror. 32. Chávez 1984–85, fig. 20; Flores
22. This head is similar to those Espinoza 1959, pl. 2c.
that adorn the handles of some Wari 33. Chávez 1984–85.
urns (see fig. 103b) (Anita Cook, 34. King 2000, 32.
personal communication, 2011).
35. Cuzco 2011.
Other handled mirrors with rows
of human heads include a Chimú- 36. See Bergh 1999, fig. 86, for the
Wari example at the Metropolitan head of a similar sacrificer rendered
Museum of Art in which many silver in tapestry weave.
nails remain embedded (1995.428;
see also Sotheby’s New York, 16 May
1995, lot 18), and one at the National
Museum of the American Indian,
Washington, DC (Keleman 1943,
298b). A history of Andean mirrors
has not been written, but versions
with handles may be a northern
tradition; several Moche and Chimú
handled examples are known.
23. See Schmidt 1929, pl. VII, lower
left for a non-Spondylus example
at Berlin’s Ethnologisches Museum
(VA41598).
24. D’Altroy and Earle 1985; Murra
1975. In colonial times, a piece of
the shell smaller than a fingernail
fetched four reales, or one-half of
a Spanish dollar (piece of eight)
(Blower 2000, 210).
Susan E. Bergh
Figure 223 [97]. One of
two offerings found at
Pikillacta in the 1920s.
Figurine offering; green-
stone, Spondylus shells,
copper or copper alloy;
H. 2 to 5.2 cm (figurines).
Colección Juan Larrea,
Museo de América,
Madrid, 8.825–64 (figu-
rines), 7.038 (rod).
233
Figurines
Many Wari figurines are sculptural master-
works in miniature, and the Wari used them
most impressively to create complex offerings
that they buried deep in the earth in acts of
devotion or dedication. All documented exam-
ples of such offerings—three, in total—come
from Pikillacta, the massive Wari outpost near
Cuzco in the southern highlands. Two of them,
each featuring forty small human figures
carved in turquoise-colored stone,1 seem to
relate closely to each other and may even have
been deposited as part of a single ritual event.
In 1927, locals discovered the two offer-
ings, now in Cuzco’s Museo Inka and Madrid’s
Museo de América,2 and we have only broad,
hearsay reports about their original contexts:
two deep, circular, stone-covered pits that
may have been located within several meters
of each other. Aside from the figurines, each
offering contained a tapering rod made of cop-
per or an alloy of copper as well as Spondylus
(thorny oyster) and Strombus (conch) shells.3
It is said that the Madrid figurines—one of
which disappeared in Peru shortly after the
offering came to light—formed a circle around
their metal rod, which had been driven into
the earth (fig. 223); the Cuzco set apparently
lay prone on a bed of sand around the rod and
shells.4 Thus, the contexts and contents of
the offerings seem to have been very similar
and the figurines themselves, all depicted in
distinctive garments and headgear of various
kinds, further imply a relationship. Overlook-
ing differences in size, sixteen of the Cuzco
figurines have an identical or similar counter-
part in the Madrid set, two from each set form
a quartet, and three others, a triplet.5 Except
for another pair within the Cuzco group,6 the
rest are individual in their traits or combina-
tions of traits.
Aside from a kneeling, bound prisoner
unique to the Madrid set, all the figurines
stand erect with their arms at their sides
although, curiously, one in each set seems to
lack hands.7 Facial features are generic—all
have wide-open, almond-shaped eyes and
minimally modeled lips parted by an inci-
sion—but variations in the features’ relation-
ships and in the overall shapes of the large
faces impart an impression of individuality
that differences in costume enhance. Among
the hats and headdresses, often settled over a
head cloth that covers the figure’s neck, an un-
decorated turban is common (fig. 224a). Deco-
rated turbans and several other types also oc-
cur (figs. 224b, 224c, 224e, 224g), some perhaps
made of an animal skin or feathers, others of
a long narrow headband (llautu) coiled many
times around the head (fig. 224d), and one of a
sling, a fiber weapon used to hurl stones.8 Most
of the figurines are clad in a single tunic-like
garment that falls to the lower legs or feet (figs.
224a, 224b, 224d, 224g). Nevertheless, a few
tunics, including the prisoner’s, are shorter,
some figurines seem to wear layers of gar-
ments (figs. 224c, 224e), and one wears only a
belt (or loincloth) and broad collar (fig. 224f).9
Many of the garments are individualized by
their incised geometric decoration—circles,
grids, zigzags, and the like. While none of the
tunics correlate securely with known Wari
tunic types, several have vertical stripes remi-
niscent of tapestry-woven examples, and the
circles on a few others may refer to tie-dyed
cloth (figs. 224b, 224g; compare with fig. 144).
Two of the matched figurines seem to carry a
bag over one shoulder,10 and a small minor-
ity, including the prisoner, wear ornaments
in their ears, nose, or around their necks (figs.
224f, 224g).11 In all the two sets contain fifty-
Figurines from the
two offerings found at
Pikillacta in the 1920s.
Photographs represent
figurines in the Madrid set.
Drawings represent figu-
rines in the Cuzco set.
Figure 224a. Two of three
similar figurines wearing
tunics with diamond pat-
terning and simple turban-
like headgear that rests
on a headcloth. Colección
Juan Larrea, Museo de
América, Madrid, 8.839;
H. 3.2 cm. Drawing: after
Valcárcel 1933, pl. I D;
H. 3 cm.
Figure 224c. Pair of figu- Figure 224d. Figurine
rines wearing layered gar- unique to the Cuzco
ments and a distinctive hat set wearing a tunic and
or helmet. Colección Juan perhaps a coiled headband
Larrea, Museo de América, (llautu). Drawing: after
Madrid, 8.858; H. 4.2 cm. Valcárcel 1933, pl. V h;
Drawing: after Valcárcel H. 3 cm.
1933, pl. III R; H. 4.3 cm.
Figure 224f. Pair of
figurines wearing broad
collars, belts or loincloths,
and large nose ornaments.
Colección Juan Larrea,
Museo de América,
Madrid, 8.840; H. 2.8 cm.
Drawing: after Valcárcel
1933, pl. II L; H. 3.4 cm.
23 4 S usan E . Bergh
Figure 224b. Pair of figu-
rines wearing decorated,
turban-like headgear and
tunics that may refer to
tapestry-woven examples.
Colección Juan Larrea,
Museo de América,
Madrid, 8.826; H. 4.7 cm.
Drawing: after Valcárcel
1933, pl. II M; H. 4.1 cm.
Figure 224e. Figurine
unique to the Cuzco set
wearing layered garments.
Drawing: after Valcárcel
1933, pl. IV W; H. 2.5 cm.
Figure 224g. Pair of figu-
rines wearing tunics with
circles that may be tie-dyed,
complex headdresses, ear
ornaments, and necklaces.
Colección Juan Larrea,
Museo de América, Madrid,
8.833; H. 3.3 cm. Drawing:
after Valcárcel 1933, pl. II
O; H. 2.8 cm.
 seven types,12 some of which may represent
women, although features that would permit
positive sex identification—such as tupu pins,
which women used to fasten their clothing—
are lacking.
What do these apparently paired, related
offerings mean? Some espouse the view that
the figurines represent not individuals but
ethnic types or groups, based on the diversity
especially of headdresses, which at least some
colonial-period native groups used to distin-
guish themselves from others.13 Building on
this idea, which has yet to be strengthened by
affiliating the figurines’ various “dress codes”
with those in use in specific regions during
the period, a few interpret the figurines as
symbols of vanquished groups and the offer-
ings, by implication, as microcosmic depic-
tions of Wari’s realm.14
In stark contrast is a reading that takes
the figurines to represent the founding ances-
tors of Wari royal lineages. According to this
approach, the offerings helped the Wari lay
claim to legitimacy and inheritance as they
established themselves in the Cuzco region.15
The best support for this idea comes not
from analysis of apparel but in part from one
colonial-period source, which reports that
people of the central coast regarded as their
first progenitor a tiny green idol of otherwise
undescribed appearance that was kept in a
shell, together with three small green stones
that were the origins (perhaps the ancestors)
of beans, “wheat,” and chili peppers.16 Also,
at least some Inca figurines, although made of
gold rather than greenstone, represented royal
lineages that descended from Manko Qhapaq,
the legendary founder of the Inca Empire.17
Finally, although the hearsay reports about
the offerings’ context do not specify the room
or even the kind of structure in which the
figurines were found, one strong possibility is
a niched hall, an important type of ceremonial
building where the Wari may have performed
rituals related to ancestor worship (see pp.
65–81, “The Wari Built Environment”).18 The
reading of the figurines as honored ancestors,
however, explains neither the presence of a
prisoner in the Madrid set nor the diversity of
garments and headdresses that the figurines
wear.19
235 F igurines
The structure of the offerings suggests that
certain numbers are also meaningful, particu-
larly two, expressed by both the paired offer-
ings and the paired figurines, and forty. The
readiest explanation for emphasis on pairs is
dualism, the Andean tendency to construct
many aspects of existence—social and politi-
cal, cosmic and religious—in terms of dialecti-
cal pairs whose reciprocal interaction makes
life possible (see pp. 103–21, “The Coming of
the Staff Deity”). Thus, among other possibili-
ties, the figurine pairs could refer in some way
to the two halves into which social groups and
communities may have been divided (perhaps
including the ancestors of the two parts),20 and
the offerings, if limited to two and deposited
together, to a ritual balancing of forces that
sanctified ground and space. Interestingly,
most of the Spondylus shells in at least one of
the offerings had been broken in two.21
While Wari interest in dualism turns up
in other ways (for example, see pp. 159–91,
“Tapestry-woven Tunics”), the number forty is
less common.22 In the figurine offerings, some
scholars relate it to practices of the later Inca,
for whom it had significance as a principle
of socio-political and ritual organization. For
example, the existence of tiers of forty Inca
officials who represented social units of vary-
ing sizes has prompted the suggestion that the
offerings reflect similar organization among
the Wari.23
The presence of shells may imply that
at least part of the offerings’ purpose was to
assure the flow of water and with it the pros-
perity of the land; at the time of the Spanish
conquest, native Andeans invested Spondylus
with a prodigious power to attract water (see
also pp. 145–57, “Shattered Ceramics and Of-
ferings,” and pp. 217–31, “Inlaid and Metal Or-
naments”). This level of meaning is compatible
with interpretation of the figurines as ances-
tors, whom Andean people, like many others
across the world, revered as sources of fertil-
ity.24 The blue-green color of the stone could
also allude to fertility concerns.25 The meaning
of the metal rod, apparently critical because
of its central placement in the offerings, is an
open question; it has often been compared to
the staff of authority that the staff deity and its
companions carry,26 even though its distinc-
Five figurines from the
lower layer of the eastern
gate offering found at
Pikillacta in 2004.

Figure 225a. A nude, Figure 225b. A warrior Figure 225c. A profile war-
kneeling prisoner with holding a circular shield rior holding an axe to his
hands bound behind his and lances (front and rear and a square shield to
back (front and back); back); gold-silver-copper the front of his body (front
gold-silver alloy?; H. 3.6 alloy?; H. 3.6 cm. After and back); greenstone; H.
cm. After Arriola Tuni and Arriola Tuni and Tesar 2.7 cm. After Arriola Tuni
Tesar 2011, 24, fig. 29. 2011, 26, fig. 31. and Tesar 2011, 25, fig. 30.

Figure 225d. A standing Figure 225e. A standing

individual with braids or individual (front and
tresses of hair tied together back); Spondylus shell; H.
at the back (front and back); 2.7 cm. After Arriola Tuni
gold-silver alloy?; H. 3 cm. and Tesar 2011, 23, fig. 28.
After Arriola Tuni and Tesar
2011, 22, fig. 26.

tive, tapering form and lack of ornamentation
are unlike that of staffs depicted artistically
(for example, see fig. 75c).27 It could well refer
to another implement, such as a digging stick
used to work the fields or a war club.28
The third Pikillacta figurine offering was
also structurally complex.29 Archaeologists
found it near the threshold of the site’s main
portal (the eastern gate), buried beneath the
floor of the primary, corridor-like “street” that
penetrates the residential portion of the site.
The offering comprised several layers, the bot-
tommost of which, more than nine feet be-
neath the corridor’s surface, contained another
tapering rod of copper or copper alloy. It too
may have been driven into the earth.
Just above the rod was a pile of forty-nine
or fifty small human figurines that origi-
nally may have been tied into a cloth bundle,
although the cloth had long since disinte-
grated.30 These figurines are marked by the
diversity of their materials; many are made
of cast or hammered metal (perhaps includ-
ing gold-silver alloys),31 some of Spondylus
shell, and three of colored stone. They fall into
several broad groups: a few bound prisoners
(fig. 225a), a larger number of warriors, iden-
tified by the shields and other weapons that
they carry (figs. 225b, c), and an even larger
contingent of unarmed individuals, who often
stand with their empty hands at their sides
(figs. 225d, e).32 Most wear simply rendered

236 S usan E . Bergh

 headgear and garments that, along with other
accouterments, can be used to sort the figu-
rines into subgroups of varying sizes, includ-
ing some pairs; a few, made of thin plaques
of shell or stone, are depicted in profile (fig.
225c). This level of the offering also contained
a few pieces of Spondylus and two Strombus
columella spirals, one carved from stone.
Immediately above were the offering’s
largest, most complex, and most spectacular
figurines, all made of cast copper or copper
alloy: two shield-bearing human warriors, one
wielding a spear-thrower (fig. 226a) and the
other, a club; two very similar felines, their
teeth bared, crouched to pounce, with tails
curled in opposite directions (fig. 226b); and
two sacrificer-like supernatural or mythical
beings. One of the last, which clutches a sev-
ered head and a long, slender rod, has the head
of a feline but the two-digit hands and feet of
a camelid or deer. The other is a snouted, com-
posite creature of unclear genealogy brandish-

Three figurines from the

upper layer of the eastern
gate offering found at
Pikillacta in 2004.
Figure 226a. Warrior hold-
ing a large rectangular
shield, a spear-thrower,
and several extra spears
(front and side). The
front portion of the spear
mounted in the spear-
thrower is bent backward;
copper alloy; H. 8.7 cm.
After Arriola Tuni and
Tesar 2011, 13, fig. 16.

Figure 226b. Crouching

feline with tail curled to
the right (front and side);
copper alloy; H. 6.4 cm.
After Arriola Tuni and
Tesar 2011, 14, fig. 17.

Figure 226c. Composite

creature brandishing a
club over the body of a
human (front and side);
copper alloy, H. 7.2 cm.
After Arriola Tuni and
Tesar 2011, 17, fig. 20.

237 F igurines
Figure 227 [93]. Figurine;
greenstone; 4.7 x 2.5 x 2.2
cm. Denver Art Museum
Collection, Funds Provided
by Jan and Frederick R.
Mayer, 1992.502.3. Photo:
© Denver Art Museum
2012.
Figure 228 [94]. Figurine;
greenstone; 3.5 x 1.9 x 1.9
cm. Denver Art Museum
Collection, Collection of
Frederick and Jan Mayer,
1994.45. Photo: © Denver
Art Museum 2012.

Figure 229 [95]. Figurine;

greenstone; 4 x 1.9 x 1.9
cm. Denver Art Museum
Collection, New World
Department Acquisition
Fund, 1997.15. Photo:
© Denver Art Museum
2012.
Figure 230 [96]. Figurine;
greenstone; 3.2 x 1.9 cm.
Denver Art Museum
Collection, Funds from
the Alianza de las Artes
Americanas, 1995.39.1.
Photo: © Denver Art
Museum 2012.

238 S usan E . Bergh

Figure 231 [82]. Two pen-
dant figurines; silver; H.
4.1 cm (each). Milwaukee
Public Museum, 34596/
9672, 34597/9672.
ing a club over the supine body of a human, its
chest slashed open to reveal the organs within
(fig. 226c). Several of the implements these
figurines carry, including the clubs, have a
tapering form similar to that of the rod in the
offering’s foundational layer.
After covering these artifacts with earth,
the Wari burned an offering in the pit. Among
the charred remains, laid out in a patterned
arrangement, were valves of Spondylus with
apexes removed, one broken in two, another
into four, and a third into eight pieces. Finally,
just beneath the corridor’s surface, a layer
of broken Wari vessels capped the offering,
which may have been covered with a stone
slab.
The time that lapsed between the deposit
of the layers is hard to know; the offering may
represent a single episode or a few that in-
cluded a later burning event aimed at recon-
secration.33 The figurines’ identities are also
elusive although emphasis on war, conquest,
and cosmically sanctioned sacrifice is clear,
leading the excavators to suggest that the of-
fering commemorates specific conquests (and
individuals) and with them Wari’s right to
settle and control the land around Pikillacta.34
Dualism may also structure this offering
239 F igurines
although it seems more equivocally expressed
than in the others. Whatever the case, some-
thing about the offering suited it for placement
in one of the most trafficked areas of the site—
near the threshold of the principal gateway
where all who visited would have passed
either around or over the offering.
Only further scientific excavation will
reveal whether other offerings occur beneath
the corridor and whether it is coincidence that
all known buried figurine offerings come from
Pikillacta. In other locales stone figurines
have been recovered from tombs and the sur-
faces of sites.35 The figurines that now reside
in collections presumably come from one of
these contexts (figs. 227–30). Most are human
and made of stone, especially blue-green stone
that, aside from any connection with fertility,
may have carried the connotation of wealth
and high status—in tapestry-woven tunics at
least, blue is the most prestigious color that
ancient weavers used (see pp. 159–91, “Tap-
estry-woven Tunics”).36 A few, however, are
made of metal, such as two very similar silver
figurines each wearing a turban headdress and
garments with belts or decorated waists; both
are pierced with a hole for suspension, per-
haps from a necklace (fig. 231). Since they may
Figure 232 [72]. Pendant
figurine (front and back);
wood, shell, turquoise, and
gold; 3.3 x 1.5 x 1.1 cm.
Pre-Columbian Collection,
Dumbarton Oaks Research
Library and Collection,
Washington, DC, PC.B.437.

have belonged to an assemblage with other,
similar bead-like figurines, it might be an er-
ror to interpret them as a pair.
Figurines with surface mosaics of metal
and brightly colored stone and shell are rare,
perhaps because of their fragility. One delicate
example (fig. 232), also pierced for suspension,
has a wood core over which a sticky, resin-like
material was applied, sometimes so thickly
that it creates three-dimensional volume. This
is the case with the hat and the unidentified
objects on the chest, which were built up with
the resin and then overlaid with gold foil.37
Another, much larger mosaic figurine was
similarly constructed: over a wood core, the
artist fashioned a handsome Wari dignitary
with carved pieces of stone, including lapis la-
zuli (the nose) and perhaps steatite (the arms
and legs), such shells as Spondylus (the upper
face and elsewhere), and now-mineralized
silver sheet (the hat). This figure, which seems
to wear a tapestry-woven tunic, also is pierced
through its shoulders and may have found use
as a pendant (see fig. 211; see pp. 217–31, “In-
laid and Metal Ornaments,” for more on this
figure). The Wari also made figurines of more
humble materials, especially fired clay.

2 4 0 S usan E . Bergh
notes 1. Inconclusive tests were conducted
in the 1930s to identify the stone
from which one set is made; it may
be sodalite rather than turquoise
(Valcárcel 1933, 23–27).
2. For the Cuzco figurines, including
brief descriptions and a complete
suite of drawings, see Valcárcel
1933. For the Madrid figurines,
see Ramos and Blasco 1977, 67–75,
97–108 (with descriptions of each
figurine and a complete set of pho-
tographs), and Trimborn and Vega
1935, 85–89.
3. Thanks to Heather Lechtman for
her help in phrasing the composi-
tion of the so-far unanalyzed metals,
which are sometimes referred to as
bronze in the literature.
4. Ramos and Blasco 1977, 68;
Trimborn and Vega 1935, 87–88;
Valcárcel 1933, 22, 33; see also
McEwan 1987, 27–28, for a few
important details, including about
the loss of one Madrid figurine.
5. Ramos and Blasco (1977, 68,
97–108) provide the only detailed
correlation of the sets, which
accounts for nineteen Cuzco figu-
rines and twenty Madrid figurines.
They seem to rank the matches by
describing them with such terms
as “igual” (the same) or, more com-
monly, “muy semejante” (very simi-
lar), “semejante” (similar), and the
like. The size discrepancies range
from none or negligible to 12 mm.
6. Valcárcel 1933, 31, figs. A and E.
7. Ramos and Blasco 1977, 74, pl.
XIh; Valcárcel 1933, 30, fig. O1. The
two are otherwise distinct but both
seem to wear layered garments,
unlike many of their compatriots. It
could be that the hands are pulled to
the inside of the garments.
8. Ramos and Blasco 1977, 71–72,
97–108; Valcárcel 1933, 29–32. Cook
(1992, 350–51, figs. 5, 7) briefly dis-
cusses and presents drawings of the
headdresses in both groups. For the
sling, see Ramos and Blasco 1977, 72,
102, pl. Xc.
9. Ramos and Blasco 1977, 73,
97–108; Valcárcel 1933, 29–32.
Cook (1992, 349–50, figs. 6, 8) again
presents a discussion and drawings
of garments.
10. Ramos and Blasco 1977, 73, 102
(no. 18).
11. Ramos and Blasco 1977, 73,
97–108; Valcárcel 1933, 29–32.
12. This count includes the prisoner.
The matching figurines account for
eighteen types; the Cuzco set con-
tains an additional twenty and the
Madrid set, an extra nineteen.
2 41 F igurines
13. Anders 1986, 895–900; Cook 24. For example, Gose 1993; Salomon
1992, 353–55; Ramos and Blasco 1995.
1977, 70–72, 75; Trimborn and Vega 25. Anders 1986, 255–57; Glowacki
1935, 87; Valcárcel 1933, 28, 33. and Malpass 2003, 442–43;
14. Arriola Tuni and Tesar 2011, Glowacki, this volume.
30; Ramos and Blasco 1977, 75; 26. For example, Anders 1986, 257;
Valcárcel 1933, 33. See Valcárcel Larrea, cited in Cook 1992, 358.
1933 and Cook 1992 for some efforts
27. See Bergh 1999, fig. 95, top for a
to link the figurines’ costumes and
rare, possible exception.
ornaments to regions.
28. For earlier Nasca representations
15. Cook 1992. McEwan (1998,
of tapered digging sticks, which lack
79–80; 2005b, 152–53; this volume)
the foot-rests that other cultures
adopts Cook’s reading. See J. Topic
sometimes added, see, for example,
and T. Topic 2001, 210, for a dissent-
Proulx 2006, 178, and figs. 5.70, 5.79.
ing view.
29. The following description is
16. The Relación del Licenciado
based on Arriola Tuni (2008) and
Felipe de Medina, cited in Valcárcel
particularly Arriola Tuni and Tesar
1933, 23. Medina incorrectly used
(2011), which contains the most
the term “wheat,” which is not an
complete photographic record of
indigenous Andean crop.
the artifacts recovered in the offer-
17. Betanzos, cited in Julien 2000, ing together with an appendix that
257, and McEwan 2005b, 153. describes each in some detail.
18. McEwan 2005b, 153; McEwan 30. It is not clear whether a crushed
1998, 79–80; see McEwan 1991, 95, metal object fused to the back of one
fig. 6, for the initial suggestion that of the figurines represents another
the offerings come from a residential figurine or something else (Arriola
patio group structure. Tuni and Tesar 2011, fig. 21, item 22).
19. Arriola Tuni and Tesar 2011, 29. 31. The results of scientific analysis
These authors believe that, in addi- of the metals have not yet been
tion to the kneeling prisoner, the published (Arriola Tuni and Tesar
Cuzco and Madrid figurines include 2011, 7).
two standing, bound prisoners.
32. See Arriola Tuni and Tesar 2011,
20. See also Anders 1986, 895–900; 21–27, for their classification of the
Cook 1992, 346, 352. figures.
21. This is the offering now in 33. Arriola Tuni and Tesar 2011,
Madrid (Trimborn and Vega 1935, 32–33.
88, and personal observation).
34. Ibid., 30–34.
Valcárcel (1933) does not mention
35. Cook 1992, 342. Gordon
broken Spondylus shells in his dis-
McEwan’s team discovered a fine
cussion of the Cuzco set.
Wari figurine in trash in Unit 36 at
22. Anders (1991, 1986), however,
the site of Choquepukio but he is of
documented a group of forty small
the opinion that the figurine was
rooms at Azángaro, an intermediate-
accidentally lost or discarded rather
size Wari site, and forty occurs in
than placed in the trash deliberately
one of the format-based number
(personal communication 2012). See
sequences in Wari tapestry-woven
National Geographic 2004, xxxii, for
tunics (see pp. 159–91, “Tapestry-
an illustration.
woven Tunics”).
36. Figurines of supernatural beings
23. Anders 1986, 895–900; Cook
are rarer; one greenstone example
1992, 358–60; see also Zuidema
depicts a sacrificer-like creature
2009, 95. Cook believes that the
(Orlando Museum of Art, 2002.025).
figurines depict “the legendary 40
Thanks to Anita Cook for calling
founding ancestors,” that twenty
this figurine to my attention.
figurines from one set have twins
37. Cook and Lechtman 1996.
in twenty from the other, and that
the latter number also has an Inca
parallel in the twenty groups (ayl-
lus) into which Cuzco’s population
was organized. It is unclear how this
interpretation squares with the pres-
ence of many more than forty figure
types in the two figurine collections,
or the fact that, due to the presence
of a triplet and a quartet, the match-
ing figurines total 39 and represent
eighteen types, not twenty (see also
notes 5 and 12 above).
Susan E. Bergh
Figure 233 [164]. Sacrificer
container (front, back, and
side views); wood and
cinnabar; 10.8 x 7 x 7.5 cm.
The Cleveland Museum
of Art, John L. Severance
Fund 2007.193.a–b.
2 43
Wood Containers and Cups
Natural fragility, climate, and time have made
ancient wood objects relatively scarce, but
surviving examples suggest that wood was an
important Wari artistic medium. Large-scale
wood objects—three-dimensional sculpture,
architectural decoration carved in relief, and
the like—have not been reported. Rather,
wood was used mostly to create small, ornate
objects of various kinds, among them elite
personal ornaments (see pp. 217–31, “Inlaid
and Metal Ornaments”) and staffs of author-
ity,1 containers, cups and spoons, and weav-
ing implements. Some have shell and colored
stone inlays or now-empty depressions that
likely once held such inlays. Find-spots are
usually unknown but preservation and avail-
able records indicate that most wood objects
were buried in the dry sands of Peru’s coastal
regions, away from the more rainy highlands
where the Wari heartland is located.
Small sculpturally elaborate containers, a
few recovered scientifically on the coast, are
among the most common surviving Wari wood
objects.2 Aside from their presumed final use
as offerings in tombs or other contexts, little
is known about the previous function of these
handsome objects, which have receptacles that
are usually cylindrical and plugged at the bot-
tom and top with separately carved stoppers.3
They have been identified as containers for
lime powder (made from a calcium carbonate
source such as shell or limestone).4 The lime
would have been added with a small spoon
or spatula to a lump of coca leaves that were
chewed, probably for their mildly stimulating
physical effects as well as for social and ritual
reasons.5 But the interiors of the containers
usually have not been examined for lime resi-
due. One exception bears no evidence of lime
on its roughly carved inner surface—only
now-invisible traces of cinnabar (mercuric
sulfide), the red-orange pigment that is more
evident on the exterior, particularly in recess-
es (fig. 233).6 Thus, at least this container prob-
Figure 234. Winged super-
natural creature container
(front and back); wood,
pigment, and shell; H. 8.5
cm. Staatliches Museum
für Völkerkunde, Munich,
NM186.
Figure 235. Prisoner
container (front and back);
wood, bone inlay, pigment;
7.6 x 3.8 x 3.3 cm. The
Metropolitan Museum of
Art, New York, Purchase,
Rogers Fund and Carol
R. Meyer and Arthur
M. Bullowa Gifts, 1977,
1977.376. Image: © The
Metropolitan Museum
of Art. Image source: Art
Resource, NY.
2 4 4 S usan E . Bergh
ably did not hold lime. It may be that contents
varied with circumstances and container type.
It also could be that the contents had a re-
lationship with the container’s artistic subject
matter, very often supernatural creatures with
fanged mouths that sometimes are winged
and commonly throttle or hold small humans,
represented either in toto or by the head alone.
One unusually complex container makes it
clear that this imagery alludes to sacrifice: a
feline-headed being, its magnificently carved
face surrounded by appendages that suggest
a connection to the staff deity, draws a knife
across the throat of the human it holds across
its lap by the hair (fig. 233). The aftermath
of this drama may be referenced by the two
identically coiffed human heads that hang
from the feline’s belt, overlapping bird-headed
bands that spring from ankle ornaments.
(Such bands emerge more clearly from the
feline’s wristlets.) A third head dangles, ap-
parently by its trachea, from the beak of a bird
head at the back of the headdress from which
descend the feline’s long tresses, each tipped
by a zoomorphic head depicted in profile.
At the sides of the headdress, trimmed with
upright elements that may represent a crown
of feathers, are additional figures with arms
raised above their heads. They seem to be hu-
man, although their faces are now eroded.7
The supernatural beings that this group of
containers portrays, then, are predators and
other details of imagery reinforce the con-
nection: the small felines occasionally found
on the shoulders or in the headdress of the
main figure; the hooked raptor-like beaks of
the bird heads used as ornaments; and, in one
instance, the axe- and shield-bearing warrior
who appears on the container’s back.8 A curi-
ous relative of the group is an example that
assumes the shape of a winged animal-headed
creature with upright ears, toothy mouth, and
open, empty hands. Identified variously as a
bat, a fox, or a feline, it carries on its back a
trachea and a heart that nests between lungs,
all rendered three-dimensionally (fig. 234).
Although the organs have not been identified
as to species, they could be human.9
The precise identities and meanings of
these figures are unknown but the sacrifice
with which they are associated may have had
 correlates in religious practices; archaeolo- creature’s feet (fig. 236). Whether this activ-
gists found human heads buried in the floor ity refers to grinding, fire making, digging,
of one Wari D-shaped temple at Conchopata, or something else is unclear.12 Occasionally
and the heads had been severed rather than re- the rod and platform are replaced by a hu-
moved from the already-dead bodies of ances- man head.13 In the container illustrated, shell
tors.10 If these heads represent sacrifices and inlays create a pelt-like pattern over the body
not war trophies, it is likely that, as in other and eye ornaments include a feather-like motif
places of the world, the sacrifice was solemnly above the eye and a “tear band” that falls from
undertaken as a renewal rite that enticed the the lower lid onto the cheek.
benevolence of cosmic forces through offering Other containers also survive and sug-
of the most precious material available (see gest that in antiquity a wider range of forms
pp. 103–21, “The Coming of the Staff Deity”). and types was common. One notable example
As the anthropologist Victor Turner observed, takes the shape of a human male seated atop a
sacrificial immolation opens a channel be- small, kayak-like boat made of lashed totora, a
tween the visible and invisible, and ritual kill- buoyant shoreline reed.14 Such boats occasion-
ing, which like the birth process involves the ally also occur elsewhere in Wari iconography
flow of blood, can be thought to give life and (fig. 103) and are still in use today in coastal
animate, even as it slays. According to Turner, regions of Peru and Bolivia’s Lake Titicaca.
in complex societies like Wari, sacrifice may A small number of wood cups, or keros,
also factor in other state-sponsored rituals and carved with intricate iconography in the Wari
with them serve as a way to regulate boundar- style are known, and they invariably feature
ies and maintain structure at the level of the supernatural imagery. For instance, encircling
state.11 Although many of the supernatural
Figure 236 [163]. Animal beings depicted in the wood containers have
container; wood, shell,
and stone; H. 4.4 cm.
associations with sacrifice, a few do not and
American Museum of instead carry staffs or other objects in their
Natural History, New York, hands.
41.2/8599. Image: cour-
Another important category of containers
tesy American Museum
of Natural History, depicts humans of various kinds, the most
Anthropology. Photo: identifiable of whom are warriors, recognized
Craig Chesek. by the weapons that they carry, and prisoners.
The fine example shown in Figure 235 kneels,
hands tied behind his back; the cotton yarn
that remains attached to the container at his
wrists may refer to a restraining rope. The
upper body is incised with a grid that could
represent a checkerboard tunic, and long
braids of hair fall onto his back from beneath
a feline-head cap or helmet. The meaning of
the designs painted on the face and the disk
of bone that inlays the chest is unknown. The
identities of other human-shaped contain-
ers are less clear. Some of these figures carry
objects including cups or other vessels in their
hands; a few wear elite ornaments but are
otherwise not elaborately attired.
Containers in the form of either felines
or foxes also exist. In one favored type, the
animal sits on its haunches with body up-
right and hands grasping a rod-like object that
rests on a small, square platform between the

2 45 Wood C ontainers and C ups

 the circumference of one is a relief register
with four winged staff-bearing figures that
in other contexts appear as the staff deity’s
profile companions (fig. 13).15 The cup’s figures
are of two types that alternate: one casts its
animal head upward so that it gazes at the
cup’s rim; the other, which has the hooked
beak of a raptor, faces forward. In both, the
wing’s feathers and the appendage that flows
backward from the neck are tipped with small
zoomorphic heads that align vertically behind
the body. Such figure pairs are also known in
246 S usan E . Bergh
tapestry-woven tunics and imply that dual-
istic thought inflected Wari religion (see pp.
159–91, “Tapestry-woven Tunics”). Other cups
feature a fanged frontal head, executed in high
relief, that sometimes has hair-like appendages
like the staff deity’s. It is not clear, however,
whether these heads represent versions of the
staff deity or its occasional associate, a sacri-
ficer. Wood cups may have been used during
feasting ceremonies to drink chicha (native
corn beer) or other beverages.
notes 1. Staffs are crucial in Wari 10. For example, Tung 2008, 294;
iconography and are assumed to Tung and Knudson 2008; Tung et al.
have been important symbols of 2007.
rulership but very few are known 11. Turner 1977, 201–2.
archaeologically, perhaps because 12. Conklin (1970, 21–22) suggests
most were made of perishable wood. that fire drilling may be depicted in
One apparent exception, a frag- one Middle Horizon tapestry.
ment today in a private collection,
has a finial carved as a standing, 13. The head is reminiscent of one
staff-bearing supernatural creature featured in ear ornaments (see fig.
flanked by felines. 203).
2. Ravines (1981, 161–62) reports 14. This container is also at the
that finely made wood containers Metropolitan Museum of Art
have been recovered from Middle (1978.412.250). See Vranich et al.
Horizon tombs at Ancón on the cen- 2005 for totora boats.
tral coast. See Lapiner (1976, 240) 15. This cup is said to come from
for several reported find-spots on Cahuachi on the south coast or,
the south and central coasts (respec- perhaps, from the Huacho-Pativilca
tively, perhaps Coyungo in the Nasca region of the central coast (Lapiner
Valley and the Huacho-Pativilca 1976, cat. 568).
region). Two containers in Berlin are
said to come from Pachacamac (VA
40419 and VA 40420).
3. The stoppers, particularly the top
one, are today often missing.
4. Trout 2006.
5. For aspects of coca’s meaning
among contemporary Quechua-
speakers of the Peruvian highlands,
see, for example, Allen 1981.
6. The analysis was performed in
2007 by Ellen Howe of the Sherman
Fairchild Center for Objects
Conservation at the Metropolitan
Museum of Art with visual micros-
copy and X-ray fluorescence spec-
troscopy (XRF).
7. A sample from this container
yielded a radiocarbon age of cal.
769–887 AD (95 percent confidence
interval). My thanks to Joerg
Haeberli for his advice about radio-
carbon issues.
8. This container is at the
Metropolitan Museum of Art
(1978.412.142).
9. Schindler (2000, 142) identifies
the creature as a bat and suggests
that the organs are animal and
related to divination. See also Cook
(this volume, p. 112). Jim Kennedy
of Bat Conservation International,
Inc., kindly provided an opinion
about the creature’s heritage based
on photographs of this container.
According to Kennedy, the face and
ears are bat-like but the creature
lacks other features limited to bats,
such as a noseleaf, and the wings are
distinctly bird-like. Thus, he could
make no positive bat identification.

2 47 Wood C ontainers and C ups

The Aftermath
William H. Isbell
and Margaret Young-
Sánchez
Figure 237 [43]. Faceneck
vessel with figure; ceramic
and slip; 17.8 x 11.4 x 11.4
cm. Brooklyn Museum,
New York, Henry L.
Batterman Fund, 41.418.
[148]. Overleaf, Four-
cornered hat with winged
creature; camelid fiber;
14.3 x 15 x 15.5 cm. The
Cleveland Museum
of Art, Purchase from the
J. H. Wade Fund 1945.378.
251
Wari’s Andean Legacy
At the dawn of the South American Bronze
Age, central Andean peoples were over-
whelmed by Wari, a new culture that spread
from its mountainous Ayacucho Valley home-
land, bringing great changes to most of high-
land and coastal Peru (see maps, pp. xiv, xv).
This revolutionary new social formation lies
deep in the ancient past, with scant hints of
its memory preserved in sixteenth-century
Inca myths and oral accounts. Consequently,
knowledge is largely limited to the archaeol-
ogy of material remains. Wari is distinguished
by certain mortuary practices, new settle-
ment patterns and architectural forms, and
the characteristic art forms discussed in this
volume.1 They define the central Andean
Middle Horizon (AD 600–1000), when Wari
established a centralized hierarchy of politi-
cal capitals,2 popularized arsenical bronze,3
organized the distribution of luxury goods
throughout its sphere of influence,4 and played
a crucial role in the northward dissemination
of a religious imagery and an associated set of
beliefs that originated in the south—southern
Peru, western Bolivia, and northern Chile.5
During Wari’s ascendency genetic variation
increased in local populations,6 implying
demographic movements and ethnic mixing.
Today, although ancient Andean archaeol-
ogists and art historians still debate the nature
of Wari political organization, most conclude
that Wari was a complex archaic state and that
it probably developed into an empire—Peru’s
first—that was ruled by a centralized and
hierarchical government, conquered and colo-
nized distant territories, and reorganized the
colonized people into administered provinc-
es.7 However, its profound influences on later
Andean cultures, arts, and events have gone
almost unrecognized. A selective re-evalua-
tion of Wari’s Andean legacy, which is so great
that many significant issues must be men-
tioned only briefly or omitted entirely here,
will help correct this situation. Enough can
be said, however, to provide an appreciation
of the immensity of Wari’s importance in the
central Andean past; it is hoped that this will
promote a new dialogue about Wari, beginning
with this groundbreaking exhibition.
The most significant transformations Wari
brought to the central Andes have gone largely
unrecognized. Together with Tiwanaku, a con-
temporary state centered on the south shore of
Lake Titicaca in Bolivia, Wari shifted the axis
of Andean political power and cultural com-
plexity through development in two funda-
mental domains, agropastoral production and
demographics on one hand, and socio-political
organization on the other. These transforma-
tions, accompanied by persuasive new reli-
gious and political ideologies, laid the founda-
tion for Wari’s better recognized achievements
in city planning, architecture, and art. Failure
to fully appreciate Wari’s contributions can be
attributed to the very recent discovery that it
was a culture separate and independent from
Tiwanaku (see pp. 31–45, “The Rise of an
Andean Empire”). The scale and complexity of
Wari’s accomplishment undoubtedly demand-
ed new political and economic systems that
likely influenced the later Inca as they built
their even more extensive empire.
Evolutionary ideal types such as “em-
pire,” however, are intellectual double-edged
swords. They stimulate us to ask more cogent
questions of the archaeological remains. But
they also provoke answers based on expecta-
tions about the ideal type. Today we must
continue to wrestle with questions that are not
easily resolved by material remains alone. For
 example, do settlements with Wari architec-
tural forms, ceramic styles, and luxury goods
found in territories far from the Wari capital
identify Wari provincial administrative cen-
ters, as most Andean archaeologists believe?
Or are alternative interpretations more con-
vincing? Theresa Lange Topic and John Topic
argue that Wari remains represent ritual prac-
tices shared by independent kingdoms partici-
pating in a great religious confederacy.8 Their
inspiration comes not from the ideal types of
comparative cultural evolution, but from early
colonial accounts of Andean principalities
under the Spanish. It is the task of archaeolo-
gists and art historians to articulate the most
convincing interpretation, verified by as much
data as can be brought to bear. But material
remains are inscrutable and subject to under-
standings that are easily influenced by the
ideas, convictions, and assumptions held by
the analysts. Knowledge of the ancient past is
more often a plausible account than a proven
hypothesis.
Shift of the Andean Axis of Power
The Wari heartland lies in the southern high-
land Ayacucho Valley. Cultural development
remained relatively simple until the final
century or two of the preceding Early Inter-
mediate Period (AD 1–600), when the rustic
local culture known as Huarpa adopted new
ceramic styles and techniques from south
coastal Nasca culture.9 This at least is what
pottery documents. The Middle Horizon is
marked by extensive Wari influence outside
of Ayacucho, primarily registered in ceramics
but also in architecture and other remains.10
Wari influences outside the homeland began
in the Nasca region but spread rapidly in all
directions. About fifty or seventy-five years
into the Middle Horizon, Tiwanakoid religious
iconography11 appeared in Ayacucho, surely
promoting the adoption of new beliefs.12 Wari
artists incorporated the religious imagery on
prestige goods such as fine ceramics, textiles,
and mosaics. Such commodities were used
and displayed at Wari itself and distributed
throughout Wari’s sphere of influence, signal-
ing political and economic power as well as
cultural prestige. Wari material culture and its
influences spread from the homeland to as far
252 W illiam H . I sbell and M argaret Young - Sánchez
north as the Piura and Chotano Rivers and, in
the south, to Cuzco and the coastal Moquegua
Valley.
Following the beginnings of central An-
dean sedentism (the process of settling down)
several millennia before the Common Era, the
focus of political complexity, demographic
density, and technological prowess was north-
ern Peru, especially the north and central
coast. The southern highlands and even the
south coast remained relatively backward. But
with Wari, Tiwanaku, and the Middle Horizon
the center shifted to the southern highlands,
altering the Andean axis of power forever.
After Wari declined, the north coast briefly
reasserted dominance with the Lambayeque
and Chimú cultures.13 But the Inca Empire,
with its capital in Cuzco, showed that cultural
ascendancy had shifted south, and into the
highlands, for the longue durée.
The southern highlands are higher, drier,
and colder than the sierra of northern Peru,
making them a more difficult environment for
human settlement. But the southern mountains
offer an abundance of land and resources to
agropastoralists skilled enough to cultivate ex-
tremely dry canyons and very steep mountain-
sides, and to exploit the expansive puna grass-
lands, so high that they are unfit for anything
but grazing. The Wari and Tiwanaku succeeded
in making these territories productive.
With the onset of the Middle Horizon, set-
tlement patterns shifted in Ayacucho.14 From
the earlier preferred locations on hilltops and
surrounded by walls, large Middle Horizon
settlements were re-sited to flat spurs and val-
ley sides adjacent to deep canyons and gorges.
Apparently they were not walled, implying a
low level of concern for defensibility. This shift
correlates with significant population increase,
suggesting that more was involved than new
appreciation of landscape. Most likely, the
shift documents diminished levels of conflict
associated with Wari’s rise to political domi-
nance, and the success of new irrigation tech-
nology that channeled water into intermediate-
elevation fields and communities through long
canals tracking hillside contours. Engineering
prowess had to be equaled by administrative
skills, and both surely required bureaucratic
specialization. More or less contemporary with

Figure 238. Rumicolca,
the Wari aqueduct. The
channel flows along the
top of a tiered wall that the
Wari built. The Inca later
broke through the wall to
create a gate and sheathed
the sides of the gate with
fine masonry in the char-
acteristic Inca style. Photo:
Susan E. Bergh.
253 Wari ’s A ndean L egac y
Wari’s ascendancy, Tiwanaku was gaining
pre-eminence in the Titicaca Basin, outstrip-
ping other settlements in size, population, and
architectural grandeur.15 Agropastoral intensi-
fication and administrative prowess were also
responsible, although emphasizing different
strategies.
Middle Horizon irrigation features are
little-studied in Wari’s heartland, although
surface reconnaissance of the territory sur-
rounding the ancient city reveals a remarkable
complex of long canals, reservoirs, lakes, and
springs constituting a vast hydraulic system
surely constructed by Wari’s urban residents.16
Furthermore, lengthy and complex canals are
documented well at Wari provincial settle-
ments, including Pikillacta and Cerro Baúl,
so there can be no question of Wari expertise
in irrigation (see pp. 65–81, “The Wari Built
Environment”).17 In fact, Rumicolca, a great
wall across the southern end of Cuzco’s Lucre
Valley, is now recognized as a Wari aqueduct
that carried water from one side of the valley
to the other, delivering the vital liquid into
Pikillacta.18 Centuries later the Inca turned the
massive aqueduct into a gateway, sheathing
the rupture in its center with megalith mason-
ry—and surely promoting Inca appropriation
of Wari’s ancient landscape (fig. 238; see also
fig. 48).
Agricultural terracing is even less docu-
mented than hydraulic facilities around Wari,
but many hillsides that are now barren appear
to have been terraced in the past. Banks of
better-preserved terraces found farther from
the ancient city are more securely dated to
Wari times, although terraces are notoriously
difficult to date conclusively. However, terrac-
ing of the colonized Sondondo Valley, which
Katharina Schreiber studied, can be attributed
to Wari with considerable security; the same
is true in the Chicha-Soras Valley.19
Wari’s contribution to production appears
to have gone beyond irrigation and terracing,
for scarce but important evidence suggests
that Wari farmers were skillful plant breed-
ers. They modified their relative dependence
on specific crops and improved plants and
animals through artificial selection. Maize
became the primary food of both the men and
women living at Conchopata, a large Wari
town in Ayacucho. This may be one of the first
central Andean populations to have subsisted
primarily on corn, at least in the southern
highlands. Aside from its productivity, food
value, and excellence for brewing into beer,
corn weathers long-term storage very well.20
Meat consumed at Conchopata was almost
exclusively from domesticated animals, hunt-
ing having declined as the regional landscape
was intensively cultivated. Guinea pigs were
consumed in abundance. Butchering grass-
fed camelids (llamas and alpacas) at a young
age maximized meat quality and body weight,
implying specialized selection of animals for
food and long-distance transport of goods.
While direct evidence for alpaca breeding is
yet to be registered, a similar selection pro-
cess for fiber production likely complemented
the programs for other dedicated herds.21
Although precisely which agricultural and
pastoral advances originated with Wari is not
known, it is apparent that under Wari admin-
istration highland agropastoralism became
more complex and productive. The resultant
abundance promoted population growth and
fostered the development of complex social
and political systems. Social complexity,
including occupational specialization and
political ranking, permitted construction of
public works and the development of sophisti-
cated art and architecture.

Figure 239. The multilevel
mausoleum in the Mon­
jachayoq sector of the Wari
capital. Photo: Susan E.
Bergh.
25 4 W illiam H . I sbell and M argaret Young - Sánchez
Wealth and Hierarchy
Prior to the Middle Horizon little evidence ex-
ists for significant differences in social status,
wealth, or political power in the southern
sierra. Indeed, Peru’s earliest excavated evi-
dence for profound social inequality appears
on the north coast, with spectacular burials at
Kuntur Wasi.22 There, women and men were
interred with numerous symbols of supreme
wealth and power, including gold crowns.
These burials are among several indications
that hierarchical rulership appeared in north-
ern Peru during the Initial Period (1800–1000
BC) and Early Horizon (1000 BC–AD 1). The
subsequent north coast Moche intensified
social hierarchy, demonstrated by the burial
of astonishingly wealthy kings. Perhaps even
more interesting are burials of intermedi-
ate elites who represent diverse new societal
ranks as government and social life became
increasingly complex.23
Elite tombs are not documented in the
southern highlands of Peru before Wari and
the Middle Horizon, although ceremonial ar-
chitecture and luxury goods are known from
Pucara, north of Lake Titicaca.24 Social strati-
fication, as evidenced by disparities in the
quantity and quality of burial goods, was pres-
ent in both Paracas and Nasca societies.25 But
these tombs are modest by comparison with
the graves of great rulers on the north coast.
However, Wari mortuary practices document
profound new differentiation in social status,
power, and wealth in the Ayacucho region.26
Evolution of socio-political complexity must
have been rapid and transformational.
Wari’s innovative mortuary rituals in-
cluded many temporal and regional varia-
tions, but heartland graves from Conchopata
and the Wari capital are sufficiently consis-
tent to distinguish a series of types based on
grave construction. Unfortunately, the number
of unlooted graves is too small to develop a
comprehensive classification based on grave
furnishings, although these materials are use-
ful for interpreting the formal tomb types. The
types, all of which frequently contain several
individuals, range from modest (excavated
pits whose furnishings have not survived)
to intermediate (pits lined and lidded with
stone, a cavity within a rock wall, or a stone
chamber attached to a wall and sometimes ac-
companied by offerings) to elaborate (special
rooms or spaces with well-prepared chambers
that contained gold artifacts and other luxury
goods). Three variants of elaborate tombs
have been identified at Conchopata, and some
have been interpreted as containing regional
elites, probably local chiefs and provincial
governors.27 Until very recently, none had been
discovered intact by archaeologists.
The most impressive type of Wari tomb is
labeled “monumental” and examples occur
only at the Wari capital. The smaller subtype
consists of multilevel chamber tombs con-
structed partially of large, beautifully cut
stones. None has been discovered intact, but
they apparently contained numerous individ-
uals and significant wealth. These dead were
probably nobles of the Wari royal court. The
larger monumental type, represented by one
severely looted example,28 is an immense mul-
tistoried complex of megalithic chambers and
galleries penetrating many meters below the
ancient surface (fig. 239). A second, probably
similar tomb complex can be recognized at
Wari as well.29 Building these elaborate, finely
crafted tombs required engineering, construc-
tion, and stone-cutting expertise, and the
mobilization of a large workforce—a huge so-
cial investment. These great burial complexes
surely contained the bodies of Wari’s highest
rulers, accompanied by lavish offerings. Thus,
Wari graves document an expansive hierar-
chy and imply the kind of social differences
described as class structure. It seems inescap-
able that Wari’s political and social hierarchy
provided at least something of a model for
organizing inequality and power among the
later Inca, an inference strengthened by the
discovery of several elite Wari graves in the
Cuzco region.
Burial practices are not the only indication
of social and political complexity at Wari and
within its sphere of influence. Wari luxury
arts are renowned for beauty and technical
excellence. Personal items and adornments
include carved wooden cups and containers,
pyrite inlaid mirrors, and ornaments with
mosaic decoration (see pp. 217–31, “Inlaid
and Metal Ornaments”). Wari fancy ceramics
are characterized by vividly colored, well­-

Figure 240. Tapestry-
woven tunic with “Inca
key” motif; ca. 1400–1540;
Inca; camelid fiber and
cotton; 76.2 x 85.1 cm. The
Cleveland Museum of Art,
Gift of William R. Carlisle
1957.136.
255 Wari ’s A ndean L egac y
executed polychrome painting, applied to both
simply shaped and sculpturally complex ves-
sels and objects.30 To judge from large deposits
of ritually smashed ceramics found at Con-
chopata and other Wari centers, ornate high-
quality ceramics were manufactured in large
quantities, probably in state-sponsored and
state-supervised workshops (see pp. 122–43,
“Archivists in Clay,” and pp. 145–57, “Shat-
tered Ceramics and Offerings”). Such work-
shops imply the existence of designers trained
in both iconographic content and stylistic
conventions, specialized potters and paint-
ers, and quality control mechanisms. Display
and use of fancy ceramics manufactured in
these workshops advertised the wealth and
authority of the individuals or officials who
possessed them. The ostentatious destruc-
tion of such valuable ceramics as offerings at
sites like Conchopata and Pacheco must have
impressed all who witnessed it.
Easily recognizable, Wari textiles are also
technically refined and brilliantly colored.
Many depict religious icons and geometric
or simplified representational motifs.31 Even
more than fine ceramics, their manufacture re-
quired tremendous quantities of highly skilled
labor, provided by spinners, dyers, weavers,
and designers, as well as a procurement sys-
tem to supply the necessary camelid and cot-
ton fiber. These visually distinctive garments
must have been owned and worn by individu-
als of high social, political, or religious rank.
Wari tapestry-woven tunics exhibit sufficient
standardization in design and manufacture to
have been interpreted as the products of state
workshops perhaps made to clothe imperial
officials (see pp. 159–91, “Tapestry-woven
Tunics”). Paradoxically, while Wari tunics are
immediately recognizable, they are not easy to
appreciate fully, aesthetically or intellectually.
Representational motifs are often abstracted
and distorted to the point of illegibility, an
effect exacerbated by complex color schemes
and compositions. Both weavers and view-
ers may have required formal instruction to
grasp the intricacies of textile aesthetics, a
level of education likely available only to the
elite. Rebecca Stone interprets small shape
and color anomalies in Wari tapestry tunics
as evidence of artistic agency on the part of
weaving specialists, who must have exercised
some individual autonomy while working
within a carefully monitored production
system.32 (See, for example, the single face in
a tunic otherwise covered with stepped frets
[fig. 148]). Stone believes the visual dynamism
imparted by the anomalies was appreciated
by viewers and added to the prestige of wear-
ers. The state is likely to have regulated the
production, ownership, and display of luxury
ceramics and textiles, although the specific
mechanisms are unclear.
Similarly, the Inca used distinctive, finely
made ceramics and textiles as visible indica-
tors of prestige and rank within their society
and their extensive empire. Ownership of
tapestry-woven tunics comparable in quality
to earlier Wari tunics (and also produced in
state-controlled workshops) was tightly regu-
lated (fig. 240). Although the tapestry tech-
nique connoted prestige in Andean societies
Figure 241. Tunic found
in a cave burial with six
individuals near Pulacayo
in the Uyuni salt flats of
the Bolivian altiplano;
ca. 500–1000; Tiwanaku;
camelid fiber. Museo de
Arte Indígena, Sucre,
Bolivia. Photo: Fernando
Maldonado, Creaimagen,
Santiago 2010.
for centuries, it seems likely that the Inca had
direct knowledge of both Wari and Tiwanaku
tapestry-woven tunics, based on technical
similarities such as horizontal warp orien-
tation, the use of very wide looms, and the
choice of extremely fine, interlocked tapestry
wefts (fig. 241).33 Indeed, some Wari garments
could well have survived the centuries as
heirlooms, and the Inca may have encountered
burials that included tapestry-woven tunics
in construction work in coastal settlements, or
when burying their dead in ancient cemeter-
ies. Interestingly, in their garments the Inca
chose to employ the single fabric construc-
tion method more associated with Tiwanaku
tunics, but they commonly used cotton warps,
a trait associated with Wari weavings. It is
difficult to know whether the Inca recognized
the two earlier weaving traditions as distinct
from one another, and whether they associated
them with specific earlier peoples. But even
if the Inca consciously imitated Wari use of
256 W illiam H . I sbell and M argaret Young - Sánchez
emblematic ceramics, textiles, and other goods
as symbols of social rank and political author-
ity, they chose not to adopt the colorful Wari
aesthetic. Inca textiles and ceramics are gener-
ally visually sober, with simple compositions
and restrained color schemes. They incorpo-
rate little overtly religious subject matter, in
contrast to Wari preferences, but frequently
repeat a limited repertoire of geometric motifs.
If these motifs encoded specific meanings,
they were probably unknown to most people,
as the information did not survive into the
Spanish colonial period. The Inca also invent-
ed a distinctive assortment of new ceramic
forms (including the aríbalo), rather than
imitate Wari vessel shapes—aside from their
distinctive drinking chalice, the kero, which
was certainly influenced by similarly shaped
Wari and Tiwanaku flagons used in official
feasting and drinking ceremonies that pro-
moted state interests (figs. 242, 243). During
Inca times, the kero was a crucial symbol of
 Inca sovereignty partly because of its mythic
associations with the past: it figured in Inca
dynastic origin myths, in strategic gifts, and
in many imperial ceremonies, including royal
investiture.34
Political Economy
The Middle Horizon is marked by the con-
struction of intrusive, architecturally distinc-
tive structures and settlements in Ayacucho
and throughout much of Peru (see pp. 65–81,
“The Wari Built Environment”).35 These
settlements appear to have been constructed
by large, well-organized labor forces accord-
Figure 242. Kero; ca. ing to detailed, predetermined plans. Some
1400–1532; Inca; wood;
are very large and complex, and seem to have
20.3 x 15.7 cm. Phoebe
Hearst Museum, University been imposed on (rather than adapted to) local
of California, Berkeley, topography. Features such as high orthogonal
4-5096. walls, a limited number of entry points, and
complex plans composed of repeated modular
compounds suggest specialized functions.
Such sites are often interpreted as adminis-
trative centers built and occupied by Wari
state officials and allied local elites. Their
construction required millions of person days
of labor so the local residents must have been
organized as a workforce. Investigation of
facilities of this and other kinds confirms the
Figure 243 [27]. Cup
with staff deity head
and warrior (front and
back); ceramic and slip;
11.7 x 9 cm. Museum
zu Allerheiligen,
Schaffhausen, Eb15182.
257 Wari ’s A ndean L egac y
Wari administrative structure, but the nature
and full scope of political activities is still the
subject of interpretation and debate.36
The Inca likely recognized the intrusive
character of Wari provincial architecture
and could have interpreted it as an earlier
state’s assertion of dominance, which may
have inspired their own practice of building
distinctive, easily recognizable structures
within the foreign territories they controlled.
In some regions the Inca built entirely new
settlements, while in others they constructed
new buildings within existing local cities. In
either case Inca architecture is easily identi-
fied by traits such as cut-stone masonry and
trapezoidal doors and niches. Such conspicu-
ous constructions fulfilled practical economic,
military, and administrative functions but
also advertised Inca political and spiritual do-
minion over the local landscape and peoples.37
Inca constructions near the ruins of Wari and
Tiwanaku may have been especially impor-
tant in this regard, laying claim not only to
land and contemporary inhabitants, but also
to ancient prestige.
How Wari’s political economy functioned
is poorly understood, but considerable evi-
dence has been presented for feasting at Con-
Figure 244. Aríbalo; ca. to be learned about the range of ways in which
1400–1532; Inca; ceramic
the Wari employed feasting, there are sugges-
and slip; 76.2 x 54.6 cm.
Denver Art Museum tions that the Wari used a state installation in
Collection, Funds from the the San Miguel Valley to manage agricultural
Burgess Trust, Walt Disney labor,41 and that Inca ceremonies, intended
Imagineering, Alianza de
las Artes Americanas, and to “reciprocate” agricultural and other kinds
Frederick and Jan Mayer, of work by commoners, owed a great deal to
1993.25. Photo: © Denver this Wari legacy. The agricultural center at
Art Museum.
San Miguel was not unique within Wari’s
hinterlands, which have not been intensively
investigated. Wari may even have developed
special state farms, another Inca institution, as
suggested by the intensive occupation of the
Sondondo Valley that included massive ter-
racing, roads, administrative facilities, ritual
complexes, and perhaps even monumental
tombs.42
It was noted earlier that Wari’s agricultural
innovations greatly increased the maize sup-
ply essential to brewing chicha (native corn
beer). Copious supplies of chicha were central
to Inca-sponsored feasting, another indica-
tion that Inca practice may have depended on
chopata and other Wari settlements (see pp. Wari innovation. Tamara Bray and Anita Cook
82–101, “The Art of Feasting”).38 Feasting was compare the forms and decorations of Wari
an important mechanism for affirming rank, and Inca jars and cups believed to have been
establishing alliances, imposing debt, con- used for serving chicha in such state-spon-
firming inequality, and mobilizing labor in the sored feasting.43 They note that Wari jars take
Figure 245. Khipu; ca. ancient Andean world.39 Maurice Godelier ar- the form of a human being (probably a state
1400–1532; Inca; camelid gued that the Inca used feasting to convert an official) who sometimes wears a tunic (fig. 237;
fiber and cotton; 85 x 108 ancient Andean institution of reciprocity into see also figs. 54, 198). Inca jars, or aríbalos,
cm. The Cleveland Museum
of Art, Gift of John Wise an institution that provided labor to govern- have a different and distinctively Inca form
1940.469. ment by elites.40 Although much more needs but are also fundamentally anthropomorphic
in concept (fig. 244). Their painted decoration
is not overtly religious but makes reference to
Inca tunic designs. Bray and Cook’s analysis
suggests that the Inca manipulated ceramic
forms and meanings in ways conceptually
similar to Wari practice in order to convey
the power, wealth, and generosity of the state.
However, they either rejected or were unaware
of the Wari state’s more literal presentation of
the same concepts.
Cuzco, the Inca capital, had many palaces,
and William Isbell44 has argued that one of the
complexes excavated at Wari was, at least ini-
tially, a palace and not a temple as its excava-
tors suggested.45 Furthermore, sites identified
as Wari’s provincial administrative centers
possess many features associating them with
Andean palaces.46 Wari administrators were

258 W illiam H . I sbell and M argaret Young - Sánchez


Figure 246. Painted textile,
from Chimú Capac; ca.
800–1000; cotton and
pigment; 170 x 89.5 cm.
Phoebe Hearst Museum,
University of California,
Berkeley, 4-7221.
259 Wari ’s A ndean L egac y
apparently housed in splendor at strategic
locations throughout the empire.
Wari political organization may have in-
cluded another feature profoundly associated
with the Inca. Although writing was unknown
in central Andean cultures, the Inca employed
khipu, devices made of multicolored strings
on which complex knots were tied to record
information (fig. 245).47 Middle Horizon khipu
are also known although they differ from Inca
khipu in important respects (see fig. 180; also
see [155], p. 276).48 A few have been found
on the coast in association with Wari style
ceramics and may be Wari. If so, they imply
an accounting system that may have required
trained specialists, as it did among the Inca.
The small number of surviving Wari khipu
suggests that they were not used in exactly the
same way, or that the Wari state’s accounting
needs were much more limited.
Religion
Some experts believe that Wari contributed
significantly to later religions, especially the
Inca pantheon of deities. What we know of
Wari’s religion is primarily derived from the
imagery represented on Wari ceramics, tex-
tiles, and other objects (see, for example, figs.
75a–f). This imagery is largely shared with
Tiwanaku, where it was also carved on stone
statues and architectural elements (see fig. 6a).
Scholars believe this religious iconography
originated in the Lake Titicaca region and far-
ther south, and was adopted first by Tiwanaku
and soon after by Wari. As discussed else-
where in this volume, Wari art consistently
represented four primary supernatural images:
a staff deity, a rayed head (probably an avatar
of the staff deity, or vice versa), staff-bearing
attendants in profile, and sacrificers (see pp.
103–21, “The Coming of the Staff Deity”). One
of the prominent features of the staff deity and
rayed head are the head emanations, which
intuitively suggest the sun. Indeed, Dorothy
Menzel identified male and female variants of
the staff deity and argued that they represent
the Wari sun and moon deities, antecedents
for the similar Inca pair.49 Dual male and fe-
male variants of the staff deity are not known
for earlier Pucara or contemporary Tiwanaku
religious art, so this pair appears to be a spe-
cifically Wari innovation.
Some fine Wari tapestry-woven tunics de-
pict musicians and elites engaged in ceremo-
nies50 that in the Inca religion were carefully
regulated by a complex calendar. Indeed, Tom
Zuidema argues that the Inca calendar inte-
grated a solar and lunar almanac, the former
managed by men and the latter by women.51
Regular astronomical observations were made
and information was stored on the knotted
strings of khipu. The patterns of repetition
on some Wari textiles imply the same solar/
lunar calendar of twelve months, centered
on the December solstice.52 Architecture and
sculpture at Tiwanaku imply similar solar and
lunar organization as well.53
Menzel suggested that, in the era imme-
diately after Wari’s decline, Wari’s religion
influenced central coast peoples, who blended
together Moche and Wari elements.54 Art from
Chimú Capac depicts a so-called Sky God, a
male shown in a frontal Wari posture with a
Moche-style arc over his head, perhaps repre-
senting the sky (fig. 246). The arc terminates in
a serpent head at each end. Rather than staffs
in both hands, the Sky God typically holds
implements such as a club, a knife, or even a
shield. These tools provoked Menzel to suggest
identification with the Inca Thunder God, de-
picted as a spectacularly dressed warrior with
shield, twirling a sling. The crack of his sling
created claps of thunder, and the flash of his
gold garments produced lightning. Numerous
filler elements complete the Sky God scene,
 including stars, animals, and an occasional
Wari-inspired profile attendant. Frontal fig-
ures, sometimes holding staffs, are common in
the ceramics and textiles of later central and
north coast cultures, including Chancay, Lam-
bayeque, and Chimú (fig. 247), harking back
(in at least a general way) to the earlier Wari
staff deity or to Moche-Wari hybrid images. It
seems clear that Wari religious beliefs were
not adopted wholesale by these later cultures
and did not displace the indigenous coastal
religions. Instead, Wari beliefs and religious
imagery were incorporated and adapted to
local religious and artistic traditions.
Some scholars believe that Pachacamac,
an Andean creator god in Inca times, was a
Wari deity brought to the coast by Ayacucho
emigrants. The deity’s principal temple and
oracle were located at the site of Pachacamac,
near Lima. In the temple was a Janus-faced
wooden sculpture carved in a Wari-influenced
style.55 However, the association of Inca-period
creator deities—who may represent a single
creator god known by different names—with
Figure 247. Panel with Wari origins may be more compelling. “Vi-
frontal figures and birds, racocha” was part of another name Andeans
from the north coast; ca. used to refer to a supreme creator, and Inca-
1100–1400; camelid fiber
and cotton; 76 x 23.7 cm.
period temples dedicated to him are often
Museum of Fine Arts, found next to Wari archaeological sites, most
Boston, Denman Waldo famously the immense Viracocha temple at
Ross Collection, 10.267.
Raqchi, south of Cuzco. Significantly, recent
Photo: © 2012 Museum of
Fine Arts, Boston. excavations at Raqchi in a set of round build-
ings long thought to be Inca storehouses
produced Middle Horizon radiocarbon dates
Figure 248. Tunic with
Wari-derived figures, from
Chimú Capac; ca. 800–1000;
camelid fiber and cotton;
160 x 106.7 cm. Phoebe
Hearst Museum, University
of California, Berkeley,
4-7827.
26 0 W illiam H . I sbell and M argaret Young - Sánchez
along with evidence that the circular struc-
tures were built by Wari settlers,56 corroborat-
ing an association of Viracocha (and probably
creator gods under other names) with Wari
and suggesting significant continuities be-
tween Wari and Inca use of ritual landscape in
creator worship.
Artistic Legacy
Wari’s impact on later coastal art styles ap-
pears to have been limited. Late Middle
Horizon coastal textiles, probably created
after Wari declined, incorporate Wari mo-
tifs, including attendant figures and frontal,
staff-bearing figures; these textiles have been
found at sites such as Huaca Malena, Pacha­
camac, Chimú Capac (fig. 248), El Castillo, and
Moche.57 While a few are executed in inter-
locked tapestry (the technique used in many
Wari textiles), most employ local techniques
such as slit tapestry or brocade. These textiles
are generally less finely woven than their
Wari prototypes and the imagery is less ac-
curate and standardized. Nor did later coastal
weavers make any real attempt to imitate
Wari tunics. Local garments have different
proportions and construction methods, and
no local coastal textiles remotely approach
the complexity and sophistication of Wari
tapestry-woven tunics with regard to composi-
tion and color scheme. Simpler Wari products
(four-cornered hats and corner-ornamented
cloths) often incorporate multiple repetitions
of small-scale motifs, with diagonally aligned
Figures 249a, 249b [159].
Mantle (view and details);
cotton and camelid fiber;
177.8 x 177.8 cm. The
Metropolitan Museum
of Art, New York, The
Michael C. Rockefeller
Memorial Collection,
Bequest of Nelson
A. Rockefeller, 1979,
1979.206.462. Image: ©
The Metropolitan Museum
of Art. Image source: Art
Resource, NY

261 Wari ’s A ndean L egac y

 color repeats (figs. 249a, 249b; see also [158],
p. 277). Such easily understood and imitated
design principles may have inspired Chimú
and Chancay textiles, especially tapestries.
Many of them were probably woven in house-
holds, or perhaps in an assortment of loosely
regulated workshops. The most sumptuous
Chimú textiles were likely woven in palace
workshops for high-ranking wearers. Detailed
study of some garments has revealed complex
encoded symbolism that likely derives from
earlier north coast systems of representation.58
Sculptural coastal objects (ceramics,
carved wood, metalwork) include rigid fron-
tal human or deity figures, but their resem-
blance to Wari sculptures is very general.
Annex
Multifigural compositions like those seen in
Lambayeque and Chimú ceramics and metal-
work, as well as Chancay woven “doll” scenes,
surely derive from earlier coastal traditions.
Wari sculptures usually portray a single fig-
ure, with information conveyed by pose and
costume rather than through narrative.
Gordon McEwan notes several formal
similarities between Wari provincial centers
such as Pikillacta and the ciudadelas of Chan
Chan, the capital of the Chimú Empire (fig.
250).59 Shared characteristics include a rectan-
gular shape (oriented roughly north-south) en-
closed by extremely high walls. Both Wari and
Chimú walls are characterized by construc-
tion sections that seem to correspond to the
labor contributions of distinct social groups.
There are few entrances, and passageways to
interior spaces are long, narrow, and easily
controlled. Compound interiors are divided
into three large sectors, and expansion needs
were accommodated by building external
annexes rather than by enlarging the original
perimeter walls.
Burial platform
Chan Chan’s ciudadelas are usually inter-
preted as the palaces of Chimú kings, built
N 100 meters
Audiencia successively to house the royal family and
retainers, store and administer goods, and
Storerooms
Figure 250. Plan of the protect each king’s tomb with its associated
Rivero ciudadela at Chan
Walk-in-well
wealth of burial offerings. McEwan believes
Chan, the Chimú capital. that as the Chimú kingdom developed in size,
After McEwan 1990, 106;
courtesy Michael Moseley. wealth, and complexity, its leaders sought an
architectural form associated with expansive,
secular state power. Wari provincial centers
became their model. McEwan notes that the

262 W illiam H . I sbell and M argaret Young - Sánchez

 buildings within the ciudadelas are gener-
ally distinct in form from those within Wari
provincial compounds. The exceptions are
small conjoined rectangular rooms, present in
all the ciudadelas and some of the Wari cen-
ters, and interpreted as storage rooms for the
Chimú; the Wari function is still unclear.
As mentioned above, Wari provincial
centers could have inspired certain aspects of
their Inca counterparts. But Wari architecture
promotes a very different experience than Inca
built environments. Wari architects virtually
ignored the lay of the land, imposing orthogo-
nal cellular grids as though the ground were
flat.60 Inca engineers, however, tailored natural
features into architectural grids as well as
individual buildings. Inca centers, especially
those located in dramatic landscapes, promote
an experience of harmony between nature and
culture.61 Conversely, Wari planned settle-
ments promote a sense of cultural domination
over nature through enclosure and imposed
order.
The Wari built fieldstone walls and cov-
ered them with shiny white plaster that must
have overwhelmed the eye with its brilliance
on sunny Andean days. The Inca manipulated
the natural color and grain of stone, contrast-
ing form and shadow with megalithic facades
that sometimes employed deeply sunken
joints.62 Yet, a few architectural forms may
link Wari and Inca architecture. The Inca
kallanka, a large roofed hall with multiple
entrances and niches set into the interior
walls, was used in religious or civic activi-
ties involving large gatherings of people. John
Topic believes that the kallanka form may
have been derived from earlier Wari niched
halls at Pikillacta and Viracochapampa.63 An-
other Wari architectural form is the D-shaped
structure or temple. Such structures were
built both at the Wari capital and at provincial
centers; excavation indicated that they were
associated with ritual activities including
both burial and sacrifice.64 Curved masonry
walls in Inca architecture, such as Cuzco’s sun
temple and the shrine at Kenko, are seemingly
also associated with sacredness, but more
evidence is necessary to establish a clear con-
nection between the Wari D-shaped structures
and Inca temples and shrines. Finally, the
263 Wari ’s A ndean L egac y
patio group, a basic unit of Wari residential
space,65 may continue as the kancha in Inca
town planning.66 Both are walled enclosures
and square, rectangular, or trapezoidal in
shape with a central patio fronted symmetri-
cally on all, or at least several, sides by a large
room, creating an apartment complex with an
open common space at its center. Patio groups
or kanchas could be placed side by side in a
grid, filling an entire sector of a community.
The two differed primarily in that Wari rooms
are attached to the outer enclosure wall and
run the entire length of the enclosure, while
Inca rooms are short and freestanding, so cor-
ners of the kancha are vacant.
Although Inca religious beliefs and calen-
drical knowledge may derive from those of
Wari and Tiwanaku, there is a remarkable
disparity in the prominence of religious
imagery in their art styles. Religious imagery
is ubiquitous in Wari and Tiwanaku ceram-
ics, textiles, and sculpture, while explicit
representation of deities is absent in Inca art.
In Wari art the supernatural figures rarely in-
teract or enact mythical episodes. Instead, the
staff-bearing deities and attendant figures (or
just their heads) are usually employed as icons
or emblems that recalled religious concepts
while identifying the objects on which they
appeared with the Wari state. Symbols such as
eight-pointed stars, “Inca keys” (fig. 240), and
stepped diamonds similarly served as visual
identifiers of Inca manufacture; some of them
may have conveyed religious content as well,
although they can no longer be “read.”
Some Inca symbols and geometric ele-
ments may derive from Wari art. For example,
the Wari and Tiwanaku used zigzag and con-
centric square motifs consistently to ornament
several of the staff deity’s and profile atten-
dants’ accessories—most prominently the staff
itself, which the two Middle Horizon cultures,
like the Inca, seem to have regarded as a pri-
mary symbol of authority (see figs. 1, 5, 6).67 In
Inca art zigzags and concentric squares com-
monly decorate keros (fig. 242) and tapestry-
woven tunics (fig. 240), both central imperial
symbols. In a colonial drawing made in about
1615, the native author Joan de Santa Cruz
Pachacuti Yamqui used a concentric square
to depict the cave from which the founders of
 the Inca state were believed to have emerged;
thus, the motif has been interpreted to mean
“place of origin.” Susan Bergh and Anita Cook
suggest that the Inca concentric square refers
to both a place in the landscape and to the
Wari and Tiwanaku, mythic ancestors whose
symbols of ancestry the Inca perhaps ap-
propriated and put to similar use. Cook also
identifies a diamond-in-a-square motif in Wari
tunics, both tapestry-woven and tie-dyed, that
she believes continues as an important element
in Inca art.68 Examples of the ubiquitous Inca
motifs known as toqapu, such designs suggest
that other, under-researched Middle Horizon
geometric images may have supplied inspira-
tion for the Inca visual vocabulary. Even if
these elements hark back to the earlier Tiwa­
naku and Wari symbol systems, they have been
thoroughly integrated into the distinctive Inca
stylistic and symbolic language. The Inca may
well have imitated Wari in the ways in which
they manipulated art, architecture, and other
aspects of material culture to legitimize state
power, both in Wari’s Ayacucho homeland and
among foreign peoples and territories.
Wari in Cuzco
Because Tiwanaku’s importance to the Inca
is recorded in Spanish colonial documents, it
was to Tiwanaku, not Wari, that researchers
of the past looked to understand the origin
of the Inca Empire, which had its capital in
Cuzco, very near Pikillacta, a major Wari pro-
vincial installation. Now that Wari and Tiwa-
naku are recognized as separate cultures and
independent polities it is time to consider the
contribution of Wari. Several issues must be
considered to identify Wari convincingly as a
major, direct inspiration of the Inca. First, how
much influence did Wari have in Cuzco, and
what was its nature? Did Wari behave like an
ideal empire, conquering, incorporating, and
reorganizing the Cuzco region as a provincial
territory composed of tribute payers and thus
imposing a technology of empire that might be
retained? Or was Wari’s influence minor, lim-
ited largely to display of symbols and partici-
pation in rituals?
Second are the questions of time and the
nature of memory. When did Wari collapse
and how many centuries separate it from the
26 4 W illiam H . I sbell and M argaret Young - Sánchez
foundation of the Inca Empire? Could actual
institutions of Wari administration survive
these centuries, or were continuities limited
to myths, images, and ritualized visits to ruins
and historic landscapes? Dorothy Menzel’s
Wari chronology, written in 1964 and 1968,
asserted the abandonment of the capital and
the collapse of imperial power around 850.
Excavations in Ayacucho now suggest that
Wari ceramic styles in the heartland survived
after 900, possibly as late as 1000 or 1050.69
This possibility takes a century or two off the
hiatus between Wari and Inca, although the
gap is still significant if we accept current
inferences that the Inca were little more than
a local chiefdom, engaged in farming and raid-
ing, until at least 1350.70
A strong Wari imperial administrative
presence is now implied in Cuzco, for Wari
elites are newly documented in widely sepa-
rated locations. The key discovery on which
this inference is based is a Wari elite tomb
at Espíritu Pampa, in the high jungles of the
Vilcabamba, excavated by Cuzco archaeologist
Javier Fonseca.71 The site is replete with Wari
pottery, and the tomb itself conforms well to
one of the elaborate Wari tomb types men-
tioned above,72 representing the highest status
Wari elites buried outside the capital. This,
the first elaborate Wari burial to be found in-
tact by archaeologists, contained many luxury
goods, especially silver artifacts along with
several gold items. (One ornament recovered
is very similar to that shown in Figure 219b.)
However, it is not the only elaborate Wari buri-
al in Cuzco. Julinho Zapata describes another,
which was looted at the site Batan Urqo.73
Less well known examples at Marcaconga and
Pomacanchi, farther south, were also looted.74
(Silver plumes from the Pomacanchi tomb
appear in Figure 12.) To the west a Wari elite
tomb was plundered at Curawasi in the Pam-
pas River drainage. While information about
looted tombs is incomplete and not entirely
reliable, most likely the graves represent local
chiefs who served as provincial Wari gover-
nors. They were buried in Wari style tombs
with Wari style symbols of authority as well as
some local-style ornaments. Another amazing
new discovery, found beneath the surface of
a corridor at Pikillacta, consists of miniature
 Wari warrior figurines along with representa-
tions of supernatural beings, captives, and
sacrificial victims (see figs. 225, 226).75 Taken
together, these finds may indicate that the
Wari depicted their occupation of Cuzco as a
military invasion, which they followed up by
establishing a formal administrative structure.
These new data help clarify the relationship
between Wari and the Cuzco region during the
Middle Horizon. Furthermore, some ethnohis-
torians believe that memory of Wari in Cuzco
was prominent enough to survive through the
intervening centuries, influencing even oral
accounts that Spanish invaders recorded in
the sixteenth century.
Inca oral history declared that a catalyst
for the founding of the Inca Empire was a
battle against the Chankas, a confederation of
enemies from the neighboring region of Anda-
huaylas. This victory initiated the career of a
young nobleman named Cusi Yupanqui (later
known as Pachacuti), who led the defense of
Cuzco, usurped the crown, and inaugurated
the Inca’s irresistible military expansion.
Celebrated Peruvian ethnohistorian Maria
Rostworowski believes that the Chanka ac-
count is semi-historical, referencing the occu-
pation of Cuzco by Wari, whose homeland also
lies west of Cuzco, not far beyond Andahuay-
las.76 Rostworowski goes on to suggest that
the name Pachacuti came from the memory of
ancient and prestigious Wari rulers. Juhua
Hiltunen argues that one chronicler, Fernando
de Montesinos, gives a list of royal dynasties
that fills the gap between the last Wari kings
and the earliest Inca, so perhaps conditions
were sufficient for more or less direct continu-
ity between the social and political institu-
tions of the Wari and the Inca.77 Conversely,
many scholars are reserved about accepting
Montesinos’s claims.
Be that as it may, a case can certainly be
made for Wari rule of the Cuzco region, and
for at least some historical memory and influ-
ence of these events among the Inca. Yet as
noted above, Inca accounts of their origins and
rise to power explicitly link them to Tiwa­
naku and Lake Titicaca, despite their greater
geographic distance from Cuzco. The legends
emphasize the Inca divine right to rule, link-
ing them to the gods whose sacred place of
26 5 Wari ’s A ndean L egac y
origin was Lake Titicaca. Inca legends do not
detail the political, religious, economic, and
administrative strategies they used to consoli-
date and integrate their empire, and shed no
light on whether the Inca invented these strat-
egies, or inherited them from earlier polities.
This silence likely reflects Inca cultural values
and the lessons they chose to repeat: the Inca
understood their greatness as a manifestation
of divine origin and sanction, rather than a
product of statecraft and administrative exper-
tise inherited from an earlier empire, whether
Wari or Tiwanaku.
Wari: One of Many Important Players
Wari seems best understood as a centralized
state and empire that had many important
influences on subsequent Andean cultures,
including the Inca. Its cultural legacy is
profound. Some of these influences have been
discussed here, but many have been omitted
for the sake of brevity. For example, it can be
argued that the Wari Empire was responsible
for spreading the Quechua languages spoken
today in Peru78 although a great deal of re-
search is required to fully evaluate the argu-
ment. Still other Wari influences likely remain
to be discovered. Certainly Wari’s most impor-
tant impact on subsequent Andean tradition
(accomplished in conjunction with Tiwanaku)
is the change it wrought on the axis of Andean
power, resulting in the rise of the southern
sierra. This transformation required major
developments in agropastoral production as
well as in the organization of people and poli-
tics. Andean demographics changed, as did
the nature and complexity of government in
southern highland Peru. The new political and
economic core that emerged is a Wari legacy.
Wari and Inca were, however, far from
identical. How much the Inca owed to di-
rect development from Wari is not apparent.
Indeed, subsequent southern highland peoples
neglected Wari’s representational imagery in
ceramic decoration and fine textiles almost
completely. Wari administrative and ceremo-
nial architecture disappeared, although there
seems to have been some continuity in sa-
cred landscapes, sites of creator worship, and
perhaps in building forms such as the Inca
kancha and kallanka.
 It is, however, clear that the Inca invented
new institutions that differed from Wari’s. For
example, Inca mortuary practices diverge in
that they emphasized permanent preservation
of the corpses of important leaders—individu-
als recognized as founders of descent groups.
These Inca leaders, male and female, were
mummified and their corpses participated in
activities of the living. The dead maintained
authority over their descendants, or panaca,
controlling its resources, affirming critical
decisions, and participating in public celebra-
tions.79 There were no royal Inca tombs but
rather revered royal ancestor mummies who
traveled from their palaces to public plazas
and temple celebrations in accord with their
social agendas. Although elite Wari tombs
contain the remains of individuals who would
have founded kin groups in Inca society, they
were securely sealed in graves that were not
particularly good environments for preserv-
ing human bodies. Holes through grave roofs,
plugged with stone stoppers, did provide for
exchange between the living and the dead, so
archaeologists should recognize some kind
of Wari ancestor veneration. As noted above,
relationships between the living and the dead
formalized the incorporation of both peoples
and places into the Wari realm, but ancestor
mummies were not active, ongoing partici-
pants in Wari ceremonial life.80
Another unique institution that distin-
guished the Inca Empire is its fascinating
tribute system, based on labor. When common-
ers paid their taxes by working for the state,
26 6 W illiam H . I sbell and M argaret Young - Sánchez
they were fed and supplied by the state; when
they performed military service, the state
clothed and armed them; when commoner
women spun and wove for the state, they
received camelid fiber from which to pro-
duce yarns and textiles. To supply and equip
tribute payers, and to feast them at the end of
their service, the Inca state constructed vast
complexes of storehouses where quantities of
goods produced for the state were deposited.
Some archaeologists believe that rows of small
Wari buildings were also storehouses and that
the Wari political economy also depended on
state-subsidized labor tribute. But excavations
have not confirmed an emphasis on storehous-
es in Wari centers. Bill Sillar, Emily Dean, and
Amelia Pérez Trujillo suggest that the rows of
circular Wari buildings at Raqchi were dormi-
tories for workers conscripted by the state; 81 if
so, Wari taxation was more forced internment
than Inca tribute. Future research will no
doubt reveal more about Wari tribute, includ-
ing the role of feasting.
Wari and its institutions require and
deserve intense archaeological investigation
to elucidate the ways in which they reshaped
central Andean cultural tradition and social
evolution. But subsequent Andean peoples
also made new inventions and innovative
adaptations that departed from Wari ante-
cedents. The Inca seem to have shared a great
deal with Wari, but also created a great deal
of their own culture. Central Andean culture
was a complex process, in which Wari was an
immensely important player, but one of many.
Notes 1. W. Isbell 2008; W. Isbell 2004a.
2. W. Isbell 2006.
3. Lechtman forthcoming; Lechtman
and Frame 1985.
4. W. Isbell 2010.
5. W. Isbell and Knobloch 2009; W.
Isbell and Knobloch 2006.
6. Sutter 2011; Sutter 2009.
7. W. Isbell 2008. The authors of
this essay disagree on how securely
Wari can be identified as an empire.
The view that the essay advances is
William Isbell’s. Margaret Young-
Sánchez is more cautious, adding
that the terms “king,” “royal,” and
“empire” have strong associations
with European political forms and
do not necessarily apply to Andean
antiquity.
8. T. Topic and J. Topic 2010a; see
also J. Topic and T. Topic 2001; J.
Topic 1994; J. Topic 1991; T. Topic
1991.
9. Anders et al. 1998; Knobloch
2005; Knobloch 1991; Knobloch
1989; Knobloch 1983; Knobloch
1976; Menzel 1977; Menzel 1964.
10. W. Isbell 1991; Menzel 1968;
Menzel 1964; Schreiber 2001;
Schreiber 1992.
11. Wari religious imagery belongs
to a great tradition that was long
associated with Tiwanaku but can
now be identified as part of an even
greater sphere of interaction and
shared beliefs that encompassed the
southern Andes. It has been pro-
posed that this tradition be known
as SAIS, the Southern Andean
Iconographic Series (W. Isbell and
Knobloch 2009).
12. W. Isbell and Knobloch 2009;
W. Isbell and Knobloch 2006.
13. Lambayeque culture is also
known as Sicán.
14. W. Isbell 2009.
15. Janusek 2008.
16. Pérez Calderón 2008; Pérez
Calderón 2003; Pérez Calderón 2001.
17. For Pikillacta, see McEwan
2005c. For Cerro Baúl, see Williams
2003; Williams 2001.
18. Valencia Zegarra 2005.
19. For the Sondondo Valley, see
Schreiber 1992; Schreiber 1991b.
For the Chicha-Soras Valley, see
Meddens and Branch 2010; Meddens
1991.
20. Finucane et al. 2006; Green
and Whitehead 2006; Sayre and
Whitehead 2003; Whitehead per-
267 Wari ’s A ndean L egac y
sonal communication.
21. Finucane et al. 2006; Rosenfeld
2006; Rosenfeld 2004.
22. Onuki 1997; Onuki and Inokuchi
2011.
23. Alva 2001; Alva and Donnan
1993; Billman 2010; Billman
2002; Castillo Butters and Uceda
2008; Castillo Butters et al. 2008a;
Chapdelaine 2001; Donnan 2007;
Donnan and Castillo Butters 1992;
Uceda and Morales 2010.
24. Chávez 2004; Conklin 2004b;
Young-Sánchez 2004b; Young-
Sánchez 2004c.
25. Proulx 2008; Silverman and
Proulx 2002.
26. W. Isbell 2004a.
27. W. Isbell and Kopissari 2011;
W. Isbell 2004a.
28. Pérez Calderón 1999.
29. W. Isbell et al. 1991.
30. Morris and von Hagen 1993,
109–23; Seville 2001.
31. Bergh 1999; Conklin 1996; A.
Rowe 2005; Sawyer 1963.
32. Stone-Miller 1992b, 38–41.
33. Oakland 1986a; Oakland 1986b.
34. Cummins 2002.
35. W. Isbell 2006; W. Isbell 2004a;
W. Isbell 1991; McEwan 2005c;
Schreiber 1992.
36. Jennings 2010c.
37. Morris and von Hagen 1993, 177.
38. Cook and Benco 2001; Cook and
Glowacki 2003; W. Isbell 1985.
39. Bray 2003a; Dietler 2001.
40. Godelier 1977a.
41. W. Isbell 1977b.
42. Schreiber 2005b; Schreiber 1992.
43. Bray and Cook 1997.
44. W. Isbell 2006; W. Isbell 2004a.
45. Bragayrac Dávila 1991; Gonzales
Carré and Bragayrac Dávila 1986.
46. W. Isbell 2006; W. Isbell 2004a.
47. Urton 2008; Urton and Brezine
2011.
48. Conklin 1982; Urton and Brezine
2011, 321, fig. 13.1.
49. Menzel 1977.
50. A. Rowe 1979.
51. Zuidema 2011a; Zuidema 2010.
52. Zuidema 2011b; Zuidema 2009.
53. Benitez 2009; Zuidema 2009.
54. Menzel 1977.
55. Dulanto 1991; Franco Jordán
2004; Kaulicke 2001; Shimada 1991.
56. Sillar et al. forthcoming.
57. For Huaca Malena, see Angeles
Falcón et al. forthcoming; for
Pachacamac, see VanStan 1967; for
Chimú Capac, see Menzel 1977, 114;
for El Castillo, see Prümers 1990,
vol. 2; for Moche, see Menzel 1977,
118.
58. Santiago 2005.
59. McEwan 1990.
60. W. Isbell 1992; W. Isbell and
Vranich 2004.
61. Dean 2010; Gasparini and
Margolies 1980; Niles 1992; Protzen
1999.
62. Like the Inca, the Tiwanaku
were famous for precisely cut
and fit megalithic stone architec-
ture. However, Wari did employ
cut-stone masonry, including
megalithic polygonal architecture.
Unfortunately, most of Wari’s spec-
tacular stonework was quarried by
the Spanish for millstones, church
walls, and other colonial construc-
tions.
63. J. Topic 1986.
64. Cook 2001a.
65. W. Isbell 1991.
66. Hyslop 1990, 10–21; Kendall
1985, 352; Kendall 1976, 92.
67. Bergh 1999, 119.
68. Cook 1996.
69. W. Isbell 2001; W. Isbell and
Cook 2002.
70. J. Rowe 1946.
71. Fonseca et al. 2011.
72. Fonseca et al. 2011; W. Isbell
2004a.
73. Fonseca et al. 2011; Zapata 1997.
74. See Chávez 1985.
75. Arriola Tuni 2008; Arriola Tuni
and Tesar 2011.
76. Rostworowski de Diez Canseco
1999, 35; Rostworowski de Diez
Canseco 1997; see also Zuidema and
Burga 1989.
77. Hiltunen 1999; see also Hiltunen
and McEwan 2004; Montesinos
[1644] 1882.
78. W. Isbell 2011.
79. W. Isbell 1997.
80. See McEwan 2005c, McEwan
and Williams 1998, and this volume
for a contrasting opinion.
81. Sillar et al. forthcoming.
Checklist of the Exhibition
NOTE TO THE READER
The vast majority of works in
this checklist date to the Middle
Horizon (AD 600 to 1000) and
belong to Wari or Wari-influenced
styles, many certainly or probably
from coastal regions and often
described as “coastal Wari.” An
asterisk (*), however, indicates
works of confirmed or possible
Middle Horizon date that do not
necessarily belong to a Wari or
Wari-influenced style. Objects
have been organized in alphabeti-
cal order by medium. The object’s
name is followed by its style (in
parentheses, ceramics only) and
any confirmed or reported pro-
venience. Dimensions are given
in centimeters, height x width x
depth or height x diameter. Where
known, radiocarbon dates are also
given parenthetically (95 percent
confidence interval). Dr. Patricia
J. Knobloch provided advice about
ceramic style attributions. For an
updated checklist and errata, see
www.ClevelandArt.org/Wari.
26 8 checklist of the E xhibition
CERAMICS
1. Bowl with bird-headed
staff-bearing creature in profile
(Conchopata style); Conchopata;
ceramic and slip; 12.7 x 33.4
cm. Museo Histórico Regional
“Hipólito Unanue,” Ayacucho,
MHRA-925. (Figure 105)
2. Fragment of a faceneck vessel
(proto-Viñaque style); Conchopata;
ceramic and slip; 43 x 39.5
cm. Museo Histórico Regional
“Hipólito Unanue,” Ayacucho,
MHRA-1778. (Figure 75c)
3. Fragment of a faceneck vessel
(proto-Viñaque style); Conchopata;
ceramic and slip; 44.5 x 48
cm. Museo Histórico Regional
“Hipólito Unanue,” Ayacucho,
MHRA-1779. (Figures 75a, 75b)
4. Fragment of a faceneck vessel
(proto-Viñaque style); Conchopata;
ceramic and slip; 53 x 48 cm.
Museo Histórico Regional
“Hipólito Unanue,” Ayacucho,
MHRA-1784. (Figure 75d)
5. Urn fragments with warriors
(Conchopata style); Conchopata;
ceramic and slip; 40 x 85 cm.
Museo Histórico Regional
“Hipólito Unanue,” Ayacucho,
MHRA-1777. (Figure 103)
[23]
6. Urn with heads of mythical
creatures (Conchopata style);
Conchopata; ceramic and slip; 34
x 64 cm. Museo Histórico Regional
“Hipólito Unanue,” Ayacucho,
MHRA-834. (Figure 62)
7. Camelid head vessel (Robles
Moqo style); Pacheco; ceramic
and slip; 15.6 x 17.8 x 12.6 cm.
Museo Nacional de Arqueología,
Antropología e Historia del Perú,
Lima, C-55032. (Figure 139)
8. Camelid skull vessel (Robles
Moqo style); Pacheco; ceramic
and slip; 17 x 22.8 x 11.8 cm.
Museo Nacional de Arqueología,
Antropología e Historia del Perú,
Lima, C-55035. (Figure 140)
9. Reclining camelid vessel (Robles
Moqo style); Pacheco; ceramic
and slip; 17.5 x 24.8 x 80 cm.
Museo Nacional de Arqueología,
Antropología e Historia del Perú,
Lima, C-55041. (Figure 138)
10. Standing camelid vessel
(Robles Moqo style); Pacheco;
ceramic and slip; 74.5 x 51.5
x 32 cm. Museo Nacional de
Arqueología, Antropología e
Historia del Perú, Lima, C-60592.
(Figure 137)
11. Cup with supernatural head
and plants (Robles Moqo style);
Pacheco; ceramic and slip; about
59 x 55 cm. Museo de América,
Madrid, 8.315bis. (Figure 133)
[29]
12. Faceneck vessel (Robles Moqo
style); Pacheco; ceramic and slip;
45.6 x 30.8 cm. Museo Nacional
de Arqueología, Antropología e
Historia del Perú, Lima, C-63067.
(Figure 134)
13. Faceneck vessel (Robles Moqo
style); Pacheco; ceramic and slip;
50 x 35.3 cm. Museo Nacional
de Arqueología, Antropología e
Historia del Perú, Lima, C-64075.
(Figure 135)
14. Faceneck vessel (Robles Moqo
style); Pacheco; ceramic and slip;
50 x 34.8 cm. Museo Nacional
de Arqueología, Antropología e
Historia del Perú, Lima, C-66969.
(Figure 136)
15. Urn with plants (Robles Moqo
style); Pacheco; ceramic and
slip; 56 x 86 cm. Museo Nacional
de Arqueología, Antropología e
Historia del Perú, Lima, C-54798.
(Figure 130)
16. Urn with staff deities (Robles
Moqo style); Pacheco; ceramic and
slip; 83.5 x 86 cm. Museo Nacional
de Arqueología, Antropología
e Historia del Perú, Lima, S/C.
(Figures 1, 5, 132)
17. Cup with axe-bearing super-
natural being (Viñaque style);
Tomb M-U1242, San José de Moro;
ceramic and slip; 15 x 7.4 cm.
Pontificia Universidad Católica del
Perú, Lima, M-U1242-C09. (Figure
30)
18. Head vessel (Viñaque style);
Tomb M-U1242, San José de Moro;
ceramic and slip; 16.1 x 9.2 cm.
Pontificia Universidad Católica del
Perú, Lima, M-U1242-C06. (Figure
38)
19. Lyre cup with supernatural
head (Viñaque style); Tomb
M-U1242, San José de Moro;
ceramic and slip; 9.5 x 8.4 cm.
Pontificia Universidad Católica del
Perú, Lima, M-U1242-C08. (Figure
37)
20. Spoon; Tomb M-U1512, San
José de Moro; ceramic; 9.8 x 3.5
x 1.8 cm. Pontificia Universidad
Católica del Perú, Lima,
M-U1512-C16. (Figure 208)
21. Bowl with mythical creature
(Viñaque style); Wari Willka;
ceramic and slip; 8 x 12.6
cm. Museo de Arqueología y
Antropología, Universidad
Nacional Mayor de San Marcos,
Lima, 3001 0041. (Figure 108)
22. Vessel with ventral animal
(Viñaque style); Wari Willka;
ceramic and slip; about 14 x
6 cm. Museo de Arqueología
y Antropología, Universidad
Nacional Mayor de San Marcos,
Lima, 3001 3453. (Figure 109)
23. Bowl with humpback animal
(Atarco style); ceramic and slip;
8.9 x 16.5 cm (rim). Michael C.
Carlos Museum, Emory University,
Atlanta, 1991.2.222.
24. Cup on serpent pedestal
(Atarco style); ceramic and slip;
14.6 x 7.9 cm. Denver Art Museum
Collection, Gift of Olive Bigelow by
exchange, 1996.37. (Figure 117)
25. Cup with staff deity
(Pachacamac style); reportedly
Pachacamac; ceramic and slip;
12.1 x 8 x 7.8 cm. Ethnologisches
Museum, Berlin, VA 19174. (Figure
122)
26. Cup with staff deity
(Pachacamac style); reportedly
Pachacamac; ceramic and slip;
12.4 x 7.9 x 8.2 cm. Ethnologisches
Museum, Berlin, VA 19167. (Figure
60)
269 checklist of the E xhibition
27. Cup with staff deity head and
warrior (Atarco or Pachacamac
style); ceramic and slip; 11.7 x 9
cm. Museum zu Allerheiligen,
Schaffhausen, Eb15182. (Figure
243)
28. Double-chambered vessel
with human (Pachacamac style);
reportedly Pachacamac; ceramic
and slip; 14.9 x 9 x 20.7 cm.
Ethnologisches Museum, Berlin,
VA 49699. (Figure 123)
29. Double-chambered vessel with
human (Pachacamac style); Casa
Grande, Chicama Valley; ceramic
and slip; 13.7 x 15.4 cm. Museo
Larco, Lima, ML031840.
30. Double-chambered vessel
with human (Pachacamac style);
reportedly Pachacamac; ceramic [30]
and slip; 12.5 x 7.7 x 21 cm.
Ethnologisches Museum, Berlin,
VA 49703.
31. Drum (Atarco style); ceramic,
slip, cotton, and animal hide; 45
x 21 cm. Staatliches Museum für
Völkerkunde, Munich, 89-311 922.
(Figure 114)
32. Bound figure vessel
(Pachacamac style); reportedly
Pachacamac; ceramic and slip;
18.5 x 12 x 15.2 cm. Ethnologisches
Museum, Berlin, VA 49645.
33. Cup-holding figure in tie-dyed
tunic and four-cornered hat
(Atarco style); ceramic and slip;
19.3 x 19.5 cm. Museo Regional de
Ica “Adolfo Bermúdez Jenkins,”
MRI-00176-01. (Figure 59)
34. Female figure (Pachacamac
style); ceramic and slip; H. 28.4
cm. American Museum of Natural
History, New York, 41.2/8596.
(Figure 92)
35. Figure in a litter (Pachacamac
style); ceramic and slip; 26.3 x 21.6
x 24.8 cm. The Cleveland Museum
of Art, Leonard C. Hanna Jr. Fund,
1997.1. (Figure 20)
[32]
36. Figure in tapestry-woven tunic
and four-cornered hat (Viñaque
style); ceramic and slip; 28.7 x 23
cm. Fundación Museo Amano,
Lima, FMAC-000020. (Figure 146)
37. Figure in tie-dyed tunic and
four-cornered hat (Viñaque style);
reportedly Anja, Mantaro Valley,
Huancayo; ceramic and slip; 30 x
22 cm. Museum Rietberg, Zurich,
RPB 320. (Figure 110)
38. Two figural vessels (Viñaque
style); ceramic and slip; 19.2 x 10.5
x 10.3 cm. Dallas Museum of Art,
1976.W.216, 1976.W.217. (Figure
120)
39. Warrior vessel (Viñaque style);
reportedly Pachacamac; ceramic
and slip; 15 x 9.5 x 7 cm. Linden-
Museum, Stuttgart, 119016. (Figure
118)
40. Vessel with litter group
(Nievería style?); reportedly Wari
Willka; ceramic and slip; 28 x 16
x 14 cm. The Cleveland Museum
of Art, Leonard C. Hanna Jr. Fund,
2011.36. (Figure 101)
41. Faceneck vessel with birds
(Pachacamac style); report-
edly Pachacamac; ceramic
and slip; 18.4 x 14.5 x 14.3 cm.
Ethnologisches Museum, Berlin,
VA 49536. (Figure 67)
[63]
270 checklist of the E xhibition
42. Faceneck vessel with felines
(Atarco style); ceramic and slip;
17.8 x 12.7 x 12.7 cm. I. Michael
Kasser Collection, KP 246. (Figure
115)
43. Faceneck vessel with figure
(Atarco style); ceramic and slip;
17.8 x 11.4 x 11.4 cm. Brooklyn
Museum, New York, Henry L.
Batterman Fund, 41.418. (Figure
237)
44. Faceneck vessel with muti-
lated nose (Pachacamac style);
reportedly Pachacamac; ceramic
and slip; 18.2 x 13.5 x 14.8 cm.
Ethnologisches Museum, Berlin,
VA 49450. (Figure 61)
45. Faceneck vessel with mythical
creatures (Atarco style); Corral
Redondo, Churunga Valley;
ceramic and slip; 83.5 x 86 cm.
Museo Nacional de Arqueología,
Antropología e Historia del Perú,
Lima, C-64874. (Figure 198)
46. Faceneck vessel with tapestry-
woven tunic (Atarco style);
ceramic and slip; 15.6 x 10 cm.
Museo Nacional de Arqueología,
Antropología e Historia del Perú,
Lima, C-54760. (Figure 54)
47. Double-spouted fish vessels
(Pachacamac style); reportedly
Pachacamac; ceramic and slip;
11.4 x 5.9 x 17.2 cm, 11.7 x 6.2 x
17.4 cm. Ethnologisches Museum,
Berlin, VA 19128, VA 19129.
(Figure 124)
48. Double-spouted sea creature
vessel (Pachacamac style);
reportedly Pachacamac; ceramic
and slip; 17 x 7 x 17.5 cm.
Ethnologisches Museum, Berlin,
VA 19127. (Figure 126)
49. Double-spouted sea cucum-
ber vessel (Pachacamac style);
reportedly Pachacamac; ceramic
and slip; 9.6 x 5 x 18 cm.
Ethnologisches Museum, Berlin,
VA 19130. (Figure 125)
50. Double-spouted snail vessel
(Pachacamac style); reportedly
Pachacamac; ceramic and slip;
12.9 x 15.1 x 10 cm. Ethnologisches
Museum, Berlin, VA 19149. (Figure
127)
51. Double-spouted feline-head
vessel (Pachacamac style);
ceramic and slip; 18 x 16 x
13 cm. Niedersächsisches
Landesmuseum, Hannover,
I/10456. (Figure 64)
52. Double-spouted head vessel
(Pachacamac style); ceramic
and slip; 15.8 x 15.3 cm. Museo
Nacional de Arqueología,
Antropología e Historia del Perú,
Lima, C-54789. (Figure 111)
53. Double-spouted skull vessel
(Pachacamac style); ceramic and
slip; 14.5 x 17.8 x 12.6 cm. The
Cleveland Museum of Art, John L.
Severance Fund, 1996.292. (Figure
113)
54. Head vessel (Robles Moqo
style); ceramic and slip; 18
x 12 cm. Museo Nacional de
Arqueología, Antropología e
Historia del Perú, Lima, C-54786.
(Figure 116)
55. Supernatural head vessel
(Pachacamac or Atarco style);
ceramic and slip; 17.6 x 17.1 x 14.7
cm. Milwaukee Public Museum,
54569/20517. (Overleaf, p. 63)
56. Container with staff deity
head and profile winged creatures
(Pachacamac style); ceramic
and slip; 16.6 x 16.9 x 21.8 cm.
The Cleveland Museum of Art,
Norman O. Stone and Ella A. Stone
Memorial Fund, 1999.2. (Figure 14)
57. Double-spouted bird-headed
creature vessel (Wari Norteño
style); ceramic and slip; 16.5 x 20.6
x 11.6 cm. Museo Larco, Lima,
ML010864. (Figure 129)
58. Double-spouted vessel
with bird-headed creature
(“Pachacamac griffin”)
(Pachacamac style); ceramic and
slip; 16 x 16.5 x 14.5 cm. Fowler
Museum at UCLA, Los Angeles,
Gift of Mr. and Mrs. Herbert L.
Lucas Jr., X86-3702. (Figure 107)
59. Double-spouted vessel with fig-
ure and birds (Wari Norteño style);
ceramic and slip; 18.6 x 19.4 x 10.5
cm. Fowler Museum at UCLA,
Los Angeles, Gift of Dr. and Mrs.
Robert Kuhn, X71.417. (Figure 128)
60. Foot vessel (Robles Moqo style);
ceramic and slip; 11.7 x 12.1 cm.
Museo Larco, Lima, ML018890.
(Figure 63)
61. Vessel with bird-headed
creature (“Pachacamac griffin”)
(Pachacamac style); report-
edly Pachacamac; ceramic
and slip; 26.8 x 20.5 x 20.9 cm.
Ethnologisches Museum, Berlin,
VA 19059. (Figure 106)
62. Vessel with bird-headed
creature (“Pachacamac griffin”)
(Pachacamac style); Temple
of Pachacamac (Painted or
Polychrome Temple), Pachacamac;
ceramic and slip; 18.3 x 17.3 x 15
cm. University of Pennsylvania
Museum of Archaeology and
Anthropology, Philadelphia, Max
Uhle, William Pepper Peruvian
Expedition, Funded by Phoebe A.
Hearst, 26709. (Figure 66)
63. Vessel with eight-limbed head
(Pachacamac style); reportedly
Pachacamac; ceramic and slip; 20
x 15 x 13 cm. American Museum of
Natural History, New York, B/476.
64. Vessel with feline head (Atarco
style); reportedly Ingenio, Nasca
drainage; ceramic and slip; 20.3
x 11.4 x 6.4 cm. The Metropolitan
Museum of Art, New York,
Purchase, Arthur M. Bullowa
Bequest and Rogers Fund, 1996,
1996.290. (Figure 91)
65. Vessel with head of mythical
creature (Atarco style); ceramic
and slip; 24.1 x 16.5 cm. Denver
Art Museum Collection, Gift
of Olive Bigelow by exchange,
1996.36. (Figure 65)
[79]
66. Vessel with humpback animal
(Viñaque style); ceramic and slip;
26.7 x 21.6 x 11.4 cm. Brooklyn
Museum, New York, Henry L.
Batterman Fund, 41.420. (Figure
97)
67. Vessel with staff deity head
(Pachacamac style); ceramic and
slip; 21.4 x 16.6 x 11.2 cm. Fowler
Museum at UCLA, Los Angeles,
Gift of Mr. and Mrs. Herbert L.
Lucas Jr., X90.488. (Figure 16)
68. Vessel with warrior (Atarco
style); ceramic and slip; 31 x 39.4
cm. Museo Regional de Ica “Adolfo
Bermúdez Jenkins,” MRI-00178-01.
(Figure 119)
INLAY
69. Ear ornament; Temple
of Pachacamac (Painted or
Polychrome Temple), Pachacamac;
wood and shell; 3.8 x 6 x 2.4
cm. University of Pennsylvania
Museum of Archaeology and
Anthropology, Philadelphia, Max
Uhle, William Pepper Peruvian
Expedition, Funded by Phoebe A.
Hearst, 26720. (Figure 203)
70. Ear ornament frontal with
staff-bearing creature in profile;
reportedly Pachacamac; shell
and stone; 5.9 x 5.8 x 1 cm.
Ethnologisches Museum, Berlin,
VA 41596. (Figure 204)
71. Pair of ear ornament fron-
tals with skulls; reportedly
Pachacamac; shell and stone; 5.9
x 5.9 x 0.7 cm. Ethnologisches
Museum, Berlin, VA 41595a, b.
(Figure 201)
271 checklist of the E xhibition
[80]
72. Pendant figurine; wood, shell,
turquoise, and gold; 3.3 x 1.5 x 1.1
cm. Pre-Columbian Collection,
Dumbarton Oaks Research Library
and Collection, Washington, DC,
PC.B.437. (Figure 232)
73. Mirror with staff deity head;
wood, stone, and shell; 23.9 x 12 x
2 cm. Pre-Columbian Collection,
Dumbarton Oaks Research Library
and Collection, Washington, DC,
PC.B.432. Cleveland only. (Figure
205)
74. Ornament with figure; shell,
stone, and metal (silver?); 6.6 x 3.6
cm. The Cleveland Museum of Art,
In memory of Mr. and Mrs. Henry
Humphreys, gift of their daughter
Helen, 1944.291. (Figure 213)
75. Figure pendant; wood, shell,
stone, and silver; 10.2 x 6.4 x 2.6
cm. Kimbell Art Museum, Fort
Worth, AP 2002.04. (Figure 211)
76. Pendant with figure; Spondylus
shell, shell, stone, and metal; 13.3
x 11.4 x 5.1 cm. Fowler Museum at
UCLA, Los Angeles, Gift of Mr. and
Mrs. Herbert L. Lucas Jr., X88-255.
(Figure 143)
77. Trumpet with figure; report-
edly the Huacho-Pativilca region;
Strombus shell, shell, and stone; H.
19.1 cm. The Dayton Art Institute,
Museum Purchase, 1970.32.
(Figure 212)
78. Spear-thrower thumb rest with
bird; bone and stone; 8 x 4 cm.
Roemer- und Pelizaeus-Museum,
Hildesheim, V. 5522. (Figure 214)
[92]
79. Spear-thrower thumb rest
with feline; bone and stone; H.
7 cm. Staatliches Museum für
Völkerkunde, Munich, NM 337.
80. Spear-thrower thumb rest
with human and camelid; bone;
7.9 x 55.4 cm. Museo Nacional
de Arqueología, Antropología e
Historia del Perú, Lima, MO-2853.
METAL
81. Pair of ear ornaments; silver
and cotton; 9.5 (with shaft) x 8.6
cm. The Metropolitan Museum
of Art, New York, Bequest of
Jane Costello Goldberg, from the
Collection of Arnold I. Goldberg,
1986, 1987.394.580–81. (Figure 218)
82. Two pendant figurines;
silver; 4.1 x 2.1 x 2.1 cm (each).
Milwaukee Public Museum,
34596/9672 and 34597/9672.
(Figure 231)
83. Mask-like ornament; reportedly
Pachacamac; silver; 20.5 x 18 x 5
cm. American Museum of Natural
History, New York, B/9450. (Figure
222)
84. Three ornaments; silver;
55 x 44 cm, 17 x 16 cm, 17 x 17
cm. Linden-Museum, Stuttgart,
M31039. (Figure 219)
85. Bird plaque; gold; 13.7 x 14.6
cm. Private collection. (Figure 220)
86. Warrior plaque; silver; 25.7
x 19.7 x 2.5 cm. Museum of Fine
Arts, Houston, Gift of Alfred C.
Glassell Jr., 2001.117. (Figure 19)
87. Winged creature plaque;
reportedly Pachacamac; gold; 13 x
15.8 cm. Ethnologisches Museum,
Berlin, VA 28787. (Figure 221)
88. Plume; reportedly Pachacamac;
gold; 21.8 x 4.6 x 0.7 cm.
Ethnologisches Museum, Berlin,
VA 31795. (Figure 216)
89. Plume; reportedly Pachacamac;
gold; 27.5 x 6.9 x 0.7 cm.
Ethnologisches Museum, Berlin,
VA 31797. (Figure 217)
90. Plume; Huaca Pucllana; silver;
28.9 x 15.3 cm. Museo de Sitio
Huaca Pucllana, Lima, MSHP-97-
156 (ME). (Figure 215)
91. Three plumes with staff deity
head; Pomacanchi; silvered cop-
per; 34.8 x 10 cm, 40 x 13.8 cm,
36.6 x 12 cm. Princeton University
Art Museum, Gift of Leonard H.
Bernheim Jr., Class of 1959, 1982-
27, 1982-29, 1982-28. (Figure 12)
STONE
92. Spear-thrower thumb rest
with warrior; stone and pigment;
8.9 x 2.9 x 8.3 cm. Brooklyn
Museum, New York, Gift of the
Ernest Erickson Foundation, Inc.
86.224.30.
93. Figurine; greenstone; 4.7 x
2.5 x 2.2 cm. Denver Art Museum
Collection, Funds Provided by
Jan and Frederick R. Mayer,
1992.502.3. (Figure 227)
94. Figurine; greenstone; 3.5 x
1.9 x 1.9 cm. Denver Art Museum
Collection, Collection of Frederick
and Jan Mayer, 1994.45. (Figure
228)
95. Figurine; greenstone; 4 x 1.9
x 1.9 cm. Denver Art Museum
Collection, New World Department
Acquisition Fund, 1997.15. (Figure
229)
96. Figurine; greenstone; 3.2
x 1.9 cm. Denver Art Museum
Collection, Funds from the
Alianza de las Artes Americanas,
1995.39.1. (Figure 230)
97. Figurine offering; Pikillacta;
greenstone, Spondylus shells,
and copper or copper alloy; H. 2
to 5.2 cm (figurines). Colección
Juan Larrea, Museo de América,
Madrid, 8.825-64 (figurines), 7.038
(rod). (Figures 223, 224)
272 checklist of the E xhibition
TEXTILES
Tapestry-woven
98. Panel; camelid fiber and cot-
ton; 77 x 109.5 cm. Royal Ontario
Museum, Toronto, 931.11.1.
(Figures 161, 251)
99. Panel; camelid fiber and cot-
ton; 99.7 x 105.1 cm. Brooklyn
Museum, New York, Frank L.
Babbott Fund, 53.147. Fort Worth
only.
100. Miniature tunic with staff-
bearing creature in profile; cam-
elid fiber and cotton; 16 x 26 cm.
Private collection. (Figure 167)
101. Miniature tunic with
weapon?-bearing creature in pro-
file; camelid fiber and cotton; 22.1
x 31.8 cm. Brooklyn Museum, New
York, Gift of Mr. and Mrs. Alastair
B. Martin, the Guennol Collection,
71.180. Cleveland and Fort Worth
only. (Figure 166)
[99]
102. Tunic fragment with staff-
bearing creatures in profile; cam-
elid fiber and cotton; 54 x 15 cm.
Ethnologisches Museum, Berlin,
VA 66028. (Figure 165)
103. Tunic fragments with bird-
headed staff-bearing creature in
profile; camelid fiber and cot-
ton; 90 x 53 cm, 89.4 x 53.5 cm.
The Cleveland Museum of Art,
Purchase from the J. H. Wade
Fund, 2005.53a, b. (Frontispiece,
Figure 152)
104. Tunic with camelid- or deer-
headed staff-bearing creature in
profile; camelid fiber and cotton;
103.7 x 108.5 cm. Deutsches
Textilmuseum, Krefeld, 12299.
Cleveland and Fort Lauderdale
only. (Figure 153)
105. Tunic with camelid- or deer-
headed staff-bearing creature
in profile; reportedly Larcay,
Yucanas Province, Department of
Ayacucho; camelid fiber and cot-
ton; 100 x 92 cm. Museo Nacional
de Arqueología, Antropología e
Historia del Perú, Lima, RT-1650.
(Figure 174)
106. Tunic with feline-headed
staff-bearing creature in pro-
file; camelid fiber and cotton;
104.7 x 102.8 cm. The Textile
Museum, Washington, DC,
Museum Purchase, 1961.3.17. Fort
Lauderdale and Fort Worth only.
(Figure 155)
107. Tunic with staff-bearing crea-
ture in profile; camelid fiber and
cotton; 97 x 144.9 cm. Brooklyn
Museum, New York, Gift of the
Ernest Erickson Foundation, Inc.,
86.224.109. Cleveland and Fort
Lauderdale only. (Figure 164)
108. Tunic fragment with figures;
Ancón; camelid fiber and cotton;
118.5 x 103.5 cm. Ethnologisches
Museum, Berlin, VA 7468 (16).
(Figure 168)

109.* Weaver’s work basket and
contents; Ancón; bone, camelid
fiber, cotton, reeds, and wood;
20 x 26 x 18 cm. Ethnologisches
Museum, Berlin, VA 5816a–t.
(Figure 170)
110. “Sumptuous Mummy Pack,”
pl. 17, The Necropolis of Ancón in
Peru by Wilhelm Reiss and Alfons
Stübel, 1880–87; color lithograph;
50.7 x 36.2 cm. Ingalls Library, The
Cleveland Museum of Art. (Figure
169a)
111. “Sumptuous Mummy Pack,”
pl. 16, The Necropolis of Ancón in
Peru by Wilhelm Reiss and Alfons
Stübel, 1880–87; color lithograph;
50 x 37 cm. Ingalls Library, The
Cleveland Museum of Art. (Figure
169b)
273 checklist of the E xhibition
[122]
112. “Sumptuous Mummy Pack,”
pl. 10, The Necropolis of Ancón in
Peru by Wilhelm Reiss and Alfons
Stübel, 1880–87; color lithograph;
50 x 37.2 cm. Ingalls Library, The
Cleveland Museum of Art. (Figure
169c, detail illustrated).
113. Tunic with sacrificer; camelid
fiber and cotton; 103.4 x 110.8 cm.
The Textile Museum, Washington,
DC, Museum Purchase, 1966.5.2.
Cleveland and Fort Worth only.
(Figure 157)
114. Tunic with sacrificer; camelid
fiber and cotton; 106.7 x 112 cm.
The Cleveland Museum of Art,
John L. Severance Fund, 2007.179.
(Figure 150)
115. Tunic with sacrificer-related
creature; camelid fiber and cotton;
105.4 x 114 cm. Pre-Columbian
Collection, Dumbarton Oaks
Research Library and Collection,
Washington, DC, PC.B.496. Fort
Worth only. (Figure 15)
116. Tunic with sacrificer-related
creature; camelid fiber and cot-
ton; 100 x 112 cm. Museum der
Kulturen, Basel, collected by
Hans Theodor Cron (1921–1964),
IVc23577. Cleveland and Fort
Lauderdale only. (Figure 151)
117. Tunic with face-fret motif;
camelid fiber and cotton; 102.2
x 102.2 cm. Dallas Museum of
Art, The Eugene and Margaret
McDermott Art Fund, Inc., in
honor of Carol Robbins’ 40th anni-
versary with the Dallas Museum of
Art, 2004.55.McD. Cleveland and
Fort Lauderdale only. (Figure 144)
118. Tunic with face-fret motif;
camelid fiber and cotton; 110.5
x 118.1 cm. Museum of Art,
Rhode Island School of Design,
Providence, Mary B. Jackson Fund
and Edgar J. Lownes Fund, 40.007.
Fort Worth only. (Figure 10)
119. Tunic with face-fret motif;
camelid fiber and cotton; 108.6 x
109.7 cm. The Textile Museum,
Washington, DC, acquired by
George Hewitt Myers in 1941,
91.343. Cleveland and Fort
Lauderdale only. (Figure 172)
120. Tunic with face-fret motif;
camelid fiber and cotton; 110 x
110.5 cm. Ethnologisches Museum,
Berlin, VA 64374. Fort Lauderdale
and Fort Worth only. (Figure 55)
121. Tunic with paired fret motif;
camelid fiber and cotton; 98 x
106 cm. Staatliches Museum für
Völkerkunde, Munich, 57-20-245
(NM 245). Cleveland and Fort
Lauderdale only. (Figure 148)
122. Tunic with paired fret motif;
camelid fiber and cotton; 102
x 98 cm. The Textile Museum,
Washington, DC, acquired by
George Hewitt Myers in 1941,
91.342. Fort Worth only.
[131]
123. Tunic with face-fret and
interlocked U-shaped motifs;
camelid fiber and cotton; 106 x 94
cm. American Museum of Natural
History, New York, 41.2/8604. Fort
Worth only. (Figure 149)
124. Tunic with stepped-cross and
interlocked U-shaped motifs; cam-
elid fiber and cotton; 96.5 x 110.45
cm. Los Angeles County Museum
of Art, M70.3.1. Cleveland only.
(Figure 52)
125. Tunic; camelid fiber and
cotton; 55 x 53.7 cm. Private
Collection. (Figure 159)
126. Tunic; camelid fiber and
cotton; 100 x 106 cm. The Textile
Museum, Washington, DC,
Museum Exchange, 1962.5.1. Fort
Lauderdale and Fort Worth only.
(Figure 160)
[141]
274 checklist of the E xhibition
127. Tunic with skulls; camelid
fiber and cotton; 220 x 115 cm.
Museo de Arte de Lima Collection,
Prado Family Bequest, IV-2.1-1241.
Conserved with the support of the
Southern Peru Copper Corporation
2001. (Figure 162)
128. Tunic with heads, insects, and
heart-lung-trachea motif; camelid
fiber and cotton; 101.6 x 105.4 cm.
Private collection. (Figure 81)
129. Bag; camelid fiber and cot-
ton; 18.7 x 16.5 cm. The Textile
Museum, Washington, DC,
Museum Purchase, 1959.10.1.
Cleveland only. (Overleaf, p. 29)
130. Headband; reportedly
Coyungo, Rio Grande Valley; cam-
elid fiber and cotton; 67 x 12 cm.
The Textile Museum, Washington,
DC, Museum Purchase, 1965.32.1.
(Figure 145)
131. Headband; camelid fiber and
cotton; 77.5 (including ties) x 10.2
cm. Private collection.
132. Glove (Moche-Wari style);
camelid fiber and cotton; 28.6
x 22.1 cm. Brooklyn Museum,
New York, Charles Stewart Smith
Memorial Fund and Museum
Collection Fund, 58.204. Cleveland
and Fort Lauderdale only. (Figure
171)
Tie-dyed
133. Tunic; camelid fiber; 182
x 112.5 cm. The Metropolitan
Museum of Art, New York, Gift
of Arthur M. Bullowa, 1980,
1980.564.2. Cleveland and Fort
Lauderdale only. (Figure 17)
134. Tunic; camelid fiber; 86.5
x 122 cm. The Textile Museum,
Washington, DC, acquired by
George Hewitt Myers in 1941,
91.341. Cleveland and Fort Worth
only. (Figure 181)
135. Tunic; camelid fiber; 112.4 x
182.3 cm. The Textile Museum,
Washington, DC, acquired by
George Hewitt Myers in 1931,
91.90. (Figure 186)
136. Tunic; reportedly Chilca;
camelid fiber; 87 x 124.8 cm. The
Textile Museum, Washington, DC,
Gift of Leo Drimmer-Lichtemberg,
1965.40.43. Fort Worth only.
(Figure 191)
 [146]
[150]
Feathered
137. Four-cornered hat; feathers,
cotton, and reed; 17 x 14 x 14 cm.
Brooklyn Museum, New York,
A. Augustus Healy Fund, 41.228.
Cleveland and Fort Worth only.
(Figure 194)
138. Ornaments; feathers
and cotton; 8 x 6.5 cm (each).
Princeton University Art Museum,
Anonymous gift 1996, 1996-228.1,
1996-228.2, 1996-228.4, 1996-228.5.
(Figure 195)
139. Panel, probably a hang-
ing; probably Corral Redondo,
Churunga Valley; feathers, cotton,
and camelid fiber; 63.5 x 208.9
cm. The Metropolitan Museum
of Art, New York, The Michael C.
Rockefeller Memorial Collection,
Bequest of Nelson A. Rockefeller,
1979, 1979.206.471. Cleveland and
Fort Lauderdale only. (Figure 197)
140. Panel, probably a hang-
ing; probably Corral Redondo,
Churunga Valley; feathers, cotton,
and camelid fiber; 68.6 x 211.5 x
2.2 cm. The Metropolitan Museum
of Art, New York, The Michael C.
Rockefeller Memorial Collection,
Bequest of Nelson A. Rockefeller,
1979, 1979.206.904. Fort Worth
only. (Not illustrated)
275 checklist of the E xhibition
[149]
141. Panel, probably a hanging;
Corral Redondo, Churunga Valley;
feathers, cotton, and camelid fiber;
81.3 x 223.5 cm. The Cleveland
Museum of Art, Andrew R. and
Martha Holden Jennings Fund,
2002.93.
142.* Tabard (radiocarbon date,
cal. AD 780–985); feathers and
cotton; 143.5 x 132 cm. Private col-
lection. (Figure 196)
143. Plate 147, Ancient Peruvian
Art by Arthur Baessler, 1902–3;
color collotype; 50.8 x 37.4 cm.
Ingalls Library, The Cleveland
Museum of Art. (Not illustrated)
Other Textiles and Fiber Objects
144. Four-cornered hat with
geometric motifs; camelid fiber
and cotton; 12.4 x 17.5 cm. The
Metropolitan Museum of Art, New
York, Gift of Arthur M. Bullowa,
1983, 1983.497.6. Fort Worth only.
(Figure 147)
145. Four-cornered hat with
geometric motifs; camelid fiber
and cotton; 13 x 18 cm. The
Metropolitan Museum of Art, New
York, Gift of George D. Pratt, 1933,
33.149.101. Cleveland and Fort
Lauderdale only. (Figure 53)
146. Four-cornered hat with heads
of mythical creature; camelid
fiber; 11.4 x 16.5 x 15.3 cm. The
Cleveland Museum of Art, Gift of
John Wise, 1947.291. Cleveland and
Fort Worth only.
147. Four-cornered hat with mythi-
cal creature; camelid fiber; 14 x
12 cm; Staatliches Museum für
Völkerkunde, Munich, 57-20-223
(NM 223). Cleveland and Fort
Lauderdale only. (Figure 11)
148. Four-cornered hat with
winged creature; camelid fiber;
14.3 x 15 x 15.5 cm. The Cleveland
Museum of Art, Purchase from
the J. H. Wade Fund, 1945.378.
(Overleaf, p. 249)
149. Headband; camelid fiber
and cotton; 9 x 15 x 15 cm.
Ethnologisches Museum, Berlin,
VA 65529.
150. Headband; camelid fiber
and cotton; 49 x 8.5 cm. Peabody
Museum of Archaeology and
Ethnology, Harvard University,
Cambridge. 42-12-30/3519.
Cleveland and Fort Lauderdale
only.
 [151]
151.* Headband; camelid fiber
and cotton; 102.9 cm. Fine Arts
Museums of San Francisco, The
Caroline and H. McCoy Jones
Collection, Gift of Caroline McCoy-
Jones, 2000.17.5. Cleveland only.
152.* Headband; camelid fiber and
cotton; 114.1 x.7 cm. Museo de
América, Madrid, 2002-5-218. Fort
Lauderdale and Fort Worth only.
153.* Cap with braids (Middle
Horizon or later, Wari-related
culture?); cotton, camelid fiber,
and human hair; L. 60 cm (cap,
15 cm; braids, 45 cm). Museum of
Fine Arts, Boston, Mary Woodman
Fund, 31.497. Cleveland and Fort
Lauderdale only.
[153]
[155]
276 checklist of the E xhibition
[152]
154.* Cap with braids (Middle
Horizon or later, Wari-related
culture?); cotton, camelid fiber,
human hair, and bast fiber; 91 x
22.2 x 6.4 cm. Brooklyn Museum,
New York, Henry L. Batterman
Fund, 41.427. Fort Worth only.
155.* Khipu (radiocarbon date,
cal. AD 690–900); cotton and cam-
elid fiber; 35 x 17 cm. American
Museum of Natural History, New
York, 41.2/7679. Cleveland and
Fort Lauderdale only.
156.* Khipu (radiocarbon date,
AD 719–981); cotton; L. 190 cm
(primary cord), 36 cm (longest sec-
ondary cord). Private collection.
(Figure 180)
[154]
 [158]
157. Bag with human face; alpaca
or llama hide, human hair,
pigment, cotton, and coca leaf
contents; H. 26 cm (bag), L. 64.7
cm (strap). The Cleveland Museum
of Art, Leonard C. Hanna Jr. Fund,
2011.35. (Figure 18)
277 checklist of the E xhibition
158. Mantle; cotton and camelid
fiber; 156.2 x 155.6 cm. Los Angeles
County Museum of Art, M.78.54.7.
Cleveland and Fort Lauderdale
only.
159. Mantle; cotton and cam-
elid fiber; 177.8 x 177.8 cm. The
Metropolitan Museum of Art, New
York, The Michael C. Rockefeller
Memorial Collection, Bequest
of Nelson A. Rockefeller, 1979,
1979.206.462. Fort Worth only.
(Figure 249)
[160]
160.* Helmet; camelid fiber
and wool?; 24.2 x 25.5 x 20.5
cm. Museum zu Allerheiligen,
Schaffhausen, Eb15032.
161.* Pair of shin guards; cam-
elid fiber, fur, and wool?; 26 x 11
cm. Museum zu Allerheiligen,
Schaffhausen, Eb15984.
162.* Sling (Middle Horizon or
later); camelid fiber; 6.4 x 2.5 x
251.5 cm. Brooklyn Museum,
New York, Gift of Ernest Erickson,
70.177.62.
[162]
278 checklist of the E xhibition
[161]
WOOD Memorial Collection, Purchase,
Nelson A. Rockefeller Gift, 1968,
163. Animal container; wood,
1978.412.214. (Figure 13)
shell, and stone; H. 4.4 cm.
American Museum of Natural 166. Mirror with sacrificer; wood
History, New York, 41.2/8599. and pigment; 28.6 x 15.9 x 1.9 cm.
(Figure 236) Private collection. (Figure 206)
164. Sacrificer container (radio- 167.* Mummy bundle head; Huaca
carbon date, cal. AD 769–887); Pucllana; wood, shell, and copper
wood and cinnabar; 10.8 x 7 x or copper alloy; 19 x 16 cm. Museo
7.5 cm. The Cleveland Museum de Sitio Huaca Pucllana, Lima,
of Art, John L. Severance Fund, MSHP-09-466 (V).
2007.193.a–b. (Figure 233)
165. Cup with staff-bearing
creatures in profile; reportedly
Cahuachi or the Huacho-Pativilca
region; wood; 11.4 x 6.4 cm. The
Metropolitan Museum of Art, New
York, The Michael C. Rockefeller
[167]
Reference List
Alcock et al. 2001
Alcock, Susan E., Terence N.
D’Altroy, Kathleen D. Morrison,
and Carla N. Sinopoli, eds.
Empires: Perspectives for Archae­
ology and History. Cambridge:
Cambridge University Press, 2001.
Allen 2002
Allen, Catherine J. The Hold Life
Has: Coca and Cultural Identity
in an Andean Community. 2nd
ed. Washington, DC/London:
Smithsonian Institution Press,
2002.
Allen 1988
. The Hold Life Has: Coca and
Cultural Identity in an Andean
Community. Washington, DC/
London: Smithsonian Institution
Press, 1988.
Allen 1981
. “To Be Quechua: The
Symbolism of Coca Chewing
in Highland Peru.” American
Ethnologist 8, no. 1 (1981): 157–71.
Alva 2001
Alva, Walter. “The Royal Tombs
of Sipán: Art and Power in Moche
Society.” In Pillsbury 2001,
223–345.
Alva and Donnan 1993
Alva, Walter, and Christopher B.
Donnan. Royal Tombs of Sipán.
Los Angeles: Fowler Museum of
Cultural History, 1993.
Anders 1991
Anders, Martha B. “Structure and
Function at the Planned Site of
Azangaro: Cautionary Notes for the
Model of Huari as a Centralized
Secular State.” In W. Isbell and
McEwan 1991b, 165–98.
Anders 1990
. “Maymi: Un Sitio del
Horizonte Medio en el Valle de
Pisco.” Gaceta Arqueológica
Andina 5, no. 17 (1990): 27–39.
Anders 1986
. “Dual Organization and
Calendars Inferred from the
Planned Site of Azángaro—Wari
Administrative Strategies.” PhD
diss., Cornell University, 1986.
279 R eference list
Anders et al. 1998
Anders, Martha B., Susana Arce
Torres, Izumi Shimada, Victor
Chang, Luis Tokuda, and Sonia
Quiroz Calle. “Early Middle
Horizon Pottery Production at
Maymi, Pisco Valley, Peru.” In
Shimada 1998, 233–51.
Andrushko and Bellifemine 2006
Andrushko, Valerie, and V.
Bellifemine. “Descripción de res-
tos óseos humanos de los contextos
funerarios de Cotocotuyoq.” Report
submitted to Instituto Nacional de
Cultura, Cuzco, Peru, 2006.
Angeles Falcón and Pozzi-Escot
2001
Angeles Falcón, Rommel, and
Denise Pozzi-Escot. “Textiles del
Horizonte Medio. Las evidencias
de Huaca Malena, Valle de Asia.”
In Kaulicke and W. Isbell 2001,
401–24.
Angeles Falcón forthcoming
Angeles Falcón, Rommel. “Tejidos
Wari, vehículos de ideología y
poder durante el Horizonte Medio;
la iconografía en los tejidos de
Huaca Malena, valle de Asia.” In
W. Isbell et al. forthcoming.
Angrand 1866
Angrand, Léonce. Lettre sur les
Antiquités de Tiaguanaco et
l’origine Présumable de la Plus
Ancienne Civilization du Haute
Pérou. Paris: DeLa Claye, 1866.
Anton 1992
Anton, Ferdinand. Brücken in
die Zukunft: Textile Kunst vor
Kolumbus. Salzburg: Galerie
Sailer, 1992.
Arnold 1975
Arnold, Dean E. “Ceramic Ecology
of the Ayacucho Basin, Peru:
Implications for Prehistory.”
Current Anthropology 16, no. 2
(1975): 183–205.
Arnold and Hastorf 2008
Arnold, Denise, and Christine
Hastorf. Heads of State: Icons,
Power and Politics in the Ancient
and Modern Andes. Walnut Creek,
CA: Left Coast Press, 2008.
Arriaga [1621] 1968
Arriaga, Pablo José de. The
Extirpation of Idolatry in Peru,
trans. and ed. by L. Clark Keating.
Lexington: University of Kentucky
Press, 1968.
Arriola Tuni 2008
Arriola Tuni, Carlos A. “Excava­
ciones Arqueológicas en Piki­­
llacta.” Saqsaywaman (Cuzco), no.
7 (2008): 14–61.
Arriola Tuni and Tesar 2011
Arriola Tuni, Carlos A., and Louis
D. Tesar. “The Pikillacta 2004
Eastern Gate Offering Pit.” Ñawpa
Pacha 31, no. 1 (2011): 1–44.
Ascher and Ascher 1981
Ascher, Marcia, and Robert
Ascher. Code of the Quipu: A
Study in Media, Mathematics, and
Culture. Ann Arbor: University of
Michigan Press, 1981.
Auther 2004
Auther, Elissa. “The Decorative,
Abstraction, and Hierarchy of Art
and Craft in the Art Criticism of
Clement Greenberg.” Oxford Art
Journal 27, no. 3 (2004): 339–64.
Aveni 1996
Aveni, Anthony F. “Astronomy and
the Ceque System.” Journal of the
Steward Anthropological Society
24, no. 1 (1996): 157–72.
Azoy 1985
Azoy, Mary Livingston, ed.
Peruvian Antiquities: A Manual
for United States Customs.
Washington, DC: Organization of
American States, 1985.
Baessler 1902–3
Baessler, Arthur. Ancient
Peruvian Art: Contributions to the
Archaeology of the Empire of the
Incas, trans. by A. H. Keane. 4 vols.
Berlin: A. Asher; New York: Dodd,
Mead, 1902–3.
Bastien 1995
Bastien, Joseph W. “The Mountain/
Body Metaphor Expressed in a
Kaatan Funeral.” In Dillehay 1995,
355–78.
Bastien 1978
. Mountain of the Condor:
Metaphor and Ritual in an Andean
Ayullu. San Francisco: West
Publishing, 1978.
Bauer 1999
Bauer, Brian S. “The Early
Ceramics of the Inca Heartland.”
Fieldiana, no. 31 (1999): 63–66.
Bauer 1998
. The Sacred Landscape of
the Inca: The Cusco Ceque System.
Austin: University of Texas Press,
1998.
Baxandall 1989
Baxandall, Michael. Painting and
Experience in Fifteenth-Century
Italy. Oxford/New York: Oxford
University Press, 1989.
Belisle and Covey 2010
Belisle, Veronique, and R.
Alan Covey. “Local Settlement
Continuity and Wari Impact
in Middle Horizon Cusco.” In
Jennings 2010c, 79–95.
Benavides Calle 1999
Benavides Calle, Mario. “Tejidos
Wari/Wari Textiles.” In Lavalle
and Lavalle de Cárdenas 1999,
353–411.
Benavides Calle 1991
. “Cheqo Wasi, Huari.” In W.
Isbell and McEwan 1991b, 55–69.
Benavides Calle 1984
. Carácter del estado Wari.
Ayacucho: Universidad Nacional
de San Cristóbal de Huamanga,
1984.
Benavides Calle 1979
. “Notas sobre excavaciones
en Cheqo Wasi.” Investigaciones
(UNSCH), no. 2 (1979): 9–26.
Benavides Calle 1978
. Yacimientos arqueológi-
cos en Ayacucho. Ayacucho:
Universidad Nacional de San
Cristóbal de Huamanga, 1978.
Benavides Calle 1965
. “Estudio de la cerámica
decorada de Qonchopata.” BA the-
sis, Universidad Nacional de San
Cristóbal de Huamanga, 1965.
Benavides Calle 1964
. “Análisis de la Cerámica
Huarpa.” Revista del Museo
Nacional (Lima), no. 37 (1964):
63–88.
Bender 1978
Bender, Barbara. “Gatherer-Hunter
to Farmer: A Social Perspective.”
World Archaeology 10, no. 2 (1978):
204–22.
Benitez 2009
Benitez, Leonardo. “Descendants
of the Sun: Calendars, Myths, and
the Tiwanaku State.” In Young-
Sánchez 2009, 49–82.
Bennett 1953
Bennett, Wendell C. Excavations
at Wari, Ayacucho, Peru. Yale
University Publications in
Anthropology 49. New Haven: Yale
University Press, 1953.
Bennett 1946
. “The Archaeology of the
Central Andes.” In J. H. Steward,
ed., Handbook of South American
Indians. Bureau of American
Ethnology 143. Washington, DC:
Bureau of American Ethnology,
1946.
Bennett 1934
. Excavations at Tiahuanaco.
Anthropological Papers of the
American Museum of Natural
History 34, pt. 3. New York:
American Museum of Natural
History, 1934.
Benson 1970
Benson, Elizabeth P., ed.
Dumbarton Oaks Conference
on Chavín. Washington, DC:
Dumbarton Oaks, 1970.
Benson and Cook 2001
Benson, Elizabeth P., and Anita
G. Cook, eds. Ritual Sacrifice in
Ancient Peru. Austin: University of
Texas Press, 2001.
Berenguer Rodríguez 2000
Berenguer Rodríguez, José.
Tiwanaku: Lords of the Sacred
Lake. Santiago: Museo Chileno de
Arte Precolombino, 2000.
Bergh 2009
Bergh, Susan E. “The Bird and
the Camelid (or Deer): A Ranked
Pair of Wari Tapestry Tunics.” In
Young-Sánchez 2009, 225–46.
Bergh 2004
. “Two Wari Ceramics.” In
Young-Sánchez 2004d, 154–57.
Bergh 1999
. “Pattern and Paradigm in
Middle Horizon Tapestry Tunics.”
PhD diss., Columbia University,
1999.
28 0 R eference list
Bergh 1993
. “Death and Renewal in
Moche Phallic-Spouted Vessels.”
RES: Journal of Anthropology and
Aesthetics, no. 24 (1993): 78–94.
Bergh forthcoming
. “The Material Sublime:
Sleeved Tunics of the Andean
Middle Horizon.” In W. Isbell et al.
forthcoming.
Bergh and Cook 2003
Bergh, Susan E., and Anita G.
Cook. “Symbols of Sovereignty:
Possible Wari Antecedents of
Inka Tukapu.” Paper presented at
the symposium “Inka, Pre-Inka,
and Colonial Tukapu,” Harvard
University. Boston, 2003.
Bernedo Málaga 1950
Bernedo Málaga, Leonidas. “El
descubrimiento de noventa y seis
mantos del arte plumario de los
antiguos peruanos en una tumba
precolombina de la Hacienda
Hispana en el distrito de Andaray,
provincia de Condesuyos, el
12 de enero de 1943.” El Deber
(Arequipa), 18 February 1950.
Betanzos [1576] 1996
Betanzos, Juan de. Narrative of
the Incas, trans. and ed. by Roland
Hamilton and Dana Buchanan.
Austin: University of Texas Press,
1996.
Betanzos [1551] 1987
. Suma y Narración de los
Incas, ed. by María del Carmen
Martín Rubio. Madrid: Ediciones
Atlas, 1987.
Biermann 2006
Biermann, Daniela. “Ofrendas
textiles de Cabezas Achatadas—un
cementerio de Nasca temprano
cerca de Huacapuy/Camaná.”
In V. Solanilla Demestre, ed.,
Actas III Jornadas Internacionales
sobre Textiles Precolombinos,
227–40. Barcelona: Grup d’Estudis
Precolombins, Departament d’Art
de la Universitat Autònoma de
Barcelona, 2006.
Billman 2010
Billman, Brian R. “How Moche
Rulers Came to Power.” In Quilter
and Castillo Butters 2010, 181–200.
Billman 2002
. “Irrigation and the Origins
of the Southern Moche State on
the North Coast of Peru.” Latin
American Antiquity 13, no. 4
(2002): 371–401.
Bird 1958
Bird, Junius B. Art of Ancient Peru.
Exh. checklist. New York: Museum
of Primitive Art, 1958.
Bird and Skinner 1974
Bird, Junius B., and Milica D.
Skinner. “The Technical Features
of a Middle Horizon Tapestry
Shirt from Peru.” Textile Museum
Journal 4, no. 1 (1974): 5–13.
Blower 2000
Blower, David. “The Many Facets
of Mullu: More Than Just a
Spondylus Shell.” Andean Past,
no. 6 (2000): 209–28.
Boone 1996
Boone, Elizabeth H., ed. Andean
Art at Dumbarton Oaks.
Washington, DC: Dumbarton Oaks,
1996.
Boone and Urton 2011
Boone, Elizabeth H., and Gary
Urton, eds. Their Way of Writing:
Scripts, Signs, and Pictographies
in Pre-Columbian America.
Washington, DC: Dumbarton Oaks,
2011.
Bourdieu 1990
Bourdieu, Pierre. The Logic of
Practice. Stanford: Stanford
University Press, 1990.
Bragayrac Dávila 1991
Bragayrac Dávila, Enrique.
“Archaeological Investigations in
the Vegachayoc Moqo Sector of
Huari.” In W. Isbell and McEwan
1991b, 71–80.
Braun 1993
Braun, Barbara. Pre-Columbian
Art and the Post-Columbian World.
New York: Abrams, 1993.
Bray 2003a
Bray, Tamara L., ed. The
Archaeology and Politics of Food
and Feasting in Early States
and Empires. New York: Kluwer
Academic/Plenum Publishers,
2003.
Bray 2003b
Bray, Tamara L. “The Commensal
Politics of Early States and
Empires.” In Bray 2003a, 1–13.
Bray and Cook 1997
Bray, Tamara L., and Anita G.
Cook. “The Art of Empire in the
Andes: A Comparison of Inca and
Wari Ceramic Iconography.” Paper
presented at 62nd annual meet-
ing of the Society for American
Archaeology. Nashville, 1997.
Brewster-Wray 1989
Brewster-Wray, Christine C.
“Moraduchayuq: An Administra­
tive Compound at the Site of Huari,
Peru.” PhD diss., State University
of New York, Binghamton, 1989.
Brewster-Wray 1983
. “Spatial Patterning and the
Function of a Huari Architectural
Compound.” In Sandweiss 1983,
122–35.
Brinckerhoff 1999
Brinckerhoff, Deborah. Weaving for
the Gods: Textiles of the Ancient
Andes. Exh. cat. Greenwich,
CT: Bruce Museum of Arts and
Science, 1999.
Brody 1977
Brody, J. J. Mimbres Painted
Pottery. Santa Fe: School of
American Research, 1977.
Browne et al. 1993
Browne, David, Helaine Silverman,
and Rubén García. “A Cache of 48
Nasca Trophy Heads from Cerro
Carapo, Peru.” Latin American
Antiquity 43, no. 3 (1993): 274–94.
Brugnoli and Hoces de la Guardia
Chellew 1989
Brugnoli, Paulina, and Soledad
Hoces de la Guardia Chellew.
Arte mayor de los Andes.
Santiago: Museo Chileno de Arte
Precolombino, 1989.
Bruhns and Kelker 2010
Bruhns, Karen O., and Nancy L.
Kelker. Faking the Ancient Andes.
Walnut Creek, CA: Left Coast
Press, 2010.
Burger 1992
Burger, Richard. Chavín and the
Origins of Andean Civilization.
New York: Thames and Hudson,
1992.
Burger 1988
. “Unity and Heterogeneity
within the Chavín Horizon.” In
Keatinge 1988, 99–144.
Calancha [1638] 1974–82
Calancha, Antonio de la. Corónica
moralizada del orden de San
Augustín en el Perú, 3:930–33.
Lima: Universidad Nacional Mayor
de San Marcos, 1974–82.
Carmichael 1994
Carmichael, Patrick. “The Life
from Death Continuum in Nasca
Imagery.” Andean Past, no. 4
(1994): 81–90.
Cases and Agüero 2004
Cases, Bárbara, and Carolina
Agüero. “Textiles teñidos por
amarras del Norte Grande de
Chile.” Estudios Atacameños, no.
27 (2004): 117–38.
Castillo Butters 2010
Castillo Butters, Luis Jaime.
“Moche Politics in the Jequete­
peque Valley, A Case for Political
Opportunism.” In Quilter and
Castillo Butters 2010, 81–107.
Castillo Butters 2009a
. “El Estilo Mochica Tardío
de Línea Fina de San José de
Moro.” In Castillo Butters and
Pardo 2009, 208–43.
Castillo Butters 2009b
. “La Arqueología del Valle
de Jequetepeque y la Colección
Rodríguez Razetto.” In Castillo
Butters and Pardo 2009, 34–61.
Castillo Butters 2005
. “Las Sacerdotisas de San
José de Moro, Rituales funerarios
de mujeres de élite en la costa
norte del Perú.” In Divina y
humana, La mujer en los antiguos
Perú y Mexico, 18–29. Ministerio
de Educación, Lima.
Castillo Butters 2001a
. “La presencia Wari en San
José de Moro.” In Kaulicke and W.
Isbell 2001, 143–79.
Castillo Butters 2001b
. “The Last of the Mochicas:
A View from the Jequetepeque
Valley.” In J. Pillsbury, ed.,
Moche Art and Archaeology in
Ancient Peru, 307–32. Studies in
the History of Art 63. Center for
Advanced Studies in the Visual
Arts, Symposium Papers XL.
Washington, DC: National Gallery
of Art, 2001.
Castillo Butters and Pardo 2009
Castillo Butters, Luis Jaime, and
Cecilia Pardo, eds., De Cupisnique
a los Incas: El Arte del Valle
de Jequetepeque, La Donación
de Pertus Fernandini al MALI,
112–53. Lima: Museo de Arte de
Lima, 2009.
Castillo Butters and Uceda Castillo
2008
Castillo Butters, Luis Jaime, and
Santiago Uceda Castillo. “The
Mochicas.” In Silverman and W.
Isbell 2008, 707–29.
281 R eference list
Castillo Butters et al. 2008
Castillo Butters, Luis Jaime, Julio
Rucabado Yong, Martín del Carpio
Perla, Katusha Berny Quiroga,
Karim Ruiz Rosell, Carlos Rengifo
Chunga, Gabriel Prieto Burmester,
and Carole Fraresso. “Ideología y
poder en la consolidación, colapso,
y reconstitucíon del estado
Mochicha del Jequetepeque: El
Proyecto Arqueológico San José de
Morro (1991–2006).” Ñawpa Pacha,
no. 29 (2008): 1–86.
Castillo Butters et al. forthcoming
Castillo Butters, Luis Jaime,
Francesa Fernandini Parodi, and
Luis Armando Muro Ynoñán.
“Imperios Hegemónicos y Estados
Oportunistas: Relaciones e interac-
ciones entre los grupos de poder
Wari y las élites Mochica.” In J.
Jennings and L. J. Castillo Butters,
eds., Interregional Interaction and
Social Change in Peru, 550–1000
AD. Lima: Fondo Editorial de la
Pontificia Universidad Católica del
Perú, forthcoming.
Chapdelaine 2010
Chapdelaine, Claude. “Moche and
Wari during the Middle Horizon
on the North Coast of Peru.” In
Jennings 2010c, 213–32.
Chapdelaine 2001
. “The Growing Power of a
Moche Urban Class.” In Pillsbury
2001, 69–88.
Chávez 2004
Chávez, Sergio. “The Yaya-
Mama Religious Tradition as an
Antecedent of Tiwanaku.” In
Young-Sánchez 2004d, 70–93.
Chávez 2002
. “Identification of the
Camelid Woman and Feline Man
Themes, Motifs and Designs in
Pucara Style Pottery.” In Silverman
and W. Isbell 2002, 35–69.
Chávez 1984–85
. “Funerary Offerings from
a Middle Horizon Context in
Pomacanchi, Cuzco.” Ñawpa
Pacha, no. 22–23 (1984–85): 1–48.
Cieza de León [1553] 1984
Cieza de León, Pedro. Crónica
del Perú, primera parte. Lima:
Pontificia Universidad Católica del
Perú, 1984.
Cieza de León [1553] 1959
. The Incas of Pedro de Cieza
de León, ed. by Victor W. Von
Hagen, trans. by Harriet de Onis.
Norman: University of Oklahoma
Press, 1959.
Clarke 2001
Clarke, Michael J. “Akha
Feasting: An Ethnoarchaeological
Perspective.” In Dietler and
Hayden 2001b, 144–67.
Classen 1993
Classen, Constance. Inca
Cosmology and the Human Body.
Salt Lake City: University of Utah
Press, 1993.
Claver 1985
Claver, F. “DinawitOgil: High Datu
of Namnam.” In A. de Guzman
and E. M. Pacheco, eds., Bukidnon
Politics and Religion, 51–114.
Institute of Philippine Culture,
Papers no. 11. Quezon City: Ateneo
de Manila Press, 1985.
Clifford 1988
Clifford, James. The Predicament
of Culture: Twentieth-century
Ethnography, Literature, and
Art. Cambridge, MA: Harvard
University Press, 1988.
Cobo [1653] 1990
Cobo, Bernabé. Inca Religion and
Customs, trans. and ed. by Roland
Hamilton. Austin: University of
Texas Press, 1990.
Cobo [1653] 1979
. History of the Inca Empire:
An Account of the Indians’
Customs and Their Origins,
Together with a Treatise on Inca
Legends, History, and Social
Institutions, trans. and ed.
by Roland Hamilton. Austin:
University of Texas Press, 1979.
Cobo [1653] 1890–95
. Historia del Nuevo Mundo.
Seville: Sociedad de Bibliofilos
Andaluces, 1890–95.
Coleman-Goldstein 2010
Coleman-Goldstein, Robin.
“Negotiating Power in the Wari
Empire: A Comparative Study of
Local-Imperial Interactions in the
Moquegua and Majes Regions dur-
ing the Middle Horizon (550–1050
CE).” PhD diss., Northwestern
University, 2010.
Conkey 1980
Conkey, Margaret. “The Identifica­
tion of Prehistoric Hunter/Gatherer
Aggregation Sites: The Case of
Altimira.” Current Anthropology
21 (1980): 609–30.
Conklin 2011
Conklin, William J. “Antes del
quipu inca: La evolución de los
sistemas informativos basados en
cuerdas.” In C. A. Hoffmann and
G. Urton, Atando Cabos, 77–94.
Lima: Ministerio de Cultura/
Museo Nacional de Arqueología,
Antropología e Historia del Perú,
2011.
Conklin 2004a
. “The Fire Textile.” Hali, no.
133 (2004): 94–102.
Conklin 2004b
. “Shaped Tapestry Sash.” In
Young-Sánchez 2004d, 86–87.
Conklin 2004c
. “Wari Tunic.” In Young-
Sánchez 2004d, 178–79.
Conklin 1996
. “Huari Tunics.” In Boone
1996, 2:375–98.
Conklin 1991
. “Tiahuanaco and Huari:
Architectural Comparisons and
Interpretations.” In W. Isbell and
McEwan 1991b, 281–91.
Conklin 1986
. “The Mythic Geometry of
the Ancient Southern Sierra.” In
Rowe 1986a, 123–37.
Conklin 1983
. “Pucara and Tiahuanaco
Tapestry: Time and Style in a
Sierra Weaving Tradition.” Ñawpa
Pacha, no. 21 (1983): 1–44.
Conklin 1982
. “The Information
System of Middle Horizon
Quipus.” In A. F. Aveni and G.
Urton, eds., Ethnoastronomy
and Archaeoastronomy in the
American Tropics, 261–81. Annals
of the New York Academy of
Sciences 385. New York, 1982.
Conklin 1978
. “The Revolutionary
Weaving Inventions of the Early
Horizon.” Ñawpa Pacha, no. 16
(1978): 1–13.
Conklin 1970
. “Peruvian Textile Fragment
from the Beginning of the Middle
Horizon.” Textile Museum Journal
3, no. 1 (1970): 15–24.
Cook 2004
Cook, Anita G. “Wari Art and
Society.” In Silverman 2004,
146–66.
Cook 2001a
. “Huari D-shaped
Structures, Sacrificial Offerings
and Divine Kingship.” In Benson
and Cook 2001, 137–63.
Cook 2001b
. “Las deidades huari y
sus orígenes altiplánicos.” In
Makowski Hanula 2000–2001,
2:39–65.
Cook 1996
. “The Emperor’s New
Clothes: Symbols of Royalty,
Hierarchy and Identity.”
Structure, Knowledge and Repre­
sen­tation in the Andes: Studies
Presented to Reiner Tom Zuidema
on the Occasion of His 70th
Birthday: Journal of the Steward
Anthropological Society, no. 24
(1996): 85–120.
Cook 1994
. Wari y Tiwanaku: Entre el
estilo y la imagen. Lima: Pontificia
Universidad Católica del Perú,
1994.
Cook 1992
. “The Stone Ancestors:
Idioms of Imperial Attire and Rank
among Huari Figurines.” Latin
American Antiquity 3, no. 4 (1992):
341–64.
Cook 1985
. “The Politico-Religious
Implications of the Huari Offering
Tradition.” Diálogo Andino (Arica,
Chile), no. 4 (1985): 203–22.
Cook 1984–85
. “The Middle Horizon
Ceramic Offerings from
Conchopata.” Ñawpa Pacha, no.
22–23 (1984–85): 49–90.
Cook 1983
. “Aspects of State Ideology in
Huari and Tiwanaku Iconography:
The Central Deity and Sacrificer.”
In Sandweiss 1983, 161–85.
Cook 1979
. “The Iconography of
Empire: Symbolic Communication
in VII Century Peru.” MA thesis,
State University of New York at
Binghamton, 1979.
Cook and Benco 2001
Cook, Anita G., and Nancy L.
Benco. “Vasijas para la fiesta y la
fama: producción artesanal en un
centro urbano Huari.” In Kaulicke
and W. Isbell 2001, 489–504.
282 R eference list
Cook and Cook 1998
Cook, Alexandra Parma, and Noble
David Cook. “Introduction.” In
Cook and Cook, trans. and eds.,
The Discovery and Conquest of
Peru by Pedro de Cieza de León,
5–35. Durham/London: Duke
University Press, 1998.
Cook and Glowacki 2003
Cook, Anita G., and Mary
Glowacki. “Pots, Politics,
and Power: Huari Ceramic
Assemblages and Imperial
Administration.” In Bray 2003a,
173–202.
Cook and Lechtman 1996
Cook, Anita G., and Heather
Lechtman. “Mosaic Figurine.” In
Boone 1996, 169–71.
Cordy-Collins 1990
Cordy-Collins, Alana. “Fonga
Sigde, Shell Purveyor to the Chimu
Kings.” In M. E. Moseley and A.
Cordy-Collins, eds., The Northern
Dynasties: Kingship and Statecraft
in Chimor, 393–419. Washington,
DC: Dumbarton Oaks, 1990.
Crawford 1946
Crawford, M. D. C. 5000 Years of
Fibers and Fabrics. Exh. cat. New
York: Brookyn Museum, 1946.
Cummins 2002
Cummins, Thomas B. F. Toasts
with the Inca: Andean Abstraction
and Colonial Images on Quero
Vessels. Ann Arbor: University of
Michigan Press, 2002.
Curatola Petrocchi 2008
Curatola Petrocchi, Mario. “La fun-
ción de los oráculos en el Imperio
inca.” In Curatola Petrocchi and
M. S. Ziołkowski, eds., Adivina­
ción y oráculos en el mundo
andino antiguo, 15–69. Lima:
Pontificia Universidad Católica
del Perú y Instituto Francés de
Estudios Peruanos, 2008.
Cuzco 2011
Espíritu Pampa, Una Nueva
Historia: Los Señores Vilca: Tras
las Huellas de los Wari en el
Antisuyu. Exh. brochure. Cuzco:
Ministerio de Cultura, Dirección
Regional de Cultura, 2011.
D’Altroy 2002
D’Altroy, Terence N. The Incas.
Oxford: Blackwell, 2002.
D’Altroy and Earle 1985
D’Altroy, Terence N., and Timothy
K. Earle. “Staple Finance, Wealth
Finance, and Storage in the Inka
Political Economy.” Current
Anthropology 26, no. 2 (1985):
187–206.
Davidson 1982
Davidson, Judith. “Ecology, Art,
and Myth: A Natural Approach to
Symbolism.” In A. Cordy-Collins,
ed., Pre-Columbian Art History:
Selected Readings, 331–43. Palo
Alto, CA: Peek Publications, 1982.
Dean 2010
Dean, Carolyn. A Culture of
Stone: Inka Perspectives on Rock.
Durham, NC/London: Duke
University Press, 2010.
DeFrance forthcoming
DeFrance, Susan D. “The Luxury
of Variety: Animals and Social
Distinction at the Wari Site of
Cerro Baúl, Southern Peru.” In S.
McCarty and B. Arbuckle, eds.,
Animals and Inequality. Boulder:
University Press of Colorado,
forthcoming.
DeHavenon 2009
DeHavenon, Georgia. “The Power
of an Icon: References to the
Gateway of the Sun from the 19th
Century to the Present.” In Young-
Sánchez 2009, 35–48.
DeLeonardis 2000
DeLeonardis, Lisa. “The Body
Context: Interpreting Nasca
Decapitated Burials.” Latin
American Antiquity 11, no. 4
(2000): 363–86.
Delgado Sumar 1980
Delgado Sumar, Hugo E. Ideología
Andina. Ayacucho: Universidad
Nacional San Cristóbal de
Huamanga, 1980.
Demarest 1981
Demarest, Arthur A. Viracocha:
The Nature and Antiquity of
the Andean High God. Peabody
Monograph 6. Cambridge, MA:
Harvard University, 1981.
D’Harnoncourt 1954
D’Harnoncourt, René. “Introduc­
tion.” In W. C. Bennett, Ancient
Arts of the Andes, 9–15. New York:
Museum of Modern Art, 1954.
Dietler 2001
Dietler, Michael. “Theorizing the
Feast: Rituals of Consumption,
Commensal Politics and Power in
African Contexts.” In Dietler and
Hayden 2001b, 65–114.
Dietler 1996
. “Feast and Commensal
Politics in the Political Economy:
Food, Power, and Status in
Prehistoric Europe.” In P. W.
Wiessner and W. Schiefenhövel,
eds., Food and the Status Quest,
87–125. Oxford: Berghahn Books,
1996.
Dietler and Hayden 2001a
Dietler, Michael, and Brian
Hayden. “Digesting the Feast—
Good to Eat, Good to Drink, Good
to Think: An Introduction.” In
Dietler and Hayden 2001b, 1–20.
Washington, DC: Smithsonian
Institution Press, 2001.
Dietler and Hayden 2001b
, eds. Feasts: Archaeological
and Ethnographic Perspectives
on Food, Politics, and Power.
Washington, DC: Smithsonian
Institution Press, 2001.
Dietler and Herbich 2001
Dietler, Michael, and Ingrid
Herbich. “Feasts and Labor
Mobilization: Dissecting a
Fundamental Economic Practice.”
In Dietler and Hayden 2001b,
240–64.
Dillehay 1995
Dillehay, Tom D., ed. Tombs for
the Living: Andean Mortuary
Practices. Washington, DC:
Dumbarton Oaks, 1995.
Dillon 1990
Dillon, Richard G. Ranking
and Resistance: A Precolonial
Cameroonian Polity in Regional
Perspective. Stanford: Stanford
University Press, 1990.
Disselhoff 1967
Disselhoff, Hans Dietrich.
Geschichte der Altamerikanischen
Kulturen. Munich: R. Oldenbourg,
1967.
Doering 1936
Doering, Heinrich. Old Peruvian
Art. New York: E. Weyhe, 1936.
Donahue 2003
Donahue, John F. “Toward a
Typology of Roman Public
Feasting.” American Journal of
Philology, no. 124 (2003): 423–41.
Donnan 2009
Donnan, Christopher B. “El Estilo
Mochica Temprano del Valle de
Jequetepeque.” In Castillo Butters
and Pardo 2009, 112–53.
Donnan 2007
. Moche Tombs at Dos
Cabezas. Monograph 59. Los
Angeles: Cotsen Institute of
Archaeology Press, 2007.
Donnan 2006
. “A Moche Cemetery at
Masanca, Jequetepeque Valley,
Peru.” Ñawpa Pacha, no. 28 (2006):
151–94.
Donnan 1992
. Ceramics of Ancient Peru.
Exh. cat. Los Angeles: Fowler
Museum of Cultural History,
University of California, 1992.
Donnan 1976
. Moche Art of Peru:
Pre-Columbian Symbolic
Communication. Exh. cat. Los
Angeles: Museum of Cultural
History, University of California,
1976.
Donnan and Castillo Butters 1992
Donnan, Christopher B., and Luis
Jaime Castillo Butters. “Finding
the Tomb of a Moche Priestess.”
Archaeology 6, no. 45 (1992):
38–42.
D’Orbigny 1839
D’Orbigny, Alcide. Voyage dans
l’Amérique Méridionale. Paris:
Pitois, Levrault, 1839.
Douglas 1996
Douglas, Mary. Natural Symbols:
Explorations in Cosmology. New
York: Routledge, 1996.
Dulanto 2001
Dulanto, Jahl. “Dioses de
Pachacamac: El Ídolo y El Templo.”
In Makowski Hanula 2000–2001,
2:159–81.
Duviols 1979
Duviols, Pierre. “La dinastía
de los Incas: monarquía o diar-
quía?” Journal de la Société des
Américanistes LVXI (1979): 76–84.
Duviols 1978
. “Un Simbolisme andin
du double: La lithomorphose de
l’ancêtre.” Actes du XLII Congres
International des Américanistes,
4:359–64. Paris: International
Congress of Americanists, 1978.
Earls and Silverblatt 1978
Earls, John, and Irene Silverblatt.
“La realidad física social en la
cosmología andina.” Actes du
XLII Congres International des
Américanistes, Proceedings of
the 42nd International Congress
of Americanists, 4:299–325.
283 R eference list
Paris: International Congress of
Americanists, 1978.
Edwards 2010
Edwards, Matthew J.
“Archaeological Investigations at
Pataraya: A Wari Outpost in the
Nasca Valley of Southern Peru.”
PhD diss., University of California,
Santa Barbara, 2010.
Eisleb and Strelow 1985
Eisleb, Dieter, and Renate Strelow.
“ ‘Nachtrag zu den Katalogen’
Altperuanische Kulturen I–III.”
Baessler-Archiv, n.s., 33 (1985).
Eisleb and Strelow 1980
. Altperanische Kulturen III:
Tiahuanaco. Berlin: Museum für
Völkerkunde, 1980.
Eisleb and Strelow 1966
. “Altperuanische
unechte Partialgewebe mit
Plangimusterung.” Baessler-
Archiv, n.s., 13 (1966): 293–308
[volume date 1965].
Eling 1987
Eling, Herbert H., Jr. “The Role of
Irrigation Networks in Emerging
Societal Complexity during Late
Prehispanic Times, Jequetepeque
Valley, North Coast, Peru.” PhD
diss., University of Texas, 1987.
Emery 1996
Emery, Irene. The Primary Struc­
tures of Fabrics: An Illustrated
Classification. Washington, DC:
Textile Museum, 1966.
Esteban-Johansson 2002
Esteban-Johansson, Gloria.
“Textiles del Valle de Acarí en las
colecciones del Museo Etnográfico
de Gotemburgo, Suecia.” In V.
Solanilla Demestre, ed., Actas
III Jornadas Internacionales
sobre Textiles Precolombinos,
217–32. Barcelona: Grup d’Estudis
Precolombins, Departament d’Art
de la Universitat Autònoma de
Barcelona, 2002.
Evans and Pillsbury 2004
Evans, Susan Toby, and Joanne
Pillsbury, eds. Palaces of the
Ancient New World. Washington,
DC: Dumbarton Oaks, 2004.
Feltham 1989
Feltham, Jane. Peruvian
Textiles. Aylesbury, Eng.: Shire
Publications, 1989.
Fewkes 1914
Fewkes, J. Walter. Archaeology
of the Lower Mimbres Valley,
New Mexico. Washington, DC:
Smithsonian Institution, 1914.
Finucane et al. 2006
Finucane, Brian, Patricia Maita
Agurto, and William H. Isbell.
“Human and Animal Diet at
Conchopata, Peru: Stable Isotope
Evidence for Maize Agriculture
and Animal Management Practices
during the Middle Horizon.”
Journal of Archaeological Science
33, no. 12 (2006): 1766–776.
Firth 1983
Firth, Raymond. “Magnitudes and
Values in Kula Exchange.” In J. W.
Leach and E. R. Leach, eds., The
Kula: New Perspectives on Massim
Exchange, 89–102. Cambridge:
Cambridge University Press, 1983.
Flannery and Marcus 1996
Flannery, Kent V., and Joyce
Marcus. “Cognitive Archaeology.”
In R. W. Preucel and I. Hodder,
eds., Contemporary Archaeology in
Theory: A Reader, 350–63. Oxford:
Blackwell, 1996.
Flores Espinoza 1959
Flores Espinoza, Isabel. “El Sitio
Arqueológico de Wari Willca,
Huancay.” Actas y Trabajos del II
Congreso Nacional de Historia del
Perú, 1:177–86. Lima: Centro de
Estudios Histórico-Militares del
Perú, 1959.
Fonseca et al. 2011
Fonseca, Javier, Julinho Zapata,
Patricia Knobloch, and William
H. Isbell. “El Señor Wari de
Vilcabamba: Description and
Implications.” Paper presented
at 30th Annual Meeting of
the Northeastern Conference
on Andean Archaeology and
Ethnohistory. Andover, MA, 2011.
Frame 2007
Frame, Mary. “Elemental Pathways
in Fiber Structures: Approaching
Andean Symmetry Patterns
through an Ancient Technology.”
In C. Bier and A. S. Perlman,
eds., Textiles Narratives and
Conversations, 406–14. Textile
Society of America 10th Biennial
Symposium. Earleville, MD:
Textile Society of America, 2007.
Frame 2005
. “Pathways to the Past:
Strategies for Investigating Ancient
Andean Textiles.” In Recuperando
el pasado . . . conservación de
textiles arqueológicos y etnográfi-
cos, 1–15. North American Textile
Conservation Conference papers.
Mexico City, 2005.
Frame 2001
. “Beyond the Image: The
Dimensions of Pattern in Ancient
Andean Textiles.” In C. Paternosto,
ed., Abstraction: The Amerindian
Paradigm, 113–36. Brussels:
Société des Expositions du Palais
des Beaux-Arts de Bruxelles, 2001.
Frame 1999
. “Nasca-Huari y otros tex-
tiles de la costa sur/Nasca-Huari
and other South Coast textiles.” In
Lavalle and Lavalle de Cárdenas
1999, 311–51.
Frame 1991
. “Structure, Image, and
Abstraction: Paracas Necrópolis
Headbands as System Templates.”
In A. Paul, ed., Paracas Art and
Architecture, 110–71. Iowa City:
University of Iowa Press, 1991.
Frame 1990
. Andean Four-Cornered
Hats: Ancient Volumes. Exh. cat.
New York: Metropolitan Museum
of Art, 1990.
Frame 1986
. “The Visual Images of
Fabric Structure in Ancient
Peruvian Art.” In A. Rowe 1986a,
47–80.
Frame forthcoming
. “From Structure to
Cognition: The ‘Logic of Models’
in the Pattern Systems of Middle
Horizon Tapestry Tunics.” In W.
Isbell et al. forthcoming.
Franco Jordán 2004
Franco Jordán, Regulo. “Poder
Religioso, Crisis y Prosperidad en
Pachacamac: Del Horizonte Medio
al Intermedio Tardío.” Bulletin
de l’Institut Français d’Études
Andines 33, no. 3 (2004): 465–506.
Franquemont and Franquemont
1987
Franquemont, Edward M., and
Christine R. Franquemont.
“Learning to Weave in Chinchero.”
Textile Museum Journal, no. 26
(1987): 55–78.
Franquemont et al. 1992
Franquemont, Edward, Christine
Franquemont, and Billie Jean
Isbell. “Awaq ñawin: El ojo del
tejedor: la practica de la cultura en
el tejido.” Revista Andina 10, no. 1
(1992): 47–80.
García 2010
García, Rubén. “Puerto Nuevo y los
orígenes de la tradición estilístico-
religioso Paracas,” 187–201.
Boletín de Arqueología PUCP
13: Lima: Fondo Editorial de la
Pontificia Universidad Católica del
Perú, 2010 [volume date 2009].
Gasparini and Margolies 1980
Gasparini, Graziano, and Luise
Margolies. Inca Architecture,
trans. by Patricia J. Lyon.
Bloomington/London: Indiana
University Press, 1980.
Gayton 1961
Gayton, Anna H. “The Cultural
Significance of Peruvian Textiles:
Production, Function, Aesthetics.”
Kroeber Anthropological Society
Papers, no. 25 (1961): 111–28.
Gayton 1927
. “The Uhle Collections from
Nievería.” University of California
Publications in American
Archaeology and Ethnology 21, no.
8 (1927): 305–29.
Giuntini 2006
Giuntini, Christine. “Precolum­
bian and Ethnographic Feather­
work from the Andes and Amazon
in the Metropolitan Museum of
Art.” Paper presented at “Feather
Creations, Materials, Production
and Circulation.” Hispanic
Society, Institute of Fine Arts, New
York, 2004. Published in 2006 as
http://nuevomundo.revues.org/
index1457.html.
Glowacki 2005a
. “Food of the Gods or
Mere Mortals? Hallucinogenic
Spondylus and Its Interpretive
Implications for Early Andean
Society.” Antiquity 79, no. 304
(2005): 257–68.
Glowacki 2005b
. “Dating Pikillacta.” In
McEwan 2005c, 115–24.
Glowacki 2005c
. “Pottery from Pikillacta.” In
McEwan 2005c, 101–14.
Glowacki 2002
. “The Huaro Archaeological
Site Complex: Rethinking the
Huari Occupation of Cuzco.” In W.
Isbell and Silverman 2002, 267–84.
Glowacki 2000
. “The Wari Ceramic Offering
Tradition: New Data from the
Huaro Valley, Cuzco.” Paper
presented at 64th annual meet-
28 4 R eference list
ing of the Society for American
Archaeology. Chicago, 2000.
Glowacki 1996
. “The Wari Occupation of
the Southern Highlands of Peru: A
Ceramic Perspective from the Site
of Pikillacta.” PhD diss., Brandeis
University, 1996.
Glowacki and Malpass 2003
Glowacki, Mary, and Michael
Malpass. “Water, Huacas, and
Ancestor Worship: Traces of a
Sacred Wari Landscape.” Latin
American Antiquity 14, no. 4
(2003): 431–48.
Glowacki and McEwan 2002
Glowacki, Mary, and Gordon F.
McEwan. “Pikillacta, Huaro, y
la gran región del Cuzco: Nuevas
interpretaciones de la ocupación
Wari de las tierras altas del sur.”
In Kaulicke and W. Isbell 2002,
31–49.
Godelier 1977a
Godelier, Maurice. “The Non-
Correspondence between Form
and Content in Social Relations.”
In Godelier 1977b, 186–95.
Godelier 1977b
. Perspectives in Marxist
Anthropology, trans. by Robert
Brain. Cambridge: Cambridge
University Press, 1977.
Goldstein 1989
Goldstein, Paul. “Omo, A
Tiwanaku Provincial Center
in Moquegua, Peru.” PhD diss.,
University of Chicago, 1978.
González Carré and Bragayrac
Dávila 1996
González Carré, Enrique, and
Enrique Bragayrac Dávila. “El área
ceremonial en la ciudad de Wari:
un hipótesis.” In El Templo Mayor
en la ciudad de Wari: estudios
arqueológicos en Vegachayoq
Moqo–Ayacucho, 7–27. Ayacucho:
Universidad Nacional San
Cristóbal de Huamanga, 1996.
González Carré and Bragayrac
Dávila 1986
. “El Templo Mayor de Wari,
Ayacucho.” Boletín de Lima 8, no.
47 (1986): 9–20.
González Carré and Soto Maguino
2004
González Carré, Enrique, and
Jorge Soto Maguino. “Una
ofrenda Wari.” In Cuaderno de
Investigación, Serie: Arqueología,
no. 2, 2da Epoca. Lima: Museo
Nacional de Arqueología,
Antropología e Historia del Perú,
2004.
Gose 2008
Gose, Peter. Invaders and
Ancestors: On the Intercultural
Making and Unmaking of Spanish
Colonialism in the Andes. Toronto:
University of Toronto Press, 2008.
Gose 1996
. “Oracles, Divine Kingship
and Political Representation in the
Inka State.” Ethnohistory 43, no. 1
(1996): 1–32.
Gose 1994
. Deathly Waters and Hungry
Mountains: Agrarian Ritual and
Class Formation in an Andean
Town. Toronto: University of
Toronto Press, 1994.
Gose 1993
. “Segmentary State
Formation and the Ritual Control
of Water under the Incas.” Society
for Comparative Study of Society
and History, no. 35 (1993): 480–514.
Green and Goldstein 2010
Green, Ulrike Matthies, and Paul
S. Goldstein. “The Nature of Wari
Presence in the Mid-Moquegua
Valley: Investigating Contact at
Cerro Trapiche.” In Jennings 2010c,
19–36.
Green and Whitehead 2006
Green, Kelsey, and William
Whitehead. “Domesticated Plant
Resources at Conchopata, a Wari
Site in the Ayacucho Valley,
Peru.” Poster presented at 71st
annual meeting of the Society for
American Archaeology. San Juan,
2006.
Greene 1991
Greene, Virginia. “An Ancient
Andean Feather Headdress.” In
R. E. Reina and K. M. Kensinger,
eds., The Gift of Birds: Featherwork
of Native South American Peoples,
16–25. University Museum of
Archaeology and Anthropology
75. Philadelphia: University of
Pennsylvania, 1991.
Grieder et al. 1988
Grieder, Terence, Alberto Bueno, C.
Earle Smith Jr., and Robert Malina.
La Galgada, Peru: A Preceramic
Culture in Transition. Austin:
University of Texas Press, 1988.
Grignon 2001
Grignon, Claude. “Commensality
and Social Morphology: An Essay
of Typology.” In P. Scholliers, ed.,
Food, Drink, and Identity: Cooking,
Eating, and Drinking in Europe
since the Middle Ages, 23–33.
Oxford: Oxford University Press,
2001.
Grossman 1983
Grossman, Joel. “Demographic
Change and Economic
Transformation in the South-
Central Highlands of Pre-Huari
Peru.” Ñawpa Pacha, no. 21 (1983):
45–126.
Guaman Poma [1615] 2009
Guaman Poma de Ayala, Felipe.
The First New Chronicle and Good
Government: On the History of the
World and the Incas up to 1615,
trans. and ed. by Roland Hamilton.
Austin: University of Texas Press,
2009.
Guss 1989
Guss, David M. To Weave and Sing:
Art, Symbol, and Narrative in
the South American Rain Forest.
Berkeley: University of California
Press, 1989.
Gutmann 1926
Gutmann, Bruno. Das Recht der
Dschagga. Munich: Beck, 1926.
Haberland 1965
Haberland, Wolfgang. “Gewebe mit
unechtem Plangi von der zentral-
peruanischen Küste.” Baessler-
Archiv, n.s., 12 (1965): 271–79
[volume date 1964].
Haeberli 2008
Haeberli, Joerg. “When and
Where Did the Nasca Proliferous
Style Emerge?” In W. Isbell and
Silverman 2006, 401–34.
Haeberli 2002
. “Tiempo y tradición en
Arequipa, Perú, y el surgimiento
de la cronología del Tema de la
Deidad Central.” In Kaulicke and
W. Isbell 2002, 89–138.
Haeberli forthcoming
. “Front Face Deity Motifs
and Themes.” In W. Isbell et al.
forthcoming.
Harth-Terre 1959
Harth-Terre, Emilio. “Pikillacta—
ciudad de positos y bastimientos
del imperio incaico.” Revista de
Museo e Instituto Arqueológico
(Cuzco), no. 1: 3–19.
Hastorf 1993
Hastorf, Christine A. Agriculture
and the Onset of Political
Inequality before the Inka. New
York: Cambridge University Press,
1993.
Hastorf 1991
. “Gender, Space, and
Food in Prehistory.” In J. M.
Gero and M. M. Conkey, eds.,
Engendering Archaeology: Women
and Prehistory, 132–62. Oxford:
Blackwell, 1991.
Hayden 2001
Hayden, Brian. “Fabulous Feasts:
A Prolegomenon to the Importance
of Feasting.” In Dietler and Hayden
2001b, 23–64.
Hayden 1995
. “A New Overview of
Domestication.” In T. Douglas
Price and Anne B. Gebauer, eds.,
Last Hunters, First Farmers, 273–
99. Santa Fe: School of American
Research, 1995.
Hiltunen 1999
Hiltunen, Juha J. Ancient
Kings of Peru: The Reliability
of the Chronicle of Fernando
de Montesinos. Bibliotheca
Historica 45. Helsinki: Suomen
Historiallinen Seura, 1999.
Hiltunen and McEwan 2004
Hiltunen, Juha J., and Gordon F.
McEwan. “Knowing the Inca Past.”
In Silverman 2004, 237–54.
Hoces de la Guardia Chellew et al.
2006
Hoces de la Guardia Chellew,
Soledad, Paulina Brugnoli, and
Carole Sinclaire. Awakhuni:
Tejiendo la historia andina.
Santiago: Museo Chileno de Arte
Precolombino, 2006.
Holmberg 1957
Holmberg, Alan R. “Lizard Hunts
on the North Coast of Peru.”
Fieldiana Anthropology 36, no. 9
(1957): 203–20.
Hugh-Jones 1978
Hugh-Jones, Christine. From the
Milk River: Spatial and Temporal
Processes in Northwest Amazonia.
Cambridge: Cambridge University
Press, 1978.
Hyslop 1993
Hyslop, John. “Factors Influencing
the Transmission and Distribution
of Inka Cultural Materials
throughout Tawantinsuyu.” In
D. Rice, ed., Latin American
Horizons, 337–56. Washington, DC:
Dumbarton Oaks, 1993.
Hyslop 1990
. Inka Settlement Planning.
Austin: University of Texas Press,
1990.
285 R eference list
Hyslop 1984
. The Inka Road System. New
York: Academic Press, 1984.
B. Isbell 1997
Isbell, Billie Jean. “De inmaduro
a duro: lo simbólico femenino y
los esquemas andinos de género.”
In D. Y. Arnold, ed., Más allá del
silencio: las fronteras de género
en los Andes, 253–300. La Paz:
Biblioteca de Estudios Andinos,
1997.
B. Isbell 1978
. To Defend Ourselves:
Ecology and Ritual in an Andean
Village. 3rd ed. Prospect Heights,
IL: Waveland Press, 1978.
W. Isbell 2011
Isbell, William H. “La Arqueología
Wari y la Dispersión del Quechua.”
In P. Kaulicke, R. Cerron-Palomino,
P. Haggerty, and D. Beresford-Jones,
eds., Lenguas y Sociedades en el
Antiguo Perú: Hacia un Enfoque
Interdisciplinario, 199–220. Boletín
de Arqueología PUCP 14. Lima:
Fondo Editorial de la Pontificia
Universidad Católica del Perú,
2011 [volume date 2010].
W. Isbell 2010
. “Agency, Identity and
Control: Understanding Wari
Space and Power.” In Jennings
2010c, 233–54.
W. Isbell 2009
. “Huari: A New Direction in
Central Andean Urban Evolution.”
In L. R. Manzanilla and C.
Chapdelaine, eds., Domestic Life
in Prehispanic Capitals: A Study
of Specialization, Hierarchy, and
Ethnicity, 197–219. Memoirs of the
Museum of Anthropology 46. Ann
Arbor: University of Michigan,
2009.
W. Isbell 2008
. “Wari and Tiwanaku:
International Identities in the
Central Andean Middle Horizon.”
In Silverman and W. Isbell 2008,
731–59.
W. Isbell 2007
. “A Community of Potters,
or Multicrafting Wives of
Polygynous Lords?” In I. Shimada,
ed., Craft Production in Complex
Societies: Multicraft and Producer
Perspectives, 68–96. Salt Lake City:
University of Utah Press, 2007.
W. Isbell 2006
. “Landscape of Power: A
Network of Palaces in Middle
Horizon Peru.” In J. J. Christie and
P. J. Sarro, eds., Palaces and Power
in the Americas: From Peru to the
Northwest Coast, 44–98. Austin:
University of Texas Press, 2006.
W. Isbell 2004a
. “Mortuary Preferences: A
Huari Case Study from Middle
Horizon Peru.” Latin American
Antiquity 15, no. 1 (2004): 3–32.
W. Isbell 2004b
. “Palaces and Politics in the
Andean Middle Horizon.” In Evans
and Pillsbury 2004, 191–246.
W. Isbell 2002
. “Reflexiones finales.” In
Kaulicke and W. Isbell 2002,
455–79.
W. Isbell 2001
. “Repensando el Horizonte
Medio: El caso de Conchapata,
Ayacucho, Perú.” In Kaulicke and
W. Isbell 2001, 9–68.
W. Isbell 1997
. Mummies and Mortuary
Monuments: A Postprocessual
Prehistory of Central Andean
Social Organization. Austin:
University of Texas Press, 1997.
W. Isbell 1991
. “Huari Administration
and the Orthogonal Cellular
Architecture Horizon.” In W. Isbell
and McEwan 1991b, 293–315.
W. Isbell 1988
. “City and State in Middle
Horizon Huari.” In Keatinge 1988,
164–89.
W. Isbell 1985
. “El origen del estado en
el valle de Ayacucho.” Revista
Andina 3, no. 1 (1985): 57–106.
W. Isbell 1984–85
. “Conchopata, Ideological
Innovator in Middle Horizon 1A.”
Ñawpa Pacha, no. 22–23 (1984–85):
91–126.
W. Isbell 1983
. “Shared Ideology and
Parallel Political Development:
Huari and Tiwanaku.” In
Sandweiss 1983, 186–208.
W. Isbell 1978
. “Cosmological Order
Expressed in Prehistoric Cere­
monial Centers.” In Proceedings
of the 42nd International Congress
of Americanists, 4:269–97.
Paris: International Congress of
Americanists, 1978.
W. Isbell 1977a
. “El Imperio Huari: Estado
o Ciudad?” Revista del Museo
Nacional (Lima), no. 43 (1977):
227–41.
W. Isbell 1977b
. The Rural Foundations of
Urbanism: Economic and Stylistic
Interaction between Rural and
Urban Communities in Eighth-
Century Peru. Illinois Studies
in Anthropology 10. Urbana:
University of Illinois Press, 1977.
W. Isbell and Cook 2002
Isbell, William H., and Anita
G. Cook. “A New Perspective
on Conchopata and the Andean
Middle Horizon.” In Silverman
and W. Isbell 2002, 249–306.
W. Isbell and Cook 1987
. “Ideological Origins of
an Andean Conquest State.”
Archaeology 40, no. 4 (1987):
26–33.
W. Isbell and Groleau 2010
Isbell, William H., and Amy
Groleau. “The Wari Brewer
Woman: Feasting, Gender, Offer­
ings, and Memory.” In Klarich
2010, 191–220.
W. Isbell and Knobloch 2009
Isbell, William H., and Patricia
J. Knobloch. “SAIS—The Origin,
Development, and Dating of
Tiahuanaco-Huari Iconography.”
In Young-Sánchez 2009, 165–210.
W. Isbell and Knobloch 2006
. “Missing Links, Imaginary
Links: Staff God Imagery in the
South Andean Past.” In W. Isbell
and Silverman 2006, 307–51.
W. Isbell and Korpisaari 2011
Isbell, William H., and Antti
Korpisaari. “Bodies of Evidence:
Mortuary Archaeology and the
Wari-Tiwanaku Paradox.” In
P. Eeckhout and L. S. Owens,
eds., Return of the Living Dead.
Funerary Practices and Models in
the Ancient Andes. Cambridge:
Cambridge University Press, 2011.
W. Isbell and McEwan 1991a
Isbell, William H., and Gordon
F. McEwan. “A History of Huari
Studies and Introduction to
Current Interpretations.” In W.
Isbell and McEwan 1991b, 1–17.
W. Isbell and McEwan 1991b
, eds. Huari Administrative
Structure: Prehistoric Monumental
Architecture and State
Government. Washington, DC:
Dumbarton Oaks, 1991.
W. Isbell and Schreiber 1978
Isbell, William H., and Katharina
J. Schreiber. “Was Huari a State?”
American Antiquity 43, no. 3
(1978): 372–89.
W. Isbell and Silverman 2006
Isbell, William H., and Helaine
Silverman, eds. Andean
Archaeology III: North and South.
New York: Springer, 2006.
W. Isbell and Silverman 2002
. Andean Archaeology
I: Variations in Sociopolitical
Organization. New York: Kluwer
Academic/Plenum, 2002.
W. Isbell and Vranich 2004
Isbell, William H., and Alexei
Vranich. “Experiencing the
Cities of Wari and Tiwanaku.”
In H. Silverman, ed., Andean
Archaeology, 146–66. Oxford:
Blackwell Publishing, 2004.
W. Isbell et al. 1991
Isbell, William H., Christine
Brewster-Wray, and Lynda
Spickard. “Architecture and
Spatial Organization at Huari.”
In W. Isbell and McEwan 1991b,
19–53.
W. Isbell et al. forthcoming
Isbell, William H., Mauricio Uribe
Rodríguez, and Anne Tiballi, eds.
Images in Action: Studies of the
Southern Andean Iconographic
Series. Los Angeles: Cotsen
Institute of Archaeology Press,
forthcoming.
Janusek 2008
Janusek, John Wayne. Ancient
Tiwanaku. Cambridge: Cambridge
University Press, 2008.
Jennings 2011
Jennings, Justin. Globalization
and the Ancient World. New York:
Cambridge University Press, 2011.
Jennings 2010a
. “Becoming Wari: Globaliza­
tion and the Role of the Wari
State in the Cotahuasi Valley of
Southern Peru.” In Jennings 2010c,
37–56.
Jennings 2010b
. “Beyond Wari Walls.” In
Jennings 2010c, 1–8.
28 6 R eference list
Jennings 2010c
, ed. Beyond Wari Walls:
Regional Perspectives on Middle
Horizon Peru. Albuquerque:
University of New Mexico Press,
2010.
Jennings 2006
. “Understanding Middle
Horizon Peru: Hermeneutic
Spirals, Interpretative Traditions,
and Wari Administrative Centers.”
Latin American Antiquity 17, no. 3
(2006): 265–85.
Jennings and Craig 2001
Jennings, Justin, and Nathan
Craig. “Politywide Analysis and
Imperial Political Economy:
The Relationship between
Valley Political Complexity
and Administrative Centers in
the Wari Empire of the Central
Andes.” Journal of Anthropological
Archaeology, no. 20 (2001):
479–502.
Jennings and Yépez Alvarez 2009
Jennings, Justin, and Willy Yépez
Alvarez. “The Inca Conquest and
Consolidation of the Cotahuasi
Valley of Southern Peru.” Ñawpa
Pacha, no. 29 (2009): 119–52.
Jiménez Díaz 2009
Jiménez Díaz, María Jesús.
Tradición de tradiciones: Tejidos
prehispánicos y virreinales de los
Andes: La colección del Museo
de América. Madrid: Museo de
América, 2009.
Johnston and Kaufman 1967
Johnston, Meda Parker, and Glen
Kaufman. Design on Fabrics. New
York: Reinhold, 1967.
Julien 2000
Julien, Katharine. Reading Inca
History. Iowa City: University of
Iowa Press, 2000.
Junker and Niziolek 2010
Junker, Laura Lee, and Lisa
Niziolek. “Food Preparation and
Feasting in the Household and
Political Economy of Pre-Hispanic
Philippine Chiefdoms.” In Klarich
2010, 17–53.
Kajitani 1982
Kajitani, Nobuko. “Andesu No
Senshoku [Textiles of the Andes].”
Senshoku no bi [Textile Art]
(Kyoto), no. 20, Autumn (1982):
10–96.
Kaulicke 2001
Kaulicke, Peter. “La Sombra de
Pachacamac: Huari en la costa
central.” In Kaulicke and W. Isbell
2001, 313–58.
Kaulicke 1997
, ed. La Muerte en el Antiguo
Perú. Boletín de Arqueología PUCP
1. Lima: Fondo Editorial de la
Pontificia Universidad Católica del
Perú, 1997.
Kaulicke and W. Isbell 2002
Kaulicke, Peter, and William H.
Isbell, eds. Huari y Tiwanaku:
Modelos vs. Evidencias. Segunda
parte. Boletín de Arqueología
PUCP 5. Lima: Fondo Editorial de
la Pontificia Universidad Católica
del Perú, 2002 [volume date 2001].
Kaulicke and W. Isbell 2001
. Huari y Tiwanaku: Modelos
y Evidencias. Primera parte.
Boletín de Arqueología PUCP
4. Lima: Fondo Editorial de la
Pontificia Universidad Católica del
Perú, 2001 [volume date 2000].
Keatinge 1988
R. W. Keatinge, ed. Peruvian
Prehistory. Cambridge: Cambridge
University Press, 1988.
Keleman 1943
Keleman, Pál. Medieval American
Art. 2 vols. New York: Macmillan,
1943.
Kendall 1985
Kendall, Ann. Aspects of Inca
Architecture: Description,
Function, and Chronology (parts 1
and 2). BAR International Series
242. Oxford: B.A.R., 1985.
Kendall 1976
. “Descripción e Inventario
de las Formas Arquitectónicas
Inca.” Revista del Museo Nacional
(Lima) 42 (1976): 13–96.
Ketteman 2002
Ketteman, W. Gregory. “New
Dates from the Huari Empire:
Chronometric Dating of the Pre­
historic Occupation of Concho­
pata, Ayacucho, Peru.” MA thesis,
State University of New York at
Binghamton, 2002.
Kidwell 1976
Kidwell, Claudia. “Bicentennial
Outlook: Riches, Rags, and
In-between.” Historic Preservation
28, no. 3 (1976): 28–33.
King 2000
King, Heidi. Rain of the Moon:
Silver in Ancient Peru. New York:
Metropolitan Museum of Art;
New Haven: Yale University Press,
2000.
King 1965
King, Mary Elizabeth. Ancient
Peruvian Textiles from the
Collection of the Textile Museum
of Washington, D.C. Exh. cat. New
York: Museum of Primitive Art,
1965.
King et al. 2012
King, Heidi, et al. Peruvian
Featherworks: Art of the
Precolumbian Era. New York:
Metropolitan Museum of Art,
2012.
Klarich 2010
Klarich, Elizabeth A., ed. Inside
Ancient Kitchens: New Directions
in the Study of Daily Meals and
Feasts. Boulder: University Press
of Colorado, 2010.
Knobloch 2011
Knobloch, Patricia J. “Who Was
Who in the Middle Horizon?”
2002, http://www.rohan.sdsu.
edu/~bharley/WWWAgents.html,
accessed August 2011.
Knobloch 2010
. “La imagen de los Señores
de Huari y la recuperación de
una identidad antigua.” In K.
Makowski Hanula, ed., Señores
de los Imperios del Sol, 197–209.
Lima: Colección Arte y Tesoros
del Perú, Banco de Crédito,
2010. English translation: http://
www-rohan.sdsu.edu/~bharley/
SenoresWari.pdf.
Knobloch 2005
. “Monkey Saw, Monkey Did:
A Stylization Model for Correlating
Nasca and Wari Chronology.”
Andean Past, no. 7 (2005): 111–34.
Knobloch 2003
. “An Early Intermediate
Period Deposit of Huarpa Style
Ceramics from the Site of Huari,
Department of Ayacucho, Perú.”
2003, http://www-rohan.sdsu.
edu/~bharley/HuarpaNPms4Web.
html (currently under revision for
Ñawpa Pacha).
Knobloch 2002
. “Who Was Who in the
Middle Horizon Andean Pre­
history.” 2002, http://www-rohan.
sdsu.edu/~bharley/WWWHome.
html.
Knobloch 2000
. “Wari Ritual Power at
Conchopata: An Interpretation
of Anadenanthera Colubrina
Iconography.” Latin American
Antiquity 11, no. 4 (2000): 387–402.
Knobloch 1993
. “Who Was Who in the Huari
Empire.” Paper presented at 33rd
annual meeting of the Institute of
Andean Studies. Berkeley, 1993.
See also http://www-rohan.sdsu.
edu/~bharley/WWWIAS93Paper.
html.
Knobloch 1991
. “Stylistic Date of Ceramics
from the Huari Centers.” In W.
Isbell and McEwan 1991b, 247–58.
Knobloch 1989
. “Artisans of the Realm.”
In S. Masuda and I. Shimada,
eds., Art of the Andean World,
107–23. Tokyo: Iwanami Shoten,
1989. English translation: http://
www.rohan.sdsu.edu/~bharley/
ArtisansoftheRealm.html.
Knobloch 1983
. “The Study of Andean
Huari Ceramics from the Early
Intermediate Period to the Middle
Horizon Epoch 1.” PhD diss.,
State University of New York at
Binghamton, 1983.
Knobloch 1976
. “A Study of the Huarpa
Ceramic Style of the Andean Early
Intermediate Period.” MA thesis,
State University of New York at
Binghamton, 1976.
Krauss 1979
Krauss, Rosalind. “Grids.” October,
no. 9 (1979): 50–64.
Kroeber 1944 and Interaction.” In W. Isbell and
Kroeber, Alfred L. Peruvian
Archaeology in 1942. Viking Fund
Publications in Anthropology 4.
New York: Viking, 1944.
Kroeber 1930
. Archaeological Explorations
in Peru: Part II, The Northern
Coast. Anthropology Memoirs 2,
no. 2. Chicago: Field Museum of
Natural History, 1930.
Kuen Lee and Zhu 2002
Kuen Lee, Yun, and Naicheng Zhu.
“Societal Integration of Religion
and Ritual in Prehistoric China.”
Antiquity 76, no. 293 (2002):
715–23.
287 R eference list
La Crónica 1943
La Crónica. “En Torno al Reciente
Descubrimiento Arqueológico
de Chuquibamba, formula inte-
resantes declaraciones para La
Crónica el doctor Luis E. Valcárcel,
director de Museos. Informe del
Subprefecto de esa circunscrip-
ción, señor Ernesto Tejada.” La
Crónica (Lima), 23 February 1943.
Lanning 1967
Lanning, Edward. Peru before
the Incas. Englewood Cliffs, NJ:
Prentice-Hall, 1967.
Lapiner 1976
Lapiner, Alan C. Pre-Columbian
Art of South America. New York:
Abrams, 1976.
Larco Hoyle 1948
Larco Hoyle, Rafael. Cronología
arqueológica del norte del Perú.
Buenos Aires: Sociedad Geográfica
Americana, 1948.
Larco Hoyle 1945
. Los Mochicas (Pre-Chimu
de Uhle y Early Chimu de Kroeber).
Buenos Aires: Sociedad Geográfica
Americana, 1945.
Larsen et al. 1976
Larsen, Jack Lenor, with Alfred
Bühler, and Bronwen and Garrett
Solyom. The Dyer’s Art: Ikat, Batik,
Plangi. New York: Van Nostrand
Reinhold, 1976.
Lau 2011
Lau, George. Andean Expressions:
Art and Archaeology of the Recuay
Culture. Iowa City: University of
Iowa Press, 2011.
Lau 2006
. “Northern Exposures:
Recuay-Cajamarca Boundaries
Silverman 2006, 143–70.
Lau 2005
. “Core-Periphery Relations
in the Recuay Hinterlands: Eco­
nomic Interaction at Chinchawas.”
Antiquity, no. 79 (2005): 78–99.
Lavalle 1984
Lavalle, José Antonio de. Arte y
Tesoros del Perú: Huari. Lima:
Banco de Crédito del Perú, 2009.
Lavalle and González García 1993
Lavalle, José Antonio de, and José
Alejandro González García, eds.
The Textile Art of Peru. Lima:
Industria Textil Piura, 1993.
Expanded and enlarged version of
Lavalle and González García 1989.
Lavalle and González García 1989
. Arte textil del Perú. Lima:
Industria Textil Piura, 1989.
Lavalle and Lavalle de Cárdenas
1999
Lavalle, José Antonio de, and
Rosario de Lavalle de Cárdenas,
eds. Tejidos milenarios del Perú/
Ancient Peruvian Textiles. Lima:
Integra, 1999.
Lechtman 2005
Lechtman, Heather. “Arsenic
Bronze at Pikillacta.” In McEwan
2005b, 131–46.
Lechtman 1999
. “El Bronce y el Horizonte
Medio.” Boletín Museo de Oro
(Bogota), no. 41 (1999): 3–25.
Lechtman 1997
. “El Bronce Arsenical y el
Horizonte Medio.” In R. Varón and
J. F. Espinoza, eds., Arqueología,
Antropología e Historia en los
Andes, 153–86. Lima: Instituto de
Estudios Peruanos, 1997.
Lechtman 1979
. “Issues in Andean
Metallurgy.” In E. P. Benson, ed.,
Pre-Columbian Metallurgy of
South America, 1–34. Washington,
DC: Dumbarton Oaks, 1979.
Lechtman forthcoming
. “Andean Metallurgy in
Prehistory.” In B. Roberts and
C. Thornton, eds., A Global
Perspective on Early Metallurgy.
New York: Springer, forthcoming.
Lechtman and Frame 2005
Lechtman, Heather, and Lisa
Frame. “Bronze Artifacts from
Conchopata, Peru.” Paper pre-
sented at 70th annual meeting
of the Society for American
Archaeology. Salt Lake City, 2005.
Lempertz 2010
Lempertz Auction 952. Pre-
Columbian Art: A Selection from
a European Private Museum.
Brussels, 30 January 2010.
Leoni 2004
Leoni, Juan Bautista. “Ritual,
Place, and Memory in the
Construction of Community
Identity: A Diachronic View from
Ñawinpukyo (Ayacucho, Peru).”
PhD diss., State University of New
York at Binghamton, 2004.
Liles 1990
Liles, J. N. The Art and Craft
of Natural Dyeing: Traditional
Recipes for Modern Use. Knoxville:
University of Tennessee Press,
1990.
Linares Málaga 1993
Linares Málaga, Eloy. Pre-Historia
de Arequipa Marzo 1990–1991.
Arequipa: CONCYTEC–UNSA,
1993.
Linares Málaga 1990
. Pre-Historia de Arequipa
Marzo 1987–1990. Arequipa:
CONCYTEC–UNSA, 1990.
Llagostera et al. 1988
Llagostera, Augustín, Constantino
Manuel Torres, and María
Antonietta Costa. “El complejo
psicotrópico en Solcor-3 (San
Pedro de Atacama).” Estudios
Atacameños (San Pedro de
Atacama, Chile), no. 9 (1988):
61–98.
Lommel 1977
Lommel, Andreas. Altamerika­
nische Kunst Mexico-Peru. Exh.
cat. Munich: Staatliches Museum
für Völkerkunde, 1977.
Lumbreras 1981
Lumbreras, Luis G. “The
Stratigraphy of the Open Sites.” In
MacNeish et al. 1981, 167–98.
Lumbreras 1977
. “Excavaciones en el Templo
Antiguo de Chavin (sector R):
Informe de la sexta campaña.”
Ñawpa Pacha, no. 15 (1977): 1–38.
Lumbreras 1974a
. Las fundaciones de Hua­
manga. Lima: Club Huamanga,
1974.
Lumbreras 1974b
. The Peoples and Cultures
of Ancient Peru. Washington, DC:
Smithsonian Institution Press,
1974.
Lumbreras 1960
. “La cultura Wari.” Etnología
y Arqueología (Lima) 1, no. 1
(1960): 130–227.
Lyon 1978
Lyon, Patricia J. “Female
Supernaturals in Ancient Peru.”
Ñawpa Pacha, no. 16 (1978):
95–140.
Mackey 1982
Mackey, Carol J. “The Middle
Horizon as Viewed from the
Moche Valley.” In M. E. Moseley
and K. C. Day, eds., Chan Chan:
Andean Desert City, 321–31.
Albuquerque: University of New
Mexico Press, 1982.
Mackey et al. 1990
Mackey, Carol, Hugo Pereyra,
Carlos Radicati, Humberto
Rodriguez, and Oscar Valverde,
eds. Quipu y Yupana: Colección de
escritos. Lima: CONCYTEC, 1990.
MacNeish et al. 1981
MacNeish, Richard S., et al., eds.
Prehistory of the Ayacucho Basin.
Vol. 2. Ann Arbor: University of
Michigan Press, 1981.
Makowski Hanula 2009
Makowski Hanula, Krzysztof.
“Royal Statues, Staff Gods, and
the Religious Ideology of the
Prehistoric State of Tiwanaku.” In
Young-Sánchez 2009, 133–64.
Makowski Hanula 2002
. “Los personajes frontales
de báculos en la iconografía
Tiahuanaco y Huari: Tema o
convención?” In Kaulicke and W.
Isbell 2002, 337–73.
Makowski Hanula 2000–2001
, ed. Los dioses del antiguo
Perú. 2 vols. Lima: Banco de
Crédito del Perú, 2000–2001.
Mannheim 1991
Mannheim, Bruce. The Language
of the Inka since the European
Invasion. Austin: University of
Texas Press, 1991.
Marcone Flores 2010
Marcone Flores, Giancarlo. “What
Role Did Wari Play in the Lima
Political Economy? The Peruvian
Central Coast in the Beginning of
the Middle Horizon.” In Jennings
2010c, 136–54.
Marcus and Williams 2009
Marcus, Joyce, and Patrick
Ryan Williams, eds. Andean
Civilization: A Tribute to Michael
E. Moseley. Monograph 63. Los
Angeles: Cotsen Institute of
Archaeology Press, 2009.
Martinez 1987
Martinez, Gabriel. “Para una
etnografía del riego en Chiapa:
Medidas y calendario.” Chungará,
no. 18 (1987): 163–79.
28 8 R eference list
Mayer 2002
Mayer, Enrique. The Articulated
Peasant: Household Economies
in the Andes. Boulder: Westview
Press, 2002.
Mays 1985–86
Mays, John Bentley. “Comment.”
American Craft 45, no. 6 (1985–86):
38–40.
McClelland et al. 2007
McClelland, Donna, Don
McClelland, and Christopher
B. Donnan. Moche Fineline
Painting from San José de Moro.
Los Angeles: Cotsen Institute of
Archaeology Press, 2007.
McEwan 2012
McEwan, Gordon F. “Indicators
of Possible Driving Forces for the
Spread of Quechua and Aymara
Reflected in the Archaeology of
Cuzco.” In P. Heggarty and D.
Beresford-Jones, eds., Archaeology
and Language in the Andes,
247–63. Proceedings of the British
Academy 173. Oxford: Oxford
University Press, 2012.
McEwan 2005a
. “Excavations at Pikillacta.”
In McEwan 2005c, 29–62.
McEwan 2005b
. “Conclusion: The Functions
of Pikillacta.” In McEwan 2005c,
147–64.
McEwan 2005c
, ed. Pikillacta: The Wari
Empire in Cuzco. Iowa City:
University of Iowa Press, 2005.
McEwan 1998
. “The Function of Niched
Halls in Wari Architecture.” Latin
American Antiquity 9, no. 1 (1998):
68–86.
McEwan 1996
. “Archaeological Investi­
gations at Pikillacta, a Wari Site in
Peru.” Journal of Field Archaeology
23, no. 2 (1996): 169–86.
McEwan 1991
. “Investigations at the
Pikillacta Site: A Provincial Huari
Center in the Valley of Cuzco.”
In W. Isbell and McEwan 1991b,
93–120.
McEwan 1990
. “Some Formal Correspond­
ences between the Imperial Archi­
tecture of the Wari and Chimú
Cultures of Ancient Peru.” Latin
American Antiquity 1, no. 2 (1990):
97–116.
McEwan 1987
. The Middle Horizon in
the Valley of Cuzco, Peru: The
Impact of the Wari Occupation of
Pikillacta in the Lucre Basin. BAR
International Series 372. Oxford:
B.A.R., 1987.
McEwan 1984a
. “Investigaciones en la
cuenca del Lucre, Cuzco.” Gaceta
Arqueológica Andina 3, no. 9
(1984): 12–14.
McEwan 1984b
. “The Middle Horizon in
the Valley of Cuzco, Peru. The
Impact of the Wari Occupation of
Pikillacta in the Lucre Basin.” PhD
diss., University of Texas, 1984.
McEwan and Couture 2005
McEwan, Gordon F., and Nicole
Couture. “Pikillacta and Its
Architectural Typology.” In
McEwan 2005c, 21–27.
Meddens 1991
Meddens, Frank. “A Provincial
Perspective of Huari Organization
Viewed from the Chicha/Soras
Valley.” In W. Isbell and McEwan
1991b, 215–32.
Meddens and Branch 2010
Meddens, Frank, and Nicholas
Branch. “The Wari State, Its Use of
Ancestors, Rural Hinterland, and
Agricultural Infrastructure.” In
Jennings 2010c, 155–70.
Medina Castro and Gheller Doig
2005
Medina Castro, María Ysabel, and
Roberto Gheller Doig. Textiles
of Ancient Peru/Tejidos del Perú
antiguo. Lima: n.p., 2005.
Menzel 1977
Menzel, Dorothy. The Archaeology
of Ancient Peru and the Work of
Max Uhle. Exh. cat. Berkeley:
Robert H. Lowie Museum of
Anthropology, 1977.
Menzel 1968
. “New Data on the Huari
Empire in Middle Horizon Epoch
2A.” Ñawpa Pacha, no. 6 (1968):
47–114.
Menzel 1964
. “Style and Time in the
Middle Horizon.” Ñawpa Pacha,
no. 2 (1964): 1–106.
Meyerson 1990
Meyerson, Julia. Tambo: Life in an
Andean Village. Austin: University
of Texas Press, 1990.
Milan 2003
Perú: tremila anni di capolavori.
Exh. cat. Milan: Electa, 2003.
Miller and Taube 1993
Miller, Mary, and Karl Taube.
The Gods and Symbols of
Ancient Mexico and the Maya:
An Illustrated Dictionary of
Mesoamerican Religion. New York/
London: Thames and Hudson,
1993.
Milliken 2006
Milliken, Charlene D. “Ritual and
Status: Mortuary Display at the
Household Level at the Middle
Horizon Wari Site of Conchopata,
Peru.” PhD diss., University of
Pittsburgh, 2006.
Millones and Onuki 1993
Millones, Luis, and Yoshio Onuki,
eds. El Mundo Ceremonial Andino.
Senri Ethnological Studies
37. Osaka: Museo Nacional de
Etnología, 1993.
Misugi 1985
Misugi, Takatoshi. The Andes—
Textiles and Ceramics: The
Collection of Ohara Gallery of Art.
Kobe, Japan: Ohara Gallery of Art,
1985.
Montesinos [1644] 1882
Montesinos, Fernando de.
Memorias Antiquas Historicales
y Politicas del Perú. Madrid:
Imprenta de Miguel Ginesta, 1882.
Monzón [1586] 1965
Monzón, Luis de. “Descripición de
la tierra del Repartimiento de los
rucanas antamarcas de la corona
real, jurisdicción de la ciudad
de Guamanga, año de 1586.” In
M. Jiménez de la Espada, ed.,
Relaciones Geográficas de Indias:
Perú, vol. 1. Biblioteca de Autores
Españoles, 183: 237–48. Madrid:
Ediciones Atlas, 1965.
Moore 1995
Moore, J. D. “The Archaeology
of Dual Organization in Andean
South America: A Theoretical
Review and Case Study.” Latin
American Antiquity 6, no. 2 (1995):
165–81.
Moraga 2005
Moraga, Vanessa Drake. Ancient
Peruvian Textiles: The Fifi White
Collection. Exh. cat. Overland
Park, KS: Johnson County
Community College, Gallery of
Art, 2005.
Morisette and Racine 1973
Morisette, Jacques, and Luc Racine.
“La hiérarchie des wamani: essai
sur la pensée classificatoire que-
chua.” Récherches Amerindiennes
au Québec 3, no. 1–2 (1973):
167–88.
Morris 2004
Morris, Craig. “Enclosures of
Power: The Multiple Spaces of Inca
Administrative Palaces.” In Evans
and Pillsbury 2004, 299–324.
Morris and Thompson 1985
Morris, Craig, and Donald E.
Thompson. Huánuco Pampa:
An Inca City and Its Hinterland.
London: Thames and Hudson,
1985.
Morris and Von Hagen 2011
Morris, Craig, and Adriana von
Hagen. The Incas: Lords of the Four
Quarters. London: Thames and
Hudson, 2011.
Morris and Von Hagen 1993
. The Inka Empire and
Its Andean Origins. New York:
American Museum of Natural
History/Abbeville Press, 1993.
Moseley et al. 2005
Moseley, Michael E., Donna J.
Nash, Patrick Ryan Williams,
Susan D. deFrance, Ana Miranda,
and Mario Ruales. “Burning Down
the Brewery: Establishing and
Evacuating an Ancient Imperial
Colony at Cerro Baúl, Peru.”
In Proceedings of the National
Academy of Sciences 102, no. 48
(2005): 17, 264–71.
Moseley et al. 1991
Moseley, Michael E., Robert A.
Feldman, Paul S. Goldstein, and
Luis Watanabe. “Colonies and
Conquest: Tiahuanaco and Huari
in Moquegua.” In W. Isbell and
McEwan 1991b, 141–64.
Murra 1975
Murra, John V. “El tráfico de
mullu de la costa del Pacífico.” In
Murra, Formaciones Económicas
y Políticas del Mundo Andino,
255–67. Lima: Instituto de Estudios
Peruanos, 1975.
Murra 1968
. “An Aymara Kingdom
in 1567.” Ethnohistory 15, no. 2
(1968): 115–51.
Murra 1962
. “Cloth and Its Function
in the Inca State.” American
Anthropologist 64, no. 4 (1962):
710–28.
289 R eference list
Nakajima [1969]
Nakajima, Akiko. Kodai Inka no
Geijutsu [Art of the Ancient Incas].
Kobe, Japan: Ohara Gallery of Art
[1969].
Nash 2010
Nash, Donna J. “Fine Dining and
Fabulous Atmosphere: Feasting
Facilities and Political Interaction
in the Wari Realm.” In Klarich
2010, 83–109.
Nash forthcoming
. “Huánuco Pampa y Cerro
Baúl: Centros de genuino diseño
imperial.” In J. L. Pino, ed., El
Camino Inca: Actas del Congreso
Craig Morris, June 2010. Qhapac
Ñan, forthcoming.
Nash and Williams 2009
Nash, Donna J., and Patrick
Ryan Williams. “Wari Political
Organization on the Southern
Periphery.” In Marcus and
Williams 2009, 257–76.
Nash and Williams 2005
. “Architecture and Power:
Relations on the Wari–Tiwanaku
Frontier.” In Vaughn et al. 2005,
151–74.
National Geographic 2004
“Geographica.” National Geo­
graphic 205, no. 4 (2004): fig. xxxii.
Niles 1992
Niles, Susan. “Inca Architecture
and the Sacred Landscape.” In
Townsend 1992, 346–57.
Niles 1987
. Callachaca: Style and Status
in an Inca Community. Iowa City:
University of Iowa Press, 1978.
Oakland 1992
Oakland, Amy. “Textiles and
Ethnicity: Tiwanaku in San Pedro
de Atacama, North Chile.” Latin
American Antiquity 3, no. 4 (1992):
316–40.
Oakland 1986a
. “Tiahuanaco Tapestry
Tunics and Mantles from San
Pedro de Atacama, Chile.” In A.
Rowe 1986a, 101–22.
Oakland 1986b
. “Tiwanaku Textile Style
from the South Central Andes,
Bolivia and North Chile.” PhD
diss., University of Texas, Austin,
1986.
Ochatoma Paravicino 2007
Ochatoma Paravicino, José.
Alfareros del Imperio Huari: Vida
cotidiana y áreas de actividad
en Conchopata. Ayacucho:
Universidad Nacional San
Cristóbal de Huamanga, 2007.
Ochatoma Paravicino and Cabrera
Romero 2002
Ochatoma Paravicino, José,
and Martha Cabrera Romero.
“Religious Ideology and Military
Organization in the Iconography of
a D-Shaped Ceremonial Precinct at
Conchopata.” In Silverman and W.
Isbell 2002, 225–48.
Ochatoma Paravicino and Cabrera
Romero 2001
. “Arquitectura y áreas
de actividad en Conchopata.”
In Kaulicke and W. Isbell 2001,
449–88.
Ochatoma Paravicino and Cabrera
Romero 1999
. “Descubrimiento del área
ceremonial en Conchopata, Huari.”
In XII Congreso Peruano del
Hombre y la Cultura Andina “Luis
G. Lumbreras,” 212–45. Ayacucho:
Universidad Nacional de San
Cristóbal de Huamanga, 1999.
O’Neale 1977
O’Neale, Lila M. “Notes on Pottery
Making in Highland Peru.” Ñawpa
Pacha, no. 17 (1977): 41–60.
O’Neale and Kroeber 1930
O’Neale, Lila M., and Alfred
L. Kroeber. “Textile Periods
in Ancient Peru.” University
of California Publications in
American Archaeology and
Ethnology 28, no. 2 (1930): 23–56.
Onuki 1997
Onuki, Yoshio. “Ocho Tumbas
Especiales de Kuntur Wasi.” In
Kaulicke 1997, 79–114.
Onuki and Inokuchi 2011
Onuki, Yoshio, and Kinya
Inokuchi. “Gemelos Prístinos:
El Tesoro del Templo de Kuntur
Wasi.” Lima: Fondo Editorial del
Congreso del Perú de Minería
Yanacocha, 2011.
Orefici 1993
Orefici, Giuseppe. Nasca: Arte
e società del popolo dei geoglifi.
Milan: Jaca Books, 1993.
Owen 2010
Owen, Bruce. “Wari in the Majes-
Camaná Valley: A Different Kind
of Horizon.” In Jennings 2010c,
57–77.
Owen 2007
. “Rural Wari Far from the
Heartland: Huamanga Ceramics
from Beringa, Majes Valley, Peru.”
Andean Past 8 (2007): 287–373.
Pärssinen 1992
Pärssinen, Martti. “Tawantinsuyu:
The Inca State and Its Political
Organization.” Studia Historica
(Helsinki), no. 43 (1992).
Pasztory 2010
Pasztory, Esther. Inka Cubism:
Reflections on Andean Art. New
York: Columbia University, 2010,
http://www.columbia.edu/~ep9/
books.html.
Pasztory 1998
. Pre-Columbian Art.
Cambridge/New York: Cambridge
University Press, 1998.
Pasztory 1990–91
. “Still Invisible: The Problem
of the Aesthetics of Abstraction
for Pre-Columbian Art and Its
Implications for Other Cultures.”
RES: Anthropology and Aesthetics
19/20 (1990–91): 104–36.
Pasztory 1984
. “The Function of Art in
Mesoamerica.” Archaeology 37, no.
1 (1984): 18–25.
Paternosto 1996
Paternosto, César. The Stone and
the Thread: Andean Roots of
Abstract Art. Austin: University of
Texas Press, 1996.
Patterson 1970
Patterson, Thomas C. “Chavín: An
Interpretation of Its Spread and
Influence.” In Benson 1970, 29–48.
Paulson 1974
Paulson, Allison C. “The Thorny
Oyster and the Voice of God:
Spondylus and Strombus in
Andean Prehistory.” American
Antiquity 39, no. 4 (1974): 597–607.
Pease 2008
Pease, Franklin. “Cieza de
León, Pedro de (ca. 1518–1554).”
In J. Pillsbury, ed., Guide to
Documentary Sources for Andean
Studies, 1530–1900, 2:143–51.
Norman: University of Oklahoma
Press, 2008.
Pérez Calderón 2008
Pérez Calderón, Ismael. “Qochas,
Puquiales, Canales y Reservorios
que Abastecían a la Metrópoli
Wari.” Investigación (Ayacucho)
16, no. 2 (2008): 50–58.
Pérez Calderón 2003
. “Estudio Arqueológico en
la Periferia del Complejo Huari.”
Investigación (Ayacucho) 40
(2003): 119–28.
Pérez Calderón 2001
. “Investigaciones en la
periferia del complejo Huari.”
In I. Pérez, W. Aguilar, and M.
Purizaga, eds., XII Congreso
Peruano del Hombre y la Cultura
Andina, 2:246–70. Ayacucho:
Universidad Nacional de San
Cristóbal de Huamanga, 2001.
Pérez Calderón 1999
. Huari: Misteriosa Ciudad
de Piedra. Ayacucho: Universidad
Nacional de San Cristóbal de
Huamanga, 1999.
Perodie 2001
Perodie, James R. “Feasting for
Prosperity: A Study of Southern
Northwest Coast Feasting.”
In Dietler and Hayden 2001b,
185–214.
Phipps 2004a
Phipps, Elena. “Cumbi to Tapestry:
Collection, Innovation, and
Transformation of the Colonial
Andean Tapestry Tradition.” In
Phipps et al. 2004, 72–99.
Phipps 2004b
. “Garments and Identity in
the Colonial Andes.” In Phipps et
al. 2004, 16–41.
Phipps et al. 2004
Phipps, Elena, Johanna Hecht,
and Cristina Esteras Martín. The
Colonial Andes: Tapestries and
Silverwork, 1530–1830. New York:
Metropolitan Museum of Art; New
Haven/London: Yale University
Press, 2004.
Pillsbury 2001
Pillsbury, Joanne, ed. Moche Art
and Archaeology in Ancient Peru.
Washington, DC: National Gallery
of Art, 2001.
Pillsbury 1996
. “The Thorny Oyster and
the Origins of Empire.” Latin
American Antiquity 7, no. 4 (1996):
313–40.
Pollack et al. 1987
Pollack, David, Mary Lucas
Powell, and Audrey Adkins.
“Preliminary Study of Mortuary
Patterns at the Larkin Site,
Bourbon County, Kentucky.” In
Current Archaeological Research
in Kentucky, 1:188–203. Frankfort:
Heritage Council, 1987.
29 0 R eference list
Pozorski and Pozorski 1986
Pozorski, Thomas, and Shelia
Pozorski. “Recent Excavations
at Pampa de las Llamas-Moxeke,
A Complex Initial Period Site in
Peru.” Journal of Field Archaeology
13 (1986): 381–401.
Pozzi-Escott 1991
Pozzi-Escott, Denise. “Conchopata:
A Community of Potters.” In W.
Isbell and McEwan 1991b, 81–92.
Pozzi-Escot et al. 1998
Pozzi-Escot, Denise, Marleni M.
Alarcón, and Cirilio Vivanco.
“Wari Ceramics and Production
Technology: The View from
Ayacucho.” In Shimada 1998,
253–81.
Proulx 2008
Proulx, Donald. “Paracas and
Nasca: Regional Cultures of the
South Coast of Peru.” In Silverman
and W. Isbell 2008, 563–85.
Proulx 2006
. A Sourcebook of Nasca
Ceramic Iconography. Iowa City:
University of Iowa Press, 2006.
Proulx 2001
. “The Ritual Use of Trophy
Heads in Ancient Nasca Society.”
In Benson and Cook 2001, 119–36.
Proulx 1999
. “Kopfjagd und rituelle
Verwendung von Tropäenköpfen
in der Nasca-Kultur [Headhunting
and Ritual Use of Trophy
Heads in the Nasca Culture].”
In J. Rickenbach, ed., Nasca:
Geheimnisvolle Zeichen im Alten
Peru, 79–87. Exh. cat. Zürich:
Museum Rietberg, 1999.
Protzen 1999
Protzen, Jean-Pierre. “Inca
Architecture.” In L. L. Minelli, ed.,
The Inca World: The Development
of Pre-Columbian Peru, A.D. 1000–
1534, 193–217. Norman: University
of Oklahoma Press, 1999.
Prümers 2001
Prümers, Heiko. “ ‘El Castillo’
de Huarmey: una plataforma
funeraria del Horizonte Medio.”
In Kaulicke and W. Isbell 2001,
289–312.
Prümers 1990
. Der Fundort “El Castillo”
im Huarmeytal, Peru. Ein Beitrag
zum Problem des Moche-Huari
Textilstils. 2 vols. Bonn: Holos
Verlag, 1990.
Quilter 2002
Quilter, Jeffrey. “Moche Politics,
Religion and Warfare.” Journal of
World Prehistory 16, no. 2 (2002):
145–95.
Quilter and Castillo Butters 2010
Quilter, Jeffrey, and Luis Jaime
Castillo Butters, eds. New
Perspectives on Moche Political
Organization. Washington, DC:
Dumbarton Oaks, 2010.
Quilter and Urton 2002
Quilter, Jeffrey, and Gary
Urton, eds. Narrative Threads:
Explorations of Narrativity in
Andean Khipus. Austin: University
of Texas Press, 2002.
Ramos and Blasco 1977
Ramos, Luis J., and María
Concepción Blasco. “Materiales
Huaris no cerámicos en el Museo
de América: Orfebrería, textiles
y pequeña escultura (Figurillas
de Pikillajta).” Cuadernos
Prehispánicos (Valladolid, Spain),
no. 5 (1977): 55–108.
Ravines 1981
Ravines, Rogger. “Prácticas
Funerarias en Ancón (Segunda
Parte).” Revista del Museo
Nacional (Lima), no. 45 (1981):
89–166.
Ravines 1977
. “Prácticas Funerarias en
Ancón (Primera Parte).” Revista
del Museo Nacional (Lima), no. 43
(1977): 327–98.
Reeves 1949
Reeves, Ruth. “Pre-Columbian
Fabrics of Peru.” Magazine of Art,
no. 42 (1949): 103–7.
Rehfisch 1987
Rehfisch, F. “Competitive Beer
Drinking Among the Mambila.”
In M. Douglas, ed., Constructive
Drinking: Perspectives on Drink
from Anthropology, 135–45.
Cambridge: Cambridge University
Press, 1987.
Rehl 2010
Rehl, Jane W. Weaving Metaphors,
Weaving Cosmos: Reflections
of a Shamanic Worldview in
Discontinuous Warp and Weft
Textiles of Ancient Peru, 300 BCE–
1540 CE. Saarbrücken: Lambert
Academic Publishing, 2010.
Rehl 2006
. “Weaving Principles for
Life: Discontinuous Warp and Weft
Textiles of Ancient Peru.” In M.
Young-Sánchez and F. W. Simpson,
eds., Andean Textile Traditions:
Papers from the 2001 Mayer Center
Symposium at the Denver Art
Museum, 12–42. Denver: Denver
Art Museum, 2006.
Rehl 2003
. “Weaving Metaphors,
Weaving Cosmos: Reflections
of a Shamanic Worldview in
Discontinuous Warp and Weft
Textiles of Ancient Peru, 300
BCE–1540 CE.” PhD diss., Emory
University, 2003.
Rehl 2001
. “The Order of Things in
Ancient Peru: Visual Metaphors in
Wari-associated DWW Textiles.”
In Approaching Textiles: Varying
Viewpoints: Proceedings of the
7th Biennial Symposium of the
Textile Society of America, 9–19.
Earleville, MD: Textile Society of
America, 2001.
Reid 1986
Reid, James W. Textile Master­
pieces of Ancient Peru. New York:
Dover Publications, 1986.
Reinhard and Ceruti 2010
Reinhard, Johan, and Maria
Constanza Ceruti. Inca Rituals and
Sacred Mountains: A Study of the
World’s Highest Archaeological
Sites. Los Angeles: Cotsen Institute
of Archaeology Press, 2010.
Reiss and Stübel 1880–87
Reiss, Wilhelm, and Alfons Stübel.
The Necropolis of Ancón in Peru,
trans. by A. H. Keane. Berlin: A.
Asher, 1880–87.
Rengifo Chunga et al. 2008
Rengifo Chunga, Carlos E., Daniela
Zevallos Castañeda, and Luis
Muro Ynoñán. “Excavaciones en el
sector norte de San José de Moro.
Las ocupaciones Mochica Medio
y Mochica Tardío.” In Programa
Arqueológico San José de Moro,
Temporada 2007, 119–61. Lima:
Fondo Editorial de la Pontificia
Universidad Católica del Perú,
2008.
Richards 1939
Richards, Audrey I. Land, Labour,
and Diet in Northern Rhodesia.
London: Oxford University Press,
1939.
Rodman and Cassman 1995
Rodman, Amy Oakland, and Vicki
Cassman. “Andean Tapestry:
Structure Informs Surface.” Art
Journal 54, no. 2 (1995): 33–39.
Rodman and Fernández Lopez
2005
Rodman, Amy Oakland, and
Gioconda Arabel Fernández
Lopez. “North Coast Style after
Moche: Clothing and Identity at El
Brujo, Chicama Valley, Peru.” In
R. M. Reycraft, ed., Us and Them:
Archaeology and Ethnicity in
the Andes, 115–33. Los Angeles:
Cotsen Institute of Archaeology
Press, 2005.
Rodman and Fernández Lopez
2001
. “Los tejidos Huari y
Tiwanaku: Comparaciones y con-
textos.” In Kaulicke and W. Isbell
2001, 119–30.
Roe 2008
Roe, Peter. “How to Build a Raptor:
Why the Dumbarton Oaks ‘Scaled
Cayman’ Callango Textile Is
Really a Jaguaroid Harpy Eagle.”
In W. Conklin and J. Quilter, eds.,
Chavin: Art, Architecture and
Culture, 181–216. Los Angeles:
Cotsen Institute of Archaeology
Press, 2008.
Rosas Rintel 2007
Rosas Rintel, Marco. “Nuevas
perspectivas acerca del colapso
Moche en el Bajo Jequetepeque.
Resultados preliminares de la
segunda campaña de investigación
del proyecto arqueológico Cerro
Chepén.” Bulletin de l’Institut
Français d’Études Andines (Lima)
36, no. 2 (2007): 221–40.
Rosenfeld 2006
Rosenfeld, Silvana. “Feasting
and Animal Offerings in the
Andes: The Case of Conchopata
(Ayacucho, Peru).” Paper presented
at 71st annual meeting of the
Society for American Archaeology.
San Juan, 2006.
Rosenfeld 2004
. “Zooarchaeology at Concho­
pata, a Huari Urban Center.” Paper
presented at 69th annual meet-
ing of the Society for American
Archaeology. Montreal, 2004.
Rösing 2003
Rösing, Ina. Religión, ritual y
vida cotidiana en los Andes, Los
diez géneros de Amarete. Madrid:
Iberoamericana, 2003.
291 R eference list
Rostworowski de Diez Canseco
1999
Rostworowski de Diez Canseco,
Maria. History of the Inca
Realm, trans. by Harry Iceland.
Cambridge: Cambridge University
Press, 1999.
Rostworowski de Diez Canseco
1997
. Pachacutec y la Leyenda de
Los Chancas. Lima: Instituto de
Estudios Peruanos, 1997.
A. Rowe 2005
Rowe, Ann Pollard. Gods and
Empire: Huari Ceremonial Textiles.
Exh. brochure. Washington, DC:
Textile Museum, 2005.
A. Rowe 1996
. “The Art of Peruvian
Textiles.” In Boone 1996, 2:329–45.
A. Rowe 1991
. “Nasca Figurines and
Costume.” Textile Museum Journal
29/30 (1991): 93–128.
A. Rowe 1986a
, ed. The Junius B. Bird
Conference on Andean Textiles.
Washington, DC: Textile Museum,
1986.
A. Rowe 1986b
. “Textiles from the Nasca
Valley at the Time of the Fall of the
Huari Empire.” In A. Rowe 1986a,
151–82.
A. Rowe 1979
. “Textile Evidence for Huari
Music.” Textile Museum Journal 18
(1979): 5–18.
A. Rowe 1978
. “Technical Features of Inca
Tapestry Tunics.” Textile Museum
Journal 17 (1978): 5–28.
A. Rowe 1977
. Warp-patterned Weaves
of the Andes. Washington, DC:
Textile Museum, 1977.
A. Rowe and Miller 2007
Rowe, Ann Pollard, and Laura M.
Miller. “Cochineal in Salasaca,
Tungurahua Province.” In A. P.
Rowe, ed., Weaving and Dyeing in
Highland Ecuador, 266–77, 287–88.
Austin: University of Texas Press,
2007.
J. Rowe 1979
Rowe, John H. “Standardization
in Inca Tapestry Tunics.” In
A. P. Rowe, E. Benson, and A.-L.
Schaffer, eds., The Junius B. Bird
Pre-Columbian Textile Conference,
239–64. Washington, DC: Textile
Museum/Dumbarton Oaks, 1979.
J. Rowe 1970
. “The Influence of Chavín
on Later Styles.” In Benson 1970,
101–24.
J. Rowe 1967
. “An Interpretation of
Radiocarbon Measurements from
Archaeological Samples from
Peru.” In J. Rowe and D. Menzel,
eds., Peruvian Archaeology:
Selected Readings, 16–30. Palo
Alto, CA: Peek Publications, 1967.
J. Rowe 1966
. “An Interpretation of
Radiocarbon Measurements on
Archaeological Samples from
Peru.” In Proceedings of the
Sixth International Conference,
Radiocarbon and Tritium Dating,
187–98. Springfield, VA: U.S.
Atomic Energy Commission,
Division of Technical Information,
CONF-650652 Chemistry (TID-
4500), 1966. Reprinted in John
Rowe and Dorothy Menzel, eds.,
Peruvian Archaeology: Selected
Readings, 16–30. Palo Alto, CA:
Peek Publications, 1967.
J. Rowe 1962a
. Chavin Art: An Inquiry into
Its Form and Meaning. New York:
Museum of Primitive Art, 1962.
J. Rowe 1962b
. “Stages and Periods in
Archaeological Interpretation.”
Southwestern Journal of
Anthropology 18, no. 1 (1962):
40–54.
J. Rowe 1960a
. “Nuevos datos relativos a la
cronología del estilo Nasca.” In R.
Matos Mendieta, ed., Antiguo Perú:
espacio y tiempo, 29–45. Lima:
Librería Mejía Baca, 1960.
J. Rowe 1960b
. “The Origins of Creator
Worship among the Incas.” In S.
Diamond, ed., Culture in History:
Essays in Honor of Paul Radin,
408–29. New York: Columbia
University Press, 1960.
J. Rowe 1946
. “Inca Culture at the Time
of the Spanish Conquest.” In
J. H. Steward, ed., The Andean
Civilization, vol. 2 of Handbook of
South American Indians, 183–330.
Bureau of American Ethnology
Bulletin 143. Washington, DC:
Smithsonian Institution, 1946.
J. Rowe and Brandel 1976
Rowe, John H., and Catherine Terry
Brandel. “El Arte Religioso del
Cuzco en el Horizonte Temprano.”
Ñawpa Pacha, no. 4 (1976): 1–20.
J. Rowe et al. 1950
Rowe, John H., Donald
Collier, and Gordon R. Willey.
“Reconnaissance Notes on the Site
of Huari, near Ayacucho, Peru.”
American Antiquity 16, no. 2
(1950): 120–37.
Ruales 2001
Ruales, Mario. “Investigaciones en
Cerro del Oro, Valle de Cañete.”
In Kaulicke and W. Isbell 2001,
359–99.
Salomon 1995
Salomon, Frank. “The Beautiful
Grandparents: Andean Ancestor
Shrines and Mortuary Ritual as
Seen Through Colonial Records.”
In Dillehay 1995, 315–54.
Salomon 1991
. “Introductory Essay: The
Huarochirí Manuscript.” In
Salomon and Urioste 1991, 1–38.
Salomon and Urioste 1991
Salomon, Frank, and George
Urioste, eds. and trans., The
Huarochirí Manuscript: A
Testament of Ancient and Colonial
Andean Religions. Austin:
University of Texas Press, 1991.
Sanders 1973
Sanders, William T. The Signifi­
cance of Pikillacta in Andean
Culture History. Occasional Papers
in Anthropology, Pennsylvania
State University 8 (1973): 380–428.
Santiago 2005
Museo Chileno de Arte
Precolombino. Chimú: Laberinto
de un Traje Sagrado [Labyrinths
of a Sacred Costume]. Exh. cat.
Santiago: Museo Chileno de Arte
Precolombino, 2005.
Sandweiss 1987
Sandweiss, Daniel H., ed.
Andean Past. Ithaca, NY: Cornell
University Latin American Studies
Program, 1987.
Sandweiss 1983
, ed. Investigations of the
Andean Past: Papers from the First
Annual Northeast Conference
on Andean Archaeology and
Ethnohistory. Ithaca, NY: Cornell
University Latin American Studies
Program, 1983.
San Román Luna 1983
San Román Luna, Wilbert. “Datos
Etnohistóricos de Pomacanchi.”
Arqueología Andina, no. 1 (1983):
66–74.
Sarmiento de Gambo [1572] 1942
Sarmiento de Gambo, Pedro.
Historia de los Incas. Colección
Hórreo (Buenos Aires), no. 10
(1942): 70.
Santoro and Ulloa 1985
Santoro, Calogero, and Liliana
Ulloa, eds. Culturas de Arica.
Chile: Departamento de Extensión
Cultural del Ministerio de
Educación, 1985.
Sawyer 1997
Sawyer, Alan R. Early Nasca
Needlework. London: Laurence
King in association with Alan
Marcuson, 1997.
Sawyer 1963
. “Tiahuanaco Tapestry
Design.” Textile Museum Journal 1,
no. 2 (1963): 27–38. Simultaneously
published as Studies, no. 3,
Museum of Primitive Art, New
York.
Sayre and Whitehead 2003
Sayre, Matthew P., and
William Whitehead. “New
Paleoethnobotanical Evidence
from Conchopata, a Huari
Site.” Paper presented at 68th
annual meeting of the Society
for American Archaeology.
Milwaukee, 2003.
Schaedel 1993
Schaedel, Richard P. “Congruence
of Horizon with Polity: Huari
and the Middle Horizon.” In
D. S. Rice, ed., Latin American
Horizons, 225–61. Washington, DC:
Dumbarton Oaks, 1993.
Schaedel 1948
. “Monolithic Sculpture of
the Southern Andes.” Archaeology
1, no. 2 (1948): 66–73.
Schaffhausen 1999
Museum zu Allerheiligen.
Vom Toten Meer zum Stillen
Ozean: Sammlung Etnöther.
Schaffhausen, Switzerland:
Museum zu Allerheiligen, 1999.
Schildkraut 1996
Schildkraut, Linda. The Wonders
of the Past. New York: Merrin
Gallery, 1996.
292 R eference list
Schindler 2000
Schindler, Helmut. The Norbert
Mayrock Art Collection from
Ancient Peru. Munich: Staatliches
Museum für Völkerkunde, 2000.
Schmidt 1929
Schmidt, Max. Kunst und Kultur
von Peru. Berlin: Propylean-Verlag,
1929.
Schreiber 2005a
Schreiber, Katharina J. “Old Issues,
New Directions.” Cambridge
Archaeological Journal 15, no. 2
(2005): 264–66.
Schreiber 2005b
. “Sacred Landscapes and
Imperial Ideologies: The Wari
Empire in Sondondo, Peru.” In
Vaughn et al. 2005, 131–50.
Schreiber 2001
. “The Wari Empire of Middle
Horizon Peru: The Epistemological
Challenge of Documenting an
Empire without Documentary
Evidence.” In Alcock et al. 2001,
70–92.
Schreiber 1999
. “Regional Approaches to
the Study of Prehistoric Empires:
Examples from Ayacucho and
Nasca, Peru.” In B. Billman and
G. Feinman, eds., Settlement
Pattern Studies in the Americas:
Fifty Years since Viru, 160–71.
Washington, DC: Smithsonian
Institution Press, 1999.
Schreiber 1993
. “The Inca Occupation
of the Province of Lucanas.” In
M. A. Malpass, ed., Provincial
Inca, 77–116. Iowa City: University
of Iowa Press, 1993.
Schreiber 1992
. Wari Imperialism in Middle
Horizon Peru. Anthropological
Papers 87. Ann Arbor: Museum
of Anthropology, University of
Michigan, 1992.
Schreiber 1991a
. “The Association between
Roads and Polities: Evidence
for Wari Roads in Peru.” In
Charles D. Trombold, ed., Ancient
Road Networks and Settlement
Hierarchies in the New World,
243–52. Cambridge: Cambridge
University Press, 1991.
Schreiber 1991b
. “Jincamocco: A Huari
Administrative Center in the
South Central Highlands of Peru.”
In W. Isbell and McEwan 1991b,
199–213.
Schreiber 1978
. “Planned Architecture
of Middle Horizon Peru:
Implications for Social and
Political Organization.” PhD diss.,
State University of New York at
Binghamton, 1978.
Seguar Llanos and Shimada 2010
Seguar Llanos, Rafael, and Izumi
Shimada. “The Wari Footprint on
the Central Coast: A View from
Cajamarquilla and Pachacamac.”
In Jennings 2010c, 113–35.
Seville 2001
Wari: Arte precolombino peruano.
Exh. cat. Seville: Fundación El
Monte, 2001.
Shady Solís 1988
Shady Solís, Ruth. “Época Huari
como interacción de las sociedades
regionales.” Revista Andina 6, no.
1 (1988): 67–99.
Shady Solís 1982
. “La cultura Nievería y la
interacción social en el mundo
andino en la epoca Huari.”
Arqueológicas, no. 19 (1982):
15–108.
Shady Solís et al. 2000
Shady Solís, Ruth, Joaquín
Narváez, and Sonia López. “La
antigüedad del uso del quipu como
escritura.” Boletín del Museo de
Arqueología y Antropología (Lima),
no. 10 (2000): 2–23.
Shea 1969
Shea, Daniel. “Wari Wilka: A
Central Andean Oracle Site.” PhD
diss., University of Wisconsin,
1969.
Shepard 1956
Shepard, Anna O. Ceramics for
the Archaeologist. Publication
609. Washington, DC: Carnegie
Institution of Washington, 1956.
Sherbondy 1982
Sherbondy, Jeanette. “El regadío,
los lagos y los mitos de origen.”
Allpanchis Phuturinga 17, no. 20
(1982): 3–32.
Shiga 2004
Ancient Civilizations of the
Americas/Amerika Kodai Bummei.
Exh. cat. Shiga, Japan: Miho
Museum, 2004.
Shimada 1998
Shimada, Izumi, ed. Andean
Ceramics: Technology,
Organization, and Approaches.
MASCA Research Papers, vol.
15 supplement. Philadelphia:
University of Pennsylvania
Museum of Archaeology and
Anthropology, 1998.
Shimada 1991
. “Pachacamac Archaeology:
Restrospect and Prospect.” In Uhle
1991, xvi–lxvi.
Shimada 1985
. “Perception, Procurement,
and Management of Resources:
Archaeological Perspective.” In
S. Masuda, I. Shimada, and C.
Morris, eds., Andean Ecology
and Civilization, 357–99. Tokyo:
University of Tokyo, 1985.
Shimada et al. 1991
Shimada, Izumi, Crystal Schaaf,
Lonnie Thompson, and Ellen
Mosley-Thompson. “Cultural
Impacts of Severe Droughts in the
Prehistoric Andes: Application of
a 1,500-year Ice Core Precipitation
Record.” World Archaeology 22, no.
3 (1991): 247–71.
Sillar et al. forthcoming
Sillar, Bill, Emily Dean, and
Amelia Pérez Trujillo. “My State or
Yours: Wari ‘Labor Camps’ and the
Inka Cult of Viracocha at Raqchi,
Cuzco, Peru.” Latin American
Antiquity, forthcoming.
Silverman 2004
Silverman, Helaine, ed. Andean
Archaeology. Malden, MA:
Blackwell Publishing, 2004.
Silverman and W. Isbell 2008
Silverman, Helaine, and William
H. Isbell, eds. Handbook of South
American Archaeology. New York/
London: Springer Science and
Business Media, 2008.
Silverman and W. Isbell 2002
. Andean Archaeology II: Art,
Landscape, and Society. New York:
Kluwer Academic/Plenum Press,
2002.
Silverman and Proulx 2002
Silverman, Helaine, and Donald
Proulx. The Nasca. Malden/
Oxford: Blackwell Publishers,
2002.
Silverman-Proust 1988
Silverman-Proust, Helaine.
“Weaving Technique and the
Registration of Knowledge in the
Cuzco Area of Peru.” Journal of
Latin American Lore 14, no. 2
(1988): 207–41.
Sinclaire Aguirre 1999
Sinclaire Aguirre, Carole, ed.
Amarras: El arte de teñir en los
Andes prehispánicos. Exh. cat.
Santiago: Museo Chileno de Arte
Precolombino, 1999.
Skar 1982
Skar, Harald O. The Warm
Valley People: Duality and Land
Reform among the Quechua
Indians of Highland Peru. Oslo:
Universitetsforlaget, 1982.
Spielvogel 1955
Spielvogel, Rosalind B. “Wari: A
Study of Tiahuanaco Style.” PhD
diss., Yale University, 1955.
Squier 1877
Squier, E. George. Peru; Incidents
of Travel and Exploration in the
Land of the Incas. New York:
Harper Brothers, 1877.
Stone 2011
Stone, Rebecca R. The Jaguar
Within: Shamanic Trance in
Ancient Central and South
American Art. Austin: University
of Texas Press, 2011.
Stone 1987
. “Technique and Form in
Huari-Style Tapestry Tunics: The
Andean Artist, AD 500–800.” PhD
diss., Yale University, 1987.
Stone 1986
. “Color Patterning and the
Huari Artist: The ‘Lima Tapestry’
Revisited.” In A. Rowe 1986a,
137–49.
Stone and McEwan 1990
Stone, Rebecca R., and Gordon
F. McEwan. “The Representation
of the Wari State in Stone
and Thread: A Comparison
of Architecture and Tapestry
Tunics.” RES: Anthropology and
Aesthetics 19 (1990): 53–80.
Stone-Miller 1995
Stone-Miller, Rebecca R. Ancient
Art of the Andes from Chavín
to Inca. London: Thames and
Hudson, 1995.
Stone-Miller 1992a
. “Camelids and Chaos in
Huari and Tiwanaku Textiles.” In
Townsend 1992, 335–46.
Stone-Miller 1992b
. “Creative Abstractions:
Middle Horizon Textiles in the
Museum of Fine Arts, Boston.” In
Stone-Miller 1992c, 34–42.
293 R eference list
Stone-Miller 1992c
. To Weave for the Sun:
Andean Textiles in the Museum of
Fine Arts, Boston. Exh. cat. Boston:
Museum of Fine Arts, Boston,
1992.
Strelow 1996
Strelow, Renate. Gewebe mit
unterbrochenen Ketten aus
dem Vorspanischen Peru/Pre-
Hispanic Peruvian Textiles with
Discontinuous Warp. Berlin:
Staatliche Museen zu Berlin, 1996.
Strong and Evans 1952
Strong, William D., and Clifford
Evans Jr. Cultural Stratigraphy in
the Viru Valley, Northern Peru: The
Formative and Florescent Epoch.
Columbia Studies in Archaeology
and Ethnology 4. New York:
Columbia University Press, 1952.
Stuhr 2008
Stuhr, Joanne, ed. Ritual Beauty:
Art of the Ancient Americas. Exh.
cat. Tucson: University of Arizona
Museum of Art, 2008.
Stübel and Uhle 1892
Stübel, Aphons, and Max Uhle.
Die Ruinenstätte von Tiahuanaco
im Hochlande des alten Perú: Eine
kulturgeschichtliche Studie auf
Grund selbständiger Aufnahmen.
Leipzig: Verlag von Karl W.
Hiersemann, 1892.
Stuiver and Reimer 1993
Stuiver, Minze, and Paula J.
Reimer. “Extended 14C Database
and Revised CALIB Radiocarbon
Calibration Program.” Radiocarbon
35 (1993): 215–30.
Stuiver et al. 2005
Stuiver, Minze, Paula J. Reimer,
and Ron W. Reimer. CALIB 6.0.
WWW Program and Documen­
tation, 1986–2005, http://calib.qub.
ac.uk/calib/calib.html.
Stumer 1958
Stumer, Louis M. “Contactos
Foraneos en la Arqueología de la
Costa Central.” Revista del Museo
Nacional (Lima), no. 27 (1958):
11–30.
Sutter 2011
Sutter, Richard C. “Dentally
Derived Phenetic Relatedness
among the Moche (A.D. 200–750)
of the North Coast of Peru: A
Comparison of the Skeletal
Populations from San José de Moro
to those from the Moche Valley.”
Paper presented at 76th annual
meeting of Society for American
Archaeology. Sacramento, 2011.
Sutter 2009
. “Prehistoric Population
Dynamics in the Andes.” In
Marcus and Williams 2009, 9–38.
Szeminski 1987
Szeminski, Jan. “Un kuraca, un
dios y una historia.”Antropología
e Historia (Juyuy, Argentina), no.
2 (1987).
Taullard 1949
Taullard, Alfredo. Tejidos y pon-
chos indígenas de Sudamérica.
Buenos Aires: Guillermo Kraft,
1949.
Tello [1931] 1970
Tello, Julio C. “Las ruinas de
Huari.” In R. Ravines, ed., 100
años de arqueología en el Perú,
519–26. Fuentes e Investigaciones
para la Historia del Perú 3. Lima:
Instituto de Estudios Peruanos,
1970. Originally published in El
Perú, Diario de la mañana (Lima),
27 August 1931.
Tello 1942
. “Origen y desarollo de
las civilizaciones prehistóricas
andinas.” In XXVII Congreso
Internacional de Americanistas
(1939), Actas y Trabajos, 1:589–720.
Mexico City/Lima: Librería e
Imprenta Gil, 1942.
Tello 1929
. Antiguo Perú: Primera
Época. Lima: Editado por la
Comisión Organizadora del
II Congreso Sudamericano de
Turismo, 1929.
Tesar 2007
Tesar, Louis S. “Wari Warfare and
Trophy Taking: More Than Just
a Skull.” Paper delivered at 72nd
annual meeting of the Society for
American Archaeology. Austin,
2007.
Thatcher 1977
Thatcher, John A. “Middle Horizon
1B Cache from Huamachuco,
North Highlands, Peru.” Ñawpa
Pacha, no. 15 (1977): 101–10.
Thompson et al. 1985
Thompson, Lonnie G., E. Moseley-
Thompson, J. F. Bolzan, and
B. R. Koci. “A 1500-Year Record of
Tropical Precipitation in Ice Cores
from the Quelccaya Ice Cap, Peru.”
Science, no. 229 (1985): 971–73.
Thornton et al. 2011
Thornton, Erin Kennedy, Susan
D. deFrance, John Krigbaum, and
Patrick Ryan Williams. “Isotopic
Evidence for Middle Horizon
to Sixteenth Century Camelid
Herding in the Osmore Valley,
Peru.” International Journal of
Osteoarchaeology 21, no. 5 (2011):
544–67. First published online 25
March 2010, http://onlinelibrary.
wiley.com/doi/10.1002/oa.1157/
abstract.
Tidball 1969
Tidball, Harriet. Peru: Textiles
Unlimited, Part I: Background.
Shuttle Craft Guild Monograph 25.
Santa Ana, CA: HTH Publishers,
1969 [cover date 1968].
Tompkins 2009
Tompkins, Peter. “Domesticity by
Default. Ritual, Ritualization and
Cave Use in the Neolithic Aegean.”
Oxford Journal of Archaeology 28,
no. 2 (2009): 125–53.
J. Topic 2008
Topic, John R. “El Santuario de
Catequil: Estructura y Agencia.
Hacia Una Comprensión de los
Oráculos Andinos.” In M. Curatola
Petrocchi and M. S. Ziołkowski,
eds., Adivinación y oráculos en
el mundo andino antiguo, 71–95.
Lima: Fondo Editorial de la
Pontificia Universidad Católica
del Perú y Instituto Francés de
Estudios Peruanos, 2008.
J. Topic 1994
. “El Izamiento del Gallardete
en Huamachuco.” In L. Millones
and M. Lemlij, eds., En el Nombre
del Señor: Shamanes, Demonios,
y Curanderos del Norte del Perú,
102–27. Lima: Biblioteca Peruana
de Psicoanálisis, 1994.
J. Topic 1992
. “Las huacas de Huama­
chuco: precisiones en torno a una
imagen indígena de un paisaje
andino.” In E. E. Deeds, ed., La per-
secución del demonio. Crónica de
los primeros agustinos en el norte
del Perú (1560), 39–99. Málaga:
Editorial Algazara; Mexico City:
Centro Andino-Mesoamérica de
Estudios Interdisciplinarios, 1992.
J. Topic 1991
. “Huari and Huamachucho.”
In W. Isbell and McEwan 1991b,
141–64.
J. Topic 1986
. “A Sequence of Monumental
Architecture from Huamachuco.”
In D. Sandweiss and D. P. Kvietok,
eds., Perspectives on Andean
Prehistory and Protohistory:
Papers from the Third Annual
Northeast Conference on Andean
Archaeology and Ethnohistory,
63–83. Ithaca, NY: Cornell
University, 1986.
J. Topic and T. Topic 2001
Topic, John R., and Theresa L.
Topic. “Hacia la comprensión del
fenómeno Huari: Una perspectiva
norteña.” In Kaulicke and W. Isbell
2001, 181–217.
J. Topic and T. Topic 1992
. “The Rise and Decline of
Cerro Amaru: An Andean Shrine
During the Early Intermediate
Period and Middle Horizon.”
In A. S. Goldsmith et al., eds.,
Ancient Images, Ancient Thought:
The Archaeology of Ideology;
Proceedings of 23rd Annual
Conference of the Archaeological
Association of the University
of Calgary, 167–80. Calgary:
University of Calgary, 1992.
J. Topic and T. Topic 1983–85
. “El Horizonte Medio en
Huamachuco.” Revista del Museo
Nacional (Lima), no. 47 (1983–85):
13–52.
J. Topic and T. Topic 1983
. Huamachuco Archaeolog­
ical Project: Preliminary Report on
the Second Season, June–August
1982. Peterborough, Ont.: Trent
University, 1983.
J. Topic et al. 2002
Topic, John R., Theresa L. Topic,
and Alfredo M. Caya. “Catequil:
The Archaeology, Ethnohistory,
and Ethnography of a Major
Provincial Huaca.” In W. Isbell and
Silverman 2002, 303–38.
T. Topic 1991
Topic, Theresa L. “The Middle
Horizon in Northern Peru.” In W.
Isbell and McEwan 1991b, 233–46.
T. Topic and J. Topic 2010a
Topic, Theresa L., and John R.
Topic. “Contextualizing the Wari-
Huamachuco Relationship.” In
Jennings 2010c, 188–212.
29 4 R eference list
T. Topic and J. Topic 2010b
. “Spondylus in the Conde­
bamba Valley.” Paper presented at
75th annual meeting of the Society
for American Archaeology. St.
Louis, 2010.
T. Topic and J. Topic 2009
. “Variation in the Practice
of Prehispanic Warfare on the
North Coast of Peru.” In A. Nielsen
and W. Walker, eds., Warfare in
Cultural Context: Practice, Agency,
and the Archaeology of Violence,
17–55. Tucson: University of
Arizona Press, 2009.
T. Topic and J. Topic 1984
. Huamachuco Archaeolog­
ical Project: Preliminary Report
on the Third Season, June–August,
1983. Trent University Occasional
Papers in Anthropology 1.
Peterborough, Ont.: Trent
University, 1984.
Torres and Repke 2006
Torres, Constantino Manuel, and
David B. Repke. Anadenanthera:
Visionary Plant of Ancient South
America. Binghamton, NY:
Haworth Herbal Press, 2006.
Torres et al. 1991
Torres, Manuel Constantino, David
B. Repke, Kelvin Chan, Dennis
McKenna, Augustín Llagostera,
and Richard Evans Schultes.
“Snuff Powders from Pre-Hispanic
San Pedro de Atacama: Chemical
and Contextual Analysis.” Current
Anthropology 32, no. 5 (1991):
640–49.
Townsend 1992
Townsend, Richard F., ed. The
Ancient Americas: Art from Sacred
Landscapes. Exh. cat. Chicago:
Art Institute of Chicago; Munich:
Prestel Verlag, 1992.
Trimborn and Vega 1935
Trimborn, Hermann, and Pilar
Vega. Catálogo de la Exposición
Arte Inca (Colección J.[uan]
L.[arrea]). Madrid: Imprenta
Martosa, 1935.
Trout 2006
Trout, Keeper. “This is Not Coca,
with Discussion of Lime and Lye.”
Entheogen Review 15, no. 4 (2006):
123, 137–38. Online edition: www.
erowid.org/plants/coca/coca_arti-
cle1.shtml.
Tsunoyama 1979
Tsunoyama, Yukihiro. Textiles of
the Andes: Catalogue of Amano
Collection. South San Francisco:
Heian International, 1979.
Tsunoyama 1977
. Textiles of the Pre-Incaic
Period: Catalogue of Amano
Collection. Kyoto: Dohosha, 1977.
Tulchin 1997
Tulchin, Lisa P. “Lateral Distortion
and Wari Design Construction
in Staff-Bearer Figure Tapestry
Tunics.” MA thesis, Emory
University, 1997.
Tung 2008
Tung, Tiffiny A. “Dismembering
Bodies for Display: A Bioarchaeo­
logical Study of Trophy Heads
from the Wari Site of Conchopata,
Peru.” American Journal of
Physical Anthropology 136 (2008),
294–308.
Tung 2007
. “The Village of Beringa
at the Periphery of the Wari
Empire: A Site Overview and New
Radiocarbon Dates. Andean Past 8
(2007): 254–86.
Tung and Knudson 2008
Tung, Tiffiny, and K. J. Knudson.
“Social Identities and Geogra­
phical Origins of Huari Trophy
Heads from Conchopata, Peru.”
Current Anthropology 49 (2008):
915–25.
Tung and Owen 2006
Tung, Tiffiny, and Bruce Owen.
“Violence and Rural Lifeways at
Two Peripheral Wari Sites in the
Majes Valley of Southern Peru.”
In W. Isbell and Silverman 2006,
435–67.
Tung et al. 2007
Tung, Tiffiny, Martha Cabrera
Romero, and José Ochatoma
Paravicino. “Cabezas Trofeos Wari:
Rituales del cuerpo en el recinto
ceremonial en ‘D’ de Conchopata.”
Revista de Investigación Nacional
de San Cristóbal de Huamanga
(Ayacucho) 15, no. 2 (2007): 216–27.
Turner 1977
Turner, Victor. “Sacrifice as Quin­
tessential Process, Prophylaxis
or Abandonment?” History of
Religions 16, no. 3 (1977): 189–215.
Ubbelohde-Doering 1966
Ubbelohde-Doering, Heinrich.
On the Royal Highways of the
Inca: Archaeological Treasures of
Ancient Peru. New York: Frederick
A. Praeger, 1966.
Uceda and Morales 2010
Uceda, Santiago, and Ricardo
Morales. Moche: Pasado y Presente.
Trujillo: Patronato Huacas del
Valle de Moche, 2010.
Uhle [1903] 1991
Uhle, Max. Pachacamac: A Reprint
of the 1903 Edition by Max Uhle.
Introduction by Izumi Shimada.
University Museum Monograph 62.
Philadelphia: University Museum
of Archaeology and Anthropology,
University of Pennsylvania, 1991.
Uhle 1915
. “Las ruinas de Moche.”
Boletín de la Sociedad Geográfica
de Lima 30, no. 3–4 (1915): 57–71.
Uhle 1903
. Pachacamac. Report of the
William Pepper, M.D., LL.D.,
Peruvian Expedition of 1896.
Philadelphia: University of
Pennsylvania, 1903.
Ulloa 2001
Ulloa, Liliana T. “Textiles prehis-
pánicos y coloniales.” MASMA
WEB, Museo Arqueológico
San Miguel de Azapa. Arica:
Universidad de Tarapacá, 2001,
http://www.uta.cl/masma/
patri_edu/textiles.htm.
Ulloa 1985
. Arica, diez mil años.
Santiago: Museo Chileno de Arte
Precolombino, 1985.
Urbano 1981
Urbano, Henrique-Osvaldo.
Wiracocha Ayar. Héroes y Fun­
ciones en las Sociedades Andinas.
Cuzco: Centro de Estudios Rurales
Andinos “Bartolomé de las Casas,”
1981.
Urbano 1980
. “Dios Yaya, Dios Churí y
Dios Espíritu: modelos trinitarios y
arqueología mental en los Andes.”
Journal of Latin American Lore 6,
no. 1 (1980): 111–27.
Urbano 1978
. “La symbolique de l’espace
andin.” In Proceedings of the
42nd International Congress of
Americanists, 4:335–45. Paris:
Musée de l’Homme/Société des
Americanistes de Paris, 1978.
Urton 2008
Urton, Gary. “The Inca Khipu:
Knotted-Cord Record Keeping in
the Andes.” In Silverman and W.
Isbell 2008, 831–43.
Urton 2003a
. Quipu: Contar Anudando en
el Imperio Inka/Quipu: Knotting
Accounts in the Inka Empire.
Santiago: Museo Chileno de Arte
Precolombino; Cambridge, MA:
Harvard University, 2003.
Urton 2003b
. Signs of the Inka Khipu:
Binary Coding in the Andean
Knotted-String Records. Austin:
University of Texas Press, 2003.
Urton 1996
. “The Body of Meaning in
Chavín Art.” RES: Anthropology
and Aesthetics, no. 29/30 (1996):
237–55.
Urton 1993
. “Moieties and Ceremoni­
alism in the Andes: The Ritual
Battles of the Carnival Season in
Southern Peru.” In Millones and
Onuki 1993, 117–42.
Urton 1985
, ed. Animal Myths and
Metaphors in South America.
Salt Lake City: University of Utah
Press, 1985.
Urton 1981
. At the Crossroads of the
Earth and the Sky: An Andean
Cosmology. Austin: University of
Texas Press, 1981.
Urton and Brezine 2011
Urton, Gary, and Carrie J. Brezine.
“Khipu Typologies.” In Boone and
Urton 2011, 319–52.
Valcárcel 1980
Valcárcel, Luis E. “La religion
incaica.” In J. Mejía Baca, ed.,
Historia del Perú, 3:73–202. Lima:
Editorial Juan Mejía Baca, 1980.
Valcárcel 1959
. “Símbolos mágico-religiosos
en la cultura andina.” Revista del
Museo Nacional (Lima), no. 28
(1959): 3–18.
Valcárcel 1933
. “Esculturas de Pikillajta.”
Revista del Museo Nacional
(Lima), no. 2 (1933): 21–35.
Valdez et al. 2006
Valdez, Lidio M., Katrina J.
Bettcher, José A. Ochatoma
Paravicino, and J. Ernesto Valdez.
“Mortuary Preference and Selected
References: A Comment on Middle
Horizon Wari Burials.” World
Archaeology 38, no. 4 (2006):
672–89.
29 5 R eference list
Valdez et al. 2002
Valdez, Lidio M., Katrina J.
Bettcher, and J. Ernesto Valdez.
“New Wari Mortuary Structures
in the Ayacucho Valley, Peru.”
Journal of Archaeological
Research, no. 58 (2002): 389–407.
Valencia Zegarra 2005
Valencia Zegarra, Alfredo. “Wari
Hydraulic Works in the Lucre
Basin.” In McEwan 2005c, 85–97.
VanStan 1967
VanStan, Ina. Textiles from
beneath the Temple of
Pachacamac, Peru: A Part
of the Uhle Collection of the
University Museum, University
of Pennsylvania. Philadelphia:
University Museum, 1967.
VanStan 1966
. The Fabrics of Peru. Leigh-
on-Sea, Eng.: F. Lewis, 1966.
VanStan 1963
. “A Problematic Example of
Peruvian Resist-dyeing.” American
Antiquity 29, no. 2 (1963): 166–73.
VanStan 1961
. “Ancient Peruvian Textile
Arts: Patchwork and Tie-dye
from Pachacamac.” Expedition
(Philadelphia) 3, no. 4 (1961):
34–37.
Vaughn et al. 2005
Vaughn, Kevin J., Dennis Ogburn,
and Christina A. Conlee, eds.
Foundations of Power in the
Prehispanic Andes. Archaeological
Papers of the American
Anthropological Association
14. Arlington, VA: American
Anthropological Association,
2005.
Vega [1609] 1987
Vega, Garcilaso de la. Royal
Commentaries of the Incas and
General History of Peru, trans.
by Harold V. Livermore. 2 vols.
Paperback ed. Austin: University
of Texas Press, 1987.
Vega [1609] 1966
. Royal Commentaries of
the Incas, and General History of
Peru: Part One, trans. by Harold
V. Livermore. 2 vols. Austin:
University of Texas Press, 1966.
Verano 1995
Verano, John. “Where Do They
Rest? The Treatment of Human
Offerings and Trophies in Ancient
Peru.” In Dillehay 1995, 189–227.
Vranich et al. 2005
Vranich, Alexei, Paul Harmon,
and Chris Knutson. “Reed Boats
and Experimental Archaeology on
Lake Titicaca.” Expedition 47, no. 2
(2005): 20–27.
Wagner 1981
Wagner, Lida. “Information
Exchange as Seen in the Middle
Horizon: Two Ceramics from the
Site of Huari, Peru.” PhD diss.,
University of Wisconsin, 1981.
Watanabe 2002
Watanabe, Shinya. “Wari y
Cajamarca.” In Kaulicke and W.
Isbell 2002, 531–41.
Whittick 1984
Whittick, Arnold. “Towards
Precise Distinctions of Art and
Craft.” British Journal of Aesthetics
24, no. 1 (1984): 47–52.
Wiessner 2001
Wiessner, Polly. “Of Feasting and
Value: Enga Feasts in a Historical
Perspective (Papua New Guinea).”
In Dietler and Hayden 2001b,
115–43.
Willey 1953
Willey, Gordon. Prehistoric
Settlement Patterns in the
Viru Valley, Peru. Bureau of
American Ethnology Bulletin 155.
Washington, DC: Smithsonian
Institution, 1953.
Williams 2006
Williams, Patrick Ryan. “Agricul­
tural Innovation, Intensification,
and Sociopolitical Development:
The Case of Highland Irrigation
Agriculture on the Pacific Andean
Watersheds.” In J. Marcus and
C. Stanish, eds., Agricultural
Strategies, 309–33. Los Angeles:
Cotsen Institute of Archaeology
Press, 2006.
Williams 2003
. “Hydraulic Landscapes and
Social Conflict in Middle Horizon
Peru.” In M. Forte and P. R.
Williams, eds., The Reconstruction
of Archaeological Landscapes
through Digital Technologies. BAR
International Series S1151. Oxford:
B.A.R., 2003.
Williams 2002
. “A Re-examination of
Disaster Induced Collapse in the
Case of the Andean Highland
States: Wari and Tiwanaku.” World
Archaeology 33, no. 3 (2002):
361–74.
Williams 2001
. “Cerro Baúl: A Wari Center
on the Tiwanaku Frontier.” Latin
American Antiquity 12, no. 1
(2001): 67–83.
Williams and Isla Cuadrado 2002
Williams, Patrick Ryan, and Johny
Isla Cuadrado. “Excavaciones
arqueológicas en Cerro Baúl,
un enclave Wari en el valle de
Moquegua.” Gaceta Arqueológica
Andina, no. 26 (2002): 87–120.
Williams and Nash 2006
Williams, Patrick Ryan, and Donna
J. Nash. “Sighting the Apu: A GIS
Analysis of Wari Imperialism and
the Worship of Mountain Peaks.”
World Archaeology 38, no. 3 (2006):
455–68.
Williams and Nash 2002
. “Imperial Interaction in
the Andes: Wari and Tiwanaku
at Cerro Baúl.” In W. Isbell and
Silverman 2002, 243–65.
Williams et al. 2008
Williams, Patrick Ryan, Donna
J. Nash, Michael Moseley, Susan
deFrance, Mario Ruales, Ana
Miranda, and David Goldstein.
“Encuentros en el Reino Wari.”
In P. Kaulicke, ed., Encuentros:
Identidad, Poder y Agencia de
Espacios Publicos, 207–32. Boletín
de Arqueología PUCP 9. Lima:
Fondo Editorial de la Pontificia
Universidad Católica del Perú,
2008 [volume date 2005].
Williams et al. 2005
Williams, Patrick Ryan, David K.
Keefer, Donna J. Nash, Kenneth
Sims, Christopher Dayton, and
Michael Moseley. “Envisioning
the Invisible: Reconstructing the
Taphonomy of Ancient Andean
Agricultural Landscapes through
GIS Slope Stability Models.” In
M. Forte, ed., The Reconstruction
of Archaeological Landscapes
through Digital Technologies II,
53–59. BAR International Series
1379. Oxford: B.A.R., 2005.
Williams et al. 2002
Williams, Patrick Ryan, Michael
E. Moseley, and Donna J. Nash.
“Empires of the Andes: A
Majestic Frontier Outpost Chose
Cooperation over War.” Scientific
American Discovering Archaeology
(March/April 2002): 69–73.
Wilson 1988
Wilson, David J. Prehispanic
Settlement Patterns in the Lower
Santa Valley, Peru. Washington,
DC: Smithsonian Institution Press,
1988.
Wilson and Rathje 2001
Wilson, Douglas C., and William L.
Rathje. “Garbage and the Modern
American Feast.” In Dietler and
Hayden 2001b, 404–21.
Woolf 2001
Woolf, Greg D. “Inventing Empire
in Ancient Rome.” In Alcock et al.
2001, 311–22.
Young-Sánchez 2009
Young-Sánchez, Margaret, ed.
Tiwanaku: Papers from the 2005
Mayer Center Symposium at the
Denver Art Museum. Denver:
Denver Art Museum, 2009.
Young-Sánchez 2004a
. “The Art of Tiwanaku.” In
Young-Sánchez 2004d, 24–69.
Young-Sánchez 2004b
. “Ornamented Gold Plume.”
In Young-Sánchez 2004d, 94–95.
Young-Sánchez 2004c
. “Pucara Stone Sculpture.”
In Young-Sánchez 2004d, 76–78.
29 6 R eference list
Young-Sánchez 2004d
, ed. Tiwanaku: Ancestors
of the Inca. Lincoln/London:
University of Nebraska Press;
Denver: Denver Art Museum, 2004.
Young-Sánchez 2000
. “Textiles from Peru’s
Central Coast, 750–1100: The
Reiss and Stübel Collection from
Ancón.” PhD thesis, Columbia
University, 2000. and the Calendar.” In Young-
Zapata 1997
Zapata, Julinho. “Arquitectura
y Contextos Funerarios Wari en
Batan Urqu, Cusco.” In Kaulicke
1997, 165–206.
Zuidema 2011a
Zuidema, R. Tom. “Chuquibamba
Textiles and Their Interacting
Systems of Notation: The Case
of Multiple Exact Calendars.” In
Boone and Urton 2011, 251–75.
Zuidema 2011b
. “Royal Tunics: Inca, Huari
and Tiahuanaco.” Lecture deliv-
ered at State University of New
York at Binghamton, October 2011.
Zuidema 2010
. El Calendario Inca: Tiempo
y Espacio en la Organización
Ritual del Cuzco, La Idea del
Pasado. Lima: Fondo Editorial del
Congreso del Perú/Fondo Editorial
de la Pontificia Universidad
Católica del Perú, 2010.
Zuidema 2009
. “Tiwanaku Iconography
Sánchez 2009, 83–100.
Zuidema 1992
. “Guaman Poma and the
Art of Empire.” In K. J. Andrien
and R. Adorno, eds., Transatlantic
Encounters: Europeans and
Andeans in the Sixteenth Century,
151–202. Berkeley: University of
California Press, 1992.
Zuidema 1990
. “Ceques and Chapas: An
Andean Pattern of Land Partition
in the Modern Valley of Cuzco.” In
B. Illius and M. Laubscher, eds.,
Circumpacífica: Festschrift für
Thomas S. Barthel, 627–43. Bern:
Sonderdruck, 1990.
Zuidema 1989
. “El Origen del Imperio
Inca.” In Zuidema and Burga 1989,
193–218.
Zuidema 1977
. “The Inca Calendar.” In
A. Aveni, ed., Native American
Astronomy, 219–60. Austin:
University of Texas Press, 1977.
Zuidema 1973
. “Kinship and Ancestor
Worship in Three Peruvian
Communities. Hernandez
Príncipe’s Account of 1622.”
Bulletin de l’Institut Français
d’Études Andines 2, no. 1 (1973):
16–33.
Zuidema 1964
. The Ceque System of Cuzco:
The Social Organization of the
Capital of the Inca, trans. by Eva
M. Hooykaas. Internationales
Archiv für Ethnographie 50, sup-
plement. Leiden: E. J. Brill, 1964.
Zuidema and Burga 1989
Zuidema, R. Tom, and Manuel
C. Burga, eds. Reyes y Guerreros:
Ensayos de Cultura Andina. Lima:
Fomciencias/Talleres Gráficos P. L.
Villanueva, 1989.
Figure 251 (detail of fig. 161).
Panel; camelid fiber and cotton;
77 x 109.5 cm. Royal Ontario
Museum, Toronto, 931.11.1.