Beyond Wari Walls : Regional Perspectives on Middle Horizon Peru [1 ed.] 9780826348692, 9780826348678

Citation preview

Latin America / Archaeology

Wari civilization swept across the central Andes during the Middle Horizon (A.D. 600–1000). The nature and importance of this civilization have long been the connections between the local, regional, and interregional changes that occurred during this period. This work needs to be done at Huari and the various Wari

Justin Jennings is associate curator of

debated by archaeologists. Many have viewed Wari as an empire governed by people living at the site of Huari in the central highlands of Peru. Some scholars, however,

New World Archaeology in

have argued that the spread of Wari artifacts, architecture, and influence can be

the Royal ontario Museum

explained by other kinds of interregional interactions.

and assistant professor of

The scholars whose work is assembled here attempt to better understand the

settlements found throughout Peru, but it

anthropology at the

nature of Wari by examining its impact beyond Wari walls. By studying a village in

also needs to be done beyond Wari walls.”

University of Toronto.

Cuzco, a water shrine in Huamachuco, or a compound on the Central Coast, these

—From the Introduction

authors provide information that cannot be gleaned either from digs around the city of Huari or work at the major Wari installations in the periphery. This book provides

Beyond Wari Walls

oooooooooo Beyond Wari Walls

Regional Perspectives on Middle Horizon Peru

“Beyond Wari Walls broadens our understanding of Wari by viewing the Wari phenomenon from a variety of different regional perspectives. As more archaeological research is being conducted in Peru, a more complex picture of the Middle Horizon is emerging. The pioneers of Peruvian archaeology were right in

no definitive answers to questions about the Wari, but it contributes to broader

identifying the period as a turning point

early cities and states throughout the world.

in the country’s prehistory. The rapid

“A broad examination of archaeological evidence regarding the nature, organization and regional importance of the Wari tradition. A provocative discussion of regional variations and their relevance for reconstructing Wari, which in turn could serve as a model for reexamining the material manifestations of other ‘empires’ in the Andes and elsewhere.” —Jerry D. Moore,

jennings

debates about interregional influences and interaction during the emergence of

urbanization of Huari forever changed the cultural landscape of the Andes. Yet, the relationship between the city and the widespread interactions that it helped generate is not as straightforward as these archaeologists first thought. To understand

professor and chair of anthropology at

what happened in Peru during the Middle

California State University, Dominguez Hills

jacket photograph: Funerary bundle B, Tomb 1–2, from the Uhle Cemetery in front of the Painted Temple (also known as Pachacamac Temple). Photograph courtesy of Izumi Shimada. jacket illustrations: Ceramic artifacts from Cerro Amaru burial structure. Illustrations courtesy of the Huamachuco Archaeological Project. jacket design: Melissa Tandysh

University of NeW MexICo Press unmpress.com 800-249-7737

Horizon, we need to continue to research

—continued on back flap

isbn 978-0-8263-4867-8

ËxHSKIMGy348678zv*:+:!:+:!

edited by

justin jennings

o Beyond Wari Walls

o

Beyond Wari Walls Regional Perspectives on Middle Horizon Peru E di t ed b y Just in Jen n in gs

University of New Mexico Press Albuquerque

© 2010 by the University of New Mexico Press All rights reserved. Published 2010 Printed in the United States of America 15 14 13 12 11 10    1 2 3 4 5 6 Library of Congress Cataloging-in-Publication Data

Beyond Wari walls : regional perspectives on Middle Horizon Peru / edited by Justin Jennings. p. cm. Includes index. isbn 978-0-8263-4867-8 (cloth : alk. paper) 1. Huari Indians—Politics and government. 2. Huari Indians—Material culture. 3. Huari Indians—Antiquities. 4. Culture diffusion—Peru—History. I. Jennings, Justin. f3430.1.h83b49 2010 985´.01—dc22 2010025871

Design and Composition: Deborah Flynn Post Text composed in Garamond Premier Pro Regular 10.5/14

s Contents

Figures  vii Ta b l e s  

ix

Chapter 1

Beyond Wari Walls  1 Just i n Jen n i n gs Ch a p t e r 2

The Nature of Wari Presence in the Mid–Moquegua Valley: Investigating Contact at Cerro Trapiche  19 Ulr i k e M at t h i es Gr een a n d Pau l S . G o ldst ei n Ch a p t e r 3

Becoming Wari: Globalization and the Role of the Wari State in the Cotahuasi Valley of Southern Peru  37 Just i n Jen n i n gs Ch a p t e r 4

Wari in the Majes-Camaná Valley: A Different Kind of Horizon  57 B ru c e O w en Ch a p t e r 5

Local Settlement Continuity and Wari Impact in Middle Horizon Cusco  79 Véro n i q u e B élis le a n d R . A l a n C ov ey

Ch a p t e r 6

Nasca and Wari: Local Opportunism and Colonial Ties during the Middle Horizon  96 Ch r ist i na A . C o n lee Ch a p t e r 7

The Wari Footprint on the Central Coast: A View from Cajamarquilla and Pachacamac  113 R a fa el Segu r a L l a n os a n d Izum i Sh i m a da Ch a p t e r 8

What Role Did Wari Play in the Lima Political Economy?: The Peruvian Central Coast at the Beginning of the Middle Horizon  136 Gi a n c a r l o M a rco n e F. Ch a p t e r 9

The Wari State, Its Use of Ancestors, Rural Hinterland, and Agricultural Infrastructure  155 Fr a n k Med d en s a n d Ni c h o l a s B r a n c h Ch a p t e r 10

Piecing Together the Middle: The Middle Horizon in the Norte Chico  171 K i t Nelso n , Nat h a n Cr a i g , a n d M a n u el P er a les Ch a p t e r 11

Contextualizing the Wari-Huamachuco Relationship  188 Th er es a L a n ge To p i c a n d Jo h n R . To p i c Ch a p t e r 12

Moche and Wari during the Middle Horizon on the North Coast of Peru  213 Claude Chapdelaine Ch a p t e r 13

Agency, Identity, and Control: Understanding Wari Space and Power  233 Wi lli a m H . Is b ell Contribu tor s 255 Inde x 259

vi

s Figures

Figure 1.1  Examples of ceramic styles from Wari heartland 4 Figure 1.2  Examples of ceramic styles from Wari heartland 4 Figure 1.3  Map of Peru showing sites discussed in the chapter 6 Figure 1.4  Orthogonal cellular architecture at the Moraduchayuq Compound at Huari 6 Figure 1.5  Site plan of Pikillacta 8 Figure 1.6  Site plans of Honcopampa and Acachiwa 9 Figure 2.1  The Moquegua frontier in the Middle Horizon 21 Figure 2.2  Settlement distribution in the Moquegua Valley 22 Figure 2.3  Air view and map of Cerro Trapiche 26 Figure 2.4  Plan view of Sector C 27 Figure 2.5  Plan view of 2003 excavations in Structure 3, Sector C 27 Figure 2.6  Ceramic distributions in excavations of the southern room in Structure 28 Figure 2.7  Wari ceramics from excavation of Structure 3 28 Figure 2.8  Wari and Huaracane ceramics from 1998 and 2004 surface collection 29 Figure 3.1  Model for ancient globalization 40 Figure 3.2  Map of the Cotahuasi Valley 43

Figure 3.3  Wari Viñaque-style designs found on locally made pots 46 Figure 3.4  Examples of postfired incision marks 47 Figure 3.5  Site plan of Tenahaha 49 Figure 3.6  Three ceramic bowls placed as offerings near the feet of an individual in Tomb 3 50 Figure 4.1  The valley of Majes and Camaná in the south-central Andes 58 Figure 4.2  Schematic models of the formation of the Inca, Tiwanaku, and Wari material culture horizons 60 Figure 4.3  Sites with Wari ceramics in the MajesCamaná drainage 62 Figure 4.4  Calibrated radiocarbon dates from Beringa, Sonay, and Toro Grande 65 Figure 4.5  Wari surface sherds from Camaná Valley sites 68 Figure 4.6  Ceramics from Beringa 69 Figure 5.1  The Cusco region 78 Figure 5.2  Late Formative pottery from Ak’awillay 81 Figure 5.3  Middle Horizon pottery styles from the Xaquixaguana Plain and Ak’awillay 82 Figure 5.4  Map of Sacred Valley village sites 83 Figure 5.5  The two clusters of EIP and Middle Horizon settlements in the Xaquixaguana Plain 85 Figure 5.6  The archaeological site of Ak’awillay 87 vii

Figure 5.7  Wari and Araway pottery from Ak’awillay 88 Figure 5.8  House plan and occupation floor of Unit C 88 Figure 5.9  A burial containing two individuals 90 Figure 6.1  Map of the Nazca drainage 99 Figure 6.2  Profile view of Tomb 5 101 Figure 6.3  Artifacts found associated with the elite Middle Horizon tombs at La Tiza 101 Figure 6.4  Examples of Type 2 and 3 tombs 102 Figure 6.5  Local Middle Horizon burial from La Tiza and associated Loro bowl 103 Figure 6.6  Middle Horizon child burial from Pajonal Alto 104 Figure 7.1  Map of the Central Coast 114 Figure 7.2  Map of Villar Córdova Compound of Cajamarquilla 116 Figure 7.3  Flood-damaged Late Lima walls in the Villar Córdova Compound 117 Figure 7.4  Abandoned irrigation canal in the Villar Córdova Compound 118 Figure 7.5  Map of Pachacamac 120 Figure 7.6  Viñaque/Atarco ceramic fragment from the Pilgrims’ Plaza 122 Figure 7.7  Diatom distributions from the Urpi Kocha Lagoon 124 Figure 7.8  Funerary bundle B, Tomb 1–2, from the Uhle Cemetery 128 Figure 7.9  Decorated kero-type cup from burial AG, Uhle Cemetery 128 Figure 8.1  Map of the Central Coast during the Early Intermediate Period 139 Figure 8.2  Aerial photo of the Little Mud-Brick Compound at Pachacamac 145 Figure 8.3  The Little Mud-Brick Compound 145 Figure 8.4  Schematic profile of the Little Mud-Brick Compound 146 Figure 8.5  Hearth adjacent to ramp in Little Mud-Brick Compound 147 Figure 8.6  Sample of plates and dish sherds recovered from the Little Mud-Brick Compound 147 Figure 9.1  Middle Horizon Epoch 2 site distribution in the Chicha-Soras Valley 156 Figure 9.2  Plan of Yako 156 viii 

Figure 9.3  D-shaped structure at the site of Yako 157 Figure 9.4  Isometric reconstruction of the central sector of Chiqnajota 158 Figure 9.5  Facade of the Charrangochayoc tomb 159 Figure 9.6  Selected pollen taxa diagram for Ayapampa peat bog 160 Figure 9.7  Middle Horizon Epoch 3–4 site distribution in the Chicha-Soras Valley 162 Figure 9.8  Cruz Muqu burial population 164 Figure 10.1  Map of the Norte Chico 172 Figure 10.2  Plan map of the site of PV39-211 located in the Pativilca Valley 175 Figure 10.3  Plan map of the site of PV39-212 located in the Pativilca Valley 175 Figure 10.4  Cluster of architecture at the upper end of the quebrada at Caldera, Huaura Valley 177 Figure 10.5  The archaeological site of Caldera 177 Figure 10.6  Impressed Middle Horizon pottery from Luriama, Huaura Valley 180 Figure 10.7  Middle Horizon polychrome pottery from the Huaura Valley 181 Figure 11.1  Map of the Huamachuco area of northern Peru showing location of major sites 192 Figure 11.2   Map of Viracochapampa 196 Figure 11.3  Construction details from Viracochapampa 197 Figure 11.4  More construction details from Viracochapampa 198 Figure 11.5  Viracochapampa gallery with septal wall cleared prior to excavation 199 Figure 11.6  Reconstruction drawing of Viracochapampa patio group 199 Figure 11.7  Ceramic artifacts from Cerro Amaru burial structure 203 Figure 12.1  Location of Moche sites on the North Coast of Peru 214 Figure 12.2  Important sites of the lower Santa Valley 221 Figure 12.3  Polychrome ceramic fragments from Huaca Chica, Guadalupito 222 Figure 12.4  Polychrome kero fragments from Huaca Chica, Guadalupito 222 Figure 12.5  Textile of the Moche-Wari tradition from El Castillo de Santa 226

s Tables

Table 2.1  Radiocarbon dates from the Moquegua Valley 24 Table 2.2  Ceramic distributions of surface collections in Sectors C, D, E, and F

28

Table 3.1  Phases of Wari globalization

41

Table 3.2  Radiocarbon dates from Middle Horizon contexts at Tenahaha 45 Table 4.1  Ceramic patterning in Inca, Tiwanaku, and Wari material culture horizons

59

Table 4.2  Sites with Wari ceramics in the valley of Majes and Camaná

63

Table 4.3  Alternative hypotheses of Middle Horizon social processes

71

Table 6.1  Middle Horizon looted tombs excavated at La Tiza Table 6.2  Burial and strontium isotopic information from La Tiza and Pajonal Alto Table 8.1  Comparison of formal categories between Cajamarquilla and the Little Mud-Brick Compound Table 11.1  Radiocarbon dates from the Huamachuco area Table 12.1  Ceramic sequence and relative chronology for the Northern and Southern Moche Table 12.2  Radiocarbon dates from Moche sites of the Santa Valley

100 108

147 190

215 218

  ix

 

Ch a p t er 1

o Beyond Wari Walls Justin Jennings

T

he Middle Horizon (AD 600–1000) was a significant turning point in Andean prehistory. For the first time, a large swath of the Central Andes was linked together through interregional exchange and shared cultural influence. These interactions were anchored by two great cities that shared basic elements of their religious cosmology—Tiahuanaco and Huari.1 The relationship between these cities and the widespread cultural phenomena that developed around them (the Tiwanaku and Wari civilizations) have been a subject of enduring debate among Andean archaeologists (Glowacki 1996:26–56; Isbell and McEwan 1991:5–10; Janusek 2008:8–17). Tiahuanaco, located in Bolivia just to the south of Lake Titicaca, was an important pilgrimage site centered on a monumental core of sculptures, sunken courtyards, and platforms. Although considered the capital of an extensive empire by many until the 1980s, most scholars now suggest that Tiahuanaco’s political control, with a few notable exceptions, did not extend much past the lake basin (Janusek 2008:23). Instead, Tiwanaku and Tiwanaku-inspired material culture spread through the colonization of a few areas, the city’s extensive trade network, and the broad appeal of its cosmology (Browman 1997; Goldstein 2004; Owen 2005; Stanish 2002).

The debate on the spread of the Wari civilization is less settled. Located in the Ayacucho basin in the central highlands of Peru, the city was a sprawling metropolis of homes, workshops, and temples. Like Tiahuanaco, Huari was long considered to be the capital of an expansive empire. Yet unlike Tiahaunaco, there is more compelling evidence to make that claim. The spread of Wari material culture throughout Peru was more pervasive than that of Tiwanaku in the south-central Andes, and the construction of sites with Wari architecture, and possibly a Wari road network, has been used to argue for the creation of an extensive imperial administrative system (e.g., Isbell 1991a; Lumbreras 1974a; Schreiber 1992; Williams and Isla 2002). Other scholars, however, have argued that the spread of Wari artifacts, architecture, and influence can be explained by other kinds of interregional interactions (e.g., Bawden and Conrad 1982; Jennings 2006a; Shady Solís 1982; J. Topic 1991; Topic and Topic 2001). Like many debates within archaeology, the argument over the role of Wari culture in the Middle Horizon is fueled by a variety of factors ranging from where the archaeologist went to school to the archaeological features that he or she finds most important (Isbell and McEwan 1991:5). The only way to resolve this debate is through more fieldwork. Thankfully, there has been a

  1

significant increase in Wari studies since the collapse of the Sendero Luminoso guerrilla movement in 1992. The bulk of these studies has continued to take place in Huari, sites in the Wari heartland of Ayacucho, or in the handful of sites with Wari architecture located throughout Peru. This makes sense—if you want to study Wari, then you should begin by studying Wari sites. Yet, the Wari phenomenon was more than the sum total of its most archetypal sites. To better understand the role of Wari in the Middle Horizon, we also need to move beyond Wari walls. Over the last few years, the expansionary dynamics of the earliest civilizations in the world have been critically reevaluated (e.g., Algaze 1993; Cusick 1998; Schwartz and Falconer 1994; Sherratt 2004; Stein 1999, 2005). This work has shown that “views from the countryside” are necessary if we hope to untangle the role that early cities like Uruk, Harappa, and Cahokia played in wider regions (e.g., Schwartz and Falconer 1994). This book moves beyond the shadows of Wari centers by bringing together seventeen authors who have conducted research on the Middle Horizon throughout Peru. The authors range from Ph.D. students fresh from groundbreaking excavation and survey work to senior scholars who have struggled with the idea of Wari for decades. These writers attempt to understand the nature of Wari through the impact of the civilization in their particular region. By studying Wari from such places as a village in Cuzco (Bélisle and Covey), a water shrine in Huamachuco (Topic and Topic), and a compound on the Central Coast (Marcone), these authors provide us with a glimpse of how Wari was experienced in the Middle Horizon, which cannot be gleaned from digs at Huari or one of its peripheral settlements. This book provides no definitive answers to the questions of Wari phenomena. While some of the authors find that their evidence supports the empire model, other authors question the validity of this model and offer alternative explanations. In this introductory chapter, I first describe the city and the spread of Wari sites, artifacts, and influence throughout much of Peru, then discuss various interpretations of Wari and the Middle Horizon, and finally introduce the chapters in 2 

J ustin J ennings

this volume. We are in the midst of an exciting time in Wari studies when old models of the Middle Horizon are being reevaluated and new models are being offered (e.g., Jennings 2006a; Topic and Topic 2001, this volume). In this book, we hope to contribute both to this dialogue and to the broader debate about the nature of the interregional interaction that occurred during the emergence of early cities and states throughout the world.

Understanding Wari Although the Spanish conquistador Pedro Cieza de León visited the highland site of Huari in 1548 (1984[1553]:249 [f. 114]), the first glimpse of Huari’s significant role in Peruvian history actually came in coastal excavations just outside of Lima in 1896. At the site of Pachacamac, Max Uhle uncovered graves with pottery, textiles, and wood carvings that were reminiscent of a style from Tiahuanaco that he had helped define several years earlier (Stübel and Uhle 1892; Uhle 1991[1903]:26–32). Although Uhle believed that some of the material belonged to the Tiahuanaco style, most was in a style that he called Epigone, which was closely related to Tiahuanaco but “inferior to its famous prototype in almost every respect” (Uhle 1991[1903]:26). He interpreted the Pachacamac material as evidence for the spread of a religious cult throughout the Central Andes (1991[1903]:48), though he remained unsure of the temporal and stylistic relationship between the Pachacamac and Tiahuanaco material (1991[1903]:22–32). By 1913, Uhle’s survey of other research being done in the Andes, combined with his own finds in other places along the coast of Peru, convinced him that there was a Tiwanaku horizon that marked the spread of an earlier civilization from the site of Tiahuanaco (Uhle 1913:341). The Tiwanaku horizon described by Max Uhle was investigated by a second generation of archaeologists like Alfred Kroeber, William Strong, John Corbett, and Rafael Larco Hoyle (Schaedel 1993:227). These scholars confirmed the spread of Tiwanakuinfluenced material across the coast of Peru, and the style became widely known as “Coast Tiahuanaco” (Kroeber 1930). While embracing the idea of a widespread civilization, these scholars also shared Uhle’s

earlier unease about the significant differences between the Coast Tiahuanaco style and the material coming from the site of Tiahuanaco. The search for a Peruvian source for the Coast Tiahuanaco style led Julio C. Tello to the site of Huari in the central highlands. His excavations in 1931 revealed ceramics that closely matched those found at Pachacamac and other sites. He would later declare that Huari was the source of the style, as well as the capital of a Wari civilization (Tello 1942:682–684). Since the site was virtually unknown before Tello’s work, his assertion was initially greeted with some skepticism. A series of site visits by other scholars (Kroeber 1944:115; Rowe et al. 1950; Willey 1945:55) and excavations at Huari by Wendell Bennett (1953) dispelled this skepticism, and, by the 1950s, Huari was broadly recognized as a monumental city that had been on par with Tiahuanaco in terms of its size and cultural importance. In 1956 John Rowe felt confident enough in the growing evidence of a Wari civilization to semantically link the Inca and Wari cultures together in a new Peruvian chronology. The later periods of prehistory in his chronology unfolded in an orderly sequence of horizons and intermediate periods with the horizon periods delineating eras of widespread civilizations and the intermediate periods marked by more regional fractionalization. The Late Horizon (AD 1476–1532) was the era of the Inca Empire, and the Middle Horizon was the period of Wari expansion (Rowe 1956:628, 1962:40). Although there was some resistance to the implicit assumptions associated with the use of the term Middle Horizon to describe the era (Schaedel 1993:225–227), Rowe’s chronology has become the standard one used in the region today.

The Wari Style The first decades of study on the Wari style were marked by confusion. When Uhle dug at Pachacamac, he illustrated ceramics and textiles in the “Epigone style” and identified some common motifs like the front-faced deity and condor (Uhle 1991[1903]:26–32). He sent many of the vessels that he found to California, where these pieces were further studied and contrasted in

greater detail with the related Nasca and Tiahuanaco styles (Kroeber 1930; Rowe et al. 1950; Willey 1948).2 At the same time, however, scholars were also noting differences within the “Coast Tiahuanaco” style, and suggestions were being made for names of regional styles under the broader umbrella of Coast Tiahuanaco (e.g., Larco 1948). By the time of Wari’s broad acceptance in the 1950s, there was considerable uncertainty in both terminology (people were using “Epigone,” “Coast Tiahuanaco,” and other names for Wari wares) and in the temporal and regional relationships between the various Wari substyles. Beginning in 1958, Dorothy Menzel began an exhaustive project to define distinct Wari ceramic styles and place these styles into a more precise chronology for the Middle Horizon. The result of this work was a Middle Horizon divided into four phases—Middle Horizon 1–4—with the first two phases related to the Wari culture and divided into A and B subphases (Menzel 1964, 1968, 1977). Menzel suggested that the Wari style emerged in Ayacucho in Middle Horizon 1A, and then spread across Peru during Middle Horizon 1B (1964:68). The principal Middle Horizon 1A styles from Ayacucho, Conchopata, Chakipampa, Ocros, and Black Decorated, developed out of local and Nasca antecedents and contained religiously charged images (also see Knobloch 1983) (Figure 1.1). By Middle Horizon 1B, Wari styles, especially Chakipampa B, were widespread throughout Peru. For Menzel, the Middle Horizon 2 was the greatest period of Wari influence, and she saw a proliferation of regional Wari styles occurring during the period (1964:35–36). In the Middle Horizon 2A, the dominant style in the highlands became Viñaque, a style perhaps best represented by depictions of the bodiless heads of the front-faced deity and angels (1964:37) (Figure 1.2). At the same time, regional Wari styles “with local peculiarities” begin to form during this period, like Pachacamac on the Central Coast and Atarco on the South Coast (1964:37). Regional differentiation in Wari styles continued in the Middle Horizon 2B when, according to Menzel, the “empire expanded very rapidly and reached its maximal extent” (1964:70). By Middle Horizon 3, the Wari Empire had collapsed.

B eyond Wari Walls

  3

Figure 1.1  Early Intermediate Period Huarpa (A), Early Middle Horizon Ocros (B), and Late Middle Horizon Huamanga (C) vessels from the Ayacucho region (photos courtesy of William H. Isbell).

Figure 1.2  Late Middle Horizon Wari Viñaque- (A, B, D) and Huamanga-style (C) vessels from the Ayacucho region (photos A, B, and D courtesy of William H. Isbell, and photo C courtesy of Patricia J. Knobloch).

Menzel’s seriation was linked with radiocarbon dates by John Rowe (1962, 1967). Using associations of pots with textiles and other carbon-based material, Rowe used the midpoint of the radiocarbon age determinations to build an (overly) exact chronology (Ketteman 2002:19). He dated the beginnings of the Middle Horizon 1 to AD 605, the Middle Horizon 2 to AD 723, the Middle Horizon 3 to AD 800, and the Middle Horizon 4 to AD 892 (Rowe 1967:24). Menzel’s Wari seriation, as well as Rowe’s chronological placement of the Middle Horizon from AD 600 to 1000, are the standards used today. This does not mean, however, that both are without flaws. The early part of Menzel’s seriation, for example, has been refined by Patricia Knobloch based on her analysis of stratigraphic associations at Huari. She found that Wari styles reflect greater Tiwanaku influence beginning only in Middle Horizon 1B (1983), an assertion that has been confirmed by excavations at Conchopata (Isbell and Knobloch 2006:342–243). Linda Wagner has also raised significant concerns about the separation of Middle Horizon 2 into two subphases (1981). The largest concerns have to do with the relationship between the seriation and the absolute chronology. The beginning of the Middle Horizon, if marked by the introduction of the principal Wari styles, likely occurred closer to AD 650–700, and the end of the Wari stylistic horizon likely occurred between AD 900 and 1000 (Ketteman 2002; Williams 2001). Within the sequence, the later part of the sequence has proved especially problematic. The Middle Horizon 4 does not seem to exist in the highlands (Menzel identified the phase only by examples from Ica and Nazca [1964:65–66]). Middle Horizon 2 styles continued to be used in many places outside of Ayacucho until around AD 1000, and the Middle Horizon 3 seems now to date to a period of abandonment of Wari centers in the heartland from AD 900 to 1000 (Isbell 2001a). In the remainder of the chapter, I will attempt to ease the chronological confusion for those unfamiliar with Menzel’s seriation by dividing the Middle Horizon into just two periods: the Early Middle Horizon, conservatively dated from AD 600 to 800, and a Late Middle Horizon, dating from AD 800 to 1000. Menzel’s Middle Horizon 1 styles fall into my Early Middle Horizon and her

Middle Horizon 2 and 3 styles fall into my Late Middle Horizon. The division is important. Menzel suggested that the Middle Horizon 1–2 transition was a critical one both stylistically and politically for Wari (1964:36). This transition, dating to around AD 800, has been confirmed more recently by other scholars (Ketteman 2002:91–92; Schreiber 2001:89; Williams 2001:80–81) and is associated with significant cultural changes in many of the chapters in this volume.

Huari and Its Environs Located in the Ayacucho Valley (Figure 1.3), Huari was likely the largest prehistoric city ever in Peru. At its height, the site covered almost 15 km2 and housed as many as 70,000 people (Isbell 2001b:106–107; Isbell et al. 1991:24). The site had been one of several competing sites in the previous Early Intermediate Period (200 BC– AD 600), which were vying for ceremonial importance in the area (Isbell 2001b:111–117; Leoni 2006). The feasts, rituals, performances, and other activities that took place at Huari seem to have won over Ayacuchanos by around AD 550 when people in the valley began to abandon rival sites and move to Huari (Isbell 2001b:117). The Early Middle Horizon was a period of rapid urbanization and architectural formalization at Huari. The monumental ritual and funerary complexes of Cheqo Wasi and Vegachayoq Moqo were constructed early in this period (Benavides 1991; Bragayrac 1991; Isbell 2001b:127–135) and created a charismatic core to the city. The construction of these complexes, combined with the creation of iconographically rich Wari ceramics styles and the practices of ancestor worship and trophy head taking (Cook 2001; Ochatoma Paravicino and Cabrera Romero 2002; Tung 2008), likely led to the development of a new religion that both underlined the importance of rituals conducted at Huari and legitimated growing class differences within the city (Menzel 1964:67). Soon after AD 700, Huari was a bustling metropolis at the height of its power and influence (Isbell 1997:186). Around this time, the 1.5 km2 core of the city was largely transformed into a series of great rectangular compounds that housed the city’s growing elite (Isbell et al. 1991). The compounds were built in a

B eyond Wari Walls

  5

1 2

0

3

7

N 0

MN

4 5 6

20 m

8 9

500

10

Kilometers Figure 1.3  Map of Peru showing the following sites discussed in the chapter: (1) Viracochapampa, (2) Honcopampa, (3) Azángaro, (4) Jargampata, (5) Huari, (6) Jincamocco, (7) Pacheco, (8) Pikillacta, (9) Acachiwa, and (10) Cerro Baúl and Cerro Mejía.

Figure 1.4  Orthogonal cellular architecture at the Moraduchayuq Compound at Huari (adapted from Isbell et al. 1991:26).

new architectural style, known as orthogonal cellular (Isbell 1991a), that was composed of repetitive modular units of long corridors and central patios (Figure 1.4). Platforms in these patios were likely used in rituals, and these rituals may have been associated with feasts where large amounts of food and alcohol were consumed (Cook and Glowacki 2003:184–189; Isbell 2001b:151). Like other ancient cities (Storey 2006), there were also areas dedicated to craft production within Huari. Unfortunately, these areas have only been identified by surface collections and remain to be fully excavated (González Carré 1981:94; Spickard 1983:153–154; von Hagen and Morris 1998:130). The valley where Huari sits has limited agricultural productivity because of its elevation (2,500– 3,000 m above sea level). Already massively terraced

and irrigated by the smaller population of the Early Intermediate Period (Isbell 1988:74–75; Lumbreras 1974b), the efforts to intensify production in the basin continued during the Middle Horizon (Ochatoma Paravicino and Cabrera Romero 2001). Maximizing agricultural production within the valley, however, could not keep up with the city’s growing population (González Carré 1981:88; Lumbreras 1974a:163), and there is ample evidence for efforts to increase agricultural yields in surrounding valleys (Browman 1999; Isbell 1977; Raymond 1992; Raymond and Isbell 1969; Vivanco and Valdez 1993). Agricultural intensification seems to have peaked at the beginning of the Late Middle Horizon when new centers, like Jargampata and Azángaro, were built in and around the Ayacucho Basin (Anders 1991; Isbell 1977). These centers mimicked the

6 

J ustin J ennings

orthogonal cellular architecture of Huari and had facilities that were likely built to store crops and other products bound for the city. This area of intensive resource exploitation may have been considerable. The Sondondo and Chicha-Soras valleys lie almost 150 km away from the site of Huari, but both sites provide considerable evidence for Middle Horizon economic transformations that were likely directed by Wari officials (Meddens 1991; Meddens and Branch, this volume; Schreiber 1992). For example, the Wari center of Jincamocco was built in the Sondondo Valley during the Early Middle Horizon and then expanded from 3.5 to 15 ha at the beginning of the Late Middle Horizon (Schreiber 2001:90). Through the construction of extensive terraces, the valley surrounding Jincamocco produced a surplus of maize that could be shipped to Huari via a Middle Horizon road that led out of the valley (Schreiber 1992:149–151, 159–160). Wari cultural influence peaked during Huari’s height from AD 700 to 900. By around AD 900, the city began a century-long decline that seems to have spurred efforts by city officials to attempt to revitalize Huari by organizing a massive construction project. As part of the project, many buildings in the monumental core of the city were abandoned, dismantled, and covered over with a layer of sand. Workers then accumulated a huge pile of building stones across Huari’s core area and likely used the pile as a raised causeway to move across the site. These workers were constructing new compounds that did not follow the previous orthogonal cellular design, but instead were made up of larger trapezoidal buildings. The architectural renovations were never finished. The site, along with others in the Ayacucho Basin, was abandoned by around AD 1000 (Isbell 1997:208–209, 2001b:160–162).

Wari Provincial Sites By the end of the Early Middle Horizon, pottery and textiles from Huari were being exchanged or emulated across much of Peru (Knobloch 1991; Menzel 1964). This interaction was paralleled by more interregional interaction in general during this period as the exchange of metals, obsidian, decorated ceramics, textiles, turquoise, and

specific marine shells (Spondylus and Strombus) increased both in terms of the quantity of items exchanged and the distance over which objects traveled (Burger et al. 2000; Lechtman 1980; Shady Solís 1988). While many imports flowed into Huari (Cabera Romero 1996:88–91; González Carré et al. 1996:100–102; Isbell 1985:70; Pérez Calderón 1995:85–86), other centers like Huamachuco (Topic and Topic, this volume) and Cajamarquilla (Segura and Shimada, this volume) were also very important players in these interactions (Shady Solís 1988). Yet, Huari was the only city that established provincial sites during this period. Scholars have identified more than twenty possible Wari sites outside of the Ayacucho Basin (e.g., Glowacki 2002; Glowacki and McEwan 2001; Isbell 1989; Jennings and Craig 2001; Knobloch 1991; McEwan 1991, 1996; Menzel 1964, 1977; Schreiber 1992, 1999, 2001; Williams and Nash 2002; Williams and Pineda 1985). These sites have been identified based on the existence of Wari or Wari-inspired ceramics, textiles, and/or architecture. Unfortunately, only a handful of these centers have been excavated. While some of these centers have proved to be Wari installations, other reported centers like Pampa de Las Llamas (Pozorski and Pozorski 1987:32), Collota (Jennings, this volume), and Chimú Capac (Valkenier 1995:279) date from other periods. There are also other possible Wari sites that remain known only from air photos or surface collections. Nonetheless, there is strong evidence for Wari installations located hundreds of kilometers from Huari. Some of these sites clearly administered their immediate surroundings. Pikillacta, for example, is a 47 ha site found near Cuzco in the southern highlands (Figure 1.5). The site’s architecture closely follows the Wari orthogonal cellular architecture style, and its ceramics are in the Wari style (Glowacki 1996; McEwan 1991, 1996, 2005). The area around the site was heavily transformed during the Middle Horizon, and these transformations seem to have been directed toward the cultivation of maize, potatoes, and other crops. New canals, reservoirs, and terraces were also constructed, and blocks of small conjoined rooms at Pikillacta were likely designed as storage facilities (Valencia 2005; but McEwan [2005] argues that these rooms were used to house ancestral mummies instead of crops).

B eyond Wari Walls

  7

Figure 1.5  Site plan of Pikillacta (adapted from McEwan 1996:173).

In other places, Wari sites appear to have been constructed on the fringes of established polities. The site of Honcopampa, for example, is a 3 ha site found in the Callejón de Huaylas Valley in the northern highlands (Figure 1.6). The site boasts at least eighteen patio groups and two D-shaped temples, as well as ceramics that are closely related to Wari examples (Isbell 1991b). While the site was perhaps built and occupied by Wari settlers, there is little evidence for Wari control over the valley (Lau 2005, 2006; Ponte 2001; see Isbell, this volume, for his new interpretations of the site). Instead, Wari goods appear to have circulated in the valley as part of a broader regional exchange of prestige goods that were coveted by local elites. Positioned along a major route out of the valley, Honcopampa might best be interpreted as more of a trading post than an administrative center (Schreiber 1999:91). 8 

J ustin J ennings

Finally, some sites, like Acachiwa in the Colca Valley (de la Vera Cruz 1987) (see Figure 1.6), were unlikely to have housed any Wari settlers since there is only limited Wari influence on the sites’ architecture and ceramics (Schreiber 1992:104). Although the Middle Horizon in the Colca Valley was a period of agricultural intensification and population expansion (Brooks 1998; Denevan 1986; Wernke 2003), there are no indications of a Wari presence in the region. The 35 ha site of Acachiwa has a rectangular enclosure wall that is only vaguely reminiscent of Wari enclosures, and the ceramics found in the enclosure and elsewhere are distant derivatives of later Wari styles (Wernke 2003). While sites like Acachiwa were not likely part of a Wari colonial or administrative network, they nonetheless attest to the widespread appeal of Wari material culture during this period.

A 100 m

0

MN

Figure 1.6  Site plans of Honcopampa (Purushmonte Sector) (A) and Acachiwa (B) (adapted respectively from Isbell 1991b:31 and de la Vera Cruz 1987:99).

B Cliff Face

MN 0

100 m

As in Ayacucho, there is a clear temporal dimension in the construction of these installations. The initial expansion of Wari influence likely occurred over a time period of less than a hundred years during the second half of the Early Middle Horizon (Schreiber 2001:85). The largest peripheral compounds—Viracochapampa in the north and Pikillacta in the south—were built at this time, as well as a handful of other smaller sites like Pacheco and Cerro Baúl. The construction of these centers was paralleled by the widespread distribution of Wari ceramics and textiles (Benavides 1999; Menzel 1964, 1977). At the end of the Early Middle Horizon, Viracochapampa was abandoned before it was occupied, the plans for Pikillacta appear to have been downsized considerably because large parts of the site were not completed, and Cerro Baúl was extensively remodeled with the construction of two D-shaped temples (we

unfortunately have little information about Pacheco since the site is now almost complexly destroyed) (McEwan 1996:181–183; Topic 1991:152; Williams 2001:79). Meanwhile, new sites like Honcopampa and Pataraya were built during the Late Middle Horizon (Isbell 1989, 1991b; Schreiber 1999, 2001). Much smaller than the massive sites built at the beginning of the Wari expansion, these later sites were often located along transport corridors on the fringes of valleys.

Interpreting Wari’s Role in the Middle Horizon Since the initial suspicions of Uhle and Tello of a Wari Empire, most scholars have envisioned Wari as an extensive, expansionist state (Isbell 1986, 1987; Isbell and McEwan 1991; Isbell and Schreiber 1978; Jennings and

B eyond Wari Walls

  9

Craig 2001; McEwan 1987, 1996, 2005; Schreiber 1978, 1987, 1992, 1999, 2001). In these models, Wari political economy is modeled in part after Inca political economy and other known states (Jennings 2006a; Topic and Topic, this volume). Huari is seen as the state capital, and the surrounding valleys make up its agricultural core. The provincial sites with Wari architecture are conceived of as Wari administrative sites that helped to organize local political economies for the extraction of desired resources. The centers stored these resources, redistributed some of this surplus in large-scale feasts to reciprocate for the labor given to the state, and then shipped the rest of these resources off to Huari. This model, often implicitly used, dates back at least to Rowe’s conceptualization of the Middle Horizon (Schaedel 1993). Since that time, the largest criticism about the model has been about the absence of a significant Wari footprint in some regions (e.g., Grossman 1983:85; Hastorf 1993:46; Lau 2002:300; Mackey 1982:330; Shady Solís 1982, 1988; Shimada 1985:361; Wilson 1988:358). This discussion has perhaps been most developed on the North Coast, where most scholars over the last three decades have suggested that the Wari presence was limited to a few imported textiles and ceramics (T. Topic 1991; Chapdelaine, this volume). Most of these scholars, however, have stopped short of questioning the overall model of Wari political economy and instead only suggest that Wari did not conquer their particular region. To better fit the Wari imperial model to the mixed Wari footprint in Peru that was emerging in the early 1990s, Katharina Schreiber developed a model that stressed the variability of state-local relationships in what she called a “mosaic of control” (Schreiber 1992:276). Schreiber suggested that Wari settlements were the anchors in a polity-wide administrative network—the pockets of direct control within the imperial mosaic. Other areas of the empire were more indirectly controlled through strategies like ousting a ruler and putting a puppet ruler in his or her place or by raising a friendly local elite person to a newly created political office (1992:17–25). In her model, the kind of control that occurred in a particular valley was determined by state interests, logistical concerns, and local conditions 10 

J ustin J ennings

(1992:276). The areas with little to no Wari material were therefore explained as either outside of Wari boundaries (such as in the case of the North Coast) or areas of indirect control that would have had few indications of Wari domination in the material record. As Schreiber was developing her model, three other archaeologists were suggesting that the evidence from their sites pointed to a more consensual relationship between Wari and local populations. Martha Anders did her dissertation fieldwork at Azángaro, the Late Middle Horizon administrative site in the Ayacucho Basin (1986, 1991). The site was a 7.8 ha rectangular enclosure divided into three sections. The North Sector was composed of patios and corridors in the orthogonal cellular style; the Central Sector was made up of thirty-eight rows of small conjoined rooms; and the South Sector was made out of irregular residential buildings. Anders argued that the site’s major function was as a ceremonial center that organized labor commitments to the state based on a calendrical system that was built into Azángaro’s architecture. Laborers resided temporarily in the Central Sector of Azángaro while they served the state, and the local lords who organized activities at the site lived permanently in the South Sector. Wari control, therefore, was primarily ritual based and heavily mediated by local leaders. While Anders did not extend her model to the Wari state in general, she did suggest that such an extension would result in a system that was “less centralized and bureaucratic” (1991:194). John and Theresa Topic shared Martha Anders’s concerns about the relationship between the Wari state and locals in the region of Huamachuco where they worked (J. Topic 1991; Topic and Topic 2001, this volume). The Topics dug at Viracochapampa, the site hypothesized by Rowe as the Wari administrative center that organized the northern extension of the Wari Empire, as well as at other settlements in the region. They found few Wari ceramics in their excavations and little to no evidence that Wari controlled the area. Instead, they felt that the “brief, but intense” relationship between Wari and Huamachuco was largely confined to mutual influence and trade (J. Topic 1991:161). For the Topics, if Viracochapampa was a Wari site, then it was built by local builders who followed many

of their own local architectural canons. Since construction of Viracochapampa was not completed, they argue that little interaction between Wari and the people of Huamachuco occurred at the site. The most intense interactions likely occurred at the water shrine of Cerro Amaru where a tomb is filled with Wari material. In their work, the Topics have highlighted the importance of interregional exchange, religious influence, and collaboration in their interpretations of the Middle Horizon in Huamachuco. While they initially confined their discussions of these themes to the valley where they worked, the Topics have more recently questioned the overarching Wari imperial model (Topic and Topic 2001:214, this volume). Other scholars have joined Anders and the Topics in questioning aspects of the imperial model. Many of these scholars express their ambivalence about the Wari Empire by calling Wari instead a “state,” “polity,” “culture,” or “phenomenon”—a few have even denied outright that a Wari state existed (e.g., Bawden and Conrad 1982:31–32; Shady Solís 1982:63–64). Yet there are few well-developed alternative models to the Wari imperial one. There are, however, some ideas. Paulsen (1989), for example, suggested that Huari’s rise could be modeled after that of Mycenae and other city-states, and Daniel Shea (1969) argued that the Wari spread through the creation of branch oracles. Unfortunately, these authors did not significantly develop their models in subsequent publications, and their ideas have been left unexplored. The only well-developed alternative model for Wari is by Ruth Shady Solís. Shady argued that Huari was only one of many cities that were interacting with each other in the Middle Horizon. Instead of a pan-Peruvian Wari phenomenon, she sees the period as one of great linguistic diversity and regional trading (Shady Solís 1982, 1988; Torero 1974). She argued that peer polity interaction, between places like Huari, Cajamarquilla, and Marca Huamachuco, caused the broad interregional interactions responsible for the cultural changes that occurred in the Middle Horizon. Shady’s argument was not well received by most Andeanists, primarily because she denied the existence of a Wari state and colonies in her writing. Her skepticism of the imperial model and her emphasis on interregional

interaction, however, has resonated with some scholars. More recent models of the Middle Horizon as a period of intense interaction have attempted to integrate Shady’s emphasis on interregional exchange into emerging models that acknowledge both the ubiquity of Wari influence and the existence of a Wari colonial network (Jennings 2006a, 2006b, this volume; Lau 2005, 2006; Marcone, this volume; Owen, this volume; Topic and Topic 2001, this volume). The debate over Wari’s role in the Middle Horizon is far from over. The monolithic vision that Uhle and Tello had of a Wari Empire sweeping across Peru is no longer accepted. Of course, this does not mean that a Wari Empire did not exist. We now know that Inca rule varied considerably from region to region (Malpass 1993), and we should expect that the material record of a Wari Empire would be similar (Schreiber 1992, 2001). The variations of the Wari footprint seen in this volume can be explained through more flexible imperial models that emphasize the complex interactions that occurred between locals and the state (Isbell, this volume). Alternative models of Wari’s role in the Middle Horizon, however, are also gaining acceptance (Jennings 2006a, 2006b; Lau 2005, 2006; Topic and Topic 2001). These models tend to emphasize that the Middle Horizon can only be understood by including interregional interactions beyond those that occurred between Huari and the rest of Peru. Some of the authors in this volume agree with these alternative Wari models, others embrace the imperial model, and still others offer other new models for consideration.

The Chapters The remaining twelve chapters in this volume explore the Wari phenomena from vantage points outside of Huari and its major Wari centers. This volume’s next eleven chapters describe fieldwork done in regions from southern to northern Peru. The authors of each chapter detail the changes that occurred during the Middle Horizon in the areas where they conduct their fieldwork and also discuss the role that Wari may have played in the broader region during this period. The final chapter is written by the noted Wari scholar William H. Isbell and discusses

B eyond Wari Walls

  11

the chapters in relation to his years of work at Wari sites in Ayacucho and elsewhere. The next chapter in this volume, written by Ulrike Green and Paul Goldstein, describes the Wari presence in the mid–Moquegua Valley. Discussions of Wari in Moquegua have tended to focus on the upper valley, where the large Wari settlements of Cerro Baúl and Cerro Mejia organized a massive agricultural project. In contrast, the slight Wari presence in the mid–Moquegua Valley was centered on a handful of structures built at the site of Cerro Trapiche. Green and Goldstein’s excavations at Cerro Trapiche have revealed a site that interacted considerably with the local Huaracane populations and provide insight into the complex cultural interactions that sometimes occurred between local people and Wari colonists on the frontiers of the civilization. The chapter on my fieldwork in the Cotahuasi Valley introduces a globalization model for the Middle Horizon. I suggest that the urbanization of Huari was the catalyst for increasing interregional interaction throughout Peru. Although there was very little direct interaction between Huari and Cotahuasi, the valley experienced rapid population gain, increased interregional trade, rising social inequality, and the widespread acceptance of the Wari style during the Late Middle Horizon. I argue that the people in Cotahuasi embraced a Wari global culture during a period of great social change in the valley and that they negotiated their new identities at the site of Tenahaha, a ceremonial center that my team has excavated. Bruce Owen offers a different model for the expansion of Wari influence in the Early Middle Horizon in the Majes/Camaná Valley and focuses in particular on ceramic assemblages from the site of La Beringa. He finds a “derived pattern” of “low-fidelity Wari core folk” finewares and local utilitarian wares without corporatestyle ceramics, which is difficult to explain using an imperial, prestige goods, or direct colonization model. Instead, he suggests that the pattern in Majes/Camaná can be explained through a stepwise budding model in which a grassroots diaspora from the Wari core created settlements in the periphery, which went through a sequence of fissioning and resettlement as populations moved farther away from Ayacucho. The ceramic styles 12 

J ustin J ennings

rapidly diverged from Wari core folk styles as successesive sets of independent colonists carried their derived ceramics farther into the periphery. While most discussions of Middle Horizon Cuzco have focused on the massive Wari sites of Pikillacta and Huaro, Véronique Bélisle and R. Alan Covey’s chapter investigates the period in the region from the bottom-up perspective of nearby villages. They argue that changes were occurring at the beginning of the Middle Horizon in Cuzco—agricultural production and participation in regional prestige goods economies were increasing—but these changes occurred largely before Wari sites were constructed in the region. Survey data, combined with data from excavations at the village of Ak’awillay, reveal some Wari influence in local affairs, but there is very little Wari pottery at these sites and considerable cultural continuity. The authors suggest that Wari was just one of many coexisting polities that were vigorously interacting with each other in the Cuzco region at this time. Christina Conlee investigates the role of Wari in the Nazca Valley. The Wari style was significantly influenced by Nasca, and one of the major Wari styles was defined based largely on the remains of smashed vessels found in the valley at the site of Pacheco. Her discussion reveals a complicated relationship between Nasca and Wari in which some people benefited through connections to Wari, but many people did not. In her excavations at La Tiza, Conlee shows how elites at the site embraced the new social identities and ritual practices that Wari offered. Yet she also recognizes that other groups in the valley seem to have fled the Wari occupation to settle in other parts of the drainage. This period of dramatic change in Nasca came to an abrupt end around AD 900 when much of the region appears to have been abandoned. The chapter by Rafael Segura Llanos and Izumi Shimada is the first of two chapters that investigate Wari’s role on the Central Coast. The authors describe excavations at two sites, Cajamarquilla and Pachacamac, that were long thought to have been bastions of Wari power on the Central Coast. The authors find that Wari influence occurred in the region only after the growth of both sites in the Early Intermediate Period and Early Middle Horizon. Segura’s excavations at Cajamarquilla

reveal that Wari influence came at a time when much of the site was abandoned, and Shimada’s work finds a similar occupational hiatus at Pachacamac during the Late Middle Horizon. The Wari materials that occur at both sites are generally found in funerary or ritual contexts. While the relationship between Wari and this cultural disruption is not clear, this work demonstrates that Wari influence was not linked to urbanization and the development of the Pachacamac cult. The second chapter on the Central Coast is by Giancarlo Marcone. Although Marcone agrees with Ruth Shady that Wari did not conquer the Central Coast, he suggests that her prestige goods model fails to explain much of what was going on during the period. From the vantage point of his excavations at the Little Mud-Brick Compound on the outskirts of Pachacamac, he suggests that the rule of local elites was buttressed by feasts and ceremonies that highlighted local beliefs and connections. Elites participated in long-distance exchanges to acquire Wari imports, and they emulated Wari styles, but they did so to buttress their position vis-à-vis each other. Their power over subordinates was based in the local economy, and they underlined these local ties at community feasts and other events. The chapter by Frank Meddens and Nicholas Branch takes readers geographically closest to Huari with their discussion of the Middle Horizon in the Chicha-Soras Valley. They argue that the valley was transformed to intensify camelid and agricultural productions by Wari officials who may have been based in the nearby Wari center of Jincamocco in the Sondondo Valley. Meddens and Branch focus their discussion on how the integration of the Wari belief system with the worship of local ancestors helped to consolidate the region. Through their long-term investigations at several sites in the valley, they show how a Late Middle Horizon D-shaped enclosure at the site of Yako fits into a preexisting landscape of communal tombs and ritual exhumations. Kit Nelson, Nathan Craig, and Manuel Perales investigate the role of Wari in the Norte Chico region of Peru’s coast. As in other regions, they see some Wari influence on local ceramics, and there have been finds of Wari textiles in the region. Nonetheless, they demonstrate considerable local continuity in settlement

patterns, ceramic styles, and architectural forms. The authors suggest that the area was a frontier zone in the Middle Horizon that was placed between polities on the North Coast, Central Coast, and highlands. The Norte Chico was increasingly tied to the surrounding region during this period, and Wari influence came into this region based on these ties. Yet, the Norte Chico was also locally defined in the sense that these various influences were mediated through long-term traditions of the people living in these valleys. The chapter by Theresa and John Topic discusses over twenty years of research in the Huamachuco Valley of the northern sierra. Building on their previous discussions of the Middle Horizon in the region, they suggest that Wari interactions with local populations in their valley and elsewhere in the region were complex and fluctuated based on changing political, ideological, and economic pressures. They suggest that Wari travelers likely came into the region interested in water, ancestors, and Spondylus shells. The site of Viracochapampa was probably built (but never completed) in conjunction with these travelers who were interested in the niched halls used for ancestor veneration at the site of Marca Huamachuco. They argue that Huamachuco was a crossroads where Wari interacted with various regions in northern Peru. Claude Chapdelaine’s chapter discusses the relationship between Moche and Wari on the North Coast of Peru. Although scholars first argued that Wari conquered the North Coast after the demise of the Moche, recent radiocarbon dates and excavation associations make it clear that there was considerable overlap between the two cultures. Taking his readers valley by valley along the coast, Chapdelaine demonstrates that there is little evidence for Wari control over Moche groups in the Early Middle Horizon and instead suggests that Wari served more as an exotic source of political and ideological knowledge that was drawn upon by North Coast elites. This relationship went largely unchanged after AD 800 when the Moche polities collapsed for endogenous and environmental reasons. The book’s final chapter by William Isbell discusses the chapters in the volume and the general implications of this research on our understanding of Wari and the

B eyond Wari Walls

  13

Middle Horizon. Isbell argues that pervasive Wari influence, combined with the construction of Wari peripheral installations, is strong evidence for a Wari Empire. He notes that the chapters in the volume demonstrate considerable variability in the Middle Horizon, but suggests that this variability should not be surprising when one considers that Wari expansion was a complicated process that unfolded over great distances. Isbell identifies two regional patterns that crosscut the chapters. To the south of Huari, agricultural expansion and intensification followed the expansion of the Wari colonial system, while to the north, changes in social complexity and resource intensification were already well under way before the Wari expansion. Isbell suggests Wari’s role in the Middle Horizon cannot be understood without recognizing broad trends like these. Beyond Wari Walls broadens our understanding of Wari by viewing the Wari phenomenon from a variety of different regional perspectives. As more archaeological research is being conducted in Peru, a more complex picture of the Middle Horizon is emerging. The pioneers of Peruvian archaeology were right in identifying the period as a turning point in the country’s prehistory. The rapid urbanization of Huari forever changed the cultural landscape of the Andes. Yet, the relationship between the city and the widespread interactions that it helped generate is not as straightforward as these archaeologists first thought. To understand what happened in Peru during the Middle Horizon, we need to continue to research the connections between the local, regional, and interregional changes that occurred during this period. This work needs to be done at Huari and the various Wari settlements found throughout Peru, but it also needs to be done beyond Wari walls.

Endnotes 1. In the hopes of disentangling sites from widespread cultural phenomena, I use the terms Tiahuanaco and Huari for the archaeological sites and Tiwanaku and Wari for the civilizations that were stimulated, at least in part, by these sites. This is a convention that is advocated by William Isbell and other Wari scholars (see Isbell 2008). 2.  Similar to the conventions separating Wari from Huari, this book uses the term Nasca for the Early Intermediate Period

14 

J ustin J ennings

culture and Nazca for the river drainage in southern Peru (as per Silverman 1988).

References Cited Algaze, Guillermo 1993 The Expansionary Dynamics of Some Early Pristine States. American Anthropologist 95:303–333. Anders, Martha B. 1986 Wari Experiments in Statecraft: A View from Azangaro. In Andean Archaeology: Papers in Memory of Clifford Evans, edited by Ramiro Matos, Solveig Turpin, and Herman Eling, pp. 201–224. Monograph 27, Institute of Archaeology. University of California, Los Angeles. 1991 Structure and Function at the Planned Site of Azangaro: Cautionary Notes for the Model of Huari as a Centralized Secular State. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 165–198. Dumbarton Oaks, Washington, D.C. Bawden, Garth, and Geoffrey W. Conrad 1982 The American Heritage. Peabody Museum Press, Cambridge, Massachusetts. Benavides C., Mario 1991 Cheqo Wasi, Huari. In Huari Administrative Structures: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 55–69. Dumbarton Oaks, Washington, D.C. 1999 Tejidos Wari/Wari Textiles. In Tejidos Milenarios del Perú/ Ancient Peruvian Textiles, edited by Jose Antonio de Lavalle and Rosario de Lavalle Cardenas, pp. 353–411. AFP Integra, Lima. Bennett, Wendell 1953 Excavations at Wari, Ayacucho, Peru. Publications in Anthropol­ogy 59. Yale University, New Haven, Connecticut. Bragayrac D., Enrique 1991 Archaeological Investigations in the Vegachayoc Moqo Sector of Huari. In Huari Administrative Structures: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 71–80. Dumbarton Oaks, Washington, D.C. Brooks, Sarah O. 1998 Prehistoric Agricultural Terraces in the Rio Japo Basin, Colca Valley, Peru. Unpublished Ph.D. dissertation, Department of Geography, University of Wisconsin, Madison. Browman, David L. 1997 Political Institutional Factors Contributing to the Integration of the Tiwanaku State. In Emergence and Change in Early Urban Societies, edited by Linda Manzilla, pp. 229–243. Plenum Press, New York. 1999 Wari Impact on the Upper Mantaro Basin. Paper presented at the 64th Annual Meeting of the Society for American Archaeology, Chicago. Burger, Richard L., Karen L. Mohr Chávez, and Sergio J. Chávez 2000 Through the Glass Darkly: Prehispanic Obsidian Procurement and Exchange in Southern Peru and Northern Bolivia. Journal of World Prehistory 14:267–362.

Cabera Romero, Martha 1996 Unidades habitacionales, iconografía, y rituales en un poblado rural de la época Huari. Unpublished thesis presented for the license of archaeology to the Univerdidad Nacional de San Critobal de Huamanga, Ayacucho, Peru. Cieza de León, Pedro 1984 [1553] Crónica del Perú, primera parte. Pontifica Universidad Católica del Perú, Lima. Cook, Anita G. 2001 Huari D-Shaped Structures, Sacrificial Offerings, and Divine Rulership. In Ritual Sacrifice in Ancient Peru, edited by Elizabeth P. Benson and Anita G. Cook, pp. 137–163. University of Texas Press, Austin. Cook, Anita G., and Mary Glowacki 2003 Pots, Politics, and Power: Huari Ceramic Assemblages and Imperial Administration. In The Archaeology and Politics of Food and Feasting in Early States and Empires, edited by Tamara L. Bray, pp. 173–202. Kluwer Academic/Plenum, New York. Cusick, James G. (editor) 1998 Studies in Culture Contact: Interaction, Culture Change, and Archaeology. Center for Archaeological Investigations, Southern Illinois University, Carbondale. de la Vera Cruz, Pablo 1987 Cambio en los patrones de asentamiento y el uso y abandono de los andenes en Cabanaconde, Valle del Colca, Peru. In PreHispanic Agricultural Fields in the Andean Region, edited by William Denevan, Kent Mathewson, and Gregory Knapp, pp. 89–127. BAR International Series Vol. 398, No. 1. British Archaeological Reports, Oxford. Deneven, William (editor) 1986 The Cultural Ecology, Archaeology, and History of Terracing and Terrace Abandonment in the Colca Valley of Southern Peru. Department of Geography, University of Wisconsin, Madison. Glowacki, Mary 1996 The Wari Occupation of the Southern Highlands of Peru: A Ceramic Perspective from the Site of Pikillacta. Unpublished Ph.D. dissertation, Department of Anthropology, Brandeis University, Waltham, Massachusetts. 2002 The Huaro Archaeological Site Complex: Rethinking the Huari Occupation of Cuzco. In Andean Archaeology I: Variations of Sociopolitical Organization, edited by W. H. Isbell and H. Silverman, pp. 267–285. Kluwer Academic, New York. Glowacki, Mary, and Gordon F. McEwan 2001 Pikillacta, Huaro y la gran región del Cuzco: Nuevas interpretaciones de la ocupación Wari en la sierra sur. In Huari y Tiwanaku: Modelos vs. evidencias, edited by Peter Kaulicke and William H. Isbell, pp. 31–49. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Goldstein, Paul S. 2004 Andean Diaspora: The Tiwanaku Colonies and Origins of South American Empire. University Press of Florida, Tallahassee. González Carré, Enrique 1981 La antigua ciudad de Wari en Ayacucho. Boletín de Lima 16–18:83–97.

González Carré, Enrique, Enrique Bragayrac Dávila, Cirilo Vivanco Pomacanchari, Vera Tiesler Blos, and Máximo López Quispe 1996 El templo mayor en la ciudad de Wari: Estudios arqueológicos en Vegachayoq Moqo-Ayacucho. Facultad de Ciencias Sociales, Universidad Nacional de San Cristóbal de Huamanga, Ayacucho, Peru. Grossman, Joel 1983 Demographic Change and Economic Transformation in the South-Central Highlands of Pre-Huari Peru. Ñawpa Pacha 21:45–126. Hastorf, Christine 1993 Agriculture and the Onset of Political Inequality before the Inka. Cambridge University Press, New York. Isbell, William H. 1977 The Rural Foundations of Urbanism. Illinois Studies in Anthropology 10. University of Illinois, Chicago. 1985 El origen del estado en el valle de Ayacucho. Revista Andina 3(1):57–106. 1986 Emergence of City and State at Wari, Ayacucho, Peru, during the Middle Horizon. In Andean Archaeology: Papers in Memory of Clifford Evans, edited by Ramiro Matos-Mendieta, Solveig A. Turpin, and Herbert H. Eling, pp. 189–200. Institute of Archaeology, University of California, Los Angeles. 1987 State Origins in the Ayacucho Valley, Central Highlands, Peru. In The Origins and Development of the State in the Andes, edited by Jonathon Hass, Sheila Pozorski, and Thomas Pozorski, pp. 83–90. Cambridge University Press, Cambridge. 1988 City and State in Middle Horizon Wari. In Peruvian Prehistory, edited by Richard W. Keatings, pp. 164–189. Cambridge University Press, Cambridge. 1989 Honcopampa: Was It a Huari Administrative Center? In The Nature of Wari, edited by R. M. Czwarno, F. M. Meddens, and A. Morgan, pp. 98–114. BAR International Series 525. British Archaeological Reports, Oxford. 1991a Huari Administration and the Orthogonal Cellular Architecture Horizon. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 293–316. Dumbarton Oaks, Washington, D.C. 1991b Honocopampa: Monumental Ruins in Peru’s North Highlands. Expedition 33(3):27–36. 1997 Reconstructing Huari: A Cultural Chronology for the Capital City. In Emergence and Change in Early Urban Societies, edited by Linda Manzanilla, pp. 181–227. Plenum Press, New York. 2001a Repensando el Horizonte Medio: El caso de Conchapata, Ayacucho, Peru. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William Isbell, pp. 6–68. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. 2001b Huari: Crecimiento y desarrollo de la capital imperial. In Wari: Arte precolombino peruano, edited by Luis Millones, pp. 99–172. Fundacíon el Monte, Seville. 2008 Wari and Tiwanaku: International Identities in the Central Andean Middle Horizon. In Handbook of South American Archaeology, edited by Helaine Silverman and William H. Isbell, pp. 731–759. Springer, New York.

B eyond Wari Walls

  15

Isbell, William H., Christine Brewster-Wray, and Lynda E. Spickard 1991 Architecture and Spatial Organization at Huari. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 19–54. Dumbarton Oaks, Washington, D.C. Isbell, William H., and Patricia J. Knobloch 2006 Missing Links, Imaginary Links: Staff God Imagery in the South Andean Past. In Andean Archaeology III: North and South, edited by William H. Isbell and Helaine Silverman, pp. 307–351. Springer, New York. Isbell, William H., and Gordon F. McEwan 1991 A History of Huari Studies and Introduction to Current Interpretations. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 1–17. Dumbarton Oaks, Washington, D.C. Isbell, William H., and Katharina J. Schreiber 1978 Was Huari a State? American Antiquity 43(3):372–389. Janusek, John Wayne 2008 Ancient Tiwanaku. Cambridge University Press, New York. Jennings, Justin 2006a Understanding Middle Horizon Peru: Hermeneutic Spirals, Interpretative Traditions, and Wari Administrative Centers. Latin American Antiquity 16:265–286. 2006b Core, Peripheries, and Regional Realities in Middle Horizon Peru. Journal of Anthropological Archaeology 25:346–370. Jennings, Justin, and Nathan Craig 2001 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 20:479–502. Ketteman, William Gregory 2002 New Dates from the Huari Empire: Chronometric Dating of the Prehistoric Occupation of Conchopata, Ayacucho, Peru. Unpublished Master’s thesis, Department of Anthropology, State University of New York, Binghamton. Knobloch, Patricia J. 1983 A Study of the Andean Huari Ceramics of the Middle Horizon 1. Unpublished Ph.D. dissertation, Department of Anthropology, State University of New York, Binghamton. 1991 Stylistic Date of Ceramics from the Huari Centers. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 247–258. Dumbarton Oaks, Washington, D.C. Kroeber, Alfred 1930 Archaeological Explorations in Peru, Part II, the Northern Coast. Memoirs Vol. 2, No. 2. Field Museum of Natural History, Chicago. 1944 Peruvian Archaeology in 1942. Viking Fund Publications in Anthropology 4. Wenner-Gren Foundation for Anthropological Research, New York. Larco Hoyle, Rafael 1948 Cronología arqueológica del norte del Perú. Sociedad Geográfica Americana, Buenos Aires. Lau, George F. 2002 Feasting and Ancestor Veneration at Chinchawas, North Highlands of Ancash, Peru. Latin American Antiquity 13(3):279–304. 2005 Core-Periphery Relations in the Recuay Hinterlands: Economic Interaction at Chinchawas. Antiquity 79:78–99.

16 

J ustin J ennings

2006 Northern Exposures: Recuay-Cajamarca Boundaries and Interaction. In Andean Archaeology III: North and South, edited by William H. Isbell and Helaine Silverman, pp. 143– 170. Springer Science, New York. Lechtman, Heather 1980 The Central Andes: Metallurgy without Iron. In The Coming of the Age of Iron, edited by Theodore A. Wertime and James D. Muhly, pp. 267–334. Yale University Press, New Haven, Connecticut. Leoni, Juan B. 2006 Ritual and Society in Early Intermediate Period Ayacucho: A View from the Site of Ñawinpukypo. In Andean Archaeology III: North and South, edited by William H. Isbell and Helaine Silverman, pp. 279–306. Springer, New York. Lumbreras, Luis 1974a The Peoples and Cultures of Ancient Peru. Smithsonian Institution Press, Washington, D.C. 1974b Las fundaciones de Huamanga. Club Huamanga, Lima. McEwan, Gordon F. 1984 Investigaciones en la cuenca del Lucre, Cuzco. Gaceta Arqueologica Andina 3(9):12–14. 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. British Archaeological Reports, Oxford. 1991 Investigations at the Pikillacta Site: A Provincial Huari Center in the Valley of Cuzco. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 93–120. Dumbarton Oaks, Washington, D.C. 1996 Archaeological Investigations at Pikillacta, a Wari Site in Peru. Journal of Field Archaeology (23)2:169–186. 1998 The Function of Niched Halls in Wari Architecture. Latin American Antiquity 9(1):68–86. 2005 Conclusion: The Function of Pikillacta. In Pikillacta: The Wari Empire in Cuzco, edited by Gordon F. McEwan, pp. 147–164. University of Iowa Press, Iowa City. Mackey, Carol J. 1982 The Middle Horizon As Viewed from the Moche Valley. In Chan Chan: Andean Desert City, edited by Michael E. Moseley and Kent C. Day, pp. 321–331. University of New Mexico Press, Albuquerque. MacNeish, Richard, Thomas Patterson, and David Browman 1975 The Central Peruvian Prehistoric Interaction Sphere. Robert S. Peabody Foundation for Archaeology Papers, Andover, Massachusetts. Malpass, Michael (editor) 1993 Provincial Inca: Archaeological and Ethnohistorical Assessment of the Impact of the Inca State. University of Iowa Press, Iowa City. Meddens, Frank 1991 A Provincial Perspective of Huari Organization Viewed from the Chicha/Soras Valley. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 215–232. Dumbarton Oaks, Washington, D.C. Menzel, Dorothy 1964 Style and Time in the Middle Horizon. Ñawpa Pacha 2:1–106. 1968 New Data on the Huari Empire in the Middle Horizon Epoch 2A. Ñawpa Pacha 6:57–114.

1977 The Archaeology of Ancient Peru and the Work of Max Uhle. R. H. Lowie Museum of Anthropology, University of California, Berkeley. Moseley, Michael E., Robert A. Feldman, Paul S. Goldstein, and Luis Watanabe 1991 Colonies and Conquest: Tihuanaco and Huari in Moquegua. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 141–164. Dumbarton Oaks, Washington, D.C. Ochatoma Paravicino, José, and Martha Cabrera Romero 2001 Poblados rurales Huari: Una visión desde Aqo Wayqo. CANO, Lima. 2002 Religious Ideology and Military Organization in the Iconography of a D-shaped Ceremonial Precinct at Conchopata. In Andean Archaeology II: Art, Landscape and Society, edited by Helaine Silverman and William H. Isbell, pp. 225–248. Kluwer Academic/Plenum, New York. Owen, Bruce 2005 Distant Colonies and Explosive Collapse: The Two Stages of the Tiwanaku Diaspora in the Osmore Drainage. Latin American Antiquity 16(1):45–80. Paulsen, Allison C. 1989 Wari and Mycenae: Some Evolutionary Comparisons. In The Nature of Wari: A Reappraisal of the Middle Horizon Period in Peru, edited by R. M. Czwarno, F. M. Meddens, and A. Morgan, pp. 214–224. BAR International Series 525. British Archaeological Reports, Oxford. Pérez Calderón, Ismael 1995 Excavaciones de salvataje y propuesta sobre conservación de estructuras descubiertas, area museo de sitio en Wari. Convenio Instituto Nacional de Cultura—Universidad Nacional San Cristóbal de Huamanga, Ayacucho, Peru. Ponte, Víctor 2001 Transformación social y política en el Callejón de Huaylas, Ancash. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William Isbell, pp. 219–251. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Pozorski, Shelia, and Thomas Pozorski 1987 Early Settlement and Subsistence in the Casma Valley, Peru. University of Iowa Press, Iowa City. Raymond, J. Scott 1992 Highland Colonization of the Peruvian Montana in Relation to the Political Economy of the Huari Empire. Journal of the Steward Anthropological Society 20(1–2):17–36. Raymond, J. Scott, and William H. Isbell 1969 Cultural Remains in the Pampas River Valley, Peru. Paper presented at the 34th Annual Meeting of the Society for American Archaeology, Milwaukee. Rowe, John 1956 Archaeological Explorations in Southern Peru, 1954–1955. American Antiquity 22(2):135–151. 1962 Stages and Periods in Archaeological Interpretation. Southwest Journal of Anthropology 18(1):40–54. 1963 Urban Settlements in Ancient Peru. Ñawpa Pacha 1:1–28. 1967 An Interpretation of Radiocarbon Measurements from Archaeological Samples from Peru. In Peruvian Archaeology:

Selected Readings, edited by John H. Rowe and Dorothy Menzel, pp. 16–30. Peck Publications, Palo Alto, California. Rowe, John, Donald Collier, and Gordon Willey 1950 Reconnaissance Notes on the Site of Huari, near Ayacucho, Peru. American Antiquity 16:120–137. Schaedel, Richard P. 1993 Congruence of Horizon with Polity: Huari and the Middle Horizon. In Latin American Horizons, edited by Don S. Rice, pp. 225–261. Dumbarton Oaks, Washington, D.C. Schortman, Edward M., and Patricia A. Urban 1998 Culture Contact Structure and Process. In Studies of Culture Contact: Interaction, Culture Change, and Archaeology, edited by James Cusick, pp. 102–125. Southern Illinois University, Carbondale. Schreiber, Katharina 1978 Planned Architecture of Middle Horizon Peru: Implications for Social and Political Organization. Unpublished Ph.D. dissertation, State University of NewYork, Binghamton. 1987 From State to Empire: The Expansion of Wari outside of the Ayacucho Basin. In Origins and Development of the Andean State, edited by J. Haas, S. Pozorski, and T. Pozorski, pp. 91–96. Cambridge University Press, New York. 1991 The Association between Roads and Polities: Evidence for Wari Roads in Peru. In Ancient Road Networks and Settlement Hierarchies in the New World, edited by C. D. Trombold, pp. 243–252. Cambridge University Press, Cambridge. 1992 Wari Imperialism in Middle Horizon Peru. Anthropological Papers 87. University of Michigan, Ann Arbor. 1999 Regional Approaches to the Study of Prehistoric Empires: Examples from Ayacucho and Nasca, Peru. In Settlement Pattern Studies in the Americas: Fifty Years since Viru, edited by Brian Billman and Gary Feinman, pp. 160–171. Smithsonian Institution Press, Washington, D.C. 2001 The Wari Empire of Middle Horizon Peru: The Epistemological Challenge of Documenting an Empire without Documentary Evidence. In Empires: Perspectives from Archaeology and History, edited by Susan E. Alcock, Terrence N. D’Altroy, Kathleen D. Morrison, and Carla M. Sinopoli, pp. 70–92. Cambridge University Press, New York. Schwartz, Glen M., and Steven E. Falconer (editors) 1994 Archaeological Views from the Countryside: Village Communities in Early Complex Societies. Smithsonian Institution Press, Washington, D.C. Shady Solís, Ruth 1982 La Cultura Nievería y la interacción social en el mundo andino en la época Huari. Arqueológicas 19:15–108. 1988 Época Huari como interacción de las sociedades regionales. Revista Andina 6(1):67–99. Shea, Daniel 1969 Wari Wilka: A Central Andean Oracle. Unpublished Ph.D. dissertation, Department of Anthropology, University of Wisconsin, Madison. Sherratt, Andrew 2004 Material Resources, Capital, and Power: The Coevolution of Society and Culture. In Archaeological Perspectives on Political Economies, edited by Gary M. Feinman and Linda M. Nicholas, pp. 79–103. University of Utah Press, Salt Lake City.

B eyond Wari Walls

  17

Shimada, Izumi 1985 Perception, Procurement, and Management of Resources: Archaeological Perspective. In Andean Ecology and Civilization, edited by Shozo Masuda, Izumi Shimada, and Craig Morris, pp. 357–399. University of Tokyo, Tokyo. Silverman, Helaine 1988 Cahuachi: Non-Urban Cultural Complexity on the South Coast of Peru. Journal of Field Archaeology 15(4):403–430. Spickard, Linda 1983 The Development of Huari Administrative Architecture. In Investigations of the Andean Past, edited by Daniel H. Sandweiss, pp. 136–160. Latin American Studies Program, Cornell University, Ithaca, New York. Stanish, Charles 2002 Formación estatal temprana en la cuenca del Lago Titicaca, Andes. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 189– 215. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Stein, Gil J. 1999 Rethinking World Systems: Diasporas, Colonies, and Interaction in Uruk, Mesopotamia. University of Arizona Press, Tucson. 2005 The Archaeology of Colonial Encounters: Comparative Perspectives, edited by Gill J. Stein. School of American Research Press, Santa Fe, New Mexico. Storey, Glen R. (editor) 2006 Urbanism in the Preindustrial World: Cross-Cultural Approaches. University of Alabama Press, Tuscaloosa. Stübel, Alfons, and Max Uhle 1892 Die Ruinenstaette von Tiahuanaco im Hochlande de Alten Perú: Eine Kulturgeschichtliche Studie Auf grund Selbstaendiger Aufnahmen. Verlag von Karl W. Heirsemann, Leipzig. Tello, Julio C. 1942 Origen y desarrollo de las civilizaciones prehistóricas andinas. Proceedings of the XXVII International Congress of Americanists, Vol. 1, Pt. 2, pp. 589–720. Lima. Topic, John R. 1991 Huari and Huamachuco. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 141–164. Dumbarton Oaks, Washington, D.C. Topic, John, and Theresa L. Topic 2001 Hacia la compresión del fenómeno Huari: Una perspectiva norteña. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 181– 218. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Topic, Theresa Lange 1991 The Middle Horizon in Northern Peru. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 233–246. Dumbarton Oaks, Washington, D.C. Torero, Alfredo 1974 El Quechua y la historia social andina. Dirección Universitaria de Investigación. Universitario Ricardo Palma, Lima. Tung, Tiffany A. 2008 Dismembering Bodies for Display: A Bioarchaeological Study of Trophy Heads from the Wari Site of Conchopata, Peru. American Journal of Physical Anthropology 136:294–308.

18 

J ustin J ennings

Uhle, Max 1913 Zur Chronologie der alten Culturen von Ica. Journal de la Société des Americanistes de Paris 10:341–367. 1991 [1903] Pachacamac: A Reprint of the 1903 Edition by Max Uhle, with an introduction by Izumi Shimada. University of Pennsylvania, Philadelphia. Valencia Zegarra, Alfredo 2005 Wari Hydraulic Works in the Lucre Basin. In Pikillacta: The Wari Empire in Cuzco, edited by Gordon F. McEwan, pp. 85–97. University of Iowa Press, Iowa City. Valkenier, Lisa 1995 New Evidence for Chimu Capac and the Early Horizon Period in the Supe Valley, Peru. Journal of the Steward Anthropological Society 23(1/2):269–286. Vivanco P., Cirilo, and Lidio M. Valdez C. 1993 Poblados Wari en la cuenca del Pampas-Qaracha, Ayacucho. Gaceta Arqueológica Andina 23:83–102. Von Hagen, Adriana, and Craig Morris 1998 The Cities of the Ancient Andes. Thames and Hudson, New York. Wagner, Linda J. 1981 Information Exchange as Seen in Middle Horizon Two Ceramics from the Site of Huari, Peru. Unpublished Ph.D. dissertation, Department of Anthropology, University of Wisconsin, Madison. Wernke, Steven 2003 An Archaeo-history of Andean Community and Landscape: The Late Prehispanic and Early Colonial Colca Valley, Peru. Unpublished Ph.D. dissertation, Department of Anthropology, University of Wisconsin, Madison. Willey, Gordon 1945 Horizon Styles and Pottery Traditions in Peruvian Archaeology. American Antiquity 11:49–56. 1948 Functional Analysis of Horizon Styles in Peruvian Archaeology. In A Reappraisal of Peruvian Archaeology, edited by Wendell Bennett, pp. 8–15. Society for American Archaeology Memoirs No. 4. Society for American Archaeology and Institute of Andean Research, Menasha, Wisconsin. Williams, Carlos, and Jose Pineda 1985 Desde Ayacucho hasta Cajamarca: Formas arquitectónicas con filiación Wari, unidad del espacio andino. Boletín de Lima 7(40):55–61. Williams, Patrick Ryan 2001 Cerro Baúl: A Wari Center on the Tiwanaku frontier. Latin American Antiquity 12(1):67–83. 2002 Rethinking Disaster-Induced Collapse in the Demise of the Andean Highland States: Wari and Tiwanaku. World Archaeology 33(3):361–374. Williams, Patrick Ryan, and Johny Isla 2002 Investigaciones arqueológicas en Cerro Baúl, un enclave Wari en el valle de Moquegua. Gaceta Arqueologica Andina 26:87–120. Williams, Patrick Ryan, and Donna J. Nash 2002 Imperial Interaction in the Andes: Huari and Tiwanaku at Cerro Baúl. In Andean Archaeology I: Variations in Sociopolitical Organization, edited by William H. Isbell and Helaine Silverman, pp. 243–265. Kluwer Academic/Plenum, New York. Wilson, David J. 1988 Prehispanic Settlement Patterns in the Lower Santa Valley, Peru. Smithsonian Institution Press, Washington, D.C.

Ch a p t er 2

o The Nature of Wari Presence in the Mid–Moquegua Valley Investigating Contact at Cerro Trapiche Ulrike M atthies Green and Paul S. Goldstein

I

n the Moquegua Valley the arrival of the Wari has been long understood through metrocentric models of imperial expansion that see the provincial centers as enforcers of centralized economic and political power of the Ayacucho core (Isbell 1978; Isbell and Cook 1987; Jennings 2006; McEwan 1991, 2005; Schreiber 1987, 1992). Based on a large-site approach, Wari influence in the Osmore drainage has thus mainly been interpreted through research at the mountaintop site of Cerro Baúl, which has become synonymous with an impressive Wari imperial display of political power (Nash and Williams 2005; Williams 2001). The research at Cerro Baúl and Mejía has not identified penetration of Wari settlement beyond these two sites and suggests a rather isolated position of Wari colonists in the valley. We feel this singular site focus only partially addresses the complex processes of colonial encounter in the Wari periphery. Therefore, in order to illuminate such complicated interactions we need to understand Wari presence in Moquegua beyond the wall of the Baúl site complex and move away from the single well-known center in the upper valley, theoretically as well as geographically. The site of Cerro Trapiche in the middle Moquegua Valley offers a view of Wari expansion that differs significantly from Cerro Baúl and allows for a complementary and more dialectic

view of the Wari experience in Moquegua. Because of its liminal geographic and cultural position, Trapiche provides a unique perspective on the colonial experience. Focusing on a site with an indigenous interface offers a view of Wari participation in cross-cultural interactions and active negotiation of social identities and political and ideological power. Trapiche can help illuminate the Wari colonization in the Moquegua Valley in the framework of a “frontier dialectic.”

From Wari Periphery to Wari Frontier Most Andean scholars view the Wari as an expansive state society, although opinions differ on the centralization and degree of Wari political and economic control in the peripheries. Isbell (1986) argued that the Wari used a centralized system of infrastructure, tribute collection, and labor taxation to connect the capital with its periphery. Wari archaeology in the 1980s suggested that the replication of parts of the urban core of Huari in provincial centers was part of an elaborate control system of administrative centers in the periphery. Schreiber (1992, 2005) also relies on evidence from administrative sites to illustrate a “mosaic model” for the Wari, which includes a number of strategies by which the Ayacucho Wari controlled different resource areas and

  19

dealt with different levels of local political complexity in the periphery. While this nuanced approach recognizes that imperial action in a given setting may be determined by existing local power structures as well as shifting imperial agendas (Schreiber 2005), it is still primarily concerned with imperial action from the decision-making perspective of a presumed unitary core. This places less emphasis on what we call a frontier dialectic: the interaction between the interests of the colonizer groups and local populations’ interests and responses to the new situation. Our abstract understanding of Wari imperial strategies has advanced enormously, but we argue that more work needs to be done on the indigenous participation in the frontier dialectic, particularly by focusing on smaller provincial sites. First, however, it is worth mentioning two interpretations of Wari administrative centers that do ascribe an active role to indigenous peoples in Wari empire building. Anders’s (1986, 1991) interpretation of the Wari empire suggests that Wari was a decentralized empire that relied on relatively autonomous local-level lords and traditional reciprocal networks to maintain integration (1986:214–216). She proposed a dual-based authority structure that integrated local-level lords and traditional reciprocal networks, emphasizing that horizontal, interdependent relationships, rather than hierarchical ones, were characteristic for Wari imperial structure. Anders interprets the site of Azángaro as a highly specialized calendrical/ceremonial center (in contrast to models of a more secular Wari state), and, of most interest to us here, she emphasizes a local collaboration with the state over the more coercive means of control implied in Isbell’s and Schreiber’s approaches. The Topics’ work at Viracochapampa in the north suggests yet another scenario for more dialectic Wari relationships with the periphery (Topic 1991; Topic et  al. 2002; Topic and Topic, this volume). They argue that Wari never controlled the Huamachuco area but that the large Wari center there was related to long-distance exchange on a pilgrimage route to the local oracle at Cerro Amaru. John Topic suggests that Viracochapampa was early in Wari expansion as its style is not as rigorously planned as the later sites at Pikillacta, Jincamocco, or Azángaro, and some of the most distinct Wari 20 

U lrike M atthies G reen and Paul S . G oldstein

architectural features, like the patio group, may possibly even be derived from earlier Huamachuco patterns (Topic 1991:158–162). Viracochapampa can be considered a hybrid Huamachuco-Huari site that suggests a complex two-way frontier dialectic between Wari colonialist groups and indigenous populations. Following these examples and approaching the Wari periphery through the lens of a frontier perspective is not only useful in illuminating the multiple facets of these interaction spheres but also in understanding the nature of Wari itself.

The Frontier Lens While models of Wari imperialism may consider multiple strategies of control, most see indigenous peoples of the peripheries as reactors, rather than actors in the dramatic growth of empire. Whether as resistors or as passive recipients of the core culture and its ideas, indigenous populations tend to be seen as a mere part of the landscape of Wari imperialism. What traditional models of Wari imperialism lack is a more refined understanding of the complicated relationship between a core and its provinces at the actual interface of the two: local contexts, in the regions of cultural contact where newcomers and local populations interact (Lightfoot and Martinez 1995). Colonial encounters are not unidirectional processes simply driven by core strategies, but are also defined by interactions in the periphery. The place of contact is a stage for a fluid process of cultural interaction between states and peripheral populations of considerably more dialectic complexity than one “core’s” choice between direct versus indirect control (Alcock et al. 2001; Cusick, ed. 1998; Dietler 1998, 2005; Rice 1998; Shortman and Urban 1998; Stein 2002; Stein, ed. 2005). Studies from the North American frontier, for example, acknowledge that provincial sites must be seen as platforms for negotiation of power and identity by local agents (Cusick, ed. 1998; Deagan 1983, 1998; Lightfoot et al. 1998; Shortman and Urban 1992, 1998). Deagan (1983, 1998) argues that cultural change in the Spanish colonies was a formation of new creolized identities through a process of transculturation and ethnogenesis; marriage between Spanish men and Native

American women formed the actual economic and cultural basis for a new colonial culture (see also Ewen 1991). Similarly, Lightfoot et al. (1998), studying the Russian fur trade enclave at Fort Ross on the north coast of California, demonstrated that Alaskan Aleut men, hired by Russians as seal hunters, married local Kashaya Pomo women and lived together in intercultural households in a distinct neighborhood adjacent to the fort. These examples illustrate that interregional imperial contact in the modern era was complex, involved interactions among different groups, and was not just the exercise of dominance of Europeans over local populations. The complexity of the interaction between colonialists and indigenous groups and the complex patterns of mestizaje and changes in social identity show that we must incorporate a bottom-up perspective that also explores the lives of the provincial populations who comprised the majority of the empire (Wolf 1982). While an interactive approach to peripheral regions has been frequently used elsewhere in the Americas, it has only found limited use in the study of

the Wari Empire. Our research in the Moquegua Valley underscores the necessity of moving toward a small-site approach for a better understanding of the complex relationships that existed between local populations and intruders on the fringes of the Wari Empire.

The Frontier Dialectic—Toward a Small-Site Approach to Imperial Interaction in Moquegua The Moquegua Valley in southern Peru occupies a unique place in the Wari Empire, not just because it was the southernmost colony, but also because during the Middle Horizon it presented an active frontier with Tiwanaku colonists from the Titicaca region (Figure 2.1). Nonetheless, interpretation of Wari imperial strategies in Moquegua has largely followed the metrocentric model: focusing on the largest provincial center at Cerro Baúl, which has been viewed as a center of Wari political power display (Nash and Williams 2005).

Wari Peru

Bolivia

0

50

100

150

Lake Titicaca

200km

Tiwanaku Moquegua

Figure 2.1  The Moquegua frontier in the Middle Horizon.

N ature of Wari Presence in the M id –M oquegua Valley

  21

We find that the important research at Baúl, Mejía, and the surrounding sites suggests an impressive, yet rather isolated, position of Wari colonists in the valley. This only partially addresses the complex processes that accompany that colonial encounter in the periphery of the Wari empire. Therefore, in order to illuminate such complicated dialectic interactions that define real-world colonial encounters, we will focus our discussion beyond the awe-inspiring Baúl site complex, on a smaller working site where Wari met and worked with indigenous local people on the ground. Cerro Trapiche represents an alternate setting because the site sits in the middle Moquegua Valley, a zone dominated by indigenous local villages and separate Tiwanaku settlement (Figure 2.2). While this setting allows us to address a number of cross-cultural interactions between these three distinct groups, the focus of this chapter is the dialectic between local peoples and Wari newcomers. Politically, local elites may want to associate themselves with new powers as clients and emulate new styles and behaviors, while indigenous com-

moners may take advantage of emerging power structures to negotiate new social settings for themselves. As Deagan (1983, 1998) and Lightfoot et al. (1998) demonstrated, people also interact outside the political power structure on a personal level, creating new social identities that are not associated with either group. New technologies may affect local production, and ideological influences may be visible in the reshaping of sacred landscapes (Schreiber 2004). Investigating Wari sites outside the bigger centers in the hinterlands can illuminate this nuanced cross-cultural interaction, if we can take into account the full context of both regional interests and indigenous host populations.

The Moquegua Frontier Setting The Moquegua Valley is part of the larger Osmore drainage system that is divided into three geographically different parts. The upper valley includes the Huaracane, Torata, and Tumilaca tributaries above the modern city of Moquegua at an elevation of 1,300 m. Drainages at these higher elevations are constricted and narrow, and most agriculture is terraced. The middle Moquegua Valley, on the other hand, is wide and flat, permitting floodplain agricultural production on its banks through the use of canal irrigation (Goldstein 2005:115). The wide coastal valley is fed by the river as it reemerges from the rocky canyons, again permitting floodplain agriculture before flowing into the Pacific north of the port of Ilo. While Moquegua is characterized by a long indigenous occupation from the Archaic to the Late Intermediate Period, the Middle Horizon presents the sole frontier between the local tradition and colonists of both Tiwanaku and Wari states. Whom the Wari Found: The Local Huaracane Tradition

Figure 2.2  Settlement distribution in the Moquegua Valley (after Owen and Goldstein 2001:176–177).

22 

U lrike M atthies G reen and Paul S . G oldstein

Systematic regional surveys by Goldstein (middle valley; see Goldstein 2000, 2005; Goldstein and Owen 2001) and Owen (coastal and upper valley; see Owen 1993; Owen and Goldstein 2001) between 1993 and 1995 revealed a continuous occupation along the river floodplain by local agrarian village settlements (see Figure 2.2). In the middle Moquegua sector, this local occupation and its ceramic tradition were named after the type

site of Pampa Huaracane (Feldman 1989a). Huaracane occupations are characterized by semicircular domestic earthen terraces, which show evidence of organic superstructures (Goldstein 2000:343, 2005:123–124). Huaracane villages were small, with a mean area of only .44 ha per domestic component, with only five settlements over 2 ha in area (Goldstein 2000, 2005). Village sites did not show evidence of defensive walls or public architecture, indicating that Huaracane villages were largely autonomous. There is no evidence supporting a regional or primary center in the valley, thus a generally low-level political and economical integration is proposed for the Huaracane tradition. Subsistence for Huaracane populations consisted of a diverse diet and nonspecialized agrarian subsistence strategy (Goldstein 2000:341, 2005). Habitation sites consistently are located on bluffs very close to the floodplain of the river, indicating simple floodplain agriculture and markedly distinguishing the Huaracane strategy from complex canal and terrace systems of politically more complex societies (Goldstein 2000). Ceramic domestic assemblages of Huaracane sites consist largely of Huaracane Arena, a formative, plain utilitarian ware with a coarse sand-tempered paste (Goldstein 2000:341). Vessels of this paste are often self-slipped in a brownish to reddish brown color. It is found primarily in neckless ollas and represents the majority Huaracane paste type at all Huaracane sites and at Cerro Trapiche. Huaracane Vegetal, a fibertempered paste with similar brown self-slip surface treatment, is associated with larger and thicker neckless olla vessel forms at many Huaracane sites. A third paste type, Huaracane Fino, represents a markedly different local ceramic technology characterized by a very fine and extremely hard, well-fired paste with almost no inclusions that exhibits a pinkish color and that is found in relatively low frequencies at most Huaracane sites. Huaracane utilitarian styles are considered to be indigenous to Moquegua, but they also exhibit generic resemblance to both altiplano and coastal styles of the Middle and Late Formative (Feldman 1989b:216; Goldstein 2000:341). Huaracane burial tradition includes three tomb styles (Goldstein 2005:125–128). Basic cylindrical cists

with individual or multiple burials have been documented at Pampa Huaracane and Tres Quebradas. The majority of Huaracane cemeteries are túmulo burials that Goldstein describes as “clustered or irregular mounds of sand or rock between 2–7m in diameter and up to 3m in height” (2005:125). Multiple burials in túmulos suggest a pattern of longtime use possibly associated with extended families or clans (Goldstein 2000, 2005). A later elite burial type, the so-called boot tomb, received its name from the resemblance of the tomb profile to a boot, marked on the surface by massive stone rings up to 4 m in diameter (Goldstein 2005:17). Boot tombs included multiple burials perhaps relating to elite kin groups, as only eight such cemeteries have been identified. Grave goods included Huaracane pottery as well as beads, wooden objects, baskets, and textiles, hinting at emerging status distinctions within the villages in the later Formative Period (2005:27). The growing argument for temporal overlap of Wari imperialism and indigenous Huaracane occupation is circumstantial, yet compelling. Dates for Huaracane occupation placed the túmulo and boot tomb burial traditions roughly between the late fourth century BC and the late fourth century AD (Goldstein 2000, 2005:128). However, recent excavation at Los Joyeros, a cemetery of a late variant of the Huaracane tradition, was dated with a 1σ range of calibration from AD 350 to 520, extending the later Huaracane tradition into the Early Middle Horizon (Table 2.1). Systematic survey by Owen has demonstrated four Huaracane sites in the upper valley on the slopes of Cerro Baúl with a minor, but indisputable, presence of Ocros- and Chakipampa-style ceramics (Owen and Goldstein 2001:175). Most recently, new dates reported by Costion (2008) for the late Huaracane occupation at Yahuay Alta confirm a Huaracane presence into the Early Middle Horizon. Collectively these data support at least some continuous occupation by the Huaracane tradition in the mid-valley that is contemporary with the arrival of Tiwanaku- and Wari-related settlements. Tiwanaku in Moquegua

The Moquegua Valley lies less than 300 km west of the site of Tiahuanaco and provides the strongest evidence for expansive Tiwanaku agricultural colonization, with

N ature of Wari Presence in the M id –M oquegua Valley

  23

Beta-189445

Beta-212298

Beta-212299

Beta-120263

Beta-212300

UCI-43615

UCI-43616

UCI-43707

UCI-43617

Cerro Trapiche

Cerro Trapiche

Cerro Trapiche

Cerro Trapiche

Cerro Trapiche

Yahuay Alta

Yahuay Alta

Yahuay Alta

Yahuay Alta

3

5

2

1

1y

1y

1w

1u

Unit

2A

3C

3

Feature 13

Level

Source: Goldstein 2000, 2005 for Trapiche; Costion 2008 for Yahuay Alta

Lab ID

Site

Table 2.1  Radiocarbon dates from the Moquegua Valley

Structural feature

Large circular subfloor storage pit

Ash deposit below living surface

Floor of storage pit

Huaracane boot tomb Cemetery

Structure 3

Structure 3

Structure 3 posthole

Context

wood charcoal

wood charcoal

wood charcoal

wood charcoal

wood charcoal

wood

wood

charred wood

wood

Material

1215 BP ± 20

1235 BP ± 20

1790 BP ± 20

1825 BP ± 20

1640 BP ± 70

186 BP 0± 70

1320 BP ± 70

1290 BP ± 60

1190 BP ± 40

14C Age

AD 775–827 (0.686308) AD 839–865 (0.313692)

AD 710–746 (0.462579) AD 766–782 (0.224629) AD 789–812 (0.237714) AD 845–856 (0.075079)

AD 177–189 (0.095667) AD 212–257 (0.733698) AD 300–318 (0.170635)

AD 138–159 (0.286646) AD 166–197 (0.433675) AD 207–227 (0.279679)

AD 339–465 (0.726695) AD 482–533 (0.273305)

AD 77-234

AD 650–772

AD 663–775

AD 799–886

Calibrated (1σ)

AD 719–742 (0.073227) AD 769–885 (0.926773)

AD 690–750 (0.415395) AD762–871 (0.584605)

AD 137–259 (0.827283) AD 284–289 (0.009228) AD 291–323 (0.163489)

AD 131–238 (1)

AD 245–564

AD 1–337

AD 605–879

AD 694–700 (0.005258) AD708–747 (0.076452) AD 765–902 (0.820074) AD 916–967 (0.098216) AD 649–878

Calibrated (2σ)

over 121 ha of residential occupation area valley-wide (Goldstein 2005). Moquegua Tiwanaku towns include extensive residential districts, plazas, spatially distinct cemeteries, and ceremonial/administrative structures, most clustered at four large towns of Chen Chen, Omo, Río Muerto, and Cerro Echeníque (see Figure 2.2). The principal Moquegua Tiwanaku towns were located near large artificially irrigated pampas (Williams 1997:90) or productive natural springs (Goldstein 1989), indicating a preference for large agglutinated settlements adjacent to optimal zones for irrigated cultivation. Indeed, unlike the Huaracane sites, Tiwanaku’s large colonial enclaves were located far from the valley and associated with the reclamation of desert pampas through extensive canals or supplementary groundwater sources. Perhaps because these towns and agricultural systems were well populated and staffed by a massive, culturally Tiwanaku immigrant population from the cities of the nearby altiplano, the Tiwanaku colonists seem to have had little to do with indigenous populations, and Tiwanaku households and tombs in Moquegua show virtually no local influence. Wari in Moquegua

As mentioned throughout, archaeological approaches to Wari strategies in Moquegua have centered on the large hilltop city of Cerro Baúl and the adjacent Cerros Mejía and Petroglífo in the upper valley (see Figure 2.2). The occupation of the colony has been described in two phases (Nash and Williams 2005). First established in the Early Middle Horizon (AD 600–800) the Wari erected a principal administration on Cerro Baúl and Mejía. Nash and Williams suggest that Wari elites came here to establish a stronghold in the Tiwanaku frontier, building ceremonial architecture associated with ritual used to legitimize Wari power in the area. An estimated several thousand people (Nash and Williams 2005) could be associated with this intrusion. However, from a settlement pattern perspective, we note that the net area of Wari sites in the region seems to be well under 20 ha, suggesting Wari site populations less than a sixth the size of the Tiwanaku colonies, and a fourth the size of the indigenous Huaracane in the mid-valley. Pending more detailed settlement pattern data for the Wari

occupation, it would appear that the residential presence was limited indeed. The second phase in the Late Middle Horizon (AD 800–1000) sees a shift in architectural construction and power relations at Baúl. Nash and Williams argue that Baúl becomes more of an institutionalized center where Wari state power is exhibited in the use of space on the summit. While Moseley et al. (2005) seek to argue for a political connection at Baúl between Wari elites and the middle Moquegua Valley Tiwanaku colonists at this time, Goldstein (2005) finds remarkably little evidence of Wari-Tiwanaku interaction through trade, and no evidence of domestic interchange whatsoever. Conversely, Nash and Williams do not report evidence of Wari interaction with local groups at or Baúl, or evidence of local occupation on the summit. They argue instead that variant ceramic styles on Baúl and Mejía are not local, but indicate that diverse Wari colonists moved from parts of the empire to this location in order to sustain the colony through agricultural production (Nash 2002; Nash and Williams 2005). The interpretation of Baúl as an intrusive Wari administrative center is predicated largely on the presence on the flanks of Cerro Baúl of extensive agricultural terraces, a typical production technique employed by the Wari and extensively documented in the homeland (Williams 2001). Many of these terraces may date to the Middle Horizon and correspond to Wari-instituted terraced agricultural production techniques. This may have placed agrarian expansion in a previously unfilled ecological niche in the upper Osmore Valley and minimized competition with Tiwanaku and local floodplain agrarian techniques in the mid–Moquegua Valley, but it is difficult to reconcile the extensive terrace networks with the size of the known Wari sites. However, until recently, little focus has been given to Wari presence and activities in other parts of valley, notably in areas of dense indigenous populations. Cerro Trapiche’s unique position in the lower Moquegua Valley allows us to investigate the actual point of interface between local populations and Wari in Moquegua, as well as the relationship between Tiwanaku and Wari within Tiwanaku demographic space and away from the formal center of Baúl.

N ature of Wari Presence in the M id –M oquegua Valley

  25

site was first examined by Moseley, Feldman, and the Programa Contisuyo team in 1983 and 1984, and based on nonsystematic surface collections and test excavations, distinct sectors of ceremonial and domestic architecture, cemeteries, and agricultural and domestic terraces were affiliated with Formative and Middle Horizon occupations. The higher domestic sectors of the site are particularly important analytically because their great separation from the Huaracane boot tomb cemetery makes cross-temporal mixing of looted material unlikely. The four Wari-occupied sectors of the site (M7C, D, E, and F) were reassessed, mapped, and surface collected during pedestrian survey and site investigation phases of the Moquegua Archaeological Survey (MAS) in 1994 and 1998. This was followed by further systematic surface collections and excavations by MAS in 2003 and 2004 and by the Proyecto Arqueológico Cerro Trapiche (PACT) in 2008.1 Figure 2.3  Air view and map of Cerro Trapiche with main sectors C, D, E, and F.

Cerro Trapiche: An Outpost in the Mid-valley Cerro Trapiche (M7) is a multicomponent site located on the slopes and peak of a rocky mountain with a commanding view of the middle Moquegua Valley. Not only is Cerro Trapiche located in the part of the valley where local Huaracane village settlements would have occupied the river floodplain, it also sits right in the middle of the Tiwanaku territory within 2 km of the contemporary Tiwanaku settlement of Cerro Echeníque and within view of the massive Tiwanaku colonies of Chen Chen, Omo, and Río Muerto (Goldstein 2005). Archaeological remains at Cerro Trapiche cover a total of 13.4 ha, and occupation extends from the lower slopes immediately above the floodplain at 1,200 m to the highest peak at 1,500 m in elevation. The most prominent feature of the site is a flat geological terrace on the middle slope (Sector C) that forms a topographical division between the upper slopes and the peak of Cerro Trapiche (Sectors D, E, and F; Figure 2.3). The 26 

U lrike M atthies G reen and Paul S . G oldstein

Sector M7C, the Ceremonial Interface Sector C at Cerro Trapiche is notable for a large area of pitted rock rubble (Figure 2.4) initially interpreted as a destroyed burial platform associated with the Pukara culture of the altiplano (Feldman 1989a). Reexamination of the M7C rock pile in 1998 and 2003 redefined this part of the site as a destroyed elite Huaracane “boot tomb cemetery” with a minority presence of Pukara trade ceramics (Goldstein 2000:349, 2005). Excavations in the cemetery produced several looted boot tombs and one intact burial of a young adult female of Huaracane affiliation (Goldstein and Barrionuevo 2005). Dates and artifacts indicate that the boot tomb cemetery was abandoned before the Middle Horizon.2 Sector C also includes a large enclosed plaza area, two rectangular structures, and slope terracing of Mid­­dle Horizon Wari affiliation that is adjacent to the Huara­ cane boot tomb cemetery (see Figure 2.4). Structure 1 is a well-preserved 7 × 5 m rectangular room with Wari-style masonry walls and internal subdivisions located west of the Huaracane cemetery. Structure 2, located north of the Huaracane burial complex, is a walled trapezoidal plaza enclosure, 32–39 × 25 m, with several stone-lined

tor

c Se

D

Structure 2

Terraces

Tumilaca cemetery

Structure 1

Cerro Trapiche Sector C

Huaracane cemetery

Structure 3

N

0

40

80m

Figure 2.4  Plan view of Sector C.

cists of Wari style having mortared and stone-lined construction, resembling similar walls and cists found in sector E.3 Structure 3 is the best-preserved building in Sec­­­­tor C and can be identified as a Wari patio complex. The construction consists of a 12 × 19 m stone-walled enclosure containing two 8 × 4 m rectangular rooms (Figure 2.5). This type of architecture occurs at Huari in various forms and sizes (Isbell 1991) and is often associated with household groups. Similar construction techniques characterize long retaining terraces on the east slope below Structure 3, possibly constructed to expand the surface of the plateau in C in order to place the structure outside of the ancient Huaracane cemetery. All the Wari structures in M7C are stratigraphically later than the Huaracane boot tomb burial complex and in some areas overlie the looted rock piles, or use material from the cemetery for their construction. Excavations by the MAS 2003 project focused on the southern room (8 × 4 m) in Structure 3 as an example of the Wari architectural occupation (see Figure 2.5). The room excavations identified two superimposed earthen floor deposits with an associated set of four pot rests and

three large, flat metates in primary context with stone manos and botanical remains of Schinus molle consistent with the production of chicha de molle in the structure (Goldstein and Carter 2004). Faunal and botanical debris found in the exterior gallery between the southern room and patio enclosure wall suggest some domestic functions, but the relative scarcity of these materials suggests a functional focus of the patio group as a chichería. Samples from the room interior’s floor, the exterior, and an interior posthole of the structure were dated. The two floor dates suggest an initial construction sometime between cal AD 605 and cal AD 879, while the posthole dates somewhat later between cal AD 694 and cal AD 967 (2σ ranges, see Table 2.1).4 These dates suggest the structure was built in the mid-seventh century and occupied and remodeled through the mid-tenth century. This coincides both with Wari occupation on Cerro Baúl and the earlier part of the Tiwanaku colonization of the Moquegua Valley. Analysis of the ceramic material from Structure 3 ​by Green (2005) revealed a consistent mix of local Huaracane and Wari wares in primary context in all excavation levels (Figure 2.6).5 The Wari ceramics in Structure 3 disproportionately represented serving wares, including flared bowls, lyre cups or tumblers, a face neck vessel

Figure 2.5  Plan view of 2003 excavations in Structure 3, Sector C.

N ature of Wari Presence in the M id –M oquegua Valley

  27

Ceramic distribution inside southern room of Structure 3

300 250

Figure 2.6  Ceramic distributions in excavations of the southern room in Structure 3.

200 Wari

150

Huaracane

100 50 0 surface

level 1

level 2

level 3

level 4

levels

white

brown

light brown

dark brown

a.

b.

0

5

10cm

(a.) Wari serving and (b.) Huaracane utilitarian vessel forms, M7 sectors C, D, E, and F, surface collection

Figure 2.7  Wari ceramics from excavation of Structure 3.

Table 2.2  Ceramic distributions of surface collections in Sectors C, D, E, and F (1998 and 2004)

 

Sector

Area (ha)

Elevation (m)

M 7C

1.061

1375

Huaracane, Pukara, Wari,Tiwanaku (Tumilaca)

M 7D

2.474

1400

Huaracane, Wari

M 7E

0.519

1480

M 7F

2.293

1500

Cultural Affiliation

Wari, Huaracane Huaracane, Tiwanaku (Tumilaca), Wari

Units

Hu

collected

count

Hu (%)

Wari count

Wari %

Tiw count

Tiw %

32

747

50.81

680

46.25

39

8

118

39.50

188

57.31

2

0.60

3

430

64.56

233

34.98

0

0.00

4

127

26.51

347

72.44

5

1.04

2.65

ari %

and a ceramic spoon, and a limited number of plainware storage vessels. Huaracane ceramics were overwhelmingly utilitarian neckless ollas (Figure 2.7). This suggests that Wari fineware was used for chicha beer consumption, while Huaracane ollas tended to be used for food preparation. The terrace retaining the slope directly below Structure 3 was tested by the PACT 2008 project. The terrace platform contained little cultural material and suggested cut-and-fill construction of retaining walls with pits dug into the sterile mountain slope. Debris in these included some discarded manos or mortars, animal bone, shell fragments, maize and seeds, and small fragments of both Huaracane and Wari plainware styles. One feature (R10) contained a concentration of Schinus molle seeds, supporting Structure 3’s connection to chicha production. The construction of a Wari patio group chichería close to what must have been a historically meaningful elite local cemetery suggests an interactive space between the two cultures in a locally revered location. While this could indicate Wari ideological control through manipulation of local sacred landscapes (Schreiber 2004), the joint consumption of molle chicha by local people and Wari newcomers could have helped negotiate new political and economic relationships and identities in a frontier dialectic. 0

5

10 cm

Sector M7D, the Domestic Terraces Rising up in the northeast behind Sector C is the slope of Sector D, a steep area of habitational terracing that covers an area of 2.47 ha. Remains in this sector consist of

Tiw count

Tiw %

Other

Other %

Total

4

0.27

1470

6.25

39

7.31

2

0.60

8

2.43

328

34.98

0

0.00

3

0.45

666

72.44

5

1.04

0

0.00

479

2.65

0

0

5

5cm

10 cm

Figure 2.8  Wari and Huaracane ceramics from 1998 and 2004 surface collection.

domestic terraces constructed from local angular stones. The use of stone-faced linear terracing for domestic occupation on steep hillsides is characteristic of Wari occupation generally (Isbell 1978; Isbell and McEwan1991; Schreiber 1992, 2001) and is locally evident on the slopes around Cerro Baúl (Nash 2002; Williams 2001). The analysis of 2004 systematic surface collections revealed a universally mixed domestic occupation in Sector D, consonant with the mixed cultural affiliation of the Wari architecture in Sector C (Green 2005; Table 2.2, Figure 2.8). In contrast, terrace test excavations in 2008 produced features indicating predominantly domestic occupation of Sector D. Tests in Terrace 4 (a 5 × 5 m unit) yielded a compacted floor surface and division of terrace space into two distinct areas. A concentration of guinea pig droppings suggests that the west side of the terrace was used to raise these animals. The east side of the terrace revealed a large storage pit feature 20 cm deep, filled with almost 23 liters of Schinus molle seeds. Associated manos and ceramic spindle whorls also provide evidence of domestic productive activities including food preparation, storage, animal raising, and textile production on this terrace. Ceramic evidence in Terrace 4 included Huaracane olla fragments in all levels, Wari plainware, and fragments of at least two decorated Wari incurving bowls. Interestingly, some of these fineware fragments

N ature of Wari Presence in the M id –M oquegua Valley

  29

have an unusual paste color that is very similar to the distinct pinkish color of local Huaracane Fino-style bowls. This could point to a local production of Waristyle pottery using Huaracane firing techniques, again suggesting a degree of emulative cultural interaction connecting the two traditions of ceramic production and consumption. While Sector D is on a slope high above the valley floor, Huaracane settlements are generally found on bluffs immediately overlooking the floodplain (Goldstein 2000). This, along with stone terrace construction and the high frequency of some Wari ceramic and lithic types, indicates this was not a traditional Huaracane domestic occupation that adopted Wari imports. We instead suggest that this pattern indicates a Middle Horizon establishment with at least some Wari presence in Sector D, together with some local Huaracane people who moved there from traditional village sites closer to the river. The predominance of Wari serving wares versus Huaracane food preparation vessels in the assemblage implies either a marked adoption of local domestic practice by Wari colonists or a multiethnic occupation across the sector and perhaps even mestizaje or multiculturalism within each household. Either pattern represents considerably more interaction between newcomers and local peoples than has been reported for Cerro Baúl.

Sector M7E, the Upper Ridge Complexes The upper ridge of the Trapiche Mountain, Sector E, forms a ridgeline above the Sector D residential terraces leading up to the fortified main peak at Sector F (see Figure 2.3). Sector E is divisible into lower and higher knolls, separated by a deep trench. The lower knoll includes the remains of one or more collapsed structures of Wari construction with 40 cm thick double masonry walls, including a walled enclosure or plaza containing two large stone-lined and plastered cists 1.5–2 m in diameter and two internal rooms. The vicinity of these buildings contains domestic terraces associated with 226 Wari sherds and a minority of 72 Huaracane ceramic sherds along with Wari projectile points. The higher knoll lacks visible architecture other than the remains of several small earthen terraces, and systematic surface collection found 30 

U lrike M atthies G reen and Paul S . G oldstein

358 Huaracane sherds and only 7 Wari sherds there (see Table 2.2). Thus the upper knoll is predominately associated with Huaracane material while the lower ridgetop represents a more predominantly Wari occupation in terms of both architecture and ceramic distribution.

Sector M7F, the Fortified Peak Sector F is the highest peak of the Trapiche Mountain and is separated from the ridge of Sector E by a dry moat and stone wall. Sector F is the most fortified part of the Trapiche complex and is completely surrounded by an enclosure wall. Wall architecture and the presence of sling-stone caches indicate a defensible Wari occupation on the high peak. The peak sector consists of masonry architecture of Wari technique on stone terrace platforms. Terraces are faced with local angular stone and subdivisions and rooms made of double-faced masonry. The ceramic assemblage represented by four systematic collection units confirmed a strong correlation with the architectural evidence for Wari affiliation, with 347 Wari sherds and 127 Huaracane sherds overall (see Table 2.2). Vessel types included both fine Wari serving wares and a higher proportion of Wari-style utilitarian storage and food preparation vessels than seen in the lower sectors, indicating perhaps an elite Wari residence on the defended summit of Trapiche. Huaracane material was also present throughout Sector F in lesser frequency than in the lower sectors, although there are no identifiable Huaracane structures on the peak. Recovered Huaracane wares in this sector are mostly sand-tempered cooking ollas, with no evidence of local Huaracane fine serving ware. Overall, our systematic collections confirmed a predominantly Wari domestic footprint, along with the use of a minority of Huaracane utilitarian ceramics. The sector’s higher location and its commanding view of the valley, including both the Huaracane-occupied floodplain and the Tiwanaku settlements of Echeníque, Omo, and Chen Chen, would have provided an excellent vantage point from which Wari colonists could observe the Moquegua frontier. The heavy Wari stone construction and defensive works in Sectors E and F, associated with heightened frequencies of Wari-style domestic materials, suggest perhaps

an elite occupation or refuge, as compared to the more locally mixed assemblage of domestic Sector D and the public/ceremonial activities of Sector C below.

Discussion: Trapiche as a Frontier Site The research presented here supports some form of contemporary joint use of the Cerro Trapiche site by Huaracane locals and Wari state colonists, migrants, or refugees during the Middle Horizon. Sector C, an already ancient burial ground of Huaracane elites who had interacted with the earlier Pukara state, seems to have been a focus of continuous special attention by both groups. As an ancestral elite cemetery it was a ritually and ceremonially charged place to local populations and probably closely connected to their representations of social status, group identity, and religious beliefs.5 The Wari occupation at M7C built three structures, effectively encircling the burial ground. This deliberate appropriation of the locally sacred space in Sector C could signal a Wari effort to legitimize their presence by taking hostage the ritual space of local ancestors. This is similar to what Schreiber (2004) describes for the manipulation of local shrines and huacas by the Wari, Inca, and Spanish colonizers of the Sondondo Valley, each of whom altered previous sacred landscapes to embed their own ideological message within local custom and practice. Moreover, the Wari complex on Sector C of Cerro Trapiche also physically controls a gateway location for access to the mixed Wari-Huaracane habitation areas on the steep slopes and peaks above. Because specialized chicha production and consumption took place in a Wari patio group adjacent to ancient sacred grounds, the second implication of Wari patrimonial feasting at this location is that important festive negotiations took place here, at an interface between Wari settlers on the mountain and lands of Huaracane natives below. Defensible habitation sectors M7D, E, and F on the mountain’s upper slopes and peaks are the Moquegua mid-valley’s only Wari residential sectors, and each provides evidence for a joint residential use of the site with Huaracane natives. Their location on the steep upper slopes and peaks of Trapiche is utterly atypical for

indigenous Huaracane settlement, as is the architecture and stone masonry techniques used for most walls, terraces, and storage cists. We feel these settlements were established, or at least greatly expanded, with the arrival of the Wari settlers. Furthermore the omnipresence of some Wari ceramics and the storage of large quantities of molle, the key ingredient in Wari chicha making, in domestic units in Sector D suggest the contribution by individual households to Wari-style patrimonial feasting either in the home or in the Sector C complex below. The use of molle chicha has been reported only for the Wari settlement at Cerro Baúl and has not been recorded for Tiwanaku or Huaracane settlements in the middle valley. While clearly contemporary and culturally affiliated, the Middle Horizon settlement at Cerro Trapiche is distinct from the Wari center at Cerro Baúl in its unique location in the lower Moquegua Valley, in territory otherwise dominated by Huaracane native settlements and Tiwanaku colonial enclaves. Our dates from Structure 3 in Sector C concur with Nash’s dating for the abandonment of Cerro Mejía in the upper valley (2002; Nash and Williams 2005). Perhaps Wari refugees moved into the lower valley during the transition between the Early and Late Middle Horizon in Moquegua, where they came into greater interaction with local people at Trapiche after the relative isolation experienced at the larger center at Cerro Baúl in the upper valley. Or perhaps interaction and some coresidence with indigenous peoples was the norm throughout the Moquegua Wari occupation, and a less-pronounced Huaracane presence at or about the Cerro Baúl provincial center has yet to be recognized (e.g., Owen and Goldstein 2001).

Conclusion How does the idea of a frontier dialectic apply to Wari at the site of Trapiche? Clearly, finer-tuned household archaeological research is needed to flesh out the relationship of Wari colonizers and indigenous peoples at this complex site. Nonetheless, like the examples we cited from the historic North American frontier, our research at Trapiche suggests interactive models of mutual accommodation that belie any core-centric image of a

N ature of Wari Presence in the M id –M oquegua Valley

  31

controlled, passive, or even vacant Wari province. The data presented here suggest a close interaction between local Huaracane and Wari newcomers in the middle Moquegua Valley and allow for a variety of scenarios to be explored in the future. One scenario might see local and Wari settlers residing in separate households within the residential sectors of the site. At Trapiche, such distinctions do not seem to be absolute, and even if there are “monocultural” houses in a multicultural site, we must expect mixed household assemblages as a result of intercultural exchange. Domestic contexts also indicate local emulation of Wari styles in new hybrid ceramic forms and the nonlocal tradition of consumption of molle chicha introduced by the Wari. However, differential frequencies in a sufficiently large sample could correspond to either local- or Wari-dominant signatures for individual households groups. This might be the case in Sector E, where surface collection results showed Wari-dominant and Huaracane-dominant assemblages in residential debris found on stone-faced and earthen terraces, respectively. Alternately, culturally mixed coresidential households might also exhibit architecture, construction techniques, and domestic practices that reflect both Huaracane and Wari preferences and styles. Coresidence of Wari and Huaracane individuals in the same household could occur in a range of social contexts from intermarriage and familial “mestizaje,” to adoption, to house societies with sharp social distinction and separation—Wari patron families with attached local laborers, retainers, and servants, perhaps. Examples from historical colonial encounters indicate that colonists are often males who marry local women, leading to creolized identities, transculturation, and ethnogenesis. In Andean households, food preparation has a strong tradition as a discrete female sphere, whereas male activities center on consumption (Weismantel 1988). In mixed households, then, Huaracane food preference and preparation might be seen in the presence of Huaracane plainwares and spatial organization of domestic production. In contrast, household food consumption could be more Wari-like as colonists introduced Wari serving wares for consumption and feasting. 32 

U lrike M atthies G reen and Paul S . G oldstein

Whether mutual acculturation occurred within households or between households, it is clear that Wari and Huaracane domestic practice was negotiated to some degree in every context in every sector of this large and complex habitation site. This differs greatly from the pattern of contemporary Tiwanaku colonies, which were invariably settled as entirely separate Tiwanaku towns without adopting any Huaracane traditions. It also differs from extant interpretations of the Wari occupation at Cerro Baúl, for which no Huaracane presence has yet been reported. Remarkably, the movement of Wari settlers to the middle valley also placed them in the middle of the Tiwanaku territory, especially close to the Tiwanaku site of Cerro Echeníque, only 2 km away. There is little evidence for interaction between these adjacent Wari and Tiwanaku colonies in the mid-valley, and it is noteworthy that Echeníque is the only fortified Tiwanaku site in the valley. While the fortifications at both sites suggest competition over productive land in this corner of the middle valley, enclaves of the two state cultures may have uneasily coexisted by occupying adjacent but different agrarian niches utilized through their distinct irrigation systems. Beyond land and water competition, labor availability played a role. Tiwanaku’s strategy of direct migration from the nearby altiplano and their creation of large, self-sufficient settlement clusters required no agrarian labor input from local populations. Wari people, on the other hand, much fewer in numbers at Trapiche and Cerro Baúl, may have needed to tap into local agricultural labor in order to support their Moquegua enterprise. Perhaps Wari “contracted” Huaracane labor by using ceremonial events at Trapiche to reciprocate connections with local elites. The joint use of domestic and ritual areas at Trapiche could indicate both intermarriage and the negotiation of labor service between Wari and local elites. While Trapiche is identifiable as a Wari settlement, its cosmopolitan domestic remains suggest an integrative relationship between Wari and local people in the middle Moquegua Valley. Contrary to Cerro Baúl, where interaction with local people is not yet documented, local populations were active participants in a cross-cultural

dialogue. Local elites may have used the arrival of Wari newcomers to renegotiate their own position in the local political power structure through feasting events at Cerro Trapiche. Future research on the Middle Horizon site of Cerro Trapiche may be viewed in light of new research at Huaracane sites in the middle valley, as well as the Wari and Tiwanaku occupations of the region. We know that the Huaracane changed this Wari settlement. Change in the nature of Huaracane lifeways during the Middle Horizon (Costion 2008) may be similarly related to the changing cultural landscape in Moquegua with the arrival of Wari and Tiwanaku groups. The multifaceted interactions that took place at the edges of the Wari Empire cannot be understood solely from the perspective of core strategies. The Moquegua area exemplifies the complexity of a frontier dialectic between colonizers and local populations in one periphery of the Wari sphere. More nuanced approaches in household archaeology can allow us to move away from unidirectional perceptions of Wari state dominance over its hinterlands and illuminate the diversity of crosscultural interaction in the Middle Horizon. Viewing at least some peripheral settlements as frontier interfaces can shed light on negotiated outcomes between multiple agents. By steering some attention away from large provincial centers to smaller Wari peripheral settings where more cross-cultural interaction is evident, we can also better understand the versatility of the expansion that took place in the Middle Horizon. Wari peoples did not expand into a vacuum, and a detailed exploration of their frontier dialectic with indigenous people and competing colonial systems will be an important area for future research.

Acknowledgments Research at Trapiche has been supported by the Pacific Rim Research Foundation of the University of California and the H. J. Heinz III Foundation (MAS 2003), CILAS (Green/MAS 2004), the F. G. Bailey Fellowship (PACT 2008), and the University of California–San Diego. We want to thank Mónika Barrionuevo for close collaboration in the 2003 season, Barbara Carbajal for her support during the 2008 field season, and Kirk Costion for

permission to cite new dates from the Yahuay Alta site. Finally we want to thank Justin Jennings for the opportunity to participate in this exciting volume and William Isbell and anonymous reviewers for insightful comments and suggestions on earlier drafts of this chapter.

Endnotes 1.  The MAS 1998 systematic surface collection in Sector C included thirty-two 10 × 10 m units. MAS 2003 excavated the southern room of Structure 3 (in 2 ×2 m units) as well as three units in the Huaracane cemetery. In 2004 systematic surface collections took place in M7D (eight 10 × 10 m units), M7E (two 10 × 10 m units), and M7F (four 10 × 10 m units). The PACT 2008 team excavated in Sector C (one 5 × 3 m unit below Structure 3 and two smaller 2 × 2 m test units were placed on other terraces for comparative analysis). In Sector D one 5 × 5 m unit was placed on a main terrace and three additional 2 × 2 m test pits were excavated on other terraces for comparison. 2.  A few Tumilaca-style Tiwanaku-derived ceramics were found in Sector C due to the nearby presence of three small, unrelated, looted cist tomb cemetery components. A few finds in Sectors D, E, and F may also be associated with the Tiwanaku site at Cerro Echeníque at a distance of about 2 km from Cerro Trapiche. 3.  While the remains of an adult female were found associated with Chiribaya ceramic vessels in Berman’s 1984 excavation, this appears to represent a later reuse. One of the circular cists (1.7 m diameter) with 1.8 m vertical walls was excavated by Berman and found empty except for a few nondiagnostic ceramic sherds and one unidentifiable bone fragment. 4.  Calibrated with the IntCa198 calibration curve (Stuiver et al. 1998) and adjusted for the southern hemisphere (Stuiver and Reimar 1993 [CALIB 5.0, southern hemisphere correction]). 5.  The condition of Huaracane sherds in all levels, as well as the prevalence of Huaracane neckless ollas, the scarcity of Huaracane fine bowls, and the absence of Pukara sherds in Wari architectural contexts, makes it unlikely that these sherds were secondary deposits from the looted Huaracane boot tombs.

References Cited Alcock, Susan, Terence D’Altroy, Kathleen Morrison, and Carla Sinopoli (editors) 2001 Empires: Perspectives from Archaeology and History. Cambridge University Press, Cambridge. Anders, Martha 1986 Wari Experiments in Statecraft: A View from Azángaro. In Andean Archaeology: Papers in Memory of Clifford Evans, edited by R. Matos, S. A. Turpin, and H. Eling Jr., pp. 201–254. Institute of Archaeology, University of California, Los Angeles. 1991 Structure and Function at the Planned Site of Azángaro: Cautionary Notes for the Model of Huari as a Centralized State. In Huari Administrative Structure: Prehistoric

N ature of Wari Presence in the M id –M oquegua Valley

  33

Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 165– 197. Dumbarton Oaks Research Library and Collection, Washington, D.C. Costion, Kirk 2008 Formative Period Social Differentiation in the Middle Moquegua Valley: A View from the Huaracane Site of Yahuay Alta. Paper presented at the 73rd Annual Meeting of the Society for American Archaeology, Vancouver. Cusick, James G. 1998 Historiography of Acculturation: An Evaluation of Concepts and Their Application in Archaeology. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology, edited by James G. Cusick, pp. 126–145. Center for Archaeological Investigations Occasional Paper 25. Southern Illinois University Press, Carbondale. Cusick, James G. (editor) 1998 Studies in Culture Contact: Interaction, Culture Change, and Archaeology. Center for Archaeological Investigations Occasional Paper 25. Southern Illinois University Press, Carbondale. Deagan, Kathleen 1983 Spanish St. Augustine: The Archaeology of a Colonial Creole Society. Academic Press, New York. 1998 Transculturation and Spanish-American Ethnogenesis: The Archaeological Legacy of the Quincentenary. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology, edited by James G. Cusick, pp. 23–43. Center for Archaeological Investigations Occasional Paper 25. Southern Illinois University Press, Carbondale. Dietler, Michael 1998 Consumption, Agency, and Cultural Entanglement: Theoretical Implications of a Mediterranean Colonial Encounter. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology, edited by James G. Cusick, pp. 288– 315. Center for Archaeological Investigations Occasional Paper 25. Southern Illinois University Press, Carbondale. 2005 The Archaeology of Colonization and the Colonization of Archaeology: Theoretical Challenges from an Ancient Medi­ terranean Colonial Encounter. In The Archaeology of Colonial Encounters: Comparative Perspectives, edited by Gil J. Stein, pp. 33–68. School of American Research, Santa Fe, New Mexico. Ewen, Charles R. 1991 From Spaniard to Creole: The Archaeology of Hispanic American Cultural Formation at Puerto Real, Haiti. University of Alabama Press, Tuscaloosa. Feldman, Robert A. 1989a The Early Ceramic Periods in Moquegua. In Ecology, Settlement and History in the Osmore Drainage, Peru, edited by D. S. Rice, C. Stanish, and P. Scarr, pp. 207–217. BAR International Series Vol. 545, No. 2. British Archaeological Reports, Oxford. 1989b A Speculative Hypothesis of Wari Southern Expansion. In The Nature of Wari, edited by R. M. Czwarno, F. M. Meddens, A. Morgan, pp. 72–97. British Archaeological Reports, Oxford. Goldstein, Paul S. 1989 The Tiwanaku Occupation of Moquegua. In Ecology, Settlement and History in the Osmore Drainage, Peru, edited by D. S. Rice, C. Stanish, and P. Scarr, pp. 219–256. BAR International Series Vol. 545, No. 2. British Archaeological Reports, Oxford.

34 

U lrike M atthies G reen and Paul S . G oldstein

2000 Exotic Goods and Everyday Chiefs: Pukara, Paracas-Nasca and Indigenous Sociopolitical Development in the South Andes. Latin American Antiquity 11(4):1–27. 2005 Andean Diaspora: The Tiwanaku Colonies and the Origins of South American Empire. University Press of Florida, Gainesville. Goldstein, Paul, and Monika Barrionuevo 2005 Moquegua Archaeological Project 2003–04: Secuencia Cultural y Ocupación Agrícola. Final Archaeological Investigation Report presented to the Instituto Nacional de Cultura, Lima. Goldstein, Paul, and T. Carter 2004 Trapiche: A Wari Outpost in Tiwanaku’s Land. Paper presented at the 69th Annual Meeting of the Society for American Archaeology, Montreal. Goldstein, Paul S., and Bruce D. Owen 2001 Tiwanaku en Moquegua: Las colonias altiplánicas. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William Isbell, pp. 139–168. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Green, Ulrike M. 2005 Controlling Food Production and Consumption: Wari Imperial Strategies at Cerro Trapiche, Moquegua, Peru. Unpublished Master’s thesis, Department of Anthropology, University of California, San Diego. Isbell, William H. 1978 Environmental Perturbations and the Origin of the Andean State. In Social Archaeology, edited by Charles Redman, pp. 303–313. Academic Press, New York. 1986 Emergence of City and State at Wari, Ayacucho, Peru, during the Middle Horizon. In Andean Archaeology: Papers in Memory of Clifford Evans, edited by Ramiro M. Matos, Solveig A. Turpin, and Herbert H. Eling, pp. 189–200. Monograph 27. Institute of Archaeology, University of California, Los Angeles. 1991 Huari Administration and the Orthogonal Cellular Architecture Horizon. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 293–316. Dumbarton Oaks Research Library and Collection, Washington, D.C. Isbell, William H., and Anita G. Cook 1987 Ideological Innovation and the Origin of Expansionist States in Ancient Peru. Archaeology 40(4):27–33. Isbell, William H., and Gordon F. McEwan (editors) 1991 Huari Administrative Structure: Prehistoric Monumental Architecture and State Government. Dumbarton Oaks Research Library and Collection, Washington, D.C. Jennings, Justin 2006 Understanding Middle Horizon Peru: Hermeneutic Spirals, Interpretative Traditions, and Wari Administrative Centers. Latin American Antiquity 17(3):265–285. Lightfoot, Kent G. 1995 Culture Contact Studies: Redefining the Relationship between Prehistoric and Historical Archaeology. American Antiquity 60:199–217. Lightfoot, Kent G., and Antionette Martinez 1995 Frontiers and Boundaries in Archaeological Perspective. Annual Review of Anthropology 24:471–492.

Lightfoot, Kent G., Antoinette Martinez, and Ann Schiff 1998 Daily Practice and Material Culture in Pluralistic Social Settings: An Archaeological Study of Culture Change and Persistence from Fort Ross, California. American Antiquity 63:199–222. McEwan, Gordon Investigations at the Pikillacta Site: A Provincial Huari Center in the Valley of Cuzco. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 93–119. Dumbarton Oaks, Washington, D.C. McEwan, Gordon (editor) 2005 Pikillacta: The Wari Empire in Cuzco. University of Iowa Press, Iowa City. Moseley, Michael E., Robert A. Feldman, Paul S. Goldstein, and Luis Watanabe 1991 Colonies and Conquest: Tiahuanaco and Huari in Moquegua. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 91–103. Dumbarton Oaks, Washington, D.C. Moseley, Michael E., Donna J. Nash, Patrick R. Williams, Susan D. deFrance, Ana Miranda, and M. Ruales 2005 Burning Down the Brewery: Establishing and Evacuating an Ancient Imperial Colony at Cerro Baúl, Peru. Proceedings of the National Academy of Sciences of the United States of America 102(48):17264–17271. Nash, Donna 2002 The Archaeology of Space: Places of Power in the Wari Empire. Unpublished Ph.D. dissertation, Department of Anthropology, University of Florida, Gainesville. Nash, Donna J., and Patrick Ryan Williams 2005 Architecture and Power on the Tiwanaku Frontier. In Foundations of Power in the Prehispanic Andes, edited by Kevin J. Vaughn, Dennis Ogburn, and Christina A. Conlee, pp. 151–174. Archaeological Papers of the American Anthropology Association Vol. 14. American Anthropological Association, Arlington, Virginia. Owen, Bruce D. 1993 A Model of Multiethnicity: State Collapse, Competition, and Social Complexity from Tiwanaku to Chiribaya in the Osmore Valley, Peru. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Los Angeles. Owen, Bruce D., and Paul S. Goldstein 2001 Tiwanaku en Moquegua: Interacciones regionales y colapso. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William Isbell, pp. 169–188. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Rice, Prudence 1998 Contexts of Contact and Change: Peripheries, Frontiers, and Boundaries. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology, edited by James G. Cusick, pp. 44–66. Center for Archaeological Investigations Occasional Paper 25. Southern Illinois University Press, Carbondale. Schreiber, Katharina 1987 Conquest and Consolidation: A Comparison of Wari and Inka Occupations of a Highland Peruvian Valley. American Antiquity 52(2):266–84.

1992 Wari Imperialism in Middle Horizon Peru. Museum of Anthropology, University of Michigan Press, Ann Arbor. 2001 The Wari Empire of Middle Horizon Peru: The Epistemological Challenge of Documenting an Empire without Documentary Evidence. In Empires: Perspectives from Archaeology and History, edited by Susan E. Alcock, Terrence N. D’Altroy, Kathleen D. Morrison, and Carla M. Sinopoli, pp. 70–92. Cambridge University Press, New York. 2004 Sacred Landscapes and Imperial Ideologies: The Wari Empire in Sondondo, Peru. In Foundations of Power in the Prehispanic Andes, edited by Kevin J. Vaughn, Dennis Ogburn, and Christina A. Conlee, pp. 137–150. Archaeological Papers of the American Anthropology Association Vol. 14. American Anthropological Association, Arlington, Virginia. 2005 Imperial Agendas and Local Agency: Wari Colonial Strategies. In The Archaeology of Colonial Encounters: Comparative Perspectives, edited by Gil J. Stein, pp. 237–262. School of American Research Press, Santa Fe, New Mexico. Shortman, Edward M., and Patricia A. Urban 1992 The Place of Interaction Studies in Archaeological Thought. In Resources, Power and Interregional Interaction, edited by Edward M. Shortman and Patricia A. Urban, pp. 3–15. Plenum Press, New York. 1998 Culture Contact Structure and Process. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology, edited by James G. Cusick, pp. 102–125. Center for Archaeological Investigations Occasional Paper 25. Southern Illinois University Press, Carbondale. Spickard, Linda 1983 The Development of Huari Administrative Architecture. In Investigations of the Andean Past, edited by Daniel H. Sandweiss, pp. 136–60. Latin American Studies Program, Cornell University Press, Ithaca, New York. Stein, Gil J. 2002 From Passive Peripheries to Active Agents: Emerging Perspectives in the Anthropology of Interregional Interaction. American Anthropologist 104(3):903–916. 2005 Introduction: The Comparative Archaeology of Colonial Encounters. In The Archaeology of Colonial Encounters: Comparative Perspectives, edited by Gil J. Stein, pp. 3–32. School of American Research, Santa Fe, New Mexico. Stein, Gil J. (editor) 2005 The Archaeology of Colonial Encounters: Comparative Perspec­ tives. School of American Research, Santa Fe, New Mexico. Stuiver, Minze, and Paula J. Reimar 1993 Extended 14C Database and Revised CALIB Radiocarbon Calibration Program. Radiocarbon 35:215–230. Stuiver M., P. J. Reimer, and T. F.Braziunas 1998 High-Precision Radiocarbon Age Calibration for Terrestrial and Marine Samples 1998. Radiocarbon 40(3):1127–1151. Topic, John 1991 Huari and Huamachuco. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 141–164. Dumbarton Oaks Research Library and Collection, Washington, D.C.

N ature of Wari Presence in the M id –M oquegua Valley

  35

Topic, John, Alfredo Melly C., and Theresa Lange Topic 2002 Catequil: The Archaeology, Ethnohistory, and Ethnography of a Major Provincial Huaca. In Andean Archaeology I: Variations in Sociopolitical Organization, edited by William H. Isbell and Helaine Silverman, pp. 303–336. Kluwer Academic, New York. Weissmantel, Mary 1988 Food, Gender, and Poverty in the Ecuadorian Andes. University of Pennsylvania Press, Philadelphia. Williams, Patrick R. 1997 The Role of Disaster in the Development of Agriculture and Evolution of Social Complexity in the South-Central Andean Sierra. Unpublished Ph.D. dissertation, Department of Anthropology, University of Florida, Gainesville.

36 

U lrike M atthies G reen and Paul S . G oldstein

2001 Cerro Baúl: A Wari Center on the Tiwanaku Frontier. Latin American Antiquity 12(1):67–83. Williams, Patrick R., and Donna Nash 2002 Imperial Interactions in the Andes: Huari and Tiwanaku at Cerro Baúl. In Andean Archaeology I: Variations in Sociopolitical Organization, edited by William H. Isbell and Helaine Silverman, pp. 243–266. Kluwer Academic, New York. Wolf, Eric 1982 Europe and the People without History. University of California Press, Berkeley.

Ch a p t er 3

o Becoming Wari Globalization and the Role of the Wari State in the Cotahuasi Valley of Southern Peru Justin Jennings

T

he Middle Horizon (AD 600–1000) was a pivotal period in pre-Columbian Peru. The period was marked by an unprecedented increase in the interregional flow of prestige goods, staple items, and ideas (e.g., Burger et al. 2000; Lechtman 1980; Shady Solís 1982, 1988). Arsenic bronze, for example, was introduced and produced extensively across the country; flexed burial became more common; and the Quechua and Aymara languages were likely more widely spoken (Heggarty 2008; Isbell 1997; Lechtman 2003, 2005). This increased interregional exchange was paralleled by the dramatic development of the Wari state in Ayacucho, the spread of a Wari-derived artistic style throughout Peru, and the construction of a handful of peripheral sites that were likely built and occupied by settlers from the Wari heartland (Isbell and Schreiber 1978; Jennings and Craig 2001; Menzel 1977; Schreiber 1992). The relationship between the Wari state and the period’s surging interregional interactions has been a subject of considerable debate since the Wari culture was first identified in the 1930s (see my introduction to this volume). As the chapters in this volume demonstrate, cultural change and Wari influence in Middle Horizon Peru was far from uniform. Archaeologists have typically tried to explain this variability through either a peer polity model, where Wari was only one of many

interacting cultures during the period (e.g., Shady Solís 1982, 1988; Topic 2003; Topic and Topic 2000; Torero 1974), or an imperial model, where Wari controlled much of Peru through a variety of administrative strategies (e.g., Isbell 1991; Lumbreras 1974; Schreiber 1992; Williams and Isla 2002). Both of these models find some support in the archaeological record, but I argue that a third model may help us better explain the nature of Wari influence away from Ayacucho. I suggest that the Middle Horizon can best be understood as an early period of globalization. The word globalization might seem quintessentially modern—evoking images of the Internet, sweatshops, or new age music. Many scholars use the term to describe how the recent significant increase in the crosscultural flow of information, goods, and individuals has transformed the relationships that people have with each other (e.g., el-Ojeili and Hayden 2006:13; Eriksen 2007:14; Held and McGrew 2000:3; Inda and Rosaldo 2008:4; Tilly 1995:1–2; Tomlinson 1999:2). I agree with these scholars, but I argue that many of the sweeping cultural changes associated with globalization today have occurred repeatedly in human history. The spread of Wari is one example. The rapid urbanization of the Wari capital was the spark that globalized the Middle Horizon by forging new relationships between people

  37

who interacted both with the Wari state and with each other. In this chapter, I discuss the possibility of ancient globalization, introduce a globalization model for Wari culture, and look at how a globalization model helps us better understand the nature of Wari-local interactions in the Cotahuasi Valley of southern Peru.

Ancient Globalization? I am not the first to argue for a connection between ancient and modern interregional interaction (e.g., Elkholm and Friedman 1985; Hall and Chase-Dunn 2006; LaBianca and Scham 2006; Nederveen Pieterse 2004). Archaeologists are most familiar with the attempts to make this connection based on the world systems approach. Pioneered in the 1970s, the approach describes global capitalism through a core-periphery model of systematic, structural connection between rich and poor nations (Wallerstein 1974). The potential application of the approach to earlier periods was quickly noted (Schneider 1977), and most of the scholars who have spent considerable time exploring the possible ramifications of interregional interactions tend to be either world system theorists or people arguing against these theorists (e.g., Abu-Lughod 1989; ChaseDunn and Anderson 2005; Chase-Dunn and Hall 1997; Frank 1993; Frank and Gills 2000). Since archaeologists are often best positioned to evaluate the evidence for ancient world systems, they were quickly drawn into these debates (e.g. Algaze 1993; Blanton and Feinman 1984; Stein 1999). After almost three decades of engagement with the idea, the world systems approach is now losing favor among many archaeologists and other scholars. There is a growing realization that ancient political economies could not work the same as modern ones (see chapters in Stein 2005b). Transportation hurdles were too high, state bureaucracies were too weak, specialization was too shallow, and exchange networks were too limited. A rejection of world systems, however, should not be tantamount to a rejection of globalization as a model for the past since the world systems approach is only one of several competing conceptions of globalization (el-Ojeili and Hayden 2006:12; Sklair 2006). Those approaches 38 

J ustin J ennings

that concentrate more on the cultural consequences of globalization may be more appropriate to understanding the ancient world. One of the most influential definitions of globalization is found in John Tomlinson’s Globalization and Culture, where he defines globalization as a result of “complex connectivity” (1999:2). “Complex connectivity” is Tomlinson’s term for the dense network of interconnections and interdependencies between people today in different parts of the world. He suggests that complex connectivity results from a significant increase in the flow of information, goods, and people across cultural and geographic boundaries. As the connections between people multiply and deepen, there are a variety of social consequences. These consequences, the indicators of complex connectivity, are often what we think of when we use the term globalization (see similar definitions in el-Ojeili and Hayden 2006:13; Eriksen 2007:14; Held and McGrew 2000:3; Inda and Rosaldo 2008:4; Tilly 1995:1–2). Globalization scholarship highlights the tensions between global and local influences that occur as a result of the increased flow of goods, people, and ideas around the world (e.g., Appadurai 1996; Eriksen 2007; Featherstone 1990; Robertson 1992). These connections create a global culture that exposes groups to a wide variety of outside influences. These connections can perhaps be seen most easily by a Maori herdsman drinking a Coca-Cola or a woman sweating out a Bikram yoga class in Los Angeles. These examples show how a global “way of being” has been created from a variety of Western and non-Western sources (Inda and Rosaldo 2008:16). Global culture also operates at a more fundamental level by instilling broadly held standards. Beauty pageants around the world, for example, are possible because of a growing consensus that beauty is a physical characteristic that can be judged on a broad set of attributes (Wilk 1995, 2004).1 Even those pageants that position themselves as countercultural, like the annual Miss Fat South Africa contest, reaffirm the normative values for beauty by positioning their pageants in opposition to other contests. Similar cultural transformations are being wrought throughout the world as people adopt more mundane global norms, such as a monetary economy,

formal education, nationalism, and the political party system (Eriksen 2007:52). The creation of a global culture does not mean that the world will become a single culture. Globalization disembeds cultures from their local roots, but it also reembeds local cultures through a process sometimes called glocalization (e.g., Giulianotti and Robertson 2007). Torn by desires for both global and local identities (Green 2002:116), people often attempt to strengthen, re-create, and even invent local cultures that are used as a counterweight to the centripetal tendencies of globalization (Eriksen 2007:142). The reembedding of culture can be seen in the reaction of the Yup’ik Eskimos of western Alaska to recent changes in their way of life, which have occurred since the introduction of snowmobiles, the availability of imported canned goods, access to cable television, and many other factors (Fienup-Riordan 1990). These changes have fueled nostalgia for an idealized old way of life where people were fit, generous, and disciplined and the world was quieter, stable, and without boundary markers. Negative aspects of Yup’ik history, like conflict, malnutrition, and isolation, are wiped out of Yup’ik history as an idealized past is created. The “invention of tradition,” a common reaction to the rapid cultural change that occurs in globalization, leads to greater cultural diversity in some regions (Hobsbawm and Ranger 1983). Groups bound together through complex connectivity find themselves in an ongoing struggle between local and global identities. They embrace some aspects of world culture, reject others, and transform other aspects to fit their own needs. Since different people within a group react to these new potential identities in different ways, globalization alternately reinforces, invents, and transforms age, class, gender, ethnic, and other divisions. The attitudes of Maya youth exemplify how identities are reshaped during globalization. More and more of these youths are being drawn into the world economy through work in rural factories that assemble apparel for export. These youths are torn in conflicting directions between “being Indian” as defined by their families and “being modern” as defined by a popular culture that valorizes Western ideas of consumerism and individualism (Green 2002:116). “Being Indian” for these youths was

not something that they thought about until confronted with another way of living. Both the local and the global are defined when globalization occurs. Most of the authors who define globalization as the result of complex connectivity argue that globalization first occurred sometime over the last six hundred years. These scholars argue that between the European voyages of discovery in the fifteenth century and our current era, a series of changes, such as capitalism, urbanization, colonialism, and the creation of new communication technologies, provided the critical push toward globalization (e.g., Castells 1996; Sklair 2002; Tomlinson 1999). This makes sense. Today we are more deeply connected to more people over larger distances than ever before, and most of these connections have developed during this period. After all, there was no Internet, Coca-Cola, or global commodities market in the ancient world. This does not necessarily mean, however, that complex connectivity is unique to the modern era. I argue that leaps in interregional interaction occurred during earlier periods, which occasionally triggered consequences broadly similar to those that we are experiencing today. If we define globalization as the result of complex connectivity, then these earlier leaps caused what should be considered the first globalizations (Jennings 2011). The first globalizations may have occurred with the emergence of some of the world’s first civilizations (Figure 3.1). Early urbanization, at places like UrukWarka, Cahokia, Harappa, and San Lorenzo, led to the flow of goods, ideas, and people across wide regions for many reasons. The increasing demands by urbanites for staple and prestige goods, for example, led to the expansion of trade networks and sometimes the establishment of trading outposts. Meanwhile, people flooded into these cities to take advantage of new job opportunities and to participate in ritual performances. These new flows, although usually quite limited in comparison to flows today, created connections across wide regions, often causing, in their own right, significant changes in local political hierarchies, economic specialization, and identities. More importantly, these flows precipitated greater interactions between outlying groups in response to the urban flows. While some of

B ecoming Wari

  39

2005; Domínguez 2002; Stein 1999, 2005a; Wells 1999). Yet the discipline seems unwilling to consider the full implications of the similarities between the cultural consequences of earlier epochs of increased interregional interaction and modern globalization. Instead, archaeologists are in danger of retreating back to older models, like diffusion and peer-polity interaction (Kohl 2007; Stein 1999), which fail to adequately examine the tension between the local and the global.

Globalization in Middle Horizon Peru

Figure 3.1  Model for ancient globalization. The dots represent societies and organization, and the dotted lines between groups represent interaction through trade, marriage, or other mechanisms. Solid lines denote an administrative link between a colony and its society of origin, and shaded areas mark cultural influence. In the global culture phase, almost the entire region is shaded to reflect the emergence of a global culture.

these relationships were with the cities, many of them were not. These new interactions, in turn, caused more and more interregional interactions to occur. In some cases, these long-distance networks were deep and dense enough to create complex connectivity and form a global culture (Jennings 2011). Archaeologists already recognize the signs of globalization in the ancient world. Those that study some of the world’s earliest civilizations have demonstrated that increased interregional interaction during these periods is often correlated with the spread of a culture that was alternately fractured, redefined, embraced, and reviled from one place to another (e.g., Lesure 2004; Pauketat and Emerson 1997; Rothman 2001). The cultural transformations that occurred at the local level are often startlingly similar to those that occur with modern globalization, and archaeologists are beginning to use the same analytic vocabulary as cultural anthropologists to describe these changes (e.g., Dietler 1990, 40 

J ustin J ennings

I argue that the expansion of the Wari culture was an example of ancient globalization. Like other examples of ancient globalization around the world (see Figure 3.1), the rise and decline of Wari can be grouped into four phases: preglobalization, expansion, global culture, and regionalization (Table 3.1). The preglobalization phase of Wari expansion occurred during the Early Intermediate Period (200 BC–AD 600) and the first part of the Middle Horizon (600–700). The Early Intermediate Period was marked by increasing regionalization in many areas because of the development of complex societies like the Moche, Nasca, and Recuay. These cultures stimulated the intensification of trade networks that ran both vertically along the coast and horizontally across the Andean cordillera (Lau 2005, 2006; Shimada 1985; Vaughn 2004). Compared to the great cultures of the period, the Ayacucho basin in the highlands of central Peru was a backwater. During the Early Intermediate Period, the inhabitants of the Ayacucho basin shared a ceramic style of motifs painted on red, white, and black slips known as Huarpa and lived in over a hundred small agricultural villages (Knobloch 1983, 2010; MacNeish et al. 1981:Table 8.16). Outside influence on the Huarpa style from the Nazca region of Peru’s South Coast increased during the later phases of the Early Intermediate Period (Knobloch 2010; Ochatoma and Cabrera 2001). Some villages, such as Ñawinpukyo, Huari, and Conchopata, expanded and likely became primate centers in fragile local political hierarchies that were maintained through feasting and ritual (Isbell 2001:111–117; Leoni 2006). If these large villages were using ceremonial displays to vie

Table 3.1  Phases of Wari globalization

Phase

Period

Description

Preglobalization

200 BC–AD 700

Development of regional interaction systems in some areas

Expansion

AD 700–800

Growth of Huari, expansion of Wari trade networks, and the founding of Wari colonies outside of Ayacucho

Global Culture

AD 800–1000

Increasing interregional interaction, widening of Wari cultural influence, and creation of global culture

Regionalization

AD 1000–1400

Huari and its colonies abandoned, end of global culture, and increased balkanization

for the hearts of Ayacuchanos, then the inhabitants of Huari seem to have won this fight by around AD 600. Rapid urbanization of the Ayacucho basin occurred at the end of the Early Intermediate Period. A number of villagers coalesced into the site of Huari by around AD 550, and the city quickly began to draw more and more people into the site (Isbell 2001:117). The growth of the city was paralleled by the creation of new polychrome ceramic styles that depicted cosmological figures, warriors, and elites (Menzel 1964). The development of this style, combined with the creation of D-shaped structures and the practices of ancestor worship and trophy head taking (Cook 2001; Ochatoma and Cabrera 2002; Tung 2008), points to the development of a new religion that both attracted people to the city and legitimated increasing class differences (Menzel 1964:67). The expansion phase of Wari globalization started around AD 700. Huari was increasing dramatically at the time in size and population. The city would soon balloon to almost 15 km2 with an architectural core of planned, densely spaced structures that covered about 250 ha (Isbell et al. 1991:24; Schreiber 2001:81). The increasing subsistence demands of the city were met through agricultural intensification in surrounding valleys (Browman 1999; Isbell 1977; Raymond 1992; Raymond and Isbell 1969; Vivanco and Valdez 1993), while the demands of Huari’s growing elite stimulated the flow of metals, obsidian, decorated ceramics, textiles, turquoise, marine shell (Spondylus and Strombus), and other prestige goods into the city from across Peru

and Ecuador (Cabrera Romero 1996:88–91; González Carré et al. 1996:100–102; Pérez Calderón 1995:85–86, 1999:78). As the city grew, its pottery and textiles were exchanged across much of Peru (Knobloch 1991; Menzel 1964). The spread of Wari objects was likely linked at least in part to the spread of Wari religion (Janusek 2008:275–276; Menzel 1964:67). New ritual and mortuary practices were introduced in many regions during the Middle Horizon, and the Wari style was widely emulated (Angeles and Pozzi-Escot 2001; Castillo 2001; Malpass 2001). Wari interaction led to the introduction of ideas in some areas, like the use of terracing for agriculture and knotted string devices for recording information (Conklin 1982; Goldstein 2005:166–167; Williams 2002). Finally, the city established a number of distant colonies (e.g. Glowacki 2002; Glowacki and McEwan 2001; Isbell 1989; Jennings and Craig 2001; Knobloch 1991; McEwan 1991, 1996; Menzel 1964, 1977; Schreiber 1992, 1999, 2001; Williams and Nash 2002; Williams and Pineda 1985). Some of these colonies established political and economic control over their immediate surroundings, while others appear to have been constructed on the fringes of established polities (Jennings 2006a; Jennings and Craig 2001; Schreiber 1992). By AD 800, Wari expansion had stimulated wideranging exchanges across Peru. People, objects, and ideas were flowing in and out of the Huari metropolis and Wari influence, although far from pervasive, was geographically widespread. Exchanges, however, were not just via

B ecoming Wari

  41

Huari—interaction was increasing across the board (e.g., Burger et al. 2000; Lechtman 1980; Shady Solís 1988). Cajamarca ceramics were being widely exchanged across northern Peru, for example, and the ritual networks of the powerful coastal shrine of Pachacamac grew enormously during this period (Kaulicke 2000; Lau 2006; but see Segura and Shimada, this volume, and Marcone, this volume, for concerns about the role of Pachacamac during the Middle Horizon). These interactions caused significant economic, political, and social changes in many regions. As local life became more and more enmeshed with the outside world, complex connectivity ensued and a global culture was created. The creation of a global culture does not mean that everyone became like the people of Huari. “Wari culture” itself was a bricolage of regional influences that developed during expansion (Janusek 2008:277–278). The galleries and niched halls seen at Huari and Wari colonies likely derive from building traditions of Huamachuco in northern Peru (Topic 1991:159), and Tiwanaku religion continued to influence Wari style and practices as the Middle Horizon went on (Cook 1994). The chapters in this volume show that groups often adopted only some Wari elements, and that these elements were adapted to local conditions. The creation of a global culture, however, does mean that most people in Peru were aware of Huari and the iconography and basic religious beliefs of the Wari culture. Many embraced Wari ideas, but even those who did not were often still engaging with Wari by intentionally purging their culture of some Wari elements (Bélisle and Covey, this volume). Wari global culture lasted for another two hundred years. The city’s colonial reach during this period would wane (Schreiber 2001:91), but its cultural influence continued to spread and intensify. Huari was likely in a slow decline—new trade networks might have circumvented the city (Shady Solís 1988; Silverman and Proulx 2002), and shrines at Pachacamac, Huaro, and even Tiwanaku might have siphoned off believers (Glowacki 2002; Kaulicke 2000). Huari and its colonies were abandoned by about AD 1000 (Isbell 2001:162). Wari influence, along with the scope and intensity of interregional influence, declined as well (Conlee et al. 2004). Nonetheless, the postglobalization landscape of Peru 42 

J ustin J ennings

was more integrated than it was before Wari expansion, and Wari architectural forms, art styles, and ideology informed many cultures of subsequent eras (McEwan 1990; Shimada 2000).

The Middle Horizon in the Cotahuasi Valley A globalization model of Wari culture helps to explain what happened during the Middle Horizon in the Cotahuasi Valley. The valley is located in the sierra of the Department of Arequipa in southern Peru (Figure 3.2). Since 1999, Willy Yépez Álvarez and I have led a team of researchers in the valley. We conducted an archaeological and geo-archaeological survey of the upper valley from the town of Cotahuasi to Puica in 1999–2000, and then returned to excavate the sites of Tenahaha and Collota from 2004 to 2007.2 Analysis of these recent excavations continues, and I should caution that the picture emerging from this work remains tentative. Like some other areas in Peru, life dramatically changed in the valley during the Middle Horizon. Population significantly increased, agricultural production expanded, interregional exchange intensified, and social stratification began to emerge. These changes occurred at a time of profound Wari influence in Cotahuasi, but there is little evidence of contact between the valley and the Wari heartland. Instead, the Wari style, and the ideas associated with it, seems to have been embraced by the people in the valley because Wari global culture helped them facilitate interregional interaction and negotiate new local identities. Before the Middle Horizon, the population was settled in ten small villages (1–3 ha) scattered across the valley. During the Middle Horizon, the population of the valley substantially increased. All existing settlements at least doubled in size, and fifteen new villages were founded (Jennings 2006b:359; Trawick 2003:40– 41, 50). The population increase corresponded with the introduction of agricultural terracing (Jennings 2002, 2006b:360). The new technology, perhaps combined with an influx of new plants during the period (see Goldstein et al. 2009; Kellner and Schoeninger 2008:238; and Orefici and Drusini 2003:109 for

7 20

7 25

7 35

7 30

7 45

7 40

Puica

450

N 0

8335

0

500

Kilometers

4500

83 30

Alca 8325

4000

24

00

35

00

30

00

25

Tenahaha

00

45

8320

Cotahuasi

3 km

0

8315

N UTM, Zone 18s 1956 Provinsional South American Datum

Figure 3.2  Map of the Cotahuasi Valley of southern Peru. The stippled area is the area of our 1999–2000 survey, and excavations were conducted in 2004–2007 at the site of Tenahaha.

examples from other regions) increased the valley’s carrying capacity. Interregional exchange to and from Cotahuasi also intensified during the Middle Horizon. The trade in the local obsidian increased, and we have the first evidence for copper and rock salt mining in the valley during the period (Burger and Asaro 1979:36; Burger et al. 2000:324; Jennings 2002). Imported pottery, rare in earlier periods, became more common, and local ceramic styles were heavily influenced by styles

from other areas. Changes in funerary customs, architecture, and ritual practices during the Middle Horizon also reflect increased interaction during the period (Jennings 2002, 2006). As population expanded in the valley, scalar stress increased to the point where preexisting social mechanisms for group cohesion were insufficient (e.g., Friesen 1999). The reaction in the valley was twofold. First, larger villages “fissioned,” and new settlements were founded

B ecoming Wari

  43

in underutilized areas of the valley (e.g., Bandy 2004). Differences in ceramic styles, settlement organization, and cranial deformations hint at how distinct village identities were developing based on both local traditions and varying regional contacts (Jennings 2002; Mummert 2006). Second, incipient stratification began to occur as village size continued to creep upward (e.g., Johnson 1982). Architectural, funerary, and artifact assemblages in the Middle Horizon suggest that a village elite was emerging (Jennings 2002, 2006b:360). A rank size plot of village size, however, suggests that there is no evidence for a valley-wide political hierarchy during the period (Jennings 2006b:361). The pottery style before the Middle Horizon, called the Aicano style, was part of a broad regional ceramic tradition in Arequipa that included the Hachas, Sopora, and Ayawala styles (Neira Avendaño and Cardonas Rosas 2001). Pots in the style are often undecorated, and when decorations are found on vessels they often include an appliqué of raised lines running parallel to the lip of the vessel, applications of a modeled human face with circular incisions for the eyes and mouth, or depressions made by the use of a thin circular stamp on the surface of the ceramic. The Aicano style was remarkably stable with very few changes in form and decoration from the introduction of the styles on the coast around 1400 BC to the Middle Horizon. The stability of Cotahuasi’s ceramic style was not a marker of cultural isolation, but rather marked a lack of desire for emulation. The local obsidian was being traded widely before the Middle Horizon, and the few imported sherds that we have speak to some contact with outside groups. Although Cotahuasinos were expressing little desire to copy other groups, it is likely that their long-distance interactions were increasing by the Early Intermediate Period. We collected seven sherds in our survey that were painted in red, purple, or black on a white or cream background (Jennings 2002:274–277). The fragments do not correspond to any published styles that I know but likely come from the central sierra and date to the end of the Early Intermediate Period (Johny Isla and Mario Ruales, personal communication 2000). Outside interaction and influence continue to be minimal in the Early Middle Horizon. Only nine fragments in Early Middle Horizon 44 

J ustin J ennings

styles have been found in Cotahuasi (0.78 percent of all Middle Horizon sherds, n = 1290). With the exception of one fragment, the fragments are local emulations of Wari styles that show a poor knowledge of Wari motifs (Jennings 2002:279, 281). The almost complete lack of Early Middle Horizon sherds in Cotahuasi is striking relative to the profound Wari influence occurring during this part of the Middle Horizon on the coast of Arequipa (Neira Avendaño 1998; Owen, this volume; Tung and Owen 2006). The cultural isolation of the valley would change quickly in the Late Middle Horizon. There was a complete change in form, decoration, and manufacturing technique of local ceramics (Jennings 2002), and Wari was by far the strongest outside influence on Cotahuasi ceramics and other material culture. Radiocarbon dates from our excavations at Tenahaha confirm the late date of Wari influence in the valley near the end of the Middle Horizon (approximately AD 850–1050) (Table 3.2). Over the space of a few generations, Cotahuasi culture opened up to the outside world. Knowledge of Wari styles was widespread. Late Middle Horizon Wari-style sherds are common at all villages, and these sherds are found in all types of contexts such as in terraces, storage facilities, and house middens. In short, by the second half of the Middle Horizon many of the pots from the valley look Wari, as do many of the metals and textiles. The only evidence for a direct connection between the valley and the Wari state, however, is four ceramic fragments that our petrographic analysis suggests might come from the Wari heartland (0.3 percent, n =1290 ceramics) and a handful of lithics of the local obsidian that was found at Huari and at several Wari colonies (Burger et al. 2000:324; Jennings 2006b:354). For the most part, Wari culture was produced locally or in the surrounding region. The Wari-style pottery found in the valley was almost always locally made (98.2 percent), with the few possible imports (other than those coming from the Huari area) likely coming from the Nazca and Arequipa regions (1.5 percent). Most of the Wari polychrome ceramics are derivatives of the regular Viñaque style (98.4 percent, n = 315). The Wari-style textiles that we found in the valley, according to a consultation by Mary Frame, are in the Nasca-Wari style (personal communication 2000),

4,5 IV

167

17 4

185

3,6 18

185

6,8,9 19 6,8,9 V

4 III

45

4 20 2 27 7 16

A-69901 4076 4076 4 A-69901

16 A-74693 16094 16094 A-74693

A-74696 17025 17025 17 A-74696

18026 18026 A-74697 18 A-74697

18163 18163 18 A-74698 A-74698

A-74714 18046 18046 18 A-74714

19089 19089 A-74700 19 A-74700

A-74701 20 20050 20050 A-74701

27 27026 27026 A-74710 A-74710

A-76040 16094.116094.1 16 A-76040

I

II

IV C

III A

III

V B

II

VI F

III

IV

IV C

I

III G'

III

VI D

IV C

Calibrated (1 δ) (1 δ) Material 14C Age Calibrated Material 14C Age

Calibrated (2 δ) (2 δ) Calibrated

Wood Wood 1191 ± 1191 35 BP± 35AD BP884–974 AD 884–974

AD 781–790 (0.0227)(0.0227) AD 781–790 AD 808–986 (0.9773)(0.9773) AD 808–986

Intact7Tomb 7 C Intact Tomb

Insect casing AD 1028–1184 (0.4858)(0.4858) BP± 34AD Insect casing 972 ± 34972 AD 1028–1184 AD 1045–1087 BP 1045–1087 AD 1105–1156 (0.5142)(0.5142) AD 1105–1156

deposit deposit Wood charcoal 35 BP± 35AD (0.3348)(0.3348) AD 891–1018 BBurned Burned Wood charcoal1141 ± 1141 BP 898–921 AD 898–921 AD 891–1018 under floor of floor of AD 943–991 (0.6652)(0.6652) under AD 943–991 ceremonial structure ceremonial structure Fill below floor of floor of Wood charcoal (0.0864)(0.0864) AD 773–978 37 BP± 37AD Fill below AD 773–978 AD 781–791 BP 781–791 Wood charcoal1222 ± 1222 ceremonial structure AD 807–897 (0.7710)(0.7710) ceremonial structure AD 807–897 AD 922–940 (0.14267) AD 922–940 (0.14267) Wood charcoal Platform fill under (0.32196) 35 BP± 35AD Wood charcoal1139 ± 1139 Platform fill under AD 898–921 (0.32196)AD 891–1018 A AD 891–1018 BP 898–921 floor floor AD 943–992 (0.6780)(0.6780) AD 943–992

35 BP± 35AD AD 782–789 (0.0130)(0.0130) Wood charcoal BP 891–974 Ceramics FCeramics AD 891–974 AD 782–789 Wood charcoal1180 ± 1180 concentration AD 810–848 (0.6455)(0.6455) concentration AD 810–848 in storage AD 856–990 (0.9224)(0.9224) in facility storage facility AD 856–990 AD 894–1021 Fill within storage Wood charcoal 1129 ± 35 BP (0.2586)(0.2586) AD 894–1021 Fill within storage Wood charcoal 1129 ± 35AD BP 899–919 AD 899–919 facility facility AD 950–997 (0.6312)(0.6312) AD 950–997 AD 1005–1016 (0.1102)(0.1102) AD 1005–1016

Fill within (0.2980)(0.2980) AD 890–1022 38 BP± 38AD Fillstorage within storage Wood charcoal AD 890–1022 Wood charcoal1134 ± 1134 AD 898–921 BP 898–921 facility facility AD 943–994 (0.6737)(0.6737) AD 943–994 AD 1009–1012 (0.0283)(0.0283) AD 1009–1012

Intact Tomb Intact7Tomb 7

Intact4Tomb 4 G' Intact Tomb

HumanHuman bone bone1098 ± 1098 (0.9641)(0.9641) (0.0923)(0.0923) AD 892–1045 42 BP± 42AD AD 892–1045 AD 903–914 BP 903–914 (0.0284)(0.0284) AD 970–1028 (0.9077)(0.9077)AD 1086–1107 AD 1086–1107 AD 970–1028 AD 1121–1127 (0.0076)(0.0076) AD 1121–1127 (0.2403)(0.2403) AD 893–1023 38 BP± 38AD AD 899–918 Intact Tomb AD 893–1023 Wood charcoal1126 ± 1126 BP 899–918 Intact5Tomb 5 Wood charcoal AD 955–955 (0.0098)(0.0098) AD 955–955 AD 961–1017 (0.7450)(0.7450) AD 961–1017 Textile AD 899–919 (0.2586) Intact Tomb 6 1120 ± 35 BP AD 894–1021 Textile Intact Tomb 6 1120 ± 35 BP AD 899–919 (0.2586) AD 894–1021 C AD 950–997 (0.6312)(0.6312) AD 950–997 AD 1005–1016 (0.1102)(0.1102) AD 1005–1016

33 BP± 33AD early early Wood charcoal (0.1293)(0.1293) AD 688–753 BP724–739 D Hearth,Hearth, (0.2636)(0.2636) Wood charcoal1280 ± 1280 AD 724–739 AD 688–753 occupation AD 770–869 (0.8707)(0.8707) AD 759–887 (0.7364)(0.7364) occupation AD 770–869 AD 759–887 of houseof house bone bone1079 ± 1079 Disturbed Tomb 1Tomb 1 HumanHuman 41 BP± 41AD (0.9887)(0.9887)AD 943–1049 (0.7866)(0.7866) BP 975–1042 Disturbed AD 975–1042 AD 943–1049 AD 1094–1096 (0.0102)(0.0102) AD 1080–1142 (0.1539)(0.1539) AD 1094–1096 AD 1080–1142 AD 898–921 (0.5953)(0.5953) AD 898–921

1048 ± 1048 35 BP± 35AD (0.77020) underneath Wood charcoal CAsh lensAsh BP992–1045 AD 992–1045 (0.77020)AD 985–1053 lens underneathWood charcoal (0.5956)(0.5956) AD 985–1053 AD 1087–1106 (0.2120)(0.2120)AD 1064–1067 (0.0060) house floor AD 1087–1106 house floor AD 1064–1067 (0.0060) AD 1123–1126 (0.1785)(0.1785)AD 1075–1148 (0.3984) AD 1123–1126 AD 1075–1148 (0.3984)

Source:Source: McCorman et al. 2004; et al. 1993 McCorman et al. Stuiver 2004; Stuiver et al. 1993 Note: The SHCal04 calibration curve was used the Calib program Note: The SHCal04 calibration curve wasinused in the5.D Calib 5.D program

7 IV

2 III

5

3,6 VI

5 III

4 IV

7 IV

5

3 III

3'3

A-69904 3545 3545 3' A-69904

3 III

23

1,2,3,5,6 12 1,2,3,5,6VI

7 4,5

2 2088 2088 A-69898 A-69898

A-69903 12060 12060 12 A-69903

7 7034 7034 A-69899 A-69899

Unit Unit FeatureFeatureAssociation Level Level Lot # LotLab Association # # LabArea # Area

Table 3.2  Radiocarbon dates from Middle Horizon contexts at Tenahaha

Figure 3.3  Wari Viñaque-style designs found on locally made pots.

and the eleven Middle Horizon metal tupus that we have chemically analyzed look Wari but are made from a copper-arsenic-nickel alloy that corresponds more closely with metals produced only in Bolivia and northern Chile (Lechtman 2005:138; Mora 2007). Wari’s important role in the changing social landscape is reflected in the active manipulation of the style by local artists. The Wari style in Cotahuasi is idiosyncratic—it uses only a limited set of Wari elements, often presents these elements in a more stylized and simplified manner, and combines these elements with local and regional designs in novel ways (Bautista 2006; Jennings 2002; Jennings et al. 2006). Artists seem to have been trying to find their place within a world of greater interregional interaction. This does not mean that Wari styles were drained of their original cosmological significance as they were brought into local schemes. Cotahuasi potters seem to have recognized religiously charged elements of the style and used them in their work (Figure 3.3). Moreover, the association of some of the finest examples of these ceramics with tombs and other ritual spaces 46 

J ustin J ennings

underlines the association of Wari with the sacred in valley (Jennings 2002; Jennings et al. 2006). The important place of Wari styles and ideas in the negotiation of new social roles in the valley is perhaps seen in the prevalence of postfired incision marks during the period (Figure 3.4). Almost completely absent in other periods, we have documented forty-seven different geometric, zoomorphic, and anthropomorphic marks on pots dating to the Middle Horizon. Of the fifty-six vessels found with incision marks, thirty of the vessels were in the locally made Viñaque style (marks were also found on two more vessels in other Wari styles). Although analysis of these marks is ongoing, it is clear that the marks were usually placed on used vessels that were being placed in graves and may have reflected some kind of identity, be it family, larger kin group, religious affiliation, and so forth. If this was the case, then the Wari style was the preferred canvas upon which Cotahuasinos negotiated their new identities. The situation in the valley does not easily fit into either of the common models for the Middle Horizon.

Figure 3.4  Examples of postfired incision marks.

An imperial model does not seem to work. There is no evidence of an imperial presence in Cotahuasi, and the flow of goods and people between the valley and the Wari heartland was limited. At the same time, Wari’s dominant stylistic and cultural influence in the valley, and throughout much of Peru for that matter, is hard to explain using peer-polity models. With its emphasis on the dynamic between global and local processes, a globalization model might best help us understand what happened in the valley. The Middle Horizon was undoubtedly an exciting time for Cotahuasinos. The number of people in the valley was rising, agricultural terracing and mining were transforming the landscape, trade was increasing, and new ideas about art, life, and the cosmos were coming into the valley from Huari, the South Coast, and other valleys in Arequipa. The changes occurring in Cotahuasi during the Late Middle Horizon would have also caused considerable conflict as people negotiated changing relationships with each other, other villages, and the outside world.

Becoming Wari at Tenahaha Embracing a shared Wari culture may have allowed Cotahuasinos to better deal with emerging local social differences during the Late Middle Horizon. The negotiation of this mixed identity was likely a daily process played out, for example, through the decoration on the bowl offered to a guest and the agricultural products brought on the backs of llamas (e.g., Bowser 2000; Flores Ochoa 1979). I believe, however, that daily negotiations were not enough to resolve the deep conflicts that would have been growing as people fought over fundamental issues like land, water, and faith. As in other regions of the world, panvalley feasts and ritual events would have also been needed to bring people together to reaffirm social solidarity (Hayden 2001; Turner 1969). There is evidence for such events occurring in Cotahuasi at the newly constructed site of Tenahaha (see Figure 3.2). Covering about 4 ha, Tenahaha was one of the fifteen new villages settled in the valley during the Middle Horizon. The site was constructed on a narrow river plain that is punctuated by a number of rolling hills.

B ecoming Wari

  47

More than two-thirds of the site has been destroyed by the modern encroachment of agricultural fields, later occupation of a portion of the site by the Incas, and the erosion of the cliff face by the river (Figure 3.5). We excavated 158.5 m2 of the preserved area of the site. The primary focus of the excavations was the collective burials, but we also excavated in domestic, storage, and ritual contexts. We also performed 60 m2 of excavations at a second site, Collota, located 800 m to the east. While three Middle Horizon fragments in the wall fall of the site’s 3 m high perimeter enclosure point to the possibility of the wall being built during that period by the residents of Tenahaha, the occupation of the site occurred only during the Late Horizon (Edwards 2006; Jennings and Yépez Álvarez 2009a). Tenahaha’s location is unusual because it is the only site from the Middle Horizon built on the valley bottom. From a subsistence perspective, Tenahaha was located in an awful place because the land around the site was not irrigated until the Inca conquest of the region (Jennings and Yépez Álvarez 2009a). The prevailing settlement logic before, during, and after the Middle Horizon was to place settlements halfway up the valley’s flanks in order to exploit spring-fed agricultural terraces and high grassland grazing areas. There were a few rain-fed terraces next to the river that date to the Middle Horizon, but agricultural yields would have been much lower than yields from the spring-fed agriculture above. Occupation layers in all Middle Horizon contexts at Tenahaha are shallow—there is usually a single floor of 8–12 cm, followed by 20–30 cm of subfloor fill resting on sterile soil. The shallow occupation of the site, combined with the narrow date range gleaned from the ceramic chronology and radiocarbon dates (see Table 3.2), suggests that the site was occupied for a short time during the Late Middle Horizon. The abandonment of the site following the Middle Horizon may be linked in part to its poor subsistence base (Jennings 2006:363–365). There was, however, a good reason to place Tenahaha in this section of the river bottom in light of the need at the time for a neutral gathering place. The site was removed from all other villages and their associated terraces, was located conveniently near the valley’s geographic center, and sat along what was likely the most important 48 

J ustin J ennings

exchange and communications route through the valley during the period (Jennings 2006:362–364). The portion of the site that remains largely intact reveals a very light resident occupation during the Late Middle Horizon relative to the site’s size and number of structures. We excavated two houses, and the culinary equipment, paleobotanical remains, hearths, and middens from the homes suggest an occupation of short duration (Capriata 2006). We had hoped to find other houses, but instead we found storage and ritual areas that seem too big for the people that lived in the preserved section of the site. To the northwest of the houses was a row of storage facilities that residents were expanding at the time of the abandonment of the site at the end of the Middle Horizon. The two excavated storage units measured 1 × 2 m. They were sunk about 30 cm below floor level and contained a variety of ceramics, lithics, and camelid bones. We also dug a pair of large open-aired structures with stamped-earth floors free of artifacts. With associated food preparation and consumption debris and burned offerings of ceramics and camelid bone that were sealed under their stamped-earth floors, the structures could be interpreted as spaces dedicated to rituals, feasting, and dancing. The sense that Tenahaha was built for the periodic gathering of large groups of people is further supported by the large number of people buried at the site. Although the resident population was low, the site is ringed by dozens of collective tombs. We dug seven of these tombs and recovered the remains of at least 171 individuals (Jennings et al. 2006). The rectangular tombs were plastered, stone-walled, semisubterranean structures with a single entrance covered by a stone that could be easily rolled away. The tombs were opened repeatedly in antiquity, and there is evidence for artifacts, and perhaps individuals, being moved between tombs and for new objects and individuals being deposited. Pieces of one vessel, for example, were found in Tombs 2, 3, and 4 (Tombs 3 and 4 were intact, while Tomb 2 was partially looted) (Jennings et al. 2006). Individuals were usually buried in a seated and flexed position with offerings placed at their feet, or more rarely on their heads (Figure 3.6). Over time, individuals were pushed to the sides of the tombs (and parts were

Figure 3.5  Site plan of Tenahaha. The approximate boundaries of funerary and domestic zones are marked on the image below while the plan above shows excavations carried out in the preserved section of the site.

 

Figure 3.6  Three ceramic bowls placed as offerings near the feet of an individual in Tomb 3.

likely removed) to make room for other individuals. The number of dead buried at Tenahaha exceeds what one would expect from the small number of people that inhabited the short-lived site. Instead, it is likely that the dead were brought from other places in the valley to be buried and honored. Burial practices in the tombs likely distinguished between age, gender, status, and group affiliation. Infants, for example, were often interred separately from adults. We uncovered the remains of twelve infants, without associated grave goods, that were placed inside of jars in Tomb 4. In a second example, 23 percent of the seventeen skulls had tabular modifications in Tomb 1—a percentage much higher than any of the other tombs that we excavated (Mummert 2006:103–104). We also found a woman in Tomb 6 who was buried wearing conspicuous wealth for the valley: a greenstone and Spondylus shell necklace. Despite these differences, all of the tombs contained both local and Wari-style ceramics. 50 

J ustin J ennings

Our excavations at Tenahaha suggest that the site served as a major ceremonial center and necropolis for the valley. When people in the Andes come together in groups today, they often celebrate both their commonalities and their differences. When a few friends drink alcohol together, they share a single glass, but the host determines who drinks and in what order they drink (Allen 2002:118–120). Neighboring communities often get together around Carnival to fight, dance, and make love (Cama Ttito and Ttito Tica 2003), and villages across the highlands of central Peru make an annual pilgrimage to Qoyllur Rit’i to celebrate a common faith and to compete in a spirited battle of bands (Salnow 1987). I suggest that the burials, feasts, and other events that likely took place at Tenahaha during the Middle Horizon functioned in a similar manner. During a time of great social change, people came together to work out their social differences within the milieu of a shared culture that included Wari elements.

The development of the Wari State in Ayacucho spurred a significant increase in interaction across Peru. While few goods actually moved between Cotahuasi and Wari, the increased flow of goods, people, and ideas in the region caused significant cultural transformations to occur in Cotahuasi, which began to accentuate social differences. Elements of Wari global culture acted as social glue and that helped the people of the valley come together as a community in the ceremonies performed at the ritual center of Tenahaha. In a sense, people in the valley were becoming Wari by embracing the Wari style and many of the ideas associated with the style. Yet at the same time, they were becoming Cotahuasinos as the ceremonies at Tenahaha cemented a panvalley identity in opposition to the global culture of Wari. They were also forging class, village, age, and gender identities relative to each other as they danced, drank, and mourned.

Wari Globalization The extent of modern globalization is unique in world history, but the experience is not. The spread of the world’s first civilizations caused sweeping cultural changes that were broadly similar to the ones that are occurring today. Ancient farmers, of course, were not dealing with the latest rock video, a new highway, or the effects of ethanol production on the price of corn. Yet a farmer in first-millennium BC Mexico during the Olmec horizon would have been dealing with new cosmologies, exotic goods at the market, and new people coming down the trail. The visceral experience of living in a new, rapidly changing, and perhaps out of control world may have been similar in both eras. Although archaeologists recognize that many of the same cultural changes were occurring during the spread of the first civilizations, they often hesitate to explicitly compare the two eras. By confining our understanding of globalization to a Western creation of the last five hundred years, we needlessly divorce the modern from the premodern and warp our understanding of both the past and the world that we live in today. Archaeologists have struggled to fit peer-polity or imperial models to what occurred in Peru during the Middle Horizon. Despite several decades of nipping and tucking, the models still sit unconvincingly on the

available evidence. I argue that the spread of the Wari culture during the Middle Horizon is best conceived as an example of early globalization. The rapid growth of the city of Huari triggered cascades of interregional interactions that created complex connectivity. From this connectivity emerged a Wari global culture that was a combination of ideas generated at Huari and ideas from across much of Peru. The cultural changes that occurred during the Middle Horizon depended on local conditions and the kinds of interregional interactions that were occurring between different places. Some places became more Wari in one way, some became more Wari in another way, some people rejected Wari and strengthened local traditions, and other people changed little because they lived in relative isolation from the Wari phenomena. The Middle Horizon in Cotahuasi was marked by rapid population increase, increased trade, rising social inequality, and widespread acceptance of the Wari style. The Wari global culture was mediated through local conditions as potters played with designs and individuals scratched rectangles, llamas, and other designs on the pots after they were fired. The Wari style and associated ideas likely provided a frame of reference that helped the people of the valley to understand their changing world and to relate to people in other valleys with whom they were coming into increasing contact. At the same time, Wari culture provided the backdrop upon which Cotahuasinos negotiated their relationships with each other. Their struggle over multiple, shifting, and overlapping identities during the Middle Horizon is not far removed from the conflicts that we face today in defining who we are in the world.

Endnotes 1.  My application of a globalization model to Wari was inspired in part by my reading of Richard Wilk’s article on globalization, the Olmec, and beauty pageants (2004). 2.  Unfortunately, Willy Yépez Álvarez and I have created considerable confusion regarding the sites of Tenahaha and Collota, a site 800 m to the east of Tenahaha. In my dissertation, the sites were treated separately, but the site of Tenahaha was called “Netahaha” ( Jennings 2002). This terminology was also used in two earlier publications ( Jennings and Yépez Álvarez 2001a, 2001b). When we mapped the sites in 2004, we decided to make

B ecoming Wari

  51

“Netahaha” a subsector of Collota as part of an eventually successful campaign to get the site recognized as a national cultural patrimony site. This way of describing the sites was used in two subsequent papers ( Jennings 2006a, 2006b). In the summer of 2005, local informants told us that the proper name of the site was Tenahaha, and excavations at Collota revealed that the site, with the important possible exception of the enclosure wall, was built in the Late Horizon ( Jennings et al. 2006; also see Edwards 2006). These new insights led us to split the sites again and to change the name of the site of Netahaha to “Tenahaha.” These new terminologies are seen in the most recent publications ( Jennings and Yépez Álvarez 2009a, 2009b). I apologize for these changes over the last few years, and I regret that it has taken us this long to get it right.

References Cited Abu-Lughod, Janet L. 1989 Before European Hegemony: The World System, AD 1250–1350. Oxford University Press, New York. Algaze, Guillermo 1993 The Uruk World System. University of Chicago Press, Chicago. Allen, Catherine J. 2002 The Hold Life Has: Coca and Cultural Identity in an Andean Community. Smithsonian Institution Press, Washington, D.C. Angeles, Rommel, and Denise Pozzi-Escot 2001 Textiles del Horizonte Medio. Las evidencias de Huaca Malena, valle de Asia. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William Isbell, pp. 401–424. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Appadurai, Arjun 1996 Modernity at Large. University of Minnesota Press, Minneapolis. Bandy, Matthew S. 2004 Fissioning, Scalar Stress, and Social Evolution in Early Village Societies. American Anthropologist 106(2):322–333. Bautista, Stefanie L. 2006 Evidence of Wari Influence in the Cotahuasi Valley, Peru: A Ceramic Analysis of the Mortuary Vessels from Tomb 3. Senior honor’s thesis, University of California, Santa Barbara. Blanton, Richard, and Gary Feinman 1984 The Mesoamerican World System. American Anthropologist 86:673–682. Bowser, Brenda 2000 From Pottery to Politics: An Ethnoachaeological Study of Political Factionalism, Ethnicity, and Domestic Pottery Style in the Ecuadorian Amazon. Journal of Anthropological Method and Theory 7(3):219–248. Browman, David 1999 Wari Impact on the Upper Mantaro Basin. Paper presented at the 64th Annual Meeting of the Society for American Archaeology, Chicago. Burger, Richard L., and Frank Asaro 1979 Análisis de rasgos significativos en la obsidiana de los Andes centrales. Revista del Museo Nacional 43:281–326.

52 

J ustin J ennings

Burger, Richard L., Karen L. Mohr Chávez, and Sergio J. Chávez 2000 Through the Glass Darkly: Prehispanic Obsidian Procurement and Exchange in Southern Peru and Northern Bolivia. Journal of World Prehistory 14:267–362. Cabera Romero, Martha 1996 Unidades habitacionales, iconografía, y rituales en un poblado rural de la época Huari. Unpublished thesis presented for the license of archaeology to the Univerdidad Nacional de San Critobal de Huamanga, Ayacucho, Peru. Cama Ttito, Máximo, and Alejandra Ttito Tica 2003 Batalas rituales: Tupay o tinkuy en Chiaraje y Tocto. In Ritos de competición en los Andes: Luchas y contiendas en el Cuzco, edited by Alejandro Ortiz Rescaniere, pp 19–50. Fondo Editorial Pontificia Universidad Católica del Perú, Lima. Capriata Estrada, Camila 2006 Excavaciones en Tenahaha, sitio de Collota. Cotahuasi, Perú. Pre-Professional Practicals Report, Pontificia Universidad Católica del Perú, Lima. Castells, Manuel 1996 The Rise of Network Society: The Information Age: Economy, Society and Culture 1. Blackwell, Oxford. Castillo, Luis Jaime 2001 La presencia de Wari en San José de Moro. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William Isbell, pp. 143–179. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Chase-Dunn, Christopher, and E. N. Anderson 2005 The Historical Evolution of World-Systems. Palgrave Macmillan, New York. Chase-Dunn, Christopher, and Thomas D. Hall 1997 Rise and Demise: Comparing World-Systems. Westview Press, Boulder, Colorado. Conklin, William J. 1982 The Information System of the Middle Horizon Quipus. In Ethnoastronomy and Archaeoastronomy in the American Tropics, edited by Anthony F. Aveni and Gary Urton, pp. 261–281. New York Academy of Sciences, New York. Conlee, Christina A., Jalh Dulanto, Carol J. Mackey, and Charles Stanish 2004 Late Prehispanic Sociopolitical Complexity. In Andean Archaeology, edited by Helaine Silverman, pp. 209–236. Blackwell, Malden, Massachusetts. Cook, Anita 1994 Wari y Tiwanaku: Entre el estilo y la imagen. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. 2001 Huari D-Shaped Structures, Sacrifical Offerings, and Divine Rulership. In Ritual Sacrifice in Ancient Peru, edited by Elizabeth P. Benson and Anita G. Cook, pp. 137–163. University of Texas Press, Austin. Dietler, Michael 1990 Driven by Drink: The Role of Drinking in the Political Economy and the Case of Early Iron Age France. Journal of Anthropological Archaeology 9:352–406. 2005 The Archaeology of Colonization and the Colonization of Archaeology: Theoretical Challenges from an Ancient Mediterranean Colonial Encounter. In The Archaeology of Colonial Encounters: Comparative Perspectives, edited by Gill J.

Stein, pp. 33–68. School of American Research Press, Santa Fe, New Mexico. Domínguez, Adolfo J. 2002 Greeks in Iberia: Colonialism without Colonization. In The Archaeology of Colonialism, edited by Claire L. Lyons, C. L. Papadopoulos, and John K. Papadopoulos, pp. 65–95. Getty Research Institute, Los Angeles. Edwards, Matthew 2006 Collota: An Inca Site in Peru. Master’s thesis, University of California, Santa Barbara. Ekholm, Kajsa, and Jonathan Friedman 1979 “Capital” Imperialism and Exploitation in Ancient World Systems. In Power and Propaganda, edited by Mogens T. Larsen, pp. 41–58. Akademish, Copenhagen. 1985 Towards a Global Anthropology. Critique of Anthropology 5(1):97–119. el-Ojeili, Chamsy, and Patrick Hayden 2006 Critical Theories of Globalization. Palgrave Macmillan, New York. Eriksen, Thomas Hylland 2007 Globalization. Berg, New York. Featherstone, Mike (editor) 1990 Global Culture: Nationalism, Globalization, and Modernity. Sage Books, New York. Fienup-Riordan, Ann 1990 Eskimo Essays: Yup’ik Lives and How We See Them. Rutgers University Press, New Brunswick, New Jersey. Flores Ochoa, Jorge A. 1979 Pastoralists of the Andes: The Alpaca Herders of Paratia, translated by Ralph Bolton. Institute for the Study of Human Issues, Philadelphia. Frank, André Gunder 1993 The World System: Five Hundred Years or Five Thousand? Routledge, London. Frank, André Gunder, and Barry K. Gills 2000 The Five Thousand Year World System in Theory and Praxis. In World System History: The Social Science of Long-Term Change, edited by Robert A. Denemark, R. A., Jonathan Friedman, Barry K. Gills, and George Modelski, pp. 3–23. Routledge, New York. Friesen, T. Max 1999 Resource Structure, Scalar Stress, and the Development of Inuit Social Organization. World Archaeology 31(1):21–37. Giulianotti, Richard, and Roland Robertson 2007 Forms of Glocalization: Globalization and the Migration Strategies of Scottish Football Fans in North America. Sociology 41(1):133–152. Glowacki, Mary 2002 The Huaro Archaeological Site Complex: Rethinking the Huari Occupation of Cuzco. In Andean Archaeology I: Variations in Sociopolitical Organization, edited by William H. Isbell and Helaine Silverman, pp. 267–285. Kluwer Academic/ Plenum, New York. Glowacki, Mary, and Gordon F. McEwan 2001 Pikillacta, Huaro y la gran región del Cuzco: Nuevas interpretaciones de la ocupación Wari de la sierra sur. Huari y Tiwanaku: Modelos vs. evidencias. Segundo parte, edited by Peter Kaulicke and William Isbell, pp. 31–49. Boletín de

Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Goldstein, David J., Robin C. Goldstein-Coleman, and Patrick Ryan Williams 2009 You Are What You Drink: A Sociocultural Reconstruction of Pre-Hispanic Fermented Beverage Use at Cerro Baúl, Moquegua, Peru. In Drink, Power, and Society in the Andes, edited by Justin Jennings and Brenda Bowser, pp. 133–166. University Press of Florida, Gainesville. Goldstein, Paul 2005 Andean Diaspora: The Tiwanaku Colonies and the Origins of South American Empire. University Press of Florida, Gainesville. González Carré, Enrique, Enrique Bragayrac Dávila, Cirilo Vivanco Pomacanchari, Vera Tiesler Blos, and Máximo López Quispe 1996 El Templo Mayor en la ciudad de Wari: Estudios arqueológicos en Vegachayoq Moqo-Ayacucho. Facultad de Ciencias Sociales, Universidad Nacional de San Cristóbal de Huamanga, Ayacucho, Peru. Green, Linda 2002 Notes on Mayan Youth and Rural Industrialization in Guatemala. In The Anthropology of Globalization: A Reader, edited by Jonathan Xavier Inda and Renato Rosaldo, pp. 101– 120. Blackwell, Malden, Massachusetts. Hall, Thomas D., and Christopher Chase-Dunn 2006 Global Social Change in the Long Run. In Global Social Change: Historical and Comparative Perspectives, edited by Christopher Chase-Dunn and Salvatore J. Babones, pp. 33–58. John Hopkins University Press, Baltimore, Maryland. Hayden, Brian 2001 Fabulous Feasts: A Prolegomenon to the Importance of Feasting. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 23–64. Smithsonian Institution Press, Washington, D.C. Heggarty, Paul 2008 Linguistics for Archaeologists: A Case-Study in the Andes. Cambridge Archaeological Journal 18(1):35–56. Held, David, and Anthony McGrew 2000 The Great Globalization Debate: An Introduction. In The Global Transformations Reader: An Introduction to the Globalization Debate, edited by David Held and Anthony McGrew, pp. 1–45. Polity Press, Malden, Massachusetts. Hobsbawm, Eric, and Terrence Ranger (editors) 1983 The Invention of Tradition. Cambridge University Press, Cambridge. Inda, Jonathan Xavier, and Renato Rosaldo 2008 Tracking Global Flows. In The Anthropology of Globalization: A Reader, 2nd ed., edited by Jonathan Xavier Inda and Renato Rosaldo, pp. 3–46. Blackwell, Malden, Massachusetts. Isbell, William H. 1977 The Rural Foundations of Urbanism. Illinois Studies in Anthropology 10. University of Illinois, Chicago. 1989 Honcopampa: Was It a Huari Administrative Center? In The Nature of Wari, edited by R. M. Czwarno, F. M. Meddens, and A. Morgan, pp. 98–114. BAR International Series 525. British Archaeological Reports, Oxford.

B ecoming Wari

  53

1991 Huari Administration and the Orthogonal Cellular Architecture Horizon. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 293–316. Dumbarton Oaks, Washington, D.C. 1997 Mummies and Mortuary Monuments: A Postprocessual Prehistory of Central Andean Social Organization. Austin, University of Texas Press. 2001 Huari: Crecimiento y desarrollo de la capital imperial. In Wari: Arte precolombino peruano, edited by Luis Millones, pp. 99–172. Fundacíon el Monte, Seville. Isbell, William H., Christine Brewster-Wray, and Lynda E. Spickard 1991 Architecture and Spatial Organization at Huari. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 19–54. Dumbarton Oaks, Washington, D.C. Isbell, William H., and Katharina J. Schreiber 1978 Was Huari a State? American Antiquity 43(3):372–389. Janusek, John Wayne 2008 Ancient Tiwanaku. Cambridge University Press, New York. Jennings, Justin 2002 Prehistoric Imperialism and Cultural Development in the Cotahuasi Valley, Peru. Unpublished Ph.D. dissertation, University of California, Santa Barbara. 2006a Understanding Middle Horizon Peru: Hermeneutic Spirals, Interpretative Traditions, and Wari Administrative Centers. Latin American Antiquity 16:265–286. 2006b Core, Peripheries, and Regional Realities in Middle Horizon Peru. Journal of Anthropological Archaeology 25:346–370. 2011 Globalizations and the Ancient World. Cambridge University Press, New York. Jennings, Justin, and Nathan Craig 2001 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 20:479–502. Jennings, Justin, and Willy Yépez Álvarez 2001a Collota, Netahaha, y el desarrollo del poder Wari en el valle de Cotahuasi, Arequipa, Perú. In Huari y Tiwanaku: Modelos vs. evidencias. Segundo parte, edited by Peter Kaulicke and William Isbell, pp. 13–29. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. 2001b Architecture, Local Elites, and Imperial Entanglements: The Impact of the Wari Empire on the Cotahuasi Valley of Peru. Journal of Field Archaeology 28(1/2):143–159. 2009a The Inca Conquest and Consolidation of the Cotahuasi Valley of Southern Peru. Ñawpa Pacha 29:119–152. 2009b El Intermedio Tardío en el valle de Cotahuasi, Peru. In Arqueología del área centro sur andina: Actas del Simposio Internacional 30 de Junio–2 de Julio de 2004, Arequipa Perú, edited by Mariusz Ziolkowski, Justin Jennings, and Luis Augusto Belen Franco, pp. 481–500. Boletín del Centro de Estudios Precolombinos No. 7. University of Warsaw, Warsaw. Jennings, Justin, Willy Yépez Álvarez, and Corina Kellner 2006 Tumbas de Tenahaha: Notas preliminares sobre contextos funerarios del Horizonte Medio en el valle de Cotahuasi. Condesuyos: Boletín de la Misión Arqueológica Andina 6:93–108.

54 

J ustin J ennings

Johnson, Gregory A. 1982 Organizational Structure and Scalar Stress. In Theory and Explanation in Archaeology, edited by Colin Renfrew, Michael Rowlands, and Brian A. Seagraves, pp. 389–421. Academic Press, New York. Kaulicke, Peter 2000 La sombra de Pachacamac: Huari en la costa central. Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William Isbell, pp. 181–217. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Kellner, Corina M., and Margaret J. Schoeninger 2008 Wari’s Imperial Influence on Local Nasca Diet: The Stable Isotope Evidence. Journal of Anthropological Archaeology 27:226–243. Knobloch, Patricia J. 1983 A Study of the Andean Huari Ceramics from the Early Intermediate Period to the Middle Horizon Epoch 1. Unpublished Ph.D. dissertation, State University of New York, Binghamton. 1991 Stylistic Date of Ceramics from the Huari Centers. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 247–258. Dumbarton Oaks, Washington, D.C. 2010 An Early Intermediate Period Deposit of Huarpa Style Ceramics from the Site of Huari, Department of Ayacucho, Perú. Ñawpa Pacha 30, in press. Kohl, Philip 2007 The Making of Bronze Age Eurasia. Cambridge University Press, New York. LaBianca, Øystein S., and Sandra Arnold Scham 2006 Connectivity in Antiquity: Globalization as Long-Term Historical Process. Equinox, London. Lau, George 2005 Core-Periphery Relations in the Recuay Hinterlands: Economic Interaction at Chinchawas. Antiquity 79:78–99. 2006 Northern Exposures: Recuay-Cajamarca Boundaries and Interaction. In Andean Archaeology III: North and South, edited by William H. Isbell and Helaine Silverman, pp. 143– 170. Springer Science, New York. Lechtman, Heather 1980 The Central Andes: Metallurgy without Iron. In The Coming of the Age of Iron, edited by Theodore A. Wertime and James D. Muhly, pp. 267–334. Yale University Press, New Haven, Connecticut. 2003 Middle Horizon Bronze: Centers and Outliers. In Patterns and Process: A Festschrift in Honor of Dr. Edward V. Sayre, edited by Lambertus van Zelst, pp. 248–268. Smithsonian Center for Materials Research and Education, Suitland, Maryland. 2005 Arsenic Bronze at Pikillacta. In Pikillacta: The Wari Empire in Cuzco, edited by Gordon McEwan, pp. 131–146. University of Iowa Press, Iowa City. Leoni, Juan B. 2006 Ritual and Society in Early Intermediate Period Ayacucho: A View from the Site of Ñawinpukypo. In Andean Archaeology III: North and South, edited by William H. Isbell and Helaine Silverman, pp. 279–306. Springer, New York.

Lesure, Richard E. 2004 Shared Art Styles and Long-Distance Contact in Early Mesoamerica. In Mesoamerican Archaeology, edited by Julie A. Hendon and Rosemary A. Joyce, pp. 73–96. Blackwell, Malden, Massachusetts. Lumbreras, Luis 1974 The Peoples and Cultures of Ancient Peru. Smithsonian University Press, Washington, D.C. MacNeish, Richard S., Angel García Cook, Luis G. Lumbreras, Robert K. Vierra, and Antoinette Nelken-Turner 1981 Prehistory of the Ayacucho Basin, Peru. Volume II, Excavations and Chronology. University of Michigan Press, Ann Arbor. McCormac, F. G., A. G. Hogg, P. G. Blackwell, C. E. Buck, T. F. G. Higham, and P. J. Reimer 2004 SHCa104 Southern Hemisphere Calibration 0–1000 cal BP. Radiocarbon 46:1087–1092. McEwan, Gordon F. 1990 Some Formal Correspondences between the Imperial Architecture of the Wari and Chimu Cultures of Ancient Peru. Latin American Antiquity 1(2):97–116. 1991 Investigations at the Pikillacta Site: A Provincial Huari Center in the Valley of Cuzco. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 93–120. Dumbarton Oaks, Washington, D.C. 1996 Archaeological Investigations at Pikillacta, a Wari Site in Peru. Journal of Field Archaeology 23(2):169–186. Malpass, Michael 2001 Sonay: Un centro Wari celular orthogonal en el valle de Camaná, Perú. In Huari y Tiwanaku: Modelos vs. evidencias. Segundo parte, edited by Peter Kaulicke and William Isbell, pp. 51–68. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Menzel, Dorothy 1964 Style and Time in the Middle Horizon. Ñawpa Pacha 2:1–106. 1977 The Archaeology of Ancient Peru and the Work of Max Uhle. University of California, Berkeley. Mora, Franco 2007 Análisis metalográfico de cuatro tupus de Tenahaha, un sitio Wari en el Valle de Cotahuasi. Senior honor’s thesis, Pontificia Universidad Católica del Perú, Lima. Mummert, Amanda 2006 A Bioarchaeological Analysis of Middle Horizon Peruvian Tombs. Senior honor’s thesis, University of California, Santa Barbara. Nederveen Pieterse, Jan 2004 Globalization and Culture. Rowman and Littlefield, Lanham, Maryland. Neira Avendaño, Máximo 1998 Arqueología de Arequipa. Chronos 1(1):9–50. Neira Avendaño, Máximo, and Augusto Cardonas Rosas 2001 El Periodo Formativo en el area de Arequipa. Boletín de la Misión Arqueológica Andina 3:27–60. Ochatoma Paravicino, José, and Martha Cabrera Romero 2001 Poblados rurales Huari: Una vision desde Aqo Wayqo. CANO, Lima.

2002 Religious Ideology and Military Organization in the Icon­ ography of a D-shaped Ceremonial Precinct at Concho­pata. In Andean Archaeology II: Art, Landscape and Society, edited by Helaine Silverman and William H. Isbell, pp. 225–248. Kluwer Academic/Plenum, New York. Orefici, Giuseppe, and Andrea Drusini 2003 Nasca: Hipótesis y evidencias de su desarollo cultural. Centro Italiano Studi e Ricerche Archeologiche Precolombiane, Lima. Pauketat, Timothy R., and Thomas E. Emerson (editors) 1997 Cahokia: Domination and Ideology in the Mississippian World. University of Nebraska Press, Lincoln. Pérez Calderón, Ismael 1995 Excavaciones de salvataje y propuesta sobre conservación de estructuras descubiertas, area museo de sitio en Wari. Convenio Instituto Nacional de Cultura—Universidad Nacional San Cristóbal de Huamanga, Ayacucho, Peru. 1999 Huari: Misteriosa ciudad de piedra. Universidad Nacional de San Cristóbal de Huamanga, Ayacucho, Peru. Raymond, J. Scott 1992 Highland Colonization of the Peruvian Montana in Relation to the Political Economy of the Huari Empire. Journal of the Steward Anthropological Society 20(1–2):17–36. Raymond, J. Scott, and William H. Isbell 1969 Cultural Remains in the Pampas River Valley, Peru. Paper presented at the 34th Annual Meeting of the Society for American Archaeology, Milwaukee. Robertson, Roland 1992 Globalization: Social Theory and Global Culture. Sage, London. Rothman, Mitchell S. (editor) 2001 Uruk Mesopotamia and Its Neighbors. Cross-Cultural Interactions in the Era of State Formation. School of American Research Press, Santa Fe, New Mexico. Salnow, Michael 1987 Pilgrims of the High Andes: Religious Cults in Cuzco. Smithsonian Institution Press, Washington, D.C. Schneider, Jane 1977 Was There a Pre-Capitalist World System? Peasant Studies 6:20–29. Schreiber, Katharina 1992 Wari Imperialism in Middle Horizon Peru. Anthropological Papers 87. University of Michigan, Ann Arbor. 1999 Regional Approaches to the Study of Prehistoric Empires: Examples from Ayacucho and Nasca, Peru. In Settlement Pattern Studies in the Americas: Fifty Years since Viru, edited by Brian Billman and Gary Feinman, pp. 160–171. Smithsonian Institution Press, Washington, D.C. 2001 The Wari Empire of Middle Horizon Peru: The Epistemological Challenge of Documenting an Empire without Documentary Evidence. In Empires: Perspectives from Archaeology and History, edited by Susan E. Alcock, Terrence N. D’Altroy, Kathleen D. Morrison, and Carla M. Sinopoli, pp. 70–92. Cambridge University Press, New York. Shady Solís, Ruth 1982 La Cultura Nievería y la interacción social en el mundo andino en la epoca Huari. Arqueológicas 19:15–108. 1988 Época Huari como interacción de las sociedades regionales. Revista Andina 6(1):67–99.

B ecoming Wari

  55

Shimada, Izumi 1985 Perception, Procurement, and Management of Resources: Archaeological Perspective. In Andean Ecology and Civilization, edited by Shozo Masuda, Izumi Shimada, and Craig Morris, pp. 357–399. University of Tokyo, Tokyo. 2000 The Late Prehispanic Coastal States. In The Inca World: The Development of Pre-Columbian Peru, A.D. 1000–1534, edited by Laura Laurencich-Minelli, pp. 49–110. University of Oklahoma Press, Norman. Silverman, Helaine, and Donald A. Proulx 2002 The Nasca. Blackwell, Malden, Massachusetts. Sklair, Leslie 2002 Globalization: Capitalism and Its Alternatives. Oxford University Press, Oxford. 2006 Competing Conceptions of Globalization. In Global Social Change: Historical and Comparative Perspectives, edited by Christopher Chase-Dunn and Salvatore J. Babones, pp. 59–78. John Hopkins University Press, Baltimore, Maryland. Stein, Gill J. 1999 Rethinking World-Systems: Diasporas, Colonies, and Interaction in Uruk Mesopotamia. University of Arizona Press, Tucson. 2005a The Political Economy of Mesopotamian Colonial Encounters. In The Archaeology of Colonial Encounters: Comparative Perspectives, edited by Gill J. Stein, pp. 33–68. School of American Research Press, Santa Fe, New Mexico. Stein, Gill J. (editor) 2005b The Archaeology of Colonial Encounters: Comparative Perspectives. School of American Research Press, Santa Fe, New Mexico. Stuiver, M., P. J. Reimer, and R. W. Reimer 1993 Extended 14C Database and Revised CALIB Radiocarbon Calibration Program. Radiocarbon 35:215–230. Tilly, Charles 1995 Popular Culture in Great Britain, 1758–1834. Harvard University Press, Cambridge, Massachusetts. Tomlinson, John 1999 Globalization and Culture. University of Chicago Press, Chicago. Topic, John R. 1991 Huari and Huamachuco. In Huari Administrative Structures, edited by William Isbell and Gordon McEwan, pp. 141–164. Dumbarton Oaks, Washington, D.C. 2003 From Stewards to Bureaucrats: Architecture and Information Flow at Chan Chan, Peru. Latin American Antiquity 14(3):243–274. Topic, Theresa L., and John R. Topic 2000 Hacia la comprensión del fenómeno Huari: Una perspectiva norteña. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William Isbell, pp. 181–217. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Torero, Alfredo 1974 El Quechua y la historia social andina. Dirección Universitaria de Investigación. Universitario Ricardo Palma, Lima.

56 

J ustin J ennings

Trawick, Paul 2003 The Struggle for Water in Peru: Comedy and Tragedy in the Andean Commons. Stanford University Press, Stanford, California. Tung, Tiffany A. 2008 Dismembering Bodies for Display: A Bioarchaeological Study of Trophy Heads from the Wari Site of Conchopata, Peru. American Journal of Physical Anthropology 136:294–308. Tung, Tiffany A., and Bruce Owen 2006 Violence and Rural Lifeways at Two Peripheral Sites in the Majes Valley of Southern Peru. In Andean Archaeology III: North and South, edited by William H. Isbell and Helaine Silverman, pp. 435–467. Springer, New York. Turner, Victor W. 1969 The Ritual Process: Structure and Anti-Structure. Aldine, Chicago. Vaughn, Kevin J. 2004 Households, Crafts, and Feasting in the Ancient Andes: The Village Context of Early Nasca Craft Consumption. Latin American Antiquity 15(1):61–88. Vivanco, Cirilo, and Lidio Valdez 1993 Poblados Wari en la cuenca del pampas—Qaracha, Ayacucho. Gaceta Arqueológica Andina 7(23):83–102. Wallerstein, Immanuel 1974 The Modern World System I. Academic Press, San Diego, California. Wells, Peter S. 1999 The Barbarian Speaks: How the Conquered Peoples Shaped Roman Europe. Princeton University Press, Princeton, New Jersey. Wilk, Richard 1995 Learning to Be Local in Belize: Global Systems of Common Difference. In Worlds Apart: Modernity through the Prism of the Local, edited by Daniel Miller, pp. 110–133. Routledge, New York. 2004 Miss Universe, the Olmec and the Valley of Oaxaca. Journal of Social Archaeology 4(1):81–98. Williams, Carlos, and Jose Pineda 1985 Desde Ayacucho hasta Cajamarca: Formas arquitectónicas con filiación Wari, unidad del espacio andino. Boletín de Lima 7(40):55–61. Williams, Patrick Ryan 2002 Rethinking Disaster-Induced Collapse in the Demise of the Andean Highland States: Wari and Tiwanaku. World Archaeology 33(3):361–374. Williams, Patrick Ryan, and Johny Isla 2002 Investigaciones arqueológicas en Cerro Baúl, un enclave Wari en el valle de Moquegua. Gaceta Arqueológica Andina 26:87– 120. Williams, Patrick Ryan, and Donna J. Nash 2002 Imperial Interaction in the Andes: Huari and Tiwanaku at Cerro Baúl. In Andean Archaeology I: Variations in Sociopolitical Organization, edited by William H. Isbell and Helaine Silverman, pp. 243–265. Kluwer Academic/Plenum, New York.

Ch a p t er 4

o Wari in the Majes-Camaná Valley A Different Kind of Horizon Bruce Owen

T

his chapter reviews evidence of the Wari material culture horizon from the MajesCamaná Valley (Figure 4.1), focusing on the initial expansion of this horizon into the region. I propose a framework for inferring social processes from the ceramics associated with a horizon style in a given region. Finally, I assess hypotheses about Middle Hori­ zon social processes in the Majes-Camaná region and perhaps beyond. The Majes-Camaná drainage lies 370 km from Huari, farther away than Pachacamac or Pikillacta. Despite the great distance from Ayacucho, provincial Wari ceramics were apparently already ubiquitous at one Majes site, and possibly at many, by early in the Middle Horizon. In addition, the patternings of corporate versus folk ceramic styles, finewares versus utilitarian wares, and local styles versus those related to a distant center differ from those implied by the usual models of Wari, Tiwanaku, or Inca material culture horizons. These differences in material patterning presumably correspond to differences in the social processes that produced them (cf. Isbell and Silverman 2006:507–510). This patterning suggests that Wari material culture in Majes and Camaná did not result from direct or indirect imperial control along Inca lines (Jennings and Craig 2001; Schreiber 1987, 1992), nor from grassroots direct colonization

such as Goldstein (2005) reconstructs for Tiwanaku in Moquegua, but instead from a third mode of geographic expansion of people and their material culture.

Categories of Ceramics Moseley defines corporate styles as those “characteristic of particular polities, religions, and organizations,” in which design and iconography are “geared to serve corporate ends” and “convey . . . corporate symbolism” (2001:78–79). Objects in a corporate style are produced by workers who are subsidized and directed by corporate institutions, which then control the distribution of the products (Moseley 2001:78–79). Corporate styles are likely to be visually distinctive and iconographically complex, conveying messages ranging from general association with the institution to specific religious, political, or social ideas. They are likely to be of high technical quality, fine craftsmanship, and high labor cost, since they should convey a positive impression of their institutional source. They may be relatively standardized, both to make them readily identifiable and interpretable and because controlled design, subsidized production, and efficiency concerns may all encourage standardization. They may embody stylistic or geochemical evidence of production in a limited number of controlled centers.

  57

Cajamarquilla

Wari influence (conventional)

Lima Pahacamac Azangaro Huari Ayacucho Jargampata Aqo Wayqo Conchopata

Cerro de Oro Cañete El Carmen, Chincha

Cuzco Pikillacta Pampas-Qaracha Chicha Soras Valley

Ica Palpa

Sondondo Valley, Jincamocco

Central Colca Cotahuasi valley, Chuquibamba Chijra valley Río Japo Beringa Aplao Arequipa Sonay Camaná Cerro Baúl

Nazca Tambo Viejo, Acarí

N

Majes-Camaná in the south-central Andes

Majes-Camaná drainage

Tiwanaku influence (conventional)

Tiwanaku

Osmore Valley

Archaeological site or area Modern city

0

50

100 150 200 250 kilometers

Figure 4.1  The valley of Majes and Camaná in the south-central Andes.

Moseley contrasts corporate styles with “folk traditions” (Moseley 2001:79), which are designed, made, and distributed in less institutional modes, such as by households or village-level specialists. Folk traditions can be divided into finewares versus utilitarian wares. Finewares are primarily for serving and display. Utilitarian wares are primarily for cooking and storage. Folk traditions can also be divided according to their geographic affinities. “Local” folk traditions develop in situ from local antecedents. Others lack local antecedents and appear to derive from traditions that developed elsewhere. Those folk traditions that resemble the folk tradition of an “origin center” (Moseley 2001:13) or core region of a material culture horizon can be called “core” folk traditions. Core folk traditions in 58 

B ruce Owen

the peripheries of material culture horizons may range from close, “high-fidelity” replications of the core tradition to derived, “low-fidelity” variants, as long as the shared features are distinctive enough to imply a historical connection. Ceramic horizons may be marked by corporate styles, core folk traditions, or both. These distinctions are merely heuristic. Complicating cases certainly occur. In particular, potters in the periphery of a material culture horizon may incorporate ideas from a corporate style or a core folk tradition into a local folk tradition or a rival corporate style, as with Chimu-Inka (Donnan 1997), Wanka-Inca (D’Altroy 2001), Estuquiña-Inca (Stanish 1992), and others. Some of these hybrid styles are clearly local folk traditions influenced by ideas from the core. For others, the

Table 4.1  Ceramic patterning in Inca, Tiwanaku, and Wari material culture horizons Corporate style Assemblage

Fineware

Core

Folk tradition Utilitarian

Fineware

Local

Utilitarian

Inca in Upper Mantaro Corporate pattern

✓

–

–

✓

✓

Tiwanaku in Moquegua Transplant pattern

–

H

H

–

–

Wari in Majes-Camaná Derived pattern

–

L

–

–

✓

Note: H indicates high fidelity to the core tradition, and L indicates low fidelity to the core tradition.

distinction between a low-fidelity core folk tradition and a local folk tradition influenced by ideas from the core may be arbitrary. This scheme results in five categories, summarized in the headings of Table 4.1: corporate-style ceramics (all of which are presumably finewares), core folk finewares, core utilitarian wares, local folk finewares, and local utilitarian wares. Corporate-style ceramics should show high fidelity to the corporate style’s canons. Core folk finewares and core utilitarian wares may range from high to low fidelity to the core tradition. These categories can make explicit some patterns in the Wari, Tiwanaku, and Inca material culture horizons.

The Inca Material Culture Horizon The Inca adjusted their strategies to fit each region they incorporated into the empire (D’Altroy 1992, 2001; Stanish 2001). Nevertheless, the Upper Mantaro Valley (D’Altroy and Hastorf 2001) can serve as a typical example. Late Horizon settlements in the Upper Mantaro have local folk finewares and utilitarian wares, sometimes augmented by Inca corporate-style ceramics. Inca corporate ceramics were visually obvious, manufactured in centers under corporate control, and distributed by the state for diplomatic and administrative ends (D’Altroy 1992, 2001). Some Inca corporate-style ceramics probably came to peripheral settlements with personnel from the core, presumably to highlight their affiliation with the empire in a strategy of direct control. Others were

probably acquired by local elites in processes of indirect control. In both cases, Inca corporate-style ceramics were concentrated in certain households or sectors, indicating that access was facilitated among people who were particularly associated with the empire. It is not clear that there was an Inca core folk fineware style distinct from the corporate style, at least in the periphery. Some Inca core utilitarian ceramics, such as pedestaled vessels and corn toasters, were apparently brought to peripheral settlements by core personnel. In general, though, utilitarian wares at peripheral sites were primarily local. At Upper Mantaro sites where Inca corporate-style ceramics are found, residents also used ceramics in a local folk fineware tradition. These ceramics sometimes incorporated Inca design concepts, forming a hybrid fineware style that nonetheless clearly derived from the local tradition. In short, ceramic assemblages in peripheral regions of the Inca material culture horizon such as the Upper Mantaro exemplify a corporate pattern, in which an intrusive corporate ceramic style was superimposed on a local folk fineware and utilitarian ware tradition, with few or no core folk finewares or utilitarian wares involved (see Table 4.1). The Inca corporate pattern suggests an empire that was administered in any given province either directly by expatriate officials from the core, or indirectly through co-opted local leaders. Both were equipped with material symbols of the empire. The process that created the Inca material culture horizon

Wari in the M ajes - C amaná Valley

  59

was a corporately organized dissemination of people and corporate-style goods and architecture from the core directly to many locations in the periphery (Figure 4.2). The historical record substantiates this interpretation (Rowe 1946).

The Tiwanaku Material Culture Horizon in Moquegua

Figure 4.2  Schematic models of the formation of three material culture horizons. The Inca corporate pattern results from corporately organized dissemination of people and corporate-style goods and architecture directly from the core to many locations in the periphery. The Tiwanaku transplant pattern results from a folk process of emigration of people bearing core material culture directly from the core to a few locations in the periphery. The Wari case involves both the corporate pattern and the derived pattern, which results from a folk process of emigration and repeated budding of settlements that carry increasingly derived versions of the core folk material culture from the core toward the periphery.

60 

B ruce Owen

Tiwanaku’s expansion also took divergent forms in different regions (Goldstein 2005; Isbell and Silverman 2006:509–510; Stanish 2002). I focus here on Tiwanaku expansion into the Moquegua area, which was quite different from the Inca case (Blom et al. 1998; Goldstein 2005; Goldstein and Owen 2001; Moseley et al. 1991; Owen and Goldstein 2001). Sites in Tiwanaku’s Moquegua periphery lack both corporate-style ceramics and local folk ceramics, instead bearing exclusively Tiwanaku core folk finewares and core utilitarian wares. There is no unambiguously corporate Tiwanaku style of ceramics in the periphery. The great quantity and within-settlement ubiquity of Tiwanaku decorated ceramics at peripheral sites argue against both corporately controlled production and controlled distribution for corporate ends. Virtually every household in the Tiwanaku settlement of Moquegua had multiple decorated ceramics. Moreover, Tiwanaku ceramics in Moquegua were not as standardized as they may appear. The numerous spatially segregated cemeteries at Chen Chen varied not only in the cranial deformation of the interred (Goldstein 2005; Hoshower et al. 1995; Moseley 2001:240) and aspects of tomb construction (Goldstein 2005:245; Owen 1995), but also in the motifs and styles of ceramics placed with the dead. This suggests that there was not a single corporate source of Moquegua Tiwanaku ceramics, but instead, a welter of minor variants, each made for or by relatively small groups in Moquegua. Finally, Tumilaca-style ceramics of the early Late Intermediate Period are so clearly derived from the Chen Chen Tiwanaku style (Goldstein 2005:171–176; Owen 2005) that the two are often difficult to distinguish. The Tumilaca style occurs not only in the former Tiwanaku province of Moquegua, but also in the coastal valley of

Ilo, the highland valley of Carumas-Calacoa, and probably other regions where Tiwanaku core institutions never reached (Owen 2005, 2009). If Chen Chen–style ceramics had been produced under corporate control, it seems unlikely that after the institution collapsed, essentially the same ceramics could have continued to be made and widely distributed with so little stylistic or technical change (Goldstein 2005:322–323; Owen 2005). Tiwanaku did have a corporate style of stone sculpture in the core. It may have had corporate styles of textiles or even ceramics in the core. Tiwanaku may even, as Goldstein (2005:320) suggests, have moved over time toward more coordinated, larger-scale institutional production of ceramics for shipment to the periphery, resulting in increasingly simplified and uniform goods. But there is no evident category of ceramics in Moquegua that clearly differs from the common folk finewares and might suggest institutional production and control. Instead, it is these core folk finewares that were widespread in Moquegua Tiwanaku sites, apparently to the exclusion of any local folk fineware tradition. The Moquegua Tiwanaku finewares are so similar to those from the core that Goldstein suggests that some may have been made in the altiplano (Goldstein 2005:320). This Tiwanaku core folk fineware tradition was not only pervasive and exclusive, but also a high-fidelity replication of the core folk fineware tradition. The utilitarian wares in Moquegua also represent a high-fidelity core utilitarian ware tradition (Goldstein 2005:158–160, 223). In short, the ceramic assemblages at Tiwanaku sites around Moquegua reflect a transplant pattern, in which people used exclusively high-fidelity core folk finewares and high-fidelity core utilitarian wares, with few or no corporate-style ceramics (see Table 4.1). Goldstein (2005:311–321) interprets this pattern of material culture—seen not only in ceramics, but also in other realms of material culture—as implying wholesale, direct settlement by families from the core. These settlers brought their entire material culture with them. Goldstein sees them as possibly organized at the ayllu level, but without formal institutional control, and thus without institutions that might produce and distribute corporate-style ceramics. The Tiwanaku material culture horizon in Moquegua arose from a minimally organized

folk process of emigration of people bearing the core folk material culture directly from the core to a few locations in the periphery (see Figure 4.2). While Goldstein suggests that the Moquegua province may have become more like a state operation over time, the horizon was originally established at a grassroots level.

Wari in Majes-Camaná: Survey and Excavated Evidence The Wari material culture horizon in the Majes-Camaná drainage differs from both the Upper Mantaro Inca corporate pattern and the Moquegua Tiwanaku transplant pattern. The Wari may have had multiple corporate ceramic styles. Conchopata and Robles Moqo styles (Menzel 1964, 1968) could have been corporate styles, as could the finer variants of Chakipampa, Ocros, Atarco, Pachacamac, and Viñaque styles (Menzel 1964, 1968). They embody specific, complex iconography that could have conveyed particular messages for institutional purposes; were made to very high standards, possibly in institutional workshops; and were distributed to, or accurately duplicated in, the distant periphery. These corporate styles may have been made and used by a single institution, such as a state, at different times or for different purposes, or by multiple institutions, such as noble houses, religious institutions, or administrations in the periphery. The less iconographically loaded, more variable and quotidian Huamanga grade of ceramic styles seems to represent a widely used folk fineware tradition (Anders 1986, 1998; Knobloch 1991:252–256; Lumbreras 1974a:181–182; Ochatoma and Cabrera 2001:152; Owen 2007:288–289; Vivanco and Valdez 1993:95–97; cf. Isbell 1977). These ceramics appear to be the finewares of rural and urban people of all statuses, even those who also had access to corporate-style ceramics. More so than the corporate styles, Huamanga-grade ceramics vary considerably from region to region, even within the core. The valley of Majes and Camaná is full of Wari ceramics. There are examples in museum collections in Camaná, Cardo, and Aplao, and in private collections. Occasional Wari ceramics turn up on the surface of

Wari in the M ajes - C amaná Valley

  61

Figure 4.3  Sites with Wari ceramics in the Majes-Camaná drainage.

many sites (Figure 4.3, Table 4.2). Many sites are covered, sometimes deeply, by volcanic ash from the eruption of Huaynaputina in AD 1600 or by debris flows from the steep valley walls. The cemeteries have been looted and picked over at least since a period of intensive commercial looting for supposedly Wari feathered textiles in the 1980s and 1990s. As a result, most sites have relatively few ceramics on the surface, and still fewer diagnostic ones. While the number of Wari sherds known from some sites is low, finding even a few under these circumstances can suggest a Wari component. About half of the drainage has been surveyed, and not all of that systematically (see Figure 4.3). Disselhoff (1968) excavated Early Intermediate Period burials from Cabezas Achatadas, part of the large looted area 62 

B ruce Owen

now called Huacapuy, near Camaná. He reported Wari sherds in a road cut near the police station of Huacapuy and illustrated two derived Wari ceramics said to have been looted from Poza Encantada, apparently near or part of Huacapuy (Disselhoff 1967:211–213), as documented by Cardona (2005). From the Majes Valley, Disselhoff (1967:211–212) illustrated Wari sherds from the looted surface of Toro Grande. Garcia and Bustamante’s (1990) survey of part of the upper Majes Valley illustrates many Wari sherds from numerous sites. Manrique and Cornejo (1990) reported exploratory findings from the Camaná Valley. Malpass (2001) mapped and excavated at the probably late Wari site of Sonay, in the Camaná Valley, and conducted a preliminary survey of some of the narrow part of the valley (Malpass et al. 1991). Cardona (2005) inventoried sites in much of the valley of Camaná for the Qhapac Ñan prehistoric road survey and has generously allowed his survey results to be included here. Manuel Garcia, Pablo de la Vera Cruz, and other staff of the Instituto Nacional de Cultura in Arequipa have visited sites in the Majes and Camaná valleys. Julio Manrique of the Universidad Católica de Arequipa conducted site visits and surface collections with archaeology students at various sites in the drainage, but no reports or collections apparently exist aside from Manrique and Cornejo’s (1990). I have also visited numerous sites in the Majes and Camaná valleys. Three sites with Wari ceramics in the Majes-Camaná drainage have been excavated. The first comprised large, collective burials at La Real, recovered in an emergency salvage project by the Universidad Católica Santa Maria de Arequipa. Garcia and Bustamante (1990) illustrate ceramics and a wooden snuff tablet from the site. Tung analyzed the osteological material (Tung 2003; Tung and Owen 2006). Tung also directed excavations in 2001 at the site of Beringa (see Figure 4.3), and I analyzed the ceramics (Owen 2007; Tung 2007; Tung and Owen 2006). Beringa was a rural village, where burials were intermixed with the residential architecture. The largely disturbed deposits contained local, Huamanga-grade Wari ceramics from burials, occupation strata, and fill. Some intact occupation deposits were also excavated. There was a small percentage of moderate quality Chakipampa

GB

GB, BDO

GB, BDO,TAT

GB

GB

GB, BDO

GB

HD

BDO

AC

MM, AC

AC

AC

AC

BDO

AC

BDO, AC

MC, BDO

AC

HD

AC

BDO

BDO

Majes

Majes

Majes

Majes

Majes

Majes

Majes

Majes

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Camaná

Bodeguillas Punta

Bodeguillas

Cerro Las Bodeguillas

Near Huacapuy/Poza Encantada

Huacapuy

Pacaecitos

Capilla de Characta

Characta

Pisques Escuela

Cerro Soto cemetary

Pisques Norte

Pisques

Sonay

No name

Just SW of Pillistay

Toro Grande

La Real

Huancarqui

Casquiña

Cerro Bilbao

Beringa

Huario

Santa Rosa II

Name

n/a

CA62 or 63

CA61

n/a

CA59

CA46 or CA47

CA39

CA38

CA33

CA51a

CA31

CA30

CA26

CA21

CA12,13,15?

n/a

n/a

M6

M2

M5

M1

M11

M14

Site No.

cem

cem

cem

cem?

cem

n/a

cem

cem

cem

cem

cem

hab

admin?

cem

cem

cem

mass cem

cem

hab

hab

cem & hab

camp

camp & cem

Type

739806

739650

740075

?

741400

741990

751200

751725

754210

751965

754725

755150

757535

756665

761800

768134

768000

769900

768150

767750

769650

772238

768471

E

8161512

8161999

8162115

?

8164240

8166775

8171279

8172000

8173364

8173745

8174140

8174360

8174755

8175500

8178273

8201583

8215850

8219700

8222050

8222400

8225300

8229515

8230788

N

1: photo (BDO 2004); Multiple: noted (Owen 2000)

1: photo (BDO 2004)

Plural: mentioned in text (AC)

Multiple sherds: mentioned in text (HD); 2 vessels: photos (HD). Same as CA59?

Looted Wari and Nasca burials mentioned in text

2: drawings (MC); 1–2: photos (BDO 2005) Site leveled for farming before 2005

Plural: mentioned in text (AC); 1: noted (BDO 2000)

Plural: mentioned in text (AC); 1: photo (BDO 2006)

Several: noted (BDO 2000); 4–5: photos (BDO 2004)

Plural: mentioned in text

Several: mentioned in text

Several: mentioned in text

4: drawings, marginally diagnostic, architecture (MM); 1 Wari, several “local Middle Horizon” in text (AC)

1: mentioned in text

3 to 5: photos (BDO 2004, 2005)

8: drawings (HD)

11: drawings (GB) (location estimated)

6: drawings (GB); Several: noted (BDO 2001)

3: drawings (GB)

A few Q’osqopa sherds, mentioned in text (GB)

5: drawings (GB); >1000: excavated (Owen 2007)

9: drawings (GB); 6: photos (BDO 2000)

1: drawing

MNI of diagnostic Wari vessels

Note: AC = Cardona 2005; BDO = Owen 2007 and Owen’s site visits in 2000, 2001, 2004, 2005, 2006; GB = Garcia and Bustamante 1990; HD = Disselhoff 1967; MC = Manrique and Cornejo 1990; MM = Malpass 2001; TAT = Tung 2007; admin? = orthogonal cellular Wari possible administrative site; camp = ephemeral campsite with rings of stones and pot drops; cem = cemetery; hab = permanent habitation site; mass cem = mass cemetery. UTM coordinates all Zone 18S, WGS 84 datum; some estimated from maps. “M” site numbers from Garcia and Bustamante 1990. “CA” site numbers from Cardona 2005.

Sources

Valley

Table 4.2  Sites with Wari ceramics in the valley of Majes and Camaná

and Ocros sherds, but there were effectively no sherds in fine Wari corporate styles. Local Huamanga-grade ceramics were ubiquitous across the site, not limited to certain areas. The architecture was substantial, varied, and repeatedly modified, and the depth and artifact density of the deposits suggest a long occupation by many people. While a few structures are reminiscent of Wari house compounds, with roughly square walled compounds enclosing narrow rooms and/or benches along the entire length of one side, they do not correspond to the orthogonal cellular architecture of planned Wari complexes (Isbell 1991), and there was no evident planning at the site level. Tung (2007) reports seven radiocarbon dates from Beringa. Two are from intrusive Late Intermediate Per­ iod burials. Of these, Burial 59 was intact and contained a single Chuquibamba-style jar (Owen 2007:Figure 21), while Burial 58 was badly disturbed and located in a dif­­ferent unit where fragments of a probably Late Inter­mediate Period canteen were found (Owen 2007:Figure 16 CID 585). A handful of other vessels and sherds could relate to these Tricolor del Sur styles (Lumbreras 1974b; Owen 2007:363; and others), but they cannot yet be stylistically dated. Of the remaining five dates, two are from in situ architectural posts in different structures; one is from a row of sticks set vertically in a fieldstone wall base; one is from a carbon crust on the exterior of an in situ subfloor storage vessel over which a stone dividing wall had been built; and the last is from the textile wrappings of a mummy bundle. These dates span cal AD 580 to 962 with 2σ confidence (Figure 4.4), calibrated using OxCal 4.0.5 and the SHCa104 calibration curve (Bronk Ramsey 1995, 2001; McCormac et al. 2004). The later part of that range is largely due to the single mortuary textile date, which is later than the others and falls on a flatter portion of the radiocarbon calibration curve, extending its calibrated uncertainty range. Since the first four dates cluster within only seventy-six radiocarbon years and all probably relate to construction or early occupation, it is reasonable to treat them as marking the initial occupation of the site, while the mortuary textile would be later. The tight clustering of dates on such disparate materials and contexts tends to rule out the old wood problem. 64 

B ruce Owen

The four construction and early occupation dates span cal AD 580 to 862 with 2σ confidence. Using OxCal to model these four dates as a single phase, their 2σ ranges fall from cal AD 664 to 775 (Figure 4.4), indicating an initial occupation of Beringa during the first half of the Middle Horizon. This is unexpectedly early for Wari influence so far from Ayacucho. The problem could be explained away if the diagnostic ceramics were actually from a later reoccupation or reuse of the site for burials, but that does not appear to be the case. Some of the ceramics certainly were disturbed mortuary offerings. However, most of the thousands of decorated sherds were small and isolated, with few that joined or matched, even though collections were taken in large units. This pattern suggests that much of the assemblage came from occupation debris or fill churned up by looters, rather than broken mortuary items, or Wari-style smashed ceramic offerings. Moreover, about 9 percent of the assemblage was excavated from undisturbed, nonmortuary contexts, including a berm of dense trash at the edge of the site, shallow floor features, pit fills rich with botanical debris, and the soil packed around subfloor storage vessels set into holes in the hard subsoil. The sherds from these intact occupation deposits include many of the characteristic features of the larger, disturbed sample, including the cream and black “lazy S and X” pattern, clustered rim ticks, parallel wavy and straight lines, various complex designs, and examples of both Chakipampa and Ocros styles (Owen 2007:Figure 22 CID 3354, 3568, 536, 532; Figure 23 CID 531). These ceramic characteristics were present when the site was occupied and were not limited to potentially later, intrusive burials. The intact occupation deposits naturally contained a lower proportion of finewares (7.0 percent slipped, 1.3 percent painted) than did the entire assemblage with its admixture of looted grave goods (19.2 percent slipped, 6.4 percent painted), but the kinds of finishes, colors, and motifs were the same. The intact deposits did not contain a different assemblage that might suggest a distinct, earlier occupation. The burials, then, contained ceramics similar to those from the intact occupation deposits. The disturbed

Figure 4.4  Calibrated radiocarbon dates from Beringa, Sonay, and Toro Grande. Bars under the probability distributions represent 2σ ranges. All calibrated using OxCal 4.0.5, southern hemisphere calibration curve SHCal04. The four construction-related dates at Beringa are modeled as a phase. The two Sonay dates are approximately recalibrated as a phase.

assemblage contained a significant amount of material, including finewares, that appears to have come from occupation deposits. If the burials, these ceramics from disturbed occupation deposits, and the ceramics excavated from intact occupation deposits all were centuries later than the dated architecture, then so was much or all of the occupation of that architecture: a possible but improbable scenario. The ceramics themselves should suggest a stylistic date. I argue elsewhere (Owen 2007) that they suggest a Middle Horizon Epoch 1B through 2A or 2B occupation. This is compatible with the early radiocarbon dates, but could also allow for a mid–Middle Horizon occupation. One reviewer noted similarities to Atarco-style vessels from the Palpa and Nazca region (Isla 2001:Figures

13, 18b, c, 19a, b, c), which are conventionally assigned to Epoch 2, but might continue into Epoch 3, allowing for a Late Middle Horizon occupation. The same reviewer suggested that the Beringa material could even relate to the Pinilla style of Epoch 3 (Lyon 1968; Paulsen 1968), but the similarity is distant. As Isla (2001:581 note 12) noted in regard to the Palpa and Nazca material, these stylistic and chronological categories need better definition. If the ceramics are associated with the radiocarbon dates, then the process that brought Wari material culture to Majes and Camaná at such an early stage of the Middle Horizon must have been very rapid, or have started very early in the Wari sequence, or both. This is especially so if the process was driven by the urbanization of Huari itself, as Jennings (this volume) suggests,

Wari in the M ajes - C amaná Valley

  65

because Wari material culture reached Majes-Camaná relatively early in Huari’s growth. The third Wari site that has been excavated is Sonay, in the narrow portion of the coastal valley of Camaná. Malpass (2001) mapped and excavated at Sonay, documenting a block of architecture that reflects some aspects of Wari orthogonal cellular canons. The structure was apparently kept clean, consistent with an administrative or other special function. Malpass suggests that Sonay was centrally located to control the entire MajesCamaná drainage, although Beringa would have been a full two days’ journey away. Excavations produced a handful of small polychrome sherds that are consistent with local Wari material at Beringa, but not really diagnostic. Unfortunately, this makes it impossible to compare the ceramics from Sonay and Beringa in terms of style, function, or chronology. Malpass (2001) reports two calibrated dates from immediately below the architecture, indicating that the structure was built shortly after cal AD 940 ± 30 to 1000 ± 50 (1σ, method and calibration curve not specified). Assuming that these dates were calibrated with a curve similar to the northern hemisphere IntCa104 curve, these dates can be approximately recalibrated for comparability to the Beringa dates using OxCal and the southern hemisphere curve. The recalibrated date ranges are cal AD 898 to 1024 and cal AD 900 to 1161, 2σ (see Figure 4.4). These dates place Sonay’s construction near the end of the Middle Horizon, and probably two centuries or more after the initial Wari occupation of Beringa. This late construction of a formal Wari installation in Camaná might be related to other Late Middle Horizon developments in Wari’s south coastal periphery, including Wari’s apparent abandonment of Nazca (Conlee, this volume) and dramatic changes in Cotahuasi, Moquegua, and elsewhere (Jennings, this volume; Williams 2001). Disselhoff (1967:211–212) reports a radiocarbon date of 970 ± 50 BP from Toro Grande in the unsurveyed lower portion of the Majes Valley, illustrating presumably associated sherds similar to those from Beringa. This date calibrates to cal AD 1025 to 1210, 2σ (see Figure 4.4) and suggests that the ceramics from Beringa represent a tradition in the Majes Valley that lasted up to the end of 66 

B ruce Owen

the Middle Horizon, contemporary with the terminal Middle Horizon center of Sonay. I recently excavated at three residential sites in the Camaná Valley, two of which are adjacent to looted cemeteries with some Wari sherds (Pillistay, by my observations, and Soto, according to Cardona 2005). However, these villages were first occupied only in the terminal Late Intermediate Period (Tantaleán and Owen 2007). To summarize, Wari ceramics have been documented at a minimum of fourteen sites in the surveyed portion of the coastal Camaná Valley (see Table 4.2, Figure 4.3). A few are close together and perhaps should be counted as single sites. At least eleven of the Camaná sites (79 percent) are cemeteries, of which two have associated habitation areas. One site is residential, one is the apparently administrative site of Sonay, and one, Pacaecitos, has been destroyed for agriculture and cannot be categorized. In addition, at least eight sites in the surveyed portion of the inland Majes drainage have Wari ceramics, for a total of twenty-two sites in the roughly 50 percent of the drainage that has been surveyed (see Table 4.2, Figure 4.3). The Majes sites are more varied and include cemeteries, residential sites, and apparent caravan stops or repeatedly occupied ephemeral camps. These twenty-two sites are probably not all of the Wari components in the drainage. Google Earth imagery shows additional cemeteries and habitation terraces in the unsurveyed areas. Even in the surveyed areas, some Wari components may not have been recognized yet. Many of these sites are buried, looted, and picked over. Diagnostic sherds are scarce, and the surface changes from year to year. Some sites in the Camaná Valley have been visited in different years by three archaeologists: Malpass, Cardona, and myself. All of us found Wari sherds on sites that one or both of the others missed. Future visits may identify Wari material at additional sites. In addition, recent agricultural and urban development has reportedly destroyed numerous archaeological sites, especially along the southern margin of the alluvial plain around Camaná. Residents mention Wari ceramics, although their stylistic attributions must be treated with caution. In sum, the map (see Figure 4.3) almost

certainly understates the original distribution of Wari ceramics in Majes and Camaná.

The Wari Material Culture Horizon in Majes-Camaná: Ceramics What exactly are the “Wari ceramics” at these sites? I analyzed the ceramics from Beringa (Owen 2007), which is the only substantial excavated sample of Wari ceramics from the Majes-Camaná drainage published so far. Surface material from other sites in the drainage looks grossly similar, so I treat the Beringa assemblage as typical. These “Wari ceramics” do not include the potentially corporate Wari styles, despite the occasional fragment of not particularly fine Chakipampa and Ocros. Fragments of one very fine, elaborately decorated canteen were found at Beringa, but it more likely relates to Nasca (Owen 2007:362). Fragments of two sculptural vessels, one a camelid (Owen 2007:362) and the other probably a monkey, might have been categorized as Qosqopa style (Lumbreras 1974c; Neira 1990) had more of the surface decoration been preserved. However, the loosely defined, Wari-related Qosqopa ceramics of the Arequipa region are too local, variable, and mediocre in quality to have been a Wari corporate style. I have seen no fine corporate Wari pieces in the four local museum collections. None are identifiable among Garcia and Bustamante’s (1990) extensive drawings, nor among the surface sherds at sites I have visited. Rare elaborate pieces may have been collected by looters, but they would presumably take only those that were intact or restorable. The absence of sherds in the excavated Beringa assemblage and in surface scatters suggests that ceramics in Wari corporate styles were very rare, at best. Instead, the diagnostically Wari ceramics on these sites (Figure 4.5; Garcia and Bustamante 1990) resemble the Huamanga-grade assemblage at Beringa (Figure 4.6; Owen 2007). Surface finds are frequently sherds of globular bowls that range from slightly restricted forms with external decoration, to shallow, open forms with internal decoration that is usually pendant from the rim. Surface and excavated ceramics also include occasional escudilla forms (cumbrous bowls), occasional beakers or keros, and rare bottle forms. External painted motifs

emphasize horizontal S and X motifs in bands below the rims of more restricted forms. Less common motifs include variants of Wari feathered wings and other geometric designs, occasional Wari-like profile felines, modeled and painted faceneck jars, and a few examples of a simplified front-face anthropomorphic head wearing a headdress with three plumes rendered like Wari feathered wings, similar to several illustrated by Garcia and Bustamante (1990). These are generally not finely crafted pieces. They typically have sloppy, incomplete burnishing and imprecise painting. The relationship to Wari Huamanga ceramics is evident, but these ceramics are not a good match for any style known in the heartland. They are a low-fidelity core folk fineware tradition. As at Moquegua Tiwanaku sites, core folk finewares are ubiquitous at Beringa. They were apparently the finewares that everyone at Beringa used. The presence of Wari sherds at many sites suggests that they were not strongly concentrated at certain intrusive or administrative sites, or at residences and burials of a few co-opted leaders, as Inca ceramics sometimes were. Unlike the high-fidelity Tiwanaku assemblages in Moquegua, however, the Beringa ceramics are a lowfidelity expression of the core folk fineware tradition. No one would suggest that these ceramics were imported from the Wari core. Nor could the Beringa assemblage as a whole be confused with that of any other peripheral Wari region, each of which tends to have its own version of Huamanga-grade folk finewares. As in Moquegua, sites with the core folk finewares have no identifiably different local folk finewares, although that may be due to the early stage of research. Nor is there any known local fineware antecedent that might have incorporated Wari ideas to form a hybrid style with local roots. A few fragments of Nasca ceramics at Beringa, Cabezas Achatadas (Disselhoff 1968), and other sites might hint at a local Nasca-related antecedent, but the evidence is thin, especially inland in the Majes Valley. The Majes-Camaná Huamanga-grade assemblage looks like an intrusive, low-fidelity core folk fineware tradition that was the exclusive folk fineware at sites where it is found. The utilitarian wares at Beringa are quite different from those in the Wari core. They seem to be a local

Wari in the M ajes - C amaná Valley

  67

Figure 4.5  Wari surface sherds from Camaná Valley sites, photographed in the field.

utilitarian ware tradition, with few recognizable similarities to core utilitarian wares. They lack flat-bottomed cooking pots in general; large, flat-bottomed, widemouthed cooking pots with horizontal strap handles (Anders 1986; Ochatoma and Cabrera 2001) or “basket” handles (Anders 1986; Bennett 1953; Ochatoma and Cabrera 2001); and wide-mouthed pitcher forms with a single high, vertical strap handle (Anders 1986; Ochatoma and Cabrera 2001). All of these characteristics of the core utilitarian ware tradition would be identifiable from sherds. Instead, Beringa’s utilitarian ceramics include generic large storage or transportation vessels, as well as small globular ollas with vertical handles, some with tripod feet, and extremely large, open-mouthed vessels with heavy, thickened square-sectioned rims (Owen 2007), all of which seem to be specifically local. Cooking vessels at Beringa tend to be either quite small, up to 68 

B ruce Owen

about 1 liter, or quite large, mostly ranging from 7 to 20 liters and more, with none in the 1 to 4 liter range (Owen 2007). Unlike other Wari-related populations, residents of Beringa apparently cooked either for one to two people, or for relatively large groups, but rarely for groups the size of a nuclear family. These cookinggroup sizes suggest, in turn, fundamental differences in social organization, labor organization, and/or domestic habitus between the people of Beringa and those of the Wari core. In addition, large vessels at Beringa were often set into deep holes laboriously dug into the hard, rocky subsoil with the neck and rim removed at ground level, a practice reported from Cahuachi (Silverman 1988:421), but apparently not common at Wari sites. In contrast to the Inca corporate pattern or the Tiwanaku transplant pattern, the Wari ceramics in the valley of Majes and Camaná form a derived pattern, in which people rarely or never saw corporate-style

Figure 4.6  Ceramics from Beringa.

ceramics, used low-fidelity core folk finewares as their only finewares, and used local utilitarian wares (see Table 4.1). While the social processes that could give rise to the corporate or transplant patterns seem fairly evident, possible causes of the derived pattern are less obvious. What specific social processes could produce the derived pattern? What social mechanisms brought a low-fidelity variant of Wari folk finewares to such a distant region, stripped of the corporate styles and the core utilitarian tradition, so quickly after the core tradition developed?

Explaining Wari in Majes-Camaná: Alternative Hypotheses Table 4.3 summarizes some hypotheses about how the derived pattern of Wari ceramics might have arisen in the Majes-Camaná drainage. The first two, “Empire” and “Direct colonization,” would have produced the Inca corporate pattern of ceramics and the Tiwanaku transplant pattern, respectively. Neither pattern obtains in Majes and Camaná, so both of these hypotheses can be rejected. If the Majes-Camaná drainage were controlled by a Wari Empire, imperial institutions would presumably have used corporate-style ceramics to reinforce Wari ideology and to reinforce the status of direct administrators or co-opted local leaders. The empire evidently used corporate ceramics this way elsewhere, but they are conspicuously scarce in Majes and Camaná. Wari artisans used tapestry textiles for conveying complex imagery like that on the corporate ceramics, but Wari tapestry, too, was absent or very rare. While looters reportedly recovered magnificent feathered textiles that they identified as “Wari,” their cultural affiliation is not certain, and their ability to convey complex concepts as corporate styles often was limited by their simple, geometric patterns. An empire that introduced core finewares to support either direct or indirect control would presumably limit access to them. Yet low-fidelity Wari core folk finewares were widespread and seem to have been the only finewares at Beringa during the Middle Horizon. Finally, no identifiable architectural facility in formal Wari style was constructed until nearly the end of the Middle Horizon. Imperial expansion as visualized by 70 

B ruce Owen

Isbell (1991), Schreiber (1992), or Jennings and Craig (2001) would likely have involved a Wari architectural center of some kind, although indirect imperial control could perhaps have functioned without one. Unless the Beringa ceramics are centuries later than the radiocarbon dates, Sonay is too late to have been part of the process that introduced Wari ceramics to the region. An empire could, perhaps, operate without a corporate style. Still, Wari institutions conspicuously exploited corporate styles in other contexts, so their absence here suggests that imperial or other institutions were not involved. With neither corporate-style artifacts nor architecture until much later, there is little to suggest direct or indirect imperial control. The distinction between corporate-style ceramics and core folk finewares is crucial here. The presence of core folk finewares does not imply an empire, as the Moquegua Tiwanaku case attests. The Majes-Camaná case suggests that the supposition of Wari imperial control, direct or indirect, should be evaluated on a regionby-region basis throughout the Wari periphery, precisely as many of the chapters of this volume do. Even if imperial models are sustained in some places, the Wari material culture horizon in other regions, including Majes and Camaná, may be better explained by other social processes. Similarly, we can reject the hypothesis of direct colonization by people from the Wari core, analogous to Tiwanaku settlement in Moquegua. The core folk finewares in Majes and Camaná are a very low-fidelity version of the core folk tradition. They are not transplanted from the Wari core. Even more striking, the utilitarian wares differ greatly from those of the core, suggesting differences in cuisine, domestic group size, and family organization, all of which imply deep differences in social identity. There are also other, nonceramic indicators of locally specific traditions, such as the ritual use of placas pintadas, which are worked sherds or flat stones bearing painted designs (Jennings 2003; Tung 2007). While the Tiwanaku colonists in Moquegua brought their entire material culture with them from the core, the people who used Wari ceramics in Majes and Camaná had a much more distantly derived material culture, combining some vague and some specific

Table 4.3  Alternative hypotheses of social processes that could explain the derived pattern of Wari ceramics in Majes-Camaná, with some expected and actual material correlates

Hypothesis

Material correlates

Present?

Evidence

Empire

Corporate-style ceramics

No

No corporate-style ceramics

Corporate-style textiles

No

No iconographically loaded Wari tapestry

Limited access to core finewares

No

Core folk finewares ubiquitous

Corporate style in architecture

Late

Not until Sonay, at end of MH

High-fidelity core finewares

No

Low-fidelity core finewares

High-fidelity core utilitarian wares

No

Local utilitarian wares

Absence of distinct local practices

No

Placas pintadas

High-fidelity core ceramic styles that develop into local variants

No

No high-fidelity core ceramics

Local variants appear late, after a period of local development

No

Local variant appears early in MH

Geographically continuous distribution of related styles

?

Majes-Camaná separated from core by nonWari regions?

Local antecedent that developed into the regional style

No?

No local antecedent

Prestigeware for local purposes

Core finewares restricted to elites

No

Core folk finewares ubiquitous

Stepwise budding

Mix of local variations and specific stylistic continuities

Yes?

Core folk finewares are low fidelity, yet have some very specific core features

Variant styles appear quickly

Yes?

Low-fidelity core folk ceramics appear early in the MH

Chain of settlements from core to periphery

?

Majes-Camaná separated from core by nonWari regions?

Beringa ceramics are much later than construction dates

No?

Possible but unlikely

Ceramic style is Epoch 3 or later

?

Possible but not necessarily

Wari sites date to different times, not contemporary

?

More research is needed

Direct colonization

Long local development

Regional interaction sphere

Terminal Wari expansion

Misleading palimpsest

 

parallels in their finewares with a profoundly different domestic habitus. The derived pattern might also result from a long period of local development following an early influx of Wari influence, perhaps in the form of direct colonists, imperial exploitation, or religious proselytizing. The material correlates of this initial influence would include a higher-fidelity Wari heartland material culture, and the derived local variant would have appeared some time later. Current evidence does not favor this hypothesis. There are no signs of a higher-fidelity Wari occupation that could have started the process. In addition, assuming that the early dating of the Beringa ceramics is correct, the finewares at Beringa were already low-fidelity versions of the core folk tradition, and the utilitarian wares were already different enough to be called local, by early in the Middle Horizon. The initial Wari influence would have to have arrived implausibly early for such a distant part of the Wari periphery, or the tradition would have to have evolved very rapidly, in order to produce the highly derived Beringa assemblage by early in the Middle Horizon. An alternative hypothesis envisions the Wari style of Majes and Camaná as not originating in the Ayacucho region at all, but instead as developing locally, in parallel with the Wari core. The similarities could be due to shared relationships with the geographically intermediate Nasca tradition or long-distance interactions linking the Wari core and the Majes-Camaná area. This hypothesis would recast the southern Pacific slope portions of the Wari material culture horizon as not the result of an “origin center → civilization horizon” process (Moseley 2001:13), but rather the record of an interaction sphere in which Wari-like ideas were widespread and Wari-like iconography developed in interacting local trajectories, from which Huari emerged as the most impressive and powerful exemplar of the tradition. Again, current evidence does not support this hypothesis. Many of the intervening areas are not known to have participated in the hypothesized regional interaction sphere, although the Andean tendency toward archipelago or scattered (salpicado) settlement patterns (Murra 1975) might account for the apparent geographic gaps. Also, with the possible exception of some scarce 72 

B ruce Owen

Nasca-like material, there is no evident local antecedent in Majes-Camaná that would have developed into the local Wari-like folk tradition. Instead, it seems to appear suddenly, without antecedents. This may be due simply to the early stage of research in the drainage. Haeberli (2001) proposes some possible local antecedents among similar materials from the adjacent valley of Sihuas. At this point, the regional interaction hypothesis remains unsupported, but not rejected. Several chapters in this volume (Chapdelaine; Marcone; Nelson et al.) suggest that in some regions, indigenous elites adopted and adapted Wari prestige wares for their own purposes within essentially local social contexts. These hypotheses do not suit the MajesCamaná case, because the finewares involved were apparently not special objects to which the elites had enhanced access. Instead, they were the finewares in general use, apparently available to most people at Beringa, if not in much of the drainage, to the exclusion of any other finewares. Another hypothesis that could account for the derived pattern involves a chain of successive folk colonization events. In this stepwise budding hypothesis, a group of people from the Wari core moved to an attractive region a bit farther away. After a period of growth and local development of their folk ceramic traditions, perhaps as brief as a single generation, some of this group split off and moved to a still more distant region. This fissioning or budding process (cf. Bandy 2004) could repeat numerous times, creating a branching chain of settlements stepping generally away from the core and accumulating increasing local variation with each step. The successive hops into available pockets of arable land could be driven by demographic expansion, political pressures, a frontier or entrepreneurial ideology, or other factors. The process might resemble ones ongoing today, such as Goldstein (2005:181) has documented in Moquegua. Each hop might encourage additional variation in the folk ceramic tradition through a cultural equivalent of the genetic founder effect, in which the few founders bring to their new settlement only the subset or variant of the source tradition that they happen to know. Practices in small frontier populations might evolve more rapidly than they would in larger populations,

where more people might constrain changes in the consensus. Moving to a different environment and building a new community might in itself encourage flexibility or change. In these ways, stepwise budding might foster rapid change from a high-fidelity core folk tradition toward a low-fidelity one, and the changes would be cumulative in each successive settlement. At the same time, kinship and other social ties would encourage continued contact with the core, potentially maintaining specific similarities between the core traditions and those at the distal ends of the budding chains. Contact with the core for goods, services, and social relationships might function in an up-the-line fashion from a distal settlement’s closest relationship to its parent community, and so on toward the core, or it could be more direct. Either process could lead to the peculiar mix in Majes and Camaná of low-fidelity and even local ceramic characteristics with some highly specific core fineware traits. Alternatively, these conservative similarities could relate to diasporic collective memory (Cohen 1997; Owen 2005). The stepwise budding hypothesis could explain some puzzling features of the Majes-Camaná case. Because it provides reasons to expect rapid, cumulative cultural change, it could explain how a low-fidelity variant of the core fineware tradition could appear so early in the Middle Horizon. It also explains how a low-fidelity core folk fineware tradition could appear in a distant province without antecedents: the intervening stages of development occurred in other locations along the way. Continuing contact with the immediate predecessor settlement, or directly with the core, could explain how some highly specific similarities to the core might be maintained even as other traits rapidly diverged. In this view, the processes that created the Wari material culture horizon would include both corporate, maybe imperial, dissemination of people bearing corporate-style goods and architecture directly to some locations in the periphery, such as Pikillacta and Cerro Baúl, and a parallel, minimally organized folk process of emigration and repeated budding bearing the core folk material culture from the core to the periphery by successive hops that facilitated accelerated, accumulating change (see Figure 4.2).

The stepwise budding hypothesis is potentially testable. There should be a chain of settlements with cumulatively more divergent ceramic assemblages stepping across the landscape from the core to the periphery. Sites in Chuquibamba might be candidates for the immediate parent settlement of the Majes-Camaná population. Cotahuasi might be a link further up the chain, although sufficiently early Wari material seems to be lacking (Jennings, this volume). An alternative route might pass through the Nazca-Palpa region. Detailed ceramic comparisons that could suggest the successive derivation of each assemblage from one closer to Huari have yet to be done. Some of the areas between the Majes-Camaná Valley and the Wari core seem to lack early Wari material (Meddens and Branch, this volume), or to have variants that are not clearly intermediate. The route across the landscape taken by the chain of budding events could have been wandering and indirect, and some of the settlements could be small and difficult to recognize. The gaps in research on Wari occupations in the intervening regions leave the stepwise budding hypothesis possible but not clearly supported. Nevertheless, it may be the most plausible explanation of the lot. If the Beringa ceramics are actually much later than the radiocarbon dates from the architecture, then they might reflect a Late or Terminal Middle Horizon intrusion into Majes and Camaná that involved building a center at Sonay, perhaps associated with faltering Wari institutions elsewhere. This terminal Wari expansion hypothesis implies diminished institutions that did not use Wari corporate styles, but could still move rural settlers and build a modest architectural complex with a Wari-style plan. The Beringa ceramics would be derived from late, possibly Atarco or Ica-Pachacamac styles that could extend into Middle Horizon Epoch 3 or later. The settlers would have come from a region where those styles were used, such as the Nazca and Palpa area (Isla 2001). The process could have been organized from Huari or might have originated with institutions in the periphery, perhaps as the Wari state was declining. While this story sounds plausible in itself, the evidence for an early Middle Horizon occupation at Beringa makes it unlikely. Finally, given the paucity of radiocarbon dates in the Majes-Camaná Valley, the settlement pattern

Wari in the M ajes - C amaná Valley

  73

in Figure 4.3 might not represent widespread use of Wari ceramics at all. Instead, it might be a misleading palimpsest of multiple settlement patterns from different moments in time. This would not change the results for Beringa and the ceramic analysis, but it might place them in a different context or offer additional clues. More research is needed.

Summary and Conclusions Wari ceramics were pervasive in the valley of Majes and Camaná. Dates from Beringa suggest that these Wari ceramics were already common in the first half of the Middle Horizon, even though the valley is far from Huari. Classifying ceramics in peripheral regions as being of corporate style versus folk tradition, core versus local affiliation, fineware versus utilitarian ware, and high fidelity to the core tradition versus low fidelity (see Table 4.1) highlights how the Wari material culture horizon in the Majes-Camaná drainage differed from the material culture horizons of the Inca in the Upper Mantaro and of Tiwanaku in Moquegua. Specifically, it makes explicit a corporate pattern of Inca ceramics that corresponds to direct or indirect imperial control, a transplant pattern of Tiwanaku ceramics in Moquegua that corresponds to a minimally organized, direct folk colonization from the core, and a derived pattern of low-fidelity Huamanga-grade Wari ceramics in the valley of Majes and Camaná that does not fit either of these social processes. The pattern in the Majes-Camaná drainage involves few or no Wari corporate-style ceramics and apparently unrestricted access to the local variant of Wari core folk finewares. There is little evidence for direct or indirect imperial control in Majes and Camaná other than the architecture of the modest room compound at Sonay, built at the very end of the Middle Horizon. If Wari ceramics were widespread in this valley without imperial control, then the same might be true elsewhere, too. The imperial model of Wari should be evaluated on a regionby-region basis. What marks the Wari material culture horizon in the Majes-Camaná drainage is the widespread use of a derived variant of Huamanga-grade ceramics, Wari’s 74 

B ruce Owen

regionally variable core folk fineware tradition. The utilitarian wares at Beringa, on the other hand, are a local tradition, responding to local foodways and social organization. The ubiquity of the core folk fineware tradition suggests that Middle Horizon residents of the Majes-Camaná Valley were not simply provincial people who somehow picked up some Wari ideas or trade goods. On the other hand, the low fidelity of their core folk fineware tradition, and their local utilitarian ceramic tradition, makes them unlikely to have been direct immigrants from the core. Nevertheless, they were closely enough connected to the core to maintain a variety of specific similarities in their otherwise divergent decorated ceramics. Specific social processes led to these patterns. With imperial control and direct colonization of Majes and Camaná rejected on the basis of the ceramics present, several other hypotheses are proposed. The stepwise budding hypothesis may be the most plausible explanation for the derived pattern in Majes and Camaná. In this process, repeated fissioning and resettlement creates a branching chain of communities that spreads increasingly divergent versions of the core’s material culture into the periphery without institutional management and creates a material culture horizon without a corporate ceramic or architectural style. The process could encourage rapid change in material culture, while maintaining some original features through continued contact with parent settlements or the core. The Majes-Camaná drainage would have been multiple steps away from the core, explaining how its low-fidelity core folk fineware tradition and its distinctive utilitarian assemblage could appear in Wari’s distant periphery fully formed, without local antecedents, and ubiquitous among the people who used it. If the ceramics are from the first half of the Middle Horizon, as the radiocarbon dates from Beringa suggest, the stepwise budding hypothesis has the additional advantage of suggesting how material culture might have evolved very rapidly en route to this early Middle Horizon settlement. In some regions, indigenous elites used Wari material culture for their own purposes, without imperial control or population intrusions (Chapdelaine; Marcone; Nelson et al., this volume). Elsewhere, sites

such as Pikillacta or Jincamocco still look like corporate, potentially imperial, establishments, where Wari traditions may have been spread through Inca-like corporate strategies (Bélisle and Covey; Meddens and Branch, this volume). In the Majes and Camaná drainage in the southern Pacific slopes of the Wari periphery, the Wari material culture horizon apparently spread through still other processes, perhaps including stepwise budding.

Acknowledgments Thanks to Tiffiny Tung, Ana Miranda, Joe Griffin, and the members of the Beringa project for making my work with the Beringa ceramics possible; to Henry Tantaleán and Merryl Martin Owen for help with many site visits; to Michael Malpass for sharing data; to Lic. Augusto Cardona for sharing his Camaná survey results and for support in Arequipa through the Centro de Investigaciones Arqueológicas de Arequipa (CIARQ), which he directed, funded by the Sociedad Minera Cerro Verde S.A.; to Lic. Marko López, Lic. Pablo de la Vera Cruz, and rest of the Instituto Nacional de Cultura in Arequipa for helpful conversations and practical support; to Prof. Augusto Mogrovejo for his help and friendship in Camaná; and to two anonymous reviewers for useful suggestions, including the “terminal Wari expansion” hypothesis and skepticism about the Beringa dates.

References Cited Anders, Martha B. 1986 Dual Organization and Calendars Inferred from the Planned Site of Azángaro: Wari Administrative Strategies. Unpublished Ph.D. dissertation, Cornell University, Ithaca, New York. 1998 El estilo Wamanga: Resistencia y subversión simbólica manifestadas en la cerámica del Horizonte Medio 2. Conchopata, Revista de Arqueología 1:138–162. Universidad Nacional de San Cristóbal de Huamanga, Facultad de Ciencias Sociales, Oficina de Investigación, Ayacucho, Peru. Bandy, Matthew S. 2004 Fissioning, Scalar Stress, and Social Evolution in Early Village Societies. American Anthropologist 106(2):322–333. Bennett, Wendell C. 1953 Excavations at Wari, Ayacucho, Peru. Yale University Publications in Anthropology 49. Yale University Press, New Haven, Connecticut. Blom, Deborah, Benedikt Hallgrimsson, Linda Keng, Maria C. Lozada C., and Jane E. Buikstra

1998 Tiwanaku “Colonization”: Bioarchaeological Implications for Migration in the Moquegua Valley, Peru. World Archaeology 30(2):238–261. Bronk Ramsey, Christopher 1995 Radiocarbon Calibration and Analysis of Stratigraphy: The OxCal Program. Radiocarbon 37(2):425–430. 2001 Development of the Radiocarbon Calibration Program OxCal. Radiocarbon 43(2A):355–363. Cardona, Augusto 2005 INC—Proyecto Qhapac Ñan Informe Final: Arequipa Costa. Unpublished survey report. Sonoma State University, Sonoma, California. Cohen, Robin 1997 Global Diasporas: An Introduction. University of Washington Press, Seattle. D’Altroy, Terence N. 1992 Provincial Power in the Inka Empire. Smithsonian Institution Press, Washington, D.C. 2001 State Goods in the Domestic Economy: The Inka Ceramic Assemblage. In Empire and Domestic Economy, edited by Terence N. D’Altroy and Christine A. Hastorf, pp. 243–264. Kluwer/Plenum, New York. D’Altroy, Terence N., and Christine A. Hastorf (editors) 2001 Empire and Domestic Economy. Kluwer/Plenum, New York. Disselhoff, Hans D. 1967 Huari und Tiahuanaco, Grabungen und Funde in Süd-Peru. Zeitschrift für Ethnologie 92(1):207–216. 1968 Oasenstädte und Zaubersteine im Land der Inka: Archäologische Forschungsreisen in Peru. Safari-Verlag, Berlin. Donnan, Christopher B. 1997 A Chimu-Inka Ceramic-Manufacturing Center from the North Coast of Peru. Latin American Antiquity 8(1):30–54. García Márquez, Manuel, and Rosa Bustamante Montoro 1990 Arqueología del Valle de Majes. Gaceta Arqueológica Andina 5(18/19):25–40. Goldstein, Paul S. 2005 Andean Diaspora: The Tiwanaku Colonies and the Origins of South American Empire. University of Florida Press, Gainesville. Goldstein, Paul S., and Bruce Owen 2001 Tiwanaku en Moquegua: Las colonias altiplanicas. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 139–168. Boletín de Arqueología PUCP No. 5. Pontificia Universidad Católica del Perú, Lima. Haeberli, Joerg 2001 Tiempo y tradición en Arequipa, Perú, y el surgimiento de la cronología del tema de la deidad central. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 89–137. Boletín de Arqueología PUCP No. 5. Pontificia Universidad Católica del Perú, Lima. Hoshower, Lisa M., Jane E. Buikstra, Paul S. Goldstein, and Ann D. Webster 1995 Artificial Cranial Deformation at the Omo M10 Site: A Tiwanaku Complex from the Moquegua Valley, Peru. Latin American Antiquity 6(2):145–164.

Wari in the M ajes - C amaná Valley

  75

Isbell, William H. 1977 The Rural Foundation for Urbanism: Economic and Stylistic Interaction between Rural and Urban Communities in EighthCentury Peru. Illinois Studies in Anthropology No. 10. University of Illinois Press, Urbana. 1991 Huari Administration and the Orthogonal Cellular Architecture Horizon. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 293– 316. Dumbarton Oaks, Washington, D.C. Isbell, William H., and Helaine Silverman 2006 Rethinking the Central Andean Co-Tradition. In Andean Archeology III: North and South, edited by William H. Isbell and Helaine Silverman, pp. 497–518. Springer, New York. Isla, Johny 2001 Wari en Palpa y Nasca: Perspectivas desde el punto de vista funerario. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 555–583. Boletín de Arqueología PUCP No. 5. Pontificia Universidad Católica del Perú, Lima. Jennings, Justin 2003 Inca Imperialism, Ritual Change, and Cosmological Continuity in the Cotahuasi Valley of Peru. Journal of Anthropological Research 59(4):433–462. Jennings, Justin, and Nathan Craig 2001 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 20(4):429–502. Knobloch, Patricia J. 1991 Stylistic Date of Ceramics from the Huari Centers. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 247–258. Dumbarton Oaks, Washington, D.C. Lumbreras, Luis 1974a Las fundaciones de Huamanga: Hacia una prehistoria de Ayacucho. Club Huamanga, Lima. 1974b Los reinos post-Tiwanaku en el area altiplanica. Revista del Museo Nacional, Lima 15:56–85. 1974c The Peoples and Cultures of Ancient Peru, translated by Betty J. Meggers. Smithsonian Institution Press, Washington, D.C. Lyon, Patricia J. 1968 A Redefinition of the Pinilla Style. Ñawpa Pacha 6:7–14. McCormac, Gerry, Alan Hogg, Paul Blackwell, Caitlin Buck, Thomas Higham, and Paula Reimer 2004 SHCa104 Southern Hemisphere Calibration, 0–11.0 cal kyr BP. Radiocarbon 46(3):1087–1092. Malpass, Michael A. 2001 Sonay: Un centro Wari celular ortogonal en el Valle de Camaná, Perú. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 51–68. Boletín de Arqueología PUCP No. 5. Pontificia Universidad Católica del Perú, Lima. Malpass, Michael A., Pablo de la Vera Cruz, Marko Lopez, Lucy Linares, Willy Yépez, and Carlos González 1991 Archaeological Reconnaissance of the Upper Camaná River Valley, Peru. Manuscript on file, Sonoma State University, Sonoma, California.

76 

B ruce Owen

Manrique Valdivia, Julio, and Manuel Cornejo Zegarra 1990 Visión sobre la arqueología del Valle de Camaná. Gaceta Arqueológica Andina 5(18/19):21–24. Menzel, Dorothy 1964 Style and Time in the Middle Horizon. Ñawpa Pacha 2:1–105. 1968 New Data on the Huari Empire in Middle Horizon Epoch 2A. Ñawpa Pacha 6:47–114. Moseley, Michael E. 2001 The Incas and Their Ancestors: The Archaeology of Peru. Rev. ed. Thames and Hudson, London. Moseley, Michael E., Robert A. Feldman, Paul S. Goldstein, and Luis Watanabe 1991 Colonies and Conquest: Tiahuanaco and Huari in Moquegua. In Huari Administrative Structure, Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 121–140. Dumbarton Oaks, Washington, D.C. Murra, John 1975 El control vertical de un máximo de pisos ecológicos en la economía de las sociedades andinas. In Formaciones económicas y políticas del mundo andino, edited by John Murra, pp. 59–115. Instituto de Estudios Peruanos, Lima. Neira A., Máximo 1990 Arequipa prehispánica. In Historia general de Arequipa, edited by Máximo Neira Vendaño, Guillermo Galdos R., Alejandro Málaga M., Eusebio Q. Paz S., and Juan G. Carpio M., pp. 5–213. Fundación M. J. Bustamante de la Fuente, Arequipa, Peru. Ochatoma P., José, y Martha Cabrera R. 2001 Poblados rurales Huari: Una vision desde Aqo Wayqo. Universidad Nacional de San Cristóbal de Huamanga, Facultad de Ciencias Sociales, Lima. Owen, Bruce 1995 Fieldnotes of the Chen Chen Cemetery Salvage Project. Notes on file, Museo Contisuyo, Moquegua, Peru. 2005 Distant Colonies and Explosive Collapse: The Two Stages of the Tiwanaku Diaspora in the Osmore Drainage. Latin American Antiquity 16(1):45–80. 2007 The Wari Heartland on the Arequipa Coast: Huamanga Ceramics from Beringa, Majes Valley, Peru. Andean Past 8:287–373. 2009 La expansión y el colapso de Tiwanaku y el papel de Arequipa. In Arqueología del área centro sur andina: Actas del Simposio Internacional 30 de Junio–2 de Julio de 2004, Arequipa Perú, edited by Mariusz Ziolkowski, Justin Jennings, Luis Augusto Belan Franco, and Andrea Drusini, pp. 431–462. Boletín del Centro de Estudios Precolombinos 7. University of Warsaw, Warsaw. Owen, Bruce, and Paul S. Goldstein 2001 Tiwanaku en Moquegua: Interacciones regionales y colapso. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 169–188. Boletín de Arqueología PUCP No. 5. Pontificia Universidad Católica del Perú, Lima. Paulsen, Allison Clement 1968 A Middle Horizon Tomb, Pinilla, Ica Valley, Peru. Ñawpa Pacha 6:1–6.

Rowe, John Howland 1946 Inca Culture at the Time of the Spanish Conquest. In Handbook of South American Indians, edited by Julian Steward, pp. 183–330. Bureau of American Ethnology, Bulletin No. 143. Smithsonian Institution Press, Washington, D.C. Schreiber, Katharina 1987 Conquest and Consolidation: A Comparison of the Wari and Inka Occupations of a Highland Peruvian Valley. American Antiquity 52(2):266–284. 1992 Wari Imperialism in Middle Horizon Peru. Museum of Anthropology, University of Michigan, Ann Arbor. Silverman, Helaine 1988 Cahuachi: Non-Urban Cultural Complexity on the South Coast of Peru. Journal of Field Archaeology 15(4):403–430. Stanish, Charles 1992 Ancient Andean Political Economy. University of Texas Press, Austin. 2001 Regional Research on the Inka. Journal of Archaeological Research 9(3):213–241. 2002 Tiwanaku Political Economy. In Andean Archaeology I: Variations in Sociopolitical Organization, edited by William H. Isbell and Helaine Silverman, pp. 169–198. Kluwer/Plenum, New York. Tantaleán, Henry, and Bruce Owen 2007 Complejos de patios con columnas de Camaná, levantamiento,

sondeos, y fechados para definir su filiación cultural y cronológica, informe de campo e informe final. Unpublished report to the Instituto Nacional de Cultura, Lima. Tung, Tiffiny 2003 A Bioarchaeological Perspective on Wari Imperialism in the Andes of Peru: A View from Heartland and Hinterland Skeletal Populations. Unpublished Ph.D. dissertation, Department of Anthropology, University of North Carolina, Chapel Hill. 2007 The Village of Beringa at the Periphery of the Wari Empire: A Site Overview and New Radiocarbon Dates. Andean Past 8:253–286. Tung, Tiffiny, and Bruce Owen 2006 Violence and Rural Lifeways at Two Peripheral Wari Sites in the Majes Valley of Southern Peru. In Andean Archaeology III: North and South, edited by W. Isbell and H. Silverman, pp. 435–467. Kluwer/Plenum, New York. Vivanco P., Cirilo, and Lidio M. Valdez C. 1993 Poblados Wari en la cuenca del Pampas-Qaracha, Ayacucho. Gaceta Arqueológica Andina 23:83–102. Williams, Patrick Ryan 2001 Cerro Baúl: A Wari Center on the Tiwanaku Frontier. Latin American Antiquity 12(1):67–83.

Wari in the M ajes - C amaná Valley

  77

Yanahuara Urubamba Maras

Sa

Chinchero

Vil

Xaquixaguana Plain Ak'awillay Izcuchaca (Anta)

Cusco Region

Calca

Cusco

cre

dV alle ca y no taUr ub am b

Peru San Salvador

aR

ive r Pikillaqta

Lucre

N

Huaro Modern Towns Archaeological Sites River

Paruro 0

15 km

Figure 5.1   The Cusco region: important Middle Horizon archaeological sites and modern towns mentioned in the text.

 

Ch a p t er 5

o Local Settlement Continuity and Wari Impact in Middle Horizon Cusco Véronique Bélisle and R . Alan Covey

A

rchaeologists have debated the administrative scope and intensity of the Wari polity for years. Some scholars argue that Wari constitutes the first Andean empire (e.g., Isbell 1989, 1997, 2000; Isbell and McEwan 1991; Lumbreras 1969; McEwan 2005a; Menzel 1964, 1968; Schaedel 1993; Schreiber 1992, 2001), while others describe kin-based confederations or the establishment of a loose state hegemony characterized by considerable local autonomy outside Ayacucho (e.g., Bawden 1982, 1983; Chapdelaine, this volume; Donnan and Mackey 1978; Jennings 2006a, 2006b; Jennings and Yépez Álvarez 2001; Mackey 1982, 1983; Marcone, this volume; Nelson et al., this volume; Segura and Shimada, this volume; Topic 1991; Topic and Topic, this volume; Wilson 1988). Significantly, the archaeological consideration of how Wari civic government might have been reconfigured and extrapolated at an imperial scale has been largely restricted to the study of Wari-style architecture and material culture, with limited survey and excavation data to contextualize Wari architectural complexes or tombs found throughout the central Andean highlands. In this chapter, we present such data for the Cusco region, where the presence of impressive Wari architectural and mortuary remains has led to its promotion by some scholars as the definitive

case study for provincial administration by an imperial Wari society. Settlement pattern data and the excavation of local houses, public architecture, and burials reveal some of the complex processes inherent to the evolution of Andean statecraft during the first millennium.

Contextualizing Wari Administrative Centers at a Regional Scale During the Middle Horizon (ca. AD 600–1000), Wari populations intensively settled the Lucre basin and Huaro Valley, located in the Cusco region of southern Peru (Figure 5.1). The scale of construction and evidence for administrative facilities at Wari sites in these areas have led scholars (Glowacki 2002; Glowacki and McEwan 2001; McEwan 2005c) to conclude that the Wari state directly controlled the Cusco region at this time. The impetus for developing provincial administration in the Cusco region was presumably the potential to intensify and centralize the regional agropastoral economy and to access coca trade routes and caravan movement into the Titicaca basin and territories controlled by the Tiwanaku state (Bauer 2004:64–69). Virtually all excavation research on the Cusco region Middle Horizon has focused on the largest Wari

  79

sites in the Lucre basin and Huaro Valley, developing perspectives on administration, elite residence, and mortuary practices. Rarely (e.g., Torres Poblete 1989) have scholars studied the smaller settlements to see the extent to which local groups across the Cusco region were affected by intrusive Wari settlement. Since imperial authorities never succeed in implementing a universal strategy of incorporation or domination, the impact of any empire often varies from one community to another within the empire and from one household to the next within the same community (Bauer and Covey 2002; Berdan et al. 1996; D’Altroy and Hastorf 2001; Hastorf 1990; Morris 1998; Schreiber 1987, 1992). If we were to investigate only the largest administrative centers, we could not determine the extent of Wari state control over the local people and resources of Cusco. To understand fully the expansion and incorporation strategies employed by any empire, it is essential to adopt a complementary “bottom-up” perspective that documents the changes in local settlement patterns and changes experienced by local families at the household level. It is also critical to be aware that different parts of the Cusco region may have been affected to varying degrees and at different times. Rather than assuming simultaneous incorporation and uniform control for the duration of the Wari presence, we must refine our chronology to assess the impact of Wari colonization on each part of the Cusco region and on all tiers of the regional settlement hierarchy. Our ongoing research is evaluating how the occupants of the Sacred Valley and Xaquixaguana Plain (see Figure 5.1) were affected by the presence of nearby Wari colonies during the Middle Horizon. Survey data from both regions provide the necessary spatial and temporal perspectives to assess continuity and change in local settlement before and after the establishment of Wari settlements in Cusco. Excavations at Ak’awillay, the largest site in the two study regions at this time, provide data on the architecture, artifact inventories, and mortuary practices of houses occupied before and during Wari colonization. These data document household changes from the pre-Wari to the Wari period and can be contextualized within broader considerations of the incorporation of the Cusco region into the Wari Empire. 80 

V éronique B élisle and R . A lan C ovey

The Wari Empire in Cusco In the Lucre basin and Huaro Valley, McEwan (1984, 1987, 1989, 1991, 1996, 2005a) and others (Barreda Murillo 1973; Glowacki 1996, 2002, 2005; Glowacki and McEwan 2001; McEwan 2005c; Zapata Rodríguez 1997) have investigated presumed Wari regional administrative centers, as well as a few Wari cemeteries. The two large installations are Pikillacta and the Huaro Archaeological Complex. Located about 30 km southeast of the modern city of Cusco (see Figure 5.1), Pikillacta adhered to Wari architectural canons, and its construction required a substantial investment of labor (McEwan 2005b; McEwan and Couture 2005). Excavators uncovered several Wari ceramic styles, and Instrumental Neutron Activation Analysis (INAA) suggests that some Wari vessels were locally produced while others were imported from the Ayacucho region (Glowacki 2005). McEwan (2005a) concludes that the site served a ceremonial and administrative function, but was not permanently occupied. Periodically Wari officials would have hosted and entertained their local subjects at feasts, which evidently functioned as asymmetrical reciprocity for labor exaction. In the niched halls at Pikillacta, the Wari could have worshipped their most important ancestors. McEwan (2005a) also believes that the small conjoined rooms of the site could have stored the mummies of the local lineages’ ancestors, partly to honor them and partly to exercise control over local populations. This model of Wari provincial administration draws from ethnohistorically known aspects of later Inca provincial governance, and it is important to note that we do not have any evidence that Wari populations mummified their dead as the Inca did. In fact, the mortuary remains at Huari and Conchopata in Ayacucho (Isbell 2004) and at the Wari cemetery of Batán Urqu in Cusco (Zapata Rodríguez 1997) all show permanent inhumations. Glowacki (2002) and McEwan (Glowacki and McEwan 2001) believe that the Wari population of Cusco lived at the nearby center of Huaro, to the east of Pikillacta (see Figure 5.1). The Huaro Archaeological Complex is a group of Wari administrative, residential, and ceremonial sites. Relying mostly on ceramics, Glowacki (2002) and McEwan (Glowacki and McEwan

2001) argue that the Huaro Archaeological Complex was built before Pikillacta and served as the first Wari administrative center in the Cusco region. They suggest that Pikillacta was built later, but never completed. These scholars have concluded that the Wari state controlled the Cusco region during the Middle Horizon and ruled directly over local populations by imposing political and economic domination (Glowacki 2002; McEwan 2005a:162–164).

(e.g., Bauer and Jones 2003). Such regional projects have systematically investigated well over 2000 km2 and identified more than twenty-eight hundred archaeological sites of all periods in the Cusco basin, the Paruro area, the Sacred Valley, and the Xaquixaguana Plain (see Figure 5.1). This paper focuses on the latter two regions, with occasional reference to published reports on other survey areas (e.g., Bauer 1992, 1999, 2004; Bauer and Jones 2003).

Middle Horizon Settlement in the Sacred Valley and Xaquixaguana Plain

The Ceramic Sequence During survey we used diagnostic ceramic fragments from surface collections to identify Middle Horizon components. Archaeological contexts associated with pottery styles utilized to identify the Middle Horizon and earlier and later periods have been radiocarbon dated (Bauer and Jones 2003:35–37). The Late Formative occupation (ca. 500 BC– AD 200) was easily recognized by its diagnostic pottery, which is fairly uniform across the Cusco basin, the Paruro area, the Sacred Valley, and the Xaquixaguana Plain. Formative pottery is characterized by thick rims,

As noted above, the interpretive argument over Wari peripheral control is based largely on site-level analyses at Wari installations. In the Cusco region, recent systematic surveys outside the Lucre basin and Huaro Valley provide data on regional settlement patterns and change through time, and on the distribution and variability of ceramic styles and architecture (e.g., Bauer 1992, 1999, 2004; Covey 2006). These surveys were followed by test excavations at selected sites of all periods in order to refine the regional chronology and ceramic typology

Figure 5.2  Late Formative pottery from Ak’awillay. Decorative techniques included incisions, painting, punctations, patterned burnishing, or a combination of these techniques.

L ocal S ettlement C ontinuity and Wari I mpact in M iddle H orizon C usco

  81

A

B

C

D

natural paste

orange

white

red

black

0

4 cm

Figure 5.3  Middle Horizon pottery styles from the Xaquixaguana Plain and Ak’awillay: Qotakalli (a), Muyu Urqu (b), Araway (c), Local (d).

burnished surfaces, and by incised, painted, punctate, or pattern-burnished decoration (Figure 5.2). The shoulder and upper body of jars and cooking pots are decorated, while bowls and plates are decorated on their interiors. Three Wari-period styles have been identified in our study area. Qotakalli ceramics are abundant in the Cusco region and easily recognized by their black-onwhite or black-and-red-on-white decoration (Figure 5.3a). This ceramic style was first produced during the Early Intermediate Period (EIP, ca. AD 200–600), and its 82 

V éronique B élisle and R . A lan C ovey

production continued throughout the Middle Horizon. Its distribution center was the Cusco basin, but it is found throughout the Cusco region (Bauer 1999, 2004; Bauer and Jones 2003). This style is characterized by thin walls and rims. Its motifs are geometric and include triangles, diamonds, dots, zigzags, and horizontal, vertical, and oblique lines. Geometric forms are sometimes filled with cross-hatching. Muyu Urqu is a polychrome ceramic style that was produced in the Cusco region toward the end of the EIP

until sometime during the Middle Horizon. It is present in the Xaquixaguana Plain but not in the Sacred Valley. Muyu Urqu is thought to be a Cusco regional style influenced by styles from the Titicaca basin (Bauer 1999, 2004; Bauer and Jones 2003). The vessels have thin walls and rims and polychrome decoration that is highly polished (Figure 5.3b). Motifs are painted in white, black, and orange on a dark red background. Most motifs are geometric and include stairs, circles, dots, nets, and lines. Zoomorphic and anthropomorphic motifs also exist, but are much rarer. Araway is a local style that was exclusively produced during the Middle Horizon. It was influenced by the Wari styles Okros (Bauer 1999:68) and Wamanga (Glowacki 2005:112), and recent INAA of Araway fragments from Pikillacta confirmed that Araway pottery was manufactured in the Cusco region (Glowacki 2005:112). Like Qotakalli, Araway is present throughout the Cusco area (Bauer 1999, 2004; Covey 2006:74–78; Glowacki 1996:199–207, 2002, 2005; Torres Poblete 1989; Zapata Rodríguez 1997) and its distribution might also have been centered in the Cusco basin. Motifs are

painted in black, white, and red on a buff background and include straight and curving lines, bands, checks, circles, triangles, dots, Xs, and crosses (Figure 5.3c).

The Sacred Valley Archaeological Survey (2000) The Sacred Valley Archaeological Project, directed by the second author, included a 400 km2 systematic survey in the Vilcanota-Urubamba Valley between the towns of San Salvador and Calca, a region in which more than four hundred archaeological sites were registered (Figure 5.4).1 A total of 131 EIP and Middle Horizon components were identified in this survey.

Disruption of Formative Settlement Patterns Formative Period settlement in the Sacred Valley consisted of village sites and hamlets commonly located on hilltops and ridges. The largest villages in the Sacred Valley survey area are 4–6 ha, with only modest evidence of a regional settlement hierarchy of large and

N

N

0

0

4 km

Small Village (1-5 ha.)

4 km

Small Village (1-5 ha.)

35

Figure 5.4  Map of Sacred Valley village sites with Qotakalli (left) and Araway (right) ceramics.

35 00

00

43

00

Patabamba 390

0

43 0

0

00

00

31

31

390 0

Willkarayan Wanka

QOTAKALLI VILLAGES

ARAWAY VILLAGES

L ocal S ettlement C ontinuity and Wari I mpact in M iddle H orizon C usco

  83

small villages. As with other parts of the Cusco region, the initiation of polychrome ceramic production (the Qotakalli style and related local variants) occurred at the same time as a significant regional settlement shift, with Formative high elevation villages being abandoned in favor of dispersed small villages and hamlets in lower elevation areas, especially in the Chit’apampa basin just to the north of the Cusco basin. More than threequarters of Formative sites in the Sacred Valley region (77 percent, fifty-four of seventy) lack evidence of continued settlement in the EIP, including nearly all sites larger than 1 ha. The Sacred Valley lacks a distinct local EIP/Middle Horizon ceramic style, demonstrating instead strong stylistic ties to the Cusco basin. Qotakalli pottery is present at forty-five sites in the region, virtually all of which lie to the south of the Vilcanota-Urubamba River. The most distant of these sites and many of the hamlets identified have significant components of what appear to be local variants of the Qotakalli style (as well as local imitations of the Araway style). Overall, the Sacred Valley exhibits a pre-Wari shift to valley-bottom agriculture, focusing on well-watered side valley areas not requiring hydraulic works. Small villages and hamlets are dispersed widely throughout the region, with low levels of population in the Sacred Valley proper.

Wari Impact on the Sacred Valley The Sacred Valley presents an important test for Wari imperial influence because of its close proximity to the Lucre basin, as well as its potential for developing highly productive hydraulic maize agriculture. A direct Wari administration of the Cusco region that controlled the local staple economy would be expected to involve intensification projects in the Sacred Valley, yet settlement in the valley proper is sparse and typically located 200–300 m above the valley floor. Araway ceramics were present at forty-two sites generally located to the south of the Vilcanota-Urubamba River, while Wari styles were identified at only two sites in the Sacred Valley. The distribution of ceramics suggests limited exchange relationships with the Wari based at Pikillacta and Huaro, with a slight tendency toward greater nucleation over time. 84 

V éronique B élisle and R . A lan C ovey

It is worth noting that Araway pottery is usually found at sites with Qotakalli ceramics, suggesting continuity from pre-Wari times. Where Qotakalli is absent, Araway is found almost exclusively at sites with a prominent Late Intermediate Period (AD 1000–1400) occupation, suggesting that deviation from the local settlement pattern is part of the post-Wari transition to a more balkanized settlement pattern.

The Xaquixaguana Plain Archaeological Survey (2004–2006) The Xaquixaguana Plain Archaeological Survey consisted of an area of approximately 600 km2, located northwest of the Cusco basin (see Figure 5.1). In the systematic survey of the plain, directed by the second author, we registered a total of 141 components dating to the EIP and the Middle Horizon.

Distribution of EIP and Middle Horizon Sites There are two clusters of EIP andMiddle Horizon sites in the Xaquixaguana Plain. The first is located in the northern part of the survey area near the modern town of Maras (Figure 5.5). It consists of thirty-two small, scattered sites, the majority of which are small hamlets less than a half hectare in size. Some sites in the northern cluster represent small Araway components at large Late Intermediate Period villages that were founded in the Maras area during the Wari decline. The second cluster of EIP andMiddle Horizon sites is located in the southern section of the Xaquixaguana Plain around the modern towns of Pucyura and Izcuchaca (see Figure 5.5). Of all forty-seven sites in that cluster, more than half (twenty-five sites) are small hamlets less than a half hectare in size that are distributed evenly throughout the cluster. There is a considerable number of small villages (twenty-one sites, with an average size of 1.4 ha) distributed throughout the cluster, and one large village, Ak’awillay, covers 10 ha. Ak’awillay is the largest known Middle Horizon settlement in the Cusco region outside the Lucre basin and Huaro Valley, and thus at the top of the regional sitesize hierarchy.

Site of the northern cluster Site of the southern cluster Modern town

Urubamba Urubamba River

N

Maras

Chinchero

Lake Huaypo

Lake Piuray

Huarocondo

Ak’awillay Izcuchaca Pucyura

Cusco 0

10 km

Figure 5.5  The two clusters of EIP and Middle Horizon settlements in the Xaquixaguana Plain.

Disruption of Formative Settlement Patterns Like the Sacred Valley, the Xaquixaguana Plain settlement pattern of the EIP and Middle Horizon represents a significant change from the Formative Period. Most EIP and Middle Horizon sites (70 percent, or fifty-five of seventy-nine) lack evidence of continued settlement from the Formative, and during the EIP Formative hilltop sites were abandoned in favor of new settlements established at lower elevations. The plain around Lake Huaypo, a higher tuber-producing area intensively occupied during the Formative, was completely abandoned in the EIP and Middle Horizon (see Figure 5.5). Like the Sacred Valley, EIP and Middle Horizon pottery from the Xaquixaguana Plain shows strong stylistic ties to the Cusco basin. Qotakalli pottery is widely distributed and present at forty-eight sites throughout the survey area. Muyu Urqu pottery is present at seventeen sites, all of which lie in the southern cluster (and thus close to the Cusco basin). Overall, the Xaquixaguana Plain exhibits a pre-Wari shift to valley-bottom agriculture, with EIP hamlets and villages established around maize-growing areas.

Wari Impact on the Xaquixaguana Plain The Xaquixaguana Plain contains fertile maize lands and lies on a principal caravan route from the Cusco region toward the Ayacucho area. If direct Wari imperial control was continuous, Wari colonies or administrative installations would be expected in the Xaquixaguana Valley, with significant impacts on Ak’awillay, the region’s largest settlement. We did not find any Wari architecture in this region, and Wari pottery is present at only two sites. Instead, there is substantial continuity of the existing local settlement pattern between the EIP and the Middle Horizon. Araway ceramics are present at thirty-three sites throughout the area, generally where Qotakalli pottery is also present. This evidence suggests that the people of the Xaquixaguana Plain obtained decorated pottery and small amounts of Wari pottery from the Cusco basin. Where Qotakalli is absent, Araway represents small components at large Late Intermediate Period sites, suggesting a post-Wari Late Middle Horizon settlement shift.

EIP and Middle Horizon Settlement Patterns in Other Areas The EIP and Middle Horizon settlement patterns of the Sacred Valley and Xaquixaguana Plain present some interesting parallels with the neighboring Cusco basin and nearby Paruro region. In the Cusco basin the EIP settlement pattern reveals a site-size hierarchy suggestive of a nonstate complex polity. In this period people started to occupy lower valley slopes where maize agriculture is possible, abandoning most hilltop sites. During the Middle Horizon, the settlement pattern of the Cusco basin changed very little, and most sites continued to be occupied (Bauer 2004:65). To the south of the Cusco basin, survey research in the Paruro region demonstrates a similar abandonment of Formative sites (twenty of thirty-one Formative sites have no EIP component) (Bauer 1999), with a comparable stability of local EIP settlements over time. Bauer identified Wari ceramics in the area closest to the Lucre basin, but found Wari to be the dominant style at only one site, which appears to have been a ceremonial location.

L ocal S ettlement C ontinuity and Wari I mpact in M iddle H orizon C usco

  85

Local Interaction with the Cusco Basin and Wari Colonies The distribution of pottery styles in the Sacred Valley and Xaquixaguana Plain suggests that communities living there interacted regularly with the Cusco basin polity and other neighboring regions, and less frequently with the Wari populations of the Lucre basin and Huaro Valley. The dominant ceramic style (apart from the local pottery) is Qotakalli. People living in both survey regions produced some local imitations of Cusco basin pottery such as Qotakalli, but fewer imitations of Wari. Qotakalli is present at ninety-three sites. The density of a site’s Qotakalli component is correlated with that site’s distance from Cusco; in other words, sites closer to the Cusco basin have more Qotakalli pottery. Groups living farther from Cusco tended to have more locally produced pottery along with imitations of Cusco basin styles. The smaller proportion of Qotakalli pottery at these northern sites may indicate sporadic or indirect contact between the Cusco basin and the northern sections of the Sacred Valley and the Xaquixaguana Plain. The Araway style is less prevalent than Qotakalli throughout the region. Where present, Araway is found with local, Qotakalli, and sometimes Muyu Urqu pottery, at sites that were already established in the EIP. Araway pottery is also found at Late Intermediate Period sites, which may represent Late Middle Horizon settlement shifts. Wari pottery is present at only two sites in the Sacred Valley and two on the Xaquixaguana Plain, including Ak’awillay. At these sites Wari pottery represents a very small portion of the Middle Horizon decorated ceramic assemblage, and local as well as Qotakalli and Muyu Urqu pottery is much more abundant. Data from the two archaeological surveys indicate that there was limited Wari “influence” in both regions—there is no Wari architecture and there is little Wari pottery at a small number of local settlements. This evidence suggests that the presence of the Wari in Cusco stimulated some contacts with the Ayacucho region. Regional production and exchange patterns were already well established centuries before Wari colonies were settled in the Lucre basin and Huaro Valley, however, and the presence of Qotakalli and Muyu Urqu pottery at EIP sites already showed an increased participation in 86 

V éronique B élisle and R . A lan C ovey

regional prestige goods economies during the EIP. The changes between Formative and EIP settlement patterns are consistent with a pre-Wari shift in regional agricultural production strategies, and the centuries of Wari occupation of the Lucre basin and Huaro area did not result in significant changes in settlement location or hierarchy in either study region.

Understanding Wari Impact on a Local Community The surveys of the Sacred Valley and Xaquixaguana Plain provide valuable regional data on changes in settlement size and distribution throughout the EIP and Middle Horizon. We saw little evidence of direct Wari administration in the settlement patterns, but we need additional evidence to determine how and if people at a local village interacted with Wari caravans and colonists. In light of these survey data and to supplement emerging regional patterns, the first author directed excavations at the village of Ak’awillay, developing a detailed view of a single community in the Cusco area. Although Ak’awillay was occupied in the pre-Wari period, it went on to become the largest site in the area during the Middle Horizon, reaching a size of 10 ha. Located approximately 20 km to the west of the city of Cusco, Ak’awillay occupies a strategic position on a low hill overlooking the surrounding Xaquixaguana Plain. Its location is near the principal Inca highway route from Cusco to Ayacucho. Schreiber (1984) has observed that the Inca rehabilitated roads originally built by the Wari in developing their administration of the central highlands. If the major Inca imperial highway to Chinchaysuyu was based on roads used in Wari times, then the llama caravans between Ayacucho and the Cusco Valley should also have passed close to Ak’awillay. To understand the nature and intensity of interaction between the occupants of Ak’awillay and the residents of the nearby Wari colonies, the first author excavated pre-Wari and Wari period houses. The comparison of these houses and their associated artifact inventories permitted the assessment of continuity and change in household organization and activities over

0

Modern Village of Piñanccay

TP-8

200 m

Excavation units Test pits TP-7

N

34

TP-6

34 34

TP-5

24

34

B

44

54 34

TP-4

3474

A

E

3484

TP-3 H

C F

3464

time. Excavations provided multiple lines of evidence for study, and we considered material culture, architecture, and mortuary contexts to determine the degree of Wari impact on local households. Using these lines of evidence, we considered three possibilities: (1) that there was no Wari administration at Ak’awillay and minimal interaction with Wari, that is, life for the people of Ak’awillay went on virtually unchanged during the Middle Horizon; (2) that there was some impact seen in the imitation of Wari pottery, access to long-distance trade items (like obsidian), and perhaps modifications to cultural identity as seen in burials and household architecture; or (3) that there was a direct Wari imperial impact at Ak’awillay that is not seen in settlement pattern shifts and surface ceramics, but evidenced in the modification of site plans and the introduction of nonlocal administrative buildings and spaces, as well as possibly religious buildings.

D G

TP-1 3504

TP-2

3494

Izcuchaca

84

34

Excavations at the Village of Ak’awillay (2006–2007) Ak’awillay is a multicomponent site and, like most preInca sites in the Cusco region, there is no architecture preserved on the site’s surface. We first made intensive collections at Ak’awillay, placing eighty 50 m2 collection units in a grid across the site. These systematic surface collections revealed spatial patterning of ceramic style distributions and densities over time, which allowed the first author to evaluate the extent and nature of the surface scatter and then select locations for excavation. The Middle Horizon occupation is concentrated on the upper slope of the hill and on the hilltop (Figure 5.6), where we collected Qotakalli, Muyu Urqu, and some Araway pottery. The Formative component is concentrated on the hilltop but also appears below Middle Horizon levels on the hillside. Test pits on the lower slope of the hill and on the plain to the north indicated that surface materials in these areas were eroded from the upper slope and hilltop. No occupation took place there. Close to the hilltop is a pre-Hispanic cemetery with tombs placed in bedrock. Few bones and no material culture remain today, and dating the cemetery was not possible at this stage of the research.

Figure 5.6  The archaeological site of Ak’awillay showing the excavation units and test pits. The shaded area corresponds to the site’s approximate Wari period boundary based on our work.

The first author excavated a total area of 295 m2 at Ak’awillay, concentrating on the upper slope of the hill and combining horizontal and vertical excavation to expose complete structures and compare households from different periods. In this area of the site we excavated two circular stone structures that proved to be houses (see Figure 5.6, Units C and F), as well as two domestic areas without architecture (Units D and E), and one public circular stone building (Unit G). In order to document changes between the Wari and preWari periods, we also excavated pre–Middle Horizon domestic structures under the public building (Unit G) and close to the hilltop (Unit H). Research is ongoing, and the first author would like to emphasize that the following results are preliminary. We do not yet have radiocarbon dates for Ak’awillay; however, the presence of some Wari and Araway pottery

L ocal S ettlement C ontinuity and Wari I mpact in M iddle H orizon C usco

  87

pit

N burial

ash

x x

x hearth

ash

ash

x x

deer bones

pit camelid bones

pit x

x xx

ash

Circular Stone House

natural paste

red

cream

black

ra 0

4 cm

Figure 5.7  Wari (top) and Araway (bottom) pottery from Ak’awillay, found in the public building of Unit G and the house of Unit F, respectively.

in our contexts dates the domestic structures and public building to the Middle Horizon (Figure 5.7). Future work on obsidian sourcing from Ak’awillay might also alter some of our interpretations.

The Houses, Outdoor Spaces, and Public Building of Ak’awillay Middle Horizon Houses The two Middle Horizon houses at Ak’awillay (Units C and F) have circular stone foundations (Figure 5.8). We found these foundations at 1.5 and 2 m below the surface, well below the plow zone, so the contents of the houses have not been modified by modern activities. The house foundations consist of one row of rectangular stones placed close to bedrock. Above the course of stones the walls were evidently built of adobe and possibly cane. These circular residences are built side by side in one area of the site. The discovery of parts of other circular walls close to these houses suggests that more houses exist in that part of the site, which will be excavated at a later date. The two residential structures are 5 m in diameter and contained the same types of features and artifacts. 88 

V éronique B élisle and R . A lan C ovey

animal bone x

stone tool

0

1m

Figure 5.8  House plan and occupation floor of Unit C.

Each house is associated with a large number of andesite and obsidian flakes and tools, animal bones, and pottery sherds. Qotakalli and Muyu Urqu ceramics are abundant, and Araway is present but much less common. These three pottery styles are never represented by cooking pots; instead, bowls, drinking vessels, and jars were used to serve foods and liquids. Villagers at Ak’awillay also imitated the Qotakalli style and produced their own coarse domestic pots and simply painted vessels (see Figure 5.3d). Decorated and undecorated Formativelike pottery also continued to be utilized in the Middle Horizon, and we found a considerable component of Formative-style fragments mixed with Qotakalli, Muyu Urqu, and Araway pottery in the two houses. In these contexts, the Formative styles correspond to cooking, serving, and storage vessels. The house of  Unit C contained several ash lenses and one square hearth made of four elongated flat stones. The house contained three deep circular storage pits dug into the ground. In these pits we found undecorated domestic and decorated fine pottery, but no macrobotanical remains. The absence of such remains, however, could be due to the poor preservation of organic materials in the region. We found several grinding stones and stone tools, including an

axe, laminated andesite knives, and obsidian projectile points. Obsidian projectile points are small (2 cm long or less) and finely made, and they usually have a concave base, although we also documented flat bases. Bones of camelid, deer, guinea pig, and bird (yet to be defined more specifically) are abundant. The house of Unit F contained a rectangular hearth made with small and large stones. Located in the center of the house, this hearth was filled with ash and contained animal bones and a few pottery sherds. Instead of circular storage pits dug into the ground, this house had a rectangular stone storage unit of 1.5 × 1.2 m. Its foundation was built with one course of rectangular stones of a size similar to the ones used for the house foundation. It had three small walls, leaving the northern side open for easy access. Similar to the house of Unit C, we found grinding stones, andesite and obsidian tools, carved bone tupus (shawl pins) and sewing needles, and camelid, deer, guinea pig, and bird bones. We also recovered a pendant made of marine shell, a material that is rare in the region for that time period.

Middle Horizon Kitchen and Patio In Unit D we recovered the remains of a cooking area. There we found two hearths within 1 m of each other. The first one was dug into the ground and filled with ash, with no stones around it. The second hearth had two burned rectangular stones on each side, which were probably used to support a cooking pot. To the south of these hearths was a storage pit dug into the ground in which we found decorated and undecorated pottery, obsidian, and a bone needle. This unit contained several ceramic spoons and ladles, some obsidian projectile points and retouched flakes, several carved bones, and numerous animal bones. The pottery of this unit is not as finely decorated as that from the two houses described previously. The Qotakalli, Muyu Urqu, and Araway styles are present, but local and Formative-style domestic pottery is much more abundant. This evidence suggests that this area of the site was involved in food processing, preparation, and storage. Unit E seems to have been an outdoor patio to the east of the house of Unit C. Here there were no hearths

or storage pits, but we recovered a large number of undecorated and decorated sherds on the two occupation floors. Qotakalli and Muyu Urqu ceramics are present, but there are no Araway fragments. This patio contained some obsidian and carved and decorated bones.

Middle Horizon Public Building Unit G corresponds to a public building. It is circular, but larger than the houses—it is 11 m in diameter. Its foundation is made of three to four rows of small circular stones. In front of and about 35 cm below these small stones in the interior wall appeared one row of large rectangular stones. These bigger stones could have formed a bench all around the structure, or they could have been a retaining wall on which posts were erected to support a roof. In some parts we found reddish clay on these large stones, suggesting that adobes were placed on top of them. This structure was associated with a large number of artifacts. The fanciest specimens of Qotakalli and Muyu Urqu, as well as most of the Araway and a small number of Wari fragments, were found in this public building. Here we recovered more serving and drinking vessels than storage or cooking pots. As in other contexts described earlier, all Middle Horizon layers also contained undecorated and decorated Formative-style pottery. This building also contained a large number of pieces of obsidian. Some projectile points and débitage were found, but most of the obsidian consists of re­­ touched flakes. We recovered some grinding stones, stone and ceramic figurines, spindle whorls, and animal bones carved into tupus, sewing needles, beads, earrings, musical instruments, and snuff tubes. This structure also contained a substantial number of camelid, deer, guinea pig, and bird bones. Unlike the houses described earlier, no hearths or storage units were associated with this structure. The absence of such domestic features makes us believe that this structure, rather than representing an elite house, was a meeting place where food was consumed, tools manufactured, and ceremonies conducted. We did not find any Wari architecture at Ak’awillay. Typical Wari elements such as D-shaped structures, multistoried buildings, niched halls, and residential patio

L ocal S ettlement C ontinuity and Wari I mpact in M iddle H orizon C usco

  89

groups with elongated galleries surrounding a central patio are absent at Ak’awillay.

EIP and Formative Houses Below the public building just described we found parts of two structures associated with EIP and Late Formative materials. These structures are similar to the Middle Horizon houses and seem to be circular. Their foundation was made with one row of large stones, and they are associated with abundant decorated and undecorated pottery. The EIP house contains some Qotakalli as well as local and Formative-style pottery, and the Formative house is associated exclusively with Formative materials. The stratigraphic data suggest that there was no break between the occupations of the Formative, EIP, and Middle Horizon periods in Unit G. Upon abandonment, the Formative house was partially covered to build the EIP house; once abandoned, the EIP house was covered with a thick layer of dirt, ash, and trash that created a leveled surface on top of which the public building was constructed. Close to the hilltop of the site we excavated another Formative house, Unit H. This structure is semisubterranean and has an irregular shape. It is built with a stone wall on its west side and uses bedrock as a natural

wall on the east side. The west wall is built with three to four rows of medium to large stones. This house is associated with one occupation floor containing a hearth, a storage bin, animal bones, and Formative pottery. We also recovered obsidian projectile points and retouched flakes, grinding stones, tupus and sewing needles made of bone, zoomorphic and anthropomorphic figurines, and spindle whorls. To the north of the house is a small contemporary trash pit that contained similar materials. Once abandoned, this house was used as a community midden; during the Middle Horizon, the same space was then repurposed as a small cemetery.

The Burials of Ak’awillay The first author excavated fifteen burials containing a total of nineteen individuals at Ak’awillay, all of which belong to the Middle Horizon. Eight of these burials (twelve individuals) come from the small cemetery (Unit H). Four burials (each containing one individual) were placed at the base of the public building (Unit G) and likely represent dedicatory offerings to that structure. One individual was buried under the floor of the house of Unit C, one was buried close to the kitchen of Unit D, and the last one was buried in Unit A, a small unit without architecture about 50 m northeast of Unit D.

Figure 5.9  A burial containing one complete seated and flexed individual (on the left) and one incomplete individual (on the right). The bodies were deposited in a pit in the small Middle Horizon cemetery of Unit H.

90 

V éronique B élisle and R . A lan C ovey

The Middle Horizon burials from Ak’awillay are very similar to the Formative burials excavated elsewhere in the Cusco region (Rowe 1944:13–14). Each burial at Ak’awillay contains from one to three skeletons (Figure 5.9). Individuals were buried in a flexed and seated position, although sometimes they were flexed but on their back or side. Some skeletons are incomplete, sometimes because of intentional omission and in other cases because of poor bone preservation. The individuals were generally buried in a circular or oval pit dug into the earth. In five cases large stones were placed around and on top of the bones of an individual, and in one case a child was located in a cist tomb. The fill used to cover the skeletons was loose dirt that contained pottery sherds and animal bones. None of the burials have ceramic offerings, but one burial included a guinea pig , and another included a bird. Wari burials from the Cusco region are generally more elaborate than the Middle Horizon burials from Ak’awillay. In the Huaro Valley, Glowacki (2002) and Zapata Rodríguez (1997) excavated several burials from Wari domestic sites and from the Wari cemetery of Batán Urqu. Tomb types include rectangular chambers, rectangular stone-lined tombs, cist burials, and unlined pits; all tombs were covered with dirt or with flat stones. Burials contain one or several individuals, and skeletons are seated and flexed or semi-flexed and lying on their side. Almost all of these Wari burials contain ceramic vessels (Wari but also Araway, Qotakalli, and other styles), and some also contain metal objects, camelid offerings, and Spondylus, lapis lazuli, and turquoise beads. No such burial was found at Ak’awillay.

Continuity at the Village Level in Middle Horizon Cusco The evidence obtained from the excavations at Ak’awillay suggests that the Wari state had limited impact on the local communities of the Xaquixaguana Plain. There is continuity from Formative to Middle Horizon houses: they have stone foundations, are generally circular, and have a similar inventory of items and features. The stratigraphic data also indicate that there was no break in the occupation of Ak’awillay—in other words, the site was

not abandoned but continuously occupied from the Formative onward. The very low frequency of Wari pottery suggests that the villagers who lived at Ak’awillay interacted infrequently with the Wari populations based at Pikillacta and Huaro or were not capable of or interested in obtaining Wari vessels from intermediate groups in the Cusco basin. If procuring Wari pottery had been difficult but desired, interested locals could have imitated the style and produced it locally. The fact that they did not suggests that they were simply not interested in this pottery. People at Ak’awillay were apparently more interested in obtaining and imitating Cusco basin pottery, as the presence of Qotakalli, local Qotakalli, and Muyu Urqu ceramics indicates. The interaction of the people of Ak’awillay with the Cusco basin polity was much more regular, intense, and long term and was not disrupted by the Wari settlement in the Cusco region. In addition to the ceramic evidence, the presence of obsidian in Formative, EIP, and Middle Horizon contexts and of marine shell in Middle Horizon contexts at Ak’awillay suggests that although the people of this village may have interacted at times with Wari caravans and colonists, they were also connected to other groups of the region and to populations still farther away. The scarcity of Wari materials does not appear to be related to the difficulty of interacting with groups from other parts of the Cusco region. Rather, it seems as though the people of Ak’awillay did belong to a wide network of exchange, and Wari traders represented one of these exchange partners. The mortuary data also show continuity from the Formative to the Middle Horizon. Although the flexed and seated position of the body is similar for Ak’awillay burials and Wari burials from Huaro, it was already in use during the Formative Period in other parts of Cusco. Burial patterns of the pre-Wari and Wari periods were similar in the Cusco region, and it appears that Wari did not influence these funerary rites.

Conclusions The changes in EIP settlement patterns preceded the establishment of Wari settlements in the Cusco region and

L ocal S ettlement C ontinuity and Wari I mpact in M iddle H orizon C usco

  91

are likely to have been stimulated by the adoption of new agricultural production strategies, which in turn encouraged the growth and expansion of complex societies. The lack of Wari sites in the Sacred Valley and Xaquixaguana Plain and the presence of limited Wari pottery at local sites along with ceramics from other parts of the Cusco region indicate that Wari impact on this region was weak and indirect. If Wari had controlled the Xaquixaguana Plain, the largest village of the region—Ak’awillay—should include Wari material culture in much larger proportions. Instead, local pottery and the ceramic styles produced in the Cusco basin are much more abundant at Ak’awillay than Wari pottery. The lack of well-defined Wari sites in the Xaquixaguana Plain region suggests a less-developed regional caravan system and a discontinuous state presence throughout the region. At Ak’awillay, continuity in architecture and burial patterns and the production and use of local ceramics indicate that Wari control in the Cusco region was limited geographically to areas in close proximity to the largest administrative centers, with more limited influence in villages lying at a distance from these sites. The appearance of non-Wari pottery styles from other parts of the Cusco area indicates that large villages like Ak’awillay had direct contact with other local centers of Cusco, and their exchange relationships with Wari might have been mediated through these local centers. People at Ak’awillay were able to acquire decorated pottery, obsidian, and marine shell without strong links to Wari exchange networks, and they maintained their own household economies and local cultural identity with little evidence of adopting the customs of their Wari neighbors. The scarcity of Wari and Wari-like material culture at Ak’awillay suggests that the villagers knew about the presence of Wari in Cusco but interacted infrequently with Wari settlers—this may also indicate a simultaneous disinterest on the part of the Wari state. By contrast, state-directed construction projects at Pikillacta indicate imperial aspirations to a more direct regional administrative control, although this was not fully mobilized before Wari centers were abandoned and burned after AD 900. Wari influence in the Cusco region appears to have developed through regional prestige goods networks, 92 

V éronique B élisle and R . A lan C ovey

leading to the establishment of farming colonies early in the Middle Horizon. Over time, the Wari state organized projects to establish a regional administrative presence that may have targeted the extraction of natural resources in the vicinity of the Lucre basin and Huaro Valley, as well as the exploitation of trade routes to obtain products from the tropical forest and the Titicaca basin. The strategy employed by the Wari state was not to annex the entire region and impose provincial administration, but rather to intensify its direct control outward from areas where it was established. This discontinuous process of empire building left some large villages like Ak’awillay largely unaffected, although it is unclear at present whether this is due to local resistance or the fact that the Wari state was not successful or interested in advancing its imperial program to the point where all local centers would be under some degree of provincial governance. The Middle Horizon in the Cusco region was a time of coexistence of various autonomous but interacting groups, including the Xaquixaguana Plain polity based at Ak’awillay, the developing installations of the Wari state, and the Cusco basin polity. Unlike the study of the large administrative centers, regional surveys and the study of a local village’s houses suggest that the local population of the Sacred Valley and Xaquixaguana Plain did interact occasionally with the Wari settlers and administrators in the area, but that this interaction was not based on asymmetrical or hierarchical terms. Our conclusions provide a rich counterpart to the perspectives emerging from the study of Pikillacta and Huaro, the two Wari centers in the Cusco region. Studying local settlements away from Wari centers is critical for developing dynamic perspectives on the growth of the Wari Empire. Only by studying both the imperial centers and the local villages can we get a complete picture of an empire’s trajectory and strategies.

Acknowledgments The first author thanks the Wenner-Gren Foundation, the National Science Foundation (BCS-0726568), the Social Sciences and Humanities Research Council of Canada, Rackham Graduate School of the University

of Michigan, and the Museum of Anthropology of the University of Michigan for funding the excavations at Ak’awillay. This research would not have been possible without the support and input of various people, including Joyce Marcus, R. Alan Covey, Kent Flannery, Carla Sinopoli, Bruce Mannheim, and Brian Bauer. In Peru, the first author counted on the valuable support of Vicentina Galiano Blanco, Rosa Galiano Blanco, Megalith Galiano Blanco, Jesús Galiano Blanco, Amanda Gamboa Cárdenas, Biviano Quispe Huallpa, Maeve Skidmore, Carlos Delgado González, David Jara, the personnel from the National Institute of Culture, and field-workers from the community of Piñanccay. The second author’s survey of the Sacred Valley was supported by a FulbrightHays DDRA Fellowship, the Wenner-Gren Foundation, the National Science Foundation (BCS-0135913), and the Heinz Foundation. The survey of the Xaquixaguana Plain was funded through grants to the second author by the National Science Foundation (BCS-0342381), the National Geographic Society, and the American Museum of Natural History. Vicentina Galiano Blanco drew the pottery. We would also like to thank Justin Jennings for organizing the SAA symposium “Beyond Wari Walls” and putting this book together, and Bill Isbell and two anonymous reviewers for providing comments on drafts of this paper. All errors remain ours.

Endnotes 1. The 2007 Calca-Yanahuara Archaeological Project continued survey research downriver to Yanahuara but identified minimal evidence of EIP andMiddle Horizon occupation (Covey et al. 2008).

References Cited Barreda Murillo, Luis 1973 Las culturas Inka y pre-Inka de Cuzco. Unpublished licensing thesis, Social Sciences Faculty, Universidad Nacional San Antonio Abad del Cusco, Cusco, Peru. Bauer, Brian S. 1992 The Development of the Inca State. University of Texas Press, Austin. 1999 The Early Ceramics of the Inca Heartland. Fieldiana New Series No. 31. Field Museum of Natural History, Chicago. 2004 Ancient Cuzco: Heartland of the Inca. University of Texas Press, Austin.

Bauer, Brian S., and R. Alan Covey 2002 Process of State Formation in the Inca Heartland (Cuzco, Peru). American Anthropologist 104(3):846–864. Bauer, Brian S., and Bradford M. Jones 2003 Early Intermediate and Middle Horizon Ceramic Styles of the Cuzco Valley. Fieldiana New Series No. 34. Field Museum of Natural History, Chicago. Bawden, Garth 1982 Galindo: A Study in Cultural Transition during the Middle Horizon. In Chan Chan: Andean Desert City, edited by Michael E. Moseley and Kent C. Day, pp. 285–320. University of New Mexico Press, Albuquerque. 1983 Cultural Reconstitution in the Late Moche Period: A Case Study in Multidimensional Stylistic Analysis. In Civilization in the Ancient Americas, edited by Richard M. Leventhal and Alan L. Kolata, pp. 211–235. University of New Mexico Press/ Peabody Museum of Archaeology and Ethnology, Harvard University, Cambridge, Massachusetts. Berdan, Frances F., Richard E. Blanton, Elizabeth Hill Boone, Mary G. Hodge, Michael E. Smith, and Emily Umberger 1996 Aztec Imperial Strategies. Dumbarton Oaks, Washington, D.C. Covey, R. Alan 2006 How the Incas Built Their Heartland: State Formation and the Innovation of Imperial Strategies in the Sacred Valley, Peru. University of Michigan Press, Ann Arbor. Covey, R. Alan, Miriam Araóz Silva, and Brian S. Bauer 2008 Settlement Patterns in the Yucay Valley and Neighboring Areas. In Imperial Transformations in Sixteenth-Century Yucay, Peru, edited by R. Alan Covey and Donato Amado González, pp. 3–17. Memoirs of the Museum of Anthropology No. 44. University of Michigan, Ann Arbor. D’Altroy, Terence N., and Christine A. Hastorf (editors) 2001 Empire and Domestic Economy. Kluwer Academic/Plenum, New York. Donnan, Christopher B., and Carol J. Mackey 1978 Ancient Burial Patterns of the Moche Valley, Peru. University of Texas Press, Austin. Glowacki, Mary 1996 The Wari Occupation of the Southern Highlands of Peru: A Ceramic Perspective from the Site of Pikillacta. Unpublished Ph.D. dissertation, Department of Anthropology, Brandeis University, Waltham, Massachusetts. 2002 The Huaro Archaeological Site Complex: Rethinking the Huari Occupation of Cuzco. In Andean Archaeology I: Variations of Sociopolitical Organization, edited by William H. Isbell and Helaine Silverman, pp. 267–285. Kluwer Academic, New York. 2005 Pottery from Pikillacta. In Pikillacta: The Wari Empire in Cuzco, edited by Gordon F. McEwan, pp. 101–113. University of Iowa Press, Iowa City. Glowacki, Mary, and Gordon F. McEwan 2001 Pikillacta, Huaro y la gran región del Cuzco: Nuevas interpretaciones de la ocupación Wari en la sierra sur. In Huari y Tiwanaku: Modelos vs. evidencias, edited by Peter Kaulicke and William H. Isbell, pp. 31–49. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima.

L ocal S ettlement C ontinuity and Wari I mpact in M iddle H orizon C usco

  93

Hastorf, Christine A. 1990 The Effect of the Inka State on Sausa Agricultural Production and Crop Consumption. American Antiquity 55(2):262–290. Isbell, William H. 1989 Honcopampa: Was It a Huari Administrative Centre? In The Nature of Wari: A Reappraisal of the Middle Horizon Period in Peru, edited by R. M. Czwarno, F. M. Meddens, and A. Morgan, pp. 98–114. BAR International Series 525. British Archaeological Reports, Oxford. 1997 Reconstructing Huari: A Cultural Chronology for the Capital City. In Emergence and Change in Early Urban Societies, edited by Linda Manzanilla, pp. 181–227. Plenum Press, New York. 2000 Repensando el Horizonte Medio: El caso de Conchopata, Ayacucho, Perú. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 9–68. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. 2004 Mortuary Preferences: A Wari Culture Case Study from Middle Horizon Peru. Latin American Antiquity 15(1):3–32. Isbell, William H., and Gordon F. McEwan 1991 A History of Huari Studies and Introduction to Current Interpretations. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 1–17. Dumbarton Oaks, Washington, D.C. Jennings, Justin 2006a Understanding Middle Horizon Peru: Hermeneutic Spirals, Interpretative Traditions, and Wari Administrative Centers. Latin American Antiquity 17(3):265–285. 2006b Core, Peripheries, and Regional Realities in Middle Horizon Peru. Journal of Anthropological Archaeology 25:346–370. Jennings, Justin, and Willy Yépez Álvarez 2001 Architecture, Local Elites, and Imperial Entanglements: The Wari Empire and the Cotahuasi Valley of Peru. Journal of Field Archaeology 28(1/2):143–159. Lumbreras, Luis G. 1969 De los pueblos, las culturas y las artes del antiguo Perú. MoncloaCampodonico, Lima. McEwan, Gordon F. 1984 Investigaciones en la cuenca del Lucre, Cusco. Gaceta Arqueológica Andina 9:12–15. 1987 The Middle Horizon in the Valley of Cuzco, Peru: The Impact of the Wari Occupation of the Lucre Basin. BAR International Series 372. British Archaeological Reports, Oxford. 1989 The Wari Empire in the Southern Peruvian Highlands: A View from the Provinces. In The Nature of Wari: A Reappraisal of the Middle Horizon Period in Peru, edited by R. Michael Czwarno, Frank M. Meddens, and Alexandra Morgan, pp. 53–71. BAR International Series 525. British Archaeological Reports, Oxford. 1991 Investigations at the Pikillacta Site: A Provincial Huari Center in the Valley of Cuzco. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 93–119. Dumbarton Oaks, Washington, D.C. 1996 Archaeological Investigations in Pikillacta, a Wari Site in Peru. Journal of Field Archaeology 23:169–186.

94 

V éronique B élisle and R . A lan C ovey

2005a Conclusion: The Functions of Pikillacta. In Pikillacta: The Wari Empire in Cuzco, edited by Gordon F. McEwan, pp. 147– 164. University of Iowa Press, Iowa City. 2005b Pikillacta Architecture and Construction Requirements. In Pikillacta: The Wari Empire in Cuzco, edited by Gordon F. McEwan, pp. 63–83. University of Iowa Press, Iowa City. McEwan, Gordon (editor) 2005c Pikillacta: The Wari Empire in Cuzco. University of Iowa Press, Iowa City. McEwan, Gordon F., and Nicole Couture 2005 Pikillacta and Its Architectural Typology. In Pikillacta: The Wari Empire in Cuzco, edited by Gordon F. McEwan, pp. 11–27. University of Iowa Press, Iowa City. Mackey, Carol J. 1982 The Middle Horizon As Viewed from the Moche Valley. In Chan Chan: Andean Desert City, edited by Michael E. Moseley and Kent C. Day, pp. 321–331. University of New Mexico Press, Albuquerque. 1983 La cerámica Chimú a fines del Horizonte Medio. Revista del Museo Nacional 47:73–91. Menzel, Dorothy 1964 Style and Time in the Middle Horizon. Ñawpa Pacha 2:1–105. 1968 New Data on the Huari Empire in Middle Horizon Epoch 2A. Ñawpa Pacha 6:47–114. Morris, Craig 1998 Inka Strategies of Incorporation and Governance. In Archaic States, edited by Gary M. Feinman and Joyce Marcus, pp. 293–309. School of American Research Press, Santa Fe, New Mexico. Rowe, John H. 1944 An Introduction to the Archaeology of Cuzco. Papers of the Peabody Museum of American Archaeology and Ethnology Vol. 27, No. 2. Peabody Museum, Harvard University, Cambridge, Massachusetts. Schaedel, Richard P. 1993 Congruence of Horizon with Polity: Huari and the Middle Horizon. In Latin American Horizons, edited by Don S. Rice, pp. 225–261. Dumbarton Oaks, Washington, D.C. Schreiber, Katharina J. 1984 Prehistoric Roads in the Carahuarazo Valley, Peru. In Current Archaeological Projects in the Central Andes: Some Approaches and Results, edited by Ann Kendall, pp. 75–94. BAR International Series 210. British Archaeological Reports, Oxford. 1987 Conquest and Consolidation: A Comparison of Wari and Inka Occupations of a Highland Peruvian Valley. American Antiquity 52(2):266–284. 1992 Wari Imperialism in Middle Horizon Peru. Museum of Anthropology, University of Michigan, Ann Arbor. 2001 The Wari Empire of Middle Horizon Peru: The Epistemological Challenge of Documenting an Empire without Documentary Evidence. In Empires: Perspectives from Archaeology and History, edited by Susan E. Alcock, Terrence N. D’Altroy, Kathleen D. Morrison, and Carla Sinopoli, pp. 70–92. Cambridge University Press, Cambridge.

Topic, Theresa Lange 1991 The Middle Horizon in Northern Peru. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 233–246. Dumbarton Oaks, Washington, D.C. Torres Poblete, Nilo 1989 Sondeo arqueológico en Araway. Unpublished licensing thesis, Social Sciences Faculty, Universidad Nacional San Antonio Abad del Cusco, Cusco, Peru.

Wilson, David J. 1988 Prehispanic Settlement Patterns in the Lower Santa Valley, Peru: A Regional Perspective on the Origins and Development of Complex North Coast Society. Smithsonian Institution Press, Washington, D.C. Zapata Rodríguez, Julinho 1997 Arquitectura y contextos funerarios Wari en Batan Urqu, Cusco. In La Muerte en el antiguo Perú: Contextos y conceptos funerarios, edited by Peter Kaulicke, pp. 165–206. Boletín de Arqueología PUCP No. 1. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima.

L ocal S ettlement C ontinuity and Wari I mpact in M iddle H orizon C usco

  95

Ch a p t er 6

o Nasca and Wari Local Opportunism and Colonial Ties during the Middle Horizon Christina A . Conlee

B

y the mid-twentieth century, when archaeologists had accepted that Wari was an independent and powerful pre-Hispanic society, research on the Nasca culture of the South Coast of Peru led to questions about the role and influence of Wari in coastal regions. Dorothy Menzel’s (1964) study of Middle Horizon ceramics was one of the first to try and define the relationship between the Wari state and the Nasca culture. Menzel recognized early on the expansive nature of Wari and the variability in the extent and type of influence it had in different areas of the central Andes. She also found that religion played a role in Wari expansion and that it was particularly important in Nasca, which had a close relationship both politically and ideologically with Wari. “Nasca thus seems to have enjoyed a special privileged position in the new empire, sharing its prestige in the provinces, perhaps somewhat in the way in which Greece shared in the prestige of the Roman Empire” (Menzel 1964:68). Menzel viewed Nasca as a center of prestige that influenced the developing Wari state, and once the state was established she thought Nasca’s prestige among other groups was enhanced because of its unique connection with Wari. Menzel’s interpretation of the relationship between Wari and Nasca was based primarily on ceramics from

96 

several highland and coastal sites. She documented the influence of the Nasca ceramic style on the contemporary Early Intermediate Period Huarpa style of the Wari heartland, and on subsequent imperial styles of the Middle Horizon. This influence included imagery, use of color, slip-painting, and high-temperature firing (Benavides C. 1971; Cook 1984–1985; Knobloch 1976; Menzel 1964). Conversely, Wari pottery (especially the Chakipampa style) also contributed elements to the local Nasca style. Menzel’s analysis included pottery excavated from the Wari site of Pacheco in the Nazca Valley where many oversized Robles Moqo–style vessels were deliberately broken, interred in adobe chambers, and used as offerings in a similar manner to the offering deposit of Conchopata near the capital city of Huari. Pacheco, with its impressive pottery, is the best known Wari site on the South Coast and has been central to interpretations of Wari expansion and control in the Nazca region. The site was first investigated by Julio C. Tello in 1927 after looters had uncovered a series of adobe rooms (Menzel 1964:24). Tello excavated over three tons of ceramics including many oversized urns. In 1930 Ronald L. Olson made some additional excavations and found an abundance of camelid bones and ceramic sherds of local and Wari nonceremonial types. Unfortunately, around 1953 Pacheco was bulldozed,

much of the site was destroyed, and it is now covered by agricultural fields (Menzel 1964:23). Menzel concluded that the site was an important ritual location, given the large offering deposit found there, and that a colony of people from the highlands lived at the settlement. Thorough scientific excavations at Pacheco were never undertaken, and the subsequent destruction of the site makes further investigation, and more detailed interpretations, difficult. Since Menzel’s initial work, much more has become known about Wari both at the capital and in the provinces. Projects in the highlands and the coast have provided a wealth of information about many aspects of the Wari state. The Nazca region has also been the focus of archaeological investigations, and recent research has focused on societies of all time periods, including the Middle Horizon, and the periods before and after Wari. This diachronic approach has aided in our understanding of local society and the impact of Wari expansion and collapse on people of the region. Menzel’s pioneering work has been crucial in guiding research on the relationship between Wari and Nasca. Recent research in Nazca supports many of Menzel’s earlier ideas; however, new data reveal that the situation was more complicated than she initially proposed, and that in Nazca, as in other areas during the Middle Horizon, there was a diversity of responses to and interactions with the Wari state.

The Nazca Drainage and the Middle Horizon Archaeological evidence of human occupation in the Nazca drainage goes back to the Middle Archaic (ca. 3500 BC), but it was in the Early Intermediate Period (AD 1–750) that the first regionally integrated complex society developed in the area. Known as the Nasca culture, this society developed out of cultural traditions from the Ica and Pisco valleys during the Early Horizon (800 BC–AD 1). The Nasca culture of the Early Intermediate Period is further subdivided into three periods: Early Nasca (AD 1–450), Middle Nasca (AD 450–550), and Late Nasca (AD 550–750). During Early Nasca a new polychrome ceramic tradition developed and was widespread over the South Coast of Peru. The

large ceremonial center of Cahuachi was also established in this period, and there was a proliferation in construction of the Nasca lines (geoglyphs). Most archaeologists agree that a politically complex polity developed at this time, although there continues to be some debate over whether to classify it as a state, chiefdom, series of chiefdoms, or a heterarchy (Reindel and Isla 1999; Schreiber 1999; Schreiber and Lancho 2003; Silverman 2002; Vaughn 2004). Middle Nasca (AD 450–550) was a transitional period during which underground aqueducts were constructed in the southern Nazca drainage, and there was movement of people into the middle parts of the valley (Schreiber and Lancho 2003). This appears to have been a period of drought that may have created stress in the region and led to new innovations (Schreiber and Lancho 2003; Silverman and Proulx 2002; Thompson et al. 1985). The ceramic style underwent a change in Middle Nasca from naturalistic to more abstract designs (Proulx 1968), and construction stopped at Cahuachi (Orefici 1993; Silverman 1993). In Late Nasca (AD 550–750) the population aggregated at large settlements (Reindel and Isla 1998; Schreiber 1999; Schreiber and Lancho 2003; Silverman 2002), and the ceramic style continued to change with warfare and warriors more commonly depicted (Proulx 1983; Silverman and Proulx 2002). Overall, Late Nasca was a time of reorganization and potentially greater political complexity, and it is during this period that the relationship developed between the Nasca and Wari people (Schreiber and Lancho 2003). By the Early Middle Horizon, Wari had established at least two settlements in the drainage, and large transformations occurred in the region. Archaeological surveys conducted in both the northern and southern Nazca drainage have located hundreds of sites of all time periods and have helped to clarify the nature of Middle Horizon settlement and the impact of the Wari state (Isla 2001; Reindel and Isla 1998; Schreiber 2001a; Schreiber and Lancho 2003; Silverman 2002). Notable differences exist between the northern valleys (Santa Cruz, Grande, Palpa, and Ingenio) and the southern valleys (Nazca, Taruga, Las Trancas) in the number and type of human settlements during the Middle Horizon. In the northern drainage there was a

N asca and Wari

  97

dramatic decrease in sites, and the majority of those identified are cemeteries with little evidence for habitation (Browne 1992; Reindel and Isla 1998; Silverman 2002). Sites documented in the north date to early in the Middle Horizon and contain Loro ceramics, the local style with Wari influence dating to Epochs 1 and 2 (Reindel and Isla 1998). Despite the low density of Middle Horizon sites in the northern drainage, new mortuary practices have been identified here, suggesting a new political and social order that was imposed by Wari (Isla 2001). The site of Tres Pallos in the Ingenio Valley, which has been destroyed (much like Pacheco), contains the remains of tombs and Wari ceramics, and it may have been a Wari center (Isla 2001:556). It is in the southern Nazca drainage that the clearest evidence of the Wari presence is found. The offering deposit site of Pacheco is located in the Nazca Valley not far from the large Early Nasca ceremonial center of Cahuachi, and Schreiber (2001a, 2001b), building on Menzel’s interpretation, thinks it was a Wari administrative center. In the upper Nazca Valley, in the Tierras Blancas tributary at 1,350 m, Schreiber has identified, and excavated, a small Wari enclosure called Pataraya associated with agricultural terraces (Schreiber 2001a, 2001b). Throughout the southern drainage Wari ceramics are found in both burial contexts and as surface artifacts at habitation sites (Schreiber 1992). During the Middle Horizon the population de­­ creased in the southern drainage but not as drastically as in the north. Habitation sites, as well as cemeteries, have been documented in each of the southern valleys. The majority of the Middle Horizon domestic sites in the south are small with the exception of Huaca del Loro in the Las Trancas Valley, the approximately 15 ha site where the local Loro ceramic style was defined (Strong 1957). The concentration of local people in the far south is thought to represent a small centralized polity that never came under direct rule by the Wari state (Schreiber 2001a). The establishment of a local center a considerable distance from the Wari-related settlements suggests resistance to the state that may have been initiated by local leaders (Conlee and Schreiber 2006; Schreiber 2001a). The resettlement of people into this area of the valley may not have been entirely out of choice; however, 98 

C hristina A . C onlee

it did create opportunities for local leaders to obtain power in new ways (Conlee 2006). The presence in Nasca of what are interpreted to be intrusive Wari sites and the dramatic shifts in local settlement patterns have led most researchers to consider that Nasca was incorporated into the Wari state; however, the nature of this integration remains debatable. The discovery of an offering deposit at Maymi in the South Coast Pisco Valley (Anders 1990), similar in some ways to the one found at Pacheco, has contributed to the view that the Wari presence on the South Coast was religious in nature. During the Early Intermediate Period, Nazca was the center of a prestigious and powerful ideology that was manifested in the widespread Nasca polychrome pottery, geoglyphs, trophy heads, and ceremonial center of Cahuachi. The adoption of Nasca iconography by Huarpa and Wari potters emphasizes that a shared belief system existed between the two regions. There is no evidence of the kind of intense, direct Wari political rule as there was in some areas of the highlands where large, standardized administrative centers were built and extensive agricultural projects were developed (Schreiber 1992). Pacheco and Pataraya were both small sites, and at Pacheco the nature of Wari presence remains unclear. However, the establishment of these sites and the dramatic changes in local settlements indicate there was more than just a shared religious tradition between the regions. Economic incentives may have played a role in Wari’s interest in Nazca and other areas of the South Coast, where crops that were desired by the state, such as cotton and coca, were grown and whose cultivation involved local people (Conlee and Schreiber 2006; Schreiber 2001b). In general, the archaeological data support Menzel’s view of a close relationship between the Nasca people and the Wari state; however, the connection was not uniform throughout region. Wari appears to have invested in some areas, in particular the Nazca Valley of the southern drainage, and not in others. Likewise, local response to Wari varied from abandonment of settlements in the north, population aggregation in the far south, emulation through adoption of the Wari style at some settlements, and possibly direct incorporation in parts of the Nazca Valley.

Figure 6.1  Map of the Nazca drainage with the sites of La Tiza and Pajonal Alto.

Middle Horizon Burial Practices in the Nazca Drainage Investigations at two sites in the southern drainage, La Tiza and Pajonal Alto, provide important information on the Middle Horizon in the Nazca region and local people’s relationship with the Wari state (Figure 6.1). Both settlements are multicomponent with local Middle Horizon habitation as well as burials. The focus here is on the burials of the Middle Horizon that include both a new type and an older burial tradition. Mortuary practices are a key component of the archaeological record and can provide insight into past populations’ religious beliefs, as well as various aspects of social and political organization.

La Tiza Several years of research and excavation at the site of La Tiza have identified a local Middle Horizon habitation area, new elite mortuary practices, and the continuation of traditional burials dating to this time period. La Tiza is located in the central Nazca Valley, near the modern

town of Nasca, in an area of abundant agricultural land and good access to water. Recently, occupation at this large settlement that spans over 30 ha was discovered to extend back to the Middle Archaic (ca. 3500 BC), although the majority of the site dates from the end of the Early Horizon (100 BC) through the end of the Late Horizon (AD 1532) and includes domestic areas and cemeteries. The Middle Horizon domestic area is approximately 2 ha in size and consists of narrow terraces on which were constructed small stone structures. There are also roughly 5 ha of Middle Horizon mortuary structures, although overall area estimates are difficult because these tombs are mixed with earlier ones, and all are heavily looted. There is a concentration of tombs just east of the Middle Horizon habitation area, and others are found scattered among later and earlier domestic areas. At La Tiza the Middle Horizon habitation is the smallest of any time period (except for the Archaic, which has not yet been fully investigated). It is substantially smaller than that of the Early Intermediate Period, when habitation and cemetery areas covered approximately 12 ha, and the Late Intermediate Period, when habitation alone spanned at least 10 ha. The ceramics from the Middle Horizon domestic contexts and tombs date to Epochs 1 and 2. The local Loro ceramic style predominates; however, the Wari imperial styles of Viñaque and Chakipampa have also been found, usually associated with the tombs. Two calibrated radiocarbon dates from the Middle Horizon habitation area (from two different structures) of AD 653–774 and AD 664–829 (2σ, CALIB5.0, Struvier et al. 1993) corroborate the ceramic data. There is no evidence of occupation (either habitation or burials) after AD 900, and reoccupation of La Tiza did not occur until about AD 1200 when new domestic areas dating to the Late Intermediate Period were established. During the Middle Horizon a new type of elite tomb was used for burial at La Tiza and was constructed partially aboveground and built of stone walls with features such as doorways and possibly niches. In addition, some of these tombs were plastered and painted on the exterior. All of these tombs are looted, but many have partially intact architecture and burials. Looters have left behind a variety of artifacts and in some cases have not disturbed

N asca and Wari

  99

the lowest levels. Several tombs that have now been cleaned and excavated have yielded a surprising amount of material and information. Three separate tomb forms have been identified and recorded, and all are associated with elaborate grave goods (Table 6.1). The first type (Tomb Type 1) consists of small, round structures that measure approximately a meter in diameter with depths between 60 cm and 1 m. The walls consist of large, shaped stones at the base and smaller, flat stones at the top, a type of construction that is also found in Tomb Type 2 (Figure 6.2). Tomb Type 1 is the smallest type and is found dispersed around Middle Horizon domestic terraces. Three tombs of this type have been excavated and recorded (Tombs 1, 2, and 5). The roofs of all the tombs have been destroyed, and their shape and material are unknown. Tomb 1 contained Middle Horizon Loro ceramics, a few folded pieces of copper, a carved stone ornament, a needle with cotton fibers, large quantities of obsidian flakes, and part of an obsidian biface. There is a possible door on the eastern side. The lowest levels of this tomb were undisturbed and exhibit the preparation taken in its construction (Noriega and Conlee 2005). A burned area at the base

may represent a symbolic preparation of the ground, which was covered with a level of fill that contained an abundance of camelid coprolites. On top of the fill was a compact mud surface that was probably the floor of the tomb on which the burials were placed. Tomb 2 is located not far from Tomb 1 and was slightly larger and had more of a D shape. Artifacts associated with this tomb included fragments of textiles and most notably nine animal and human copper figures (Figure 6.3). A burned area was also identified, on top of which the tomb was constructed. Tomb 5 was located on a bedrock ridge to the east of Tombs 1 and 2, surrounded by several other Middle Horizon tombs. It contained several fragments of textile including part of a bag, two tupu pins, a grinding stone, and a few Middle Horizon ceramics mixed in with some Nasca sherds. There was evidence of white plaster on a small area of the exterior. All of these tombs (as well as all of those of the other two types) contained metal objects, and it has been noted that metal artifacts were much more commonly used in Middle Horizon burials than in those of other time periods in the Nazca region (Isla 2001; Menzel 1968). Fragments of human bone were found

Table 6.1  Middle Horizon looted tombs excavated at La Tiza Tomb #

Type

Diam/Size (m)

Depth (m) MNI

Features

Grave Goods

Tomb 1

1

1.00

0.60

Unknown

Possible door

Loro pottery, folded copper, ornament, needle, obsidian flakes, obsidian biface

Tomb 2

1

1.50

1.00

6

Tomb 3

2

2.90

1.25

3

Plaster and paint

Loro pottery, shell bird pendants, copper ornaments, shell beads, textile fragments

Tomb 4

3

2.60 × 2.10

0.90

2

Possible door, plaster and paint

Loro pottery, weaving implements, spindle whorls, miniature bottle, shell beads, copper llama head

Tomb 5

1

0.80

0.90

3

Plaster

Loro pottery, textile fragments, tupu pins, grinding stone

Tomb 6

3

2.70 × 2.40

1.30

6

Plaster and paint

Loro and Viñaque pottery, partial mummy bundle, Spondylus shell, textile fragments, shell beads, worked bone, spindle whorls

100 

C hristina A . C onlee

Textile fragments, copper animal and human figurines

Figure 6.2  Profile view of Tomb 5 showing the construction technique used in Tomb Type 1 and Tomb Type 2.

Figure 6.3  A sample of artifacts found associated with the elite Middle Horizon tombs at La Tiza.

 

associated with all three tombs, and while Tomb 1 did not have enough intact bone to estimate a minimum number of individuals (MNI), Tomb 2 had a MNI of six, which included four adults, one child, and one adolescent (Michele Buzon, personal communication 2009). Tomb 5 had a MNI of three, which included one adult, one child, and one adolescent (Sarah Cross, personal communication 2008). Tomb Type 2 has construction similar to Type 1 with a round shape, large stones at the base, and small, flat stones at the top (see Figure 6.2), but it is larger in diameter (Figure 6.4). Only one of these (Tomb 3) has been excavated, and it was 2.90 m in diameter with a depth of 1.25 m. This tomb was located to the west of the smaller round tombs in an area of Late Intermediate habitation. Associated artifacts included shell bird pendants, copper ornaments, shell beads, textile fragments (including part of a belt or headband), and Middle Horizon ceramics (primarily Loro) mixed with Late Intermediate ceramics from the domestic terraces (see Figure 6.3). Pieces of white mud plaster that were painted red were found on the exterior of the tomb. Judging from the skeletal remains associated with the structure, there were at least three adult individuals interred in the tomb (Sarah Cross, personal communication 2008). The third tomb type is square to rectangular in shape and constructed with double-coursed stone walls (see Figure 6.4). Two tombs of this type were excavated, both of which were located on the ridge east of the Middle Horizon habitation terraces where several aboveground tombs were constructed on bedrock. Both tombs were located in an elevation of the site lower than the round tombs. Tomb 4 (2.60 × 2.10 m) was almost a meter deep and had a doorway on the east side. It contained many grave goods including weaving implements, spindle whorls, a miniature bottle, small shell beads, a copper llama head, and Loro pottery. Remnants of white plaster were found on the exterior along with red and yellow paint. It was estimated that at least two adults were buried inside. Tomb 6 (2.7 × 2.4 m), located nearby, was also square, plastered white, painted red and yellow, and 1.3 m deep. This was the most heavily looted of all the tombs, and there was no intact floor. A partial mummy bundle that consisted of part of a torso wrapped 102 

C hristina A . C onlee

Figure 6.4  Tomb 3 representing Tomb Type 2 (upper) and Tomb 4 representing Tomb Type 3 (lower).

in cotton batting and textiles was found inside the tomb. There was also a complete skull of a female aged twentyfive to thirty-five years old (Conlee et al. 2009). Several pieces of worked Spondylus and many fragments of textiles were found associated with the tomb (see Figure 6.3). In addition, there were polishing stones, spindle whorls, shell beads, and worked bone. There were several types of Middle Horizon 1 and 2 ceramics including Loro and Viñaque. The remains of disturbed skeletal

Figure 6.5  Local Middle Horizon burial from La Tiza and associated Loro bowl.

material in this tomb have an MNI of six, including the adult female, one infant, two children, one adolescent, and one older adult. In addition to the elaborate aboveground tombs constructed in the Middle Horizon, there was also a continuation of local burial practices at La Tiza. A Middle Horizon individual was found buried in a pit inside of a domestic structure in a seated, flexed position facing south with a complete Loro polychrome bowl (Figure 6.5). It is interesting to note that the fill associated with the body contained an abundance of camelid coprolites similar to those found in Tomb 1. The individual was a male between thirty and forty years old. A calibrated

radiocarbon date of AD 661–859 (2σ, CALIB5.0, Struvier et al. 1993) was obtained from a hearth associated with the burial. This type of burial in a pit with the individual placed in a seated and flexed position was the most common type of burial in the region from the beginning of the Nasca culture through the end of the Late Intermediate, a period of almost fifteen hundred years. At La Tiza, burials from the Nasca culture, Middle Horizon, and Late Intermediate have been found in this style, indicating that it is the most traditional and enduring burial practice in the region. During the Middle Horizon at La Tiza, this traditional regional burial style was practiced along with the new elite burials.

N asca and Wari

  103

Pajonal Alto The small village of Pajonal Alto, located in the Taruga Valley, one valley to the south of La Tiza (see Figure 6.1), also contains Middle Horizon habitation and burials. This settlement was first inhabited during a transitional period between Late Nasca and the Early Middle Horizon and was occupied through Middle Horizon Epochs 1 and 2. The village was subsequently abandoned and then reoccupied in the middle of the Late Intermediate Period (ca. AD 1300), and the majority of the habitation dates from this period through the Late Horizon (Conlee 2003). There is evidence for a small domestic area in the Middle Horizon, and two adjacent cemeteries appear to contain some burials dating to this period. Only one unit was excavated at the village that contained Middle Horizon domestic debris, although pottery of this period was found in several areas on the surface and in sections of the eroded river cut on the southern edge of the site (Conlee 2000). Because of the limited Middle Horizon domestic contexts that were excavated, no general conclusions could be made

Figure 6.6  Middle Horizon child burial from Pajonal Alto.

104 

C hristina A . C onlee

except that the settlement was much smaller during this period than it was in the later Late Intermediate and Late Horizon. One Middle Horizon burial, a child aged three to five years old, was found underneath a midden that dated to the Late Intermediate and Late Horizon. The child was placed in a seated position facing west and was buried with large fragments of a ceramic vessel (Figure 6.6). This was a typical local burial practice for children beginning in Early Nasca and is known to have continued through the Late Intermediate Period. There were several indications that the child was not healthy at the time of death. Cribra orbitalia, a porosity of the bone that is usually a result of infection, was found in the eye orbits, and one of the lesions was active at the time of death. Periostitis was present on the long bones, and almost all of these lesions were active at the time of death. The distribution of these reactions probably indicates the effects of a nonspecific generalized disease and not injury or localized infection (Conlee 2000; Kellner and Conlee 1998). There were also carious lesions and an abscess in the deciduous teeth of

the child. All of these indicators suggest that the child was under a lot of physiological stress.

Regional Transformations in Burial Practices The burials found at La Tiza and Pajonal Alto reflect changes on a broader, regional level that coincided with the Wari presence in the Middle Horizon. Elsewhere in the Nazca drainage, studies by Carmichael (1988, 1995) and Isla (2001) have documented transformations in mortuary traditions in the Middle Horizon. During this period traditional burial practices were continued by some segments of the population, and new forms were used by others. Carmichael (1988, 1995) found that burial form and offering type were consistent throughout the Nasca culture of the Early Intermediate Period, indicating that burial patterns were conservative. Burials in pits, most commonly with the individual in a flexed position and accompanied by at least one whole pot, were the most widespread burial type of the Nasca culture (Carmichael 1988), and this tradition continues in the Middle Horizon as seen in the adult male burial at La Tiza. Child burials in large ceramic vessels (or with large fragments of these) were common in the Nasca culture for children six years of age and younger, and this practice continued in the Middle Horizon but was less frequent (Isla and Reindel 2006). The Middle Horizon child burial at Pajonal Alto is an example of this tradition. The most elaborate type of Nasca culture burials was in barbacoa tombs that were oval or square chambers with log roofs. These had varying depths, with the deepest approximately 8 m reported at the site of La Muña in the northern drainage (Isla and Reindel 2006:385). These tombs were for the highest status individuals of the Nasca culture, and all those recorded contained one individual (Isla and Reindel 2006). This tomb type continued to be used into the Middle Horizon, but the roofs were more substantial, and the tombs were not filled in, probably in order to more easily reenter and inter additional individuals (Isla 2001). In addition, archaeologists have identified a new tomb type that consisted of square or rectangular tombs with large stone-slab roofs that have been found at the Wari site of Pataraya in the upper Nazca Valley and also in

the upper elevations of Las Trancas Valley (Isla 2001; Schreiber 2001b). The use of plaster and paint on the exterior of the tombs at La Tiza appears to be a new practice during the Middle Horizon. Isla (2001:565) has noted that some tombs found at the Middle Horizon sites of Atarco and Huaca del Loro in the southern Nazca drainage had white or yellow paint on the walls. Plaster and paint were commonly used on buildings at the Early Nasca ceremonial center of Cahuachi (Orefici 1993), although no specific mention of this treatment on Early Intermediate Period tombs has been made. There is evidence of white plaster being used on the exteriors of compounds at the capital of Huari, and some of the ceremonial buildings at the city were painted red (Isbell and Vranich 2004). Other architectural traditions of the Middle Horizon on the coast also used white plaster and red and yellow paint (Nelson et al., this volume). The most dramatic changes in Middle Horizon burial practices in the Nazca drainage were in the treatment of the body and the number of persons interred together. Both Carmichael (1988, 1995) and Isla (2001; Isla and Reindel 2006) found that during the Early Intermediate most tombs contained one individual. Isla (2001:578) calculates that 99 percent of the recorded burials of this period were of a single individual. In the Middle Horizon this changed, and of the tombs that Isla has documented, only 65 percent of tombs contained one individual, and the rest had two or more people. We do not know if the multiple burials found in the region, including those from La Tiza, were individuals buried together at the same time or sequentially interred over many years. Even with un-looted contexts and well-controlled dating this can be difficult to determine. Another important change in the Middle Horizon is the first evidence of mummification in Nazca, and mummy bundles become common. Isla (2001) has observed evidence of mummy bundles in many cemeteries that are associated with the local Middle Horizon Loro culture and in particular dating to Epoch 2. This is a trend that extends to other coastal valleys where the large and elaborate mummy bundles of the falsa cabeza, or false head, type are found. Elaborate mummy bundles were used for the first time on the Central Coast during Middle Horizon 1B and

N asca and Wari

  105

are found at sites including Pachacamac, Nievería, and Ancón, with most dating to Epochs 2 and 3 (Angeles and Pozzi-Escot 2001; Gayton 1927; Kaulicke 1997; Segura and Shimada, this volume; Uhle 1903). Tung (2007) has also reported Middle Horizon mummy bundles at the site of Beringa in the Majes Valley on the far South Coast. These changing burial practices are reflected at La Tiza where evidence of multiple burials and mummy bundles is found for the first time associated with the new elite Middle Horizon mortuary structures. Multiple burials were also a common practice in the Wari heartland during the Middle Horizon and replaced the individual graves and cemeteries of the Early Intermediate Period (Isbell 2001, 2004; Isbell and Cook 2002; Ochatoma and Cabrera 2001). There is evidence that some tombs were reopened and additional bodies interred, and sometimes bones were removed and perhaps grave goods were changed (Isbell 2004; Isbell and Cook 2002). The capital city of Huari had evidence of multiple burials in the Moraduchayuq compound (Isbell et al. 1991) and in the royal subterranean slab tombs in the Monjachayuq Sector (Isbell 2004). The elaborate Wari tombs with multiple burials and evidence of reentry are thought to be indicators of ancestor veneration or worship, which was focused on the deceased elites (Isbell and Cook 2002:287–288). Other Wari sites in the central highlands contained multiple burials, including Azángaro (Anders 1986), Batan Urqu at the Huaro complex (Zapata 1997), and Jincamoco (Schreiber 1992). Multiple burials in tombs associated with Wari ceramics are also reported outside of the central highlands during the Middle Horizon. In the Cotahuasi Valley in Arequipa large tombs containing multiple individuals were built at this time, and there is evidence both that they were repeatedly reentered with objects and that individuals were moved into and out of these structures (Jennings, this volume). The tomb shape and construction technique at La Tiza have some similarities to those found at Conchopata where one Wari tomb type was constructed by excavating into bedrock using cracks in the rock, and these tombs always contained multiple burials (Isbell 2004; Isbell and Cook 2002). At La Tiza one concentration of Middle Horizon tombs is along a ridge where they were 106 

C hristina A . C onlee

constructed on the bedrock. Burial goods in the Wari heartland contained similar funerary items to those found in the tombs at La Tiza and elaborate tombs of the Middle Horizon documented elsewhere in the Nazca drainage. These items include Spondylus, copper tupu pins, and shell beads (Isbell and Cook 2002; Isla 2001). During the Middle Horizon, in many areas that were interacting with the Wari state, there is a dramatic change in mortuary practices most often associated with the introduction of multiple burials. Isbell (this volume) has cautioned that the mortuary variation in the Middle Horizon is too great to attribute it to centralized Wari political control. However, there appears to have been some type of new and widespread ideology that expressed itself in the treatment of the dead. The mortuary variability documented in the Middle Horizon may itself be the commonality in many places during this time. In the Nazca region burial practices were conservative throughout the approximately seven hundred years of the Nasca culture, and new forms were not adopted until the Middle Horizon associated with Wari influence.

Middle Horizon Immigrants and Population Movements Given the presence of Wari sites in the Nazca region, and the documented changes in settlement patterns and burial practices, we would expect that there was an influx of foreigners from the highlands into the Nazca region during the Middle Horizon. In addition to using material correlates such as architecture, ceramic style, and burial practices to detect immigrant populations, isotopic analysis of human remains is another way to identify local versus foreign individuals. Strontium isotope analysis was conducted on burials from La Tiza and Pajonal Alto and has provided important information about the origins of populations and migrations in the Nazca region (Conlee et al. 2009). The burials that were analyzed from La Tiza date to the Nasca culture, the Middle Horizon, and the Late Intermediate Period, and the one burial from Pajonal Alto dates to the Middle Horizon. Although strontium isotope analysis is a relatively new technique, research has already illustrated the feasibility and the potential of this type of analysis to elucidate

residential mobility in the archaeological record (e.g., Ambrose and Krigbaum 2003; Bentley 2006; Burton et al. 2003; Knudson and Price 2007; Price et al. 2008). Strontium is found in rock, groundwater, and soil, and the concentrations vary according to local geology so that regions are distinct (Faure 1986). Because the strontium present in soil and groundwater is incorporated into the plants and animals of a region, the strontium composition of an individual’s diet will be reflected in his or her skeletal material and will mirror the isotopic composition of the geological region in which a person lived before death. Tooth enamel forms during early childhood and will reflect the isotopic composition of the area where a person lived during this time. Through analysis of strontium in teeth and skeletal material it is possible to determine where a person lived as a child and as an adult (Bentley 2006). This technique allows for the identification of first-generation immigrants from geologically distinct areas. The local strontium signature of the region around the sites of La Tiza and Pajonal Alto was defined by testing soil, archaeological faunal remains (mice and guinea pigs), and modern faunal remains (mice and guinea pigs) (Conlee et al. 2009). Using the standard method of faunal mean ± 2 standard deviations, the local range based on these samples is 0.70559–0.70727. Fieldwork at La Tiza between 2004 and 2006 resulted in the excavation of eight intact burials including the Middle Horizon individual buried in the local style. The skeletal material and teeth of each of these burials were analyzed to determine their strontium ratios and compare them with the local range. Strontium analysis was also conducted on the teeth of the female individual found in looted Tomb 6 dating to the Middle Horizon and on the Middle Horizon child burial from Pajonal Alto (Table 6.2). The results of the study have documented that all seven individuals who date to the pre-Wari Nasca culture and the post-Wari Late Intermediate Period at La Tiza fall into the local strontium range (Conlee et al. 2009). The Middle Horizon individual at La Tiza who was buried in the local tradition also had a local signature. However, the other two Middle Horizon individuals, the child from Pajonal Alto and the woman from the looted elite tomb at La Tiza, fall outside of the local range,

indicating that both were born outside of the local area. Because the cranium from the looted tomb could not be directly associated with any particular skeletal material, it was not possible to test if the woman had spent much of her adult life in the local region or if she had very recently moved to Nazca. At present, the precise region where both individuals were born is unknown. They fall closest in range to the local signature at Chokepukio in the Cuzco region, but at this time no definite conclusion about their place of birth can be made (Andrushko et al. 2009; Conlee et al. 2009). The presence of a nonlocal adult female buried in the new elite tomb type suggests that this mortuary practice may have been brought in by people who were not originally from the Nazca region. This woman may have married into the community, or she may have been a member of a larger group that settled in the area. This is similar to a discovery at the Central Coast site of Ancón where a young woman from an elite burial was wrapped in elaborate textiles, buried with Chakipampa ceramics, and had a strontium signature similar to that of the Wari heartland (Slovak et al. 2009). In contrast, the child at Pajonal Alto was buried in a local style despite the fact that the strontium signature indicates he or she was not originally from the village. In addition, the evidence of infection and possible malnutrition suggests the child was under stress, possibly as a result of immigration to the Nazca region. Previous investigations on Middle Horizon settlements in the Nazca region had indicated this was a time of great change and population movement. This appears to have been in part initiated by local people, as seen in the abandonment of settlements in the northern drainage and a congregation of people in the far south at the site of Huaca del Loro. The new Wari sites in the middle and upper Nazca Valley along with the new mortuary practices found throughout the drainage also indicate a foreign intrusion into the region during this period. This foreign influence was not just the result of trade in ideas, commodities, or religious practices but was also the result of foreign people in the region. There was likely settlement by some Wari people, who may have held administrative or economic leadership positions, and they lived at the Wari sites. There was probably an alliance of

N asca and Wari

  107

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

La Tiza

Pajonal Alto

Pajonal Alto

1

2

2

3

3

4

4

5

6

6

7

7

8

8

9

10

10

Bone

Enamel (M2)

Enamel (M1)†

Bone

Enamel (C)

Enamel (C)

Bone

Enamel (P4)

Bone

Bone*

Enamel (I1)

Bone

Enamel (P4)

Bone

Enamel (P4)

Bone

Bone*

Material

Middle Horizon

Middle Horizon

Middle Horizon

Early Nasca

Early Nasca

Early Nasca

Early Nasca

Late Intermediate

Late Intermediate

Late Intermediate

Late Intermediate

Late Intermediate

Late Intermediate

Late Intermediate

Middle Horizon

Middle Horizon

Middle Nasca

Cultural Affiliation

3–5

25–35

9–10

45+

18 months

14–15

12–18 months

25–35

30–40

22–29

Age

Source: Conlee et al. 2009. *No tooth samples were available because the burial was headless. † No bone sample because the cranium could not be associated with particular skeletal material.

Site

Burial

Table 6.2  Burial and strontium isotopic information from La Tiza and Pajonal Alto

Juvenile

Female

Juvenile

Female

Juvenile

Probable male

Juvenile

Possible female

Male

Male

Sex

Pit

Looted tomb

Wall

Pit in domestic structure

Inside of large olla

Pit in domestic structure

Pit above Burial 6

Pit in domestic structure

Pit in domestic structure

Round stone-lined tomb

Burial Location

Sr/86Sr

0.70733

0.70770

0.70747

0.70691

0.70677

0.70655

0.70635

0.70643

0.70655

0.70643

0.70682

0.70686

0.70612

0.70662

0.70640

0.70667

0.70690

87

these Wari people with local elites who benefited from the relationships through acquisition of material goods such as fine Wari pottery, Spondylus, and metal artifacts. The burial data and strontium analysis also suggest that not all of the foreigners were elites. Instead, the influx of people from the highlands may have also included support personnel or people looking for better economic opportunities who were not under state supervision.

A Reassessment of Nasca and Wari Relationships The changes in mortuary practices in the Middle Horizon as documented in this and other studies provide important insight into the impact the Wari state had on the region. The relationship between the Nasca and Wari people was a complicated one. As Menzel noted in her analysis, the pottery styles of the two regions reveal a closeness that may represent shared religious beliefs, economic ties, and political integration. The Nasca people likely obtained prestige in the Middle Horizon through their connection to Wari as Menzel proposed; however, the evidence suggests only certain Nasca people benefited from this relationship. There appear to have been close ties between some segments of the Nasca population who lived in the Nazca Valley of the southern drainage and the Wari state. Here at La Tiza new mortuary practices reflect the development of new statuses during this period. This new mortuary tradition was practiced alongside older local burial styles, and only certain people were being buried in the new style. The presence in these new tombs of more elaborate grave goods suggests the people buried in them were of high status and had a close connection with the Wari state. Wari appears to have created new opportunities for local people, and new types of elites emerged. At least some of the people buried in these tombs were recent immigrants to the region as suggested by the strontium signature of the woman buried in one of the tombs. Additional isotopic analysis needs to be conducted in order to make more concrete conclusions; however, the preliminary data indicate that there was not just an influx of foreign ideas and artifacts during the Middle Horizon, but also an influx of foreign people.

Not everyone in Nazca had ties with the Wari people or benefited from their presence. Previous research in the Nazca region documented dramatic changes in settlement patterns in the Middle Horizon, including a decrease in sites, especially in the northern drainage, and an aggregation of people in the southern drainage to the far southern Las Trancas Valley away from the Wari sites in the Nazca Valley. The research at La Tiza and Pajonal Alto corroborates this pattern with smaller Middle Horizon habitation areas than those found in the previous or subsequent periods. Another important finding from the excavations at La Tiza and Pajonal Alto is that both sites appear to have been abandoned after Middle Horizon 2, and there is no evidence of occupation at either site after AD 900 until reoccupation between AD 1200 and 1300. This is a pattern that is reflected in the region as a whole where most of the evidence for Wari influence comes from Middle Horizon 1 and 2. There is little evidence for a later Wari presence as has been documented in other areas, and there appears to have been an abandonment of most sites after Epoch 2. Interactions between Wari and Nasca began in the Early Intermediate Period during the seventh century AD, and this relationship that began early also ended early. This suggests a major disruption that occurred in late ninth or early tenth century AD and which impacted all people of the Nazca drainage whether they had close ties with Wari or not. In the subsequent Late Intermediate Period the elite burial practices that were new in the Middle Horizon were abandoned, as were the ceramic and artifact styles associated with Wari. Research on the transformations that occurred in the Middle Horizon, and the relationship between the Wari state and people in diverse areas of the central Andes, has developed greatly in the last fifty years. Menzel’s pioneering work helped to establish the ceramic chronology for the period and provided a framework to study interactions between Wari and other polities. After several decades of research in Nazca and the Wari heartland, we see that the relationship between the two societies appears to have been more complex than initially proposed and mirrors the contemporary view of Wari in various areas of the Andes. In Nazca, as elsewhere, the

N asca and Wari

  109

complicated interactions between local people and the Wari state helped to create the dynamic and innovative culture of the Middle Horizon.

Acknowledgments Funding for the research at La Tiza was supported by the National Science Foundation (BCS-0314273), the John Heinz III Fund Grant Program for Latin American Archaeology, and the Research Enhancement Program at Texas State University. Research at Pajonal Alto was provided by a National Science Foundation Dissertation Improvement Grant and a Humanities and Social Science Grant from the University of California, Santa Barbara. Thanks go to Aldo Noriega who is the codirector of the La Tiza project and who helped to excavate and analyze the tombs discussed in this paper. Michele Buzon, Sarah Cross, and Corina Keller provided bioarchaeological analysis of the burials used in this study.

References Cited Ambrose, Stanley H., and John Krigbaum 2003 Bone Chemistry and Bioarchaeology. Journal of Anthropological Archaeology 22:193–199. Anders, Martha B. 1986 Dual Organization and Calendars Inferred from the Planned Site of Azángaro: Wari Administrative Strategies. Unpublished Ph.D. dissertation, Cornell University, Ithaca, New York. 1990 Maymi: Un sitio del Horizonte Medio en el valle de Pisco. Gaceta Arqueológica Andina 5(17):27–39. Andrushko, Valerie A., Michele R. Buzon, Antonio Simonetti, and Robert A. Creaser 2009 Strontium Isotope Evidence for Prehistoric Migration at Chokepukio, Valley of Cuzco, Peru. Latin American Antiquity 20(1): 57–75. Angeles, Rommel, and Denise Pozzi-Escot 2001 Textiles del Horizonte Medio. Las evidencias de Huaca Malena, Valle de Asia. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 401–424. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Benavides C., Mario 1971 Análisis de la cerámica Huarpa. Revista del Museo Nacional (Lima) 37:63–88. Bentley, R. Alexander 2006 Strontium Isotopes from the Earth to the Archaeological Skeleton: A Review. Journal of Archaeological Method and Theory 13:135–187. Browne, David M. 1992 Further Archaeological Reconnaissance in the Province

110 

C hristina A . C onlee

of Palpa, Department of Ica, Peru. In Ancient America: Contributions to New World Archaeology, edited by Nicholas J. Saunders, pp. 17–116. Oxbow Books, Oxford. Burton, James H., T. Douglas Price, Laura Cahue, and Lori E. Wright 2003 The Use of Barium and Strontium in Human Skeletal Tissues to Determine Their Geographic Origin. International Journal of Osteoarchaeology 13:88–95. Buzon, Michele R., Antonio Simonetti, and Robert A. Creaser 2007 Migration in the Nile Valley during the New Kingdom Period: A Preliminary Strontium Isotope Study. Journal of Archaeological Science 9:1391–1401. Carmichael, Patrick H. 1988 Nasca Mortuary Customs: Death and Ancient Society on the South Coast of Peru. Unpublished Ph.D. dissertation, Department of Anthropology, University of Calgary. 1995 Nasca Burial Patterns: Social Structure and Mortuary Ideology. In Tombs for the Living : Andean Mortuary Practices, edited by Tom Dillehay, pp. 161–188. Dumbarton Oaks, Washington, D.C. Conlee, Christina A. 2000 Late Prehispanic Occupation of Pajonal Alto, Nasca, Peru: Implications for Imperial Collapse and Societal Reformation. Unpublished Ph.D. dissertation, University of California, Santa Barbara. 2003 Local Elites and the Reformation of Late Intermediate Period Sociopolitical and Economic Organization in Nasca, Peru. Latin American Antiquity 14(1):47–65. 2006 Regeneration as Transformation: Post-Collapse Society in Nasca, Peru. In After Collapse: The Regeneration of Complex Societies, edited by Glenn M. Schwartz and John J. Nichols, pp. 99–113. University of Arizona Press, Tucson. Conlee, Christina A., Michele R. Buzon, Aldo Noriega Gutierrez, Antonio Simonetti, and Robert A. Creaser 2009 Identifying Foreigners versus Locals in the Burial Population of Nasca, Peru: An Investigation Using Strontium Isotope Analysis. Journal of Archaeological Science 36:2755–2764. Conlee, Christina A., and Katharina Schreiber 2006 The Role of Intermediate Elites in the Balkanization and Reformation of Post-Wari Society in Nasca, Peru. In Intermediate Elites in Pre-Columbian States and Empires, edited by Christina Elson and Alan Covey, pp. 94–111. University of Arizona Press, Tucson. Cook, Anita 1984–1985 The Middle Horizon Ceramic Offerings from Conchopata. Ñawpa Pacha 22–23:91–126. Faure, Gunter 1986 Principles of Isotope Geology. John Wiley, New York. Gayton, Anna H. 1927 The Uhle Collections from Nievería. University of California Publications in American Archaeology and Ethnology Vol. 21, No. 8. University of California Press, Berkeley. Isbell, William H. 2001 Repensando el Horizonte Medio: El caso de Conchopata, Ayacucho, Perú. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 9–68. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima.

2004 Mortuary Preferences: A Wari Culture Case Study from Middle Horizon Peru. Latin American Antiquity 15(1):3–32. Isbell, William H., Christine Brewster-Wray, and Lynda Spickard 1991 Architecture and Spatial Organization at Huari. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 19–53. Dumbarton Oaks, Washington, D.C. Isbell, William H., and Anita G. Cook 2002 A New Perspective on Conchopata and the Andean Middle Horizon. In Andean Archaeology II: Art, Landscape and Society, edited by Helaine Silverman and William H. Isbell, pp. 249– 305. Kluwer Academic/Plenum, New York. Isbell, William H., and Alexei Vranich 2004 Experiencing the Cities of Wari and Tiwanaku. In Andean Archaeology, edited by Helaine Silverman, pp. 167–181. Blackwell, Malden, Massachusetts. Isla, Johny 2001 Wari en Palpa y Nasca: Perspectivas desde el punto de vista funerario. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 555–584. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Isla, Johny, and Markus Reindel 2006 Burial Patterns and Sociopolitical Organization in Nasca 5 Society. In Andean Archaeology III: North and South, edited by William H. Isbell and Helaine Silverman, pp. 274–400. Springer, New York. Kaulicke, Peter 1997 Contextos funerarios de Ancón. Esbozo de una síntesis analítica. Pontificia Universidad Católica del Perú, Lima. Kellner, Corina, and Christina A. Conlee 1998 Evidence of Physiological Stress in a Child Burial at Pajonal Alto, a Late Prehistoric Site on the South Coast of Peru. Paper presented at the 25th Annual Meeting of the Paleopathology Association, Salt Lake City. Knobloch, Patricia J. 1976 A Study of the Huarpa Ceramic Style of the Andean Early Intermediate Period. Unpublished Master’s thesis, Department of Anthropology, State University of New York, Binghamton. Knudson, Kelly J., and T. Douglas Price 2007 Utility of Multiple Chemical Techniques in Archaeological Residence Mobility Studies: Case Studies from Tiwanaku- and Chirabaya-Affiliated Sites in the Andes. American Journal of Physical Anthropology 132:25–39. Menzel, Dorothy 1964 Style and Time in the Middle Horizon. Ñawpa Pacha 2:66–105. 1968 La cultural Huari. Compañía de Seguors y Reaseguros PeruanoSuiza, Lima. Noriega, Aldo, and Christina A. Conlee 2005 Informe del Proyecto La Tiza 2005. Unpublished report submitted to the Instituto Nacional de Cultural, Lima. Ochatoma, José, and Martha Cabrera 2001 Arquitectura y áreas de actividad en Conchopata. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 449–488. Boletín

de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Orefici, Giuseppe 1993 Nasca: Arte e societa del popolo dei geoglifi. Jaca, Milan. Price, T. Douglas, James H. Burton, Paul D. Fullagar, Lori E. Wright, Jane E. Buikstra, and Vera Tiesler 2008 Strontium Isotopes and the Study of Human Mobility in Ancient Mesoamerica. Latin American Antiquity 19(2):167– 180. Proulx, Donald A. 1968 Local Differences and Time Differences in Nasca Pottery. University of California Publications in Anthropology 5. University of California Press, Berkeley. 1983 The Nasca Style. In Art of the Andes: Pre-Columbian Sculptured and Painted Ceramics from the Arthur M. Sackler Collections, edited by L. Katz, pp. 87–105. Arthur M. Sackler Foundation and the AMS Foundation for the Arts, Sciences and Humanities, Washington, D.C. Reindel, Marcus, and Johny A. Isla Cuadrado 1998 Proyecto Arqueológico PALPA. Informe Final. Instituto Nacional de Cultura, Lima. 1999 Das Palpa-Tal: Ein archiv der vorgeschichte Peru. In Nasca: Geheimnisvolle zeichen im alten Peru, edited by Judith Rickenbach, pp. 177–98. Museum Rietberg Zurich, Switzerland. Schreiber, Katharina J. 1992 Wari Imperialism in Middle Horizon Peru. Anthropological Papers No. 87. Museum of Anthropology, University of Michigan, Ann Arbor. 1999 Regional Approaches to the Study of Prehistoric Empires: Examples from Ayacucho and Nasca, Peru. In Settlement Pattern Studies in the Americas: Fifty Years since Virú, edited by Brian R. Billman and Gary M. Feinman, pp. 160–171. Smithsonian Institution Press, Washington, D.C. 2001a The Wari Empire of Middle Horizon Peru: The Epistemological Challenge of Documenting an Empire without Documentary Evidence. In Empires, edited by Susan E. Alcock, Terrence N. D’Altroy, Kathleen D. Morrison, and Carla Sinopoli, pp. 70–92. Cambridge University Press, Cambridge. 2001b Los Wari en su contexto local: Nasca y Sondondo. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 425–447. Boletín de Arqueología PUCP No. 5. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Schreiber, Katharina J., and Josué Lancho Rojas 2003 Irrigation and Society in the Peruvian Desert: The Puquios of Nasca. Lexington Books, Lanham, Maryland. Silverman, Helaine 1993 Cahuachi in the Ancient Nasca World. University of Iowa Press, Iowa City. 2002 Ancient Nasca Settlement and Society. University of Iowa Press, Iowa City. Silverman, Helaine, and Donald Proulx 2002 The Nasca. Blackwell, Malden, Massachusetts. Slovak, Nicole M., Adina Paytan, and Bettina A. Wiegand 2009 Reconstructing Middle Horizon Mobility Patterns on the Coast of Peru through Strontium Isotope Analysis. Journal of Archaeological Science 36:157–165.

N asca and Wari

  111

Strong, William Duncan 1957 Paracas, Nazca, and Tiahuanacoid Cultural Relationships in South Coastal Peru. Memoirs of the Society for American Archaeology 13. Society for American Archaeology, Salt Lake City. Stuiver, M., P. J. Reimer, and R. W. Reimer 1993 Extended 14C Database and Revised CALIB Radiocarbon Calibration Program. Radiocarbon 35:215–230. Thompson, Lonnie G., E. Mosley-Thompson, J. F. Bolzan, and B. R. Koci 1985 A 1500-Year Record of Tropical Precipitation in Ice Cores from the Quelccaya Ice Cap, Peru. Science 229:971–973. Tung, Tiffiny A. 2007 The Village of Beringa at the Periphery of the Wari Empire: A Site Overview and New Radiocarbon Dates. Andean Past 8:253–286.

112 

C hristina A . C onlee

Uhle, Max 1903 Pachacamac: Report of the William Pepper, M.D., LL.D, Peruvian Expedition of 1896. Department of Archaeology, University of Pennsylvania, Philadelphia. Vaughn, Kevin 2004 Households, Crafts, and Feasting in the Ancient Andes: The Village Context of Early Nasca Craft Consumption. Latin American Antiquity 15(1):61–88. Zapata, Julinho 1997 Arquitectura y contextos funerarios Wari en Batan Urqu, Cusco. In La muerta en el antiguo Perú, edited by Peter Kaulicke, pp. 165–206. Boletin Arqueologica PUCP Vol. 1. Pontifica Universidad Católica del Peru, Lima.

Ch a p t er 7

o The Wari Footprint on the Central Coast A View from Cajamarquilla and Pachacamac R afael Segura Llanos and Izumi Shimada

M

uch of the debate over the last four decades on the nature of and the interaction between the societies of the Central Coast and Wari during the Middle Horizon has been polarized by two positions that could be characterized as the Ayacucho-based hegemonic or imperialist view and the locally based mercantile model. The former sees the Central Coast directly or indirectly dominated by the Wari Empire (e.g., Agurto 1984; Isla and Guerrero 1987; Lumbreras 1969; Menzel 1964, 1968, 1977; Vásquez 1984), while the latter argues that the central coastal polities were commercial states strong enough to interact with their Wari peers without being politically dominated or assimilated (e.g., Shady 1982, 1988, 1989; Torero 1970). In a review of the interpretative models for understanding the Wari expansion, Isbell and McEwan (1991:5) suggested that discrepancies among them stem largely out of different theoretical commitments, geographical areas of research, and scientific ethnocentrisms of the archaeologists concerned with the topic. In his comment on an article by Shady (1988), Isbell (1988a:106) argued that the Middle Horizon cannot be correctly explained on the basis of either or both of the two aforementioned models and that a range of intermediate possibilities must be considered.

While we agree with this assessment, we consider that the lack of new and varied archaeological data in the Central Coast is another key factor that has prevented us from elaborating new alternative models. For example, archaeologists concerned with the local Middle Horizon continue to base many of their hypotheses on the same tombs and ceramics excavated by Max Uhle (1903, 1913, 1998 [1910]) at Ancón, Pachacamac, and sites in the Chancay Valley and around the city of Lima over or nearly a century ago (e.g., Kaulicke 2000). Studies on regional demography, trade networks, craft technology, land use and settlement patterns, and environmental phenomena and their societal impacts remain rare or unrealized (Segura 2007). Only a few bioanthropological studies have been recently completed (e.g., Slovak 2007). While in many cases pertinent evidence has not been found, it is also true that it has not been systematically or appropriately sought. An alternative to this monolithic or mutually exclusive unimodal thinking is a more nuanced, situationally adaptive, multimodal, and multiphase model of interaction that considers a negotiated coexistence or co-option depending on the scale and strength of the local polity involved, missionary activities, trading, or conquest (e.g., Jennings 2006; Shimada 1994). Yet such a model needs

  113

new data obtained either opportunistically or through theoretically guided, problem-oriented research. This article presents an example of how data collected under both modalities can be used for formulating a new explanatory hypothesis of the cultural dynamics of the Central Coast during the Middle Horizon.

Research Setting and Data Sources Our assumptions and beliefs about the Middle Horizon on the Central Coast, particularly in reference to its core area of the Chillón, Rímac, and Lurín valleys, can be reassessed from the vantage point of two of its most important sites, Cajamarquilla and Pachacamac (Figure 7.1). We believe that such an approach is advantageous for a number of reasons. First, both Cajamarquilla (ca. 160 ha) and Pachacamac (ca. 500 ha) exemplify very well the complexity of pristine “urban society” in the Andes, a topic commonly raised in discussing the Middle Horizon cultural innovations (e.g., Isbell 1988b; Lumbreras 1989; Makowski 2000; Rowe 1963; Schaedel 1966; Von Hagen and Morris 1998; Williams 1980). Second, both centers were occupied roughly contemporaneously from the Early Intermediate Period to the Late Intermediate Period, which suggests that they experienced similar regional and interregional cultural and environmental events and processes. Third, they occupy complementary ecological and economic areas; Cajamarquilla is located at a key traffic node and intensive cultivation area in the middle Rímac Valley with easy access to the highlands through the Jicamarca quebrada (dried side valley); Pachacamac lies in the lower Lurín Valley where three major bodies of water converge: the Lurín River, the Pacific Ocean, and the Urpi Kocha Lagoon. Fourth, they are the only wellpreserved, major civic-ceremonial centers of the LurínRímac hydrological-cultural system as suggested by the distribution of shared ceramic, textile, and architectural features (Schaedel 1951). Unfortunately, the third major center of this system, Maranga in the lower Rímac Valley, no longer preserves much of its original layout. Finally, the cemeteries associated with both Cajamarquilla and Pachacamac were the type sites for two important Wariderived ceramic styles in the Central Coast: Nievería and Pachacamac, respectively (Menzel 1964, 1968, 1977). 114 

R afael S egura L lanos and I zumi S himada

Figure 7.1  Map of the core area of the Central Coast showing key archaeological sites discussed in the chapter.

The primary data discussed in this article come from two projects independently carried out in Cajamarquilla and Pachacamac between 2000 and 2005. At Cajamarquilla, Segura had the opportunity to carry out excavations as the temporary director of the Cajamarquilla Archaeological Project in 2000–2001. Fieldwork by Segura aimed to establish the occupational history of the Pedro Villar Córdova Compound, an area of nearly 13 ha located in the northwestern end of Cajamarquilla. In Pachacamac, Shimada has concluded the first phase (2003–2005) of his long-term Pachacamac Archaeological Project, with the purpose of elucidating the social and environmental foundations of the famed site. As will be seen, our projects provide evidence for questioning the aforementioned polarized interpretative models and hence challenging our understanding of the Middle Horizon on the Central Coast. Beyond a systematic

assessment of ceramic, architectural, and funerary data from the period prior to as well as during and after the local-Wari interaction, we offer new environmental data over the same time span that suggest this interaction cannot be properly understood without an appreciation of social responses to dynamic and adverse environmental processes.

Cajamarquilla Cajamarquilla, located on the middle Rímac Valley 21 km from the center of Lima, lies on the north bank alluvial terrace near the mouth of a quebrada called Jicamarca or Huaycoloro. Occasionally, when El Niño occurs, the quebrada carries muddy torrents from the western Andean slopes to the Rímac River, flooding and causing serious damage in and around the site (Mogrovejo and Makowski 1999; Segura 2001). At the same time, much of the south side of the site is bordered by the pre-Hispanic Nievería canal, which is still in use and whose current intake is located 4.5 km upstream on the north bank of the Rímac River. Associated Late Lima ceramics suggest that Cajamarquilla was built by AD 500–600, and since then intermittently occupied, probably until the establishment of the Inca administration on the Central Coast (Mogrovejo and Segura 2000; Narváez 2006; Segura 2001). With its almost 120 ha of area built with tapia (rammed-earth block), Cajamarquilla contains a number of major mounds surrounded by numerous streets, plazas, houses, cemeteries, storerooms, hydraulic facilities, etc. Various features of the site, such as tapia technique, orthogonal organization that implies planning, and vessels stylistically Wari influenced in the nearby cemeteries of Nievería, have led many archaeologists to think that Cajamarquilla was a provincial Wari center with a large population, probably governed by intrusive Wari elites (e.g., Agurto 1984; Bueno 1974–1975). The alternative hypothesis is put forth by Ruth Shady, who sees the site as a commercial center strategically located between the highlands and the lowlands (Shady 1982, 1988, 1989). Primarily based on stylistic features of ceramics, Shady argues that the growth of Cajamarquilla during the Middle Horizon Epoch 1 can

be explained by the production of market items under the leadership of local elites. She suggests further that an economic depression occurred during the Middle Horizon Epoch 2, leading to the abandonment of Cajamarquilla and the establishment of a new commercial center at Pachacamac. As noted previously, fieldwork at Cajamarquilla focused on the Pedro Villar Córdova Compound (Figure 7.2). Given the observable superposition and mixture of different architectural designs, as well as variations in their degree of preservation, excavations were undertaken here to establish its occupational history. In addition to the cleaning of two extensive areas, forty-one excavation units, ranging from 9 m2 to 48 m2 and distributed in all the seven sectors of the compound, provided enough stratigraphic information and associated materials for reaching reasonable chronological conclusions. Two major architectural phases have been documented: Lima and Ychsma. During the Lima phase (late Early Intermediate Period and Early Middle Horizon, AD 500–700), the compound area had a trapezoidal shape defined by a long tapia wall that was blurred by later occupations and erosion. A dominant multilevel platform mound (sector I) that has overall dimensions of 18 m in height, 212 m in length, and 110 m in width was erected with tapia and thick fills of mud blocks extracted from the underlying alluvial deposits. A big lateral platform (sector II) that measures 4 m in height, 100 m in length, and 52 m in width is located on the northeast side and is believed to have been used as a funerary area. At the bottom of the mound, and surrounding it (sectors III and IV), large courtyards and rooms were built in orthogonal arrangement by following the orientation of the mound. Our architectural analysis indicates that the mound grew not only vertically, but also horizontally through more than ten construction episodes, including the addition of the lateral platform. As a result, the overall building achieved an L shape. Reports from other archaeological projects indicate that the horizontal growth and the L shape are found in other contemporary mounds in the Rímac Valley, particularly in Maranga (Narváez, personal communication 2008) and Huaca

T he Wari F ootprint on the C entral C oast

  115

Figure 7.2  Map of Villar Córdova Compound of Cajamarquilla (courtesy of Luis Cáceres).

Pucllana (Flores 2005). Thus, the main mound of the Villar Córdova Compound seems to exemplify a Lima monumental architectural tradition distributed in the middle and low Rímac Valley. Fills, floors, and associated features like fire and offering pits were typically associated with Late Lima ceramics (Maranga or Lima 7–9 and Nievería styles). For example, in front of the northwestern face of the mound (the hypothesized earliest section of the building), we were able to document in 2001 (excavation unit 9) a sequence of three superimposed Lima floors (5–10 cm in thickness each) successively associated with three tapia walls that formed the base of the lowest platform of the building. The oldest floor exhibited fortythree offering pits associated with at least two quincha (wattle-and-daub) rooms. The pits, which ranged from 30 cm to 70 cm in diameter and from 10 cm to 40 cm in 116 

R afael S egura L lanos and I zumi S himada

depth, contained badly preserved remains of plants, ash, and abundant fish bones. Although detailed analyses have not yet been carried out, peanut, squash, and a type of sardine were identified during excavations. Likewise, thirty-four of these offering pits contained small and eroded Late Lima ceramics (Lima 8–9 and Nievería styles). While less complex and preserved, these offering pits resemble those documented by Segura (2001) at a similar location in the Julio C. Tello Compound, the contemporary architectural complex that lies on the opposite end of the site of Cajamarquilla. Our excavation of unit 9 was finished with a test pit that reached 2 m below the offering level. The long sequence of culturally sterile geological strata confirmed that the mound was built no earlier than Lima phases 7–9. Additional test pits around the mound strengthened this conclusion.

The stratigraphic records also indicate that a number of flooding events impacted the compound during the Late Lima phases. While individual flooding events were not identified, damage produced by floods seems to have been particularly severe in two moments. Firstly, sand and silt deposits of 8 cm in thickness covered directly the oldest floor of sector III, a plain area on the north side of the compound exposed to quebrada discharges during wet periods. Similarly, the east and north flanks of the mound (sector I) facing the mouth of the Jicamarca quebrada exhibited evidence of eroded walls and adobe fused by water (Figure 7.3). Importantly, new floors in sector III and the absence of similar damage in the lateral platform (sector II) adjacent to the east flank of the mound suggest that Lima inhabitants recovered from these hydrological damages. A major second sequence of floods is represented by new sand and silt deposits over some of the last Lima courtyards without any evidence of subsequent architectural reconstruction. We did not find continuity between Lima occupation and the second occupational phase, which occurred

some five hundred years later, during the Ychsma period (ca. AD 1100–1470). In fact the Ychsma buildings overlie earlier Late Lima structures without intervening constructions. The second phase is defined by the construction of agglutinated rooms. Buildings tend to cover horizontally all the available space but are not voluminous. In many cases, rooms are organized around patios or plazas with terraces oriented to west. As hypothesized by Juan Mogrovejo (personal communication 1997), second-phase tapias change in preparation technique and morphology, becoming thinner than those of Late Lima and exhibiting distinctive horizontal grooves (Narváez 2006). Associated artifacts are much more diverse than those of Lima phase occupations, since we were able to document a wider range of contexts, including domestic areas, offerings, and burials. Significantly, potsherds coming from floors and complete vessels from offerings and burials are stylistically the same. Our collection is primarily made up of Ychsma style, but a poorly known,

Figure 7.3  Flood-damaged Late Lima walls at the bottom of the mound (Sector I) in the Villar Córdova Compound.

T he Wari F ootprint on the C entral C oast

  117

Figure 7.4  Abandoned irrigation canal (Sector VI) used during phase 2, probably built during phase 1, Villar Córdova Compound.

brownish, coarse ceramics named Cuculí (Guerrero 2004) and a few Chancay ceramics are also present. A good example of the new architectural style is observed on the west side of the Lima mound in sector IV, where a number of multiroom complexes surrounded by perimeter walls are prominent. Cleaning and excavation of one of these complexes revealed a large patio with a terrace and stairway, a room for processing food (containing fire pits, a storage pit containing sand and maize cobs, and grinding stones), another room with a bench and pits filled with organic refuse and broken whole vessels, and a rectangular funerary chamber (4.00, 2.64, and 1.40 m in length, width, and height, respectively) that had been looted. Other offering pits found in the patio included some gourd vessels and an animal tentatively identified as a monkey wrapped in a striped textile. Judging from these features, the multiroom complex seems to have been an elite residence. Much less well preserved, smaller rooms are found spread out in sector VI, an area on the west end of the 118 

R afael S egura L lanos and I zumi S himada

Villar Córdova Compound characterized by abandoned small branches of the Nievería canal (Figure 7.4). Five burials tentatively dated to the Late Intermediate Period on the basis of the iconography on associated textiles were documented here. One to nine individuals wrapped in multiple layers of textiles were buried in simple pits excavated on the border of the canal branches or inside some of these smaller rooms. Associated artifacts included ceramic figurines and vessels, metal ornaments and tools, and shells. The branches of the Nievería canal are associated with two important features of adjacent sector VII at the far northwestern end of the compound: a very big depression that appears to have been a type of reservoir or cocha, as well as abandoned raised fields or camellones. Each field measures on average 4 m in width and 20 m in length, and together they cover an area of approximately 0.36 ha. Although these fields have been reused in modern times, no comparable agricultural fields are documented in the entire Rímac Valley. These fields were

undoubtedly irrigated by an abandoned branch of the Nievería canal, although whether they date back to the Late Lima phase is still unclear. Dating the end of the Ychsma occupation is problematic, but new deposits of mud and sand overlie many of their structures. Ychsma usage of the structures after the flooding was limited to burial purposes, and only minor new constructions took place until the Inca arrival. The overall architectural sequence documented at the Villar Córdova Compound is corroborated by similar findings in the Julio C. Tello Compound and at contemporary sites associated with the Nievería canal. The Julio C. Tello Compound (Mogrovejo 2002; Mogrovejo and Segura 2000), the second largest compound of Cajamarquilla, exhibits a much better-preserved trapezoidal layout, resembling compounds at earlier Lima sites such as Cerro Culebras in the lower Chillón Valley (Paredes 1992). Six construction episodes were documented in the Julio C. Tello Compound. These episodes encompass, from the earliest to the latest, the Lima origin (Phases A–B) of the compound, its vigorous early development (Phases CI–III), Lima abandonment and occasional reuse during much of the Middle Horizon, its reoccupation by the Ychsma during the Late Intermediate Period (Phase D), and final abandonment before the Inca intrusion in the valley (Mogrovejo 2002; Mogrovejo and Segura 2000; Segura 2001). In terms of absolute chronology, we do not know when exactly the hiatus prior to Phase D took place, but it undoubtedly separates two major phases of occupation, each associated respectively with Late Lima and Nievería, and with Ychsma and Chancay ceramics. Considering that the Villar Córdova and Julio C. Tello compounds together represent nearly 40 percent of the visible, built area of the site, and that other archaeological projects also failed to detect any intervening occupations between those of Lima and Ychsma in other sectors, it can be argued that there were no extensive or intensive occupations during much of the Middle Horizon, especially Epoch 2 with an estimated time span of AD 700–850. We reach a similar conclusion from a review of archaeological remains in the general area surrounding Cajamarquilla. The Nievería canal, which supplied

water to the site, was part of an irrigation system formed by three major canals (Huachipa, Nievería, and Caraponguillo) on the north bank of the middle valley (Segura and Habetler 2008). Significantly, none of the twenty-two sites associated with this system (Milla 1974; FAUA–UNI–Fundación Ford 1994) exhibits an occupation between the Middle Horizon Epoch 1 and the Late Intermediate Period. While we do not suggest a complete depopulation of the area during the second part of the Middle Horizon, it is clear that occupations of this period are either quite sparse, ephemeral, or less physically visible than those of the late Early Intermediate Period (Lima), Late Intermediate Period (Ychsma), and Late Horizon (Inca). Beyond a few burials with Wari-derived items from some cemeteries associated with Cajamarquilla (D’Harcourt 1922; Mogrovejo and Segura 2000; Sestieri 1971), material remains dated to the Middle Horizon 2–4 are virtually “invisible.”

Pachacamac As is well known, since the pioneering work of Max Uhle (1903) at the end of the nineteenth century, the occupational sequence of Pachacamac has been primarily constructed on the basis of three kinds of interrelated archaeological data: monumental architecture, elite tombs, and high-quality decorated ceramics. Particularly the Lima (middle–late Early Intermediate Period), Ychsma (Late Intermediate Period), and Inca (Late Horizon) occupations of the site exhibit a good deal of material evidence. Unfortunately, the evidence for the Middle Horizon occupation is much more limited. In fact, the buildings presumably of this period are insufficiently studied and dated. Further, aside from the tombs Uhle excavated in 1896–1897, there have been hardly any new Wari or Wari-affiliated tombs scientifically excavated. Lastly, the diagnostic ceramics for this period with secure proveniences are extremely scarce. In spite of these deficiencies, Pachacamac has been postulated as having had a panregional religious influence and even political rivalry with Wari at least since the Middle Horizon Epoch 2 (Menzel 1964, 1968, 1977). In order to redress the extant, skewed knowledge and understandings of the site that rely excessively on

T he Wari F ootprint on the C entral C oast

  119

the elite remains and the highly visible monumental constructions, our research has focused on areas that have been scarcely explored thus far but nonetheless hold much potential for elucidating the social and environmental foundations of the site, two dimensions that we consider to be critical to understand Pachacamac’s longevity, preeminence, and attraction that crosscut ethnic, social, and political barriers. Two of these areas are the Pilgrims’ Plaza and the Urpi Kocha Lagoon (Figure 7.5).

Pilgrims’ Plaza Our best record of construction/occupation episodes has been obtained in the famed Pilgrims’ Plaza, a rectangular area of 65 × 320 m located just below the oldest monumental architectural sector of Pachacamac. By combining surface survey, seven large trenches (up to 10 × 10 m in extent), and large-scale ground-penetrating radar surveys (GPR) between 2003 and 2005, we were able to define a dense occupation of the area in the plaza

Figure 7.5  Map of Pachacamac showing the location of Pilgrims’ Plaza and Urpi Kocha Lagoon (courtesy of Go Matsumoto).

120 

R afael S egura L lanos and I zumi S himada

area represented by almost 3 m of stratified deposits rich in organic and artifactual materials together with superimposed occupational surfaces and architectural remains (Shimada et al. 2003; Shimada et al. 2005). Our excavations confirmed the hypothesis that the presently visible Pilgrims’ Plaza was a result of a largescale Inca transformation of the pre-Inca sacred landscape. To create the plaza, a thick fill (30–40 cm) of gravel and loose sand was used for burying the pre-Inca occupation and creating a new leveled surface. Then, a compacted earthen occupation surface was built and renewed at least a couple of times. Despite its poor preservation, a number of postholes, fire pits, and small offerings were documented. Finally, the entire plaza or at least much of it was “sealed” with a 5 cm thick cover of mud and small pebbles. Overall, these three layers account for the construction, use, and abandonment of the plaza during the Inca hegemony and collapse. Underneath the Inca occupation lies a truly complex sequence of Ychsma floors and fills that are associated with hundreds of offering pits and enclosures that span the entire Late Intermediate Period (Shimada et al. 2003; Shimada et al. 2005; Shimada et al. 2006a). A sense of the intensity, complexity, and continuity of these construction and ritual activities can be gained from the over eight hundred features in Trench 1 (T-1) that we documented in a 10 × 10 m area approximately 2 m in depth. Similarly, Trench 8, 30 m east of T-1, yielded 465 features of similar nature and scale in a 7.0 × 5.5 m area less than 1.5 m in depth. However, these findings were not homogeneously distributed. Unlike the literally thousands of offerings and associated structures concentrated in the eastern half of the plaza, considerably less dense cultural remains and a simpler stratigraphy were found in the western half. The overall pattern of use of these Ychsma levels consisted of adobe-brick and quincha rooms that delimited ritual spaces built time after time with few modifications in layout and location. Some looted dispersed cemeteries were also found underneath the plaza surface in 2003. Offerings consisted mostly of complex arrangements of presorted ceramic fragments or stones; one or more cántaros (utilitarian ceramic jars with ellipsoidal to spherical bodies) of different sizes; ceramic figurines

(invariably damaged); stone tools; whole fish of varied sizes and species; whole cuyes (guinea pigs), peanuts, maize kernels, ajíes (ají peppers), and other agricultural produce; and wildflowers (Shimada et al. 2006a). These items, usually separated by types, were mostly placed in intrusive conical pits or sunken adobe- or stone-lined and plastered chambers, sometimes with elaborate roofs. Interestingly, roofed chambers resemble smaller versions of tombs found nearby in front of the famed Painted or Pachacamac Temple, suggesting that they may have served as accessible loci of veneration of the deceased (Shimada et al. 2006a; Shimada et al. 2010). A more detailed analysis of features and their associations provided us with important insights on the nature of the Ychsma occupation. Firstly, we are certain that the use of the pre-Inca plaza had a strong seasonal character. Numerous burrows created by wasps and scorpions in patches of preserved occupational surfaces and prepared floors suggest that, during the late spring to the hot summer months, much of the plaza was not utilized. At the same time, analysis of paleobotanical remains indicates the noteworthy presence of plants in the prime of their reproductive season, indicating late winter to early spring occupation (Shimada et al. 2006a). Second, zooarchaeological data indicate an impressive array of marine and terrestrial animals, ranging from llamas, cuy, dog, a variety of rodent, and deer to seals and large and small marine fish (Shimada et al. 2006a). Our examination suggests that processing and consumption of fauna, including llamas, was quite limited, and that remains were mostly left in the plaza as offerings. Third, there is no doubt that different social groups were responsible for these ritual activities, as indicated by formal and constructive variability in roofed chambers, stylistic and technique variations of ceramic figurines, and local versus nonlocal ceramic groups identified by archaeometric analyses (neutron activation and Mössbauer analyses; Shimada et al. 2006a; Shimada et al. 2010). Likewise, the strontium isotope signatures (87Sr/86Sr) of dental and skeletal samples from twenty-six Ychsma-era individuals excavated from the Max Uhle cemetery (cemetery I in Figure 7.5) in front of the Pachacamac (also known as Painted) Temple, just south of the Pilgrims’ Plaza, are also heterogenous, suggesting that they spent the first years of

T he Wari F ootprint on the C entral C oast

  121

their lives in a variety of different locales (Knudson et al. 2008). Mitochondrial DNA analysis of the same group further shows considerable haplotype and haplogroup variabilities, suggesting that some of them may have pertained to highland population(s) (K. Shinoda, personal communication 2008; Shimada et al. 2010). Finally, radiocarbon dating of a number of features associated with well-defined floors indicates that Ychsma occupation started around AD 1100. For example, charcoal from a deep fire pit (ca. 1.7 m below the surface) provided a date of 950 ± 50 BP (2σ cal AD 1029 to 1219; Beta-184644). In the Pilgrims’ Plaza stratigraphy, the intrusion of the Ychsma offerings into the underlying levels resulted in a significant mixture of chronological markers. Thus, some Late Lima and Nievería, just a very few Middle Horizon Epoch 2 provincial Wari, and what Uhle (1903) called “Epigonal” style sherds were dislodged from their original stratigraphic settings. Overall, we were surprised by the extreme scarcity of post–Middle Horizon Epoch 1 ceramic fragments. Only one fragment relatively well preserved, belonging to a cup, could be reliably dated to Epoch 2 on the basis of its Atarco/Viñaque motifs (Figure 7.6). Figure 7.6  Viñaque/Atarco ceramic fragment from the Pilgrims’ Plaza.

122 

R afael S egura L lanos and I zumi S himada

The earliest Ychsma floors, which often directly overlie Lima adobito constructions, contained some human burials, mostly infants buried in simple small pits with very few associated artifacts. Presumably, these individuals might have been placed there as either dedicatory or propitiatory offerings that marked the onset of the Ychsma occupation. One secondary burial (T-1-’04 burial 1) corresponded to a male adult buried in a seated-flexed position but with the head forcibly oriented to the southwest, that is, to the Pachacamac Temple. Radiocarbon analysis of charcoal from a fire pit associated with a floor close in time to the burial provided a date of 1,277 ± 41 BP (2σ cal AD 679 to 894; AA65269). This burial associated with plain utilitarian vessels is the only undisturbed Middle Horizon feature that was documented in three seasons of excavations at the Pilgrims’ Plaza. Even though badly disturbed by Ychsma activities, it was possible to recognize that the Lima buildings consisted of platforms, ramps, and floors between 2.80 and 2.00 m below the surface of the plaza. We knew that Lima buildings could be massive and very deep as suggested by our ground-penetrating radar surveys and defined by our excavations at the Urpay Wachak Temple in 2003. Given the monumental character of the Lima architecture, our excavations stopped at this level in Trench 1 (2003–2004). However, we succeeded in reaching the sterile level in our adjacent Trenches 8 and 10 (2005), where evidence of Lima occupation was less intensive. Hence, it became evident that Lima occupation, at different depths, lies immediately below Early Ychsma levels and on sterile deposits, and that the nature of the Lima occupation differed greatly from that of the Ychsma period. While our ceramic collections from Pachacamac have a limited temporal extension, it is worth noting that the collections obtained by Strong and Corbett (1943) near the base of the Inca Temple of the Sun, which together represent the most complete ceramic sequence from Pachacamac up to now, exhibited a similar stylistic composition and distribution; this is to say, there are numerous diagnostic ceramics for the Late Intermediate Period and Late Horizon (Ychsma and Inca) and for the Early Intermediate Period (basically Lima and related

styles), but a scarce presence of sherds of the Wari and Wari-related styles. Particularly notable is a conspicuous absence of the emblematic Pachacamac style that Menzel (1964, 1968, 1977) defined using looted ceramics that are found today in museums in Hamburg and Berlin. As in Cajamarquilla, we do not have evidence that indicates any extensive or intensive Middle Horizon occupation at Pachacamac.

The Urpi Kocha Lagoon Of particular interest to our research is the “resilience” of Pachacamac as a religious center, not only to deal with political and religious pressures from a series of regional and pan-Andean polities (e.g., Wari, Middle Sicán, Ychsma, and Inca) and endure as a largely autonomous entity (Menzel 1964, 1968, 1977; Patterson 1985; cf. Segura and Shimada 2009 for the possible Middle Sicán co-option of Pachacamac), but also to face environmental disruptions, such as suggested by rain-damaged architecture documented in the Old Temple of Pachacamac (Franco and Paredes 2000) and in some sections of the Third Wall of the site (Paredes and Ramos 1992). Given the recognized evidence of severe El Niño events and droughts that occurred by the beginning of the Middle Horizon and the Late Intermediate Period (Shimada et al. 1991; Thompson et al. 1984; Thompson et al. 1985), we are most interested in seeing if such climatic abnormalities impacted the local populations and, if so, in determining their potential role in the extent and intensity of the site occupation. In order to collect and analyze appropriate paleoenvironmental data to this end, we selected one of the most important bodies of water associated with the site: the Urpi Kocha Lagoon located along the northwest corner of Pachacamac about 1 km inland from the Pacific coast. The lagoon, which covered about 22,000 m2 and held enough water for swimming and supporting small fish and ducks until some decades ago, now lacks standing water and is overgrown by reeds and other plants. While our long-term paleoenvironmental reconstruction integrates complementary data from diatom, pollen, macrophytic, and archaeological analyses, we summarize in this section only pertinent results from the diatom

analysis carried out on sediment cores extracted from the lagoon (Winsborough et al. 2010). Diatoms are critical for our effort at paleoenvironmental reconstruction (Winsborough et al. 2005). These are highly varied, short-lived, single-celled algae, each taxon of which has specific habitat requirements and preferences. Thus, the composition and relative frequencies of varied diatom taxa in archaeological context serve as sensitive indicators of the local environment. Depending on temporal resolution, changes in the composition of the diatom assemblage may reflect differentscale phenomena, such as individual meteorological events and longer-term climate changes. At the finest resolution, an abrupt change in the diatom assemblage, to one characteristic of a flooded, turbid lagoon, reflects an immediate diatom response to a short-term disturbance such as a heavy rain or a tsunami. If such a change in diatom composition is abrupt but long lasting it may indicate more severe, long-lasting storms associated with El Niño Southern Oscillation (ENSO) events. A gradual, sustained trend in diatom populations toward one favoring increasing salinity, on the other hand, is indicative of a long-term change in climate, such as a drought. Our study reveals a complex depositional history over last four thousand–plus years during which the lagoon was subjected to periods of intense storms and sustained drought. There is one particularly severe and sustained flooding event recorded in the sediment cores that are of interest to us. Radiometric dating (Accelerator Mass Spectrometry) of peat samples underlying and overlying the flood in question allows us to bracket and estimate its time of occurrence to approximately AD 600 (1540 ± 40 BP; 2σ cal AD 436 to 489, 513 to 515, and 530 to 651; Beta185162). The flood is represented by the thickest flood deposits in the two analyzed cores and is associated with an abrupt and permanent shift in the lagoon’s water quality from brackish to less saline and increasingly eutrophic (Figure 7.7). Further, this flood appears to have closely followed a prolonged drought recorded in the fifteenhundred-year precipitation records from the Quelccaya glacier (Shimada et al. 1991; Thompson et al. 1984). There is an independent documentation of this inferred ENSO-related flood at Pachacamac. As noted earlier, Franco and Paredes (2000) identified a thick

T he Wari F ootprint on the C entral C oast

  123

Figure 7.7  Chart showing diatom distributions from the Urpi Kocha Lagoon (courtesy of Barbara Winsborough).

layer of mud atop the Old Pachacamac Temple deposited by a substantial amount of rain that fell at the end of the Lima culture, ca. AD 600, which led to its “abandonment” and a major subsequent remodeling associated with the Pachacamac style. The dating makes it probable that this flood is one and the same as that documented at some Moche sites of the North Coast (Moseley et al. 1981; Moseley et al. 1983; Uceda 1992). The inferred flood certainly ranks it as one of the mega–El Niño events that Wells (1987) considers to occur only about every five hundred years. Another such event occurred around AD 1050–1100 (Shimada 1990; Winsborough et al. 2010). Overall, we must carefully consider short- and long-term environmental and social ramifications of the extraordinary conjunction of a mega–El Niño occurring near the end of the sustained, severe three-decade drought. It is difficult to model cultural response to this conjunction as there is nothing comparable in the recorded history in the Andes. Significant panregional changes in both social and physical landscapes would 124 

R afael S egura L lanos and I zumi S himada

be expected. The local-Wari interaction on the Central Coast thus needs to be considered within this extraordinarily fluid social and physical context.

Discussion What becomes apparent from the preceding is that our finds do not match well with postulates of the current models for explaining the Wari or Wari-related materials of the Central Coast. At the same time, we have good documentation at both Cajamarquilla and Pachacamac of many buildings and floors dating from the late Early Intermediate Period (Lima) to Late Horizon (Inca). We have, however, difficulty in locating any extensive or intensive occupations dating reliably to the Middle Horizon Epoch 2 on the base of Wari diagnostic elements, the quintessential chronological markers for this time. An exhaustive review of the occupations identified in other contemporary sites of the Central Coast core area (e.g., Copacabana, Huaca San Marcos, Huaca

Pucllana) leads to a similar conclusion. Indeed, independently from the Wari influences, one would expect to find the occupation remains that would correspond to the span of time between the first part of the Middle Horizon and the Late Intermediate Period. Yet, archaeological data available to identify and characterize those occupations is still quite scanty. Ideally, cultural occupations within large settlements can be reliably identified and dated by means of the combination of architectural features, artistic representations, associated materials, and radiocarbon or other radiometric measurements. Architectural episodes are especially relevant as indicators of the labor available during periods of intense cultural interaction or homogenization. These indicators are variably available from a number of sites belonging to both the Early and Late Horizon in the Central Coast. As for the Early Horizon, pre-Chavín U-shaped temples were remodeled when interactions with Chavín started to intensify, such as is observed in Huacoy in the Chillón Valley (Silva 1998) and Garagay in the Rímac Valley (Ravines and Isbell 1976; Ravines et al. 1984). As for the Late Horizon, Incarelated buildings were erected within or close to earlier public centers, such as Maranga (Canziani 1987) and Pachacamac (Hyslop 1990; Shimada 1991). In most mentioned cases, portable artifacts (ceramics, figurines, textiles, etc.) found in contexts other than burials supported inferences based on architectural data. Hence, we see that both in a context of dynamic religious interrelationship during the Early Horizon and direct political domination during the Late Horizon the local coastal inhabitants left a clear footprint of their presence in the places where they interacted with outsiders. Surprisingly, this is not the case for the Middle Horizon occupations. Until Epoch 1, as dated by the presence of Nievería ceramics, we see a lasting local architectural tradition characterized by platform-based mounds and small adobe-brick (adobitos) and tapia walls (e.g., Agurto 1984; Jijón y Caamaño 1949; Kroeber 1954). Afterward, a simultaneous and enigmatic occupational gap extends until the beginning of the Late Intermediate Period. In fact, the discontinuity observed in Cajamarquilla and Pachacamac seems to be replicated in other sites.

Two of the best-known Middle Horizon coastal sites are Ancón and Chimú Capac, and much of our knowledge concerning the major trends and characteristics of the pre-Hispanic funerary tradition, for example, was derived from burials excavated at Ancón since the late nineteenth century (Kauffman 1996; Kaulicke 1997; Ravines 1977, 1981; Reiss and Stübel 1880–1887; Slovak 2007; Tello 1946; Uhle 1998 [1910]). In spite of its significance to Peruvian archaeology, Ancón has provided us basically just funerary information; we cannot productively compare it with contemporaneous sites with more diverse lines of information. Similarly, Chimú Capac on the mouth of the Supe Valley has been an important part of the Middle Horizon discussions due to the burials excavated by Uhle at the Hacienda San Nicolás (Kroeber 1925; Menzel 1977), but the traces of the people living at the site during the Middle Horizon are still elusive (Valkenier 1995). The Chillón Valley provides us some betterdocumented cases. For instance, we know that the architecture in Copacabana (Falcón 2001:136) and Cerro Culebras (Paredes 1992; Silva et al. 1988), two of the most important Early Intermediate Period sites in the lower valley, was renewed and maintained between Middle Lima and Late Lima phases. No new architectural projects, however, are identified after the end of the Early Intermediate Period and the beginning of the Middle Horizon (Silva et al. 1988:29). More importantly, in a settlement patterns study, Silva (1996:386) found that newly established Middle Horizon sites are practically nonexistent in the valley. In the Rímac Valley, construction episodes in Huaca Pucllana ended during Middle Horizon Epoch 1. Since then the “pyramid” and other sectors of the site were used only as a cemetery (Flores 2005:86–88). A similar process is observed in Huaca Huallamarca during the same time span (Valladolid 1992). In Huaca San Marcos, one of the main mounds of Maranga, substantial architectural projects seem to have been carried out only until Middle Horizon Epoch 2 with no significant subsequent occupation until the Ychsma phase (Shady and Narváez 2000). This fact is corroborated by data from Huaca 20, a small mound within the Catholic University of Peru campus abandoned during the Middle Horizon Epoch 1,

T he Wari F ootprint on the C entral C oast

  125

as indicated by the presence of Chakipampa ceramics in its last architectural episode (MacKay and Santa Cruz 2000:590). Huaca Trujillo and Pirámide de Nievería, two sites near Cajamarquilla, also present a reduced occupation during the Middle Horizon 2–4. On the basis of limited excavations, Silva (1992:71) was able to identify an almost complete occupation sequence from the Early Intermediate Period to the Late Horizon in both sites, arguing that this might be due to a populational concentration at Cajamarquilla. However, we know now that Cajamarquilla did not house a significant population after Middle Horizon Epoch 1 until the Ychsma occupation in the Late Intermediate Period. As for the Lurin Valley, the entire discussion about the Middle Horizon is centered on the site of Pachacamac. As pointed out by Marcone (this volume), the most relevant valley-wide studies deal only with the Early Intermediate Period occupations (e.g., Earle 1972; Patterson et al. 1982), a fact that seems to be explained by absence of post-Lima public buildings and difficult-toidentify domestic sites of the same period. South-central coast valleys present a similar situation. Cerro del Oro in the Cañete Valley shows three occupation phases, with the last one dating to the Middle Horizon Epoch 1, followed by an exclusively funerary use (Ruales 2000:369). A similar case is observed in Huaca Malena in the nearby Asia Valley, where post–Middle Horizon Epoch 1 constructions have not been clearly identified. Instead, Wari burials as identified by their textile associations were intrusive to the earlier architecture (Angeles and Pozzi-Escot 2000:403–404). In the Mala Valley site of Cerro Salazar, the monumental sector was buried ex professo by the end of the Early Intermediate Period. Only a small domestic occupation continued to exist during the Middle Horizon (Gabe 2000:49–50). These data confirm that the funerary use of mounds and platforms by the end of the Middle Horizon 1 was a wide phenomenon on the Central Coast (MacNeish et al. 1975:54). From the above, we see a widespread agreement among archaeologists of different generations, working with distinct methodologies at sites of varied scale and nature in different valleys in and close to the Central 126 

R afael S egura L lanos and I zumi S himada

Coast. This consensus may be explained in terms of some pervasive methodological or sampling biases and limitations or, alternatively, by widespread factors and roughly synchronous processes that elicited similar social behaviors among divergent prehistoric local populations. The former may include scarce radiocarbon dates, excessive emphasis on burials and fancy ceramic styles as chronological markers, and difficulties in recognizing postLima architecture. The latter includes threat of or actual Wari invasion. However, as seen already, the evidence for the post–Middle Horizon Epoch 1 Wari-related occupations is weak and problematic in various regards. All of the Pachacamac-, Viñaque-, and Atarco-style fragments that have been recovered on the Central Coast are, in reality, just a handful in number and from secondary context fills. In the Old Temple of Pachacamac, Franco and Paredes (2000:625, Figures 22, 23) found a few Middle Horizon Epoch 2 sherds in a fill used for burying the temple during Middle Horizon Epoch 3 (Franco and Paredes 2000:613). Likewise, there is no consensus on the chronology and stylistic definition of the miniature ceramics that the authors found as offerings in the temple (e.g., Franco and Paredes 2000 versus Kaulicke 2000). In Huaca San Marcos, some Pachacamac sherds, as well as decorated gourd vessels, were found only in a fill sealing a passageway of Platform 2 (Narváez 1999:9–10; Shady and Narváez 1999:7). Even Socos, the formerly hypothesized Wari administrative center in the middle Chillón Valley (Isla and Guerrero 1987), represents a problematic case. In fact, Wari-related ceramics and textiles presented by Isla and Guerrero are surface collected, and the identification of the presumed Wari-affiliated architecture of Socos has been recently questioned (Jennings 2006). While the presence of Middle Horizon Epoch 2 Wari or Wari-related ceramics in the aforementioned sites could be explained by any of a variety of interrelationships between the central coastal societies and the empire (e.g., limited trade, local emulation of the prestigious or innovative Wari religious art for elite or ritual purposes, etc.), what is notable about them is their paucity and limited range of context of occurrence. They form such a notable contrast with the much more abundant, varied,

and reliable context occurrences of the earlier and later styles (Lima, Nievería, Ychsma, and Inca). We ask where the sherds found in architectural fills originated. Even though generalizations are difficult, we believe, based on our experience at Cajamarquilla and Pachacamac, that many of the vessel fragments come from postinterment removal of funerary goods. This discussion leads us to an interesting paradox. The paucity of reliable evidence of extensive or intensive Middle Horizon occupations, particularly those of Epoch 2, differs from the relative abundance of pertinent funerary evidence. Starting in the Middle Horizon 1B, individuals were buried in flexed position and wrapped with successive layers of textiles. The resulting bundles were accompanied by artifacts that included vessels in some of the Wari or Wari-inspired styles (e.g., Atarco, Viñaque, Pachacamac). The new type of burial replaced the Lima funerary tradition that was primarily but not exclusively characterized by extended individuals and associated with more local and conservative ceramic styles. The new burials are well known from a number of Central Coast sites, such as Ancón (Ravines 1977, 1981; Reiss and Stübel 1880–1887; Uhle 1913), Nievería (Gayton 1927; Uhle 1998 [1910]), Vista Alegre (Stumer 1957), and Pachacamac (Uhle 1903). Less impressive but similarly important are a few looted Wari-related burials recently found in Huaca Pucllana (Flores 2005). In the case of Cajamarquilla, besides a Middle Horizon Epoch 2 burial found by the Italian Archaeological Mission atop the Julio C. Tello multilevel platform mound (Sestieri 1971), seven looted funerary chambers dating to the same epoch were documented in 1996–1997 at the same mound (Mogrovejo and Segura 2000). Importantly, despite the severe destruction of such contexts, some Pachacamac, Moche V-inspired, and Huamanga vessels were recovered from the debris left by the looters. At Pachacamac, our Trench 7 excavated in 2005 in the cemetery in front of the Painted Temple (Temple of Pachacamac) provided us nine fairly complete burials and the remains of twenty-two severely damaged ones that together spanned from the Middle Lima phase (ca. AD 300–500) to the Ychsma-Inca phase. In light of the described Wari impact on the regional funerary

practices, we were very interested in documenting a number of intact or disturbed Wari or Wari-related burials. Our findings included a large, figure eight–shaped collective tomb with masonry walls and a solid beamed roof (Tomb 1–2) that contained at least thirty-four bundles (Shimada et al. 2006b; Shimada et al. 2010). Most of the bundles were placed concentrically around the largest and presumably the oldest and most important bundle (Bundle B) that had a wooden mask whose style is conventionally dated somewhere between Middle Horizon Epoch 2 and 3 (Figure 7.8; Kaulicke 1997). No Middle Horizon pottery, however, was found in the tomb. A nearby burial (Burial AG) that had been severely damaged by the later construction of the Tomb 1–2, however, did contain an intact kero-type cup with a polychromic anthropomorphic representation that is commonly assigned to the Middle Horizon 3 (Figure 7.9). The incongruity between the stylistic dating of the cup and the wooden mask (Middle Horizon 3 and 2 or 3, respectively) on one hand, and the interment sequence of Tomb 1–2 and Burial AG, on the other, highlights the current troubling situation of the Middle Horizon chronology on the Central Coast. An AMS dating of a gourd vessel that covered the kero cup of Burial AG yielded a date of 940 ± 40 BP (2σ cal AD 1041 to 1217; Beta-244626). AMS dating of a sample each of an outer (Layer 2) and an inner (Layer 5) cloth wrapping the adult male inside Bundle B of Tomb 1–2 yielded dates of 760 ± 40 BP (2σ cal AD 1224 to 1319 and 1352 to 1384; Beta-244627) and 690 ± 40 BP (2σ cal AD 1286 to 1394; Beta-244628), respectively. These dates are much later than those commonly assumed for Middle Horizon Epochs 2 and 3. While we cannot dismiss the possibility that wooden mask of Bundle B may have been curated from an earlier bundle, an AMS dating of fibers attached to a similar wooden mask from a disturbed bundle in a small cemetery in the Pilgrims’ Plaza just northwest of the Painted Temple yielded much the same date of 708 ± 34 BP (2σ cal AD 1280 to 1329 and 1336 to 1391; AA65273). It is clear that for a meaningful discussion of Middle Horizon cultural dynamics, we need many more radiocarbon dates for samples from well-documented contexts, as well as caution in our use of stylistic dating.

T he Wari F ootprint on the C entral C oast

  127

Figure 7.8  Funerary bundle B, Tomb 1–2, from the Uhle Cemetery in front of the Painted Temple (also known as Pachacamac Temple).

Figure 7.9  Decorated kero-type cup from burial AG, Uhle Cemetery (drawing by César Samillán Torres).

Our Trench 7 reached the sterile soil starting some 2.2 m below the current surface of the cemetery. In general, in spite of the large quantity of disturbed ceramics recovered from the trench, only a few sherds of either provincial Wari or Wari-inspired styles were obtained. A similar observation can be made in relation to the most widely discussed Middle Horizon funerary items excavated by Uhle (1903) just a few meters away. During the problematic Middle Horizon Epoch 2, we suspect that the number of decorated ceramic vessel offerings per burial declined and the number of decorated textiles per burial increased. If this were the case, one productive way to better identify contexts belonging to this epoch would be to perform detailed textile analyses such as those undertaken by Angeles and PozziEscot (2000) or Prümers (1989, 2000). Yet, analyses of textiles excavated by Uhle have identified only a few Middle Horizon pieces (Engelstad 1984; Kohut 2005; Van Stan 1967), a fact that leads us to ask about the identity of individuals buried in the cemetery in front of the Painted Temple. We are aware that Menzel’s cornerstone Middle Horizon pottery chronology (Menzel 1964, 1968) is being reevaluated on the basis of new ceramic studies and radiometric measurements (Isbell 2000). Results suggest that in terms of radiocarbon or calibrated calendrical dates, Middle Horizon Epoch 1 styles lasted into Epoch 2. Similarly, recently acquired radiocarbon dates suggest that the last two Moche ceramic phases (IV and V) correspond to Menzel’s Middle Horizon Epoch 1 and 2, respectively (Chapdelaine 1999, 2003, this volume; cf. Table 2 in Shimada 1994:4). Given the known presence of Nievería and Pachacamac ceramics in Late Moche burials (e.g., Castillo 2000), in calendrical dates, Nievería ceramics may well have lasted until the Middle Horizon Epoch 2 as well. In turn, this would imply that drastic changes observed in the use of central coastal civic-ceremonial sites occurred in the second half of the Middle Horizon. While the scarcity of reliable radiocarbon dates from well-defined Middle Horizon archaeological contexts prevents us from being certain, we hypothesize that the cessation of the Nievería style reflects the collapse of sociopolitical foundation of the central coastal elites.

We argue that a comprehensive understanding of the Middle Horizon cultural dynamics on the Central Coast requires an in-depth consideration of an array of evidence (e.g., radiocarbon dating, isotopic analysis of human bones, trace element analysis of associated materials, etc.) in addition to mortuary data that are often poorly documented or analyzed. One line of evidence that merits our careful consideration is paleoenvironmental. There are enough indications that paleoenvironmental disturbances played an important role in the cultural processes during the Middle Horizon. Major hydrological, economic, and social changes that would be expected from severe and widespread floods and droughts should be integral parts of any comprehensive explanatory model of the Middle Horizon cultural dynamics. Reliable and informative indicators of these natural phenomena are often not readily found in conventional archaeological settings such as residences and burials. Pertinent to this volume is the documentation of severe and protracted droughts that impacted much of the Andes during the latter half of the sixth century and the first half of the eleventh century (e.g., Dillehay and Kolata 2004; Kolata 2000; Moseley 2002; Shimada 1994; Shimada et al. 1991; Thompson et al. 1984; Thompson et al. 1985; Winsborough et al. 2010), as well as past floods as detected by certain specific deposits and depositional patterns at diverse coastal sites such as Cajamarquilla (Mogrovejo and Makowski 1999), Galindo (Moseley et al. 1981), Pampa Grande (Shimada 1994), and Cahuachi (Grodzicki 1994). In some cases, archaeological correlations and radiocarbon dates from detritical charcoal associated with flood deposits (e.g., Wells 1987, 1990) can allow us to bracket these events to the time periods of interest to this volume. In the case of Cajamarquilla, the preliminary study of a channel branching from the Huaycoloro quebrada, near the Julio C. Tello Compound (Mogrovejo and Makowski 1999), suggests a complex sequence of consecutive water overflows and flow declinations by the end of the Lima occupation. While deposits have not been radiometrically dated, much of this dynamic would have taken place by the critical sixth–seventh centuries. As for our own field data, there is no doubt that at least three

T he Wari F ootprint on the C entral C oast

  129

major flooding sequences impacted the site, one damaging the Late Lima structures, the second sealing those structures, and the last eroding the subsequent Ychsma buildings. We hypothesize that the second sequence would correlate with the fluvial period that followed the thirty-two-year drought at the end of the sixth century, whereas the last one would correspond to similar events occurring by the eleventh century (Shimada et al. 1991; Thompson et al. 1984, Thompson et al. 1985). Pluvial/interpluvial sequences are more accurately observed at Pachacamac, where diatoms from the Urpi Kocha Lagoon provide a fine-grained paleoenvironmental record (Winsborough et al. 2010). Diatom composition indicates a number of intense floods during or immediately after dry periods. The effects of two of these events are particularly noteworthy—one dated to the eleventh century (not discussed in this article) and the other dated to the sixth century—and produced a drastic shift in the water quality of the lagoon. Of crucial importance to issues at hand is the fact that floods can be much more devastating if they occur as punctuating phenomena during or after processes of environmental deterioration as represented by prolonged droughts (Moseley 1997). We wonder if such conditions underlay the massive abandonment of Lima sites during the first part of the Middle Horizon and the virtual absence of civic-ceremonial centers during the rest of that period. In this conception, the Wari intrusion or influence into the Central Coast was preconditioned by the inferred social, political, and economic disarray resulting from a severe drought-flood sequence. While it is simple to invoke environmental disturbances such as torrential rains that commonly accompany ENSO events and resultant flooding as causal factors for observed cultural changes, it is in reality quite challenging to document or demonstrate their magnitude, duration, timing, and spatial coverage, as well as how and why they would lead to the observed changes. In fact, we need more in-depth, multidisciplinary investigations into pre-Hispanic paleoenvironmental conditions and processes and their interaction with local populations on the Central Coast.

130 

R afael S egura L lanos and I zumi S himada

Conclusions In their original versions, both the local, commercial and the foreign, imperialist models argued that the center of regional political power moved from Cajamarquilla to Pachacamac at the beginning of Middle Horizon Epoch 2. Further, it has been stated that Cajamarquilla was definitively abandoned (Shady 1988, 1989), and Pachacamac became an ideological and probably political supraregional power (Franco and Paredes 2000; Menzel 1964, 1968). Our excavations do not support this scenario. Cajamarquilla was certainly abandoned by the onset of the Middle Horizon 2, but neither totally or permanently. The site experienced a resurgence during the Late Intermediate Period, becoming as extensive as during the Lima occupation. At Pachacamac, contrary to what one would expect as the inferred regional, political, and ideological center during the Middle Horizon Epoch 2, no reliable evidence of either extensive or intensive occupations of this epoch has emerged from any fieldwork, including ours, over the past century. Although Menzel (1964, 1968, 1977) postulated a powerful Pachacamac polity that restricted, if not prevented, the dominance of the Wari Empire on the Central Coast during this epoch, we are not even certain if the Pachacamac-style artifacts bearing its emblematic griffin were produced and disseminated from the site of Pachacamac. In fact, basic questions about Pachacamac’s geographical, social, economic, and political bases remain unanswered. Overall, our preliminary conclusions can be extended to the entire core area of the Central Coast; physical evidence for Middle Horizon 2 occupations in this area seems too diffused or localized or limited in size and quantity to be readily detected. There seem to have been no institutions powerful or effective enough to mobilize sufficient human and material resources to undertake public projects of lasting quality and notable scale and size during this problematical epoch. What could explain this apparent occupational hiatus or archaeological elusiveness? While this may have resulted from our own methodological limitations and skewed sampling, we cannot ignore, at the same time, some factors such as severe and long-term environmental

disturbances during the first part of the Middle Horizon (Shimada et al. 1991; Winsborough et al. 2010). The cumulative effects of these new environmental conditions could have forced the regional population to be dispersed and politically fragmented during the Middle Horizon Epoch 2. In this regard, we may consider any Wari invasion or influence on the Central Coast to be related to Ayacucho’s social response to the postulated environmental disturbances of the period, and imperial actions would have taken place amid the social disarray of a regional population faced with the same environmental challenge. By integrating local social and environmental conditions, we believe this working hypothesis is more effective in explicating the local-Wari interaction during the Middle Horizon than the unimodal models described at the outset of this article.

Acknowledgments We want to thank members of the Cajamarquilla Ar­­ chae­­ological Project and the Pachacamac Archaeological Project for their assistance in our fieldwork projects. Our work at Cajamarquilla (2000–2001) received the financial support of the Refinería de Zinc de Cajamarquilla S.A. and was sponsored by the Instituto de Conservación y Restauración Yachay Wasi. For their generous support of our research at Pachacamac (2003–2005), we are grateful to the National Science Foundation (BCS-0313964, 0411625), National Geographic Society (7472–03, 7668– 04, 7724–04), and John Heinz III Foundation. All of our fieldwork was authorized by the Instituto Nacional de Cultura del Perú. We thank Melody J. Shimada for her valuable comments and editorial assistance.

References Cited Agurto, Santiago 1984 Lima prehispánica. Municipalidad de Lima-FINAMPRO, Lima. Angeles, Rommel, and Denise Pozzi-Escot 2000 Textiles del Horizonte Medio. Las evidencias de Huaca Malena, valle de Asia. In Huari y Tiwanaku: Modelos vs. evidencias. Segunda parte, edited by Peter Kaulicke and William H. Isbell, pp. 401–424. Boletín de Arqueología PUCP No. 4. Fondo editorial de la Pontificia Universidad Católica del Perú, Lima.

Bueno, Alberto 1974–1975 Cajamarquilla y Pachacamac: Dos ciudades de la costa central del Perú. Boletín Bibliográfico de Antropología Americana 36:171–201. Canziani, José 1987 Análisis del complejo urbano Maranga Chayavilca. Gaceta Arqueológica Andina 14:10–17. Castillo, Luis Jaime 2000 La presencia de Wari en San José de Moro. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 143–179. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Chapdelaine, Claude 1999 Struggling for Survival. The Urban Class of the Urban Site of Moche, Peru. In Environmental Disaster and the Archaeology of Human Response, edited by Garth Bawden and Richard M. Reycraft, pp. 121–142. Maxwell Museum of Anthropology Anthropological Papers No. 7. University of New Mexico, Albuquerque. 2003 La ciudad Moche: Urbanismo y estado. In Moche: Hacia el final del milenio, Vol. 2, edited by Santiago Uceda and Elias Mujica, pp. 247–285. Universidad Nacional de Trujillo-Pontificia Universidad Católica del Perú, Lima. D’Harcourt, Raoul 1922 La céramique de Cajamarquilla-Nievería. Journal de la Societé des Americanistes 14:107–118. Dillehay, Tom D., and Alan L. Kolata 2004 Long-Term Human Response to Uncertain Environmental Conditions in the Andes. Proceedings of the National Academy of Sciences of the United States of America 101(12):4325–4330. Earle, Thimoty K. 1972 Lurin Valley, Peru: Early Intermediate Period Settlement Development. American Antiquity 37(4):467–477. Engelstad, Helen 1984 A Group of Grave Tablets and Short Fragments from Pachacamac. Ñawpa Pacha 24:61–72. Falcón, Víctor 2001 Copacabana un centro urbano de la cultura Lima en la costa central. In XII Congreso Peruano del Hombre y la Cultura Andina, Vol. 2, edited by Ismael Pérez, Walter Aguilar, and Medardo Purizaga, pp. 126–138. Universidad Nacional San Cristóbal de Huamanga, Ayacucho, Peru. FAUA–UNI–Fundación Ford 1994 Inventario del patrimonio inmueble de Lima. Valles del Chillón, Rímac y Lurín. FAUA–UNI–Fundación Ford, Lima. Flores, Isabel 2005 Pucllana, esplendor de la cultura Lima. Instituto Nacional de Cultura, Lima. Franco, Régulo, and Ponciano Paredes 2000 El Templo Viejo de Pachacamac: Nuevos aportes al estudio del Horizonte Medio. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 607–630. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Gabe, Carmen 2000 Investigaciones arqueológicas en el Cerro Salazar, Mala. Serie de Investigaciones CEAMA 1, Lima.

T he Wari F ootprint on the C entral C oast

  131

Gayton, Anna 1927 The Uhle Pottery Collections from Nievería. University of California Publications in American Archaeology and Ethnology Vol. 21, No. 8. University of California Press, Berkeley. Grodzicki, Jerzy 1994 Nasca: Los síntomas geológicos del fenómeno El Niño y sus aspectos arqueológicos. Centro de Estudios Latinoamericanos, Universidad de Varsovia, Warsaw. Guerrero, Daniel 2004 Cronología cerámica y patrones funerarios del valle del Rímac: Una aproximación a los períodos tardíos. In Puruchuco y la sociedad de Lima. Un homenaje a Arturo Jiménez Borja, edited by Luis Felipe Villacorta, Maria Luisa Vetter, and Carlos Ausejo, pp. 157–177. CONCYTEC, Lima. Hyslop, John 1990 Inca Settlement Planning. University of Texas Press, Austin. Isbell, William H. 1988a Comment to “La época Huari como interacción de las sociedades regionales” Ruth Shady Solís. Revista Andina 6(1): 105–108. 1988b City and State in Middle Horizon Wari. In Peruvian Prehistory, edited by Richard W. Keatinge, pp. 164–189. Cambridge University Press, Cambridge. 2000 Repensando el Horizonte Medio: El caso de Conchopata, Ayacucho, Perú. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 9–68. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Isbell, William H., and Gordon F. McEwan 1991 A History of Huari Studies and Introduction to Current Interpretations. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 1–17. Dumbarton Oaks Research Library and Collection, Washington, D.C. Isla, Elizabeth, and Daniel Guerrero 1987 Socos: Un sitio Wari en el valle del Chillón. Gaceta Arqueológica Andina 14(4):23–28. Jennings, Justin 2006 Understanding Middle Horizon Peru: Hermeneutic Spirals, Interpretative Traditions, and Wari Administrative Centers. Latin American Antiquity 17(3):265–285. Jijón y Caamaño, Jacinto 1949 Maranga, contribución al conocimiento de los aborígenes del valle del Rímac. La Prensa Católica, Quito, Peru. Kauffman D., Federico 1996 Proyecto arqueológico tumbas de Ancón (1). Arqueológicas 23:9–177. Kaulicke, Peter 1997 Contextos funerarios de Ancón. Esbozo de una síntesis analítica. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. 2000 La sombra de Pachacamac: Huari en la costa central. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 313–358. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima.

132 

R afael S egura L lanos and I zumi S himada

Knudson, Kelly, Martha Palma Málaga, Izumi Shimada, and Rafael Segura Llanos 2008 Geographic Origins and Residential Mobility at the Pilgrimage Center of Pachacamac, Peru. Paper presented at the 77th Annual Meeting of the American Association of Physical Anthropologists, Columbus. Kohut, Lauren 2005 Weaving Culture: Burial Textiles from Pachacamac. Unpublished Bachelor’s thesis, Department of Anthropology, Bryn Mawr College, Bryn Mawr, Philadelphia. Kolata, Alan L. 2000 Environmental Thresholds and the “Natural History” of an Andean Civilization. In Environmental Disaster and the Archaeology of Human Response, edited by Garth Bawden and Richard M. Reycraft, pp. 163–178. Maxwell Museum of Anthropology Anthropological Papers No. 7. University of New Mexico, Albuquerque. Kroeber, Alfred 1925 The Uhle Pottery Collections from Supe. University of California Publications in American Archaeology and Ethnology Vol. 21, No. 6. University of California Press, Berkeley. 1954 Proto-Lima, a Middle Period Culture of Peru. Fieldiana Anthropology Vol. 44, No. 1. Chicago National History Museum, Chicago. Lumbreras, Luis G. 1969 De los pueblos, las culturas y las artes del Perú antiguo. MoncloaCampodónico Editores Asociados, Lima. 1989 El estudio arqueológico del estado. Gaceta Arqueológica Andina 4(16):3–5. MacKay, Martín, and Raphael Santa Cruz 2000 Las excavaciones del proyecto arqueológico Huaca 20 (1999 y 2000). In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 583– 595. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. MacNeish, Richard, Thomas Patterson, and David Browman 1975 The Central Peruvian Prehistoric Interaction Sphere. Papers of Robert S. Peabody, Foundation for Archaeology Vol. 7. Robert S. Peadody Foundation for Archaeology, Andover, Massachusetts. Makowski, Krzysztof 2000 El sindrome de Catal Huyuk: Observaciones sobre las tendencias aglomerativas tempranas. Arqueologia y Sociedad 13:99–118. Menzel, Dorothy 1964 Style and Time in the Middle Horizon. Ñawpa Pacha 2:1–105. 1968 New Data on the Huari Empire in Middle Horizon Epoch 2A. Ñawpa Pacha 6:47–114. 1977 The Archaeology of Ancient Peru and the Work of Max Uhle. R. H. Lowie Museum, Berkeley, California. Milla, Carlos 1974 Inventario, catastro y delimitación del patrimonio arqueológico del valle del Rímac y Santa Eulalia. Centro de Investigaciones y Restauración de Bienes Monumentales. Instituto Nacional de Cultura, Lima. Mogrovejo, Juan 2002 Cajamarquilla y el fin de la cultura Lima. Boletín del Instituto Riva Agüero 26:227–243.

Mogrovejo, Juan, and Krzysztof Makowski 1999 Cajamarquilla y los Mega Niños en el pasado prehispánico. Íconos 1:46–57. Mogrovejo, Juan, and Rafael Segura 2000 El Horizonte Medio en el conjunto arquitectónico Julio C. Tello de Cajamarquilla. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 565–582. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Moseley, Michael E. 1997 Climate, Culture, and Punctuated Change: New Data, New Challenges. The Review of Archaeology 18(1):19–27. 2002 Modelling Protracted Drought, Collateral Natural Disaster, and Human Responses in the Andes. In Catastrophe and Culture. The Anthropology of Disaster, edited by Susanna H. Hoffman and Anthony Oliver-Smith, pp. 187–212. School of American Research Press, Santa Fe, New Mexico. Moseley, Michael E., Robert A. Feldman, and Charles R. Ortloff 1981 Living with Crises: Human Perception of Process and Time. In Biotic Crises in Ecological and Evolutionary Times, edited by Matthew Nitecki, pp. 231–267. Academic Press, New York. Moseley, Michael E., Robert A. Feldman, Charles R. Ortloff, and Alfredo Narváez 1983 Principles of Agrarian Collapse in the Cordillera Negra, Peru. Annals of Carnegie Museum 52:299–327. Narváez, Joaquín 1999 Proyecto de investigaciones arqueológicas en la Huaca San Marcos. Resultados preliminares. Boletín del Museo de Arqueología y Antropología 5:5–10. Universidad Nacional Mayor de San Marcos, Lima. 2006 Sociedades de la antigua ciudad de Cajamarquilla. Investigaciones arqueológicas en el Sector XI del Conjunto Tello y un estudio de la colección tardía del Conjunto Sestieri. Avqi Ediciones, Lima. Paredes, Juan 1992 Cerro Culebra: Nuevos aportes acerca de una ocupación de la cultura Lima (costa central del Perú). Gaceta Arqueológica Andina 22:51–62. Paredes, Ponciano, and Jesús Ramos 1992 Evidencias arqueológicas del “Niño” en las excavaciones de Pachacamac. In Paleo-ENSO Records International Symposium: Extended Abstracts, edited by Luc Ortlieb and José Macharé, pp. 225–233. ORSTOM, Lima. Patterson, Thomas C. 1985 Pachacamac—An Andean Oracle under the Inca Rule. In Recent Studies in Andean Prehistory and Protohistory, edited by D. Peter Kvietok and Daniel H. Sandweiss, pp. 159–176. Latin American Studies Program, Cornell University, Ithaca, New York. Patterson, Thomas, John P. McCarthy, and Robert Dunn 1982 Polities in the Lurin Valley during the Early Intermediate Period. Ñawpa Pacha 20:61–82. Prümers, Heiko 1989 Tejidos del Horizonte Medio del valle de Huarmey. In The Nature of Huari. A Reappraisal of the Middle Horizon Period in Peru, edited by Richard M. Czwarno, Frank M. Meddens, and Alexandra Morgan, pp. 188–213. BAR International Series 525. Bristish Archaeological Reports, Oxford.

2000 “El Castillo” de Huarmey: Una plataforma funeraria del Horizonte Medio. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 289–312. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Ravines, Rogger 1977 Prácticas funerarias en Ancón (primera parte). Revista del Museo Nacional 43:327–398. 1981 Prácticas funerarias en Ancón (segunda parte). Revista del Museo Nacional 45:89–166. Ravines, Rogger, and William H. Isbell 1976 Garagay: Sitio temprano en el valle de Lima. Revista del Museo Nacional 41(1975):253–272. Ravines, Rogger, Helen Engelstad, Victoria Palomino, and Daniel Sandweiss 1984 Materiales arqueológicos de Garagay. Revista del Museo Nacional 46(1982):135–233. Reiss, Wilhelm, and Alphons Stübel 1880–1887 Peruvian Antiquities: The Necropolis of Ancon in Peru; A Series of Illustrations of the Civilization and Industry of the Empire of the Incas, translated by A. H. Keane. 8 parts. A. Ascher, New York. Rowe, John H. 1963 Urban Settlements in Ancient Peru. Ñawpa Pacha 1:1–27. Ruales, Mario 2000 Investigaciones en Cerro del Oro, valle de Cañete. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 359–399. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Schaedel, Richard 1951 Major Ceremonial and Population Centers in Northern Peru. In The Civilization of Ancient America: Selected Papers of 29th International Congress of Americanists, edited by Sol Tax, pp. 232–243. University of Chicago Press, Chicago. 1966 Incipient Urbanization and Secularization in Tiahuanacoid Peru. American Antiquity 31:338–344. Segura, Rafael 2001 Rito y economía en Cajamarquilla. Pontificia Universidad Católica del Perú, Lima. 2007 Old Problems and New Issues and Evidence for the Archaeology of the Central Coast: The Case of Cajamarquilla. Paper presented at the 35th Annual Meeting of the Midwest Conference on Andean and Amazonian Archaeology and Ethnohistory, Carbondale. Segura, Rafael, and Patricia Habetler 2008 Pre-Inka Irrigation Canals and Settlements on the North Bank of the Rímac River. Paper presented at the 36th Annual Meeting of the Midwest Conference on Andean and Amazonian Archaeology and Ethnohistory, Madison. Segura, Rafael, and Izumi Shimada 2009 Examining Middle Sicán–Central Coast Interaction, ca. A.D. 1000. Paper presented at the 74th Annual Meeting of the Society for American Archaeology, Atlanta. Sestieri, Pellegrino 1971 Cajamarquilla, Peru. The Necropolis on the Huaca Tello. Archaeology 17(1):12–17.

T he Wari F ootprint on the C entral C oast

  133

Shady, Ruth 1982 La cultura Nievería y la interacción social en el mundo andino en la época Huari. Arqueológicas 19:15–108. 1988 La época Huari como interacción de las sociedades regionales. Revista Andina 6(1):67–99. 1989 Cambios significativos ocurridos en el mundo andino durante el Horizonte Medio. In The Nature of Huari. A Reappraisal of the Middle Horizon Period in Peru, edited by Richard M. Czwarno, Frank M. Meddens, and Alexandra Morgan, pp. 1–22. BAR International Series 525. British Archaeological Reports, Oxford. Shady, Ruth, and Joaquín Narváez 1999 La Huaca San Marcos y la antigua ciudad de Maranga-Lima. Museo de Arqueología y Antropología de la Universidad Nacional Mayor de San Marcos, Lima. 2000 Historia prehispánica de Lima: Arqueología de la Huaca San Marcos. Museo de Arqueología y Antropología de la Universidad Nacional Mayor de San Marcos, Lima. Shimada, Izumi 1990 Cultural Continuities and Discontinuities on the Northern North Coast, Middle-Late Horizons. In The Northern Dynasties: Kingship and Statecraft in Chimor, edited by Michael E. Moseley and Alana Cordy-Collins, pp. 297–392. Dumbarton Oaks, Washington, D.C. 1991 Pachacamac Archaeology: Retrospect and Prospect. In Pachacamac, A Reprint of the 1903 Edition by Max Uhle, edited by Izumi Shimada, pp. XII–LXVI. University Museum Monograph 62. University Museum of Archaeology and Anthropology, University of Pennsylvania, Philadelphia. 1994 Pampa Grande and the Mochica Culture. University of Texas Press, Austin. Shimada, Izumi, Crystal Barker Schaaf, Lonnie G. Thompson, and E. Mosley-Thompson 1991 Cultural Impacts of Severe Droughts in the Prehispanic Andes: Application of a 1,500-Year Ice Core Precipitation Record. World Archaeology 22:247–270. Shimada, Izumi, Rafael A. Segura Llanos, David J. Goldstein, Kelly J. Knudson, Melody J. Shimada, Ken ichi Shinoda, Mai Takigami, and Ursel Wagner 2010 Un siglo después de Uhle: Reflexiones sobre la arqueología de Pachacamac y Perú. In Max Uhle (1856–1944): Evaluaciones de sus investigaciones y obras, edited by Peter Kaulicke, Manuela Fischer, Peter Masson, and Gregor Wolff. Fondo Editorial de la Pontificia Universidad Católica del Perú and Fritz Thyssen Stiftung für Wissenschaftsförderung, Lima. Shimada, Izumi, Rafael Segura, David Goldstein, Melody Shimada, Robert Speakman, Ursel Wagner, and Hirokatsu Watanabe 2006a What Did People Do at Pachacamac? Identity, Form, Timing and Meaning of Offerings. Paper presented at the 34th Midwest Conference on Andean and Amazonian Archaeology and Ethnohistory, Nashville. Shimada, Izumi, Rafael Segura, John Jones, María Rostworowski, and Hirokatsu Watanabe 2003 The Pachacamac Archaeological Project: Results of the First Season and Their Implications. Paper presented at the 22nd Northeast Conference on Andean Archaeology and Ethnohistory, Cambridge.

134 

R afael S egura L lanos and I zumi S himada

Shimada, Izumi, Rafael Segura, and Go Matsumoto 2006b Living with the Dead: Conception and Treatment of the Dead at Pachacamac. Paper presented at the 34th Midwest Conference on Andean and Amazonian Archaeology and Ethnohistory, Nashville. Shimada, Izumi, Rafael Segura, María Rostworowski, and Hirokatsu Watanabe 2005 Una nueva evaluación de la Plaza de los Peregrinos de Pachacamac: Aportes de la primera campaña 2003 del Proyecto Arqueológico Pachacamac. Boletín del Instituto Frances de Estudios Andinos 33(3):507–538. Silva, Jorge 1992 Ocupaciones postformativas en el valle del Rímac: HuachipaJicamarca. Pachacamac 1:49–74. 1996 Prehistoric Settlement Patterns in the Chillon River Valley, Peru. Ph.D. dissertation, University of Michigan, Ann Arbor. University Microfilms, Ann Arbor. 1998 Una aproximación al periodo Formativo en el valle del Chillón. In Perspectivas regionales del periodo Formativo en el Perú, edited by Peter Kaulicke, pp. 251–268. Boletín de Arqueología PUCP No. 1. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Silva, Jorge, Daniel Morales, Rubén García, and Enrique Bragayrac 1988 Cerro Culebras, un asentamiento de la época Lima en el valle de Chillón. Boletín de Lima 56:23–34. Slovak, Nicole M. 2007 Examining Imperial Influence on Peru’s Central Coast: Isotopic and Cultural Analysis of Middle Horizon Burials at Ancón. Unpublished Ph.D. dissertation, Department of Anthropological Sciences, Stanford University, Stanford, California. Strong, William D., and John M. Corbett 1943 A Ceramic Sequence at Pachacamac. In Archaeological Studies in Peru, 1941–1942, edited by William D. Strong, Gordon R. Willey, and John M. Corbett, pp. 27–122. Columbia University Studies in Archaeology and Ethnology Vol. 1, No. 2. Columbia University Press, New York. Stumer, Louis A. 1957 Cerámica negra del estilo Maranga. Revista del Museo Nacional 26:272–289. Tello, Julio C. 1946 Los tesoros arqueológicos de Ancón. El Comercio (Lima), 17 March. Thompson, L. G., E. Mosley-Thompson, J. F. Bolzan, and B. R. Koci 1985 A 1500-Year Record of Tropical Precipitation in Ice Cores from the Quelccaya Ice Cap, Peru. Science 229:971–973. Thompson, L. G., E. Mosley-Thompson, and B. Morales Arnao 1984 El Niño-Southern Oscillation Events Recorded in the Stratigraphy of the Tropical Quelccaya Ice Cap, Peru. Science 203:50–53. Torero, Alfredo 1970 Lingüística e historia de la sociedad andina. Anales Científicos de la Universidad Nacional Agraria 8(2):231–264. Uceda C., Santiago 1992 Evidencias de grandes precipitaciones en diversas etapas constructivas de la Huaca de la Luna, costa norte del Perú. In Paleo-ENSO Records International Symposium: Extended

Abstracts, edited by Luc Ortlieb and José Macharé, pp. 315–318. ORSTOM, Lima. Uhle, Max 1903 Pachacamac, Report of the William Pepper, M. D. LL. D., Peruvian Expedition of 1896. Department of Archaeology, University of Pennsylvania, Philadelphia. 1913 Die Muschelhügel von Ancon, Peru. Proceedings of the XVIII session, Part 1, pp. 22–45. International Congress of Americanists, London, 1912. 1998 [1910] Acerca de las culturas tempranas de Lima y sus alrededores (1910). In Max Uhle y el Perú antiguo, edited by Peter Kaulicke, pp. 231–254. Fondo Editorial de la Pontificia Universidad Católica del Perú , Lima. Valkenier, Lisa 1995 New Evidence for Chimu Capac and the Early Horizon Period in the Supe Valley. Journal of the Steward Anthropological Society 23(1):269–286. Valladolid, Clide 1992 Huallamarca. Pachacamac 1:133–134. Van Stan, Ina 1967 Textiles from Beneath the Temple of Pachacamac, Peru: A Part of the Uhle Collection. University of Pennsylvania Press, Philadelphia. Vásquez, Segundo 1984 La Waka Pucllana. Gaceta Arqueológica Andina 3(9):8–9. Von Hagen, Adriana, and Craig Morris

1998 The Cities of the Ancient Andes. Thames and Hudson, London and New York. Wells, Lisa E. 1987 An Alluvial Record of El Niño Events from Northern Coastal Peru. Journal of Geophysical Research 92(C13):14463–14470. 1990 Holocene History of the El Niño Phenomenon as Recorded in Flood Sediments of Northern Coastal Peru. Geology 18:1134–1137. Williams, Carlos 1980 Arquitectura y urbanismo en el antiguo Perú. In Historia del Perú, Vol. 8, pp. 369–575. Editorial Juan Mejía Baca, Lima. Winsborough, Barbara, John Jones, Lee Newsom, Izumi Shimada, Rafael Segura, and María Rostworowski 2005 Paleoenvironmental Reconstruction at Pachacamac: Integrating Diatom, Pollen, Macrobotanical, and Archaeological Data. Paper presented at the 33rd Annual Meeting of the Midwest Conference on Andean and Amazonian Archaeology and Ethnohistory, Columbia. Winsborough, Barbara, Izumi Shimada, John Jones, Lee Newson, and Rafael Segura 2010 Paleoenvironmental Catastrophes on the Peruvian Coast: Multidisciplinary Analysis of Sediment Cores from Pachacamac. Manuscript on file, Southern Illinois University, Carbondale.

T he Wari F ootprint on the C entral C oast

  135

Ch a p t er 8

o What Role Did Wari Play in the Lima Political Economy? The Peruvian Central Coast at the Beginning of the Middle Horizon Giancarlo M arcone F.

A

rchaeological research has increasingly disputed the centralized character of the spread of Wari in the Andes. Field data continue to challenge style-based chronologies, showing that strong local developments, traditionally thought to belong to the Early Intermediate Period, are at least partially contemporaneous with the pan-Andean expansion of Wari traits. Recently, the coexistence of strong local societies with the pan-Andean distribution of Wari or Wari-like objects has been explained using models that focus on prestige exchange networks. According to these models, international prestige or prestige exchange networks explain the presence of Wari or Wari-emulated objects in each area. These explanations are at least partially replacing the traditional vision of Wari’s sphere of influence in some regions. Whereas Wari influence in the past was viewed as a “mosaic” of imperial strategies that ranged from direct (high cost) to indirect (low cost) (Schreiber 1992), it is now increasingly viewed as the voluntary emulation of Wari features, which occurred as a result of the self-centered search for prestige by local elites. However, the utility of these explanations for understanding local sociopolitical contexts may be limited because of their focus on prestige accumulation as the central strategy used by the elites to uphold a privileged position in society.

136

This chapter discusses the available data on sociopolitical transformation in the Lima culture from the end of the Early Intermediate Period to the beginning of the Middle Horizon. Available data support the idea that a multivalley polity was forming on the Peruvian Central Coast at this time. The process of increasing political centralization was initially centered on agricultural intensification and the expansion of agricultural land. With this intensification came a centralized management of production by elites in the lower Chillón and Rímac valleys, and the emergence of “rural” and decentralized elites in the Lurín Valley and the chaupi yunga.1 The inclusion of these rural elites into the ChillónRímac lower valley political formation was sustained by the management of local (Lima) symbols. Wari-related foreign materials played a role only at the higher levels of sociopolitical practices helping to build both the cohesion of several of these lower valley elites and separate these elites from a second level of more decentralized rural elites in the Lurín Valley and elsewhere. It is not until we understand these local transformations—during and after the Early Intermediate Period—that we can recognize what role Wari symbols played in the preexisting Lima political economy.

Explanations of Interactions with Wari It is now common in Andean archaeology and elsewhere to critique centralized models of interregional interactions (Cusick 1998; Goldstein 2000; Jennings 2006a, 2006b; Schortman et al. 2001; Stein 2002, 2005). These critiques are based on the assertion that centralistic models, like core-periphery or assimilation, are top-down approaches that focus mainly on processes related to the core area of the state and give local polities only the capacity to react to overarching state orders (Goldstein 2000:336). One problem with such centralistic perspectives is that they assume that any evidence of contact with an empire (or state) is proof of hierarchical interregional relations (Jennings 2006a). The process is understood only in relation to the center, relegating local trajectories to second place and biasing the interpretations of these independent trajectories to fit the regional scenario (Shady 1988). As an alternative to centralized models,2 several Andean archaeologists (e.g., Burger and Matos 2002; Covey 2000; Goldstein 2000; Jennings and Yepez 2001; Lau 2005) are incorporating explanations that reinterpret the emulation models proposed in the 1970s.3 In these agent-based models, the larger polity’s existence provides new settings and opportunities for emergent elites. Elites emulate selected aspects of ritual and ideology voluntarily, with little or no intervention from the other society, in order to meet their own objectives of power accumulation. The materials and practices that resemble the empire or state help to legitimate social leadership and even contribute to the creation of new elite identities (Schortman et al. 2001; Stein 2002:903). In these explanations, the need to be a part of the regional prestige system compels elites to intensify production in local societies (Burger and Matos 2002; Jennings and Yepez 2001; Lau 2005). Without denying the multiple insights gained from these approaches to ancient societies, these models can be critiqued for pushing economic factors aside and assuming that any local change in social leadership and political centralization occurs because of the quest for elite prestige (Smith 2004:76; Stein 2002:903). Under these models, elite legitimization rests on the

domination of ritual production, communal ceremonial practices, and positions within hierarchical kinship systems (DeMarrais et al. 1996; Goldstein 2000; Schortman et al. 2001), and the control over economic factors like land, production, or labor comes as a derivative of this new prestige. There is a growing group of scholars who point out that the heavy focus on external prestige relations in this approach needs to be complemented by concern with internal economic developments. These authors (e.g., Cobb 1993; Smith 2004; Wells 2006) seek to complement such perspectives by focusing on the “variability in the relationship between politics and economics” (Smith 2004:77). Recent investigations on the nature of Wari influence in polities outside of Ayacucho are reviving some of Shady’s (1988) earlier ideas about the existence of a prestige-goods exchange network during the Middle Horizon (e.g., Jennings 2006a, 2006b, this volume; Lau 2005). Proposing a scenario in which several peer regional polities, or “emporiums,” interacted through prestige-goods exchange networks (Shady 1988:68), Shady argued that these networks allowed for the emergence of state-level formations on the Peruvian Central Coast and elsewhere. The revived prestige exchange networks models argue that many of the sociopolitical changes occurring around the end of the Early Intermediate Period and the beginning of the Middle Horizon were largely endogenous to each region—a result of the entry of local elites into a wider, central Andean elite political culture in which Wari images and canons were cultural capital, rather than the result of Wari political or economic hegemony (see Jennings, this volume; Lau 2005). Jennings (2006a) concludes that emerging elites appear to have seized upon the international identity of the Wari state, less in a desire to associate themselves with a vaguely understood distant state, and more in a bid to forge a shared identity with neighbors. Elites monopolized and restricted access to Wari canons in objects and architecture and used them as prestige symbols in diacritical feasting activities and other public displays. Such feasting and public displays finally allowed them to mobilize labor and consolidate sociopolitical power.

W hat Role D id Wari P lay in the L ima P olitical E conomy?

  137

The focus on elites seeking prestige through prestige exchange networks, however, ends up highlighting international relations at the expense of homogenizing local responses to fit international models. The resultant models obscure fundamental local developments. I argue that this has been the case in the study of the Peruvian Central Coast between the Early Intermediate Period and the Middle Horizon. Although the precise nature of Wari presence on the Central Coast is still controversial, the area is continually presented as part of the Wari Empire, and local political developments are disregarded in several syntheses about the Middle Horizon and Andean archaeology (e.g., Conlee and Ogburn 2004:6; Isbell 2004:5; Richardson 1994:122; Schreiber 2001:78– 79, 2004:132; Von Hagen and Morris 1998). This chapter seeks to move away from futile discussion on whether the area was or was not part of a Wari Empire. It is time to incorporate the political economy of the Lima culture into Wari studies in order to gain a better understanding of the Central Coast during the Middle Horizon.

Wari and Lima Culture Interaction at the Beginning of the Middle Horizon The archaeological recognition of the Lima culture began with the pioneering work of Max Uhle in Pachacamac (Uhle 1903; see Kaulicke 2001; Shimada 1991). Based on materials recovered in Pachacamac and later in Cerro Trinidad, he was able to define the presence of material remains belonging to a society that he called Proto-Lima as an antecedent of the Tiwanaku civilization in the Lima region. Later studies focused primarily on ceramic style were able to identify two major phases within the Lima culture (Kroeber 1926; Strong and Corbett 1943; Willey 1943). The earlier phase was named Playa Grande, and the later phase was named Maranga, although other terms like Interlocking and Proto-Lima have been used for these phases in the Lima style (see Flores 1981; Kaulicke 2001; Montoya 1995). During the Playa Grande phase, the Lima culture was centered in the Chillón and Chancay valleys (Kaulicke 2001), while the Rímac Valley was the heartland of the Lima culture during the “Maranga” phase (Kaulicke 2001) (Figure 8.1). Patterson (1966) refined these initial stylistic observations and 138 

G iancarlo M arcone F.

developed a nine-phase stylistic chronology in which the last three phases (7, 8, and 9) correspond roughly with what was identified by earlier scholars as Maranga (Kaulicke 2001; Segura 2004).4 After Wari was identified as the source of Tiwanakulike influence on the Central Coast, Patterson (1966) and Menzel (1964) identified a style called Nievería, which incorporates local and Wari-related iconographic features. Nieveria was assigned to the Middle Horizon Epoch 1B, replacing phase 9 of the Lima style and inaugurating the period of Wari influence on the coast. In Middle Horizon Epoch 2, the Nievería style was in turn replaced by another locally produced Wari-related style named Pachacamac (Kaulicke 2001; Menzel 1964). Although the orderly transition between styles created the impression that these styles represented different cultures at different times, this was not the case. Recent work in the area is proving stratigrafically that Late Lima (Patterson’s last three phases) and Nievería are at least partially contemporaneous and correspond with the beginning of the Middle Horizon Period (Ccencho 2001; Guerrero and Palacios 1994; Kaulicke 2001; Mac Kay and Santa Cruz 2000; Marcone 2001; Mogrovejo and Segura 2001; Montoya 1995; Narváez 2006; Segura 2001, 2004; Shady and Narváez 2000). The presence of Wari-related styles like Nievería and Pachacamac is clear proof that at least some interaction between the Central Coast and the region of Ayacucho did occur. What is not clear is what this interaction meant to local sociopolitical transformation in the region. Lima Tardío and Nievería ceramics styles are associated with evidence of a major sociopolitical transformation in the area. Monumental sites like Maranga (Canziani 1987; Shady 1982; Shady and Narváez 2000), Cajamarquilla (Mogrovejo y Segura 2001), Huaca Pucllana (Ccencho 2001; Flores 1981; Montoya 1995), and Pachacamac (Marcone 2001; Shimada 1991) grow in expanse and monumentality. There is also an expansion of Lima-style sherds distributions to the chaupi yunga of the Chillón and Rímac valleys, as well as in the neighboring valley of Lurín (Patterson et al. 1982). Numerous archaeologists interpret these social transformations as evidence for the emergence of a state-level society (Dillehay 1976,

Figure 8.1  Peruvian Central Coast during the Early Intermediate Period (drawing based on Agurto Calvo 1984; Narváez 2006; Patterson et al. 1982; and Villacorta et al. 2004).

1979; Earle 1972; Kaulicke 2001; MacNeish et al. 1975; Patterson et al. 1982; Shady 1982, 1988).

Models of a Lima State: Local Developments versus External Stimuli Despite agreement on the formation of a state-level society in the area, characterizations of the development of the Lima polity have been formulated between two opposite extremes. First, some archaeologists argue that the development of a Central Coast state occurred during the Middle Horizon because of interactions with Wari. These authors stress the role of external ties, with either elite status increasingly tied to (1) participation in exchange networks of ideologically charged prestige

items (MacNeish et al. 1975; Shady 1982, 1988); (2) emulation of Wari elites; or (3) “clientage” relationships with the Wari Empire (Isla and Guerrero 1987; Kaulicke 2001; Menzel 1964). These explanations have a marked emphasis on the study of elite behavior, focusing especially on the study of ceramics styles present in public buildings, offerings, and burials. The other extreme in the explanations of Lima state development suggests that its development occurred during the Early Intermediate Period and was based on local processes of economic centralization. In these later models, Wari presence in the area was more linked to the demise of the Lima state than with its rise (Stumer 1954, 1956). In contrast to prestige models that are supported largely by those who study elite ceramics, researchers in

W hat Role D id Wari P lay in the L ima P olitical E conomy?

  139

the area who focus on the study of settlement patterns or households tend to deny the influence of Wari and propose a gradual process of economic differentiation that culminates in the formation of a Lima state. Under these models, elite status is linked to agricultural intensification and surplus mobilization, perhaps through their control of (1) prime land and canals (Dillehay 1979; Earle 1972; Patterson et al. 1982); (2) the flow of agricultural goods inside a vertical exchange system (Dillehay 1979:25); or (3) kin-based labor mobilization organized by hereditary community leaders (Patterson et al. 1982:64). In another possible scenario usually overlooked on the Central Coast, elites would dominate ritual practices in relation to local traditions of style and material culture instead of external or foreign objects. According to this perspective, elites based their power on and conveyed support using local symbols as a source of prestige. Local objects were therefore prestigious enough to earn support from the general population. Vaughn (2004) has made a similar point in relation to the distribution of prestige symbols by Cahuachi in the Nazca region, where local elites strengthened their positions by distributing ceramics on the village level.

Lima Culture Transformations from the Early Intermediate Period to the Middle Horizon The archaeological remains of the Central Coast are under constant threat from the urban expansion of modern Lima. One of the consequences of this urban pressure is that archaeological research is focused on monumental buildings, burials, and more “valuable” sites that are in danger of damage by the encroaching city. In an effort to rescue these more valuable sites, less visible contexts like domestic and productive sites are being systematically destroyed. There are therefore only a handful of studies outside of elite contexts that give archaeological evidence of the sociopolitical and economic transformations of Lima culture between the Early Intermediate and the Middle Horizon. These studies are, in general, incomplete or poorly published. However it is possible to reconstruct a tentative picture of Lima’s sociopolitical 140 

G iancarlo M arcone F.

organization during the time of such transformations, which can help us to understand the role that Wari played in Central Coast political economy.

Settlement patterns Population rose significantly in the Chillón, Rímac, and Lurín valleys during the Early Intermediate Period (Agurto Calvo 1984; Earle 1972; Silva 1992, 1996; Stumer 1954). This population increase ended at the beginning of the Middle Horizon, and the decrease of the population in the area is reflected in the complete abandonment of Lima sites and the lack of archaeological structures associable with the second period of the Middle Horizon (Segura and Shimada, this volume; Silva 1992:399–400; Stumer 1954). Stumer (1954) and Silva (1992, 1996) proposed that the population change between the Early Intermediate Period and the Middle Horizon was linked to increased centralization in the Rímac and Chillón valleys during the Early Intermediate Period (Stumer 1954:144). This process of increasing centralization culminated with the spread of the Lima culture to the adjoining Lurín Valley (Stumer 1954:144). Stumer also suggested that the area experienced a general increase in the construction of public buildings at this time. A two-tiered settlement hierarchy developed with sites containing a combination of public buildings and residential areas in the first tier (Agurto Calvo 1984:87; Silva 1992:399). These first-tier sites, located exclusively in the lower Chillón and Rímac valleys, were surrounded by second-tier sites that only had a residential component (Silva 1996).5 The idea of first-tier sites centralizing populations as early as the middle phases of the Early Intermediate Period on the Central Coast has long been discussed (Kaulicke 2001; Shady 1988; Silva 1992; Stumer 1954). There is limited to no evidence, however, for a clear center or capital of the Lima culture during the Early Intermediate Period. For example, in the Chillón Valley, Silva was unable to identify one site as the center or “capital,” but instead suggested that Cerro Culebras, Playa Grande, and Copacabana worked together as a sort of nuclei for the early and middle periods of the Early Intermediate Period in the valley (Silva 1996:147). In the

Rímac Valley, Stumer proposes that sites like Maranga, Cajamarquilla, Huaca Juliana, and Vista Alegre were at the top of the settlement hierarchy (Stumer 1954:132). The proposed coexistence of these sites led Kaulicke (2001:325) to discuss briefly the possibility that each site represented the head of a city-state. Agurto Calvo (1984) has a different interpretation. He considers Maranga the initial center or “capital” for Lima in the Rímac Valley, which was then replaced by Cajamarquilla at the beginning of the Middle Horizon (Agurto Calvo 1984:78). In the chaupi yunga and the Lurín Valley, a different pattern was registered. Settlements in the area were dispersed and composed almost exclusively of residential units (Agurto Calvo 1984; Guerrero and Palacios 1994; Silva 1996). In the Lurín Valley, Earle (1972) and Patterson and colleagues (1982) distinguished a change in the domestic settlement pattern at the beginning of the Early Intermediate Period, when sites were moved up to the tops of hills as a strategy to make more agricultural land available on the valley floor (Earle 1972). After maximum agricultural land potential was reached in the middle of the Early Intermediate Period, settlements became more agglutinated, and elite compounds appeared in these villages (Earle 1972:475) that by the Middle Horizon were integrated into the Lima state (Earle 1972). The political situation began to change toward the end of the Early Intermediate Period as power shifted to the south. Several public buildings in the lower Chillón Valley fell into disuse. The abandonment of public buildings in the lower valley does not extend to the residential units, which continued to be used (Paredes 1992; Silva 1996; Silva et al. 1988). Meanwhile in the Rímac Valley, sites like Maranga and Cajamarquilla grew in extension and monumentality, and Cajamarquilla even acquired characteristics that have been identified as evidence of urbanism (Kaulicke 2001; Segura and Shimada, this volume). In the Lurín Valley, the construction of public buildings seems to begin at the end of the Lima sequence. This pattern is in agreement with the formation of a multivalley polity (centered in the Rímac Valley) that would go on to include the Lurín Valley by the end of Lima’s cultural sequence (Earle 1972; Kaulicke 2001). Simultaneous with the expansion of Lima public architecture in the Lurín Valley at the end of the Early

Intermediate Period, the extension of Lima culture traits into the highlands suggests Lima population movements up the valley (Dillehay 1979:26; Silva 1996). Agurto Calvo, for example, made reference to Lima ceramics found as high as San Pedro de Casta in the middle-upper region of the Rímac Valley (Agurto Calvo 1984:91). The sporadic presences of Lima ceramics in the middle and upper parts of the valleys—traditionally associated with highland sociopolitical formations—suggests that the chaupi yunga (or upper low valley) sites were under the political control of the people down the valley by the end of the Lima sequence (Dillenay 1979; Earle 1972; Patterson et al 1982; Silva 1992:401). Within the three valleys, it is difficult to identify domestic or public sites that date to the second phase of the Middle Horizon (Agurto Calvo 1984:105; Patterson et al. 1982; Silva 1992; Stumer 1954:142). In terms of residential architecture, Stumer argues that either we lack the necessary diagnostic elements to identify changes, or that there were no residential structures in the Middle Horizon (Stumer 1954:136). Since Segura and Shimada (this volume) have similar problems identifying a Middle Horizon occupation after the fall of Lima society in highly public sites like Cajamarquilla and Pachacamac, the second phase of the Middle Horizon on the Central Coast remains poorly understood.

Household Social Differentiation Little research has been carried out at the site level regarding household differentiation as a way to understand social and political differences, with the notable exception of Cerro Culebras in the lower Chillón Valley where domestic structures have received more attention than at any other Lima site (Paredes 1992; Silva et al. 1988). At Cerro Culebras, Silva and colleagues identified two different sectors of households, one in the northern part of the site constructed primarily from canes, and a southern part, constructed using small stones and adobitos (little mud bricks) (Silva et al. 1988:27–29). There are no ascertainable differences in the ceramic assemblages of these different household constructions, which leads researchers to believe that such differences are more related to economic

W hat Role D id Wari P lay in the L ima P olitical E conomy?

  141

specialization than just wealth or status differences between households (Silva et al. 1988). Silva and his colleagues (1988) carried out excavations in each of these sectors and found that stone residences were associated with abundant shellfish remains. They proposed that the people who lived in these residences were specialized fishermen with selective predilection for species like Mesodesma donacium, Perumytilus purpuratus, and Semimytilus algosu (Silva et al. 1988:33). In the residences made of canes, they found evidence of hearths, deposits of “ollas,” manos, batanes, cottonseeds, and other remains that they associate with agricultural activities (Silva et al. 1988:30–31). They conclude that the differences between the houses can be explained in terms of economic specialization that first occurred around the middle of the Lima sequence. This specialization was the major impetus for the later development of power differentials in the area (Silva et al. 1988). Through intersite comparison with other low valley sites, Silva and his colleagues show both that cane houses are more common in the valley and that stone residences are only present in sites with public buildings. Based on this distribution they conclude that being a fisherman constituted a specialized activity with more prestige in Lima society (Silva 1992:401; Silva et al. 1988:31). The idea is provocative but needs to be investigated in more depth. One potential issue with their data, for example, is that varying quantities of shellfish could correspond to differences in diet rather than occupation since no fishing tools were reported, only increased quantities of shells. Guerrero and Palacios (1994) briefly studied several village-level (El Vallecito, Huampani, and Huachipa) sites in the chaupi yunga of the Rímac Valley. They found that residential units were generally similar in shape, organization, and building materials to those described by Silva (1996) in the Chillón Valley. They note a change in these villages in phases 5–6 in Patterson’s sequence as the agricultural frontier was expanded to the quebrada of Huachipa and more storage deposits were constructed for agricultural products (Guerrero and Palacios 1994:302). Guerrero and Palacios (1994:280) did not find major interhousehold variability between sites throughout the Early Intermediate Period in terms of architecture or associated objects. The only exception was Sector C of 142 

G iancarlo M arcone F.

the site El Vallecito, where they found rooms that were bigger and with “better” architectural details than those found at the other sites (Guerrero and Palacios 1994:280). This sector included a rectangular building that the authors think could have served public functions. It is in this sector, as well as from burials at other sites, that they identified forms, wares, and decorations that they interpret as antecedent of the Nievería style (Guerrero and Palacios 1994:306–308). The presence of Nieveríalike ceramics in such contexts led the authors to propose that the development of Nievería began as a means to appropriate ritual practices in order to control expanding agricultural production, as shown in the increasing existence of several types of storage facilities at these sites (Guerrero and Palacios 1994:299–300, 306).

Agriculture Intensification Agurto Calvo (1984) and Shady (1982) proposed that elites controlled water and irrigation channels in the lower valleys. For example, in the Rímac Valley they proposed that each major canal was under the control of a major site: the Río Magdalena canal was controlled by Huaca Huantilla, the Río Huatica canal by Limatambo, and the Río Surco canal, which irrigates the modern-day neighborhoods of Miraflores, Surco, and Chorrillos, by Huaca Pucllana (Agurto Calvo 1984:84–85; Shady 1982). These authors see the relation between public sites and major canals as evidence of a centralized and almost bureaucratic management of water resources (Shady 1982). In contrast, each village elite in the chaupi yunga controlled and managed his own canals and ditches in a decentralized way until the area was incorporated into the overarching Lima order at the beginning of the Middle Horizon (Earle 1972:496; Silva 1996). Craft Production As in many cultures during the Early Intermediate, there was also an increase in craft specialization during the Lima culture (DeLeonardis and Lau 2004). For example, there is evidence that points to an increased specialization in the production of textiles, in particular those produced in the tapestry technique (Agurto Calvo 1984;

Mogrovejo 1995). The designs in these textiles are based on local iconographic traditions similar to the ones manifested on ceramics and in frisos that cover public architecture (Agurto Calvo 1984). These fine textiles have been recovered in association with public buildings and burials (Mogrovejo 1995). But so far there is no evidence of how or where this textile production was organized. Other evidence for increasing specialization comes from the standardization of wares during Late Lima phases. First, as Segura (2004) points out, Patterson defined eight wares for the whole Lima sequence. Two of them are constant and present in all the phases of the sequence, but these two wares only become predominant in the later part of the sequence (after phase 7, but this could be biased by how Patterson’s samples were taken) (Patterson 1966; Segura 2004). Second, Earle’s studies of plainwares based on sites in the middle Lurín Valley show a high variability both within and between sites for most of the Early Intermediate sequence. This situation changed drastically toward the end of the sequence when a tendency toward the unification in plainwares began (Dunn 1979:27; Earle 1972:469–470). Signs of standardization may reflect centralized management by the polity by the end of the sequence (Earle 1972:476).

Interregional Exchange The people of the Central Coast likely participated in long-distance exchange networks at least from the first millennium BC onward (Browman 1975:325; MacNeish et al. 1975). By the time of Wari contact, Dillehay (1976) suggests that there existed significant exchange, not of prestige goods, but of agricultural and staple goods between Lima coastal elites and the neighboring populations from the highlands, in particular the Mantaro Valley (Browman 1975; MacNeish et al. 1975). Guerrero and Palacios (1994:299) found evidence that corroborated this suggestion when they discovered that the only evidence of exotic ceramics in the upper Rímac Valley is from the Mantaro area. Yet, the strongest evidence for Central Coast participation in a pan-Andean exchange network comes from Nievería-style ceramics. Nievería-style ceramics have a wide distribution in the Andes. They appear in regions

like the North Coast (Castillo 2007), central highlands (Lau 2005), and Huari (Knobloch 1991), and there is even a loose reference to Nievería ceramics in Manta, Ecuador (Jijón y Camaaño 1949, quoted in Agurto Calvo 1984:87). The presence of Nievería style in the core of Wari in preimperial epochs (Knobloch 1991) lends support to the idea that Nievería style represents the adoption of an international canon, more than a political or economic imposition by Wari (Jennings, this volume). Knobloch suggests: “The early Epoch 1A presence of a few Nievería style shards at Huari with Chakipampa and Ocros refuse followed by the later Epoch 1B presence of fancy Chakipampa animal icons on Nievería pottery indicates a rather long and rather peaceful coexistence between the two areas” (1991:2, my emphasis). The distribution of imported ceramics in Lima society was restricted. Foreign or exotic vessels are reported in burials and public sites, but not in domestic buildings. The available surveys do not report evidence of exotics outside the highest level of public buildings. For example, there are no exotic ceramics reported for the chaupi yunga. Such patterns suggest a restricted and centralized control over long-distance exchange practices by the Lima elite living in the lower valley. If the existence of long-distance exchange networks has a long trajectory, then the participation of Lima elites in this exchange cannot be considered an explanation for political changes that occurred. In general, long-distance exchanges are a common and constant feature in Andean prehistory, not anomalies or unusual characteristics of the Middle Horizon. The explicative power of prestige networks as the primary vectors of political change on the Central Coast needs to be reevaluated.

Lima Elite Practices at the Little Mud-Brick Compound of Pachacamac The previous reconstruction provides evidence of a scenario in which economic intensification and political centralization were at work before any contact with Wari. By the end of the Early Intermediate Period and beginning of the Middle Horizon, elites in the lower valleys had established themselves on the basis of control over agriculture and not through their privileged access

W hat Role D id Wari P lay in the L ima P olitical E conomy?

  143

to exchange networks of exotics goods. These elites likely controlled or influenced a group of decentralized elites in the Lurín Valley. This control was not linked with a distribution of imported Wari ceramics or emulations, but is marked instead with an increased distribution of local Lima ceramics. In the beginning of the Middle Horizon, Warirelated symbols, especially the Nievería style, were associated with the highest level of public Lima life. The most common locations for this style of ceramics are in public contexts, offerings, and burials (Kaulicke 2001). These restricted distributions support the idea that the style was used to further sustain a restricted identity. Meanwhile, the Late Lima style appears in all contexts and sites, including in the newly integrated areas (the chaupi yunga and the neighboring Lurín Valley) where Nievería ceramics are not found. The lone exception to this trend occurs in the Rímac Valley, where Guerrero and Palacios (1994) report evidence that the area close to Cajamarquilla, near the geographical start of the chaupi yunga in the Rímac Valley, could be the location where the Nievería style began to be produced. One of the highly public contexts where Nievería ceramics are found corresponds to feasting activities. Segura (2001) reported activities of ceremonial drinking in one of the political centers of the area—Cajamarquilla. Cajamarquilla is located at the end of the lower section of the Rímac Valley. The site was considered the urbanadministrative center for Middle Horizon in the Central Coast (Segura 2001; Segura and Shimada, this volume). At this site, Segura identified evidence of feasting events where Lima- and a “small” amount of Nievería-style vessels were used to consume chicha (Segura 2001, 2004:101). Segura (2001; Mogrovejo and Segura 2001) identified several contexts where pots in Lima and Nievería styles were intentionally broken, deposited, and sealed in pits after use. These findings prove that the two styles are contemporaneous and show how Nievería ceramics are associated with feasting activities among Cajamarquilla elites. Evidence of similar feasting has been reported for other Lima monumental sites in the Rímac Valley—sites like Maranga and Huaca Pucllana (Flores 1981:68–69; Segura 2004)—which seem to have been elite events. Recent excavations of a portion of a Lima building at 144 

G iancarlo M arcone F.

Pachacamac known as the Little Mud-Brick Compound, however, revealed evidence of a different feasting pattern that encouraged commoner support by providing abundant amounts of commonly consumed foods (e.g., LeCount 2001:935).  

The Little Mud-Brick Compound The Little Mud-Brick Compound is a Lima building located at the periphery of the Pachacamac Sanctuary (Figure 8.2; see Segura and Shimada, this volume). The building was initially excavated in the 1960s by Arturo Jimenez Borja and Alberto Bueno. The results of this work were only partially discussed by the excavators (Bueno 1982), and the ceramic assemblage from the site was reexamined by Lavallée (1966) who determined that the material was closely linked to Maranga ceramics from the Rímac Valley. In 2000 we excavated another portion of the compound. The goals of our excavations were limited to the preservation of the site and the habilitation of new areas to attract visitors. The excavation was designed as an extension of the earlier work carried out by Jimenez Borja and Bueno. The project opened a broad area, approximately 18 m by 34 m (Figure 8.3). The excavation of this building revealed three successive major architectural phases. Each phase represents a reconstruction of the previous structure—re-using the mud bricks or adobitos from the older dismantled building and respecting the same general architectural layout. The sequence is clear—the walls of the previous building were dismantled, the remaining architecture was covered by a layer of sand, and on top of this layer a new floor was built with reused materials for the walls. Each of these phases presents evidence of constant repair, such as floor patch ups and wall fixings, suggesting an intensive use of the compound (Figure 8.4). The ceramics associated with each of these three major architectural phases are similar and correspond to Lima Tardío or Maranga. Based on the ceramic association, the reuse of building materials, and the relative continuity in the architectural layout, I believe that the Little Mud-Brick Compound at Pachacamac was constructed and remodeled in a relatively short amount of time (Marcone 2001).

Figure 8.2   Aerial photo (SAN 1945) showing the location of the Little Mud-Brick Compound in the Pachacamac Sanctuary.

Figure 8.3  The Little Mud-Brick Compound.

Figure 8.4  Schematic profile of the Little Mud-Brick Compound.

The Little Mud-Brick Compound and the Lima Occupation of Pachacamac As a working hypothesis, I proposed elsewhere (Marcone 2001) that the structures (like the Little Mud-Brick Compound) located on the western periphery of Pachacamac were built during the Late Lima phases and represent a growth in the extension and importance of that site. I believe the construction of these peripheral structures likely correlates with the construction and remodeling of Pachacamac’s Old Temple, another public building reported by Strong and Corbett (1943) under the Temple of the Sun and the Urpiwachac Temple tentatively dated to the Late Lima period (Paredes 1991). These contexts do not represent the oldest Lima cultural traits identified at the site, but correspond to a significant change in Pachacamac’s layout and organization, possibly as a result of the consolidation of a multivalley Lima polity at the transition between the Late Intermediate Period and Middle Horizon. Feasting Activities at the Little Mud-Brick Compound An understanding of some of the activities carried out at Pachacamac can be gleaned from our excavations at the Little Mud-Brick Compound. Despite the limited scope of our work, we were able to identify the presence of a ramp in the second architectural phase (a typically Andean feature of public/administrative architecture). This ramp is associated with a lateral room that was almost completely covered by an intensely used hearth 146 

G iancarlo M arcone F.

and a bigger open space that was only partially excavated (Figure 8.5). The ceramics recovered from the hearth, and in the building in general, contain a relatively high percentage of plates and cooking pots and a smaller percentage of jars, bottles, cántaros, and other functional drinking categories (Figure 8.6). The distribution of functional ceramic categories (Table 8.1), the evidence of intensive cooking activities, and the amount of ceramic sherds (in contrast to intentionally broken complete vessels) recovered from the hearth can be used as evidence that activities held at the compound were different than those described by Segura (2001) at Cajamarquilla.6 Another important difference between the two feasting patterns is the almost complete absence of Nievería or other Wari-style ceramics in the Little Mud-Brick Compound (we found just one Nievería-style sherd, and Lavallée’s [1966] ceramics analysis does not report any at all). Several archaeologists recognize that a society may use a variety of feasting activities (e.g., Dietler 2001; Hayden 2001; LeCount 2001). LeCount (2001) for example, distinguishes two broad feasting patterns. According to her, diacritical feastings were limited to those who “command social and economic attention” (LeCount 2001:935) and reflect the strategies used by powerful and wealthy individuals to naturalize social status differences (LeCount 2001:935, following Dietler 1996). In contrast she also identifies a broad category of inclusionary feasting used to support a wide range of activities like patron-client relations and tribute collection. In inclusionary feastings, the intention is not

Figure 8.5  Hearth adjacent to ramp built during the second architectural phase of the Little Mud-Brick Compound.

Figure 8.6  Sample of plates and dish sherds recovered from the Little MudBrick Compound.

Table 8.1  Comparison of formal categories between Cajamarquilla and the Little Mud-Brick Compound Plates and open vessels (%)

Pot (without neck) (%)

Pot and jug (with neck) (%)

Bottles (%)

others/NI (%)

R-105 Cajamarquilla

10.91

7.01

68.15

1.91

12.02

Little Mud-Brick Compound Pachacamac 2000

25.10

15.32

19.81

2.90

36.86

Source: R-105 Cajamarquilla after Segura 2001:Cuadro 15, 69.

 

to solidify social differences but to promote solidarity and equality between community members while at the same time gaining support and corvée labor (LeCount 2001:935–936). This type of feasting has been registered in the Andes as early as the Late Archaic Period (VegaCenteno 2007) and continued into the time of the Inca Empire (Bray 2003; Moore 1989). For the Maya area, LeCount (2001:946) associates restricted drinking events with diacritical feasts that linked elites to the outside world through foreign symbols and big parties with more common food and autochthonous drinking vessels with inclusionary events that brought the greater community together. Similarly, Cook and Glowacki (2003) propose—based on analogies with the Inca Empire—that two different feasting patterns were used at Wari sites. The first was heavily oriented toward elites and linked to higher ratios of drinking vessels, while the other pattern was used to obtain labor and support from commoners. This second pattern was associated with a higher ratio of bowls and more evidence of food consumption (Cook and Glowacki 2003:194–195). Following these studies, the feasting activities as reported in the Little Mud-Brick Compound correspond to a locally based pattern of inclusionary feasting. The high ratio of serving plates and cooking pots and the local Lima iconography found on the vessels support this hypothesis. I argue that the events held at the Little Mud-Brick Compound were related to the consolidation of the Lurín Valley into an overreaching Lima polity at the beginning of the Middle Horizon. Cajamarquilla’s feasting pattern could correspond to what we are calling here diacritical feasts. These celebrations helped to solidify an elite identity centered in the lower Rímac using only partially “international” symbols. Elite power over commoners, however, was solidified through inclusionary feasts like those that took place in the Little MudBrick Compound.

Challenging Assumptions about Pachacamac Since Menzel’s (1964) stylistic study of the Middle Horizon, scholars have commonly believed that Pachacamac was co-opted by Wari in the first phases of the Middle 148 

G iancarlo M arcone F.

Horizon. Through the initial co-option of the site, Wari was able to use Pachacamac’s regional religious importance to impose itself over the region (Menzel 1964; see also Kaulicke 2001), and, in the second part of the Middle Horizon, the site became an important node in the spread of Wari religion throughout Peru. These ideas have sometimes been taken as a given by archaeologists working on the Middle Horizon and used to build pan-Andean scenarios of a strong Wari influence (e.g., Glowacki and Malpass 2003; Isbell 2004). New evidence gathered at Pachacamac (see also Segura and Shimada, this volume), such as the feasting data described earlier, forces us to reevaluate the idea of Pachacamac as a Wari oracle or sanctuary. The first idea in need of reevaluation is the belief that the Wari presence on the Central Coast is based on the ideological cooptation of Pachacamac. In reality, Wari symbols were more likely to appear in secular sites like Cajamarquilla. The association of these labor-oriented feasting activities at Pachacamac with local Lima Tardío/Maranga ceramics, and almost no evidence of Nievería wares, supports a scenario where the control of the population was based on local traditions, while foreign emulated features were used for interelite political activities. Instead of supporting the idea that Pachacamac was a religious satellite of the Wari cult, this evidence suggests that Pachacamac was a locally orientated site. Second, it is necessary to reevaluate the idea that Pachacamac’s political and ideological influence peaked in the second half of the Middle Horizon (Kaulicke 2001; Menzel 1964). These ideas, which are based on the distribution of Pachacamac-style vessels (a style of the second half of the Middle Horizon) in high-status contexts and burials through the central Andean region (Kaulicke 2001), contradict the evidence at the site that shows it first peaked in extension and monumentality between the end of the Early Intermediate Period and the beginning of the Middle Horizon. By the second part of the Middle Horizon, the site had contracted. Lima buildings fell into disuse, and it becomes difficult to identify a significant occupation at the site during the second part of the period (Segura and Shimada, this volume). The Pachacamac style is so-named because the earliest examples of the style were found at Pachacamac. Yet,

there is no evidence that the Pachacamac style was originally from the site or even from the Peruvian Central Coast (Kaulicke 2001). Finally, the interpretations of Pachacamac as a Wari oracle or sanctuary are based on projections from ethnohistoric accounts of how the site functioned. By projecting into the past, archaeologists thus assume the site shared similar functions and importance for both the Wari and the Incas. Yet at present there is no evidence from Pachacamac to support the assertion that it functioned in the same manner during earlier periods (also see Segura and Shimada, this volume).

Discussion It is not unusual to find a diacritical feasting pattern that uses nonlocal canons coexisting with the kind of inclusionary feasting evident at the Little Mud-Brick Compound in interregional interactions. The understanding of these feasting patterns and the emulation of Wari features need to be situated within the local economy in order to address the political role this emulation played in local society. Wari emulation did not play a role in driving social complexity in the region because it first occurs only at the end of the development of the Lima state. The explanations of sociopolitical changes on the Central Coast at the beginning of the Middle Horizon are divided into two polar extremes. I argue that viewing internal-economic models and external-prestige models as antagonistic perspectives only obscures our understanding of each region. As the Lima case shows, these social and political transformations were a gradual process that occurred during the late Early Intermediate Period and Early Middle Horizon. This process likely involved (1) economic differentiation (ranging from craft specialization to staple production to wealth accumulation); (2) elite domination of ritual practice based in local symbols; and (3) elite emulation of the activities and stylistic preferences of elites in Ayacucho and other regions. To prioritize one of these three interconnected power strategies generates a partial and skewed understanding of the political and social transformation that occurred on the Central Coast.

The formation of this multivalley polity was spawned by the progressive intensification of agriculture and the expansion of agricultural land. This intensification allowed for both the centralized management of production by the elites of the lower Chillón and Rímac valleys and the emergence of rural and decentralized elites in the Lurín Valley (Makowski 2002) and the chaupi yunga. It was only toward the end of the Lima sequences that elites in the Lurín Valley were brought into the structure of the Lima state. The inclusion of these rural elites into an overarching political system was apparently sustained by the management of Lima symbols and canons in rituals, art, and feasts, as well as the wide distribution of finely decorated Lima ceramics. Foreign materials, like Wari objects or Wari-related symbols (either imposed or voluntarily accepted), were used only at the highest sociopolitical levels of Lima culture. The Wari-related Nievería style was probably used to define a new eliteness that originated through elites’ interactions in the Rímac Valley. The Wari-related symbols helped to both build the cohesion of lower Rímac Valley elites and separate this group from a second level of rural elites in the neighboring Lurín Valley, who were using Lima symbols. The fact that Nievería-style ceramics only appear in association with the highest level of elites can be explained either in terms of these elites participating in interregional networks or as a result of the indirect imposition of a nascent Wari Empire. These options are not necessarily antagonistic, and the elite adoption of the Nievería style does not change our understanding of the trajectory of this polity on the Central Coast. The role of Wari in the formation of this polity was limited, and the rise of a Lima state cannot be explained in terms of a pan-Andean process. In the end, the focus on panAndean influence obscures a long line of antecedents in local development evident in this area.

Final Conclusions There are three broader conclusions that can be drawn from my discussion of Wari and the development of the Lima polity. First, the selective emulation of Wari features neither implies core control over an interregional

W hat Role D id Wari P lay in the L ima P olitical E conomy?

  149

network (e.g., Stein 2002:903), nor does it necessarily suggest the absence of such control. Wari materials and emulations will be present in regions where Wari had some level of direct administration. Other times these emulations reflect the existence of a pan-Andean exchange network, where relations of dependency between locals and Wari were the catalyst for social changes at the local level. At other times one can find these emulations in a relationship—like those between Wari and polities like Lima, Moche (Chapdelaine, this volume) or Huamachuco (Topic and Topic, this volume)—where there is no evidence that Wari enjoyed political, economic, or ideological power over local groups. Without examining how local production was organized, who controlled exchange, and how feasts were celebrated, we cannot reconstruct power relations within and between societies (Stein 2002:903, 2005). We need to compare the social contexts in which foreign material culture, foreign knowledge, and foreign symbols were used, as well as those contexts where they were not used. Such an analysis makes it possible to determine more precisely the relationship of local groups to broader-scale local processes and interpolity interactions. The second conclusion that can be drawn from this chapter is that there is often an overemphasis on prestige in many of the current explanations that challenge the centralized character of Wari. The Lima case clearly shows that if we construct our explanations focusing only on the basis of elite styles and public contexts, we end up highlighting the international agency of these elites (in their quest for prestige) and neglect the economic foundation of their power. If we focus solely on the economic strategies of these elites, then we tend to overlook the functions that prestige goods could have played in local political developments. We need to develop better models for interregional interaction that incorporate both of these explanations. Finally, this chapter shows that we should be cautious in interpreting the Middle Horizon as a unitary pan-Andean phenomenon. Each case should be examined independently to avoid forcing our interpretations to fit our models. Current models that focus on prestige goods risk overlooking local variations because of the

150 

G iancarlo M arcone F.

central position attributed to Wari in these models. The explanation that Wari created long-distance exchange networks that spurred regional sociopolitical transformations continues to place the weight on external factors in local transformations. In the end, these models may actually reinforce the idea of the centralized pan-Andean Wari sphere that they set out to critique.

Acknowledgments An earlier version of this paper was presented to fulfill a partial requirement for the master’s degree at the University of Pittsburgh. The excavation of the Little Mud-Brick Compound was carried out as an “investment” project on behalf of the Pachacamac Site Museum—Instituto Nacional de Cultura, Peru. I would like to thank Marc Bermann for his help in the master’s paper and Rafael Vega-Centeno, Justin Jennings, Alex Martin, and Mauricio Murillo for their comments at different stages of the development this chapter. Of course, all errors or omissions are the responsibility of the author. Finally I would like to thank Amy for everything, especially her patience.

Endnotes 1. “In Quechua . . . the Peruvian coast was divided into yunga and chaupi-yunga. The yunga, or coast proper, is an arid strip along the ocean, rarely extending inland more than 50 km. Beyond that point, it merges with the chaupi-yunga, a piedmont zone at the base of the Andes. The chaupi yunga is cut by stream canyons supporting trees, shrubs and grasses that are rare to absent in the yunga” (Marcus et al. 1999:6564). Specifically in this paper, I identify the chaupi yunga as the upper part of the lower valley. This area is located between the start of the dejection cone of the coastal river to approximately 1,000 m above sea level. This region is usually proposed as the area of interaction between coastal and highland populations, and their cultural characteristics differ from the ones found in the middle or lower coastal valleys. 2.  It is possible to combine these two perspectives in one explanation. For example, see Covey’s (2000) discussion of Inca presence in Moquegua. 3.  See Flannery (1968), Renfrew (1975), and Helms (1979) for their original formulations of emulation models of interregional interactions. 4.  For the most recent discussion of the Lima culture’s public architecture, ceramic styles, and burial patterns, see Kaulicke 2001.

5.  Media Luna, Cerro Culebra, Copacabana, La Uva, Huaca Santa Rosa, Huarangal, Algodonales, Coyalta, Pv46–923, Pv46– 924, and Playa Grande are some of the sites with public architecture in the lower Chillón Valley. In the Rímac Valley, the best known sites with public architecture are Maranga, Cajamarquilla, Huaca Juliana, Limatambo, and Vista Alegre. 6.  Segura (2001) excavated several contexts in the interior of the room R-105 in the Julio C. Tello Compound in Cajamarquilla. His analysis is based on the MNI (minimum number of individuals) number of complete vessel ceramics. Despite the differences in context and assemblages, I believe that comparisons of the different percentages of formal categories can be taken as relative indicators of different activities at these two sites.

References Cited Agurto Calvo, Santiago. 1984 Lima prehispánica. FINANPRO, Municipalidad de Lima, Lima. Bray, Tamara L. 2003 Inka Pottery as Culinary Equipment: Food, Feasting, and Gender in Imperial State Design. Latin American Antiquity 14(1):3–38. Browman, David L. 1975 Trade Patterns in the Central Highlands of Peru in the First Millennium BC. World Archaeology 6:322–329. Bueno, Alberto 1982 El antiguo valle de Pachacamac: Espacio, tiempo y Cultura. Separata del Boletín de Lima 24:3–52. Burger, Richard L., and Ramiro Matos Mendieta 2002 Atalla: A Center on the Periphery of the Chavín Horizon. Latin American Antiquity 13(2):153–177. Canziani, José 1987 Análisis del complejo urbano Maranga Chayavilca. Gaceta Arqueológica Andina 4(14):10–17 Castillo, Luís Jaime 2007 Las señoras de San José de Moro: Rituales funerarios de mujeres de elite en la Costa Norte del Perú. Summa Humanitatis 1(1):1–15. Ccencho Huamani, José 2001 Alfarería Pucllana propuesta de una metodología de clasificación y algunos aportes para el entendimiento de la cultura Lima. Paper presented at the XII Congreso del Hombre y La Cultura Andina, Ayacucho. Cobb, Charles R. 1993 Archaeological Approaches to the Political Economy of Non-Stratified Societies. Archaeological Method and Theory 5:43–100. 2003 Mississippian Chiefdoms: How Complex? Annual Review of Anthropology 32:63–84. Cohen, Mark 1978 Population Pressure and the Origins of Agriculture: An Archaeological Example from the Coast of Peru. In Advances in Andean Archaeology, edited by David L. Browman, pp. 91–132. Mouton, The Hague.

Conlee, Christina A., and Dennis Ogburn 2004 The Foundations of Power in the Prehispanic Andes: An Introduction. In Foundations of Power in the Prehispanic Andes, edited by Kevin J. Vaughn, Dennis Ogburn, and Christina A. Conlee, pp. 1–12. Archaeological Papers of the American Anthropological Association 14. American Anthropological Association, Arlington, Virginia. Cook, Anita G., and Mary Glowacki 2003 Pots, Politics, and Power. In The Archaeology and Politics of Food and Feasting in Early States and Empires, edited by Tamara L. Bray, pp. 173–202. Kluwer Academic/Plenum, New York. Covey, R. Alan 2000 Inka Administration of the Far South Coast of Peru. Latin American Antiquity 11(2):119–138. Cusick, James G. 1998 Introduction. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology, edited by James G. Cusick, pp. 1–20. Center for Archaeological Investigations, Southern Illinois University, Carbondale. DeLeonardis, Lisa, and George Lau 2004 Life, Death, and Ancestors. In Andean Archaeology, edited by Helaine Silverman, pp. 77–115. Blackwell, Malden, Massachusetts. DeMarrais, Elizabeth, Luis Jaime Castillo, and Timothy K. Earle 1996 Ideology, Materialization, and Power Strategies. Current Anthropology 17:15–31. Dietler, Michael 1996 Feast and Commensal Politics in the Political Economy: Food, Power, and Status in Prehistoric Europe. In Food and the Status Quest: An Interdisciplinary Perspective, edited by Paul Wiessner and Wulf Schiefenhovel, 87–125. Berghahn, Oxford. 2001 Theorizing the Feast: Rituals of Consumption, Commensal Politics, and Power in African Context. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 65–113. Smithsonian Institution Press, Washington, D.C. Dillehay, Tom D. 1976 Competition and Cooperation in a Prehispanic Multiethnic System in Central Andes. Unpublished Ph.D. dissertation, Department of Anthropology, University of Texas. 1979 Pre-Hispanic Resource Sharing in the Central Andes. Science, New Series 204(4388):24–31. Dunn, Robert A. 1979 Early Intermediate Period Ceramic Assemblages from Lurín Valley, Peru. Unpublished Master’s thesis, Department of Anthropology, Temple University, Philadelphia, Pennsylvania. Earle, Timothy 1972 Lurín Valley, Peru: Early Intermediate Period Settlement Development. American Antiquity 37:467–477. Flannery, Kent V. 1968 The Olmec and the Valley of Oaxaca: A Model of Interregional Interaction in Formative Times. In Dumbarton Oaks Conference on the Olmec, edited by E. P. Benson, pp. 79–110. Dumbarton Oaks, Washington, D.C. Flores, Isabel 1981 Investigaciones arqueológicas en la Huaca Juliana, MirafloresLima. Boletín de Lima 13:65–70.

W hat Role D id Wari P lay in the L ima P olitical E conomy?

  151

Franco, Régulo 1993 El centro ceremonial de Pachacamac, nuevas evidencias del Templo Viejo. Boletín de Lima 86:45–62. Franco, Régulo, and Ponciano Paredes 2001 El Templo Viejo de Pachacamac: Nuevos aportes al estudio del Horizonte Medio. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 607–630. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Glowacki, Mary, and Michael Malpass 2003 Water, Huacas, and Ancestor Worship: Traces of a Sacred Wari Landscape. Latin American Antiquity 14(4):431–448. Goldstein, Paul 2000 Exotic Goods and Everyday Chiefs: Long Distance Exchange and Indigenous Sociopolitical Development in the South Central Andes. Latin American Antiquity 11:335–362. Guerrero, Carlos D., and Jonathan B. Palacios 1994 El surgimiento del estilo Nieveria en el valle del Rimac. Boletín de Lima 16(91–96):275–311. Hayden, Brian 2001 Fabulous Feast: A Prolegomenon to the Importance of Feasting. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 23–64. Smithsonian Institution Press, Washington, D.C. Helms, Mary 1979 Ancient Panama: Chiefs in Search of Power. University of Texas Press, Austin. Isbell, William H. 2004 Mortuary Preferences: A Wari Culture Case Study from Middle Horizon Peru. Latin American Antiquity 15(1):3–32. Isla, Elizabeth, and Daniel Guerrero 1987 Socos: Un sitio Wari en el valle de Chillón. Gaceta Arqueológica Andina 4(41):23–28. Jennings, Justin 2006a Core, Peripheries, and Regional Realities in Middle Horizon Peru. Journal of Anthropological Archaeology 25:346–370. 2006b Understanding Middle Horizon Peru: Hermeneutic Spirals, Interpretative Traditions, and Wari Administrative Centers. Latin American Antiquity 17(3):265–286. Jennings, Justin, and Willy Yepez 2001 Architecture, Local Elites, and Imperial Entanglements: The Wari Empire and the Cotahuasi Valley of Peru. Journal of Field Archaeology 28:143–159. Jijón y Camaaño, Jacinto 1949 Maranga: Contribución al conocimiento de los aborígenes del valle del Rimac, Perú. La Prensa Católica, Quito, Peru. Kaulicke, Peter 2001 La sombra de Pachacamac: Huari en la Costa Central. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 313–358. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Knobloch, Patricia J. 1991 Huari and Nieveria: A Reassessment of a Coastal and Sierra Interaction. Paper presented at the 31st Annual Meeting of the Institute of Andean Studies, Berkeley.

152 

G iancarlo M arcone F.

Kroeber, Alfred 1926 The Uhle Pottery Collections from Supe. University of California Publications in American Archaeology and Ethnology Vol 21, No. 6. University of California, Berkeley. Lau, George 2005 Core-Periphery Relations in the Recuay Hinterlands: Economic Interaction at Chinchawas, Peru. Antiquity 79:78–99. Lavallée, Daniele 1966 Una colección cerámica de Pachacamac. Revista del Museo Nacional 34:220–246. LeCount, Lisa J. 2001 Like Water for Chocolate: Feasting and Political Ritual among the Late Classic Maya at Xunantunich, Belize. American Anthropologist 103(4):935–953. Mac Kay, Martin and Rafael Santa Cruz 2001 Las excavaciones del proyecto arqueológico Huaca 20. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 583–595. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. MacNeish, Richard S., Thomas C. Patterson, and David L. Browman 1975 The Central Peruvian Prehistoric Interaction Sphere. Papers of the Robert S. Peabody Foundation for Archaeology Vol. 7. Phillips Academy, Andover, Massachusetts. Makowski, Krzysztof 2002 Power and Social Ranking at the End of the Formative Period: The Lower Lurín Valley Cemeteries. In Andean Archaeology I: Variations in Socio Political Organization, edited by William Isbell and Helaine Silverman, pp. 89–119. Kluwer Academic/ Plenum, New York. Marcone, Giancarlo 2001 El complejo de los adobitos y la cultura Lima en el Santuario de Pachacamac. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 597–605. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Marcus, Joyce, Jeffrey D. Sommer, and Christopher P. Glew 1999 Fish and Mammals in the Economy of an Ancient Peruvian Kingdom. Proceedings of the National Academic of Sciences of the United States of America 96:6564–6570. Menzel, Dorothy 1964 Style and Time in the Middle Horizon. Ñawpa Pacha 2:1–106. Mogrovejo, Juan 1995 Tapicería en la cultura Lima: Un hallazgo textil en Cerro Culebras. Gaceta Arqueológica Andina 24:63–72. Mogrovejo, Juan, and Rafael Segura 2001 El Horizonte Medio en el conjunto arquitectónico Julio C. Tello de Cajamarquilla. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 565–582. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. Montoya, Huayta 1995 Análisis de fragmentaria cerámica excavada en un relleno de clausura. Complejo arqueológico “Huaca Pucllana.” Unpublished licensing thesis, Pontificia Universidad Católica del Perú.

Moore, Jerry D. 1989 Pre-Hispanic Beer in Coastal Peru: Technology and Social Context of Prehistoric Production. American Anthropologist 91(3):682–695. Narváez, Joaquín J. 2006 Una visión general del desarrollo de las sociedades prehispánicas del valle del Rimac. Contextos. Estudios e Investigaciones en Arqueología Peruana 1:111–164. Paredes, Juan 1992 Cerro Culebras: Nuevos aportes acerca de una ocupación de la cultura Lima. Gaceta Arqueológica Andina 6(22):51–62. Paredes, Ponciano 1991 Pachacamac. In Los Incas y el antiguo Perú, 3000 anos de historia, edited by Sergio Purin, pp. 364–383. Sociedad Estatal Quinto Centenario, Barcelona/Madrid. Patterson, Thomas 1966 Pattern and Process in the Early Intermediate Period Pottery of the Central Coast of Peru. Publications in Anthropology 3. University of California, Berkeley. Patterson, Thomas, John P. McCarthy, and Robert A. Dunn 1982 Polities in the Lurín Valley, Peru, during the Early Intermediate Period. Ñawpa Pacha 20:61–82. Renfrew, Colin 1975 Trade as Action at a Distance. In Ancient Civilization and Trade, edited by Jeremy Sabloff and C. C. Lamberg-Karlovsky, pp. 1–60. School of American Research/University of New Mexico Press, Albuquerque. Richardson, James B., III 1994 People of the Andes. St. Remy Press/Smithsonian Institution, Washington, D.C. Schortman, Edward M., Patricia A. Urban, and Marne Ausec 2001 Politics with Style: Identity Formation in Prehispanic South­ eastern Mesoamerica. American Anthropologist 103:312–330. Schreiber, Katharina 1992 Wari Imperialism in Middle Horizon Peru. Anthropological Papers 87. University of Michigan, Ann Arbor. 2001 The Wari Empire of Middle Horizon Peru: The Epistemological Challenge of Documenting an Empire without Documentary Evidence. In Empires, Perspectives from Archaeology and History, edited by Susan Alcock, Terence D’Altroy, Kathleen Morrison, and Carla Sinopoli, pp. 70–92. Cambridge University Press, New York. 2004 Sacred Landscapes and Imperial Ideologies: The Wari Empire in Sondondo, Peru. In Foundations of Power in the Prehispanic Andes, edited by Kevin J. Vaughn, Dennis Ogburn, and Christine Conlee, pp. 131–150. Archaeological Papers of the American Anthropological Association Vol. 14. American Anthropological Association, Arlington, Virginia. Segura, Rafael 2001 Rito y economía en Cajamarquilla. Investigaciones arqueológicas en el conjunto arquitectónico Julio C. Tello. Pontificia Universidad Católica del Perú, Lima. 2004 La cerámica Lima en los albores del Horizonte Medio y algunas notas para el debate. In Puruchuco y la sociedad de Lima: Un homenaje a Arturo Jimenez Borja, edited by Luis Felipe Villacorta, Luisa Vetter, and Carlos Ausejo, pp. 97–118. Concytec, Lima.

Shady, Ruth 1982 Cultura Nieveria y la interacción social en el mundo andino en la época Huari. Arqueológicas 19:5–18. 1988 La época Huari como interacción de las sociedades regionales. Revista Andina 6(1):67–99. Shady, Ruth, and Joaquín J. Narváez 2000 Historia prehispánica de Lima: Arqueología de la Huaca San Marcos. Museo de Arqueología y Antropología de la Universidad Nacional Mayor de San Marcos, Lima. Shimada, Izumi 1991 Pachacamac Archaeology: Retrospect and Prospect. In Pachacamac: A Reprint of the 1903 edition by Max Uhle, edited by Izumi Shimada, pp. X–LXVI. University Museum Monograph 62. University Museum of Archaeology and Anthropology, University of Pennsylvania, Philadelphia. Silva, Jorge 1992 Patrones de asentamiento en el Valle del Chillón. In Estudios de arqueología peruana, edited by D. Bonavia, pp. 395–415. Fomciencias, Lima. 1996 Prehistoric Settlement Patterns in the Chillón Valley, Peru. Unpublished Ph.D. dissertation, Department of Anthropology, University of Michigan, Ann Arbor. Silva, Jorge, Daniel Morales, Ruben Garcia, and Enrique Bragayrac 1988 Cerro Culebras, un asentamiento de la época Lima en el Valle de Chillón. Boletín de Lima 10(56):23–33. Smith, Michael E. 2004 The Archaeology of Ancient State Economies. Annual Review of Anthropology 33:73–102. Stein, Gil J. 2002 From Passive Periphery to Active Agents: Emerging Perspectives in the Archaeology of Interregional Interaction. American Anthropologist 104(3):903–916. 2005 Introduction. In The Archaeology of Colonial Encounter: Comparative Perspectives, edited by Gil J. Stein, pp. 1–29. School of American Research Press, Santa Fe, New Mexico. Strong, William D., and John M. Corbett 1943 A Ceramic Sequence at Pachacamac. In Archaeological Studies in Peru, edited by William. D. Strong, Gordon R. Willey, and John M. Corbett, pp. 27–122. Columbian Studies in Archaeology and Ethnology Vol. 1, No. 3. Columbia University Press, New York. Stumer, Louis M. 1954 Populations Centers of the Rimac Valley, Peru. American Antiquity 20(2):130–148. 1956 Development of Peruvian Coastal Tiahuanacoid Styles. American Antiquity 22(1):59–69. Uhle, Max 1903 Pachacamac: Report of the William Pepper, M.D., LL. D., Peruvian Expedition of 1986. Department of Archaeology, University of Pennsylvania, Philadelphia. Vaughn, Kevin J. 2004 Households, Craft, and Feasting in the Ancient Andes: The Village Context of Early Nasca Craft Consumption. Latin American Antiquity 15(1):61–88. Vega-Centeno, Rafael 2007 Construction, Labor Organization, and Feasting during the Late Archaic Period in the Central Andes. Journal of Anthropological Archaeology 26:150–171.

W hat Role D id Wari P lay in the L ima P olitical E conomy?

  153

Villacorta, Luis Felipe, Luisa Vetter, and Carlos Ausejo (editors) 2004 Puruchuco y la sociedad de Lima: Un homenaje a Arturo Jiménez Borja. Concytec- Minera Buena Ventura, Lima. Von Hagen, Adriana, and Craig Morris 1998 The Cities of the Ancient Andes. Thames and Hudson, New York. Wells, E. Christian 2006 Recent Trends in Theorizing Prehispanic Mesoamerican Economies. Journal of Archaeological Research 14:265–312.

154 

G iancarlo M arcone F.

Willey, Gordon R. 1943 Excavations in the Chancay Valley. In Archaeological Studies in Peru, edited by William D. Strong, Gordon R. Willey, and John M. Corbett, pp. 123–195. Columbian Studies in Ar­­ chaeology and Ethnology Vol. 1, No. 3. Columbia University Press, New York.

Ch a p t er 9

o The Wari State, Its Use of Ancestors, Rural Hinterland, and Agricultural Infrastructure Frank Meddens and Nicholas Branch

W

hile the other chapters in the volume describe the changes that occurred during the Middle Horizon in places outside of the central highlands of Peru, this chapter describes what happened in a valley located relatively close to the Wari heartland. Evidence from the Chicha-Soras Valley indicates that the first appearance of Wari-affiliated sites set in large-scale terraced agricultural systems occurred during Middle Horizon Epoch 2 (Knobloch 1991). The dating of this event is based on the presence of Middle Horizon Epoch 2 ceramic styles, such as Viñaque, Huamanga, and Black Decorated, as well as 14C dates from terrace construction sequences (Branch et al. 2007; Kemp et al. 2006). The Wari expansion into the area was associated with rapid population increase and intensified exploitation of the little-used region. Pollen records from the Chicha Valley indicate a noticeable increase of Chenopodiaceae/Amaranthaceae and the presence of a lesser Zea mays component. These data are likely symptomatic of agricultural terraces used to grow quinoa and, to a lesser extent, maize (Branch et al. 2007; Kemp et al. 2006; Silva 2005). The maintenance of the terraces continued in this area beyond the lifespan of Wari. Since we have discussed the changes in agricultural production elsewhere (Keeley and Meddens 1993; see

also Branch et al. 2007; Kemp et al. 2006), the discussion in this chapter focuses on how the area was administered through the manipulation of ritual practices without the benefit of a Wari administrative center. The management of the Chicha-Soras Valley’s agricultural base appears to have been rooted in a religious political structure emphasizing the importance of kin relations and ancestor veneration. This is evidenced by the early presence of a D-shaped temple structure at the site of Yako, its association with burials, and the establishment of the prominent communal cave tomb of Charrangochayoc. The swift abandonment of Yako at the end of Middle Horizon Epoch 2, as well as that of several minor sites elsewhere in the valley, appears to be coupled with the collapse of central Wari control. This loss of political cohesion is particularly manifest at Yako, a state installation that linked the periphery to the core by manipulating concepts of sacred space. Although abandonment of all occupation sites and the complete disruption of the agricultural infrastructure appears to have been widespread elsewhere in Ayacucho and Apurimac, the disruption of the agricultural infrastructure and wider demography of the Chicha-Soras Valley at the end of Middle Horizon Epoch 2, though indicative of the Wari loss of central control, was relatively limited.

  155

The Chicha-Soras Valley under Wari Control Although a few Middle Horizon 1B sherds have been found in the Chicha-Soras Valley, it appears that habitation sites and the introduction of terrace-supported agricultural systems date to Middle Horizon Epoch 2 (Branch et al. 2007; Kemp et al. 2006; Knobloch 1991). Evidence for dating these remains to this epoch is particularly strong at Chiqnajota and Yako, but a small number of dispersed hamlets, extensive terraced agricultural systems, and tombs on the valley slopes also date to this period.

Middle Horizon Epoch 2 Sites Yako is located on the west bank of the Chicha River, at an altitude of about 3,330 m above sea level and at 14º 11" 03' S, 73º 33" 22' E (Figure 9.1). It measures approximately 0.65 ha with the remains of four structures present (Figure 9.2). Two of these building are almost circular and have diameters of approximately 7 m and 8.5 m. In between these two, the corner of an additional structure can be detected. There are also about fifteen mounds of rubble, broken pottery, and stone tools. The mounds of rubble measure from 5 to 7 m in diameter and stand up to 3 m above the surrounding ground surface. A D-shaped structure is the largest surviving building. It was originally observed during fieldwork carried out between 1978 and 1982 prior to D-shaped structures having been recognized as a distinct type of Wari architecture. When first reported, no conclusions were reached regarding the building’s shape and configuration (Meddens 1985:114, 1991:215). Upon reexamination, some of the structural elements originally recognized proved to be parts of wall collapse rather than original construction. Though the outline of a niche was observed, its state of preservation was so bad that the measurements taken at the time later proved to be incorrect. In 1999 the site was revisited. Continued displacement of the rubble in the intervening years had exposed one niche completely and allowed its measurements to be confidently recorded. The presence of a second has also become clear (Figure 9.3), and we know more about some details of the walls. These new data have allowed Cirilo Vivanco P. and 156 

F rank M eddens and N icholas B ranch

Figure 9.1  Middle Horizon Epoch 2 site distribution in the Chicha-Soras Valley.

Figure 9.2  Plan of Yako.

Figure 9.3  D-shaped structure at the site of Yako.

Ismael Perez C., archaeologists who worked on excavations of similar buildings at Conchopata and Huari, to confirm the identification of the deteriorated structure at Yako as a Wari D-shaped structure. It measures about 13.25 m across, and the curved side of the D faces south. This curved section is best preserved, and of the two niches along its interior face, one measures 0.99 m wide by 1.05 m high and has a depth of 0.80 m. It is separated from the next one by a section of wall constructed of thin, equal-sized horizontal stone slabs (see Figure 9.3). The type of masonry used in the building has been classified as “ordinary masonry” or “ordinary rubblework” at the site of Huari (Benavides 1984:44, Figure VII b, 1991:58). It is comprised of rough fieldstones set in mud mortar with lines of thin, slablike stones with flat faces irregularly spaced throughout the wall, and the walls measure approximately 1.05 m in thickness. Fragments of smooth mud plastering adhere to portions of the wall. The D-shaped structure’s “ordinary

masonry” is only found at Yako in the Chicha-Soras Valley, and the activities carried out inside it would have been for a restricted audience. In 1999 we dug a trench across the full width of a terrace to the southeast of the D-shaped structure. Immediately below the plough zone, a wall was found that was not part of terrace construction. It may have been aligned with the straight front wall of the D-shaped temple structure, in which case it could be part of the foundation of an enclosure wall associated with this building. The assemblage of cultural material recovered consisted of Ocros, Viñaque, Black Decorated, and Huamanga material; obsidian tools consisting primarily of large blades and occasional bifacial points; basalt agricultural tools (such as picks and spades); pebbles; metal-working slag; and animal bone including camelid (llama and alpaca) and deer remains. With the exception of the few Ocros sherds the material derives largely from Middle Horizon Epoch 2B.

T he Wari S tate , I ts U se of A ncestors , Rural H interland, and Agricultural I nfrastructure

  157

The lack of ceramics that could be identified as local or non-Wari in origin or cultural affiliation is notable. The assemblage represents the total transplant of a complete material tradition that appears to arrive fully developed and completely supplanted what may have been there before. The lack of archaeological antecedents for a Middle Horizon population in the region suggests that the region was not intensively exploited. The fact that maize pollen was identified in marsh deposits of Early Intermediate Period date indicates that some agricultural activity at least was taking place, but the associated population appears to have been limited. The remains of an undated irrigation canal that came from the altiplano south of the site may have supplied at least some of the terracing at Yako with water. The largest Middle Horizon settlement site in the Chicha-Soras Valley is at Chiqnajota (see Figure 9.1). Near the confluence of the Pachachaca and Chicha rivers, the 9 ha site is found on the margins of the hamlet of Chicha at an altitude of 3,450 m above sea level (asl). There are some two hundred circular, square, subrectangular, and rectangular standing structures arranged around three sunken plazas and two hills (Figure 9.4). The architecture is rustic in style, and the associated ceramic assemblages demonstrate it remained occupied from Middle Horizon Epoch 2 through the Late Horizon. One of the site’s excavated features consisted of a narrow rectangular room subdivided into small cellular units constructed during Middle Horizon Epoch 2. It measured 12.7 m in length, tapering from 1.10 m to 0.58 m in width at floor level. The walls were constructed of roughly dressed fieldstone covered with a clay plaster. It overlooked the principal plaza from the western side (Meddens 1994). The Charrangochayoc tomb is the most important, best built, and largest of such constructions that date to the Middle Horizon in the Chicha-Soras Valley. It consists of small rectangular rooms, with white plastered walls and rectangular windows (Figure 9.5), that were built in a rock shelter on the western margin of the valley. The tomb is at the base of a cliff of columnar basalt that forms the margin between the surrounding altiplano and the slope down to the valley floor. The remains of

158 

F rank M eddens and N icholas B ranch

Figure 9.4  Isometric reconstruction of the central sector of Chiqnajota (Lu5).

about two hundred individuals were found in this tomb during its first survey carried out in 1979. The site of Tincoq (AP2-18) at an elevation of 3,600 m is on the east side of the Chicha River overlooking its tributary, the Río Yanamayo, from the north. It has some seventy circular structures clustered in groups of three, separated by low fieldstone walls. Pottery from the site is comprised of Middle Horizon Epoch 2B Viñaque material and Middle Horizon Epoch 3–4 Chicha-style ceramics (Meddens 1985). The remaining sites in the valley are smaller and known only by our field numbers. A16 is located on the east bank of the Chicha River just south of Iscayhuanca at 3,445 m asl and is comprised of fifteen mounds of rubble averaging about 3 m in diameter. The mounds are associated with Middle Horizon Epoch 2B material.

Figure 9.5  Facade of the Charrangochayoc tomb (Lu3).

A22 is on the east bank of the Chicha River, north of Chicha at 3,400 m asl. It comprises a single mound with rubble, lithics, and pottery of Middle Horizon Epoch 2 date. LU13 lies on the south bank of the Río Yanamayo about 500 m southeast of Chiqnajota at an altitude of 3,480 m asl. The site consists of two approximately 50 m long north-south-orientated parallel crescent-shaped mounds, each overlaid by four smaller mounds. Some wall foundations are visible. The surface material dates to Middle Horizon Epoch 2B and is dominated by Viñaque material. LU6 occupies the west bank of the Río Chicha at 3,450 m asl. The site is just west of the road from Chicha to Larcay and northeast of the Late Horizon site of Imglesiachayoc. It is on a hill and covers an area of 200 by 300 m. The site is terraced, and there are about twenty mounds of rubble. Two probably later rectangular structures are present. Surface material dates to Middle Horizon Epoch 2. LU10 is found west of Chicha and south of Imglesiachayoc. The site is comprised of a 3 m wide causeway defined by two double-faced, rubble-filled retaining

walls. These walls run across a dry lake bed and are about 1 m high for much of their length before fading out on the southern end. Pottery from the rubble fill dates to Middle Horizon Epoch 2, but it is possible that this material is residual, having been redeposited as a result of the site’s construction.

The Environmental and Climate Evidence The period of Wari expansion in the southeastern sector of the Department of Ayacucho was linked to the construction of large-scale systems of agricultural terracing. Pollen-stratigraphic evidence for the cultivation of Zea mays during the Middle Horizon in the ChichaSoras Valley is limited, with concentrations of maize pollen being generally low for both the Tocsaqocha and Ayapampa marsh sequences (Branch et al. 2007; Kemp et al. 2006; Silva 2005). Carbonized maize cobs were found at the site of Chiqnajota in direct association with a Middle Horizon Epoch 2 offering deposit made up of Huamanga-style pottery vessels, obsidian flakes, and artifacts and a large deposit of carbonized material

T he Wari S tate , I ts U se of A ncestors , Rural H interland, and Agricultural I nfrastructure

  159

Figure 9.6  Selected pollen taxa diagram for Ayapampa (Tocctoccasa) peat bog (marsh). The core samples were recovered from the peat bog using a Russian D-section peat sampler. The pollen grains were extracted using conventional methods, involving dispersal in sodium pyrophosphate (1 percent), sieving through 200 µm and 10 µm mesh sizes, oxidation of organic matter using acetic anhydride and sulfuric acid (ratio 9:1), and heavy liquid separation using sodium polytungstate (2 g/cm3). The pollen count consisted of approximately

(Meddens 1985, 1991). These data suggest that maize may have been more prevalent in the valley, and the poor representation of maize in the pollen record could be due to taphonomic variables, such as pollen productivity and dispersal distance. On the other hand, maize may not have been a dominant crop in the valley at this time. This alternative explanation is more likely based upon the high pollen-stratigraphic values of Chenopodiaceae/ Amaranthaceae, which suggest that quinoa cultivation formed the main focus of plant-based subsistence practices (Figure 9.6). Most terrace construction and Zea mays and Chenopodiaceae/Amaranthaceae cultivation occurred during the Middle Horizon and Late Intermediate Period. Both Tocsaqocha and Ayapampa record a complex lithostratigraphy that provides evidence for fluctuating water levels that in some cases may be attributed 160 

F rank M eddens and N icholas B ranch

three hundred pollen grains and spores. The data are expressed as a percentage of total pollen, with only the most important taxa illustrated. The y-axis displays depth, in centimeters from ground surface, while the x-axis displays the name of the selected pollen taxa and percent values. The pollen diagram illustrates variations in pollen percentages for each taxon over time (shown as calibrated radiocarbon ages). The organic matter values are expressed as a percentage and were determined by the loss-on-ignition (thermal oxidation) method.

to changes in the hydrological status of the mire due to human activity. At the Tocsaqocha marsh, one event was characterized by the deposition of a narrow clay band that coincided with peaks in Chenopodiaceae/ Amaranthaceae and Zea mays pollen (see Figure 9.6). This reflects an erosion episode, which occurred between cal 210 BC–AD 70, 2075 ± 60 BP (WK-12224; peat; δ13C = -28.2‰) and cal AD 1150–1330, 755 ± 64 BP (WK-12223; peat; δ13C = -27.0‰) (all 14C dates are calculated at 2σ). Terrace construction adjacent to the mire was radiocarbon dated to cal AD 615–695, 1368 ± 25 BP (SUERC-1531; charcoal; δ13C = -25.2‰) from charcoal obtained from within the terrace palaeosol. This suggests that the erosion event evident in the clay band is probably Wari in age. These dates are calibrated to the southern hemisphere radiocarbon calibration curve (McCormac et al. 2004).

The erosion of the slopes surrounding the basin appears linked to a rise in water level. The stage associated with the erosion episode spans a period of wetter, colder climate and then drier, warmer climate, approximately AD 600–650 and AD 650–700 in the Quelccaya ice-core records (Thompson et al. 1985). Though the date range for the deposits that include the erosion event is unfortunately broad, on balance, we consider the erosion most likely to be associated with the period of terrace construction and the trend toward drier, warmer conditions. We argue that the water levels were artificially raised to mitigate the regional drought, at a time when efforts toward cultivation on the surrounding terracing were intensified (Silva 2005). This interpretation is supported by evidence for human control of the mire hydrology, in the form of stone walls constructed across the drainage outflows, at both Tocsaqocha and Ayapampa. At both marshes the irrigation canals feeding the terracing downstream from the marsh pick up water from immediately below the walls across the drainage outflows. It has been suggested that prior to the Late Horizon, the maize crop in ancient highland Andean societies was limited to ceremonial and hospitality uses rather than being a staple in the diet (Finucane et al. 2006; Murra 1975:57). By the Late Horizon the Inca army had maize as its preferred ration (Murra 1975:56), and maize could be considered a staple crop. Recent stable isotope work on human, camelid, and guinea pig remains of Middle Horizon date from the site of Conchopata in the Wari heartland suggests that maize was also the mainstay of their diet (Finucane et al. 2006). Within the camelid population, there were two distinct dietary groups. The first group was foddered predominantly on maize or maize stalks whereas the second clearly had been limited to puna grazing. With the trivial amount of paleoenvironmental evidence for maize cultivation from the Cuzco, Apurimac, and Ayacucho hinterland for this period, this begs the question of whether maize consumption within the Wari state was limited to selected groups of people and animals in the Wari heartland, rather than being a cross-population staple. The dramatic establishment of occupation sites and the intensification of agricultural terracing during

Middle Horizon Epoch 2 in the Chicha-Soras Valley are indicative of the arrival of a new population, probably colonists, and are clearly linked to changes associated with the expansion of the Wari. The expansion of the Wari state in the Chicha-Soras Valley is empirically demonstrated by its expansion into an area with low population density and by its introduction of terraced agricultural infrastructure, habitation sites, and road construction.1 No cultural remains dating to the Early Intermediate Period are known from the area, but pollen data from the Tocsaqocha and Ayapampa mires on the west and east sides of the river, respectively, provide evidence for maize cultivation in the Early Intermediate Period (see Figure 9.6). Wari had little difficulty in establishing a colonist population under its direction in the Chicha-Soras Valley.

Abandonment and Continuity The abandonment of Yako and some of the dispersed hamlets suggests that major changes occurred in the Wari socioreligious complex at the end of Middle Horizon 2. The abandonment of Yako at this time is shared by Wari sites in the Ayacucho, Sondondo (Schreiber 1987), and Qaracha valleys (Valdez and Vivanco 1994). Yako’s desertion may be related to its nature and role as a state installation. This abandonment, however, did not mean that everyone fled the valley. On the contrary, based on ceramic data and stratigraphic evidence, some sites continued to be occupied, and there appears to have been a significant increase in the number and size of occupation sites in the valley (from five in Middle Horizon 2 to nine in Middle Horizon 3–4; Meddens 1985) (Figure 9.7). Unlike in the Ayacucho, Sondondo (Schreiber 1987), and Qaracha valleys (Valdez and Vivanco 1994), the political events at the end of Middle Horizon 2 in part bypassed the Chicha-Soras Valley. Gradual changes in pottery style and iconography during Middle Horizon 3–4 and into the Late Intermediate Period show that the disappearance of Wari central authority had little immediate effect on the cultural integrity of the Chicha-Soras area, and there was no apparent rejection of the value structures of the preceding period since sites like the burial tombs at Charrangochayoc continued to be used.

T he Wari S tate , I ts U se of A ncestors , Rural H interland, and Agricultural I nfrastructure

  161

administrative center and several subsidiary Wari installations were constructed more or less at the same time. How then did Wari manage to extract resources from the Chicha-Soras Valley without imposing an aggressive Wari footprint like the one found in neighboring Sondondo? We argue that the management of ChichaSoras’s agricultural base appears to have been rooted in a religious political structure that emphasized the importance of kin relations and ancestor veneration of a group of Wari colonists who on balance are not likely to have originated in the Ayacucho Valley. To make this argument, we will first describe the human remains found in valley and elsewhere during the Middle Horizon and then discuss how these were used in ancestor worship.

Middle Horizon Human Remains in the Chicha-Soras Valley

Figure 9.7  Middle Horizon Epoch 3–4 site distribution in the Chicha-Soras Valley.

Although the loss of Wari central authority likely led to the abandonment of Yako because the site served as a state installation, the local elite managed to continue to build on the existing agricultural infrastructure during a phase of extensive remodeling during the Late Intermediate Period (Branch et al. 2007; Kemp et al. 2006).

Human Remains, Ancestor Worship, and Wari Administration of the Chicha-Soras Valley The data presented earlier provide strong evidence for agricultural intensification during the Middle Horizon in the Chicha-Soras Valley, which was directed by the Wari Empire. Yet, the evidence for direct administration of the valley is weaker than that for the neighboring Qarwarazu-Sondondo Valley where a 15 ha Wari 162 

F rank M eddens and N icholas B ranch

At Yako a small 2 × 4 m trench was dug in wall collapse against the exterior face on the southeast side of the D-shaped temple. The foundation of the D-shaped structure was built directly on bedrock without a foundation trench. Immediately below the wall collapse, we found an offering in a small pit against the outside of the wall that contained plain vessels in the Huamanga style. The rubble to the north of the offering included the disturbed, incomplete skeletal remains of at least six individuals. These were so near the surface that it was not possible to be certain whether they had been redeposited in field clearance, looting, or had originally been placed inside the walls of the D-shaped building. The latter appears more likely as human remains have also been found associated with D-shaped structures at Huari, Conchopata, and Ñawinpukyo in the Ayacucho Valley (Meddens and Cook 2000; Tung and Cook 2006). Secondary burials are also known from within the walls of Viracochapampa and Huamachuco in the Cajamarca region (Topic 1985; Topic and Topic, this volume). No evidence of human remains was observed anywhere else at Yako. Three of the individuals from Yako were male, and all were adults with the youngest individual about nineteen years of age. One of the skulls had a deformed vault and included a circular frontal trepanation; no artifacts of any kind were associated. One of the skeletons

was virtually complete, and this completeness suggests that either this body was not moved any great distance from its original location of deposition or that it was sufficiently intact as a result of mummification of the corpse when it ended up in the wall collapse. At Chiqnajota there is a stepped, irregular platform on the same alignment as the cellular building on the western margin of the plaza. The platform was likely subrectangular in shape originally and would have measured 26.0 × 8.0 × 2.6 m. Fragmentary skeletal remains of a human adult male were located in a shallow cut between the platform and the terrace wall delimiting the eastern limit of the plaza. Predominately associated with Middle Horizon Epoch 3–4 sherds, the platform may be an antecedent to the later Inca ushnus that were used in a political religious context to enable rulers to mediate between their populations, deities, and ancestors. Ethnohistorical sources link these later Inca platforms with human sacrifice and community gatherings involving public generosity and redistribution on the part of the state (Meddens 1997; Meddens et al. 2008; Zuidema 1989). The Late Intermediate platform might owe its genesis to an earlier ceremonial focal point at the site. One of the excavations at Chiqnajota focused on a mound of broken pottery of Middle Horizon Epoch 3–4 date. Underlying this was an offering deposit dating to Middle Horizon Epoch 2B. The offering surrounded two sides of a large pit-marked boulder of volcanic tuff interpreted as a huaca. It comprised pottery vessels, obsidian tools, and a sequence of ash layers, potsherds, and carbonized material in a deep pit. This excavation also uncovered remains of a 5.6 m diameter circular structure and the southeast side of a smaller 1.6 m diameter building that abutted the circular structure. We found a pair of adult human legs side by side in the foundation cut of the smaller of the two circular structures. The ceramic styles found in the Charrangochayoc burial cave include Viñaque, Huamanga, and Black Decorated wares as well as a local Chicha-style material. The latter becomes dominant during Middle Horizon Epoch 3–4 after Wari state power wanes (Meddens 1985, 1991). The human remains present included individuals with cranial deformation similar to the type found at Yako and characteristic of forms found in the Nazca

area. Fragments of rope from mummy bindings and textile fragments were also common. Preservation was good enough for some body parts to still have skin and hair surviving. The valley-wide importance of the burial cave is demonstrated in a number of ways. The curved wall of the D-shaped temple at Yako was orientated toward Charrangochayoc, and the tombs are close to the drop point where the irrigation canal that likely provided water to the site of Yako came into the valley (Meddens and Cook 2000). Furthermore, Charrangochayoc’s location was widely visible across the valley since the dark mouth of the cave stood out at the interface between the columnar basalt rock face at the top of the valley and the gradually eroding terraced volcanic tuff slope descending to the valley floor. The whitewashed plaster face of the tomb itself would have further increased the tomb’s visibility as at certain times of the year its interior would have been lit up by the sun rising in the east.

The Disposal of Middle Horizon Human Remains and the Issue of Ancestor Worship Disposal of the dead and manipulation of human remains during the Middle Horizon took many forms, likely served a range of purposes, and have been linked to a variety of beliefs. In order to better understand Charrangochayoc and its role in the Chicha-Soras Valley, we need to first better define those aspects of Middle Horizon burial ritual that were linked to ancestor veneration and worship. In a broad sense the concept of ancestor veneration or worship as employed here is based on the transmission of ancestral rights and privileges to resources and the development of inequalities in access to these privileges and resources. Following McAnany, ancestor worship can be defined as traditions comprising “rituals and practices surrounding the burial and commemoration of apical ancestors of kin groups” (McAnany 1995:11). Ancestor worship can be recognized in protracted burial rites such as the mummification and preservation of the remains of the deceased, the repeated access to and feeding of, or sacrifice to, these physical remains, and the participation of ancestral relics in the

T he Wari S tate , I ts U se of A ncestors , Rural H interland , and Agricultural I nfrastructure

  163

rituals and ceremonies of the living (McAnany 1995:10– 18). It should be noted that during the Late Horizon even the remains of relatively young members of kin groups could in certain contexts be viewed as being part of the world of the ancestors, such as in the Capac Hucha ritual (Meddens et al. 2008:318). Ancestor worship has long been recognized and accepted as a major aspect of Late Horizon and colonial native belief systems. It is an indigenous tradition in the region and may have deep roots. In the Middle Horizon, the treatment of some of the dead suggests the existence of belief structures that were similar or related to those held during the Late Horizon. One example of a similar practice is the use of the tocco element in some cists at Conchopata. This appears to have served a function of enabling the living to provide offerings or sacrifices to the dead (Tung and Cook 2006). Within the Middle Horizon, however, there were critical differences between how the dead were treated in the Department of Ayacucho during the period that can be used to better understand the relationship between the people living in Chicha-Soras and the city of Huari.

Evidence for Communal Cave–Type Burial Sites in the Department of Ayacucho One of the ways in which people were buried during the Middle Horizon was in communal cave burials. Since little work has been carried out on the cave and rock shelter burials in the Ayacucho and Apurimac regions, there are few comparisons for the Charrangochayoc cave in the Chicha-Soras Valley. Nonetheless, Isbell describes two similar sites at Huanca Sancos and Sarhua (1997:183– 184). He defines them as “Open Sepulchre” type tombs, and the associated cultural material appears to date from Middle Horizon Epoch 2 to the Late Horizon. A limited survey by Jo Hurst and Cirilo Vivanco Pomacanchari of the Cruz Muqu 1, 2, and 3 burial populations in caves at Sarhua concerned a population of between 140 and 180 individuals distributed over these three caves (Hurst 2007). The remains reflect a “normal” rural settlement cemetery age and sex distribution with approximately half of the individuals showing evidence of cranial deformation, mostly of the frontal-occipital type (Figure 9.8). 164 

F rank M eddens and N icholas B ranch

The dental health of the Cruz Muqu 1 group was poor in comparison with that of the individuals in Cruz Muqu 2 and 3. There was no evidence of antemortem trauma or fractures to the bones except for one healed wound on a female cranium, caused by a pointed object. There was no evidence of dietary deficiencies or osteoporosis. Some of the vertebrae showed evidence of pathological changes, including osteophytes and porosity indicative of damage caused as a result of regularly carrying heavy loads (Hurst 2007). The cranial deformation in the Sarhua material is the same as that found in individuals from the Charrangochayoc site and at Yako. This imperfect evidence available for the populations found in the open sepulchre–type tombs in the Ayacucho and Apurimac areas suggests that these sites represented large communal tombs used by a broad cross

Figure 9.8  Cruz Muqu burial population (photo by Joanna Hurst).

section of the local population. Wari tapestries and fine textiles from these sites demonstrate access to prestige goods, but the presence of coarser fabrics and animal skin wrappings suggests that lower status individuals were also placed in the tombs (Isbell 1997:183–184; Meddens 1985, 2001:207). The open sepulchre–type tomb has a restricted distribution that is limited to the south of the Departments of Ayacucho and parts of Apurimac. Neither this tomb type nor the frontal occipital types of head deformation have been identified in the Wari heartland (W. H. Isbell, personal communication 2008). Although this form of cranial deformation and similar mummification practices have been recognized in contemporary burials from the Nazca area (Conlee, this volume), the dead are not buried in tombs of the open sepulchre type here. This admittedly limited evidence suggests that the Wari colonists of the Chicha-Soras Valley may not have originated in the Ayacucho Valley.

Evidence for Wari State Burial Practices in the Department of Ayacucho and Elsewhere There is a bewildering array of methods for disposing of the dead and manipulation of human remains at Wariaffiliated sites. Our best understanding of this variation comes from the site of Conchopata in the Ayacucho Valley where eight distinct types of burial were recognized by Isbell, excluding the disposal of defleshed bodies on temple floors. These include individual and multiple interments in unlined pits; cist interments in cylindrical stone-lined pits frequently with a ttoco or offering passage or opening (Isbell 2004:9); bedrock cavity tombs used for multiple interments over periods of time; mortuary room interments with multiple circular or rectangular stone-lined cists with the access lid or plug covered by a small mortuary house; chambers cut into thick walls for multiple interments; and communal sacrificial group burials and multiple story and chamber-cut stone ashlar tombs with multiple burials (Isbell 2004:6). Ochatoma recognizes several more variations in burials at the site. One variation involves the placement of one or more infants’ or children’s skulls, often covered with a bowl or cup, in pits under walls. Another burial type is children and infants buried inside large

reused pottery vessels that were placed in pits. The single example of cremated human remains at Conchopata was also found in a pit in a patio (Ochatoma 2007:13–53). Secondary burial and the manipulation of remains is a common feature of these burials (Isbell 2004; Ochatoma 2007), and interaction with the dead was a vital aspect of Wari culture (Ochatoma 2007:54). This suggests that the dead entered the world of the living in an ever-present manner, in which they were inseparable from the world of the living. Possible provision of offerings to the dead subsequent to interment suggests that a form of ancestor veneration was part of this particular variant of disposal of the dead. Tung and Cook (2006) conclude in their discussion of Conchopata that Wari rulers either tolerated diversity or that the variety of disposal of the dead practices demonstrates conflict over the rituals associated with them. Hints of this diversity are also seen at Wari-affiliated sites in the periphery. The niched hall, for example, forms an integral element of the two largest Wari peripheral compounds, Viracochapampa and Pikillacta. This architectural form has been interpreted as having functioned as a “religious or ceremonial building related to ancestor worship” (McEwan 2005:149–158). Such an interpretation makes ancestor worship a core function of the operation of these key Wari administrative sites. It should be noted that the human remains from these contexts were found in secondary contexts. At Pikillacta, a cache of human skulls was found in a pit cut into a niched hall’s plaster floor (McEwan 2005). The heads may therefore not be related to the primary function of the building. Similarly, the human remains at Viracochapampa, and likewise at Huamachuco, comprised human body parts in a secondary context where defleshed bones appear to have been immured in the corners of the walls of the structures themselves (Topic 1985:148). Though heads have been directly linked to ancestor worship across time in the Andes (Arnold and Hastorf 2008), the link between these human remains and possible ancestor rituals in the halls can certainly be questioned. Human remains have also been found at other Wari-affiliated sites. At the rectangular enclosure of Azángaro, a shaft tomb was associated with an elite sector of the site as well as with drainage features. This link

T he Wari S tate , I ts U se of A ncestors , Rural H interland, and Agricultural I nfrastructure

  165

with water was also a characteristic of a further burial, near a water conduit. These tombs were interpreted as related to ancestor worship and possibly sacrifice (Anders 1986:616–621, 737–738, 924–931). At Jincamocco, a single human skull with evidence of trepanations belonging to an elderly female was found in an offering context in a bench on the inside of a large patio in the enclosure (Schreiber 1992:182–184). Further human remains found here comprised a multiple burial of two adult individuals in flexed seated positions in a small cist found against the wall of an earlier building underneath a patio floor (Schreiber 1992:249–250). The D-shaped temple structures found at Huari at Cheqo Wasi and Vegachayoc Moqo, as well as those found at Conchopata and Ñawinpukyo in the Ayacucho Valley, are associated with human remains, tombs, and burials (Cook 2001; Meddens and Cook 2000). One D-shaped temple at Conchopata included a cache of disarticulated human heads (Cook 2001:147). These comprised six burned skulls with holes in them, suggesting that they originally had cords attached, similar to the Nasca “trophy head” tradition. These were found arranged within a semicircle of stones and covered by the articulated skeletal remains of a juvenile camelid (Ochatoma 2007). Similar associations between D-shaped structure, burials, and offerings occur throughout Peru (Cook 2001; Meddens and Cook 2000; Ochatoma and Cabrera 2001:61–63; Tung and Cook 2006; Williams 2001:71–74). The association of human remains with D-shaped temples and niched halls may relate to a long-standing ritual tradition involving animated disembodied heads. Human heads and skulls are frequently represented on Middle Horizon Wari ceramics, notably in the Viñaque and Atarco styles. The earlier south coastal Nasca tradition that influenced the development of the Wari styles also included belief systems involving human “trophy heads.” Furthermore, present-day Andean cultural concepts include a series of myths concerning disembodied animated human heads such as cabezas voladores (flying heads) and condenados (spirits of the condemned). These represent malevolent beings that wander around at night and are capable of taking the spiritual essence of unsuspecting people (Ansión 1987). Reported practices include a reference by Hocquenghem (1979:90) that in Cusco the 166 

F rank M eddens and N icholas B ranch

skull of an ancestor might be kept in a niche in the house for the safekeeping of the household. Andean messianic concepts include a belief that a new body may grow from a head, which constitutes the regenerative part of the body (B. J. Isbell reported in Anders 1986:931).

Returning to Chicha-Soras What is clear from the above is that the practices associated with disposal of the dead and manipulation of human remains in the Chicha-Soras Valley are not equivalent to practices at Huari and peripheral Wari centers. The human remains found around Yako’s D-shaped structure fit well within the practices commonly found at Wari sites. Yet, the valley’s most visible monument for both the ancient occupants of the valley and the modern visitor were and are the tombs of Charrangochayoc. Since this tomb type is clearly not associated with the immediate core area of Wari power in the Ayacucho Valley itself, the most important sacred site in the valley is not, in a strict sense, Wari. Future research, especially isotopic studies of human bone, may help us better understand the relationships between Huari, the Wari state, and the residents of the Chicha-Soras Valley. One possibility is that people were moved into the valley by the state and brought their ancestors with them in a pattern similar to the relocation of mitimae groups by the Inca.

Conclusions The expansion and incorporation of the southern part of the Department of Ayacucho into the Wari state may have been part of a wider extension of Wari control into the neighboring Department of Arequipa during Middle Horizon Epoch 2. This expansion has been interpreted as a response of Wari to a prolonged period of drought whereby the state attempted to gain control over sustainable arable and grazing lands and water sources and cosmological power over water (Glowacki and Malpass 2003). Wari influence, if not Wari administrative control, has been documented in many of the highland valleys in this region (Jennings, this volume), and the

paleoenvironmental evidence from ice core and paleoecological records suggests that the extension of the Wari state into southern Ayacucho and the sierra of Arequipa corresponds to a comparatively dry and stable period in Andean prehistory. However, correlations between the terrestrial proxy data and ice core records are somewhat hindered by both the absence of precise chronological models for each sampled marsh sequence and the difficulty of distinguishing climatic fluctuations from human impact in the paleoecological records. Whatever caused the establishment and eventual decline of the Wari-managed population in the ChichaSoras area; it was unlikely to be directly related to shortterm climate change. Expansion into an underutilized agricultural area to maximize agricultural potential by introducing large-scale terraced agriculture would have been a relatively low risk investment for the Wari state. Wari’s purpose in the Chicha-Soras Valley would have likely been to create agricultural resources to be transferred to the larger state facilities located at Jincamocco, some 30 km to the west. The state’s relationship with pastoral activities in this region has been more difficult to demonstrate. Since there is strong evidence for use of camelid wool in Middle Horizon Wari textiles, camelids in Middle Horizon Wari iconography, and the presence of significant quantities of llama and alpaca remains in Middle Horizon animal bone assemblages, it is clear that camelid husbandry was of great importance in the Wari state. Limited survey work in the surrounding altiplano both east and west of the Chicha-Soras Valley, however, has picked up evidence for significant animal husbandry complexes only of Late Intermediate/Late Horizon date. The only possible evidence of Middle Horizon activity is a single fragment of a Huamanga-style face neck jar retrieved from the important Late Horizon Shrine at Usccunta beyond Qarwarazu to the west. Evidence for Wari pastoral activities in the grazing lands of the Arequipa altiplano has also been elusive (M. S. Ziolkowski, personal communication 2007). The seeming contradiction between the importance of camelid products in the state and the lack of state-sponsored pastoralism could be resolved through a better understanding of local ceramic chronologies. There is some

evidence, for example, of Middle Horizon herding practices in the altiplano of the upper Colca Valley where local pottery and obsidian-working debris are associated with seasonal transit camps of specialist herders (Tripcevich 2007:621, 640, 645, 658, 782, passim; Willy Yépez, personal communication 2007). The presence of Wari sites like Tajra Chullo at ap­­ proximately 4,100 m in altitude in the Virginiyoc Valley in the Department of Cusco demonstrates that the Wari state was perfectly capable of exercising and projecting a presence in the high altitude pastoralist eco-zones (Meddens 1989). The lack of evidence for its exploitation of this resource on the altiplano surrounding the Chicha-Soras area and elsewhere therefore suggests that the state chose not to directly manage these resources. Instead, Wari’s priorities were in agricultural exploitation and, in at least the case of the Chicha-Soras Valley, focused firmly on the production of Chenopodiaceae/ Amaranthaceae (quinoa) and to a significantly lesser extent Zea mays (maize) (Branch et al. 2007; Kemp et al. 2006). If maize constituted a crop reserved for specific limited sections of the Wari population rather than a staple for the masses, then its production, storage, and redistribution would not be logistically complicated or costly. It appears that the large-scale open sepulcher–style tombs, such as the ones at Charrangochayoc and Cruz Muqu 1, 2, and 3, may have served a distinct function as large communal tombs to set up the foundation of a publicly visible, kin-based ancestor veneration tradition, around which a local, Wari colonist–based political structure could function. In one sense, the political focus of the Wari state in the Chicha-Soras area was the site of Yako with its D-shaped structure and closer allegiance to Wari state practices. Yet, Charrangochayoc was perhaps more important politically, based on its high visibility, communal characteristics, and the demographic breadth of its mummified remains. Yako, after all, was oriented to face the Charrangochayoc tomb, and the importance of the cave site appears to be rooted in the generation of an agricultural surplus. The Wari state, in this case, seems to have invested in maintaining, if not strengthening, kin/ancestor religious practices that were different than those practiced at Huari and other Wari state sites.

T he Wari S tate , I ts U se of A ncestors , Rural H interland , and Agricultural I nfrastructure

  167

The visibility of Charrangochayoc and other open sepulcher tombs and the rituals associated with them are particularly important in a pre- or protoliterate cultural context, where such landscape referents functioned as a daily reminder of the fundamentals of social interaction and tradition. Within an environment where surpluses can be plentiful and stored assets can be used as buffers against crop failures, traditional means of food sharing ensure political stability. This is particularly important when frequent unpredictable climatic oscillations and seismic events can have a dramatic impact on available food stocks and infrastructure. Rapidly fluctuating climatic events appear at least in part to be a consequence of the cyclical El Niño Southern Oscillation (ENSO), which results in dramatic changes in precipitation between the coast and the mountains. The limited distribution of the open sepulcher tomb type and its continued use in subsequent periods demonstrate that even in an area that appears to have been under direct control of the Wari state there existed room for considerable social and cultural variation from what was the norm at the center of state power (Tung and Cook 2006). .

Acknowledgments The fieldwork was carried out between 1979 and 1982 and received support from the Cusichaca trust in 1999 and 2000. The Natural Environment Research Council provided backing, and we are grateful to the Instituto Nacional de Cultura for facilitating the work. The authors wish to thank Ann Kendall for her encouragement; George Bankes for his thoughts on the text; Bill Isbell for his considered comments on, and discussion of, an earlier version of this paper; and Justin Jennings for his invitation to participate in the SAA symposium and for his editorial input. We are also grateful to Jenny Kynaston for the maps and Bert van de Hee for the isometric illustration. We thank the people of Pampachiri, Larcay, and Chicha, without whose support and friendship the work would not have been possible.

168 

F rank M eddens and N icholas B ranch

Endnotes 1.  There is no evidence for the storage of surpluses in the Chicha-Soras Valley, and indeed the lack of these facilities at Middle Horizon Wari sites has been a subject of debate. Although structures in architectural plans can sometimes be interpreted as serving a storage function, these buildings, with few exceptions, have failed to conform to a storage interpretation upon excavation (Glowacki 1996; McEwan 1991, 1996; Schreiber 1987). A few of the cellular structures associated with patio groups at Pikillacta appear to have been used for the storage of foodstuffs, although others appear to have been used in food preparation (McEwan 2005:152). Potential storage structures of Middle Horizon date investigated at Raqchi have proven to have served a habitation function (Bill Sillar, personal communication 2007). The cellular architecture we know from most of the rectangular enclosure–type sites was manifested as multistory structures, though the upper stories do not survive. These could have served as well-ventilated storage facilities. Indeed the architectural model found at Conchopata (Isbell 2001:Figure 20) as well as a possible multistory structure illustrated on an oversize offering vessel (Isbell 1977:49, Figure 7) have features that may be interpreted as ventilation vents. If the upper stories of these compounds served a storage function, then extensive facilities would have been available.

References Cited Anders, Martha B. 1986 Dual Organisation and Calendars Inferred from the Planned Site of Azangaro—Wari Administrative Strategies. Un­­ published Ph.D. dissertation, Department of Anthropology, Cornell University, Ithaca, New York. Ansión, Juan 1987 Desde el Rincón de los Muertos, el pensamiento mítico en Ayacucho. Gredes, Lima. Arnold, Denise Y., and Christine A. Hastorf 2008 Heads of State, Icons, Power and Politics in the Ancient and Modern Andes. Left Coast Press, Walnut Creek, California. Benavides Calle, Mario 1984 Caracter del estado Wari. Universidad Nacional de San Cristóbal de Huamanga, Ayacucho, Peru. 1991 Cheqo Wasi, Huari. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 55–69. Dumbarton Oaks, Washington, D.C. Branch, Nicholas P., Rob A. Kemp, Barbara Silva, Frank M. Meddens, Alan William, Ann Kendall, and Cirilo Vivanco Pomacanchari 2007 Testing the Sustainability and Sensitivity to Climatic Change of Terrace Agricultural Systems in the Peruvian Andes: A Pilot Study. Journal of Archaeological Science 34:1–9. Cook, Anita G. 2001 Huari D-shaped Structures, Sacrificial Offerings and Divine Rulership. In Ritual Sacrifice in Ancient Peru, edited by Elizabeth P. Benson and Anita G. Cook, pp. 137–164. University of Texas Press, Austin. Finucane, Brian, Patricia M. Agurto, and William H. Isbell 2006 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 33:1766–1776. Glowacki, Mary 1996 The Wari Occupation of the Southern Highlands of Peru: A Ceramic Perspective from the Site of Pikillacta. Unpublished Ph.D. dissertation, Department of Anthropology, Brandeis University, Waltham, Massachusetts. Glowacki, Mary, and Michael Malpass 2003 Water, Huacas, and Ancestor Worship: Traces of a Sacred Huari Landscape. Latin American Antiquity 14(4):431–448. Hocquenghem, Anne M. 1979 Rapport entre les mort et les vivants dans la cosmovision Mochica. Objets et Mondes 19:85–95. Hurst, Joanna 2007 Human Skeletal Analysis of Cruz Muqu 1, 2 and 3, Sarhua, Peru. Research report on file, University of London, London. Isbell, William H. 1977 The Rural Foundations for Urbanism, Economic and Stylistic Interaction between Rural and Urban Communities in EighthCentury Peru. Illinois Studies in Anthropology No. 10. University of Illinois Press, Urbana. 1997 Mummies and Mortuary Monuments: A Postprocessual Prehistory of Central Andean Social Organisation. University of Texas Press, Austin. 2001 Repensando el Horizonte Medio: El caso de Conchopata, Ayacucho, Peru. In Huari y Tiwanaku: Modelos vs. evidencias. Primera parte, edited by Peter Kaulicke and William H. Isbell, pp. 9–68. Boletín de Arqueología PUCP No. 4. Fondo Editorial de la Pontificia Universidad Católica del Perú, Lima. 2004 Mortuary Preferences: A Wari Culture Case Study from Middle Horizon Peru. Latin American Antiquity 15(1):3–32. Keeley, Helen M., and Frank M. Meddens 1993 Prehispanic Agricultural Terraces in the Chicha/Soras Valley Peru. Institute of Archaeology Bulletin (London) 29:121–138. Kemp, Rob, Nicholas Branch, Barbara Silva, Frank Meddens, Alan Williams, Ann Kendall, and Cirilo Vivanco 2006 Pedosedimentary, Cultural and Environmental Significance of Paleosols within Pre-Hispanic Agricultural Terraces in the Southern Peruvian Andes. Quaternary International 158:13–22. Knobloch, Patricia J. 1991 Stylistic Date of Ceramics from the Huari Centres. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 247–258. Dumbarton Oaks, Washington, D.C. McAnany, Patricia A. 1995 Living with the Ancestors: Kinship and Kingship in Ancient Maya Society. University of Texas Press, Austin. McCormac, F. G., A. G. Hogg, P. G. Blackwell, C. E. Buck, T. F. G. Higham, and P. J. Reimer 2004 SHCa104 Southern Hemisphere Calibration, 0–11.0 cal kyr BP. Radiocarbon 46(3):1087–1092. McEwan, Gordon F. 1991 Investigations at the Pikillacta Site: A Provincial Huari Centre in the Valley of Cuzco. In Huari Administrative Structures: Prehistoric Monumental Architecture and State Government, edited by William H. Isbell and Gordon F. McEwan, pp. 93–120. Dumbarton Oaks, Washington, D.C.

1996 Archaeological Investigations at Pikillacta, a Huari Site in Peru. Journal of Field Archaeology 23:169–186. McEwan, Gordon F. (editor) 2005 Pikillacta: The Wari Empire in Cuzco. University of Iowa Press, Iowa City. Meddens, Frank M. 1985 The Chicha/Soras Valley during the Middle Horizon: Provincial Aspects of Huari. Unpublished Ph.D. dissertation, Institute of Archaeology, University of London. 1989 Implications of Camelid Management and Textile Production for Huari. In The Nature of Huari: A Reappraisal of the Middle Horizon Period in Peru, edited by Michael Czwarno, Frank Meddens, and Alexandra Morgan, pp. 146–165. BAR International Series 525. British Archaeological Reports, Oxford. 1991 A Provincial Perspective of Huari Organisation Viewed from the Chicha/Soras Valley. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by Gordon McEwan and William H. Isbell, pp. 215–231. Dumbarton Oaks, Washington, D.C. 1994 Mountains, Miniatures, Ancestors and Fertility: The Meaning of a Late Horizon Offering in a Middle Horizon Structure in Peru. Institute of Archaeology Bulletin (London) 31:127–150. 1997 Function and Meaning of the Ushnu in Late Horizon Peru. Tawantinsuyu 3:5–14. 2001 La sequencia cultural en la cuenca de Chicha-Soras. In XII Congreso Peruano, “Luis Lumberas,” vol. 2, edited by Ismael Pérez C., Walter Aguilar S., and Medardo Purizaga V., pp. 200– 210. Universidad Nacional de San Cristóbal de Huamanga, Ayacucho, Peru. Meddens, Frank M., Nicholas P. Branch, Cirilo Vivanco Pomacanchari, Naomi Riddiford, and Rob Kemp 2008 High Altitude Ushnu Platforms in the Department of Ayacucho, Peru: Structure, Ancestors and Animating Essence. In Pre-columbian Landscapes of Creation and Origin, edited by John Staller, pp. 315–355. Springer, New York. Meddens, Frank M., and Anita Cook 2000 La administración Wari y el culto a los muertos: Yako, los edificios en forma “D” en la sierra sur-central del Peru. In Wari: Arte precolombino peruano, edited by Luis Milliones, pp. 212– 228. Fundación El Monte, Seville. Murra, John V. 1975 Maís, tubérculos y ritos agrícolas. In Formaciones económicasa y políticas del mundo andino, edited by John V. Murra, pp. 45–58. Instituto de Estudios Peruanos, Lima. Ochatoma Paravicino, Jose 2007 Culto a los muertos y sistemas de enterramiento en la época Huari, una visión desde Conchopata. In El desarrollo de la ciencias sociales en Ayacucho, edited by Luis Milliones, Jefrey Gamarra, and Jose Ochatoma, pp. 12–60. La Universidad de San Cristóbal de Huamanga, Universidad Nacional Mayor de San Marcos, Fondo Editorial de la Facultad de Ciencias Sociales, Ayacucho, Peru. Ochatoma Paravicino, Jose, and Martha Cabrera Romero 2001 Poblados rurales Huari, Una vision desde Aqo Wayqo. CANO Asociados SAC, Lima. Schreiber, Katherina J. 1987 From State to Empire: The Expansion of Wari Outside of the Ayacucho Basin. In Origins and Development of the Andean

T he Wari S tate , I ts U se of A ncestors , Rural H interland , and Agricultural I nfrastructure

  169

State, edited by Jonathan Haas, Sheila Pozorski, and Thomas Pozorski, pp. 91–96. Cambridge University Press, New York. 1992 Wari Imperialism in Middle Horizon Peru. Anthropological Papers, Museum of Anthropology No. 87. University of Michigan, Ann Arbor. Silva, Barbara N. 2005 Late Glacial and Holocene Environmental Changes in the Peruvian Andes: Implications for Human Society. Unpublished Ph.D. dissertation, Royal Holloway, University of London. Thompson, Lonnie G., Ellen Mosley-Thompson, J. F. Bolzan, and B. R. Koci 1985 1500 Year Record of Tropical Precipitation in Ice Cores from the Quelccaya Ice Cap, Peru. Science 229:971–973. Topic, John 1985 Huari and Huamachuco. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by Gordon McEwan and William H. Isbell, pp. 141– 164. Dumbarton Oaks, Washington, D.C. Tripcevich, Nicholas

170 

F rank M eddens and N icholas B ranch

2007 Quarries, Caravans, and Routes to Complexity: Prehispanic Obsidian in the South-Central Andes. Unpublished Ph.D. dissertation, University of California, Santa Barbara. Tung, Tiffiny A., and Anita G. Cook 2006 Intermediate Elite Agency in the Wari Empire: The Bio­­ archaeological and Mortuary Evidence. In Intermediate Elites in Pre-Colombian States and Empires, edited by Chistina M. Elson and R. Alan Covey, pp. 68–83. University of Arizona Press, Tucson. Valdez, Lidio, and Cirilo Vivanco P. 1994 Arqueología de la cuenca de Qaracha, Ayacucho (Perú). Latin American Antiquity (5)2:144–157. Williams, Patrick R. 2001 Cerro Baúl: A Wari Center on the Tiwanaku Frontier. Latin American Antiquity 12(1):67–83. Zuidema, Reiner T. 1989 El ushnu. In Reyes y guerreros, Ensayos de cultura andina, edited by Manuel Burga, pp. 402–454. Grandes Estudios Andinos. Fomciencias, Lima.

Ch a p t er 10

o Piecing Together the Middle The Middle Horizon in the Norte Chico Kit Nelson, Nathan Craig, and M anuel Perales

T

he data for the Norte Chico do not fit expectations of social or political control by the Wari during the Middle Horizon (MH). Instead, they reveal participation of local polities in at least two panregional iconographic phenomena as well as in their own localized systems. These levels of participation are visible in architecture and pottery stylistic features in both the Pativilca and Huaura valleys. Stylistic similarities in architecture document local solidarity by the presence of repeated room forms, invoking shared concepts of architecture and space. Pottery styles represent a duality that is indicated by the inclusion of both Central Coast press-mold pottery and more general MH polychromes. Both architecture and pottery, which each represent multilevel and multimedia systems of style, come together to form the complex MH of the Norte Chico. Previous notions of the incorporation of the Norte Chico in the Wari Empire have been based on expansionist models that include the north-central coast in the broad area of control. These assumptions of Wari dominance were based on characterizations of the Wari as having broad control over vast regions (Isbell and McEwan 1991; Schreiber 1992) and on isolated finds of Wari-like architecture and pottery in the region. The Norte Chico was included within the Wari Empire

because of its location close to sites such as Pachacamac (Rostoworski de Diez Canseco 1992; Shimada 1991; Uhle 1903) and Socos (Isla and Guerrero 1987), which were argued to have Wari components. The interpretation of the Norte Chico as one of the many areas under Wari domination was sealed by a few tantalizing descriptions of Wari-type artifactual remains discovered in the Supe Valley (Menzel 1977; Reiss and Stuebel 1880; Uhle 1925) and at the coastal site of Végueta in the Huaura Valley (Shady and Ruiz 1979). As a result, the Norte Chico is assumed to have been under Wari domination even though there was scant empirical evidence of direct control. Using broadscale patterning of traits, we explore the stylistic elements that compose the MH complex in the Norte Chico. We use these data to document different levels of stylistic spheres, including panregional, regional, and local distributions. These stylistic modalities are displayed in a variety of media, including architecture and pottery, and detail the complex stylistic systems of this frontier zone during the MH. This reevaluation of the MH in the Norte Chico raises larger questions of the role of the Wari in this and other intermediary or frontier areas. These results support a dynamic model of interaction that includes the importance of local systems of power and authority that

  171

in some cases are stronger outside of, the Norte Chico. Archaeological research documents a long human occupation of the Norte Chico. Although focused on the earliest and latest prehistoric periods, this work has revealed a range of site types and occupational periods in this region. Research in the Norte Chico began in the 1920s with the visits by Uhle (1925) to the Supe Valley and included basic published descriptions of the archaeology. In 1943 preliminary excavations of the site of Puerto de Supe and other sites outside of the Norte Chico were carried out by Strong and Willey (1943). Following these

develop within the context of negotiating the incorporation of specific components of the political ideology that likely underlie the larger stylistic systems.

Previous Research The Pativilca and Huaura valleys are located on the Central Coast in an area referred to locally as the Norte Chico (Figure 10.1). Although identified as a distinct region, the various cultural connections among valleys in the Norte Chico changes through time. During some periods, these connections articulate beyond, and

-77

a

-78

Fo Rio rta lez

PV39-211

-75

-74

PV39-212

Rio Supe

Vegueta

-11

-76

Rio ilca Pativ

Caldera

ur

ua

oH

Ri

a

Rio

cay

an

Ch

o

Ri

Socos

illo

Ch

n

imac

Rio R

-12

Pachacamac

0

-13

25

50

100

150

200

Wari

Rio

15

20

20

25

2500

1500

vilca Pati

1500

150

00 2000

25

0 100

PV39-212 PV39-211

Ri o

-10.6

0

00

00

lez

10

Fo rta

0

15

20

a

0

Kilometers

00

35

00

00 0

20

450

0

0 50

-10.7

0

4500

0 400

30 0

-10.8

0 450

-10.9

Rio

pe

Su

2000

6

12

18

-77.8

Rio

Hua

ura

0

Kilometers

24

15

3

500

0

0

Caldera

0

10 0

20 -11.1

00

50 0

Vegueta

-11

172 

00

-77.7

-77.6

-77.5

-77.4

K it N elson , N athan C raig , and M anuel P erales

-77.3

-77.2

Figure 10.1  Map of the Norte Chico marked with sites discussed in the text.

early studies, more detailed projects were conducted. These later investigations are dominated by a focus on sites dating to the Preceramic (or Late Archaic) Period. General discussions concerning the Preceramic in the Norte Chico cover social organization, consumption, control, and other diverse topics (Engel 1957a, 1957b; Feldman 1983, 1985, 1987; Haas and Creamer 2001, 2004; Haas et al. 2004a; Moseley 1975; Vega-Centeno SaraLafosse 2005a; Vega-Centeno Sara-Lafosse et al. 1998; Zechenter 1988). Detailed investigations have occurred at several Preceramic sites including Aspero (Feldman 1980; Moseley 1975) and Lampay (Vega-Centeno SaraLafosse 2005b, 2006, 2007; Vega-Centeno Sara-Lafosse et al. 2006), and projects are ongoing at the sites of Caral (Shady 1993, 1995, 1997, 1999a, 1999b, 1999c, 1999d, 2000a, 2000b, 2000c; Shady et al. 2001), Caballete (Haas et al. 2004b), Huaricanga (Haas et al. 2007), and Banduria (Chu Barrera 2006; Fung 1974). The later periods of the Norte Chico have received less attention (Ruiz Estrada 1999). Surveys of the lower and middle valleys have provided basic information such as site location and general time period. These surveys provide the basis for understanding change through time and period-specific settlement patterns. The first survey of sites was carried out in the Huaura Valley using air photos by Cárdenas (1977, 1977–1978, 1988), which resulted in a list and descriptions of sites based on Cárdenas’s study by Miasta Gutiérrez and Merino Jiménez (1986). Several more recent comprehensive surveys of the Huaura (Nelson and Ruiz 2004), Pativilca (Perales Munguía 2006), and Fortaleza valleys (Perales Munguía 2007) incorporate new data collection methods including Global Positioning System (GPS) and geographic information system (GIS) technology and provide improved coverage of the lower and middle valleys of the Norte Chico. The majority of excavation projects at late period sites in the Norte Chico has been focused on Late Intermediate Period (LIP) sites and their association to the Chancay. These include descriptions of the sites of Centinela and Vilcahuaura by Kosok and Schaedel (Kosok 1965) and small-scale excavations at Rontoy, Quipico, Centinela and Chambara (Nelson and Ruiz Estrada 2010; Ruiz Estrada and Nelson 2008), and Casa

Blanca and Quintay (Krzanowski 1991) in the Huaura Valley. The fortress of Acaray has both an LIP and Early Horizon occupation that has been studied in more detail (Brown Enrile 2005, 2006; Brown Enrile and Rivas Panduro 2004; Brown Vega 2008, 2009; Horkheimer 1962; Ruiz Estrada and Domingo Torero 1978), including large open excavations and radiocarbon dating confirming the timing of use (Brown Vega 2009). Of specific interest to this paper are the studies of MH sites. The MH in the Norte Chico is largely unstudied; few published reports are available, and these represent a small number of sites, all of which are focused on individual finds or descriptions of single sites. Shady and Ruiz (1979) discuss the multiple design styles present on pottery from MH-period burials from the site of Végueta located in the coastal zone of the Huaura Valley, stressing the importance of local polities in the Norte Chico during the MH. The site of Caldera was examined by Stumer (1952), who provides a general site map and discusses the types of pottery found, including both MH and LIP pottery styles. More recent work at the MH sites of Caldera and El Carmen in the Huaura Valley include targeted excavations within MH architecture (Heaton et al. 2010; Pierce Terry et al. 2010). In all, although the valleys of the Norte Chico are rich in sites dating to all periods, very little is known about the archaeology of this area.

Assessing Architectural Style This study is based on a combination of valley survey and stylistic analysis of traits. These data are preliminary and based on analysis of data collected during an initial valley survey and small amounts of additional site documentation. In the Pativilca Valley and Huaura Valley, initial steps of systematic regional investigation were operationalized through siteless survey (Dunnell and Dancey 1983; Thomas 1975) using Differential GPS (DGPS)–enabled mobile GIS (Tripcevich 2004a, 2004b, 2006) to systematically document artifact and architectural distributions on the landscape (Craig et al. 2007). Shuttle Radar Topographic Mission (SRTM) (van Zyl 2001) 90 m digital elevation models (DEMs) and historic aerial photographs from the Servicio Aerofotográfico Nacional

P iecing T ogether the M iddle

  173

(SAN) were added to the GIS database to create a more detailed picture of the sites and their relationships to the valley topography. These data provided the distribution of sites by general time period as well as very basic maps of each site. Additional data were also collected, including general pottery information based on surface scatters and types of construction techniques utilized. The second phase, still in progress, is designed to further develop the valley chronology using radiocarbon dating and ceramic analysis. This phase includes additional mapping, small-scale excavation, and artifact analysis of sites throughout the valleys. MH sites were documented during the survey, and the site of Caldera was chosen for a small architectural study due to its high level of preservation. Located in the middle of the lower valley on the north side of the river, Caldera rests in a small quebrada and is visibly prominent from the valley floor. A map of selected architecture at the site was created using the DGPS-enabled mobile GIS, and data on each room, including presence and description of doorway embellishments, wall construction, plaster color, doorway width, wall height, number of stories, and type of subdivisions, were recorded for each of the thirteen buildings studied. Although this is by no means a complete record of the site of Caldera, it is a sample that represents some of the diversity in architecture at a well-preserved MH site that is typical of the architecture of this period in the Huaura Valley, in contrast to the Pativilca Valley, where no such well-preserved sites were identified during survey. The detailed study of the site of Caldera was then compared to observational data collected during survey and used to more broadly define the MH architectural traditions. Special focus was given to architectural stylistic elements that were present at all sites and features that defined the MH and were not present at sites dating to periods directly before and following the MH. Detailed features at Caldera include embellished doorways, specialized wall form, and distinctive wall color. These features were present at sites in the Huaura Valley and not present in the Pativilca Valley. Sites in the Pativilca Valley shared fewer and more general features. In both cases, architecture represented localized expressions of ideas of space, function, ideology, and aesthetic (Moore 1996). 174 

K it N elson , N athan C raig , and M anuel P erales

Architecture as Style and the MH of the Norte Chico The patterning of architectural types within the Pativilca and Huaura valleys reveals the localized nature of style and possibly the extent of local polities in the Norte Chico during the MH. The regularity of architecture in the Huaura Valley, and to a lesser extent in the Pativilca Valley, presents and represents shared concepts of space (Hillier and Hanson 1984; Lawrence and Low 1990) as documented in the repetition of layout and architectural features at both the intersite and intrasite levels. In the Norte Chico, the MH settlement pattern consists of small (