An Archaeology of Taki Onqoy:
Revitalization and Entanglement in Colonial Peru
By
Scotti Michelle Norman
Dissertation
Submitted to the Faculty of the
Graduate School of Vanderbilt University
in partial fulfillment of the requirements
For the degree of
DOCTOR OF PHILOSOPHY
in
Anthropology
August 31, 2019
Approved:
Steven A. Wernke, Ph. D.
John W. Janusek, Ph.D.
Beth Conklin, Ph.D.
Jane Landers, Ph.D.
Copyright © 2019 by Scotti Michelle Norman
All Rights Reserved
ii
To victims and survivors, the fight continues.
iii
ACKNOWLEDGEMENTS
Funding for the initial phases of this research was provided by the Vanderbilt Center for
Latin American Studies and the Tinker Foundation, and the Vanderbilt College of Arts and
Science. Intensive field research was generously supported by a Dissertation Research Grant
from the Wenner-Gren Foundation for Anthropological Research and the Fulbright-Hays
Doctoral Dissertation Research Abroad Program. Subsequent funding for more in-depth analysis
was awarded by the Field Museum Elemental Analysis Facility Research Grant and the
Department of Anthropology at Vanderbilt University. The final write-up stages of this research
would not have been possible without the support of the Vanderbilt University Robert Penn
Warren Center for the Humanities Fellowship and the Dumbarton Oaks Junior Fellowship in
Precolombian Studies.
First and foremost, thank you to my colleagues and friends associated with the Proyecto
Arqueológico de Taki Onqoy. My intrepid co-director, Stephanie Peña del Castillo, was a
steadfast source of support. As two new field directors on our first solo projects, Steph and I
sustained one another, and became a wonderful team. Thank you to the group of students from
Ica, Ayacucho, and Andahuaylas: Alcides Berrocal Gonzalez, Marco Antonio Naveros Guizado,
Gaby Rocio, Ronald Mendoza, and Rubi Zanabria. Gracias a todos los comuneros de Chicha,
Larcay, and Pampachiri, especialmente Heraclio Barraza Arenaza, Rafael Mallco Huarcaya,
Paulina Astoquillca, Waldir Padilla, la Sra. Mercedes, y Ingrid Hernández Sotelo. Gracias al
Ministerio de Cultura de Ayacucho y José Ochatoma. Thank you to my team of strong women:
Sarah Jolly was an ace in the field and helped despite a touch of la bicicleta. Anna Gurevitz did
an incredible job analyzing the bioarchaeological remains and Sarah Kennedy was fast and
thorough in her analysis of the zooarchaeological remains.
I am sincerely grateful to a number of individuals who have supported me throughout this
long journey. My committee members Steve Wernke, John Janusek, Beth Conklin, and Jane
Landers have all provided thoughtful feedback throughout various stages of developing this
research project. I am especially thankful to my advisor, Steve Wernke, who encouraged me to
pursue my own research interests and to trust my instincts during fieldwork. I will always fondly
remember the first meeting in which we discussed my interests in studying Taki Onqoy, and the
widening of his eyes in our subsequent meeting, in which I told him I had found a potential field
site. Dr. Wernke helped me throughout the countless proposal revisions, and has been both a
supportive mentor and a tough critic of my work. My understanding of archaeology and the
improvement of my writing throughout the years owe much to him. John Janusek has always
been an optimistic and thoughtful mentor, and I am honored to call him my friend. Beth
Conklin’s course on Amazonia was inspirational in shaping the way I have thought about
revitalization and millenarian movements, and her ethnographic experience has been critical in
helping me understand the concepts of illness and the body. Jane Landers has always been
extremely patient and willing to help my research—she was responsible for my initial forays into
documentary research and Spanish paleography, and sat next to me several times helping me
read each account line by line.
I am indebted to a number of other faculty members from various universities who are
responsible for shaping my career path. When I was an undergraduate student at UCLA, I
emailed every graduate student listed as an archaeologist on the Department of Anthropology
iv
web page. Two (then) graduate students responded. Anthony Graesch invited me to work in his
lab processing the micro-artifacts from his research. He has continued to support me over the last
decade, and I always look forward to our chats at the SAAs. Colleen Zori asked if I would like to
help out with her dissertation field research in Chile, and my three months living with her and
others in the Atacama Desert marked my first time completing an extended field season. The
generosity of Anthony and Colleen during my undergraduate years shaped who I want to be as a
mentor and who I am as an archaeologist. Thank you also to several women in archaeology who
have offered advice on succeeding in a male-dominated field: it was a pleasure to work with
Elizabeth Arkush in the field and I have always looked up to her as a model of who I would like
to emulate as an advisor in the future. Tiffiny Tung and Tamara Bray have been wonderful
mentors, and I always value our discussions about the role of women in the field, how to project
my voice and protect myself, and strategies for career development. Thank you also to Uzma
Rizvi and Chelsea Blackmore, who both visited Vanderbilt University and have offered integral
advice on diversifying my theoretical orientations and have been sources of support. One final
thanks to Monica Barnes, who has graciously offered her advice and much of her research
regarding the Chicha-Soras Valley.
Two of my best friends, Douglas Smit and Carla Hernandez, have been my biggest
sources of support throughout this process. Doug and I developed our projects simultaneously,
co-chaired the first SAA session we ever organized, and we have been there for one another in
the darkest days. I cannot wait to collaborate with him in the future. Despite my initial poor
impression on Carla Hernandez, we have become more like family rather than merely friends.
She has been there for me always, maintaining my safety, sanity, and improving my
archaeological education. Together with my amazing friend and co-host Gabriela Ore, Doug and
Carla saved me from some terrible situations, and I am eternally grateful.
I have so many other colleagues I would like to thank. First, thank you to my cohort of
delinquents, especially Tatiana Paz Lemus, Sarah Eckhardt, and Brian McCray. So much love to
you all, we even survived living together at the Anthro House. Other friends and colleagues were
instrumental in my finishing this dissertation: Tizoc Chavez, Anna Guengerich, Ellen Lofaro,
Kylie Quave, Keitlyn Alcantara, Matt Velasco, Brendan Weaver, Rebecca Bria, Terren Proctor,
Amanda Bishop, Emma Banks, James Zimmer-Dauphine, Gloria Suarez Fuentes, and Dan
O’Maley. Thank you to Ran Boytner, Sam Connell, and Chad Gifford for giving me my first
archaeological opportunities and more importantly, ridiculous memories which I will cherish
forever. Thank you to the other PAP crew, including Amber Anderson, Vanessa Muros, Eric
Fries, Darryl Wilkinson, Beccah Landman, and honorary members Emma Koenig and Aaron
Coons. Thank you to the Dumbarton Oaks Mandatory Friendship Group and Precolumbian
Studies Cohort, and especially to Adrienne Vartimidis and Colin McEwan.
Finally, thank you to my assorted other friends who have been by my side: Megan
Reiger, Linda Nyberg, Yuliya Goldshteyn, Kaleigh Shaw, and Liz Wilgenburg. And to my
family, my mother Mindy Norman and my sister Sierra Norman. And there are two other
individuals I must acknowledge: my dog, Fritz, who has been my neurotic ferocious pitbull
support system, and my 2001 Chevy Astrovan Tiny Tim. You both have been by my side
without complaint (though you both lack the language to complain), and I am grateful.
v
TABLE OF CONTENTS
Page
Acknowledgments………………………………………………………………………………..iv
Table of Contents…………………………………………………………………………………vi
List of Tables……………………………………………………………………………………..xi
List of Figures…………………………………………………………………………………...xii
Chapter
1. Conquest, Conversion, and Revitalization .................................................................................. 1
Taki Onqoy Source Critiques and Historiography: A Brief Assessment .................................... 3
Archaeological Interventions ...................................................................................................... 8
Theoretical Orientation ............................................................................................................. 11
Organization of the Dissertation ............................................................................................... 15
2. The History and Archaeology of Taki Onqoy ......................................................................... 18
Introduction ............................................................................................................................... 18
Things, Materiality, Animacy, and Entanglement .................................................................... 21
Andean Ontologies and Western Ontologies ............................................................................ 23
Andean Religion Prior to Inka and Spanish Conquest .............................................................. 24
Animate Andean Things and the Transformative Properties of Huacas ............................... 26
Inka Religion and the Establishment of the Cult of the Sun ..................................................... 28
The Inka Dead as Huacas ...................................................................................................... 29
Conversion in the Andes ........................................................................................................... 30
The Problems of Language Diversity at the Time of Conquest ................................................ 32
Evangelization after Conquest (1532-1569) ............................................................................. 33
Taki Onqoy................................................................................................................................ 37
Primary Sources ..................................................................................................................... 38
The First Account—a Letter from Guerrero to Luis de Olvera (1564) ................................. 39
Cristóbal de Molina and Cristóbal de Albornoz .................................................................... 42
Alternative Accounts: Álvarez and Sotelo ............................................................................ 58
Scholarly Debate of Taki Onqoy............................................................................................... 63
Historiography of Taki Onqoy Scholarship........................................................................... 64
Theme One: What was the Nature of Taki Onqoy? .............................................................. 65
vi
Theme Two: What was the Geographic Expanse of Taki Onqoy? ....................................... 73
Theme Three: Was Taki Onqoy Connected to Vilcabamba? ................................................ 75
Theme Four: Was Taki Onqoy an Exaggeration Levied by Spanish Priests? ....................... 78
Models for an Archaeological Study of Taki Onqoy ................................................................ 85
3. An Andean Archaeology of Revitalization ............................................................................... 88
The Utility of the Resistance Concept....................................................................................... 90
Revitalization Movements......................................................................................................... 93
Revitalization Theory and Taki Onqoy ................................................................................. 98
Archaeology of Revitalization ............................................................................................. 108
Divided Religions, Entangled Practices: The Utility of Considering Taki Onqoy as a
Revitalization Movement .................................................................................................... 113
Discussion ............................................................................................................................ 116
Taki Onqoy as Performance .................................................................................................... 117
Do Taki Onqoy Practices Constitute a Novel Response to Spanish Presence?................... 127
Reframing considerations of Taki Onqoy: A Practice Theory Approach ............................... 134
Discussion and Conclusion ..................................................................................................... 137
4. Historical and Archaeological Background of the Chicha-Soras Valley................................ 138
Introduction ............................................................................................................................. 138
Geography and Climate........................................................................................................... 139
Valley Divisions and Ecozones ........................................................................................... 144
Human Interventions in Agricultural Productivity .............................................................. 148
Prior Archaeological Research ................................................................................................ 150
Early Settlement: The Preceramic (?—1800 BCE), Initial (1400—1000 BCE), Early
Horizon (~1000 BCE – 200 BCE), and Early Intermediate Periods (200 BCE – 550 CE) 153
The Fluorishing Middle Horizon (550 CE - 1000 CE)........................................................ 158
Late Intermediate Period (1000 CE - 1400 CE) .................................................................. 169
Late Horizon (1400 CE – 1532 CE) .................................................................................... 174
Discussion of Archaeological Investigation in the Chicha-Soras Valley ............................ 186
An Ethnohistoric Perspective of the Chicha-Soras Valley...................................................... 187
The Soras and the Chanka ................................................................................................... 188
The Soras and the Inka ........................................................................................................ 191
The Soras after Spanish Conquest ....................................................................................... 197
vii
Recent History of the Chicha-Soras Valley ............................................................................ 204
Discussion: Patterns in the History and Archaeology of the Chicha-Soras Valley ................ 209
5. Research Questions and Methodology.................................................................................... 212
Research Questions and Hypotheses ....................................................................................... 212
Materiality of Taki Onqoy in the Non-Religious Sphere .................................................... 213
Taki Onqoy in the Religious Sphere.................................................................................... 215
Taki Onqoy and Catholic Spaces......................................................................................... 217
Research Design ...................................................................................................................... 218
Methodology ........................................................................................................................... 226
Phase One: Preliminary Mapping and Architectural Survey............................................... 226
Phase Two: Excavation at Iglesiachayoq ............................................................................ 232
Phase Three: Artifact and Materials Analysis ..................................................................... 248
Phase Four: Spatial Analysis ............................................................................................... 261
Summary of Methods .............................................................................................................. 261
6. Household Archaeology at Iglesiachayoq .............................................................................. 263
Early Colonial Period Materials .............................................................................................. 265
Animals and Plants .............................................................................................................. 265
Ceramic and Lithic Technologies ........................................................................................ 268
Personal Goods and Portable Items ..................................................................................... 271
Sectorization of Iglesiachayoq ................................................................................................ 272
Sector 1 ................................................................................................................................ 275
Sector 2 ................................................................................................................................ 285
Sector 3 ................................................................................................................................ 287
Archaeological Occupations at Iglesiachayoq ........................................................................ 289
Architectural Styles and Forms ............................................................................................... 290
Ceramic Vessel Types and Pastes ........................................................................................... 296
Excavation Results .................................................................................................................. 299
Sector 1 Domestic Structure Excavations ........................................................................... 299
Sector 2 Domestic Structure Excavations ........................................................................... 322
Sector 3 Domestic Structure Excavations ........................................................................... 334
Units 11-15: Empty Patio Spaces ........................................................................................ 345
Unit 16—Stone Fragmentation and Burned Soil in the Plaza ............................................. 348
viii
Comparative Assemblage Analysis by Sector ........................................................................ 352
Comparative Ceramic Data: Differential Status and Activities by Sector .......................... 352
Faunal Data .......................................................................................................................... 358
Comparative Lithic Data by Sector ..................................................................................... 359
Discussion and Conclusion ..................................................................................................... 362
7. Mortuary Politics at Iglesiachayoq ......................................................................................... 364
The Living Bury the Dead, the Dead Transform the Living ................................................... 366
The Andean Animate Dead and the Spanish Corruptible Body.............................................. 368
Andean Burial Practices in the Early Colonial Period ............................................................ 373
Catholic Treatment of the Dead .............................................................................................. 376
Catholic Church Mandates During the Early Colonial Period ................................................ 378
Hypotheses for Mortuary Practice at Iglesiachayoq ............................................................... 380
Mortuary Practices at Iglesiachayoq ....................................................................................... 387
General Findings and Burial Descriptions........................................................................... 388
Burial Types and Spatial Location ...................................................................................... 390
Discussion of Spatial Location and Burial Styles ............................................................... 419
Readdressing Hypotheses........................................................................................................ 430
Conclusion............................................................................................................................... 432
8. Spatial Analysis of Visibility and Walking Paths at Iglesiachayoq ........................................ 434
Taki Onqoy Spaces and Catholic Spaces ................................................................................ 436
Visibility and Invisibility ..................................................................................................... 437
Proximity to Sacred Landscapes .......................................................................................... 438
Built Environment and Architectural Styles at Iglesiachayoq ................................................ 440
The Central Church Complex (Sector 1) ............................................................................. 440
The Upper Prehispanic Plaza Complex (Sector 3) .............................................................. 448
Sector 2 Layout: A Haphazard Grid? .................................................................................. 450
Synthesis of Iglesiachayoq Building Phases and Sectorization........................................... 454
Modeling Visibility at Iglesiachayoq—Viewsheds as Indices of Privacy .............................. 455
Viewshed Modeling at Iglesiachayoq.................................................................................. 458
Model 1: Sector 1 and Church Complex ............................................................................. 458
Model 2: Local Plaza ........................................................................................................... 460
Model 3: Wallpa Wiri .......................................................................................................... 462
ix
Discussion of Viewsheds and Visibility .............................................................................. 464
Modeling Movement through Iglesiachayoq—Walking Times as Indices of Proximities to
Sacred Landscapes .................................................................................................................. 466
Spatial Network Analysis Modeling at Iglesiachayoq ........................................................ 468
Model 1 Results: Rectilinear Structures to Church Plaza.................................................... 471
Model 2 Results: Sector 3 Structures to Central Plaza ........................................................ 474
Model 3 Results: Rectilinear Structures to Prehispanic Plaza ............................................. 476
Model 4 Results: Sector 3 Structures to Prehispanic Plaza ................................................. 478
Discussion: Taki Onqoy and movement through Iglesiachayoq ............................................. 481
Discussion and Conclusions .................................................................................................... 482
9. Summary and Conclusions ..................................................................................................... 484
Taki Onqoy Sources and Practices .......................................................................................... 485
Results: Household Archaeology ............................................................................................ 489
Results: Mortuary Archaeology .............................................................................................. 491
Results: Spatial Analysis ......................................................................................................... 493
Re-evaluation of Hypotheses .................................................................................................. 494
Was Taki Onqoy a Unique Movement, or a Reflection of Pan-Andean Sixteenth-Century
Practices? ................................................................................................................................. 498
Contributions and Future Directions ....................................................................................... 500
Appendix
A. Faunal Analysis ...................................................................................................................... 505
B. Stable Isotope Analysis Report .............................................................................................. 508
C. Ceramic Analysis ................................................................................................................... 514
D. Exported Samples and Radiocarbon Dates ............................................................................ 554
E. Tupus and Metal Artifacts ...................................................................................................... 558
F. Unit Summaries ...................................................................................................................... 562
References ................................................................................................................................... 649
x
LIST OF TABLES
Table
Page
2.1. Taki Onqoy correlates from the 1564 Guerrero letter. .......................................................... 42
2.2. Taki Onqoy correlates from Cristóbal de Molina. ................................................................. 47
2.3. Taki Onqoy correlates from Albornoz. .................................................................................. 57
2.4. Taki Onqoy correlates from Álvarez. .................................................................................... 62
4.1. Accounts of Albornoz's activities in the Soras region. ........................................................ 199
4.2. Towns of the 1586 Soras repartimiento; adapted from Meddens and Schreiber 2010. ....... 200
5.1. Distribution of structure forms by Sector. ........................................................................... 221
5.2. Excavated Units at Iglesiachayoq. ....................................................................................... 223
5.3. Breakdown of age categories used by Gurevitz in osteological analysis. ........................... 254
6.1. Flora and fauna in the Chicha-Soras Valley in the Early Colonial Period. ......................... 268
6.2. Possible ceramic types in the Early Colonial Period in the Chicha-Soras Valley. .............. 271
6.3. Possible personal items in Early Colonial Period assemblages in the Chicha-Soras Valley.
..................................................................................................................................................... 272
6.4. Structure forms by sector at Iglesiachayoq. ......................................................................... 291
6.5. Comparison of average wall heights and dwelling areas by structure form, n=91 structures.
..................................................................................................................................................... 295
6.6. Vessel types in Unit 1. ......................................................................................................... 308
6.7. Paste types in Unit 1. ........................................................................................................... 309
6.8. Unit 1 recovered fauna by Minimum Number of Individuals (MNI) and site-wide
percentages. ................................................................................................................................. 309
6.9. Vessel types in Unit 2. ......................................................................................................... 312
6.10. Paste types in Unit 2. ......................................................................................................... 312
6.11. MNI of fauna and site-wide percentage for Unit 2. ........................................................... 313
6.12. Vessel types, counts, and percentage of unit assemblage in Unit 3. .................................. 315
6.13. Paste types from Unit 3. ..................................................................................................... 316
6.14. Faunal MNI recovered from Unit 4. .................................................................................. 319
6.15. Numbers and Percentages of Vessels from Unit 4............................................................. 321
6.16. Paste Types from Unit 4. ................................................................................................... 322
xi
6.17. Vessel Types and Counts from Unit 5. .............................................................................. 324
6.18. Paste types from Unit 5. ..................................................................................................... 324
6.19. Counts of vessels from Unit 6............................................................................................ 328
6.20. Paste Types from Unit 6. ................................................................................................... 328
6.21. Vessel types and counts from Unit 7. ................................................................................ 331
6.22. Paste Types from Unit 7. ................................................................................................... 331
6.23. Proportions of Vessels Recovered from Unit 8. ................................................................ 335
6.24. Vessel Pastes from Unit 8. ................................................................................................. 335
6.25. Vessel types and proportions from Unit 9. ........................................................................ 339
6.26. Paste types of vessels from Unit 9. .................................................................................... 339
6.27. Vessel types and proportions from Unit 10. ...................................................................... 343
6.28. Vessel paste types from Unit 10………………………………………………………….344
7.1. Burials and mortuary contexts in churches from the Colonial Period Spanish Southeast,
North America. Data adapted from Larsen 1993: 336. ............................................................... 385
7.2. Summary of data from burial loci at Iglesiachayoq. ............................................................ 420
7.3. Age breakdown of individuals at Iglesiachayoq. ................................................................. 426
8.1. Four walking models between starting and ending points. .................................................. 469
8.2. Summary of walking times for each model. ........................................................................ 479
xii
LIST OF FIGURES
Figure
Page
1.1 Map of Chicha-Soras Valley with modern towns and Iglesiachayoq (red triangle). ................ 9
1.2. Sign in Pampachiri depicting "las rutas del Taki Unquy" ..................................................... 15
2.1. View of Iglesiachayoq looking north with casa de kuraka (left) and church (right). ............ 55
2.2. Map of all provinces and departments in Peru mentioned as being home to Taki Onqoy
practices. ....................................................................................................................................... 74
4.1. Location of Iglesiachayoq and the Chicha-Soras Valley between Ayacucho and Apurímac.
..................................................................................................................................................... 141
4.2. Bosque de Piedras. Source: Javier Vargas. .......................................................................... 142
4.3. Photo of Carhuarazo. Source: Google Earth. ....................................................................... 143
4.4. Location of the different zones of the Chicha-Soras Valley. ............................................... 146
Figure 4.5. Map of all sites documented by Meddens and Mallco within the Chicha-Soras Valley,
n=61. ........................................................................................................................................... 152
4.6. Preceramic and Initial Period sites in the Chicha-Soras Valley, n=6. ................................. 154
4.7. Examples of Muyo Moqo style Phase C-D pottery. Photos by Joel Grossman. .................. 155
4.8. Early Horizon and Early Intermediate Period sites in the Chicha-Soras Valley, n=2. ........ 157
4.9. Middle Horizon Epoch 1 Sites in the Chicha-Soras Valley, n=4. ....................................... 159
4.10. Middle Horizon Epoch 2 Sites in the Chicha-Soras Valley, n=9. ..................................... 163
4.11. Plan map of Chiqna Jota. Map by Meddens (2010)........................................................... 164
4.12. Outside facade of Charrangochayoq (Photo from Kurin 2016). ........................................ 165
4.13. Human remains on the surface at Charrangochayoq (Photo from Rafael Mallco 2013). .. 166
4.14. Middle Horizon Epochs 3-4 Sites in the Chicha-Soras Valley, n=14. .............................. 168
4.15. Late Intermediate Period 1 sites in the Chicha-Soras Valley, n=25. ................................. 171
4.16. Late Intermediate Period 2 sites in the Chicha-Soras Valley, n=30. ................................. 172
4.17. Late Horizon sites in the Chicha-Soras Valley and environs, n=18. ................................. 175
4.18. Sectorization of Iglesiachayoq. .......................................................................................... 177
4.19. Church altar facing east. .................................................................................................... 179
4.20. Cut-stone Inka masonry, photo facing north...................................................................... 180
4.21. Parallel structure in the "Wallpawiri" Sector 2 of Iglesiachayoq, facing north. ................ 181
xiii
4.22. Side entrance of the church at Soras. Notice the Inka architecture lining the accessway.
Photo by Kathe Schreiber 2010. ................................................................................................. 184
4.23. Map showing potential Chanka confederation and relationship to Soras. Map from Kellett
2010............................................................................................................................................. 191
4.24. Area of research with modern towns, tourist attractions, and Iglesiachayoq. See Soras and
San Pedro de Larcay in the northern part of the valley............................................................... 204
5.1. Sectorization of Igleiachayoq. ............................................................................................. 220
5.2. Excavation Units at Iglesiachayoq. ...................................................................................... 224
5.3. View of Iglesiachayoq looking north. Notice the tilled fields in the foreground south of the
church. Photo by Monica Barnes. ............................................................................................... 226
5.4. Example of Google Earth imagery of Iglesiachayoq. Church and kuraka house in the center,
surrounded by heavily-terraced hillsides. ................................................................................... 227
5.5. Geogreferenced Barnes map--while spatial relationships are more or less rendered
accurately, the map itself is skewed. Structures mapped by PATO are in red. .......................... 228
5.6. Work space of map in GIS Pro, attributes on the left. ......................................................... 230
5.7. Units, sizes, and UTM coordinates for Sector 1 excavations. ............................................. 235
5.8. Unit 1 with quad subdivisions.............................................................................................. 237
5.9. Image of the test pit placed at the entrance to the kuraka house faing north. Photo by Frank
Meddens. ..................................................................................................................................... 239
5.10. Units, sizes, and UTM coordinates for Sector 2 excavations. ........................................... 242
5.11. Units, sizes, and UTM coordinates for Sector 3 excavations. ........................................... 245
5.12. Prehispanic or "local” plaza in Sector 3 facing north. ....................................................... 246
5.13. Example of a metal artifact drawn in Adobe Illustrator. ................................................... 252
5.14. Map of structures and water sources at Iglesiachayoq....................................................... 258
6.1. Majolica sherds produced in Peru. Picture from Kelloway et al. 2018. .............................. 269
6.2. Early green glaze wares from Peru. Photo from VanValkenburgh et al. 2017. ................... 269
6.3. Sectorization of Iglesiachayoq. ............................................................................................ 274
6.4. Some of the 16 aligned niches, looking north...................................................................... 276
6.5. Exterior of the northern wall of the church, photo looking south. ....................................... 277
6.6. Three-step altar facing east, with window at the top of the gable. ...................................... 278
6.7. Possible remains of the bowl which contained holy water at Iglesiachayoq. ...................... 280
xiv
6.8. Font at Iglesia de Pampachiri. Notice the similarities in shape and size with the bowl in
Figure 6.7. ................................................................................................................................... 281
6.9. The 100m radius is in green. The structures in red are the ones with squared corners
distributed throughout Iglesiachayoq. ......................................................................................... 284
6.10. 16-niched hall in Sector 2, facing west. ............................................................................. 286
6.11. Circular structure surrounded by semi-circular "local" plaza. ........................................... 288
6.12. View of the northern retaining wall of the "local" plaza. Notice incorporation of large
stones and inability to see beyond hill. ....................................................................................... 288
6.13. Typical uncoursed masonry style with faced stones lining the entrance, photo facing north.
..................................................................................................................................................... 291
6.14. Inka cut-stone architecture stacked vertically lining the entrance of an important household,
photo facing north. ...................................................................................................................... 292
6.15. Example of a well-preserved ovoid structure with access to the northeast. ...................... 292
6.16. Church with visible gables, photo facing north. ................................................................ 295
6.17. Boxplot comparing the three structure forms by dwelling area, ovoid n=23; quadrangular
with rounded corners n=58; quadrangular n=10; total population with 100% coverage n=91. . 295
6.18. Boxplot comparing the three structure forms by wall height, ovoid n=23; quadrangular with
rounded corners n=58; quadrangular n=1-; total population with 100% coverage n=91. .......... 296
6.19. Layout of quads in Unit 1. Fully excavated quads included 1, 3, 5, 7, 8, 10, and 12. ....... 301
6.20. Caret-head nail found in the northwest corner of the kuraka house. ................................. 303
6.21. Anthropomorphic mango handle from a plate. .................................................................. 303
6.22. Front side of 1892 cross. .................................................................................................... 304
6.23. Reverse side of the 1892 cross. .......................................................................................... 305
6.24. Drawing of 1892 cross. ...................................................................................................... 305
6.25. Fish head mango characteristic of Inka or Soras plates. .................................................... 307
6.26. Example of mango from Unit 2. ........................................................................................ 310
6.27. Blistered/vitrified ceramic fragments with Type E paste from Unit 2............................... 311
6.28. Sherds per excavated cubic meter in ten household units. Notice Units 2 and 3 with the
least material. .............................................................................................................................. 314
6.29. Weight in grams of ceramic per excavated cubic meter in household units. ..................... 315
6.30. Figurine from Unit 3. ......................................................................................................... 316
xv
6.31. Lithic possibly made of silex. ............................................................................................ 317
6.32. Carbonized maize and camelid vertebrae in situ. .............................................................. 318
6.33. Chert blades from Unit 4.................................................................................................... 320
6.34. Red and cream aribalo (left), red and black aribalo (right)................................................ 320
6.35. Two ollas with diamters over 35cm. .................................................................................. 321
6.36. Unit 5 after excavations, showing the shape of the structure. ........................................... 323
6.37. Metal tupu head recovered in Unit 5.................................................................................. 325
6.38. Possible hematite. .............................................................................................................. 326
6.39. Ceramic fragment with oxidation. ..................................................................................... 327
6.40. Quartz knife from Unit 6.................................................................................................... 329
6.41. Small metal pendant from Unit 6. ...................................................................................... 329
6.42. Straight-walled vessel, perhaps a kero. .............................................................................. 332
6.43. Middle Horizon 2 fragment. .............................................................................................. 332
6.44. Iron door hardware. ............................................................................................................ 333
6.45. Two complete vessels found in Unit 9, left: olla, right: jar. .............................................. 337
6.46. Prepared floor in southern half of Unit 9 facing south. ..................................................... 338
6.47. Two views of the polished quartz. ..................................................................................... 340
6.48. Locus 153 hearth with stained soil and containing a camelid mandible and carbon. ........ 342
6.49. Caret-head iron nail from the first half of the sixteenth century. ....................................... 343
6.50. Example of welded tuff collected in high density from Unit 16........................................ 349
6.51. Sandstone slab with divets. ................................................................................................ 350
6.52. Percentage of identified vessel types by sector.................................................................. 356
6.53. Percentage of identified vessel types by structure form. ................................................... 357
7.1. Mummified Inka king being carried on a litter by the Inka. Source: Guaman Poma de Ayala
1615: 258v. ................................................................................................................................. 369
7.2. Locations and burials in church units. ................................................................................. 388
7.3. Primary extended burials (in blue) recovered at Iglesiachayoq. .......................................... 392
7.4. Primary extended burial and metal pendant......................................................................... 393
7.5. Burial goods interred with extended burial. ......................................................................... 393
7.6. Sarcophagus with leg bones showing. ................................................................................. 394
7.7. Location of primary flexed burials in the church................................................................. 396
xvi
7.8. Primary flexed interment abutting western wall of church. ................................................. 397
7.9. Primary flexed interment with tupus and teacup. ................................................................ 398
7.10. Left: matching tupu pins. Right: miniature ceramic teacup. .............................................. 398
7.11. Secondary extended burial beneath the altar. .................................................................... 400
7.12. Secondary extended individual. ......................................................................................... 401
7.13. Secondary flexed individuals in the center of the church. ................................................. 402
7.14. Left: Photo of interment, right: drawing of interment. ...................................................... 403
7.15. Secondary flexed burial. .................................................................................................... 404
7.16. Locations of removed burials int he western portion of the church furthest from the altar.
..................................................................................................................................................... 405
7.17. Tupus associated with the disinterment of Individual 909. ............................................... 406
7.18. Left: indented area of loose sand among the compact soil matrix. Right: excavated pit filled
with larger stones. ....................................................................................................................... 407
7.19. Disturbed burials in clue and indeterminate burials in red. ............................................... 409
7.20. Left: inverted and backwards cranium. Right: articulated, extended lower extremities. .. 410
7.21. Left: photo of Individual 962. Right: drawing of remaining disturbed bones from 962. .. 412
7.22. Left: Photo of Individual 1005. Right: Drawing of 1005. ................................................. 413
7.23. Vertical orientation of juvenile cranium, adult leg bones, and tupus. ............................... 415
7.24. Top: sewing needle. Bottom: unorthodox tupu.................................................................. 416
7.25. Mixed cranium and dental fragments, with tupu pin in the center. ................................... 418
7.26. Burial types throughout the church. ................................................................................... 421
7.27. Burial types by spatial location. ......................................................................................... 422
7.28. Number of Burials by age class and unit. .......................................................................... 427
8.1. Map of central church complex with structures and plaza mentioned in the text................ 446
8.2. Prehispanic plaza in southern sector (Sector 3). Notice the large boulders incorporated into
the built environment. ................................................................................................................. 449
8.3. Floor-level arched niches and rectangular niches in Structure 107, photo looking west. ... 452
8.4. Sector 2 layout with Structure 107 and units. ...................................................................... 453
8.5. Multiple viewshed from all structures in Sector 1. Red circle: local plaza and location of
proposed Taki Onqoy performances. .......................................................................................... 460
8.6. Multiple Viewshed results from Sector 3. ........................................................................... 462
xvii
8.7. Sector 2 multiple viewshed. ................................................................................................. 463
8.8. Cumulative viewshed from all structure accessways to all portions of the site. Those regions
which were visible from the most other observer points are in red, and those which were visible
from the least observer points are in red. .................................................................................... 465
8.9. Central plaza with three accessways designated by stars. ................................................... 470
8.10. Prehispanic plaza with entrances designated by stars. ....................................................... 471
8.11. Left: Map of routes used by rectilinear structures to the central plaza, n=9/9. Right: walking
times from structures to plaza. .................................................................................................... 473
8.12. Boxplot distribution of walking times from rectilinear structures to the central church, one
outlier removed, structure 107 with a time of 278 seconds; n=8. ............................................... 474
8.13. Left: Routes from the southwest sector to the church, all utilizing the eastern accessway
and being directed to walk the length of the church. Right: walking times from Sector 3
structures to the church, n=34/34 structures. .............................................................................. 475
8.14. Boxplot distributions comparing walking times to the central church between rectilinear
structures and Sector 3 structures, rectilinear structures n=9; sector 3 structures n=34. There was
a significant difference in the times for rectilinear structures (mean =73.43s) and Sector 3
strucures (mean = 257.23s); t=-7.65021, p-value is <.00001. .................................................... 476
8.15. Left: Map of routes from rectilinear structures to prehispanic plaza. Right: travel times
from rectilinear structures to local plaza, n =10/10. ................................................................... 477
8.16. Boxplot distribution of walking times from rectilinear structures to prehispanic plaza, one
outlier removed, structure 107 with a walking time of 393 seconds. ......................................... 478
8.17. Left: Map of routes from Sector 3 structures to prehispanic plaza. Right: travel times from
Sector 3 structures to prehispanic plaza, n =34/34. .................................................................... 479
8.18. Boxplot distributions comparing walking times to the prehispanic plaza from rectilinear
structures and Sector 3 structures, rectilinear structures n=10, sector 3 structures n=34. A Mann
Whitney U Test was conducted to compare walking times to the prehispanic plaza from
rectilinear structures and sector 3 structures. There was a significant difference in the scores,
with a z-score of -4.38288, and thus the difference in walking times was significant at p<.05.. 480
9.1. Sign in Pampachiri advertising "las rutas del Taki Unquy."…………………………….500
xviii
CHAPTER 1
CONQUEST, CONVERSION, AND REVITALIZATION
European colonization of the Americas was undoubtedly one of the most impactful and
disruptive processes in human history, spurring the traumatic emergence of a new global
economy, while indelibly reshaping indigenous lands and cultural traditions. European invasion
was a protracted event which caused ruptures in indigenous practices: in the years immediately
following conquest, European and indigenous groups entered into a process of mutual
contestation and negotiation over the structures of new government, religious systems, cultural
norms, and daily practice. Scholarly research on the colonial Andes has relied overwhelmingly
on written documents, which systematically exclude native voices. Reflecting these biases,
twentieth-century scholars monolithically cast Europeans as dominant conquerors and native
peoples as either resistors or assimilators to European cultures (Liebmann and Murphy 2011;
Restall 2003). More recent archaeological research in the Americas recognized the diverse array
of indigenous responses to European colonization. This dissertation examines the household and
religious practices constituting a nonviolent native response to Spanish colonialism in highland
Peru—the 1560s revitalization movement known as Taki Onqoy (Quechua: “dancing sickness”).
The Spanish invasion of South America in 1532 triggered radical, irreversible
transformations as people with profoundly different cultural practices confronted a new “other.”
The Spaniard Francisco Pizarro led 168 conquistadores in the swift capture, ransom, and
eventual execution of the Inka ruler, Atahualpa, essentially ending the powerful Inka Empire and
its control over diverse Andean groups stretching across Western South America (D’Altroy
1
2015; Hemming 1970; Stern 1993). Yet, defeating the Inka was only the first step for the Spanish
in a protracted process of learning how to conquer, govern, and evangelize Andean people. In the
subsequent decades, Spanish governmental and church authorities and Andeans became
entangled in seemingly opposing belief systems, cultural norms, and daily praxis. Understanding
of these processes of cultural negotiation during the sixteenth century has been shaped by the
affordances and limits of extant Spanish accounts, governmental documents, and ecclesiastical
texts. In the first decades after Spanish invasion, this corpus of documents was particularly onesided, since it took time for Andeans to learn the Spanish system of writing and legal procedures
(Durston 2007; Salomon 2011).
Over the last several decades, archaeological studies of colonial-era and transconquest
(sensu Wernke 2007) settlements have supplemented the critical gaps in colonial documentation
(Abraham 2017; deFrance 1996, 2003; deFrance et al. 2016; Kennedy and VanValkenburgh
2016; Klaus 2008, 2009, 2013; Murphy et al. 2011; Quilter 2011; Rice 1993, 1994, 1996, 2012a,
2012b; Van Buren 1999, 2010; VanValkenburgh 2012, 2017a, 2017b, 2019; Wernke 2012,
2013). These studies allow more dedicated interpretation of the daily practices of conquered
individuals, and specifically the ways in which these individuals navigated the changing colonial
landscape. Archaeological studies of the colonial period demonstrate that colonialism was not a
unilateral acculturative process for conquered groups. European invasion created an entirely new
South American way of life, in which Andean practices were adopted by colonizers and Andean
traditions were entangled with Catholic and European praxes.
This dissertation builds upon and adds to a growing body of research on the tumultuous
transconquest (Late Horizon through Spanish invasion and colonial resettlement) era through an
archaeological study of Taki Onqoy. Heretofore known only through a suite of documents
2
produced by Spanish priests in the second half of the sixteenth century, Taki Onqoy was a
religious revitalization movement enacted by Andean preachers in explicit engagement with and
purported rejection of Catholic evangelization. According to these documents, Taki Onqoy
preachers traveled throughout the highlands of south-central Peru, advocating for the rejection of
Catholicism and other Spanish traditions including Spanish foods, dress, and cultural practices
(Albornoz 1990 [1569, 1570, 1577, 1584]; Álvarez 1998 [1588]; Molina 2010 [1574]; Yaranga
1978). These preachers asserted that Andean huacas (local deities)1 had been neglected by their
communities and/or destroyed by Spanish conquistadores. Three decades after Spanish invasion,
these huacas were resurrecting in order to unite and fight a battle to defeat the Spanish Catholic
God, thereby returning Andeans to an imagined utopian time prior to Spanish colonialism. The
stakes were high: by some accounts, Taki Onqoy was on the cusp of toppling the colonial order.
But given the total absence of archaeological research on Taki Onqoy, almost nothing is known
about its concrete manifestation, or its material and spatial dimensions more broadly. This
dissertation addresses this issue, presenting the results of my excavations at Iglesiachayoq
(Chicha-Soras Valley, Ayacucho, Peru), a Late Horizon through Early Colonial Period site
which was branded by documentary records as a Taki Onqoy center (Albornoz 1990). The
findings of these excavations both inform us about the materiality of Taki Onqoy and enrich the
hermeneutic circle for interpreting the documentary sources. An overview of historiographical
debates about the movement follows below.
Taki Onqoy Source Critiques and Historiography: A Brief Assessment
1
For the introduction, I use this simple clarification for huaca, but see Chapter 2 for a more thorough discussion of
pre-conquest Andean religion.
3
In my study of Taki Onqoy primary sources and secondary debates, I have assessed the
works of four authors—all Spanish priests—who use the phrase “Taki Onqoy” to refer to the
movement. The first direct mention of Taki Onqoy was written by a priest named Guerrero2 in
1564 in a letter to the priest Luis de Olvera (Yaranga 1978). This letter was rediscovered in
January of 1950 in the butler’s quarters of the church of Huancaraylla, in the Department of
Ayacucho. In this letter, Guerrero demonstrated concern for a growing population of “Indians”
who worshipped their huacas and whose bodies housed these huacas, manifested through
heaving, trembling and drinking. He also warned that “they offered chicha and at midnight they
performed ceremonies which they call Taki Onqoy, singing in a round, and preached great
abominations against our god, against our religion, and against the Spaniards” (Yaranga 1978:
168).3
By far, the lengthiest sources discussing Taki Onqoy were written by Cristóbal de Molina
and Cristóbal de Albornoz, and it is largely from these two authors that we understand the
practices associated with Taki Onqoy. In the final chapter of his Relación de las fábulas y ritos
de los incas (2010 [1574], hereafter Relación), Molina briefly discusses the motivations and
practices indicative of Taki Onqoy.4 He suggests that the movement began in the Inka holdout of
Vilcabamba, and that it was defined by the rejection of Catholic faith, fasting (from salt, pepper,
sex, colored maize, or Spanish foodstuffs), boycotting Catholic church, sweeping and preparing
houses for the huacas, trembling and rolling on the floor, huaca spirit possession, utilization of
“Guerrero” is the only reference I have been able to find to this priest—I do not know his full name.
Translated from “quemaban los bollos y derramaban chicha y a media noche fazían ceremonias que llaman taqui
ongo, cantando a la rronda, y predicavan grandes abonminaciones contra dios nuestro señor, contra nuestra religión,
y contra los españoles” (Yaranga 1978: 168), translation mine.
4
While he does not give the movement the name Taki Onqoy in this chapter, he acts as a witness for Albornoz in his
Information of Services (1990 [1577; 1584]), where he identifies the movement by name. Most scholars attribute the
“apostasia” Molina refers to in his Relación as Taki Onqoy (Bauer 2011).
2
3
4
special rooms with straw and blankets for possessed individuals, consumption or worship of
rams, colle, chicha, llipta, mollo, dancing and drinking without sleeping, and sacrifice of rams
and cuyes (Molina 2010: 97).5 Molina’s account is broad, and based mostly on conversations
with the parish priest of Parinacochas, Luis de Olvera, who is credited by Molina to be the
discoverer of the movement (Molina 2010 [1574]: 94).
While Molina’s account is stereotypic and general in its descriptions of Taki Onqoy
participants, that of Cristóbal de Albornoz is meticulous and exhaustive. In his Informaciones de
servicios (1990 [1569, 1570, 1577, 1584], hereafter Informaciones) and Instrucción para
descubrir todas las guacas del Pirú y sus camayos y haziendas (1989 [1581-85], hereafter
Instrucción), Albornoz records the testimonies of hundreds of witnesses and the means by which
one might discover and destroy specific huacas, respectively. Though these witnesses answer a
prescribed set of questions, they often add additional details regarding performative aspects of
the movement, its geographic extent, and Albornoz’s role in extirpating the movement. Albornoz
himself also recorded how he found huacas, and his punishments for the various ranks of
takiongos (to be discussed thoroughly in Chapter 2).
After the publication of Albornoz’s Informaciones in 1964 and 1973 (Millones) and his
Instrucción in 1971 (Duviols), scholars of Peruvian studies began to engage in a spirited debate
regarding the “true nature” of Taki Onqoy, its territorial limits, what it meant to Andeans, and
even whether or not it was an actual, impactful movement. Utilizing only these few primary
sources, Taki Onqoy scholars have arrived at diverse interpretations of the movement given the
vagaries of the Spanish documentation. As will be discussed in depth in Chapter 2, secondary
5
These practices (and Molina’s discussion of them) are discussed in more detail in the following chapter.
5
scholars initially took Molina and Albornoz at face value—they interpreted the movement to
have been of wide geographic expanse, associated with the Inka of Vilcabamba, and one of the
greatest threats to the new overseas Spanish colony (Duviols 1971; Millones 1964 and 1973).
Millones and Duviols also viewed the movement as a predominately Inka phenomenon, and one
that was explicitly against both Spanish and Andean religious traditions.
Later Taki Onqoy historians turned away from the Inka-centric interpretation of Taki
Onqoy for several reasons. First, colonial historians shifted toward cultural histories,
emphasizing bottom-up historical narratives, and scholars began to deemphasize state-based
studies of the preconquest Andes in favor of investigation of specific Andean groups (Mumford
1998, see Spalding 1984 and Stern 1993 for examples). Second, archaeological evidence of the
Inka Empire affirmed that the empire was relatively short-lived. Lastly, historians of Taki Onqoy
recognized the absence of the Inka Sun God in Taki Onqoy documents (Mumford 1998: 157).
Reflecting the de-emphasis of the Inka Empire in favor of the study of regional Andean
ethnic groups, the subsequent generation of work on Taki Onqoy viewed the movement in a
variety of ways. Some broadly cast the movement as a resurgence or revitalization of Andean
cultural values prior to Inka and Spanish conquest (Curatola 1976; MacCormack 1991; Pease
1973). Conversely, other scholars argued that Taki Onqoy was a form of cultural illness that
proved ultimately ineffectual (Lemlij et al. 1990; Stern 1993; Wachtel 1977). In 1990, Peruvian
historian Luis de Millones published an edited volume entitled El Retorno de las Huacas in
which two of the contributors (Guibovich 1990; Varón 1990) put forth detailed and groundbreaking biographical studies of Cristóbal de Albornoz. These two authors suggested first, that
Albornoz may have exaggerated parts of his accounts of Taki Onqoy in order to receive a
6
promotion from the Spanish Crown, and second, that there was actually no connection between
Taki Onqoy and the Inka holdout at Vilcabamba.
In 1992, Juan Carlos Estenssoro explicated the phenomenon of dance emphasized by the
sources, and depicted the movement as the result of a feud between secular clergy (like
Albornoz) and their Dominican counterparts. In the initial years of evangelization, Dominican
clergy were criticized by secular (state-appointed) clergy for being too accommodating in their
conversion practices—Dominicans initially allowed autochthonous practices, dances, and
languages to be incorporated into Catholic religious rites. Perhaps indicative of a broader contest
between secular and mendicant clergy over who had the right to convert Andeans, and thus, who
had ecclesiastical power, the more accomodationist Dominicans were blamed by subsequent
secular clergy for the continuation of Andean religious practices considered idolatrous. Similar
to the revisionist perspective of Estenssoro, Gabriela Ramos proposed that Taki Onqoy may not
have ever existed but that instead, it was entirely fabricated by Albornoz (Ramos 1992).
Albornoz’s Informaciones are organized into four letters he wrote and sent to the Spanish
Crown, asking for promotion into specific clerical positions in the New World (Heilman 2002).
In these four accounts, Albornoz provides testaments as to his extirpation actions in the regions
of Huamanga, Arequipa, and Cuzco as evidence that his experience in Peru and his dedication to
spreading Catholic faith (and simultaneously stamping out Andean religious practices) made him
an ideal candidate for higher positions within the church hierarchy. Ramos utilizes these letters
to argue that Albornoz would have fabricated much of his exploits in order to achieve a
promotion.
After this critical period of scholarly interest in the sixteenth-century movement, there
were a few other text-based interventions—in 2001, Ranulfo Cavero Carrasco contributed an
7
ethnohistorical text called Los Dioses Vencidos which analyzed tropes in the Albornoz
documents. Subsequently, in 2002, Jaymie Heilman countered Ramos’s pointed piece from
1992, and Ramos quickly responded, doubling down on her initial argument (Heilman 2002;
Ramos 2002). Most recently, Peter Gose’s 2008 work Invaders as Ancestors reaffirms the initial
arguments by Millones and Duviols, to the effect that Taki Onqoy was definitively related to the
Vilcabamba revolt, thus bringing historical inquiry into Taki Onqoy full circle.
Archaeological Interventions
The extensive debate amongst historians over multiple dimensions of Taki Onqoy reflects
long-term trends in Peruvian ethnohistorical historiography, and provides an array of
interpretations and viewpoints derived from the documents. However, it is fair to say that the
debate has met impasses that reflect on the limits of the textual sources themselves.
Archaeological investigation can help break open this epistemological loop. This dissertation
makes this contribution as the first archaeological study of Taki Onqoy, focusing on an avowed
center of Taki Onqoy activity, the site known today as Iglesiachayoq (Ayacucho, Peru), which
corresponds to the historical settlement of Chicha in sixteenth-century documentation (Figure
1.1).
8
Figure 1.1 Map of Chicha-Soras Valley with modern towns and Iglesiachayoq (red triangle).
Albornoz specifically identified Chicha as a center of Taki Onqoy belief and practice, and
the relict settlement is still remarkably well-preserved, with standing fieldstone buildings,
including an early colonial church. Indeed, the church is the namesake of the site today:
Iglesiachayoq being an appropriately hybrid Spanish/Quechua word combining the Spanish
Iglesia (church) with the Quechua nominal diminutive suffix –cha and the Quechua comitative
suffix –yoq, thus “place with the little church”. Given its combined archaeological endowments
and relevant written textual sources, Iglesiachayoq provides a unique opportunity to consider the
materiality of Taki Onqoy. Several lines of evidence indicate that Iglesiachayoq is the historical
9
settlement of Chicha; these are presented in Chapter 2. By completing the first archaeological
project researching Taki Onqoy, I generate a critical new data source for understanding the
movement. My research methodology included architectural survey to enable exploration of the
spatial practices of Taki Onqoy, targeted excavation of 19 units in domestic and ritual contexts to
address questions of the materiality of the movement, post-field lab analysis of portable material
culture, and spatial analysis of the settlement and its surrounding landscape. My findings
demonstrate a diversity of dispositions and orientations concerning Taki Onqoy: the movement
was not a monolithic front that was adopted uniformly by community members. Excavation
results demonstrate remarkable continuity in domestic contexts from the Late Horizon (14501532 CE) to the Early Colonial Period. In contrast, evidence from burials in the church of
Iglesiachayoq suggests ambivalence or subversion of Catholic mandates. I argue that Taki
Onqoy was an ambiguous, and seemingly contradictory movement—it was anti-Spanish and
Catholic, yet shaped by Spanish Catholicism. The entangled interactions between Spanish and
Andean materials, places, and people, produced a variety of meanings to Andean participants,
and that these participants only variably adhered to Taki Onqoy dictates. Documentary records
and archaeological findings both affirm that the practices central to the movement were adopted
only by a subset of people in the region. Additionally, these practices were not uniform across all
spheres of human existence, but were most impactful in influencing mortuary treatment of the
dead. Andean peoples took a variety of stances in their negotiation of indigenous and Spanish
traditions.
10
Theoretical Orientation
As discussed in the previous section, historians who study Taki Onqoy have defined the
movement as made up of and reflecting a wide variety of social phenomena—nativism, illness,
revivalism, etc. I approach Taki Onqoy as a revitalization movement. Here I follow Wallace’s
influential early definition of a revitalization movement, as a “deliberate, organized, conscious
effort by members of a society to construct a more satisfying culture” (Wallace 1956: 265).6
Revitalization movements are explicit attempts by specific groups to reformulate their current
social circumstances such that societal norms return to a perceived idyllic past. These
movements are generally led by charismatic preachers promising a “revival” and have been
common throughout history—as Wallace affirms, “probably few men have lived who have not
been involved in an instance of the revitalization process…both Christianity and
Mohammedanism, and possibly Buddhism as well, originated in revitalization movements”
(Wallace 1956: 267). Thus, revitalization movements can be attributed to generating some of the
most influential and durable religious traditions. Since revitalization movements have the ability
to create dynamic social change within a compressed time scale and have been responsible for
numerous new religious traditions or social organizations throughout history, the concept is
useful for exploring Taki Onqoy.
Revitalization movements are generative and have explicit goals to create more equal
societies, but anthropologists have focused on these overarching objectives rather than
considering the associated practices enacted by followers of these movements (see Liebmann
2008 for discussion of material correlates of revitalization). Similarly, the societal conditions
Revitalization movements are often alternatively categorized as “nativistic movements, reform movements, cargo
cults, religious revival, messianic movements, utopian communities, revolutions, etc.” (Wallace 1956).
6
11
which cause these movements are often critically assessed while overlooking the movement
itself and its material signatures (Liebmann 2012). Yet, material media are central to cultural
change and thus documented revitalization movements should also be assessed archaeologically
in order to better understand the ubiquity of these movements today. This dissertation recognizes
revitalization as an option for performative, generative resistance, albeit one that was neither
uniformly adopted by those existing under colonial rule nor one that penetrated all aspects of
daily life. Specifically, this research examines how sixteenth-century Andean participation in the
documented revitalization movement of Taki Onqoy was variably adapted and in which realms it
was salient. In my work at Iglesiachayoq, I investigated the material markers of Taki Onqoy, and
more importantly, how the movement would have been enacted on the ground.
Building on the primary accounts of Albornoz and Molina, previous archaeological
research and theoretical debates on revitalization movements, and archaeological research of the
Early Colonial Period in Peru, my fieldwork included a detailed architectural survey, targeted
excavation, and artifact and spatial analysis to answer the following questions: Were material
and spatial signatures associated with Taki Onqoy located in private areas away from central
Spanish religious spaces? Did Taki Onqoy practitioners repurpose the use of Catholic objects in
new ways? Did they generate new or hybrid forms of material culture? Did Taki Onqoy
influence daily life and/or ritual realms? Was the Spanish response to Taki Onqoy at
Iglesiachayoq overtly violent and enacted in a public space? At a broader level, this project was
undertaken to provide a new perspective on how revitalization movements can be studied
archaeologically and how rapid religious change can cause ruptures in imperial agendas.
In this dissertation I argue that participation in the Taki Onqoy movement was uneven
and partial, and that daily domestic practices were unaltered by the movement. At first glance,
12
my findings demonstrate an apparent paradox in the realms of Taki Onqoy participation: in
private households away from the Spanish church, we found little evidence of Taki Onqoy
practice. However, it was in the region most visible to Spanish authorities (at least nominally),
the Catholic church, where Taki Onqoy practices evidenced a wide array of unorthodox
positionalities in mortuary treatment. Taki Onqoy was a movement fraught with ambiguity and
contradiction—it was both anti-Catholic, yet its premises were influenced by Catholic tenets. It
attempted to elide factional, ethnic, and status difference by appealing to an essential distinction
between “Spaniard” and “Andean,” thereby recasting the categorical foundation of Spanish
colonialism (Spaniard/Indian) as a means of emancipatory unification.
If the movement was entangled in the foundational ideology of Spanish colonialism in
the Americas, it was also constituted by entangled materiality. It was produced in situ through
people, places, and things interacting in assemblages that were both already and becoming
further entangled through the work of colonialism. During the colonial period, the meanings of
places and things transformed—places of Andean or Inka veneration could also be associated
with Catholicism. Religious practices often constituted both Andean and Catholic belief systems.
The meanings of things, and the ways in which things acted on people were predicated on
entangled contexts: in the Early Colonial Period, European and Andean things interacted and
influenced one another such that the very characteristics which made an object “European” or
“Andean” were shaped by the interaction between both cultural frameworks. Individuals and
daily praxis were also a product of this new entangled reality, and associated practices were
altered depending on contexts and goals. The shifting and amorphous Taki Onqoy—a movement
itself a product of the post-invasion negotiation of cultural frameworks at a local level—should
thus be explored through its entanglements and its contradictions. Material media and its
13
interaction with people and places were the means through which Taki Onqoy was constituted,
and my dissertation investigates these contextual, entangled materialities through archaeological
analysis.
Over the short term, Taki Onqoy was remarkably effective in redeploying Andean and
Spanish colonial practices to mount resistance to its depredations on some of the most closely
held cultural postulates regarding order, patrimony, and the right of rule. Over the long term,
internal contradictions (fluidity between Taki Onqoy and Catholicism, and disagreements
between kurakas and lower-status takiongos) interacted with external forces (extirpation
campaigns, the reducción program, and implemented sanctions from the Councils of Lima) to
bring Taki Onqoy to its public end. The multi-day dances, chanting, and drinking, along with
other Taki Onqoy practices, were driven underground by the end of the sixteenth century (see
Álvarez 1998 [1588] and Gose 2008 for later examples of idolatry reminiscent of Taki Onqoy).
Yet, even as Taki Onqoy was formally defeated as a social movement, its legacy reverberates to
the present.
In the Departments of Huamanga, Huancavelica, and Apurímac, the Taki Onqoy
movement is still a salient topic in contemporary consciousness (Figure 1.2). As study of Taki
Onqoy flourished in the second half of the twentieth century, Andeans in these regions adopted
the movement as an indigenous rallying cry, one which spoke to ongoing neglect and abuses by
the Peruvian state and the resistance of those who lived in these three poorest departments. My
conversations with those who live in this region found that people discuss Taki Onqoy and
Shining Path as thematically similar. Both movements arose in a time of economic and political
systemic oppression, and the survival of those who live in the region speaks to their resilience.
Taki Onqoy dances, and more broadly, prehispanic takis are regularly performed in the highlands
14
of Peru, and are heralded as authentic examples of the ingenuity and diversity of prehispanic
cultural systems. Many of the entanglements reflected in Taki Onqoy are enacted in
contemporary Peruvian festivals in honor of Catholic saints, and particularly the Corpus Christi
celebration (Dean 1999). Thus, in some ways Taki Onqoy was a foundational movement which
served as a blueprint for future Andean/Catholic performances.
Figure 1.2. Sign in Pampachiri depicting "las rutas del Taki Unquy".
Organization of the Dissertation
In Chapter 2, I survey the polarized and heated debates regarding Taki Onqoy, and
review its varied historiographical trends. In this chapter, I also generate archaeological
correlates for Taki Onqoy practices, according to the various source documents. In Chapter 3, I
15
assess Taki Onqoy as a revitalization movement, and consider what utility the revitalization
concept offers in understanding how the movement manifested in daily practices. I combine this
theoretical perspective with performance theory and theory of peasant resistance in order to
recognize how the movement would have been performed and experienced. Taki Onqoy was not
uniformly practiced nor did it take the same form in each location where it was adopted.
Moreover, Taki Onqoy was not performed by all individuals in a given settlement—differences
in social status, kuraka influence, and the consistency of Spanish presence all affected how and
why the movement emerged where it did. Finally, one of the major challenges in this dissertation
was considering which practices defined Taki Onqoy as its own distinctive and unique
movement as opposed to being part of a broader culture of pan-Andean practices which recalled
prehispanic belief systems. I contend that through application of theories of revitalization,
performance, and resistance theory, we can begin to recognize those practices which were
developed or intensified as a part of Taki Onqoy and demonstrate how Andean peoples generated
new behaviors in response to the hardships faced during Spanish conquest and control.
Chapter 4 reviews the documentary and historical record of the Chicha-Soras Valley and
relates the reasons behind the dearth of archaeological investigations into this area. Chapter 4
also considers the affordances and disadvantages of settlement in the Chicha-Soras Valley,
outlining the various waves of imperialism into the valley. Chapter 5 presents research questions
and methodology, and justifies why my multi-phase research design is ideally suited for my
research questions. This research methodology included 1) architectural survey, 2) targeted
excavation of households, plazas, and the central church, and 3) artifact, contextual, and spatial
analysis.
16
Chapters 6, 7, and 8 present the results of my archaeological excavations and spatial
analysis. In Chapter 6, I take a contextual approach and present the results of excavation in
domestic structures and patios at Iglesiachayoq. I argue that ceramic, lithic, and faunal data
demonstrate relative continuity in daily life throughout the site’s occupation, albeit with clear
differences in activities between each sector. While we did recover evidence of Old World fauna
(horse, cow, sheep, etc.), the quantities of these remains suggest isolated instances of
consumption rather than a pattern of changing foodways. In Chapter 7, I present the results of
excavation in the Catholic church at Iglesiachayoq. I contend that the variety of interment
styles—primary or secondary, flexed or extended, disturbed and removed—indicate a multitude
of positions toward Catholicism in relation to Taki Onqoy. Finally, Chapter 8 integrates results
of architectural survey, excavation data, and spatial analysis to propose that Taki Onqoy rituals
were likely clustered in the local portion of the site where they would be both nearer to natural
features of local importance and out of view from Spanish authorities or Catholic converts.
Chapter 9 concludes this dissertation with a synthetic discussion regarding Taki Onqoy
materiality and practices.
17
CHAPTER 2
THE HISTORY AND ARCHAEOLOGY OF TAKI ONQOY
Introduction
Taki Onqoy was a sixteenth-century revitalization movement predicated on the strategic
veneration or rejection of things, practices, and belief systems. Andean ontological precepts held
that specific objects could be huacas who had the power to bring about a cultural revolution
culminating in a return to pre-Inka Andean independence (Bray 2009, 2012, 2015; Chase 2016;
Mannheim 1986, 1991). Other objects, however, were constitutive of “Hispanic” culture and
were rejected by Taki Onqoy practitioners. Through performance of Andean takis (Quechua:
dance/chant; Estenssoro 1992), and adoration of the materials associated with those takis,
takiongos (Taki Onqoy practitioners) promised a world reversal in which Andean huacas would
“resurrect” and defeat the Catholic God (Molina 2010 [1574]).7 While the movement itself was
enacted and performed through specific materials in uniquely constructed spaces, study of the
Taki Onqoy has heretofore exclusively been relegated to Spanish documentation from the late
1500s produced by several Spanish priests trying to extirpate the movement. In this chapter, I
briefly discuss Andean religion prior to Inka and Spanish conquest, the changes in this suite of
religious practices which were caused by Inka conquest, and then introduce the historical context
of Taki Onqoy and how it related to Spanish Catholic attempts at evangelization and conversion.
Next, I relate the central themes in primary accounts of the movement, especially the different
materials associated with Taki Onqoy, and how these materials and their usage constituted the
7
Taki Onqoy practitioners would not have been old enough to have existed during both Inka and Spanish conquest,
so their “return” to prehispanic religious practices was predicated on oral tradition and influenced by several decades
of Spanish presence, rather than a first-hand remembrance of what these practices consisted.
18
movement. Then, I outline debate regarding Taki Onqoy among contemporary scholars. These
primary accounts, though ambiguous and written from an imperial perspective, also provide
clues as to the materials and practices indicative of the movement. Through this historical
background, I argue that the Taki Onqoy movement (and Andean religions in general) must be
studied through things, and not exclusively through the lens of sixteenth- and seventeenthcentury chronicles. By investigating the materiality of Taki Onqoy, scholars can build upon the
critical work of historians of this time period through the advancement of knowledge of native
and Spanish practices in the immediate period following conquest, the strategies of the initial
period of evangelization, and understanding of the Andean practices which constituted Taki
Onqoy.
The Taki Onqoy revitalization movement has long symbolized indigenous “resistance” or
“resilience” in the face of Spanish conquest and conversion (Duviols 1971; Millones et al.
1990).8 Yet, a bottom-up perspective on how the movement was implemented and practiced, and
how materials were essential to the movement is curiously absent. Archaeological practice
allows the recovery of things and objects, but the significance of these things was not limited to
their physical characteristics—meaning is understood through cultural structures and interactions
with objects. Utilizing the contexts provided by documentary records and archaeological
investigation of the period, I argue that a nuanced exploration of Taki Onqoy must consider
performance and materiality. By studying how individuals interacted with materials and places,
scholars can ask more specific questions about how Taki Onqoy was performed, where these
8
Its connection with the danza de las Tijeras has strengthened the link between specific dances and materials as
symbols of resistance within especially poor departments such as Ayacucho, Apurímac, and Huancavelica
(Arguedas 1962).
19
performances took place, and perhaps even who were performing Taki Onqoy rituals in the
Andean highlands.
I place primary documentation, historical scholarship on evangelization, and
archaeological study of revitalization in dialogue to consider how community members
interacted with Andean and Spanish things at Iglesiachayoq, thereby constituting the Taki Onqoy
movement. I begin by defining my terms—things, materiality, animacy, and entanglements—and
then provide a review of pre-conquest Andean religion, Inka-era religion, and Spanish
evangelization. In this chapter, I also relate difficulties and problems with conversion, and how
the Councils of Lima sought to address these issues. Next, I will utilize the primary sources and
secondary debate on Taki Onqoy to propose models and hypotheses for how to learn about Taki
Onqoy from its materials. These models and hypotheses are grounded in the primary sources,
revitalism theory, and other colonial era archaeology which has been undertaken throughout the
Americas. Though my study is situated in the central highlands of Peru, revitalization
movements have been documented throughout the Americas during colonial encounters—
ethnographic accounts and archaeological investigations of these movements are invaluable in
understanding Taki Onqoy materials.9 By its nature, archaeology cannot provide evidence of the
thought processes or religious motivations behind Taki Onqoy. However, archaeology can lend
insight into the practices associated with the movement which structured and shaped cultural
frameworks. Was Taki Onqoy practiced in distinctive sectors of the site? Is there evidence of
Taki Onqoy in particular households? Was it a covert or an overt rebellion? Was “rejection” of
Spanish goods and religion all-encompassing or did takiongos object to specific materials,
practices, bodily consumption, etc.?
9
For Ghost Dance, see Andersson 2008; Du Bois 2007; Harkin 2004; Kehoe 1989. For the Pueblo Revolt see
Liebmann 2007, 2008, 2012.
20
Things, Materiality, Animacy, and Entanglement
To explore the relationship between things, humans, and places, and their dialectic
relationships to one another, defining how I use these terms is necessary. In considering things, I
draw on Hodder’s 2012 definition of a thing as “an entity that has presence, a configuration that
endures, however briefly” (Hodder 2012: 8, see also Heidegger 1971, 1973, 2002). Things work
to connect with other things, people, and places. They have life histories in which these
connections are changed, when the thing comes into use or falls out of it—things are relational
and constructed through their effects on other things and entities. Within my discussion of my
archaeological research, I consider things as entities made or utilized by human actors (following
Hodder 2012: 15). It is the interaction between humans and things which allow humans to exist,
to be social beings, and to enact belief systems and practices.
Things are connected to humans through materiality. Following Meskell, materiality is “a
set of cultural relationships…imbued matter and embodied objects exist in relationship to the
specificities of temporality, spatiality, and sociality” (Meskell 2005: 6). Things, then, are not
reducible to their physical properties, but instead must be studied in specific cultural moments
“to understand particular contextual notions of the material world and its propensity to forge,
shape, interpolate, and possibly even challenge and undermine social relations and experiences”
(Meskell 2005: 6). Taki Onqoy was enacted in a time wrought by dramatic cultural change—
Andean things took on new meanings, becoming signifiers of not just Andean-ness, but explicit
indicators of resistance to Spanish-ness. By focusing on the thing-ness of Taki Onqoy artifacts,
how these things were imbued with new meanings in the 1560s, and how these meanings were
both influenced by the Spanish, but also anti-Spanish, I aim to explore the nuances of Taki
21
Onqoy—how it was constituted by and acted through things—and thus offer a new, updated
perspective on the movement apart from its documentary presence.
Apart from their mere existence or presence, things can also be animate, a trait which is
particularly common in the Andes (Abercrombie 1998; Allen 2002; Bray 2015; DeMarrais 2017;
Swenson 2014). In other (non-western) ontological orientations, to be animate or to have
animacy offers that things are alive and imbued with agency (Allen 2015). Things can have their
own points of view, experiences, histories, and the ability to be beings and persons. In the Andes,
this animacy is a trait not only of objects (cloth, stones, vessels, etc.), but also of places
(mountain peaks, lakes, rivers) and of animals. Animacy of things, like materiality, must also be
studied in context—things can be animate at different times, and their animacy can be
transformed as the thing itself is transformed. For example, as discussed below, during Taki
Onqoy, Andean huacas in the forms of stones were transformed such that they inhabited humans,
thereby transferring their power to these humans—the embodied individuals literally were
huacas.
Building on recent archaeological perspectives exploring interactions between things,
places, and humans, I utilize the term entanglement as the medium through which human and
non-human entities are constructed and transformed (Der and Fernandini 2016; Hodder 2012;
Silliman 2005, 2016). Entanglement as a concept or metaphor has been employed recently in
archaeological studies of colonial encounters, modifying or replacing concepts such as
“hybridity” or “syncretism,” terms which connote an unspecified combination of cultural factors
intermixing to form something different (see Dietler 2010; Jordan 2014; Hodder 2012; Silliman
2016). Moreover, entanglement allows archaeologists to recognizing things and people as
shifting and contextual—though artifacts may have been made by European actors, their use by
22
native peoples denies the simple classification of “European” (Silliman 2015). Most importantly
for my research, the concept of entanglement “can provide a view of how things, peoples, and
practices intertwine to produce new forms without requiring them to remain that way, and
without applying a term to the product that binds it to an entangled state” (Silliman 2016: 39).
That is, the concept of entanglement allows the archaeologist to consider specific things and their
interactions with people in a specific place and time—these moments of entanglement need not
be permanent, but can be recognized as part of a thing’s life history. Additionally, the concept of
entanglement allows recognition of network theory (Latour 2005). Entanglement thus shifts
questions about physical characteristics (color, weight, function) of things or identity (Spanish or
Andean, Catholic or Taki Onqoy) of things to more practical, essential notions: why were these
objects used, who were they used by, and how did they act on the user?
Andean Ontologies and Western Ontologies
To study Andean religion—its practices, its materiality, and how it changed throughout
conquest—is to recognize fundamentally different ontological orientations. Recent
archaeological studies have explored ontologies through emphasis on things (Bray 2009, 2015;
Henare et al. 2007; Meskell 2005). In the Andes, the ontological turn means to renounce Western
notions of dualist “animate” and “inanimate” categories of being and to understand objects as
subjects with their own points of view and their own agency (Allen 2015; Bray 2009, 2015;
Chase 2015, 2016; Mannheim 1991; Salomon 1991). These objects were not “imbued” with
power or animacy, but rather they innately contained a life force which manifested in a variety of
physical forms. Archaeologically, the ontological turn centers on materiality—how objects could
have acted on people—not merely on identifying what might be “sacred” from an Andean
23
perspective. In the following section, I establish the baseline religious practices and materials of
Andean religion in order to understand how Taki Onqoy renewed and simultaneously changed
these practices.
In the first decades after the conquest, Spanish authors were unable to comprehend
Andean ontology, and were particularly stymied by Andean epistemologies of animism and
personhood. In early documents, these authors struggled to convey the meanings of the Andean
concepts of camay and huaca—there were no equivalent terms in Spanish for the Quechua
concepts (Bray 2015; Salomon 1991). Salomon defines camay as “a concept of specific essence
and force, ‘to charge with being, to infuse with species power’” (1991: 16). Camay is not the act
of creating, but rather is an existing essence or power that may be embodied, reformed, or
transformed (Allen 2002, 2015; Bray 2009; Salomon 1991). The inability of the Spanish to
comprehend camac is demonstrated in the records of extirpation officials (Albornoz 1990;
Arriaga 1968, 1999 [1621]; and Salomon et al. 1991) who gave numerous accounts regarding
their attempted destruction of huacas. In these accounts, they lamented that even when burning
these huacas, the place where the huaca was or its ashen remains were still worshipped as the
huaca itself. In these cases, the camac remained as the huaca, just in a different form. Spanish
confusion in understanding Andean ontological views was a contributing factor to the initial
chaos of evangelization (Durston 2007; Estenssoro 2001, 2003; MacCormack 1991).
Andean Religion Prior to Inka and Spanish Conquest
The archaeology of religion—belief, sacrality, and even practices—has been until
recently understudied and undervalued as somewhat of a “black box” in archaeological pursuits
(Insoll 1999, 2001, 2004). Because of religion’s perceived immateriality, archaeologists have
24
focused instead on more quotidian and secular aspects of ancient peoples, spanning research
queries into subsistence strategies, political organization, or social status, to name a few.
However, as Bray argues, while there are perceived barriers to understanding religious beliefs
and practices, “if we fail to consider and theorize the influence of the sacred on peoples in the
past, then many of the questions we frame—as well as the answers we derive—are likely to be
incomplete” (Bray 2015:3). As the field of archaeology moved away from exclusively positivist
epistemology in the mid- to late-twentieth century, post-processual and contextual archaeologists
recognized material assemblages which might constitute religious beliefs or practices (Hodder
2000; Insoll 2004; Renfrew et al. 2011). Additionally, ethnographic study and historical
documents have provided contexts or frameworks through which the archaeologist might
determine religious practices from a recovered assemblage.
Scholars who study the pre-contact Andes have utilized both ethnography and colonial
era documents for interpreting ancient religion and the relationships between people and things.
These other data sources have provided invaluable insights into “Andean” religion as a panAndean phenomenon, yet often the differences between Inka religion and pre-Inka or non-Inka
Andean religion have been obscured.10 In some cases, “Inka” and “Andean” are used
synonymously (Salomon 1991: 4). This notion of interchangeability is a relic of colonial era
documentation, which conflated Inka and Andean religious hierarchies. For example, Viracocha,
the Inka “creator” god, was centered on Lake Titicaca and specifically the Island of the Sun, and
served as a convenient analogy for Spanish church authorities when attempting to convert local
peoples (Brosseder 2014; Gose 2008; MacCormack 1985; Salomon 1991). Similarly, when the
For this dissertation, I use “Andean” to refer to people of the Andes who are not explicitly of Inka ethnicity,
though they lived concurrently. When discussing those who lived in the Chicha-Soras Valley, I will use “Soras”
since this is the name of the Andean group in this region throughout the LIP, Late Horizon, and Early Colonial
Period.
10
25
Spanish invaded Peru, they found a diverse Andean landscape with varying hierarchies of cults,
which were all organized under Inka numina (D’Altroy 2015; Gose 2008). In reality, however, as
Salomon writes, “the religious life of most of the people who made up Tawantinsuyu’s
innumerable subject ‘nations’ had little to do with abstract or universalizing expressions.
Worship usually focused on sacred beings peculiar to particular kin groups, villages, mountains,
canals, and so forth. In fact, religious particularism, expressed in terms of place and descent, lies
at the heart of much Andean myth” (Salomon 1991: 4). That is, an Inka strategy of conquest and
consolidation was to “map” onto existing local deity hierarchies, while pre-Inka religion was
based primarily on local kinship and natural features of provincial landscapes. In the following
section, I discuss religious practices in pre-Inka, Inka, and sixteenth century early Spanish
colonial eras. Since Taki Onqoy’s central belief was in the resurrection of huacas, my main focus
here is on the concept and definitions of huacas as scholars best understand them through
archaeology, ethnography, and colonial era texts.
Animate Andean Things and the Transformative Properties of Huacas
In its most basic definition, a huaca is any “sacred thing” (Bray 2009, 2015)—the
materiality and sacredness of these “things” highlights the ontological differences between
Andean and Western worldviews, especially in regards to animism (Mannheim 2015). More
specifically, prior to Inka conquest, a huaca “was any material thing that manifested in the
superhuman: a mountain peak, a spring, a union of streams, a rock outcrop, an ancient ruin, a
twinned cob of maize, a tree split by lightning. Even people could be huacas” (Salomon 1991:
17). Huacas were venerated through gift-giving in return for advice or good fortune, and there
were cults for specific huacas based on lineage or ayllu. Huacas had life histories and
26
personalities, and were animate—they could marry other huacas or people, they ate and drank,
and they could change shapes or even mediums (Salomon et al. 1991). Huacas could be static in
the landscape, as in mountain peaks or lakes, or they could be smaller and portable, as in idols or
mummified kin (Gose 2008). Huacas were also “partible” in that they could be distributed
through space, and that pieces of huacas could be worshipped as that huaca (Chase 2015).
The fluidity of huaca and its ability to change shape or substance complicates
archaeological findings. To this end, the 2015 volume The Archaeology of Wak’as (Bray, ed.)
theorizes that “power is construed not in some abstract or ideal sense but rather as a type of
natural force having a specific and immediate local referent. Approaching wak’as as physical
embodiments of power, rather than as representations of other-worldly beings, highlights the
importance of their materiality” (Bray 2015:8). Huacas, then, are persons who are embodied with
power through their materiality. There are several potential ways in which the archaeologist
might study the sacred huaca in an Andean context, and more importantly, explicate the ways
this huaca acted on its biologically human counterparts.
The study of pre-Inka, Andean huacas is problematic because much scholarly knowledge
of huacas is refracted through Spanish texts describing Inka huacas. Though tempting to project
this knowledge to pre-Inka groups, explicit notions of huaca as avenues for scholarly
investigation in pre-Inka time periods have been only sparsely theorized (Cook 2015; Janusek
2015). Recent work has argued that for the Wari Empire (~650-1000 CE), D-shaped architecture
seen as sacred huacas was a tradition which built upon circular domestic forms in the Early
Intermediate Period (1-650 CE) (Cook 2015). The activities that took place in these D-shaped
structures, as well as the form and materials which constituted these structures may have
reaffirmed connections to ancestors (Cook 2015: 323-325). In the same way, the materiality of
27
stone has been conceptualized as possibly constitutive of ancestral huacas (Dean 2010, 2015;
Janusek 2015; Meddens et al. 2010). At Khonkho Wankane, the intentional destruction of some
of the monoliths emphasizes their potential status as powerful beings (Janusek 2015). Both of
these studies being from a perspective that nonhuman objects—temples and monoliths—were
interacted with and worshipped in ways that centered on their materiality.
Inka Religion and the Establishment of the Cult of the Sun
Andean archaeologists who study religion primarily focus on Inka religious practices,
since these are the most clearly delineated in Spanish chronicles. As the Inka conquered and
incorporated diverse ethnic groups throughout the Andes, they instituted a cult of the Inka Sun
(Betanzos 1987 [1551]; Cobo 1956 [1653]; D’Altroy 2015; MacCormack 1991; Molina 2010
[1574]) which was the supreme deity above all local hierarchies. In emplacing this deity, the
Inka asserted the differences between themselves and uncivilized “other,” manifesting these
social variances in intensive restructuration of people and environment (Kosiba 2012, 2015). For
example, at the pilgrimage center of Pachacamac, the Inka constructed their own temple to the
sun near the Pachacamac oracular shrine, “thus giving cultic and religious expression to the
political role of the Inkas in the region” (MacCormack 1991: 59; Makowski 2015). By installing
their own religious architecture at Pachacamac, the Inka legitimized their own role in the Andean
religious hierarchy.
The center of the Inka religious geographical landscape was the Coricancha in Cuzco, a
literal monument to the sun, from which at least 42 ceques (loosely translated as the Spanish
rayas, or lines) radiated, organized by 328 huacas (Bauer 1998; Cobo 1956 [1653]; Rowe 1979,
28
1985; Zuidema 1964).11 While some of these huacas were identified as landscape features
(caves, boulders, springs) as in pre-Inka Andean religion, other huacas were places associated
with Inka origin myths or important places in Inka history, such as locations commemorating the
victory in critical battles (Rowe 1979). Thus, while some Inka huacas—akin to Andean huacas—
were natural features, many huacas during the Late Horizon Cuzco region were specific to the
Inka. Bray (2009) suggests that Inka huacas were either iconic (as in the miniature
anthropomorphic or camelid figures used in Capacocha) or aniconic (non-anthropomorphic,
although possibly still representative). In both cases, these huacas were animate beings and
treated as such through being dressed, fed, and queried or adored.
The Inka Dead as Huacas
In addition to natural features, places of mythical or historical importance, and the cult of
the sun, the mummified ancestors of Inka rulers (and the rulers themselves) also played critical
roles in both politics and religion. In Cuzco, these mummies were worshipped, presented with
food and drink, and cared for by attendants (MacCormack 1991: 68). When an Inka ruler, or
sapa Inka (unique Inka) died, rulership was usually transferred to his oldest son, while the
remaining descendants of the sapa Inka formed a royal descent group called a panaca (Bauer
1998; D’Altroy 2015). Through the panaca, deceased Inka rulers still held much political power,
and retained control over their resources (herds, foodstuffs, lands, etc.). They were also able to
consult with the living via their attendants (MacCormack 1991). That the Inka regularly
interacted with their dead ancestors in political and religious arenas startled the Spaniards when
In fact, the Spaniards referred to the Coricancha literally as the “Templo del Sol” or Temple of the Sun (Bauer
1998: 3).
11
29
they arrived in Cuzco—these Spanish travelers were “convinced that the devil spoke to Andeans
in the guise of their deceased ancestors” (MacCormack 1991: 29).
Inka religion and Inka huacas acted alongside and in conversation with Andean religion
and huacas, and both groups considered these huacas as animate powerful beings. By installing
their own creation myths and beliefs in Andean hierarchies, the Inka were able to use schema
which were legible to Andeans, while simultaneously allowing for the continuous worship of
Andean huacas. There was no need to convert Andeans from one religion to another. The Inka
were content to have their own deities worshipped with preexisting Andean deities. This
agglutinative strategy to religious practice which worked so well under the Inka would come
under scrutiny during the first attempts at evangelization by the Spanish. Moreover, Spanish
authorities interpreted ontological differences in the notions of animacy surrounding the concept
of huaca as evidence of demonic influence on Andean peoples.
Conversion in the Andes
Justification for the conquest of Peru was predicated on “converting” Andeans to
Catholicism. Yet, would a “converted” Andean look like? What would they believe? How
would they act? Conversion is a change of internal state in the soul of the subject paired with a
shift to Catholic orientations of divinity. These principles of conversion could never be fully
measured in Andean populations—scholars have even argued that as Andeans began to behave
more and more like converted Catholics, the greater the tendency for Spaniards to consider their
practices as demonic, influenced by idolatry, and thus unacceptable (Estenssoro 2001, 2003;
Gose 2008). In the initial decades of evangelization, the chaotic period coined as the primera
evangelización by Estenssoro, the Spanish program of religious practice was far from organized
30
(Durston 2007; Estenssoro 2001, 2003; Gose 2008; MacCormack 1991). In fact, it was not until
Francisco de Toledo’s appointment to the position of Viceroy (1569-1581) and the
implementation of the Third Lima Council edicts (1582-1583) that the Spanish conversion
program in Peru was organized and manifested in a streamlined manner (Durston 2007,
Estenssoro 2001, 2003). The 1530s and 1540s were characterized by the chaos of conquest,
during which those Spaniards in Peru were both infighting amongst themselves, while also
embarking on several plundering campaigns (Durston 2007). These campaigns were aimed at
obtaining enough precious materials and items to settle debts of the conquistadores, and to justify
the mission monetarily to the Spanish Crown.
During this time of exploration and plunder, mendicant orders of religious clergy arrived
in Peru, and were largely responsible for early attempts at conversion (Durston 2007; Estenssoro
2001, 2003). Specifically, the Dominicans initiated these conversion campaigns, as they were the
first to arrive and subsequently converted the Inka Coricancha into a functioning Catholic
church. The first diocese in Peru was established in Cuzco in 1536, and the first Bishop of Cuzco
was the Dominican Vincente de Valverde, who had accompanied Pizarro throughout the
conquest (Estenssoro 2003). Along with the Dominicans (who first officially founded a parish in
1540), orders of Franciscans (1552), Augustinians (1551), Mercedarians (1560), and eventually
the Jesuits (who arrived in 1568) settled in various geographic locations and enacted pastoral and
evangelization activities independently from one another and from the Spanish Crown
(Estenssoro 2003: 23). The following sections detail the initial periods of Spanish evangelization,
with particular emphasis on the disjunctures between official decrees or Councils, and their
implementation (or lack thereof) by the mendicant orders throughout the Andes. In this
discussion, I return to the materiality of religious practices of Andeans at this time, and
31
demonstrate how these practices were perceived by Spanish clergy. In this way, I set the stage
for understanding how Taki Onqoy was practiced and performed, and how it was interpreted by
Spanish authorities, as seen in their writings on the subject.
The Problems of Language Diversity at the Time of Conquest
Throughout the early years of this conversion program, attempts at converting Andean
beliefs were disorganized and piecemeal. One major problem for Spanish religious authorities
involved the diverse array of linguistic differences across the Andes at the time of conquest.
These linguistic differences can be explained by several main policies and practices common
during the Inka regime and in general Andean traditions. Essentially, there was no
correspondence between language, territory, and polity (Mannheim 1991:31-60). First, the
Andean economic tradition of traveling between and having familial ties in several different
ecological zones, or vertical complementarity (Aldenderfer 1993), created highly mobile
populations and spread language groups both vertically and horizontally. Second, Inka policies
of mitmaqkuna, or the movement of troublesome groups to different areas of the Empire, created
linguistic patterns in which the language of one particular small group would intrude upon a
broader language tradition in another region (D’Altroy 2015). Thus, across the Andes, polities
could contain any number of individuals who spoke different languages—this manifested itself
in differences between elite and local lingua franca, as well as a general historical pattern of
linguistic diversity caused by changing political alliances or divisions throughout time
(Mannheim 1991). Even if some Spanish mendicants had become excellent Quechua or Aymara
speakers, they still would have been unable to account for the wide array of linguistic diversity
encountered as they spread their attempts at conversion.
32
Durston argues that in order to explain the structures of new languages at the time of
conquest, Spanish thinkers employed two explanatory models: the genetic model “which focused
on the origins of languages and was based on the assumption that all ultimately proceeded from
God,” or the conventional model “in which languages were understood as reflections of the
societies that spoke them” (Durston 2007:42). In the first model, the understanding of language
was biblical—Spaniards believed that Adam and Eve had spoken a divine language and that 72
other languages had been distributed into different groups of people. The second model did not
have to contrast with the first model, since languages could change and degrade over time. The
broad diversity of languages in the Andes was interpreted by the Spanish as a sign of barbarism.
The Spanish understood this array of linguistic diversity as a sign that Andean peoples were
more simplistic and that their languages were not sophisticated enough to consider and reflect
Catholic doctrine. That is, not only did Andean languages lack the ability to discuss specific
Catholic beliefs, but they also were not equipped with the words or concepts used to consider
abstract concepts like faith, god, belief, and other philosophical topics.
Evangelization after Conquest (1532-1569)
Led by Francisco Pizarro, Spanish conquistadores spent the first five years in Peru allying
with or rebuffing Inka descendants from Wayna Capac’s lineage while simultaneously fighting
against the Inka in several skirmishes (specifically Manko Inka’s rebellion in 1536). It was not
until 1537 that Manko Inka led the remaining Inka to the jungles of Vilcabamba (Department of
Cuzco), thereby allowing the Spanish to seize full control of the Andean heartland (Hemming
1970). The broad military and historical background of the initial years of conquest is known
through Spanish accounts, but knowledge and understanding regarding evangelization in this
33
period is lacking (Estenssoro 2003). In what he calls the primera evangelización, Estenssoro
affirms that religious teachings at this time were not streamlined, and were documented in
handwritten manuscripts which differed between orders and of which none have survived
(Durston 2007; Estenssoro 2001). In the first half-century after conquest, Spanish clergy held
variable opinions toward the capability of Andeans to comprehend Catholicism and by
extension, how Andean mental and cultural capacities were adaptable to conversion. Debates
over whether conversion should be predicated on faith (the idea that God already existed in the
hearts of everyone) or reason (that voluntary baptism and adoption of Catholicism was
necessary) influenced practical ways of evangelizing (MacCormack 1985). In the 1530s and
1540s, some priests had to draw analogies between Andean and Catholic practices in order to
make them comprehensible. Thus, mendicant orders during the primera evangelización were
forced to map onto these practices and incorporate Andean terms, rituals, and materials in
evangelization (Durston 2007; Estenssoro 2001, 2003; MacCormack 1985; Wernke 2018).
It was not until 1545 that the first Archbishop of Lima, a Dominican friar named
Jerónimo de Loayza (who served from 1543-75), that the first guidelines were established for
religious conversion. In this initial period of evangelization, catechesis consisted of natives
memorizing the Ten Commandments, the basic prayers, and attending church services on
Sundays (Durston 2007). As part of these services, priests were to perform simple sermons called
platicas—the forms and languages of these platicas were initially very diverse, and differed
between different mendicant orders and native languages. Loayza’s 1545 Instrucción de la orden
que se a de tener en la doctrina de los naturales is considered the first surviving doctrinal primer
issued in the Andes, and has been characterized (in part) as a response to the diversity of
doctrinal instruction manuals in native languages (Durston 2007: 67; Estenssoro 2003; Wernke
34
2018). Loayza’s Instrucción asserted that all Andean peoples were already allowed in the
Kingdom of God (Estenssoro 2003; Wernke 2018). It also addressed a prevalent concern that
many of these doctrinal manuals had not been checked for orthodoxy against the Latin (or
Spanish) originals (Durston 2007). Further, Loayza warned that errors or discrepancies between
catechesis in native languages and catechesis in Latin or Spanish could suggest to native
populations that religious doctrine was diverse and variable (Durston 2007; Estenssoro 2003;
Wernke 2018).
In an effort to standardize aspects of religious evangelization, Loayza also called and
presided over the First Council of Lima (1551-52) (Estenssoro 2003: 26; Vargas Ugarte 1951).
Specifically, this Council reflected a greater desire for the secular church to have influence over
evangelization procedures in the New World. However, the set of decrees by the First Lima
Council contradicted aspects of Loayza’s 1545 Instrucción, inadvertently increasing the diversity
in doctrinal materials (Estenssoro 2003; Wernke 2018). The First Council emphasized a
distinction between baptized and unbaptized individuals, affirming that it should be
communicated to native peoples that “sin el baptism ninguno se puede salvar” (Estenssoro 2003:
421, as quoting the First Lima Council). This edict served to contradict Loayza’s first Instrucción
and also undermine or destabilize Andean claims based on ancestral authority, since these
ancestors were no longer welcome in the Kingdom of God.
The First Council of Lima’s Instrucciones did not have much impact on actual
evangelization practices and were essentially supplanted by the Dominican friar Domingo de
Santo Tomas’s Platica para todos los indios (Durston 2007; Estenssoro 2003; MacCormack
1985). Domingo de Santo Tomas was a follower of Bartolomé de las Casas, who argued on
behalf of the intellectual and spiritual capabilities of native peoples. Following las Casas, Santo
35
Tomas asserted that the Quechua language was evidence of a “civilized” population of Andeans,
and thus that Christian terms could be expressed by preexisting Quechua phrases (MacCormack
1985: 449). Santo Tomas’s instructions, distributed through his Platica, allowed more
accomodationist practices by Andean peoples and a generous reading of what it meant to be
Christian (Durston 2007).
While the priests of the primera evangelización were variably performing catechism and
liturgical doctrine as they saw fit in the New World, the Council of Trent (1545-1563) was
developing a more orthodox and standardized protocol for catechetical instruction (Durston
2007; Estenssoro 2003). The reception of this protocol was received by Archbishop Loayza in
October of 1565, who then called a Second Council of Lima (1567-68) (Durston 2007). In
general, the Second Council had a more vested interest in Andean “idolatry”—while local
practices had been more or less accommodated and allowed in the First Council, the Second
Council denounced and prohibited local traditions, specifically in regards to burial rites. The
Second Council also officially established the parish system, which corresponded to a
specifically defined territory and group of people who were assigned to a certain priest. The head
priest of each parish was to be responsible for monitoring individual members, and had up to 400
heads of households under his care (Durston 2007). Next, the Second Council became much
stricter regarding sermon and catechesis attendance—these sessions became mandatory on
Wednesdays and Fridays, as well as a feasting session on Sundays. Lastly, and perhaps most
importantly, priests of Indian parishes were required to learn Quechua in order to better serve
and receive their corresponding populations (Durston 2007: 72). As was true of Archbishop
Loayza’s 1545 Instrucciones and the decrees of the First Lima Council, the orders put forth by
the Second Lima Council were also unable to be enforced. It would not be until the Third Lima
36
Council’s (1582-1583) standardized set of vernacular texts in combination with Viceroy
Francisco de Toledo’s resettlement program that an organized, uniform set of catechetical
materials would be implemented (Durston 2007; Estenssoro 2001, 2003; MacCormack 1985,
1991; Wernke 2018)
In the preceding section, I have given a brief overview of some of the conflicts and
problems with Spanish conversion in the New World. The initial phase of Catholic
evangelization was not unified—itinerant friars developed diverse tactics and were responsible
for conversion of entire swaths of populations. Because of the geographic range and problems of
linguistic differentiation, Andeans were initially only required to perform certain rites or attend
catechisms a few times per year. Furthermore, linguistic barriers between Andean peoples and
priests led only to partial conversion in most cases—Andeans did not substitute the Christian
deity for all of their landscape deities. Instead, the Christian deity became an additional force in
their pantheon. In this time of diverse evangelization strategies, a “disaffection” began to spread
through the Andean highlands characterized by a rejection of Spanish religion and culture in
favor of a return to prehispanic huaca worship. The following section will highlight the material
practices of Taki Onqoy as described through sixteenth-century documentary records.
Taki Onqoy
The Taki Onqoy movement has been studied utilizing a corpus of documents primarily
authored by two Spanish priests who were attempting to extirpate it. However, I argue that
secondary explications on these texts could be supplemented by a practice approach grounded in
archaeological investigation. The two Spanish priests—Cristóbal de Molina and Cristóbal de
Albornoz—provide descriptions of the practices of takiongos, including the materials they used
37
and how they used them. In this section, I will detail the primary sources which discuss Taki
Onqoy, their descriptions of associated behaviors and geographic expanse of the movement, and
the material implications for the behaviors they describe. I show how material media were central
to the constitution of the movement, and how a study of this media can allow new avenues for
inquiry which cannot be achieved from the documents alone.
Primary Sources
All Taki Onqoy primary source documents were written by Spanish priests who were
invested in extirpating the movement, and thus reflect somewhat myopic views refracted through
the lens of the Spanish agenda. Yet, these texts are incredibly valuable in their accounts of the
practices which constituted Taki Onqoy. Four authors—Guerrero (1564, as printed in Yaranga
1978), Cristóbal de Molina (2010 [1574]), Cristóbal de Albornoz (1990 [1569, 1570, 1577,
1584]), and Bartolomé Álvarez (1998[1588])—refer to the movement directly, while one other
author alludes to a movement which likely points to Taki Onqoy—Gaspar de Sotelo (1565, as
printed in Lohmann Villena 1941). Of these five authors, Guerrero’s and Sotelo’s descriptions are
from letters warning of idolatrous practices and preparations for militaristic revolt, respectively.
Cristóbal de Molina’s account is the last chapter of his Relación de las fabulas y ritos de los incas
(hereafter Relación), and narrates the general beliefs and practices indicative of Taki Onqoy.
Álvarez’s De las costumbres y conversión de los indios del Peru, Memorial a Felipe II (hereafter
Memorial) contains a full section on the idolatries of native peoples, specifically from his time in
and around the town of Pampa Aullagas (Audiencia de Charcas, in what is today Bolivia).
Álvarez’s sections on Taki Onqoy are particularly fascinating because they offer a view outside of
the Cuzco and Ayacucho heartland—Álvarez even provides an Aymara term for the movement,
38
which he calls talausu. Finally, Cristóbal de Albornoz wrote several documents detailing his
actions rooting out and destroying idols and native idolatrous practices. Albornoz’s work has two
major components: first, his Informaciones de servicios (hereafter Informaciones) were letters to
the Spanish Crown written in 1569 (from Cuzco), 1570 (Huamanga), 1577 (Cuzco), and 1584
(Cuzco); and second, his La instrucción para descubrir todas las guacas del Pirú y sus camayos y
haziendas (hereafter Instrucción ~1580, as cited in Duviols 1989). Together, the accounts of these
five authors make up the entire known corpus of Taki Onqoy primary written materials, and offer
important observations about the movement which I will detail below.
The First Account—a Letter from Guerrero to Luis de Olvera (1564)
Chronologically, the first known mention of Taki Onqoy as a movement is found in the
1564 letter from a parish priest named Guerrero to Luis de Olvera, the vicar of Parinacochas. The
Peruvian cultural historian, Abdon Yaranga Valderrama, found this letter in the community of
Huancaraylla (Ayacucho) in January of 1950. As a document transcribed in the initial years of the
movement, Guerrero’s letter is critical in understanding the practices central to Taki Onqoy
during its rise. Guerrero warns Olvera, who was already known for his extirpation practices (Gose
2008; Guibovich 1991) that Andeans in the province of Vilcashuaman were practicing an idolatry
called “cuyllur Onqoy” (Yaranga 1978: 167).12 Guerrero attests that the Andeans in the towns of
Guancaraylla, Guamanquiquia, and Sacsamarca believed:
That all of the guacas and guamani and idols destroyed by the Christians after
Pizarro’s arrival had resurrected and were entering the bodies of the Indians and
that these guacas and guamanis and idols of Collasuyos and Cuntisuyos had
united at the guaca of Titicaca and that the guacas and guamanis and idols of
Chinchasuyos and Condesuyos had united at the guaca of Pachacamac. These
According to Holguin’s Quechua-Spanish 1608 Vocabulario de la Lengua General de todo el Peru, the word
cullyur or coyllur means estrella, or bright star (Holguin 1989 [1608]).
12
39
guacas were on the warpath to exact revenge and bring death to God and the
people of Castile in this land would be finished because the guacas condemned
them to sickness and disease…the guacas were angry with the Indians because
they had taken another god and Christian names and that if the Indians did not
want to die like the people of Castile, but instead prosper and increase their
wealth, they would renege Christianity and would not enter the churches nor
worship the crosses. (Yaranga 1978: 168).13
In this early account of Taki Onqoy, many of the central beliefs had already been
established that would become hallmarks in the tellings of Molina and Albornoz. During the
1530s and 1540s, the Spanish campaign of plunder and destruction angered the huacas, who
were now poised to return to life and form armies under the shrines of Titicaca and Pachacamac.
These armies would fight and defeat the Spanish God. If the Andeans wanted to benefit from the
return of the huacas, they would have to renounce their Christian teachings, avoid churches, and
refrain from worshipping crosses.
There are two very important aspects to this passage. First, Guerrero’s letter demonstrates
the materiality of the huacas—in this account, huacas are not inanimate lakes or streams or
rocks; rather, they are powerful beings who have come to exact revenge on the Spanish and the
Andeans who had neglected them. In practice, these huacas entered the bodies of Andeans in a
form of spirit possession in which the Andean individual became the camasca of the huaca,
thereby inheriting its power. Second, this initial passage provides two practices which would
leave an absence of evidence archaeologically. Andean takiongos were not to enter churches, nor
Translated from “Todas las guacas y guamani y ydolos destruidos por los cristianos despues de la llegada del
marques Picarro avian rresucitado y se metian en el cuerpo de los yndios y estas guacas y guamanys y ydolos de los
collasuyos y cuntisuyos se avian juntado en la guaca del titicaca y las guacas y guamanys y ydolos de los
chinchasuyos y condesuyos se avian juntado en la guaca del pachacamac. Estas dichas guacas estaban en pie de
guerra para dar batalla, vencelle y dar muerte a dios nuestro senor y que las gentes de Castilla de esta tierra se
acabarian porque las guacas les condenaban a enfermedades para matalles y que las guacas estaban enojadas con los
indios porque habian tomado otro dios y nombres cristianos y que si los indios no querian morir como las gentes de
Castilla, sino todo salud y aumento de bienes, renegasen al christianismo y que no entrasen en las iglesias ni
adorasen las cruzes” (Yaranga 1978: 168), translation mine.
13
40
worship crosses—materially, this would manifest in an absence of Catholic religious objects
within Taki Onqoy households.
In the second half of Guerrero’s letter, there are several more descriptions of Taki Onqoy
practices, which indicate that by invoking specific chants and huacas, takiongos were becoming
huacas:
The Indians get together in the mountain top and call out ‘tayta guamani tinca’
and there, shake saying that they have the guacas in their bodies and they chant
and many of them paint their face with paria and llacsa…and they bury rams in
the ground, sanco, parpa, gold, silver and cuca and yllapas, and burn the bundles
and pour chicha and at midnight, they perform ceremonies that they call taqui
ongo, singing in rounds and committing abominations against our Lord, against
our religion, and against the Spanish. The leaders of this idolatry move in groups
of three: two men and one woman, the men chant ‘catun colla cocha’ ‘camac
pacha’ and the woman ‘guamani tinca’ o ‘guamani carguaras. (Yaranga 1978:
168).14
In the second half of Guerrero’s letter, the Andeans are described as manipulating several
materials in the idolatrous performances. First, Guerrero attests that they paint their faces with
“paria” and “llacsa.” Here, paria is referring to a red pigment, likely cinnabar or bija, and llacsa
indicates bronze.15 The use of red and bronze pigments in Taki Onqoy ritual demonstrate the
process of transformation from human to huaca, and also provide positive evidence of Taki
Onqoy practices. Archaeologically, these practices could be inferred from the presence of Bixa
Orellana paleoethnobotanic remains, cinnabar pigments, and bronze metalworking. Both sanco
and parpa are associated with prehispanic ritual performances and offerings to Andean huacas.
Translated from “se juntan en el cerro tinca y alli tiemblan diziendo que tenian las guacas en el cuerpo y los
hazian hablar y a muchos de ellos les pintaban el rostro con paria e llacsa y alli iban los yndios a los adorer port al
guaca e guamani e ydolo que dezian que se le avia metido en el cuerpo y les ponian carneros de la tierra, sanco,
parpa, oro, plata y cuca e yllapas y quemaban los bollos y derramaban chicha y a media noche fazian ceremonias
que llaman taqui ongo, cantando a la rronda, y predicaban grandes abominaciones contra dios nuestro señor, contra
nuestra religión y contra los españoles. Los maestros predicadores de esta idolatría andavan en grupos de tres: dos
ombres y una mujer, los ombres se acian llamar catun colla cocha y camac pacha y la mujer guamani tinca o
guamani carguaras” (Yaranga 1978: 168), translation mine.
15
As Bauer et al. suggest, “Molina uses the word “enbixauan,” the root of which comes from the plant “Bija” (Bixa
Orellana) from which a red paint (achiote) is made” (Bauer et al. 2011: 120).
14
41
Both terms refer to small maize cakes, with the former “often prepared with blood offerings for
the Andean divinities,” and the latter cooked with salt or aji (Mills 1997:292, 57). Sanco and
parpa were both cooked with exclusively Andean products, which may represent both a renewal
of Andean cultural traditions and a rejection of Spanish foodstuffs.
The final portions of this passage describe the sacrifice and subsequent interment of rams
and coca, and reflect the renewal of prehispanic huaca worship, as seen in both Inka (Molina
2010 [1574]) and pre-Inka societies (Garcilaso 2000, Gose 2008). The term yllapa has several
meanings in sixteenth-century Quechua: yllapa’s most common associations are with “thunder”
or “lightning” (MacCormack 1991, Molina and Bauer 2011). Yet, during Taki Onqoy, this term
may have been referring to Inka mummies, as suggested by both Albornoz and Guaman Poma
(Albornoz 1967: 21; Guaman Poma 2015 [1615]: 287 [289]; Heaney 2018; MacCormack 1991).
MacCormack even suggests that “such terminology was meaningful in the Andes because
lightning and the dead epitomized a cosmic imbalance tending either to generation or
destruction” (1991: 286). Table 2.1. lists positive and negative archaeological correlates of Taki
Onqoy practices.
Source
Year Positive Signature
Guerrero
1564 Red pigment--cinnabar
or achiote (Bixa
orellana)
Bronze pigment
Absence of
Evidence
No Catholic
objects or
imagery in TO
households
Table 2.1. Taki Onqoy correlates from the 1564 Guerrero letter.
Cristóbal de Molina and Cristóbal de Albornoz
42
No Signature
Avoidance of
churches
Cristóbal de Molina and Cristóbal de Albornoz were the two authors who produced the
lengthiest accounts of Taki Onqoy, although both were writing with a clear agenda vis a vis the
Spanish Crown. Additionally, the two priests were colleagues and friends who regularly
interacted with one another—Molina even twice testified on behalf of Albornoz, attesting to
Albornoz’s extirpation activities in the central highlands. Cristóbal de Molina, commonly known
as “el Cusqueño,” arrived in Peru sometime after 1556, and was appointed to be an ecclesiastical
visitor by Viceroy Toledo around 1572 (Urbano 2008: 427). Molina was known for his Quechua
language skills, and was also tasked by Sebastián de Lartaún, the third Archbishop of Cuzco, to
write a chronicle of the history and religion of the Inkas (Bauer et al. 2011). Molina’s Relación is
a month-by-month description of Inka religion and rituals, recounting information which he
recovered from several interviews with Cuzco’s oldest indigenous inhabitants.
Since the Relación was commissioned by Archbishop Lartaún, some scholars consider
the work as the official church version of events, thereby questioning its validity as an
independent source (Urbano 1989). Others have even argued that the final chapter of Molina’s
work was a later addition to the text, likely a supplemental explication written in 1583 (Ramos
1992). However, most scholars agree that the text was written between 1573 and 1575, and is a
valuable source for the study of Inka religion in general, and also of Taki Onqoy (Bauer 2011;
Heilman 2002). In terms of Taki Onqoy beliefs, materials and practices, Molina echoes much of
what Guerrero reported in his 1964 letter. Molina did not witness any Taki Onqoy practices
himself, and thus his report is based on hearsay from other priests, particularly Luis de Olvera.
The debate of the veracity or reliability of Molina’s account underscores broader challenges in
study of Taki Onqoy—since all of these authorities knew each other and regularly interacted, the
different accounts cannot be considered as independent contributions to a broader literature. This
43
dissertation is innovative in that it shifts the discussion from textual reports and biographical
study of a centuries-old movement to a new emphasis on materials and alternative ways to
understand this movement.
Molina reiterates the tenets attributed to Taki Onqoy in the initial portions of his chapter.
After the alliances between indigenous peoples and Spanish authorities began to disintegrate and
splinter, embittered local leaders started to blame Spanish presence for much of local displeasure
and hardships occurring at that time (Stern 1993). As a result, Andeans were “departing from the
Catholic faith they had received” and “returning to the idolatry that they had practiced in the time
of their infidelity” (Molina et al. 2011:84). Molina’s use of “returning” here suggests that the
1560s Taki Onqoy beliefs were a resurgence or revitalization of prehispanic belief systems in the
Andes. He reaffirms this idea when discussing the major conviction of the takiongos:
And returning to the resourcefulness of the devil in diverting these poor souls, it
happened that they believed that all the huacas of the kingdom that the Christians
had demolished and burned had come back to life, and had formed themselves into
two sides: some had joined with the huaca of Pachacamac, and the others with the
huaca Titicaca. They said that all of them were flying around in the air, ordering the
people to give battle to God and defeat Him. (Molina and Bauer 2011:85).
Here, Molina outlines the central tenet of Taki Onqoy: the huacas which had been destroyed by
the Spaniards or neglected by Andeans were reclaiming their power and aligning with the major
huacas at either Titicaca or Pachacamac to overthrow the new Spanish religion and regime.
Molina does not mince words—he is direct in affirming that the movement is truly local interests
and beliefs versus Spanish interests and beliefs.
In his Peru’s Indian Peoples and the Challenge of Spanish Conquest, Steve Stern
expands on these ideas of Spanish versus native. Stern suggests that Taki Onqoy was focused on
uniting local peoples against Spanish traditions. He writes that “the huacas and takiongos
expressed gnawing anxiety that natives who assumed Hispanic styles of religion, clothing, food,
44
and custom were becoming enemies tied to colonialism and Christianity” (Stern 1993:59).
Rather than ascribing to prehispanic local political configurations, Taki Onqoy practitioners
wanted to unify the Andean population. They sought to elide ethnic, religious, and kin
differences to create a larger, more unified population of Andean believers who rejected
Catholicism. As Spanish colonialism was founded on the categorical distinction between Spanish
and “Indian,” Taki Onqoy preachers utilized this dichotomy in their attempts to unite indigenous
populations. Takiongos—whether wittingly or not—were redeploying these binary categories as
a means of resistance. Thus the Taki Onqoy movement was indelibly influenced by Spanish
colonial tactics and Catholic dictates, entangling colonizer and colonized.
For example, when describing the practices associated with Taki Onqoy, Molina’s
descriptions depict a syncretic mix of Catholic and Andean beliefs, manifesting themselves in a
set of actions which were inspired by both religious traditions. First, Molina argues that Taki
Onqoy was spread by wandering preachers, who “went about preaching this resurrection of the
huacas, saying that the huacas now were flying through the air, dried out and dying of hunger,
because the Indians no longer made sacrifices nor poured chicha to them” (Molina and Bauer
2011:85-86). The notion of traveling preachers wandering through the Andean highlands was a
practice borrowed from Catholicism—prior to Spanish contact, religious doctrine was not spread
through individuals preaching the message of the Gods (Gose 2008). Second, Molina describes
individual huaca possession as a crucial aspect of Taki Onqoy: “Accordingly, there were many
Indians who trembled and threw themselves on the ground…everyone would celebrate in the
town for two or three days, dancing, drinking, and invoking the huaca that the possessed person
represented and said it was inhabiting his body” (Molina 2011: 86). While possession of huacas
was an integral part of prehispanic worship (Gose 2008: 97), during Taki Onqoy, huacas were no
45
longer embodied in the stones or landscape markers of the past, but rather in the adepts who
believed in them. As Gose argues, “that Takiongos made offerings directly to the mediums
shows that they saw them as the huacas’ principal embodiments” (Gose 2008: 97). The degree of
the breakdown between Andeans and their huacas could only be resolved through bodily
possession of these spirits.
Besides wandering preachers and huaca possession, Molina describes several other
practices associated with Taki Onqoy: preaching advocating the resurrection of the huacas;
fasting; avoidance of Christian foods, clothing, churches, and prayer; huaca possession of
Indians; ritual cleaning of houses; trembling, dancing, and drinking; use of red body paints;
huaca worship; sacrifice of rams and cuyes; Indian sorcerers using colored maizes and spondylus
shells to cure afflicted people; ancestor worship; and the construction of outdoor enclosures for
Taki Onqoy preaching and dancing (Molina 2011, Albornoz 1990:191) (Table 2.2). Much of
Taki Onqoy was thus based on performance; that is, the oracular possession of Taki Onqoy
adepts and their subsequent dancing, trembling, fasting, and attempting to heal those who were
also afflicted, suggests that the movement was a visible, almost tangible way of interacting with
and pleasing the prehispanic huacas which had been forgotten during the Spanish conquest.
46
Source
Year
Molina
157375
Positive Signature
Red pigment--cinnabar or achiote (Bixa
orellana)
Sacrifice of rams and cuy
Dark purple maize or chicha
Llipta
Spondylus shell
White corn flour
White, black, red, and yellow maize
Coca
Gold and silver
Chicha
Cleaned houses
Absence of
Evidence
No salt or aji
Absence of
Spanish
foodstuffs,
food, and
clothing
No
Signature
Fasting
Possession
Dancing
Preaching
and sermons
Table 2.2. Taki Onqoy correlates from Cristobal de Molina.
Molina provides several other details regarding the geographic extent of Taki Onqoy and
the Spanish response to the movement. While the geographic area and relative “importance” of
the movement are matters of considerable debate in the twentieth-century, Cristóbal de Molina
definitively records Olvera’s testimony regarding the afflicted provinces.16 Molina writes that
“not only in that province [Parinacocha], but in all the other provinces and cities of Chuquicaca,
La Paz, Cuzco, Humanaga, and even Lima and Arequipa, most of the people had fallen into great
apostasies” (Molina 2011: 84). Olvera, and by extension Molina, thereby name the largest
provinces and regions in Peru at that time as having a definite population of Taki Onqoy adepts.
Molina’s testaments regarding the geographic range of Taki Onqoy cannot be taken as fact,
however. His information is second-hand and unverified, and he may have had a personal
interest in exaggerating the spread of Taki Onqoy in the eyes of the Spanish crown.
16
For sources on this debate, see Duviols 1977; Varon and Albornoz 1990; Millones, Castro-Klaren, and Albornoz
1990.
47
Finally, Molina’s account provides few details regarding the Spanish authorities’
response to Taki Onqoy. Molina writes that the movement lasted “more than seven years” and
that “because the Vicar Luis de Olvera began to punish that province and that of Acari, and he
informed the Royal Audience of Lima, the archbishop and the bishop of the Charcas, and other
parts…the movement began to weaken” (Molina 2011: 87). Molina is very imprecise when
defining the Spanish actions against Taki Onqoy. While he refers to “punish[ment]” of takiongos
in certain regions, he is unclear as to what this punishment may have been, physically or
otherwise. Furthermore, Molina only names Luis de Olvera as one trying to eradicate the
apostasy—he neglects to mention Cristóbal de Albornoz’s extirpation campaigns in the
highlands as well as Francisco de Toledo’s reducción campaign which began in 1572 (Mumford
1998). For a much more detailed discourse on Taki Onqoy punishments, Cristóbal de Albornoz’s
Informaciones de servicios (1569, 1570, 1577, and 1584) and La Instrucción para descubrir
todas las huacas del Pirú y sus camayos y haziendas (1580s) are more specific in their
discussion of people and places associated with Taki Onqoy.
Cristóbal de Molina’s Relación thus provides a general description of the motivations of
and practices associated with Taki Onqoy. While his chapter is brief, Molina’s descriptions of
Taki Onqoy traditions and ritual are invaluable in understanding how takiongos became huacas,
and how the materials associated with prehispanic huaca worship were again the media through
which idolatry was performed. While Molina focuses more on the behaviors associated with Taki
Onqoy, Cristóbal de Albornoz’s Informaciones and Instrucción give testimonials and personal
accounts of the extirpation campaign, the towns visited by Albornoz, the people punished in each
down, and the huacas destroyed. I will now give a brief biography of Albornoz, highlight the
new information in his two works, and outline some of the potential biases in his accounts.
48
Though Cristóbal de Albornoz was the Spanish priest who ostensibly most interacted
with takiongos—and by extension wrote the most on Taki Onqoy—the veracity and reliability of
his accounts have been subject to considerable debate. More specifically, Albornoz’s overt career
ambitions and the ongoing tensions between secular and mendicant clergy have prompted
scholars to suggest that the geographical expanse of Taki Onqoy was highly exaggerated by
Albornoz (Estenssoro 1992; Varón 1990), or even that the movement was a creation of Albornoz
(Ramos 1992). Much of the secondary Taki Onqoy debate hinges on Albornoz as a narrator—the
gaps, silences, and contradictions in his texts have been subject to extreme scrutiny, of which
questions regarding his motivations or honesty cannot be fully answered through his works or
biographical study of his life. In the following section, I discuss Albornoz’s life, his works, the
new information provided by his works, and I highlight the contradictions in secondary
scholarship on Albornoz as he relates to the Taki Onqoy movement.
In 1964, Luis Millones published a short article analyzing Cristóbal de Albornoz’s
Informaciones entitled “Un movimiento nativista del siglo XVI, el Taki Onqoy (Millones
1964:138-140), which generated initial interest in historical research into the movement. In 1971,
Millones published Albornoz’s Informaciones; however, I will cite the 1990 version of
Albornoz’s writings, transcribed and edited by Millones et al. in El retorno de las huacas:
Estudios y documentos sobre el Taki Onqoy, Siglo XVI. In 1967, Pierre Duviols brought to light
another of Albornoz’s accounts, the La Instrucción para descubrir todas las huacas del Pirú y
sus camayos y haziendas, which was written in the early 1580s. In general, Albornoz’s
Informaciones is a collection of four separate testimonials (1569, 1570, 1577, and 1584) in
which Albornoz provides a list of questions posed to a variety of priests and authorities and their
responses. This set of documents served mostly to discuss, in great detail, Albornoz’s range of
49
activities in his extirpation campaigns. Similarly, Albornoz’s Instrucción reads as a how-toextirpate-idolatry manual founded in his own experiences and actions.
Albornoz was a secular priest educated in Spain who arrived in Paita in the Viceroyalty
of Peru in 1567 after having been in Santo Domingo and Nueva Granada (Guibovich 1991;
Millones 2008). Upon arriving, Albornoz was assigned to a doctrina in Collao, but later was
appointed as an ecclesiastical visitador in charge of Arequipa and its district. From 1569 through
1571, Albornoz continued to work as a visitador, continuing from Arequipa to the areas of
Huamanga and Cuzco (Millones 2008). During this time, Albornoz tasked himself with rooting
out and punishing those who he caught in religious performances which he perceived as
idolatrous to Catholic belief, in part to bolster his case for promotion. To this end, each of the
letters in his Informaciones was the physical manifestation of a request for promotion to specific
ecclesiastical offices, and their contents reflect the unique demands for each post.
Although the testimonials of Albornoz’s witnesses are very repetitive—often with the
witnesses simply repeating and rephrasing the questions posed to them—the individuals still
provide new information not found in Molina’s writings. For example, while Molina’s work
never directly gives the “apostasy” a name, Albornoz identifies the movement as having two
names: Taki Onqoy and Aira. These formal names are absent in Albornoz’s 1569 account, but
instead are cited in the 1570 account, thus indicating that defining characteristics of the
movement were not known to Albornoz until the 1570s.17 In one testimony, the Spaniard Don
Phelipe attests that Albornoz visited cities in the provinces of Huamanga and Cuzco, where “had
been discovered the sect and apostasy and predication that they call Taki Onqoy or by another
17
Ramos (1992) suggests that the fact that Taki Onqoy is not mentioned in the 1569 letter casts doubt over all of his
accounts, while Heilman (2002) counters this and offers that Albornoz had only just arrived in Peru at the time of
the 1569 letter, and thus the oversight in the 1569 letter is explicable.
50
name, Aira” (Albornoz 1990:61).18 This account is the first time that the name Aira is mentioned
in any of the primary sources.
The subsequent witness testimonies in the following pages of Albornoz’s Informaciones
generally align with Molina’s descriptions of the movement, its motivations, and its practices. In
the same 1570 account, the witness Baltasar de Hontiveros testified on behalf of Albornoz that
“they said that they did not believe in God nor worship the crosses nor the saints, nor enter the
churches, nor confess with the clerics or friars, but instead within them, perform rites and
ceremonies of the time of the Inkas on certain days of the week, and offer the huacas gold, silver,
and clothes” (Albornoz 1990 [1570]: 75)19. This negation of Spanish tradition and Catholic
religion was partnered with a revitalized belief in various ceremonies and rituals that were
performed in the time of the Inka, as well as restored and fortified huaca worship and adoration.
Like Molina’s account, the witnesses in Albornoz’s Informaciones attest to native rejection of
Spanish religious teachings in favor of a return of the huacas and rituals which dominated in
prehispanic times.
Although the accounts of Molina and Albornoz broadly align, where Molina is very
general and vague in his descriptions of Taki Onqoy, Albornoz is extremely specific in his
passages and testimonials, naming specific places, huacas, and individuals. In his 1577 account,
Cristóbal de Albornoz consigns Cristóbal de Molina to testify on his behalf. In this account,
Molina offers more information than in his Relación, arguing that “in the provinces of Soras and
Apcara and Lucanas, I find a large sum of instigators and performers of Taki Onqoy” and that
Translated from “havia descubierto la seta e apostasia y predicacion que llaman Taki Ongoy por otro nombre
Aira” (Albornoz 1990: 61), translation mine.
19
Translated from “dezian que no creyesen en Dios ni adorasen las cruzes ni a los santos, ni entrasen en las yglesias,
ni se confesasen con los clerigos ni frailes, sino con ellos, e que ayunasen ciertos dias de la semana conforme a los
ritos e ceremonias del tiempo de los yngas, y que ofrescian en las guacas que tenían oro, e plata, e ropas” (Albornoz
1990 [1570]: 75), translation mine.
18
51
“the principal instigators were two men and one woman” (Albornoz 1990: 180-182).20 Here,
Albornoz and Molina work together, with Molina attesting that the majority of takiongos were
from the provinces of Soras, Apcara, and Lucanas, and even describing the leaders of the
movement. Later, in Albornoz’s 1584 Información, the extirpator interviews Molina for a second
time. It is in Molina’s account of 1584 that the preacher-leader of Taki Onqoy is named: “the
most guilty were two natural Indians and one indian named don Joan Chocne or Joan Chocna”
(Albornoz 1990: 225).21 While the other two leaders remain unnamed at this point, witnesses in
the 1570 Información document suggest that some of the leading women may have been called
“Santa Maria and Santa Maria Magdalena” (Mumford 1998:154). While it is in Albornoz’s
Informaciones that we receive much more detailed information about the leadership and
geographic extent of the movement, the actual names of the leadership are provided by Cristóbal
de Molina.
There are several other discrepancies in the witness testimonials of the Informaciones,
which again have been utilized for opposite interpretations of Taki Onqoy in the works of
historians of the movement. To understand these oversights, generalizations, and the veracity of
the accounts, one must parse the function of each letter to the King (1569, 1570, 1577, 1584), its
contents, and consider the letters in context with Albornoz’s life goals.22 Albornoz controlled
much of what was said in the interviews—his letters all have formulaic questions, which often
merely demanded only an affirmative or negative response from the witness (Albornoz 1990).
Yet, throughout the Informaciones, witnesses also provide unsolicited, additional information
Translated from “en las provincias de los Soras y Apcara y Lucanas, hallo gran suma de docmatizadores y
maestros del Taki Onqo” and that “los mas principals docmatizadores heran dos hombres e una mujer” (Albornoz
1990: 180-182), translation mine.
21
Translated from “los mas culpados fueron dos yndios naturales e una llamado el un yndio don Joan Chocne o Joan
Chocna” (Albornoz 1990: 225), translation mine.
22
To this end, Heilman 2002 gives an excellent, contextual description of each letter and its most proximate goals.
20
52
about practices they had seen or heard (Heilman 2002). Some witnesses even undermine the
severity of Taki Onqoy, suggesting that its practitioners were fickle, and often resorted to prayer
to Catholic saints in times of strife (Stern 1993; Albornoz 1990). The disarticulations and
discrepancies between the witness testimonies actually supports Albornoz’s reliability—had he
fully created the movement for his own benefit, the array of witnesses would likely all have held
similar accounts.
Finally, one of the most valuable aspects of Albornoz’s Informaciones is his Anexos de la
información (1584): Relación de la visita de extirpación de idolatrías, which was published as
an addendum to Albornoz’s 1584 document. Here, Albornoz lists the towns he visited during his
1569-1571 time in Huamanga, the Indians punished in each town, and the names of the idols or
huacas he destroys. This section of Albornoz’s manuscript is very thorough and expansive—so
much so that some secondary authors have argued that he exaggerated his actual presence and
work in the area (Ramos 1992). One of the towns Albornoz identifies is called Chicha—several
lines of evidence suggest that Iglesiachayoq was the town of Chicha in the sixteenth century: 1)
“Chicha” is listed in the Soras province near the pueblo of “Anansora” (modern location 12.5m
northwest of Iglesiachayoq) (Albornoz 1990 [1584]: 270); 2) the site has a large rectangular
structure constructed as an early Catholic church; 3) the site is not listed in the 1586 census,
suggesting that the people had been relocated to the reducción of San Pedro de Larcay (5.8km
northwest of Iglesiachayoq, Meddens and Schreiber 2010); 4) the modern hamlet of Chicha is
located only 1.25km southeast of the site, yet is small and lacks any Inka-era or early colonial
architecture; and 5) all other Soras-area sites listed by Albornoz have been identified and are
located near their modern locations. Based on their work in the region, other scholars have also
independently concluded that Iglesiachayoq was the village known as Chicha by the Spaniards
53
(Barnes 1985; Meddens 1994: 142-143). For the particular purposes of the current research, I
provide the accounts of Albornoz in which he specifically discusses those he punishes in the
town of Chicha:
[Margin]: 12 indians punished for the said sect from the pueblo of Chicha.
The ayras takiongos punished were Hernando Marca Susma, Mateo Supa, Joan
Mollo, and nine more natural Indians from the town of Chicha, subjects of don
Joan Hachi, were bound and whipped, sheared, and served as it was necessary
during the production of the largest church in this town and the province of
Atunsora, and three days a week had to appear before the priest for them to be
taught in the things of our holy Catholic faith. (Albornoz 1990 [1584]: 270).23
This passage serves as an example of Albornoz’s addendum, in which all of the entries are
extremely specific, naming individual people and places, and describing in great detail their
punishments. In Chicha, Albornoz identifies three names of those he punishes, attests that
beyond these three, he punished nine more, gives the name of the kuraka of Chicha (Joan Hachi),
and says that the Indians were whipped and their hair was cut. The whipping and hair cutting
would have been humiliating for these takiongos; in addition, they were also forced to help
construct the church at Chicha, as well as visit the priest a few days a week to study and practice
Catholicism (Figure 2.1). This type of punishment is commonly listed throughout Albornoz’s
addendum.
23
Translated from “[Al margen:] 12 yndios castigados por la dicha seta del pueblo de Chicha.
Fueron castigados por ayras taquiongos Hernando Marca Susma, Mateo Supa, y Joan Mollo y otros nueve indios
mas naturales del pueblo de Chicha, subjetos a don Joan Hachi; fueron acotados (whipped) y trasquilados (hair cut),
y que mientras durase la fabricación de la iglesia mayor de este pubelo y provincia de Atunsora, sirviesen en ella en
lo que fuese necesario, y que tres días en la semana se presentasen ante su cura para que fuesen ensenados en las
cosas de nuestra santa fe católica.” (Albornoz 1990 [1584]: 270), translation mine.
54
Figure 2.1. View of Iglesiachayoq looking north with casa de kuraka (left) and church (right).
Additionally, Albornoz describes the violent manner in which he destroys the huacas and
idols of each town. As in the section above, I will record the entire passage for the town of
Chicha, so that I can convey the type of information given in these accounts, as well as the exact
names of the destroyed huacas in Chicha.
[Margin:] 32 idols of the kurakas of the town of Chicha
Found and discovered 32 idols of the kurakas of the town of Chicha, Joan Hacha,
head of the town, those which were called Supayco, Guamancapcha, Quilca, their
pacarinas Coalla visa, Malloalca, Tuputa Anpalla, Vilucha, Guayoa Supaico,
Guman Capcha, Guamanilea, Anra, Yachi, Guahucata, Ronto, Angas Guacsa,
Mollomolla, Quipchiparca, Taxcarima, Apaxallaquio, Caxapi Guazmi, Ochuylla
55
Poboznos seven llamayllas, three mamasaras, one Vilca;24 almost all, on the order
of Extirpator Cristóbal de Albornoz were brought and publicly burned like the
others, and those which were not brought because they were too big were burned
by Joan Cocha Quispe. (Albornoz 1990 [1584]: 275).25
Again, this part of Albornoz’s manuscript is extremely useful in learning the names of the huacas
in each town. Some of these names may be linked with toponyms in the region today.
Additionally, while some of the accounts provide descriptions in the punishment of people who
practiced Taki Onqoy, the account above also confirms that the portable huacas were brought to
Albornoz and burned publicly, while those that were too large to move were burned by Jon
Cocha Quispe. While Molina provides general details about the Spanish response to Taki Onqoy,
Albornoz instead gives direct, specific explanations of his role in punishing both Indians and
huacas for their roles in Taki Onqoy.
Albornoz’s final work, the Instrucción, does not rely on the testimonies of different
Spaniards throughout Peru, but rather is a description of how to find and identify various types of
huacas across the Andes. Published by Duviols in 1967 in an article entitled “Un Inedit de
Cristóbal de Albornoz: La Instrucción para Descubrir Todas Las Guacas del Piru y Sus Camayos
y Haziendas,” this document contains notable contributions to the study of Taki Onqoy
(Albornoz and Duviols 1967). First, there is a narrative which recounts all types of huacas
known to Albornoz, where to find them, and how to eradicate them. Second, Albornoz mentions
the types of dances and songs associated with particular ceremonies, along with the time of year
According to my research, “llamaylla” may be referring to llama-illas, or idols in the forms of llamas. Mamasaras
are corn cobs or potatoes, and vilcas are a type of poisonous, perhaps hallucinogenic fruit (Albornoz 1989 [1580])
25
Translated from “[Al margen:] 32 ydolos de los caciques del pueblo de Chicha
Fueron halladas y descubiertas las guacas de don Joan Hacha, principal del pueblo de Chicha, las quales se llamaban
Supayco, Guamancapcha, Quilca, sus pacarinas Coalla visa, Malloalca, Tuputa Anpalla, Vilucha, Guaoya Supaico,
Guman Capcha, Guamanilea, Anra, Yachi, Guachucata, Ronto, Angas Guacsa, Mollomolla, Quipchiparca,
Taxcarima, Apaxallaquio, Caxapi Guazmi, Ochuylla Poboznos siete llamayllas, tres mamasaras, una Vilca; casi
todas ellas por mandado del dicho señor Visitador fueron traídas y se quemaron públicamente según que las demás,
y las que no se truxeron por ser grandes las fueron a quemar don Joan Cocha Quispe.” (Albornoz 1990 [1584]: 270),
translation mine.
24
56
these rituals were performed. Finally, Albornoz provides general descriptions and names of the
important huacas and landscape features in numerous provinces.
Archaeologically, the witness testimonies in the Informaciones and the general
information regarding huacas and huaca destruction in the Instrucciones provide supplemental
information especially with regard to punishment for Taki Onqoy practices. Evidence of
punishment for these practices allows scholars to broaden what archaeological signatures might
look like and by extension, affirm Taki Onqoy practices. As Table 2.3 shows, these practices
include positive signatures—burned fragments of huacas (ash lenses or stone fragments in the
central plaza), or the physical embodiment of punishments (evidence of broken bones and
defensive wounds on Taki Onqoy skeletons). Additionally, practices which would leave no
material signature would be the shaving of hair and the forced attendance at church.
Source
Year
Positive Signature
Albornoz
1569
1570
1577
1584
Offerings to huacas of gold, silver, clothing
Physical injuries on skeletons
Evidence of burning and destruction of
huacas in central plaza
Absence of
Evidence
No crosses
or Catholic
objects
No
Signature
Avoidance
of churches
No
confession
Table 2.3. Taki Onqoy correlates from Albornoz.
The accounts of the two Cuzco priests thus are the main fodder for secondary analyses of
Taki Onqoy. From their works, scholarship has thoroughly assessed the gaps and overlaps, yet
historians still mostly utilize an existence/negation approach to Taki Onqoy: what was the nature
of this movement? What was the geographical extent? Did it exist or was it a construction of
Albornoz? In clearly laying out the material correlates of Taki Onqoy reported by these two
authors, I have created a baseline for material interpretation of the movement, which will be
57
addressed in a following section. In the subsequent section, I discuss the two new “alternative”
sources brought to light by Peter Gose in his 2008 Invaders as Ancestors.
Alternative Accounts: Álvarez and Sotelo
While Molina, Albornoz, and Guerrero constituted the major sources for twentieth
century secondary debates on Taki Onqoy, Gose’s 2008 work introduces two new sources which
have heretofore not been thoroughly assessed. The first of these, Bartolomé Álvarez’s Memorial
(1991 [1588]), was not published until after the 1990 Retorno de las Huacas, and thus represents
a new source written by a Spanish priest who worked in what is today Bolivia. The second of
these sources, a letter from Gaspar de Sotelo to the Cabildo of Cuzco, was published by the
Peruvian historian Lohmann Villena in 1941, but has been considered a part of the Vilcabamba
uprising literature rather than the Taki Onqoy literature (Gose 2008). Gose uses both sources to
argue that Taki Onqoy was indeed the religious counterpart to the Vilcabamba uprising, yet both
sources can also provide more clues for understanding the materiality of Taki Onqoy.
Prior to the surfacing of his manuscript in Madrid in 1991, Álvarez was a relatively
unknown figure in Peruvian history (Villarias Robles and Martin Rubio 1998). Álvarez worked
as a parish priest in a small town called Pampa Aullagas, in what is today Oruro, Bolivia during
the time he wrote this manuscript. He also spent time in Sabaya and Potosí, and his work
demonstrates his frustrations with indigenous peoples who regularly participated in what he
deemed idolatrous practices. The Memorial was a letter to Felipe II with the intent of warning
the King about the failures of evangelization and conversion; throughout his entire document, he
denigrates Andean peoples and refers to them as “bestias” or “sucios” (Álvarez 1998 [1588]). As
such, it negates much of Albornoz’s arguments about his role in successfully eliminating
58
idolatrous practices in Huamanga, and suggests that rather than two distinctive extirpation
campaigns (one with Albornoz in the 1570s and another in the 1600s), the documentation of
attempted eradication of idolatrous practices was ongoing through the last decades of the
sixteenth century.
Álvarez’s work represents an “alternative” source because it was written at a later date
than the primary corpus of source documents (1587-1588) and from a town which was close to
1,000 kilometers outside of the Taki Onqoy central region. However, if we accept his work as a
legitimate new source (as do Villarias Robles and Martin Rubio, the editors and contributors to
the 1998 volume), then Álvarez also provides critical new details about Taki Onqoy specifically,
and also about huaca worship and idolatry in general. Throughout the Memorial, Álvarez argues
that indigenous peoples continue to worship the idols and ancestors of their past, and that they in
turn instruct young Andeans in the ways of the ancestors as well. Specifically, he gives thorough
descriptions of Taki Onqoy:
During this ceremony, either in a group or sometimes only one or two people who
want to do the ceremony, begin to sing a song that is not words nor reasons nor
sentences that can be understood as saying something. It only sounds like ‘u, u, u,
u’: it is necessary to hear it and see it to understand it, and as such it cannot be
written. And with this high-pitched song, they clap their hands and stomp their
feet and move the head from side to side, such that the whole body is working.
And they stop after three or four days and nights, as long as the forces last, and
they do not stop if they need to defecate or urinate (Álvarez 1998 [1588]: 124125).26
Translated from “Es la fiesta que, juntos dellos la cantidad que se conciertan – y a veces uno o dos solos que
quieren hacer la cerimonia- comienzan a cantar un cantar que no es palabras, ni razones ni sentencias ni cosa que se
pueda entender que dicen algo. Solo suena ‘u, u, u, u’: es menester oirlo y verlo para entenderlo, que es tal que no se
puede escribir. Y con este canto muy alto estan de pie, dando de pie y mano, alzando un pie y abajando otro, y
asimismo hacienda con las manos, los punos cerrados, meneando la cabeza a un lado y a otro, de suerte que con todo
el cuerpo trabajan. Y paran en este canto tres o cuatro dias con sus noches y mas: lo que las fuerzas les duran, que no
cesan si no es que les venga necesidad de hacer camara o de orinar; que a esto salen, y luego vuelven a la tahona del
demonio” (Álvarez 1588: 124-125), translation mine.
26
59
The practices described by Álvarez indicate how takiongos would have performed their
songs and dances, and especially emphasizes how taxing the performance was on the body. In
conjunction with this song and dance, Álvarez goes on to say that:
They eat almost nothing or nothing, the most is coca, which they ordinarily have
in their mouth for this exercise. They drink as much as they want; the result of
this drinking and the whole body working and swaying with the head – thin from
their exhaustion and lack of food and drunkenness, lacking nourishment and
strength—fall on the floor among the others who look similar, so that little by
little they all fall down…If they die afterwards, they get them together with gold
and silver and feathers and coca and chaquiras27 that the women bring as chokers
that they call mollo and capa and they put them there together (Álvarez 1998
[1588]: 125).28
With this passage, Álvarez affirms the practices of excessive drinking and fasting.
Materially, assemblages might contain evidence of large chicha vessels, evidence of the chicha
itself, and remains of coca. In this passage, Álvarez also mentions the exhaustion and danger
provoked by fasting, dancing, and drinking—he goes on to affirm that those takiongos who
drank the most, who passed out, or even those who died, were the most respected and were
considered deities. If takiongos died during their performances, they would be interred with coca,
gold, silver, and feathers, and also “mollo” and “capa.” Here, “mollo” is the spondylus shell and
“capa” likely refers to the sound that chewing makes (Álvarez 1998 [1588]: 126). With these
clues, we are again provided the material practices and materials associated with Taki Onqoy
performances.
Chaquiras are defined by the authors as “que llamamos gargantillas de las mujeres” (Álvarez 1998 [1588]: 415).
Translated from: “No comen casi nada o nada; lo mas es coca, que en la boca tienen de ordinario para este
ejercicio. Beben todas las veces que quieren, sin gana o con ella: lo cual es causa de que con el mucho beber y aquel
ordinario trabajar con todo el cuerpo, y aquel devanear con la cabeza, a que – desflaquecidos por el cansancio y la
falta de comida y borrachera, faltos de aliento y fuerza- caigan en aquel suelo entre los otros; los cuales todos estan
casi de aquella figura, de suerte que poco a poco van cayendo hasta que todos tumban…Si muere luego, juntan un
poco de oro y plata, y plumas –de las que ya he dicho- y coca y chaquiras –de las que traen las mujeres por
gargantillas, que llaman “mollo” y “capa” – y se lo ponen allí junto.” (Álvarez 1998 [1588]: 125), translation mine.
27
28
60
Finally, in the most illuminating passage of Álvarez’s account, the Spanish priest asserts
that Taki Onqoy also had an Aymara name:
They call this exercise in Aymara talausu and in the tongue of Cuzco taquiongo
which means “song sickness.” When the evil of the Indians began to be
understood, that many years ago the whole earth was contaminated so that some
people died of nonsense, the remedy for it was so weak that it did not stop the
idolatry; and, if something stopped, it was only in name because the Indians still
perform it in secret, as in all of the other things that I have said. And so I
confessed in the confession: that, as it is common for them to confess the sins of
others and not mention theirs, if they met others who performed that diabolic
song, or watched or acted, they say what they saw and not what they did. (Álvarez
1998 [1588]: 126-127).29
The most important aspect of this passage is that the movement was well-known even in Bolivia,
so much so that those who spoke Aymara had their own name for the movement, called talausu.
Archaeologically, other evidence in the above passage speaks to where the movement was
enacted. Álvarez argues that the punishment (likely referring to those levied by Albornoz in the
Huamanga region) was not enough to prevent Andeans from continuing their performances.
Rather, the movement itself was still practiced in secret, likely closed, private spaces away from
the watchful eyes of Spanish authorities. Álvarez’s accounts thus support earlier testimonies
from Molina, Albornoz, and Guerrero, but also supply other details at a later time and from a
town in central Bolivia, far from the Taki Onqoy heartland. That Álvarez affirms its existence
nearly 20 years after its apogee and 1000 kilometers away from its center suggests that Taki
Onqoy had spread to far extents of the Andes. Moreover, initial extirpation attempts by Olvera
29
Translated from: Tienen despues a estos tales en veneracion, como a hombres dedicados a su diabolico culto;
llaman a este ejercicio en lengua aimara talausu, y en lengua del Cuzco taquiongo, que quiere decir “canto enfermo.”
Cuando se comenzó a entender la maldad de los indios, que ha muchos anos estaba toda la tierra contaminada de
suerte que del disparate morían algunos, el remedio que a esto se puso fue tan liviano que no basto a que cesase; y si
en algo ceso, no fue mas que en la publicidad porque en lo secreto se hace, como en todas las otras cosas que he
contado. Y asi en la confesión lo he sacado: que, como es ordinario en ellos confesar los pecados ajenos y callar los
suyos, si se hallaron con otros que hacían ese canto diabólico, o mirando u obrando, dicen lo que vieron y no lo que
hicieron” (Álvarez 1998 [1588]: 126-127), translation mine.
61
and Albornoz were successful in name only—in these later years of the movement, it was
practiced in private spaces which would have been hidden from Spanish authorities. Table 2.4
lays out the new information from Álvarez’s works, and reiterates some of the positive correlates
and non-signature practices.
Source Year
Positive Signature
Álvarez 1588 Coca
Chicha and chicha vessels
Gold, silver, feathers, and chaquiras
Mollo
Private spaces
No Signature
Singing and chanting
Clapping and stomping
Defecation and urination
Table 2.4. Taki Onqoy correlates from Álvarez.
The last primary source document related to Taki Onqoy, a letter from Gaspar de Sotelo
to the Cabildo of Cuzco, does not mention the movement by name, but does suggest that there
existed a religious counterpart to the Vilcabamba Revolt in the 1570s. Within the secondary
literature on Taki Onqoy, this source (as with the Álvarez) is only utilized by Peter Gose, who
cites Sotelo’s letter as evidence that Taki Onqoy was connected to the Vilcabamba militaristic
revolt. Sotelo warns that the arrival of
those of Parinacochas because the Cacique of Villagran who is named don Juan
Chancabilca sent a son of his whom I know dressed in scarlet like a Spaniard to
this land (Jauja) to preach the sect. They captured him in Huánuco and he said
everything that I write. Through him, I learned about this wickedness because
what was said in this city about the resurrection of Pachacamac began in
Parinacochas, and they made great sacrifices and offered much livestock to the
Devil of Pachacamac” (Lohman Villena 1941:3, as quoted in Gose 2008: 85,
emphasis mine).
There are two critical parts of this passage: first, Sotelo warns that individuals from Parinacochas
had arrived to “preach the sect.” Sotelo thus confirms Guerrero’s warnings about the apostasy
being performed in Parinacochas, and that it was spread through charismatic wandering
62
preachers. Second, as I have emphasized in the passage above, Sotelo affirms that these
preachers were calling for the resurrection of Pachacamac through a renewal of worship which
took the form of sacrifices to the neglected huacas. These practices were indicative of Taki
Onqoy, even at this relatively early phase in the movement (1565).
Utilizing the accounts of Guerrero, Molina, Albornoz, Álvarez, and Sotelo, I have first
mapped out the material practices associated with the movement. Through understanding the
basic materials associated with Taki Onqoy, I have demonstrated what types of anthropological
questions can thus be drawn and how we might be able to understand material correlates of those
questions (to be discussed later in the chapter). While these primary sources are useful in
elucidating the archaeological correlates of Taki Onqoy, secondary debates on the movement
have instead focused on broad questions related to the movement, rather than how it was enacted
and performed by Andeans. In the following section, I highlight some of the themes in this
debate and suggest that while useful to consider, they do not offer any greater understanding of
how the movement was actually practiced.
Scholarly Debate of Taki Onqoy
Although there are only a limited number of primary Spanish authors who penned
accounts of Taki Onqoy, the works (and especially those of Molina and Albornoz) have
prompted a spectrum of secondary responses by twentieth-century historians, indicative of the
gaps and allowances of documentary sources. Though this debate has exhausted the works of
both Molina and Albornoz, I argue that this secondary literature has not considered the
movement from a material, Andean ontological perspective, but has instead focused on four
predominant themes: 1) the “nature” of the movement, 2) the geographic expanse of the
63
movement, 3) Taki Onqoy’s connection with Vilcabamba, and 4) the personal motivations in
Molina and Albornoz’s accounts. These four themes are critical to consider, but would be better
understood through an understanding of how practices shaped Taki Onqoy and what these
practices were doing in terms of structuring local culture. In the following sections, I discuss
each of these themes, outlining the major debates which have framed how Taki Onqoy is thought
of today.
Historiography of Taki Onqoy Scholarship
There are two notable trends in the methods and tactics of those authors who have written
about Taki Onqoy in the twentieth and twenty-first centuries—those who take the documents
more or less at face value, and those who prefer a revisionist explanation of the movement,
grounded primarily in biographical studies into Albornoz and Molina. In the 1960s, Luis
Millones initiated the debate on Taki Onqoy through the publication of two articles on the
movement, which he published in Lima journals (Millones 1964, 1973; Mumford 1998).
Founded on analysis of Molina’s work and the never-before-studied work of Albornoz, Millones
sought to introduce what appeared to be a very important cultural movement in the 1560s. After
the publication of Millones’s two articles, there was a renewed interest in finding and
discovering historical Spanish documents related to this critical time of conquest and conversion.
To this end, Pierre Duviols discovered, transcribed, and published Albornoz’s Instrucción in
1971. Subsequent publications on Taki Onqoy reflected the general trends in history as a
discipline, particularly throughout the 70s, 80s, and 90s. After the seminal 1990 publication El
Retorno de las Huacas, compiled by Luis Millones with other contributions from Peruvian
historians, discussion of Taki Onqoy faded until Gose’s 2008 Invaders as Ancestors. Though
64
Gose introduced two new sources, the debates were exhausted—through my emphasis on
materiality of Taki Onqoy, I hope to reinvigorate research into the movement and also generate
new questions.
Theme One: What was the Nature of Taki Onqoy?
The first theme that secondary scholarship has thoroughly debated is the “true nature” of
Taki Onqoy. The authors who consider this theme question what the movement signified for the
people practicing it—a cultural renewal, a cultural renaissance, a symptom of cultural illness, or
even a psychological coping mechanism. When the Taki Onqoy movement was brought to the
fore in the second half of the twentieth century, the modern people of what is today Huamanga,
Apurímac, and Huancavelica (the heartland of Taki Onqoy) embraced the movement as a symbol
of indigeneity and resistance (Arguedas 1968; Bush 2012). With this new significance, the
secondary scholarship on Taki Onqoy logically pursued the question of what the movement
meant for those practicing it. Yet, since all of the evidentiary accounts were written by Spanish
authorities, this question is somewhat misguided. In the first interpretations of Taki Onqoy,
Millones and Duviols, both largely took Molina and Albornoz’s accounts at face value. Millones,
drawing from Molina’s Relación and Albornoz’s Informaciones suggested that Taki Onqoy was
a reconquest of Andean religion. Similarly, Duviols, while primarily relying on Albornoz’s
Instrucción, considered Taki Onqoy to be a revitalization of Inka religion.30 After Millones’ and
Duviols’ publications in the 1960s, scholars who debated the movement generally took one of
two major perspectives on the movement: that it was a revitalization of either the Andean or Inka
30
I will discuss the connection with the Inka more thoroughly in Theme Three.
65
state, or that it was an irrational response to mass death and heavy labor and tribute demands
(Mumford 1998: 159).
In 1979, Guillermo Cock and May Eileen Doyle wrote a chapter called “Del culto solar a
la clandestinidad de Inti y Punchao” in which they supported Duviols and argued that Taki
Onqoy was indeed a resurgence of Inka belief systems. Cock and Doyle provide several
instances in which the cult of the sun was discussed in the seventeenth century and linked to
other divinities, explicitly in opposition to European belief systems (Cock and Doyle 1979: 55).
They argued that these links between the solar cult and clandestine Andean religion persisted
underground for at least two hundred years. Though Cock and Doyle make an interesting case
for the continuation of Inka religion in some regions of the Andes, they do not fully elucidate the
significance of these connections to Taki Onqoy.
Apart from Cock and Doyle’s affirmation of the Inka State, the mid-1970s witnessed a
spreading trend in the study of history and historical documents moved away from political
history (a top-down approach) and shifted to a focus on bottom-up social histories, in which nonelites became the subjects of interest (Mumford 1998: 157). This movement from political
histories to social histories led to a shift in focus for scholars of Andean studies. Coupled with a
proliferation of archaeological and ethnohistorical studies into the Inka Empire which affirmed
its relatively short duration, the turn to social histories deemphasized the Inka Empire in favor of
Andean groups (Spalding 1984; Stern 1993). With regard to studies of Taki Onqoy, Duviols’
argument that the movement represented a resurgence of the Inka State was discredited,
supported by the fact that the Inka Sun God was not listed as a central aspect of the movement
(Mumford 1998).
66
In the mid- to late- 1970s, Franklin Pease and Marco Curatola discussed Taki Onqoy as a
distinctly Andean, (and explicitly not Inka) revitalization movement. In his El dios creador
andino and his article “El mito de Inkarri y la vision de los vencidos,” Pease argued that:
The Taki Onqoy is a clear expression of a cyclic Andean messianism. The Spaniards
are part of an image of chaos and a new cosmos that must be organized and that
means a clear return to the previous world, the huacas, as in the movement of 1565
was not a return to ‘the time of the Inka’ and there is no link with the ‘past’ as ‘Inka
past’ (Pease 1973: 450).31
Here, Pease states his opinions on the nature of Taki Onqoy very clearly: instead of affirming a
link between the Inka past and the ‘past’ referred to by takiongos, Pease suggests that the
movement was actually a revitalization of the pre-Inka past. Pease suggests that pre-Inkan
Andean beliefs in the cyclical nature of time influenced the practice of Taki Onqoy—the Spanish
conquest and simultaneous chaos which accompanied it fit into an Andean conception of time. If
the cycle continued as it had throughout Andean history, then the Spanish reign would eventually
pass and return to a time of huaca worship and Andean tradition.
Like Franklin Pease, Marco Curatola supported the idea that Taki Onqoy was not
founded in Inka religion, but rather a grassroots Andean cultural tradition. Curatola argued that
Taki Onqoy represented the “consciousness of the Indians in the Andes, a self-identification as a
unique nation” (Curatola 1976: 85). Curatola’s argument is slightly different from Pease’s—
while Pease viewed Taki Onqoy as a version of the messianic Andean cycle, Curatola instead
Translated from “El Taki Onqoy es una clara expresion de un mesianismo andino ciclico (cuya existencia
advirtieron entre otros Imbelloni y Zuidema). Los españoles forman parte de una imagen de caos previa al nuevo
cosmos que debe organizarse y que significa un claro retorno al mundo anterior, al de las guacas, pues en el
movimiento de 1565 no se intenta un retorno ‘al tiempo del Inka,’ no ay vinculación con el ‘pasado’ en cuanto
‘pasado incaico’” (Pease 1973: 450), translation mine.
31
67
saw Taki Onqoy as a larger social movement, one in which Andean peoples sought to define and
identify themselves as a distinctive culture.
In the 1980s, the social history, bottom-up focus of the 1970s continued logically through
the investigations of several Andean scholars, who completed works founded in specific local
groups or regions rather than lo andino as a whole (MacCormack 1991; Spalding 1984; Stern
1993 [1982]). In her 1988 article “Pachacuti: Miracles, Punishments, and Last Judgment,”
Sabine MacCormack described the rituals and rites associated with Taki Onqoy as “attempting to
cure the disease of invasion as expressed in the alien government of Spain and the alien religion
of Christianity…Andean rituals and beliefs that articulated both the Taki Onqoy and subsequent
responses to Hispanization accordingly confronted and refuted the arguments and propositions of
missionaries” (MacCormack 1988: 984). In other words, Taki Onqoy was a nationalist
movement which attempted to “cure” the problems and concerns of Spanish presence through
Andean resurgence.
In Karen Spalding’s Huarochiri: An Andean Society Under Inca and Spanish Rule, the
author provides a description and brief history of perceived indigenous heresy in the 1540s and
1550s. In the first generations of Spanish invasion of the New World, she argues that the
justification of conquest was Christian conversion. However, using the excuse of conversion as
an explanation for conquest did not precede Spanish takeover, but followed it. Thus, for the first
two decades after Spanish conquest, Andean ritual and state structure was tolerated. In the 1550s
however, in a time of growing rebellion marked by the Inka in Vilcabamba and the
disillusionment of local groups and kurakas, Spanish authorities began to take a more active role
in regulating and legislating proper means of conversion. Church authorities sought to convince
local peoples that ancestor worship (and specifically the worship of Andean ancestors as opposed
68
to European saints) was a form of allegiance with the devil (Spalding 1991: 245). In return,
Spalding argued that this attack on local beliefs did not go unanswered, and that Taki Onqoy was
“a struggle to reestablish the old traditions and relationships” (Spalding 1991: 246). That is, Taki
Onqoy was a movement which sought to reaffirm the power of the huacas and local religion. The
rise of the huacas went hand-in-hand with the rejection of Spanish religion and lifeways. In order
to support the huacas, Andean peoples also had to rebel against Spanish presence and its various
forms.
All of these authors viewed Taki Onqoy as a nationalist resurgence, one which sought to
reestablish Andean or Inka tradition while simultaneously rejecting Spanish culture and religion.
For these authors, it was a reclamation of Andean culture and beliefs. At the opposite end of the
spectrum, many authors writing in these same periods instead viewed the movement as a symbol
of cultural illness and despair. Rather than a form of nationalist expression, these authors
considered the movement to be a reflection of an irrational attempt to counter the widespread
death and disease at the time. During the 1970s, 80s, and 90s, these authors were all utilizing the
same two sources, yet these interpretations—that the movement was either a symbol of cultural
revitalization or a symbol of cultural illness—are contradictory and ultimately unresolvable
through the existing data sources.
The second camp debating the “nature” of Taki Onqoy viewed the movement as a symbol
of cultural illness, a psychological irrationality, or a crisis cult. The first of these authors was
Nathan Wachtel, who published The Vision of the Vanquished (Wachtel 1977).32 Wachtel’s main
contribution to the study of Taki Onqoy was his psychological interpretation of the movement—
32
Originally published in 1971.
69
the Spanish invasion caused massive ruptures in Andean society, and Taki Onqoy was a form of
defense against these ruptures. When discussing the motivations and practices of Taki Onqoy,
Wachtel writes that “Indians expected their liberation to come from the victory of the huacas
over the Christian God,” and that these “expectations of divine intervention added up to a praxis
of an essentially magic and religious nature” (Wachtel 1977: 183). Taki Onqoy was thus actually
“the expression of an Indian theory of the state of the disjunction between Indians and Spanish,
together with the hope that it would be transcended” (Wachtel 1977: 183). For Wachtel, then,
Taki Onqoy was not a nationalist reclamation of local beliefs, but essentially an internal,
individual reaction and defense against a changing and chaotic period of conquest. Taki Onqoy
did not emerge from a point of cultural pride, but instead was driven by guilt and fear.
In his magisterial, documentary study of the aftermath of Spanish conquest, Steve Stern
argued that Taki Onqoy was fundamentally a local movement which arose from a sense of
distance and disillusionment for the indigenous people of Huamanga (Stern 1993 [1982]). Far
from the Inka rebellious factions at Vilcabamba and located in rather distant, highland towns, the
Huamanginos were neglected by both Spanish and Inka authorities. In fact, Stern argued that
Huamanga was relatively less affected by Spanish disease, with the region’s mortality rate at
only 0.5% per year, compared with much higher percentages in other regions (Stern 1993: 44).
Local elites in Huamanga were additionally very strategic in their alliances with Spanish
authorities, initially benefitting through gifts of land and luxury goods. However, being removed
from intensive Inka and Spanish presence caused less of these state influences in the Humanga
region, and promoted a continuation of local practices. As Stern argues, “what distinguished the
millenarian crisis which inflamed southern Huamanga from other anticolonial movements was
this predominance of an inward-looking moral drama” (Stern 1993: 69). Stern suggested that the
70
movement was primarily internal and thus irrational—it attacked the misdeeds of native peoples
who had abandoned their huacas instead of the Spanish atrocities, and thus was ultimately
defeatist and bound to fail.
Stern’s argument that Taki Onqoy was largely internal contradicts the accounts in the
primary sources, which suggest the movement was as much an attempt to thwart Spanish
evangelization through practices upholding the huacas. Furthermore, Stern seems to affirm that
the only “success” of Taki Onqoy would be an overthrow of the Spanish government: “One
could argue with some truth that Taki Onqoy’s inward emphasis on moral purification was
politically defeatist, that rejection of Spanish society was ambivalent, that millenarian dreams
failed…that [the movement] provided Huamanga’s peoples renewed spirit and hope, but not a
path toward effective liberation” (Stern 1993: 70). Stern’s assertion that Taki Onqoy was not a
means of “effective liberation” neglects the material practices of the movement—by returning to
huaca worship, to the clothing, feeding, and dancing with the huacas, Andeans were liberating
themselves from the confines of exclusionary Catholic religion.
One last author who considered the “true nature” of Taki Onqoy to be associated with a
cultural disorder was Moises Lemlij, who utilized Albornoz’s works to argue for a movement
caused by psychological guilt. Lemlij’s article “El Taki Onqoy: reflexiones psicoanaliticas” cited
the fact that some of the leaders of Taki Onqoy utilized Spanish saints’ names as evidence that
Andeans felt remorseful about their participation in Spanish culture (Lemlij and Albornoz 1990:
427). Lemlij et al. write that “we see that the fragmentation and internal and external death
reproduce a closed circle, or that the forces of life regroup and recover in the last bastion: the
body” (1990: 430). Inspired by Freud’s psychoanalytical perspectives, Lemlij et al. argue that
when the internal and external worlds of the Andeans were breaking down, their ultimate refuge
71
was their own bodies. Thus, Taki Onqoy’s central practice was dancing, which took the form of
possession or illness in the body, it would be reasonable to assume that Andean bodies were the
vessel through which takiongos practiced their resistance. Indeed, these sorts of bodily
resistances are present in several other millenarian movements (Brown and Fernandez 1991;
Harkin et al. 2004).
All of the authors I have discussed thus far focus on the theme of the “nature” of Taki
Onqoy, and all use particular aspects of Molina’s and Albornoz’s documentary accounts to
support their claims. Some of these authors even consider the practices associated with Taki
Onqoy—specifically huaca worship and spirit possession—as indicators of either cultural
revitalization or ineffectual, personal subversive acts. However, discussions of the materials of
the movement, those I have highlighted above, are noticeably absent. Given the ontological
frameworks of Andean belief systems, this absence is striking, and assessment and understanding
of the materials which structured practices has the potential to offer new avenues of investigation
for this fraught time in Andean history.
The central problem with debating the “true nature” of Taki Onqoy is that it requires
assumptions regarding the state of mind of Andeans during the 1560s, and it homogenizes the
diverse reasonings behind Taki Onqoy practices to one motivation. Aspects of this debate thus
fall into the “resistance/dominance” trap (Chapter 3), in that they attribute the same catalyzing
factors to the entire population of takiongos rather than considering why the movement arose in
certain places, and how people performed it differently.33 By shifting to a material practice
33
See Ortner 1995 and Brown 1996 for a more thorough discussion of the problems of the resistance concept and
the homogenization of whole populations.
72
perspective and investigating where these practices emerge in different places within the Andes,
we can then ask more specific questions which are actually answerable.
Theme Two: What was the Geographic Expanse of Taki Onqoy?
The second major theme in scholarly debates of Taki Onqoy is related to its areal extent,
and has important implications for general comprehension of the primera evangelización, and
the array of Andean responses to this period. The geographic spread of Taki Onqoy is also a
central theme of debate because its main practices—huaca worship/possession, drinking, and
dancing—were all considered idolatrous and were a contributing factor to Viceroy Francisco de
Toledo’s 1570s mass resettlement program (Toledo 1986). From those authors who exclusively
relied and Molina and Albornoz, the Taki Onqoy movement began in Vilcabamba, in the
Department of Cuzco, but was discovered in Parinacochas (at that time also in the bishopric of
Cuzco) (Albornoz 1990; Duviols 1971; Millones 1964, 1973; Molina 2010). Molina and
Albornoz (as well as Álvarez) all asserted that the movement was incredibly extensive, and
encompassed Chuquicaca, La Paz, Cuzco, Huamanga, and even Lima and Arequipa.
73
Figure 2.2. Map of all provinces and departments in Peru mentioned as being home to Taki
Onqoy practices.
More recent debates on Taki Onqoy have reduced the proposed area of affectation,
evidenced by the belief that Molina’s account was based primarily on hearsay, and that
Albornoz’s letters were exaggerations (Ramos 1992). This debate, as with the others I discuss in
this section, is another that cannot be solved through the textual sources alone. Materially,
studying the geographic spread of Taki Onqoy is difficult: many of the positive signature
material practices of the movement would leave the same evidence as would general prehispanic
Andean religious practices throughout the Andes, even at this relatively later date. Similarly, the
74
practices which were especially unique to Taki Onqoy—spirit possession, rejection of Spanish
culture, and specific dances and chants—all leave either no signature or negative evidence, and
so are difficult to evaluate from a purely materials-based perspective.
Theme Three: Was Taki Onqoy Connected to Vilcabamba?
Like the previous two themes, the question of whether Taki Onqoy was connected to the
militaristic Inka uprising in Vilcabamba has also been an issue of great debate, and ultimately
one that is unsolvable from documentary resources, and which would also benefit from a study
of materials. More specifically, if Taki Onqoy and the Vilcabamba uprising were connected, we
would expect to find evidence of Taki Onqoy practices—those I have discussed in the previous
sections—in conjunction with or at the same sites as pikes and weapons caches near Vilcabamba.
The potential connection between Taki Onqoy and Vilcabamba is significant in that if they were
known to one another and even allied, the movement likely was much more organized and
expansive than the Spanish originally thought. Taki Onqoy and the Vilcabamba uprising were
contemporaneous events, and so their connection likely seemed logical to the Spanish priests
writing about both happenings.
The works of Molina and Albornoz both affirm that Taki Onqoy began in Vilcabamba,
and was spread by traveling Inka priests moving throughout the highlands (Albornoz 1990;
Molina 2010). Millones and Duviols echoed this affirmation, utilizing the accounts of these two
primary authors to verify this connection (Duviols 1971; Millones 1973). After the initial
publications by Millones and Duviols, the links between Vilcabamba and Taki Onqoy were
largely dismissed as a part of the bottom-up historical focus discussed in the previous section.
Since there was no mention of the sun deity, the Inka apical solar deity mentioned in connection
75
with Taki Onqoy in primary sources, scholars argued that there was likely no explicit link
between the Inka and Taki Onqoy (Curatola 1976; Pease 1973). However, in his recent Invaders
as Ancestors, Peter Gose staunchly reaffirms that Taki Onqoy and the Vilcabamba rebellion,
bolstering his claim with the additional letter from Gaspar de Sotelo.
Gose introduces a new source to his discussion of the early roots of Taki Onqoy, a letter
from Gaspar de Sotelo, an encomendero and trusted relative of the Viceroy Nuñez Vela, written
on March 24, 1565. This letter discusses both an uprising in Jauja and its links to Vilcabamba
and explicitly connects the Vilcabamba uprising with Taki Onqoy. In this letter, Sotelo makes
references to “Pachacamac’s resurrection and the ‘sect’ with an epicenter in Parinacochas” (Gose
2008: 85). Both details reiterate those mentioned by Molina and Albornoz regarding Taki
Onqoy: specifically that it was centered in Parinacochas and that Pachacamac was going to rise
again. Gose reaffirms the link between Vilcabamba and Taki Onqoy, writing that “seemingly,
extensive logical preparations for an uprising were in place, and the Taki Onqoy was their
ideological counterpart” (Gose 2008: 85).
Gose’s central contention is that Taki Onqoy was affiliated with Tupac Amaru, the
brother of Titu Cusi. While Titu Cusi was more willing to utilize Spanish written documentation
and work with Spanish forces, Tupac Amaru took a hardline attitude toward the Spaniards. Gose
invokes Albornoz’s accounts, which suggest that Taki Onqoy preachers came from Vilcabamba
and explicitly promoted Tupac Amaru—thus, Taki Onqoy may have “articulated not only a hard
line against Spanish colonialism, but also fraternal tensions within the Vilcabamba resistance”
(Gose 2008:89). Here, Gose suggests a brand new theory of Taki Onqoy: not only was Taki
Onqoy linked to the Vilcabamba revolt, but it also exacerbated internal tensions in the leadership
76
of this revolt. Viewing Taki Onqoy from this perspective, the movement became a “rallying cry
for Tupac Amaru’s militants” (Gose 2008: 91).
Gose’s argument about Taki Onqoy’s connection to Vilcabamba thus refutes the
revisionist school which deemed Taki Onqoy as an exaggeration or a figment of Spanish priests,
rather than a movement in its own right (see next section). His case is compelling and logical,
but would be better supported through a discussion of the materiality of the practices for which
he argues. While the religious beliefs of Taki Onqoy—specifically that of the rise of Pachacamac
and Titicaca—do indicate some sort of Inka influence, this could also be explained by the fact
that Andean society had been under Inka rule for the previous century. Those that were
practicing Taki Onqoy grew up under Inka influence or during Spanish conquest, so their belief
systems would have been impacted by Inka beliefs. If Taki Onqoy and Vilcabamba had the same
goal—to reconquer the Spaniards in planned uprisings across the Andes—then a potential
material correlate could be evidence of stockpiled weapons. In this case, the probability of
finding this sort of assemblage is very low and there is an absence of evidence.
If Taki Onqoy and Vilcabamba were actually linked movements, we would expect to see
related archaeological correlates in both Vilcabamba and the highlands. For example, if Taki
Onqoy’s message was also practiced in Vilcabamba, I would expect to see rejection of Spanish
goods and Catholicism evidenced at the physical site of Vilcabamba. To this end, there have
been few archaeological investigations into Vilcabamba. As reported by Bauer et al., there are
numerous findings of Spanish goods and artifacts in addition to a structure which appears to be a
Catholic church (Bauer et al 2015). Thus, though the archaeology of this time period is still in
nascent stages, the material findings from work at Vilcabamba do not support a link between
Taki Onqoy and the militaristic revolt.
77
Theme Four: Was Taki Onqoy an Exaggeration Levied by Spanish Priests?
A fourth major theme in Taki Onqoy scholarship was part of a revisionist take on the
movement which heavily critiqued Albornoz and Molina. Building on research into Albornoz’s
personal biography and motivations, these authors viewed Albornoz as an opportunist, one who
exaggerated the movement’s importance and his own role in it. Millones’ updated 1990
publication El Retorno de las Huacas not only contained a version of Albornoz’s Informaciones,
but also put forth several new essays regarding the priest’s life. This revisionist history uses the
disjunctures between Albornoz’s letters and the vagaries of Molina’s chapter as evidence that
Taki Onqoy was greatly exaggerated by the priests who wrote about the movement. The
revisionist takes on Taki Onqoy are important in that they highlight potential problems with
taking the texts at face value; as in Theme One, however, I again suggest that relying exclusively
on these two authors and the broader sociopolitical atmosphere limits understanding of how and
why the movement manifested through material practices.
In his “Nota preliminar al personaje historico y los documentos,” Pedro Guibovich Perez
contributes the first biographical study of Albornoz, albeit conceding that the references that
exist about the great extirpator of idolatry are “scarce and fragmentary” and largely come from
Albornoz’s own publications (Guibovich 1990: 23). Guibovich highlights important
contradictions between Albornoz and Molina: for example, while Molina attributes discovery of
the movement to the priest Luis de Olvera, Albornoz himself claims to be the original discoverer
of Taki Onqoy, and several witnesses support this claim (Albornoz 1990 [1577]). However,
Guibovich shows that Albornoz actually arrived in Peru in 1567, which postdates the letter from
Olvera in 1564. Additionally, Albornoz’s first letter (1569) contains no mention of Taki
78
Onqoy—instead, the name and descriptions of the movement first appear in his 1570 letter,
suggesting perhaps that Albornoz actually had no knowledge of the movement until this time.
Guibovich expanded his introduction in El Retorno in a 1991 article entitled “Cristóbal
de Albornoz y el Taki Onqoy” in which he gives a much more in-depth biographical study of
Albornoz, tracing Albornoz’s movements from his birth in Huelva, Spain, to his arrival in Peru
(Guibovich 1991:210). Guibovich argues that Albornoz was anxious to receive a promotion, and
thus utilized Taki Onqoy as a means of demonstrating the dire threats to evangelization and the
Spanish Crown and his own role in extirpating the movement. Guibovich writes “for this man
who was anxious to make merits to win a prompt ecclesiastical promotion, Taki Onqoy was the
occasion that presented itself and he [Albornoz] decided to conveniently exploit it for his own
benefit” (Guibovich 1991: 211-212). The letters do not reflect a growing, legitimate urgency in
the need for proper evangelization and conversion, but rather are representative of desires to earn
a bishopric or archbishopric from the Spanish crown (Guibovich 1991; Heilman 2002; Ramos
1992; Urbano 1992).
In a longer article in the same El Retorno volume, Rafael Varón seconds Guibovich’s
affirmation that Albornoz exaggerated his role in extirpating the Taki Onqoy movement. Varón
suggests that the sources are vague about several aspects of Taki Onqoy, especially in regards to
the movement’s connections to Vilcabamba and its relationship to local kurakas and Andean
officials (Mumford 1998). Varón confirms that Albornoz’s manuscripts are invaluable because
they contain lists of towns visited, names of takiongos punished, and names of huacas destroyed.
However, Varón argues that when analyzed with a more critical approach, “we should have in
mind that the document is abundant in exaggerations, and, like real life, many of the affirmations
of the witnesses are inexact” (Varón and Albornoz 1990:334). Varón essentially argues that if at
79
all possible, Albornoz’s works should be compared with other sources such as Molina, in order
to arrive at a more nuanced and truthful understanding of the movement.
A second major issue identified by Varón is the vagaries of the practices of Taki Onqoy,
and the way these practices and motivations are described by Molina and Albornoz. Both Molina
and Albornoz affirm that Taki Onqoy’s main manifestation was a frenetic dancing inspired by
contagion or huaca possession. The two authors also describe the movement as “a resurgence of
Andean huaca worship”—Varón, however, suggests that this supposed resurgence “did not have
the characteristics of a homogeneous cult practiced in all of Peru” (Varón 1990: 343). That is, the
main behaviors associated with Taki Onqoy, and with prehispanic huaca worship in general,
varied from province to province. Varón argues that there was no unified type of worship during
either prehispanic times or Taki Onqoy, and that thus the generalized idolatry described by
Molina and Albornoz is likely not completely correct.
Varón’s most striking argument hypothesizes that Taki Onqoy was the product of intraorder disagreements and rivalries between secular clergy (like Cristóbal de Albornoz) and the
Dominican Order (Mumford 1998). Varón suggests that Jesuit religious officials sought to
evangelize in the Huamanga region, which was already occupied by Dominican religious
officials. At this time, Dominican religious clergy were being criticized for their accomodationist
tactics—instead of asserting one particular “correct” way to practice Catholicism, Dominicans
allowed local Andeans to continue with their dancing, singing, and other traditional practices
(Mumford 1998: 162; Varón 1990: 402). Thus, Varón attests that the tension between lax
Dominican conversion practices and the stricter methods practiced by secular contemporaries
may have prompted the exaggeration of Taki Onqoy.
80
In 1992, Juan Carlos Estenssoro followed up on Varón’s conclusion and expanded on the
rivalry between religious orders in his “Los bailes de los indios y el proyecto colonial”
(Estenssoro 1992). Using Taki Onqoy’s central practice of dancing as evidence, Estenssoro
situates the Taki Onqoy movement in a larger contention between Dominicans and secular
priests. From the start of conquest, Spanish policy in the New World had questioned indigenous
traditional practices—singing, dancing, ritual, etc.—and more importantly, whether or not these
practices were allowable during the era of conversion (Durston 2007; Estenssoro 2003). For
example, Estenssoro cites González de Cuenca, a personal contact of Juan de Matienzo, who
wrote in the 1560s regarding politics and public administration: “Item. That the Indians do not do
dances nor get drunk and if some of the dancers wanted to spend the day in parties, that they do
so in public places where the Corregidor or priest of the valley or doctrina would warn them that
these dances would be punished by shearing” (Estenssoro 1992: 363).34 In this work from 1567,
we see that dancing was considered a problem even before it was discussed by Molina or
Albornoz. It is especially telling that González de Cuenca refers to dancing specifically as
“takis” and makes a point to differentiate between dancing in public and dancing in private.
In general, indigenous dancing—both in secret and in public—was heavily debated in the
sixteenth century. While some thought this native practice was not harmful and was in fact a
promising way to learn new traditions, others thought that dancing was inherently idolatrous.
Inspired by the teachings of Bartolomé de las Casas, which were sympathetic to indigenous
traditions, the Dominican order allowed Andeans to continue their local practices (Estenssoro
1992; Las Casas 1986; Mumford 1998). However, secular clergy who were practicing in the
Translated from “Yten. Que los yndios no hagan taquies ni borracheras y si algunos bailes quisieren hazer sea de
dia y en fiestas y en lugar public donde lo pueda ver el Corregidor del valle y el sacerdote de la dotrina con
apercibimiento que haziendo los tales bayles de otra manera sean tresquilados” (Estenssoro 1992: 363), translation
mine.
34
81
sixteenth century in Peru resented the Dominican order for allowing these behaviors, viewing
them as openly and unambiguously idolatrous. Estenssoro argues that rather than a symbol of
nationalist revival or Andean psychological torment, Taki Onqoy may have simply been a
strategic way to discredit Dominican clergy. As Estenssoro suggests when discussing Toledo’s
role in Taki Onqoy, “One of the institutions that Toledo openly attacked was the powerful
Dominican order, which was disconnected from episcopal management” (Estenssoro 1992: 366).
That is, Toledo had an open goal to discredit and attack the Dominican clergy. Thus, if Albornoz
did exaggerate Taki Onqoy, it was not in a vacuum of the dreams of one man, but rather may
have reflected broad, governmental attempts to disgrace Dominicans at the time. Taki Onqoy
was discovered by two secular clergy in Dominican-controlled regions in Huamanga. These
clergy could have exaggerated the threat of the movement in order to shame their Dominican
colleagues.
Finally, in the most categorical denial of the existence of the movement, Gabriela Ramos
argues that Taki Onqoy was never an actual movement. Instead, it was a false creation by
Cristóbal de Albornoz, which he exaggerated in order to advance his own career and receive a
promotion to bishop or archbishop from the Spanish Crown. In her “Politica eclesiástica y
extirpación de la idolatría: Discursos y silencios en torno al Taki Onqoy,” Ramos manipulates
existing historiography about both Molina and Albornoz. She suggests that Molina’s Relación
was actually written and published at a later date than the agreed-upon 1574, which has since
been affirmed by several historians (Bauer 2011; Mumford 1998; Urbano 2008). Specifically,
Ramos writes that Molina’s final chapter on Taki Onqoy was clearly an addition to the text,
rather than having been included in the first version of the account (Ramos 1992:162). In
addition to this claim, Ramos relies on the information from Albornoz’s 1569 Información—
82
since he neglects to mention Taki Onqoy some four or five years after it started, Ramos thus
finds fault in all of Albornoz’s claims (see Ramos 1992: 150 for further explanation). With such
doubt regarding both Molina and Albornoz, Ramos inevitably concludes that Taki Onqoy never
existed, but was completely created in the head of Albornoz in order to get his promotion. As
stated in the opening pages to her article, “In the first place, Taki Onqoy and the affirmation of
its existence was a matter which was progressively built, the pair were increased aspirations of
the cleric Cristóbal de Albornoz in an attempt to climb the hierarchy of the church” (Ramos
1992: 149). Thus, utilizing the same two sources, Gabriela Ramos instead puts forth a scathing
revision of Taki Onqoy, in which she discredits Molina, Albornoz, and several other secondary
scholars who have studied the movement. She concludes by suggesting that new sources must be
found and studied before any agreement can be reached on the motivations of Taki Onqoy, and
indeed, on its very existence.
In 2002, Jaymie Heilman published a response to Gabriela Ramos, entitled “A Movement
Misconstrued? A Response to Gabriela Ramos’s Interpretation of Taki Onqoy” in which she
refuted Ramos’s claims. While Heilman agreed that Albornoz had sole control over his
Informaciones, she instead justified the gaps and silences in Albornoz’s accounts as intentional
or at the very least, explicable. For example, while Ramos uses Albornoz’s failure to note Taki
Onqoy in his 1569 Información as evidence that he only created the movement when it became
beneficial to him, Heilman instead suggested that this was a reasonable omission—having just
arrived in Peru in 1567 and only traveling to Huamanga in 1569, it is plausible that Albornoz had
not yet established himself in the region (Heilman 2002:128). Furthermore, as an extirpator of
idolatry, Albornoz would have been responsible for punishing not only the Andeans who
83
practiced the movement, but also the Spaniards who turned a blind eye. As such, he was in a
predicament, caught between the Spanish clergy and the Spanish Crown.
Since Albornoz had only recently began his tenure as visitador, and additionally, since he
was already combating the somewhat negative perception of the secular clergy in Huamanga,
Albornoz was forced to tread lightly. He could not condemn all other clergy and Spanish
encomenderos in the region from the instance he arrived in Huamanga. As Heilman writes, “to
proceed hastily with mass punishments against priests, encomenderos, and average Spaniards
would risk angering and alienating a huge proportion of Huamanga’s Spanish population”
(Heilman 2002: 129). As such, Albornoz’s first letter to the Crown was restrained—Albornoz
thought he could earn a promotion based on his own merits at the time. When he failed to earn
these positions, Albornoz revised his second (1570) letter, this time lengthening the text and now
including his role in combating Taki Onqoy. Heilman goes on to convincingly explain several of
the problems Ramos raises regarding Albornoz’s texts, the silences among them, and the
contrasting strategies Albornoz may have used when constructing his letters.
Finally, Heilman looks to other sources in order to support her counter-argument that
Taki Onqoy was a real movement in the 1560s. For instance, in his correspondence to the crown
in 1570, Viceroy Francisco de Toledo mentions the links between failure to convert Andeans and
their possession and dancing (Heilman 2002: 134). Heilman also refers to Guaman Poma, who
assisted Cristóbal de Albornoz in his 1569-1571 visita, affirming that the native author mentions
Taki Onqoy and condemns the movement. Heilman makes a similar argument to the one I make
here—through the two major primary sources on Taki Onqoy, scholars have arrived at radical
interpretations regarding nearly every aspect of the movement. When the same words on a page
84
can be interpreted in so many different ways, it is both impossible to determine who is “correct”
and also crucial to investigate the movement in other ways.
Models for an Archaeological Study of Taki Onqoy
In the previous sections, I have given a history of the conquest of the Inka, described
some of the early problems for Spaniards during conversion, outlined the primary sources on
Taki Onqoy, and given an extensive account of the current state of the field in the secondary
literature. Until my 2015 field research, there had been no systematic investigation into Taki
Onqoy. In fact, general archaeological studies centered on revitalization in other regions (see
Liebmann 2008, Fry 1985) have been sparse and difficult to replicate. It is clear, however, that
the Taki Onqoy movement was predicated on the entanglements between Andean and Spanish
things, practices and places. In my own materials-based study of the movement, then, I have
avoided asking totalizing questions which mirror those themes discussed above—it is impossible
to address those themes in a cohesive manner, since Taki Onqoy was defined by its
contradictions and the variable nature of participation in or adherence to the movement. Instead,
my hypotheses are grounded in the study of things, materiality, animacy, and entanglements
defined at the beginning of this chapter. Given the particularities of the practices associated with
Taki Onqoy, I relied largely on Molina and Albornoz to generate archaeological hypotheses in
my own investigation. Building on primary accounts of Albornoz (1990 [1570, 1577, 1584]) and
Molina (2011 [1574]), previous archaeological research and theoretical debates on revitalization
movements, and archaeological research on colonialism in Peru, my fieldwork investigated the
following hypotheses (which are discussed more thoroughly in Chapter 5):
85
1. If Taki Onqoy practices did not entirely proscribe the use of Spanish things, they will be
characterized by the repurposing of Catholic or Spanish things in new ways or hybrid
forms of material culture.
2. If Spanish responses to Taki Onqoy at Iglesiachayoq were violent, they will be marked
by destruction of huacas in public arenas and/or physical punishment of takiongos in
public spaces.
3. If the inhabitants of Iglesiachayoq were challenging Catholic traditions in burial practices
in favor of those supported by the Taki Onqoy movement, then interments will reflect
Andean traditions rather than standardized Catholic norms in terms of position and burial
goods.
4. If Taki Onqoy performances were an explicit revitalization of prehispanic cultural
traditions, then their associated materials would reflect practices common in the region
prior to Spanish invasion.
5. If Taki Onqoy was a clandestine movement, its material and spatial signatures will be
uncovered in private areas away from central religious spaces. If there are public spaces
of Taki Onqoy, these will be located away from places of Spanish administration and
Catholic practices.
More broadly, my project investigated instances of incorporation or rejection of Spanish objects
in indigenous households, as well as spatial distinctions between the distributions of these
objects.
Finally, I will speak briefly as to the notion of rejection. Throughout the primary sources,
the secondary literature, and the previous discussion in this dissertation, I have referred to
“rejection” as a uniform action. However, “rejection” can take many forms and is likely related
86
to specific aspects or practices, rather than culture as a whole. For example, why did Andeans
reject Spanish Old World foodstuffs and vessels? It is plausible that since Andeans had fallen ill
after the arrival with the Spaniards, they considered ingesting Spanish items as a very dangerous
activity (for more research on body and consumption, see Conklin 2001). More generally, what
aspects of Spanish objects or religion denoted them as “Spanish” or “Catholic”? These sorts of
questions are critical for a more nuanced understanding of Taki Onqoy, one that builds upon the
traditional “Did Taki Onqoy exist or not”? In the following two chapters, I will provide the
regional background and methodology for my archaeological research. In the final chapters of
this dissertation, I will return to answer these questions, armed with archaeological data.
87
CHAPTER 3
AN ANDEAN ARCHAEOLOGY OF REVITALIZATION
Revitalization movements can be defined as “calculated, methodical efforts to reform
culture and society that frequently occur in colonial situations” (Liebmann 2012:14). These types
of movements have been documented across time and space, and are common in colonial or
imperial situations (Adas 1979; Linton 1943; Worsley 1957). However, the ubiquity of
revitalization movements in situations characterized by extreme power differentials should not
reduce the unique aspects of Taki Onqoy into the diluted explanatory category of “revitalization”
as a means of “resistance.” From the Spanish perspective, participation in the 1560s Taki Onqoy
movement was an organized threat to the Spanish Empire in the New World. Andean individuals
and communities were intentionally disregarding religious and Spanish mandates in favor of a
return to local prehispanic huaca worship.
This chapter proposes a theoretical framework for understanding the Taki Onqoy
movement as heterogeneous and multivocal, and historically situates the defining practices of the
movement. First, I discuss Taki Onqoy as a revitalization movement, demonstrating how nativist
beliefs and Catholic practices intertwined to help shape the movement’s structure and goals. Taki
Onqoy preachers explicitly described their beliefs as a “resurrection of the huacas,” yet these
preachers had also existed under Spanish rule and evangelization throughout their lives
(Albornoz 1990; Molina 2010). Some of the practices deemed as “prehispanic” by these
preachers, then, were actually reminiscent of or structured by religious practices performed by
Spanish clergy. Next, I consider performance theory in archaeology to posit how covert
performance disseminated the movement’s ideological tenets. Then, utilizing conceptions of
modern forms of “weapons of the weak” (Scott 1985), I suggest the means by which Taki Onqoy
88
would have played out on-the-ground, considering the different social groups and cultural
interests which clashed with one another in the 1560s. Finally, I will briefly address the
secondary debates regarding Taki Onqoy: specifically, whether Taki Onqoy was its own
phenomenon, or if it was one example in a series of religious uprisings in the Andes which
should not be afforded its own significance.
A strong theoretical framework is necessary in implementing archaeological investigation
of Taki Onqoy. By using a practice theory approach to ground study of the material aspects of
Taki Onqoy, I avoid merely engaging in “text-aided archaeology,” and contribute new questions
and new methods and materials for the study of this movement. In doing so, I partake in a
growing dialogue amongst archaeologists who study colonial encounters—archaeology can
“furnish kinds of evidence that are qualitatively different from and independent of the colonial
texts that constitute the vast bulk of evidence available to historians…archaeology offers access
to the material dimension of the encounter and to the processes of daily life through which the
colonial situation was experienced and worked out by ordinary people” (Dietler 2005: 50,
emphasis original). Though archaeological data sources have their own respective indices of bias
(such as preservation and the types of questions which can be asked), the study of materials
promotes understanding of those themes, peoples, and practiced rarely discussed in colonial
texts.
In this chapter, I argue that the presence of Spaniards intent on converting local
individuals to Catholicism was mitigated by local spaces and practices in which those who
“resisted” the new religion could subvert Spanish teachings in private spheres. However, it
would be a broad generalization to suggest that all of the local and Inka residents of
Iglesiachayoq participated in Taki Onqoy—it is likely that many individuals went about their
89
daily lives negotiating and interacting with both religious traditions. While some Andeans likely
identified as full Catholic converts, others actively rejected Catholic teachings and Spanish
cultural materials. Still a third group likely existed in between these two poles of religious
orientations—these individuals may have appreciated both Catholic and Taki Onqoy belief
systems, and called upon the practices of either or both depending on their needs. Through a
discussion of revitalization and everyday forms of resistance, I take an intersectional approach to
demonstrate how Taki Onqoy was understood, practiced, resisted, or forced on different aspects
of the population.
The Utility of the Resistance Concept
Anthropological and archaeological literature on the topic of European colonialism has
recently broadened from an initial tendency to understand colonial processes as artefacts of
European hegemony (Given 2004; Gosden and Knowles 2001; Stein 2005), to a variety of more
nuanced perspectives which center native experiences or actions and the importance of local
sociopolitical schema (Brown 1996; Dietler 2005; Liebmann and Murphy 2011; Ortner 1995;
Wernke 2013). Initially, literature studying colonial encounters viewed domination as
essentially an all-encompassing form of institutionalized power, while resistance was thought of
as any sort of opposition to this power (Du Bois 1945; Fanon 1952; Memmi 1965; Said 1978).
Much of this dichotomy was derived from a biased body of documents written by the “victors,”
such that colonial powers were seen as unflinching and all-encompassing, and that any
undesirable or rebellious behavior challenging this power was categorized as “resistance.” While
the terms “resistance” and “domination” still have utility as blanket terms, scholars have
expanded and clarified these concepts in recent decades.
90
The terms “domination” and “resistance” are each problematic as analytical tools in
studying colonialism. One issue with the notion of colonial “domination” is its tendency to imply
a one-way transfer of colonial structures onto colonized societies, thereby reaffirming the
structures of colonial documentation. “Domination” as a concept is all-encompassing because it
signifies an implicit power structure, one in which the colonizing group is already positioned as
the one with agency and control. The very notion of “domination” thus frames interpretations of
colonialism as responses to domination, and is insufficient in allowing for interpretation of new
cultural frameworks generated by colonialism. In the late twentieth century, ethnohistory became
a field in its own right and historians began to explicitly focus on native groups during colonial
encounters (Spicer 1961; Wachtel 1977). However, the domination framework still swayed
interpretations such that “they often perpetuated a view of indigenous peoples as passive subjects
caught in a process of acculturation that they had no ability to shape” (Liebmann and Murphy
2011: 6). In the 1980s, historians began to challenge perceived one-way processes of domination
by explicitly searching for forms of native resistance in the historical record (Spalding 1984;
Stern 1993 [1982]).
The term “resistance” is a useful retort to models of colonial domination in that it affords
native communities agency, and recognizes that the historical record teems with examples of
native populations pushing back against their European colonizers (Deagan 2004; Silliman 2001;
Van Buren 2010; Wernke 2011). It also contradicts the stereotypical “myths” created by colonial
documentation which assert that the Europeans were “powerful exceptional men” and “superior”
to indigenous peoples (Restall 2008). The term itself can be applied to any power differential
throughout time and space—initially, “resistance” was used in both history and archaeology to
categorize highly visible, often armed militaristic responses to colonialism (Liebmann and
91
Murphy 2011; Wernke 2011). More recently, concurrent with the spread of household
archaeologies and theoretical perspectives from feminist theory, “resistance” has applied to the
minute acts of daily subversion, most famously categorized in Scott’s Weapons of the Weak
(Scott 1985, 1990).
While the resistance concept has been useful in recognizing the array of responses to
colonialism, the ubiquitous application of the term in anthropology and archaeology has been
criticized as being overused (Brown 1996; Ortner 1995). Sherry Ortner’s 1995 article
“Resistance and the Problem of Ethnographic Refusal” argues that conceptions of domination
and resistance are “thin” and uniform because the ethnography required to understand these
power struggles is not thorough. In order to avoid reaffirming the binary division of
“domination” and “resistance,” it is critical to not only consider the broad politics between these
two sides, but also the internal strife between classes, ethnicities, and genders within each side.
As Ortner writes:
If we are to recognize that resistors are doing more than simply opposing
domination, more than simply producing a virtually mechanical re-action, then we
must go the whole way. They have their own politics—not just between chiefs
and commoners or landlords and peasants, but within all the local categories of
friction and tension: men and women, parents and children, seniors and juniors;
inheritance conflicts among brothers; struggles of succession and wars of
conquest between chiefs; struggles for primacy between religious sects, and on
and on. (Ortner 1995:177).
It is only by studying both the internal divisions and politics of both superordinate and
subordinate groups that we can fully understand how and why forms of control are levied, and
how and why subject groups push back against these forms of control. Ortner goes on to argue
that both individual and unified resistance practices are not universally supported, and that the
92
internal divisions within subaltern groups often cause resistance acts which are “conflicted,
internally contradictory, and affectively ambivalent” (Ortner 1995:179).
Building on Ortner’s essay, Brown’s 1996 article “On Resisting Resistance” suggests that
cultural anthropologists have over-utilized the concept of resistance such that it has lost its
analytical value. Brown even argues that cultural anthropology has narrowed its focus so that the
common framework of study is the dominance versus resistance model, instead of
acknowledging the wide range of social settings in which humans actually exist (Brown
1996:734). Although still a growing field, archaeological studies of resistance have followed
trends in anthropology, such that archaeologists often invoke “resistance” to categorize native
responses to conquest (Liebmann and Murphy 2011:10). The problem with explaining material
patterns as evidence of resistance is that these sorts of explanations flatten behaviors and use
resistance as a conclusion rather than a starting point. With anthropological and archaeological
criticisms of the resistance concept in mind, I strive to consider how Taki Onqoy was not merely
a response to Spanish conquest, but also to demonstrate how it was nativistic, generative,
creative, and how it outwardly promoted alliance amongst politically and ethnically diverse
Andean groups. I argue that Taki Onqoy was one of many resistive responses to Spanish
colonialism, a religious revitalization movement.
Revitalization Movements
Revitalization movements are intentional efforts to create rapid cultural, political, or
social change, often occurring in colonial situations. In 1956, Anthony F. C. Wallace published
his landmark paper “Revitalization Movements” which was an overt attempt to create a
framework for explicit “innovations of whole cultural systems” (Wallace 1956:264). Wallace
93
documents various millenarian movements that were undertaken by subjugated peoples,
suggesting that “the persons involved in the process of revitalization must perceive their culture,
or some major areas of it, as a system; they must feel that this cultural system is unsatisfactory;
and they must innovate not merely discrete items, but a new cultural system, specifying new
relationships as well as new traits” (Wallace 1956: 265). In contrast with steady, often
subconscious culture change over time, revitalization movements are deliberate attempts at
change by subordinate groups. Wallace writes that revitalization movements are very common,
and are often the impetus for establishing new religious communities.35 Revitalization
movements have been documented across time and space, and are often compared to similar
types of movements known as cargo cults, millenarian movements, utopian movements,
messianic sects, or apocalypticism (Adas 1979; Linton 1943; Worsley 1957).
Wallace’s article is important in that it popularized the term “revitalization” and provides
a framework for understanding the basic characteristics or practices of these types of movements.
He argues that while there are different names for these movements—messianic movements,
revolutions, reform movements, cargo cults, etc.—in reality, all are examples of revitalization
movements that are only characterized differently due to disciplinary variation. Rather than a
natural change of a culture over time, Wallace insists that revitalization movements are
deliberate attempts at change by subaltern groups (Spivak 1996; Wallace 1956). Wallace
suggests that all of these types of movements follow the same processual structure, and then
creates a five-stage teleological path which he argues is present in all revitalization movements if
they are successful. The first stage is the “steady state,” which represents the population existing
in prolonged, but manageable levels of stress. While the cultural situation is not ideal in this
35
Wallace discusses Christianity, Mohamedanism, and Buddhism as examples of revitalization movements (1956:
267).
94
stage, it is tolerable for the majority of the population. The second stage is called “the period of
increased individual stress,” and is characterized by a number of years or experiences which
cause severe stress in the culture. Here, Wallace lists examples such as epidemics or political
subordination as elements which cause stressors to the cultural balance. The third stage is “the
period of cultural distortion” in which the prolonged stressors begin to create anxiety within
individuals in the population. Stage four is the “period of revitalization”: here, the society is
faced with either dire events such as population death or the end of traditional ways of life unless
the society undergoes the revitalization movement. The last stage is the “new steady state” in
which cultural transformation has occurred and a new set of schema is established.
Wallace’s five stages are very basic, and are plausibly applicable to any process of
cultural change. However, he delves more deeply into stage four, the period of revitalization, and
suggests that revitalization movements which are religious in nature must undergo six further
tasks (Wallace 1956: 270). The first task is the “mazeway reformulation,” or, the idea of a more
idealized society. For Wallace, the conception of this idealized society often originates as a
prophecy in one charismatic individual—this is certainly the case for the Pueblo Revolt, whose
prophet Po’Pay preached his visions to the extended population (Liebmann 2008). In the case of
Taki Onqoy, there was not one lone prophet, but rather multiple individuals who roamed the
countryside to spread their message (Molina 2010 [1574]). The second task is “communication,”
in which the prophet spreads his revelations to the rest of the population. The third task is
“organization,” which represents the goals of the leader to convert his/her followers. The fourth
task is “adaptation,” where Wallace suggests that the movement inevitably receives resistance
from either a powerful group within the society, or an external dominant group. In the case of
Taki Onqoy, this phase represents the extirpation campaigns of Cristobal de Albornoz,
95
undertaken from the 1560s to the 1580s, as well as the Toledan reducción resettlement, instituted
in the 1570s. The fifth task is “cultural transformation” in which the population or society
accepts the new status quo. Finally, the sixth task is “routinization” in which the movement
“maintains responsibility only for the preservation of doctrine and the performance of the
ritual—it becomes a church” (Wallace 1956: 275).
While the phases and tasks Wallace introduces are useful for comparative purposes, they
are also restrictive. Wallace acknowledges that “the outline of stages is properly applicable to a
revitalization movement which is completely successful” (1956: 278). However, Wallace is
unclear as to what constitutes full success or complete failure. I would argue that casting
movements as either “successful” or “failing,” replicates the same problems with dichotomies
like “dominance” and “acculturation.” That is, I do not see a purpose in categorizing movements
as successful or failing, but rather suggest that movements and events should be analyzed on
their own merits according to the practices which constitute them. Even if the Taki Onqoy
movement was ultimately “unsuccessful” in that it did not entirely reverse Catholic religious
indoctrination and overthrow the Spaniards, the threat of its success prompted Spanish
authorities to make major governmental changes in the 1570s. Furthermore, there are lasting
effects in regards to religious beliefs such that many people in the highlands today still make
sacrifices to the apus while simultaneously attending Catholic churches. In sum, while I do view
Wallace’s revitalization model as useful in interpreting Taki Onqoy, I find problems with his
rigid stages and tendency to cast revitalization movements in a dichotomized binary.
Historically, revitalization theory was a product of its time—it relied on a linear model of
social progress which “attempted to account for the entire range of human behavior with
reference to one or two driving principles” (Harkin 2004: xviii). Additionally, the model
96
attempted to explain human behavior rather than interpret this behavior—the stages were strict
categories within which humans moved in a consistently forward and linear matter, and were
unable to account for struggles or reflexivity within these stages. Furthermore, Wallace’s model
was based on the idea of “equilibrium” within a society. His idealized state for a society is a
“steady state” which is true for the “vast majority” of the population—however, Wallace fails to
define his version of a society or what constitutes the “vast majority” (Siikala 2004). Jukka
Siikala argues that Wallace’s system-based model “presupposes harmony among its elements”
and writes that “the collective coherence of individual incoherences is attained through culture”
(2004: 89). That is, although the actions of individuals are discordant, their desired goal is
identical.
Beyond problems with the model itself, there are also an array of difficulties facing
anthropological interpretation of past revitalization movements. First, like broader problems with
studying colonial encounters, many of the documents which describe religious movements were
written by colonizing populations. Christian European authors often were the people recording
the movements, and their cultural and religious beliefs surely shaped the interpretations of the
people participating in these movements (Brown 2004, Harkin 2004). Second, as Brown argues,
“in searching for effective ways to formulate and conceptualize prophetic movements in terms
considered effective and intelligible within our academic disciplines, we risk exoticizing them,
distancing ourselves from the people involved, and neglecting the historical and cultural
perspectives they could offer” (Brown 2004: 104). That is, as researchers or theorists, we risk
perpetuating the same process of “othering” created by the original authors. On the reverse side
of this point, attempts to create better societies are often idealized by scholars, who cannot resist
97
heralding native resilience or utopian societies, even when the attempts to create these societies
may be detrimental to some aspects of the population.
Despite the problems discussed above, the concept of revitalization is still a useful
theoretical tool, albeit with some necessary manipulation. To complete my model, I build upon
Michael E. Harkin’s 2004 chapter entitled “Revitalization as Catharsis.” In this chapter, Harkin
suggests that while Wallace’s model is problematic in its teleological, linear model which ends
in a new cultural norm or failure, the more plausible interpretation is that “many such
movements were not in the end about ‘revitalization,’ that they were not centered on the
restoration of vitality to a cultural pattern” (Harkin 2004: 144). Rather, revitalization movements
can serve a different purpose—to “offer emotional release from the dreary reality of death,
disease, starvation, and acculturation” (Harkin 2004: 157). Revitalization movements, then, and
the practices which are manifested during these movements, replace individual suffering with
collective rites and actions, which can serve to construct new senses of community rooted in past
notions of tradition.
Revitalization Theory and Taki Onqoy
Although there are several problems with revitalization theory as put forth in 1956 by
Wallace, the concept itself is extraordinarily useful in devising a framework for analyzing these
types of movements in a global setting. Primarily, revitalization theory is accurate in explaining
the format of cultural transformation—however, I argue that Wallace’s conception of “stages” is
not useful, and is restrictive in its false boundaries. I will now create a model of Taki Onqoy
borrowing both from Wallace’s original model and from some of the critiques and
reformulations (Harkin et. al 2004). Wallace’s model begins with a “steady state” in a given
98
society—I see this state as restrictive and not applicable to Taki Onqoy. Taki Onqoy did not arise
in a time of relative psychological contentment and satisfaction in the Andes. Those who
practiced Taki Onqoy had likely known only a life of controversy, having existed under
protracted Spanish rule, or having experienced Inka imperialism followed by Spanish takeover.
When Taki Onqoy was visualized and practiced in the 1560s, its practitioners were fully
disillusioned from promises of Spanish benevolence, and had endured plagues and violenceinduced deaths of thousands of Andeans in the first half of the sixteenth century. For those living
in the 1560s, this disillusionment had been occurring for some time, such that many of those who
believed in Taki Onqoy likely had never experienced a “steady state” in their lifetimes. Indeed,
this lack of steady state matches with some criticisms put forth by Harkin et al. 2004 wherein
Wallace’s basic conception of the first stage is flawed. It is not necessary to first have a state of
relative comfort prior to a stage of disillusionment—the stories and oral histories of this state of
wealth and abundance are sufficient enough to prompt dissatisfaction. For example, Taki Onqoy
practitioners likely did not live during the time of Inka rule—or if they had, they were children
during this time. However, oral histories put forth of prehispanic religious traditions and selfgovernance were enough for those takiongos to create a movement which preached a return to
this time.
Colonial encounters themselves are times of great discordance, with two distinctive
cultures interacting and those living through these times forced to find ways in which to both
shape and survive the emerging governance. I argue that Taki Onqoy begins in Wallace’s
conception of Stage Two—a prolonged period of increased individual and cultural stress,
exacerbated by the extirpation of many indigenous lifeways and the devastation of deadly
epidemics which wiped out large swaths of the Andean population. Stage two and stage three are
99
very similar in Wallace’s model, with Wallace’s third stage being the “period of cultural
distortion.” I do not see a need to differentiate between stages two and three according to
Wallace—for the sake of Taki Onqoy, they can be subsumed into one another such that the
movement arises in a time of prolonged cultural stress on a communal level. There was not
necessarily a specific “trigger” or “loss” for takiongos—rather, the movement emerged as a
mechanism for countering restrictive and extractive Spanish governance. My model for Taki
Onqoy begins here—indigenous people were living through a period of ever-increasing stress
and demands, such that they needed to develop a way to “fight” or at least endure these stressors
in a non-violent way. Andean peoples did not have the technologies, strength, or numbers in the
second half of the sixteenth century to put forth a feasible combative rebellion or resistance.
Instead, they invented an ideological defense mechanism to confront the colonial environment.
Although the initial stages of Taki Onqoy do not align with Wallace’s model, the stage
four “period of revitalization” provides some explanatory guidance. For Wallace, this period is
denoted by several sub-stages related to spreading the vision of a prophetic individual. For Taki
Onqoy, Albornoz (1990 [1584]) and Molina (2010 [1574]) both suggest that the ideals of the
movement are spread through several wandering preachers. The leaders and most emphatic
preachers are identified by both Albornoz and Molina—in his 1584 Información document,
Albornoz records the testimony of Cristobal de Molina, whose account affirms that the “most
guilty were two natural Indians and one indian named don Juan Chocne or Joan Chocna”
(Albornoz 1990 [1584]: 225). In his 1570 document, Albornoz records a witness saying that
some of the leading women may have been called “Santa Maria and Santa Maria Magdalena”
(Albornoz 1990 [1570]). Through these two documents, we are given the names of the leaders
100
(and likely prophets) of Taki Onqoy. Much like the leader Po’Pay of the Pueblo Indian Revolt,
these three individuals were responsible for much of the spread of the movement.
With Wallace’s notions of “mazeway reformulation” accounted for in the promises of a
return of the huacas put forth by Taki Onqoy leaders, these traveling preachers then
“communicated” their message through conversion and persuasion throughout the highlands.
Attested to by both Molina and Albornoz, these preachers traveled through the highlands and
convinced indigenous peoples to participate in the movement through dramatic performative
rituals. These rituals involved purification, intoxication, and huaca possession, such that
prehispanic Andean huacas were believed to have inhabited individuals, who danced and sang to
rid themselves of Spanish-induced illnesses. Taki Onqoy participants spread the ideals of the
movement (aligning with Wallace’s notion of “organization”) so that the wandering preachers
could convert a few individuals in each town, who would in turn spread their propaganda to the
rest of the local inhabitants.
Wallace’s principle of “adaptation” is somewhat of a misnomer—in this stage, the
movement faces resistance by a dominant population. For Taki Onqoy, the resistance faced was a
two-pronged attack led by the Viceroy Francisco de Toledo, and the extirpator of idolatry
Cristóbal de Albornoz, who visited numerous towns in the highlands where he destroyed huacas
and punished Taki Onqoy adepts. Those Andean peoples who were punished faced humiliation
and pain through public whipping and shaving of heads, while all the while being forced to
attend church functions and help construct the churches in certain towns. The destruction of
huacas may have not seemed as catastrophic as one might think for Andeans—in the sixteenth
century conceptions of huacas, the prehispanic gods were mobile and could inhabit various living
organisms. As such, the destruction of portable huacas may have signified the huacas inhabiting
101
people, other rocks, or other natural features (Albornoz 1990 [1584]; Gose 2008; Molina 2011
[1574]; Stern 1993).
Viceroy Francisco de Toledo’s reducción resettlement campaign was a much more active
and aggressive attempt to stamp out Andean idolatrous practices. This resettlement involved the
deliberate and mass relocation of about 1.4 million Andeans to some 840 nucleated, planned
towns across the Andes (Mumford 2012). By removing Andeans from their original settlements,
Toledo severed ties between Andeans and their local huacas, and also created more “visible”
settlements for the Spanish. That is, Spanish authorities were better able to control and convert
Andean peoples through these planned towns. The Viceroy Francisco de Toledo clearly
designates the purpose of reducciones as “to extirpate idolatries, sorcerers, dogmatizers so that
the Evangelical teaching would fall well-disposed upon ground where it could bear fruit”
(Toledo 1986:36). Toledo’s clearly-stated goals of reforming idolatrous practices indicate the
worries Spanish religious authorities had regarding the Taki Onqoy movement. Furthermore, the
establishment of reducciones followed a strict format wherein the church and central plaza were
the literal center of these towns. The gridded streets expanding from the center allowed Spanish
officials to ensure church attendance in a timely manner.
Taki Onqoy participants never fully went through all of the stages conceived of by
Wallace. His conceptions of “cultural transformation” and “routinization” are neither formally
nor fully fulfilled. With the implementation of Toledo’s reducción resettlements in the second
half of the sixteenth century, Andeans had little formal recourse in their religious practices.
However, Taki Onqoy should not be considered as a “failed” revitalization movement. There are
several lines of evidence which attest to the lasting importance of Taki Onqoy as a religious
revitalization movement. First, Taki Onqoy served as the initial example in a long line of
102
revitalization movements or resistances put forth in the Andes which all had similar frameworks.
Although Taki Onqoy struck fear into the hearts of concerned Spanish religious authorities, its
beliefs were not revolutionary or rare. Jamie Heilman’s 2002 response to Gabriela Ramos’s
denial of Taki Onqoy’s existence cites similar instances of “religious recidivism or indigenous
peoples” throughout New Spain, including the Yucatan, Chancay, and the parish of Atenango
(for primary discussions of these movements see Clendinnen 1987, Silverblatt et al. 1993, and
Taylor 1996).36 With respect to Colonial Peru, Heilman cites several further examples of
religious idolatry that paralleled Taki Onqoy which took place nearly a century after Taki
Onqoy’s rise. She writes that “if Albornoz in fact invented many of the details of the Taki Onqoy
movement, he proved incredibly prescient in his imaginings of what an Andean religious
rebellion would look like” (Heilman 2002: 125; see also Gose 2008). Thus, if we are strictly
adhering to Wallace’s account of revitalization movements, the Taki Onqoy movement was a
complete failure because it did not create an upheaval of the Spanish religion and a new church.
However, by examining the ramifications of Taki Onqoy throughout Peru and greater New
Spain, I suggest that Taki Onqoy did create a model movement from which many local cults
derived their motivations. If Taki Onqoy held little impact on future Andean movements, then
the similarities in the shapes of these movements is very coincidental.
Second, local religion in the Andes continued to reflect ideals of the Taki Onqoy
movement. While many scholars suggest that the movement permanently ended in the 1570s,
when Toledo’s resettlement program was instituted, the movement (or parallel practices) likely
continued to exist well into the seventeenth century (Gose 2008). As Urton attests, “into the early
Heilman writes “From Diego de Landa in the Yucatan to Juan Sarmiento de Vivero in Chancay to Hernando Ruiz
de Alarcon in the New Spain parish of Atenango, colonial clerics found themselves in a continuing struggle against
the religious recidivism of indigenous peoples” (Heilman 2002: 125).
36
103
seventeenth century, as the clergy moved into remoter regions of the Andes, it became
increasingly clear to them that the native peoples in these scattered mountainous settlements
were continuing the worship of mountains, ‘pagan’ creator deities, and ancestral mummies, as
well as the sun, moon, and stars” (1999: 69). That is, in the early 1600s, there was documented
evidence of indigenous idolatry in the Andes, even as the reducción resettlement program was
taking place. In Gose’s 2008 analysis of Taki Onqoy, he argues that the legacy of the movement
continued, but no longer as a “bounded and internally consistent entity,” but instead as “an array
of initiatives, some of which continued to orient a more diffuse and emergent colonial
indigenous culture” (108-109). While the narrow preachings of the Taki Onqoy prophets no
longer flourished under the name of Taki Onqoy, the practices and prehispanic religious
movement continued across the Andes. While the conception of Taki Onqoy as an activist
movement waned in the 1570s, its ideological providence has held politico-religious
ramifications until today.
The prolonged paranoia of Spanish religious officials prompted by Taki Onqoy and
further idolatrous practices led to a wave of more directed extirpation campaigns, often centered
on the priest Francisco de Ávila (Charles 2010). In 1609, Ávila traveled to Lima bringing Checa
religious objects and mummified ancestors and presented the display to the viceroy. Soon after,
Ávila, accompanied by hundreds of the city’s Spanish and Indian residents, “denounced from the
pulpit the crimes of idolatry and superstition that still endured eighty years after the Europeans’
initial invasion” (Charles 2010: 132). Ávila’s descriptions of the motivations behind this idolatry
are identical to those of Albornoz and Molina written nearly 40 years earlier—Ávila viewed the
indigenous people as practicing “revitalized pre-Columbian beliefs and practices in defiance of
the Church…attributed the survival of autochthonous beliefs and models of identity to native
104
resilience in the face of Spanish colonial authority” (Charles 2010: 133). Like Albornoz’s and
Molina’s descriptions of Taki Onqoy in the 1570s, Ávila’s positions on indigenous idolatry in
the 1600s viewed this idolatry as intentionally antagonistic. Rather than simply a failure to
“fully” convert Andean peoples on the part of the Spanish religious authorities, Ávila saw
ongoing idolatry as a deliberate defiance to Spanish authority.
It was the discovery of idolatrous practices related to burial and treatment of the dead
which prompted Ávila’s extirpation campaign in the early 1600s (Gose 2008; Mills 1997;
Salomon et al. 1991). The following harsh wave of anti-idolatry campaigns from 1609 to 1622
was designed to eradicate any remnants of Taki Onqoy (Duviols 1971). Pablo Jose de Arriaga’s
La extirpación de la idolatría en el Píru (1999 [1621]) is a pinnacle treatise on extirpation,
written by a Jesuit priest as a manual in the findings and actions of the campaigns at this time.
Arriaga’s 1621 work would provide information and suggestions for later campaigns. The
second wave of extirpation campaigns was undertaken from 1649 through 1671, and consisted of
a search for the remnant superstitions and beliefs of Andean peoples (Cobo 1956 [1653]). Many
of these campaigns focused on specific “sorcerers” and “illicit arts” rather than blatant idolatry.
Mills (1997) affirms that the second round of anti-idolatry campaigns found mostly syncretic
instances of a distinctly “Andeanized” Christianity rather than a direct threat to Spanish notions
of religion. Further studies of idolatry extirpation affirm that the 1660s were a transformation for
Andean spirituality, such that most documented instances of deliberate idolatry peter out by the
end of the seventeenth century (Burga 2005 [1988]; Cock Carrasco 1980; Duviols 1967, 1971,
1986; Kubler 1946; MacCormack 1991). Thus, while there was no cultural “revolution” as
conceived of by Wallace, Taki Onqoy did promote native religious practices which lasted for a
century after the movement was supposedly quashed. While most of our information about these
105
practices is written by Spanish religious visitadores who were working to extirpate these
idolatries, their affirmations that so-called pagan practices were still embraced throughout
highland towns indicates that the practices associated with Taki Onqoy were still performed in
community settings.
A final visible practice which ties to Taki Onqoy practices is that of the “danza de las
Tijeras,” or “scissors dancers” in the highlands of Peru. In the past fifty years, many scholars
have connected the modern dance directly to Taki Onqoy dancing. As Bush (2012) writes, “once
repudiated by colonial and national elites precisely because it embodied the irrationality of
indigenous ritual practices, this cultural form is now a celebrated emblem of Peru’s cultural
diversity and the perseverance of Andean spiritual values in the modern world” (Bush 2012: 2).
In its modern form, the scissors dance is performed in the regions of Ayacucho, Huancavelica,
and Apurímac, some of the poorest departments in Peru, and is “a masculine Andean ritual dance
that originated in Pre-Columbian rituals of the Chanka region in the south-central Andes of Peru”
(Bush 2012: 9, for references see Barrionuevo 1988; Millones and Tomoeda 1998; Vivanco
1976). The scissors dance is seen as an authentic representation of indigenous identity in which
dancers hold two metal blades in their right hands and make a metallic sound while dancing. The
festivals in which the scissors dance is performed last from five days to a week, during which the
dancers perform without rest (Bush 2012). This prolonged exhaustive performance emulates the
dances which would have been performed during Taki Onqoy, where dancers would fast and go
without sleep for days on end in order to achieve a highly-praised trance state. Furthermore,
scissors dancers compete in physical acts of valor and make offerings to the apus during their
dances, again reminiscent of Taki Onqoy practices (Bush 2012). Bush’s 2012 thesis on scissors
dance relates the links between the dance and the Taki Onqoy movement, suggesting that the
106
strongest link identified between the two involves the trance-like states (Barrionuevo 1988; Bush
2012, Castro Klaren 1990, Vivanco 1976). A second parallel relates to the ritual cleaning of
houses and spaces performed by Andean peoples who viewed takiongos as “agents of
purification” (Molina 2011). For the scissors dancers, domestic locations where they are housed
during festivals are ritually cleaned after the festivals are over, suggesting a potential reversal of
this belief from the sixteenth century (Bush 2012: 55-56). In the area in which I work, scissors
dance is performed on a regular basis, and Taki Onqoy is commonly discussed as a proud
religious rebellion of which the ancestors of local people were a part.
While under Wallace’s model, Taki Onqoy would be considered as a “failed” movement,
I instead suggest that the movement held critical ramifications for Andean religious practices—
while the significance of these practices is debated, Taki Onqoy cannot only be considered
within the strict 10-15 year period in which it is typically bracketed. Arising out of a time of
confusion and consternation, the movement was put forth and spread by local preachers and
prophets, resulting in the deliberate Toledan reforms which were an attempt to stamp out this
perceived idolatry. The performances and idolatrous practices attributed to Taki Onqoy inspired
and shaped further idolatrous practices in the Andes and the greater New World. There is
documented evidence of prophetic individuals preaching a return to huaca worship throughout
the seventeenth century (Gose 2008; Heilman 2010; Mills 1997).
In the study of revitalization movements, then, I propose a few changes which broaden
the scope of these movements and refrain from bracketing certain stages in a teleological
fashion. Using Taki Onqoy as an example, revitalization movements can be thought of as
deliberate attempts to change challenging socio-political environments. They are spread by a
prophet or prophetic individuals, and they encourage the participation of local peoples in the
107
quest for a more healthy and fruitful existence. Inevitably, these movements are challenged by
the dominant or dominating power who view them as a threat to established norms. Tensions
eventually come to a head when the colonizers view these movements as untenable to their
governance or when the movements gain enough support to overthrow the existing establishment
and begin a new way of life (as in the case with the Pueblo Revolt, Liebmann 2012). However,
there is not a need to classify revitalization movements as “failures” or “successes” based strictly
on the establishment of a new church. Rather, these movements should be studied as to how they
were practiced through a practice theory approach.
Archaeology of Revitalization
Revitalization movements have been documented throughout time and space and are a
remarkably common occurrence,37 but there is a conspicuous absence of documented
revitalization movements in archaeological contexts. As Liebmann affirms, “the lack of
identified instances of revitalization in the archaeological record is fueled in part by the
assumption that the core characteristics of these phenomena—charisma, revelation, and
prophecy—are immaterial and therefore archaeologically invisible” (2012:16). However, cultural
transformation manifests itself through the practices of all groups, and material media were
central to the practices of Taki Onqoy. Thus, by studying the materiality of Taki Onqoy
practices, we can ask questions unanswerable in texts. While there are many ethnographic
studies of revitalization movements, (specifically in regard to Ghost Dance38 or cargo cults in the
South Pacific39), there have only been three examples of archaeological investigations of
37
For example, Christianity, Islam, Buddhism, Church of Jesus Christ of Latter-Day Saints all initially started as
revitalization movements (Wallace 1956).
38
For work on Ghost Dance see (Andersson 2008; Du Bois 2007; Kehoe 1989).
39
For work on cargo cults in the South Pacific, see (Christiansen 1969; Worsley 1957).
108
revitalization movements: Fry’s 1985 work with the Lowland Maya in the Postclassic and
archaeology of the southwest, including the seventeenth-century Pueblo Revolt (Liebmann 2007,
2008, and 2012) and the thirteenth-century depopulation of the Central Mesa Verde Region
(Glowacki 2011).
In his 1985 article about revitalization in the Postclassic Lowland Maya, Robert E. Fry
outlines potential archaeological correlates for revitalization movements, and posits how rapid
cultural change might be identified in the archaeological record. For example, Fry suggests that
followers of the new revitalization movement could show distinctive patterns of dress or
practices in public and private contexts. These patterns of dress and behavior are often “reversals
of ‘acceptable dress style…and may involve syncretism of traditional styles with those exhibited
by members of the superordinate culture” (Fry 1985:127-128). Furthermore, revitalization
movements may manifest themselves in the intentional destruction or erasure of aspects of the
dominant cultural framework, as is seen with the destruction of crosses and bells in the Pueblo
Revolt and deliberate rejection of Spanish culture during the Taki Onqoy movement. Fry
hypothesizes that revitalization movements have not been largely explored in archaeological
contexts due to the rapid nature of change inherent in these types of movements—revitalization
movements often have a timescale corresponding with years rather than decades, and so these
brief flashes of time are often invisible in the archaeological record.
Revitalization movements also take many forms and can be primarily ideological or
political—there is no universal set of archaeological correlates of “revitalization.” Fry posits that
an archaeology of revitalization would demonstrate the razing or dramatic alteration of public
buildings if the movement is successful (Fry 1985: 129). Taki Onqoy’s ultimate inability to
produce drastic reversal of Spanish lifeways and religion suggests that destruction of existing
109
visible norms on a large scale would not be present in the sixteenth-century Peruvian highlands.
However, while Taki Onqoy did not produce a dramatic overhaul of Spanish religion, it was a
contributing factor to Toledo’s mass resettlement project in the 1570s (Gose 2008; Toledo et al.
1986). In this sense, it is possible to argue that Taki Onqoy did, in fact, play a part in creating a
mass change in societal norms—just not by those who practiced the movement. Furthermore,
although Taki Onqoy may not have produced the dramatic material destruction seen in Fry’s or
Liebmann’s archaeological projects, aspects of the movement are still seen today in a variety of
entangled religious practices and native dances. While Fry uses revitalization to explain
archaeological evidence of rapid change after excavation, Taki Onqoy was instead a known
revitalization movement whose very existence is debated, and so for this dissertation, the goal is
not to “find” revitalization or Taki Onqoy in the archaeological record. Instead, I combine
archaeological remains and historical documents to consider how the people at Iglesiachayoq
would have enacted and experienced daily and religious life during the time of Taki Onqoy.
Liebmann’s study of the seventeenth-century New Mexico Pueblo Revolt has many
commonalities with Taki Onqoy, as both movements sought to throw off the yoke of Spanish
colonial oppression in favor of a “return” to prehispanic cultural traditions. Like Fry, Liebmann
utilizes the construction of new architectural styles as an example of revitalization. He argues
that the two Jemez settlements constructed after the Pueblo Revolt mirror prehispanic settlement
layouts rather than Spanish-style sites. As Liebmann writes, “nativism (defined as the
elimination of foreign elements) was here manifested through the rejection of the dispersed
architectural plan of the mission pueblo, and conversely revivalism (the introduction of cultural
practices thought to have been characteristic of previous generations but not recently present in a
social group) was materialized through the return to aggregated, compact village forms”
110
(Liebmann 2008:364). That is, the Pueblo constructed new settlements in the forms of perceived
past settlements, much like the takiongos developed practices of huaca worship based on an
imagined set of prehispanic practices.40 Liebmann also provides a critical caveat, affirming that
not all Pueblo settlements constructed after the revolt utilized the same prehispanic architectural
forms. Instead, Liebmann’s work emphasizes that revitalization movements are not monolithic
nor do they reflect identical responses from all inhabitants of a region. I would take this
argument even further, and suggest that not all inhabitants of a single site will demonstrate
uniform reactions to revitalization movements, and that participation in these movements is often
partial.
Liebmann suggests that revitalization movements do not necessarily completely revive
previous traditions, but also that they are innovative in terms of practices and material culture.
Liebmann writes that after the revolt, indigenous artists ceased the production of prehispanic
Black-on-White pottery, instead embracing new ceramic types which incorporated traded goods
and glazewares (Liebmann 2007). Furthermore, the potters created a new type of pottery called
“Historic Red.” Liebmann argues that the integration of Black-on-White pottery during Spanish
control resulted in several Spanish forms or schemas which were introduced to this pottery, and
eventually became conflated. This conflation led to the association of Black-on-White pottery
with colonialism rather than with prehispanic art forms. One of Liebmann’s major contributions
is the assertion that revitalization movements—though their explicit motivations may indicate a
return to pre-colonial traditions—actually produce novel, new cultural traditions. We see these
new traditions in Taki Onqoy through the preaching and performance of huaca worship, the
40
Most of those takiongos practicing in the 1560s had not yet been born when the Spanish arrived in Peru, and
certainly none of them were alive prior to Inka takeover. Thus their avowal to practice local huaca worship in the
way of the ancestors was derived from stories of the ancestors combined with aspects of Catholicism.
111
emergence of personal huaca possession, and the diversity in burial practices in the central
church at Iglesiachayoq. Furthermore, like the Pueblo conflated aspects of colonial tradition with
prehispanic tradition, Taki Onqoy practitioners were reviving a perceived idyllic past—however,
those at the forefront of the Taki Onqoy movement were not actually alive in this past, and so
much of their “anti-Spanish” practices are heavily influenced by Spanish presence.
Finally, in the most recent iteration of the use of revitalization theory in archaeological
investigation, Donna M. Glowacki argues that thirteenth-century innovations in settlement,
architecture, and pottery in Central Mesa Verde indicate revitalized religious practices. Glowacki
writes that “these changes came on the heels of a difficult time with severe drought, violence,
and a breakdown of key aspects of the Chaco system. Even though revitalization is not an
inevitable outcome of crises, periods of crises, both natural and social, do create disjunctions that
are often addressed through religious change and revitalization” (2011: 78). Glowacki’s article is
a step forward in utilizing cultural models to understand rapid change, rather than deferring to
common interpretations such as environmental crises or violence. In investigating why societies
changed rapidly, it should be acknowledged that these changes may have been directed and
intentional rather than a response to external stimuli.
Taken together, the works of these authors highlight several very important points. First,
the material changes associated with revitalization are not uniform, and the archaeologist cannot
create a “blueprint for revitalization” as one might for the identification of cooking, warfare,
ceramic production, etc. Second, participation in revitalization movements is not uniform or
monolithic. Finally, the authors provide evidence to suggest that archaeologically documented
instances of rapid change may not necessarily be due to outside forces (such as environmental
112
catastrophe or imperial conquest), but rather can be driven from internal strife or intentional
creations of new cultural practices.
Divided Religions, Entangled Practices: The Utility of Considering Taki Onqoy as a
Revitalization Movement
Taki Onqoy was one response to Spanish conquest and conversion in a milieu of local
confusion and ambivalence to Spanish control in the sixteenth century. In this section, I will first
discuss the competing internal factions and overall goals in the Taki Onqoy movement and then
demonstrate how the “return to prehispanic huaca worship” was developed in an era of religious
turmoil. Although the leaders of Taki Onqoy claimed to preach a separatist, prehispanic message
to their followers, these leaders personally and outwardly assumed aspects of Catholicism in
their identities and in the practices of the movement.
Internal Divisions of Taki Onqoy
The leaders of the Taki Onqoy movement proclaimed a foreboding cataclysm in society,
such that all local individuals needed to resist Spanish religion and culture (Chapter 2), yet the
movement was not uniformly practiced. Taki Onqoy could not ease longstanding preexisting
conflicts between local groups or classes, and Andean peoples were hesitant to jeopardize
tenuous alliances with Spanish officials. As Stern writes, “Taki Onqoy exacerbated the moral
crisis of Indian society precisely because at a moment of severe disillusion and misgiving, it
condemned strategies, adaptations, and traditions which, in practice, too many natives would find
difficult to renounce” (Stern 1993:62). That is, by encouraging Andeans to actively reject aspects
of Spanish culture and religion, Taki Onqoy leaders discouraged ingrained survival strategies
such as appealing to multiple deities and allying with powerful leaders. In some cases when it
113
appeared Spaniards would be dominant for the foreseeable future, Taki Onqoy leaders
themselves even went in search of benevolent Catholic saints who would hear their appeals. As
Stern writes, in a population of some 120,000 people, 8,000 were identified as participants in
Taki Onqoy, indicating that the majority of the people in Soras and Lucanas actually were not
practitioners.
Just as tensions between local groups were not eased by the Taki Onqoy movement, the
popular support of Taki Onqoy exacerbated the tenuous relationships between kurakas and the
general population. Non-elites put obvious pressure on community leaders to support the
movement, but it was these leaders who had the most to lose if they supported a movement that
was ultimately unsuccessful. Community leaders also could not control the popular outbursts of
the local people, thereby challenging traditional power structures in religion and governance.
Since non-elites often relied on kurakas and community leaders to navigate favorable tribute
deals and relationships with Spaniards, the ambivalence portrayed by these leaders inevitably
weakened the movement (Stern 1993:66). Albornoz inadvertently affirms this ambivalence in his
1584 visita, casting the kurakas as “complicit” or “hiding” the movement, rather than actually
participating in it in most cases.
The complicated relationship between native Andean religion and Spanish Catholicism is
apparent in the contradictions and entanglements that shaped Taki Onqoy. Although clearly
designed to be a “revitalization” of prehispanic huaca worship, the movement itself—whether
consciously or otherwise—adopted aspects of Spanish Catholicism in its everyday practice. The
very notion of exclusionary religion was an artifact of Spanish religion, and fully contradictory
to traditional Andean practices of polytheism. That Taki Onqoy preached full rejection of all
aspects of Spanish religion was anathema to local peoples—had the movement been a
114
revitalization in its purest form, then worship of multiple deities would have been embraced. The
entanglement of Catholicism with Taki Onqoy also reverberated in more overt ways which were
often related to gender differences in local groups. Stern’s quantitative analysis of Albornoz’s
accounts finds that of 500 takiongos amongst the Soras, 55% of these were women (Stern
1993:223).41 Albornoz also names several women as leaders of Taki Onqoy. Perhaps women
took an active role in the revitalization movement, seizing the opportunity to resist Spanish rule
in a covert and nonviolent way.
Albornoz affirms that several of the female takiongos did not deny the power of the
Catholic deity, but rather utilized this inherent power to declare themselves “saints.” Taki Onqoy
preached rejection of Spanish religion, but also reaffirmed the power of God—this God had to be
strong in order to have aided the Spaniards in their victory over Andean peoples. Takiongos were
thus weary of fully rejecting God. The self-declared Taki Onqoy female “saints” utilized the
powerful reputation of God and the Catholic saints, and deliberately adopted versions of the
names “Mary” and “Magdalena” (Stern 1993:66).42 Although they still argued for an overthrow
of the Catholic God, these female prophets allied with female Catholic saints, whom they viewed
as part of a broader pantheon. Perhaps most importantly, at the height of Taki Onqoy’s breadth
and strength, some leading takiongos (including two female leaders Santa Maria and Santa Maria
Magdalena) publicly confessed that the movement was a fraud (Albornoz 1990: 181, 225; Gose
2008:115). Both Stern and Gose attribute these sorts of confessions and the tendency for
takiongos to revert to practicing Catholicism to true ambivalence within the leaders of the
movement. Taki Onqoy was thus not perpetuated as a fraudulent movement, but rather was a
41
My own research on the participation of women in Taki Onqoy is ongoing and will be presented in forthcoming
publications.
42
Though I cite Stern here, there are several passages recorded by Albornoz which explicitly discuss this trend.
115
survival tactic put forth by the subjugated Andeans in the face of Spanish conquest and
conversion and its subsequent consequences.
Discussion
Although Taki Onqoy was definitively heralded as a return to prehispanic huaca worship,
the movement itself was not purely nativist in its adoption of Catholicized practices. The
emphasis on possession and the need to exclude the Catholic god were both practices influenced
by Catholicism, and their adoption ironically undercut Taki Onqoy’s explicit goals. Andean
peoples were ambivalent about rejecting the Catholic deity, as he had demonstrated his power
during the conquest and its aftermath. Similarly, takiongos viewed Catholicism as a pantheon
rather than a monotheistic religion—the lower saints could be worshiped in their own right, and
many takiongos would look to these saints in times of crisis. By adopting the names of popular
female saints, women takiongos were analogies to both the nativist revival and the new Catholic
religion.
While undoubtedly a form of preserving local culture against a powerful, superordinate
power, Taki Onqoy was not devoid of factionalization, which broke down the unity of the
movement along preexisting tensions amongst various groups. Local groups which were at odds
with one another remained at odds, although they may have superficially united in the Taki
Onqoy movement. Perhaps the biggest, most damaging rift remained between kurakas and elite
leaders and common Andeans—since these leaders had the most to lose if Taki Onqoy failed,
they were often ambivalent at fully embracing Taki Onqoy in order to preserve their
relationships with Spanish officials (Stern 1993). Taki Onqoy was thus not a uniform, united
116
front as some Spanish officials would indicate to the Crown, but rather suffered from its own
fractious politics which ultimately weakened the movement.
I do assert that the revitalization concept is a useful framework for investigating Taki
Onqoy: first, the framework explicitly considers chronological time and material culture. Second,
although revitalization movements have been rarely archaeologically investigated, they are
common occurrences throughout history. Revitalization movements are documented to have
occurred globally in all types of societies and thus have the potential to shape modern society,
despite the brief time scale in which they form. If historians are able to identify revitalization
movements in texts, so should the archaeologist consider revitalization as a concept in explaining
rapid change in various archaeological contexts. Third, revitalization movements are prevalent in
colonial encounters—these movements often erupt in eras of ambivalence and “cultural
distortion” (Wallace 1956:269), where longstanding cultural practices (such as huaca worship)
have been changed or abandoned. Finally, by analyzing Taki Onqoy as one of the first
documented revitalization movements in the Andes, we can begin to understand how future
movements were influenced by Taki Onqoy.
Taki Onqoy as Performance
Taki Onqoy was a popular outburst which was disseminated through intricate
performances involving cleaning, fasting, drinking heavily, and dancing for days in order to
create a fervency which spread throughout the highlands. Taki Onqoy consisted of both
performances—presented and acknowledged acts which followed standardized cultural scripts—
and performatives, or “the class of semiotic acts that do or can make a change in the world by
their execution and reception” (Chase 2016: 48; see also DeMarrais 2014 and Inomata and
117
Coben 2006). By dancing, chanting, and falling into trances, takiongos were not just acting like
huacas, they were huacas. Following Inomata and Coben’s 2006 introduction, I agree that
“public events and theatrical performance are important for the operation of any society, and
they had critical implications and consequences for the development of centralized polities”
(Inomata and Coben 2006: 3). Performance and spectacle are integral not only to the creation and
maintenance of political relationships, but also to the subversion of political norms, as is the case
with Taki Onqoy. As a spectacle, Taki Onqoy would have drawn the attention of local
participants and observers, and it was deemed radical enough to worry Spanish authorities.
Performative acts are shaped by cultural structures, and also have the ability to alter these
structures through changes in presentation and reception. At the broad end of the spectrum,
performance has been defined as “all the activity of an individual which occurs during a period
marked by his continuous presence before a particular set of observers and which has some
effect on the observers” (Goffman 1959:22). By this open definition, performance plausibly
consists of any social interaction in everyday life. Conversely, on the opposite side of the
spectrum, performance is defined in a narrower sense so that it “encompasses highly
circumscribed and prescribed acts in formalized theaters, in which the performers and audience
are consciously concerned with the theatricality of the acts and settings” (Inomata and Coben
2006:8). Within these more specific performative settings, the audience and the performers can
take on different roles so that they are at various times acting as observers or participants. I
suggest that Taki Onqoy should be considered in the narrower sense, as an explicit type of
performance in which both participants and observers understood the theatricality and goals of
the enactment.
118
According to Schechner, performance and ritual share several guiding characteristics
including predictable ordering of time, imbued value of material objects, non-productivity (as
goods), and rules (Schechner 2003). Performative acts have distinctive scripts and arenas and are
inherently related to politics, power, and hegemony. Moreover, performances are generally
instances of transformation. Victor Turner’s 1974 work discusses how social crises can be
countered through performances. He writes that these sorts of performances are enacted in
conflict situations and that:
They have four main phases of public action…these are 1. Breach of regular,
norm-governed social relations…2. Crisis during which there is a tendency for the
breach to widen…each public crisis has what I now call liminal characteristics,
since it is a threshold between more or less stable phases of the social process…3.
Redressive action to resolve certain kinds of crisis or legitimate other modes or
resolution, the performance of public ritual…Redress too, has its liminal features,
its being ‘betwixt and between,’ and, as such, furnishes a distanced replication
and critique of the events leading up to and composing the ‘crisis.’ 4. The final
phase…consists either of the reintegration of the disturbed social group or of the
social recognition and legitimization of the irreparable schism between contesting
parties. (Turner 1974:37-41)
Turner’s explication of crisis or conflict situations mirrors Wallace’s discussion of revitalization
movements in its creation of multiple teleological phases in order to explain a particular human
phenomenon. However, aspects of this set of stages are useful when considering the conquest of
the Andes. Employing Turner’s model, Taki Onqoy performances were enacted in order to
resolve the crisis of Spanish conquest and conversion. These performances themselves were a
liminal space in which takiongos were able to critique the problems of Spanish presence through
the bodily possession of huacas. Takiongos thus changed social norms through their dramatic
dance performances—by ritually transforming their physical bodies, the performers existed in
another time and place. The transformation of the performers, then, was also experienced by the
spectators—the dances completed by takiongos would have been awe-inspiring and at least a
119
temporary resolution (or at least an agentive and strategic strategy) to combatting the problems
of Spanish conquest.
Performance can be difficult to study archaeologically—the event itself is often
ephemeral. Yet, “because performances are usually subjunctive, liminal, dangerous, and
duplicitous, they are often hedged in with conventions and frames: ways of making the places,
the participants, and the events somewhat safe” (Schechner 2003:xiv). That is, performance has
the ability to both establish and transform social orders and can affect change in both secular and
religious realms. Performance is therefore a form of communication, and archaeology can
delineate aspects of this communication through “understanding of a ritual space or theater and
the ways that its layout affected the relationships between performers and an audience”
(DeMarrais 2014). Archaeology is well-suited to consider how architectural space would have
structured movement and affected bodies. Perhaps most importantly, archaeology can be used to
understand the past practices of performances—since archaeology cannot presume to understand
past belief systems, studying the performances and practices can shed light on how culture
changed during colonial encounters.
Performance is often used as a medium through which individuals are more able to
converse with deities or the supernatural. However, performance can also be used to portray or
put forth specific messages and meanings to both observers and participants. An explicit goal in
the archaeological study of performance, then, should be to consider how these practices
communicated and signified meanings, but also to think of how meanings may have varied to the
individual observers. If each Taki Onqoy performance is considered its own event, then one must
try to account for the undoubtedly multiple reactions to this event, and the meanings for
individual participants.
120
Taki Onqoy’s message and its practices were clearly delineated in the works of Albornoz
and Molina, and will be discussed in depth in the next chapter. Here, however, I want to theorize
how Taki Onqoy practices would have functioned as performances, and what sorts of messages
they may have conveyed to observers and participants. Taki Onqoy is a Quechua phrase meaning
“dancing sickness,” and its primary performative aspects were uncontrollable singing and
dancing, in which the takiongos became prehispanic huacas. Those possessed by the huacas
would have seizures and emerge “purified” of the sickness, preaching a message of overt
rejection of the Spanish God and broader Spanish culture. Takiongos often performed in
specially-designed circular enclosures where they would fast, drink, dance, and feast (Albornoz
1990:191). These enclosures became the stage for elaborate Taki Onqoy performances, during
which takiongos supervised the performances and simultaneously received offerings from those
participants (Stern 1993:53). In Bartolomé Álvarez’s (1998 [1588]) account of Taki Onqoy, he
writes that adepts “danced in groups, raising and lowering their legs and arms in a vigorous
stationary march with clenched fists, moving their heads from side to side with each step and
making a high-pitched sound: ‘u, u, u, u’” (Álvarez 1998 [1588]: 124-125).43 These
performances went on for days, and participants abstained from food, drink, and sleep such that
some eventually collapsed or even died. Álvarez suggests that those performers who pushed
themselves the furthest were the most respected and revered.
Álvarez (1998[1588]) writes: “Es la fiesta que, juntos dellos la cantidad que se conciertan—y a veces uno o dos
solos que quieren hacer la cerimonia—comienzan a cantar un cantar que no es palabras, ni razones ni sentencias ni
cosa que se pueda entender que dicen algo. Sólo suena “u, u, u, u”: es menester oírlo y verlo para entenderlo, que es
tal que no se puede excribir. Y con este canto muy alto están de pie, dando de pie y mano, alzando un pie y abajando
otro, y asimismo haciendo con las manos, los puños cerrados, meneando la cabeza a un lado y a otro, de suerte que
con todo el cuerpo trabajan. Y paran en este canto tres o cuatro días ocn sus noches, y mas: lo que las fuerzas les
duran, que no cesan si no es que les venga necesidad de hacer cámara o de orinar, que a esto salen, y luego vuelven a
la tahona del demonio” (125).
43
121
Consider the spectacle as an outsider: groups of native peoples chanting loudly in
Quechua, and performing dances in unison which went on for days. The endurance of these
performers must have been astonishing, and their ability to continue for extended periods without
food or drink might even have made them seem as though they had supernatural abilities. The
respect and reverence given to those who collapsed or even died would have reinforced the role
of these individuals as spiritual leaders. The destruction of Catholic symbols—a practice
sometimes performed during these dances—would have alarmed and threatened those Catholic
converts. Spanish priests—when in residence at Taki Onqoy towns—could have experienced
Taki Onqoy performances as an overt expression of idolatry; however, the dramatic length and
exertion required to participate in these dances likely would have prompted a more visceral
response in Spanish observers. The fact that these performances often took place in special
enclosures away from church spaces would have made the dances more surprising, as they were
located in spaces away from the sanctity and protection of Catholic realms. Spanish officials
were perpetually concerned about the unpredictability of Taki Onqoy—it could strike anyone at
any time, regardless of age, gender, or class. Confining the spread of Taki Onqoy was thus
challenging and unrealistic for the Spaniards, as they could not predict who would be “afflicted”
by the dancing sickness. Additionally, Spanish religious authorities had limited resources at this
time—it is likely that Taki Onqoy performances were practiced when the Spanish church
authorities were not physically present in certain towns—and thus felt powerless to halt the
spread of the idolatrous practices. It is unsurprising that Taki Onqoy greatly alarmed the Spanish
officials—the eerie performances had existed for some years before the Spanish were alerted to
them, and the literal popular outbursts were unpredictable and apparently, had spread rapidly.
122
Taki Onqoy performances likely had very different meanings for Andeans, whether they
were priests, performers, observers, or abstainers. Taki Onqoy was clearly not a uniform front
practiced by all local peoples against the Spanish: Albornoz divides the idolaters he identifies
into several categories depending on their degree of involvement with the movement. The most
powerful takiongos were the “maestros,” or leaders of the movement—the punishment levied on
these individuals included flogging, shaving their heads, and often exiling them to work in the
churches or hospitals of Huamanga. These leaders were in charge of the feasting and fasting
associated with Taki Onqoy, and also utilized these opportunities to preach their message of
revitalization. Taki Onqoy preachers would have been the major performers of the ritual
dances—in order to gain respect, these preachers would have gone to dramatic lengths to outlast
other performers, and maintain their fatigued trance for days. For the preachers, the
performances took on multiple meanings: first, they were intended to draw people into
participating in the movement. The dramatic and sensory experience of the performances would
likely have impressed observers, and the preachers would have utilized the dancing to spread
their message. In a more practical sense, the Taki Onqoy preachers utilized these performances
to control portions of the population and increase their own personal wealth. Participants and
observers would often bring gifts to the preachers, as these individuals had already proven
themselves to have been cleansed by huaca possession. Thus, for these preachers, performing in
Taki Onqoy dances served both a public and a private purpose: publicly, the preachers could
spread the message of the huaca revival, and privately, these leaders were gaining wealth and
prestige within the Taki Onqoy movement.
A second category of people associated with Taki Onqoy were known as accomplices
and/or participators in the “mala seta apostasia” (Albornoz 1990:260). These individuals were
123
those who either were caught dancing or were complicit in hiding the movement from Spanish
officials. Like the preachers, punishment included shaving of the head and flogging, as well as
forced labor demands to help build the churches of the region. In some cases, those deemed
participators were also mandated to meet with the Catholic priests three times a week in order to
be instructed in Catholic doctrine. As participators in Taki Onqoy performances, these
individuals likely were affected by the religious fervor, such that they were willing to bring gifts
and tribute to the preachers leading these events. These individuals also could have used Taki
Onqoy performances strategically in order to increase their own status within the movement. For
those considered accomplices, or observers, the performances may have prompted an ambivalent
response. By watching the performers but not participating, observers could plausibly argue both
support for Taki Onqoy and support for Catholic doctrine—they may have employed very
different public and private transcripts in order to navigate the shifting politics at this time.
A third and final category of people who were punished by Albornoz were the kurakas,
who also took a variety of positions relative to the Taki Onqoy performances. As discussed
briefly above, kurakas had the most to lose if Taki Onqoy ultimately failed—their tenuous
alliances with Spanish officials were at risk. Most kurakas thus took a position of ambivalence
toward the movement and the performances. Some were punished for being “complicit” or
“concealers” while others were punished for outright practice and support of the movement. For
these kurakas, dancing events may have been wrought with stress—concealing the movement
would protect the individuals whom they supervised, but also jeopardized their standing vis-á-vis
Spanish authorities. However, the reverse was also true—by reporting the performances, kurakas
put their own people in danger of punishment, and also risked losing authority by not backing the
popular uprising.
124
Thus, there was a diverse array of perspectives from different groups who would have
either participated in, observed, or concealed Taki Onqoy dance performances. What purpose did
these performances serve in general? Public spectacles and performances “tolerate a degree of
internal resistance and indifference among the participants, while requiring their external
consent…it grounds and displays a sense of community without overriding the autonomy of
individuals” (Inomata and Coben 2006:16-17). By the lists and witness testimonies provided by
Albornoz, it is clear that not all highland individuals were participating in Taki Onqoy. However,
by defining the movement through fervent, multiday performances, takiongos brought people
together to witness the spectacle. Publicly, the movement could have appeared as a united
front—peoples’ external support could easily have masked their internal confusion, ambivalence,
or indifference.
I suggest that in the 1560s in highland Peru, there were parallel performative religious
practices which served opposing goals. For the Spanish, the forced conversion of Andean
peoples and their subsequent participation in church sermons, teachings, and rituals, served to
integrate Andeans into a formalized Catholic discourse. Practiced for centuries, these church
rituals and performances were put forth by the state and Spanish religious authorities, and their
explicit goal was to teach Catholicism to Andean converts. At a time when the Spanish lacked
the physical capabilities to oppress new subjects, “one cannot overemphasize the importance of
performance in establishing, affirming, manipulating, and maintaining power relations between
elites and nonelites” (Inomata and Coben 2006:18). That is, the Spanish utilized religious
performative ceremonies to create new, blended communities and reaffirm newly established
power differentials in the sixteenth century. Since Andean and Inka peoples could not read or
write in Spanish or Latin, music, songs, and public practices were often used to transmit these
125
religious teachings (Abercrombie 1998; Durston 2007; Estenssoro 2001, 2003, Gose 2008, see
also Chapter 2).
While many Andeans undoubtedly publicly ascribed to these Catholic performances,
Taki Onqoy was an opposing, yet parallel performance-based movement whose goal was to
subvert the dominant Catholic ideologies. From 1558-1561, the Corregidor del Cuzco, Juan Polo
de Ondegardo y Zarate, sought to destroy Cuzco huacas and mummies—during this time, he also
outlawed “public and solemn drinking sessions” yet allowed drinking “within private houses”
(Acosta, as quoted in Abercrombie 1998: 48). Abercrombie posits that this new law was derived
from Spanish opinions of public drunkenness, and that it was associated with worshipping idols.
This new policy subverted the previous dictum which allowed kurakas to manage their people
however they desired: “priests and colonial officials permitted idolatries, drunkenness, and other
sins in exchange for satisfaction of their economic interests” (Abercrombie 1998:216). In
forbidding dancing and drinking in public spaces, Polo effectively removed the kuraka’s ability
to watch over the activities of his kin, inadvertently encouraging private spaces such as
households to be used for native traditional practices such as drinking and dancing, while
Spanish spaces were delineated as being located in the public sphere.
Inhabitants of Iglesiachayoq and other highland sites in Soras and Lucanas were thus
enmeshed with two competing performance-based religious ideologies. Local peoples were also
ill-equipped to put forth a successful violent rebellion, so the Taki Onqoy movement served to
subvert Catholic teachings and promote the past prehispanic huaca worship. Although Spanish
authorities claimed that Taki Onqoy presented a somewhat united front which could afflict
anyone at any time, these authorities were also assessing the movement as to its public
performative aspects rather than the private beliefs of the takiongos. In reality, Andeans were
126
likely situationally moving between these two religious traditions—though they may have
publicly participated both in a way which provided external consent and support, their private
beliefs may have been laced with ambivalence and fear. At this time, Andean popular religion
was transactional in nature: Andeans would provide gifts to different huacas or to the Catholic
God in order to gain protection, good harvests, health, etc. They thus would have been unlikely
to fully renounce a powerful deity which had demonstrated its dominance.
In this section, I have tried to elucidate the shifting practices, beliefs, behaviors, and
transcripts which were overlapping and intertwining at Iglesiachayoq in the sixteenth century.
Instead, I situate how Taki Onqoy could have been practiced simultaneously with Catholicism,
and how ambivalent Andeans likely navigated these two extremes. Just as the ethnohistoric
analysis must be nuanced, account for factions and internal conflict, and propose multiple
interpretations of events, so too the archaeology must embrace this ambivalence. As Silliman
continues, “the difficulty in recognizing Indigenous people in distinctly colonial settings of the
past lies in the fact that artifacts and spaces in colonial worlds are fraught with ambiguity,
alternate functions, and multiple users” (Silliman 2010:32). This perspective is even more
complicated during Taki Onqoy, as the goal is to understand how Andeans actively navigated,
participated in, and shaped both realms.
Do Taki Onqoy Practices Constitute a Novel Response to Spanish Presence?
In the final section of this chapter, I will address the problem of Taki Onqoy as a novel
movement. Much of the secondary literature views Taki Onqoy as the creation of Spanish
authorities, who could have invented the movement for their own benefit (Chapter 2). They
suggest that Taki Onqoy practices were not distinctive, and that they merely consisted of the
127
same indigenous resistive practices which had been documented throughout Inka and Spanish
conquest. The question thus becomes, was Taki Onqoy merely a continued set of Andean
practices utilized in times of stress? Or was Taki Onqoy a unique movement, one which was
created and practiced in direct response to Spanish religion and rule? To fully consider the
possibilities, I turn the 1550s Relación de los Agustinos de Huamachuco. In the remainder of
this section, I will clarify which aspects of the Taki Onqoy movement were novel and newly
developed, which practices were explicit reinstitutions of general prehispanic Andean traditions,
and which beliefs continued to influence future revitalization movements.
Taki Onqoy Practices as a Revival of Prehispanic pan-Andean Tradition
One of the earliest documentary sources regarding Andean religious practices during
Catholic evangelization is the 1550-1560 Relación de los Agustinos de Huamachuco, written as a
somewhat ethnographic study of Andean peoples in the Huamachuco region (Castro de Trelles
1992). Though brief, the Relación is critical for the research presented here because it details a
suite of Andean religious responses to Catholicism prior to the generally-accepted time period
assigned to Taki Onqoy (1560-1580). Though little is known about the author of this Relación,
he seems to be one of the first twelve Augustinian priests who arrived in Peru in 1551 (Castro de
Trelles 1992: XII). The author of the Relación is clearly frustrated with the idolatrous practices
of the people in Huamachuco. He often complains that “yndios” who appeared to be “buen
Cristiano” secretly worshipped the huacas (Relación 1551:10). For example, when describing the
adoration of these huacas, he writes:
They had huge corrals with some of the walls built up very tall so that they could
have their parties, and in the middle they put a stick and moved around it, and that
which was to be sacrificed was placed beneath the stick dressed in white clothing,
128
and they killed a cuy and offered the blood…others killed sheep and threw the
blood at the stick…in the walls there were many small shelves to save the relics
that were left from the sheep or goat…and in these corrals they had huge parties
and their sacrifices would last five days and they performed huge taquis and sang
dressed in their best clothes…sometimes they did not eat aji nor salt and only
drank chicha, and did not sleep with their women.44 (Relación 1551: 12-14)
The Augustinian priest essentially is outlining a collection of Andean practices which are similar
from pre-Spanish conquest throughout the evangelization period. Throughout the Late
Intermediate Period and the Late Horizon, Andean groups often participated in rituals which
involved fasting, heavy drinking, dancing, singing, and animal sacrifice. During the Taki Onqoy
movement, these same actions are reported by Molina, who describes them in much the same
way as does the Augustinian priest: “And to return to these times [pre-Spanish conquest], some
days they didn't eat salt nor aji, nor did they sleep man with a woman, nor did they eat different
colored corn” (Molina 2010: 96).45 It is perhaps suspicious that Molina uses nearly the exact
terminology to describe Taki Onqoy practices as the Augustinian priest—however, the practices
being described made up many shared cultural tendencies commonly attributed to lo andino.46
During the dancing events, afflicted Andean peoples ate nothing other than coca, and
then drank heavily. The fasting and subsequent overconsumption are connected with the
movement’s ideals of purification—Taki Onqoy practitioners were attempting to rid themselves
Translated from “Tenyan grandes corrales y estos tenyan por una parte la pared muy alta y tenyan dentro unos
hoyos donde hincaban unos palos para hacer las fiestas, y en medio ponyan un palo y revolbianle con paja y
atabanle, y el que avia de sacrificar subia encima del palo, vestido de unas vestiduras blancas, y matavan un coy y
ofrecia la sangre…y otros matavan ovejas y echavan la sangre al palo…avia en las paredes munchas poyatillas para
guardar las reliquias que de la oveja o carnero quedavan…y en estos corrales hazian grandes fiestas en sus
sacrificios que duravan cinco días y hazian grandes taquis y cantos vestidos lo mejor que podían; hay grandes
borracheras…el ayuno es que no a de comer agi ni sal, beber azua o chicha…no han de dormir sus mujeres.”
(Relación 1550:12-14), translation mine.
45
“Y que para bolver a ellos ayunasen algunos días no comiendo sal ni ají, ni durmiendo hombre con mugger, ni
comiendo maíz de colores, ni comiendo cosas de Castilla, ni usando d’ellas en comer ni en vestir, ni entrar en las
yglesias, ni reçar, ni acuda al llamamiento de los padres curas, ni llamarse nombre de Cristiano” (Molina 2010),
translation mine.
46
I am using this term here to refer to common practices which occur in geographically diverse areas at different
times. I do not want to essentialize Andean groups, but I think recognizing the similarities in these practices is
critical for Taki Onqoy studies.
44
129
of the sicknesses brought by the Spanish, and their bodily practices aligned with these ideals.
However, as Gose argues, these consumption practices were a recycling of previous Andean
traditions: Gose sites Guaman Poma and other sources as referring to Taki Onqoy as a disease,
without directly linking it to the movement of Taki Onqoy (Gose 2008: 96). Similarly, Molina
provides thorough descriptions of prehispanic calendrical rituals, many of whose practices
resonated with Taki Onqoy. For example, when discussing the rituals in the month of July,
Molina writes “The tarpuntaes, who are a group of people like priests, took care to fast from
when they planted the maize until it erupted from the soil about a finger in height. And in this
time they would not have sex with their wives. Likewise their wives and children would fast”
(Molina 2010: 50, translation mine).47 At times, Molina conflates Andean and Inka practices;
however, since Inka conquest was founded on a shared cultural legibility with Andean groups, it
is plausible that aspects of these rituals had been practiced by non-Inka Andeans in the preceding
centuries. This notion of the cleansing of the homes and bodies of Andeans prepared them to
symbolically renounce Spanish traditions in preparation for embodiment of huacas.
The embodiment of huacas in human hosts is similar to the Catholic tradition of oracular
possession—the combination of fasting and extreme physical exertion would have created a
trancelike status such that takiongos could have interpreted this trance as huaca possession. This
practice of huaca possession was customary in prehispanic Andean groups, although several Taki
Onqoy scholars have neglected to investigate this further. During seasonal rituals, Andean people
would interact with their deceased (often mummified) ancestors. The ancestors could provide
advice or consultation by speaking through living mediums (Gose 2008: 97). In the Relación,
“Los tarpuntaes, que es una jente como sacerdotes, tenýan cuydado de ayunar desde que sembravan el maíz hasta
que salía de la tierra como un deo en alto. Y en este tiempo no se juntavan con sus mugeres. Y asimismo ayunavan
sus mugeres e hijos d’éstos” (Molina 2010)
47
130
this ancestor worship appears to continue. For example, the author argues that “these Indians had
a powerful Inka captain called Condor who had been dead 43 years, and his son was the señor de
Huamachuco, the Indians said at the time that they had destroyed and burned his body…this
body they had hidden in some heaps of corn where they said the preservation was better—their
clothes were well-preserved” (Relación 1550: 30).48 During Taki Onqoy, these practices were
somewhat transformed in that landscape huacas (as opposed to the ancestor cult) were thought to
have become mobile, flown through the air, and inhabit various individuals. Like the common
Andean practice of speaking with the deceased for advice, in the 1560s all huacas were
possibilities for possession and subsequently, consultation. A second potential explanation for
the new mobility and possession of huacas was that early Spanish extirpation attempts destroyed
many of these landscape huacas, breaking them into numerous pieces which then became
mobile.49
In any case, oracular possession and speaking with the huacas both had precedents in the
broader suite of pan-Andean practices. It is striking, then, that several secondary scholars
neglected to trace this tradition’s lineage. By omitting the Andean roots of this practice,
secondary authors could more easily argue that Taki Onqoy was a creation by Spanish authorities
rather than a truly Andean movement. For example, in Millones’ 1990 influential work on Taki
Onqoy, the contributor Rafael Varón discusses the Andean antecedents of Taki Onqoy, yet omits
“Tenyan estos yndios un gran capitan del ynga que se llamava Condor con un hijo suyo el qual era senor de
Guamachuco, que avia que murio quarenta y tres anos, digo al tiempo que se hallo y quemo su cuerpo. A esta
mochaban los indios, adoraban y hazian grandes fiestas; a estos cuerpos o por mejor decir, a el demonyo en ellos
hazian grandes sacrificios quemando coca, mais, y las otras cosas. Estos dos cuerpos los tenyan escondidos en unos
montones de mayz que dizen que allí se conservaban mejor; tenyanlos bien compuestos de ropa y otras cosas…”
(Relación 1550: 30), translation mine.
49
Here, again see the Relación of Huamachuco because it has several passages regarding the destruction of huacas.
The priests report that the Andeans would worship destroyed huacas, saying that the dust rose to the sky and stayed
with their deities (Relación: 17; Arriaga 1620).
48
131
the theme of ocular possession (Varón and Albornoz 1990).50 More significantly, in her
denouncement of the existence of Taki Onqoy, Gabriela Ramos negates the practice of oracular
possession prior to Spanish presence and instead suggests it is an “invention” of Catholic beliefs
regarding demonic possession (Ramos 1992: 153-154). Ramos thus casts this practice as a new
tradition brought by the Spaniards in order to bolster her argument that Taki Onqoy was a great
creation and/or exaggeration by Spanish religious officials.
The foundations of Taki Onqoy, then, can be linked to prehispanic Andean ritual
practices throughout Peru. Numerous documented rites and seasonal celebrations had observed
the same fasting, cleaning, dancing, drinking, and singing popular during Taki Onqoy. Similarly,
sacrifice of rams and cuyes and the use of red pigment to paint faces were also common
traditions in Andean ritual. In building on these specific Andean ideas about ancestral authority,
cleanliness, and performance, the Taki Onqoy movement was legible to Andean groups in its
delimiting practices. However, in its response to Spanish religious conversion and conquest, Taki
Onqoy was a specific sociopolitical movement designed to thwart this control. Despite all of the
revitalized practices from the prehispanic period, the Taki Onqoy movement also preached
dictates which would have been at odds with native beliefs. Specifically, the exclusionary nature
of Taki Onqoy was in direct contradiction with the inclusive nature of Andean religion prior to
Spanish arrival.
Taki Onqoy Messaging: Anti-Spanish but Shaped by Spanish Teachings
With the introduction of monotheistic, exclusionary religion at the cost of destruction of
local huacas, Spanish Catholics thrust local religious practices into turmoil. Andeans interpreted
50
Although he omits thorough discussion of ocular possession, his chapter and analysis are brilliant.
132
the destruction of their huacas as a bodily sickness (Brosseder 2014:75). This interpretation was
not just a figurative one—Andean groups perceived the destruction of their ancestors and their
landscape deities as the cause of the ailments in the second half of the sixteenth century. It was
in this antagonistic milieu in which Taki Onqoy developed, and it is why Taki Onqoy’s central
tenet was to worship those huacas who had been destroyed. In the Andean religious framework,
individuals could worship both their traditional idols while simultaneously acknowledging the
power of the Catholic god. Yet, these Andean takiongos also must have been influenced by the
very teachings of the Catholic Church in that the most novel practice associated with Taki
Onqoy was the rejection of all things Spanish. As Stern affirms, “the huacas and takiongos
expressed the gnawing anxiety that natives who assumed Hispanic styles of religion, clothing,
food, and custom were becoming enemies tied to colonialism and Christianity” (Stern 1993:59).
Taki Onqoy preachers were insistent that those Andean peoples who had actively embraced
“Hispanic” orientations were thus enemies of the native communities. In order to regain the
alliances and trust of their fellow Andeans, those who practiced Catholicism had to renounce
these beliefs and instead join the Taki Onqoy movement.
However, disbelieving in the Catholic god was not ideal for Andean individuals—this
god had demonstrated his power in helping the Spaniards triumph over the Inka, and thus was a
deity to be feared and admired. Taki Onqoy preachers creatively did not denounce the power of
this god; instead, they affirmed the power, but preached that it did not apply to Andean peoples.
While the Spanish god had given Spaniards their successes in Spain and in the New World, it
was the huacas which supported the Andeans—Andean peoples had to recommit to worshipping
the huacas, providing them with gifts and food, in order to win back their support which they had
lost. Still, this notion of exclusionary religious practice was anathema to local Andeans—
133
avoiding the Catholic god would have been seen as risky to local peoples, especially since this
god had clearly demonstrated his power. In practice, the rejection of the Spanish god and other
Spanish traditions was not evenly supported, nor was it fully enacted by Taki Onqoy preachers
themselves. As will be discussed in the next chapter, Taki Onqoy priests were known to shift to
praying to benevolent saints when it seemed as though Spanish officials were going to be the
ultimate new ruling body (Albornoz 1990).
While it was simple enough to preach the rejection of all Spanish-associated beliefs and
objects, in practice it would have been difficult to control obtainment of these goods or hidden
practices of Catholicism. Similarly, denouncing the Catholic god was risky for Andean peoples,
who understood his strength and capabilities. Spanish authorities taught exclusionary notions of
religious devotion—they expected Andeans to worship their god and no other deity. This very
notion of exclusivity clearly influenced the on-the-ground messaging of Taki Onqoy. Takiongos
easily could have promoted a nationalist movement and resurgence of huaca worship without
also demanding rejection of Spanish culture. It is this concept of rejection that marks an
incredibly novel concept in Andean religion, one that was specific to Taki Onqoy and would
have an effect on revitalization movements to come.
Reframing considerations of Taki Onqoy: A Practice Theory Approach
In secondary literature, many authors take an either/or position on the practices indicative
to Taki Onqoy—if these practices are deemed as “native” and “historical,” then the movement is
considered a resurgence of native culture. However, this argument also allows scholars to argue
that Taki Onqoy was not a “real threat” to the Spanish Crown because it represented the
continuation of pan-Andean traditions rather than any new, novel movement. Conversely, other
134
scholars of Taki Onqoy suggest that the associated practices were new and heavily influenced by
the Spanish—they use this data to claim that Spanish officials must have created the movement
since so many of the practices mirrored Catholic practice. Both of these designations can lead
scholars to the interpretations of Taki Onqoy that they themselves desire. Instead of trying to find
the exact root or long duree history of each Taki Onqoy practice here, I have tried to delineate
the entanglements and demonstrate which practices may have been influenced by Spanish
authorities or Spanish religion.
In order to understand the performance of any typical practice in the past, scholars must
not only investigate the final product, but also the structuring factors of said practice, and any
ruptures in this structure which altered its realization. Event, structure, and conjuncture are useful
in the study of colonial encounters because these dramatic occurrences force groups to use their
own cultural frames of reference to interpret novel stimuli produced by the “other.” In reading
and interpreting Taki Onqoy practices, and specifically whether they represent novel phenomena
or continuations of the past, I suggest we use the concepts of structure, event, conjuncture, and
“structure of conjuncture” to explain the long-term processes, short-term occurrences, and the
relationship between the two (Beck Jr. et al. 2007; Ohnuki-Tierney 1990; Sahlins 1981, 1985,
2004; Sewell 2005; Whitehead 1995). Broadly speaking, cultural structures are those guiding
principles which govern daily practices and interpretations of novel phenomena. Events are “that
relatively rare subclass of happenings that significantly transforms structures” (Sewell
2005:100). In Sewell’s conception, structures are flexible, and are shaped by important events. In
this scenario, events are attributed their own agency—they have the power to challenge the
cultural constructs which govern human societies. In order for a historical event to become
transformative, it must be a “ramified sequence of occurrences that is recognized as notable by
135
contemporaries, and that results in a durable transformation of structures” (Sewell 2005:228). In
this case, the arrival of the Spanish in the New World and their rapid expansion into Peru was an
indelible event which altered the previous prehispanic guiding cultural structures.
Marshall Sahlins’s theory of the “structure of conjuncture” can be applied to Sewell’s
model in order to create a comprehensive theory which encapsulates structure, event, the
combination of the two, and provides temporal guides for social change. Sahlins defines the
structure of conjuncture as “the practical realization of the cultural categories in a specific
historical context, as expressed in the interested action of the historic agents, including the
microsociology of their interaction” (Sahlins 1985:xiv). Sahlins adds a specific situational
context from which to interpret the juxtaposition of structures—his emphasis on “interaction”
recognizes the clash of two schemas and the resulting, changed structure. The structure of
conjuncture, then, would analyze the specific circumstances in which an event occurs and how
this event changes basic frameworks in society. As Sewell elaborates, this “structure of the
conjuncture” is reversible such that it may also be thought of as a “conjuncture of structures”:
“what makes possible the peculiar dynamic that characterizes events is the conjoining in a given
situation of structures that previously either had been entirely disjointed or had been connected
only in substantially different ways” (Sewell 2005:221).
When the Spanish thrust Catholicism upon the people of the New World, indigenous
groups would have understood this new religion from the perspective of their Andean
worldview. The Spanish initially likely were seen as another imperialist group—albeit one
without a shared cultural realm. Yet, the idea of the Catholic god and worshipping a novel deity
were not new practices to Andean peoples. The event of the Spanish arrival, however, indelibly
altered practices which were uniquely Andean—the practice of religious exclusion is the major
136
example of this alteration. Thus, I argue that rather than being untouched, cultural traditions
dating to the prehispanic Andes, the unique practices indicative of Taki Onqoy were influenced
by Spanish presence, teachings, and religious practices.
Discussion and Conclusion
In this chapter, I have tried to shift discussion of Taki Onqoy away from the ongoing
debate in secondary sources—there is no utility in continuing to ask if Taki Onqoy did or did not
exist or what its “true” motivations were. This debate has already been thoughtfully completed
and is ultimately untenable from an archaeological perspective. Instead, through the application
of theories of revitalization, performance, and practice theory, I have attempted to show how
colonial power differentials, cultural forms, and active and passive forms of resistance interacted
to shape the movement.
Taki Onqoy, like any other popular religion, would have been understood, implemented,
performed, and resisted in different ways by individual Andean peoples. The movement
exacerbated preexisting social tensions between kurakas and their subjects, Spanish and Andean
groups, and Andean Catholics and Andean takiongos. Moreover, in a time of extreme stress and
rapid cultural change, local individuals would have been primarily concerned with ensuring the
survival of themselves and their kin. In the sixteenth century, these precautions likely were
performed through participation in both huaca worship and Catholicism, as individuals would
have used all of their “spiritual currency” to protect their families.
137
CHAPTER 4
HISTORICAL AND ARCHAEOLOGICAL BACKGROUND OF THE CHICHA-SORAS
VALLEY
Introduction
The sociopolitical and geographic climate in which Taki Onqoy thrived was the
culmination of over 3,000 years of prehispanic human habitation, abandonment, and
resettlement. The Chicha-Soras Valley, the dividing frontier between the Departments of
Apurímac and Huamanga, has been subject to human habitation since the Preceramic Period (?1800 BCE), and was a key area in the Middle Horizon (~500-1000 CE) and Late Intermediate
Period (~1000-1438 CE) for intensive agricultural production as evidenced by the extensive
investment in terrace systems (Branch et al. 2006; Keeley and Meddens 1992; Kemp et al. 2006;
Meddens 1985). As a desirable region for agricultural production and animal husbandry, the
Chicha-Soras Valley experienced successive takeovers by both the Wari and the Inka. Yet, the
valley is relatively isolated from the major cities of Cuzco and Ayacucho, and was historically a
region which was difficult for Spanish authorities to reach—and by extension to control—prior
to Toledo’s resettlement program (Albornoz 1990; Meddens and Schreiber 2010). Today, the
region remains politically isolated and encompasses some of the poorest communities in Peru.
Archaeological research into the Chicha-Soras Valley (and more broadly Apurímac and
Huamanga) reflects this isolation: in addition to the lack of road infrastructure and difficulties in
merely traveling to the region, scholarly research stagnated in the last three decades during the
rise and zenith of Maoist terrorism (Shining Path or Sendero Luminoso) and the government’s
response to this terrorism (Comisión de la verdad y reconciliación 2003). The people in
138
Huamanga and Apurímac are not inherently bellicose or rebellious, but the geographic and
sociopolitical trends in the region have fostered a culture of governmental neglect and
subsequent subversion of cultural norms.
This chapter discusses the geographic and cultural context of the landscape and people of
the Chicha-Soras Valley. Through assessment of these foundational frameworks, I aim to create
a background for understanding how and why Taki Onqoy was adopted in this region. In the first
portion of the chapter, I outline the parameters for human settlement of the Chicha-Soras Valley,
and specifically how these physiological features made the region desirable for multiple imperial
projects. Next, I relate the archaeological background of study in the Chicha-Soras Valley,
demonstrating how the region has been home to multiple settlements over an extended time
scale. Specifically, I show how patterns of settlement and abandonment related to one another,
and how the relative independence of the people in the Chicha-Soras Valley created an ideal
sociopolitical atmosphere for the Taki Onqoy movement. Finally, since the Early Colonial Period
is the critical time period for my study, I interweave different types of Spanish documentation
(chronicles, ecclesiastical records, political visitas) to show what Spanish authorities knew about
the people of the region, why Taki Onqoy thrived here, and why the first Extirpation Campaigns
of the 1560s were levied in Huamanga and Apurímac.
Geography and Climate
The Chicha-Soras Valley defines the border between the departments of Ayacucho and
Apurímac in the southern highlands of Peru (province of Sucre), and spans from 2,600 masl on
the valley floor to 4,000 masl at the broad tableland surrounding the valley (Figure 4.1). The
Chicha-Soras River flows down the valley, eventually joining the Rio Pampas, which finally
139
becomes a tributary of the Rio Apurímac. The rocks in the valley are volcanic in nature,
consisting mainly of deposits of basalt or tuff (Keeley and Meddens 1992: 121). These deposits
are very extensive, suggesting multiple lava flows which hardened over long periods of time,
punctuated by shorter, more explosive events (Eastman 1980). Several of the basalt deposits are
columnar in nature, indicating rapid flow and slow cooling, with volcanic activity in the region
dating to between the Ordovician Age (488.3 – 443.7 million years ago) and Devonian Age
(419.2 – 358.9 million years ago) (Mapa Geológico Del Peru 1975).
The Chicha-Soras Valley itself was formed over millions of years through the actions of
volcanic activity, flowing water, frost, and erosion (Eastman 1980). These processes resulted in a
broad, high, flat tableland or mesa punctuated by very steep valleys. The surface geology of the
Chicha-Soras Valley is of volcanic origin, and geothermal activity continues today as evidenced
by the hot springs near San Pedro de Larcay (Meddens and Schreiber 2010). As a result of these
ancient formation processes, some 65% of the territory is made up of massive, flat pampas such
as Sicuani Pampa, Qatunrumi Pampa, Soras Pampa, etc. (Mallco 2013). Conversely, 35% of the
territory is made up of steeper slopes and quebradas, where the small pueblos of Chicha, Soras,
Pampachiri, and San Pedro de Larcay are located (Mallco 2013, Figure 4.1).
140
Figure 4.1. Location of Iglesiachayoq and the Chicha-Soras Valley between Ayacucho and
Apurímac.
Periodic eruptions from the now-extinct volcanoes of Sotaya and Carhuarazo some four
million years ago helped to shape parts of the upper valley (Mapa Geológico Del Peru, 1975).
Sotaya’s activity shaped Pampachiri’s Bosque de Piedras, a very popular tourist location in the
Chicha-Soras Valley (Figure 4.2). This Bosque de Piedras consists of numerous conical rock
formations which range from six to eight meters in diameter, and some over ten meters in height
(Mallco 2013). West of Sotaya and located on the upper tableland at an altitude of approximately
4600 masl, the extinct volcano Carahuarazo dominates the terrain and is considered to be an apu
141
by those who live in the region today (Figure 4.3). Carahuarazo is also the target of a proposed
mining project which is currently underway in the Chicha-Soras region.
Figure 4.2. Bosque de Piedras. Source: Javier Vargas.
142
Figure 4.3. Carhuarazo. Source: Google Earth.
Like other locales in the western cordillera of the southern Peruvian Andes, the ChichaSoras Valley has two major seasons, a season of high precipitation which runs roughly from
November to April, and a dry season which runs from May to October. Average rainfall is
around 70 mm per month during the rainy season, 3 mm per month during the dry season, and a
total of 500 mm yearly average. This pattern of rainfall creates a climate system which allows for
water to be provided through the natural environment, and evidence for manmade irrigation
canals is scarce in the Chicha-Soras Valley (Kemp et al. 2006). The heaviest rainfall usually
occurs in December, January, February, and March, and there is almost no accumulation of
143
rainfall between May and August.51 During this dry season, diurnal temperature fluctuations shift
drastically, varying between negative degrees Celsius through temperatures as high as 18 or 20.
These fluctuations can prompt frost, freezing, and snow, and thus can act as a challenge to
agricultural production.
Like other regions in Peru, the Chicha-Soras Valley is concomitantly affected by the
weather system known as the Intertropical Convergence Zone (ITCZ) and the El Niño Southern
Oscillation (ENSO) (Kemp et al. 2006). The ITCZ is the equatorial region where northern and
southern tradewinds collide, capable of altering wet and dry seasons periodically. El Niño is an
oceanic-atmospheric event which occurs roughly every 2-7 years in which eastern trade winds
weaken, reducing surface water deflection and the upwelling of the cold water from the
Humboldt Current (NOAA 2019). El Niño effects can cause both heavy rains and flooding,
particularly on the north coast of Peru, while simultaneously resulting in a lack of rainfall in
higher areas. Taken together, these two weather systems can alter the patterns of wet and dry
seasons, creating an unreliable climate within the region (Engel et al. 2014). These weather
systems have not been thoroughly studied in the Chicha-Soras Valley, though initial
paleoecological studies of lakes and mire basins have begun to investigate the local effects in
conjunction with human activities (Kemp et al. 2006).
Valley Divisions and Ecozones
The Chicha-Soras Valley can be divided into the upper, central, and lower sections
(Figure 4.4). The upper valley begins in the Southeast near the modern town of Pampachiri and
51
Since there has been little to no intensive climate research in the Chicha-Soras Valley specifically, the values I use
here are taken from Pampachiri at: https://en.climate-data.org/south-america/peru/apurimac/pampachiri-876217/
144
the Bosque de Piedras, and consists of the highest elevations. Here, the tableland is over 4000
masl in elevation and the valley floors are around 3600 masl. As one moves northeast into the
Central Valley (broadly located between the sites of Taccarampa and Aukimarka), the altitude
decreases, and the valley floor is close to 3000 masl. Finally, the lower valley is located to the
northwest and is defined from the site of Aukimarka to the confluence of the Rio Chicha and the
Rio Pampas. The lower valley has not been extensively studied archaeologically—the two
research surveys completed in the region were bracketed by San Pedro de Larcay in the North,
and Pampachiri in the South, so I will focus on the central and upper valleys in this chapter.
Iglesiachayoq, the site of my excavations, is located in the upper valley.
145
Figure 4.4. Location of the different zones of the Chicha-Soras Valley.
The upper valley consists of the high tableland puna zone (above 3800 masl), where the
soil depths are very thin and cannot be used for agricultural activities. Instead, this flat tableland
is used for animal husbandry, as the grasslands provide pasturage for llama (Camelidae), cow
(Bos taurus), and sheep (Ovis aries) herds. The animals which are native to the region and zone
include camelid, condors (Vultur gryphus), viscachas (Lagidium visacia), foxes (Lycalopex
culpaeus), pumas (Puma concolor), and two types of deer (Hippocamelus) (Meddens 1985: 27).
It is unknown when the Spanish brought cow, sheep, and goat (Capra aegagrus hircus) to the
146
region, but these animals thrive in the puna of the Chicha-Soras Valley (see Chapter 5 for more
discussion).
The steep slopes of the upper valley below the puna are located within the kichwa zone
(3300-3800 masl), which has a slightly warmer climate that allows cultivation of multiple crops.
At the higher elevations of the kichwa (3400-3800 masl), people grow tubers (Solanum
tuberosum) and ulloco (Ullucus tuberosus), while at the lower elevations of this zone, they grow
predominately maize (Zea mays), tarwi (Lupinus mutabilis), and quinoa (Chenopodium quinoa)
(Branch et al. 2007: 7). Because of the steep grades of the slopes in the kichwa zone, the entire
valley is heavily terraced—not all of these terraces were constructed for agriculture, others were
used for leveling out areas for towns or other settlements (Keeley and Meddens 1991: 126).
Agricultural work begins when the rainy season ends, marked by an initial plowing, and
agricultural fields are located on extensive terracing which lines the steep slopes of the valley.
From June until August, the temperature is often below freezing at night, and the communities in
the region begin to produce chuño, or freeze-dried potatoes (Mallco 2013). There is a second
plowing which occurs in September or October. Minga labor, a form of group organization, is
still utilized during planting and harvesting of major crops (Meddens 1985). Fields are utilized
once every six years, and the cultivated fields are irrigated through natural water sources, rather
than manmade canals (Keeley and Meddens 1992). Many of the residents of Chicha utilize the
interior spaces of archaeological structures in order to cultivate their crops. While damaging to
the archaeological remains, these sheltered areas give some protection to the vulnerable crops
during the extremely cold and frosty nights.
147
Human Interventions in Agricultural Productivity
Since the broad tablelands of the Chicha-Soras Valley lack arable land and the kichwa
zone is naturally steep, humans have constructed terraces throughout the valley slopes within
every possible surface (Keeley and Meddens 1991: 126). The earliest phase of terracing is
associated with the first part of the Middle Horizon (550-700 CE) at the site of Yako (Keeley and
Meddens 1991; Meddens 1985). The northern part of the valley, which has steeper slopes and
more dramatic topography, was heavily terraced during the Late Intermediate Period. Terrace
systems can be difficult to date and require excavation, and to this end, Keeley and Meddens
(1992), Kemp (2007), and Branch (2006) have all contributed soil studies in the Chicha-Soras
Valley in the last twenty years. Most notably, excavations of these terrace systems demonstrate
evidence of anthropogenic modification, or manmade soils—they contain bits of ceramic, bone,
and charcoal, though pollen and other botanicals have not preserved (Keeley and Meddens 1991:
128). In all of the terraces, these manmade soils suggests the intentional use of matrices to
promote water drainage. The pH levels of the soils indicate that they were not especially
nutrient-rich. However, Keeley and Meddens suggest that “the amount of land under cultivation
was more than could have been maintained and cultivated by the inhabitants of the place…This
suggests that cultivators were brought in seasonally and that any surplus produced would have
served a population outside the immediate valley” (1991: 128). That is, considering the amount
of land terraced and the overall area of cultivation, the estimated yields would have been
sufficient to either carry a surplus or to support pastoralists or villagers from nearby towns.
Investigations into the terracing of the Chicha-Soras Valley notably found no evidence of
manmade irrigation canals, suggesting that inhabitants of the Valley relied on rainfall for
agricultural purposes (Branch et al 2007).
148
The geography and climate of the western cordillera of the Southern Andes, and more
specifically, the dramatic steep slopes of the Chicha-Soras Valley, has historically created a
number of challenges for human habitation. First, the broad areas of the tableland are the only
naturally flat areas in the Chicha-Soras Valley, yet the shallow soil deposits and high elevations
have prevented any intensive agricultural activities here. Second, though the kichwa zone
contains the soils and elevation necessary for production, the steep slopes required intensive
human intervention in order to be both inhabitable and arable. Third, although wet and dry
seasons are generally distinct and reliable, periodic El Niño events can drastically shift wet and
dry seasons, putting crops at risk. Finally, the rapid and extreme temperature deviations can lead
to frost and snow, again putting crops at risk.
On the other hand, the Chicha-Soras Valley also had a number of factors which attracted
inhabitants over the last three thousand years. First, the broad pampas on either side of the valley
were an ideal location for animal herding and pastoralism, and has even been a theory behind the
Wari takeover of the region (Meddens and Schreiber 2010). Second, the Chicha-Soras River
which divides the valley was an important source of water and could have supplemented water
supplies during unexpectedly dry months. Third, since many of the terraces were constructed
during the Middle Horizon, they would have been a convenient existing feature for later groups,
particularly in the extremely dry Late Intermediate Period (Bauer and Kellett 2010; Engel et al.
2014; Kellett 2010; Kellett et al. 2013; Kurin 2016). Stone is plentiful throughout the valley and
used in all construction, and there are obsidian sources within a few days’ walk from the ChichaSoras Valley (Burger 2006, Meddens and Schreiber 2010). Finally, while freezing temperatures
can be detrimental to crop production, they can also create an ideal atmosphere for freeze-drying
goods such as chuño, which is common in the Chicha-Soras Valley today.
149
The geography and climate of the Chicha-Soras Valley thus presented a number of
benefits and challenges to human settlement, and this geography was an ideal setting for Taki
Onqoy performances. Those who practiced Taki Onqoy were accustomed to the seasons of heavy
and light precipitation, and could have utilized existing agricultural systems to maintain the
maize crops necessary for Taki Onqoy performances. However, the unpredictable shifts in
seasonal rainfall caused by the ICTZ and ENSO could have been an instigating factor in the Taki
Onqoy tenet that the huacas were angry with Andeans due to their neglect. Moreover, the
valley’s dramatic elevation changes and the lack of accessibility between the Chicha-Soras and
other regions would have created an environment of relative independence from Spanish
religious authorities, thus allowing prehispanic religious practices to flourish. Finally, the
Chicha-Soras Valley provided an environment which fostered all of the materials commonly
utilized in Taki Onqoy performances (see Chapter 2).
Prior Archaeological Research
This dissertation builds on previous archaeological research completed by scholars prior
to and after Shining Path limited access to the region. The Chicha-Soras Valley, relatively
inaccessible to governmental authorities in the sixteenth century, remains remote from any large
city, and this isolation has been an additional factor in limiting the archaeological research in the
region. Furthermore, the majority of archaeological inquiry in the region has been limited to
survey—as a result, basic ceramic chronologies or knowledge of past daily life is still unknown.
In 1978, Frank Meddens and Monica Barnes undertook the first scientific survey of part of the
Chicha-Soras Valley, identifying the sites of Qasapampa, Llaku, Mullurmachay, Chicha Qasa,
and Chumado (Mallco 2013). This survey was primarily designed to systematically explore and
150
map evidence of Wari expansion into the region. In 1980, Meddens and Barnes expanded on
their early survey, excavating the archaeological sites of Chicha Qasa (located in Pampachiri,
Apurímac) and Chiqna Jota (located in Larcay, Ayacucho), again with the intent of
demonstrating the Wari influence in the region. In 1981, Barnes continued surveying the ChichaSoras Valley, finding many sites with Wari ceramic (Mallco 2013). In 1982, Frank Meddens and
his wife Beverly Meddens continued excavation at Chiqna Jota, confirming the presence of Wari
at this site.
In 1984, the anthropologist Hector Espinoza, native of Larcay and field assistant to Frank
Meddens, registered nearly 80% of the sites in Larcay (Mallco 2013; Espinoza Martinez 1984).
In 1990, Frank Meddens returned to the region and excavated the cemetery of Charangochayoq,
a known Wari burial cave. In 1999, Meddens returned to Chiqna Jota, finding Wari ceramic in
several different styles. In 2009 and 2010, Rafael Mallco and Luis Angulo documented 28 sites
with architecture, identifying several sites dating from the Middle Horizon to the Late Horizon
(Mallco 2013). Sites recorded by either Meddens or Mallco and mentioned in this chapter are
displayed in Figure 4.5. Since the majority of the research completed in the Chicha-Soras region
was completed by either Frank Meddens or Monica Barnes, settlement pattern data and
archaeological occupations in the region are taken only from a few sources, especially including
Meddens’ dissertation and subsequent publications.
Settlement of the Chicha-Soras Valley was punctuated by two imperial projects—the
Wari and the Inka—between which the local ethnic group known as the Soras thrived during the
Late Intermediate Period. In the time of Taki Onqoy, those who inhabited the region were still
producing and utilizing local ceramics, and their masonry styles and forms also followed local
traditions in the valley. The material signatures of Inka control are very sparse in the valley,
151
perhaps indicating that Taki Onqoy was levied and performed by indigenous actors of Soras
ethnicity throughout the Chicha-Soras. In the subsequent sections, I discuss the habitation
patterns in the Chicha-Soras Valley (as identified by Meddens, Barnes, Mallco, and Espinoza) to
explore the archaeological and cultural-historical background of Taki Onqoy, as it was
performed in this region.
Figure 4.5. Map of all sites documented by Meddens and Mallco within the Chicha-Soras Valley,
n=61.
152
Early Settlement: The Preceramic (?—1800 BCE), Initial (1400—1000 BCE), Early Horizon
(~1000 BCE – 200 BCE), and Early Intermediate Periods (200 BCE – 550 CE)
Meddens documented a Preceramic occupation (?—1800 BCE) in the Chicha-Soras
Valley at two sites primarily inhabited in later periods. At the Middle Horizon site of Chiqna Jota
(Figure 4.6), Meddens studied a road cut where he found a floor covered with obsidian flakes
and a high density of waste. Meddens categorized this occupation as Preceramic since the flakes
showed evidence of human modification but there were no associated ceramic fragments. At the
Late Horizon/Early Colonial site of Iglesiachayoq, Meddens documented basalt and andesite
artifacts located about 1.5m down in a looter’s pit; however in my own work at this site, we did
not find any evidence of occupations predating the Late Horizon. While no specific Preceramic
sites have been documented in the region, the presence of stone tools without ceramic artifacts
indicate that the region was populated prior to 1800 BCE.
153
Figure 4.6. Preceramic and Initial Period sites in the Chicha-Soras Valley, n=6.
The Initial Period (1400-1000 BCE) occupation in the Chicha-Soras Valley is manifested
in four archaeological sites, dated through ceramic material (Figure 4.7). Since there have been
no real targeted ceramic studies in the Chicha-Soras Valley, ceramic sequences are based on
those in the Andahuaylas region. Thus, Initial Period sites were identified by the presence of
Muyu Moqo ceramic pottery, as defined in Grossman’s publication on ceramic sequencing in the
Andahuaylas region (Grossman 1972). The largest sample of this pottery, dating to the Muyu
Moqo C-D phase (1400-1000 BCE) was found at site A 23, located on a hill east of the Soras
154
River. Here, stone tools and lithic debitage were recovered, although no structural remains are
present at the site. Other evidence of Initial Period ceramic production was found at sites A 7
(Anccoccahua) and A 21, located east of the Chicha River. Meddens recovered only three Muyu
Moqo C-D period sherds at Anccoccahua, with the majority of the site dating to the Late
Intermediate Period (Meddens 1985). At site A 21, a single Muyu Moqo C-D sherd was
recovered near a modern structure. Finally, site Lu 16, located west of the Soras River, contained
a collection of Muyu Moqo C-D material, although like Anccoccahua, the majority of the site
(including some 50 to 70 structures) dates to the Late Intermediate Period.
Figure 4.7. Examples of Muyo Moqo style Phase C-D pottery. Photos by Joel Grossman.
While the evidence for Initial Period occupation is sparse and dependent on a small
number of ceramic sherds, the dispersion of these sherds across various parts of the valley
suggests that people populated the region between 1400 and 1000 BCE. Unlike the Initial Period
155
ceremonial centers found in the Casma Valley (Sechin) or the Callejon de Huaylas, no evidence
of these ceremonial centers have been found in the Chicha-Soras to date (Willey 1971).
However, the Muyu Moqo C-D phase ceramics share similarities with Initial Period Hacha
ceramics on the south coast of Peru and have also been found at the site of Waywaka, located on
a hill just south of the city center of Andahuaylas (Grossman 1972).
Like the lack of Initial Period sites in the Chicha-Soras Valley, Meddens did not recover
a single site dating to either the Early Horizon (~1000 BCE – 200 BCE) or the Early
Intermediate Period (200 BCE – 550 CE). Meddens suggests that during this time, the valley was
either uninhabited, or that its material signatures were so faint that they were imperceptible from
surface survey (Meddens 1985:112). More broadly, the evidence for Early Horizon and Early
Intermediate Period sites in the Department of Apurímac is scarce, also possibly caused by a lack
of investigation in this region. Indeed, later work by Espinoza (1984 and 1995) and Meddens
(1998) identified a ceramic style called “Chicha de color rojo” which contained lined painting
and was burnished, similar to the Santa Rosa style in Pampachiri. This ceramic style dates to the
Early Horizon and was found in the southern middens at Chiqnajota (Figure 4.8). Mallco affirms
that these fragments are similar to the South Coast Paracas style, and were found associated with
obsidian points, thus suggesting presence of an Early Horizon occupation (Mallco 2013: 253).
While initial research in the 1980s did not produce evidence of Early Intermediate
occupation, Mallco’s 2010 survey produced evidence of Huarpa-style ceramic in the valley of
Chicha-Soras, at the site of Tororayoq-Huahuerqa, which are cream-colored vessels with red or
black designs (Figure 4.8, Mallco 2013: 255). Mallco argues that these groups arrived in the
Chicha-Soras Valley from northern Peru. The site is designated by circular structures near
agricultural fields, but they are not well-preserved and they are difficult to see. While more
156
recent findings have been described by Meddens’ 1998 article and Mallco’s 2010 survey, there is
still very sparse evidence for a strong Early Horizon or Early Intermediate Period Occupation in
the Chicha-Soras. Given the continuous occupation of the Chicha-Soras Valley from the Middle
Horizon onward, the lack of Early Horizon and Early Intermediate Period sites seems odd (a time
period of 1500 years), and could have been a product of survey strategy rather than a gap in
habitation at that time.
Figure 4.8. Early Horizon and Early Intermediate Period sites in the Chicha-Soras Valley, n=2.
157
The Flourishing Middle Horizon (550 CE - 1000 CE)
In the Middle Horizon (550 – 1000 CE), sites in the Chicha-Soras Valley flourished and
demonstrate a pattern of intense settlement by the Wari. Centered on their primary settlement at
Huari in Ayacucho, the Wari are widely considered to be the first state in the Andes (Jennings
2011, Isbell 1991; Meddens 1985; Menzel 1968; Schreiber 1992). The Wari streamlined
religious traditions to form a cult which emphasized sacrifice and ancestor worship (Cook 2001;
Tung 2008). As the city of Huari rapidly grew (with its height at 700 CE), the subsistence needs
of the Wari expanded, and the Wari transformed the surrounding valleys through canals, terraces,
and storage systems (Jennings 2011; Isbell 1977). Though the Chicha-Soras Valley is
approximately 150km southeast of Huari, research in the region has confirmed that the Wari
expanded into this valley in order to supplement their economic intensification (Meddens 1991;
Meddens and Schreiber 2010).
Meddens’s work in the Chicha-Soras Valley was initially focused on identifying and
studying Middle Horizon sites, so scholarly knowledge of this time period is much more
thorough than of earlier or later periods. Meddens divides the Middle Horizon into four epochs:
the Epoch 1 was from 550-750 CE, Epoch 2 was from 750-900 CE, and Epochs 3-4 were from
900-1000 CE. I utilize Meddens’s chronology here in my discussion of Middle Horizon sites. In
general, during the first Epoch of the Middle Horizon, Wari architecture consists of a broadening
of administrative facilities and structures made up of large, rectangular compounds which are
divided by avenues (Meddens 1985:67). The settlement pattern for the early part of the Middle
Horizon in the Chicha-Soras Valley marks an increase in sites which are generally larger and
contain more structures than earlier occupations.
158
Figure 4.9. Middle Horizon Epoch 1 Sites in the Chicha-Soras Valley, n=4.
One of the earliest Middle Horizon sites in the Chicha-Soras Valley is named Sillacha,
and was identified by Mallco in his 2010 survey of the region (Figure 4.9, Mallco 2013: 255).
Sillacha is two hectares in size and is made up of platforms and Wari-style ceramics. These
fragments consist of bowls, plates, and cups of the styles Ocros, Viñaque, and Huamanga
(Mallco 2013: 239). Mallco attests that this site was used as a habitation area where people
practiced various activities including religious ceremonies, as evidenced by ceramic smashes
which appear to have been intentional. The second Middle Horizon site is called Yako (Llaku),
159
and is also indicative of the first phase of the Middle Horizon in the valley. While only 0.65
hectares in size, there are remains of three structures on the surface (including a typical Wari Dshaped structure (Mallco 2013: 255), with an average diameter of 7-8.5m. The structures contain
niches and have evidence of smooth mud plaster on the interior walls (Meddens 1985: 114).
Yako is heavily terraced, although there are no associated irrigation canals, thus suggesting that
rainfall or run-off from the high altiplano was the major water supply for this particular site.
Meddens sampled these terraces in three different units, finding that “soil had been brought in
and dressed, drainage taken into account, and rubble layers constructed to facilitate
it…containing exclusively Middle Horizon Epoch 1 pottery” (Meddens 1985: 166; Keeley and
Meddens 1992). Meddens also excavated a 2 x 4m unit in one of the structures, finding an
offering of pottery vessels and comingled skeletal remains of at least six people, although the
context of these remains is tenuous because they were clearly redeposited. Mallco identifies two
other sites, Turkuymalla and Soqia, as dating to the first Epoch of the Middle Horizon.
Turkuymalla contains two sectors, one which consists of poorly preserved residential structures,
and the other which is made up of collective tombs deposited beneath large stone blocks.
Similarly, Soqia is another collective tomb—Mallco’s survey and preliminary analysis
confirmed both male and female individuals of all ages, and some with evidence of cranial
deformation (Mallco 2013: 238).
Meddens posits that the central motivation for incorporating the Chicha-Soras Valley into
the Wari Empire was for the Wari to obtain access to ideal environments for camelid herding and
by extension textile production (Meddens 1985: 126). The agriculture and terracing would have
provided camelid herders food and produce, and thus the region was ripe for herding and food
production. In addition, Wari groups also clearly utilized the extensive caves and crevasses in the
160
rocky terrain in order to collectively inter their dead—in this way, the landscape was more than
just useful for sustenance purposes, but also was able to produce and reproduce meaningful
ideological spaces for the Wari.
In the Middle Horizon Epoch 2, the Wari state and its infrastructure fully developed, with
Wari influence spreading on the south coast. Simultaneously, the central coast site of
Pachacamac gained importance (Meddens 1985: 75). The number of Wari administrative centers
also increases—in the southern highlands, Pikillacta was constructed, and was likely a garrison
or a residence for Wari administrators (McEwan 1984). At the time of Meddens’ completed
dissertation, some 22 Middle Horizon Wari sites had been identified in the Ayacucho Valley,
including seven which were characterized by rectangular enclosure compounds. Meddens and
Lumbreras speculate that the motivations behind the Wari expansion may have been in order to
have access to wool and cotton in order to craft specialized cloth and textiles (Lumbreras 2000;
Meddens 1985).
In the Middle Horizon Epoch 2 in the Chicha-Soras Valley, Meddens documented eight
sites including Lu 13, A 16, Lu 6, A 22, Lu 10, Tincoq, Charrangochayoq, and Chiqna Jota, and
Mallco added the funerary site of Sullumachay (Figure 4.10).52 Out of these, Meddens
investigates only Chiqna Jota and Charrangochayoq because the other five were very disturbed.
Chiqna Jota covers nine hectares, but Meddens hypothesizes that including surrounding areas,
the site may have ranged up to sixteen hectares. Meddens identified 200 structures on the
surface, which date from the Middle Horizon Epoch 2 through the Late Horizon (Meddens 1985:
129). Chiqna Jota is located where the Chicha and Pachachaca rivers merge, and is very near the
52
Meddens also documents a site called Tincoq which is outside of the Chicha-Soras Valley proper, yet contains 70
circular structures. Mallco identifies a site called Sullumachay which also dates to MH 2. While I will not discuss
either of these sites directly in the text, they are included in the map.
161
present-day town of Chicha, where I completed my fieldwork. Chiqna Jota is also heavily
terraced, and ceramic evidence affirms that the majority of pottery at the site is of a later Middle
Horizon date, therefore confirming that the site was occupied throughout the duration of the
Middle Horizon (Meddens 1985: 131). Meddens argues that parts of the site appear to have been
an original planned concept, arranged around three sunken plazas (Meddens 1994: 127, Figure
4.11). While part of the site was constructed in the Middle Horizon Epoch 2, the fluorescence of
the site was in the Late Intermediate Period, during which raised pathways were constructed in
order to separate the site into different sectors.
162
Figure 4.10. Middle Horizon Epoch 2 Sites in the Chicha-Soras Valley, n=9.
163
Figure 4.11. Plan map of Chiqna Jota. Map by Meddens (2010).
One other very important Wari site which dates to the Middle Horizon Epoch 2 is a burial
site called Charrangochayoq (Kallpa). Charrangochayoq is a mortuary site located in a high
altitude cave, approximately 3,400 m in elevation, in the bluffs of the upper regions of the valley.
Charrangochayoq is a structure built of dressed fieldstone walls and topped with mud mortar,
constructed within a small cave. The site primarily functioned as a burial location for Wari
164
groups in the Middle Horizon Epoch 2, with several small chambers and evidence of human
bone covering the surface (Figure 4.12 and 4.13). There are two separate cave chambers—one of
the caves is made up of circular chullpas distributed throughout the interior, which are also in a
poor state of conservation (Mallco 2013: 231), while the other cave is filled with the mud mortar
structures, some reaching three levels. In his 2010 survey, Mallco registered the remains of at
least 64 individuals in the first cave, the majority comprised of male adults, with less of the
sample made up of women and children. These human remains were found disarticulated and
comingled. In the second cave, Mallco registered some 234 separate individuals, and
hypothesizes that in total, Charrangochayoq could have had up to 500 separate individuals
interred during the Middle Horizon (Mallco 2013: 233).
Figure 4.12. Outside facade of Charrangochayoq (Photo from Kurin 2016).
165
Figure 4.13. Human remains on the surface at Charrangochayoq (Photo from Mallco 2013).
During the Middle Horizon Epoch 2, Wari groups heavily settled in the Chicha-Soras
Valley, as demonstrated by an increase in number of sites, and a general increase in the terracing
of the valley, suggesting more investment in agriculture. Meddens suggests that the architecture
of Wari sites in the Chicha-Soras would not be considered “typical Wari” for the highlands, as
rectangular enclosures are largely absent and the use of circular structures continues, likely a
vestige of local traditions in the region (Meddens 1985: 150). Beyond Chiqnajota and
Charrangochayoq, other evidence of Middle Horizon Epoch 2 sites in the Chicha-Soras Valley
were in a state of very poor preservation, and thus were not mapped nor studied by the Meddens
survey (Meddens 1985: 145).
166
At the end of the Middle Horizon Epoch 2, the Wari state began to collapse, and areas
which had once been under Wari control regained their independence, resulting in ceramic styles
becoming more localized and regional. The site of Huari is abandoned at the end of this period
(Menzel 1968), and overarching influence wanes. In the Middle Horizon Epochs 3 and 4, there
has been less focus on archaeological projects in the highlands, but coastal Wari settlements
which had been powerful in Middle Horizon Epoch 2 continue to decline. In the Chicha-Soras
Valley, Middle Horizon Epochs 3 and 4 will be treated as one continuous period, because
Meddens was unable to distinguish between the two types of assemblages and sites (Meddens
1985: 85).
Meddens recorded ten sites for the Middle Horizon Epochs 3 and 4 in the Chicha-Soras
Valley (as compared to eight in the Middle Horizon Epoch 2), six having associated structural
remains, and Mallco identified Obrajeria, Sullumachay, Marcalo, and Tororayoq for a grand total
of fourteen sites (Figure 4.14).53 Chiqnajota and Charrangochayoq are continuously occupied or
utilized throughout these two periods, while new sites including Chumayoq and Chichaqasa
appear, each having about 70 structures (Meddens 1985: 164). Located on the east bank of the
Chicha River near Pampachiri, Chumayoq contains some 70 circular structures, situated in
groups of 2-4 around open areas. However, none of the buildings appear to have been utilized for
administrative functions, and the site seems to have been largely residential. Chichaqasa is
southeast of the Chicha River, and was occupied from the Middle Horizon Epochs 3-4
throughout the Late Horizon. The structures at Chichaqasa are circular, rectangular, and
rectangular with rounded corners, much like the variety of structures at Iglesiachayoq. In
53
These sites include: Chiqna Jota, Charrangochayoq, Tincoq, Chumayoq, Chicha Qasa, A 14, A 12, A 13, A 18, A
5 (Meddens 1985: 91).
167
Meddens’ excavations at Chichaqasa, he recovered pottery dating from Middle Horizon Epochs
3 and 4 into the Late Horizon. Furthermore, like other sites in the valley, Chichaqasa is heavily
terraced, and it appears that some of the terraces were constructed during the Middle Horizon
while others dated to the Late Horizon (Meddens 1985:169). There is a large boulder at the site,
which may have been worshipped as a huaca—this idea is supported by the large amounts of
drinking vessels dumped at the site (Meddens 1985:171).
Figure 4.14. Middle Horizon Epochs 3-4 Sites in the Chicha-Soras Valley, n=14.
168
In general, the Middle Horizon in the Chicha-Soras Valley was demarcated by a rise in
Wari habitation, peaking during Epoch 2, and a decline in Epochs 3 and 4. During Epochs 1 and
2, the Wari heavily terraced portions of the valley, possibly to sustain their growing population
in neighboring Ayacucho. These terraces would be continually revamped and reutilized in the
following two chronological periods, and perhaps were responsible for the reinhabitation of the
valley during the Late Horizon. There is evidence for both Wari religious and administrative
settlements in the Chicha-Soras Valley, designated through D-shaped structures and sweeping
sites which housed many residential structures as well (Cook 2001).
Late Intermediate Period (1000 CE - 1400 CE)
Well-documented across the Andes, the Late Intermediate Period was characterized by
political turbulence and de-unification, along with the rise of small, local groups which often
warred with one another, but also could align as allies (Arkush 2011). While other regions
demonstrate a downfall in populated centers in the Late Intermediate Period, the Chicha-Soras
Valley actually shows evidence of continuous development and occupational settlement from the
Middle Horizon Epoch 1 until the end of the Late Intermediate Period. In the Chicha-Soras
Valley, the Late Intermediate Period was dominated by the Soras ethnic group, which had
principal regional centers at Chiqnajota, Qasamarka, and Taccarampa, with smaller settlements
at Chichaqasa, Puka Orqo, Yako, Onqa, Raqayniyuq, and Karpani (Mallco 2013: 255). Many of
these sites will be discussed in more detail in the next section, as the sites are continuously
occupied throughout the Late Horizon and some take on more importance during this period. In
general, Late Intermediate Period Soras sites are characterized by circular domestic residences
distributed without planning or layout (Mallco 2013).
169
Meddens divides the Late Intermediate Period into two phases in the Chicha-Soras
Valley, the first of which is characterized by the Chicha style of pottery (influenced by Wari
styles), and the second in which the Soras style of pottery replaces the Chicha style. The Soras
style is characterized by smudged yellow brown wares, most of it lacking decoration. Meddens
and Mallco both document an increase in the number of archaeological settlements during the
Late Intermediate Period 1, rising to 25 after the 14 collectively documented in the Middle
Horizon Epochs 3 and 4 (Figure 4.15, Meddens 1985: 226).54 In the Late Intermediate Period 2,
Meddens and Mallco document another increase in number of sites, from 25 to 30 (Figure
4.16).55
54
For LIP 1, Meddens names Chiqna Jota, Charrangochayoq, Chicha Qasa, Chumayoq, A 14, A 13, A 5, A 8,
Taccarampa, Lu 7/Lu 8, A 15, A 1, Lu 19, and A 31 while Mallco identifies Obrajeria, Puka Orqo, Raqyniyoq,
Onqa, Pachachaka, Potongo, Sullumachay, Karpani, Laymi Qocha, Pulluhuay, and Marcalo.
55
For LIP 2, Meddens names Chiqna Jota, Charrangochayoq, Qasamarca, Laymi Qocha, Chicha Qasa,
Anccoccahua, Lu 16, A 8, Taccarampa, Lu 7-8, Lu 19, A 31, Auquimarca, A 20, A 2, Lu 9, A 33, A 34, A 35,
Puyka, and Lu 15 while Mallco identifies Puka Orqo, Raqayniyoq, Onqa, Pachachaka, Ptongo, Sullumachay,
Karpani, Pulluhuay, and Marcalo.
170
Figure 4.15. Late Intermediate Period 1 sites in the Chicha-Soras Valley, n=25.
171
Figure 4.16. Late Intermediate Period 2 sites in the Chicha-Soras Valley, n=30.
It is during the second half of the Late Intermediate Period that the site patterning in the
Chicha-Soras region mirrors that in other parts of the highlands, with people shifting to defensive
hilltop settlements such as Aukimarka and Puyka. Aukimarka is a high-altitude (4100 masl)
fortress located on the east bank of the Soras River, and extending over 12 hectares in area. It
consists of over 1000 circular and oval structures, and is protected by a series of defensive walls
encircling the hilltop. Along one ridge of the hill, there are around 30 chullpas (Meddens 1985:
97). There are noticeable pathways along Aukimarka, suggesting that movement may have been
172
planned or dictated throughout the site, and the buildings are made up of fieldstone without
mortar, which is diagnostic of both Middle Horizon and Late Horizon sites. Diagnostic ceramic
found during surface collection dates to the Late Intermediate Period 2 throughout the Late
Horizon, affirming that there may have been a brief Inka presence at Aukimarka as well.
One of the most important developments in the Late Intermediate Period in the ChichaSoras Valley is the reutilization of terrace systems, and the creation of new terracing north of
Larcay in the Middle Valley. This pattern suggests that Late Intermediate Period groups were
threatened by drought or increased population, such that they needed extra food sources.
Moreover, the development of new terraces also is consistent with ice-core records, which assert
the occurrence of a “Little Ice Age” in the Late Intermediate Period. During this time,
agricultural activities specifically related to maize cultivation were levied in the Chicha-Soras
Valley (Branch et al. 2007), perhaps as a means of tempering the unpredictability of the
environment.
In sum, the Late Intermediate Period in the Chicha-Soras Valley seems to have been
characterized by two separate phases, while the overall number of sites steadily increases. In the
first phase, represented by the Chicha ceramic style, residential centers were located throughout
the valley, marked by circular houses. In the second phase, small groups started to build
defendable hilltop fortresses in the upper altitudes of the slopes, conforming to a pattern seen
elsewhere just before the Inka incursion (for the Titicaca Basin see Arkush 2011; for
Andahuaylas see Kurin 2016; for the Colca Valley see Kohut 2016). The reutilization and
establishment of agricultural terraces also lends evidence that agricultural production was
intensifying, perhaps in order to support the population during a time of environmental stress.
More research is needed in order to understand the Late Intermediate Period occupations—this
173
period has lacked investigation thus far in the region. I will now describe the Late Horizon
settlements in the Chicha-Soras Valley in order to set the stage of my excavations at
Iglesiachayoq.
Late Horizon (1400 CE – 1532 CE)
While the overall number of sites greatly declines during the Late Horizon (going from
30 in the Late Intermediate Period 2 down to 18), many of the sites that were important in the
Late Intermediate Period remain occupied or even flourish during the Late Horizon (Figure
4.17). One reason for this decrease in sites could be a result of sampling bias—Meddens only
includes sites that had Inka pottery or polygonal ashlar masonry, so sites which lacked either of
these aspects were not included in the overall total of Late Horizon sites despite the possibility
that they could have been continuously occupied. Conversely, Cieza suggests that the Inka
conquest of the region resulted in a great population loss, which could explain the decline in sites
at this time (Cieza de Leon 1985: 376-377).
174
Figure 4.17. Late Horizon sites in the Chicha-Soras Valley and environs, n=18.
I will now discuss the evidence for Late Horizon sites in the Chicha-Soras Valley in
relatively more detail, in order to create an understanding of the landscape at the time of the
Spanish Conquest. The Middle Horizon Wari center of Chiqnajota, located at the southern end of
the upper valley, grew into a bigger center in the Late Intermediate Period (some 200 houses),
and was reduced in size to 80 houses in the Late Horizon (Meddens 1994, Meddens and
Schreiber 2010: 140). Houses are circular, rectangular with rounded corners, and rectangular,
largely constructed out of the local masonry style with few examples of Inka ashlar masonry.
175
There are two larger Inka buildings constructed at the southern end of the site, one which is
trapezoidal, 12m long and 11m wide with an entrance on the northern façade. Inside are five
trapezoidal niches and three trapezoidal windows (Meddens and Schreiber 2010: 140). The other
Inka structure is round, 10.5m across—Meddens and Schreiber hypothesize that these two
structures may represent the temples of the sun and the moon due to their size and form (2010:
141).
The second major Late Horizon site is Iglesiachayoq, the location of my 2015
excavations. Meddens and Schreiber suggest that Iglesiachayoq (Quechua: “church over there”)
was erected to take over the authority of the valley from Chiqna Jota. The site is located on the
west bank of the Chicha River at 3420 masl, and can be separated into three sectors (Figure
4.18): Sector 1 houses a large rectangular structure known colloquially as the “iglesia,” and was
the Early Colonial Period religious center of the site. Sector 2 is northeast of Sector 1, separated
by a large hill, and is locally known as the site of “Wallpawiri.” While Iglesiachayoq and
Wallpawiri are often separated as two distinct, separate sites (Mallco 2013), their proximity to
one another and the similarities in architecture and masonry style lead me to confirm that both
areas were occupied at the same time, are merely divided by a hill, and are thus part of the same
site. Sector 3 is located on another hill south of Sectors 1 and 2, and appears to be an older
portion of the site with local ovoid architecture and smaller structures. My architectural survey in
2013 and 2014 identified 91 structures at Iglesiachayoq, distributed across all three sectors. In
general, structures are ovoid, rectangular with squared corners, and rectangular with rounded
corners, and many of the houses are grouped around patios or small platforms. The masonry is
typical for the area, consisting of roughly dressed field stone with mud mortar, with classic Inka
176
polygonal ashlar masonry lining the entrances of several of the structures. Many structures
include trapezoidal doorways and niches.
Figure 4.18. Sectorization of Iglesiachayoq.
177
Like Meddens (1985) and Meddens and Schreiber (2010), I identify a major center of the
site in Sector 1, and a sub-center mirroring aspects of Sector 1 in Sector 2. The ceremonial center
of Iglesiachayoq consists of a long, large rectangular structure in the central plaza which
measures 33 x 9m in dimension and is oriented east-west. Some scholars have argued that there
are two doors and several trapezoidal windows, thus leading them to conclude that the structure
is an Inka kallanka (Mallco 2013; Meddens 1985; Meddens and Schreiber 2010). However, my
excavations confirmed that the structure is actually an Early Colonial church, with its altar to the
east and one doorway to the north (Figure 4.19). The windows are not trapezoidal, but
rectangular in shape. There are large gables on the eastern and western ends of the church, which
extend 5.5m above the normal wall height, and contain a circular window near the top. In each of
the interior corners, there are two horizontal wooden posts most likely used to support the walls.
The interior and exterior of the building are covered in mud plaster, with the interior western
wall still containing mortar decorated with red, cream, and black pigment which may have been
a mural. To the east and west of the entrance, I recovered pedestals and fragments of a
benediction bowl, comparable to contemporary churches in the region. The church door faces a
small, low-walled plaza which marks the geographical center of the site. Interrupting the western
wall of this plaza is another rectangular building (4.5m x 3.5m) which likely served as a chapel
in the Early Colonial Period (a similar pattern to other Early Colonial Period sites in Peru, such
as Machu Llacta in the Colca Canyon). To the west of the church is another large building,
rectangular with rounded corners, with dimensions 16m x 11m, oriented north/south. This
structure was clearly constructed during the Late Horizon, as the accessway is framed by Inka
polygonal ashlar masonry and there are 18 trapezoidal niches on the interior walls (Figure 4.20).
178
There are also three windows, one in the center of each wall which does not contain the southfacing doorway.
Figure 4.19. Church altar facing east.
179
Figure 4.20. Cut-stone Inka masonry, photo facing north.
The sub-center at Iglesiachayoq is located in Sector 2 (Wallpawiri), and mirrors the
second structure discussed above (Figure 4.21). Although rectangular with squared corners in
form and measuring 12 x 7.1m, this structure is also oriented north/south, and it contains the
same number of niches (n=18) as the structure mentioned above. There are entrances on both the
north and south ends of this structure which were also lined by Inka cut stone masonry. Wooden
stakes are also apparent in the corners of this structure. While the three structures discussed here
180
contain many similarities in construction style, their uses were different and distinctive. This
leads me to propose that all three were constructed by Inka populations, but in different periods
(the two with niches in the Late Horizon prior to Spanish contact, and the church during the
Early Colonial Period). Iglesiachayoq is a massive regional administrative center in the Late
Horizon, and its sectors and changes in the Early Colonial Period will be discussed in greater
detail in Chapter Six.
Figure 4.21. Parallel structure in the "Wallpawiri" Sector 2 of Iglesiachayoq, facing north.
The next site, Chichaqasa, has also already been discussed in the Middle Horizon Epoch
2 section, but was continuously occupied through the Late Horizon. Here, 70 structures were
identified by Meddens in 1985, also of roughly dressed fieldstone with mud mortar. Local Soras
181
ceramics dating to the Late Intermediate Period and the Late Horizon have been found at the site
(Meddens and Schreiber 2010). The next site is called Qasapampa, is located northwest of
Iglesiachayoq, on the west bank of the Chicha River at 3490 masl (Meddens and Schreiber 2010:
142). The site of Qasamarca includes a hilltop by the same name, where chullpas and circular
structures, as well as defensive walls have been documented and date to the Late Intermediate
Period. Some 70 circular structures have also been recorded at this site, of the same roughly
dressed fieldstone which makes up most of the structures in the region. Small trapezoidal niches
and trapezoidal doorways, as well as Late Horizon Inka pottery has also been found at the site,
affirming its occupation in the Late Horizon.
Laymi was occupied and constructed during the Late Intermediate Period, but gained
importance in the Late Horizon. Located on the western side of the Chicha-Soras Valley at 3445
masl, Laymi is defined through extensive agricultural terracing and the construction of around 60
circular structures of dressed field stone (Meddens and Schreiber 2010: 143). Many of the
structures have trapezoidal entrances and interior niches, and there are also square and
rectangular chullpas at the site. Meddens and Schreiber suggest there is an Early Colonial Period
occupation at the site, although I have not been to the site myself to confirm. Unlike earlier
Middle Horizon sites in the valley there is evidence for irrigation at Laymi, consisting of a large
irrigation canal which initially feeds the terracing above the site and then enters three artificial
basins at lower levels. The irrigation canal is associated with cup-marked stones, which Meddens
suggests had an importance for water management in the Late Horizon (Meddens 2006). At the
southern edge of the site is an Inka ushnu constructed in at least two levels and containing one or
two staircases on the northern side. Ushnus were generally symbols of Inka authority, stepped
platforms which often had astrological alignment and used for performing important religious
182
ceremonies (Meddens et al. 2008). The presence of an ushnu at Laymi suggests that the Inka
found spiritual significance at the site and that it was utilized for ceremonial purposes rather than
solely residential purposes. The irrigation canal passes very near to the ushnu, and it is possible
that the canal and the water it carried may have been used during these ceremonies.
The next Late Horizon site is located in the upper valley, and is simply designated as “A20” by Meddens and Schreiber (2010), northwest of Chichaqasa at 3520 masl. Unlike the other
Late Horizon sites discussed thus far, A-20 only contains one large rectangular structure (22.28 x
5.13m) which is oriented northeast/southwest, and 15 other smaller structures which are also
rectangular or square in nature. This pattern is distinct from local sites in the region, which are
largely comprised of circular structures in the typical Chicha and Soras styles. The large
structure has a trapezoidal doorway on the east wall and a rectangular doorway on the west wall,
with several trapezoidal windows in the walls. As is typical at Iglesiachayoq and other Late
Horizon sites in the region, the masonry style consists of roughly dressed fieldstone with mud
mortar (Meddens and Schreiber 2010: 144). In the large structure, the doorways are offset from
one another, which is rare in other Inka constructions (Protzen 1993: 220)—Meddens and
Schreiber thus suggest that the structure may have had a religious function which emphasized
Inka control.
Moving into the Middle Valley, I will discuss three more sites which date to the Late
Horizon. First, Meddens and Schreiber 2010 suggest that the modern-day town of Soras
(discussed by several sources, see above) was not relocated during the 1570s Toledan reforms,
and possibly overlies the original local Inka center (Meddens and Schreiber 2010: 145). The
church at Soras was constructed to include existing Inka architecture, or Inka populations
incorporated their own style into the original construction of the church. This is evidenced by a
183
northern entrance which is fabricated with Inka polygonal ashlar stones, and is 2.6m wide and
3.52m tall (Figure 4.22). Meddens and Schreiber argue that the doorway “was built by masons
who were experienced in using Inka building technology, almost certainly sometime early in the
Colonial period” (Meddens and Schreiber 2010: 146). Like the construction of the church at
Iglesiachayoq, the church at Soras also appears to be an Early Colonial construction, likely built
by contemporary Inka groups who were experienced in building in both Local and Imperial Inka
styles.
Figure 4.22. Side entrance of the church at Soras. Notice the Inka architecture lining the
accessway. Photo by Kathe Schreiber 2010.
Taccarampa is on the western side of the valley at 3500 masl, and is about 2km south of
Soras. Like many other sites in the Chicha-Soras Valley, the slopes surrounding the site are
heavily terraced, and there are around 70 circular structures of a similar style as other Soras
184
structures in the valley. There are also examples of polygonal ashlar stonework at the site, and
many of the structures have trapezoidal windows, niches, or doorways.
Mallco (2013) identifies several other Late Horizon sites which were developed between
1400 and 1532. Of these, four surround the peak and apu which dominates the physical
landscape, Carhuarazo. These four sites—Kullkuncha, Kusmairo, Doce corrales, and Alala—are
not technically a part of the Chicha-Soras Valley proper, but were instead constructed on the flat
pampa between the Chicha-Soras and the greater Sondondo Valley. These sites are important
because they could have been midway points on the routes traveling between these two valleys.
Soras and Lucanas groups are both identified in Albornoz’s accounts as participators in the Taki
Onqoy movement, so the pampa sites documented by Mallco are important in recognizing the
broader links between these two ethnic polities. Of the four sites, Kullkuncha is the most
substantial and likely was an Inka tambo (way-station). Mallco affirms that Kullkuncha consists
of twelve medium-sized rectangular structures, seven associated corrals, and a kallanka which
measures 24m x 5.5m (2013: 246). Although not thoroughly studied, identification of this Inka
tambo is critical in that it creates a model for interaction between the Sondondo and ChichaSoras Valleys. Kusmairo, Doce Corrales, and Alala are the remaining Inka sites, and all three
seem to be somewhat related to Kullkuncha. Kusmairo consists of rectangular tombs atop a small
hill. Additionally, there is evidence of other types of tombs which appear to have housed
collective burials.56 Doce Corrales irritatingly consists of 14 poorly preserved rectangular
structures associated with obsidian and Inka ceramics. Mallco hypothesizes that this site allowed
56
Though these burial patterns sound similar to the Middle Horizon tombs discussed earlier, all recovered ceramic
fragments were dated to the Late Horizon (Mallco 2013: 249).
185
for commercial exchange between social groups. Finally, Alala is the site nearest to Carhuarazo,
and is made up of several corrals and was likely used by herders.
Discussion of Archaeological Investigation in the Chicha-Soras Valley
While there have been few archaeological projects in the Chicha-Soras Valley, it is clear
from the extensive settlement dating from the Middle Horizon through the Early Colonial Period
that the region was desirable and inhabited by numerous groups throughout its long history. In
general, there is a steady increase in occupation and number of sites beginning with the Middle
Horizon Epoch 1 and growing up until the Late Intermediate Period. The Chicha-Soras Valley
clearly held enticed the Wari in the Middle Horizon, who settled in the region in order to
increase access to agricultural resources, as evidenced by the extensive terracing in the valley
slopes, as well as for access to herding lands at higher altitude. Given the proximity of
Ayacucho, the center of the Wari Empire, the Wari occupation in the Chicha-Soras Valley is not
surprising and was the focus of Meddens’s excavations over the last three decades. Chiqna Jota
was the clear center of Wari occupation in the Chicha-Soras, with a great amount of structures
and evidence for religious activity during the Middle Horizon. With the conquest of the region in
the Late Horizon by the Inka, the number of sites greatly decreases, with several sites
continuously inhabited from the Late Intermediate Period, while the Inka also constructed the
new site of Iglesiachayoq as a regional administrative center from which they could control
Soras populations in the valley.
At the time of the Spanish arrival in the Chicha-Soras Valley (likely sometime during the
1540s), there were two main populations coexisting: the Inka and the Soras. The region was thus
populated by both Andeans and Inkas, two groups which were each implicated in the Taki
186
Onqoy movement. The Inka established their administrative center at Iglesiachayoq, but daily
life appears to have been more or less unaltered from the Late Intermediate Period. While we
have some evidence of Inka conquest in the region, the majority of knowledge of the Late
Horizon and Early Colonial Period is derived from a corpus of Spanish documentation,
consisting of chronicles, ecclesiastical documents, and governmental visitas. In the next section,
I relate how these documents allow us to understand the sociopolitical atmosphere during the rise
of Taki Onqoy, and hypothesize as to why the movement thrived in the Chicha-Soras Valley.
An Ethnohistoric Perspective of the Chicha-Soras Valley
While physical remains of site settlements lend information into imperial processes in the
Chicha-Soras Valley, Spanish documentary sources written in the sixteenth century provide more
nuanced views of the Soras, the local ethnic group which coexisted with the Inka in this region.
There are three main types of sources which I utilize here to create an understanding of the
Soras, their affiliation with the archetype of the Chanka, and the religious extirpation campaigns
which occurred in the region in the late sixteenth century—Spanish chronicles, ecclesiastical
documents, and governmental visitas. The chronicles were written by Spanish travelers who
moved through Peru and recorded Inka histories through oral interviews with Andeans and the
Inka. These types of documents generally describe cultural details, albeit in stereotypical forms
which reflected the political leanings and lineages of their informants. However, they can still
provide evidence of cultural practices and insight into ethnic affiliations, important events, and
mythohistories.
A second critical body of documentary information comes from ecclesiastical documents
drafted primarily by Cristóbal de Albornoz. The benefits and problems of Albornoz’s work are
187
more thoroughly discussed in Chapter 2, yet I will briefly highlight some of his work here
because he offers so many details regarding the specificities of people, practices, and sacred
huacas from both witness accounts and his personal experiences in the Chicha-Soras Valley.
Finally, a third source of Spanish documentation comes from the 1586 visita undertaken by the
Soras corregidor Luis de Monzón. While Monzón’s description of the Chicha-Soras is useful in
that it relates the post-reducción makeup of the valley, it does not provide many details about the
Soras themselves, and it also does not mention the site of Iglesiachayoq (Chicha), suggesting that
the site was not habited at the time of the visita. By placing these three types of sources in
conversation with one another, I aim to consider the totality of evidence discussing who the
Soras were and why Taki Onqoy was practiced in the region.
The Soras and the Chanka
In order to understand the Soras and their role in the Inka and Spanish conquests of the
Chicha-Soras Valley, it is critical to first consider the Chanka and their archetypal status in
Spanish chronicles. For most early Spanish authors, the Chanka represented the dynamic foe of
the Inka, whose defeat predicated the Inka expansion and conquest of the other ethnic groups in
the Andes. While often linked in the chronicles, the Soras and the Chanka are very different
groups: the Chanka were the bellicose “confederation” located in the Apurímac/Andahuaylas
region east of the Río Apurímac. The Chanka were depicted as the major foe of the Inka, whose
defeat at the hands of Pachakuti marked the rise and expansion of the Inka under Pachakuti’s
reign.57 While very near in geographic proximity, the Soras were a distinct ethnic group existing
57
See Bauer et al. 2010 The Chanka: archaeological research in Andahuaylas (Apurimac), Peru for a thorough
discussion of ethnohistoric accounts of the Chanka and the archaeological evidence to support or refute these
accounts.
188
at the same time as the Chanka, with distinctive material culture and architectural styles.
Likewise, in the documentary record, the Soras are often described as a group similar to the
Chanka, but conquered after the Chanka.
Chroniclers suggest that the Chanka rose to power after Wari collapse during the Late
Intermediate Period, and that their home lands were either in Humanga, Huancavelica, or
Apurímac (the three main provinces for Taki Onqoy performances). The Chanka are always cast
as fierce warriors who attempted to conquer the Inka in Cuzco several times before their final
defeat at the hands of Inka Pachakuti Yupanki (Bauer 2010; Cieza de Leon 1985 [1553]; Kellett
2010; Kurin 2012; Meddens and Pomacanchari 2002; Sarmiento 2007 [1572]). In their mythohistory, the Chanka, led by Uscovilca, gathered an infantry to march toward Cuzco and fight the
Inka Viracocha sometime in the early 1400s. Afraid for his life, Viracocha fled from Cuzco,
leaving his son Pachakuti Inka Yupanki to defend the Inka city (Betanzos 1987 [1551]). After a
long, bloody battle, Yupanki was eventually supported by purunraucas, or large stones which
were turned into mythic warriors, and was able to defeat the Chankas (Meddens and
Pomacanchari 2002; Bauer et al. 2010). After defeating the Chankas, Yupanki paraded the heads
of Chanka warriors on pikes and began to expand and consolidate the Inka Empire.
It is beyond the scope of this dissertation to re-litigate tellings of the Chanka-Inka War
and assess how these accounts differ between Spanish chroniclers.58 However, the Chanka are
important to study of the Soras because they are often described as being allies; in fact, early
archaeological studies of the Chanka argued that the Chanka ethnic group may have occupied the
entire regions of Huancavelica, Ayacucho, and Apurímac (Bauer 2010; Carré 1979, 1992;
58
However, see Meddens and Pomancanchari 2002, Kellett 2010, Kurin 2012, Bauer et al. 2010 for thorough
descriptions.
189
Lumbreras 1959, 1974). More recent archaeological research completed by Lucas Kellett and
Brian Bauer has argued for a more limited Chanka territory (only Andahuaylas), asserting that
the surrounding territories were not Chanka, but rather sometimes-affiliated diverse ethnic
groups (Bauer et al. 2010). One of these affiliated groups was the Soras (Meddens and
Pomancanchari 2002).
In the descriptions of the Soras from early chroniclers Juan de Betanzos and Cieza de
León, the Soras are often described in relation to the Chanka, sometimes as having gathered a
large number of soldiers in alliance with this group (Betanzos 1996 [1557]: 85), and other times
depicted as fighting against the Chankas (Monzón 1965 [1586]: 222). Although the debate over
whether or not the Soras and the Chanka were allies will likely never be solved, it is still critical
to understand the geographical proximity between the two groups. The Inka defeat of the Chanka
was cast in all chronicles as the catalyzing event which spurred Inka conquest, and their next
target was usually considered to be the Soras (Betanzos 1987 [1551]; Cieza de Leon 1968
[1551]; Sarmiento 1999 [1572]; Cabello Valboa 1951 [1586]; Garcilaso de la Vega 2000;
Guaman Poma 1980 [1615]; Meddens and Pomancanchari 2002).
190
Figure 4.23. Map showing potential Chanka confederation and relationship to Soras. Map from
Kellett 2010.
The Soras and the Inka
Primary sources of knowledge regarding the Soras as a distinct cultural group as well as
their interactions with the Inka come from accounts of early chroniclers Juan de Betanzos and
Cieza de León, later chroniclers such as Sarmiento de Gamboa, and native authors including
Garcilaso de la Vega and Guaman Poma de Ayala. None of these authors specifically mention
Taki Onqoy, and all are concerned with their own perspectives on the history of Peru and the
Inka conquest of the Andes. The accounts of early chroniclers Juan de Betanzos and Cieza de
191
Leòn affirm that the Soras region was desired by the Inka, likely for its agricultural potential and
its position near a major route between Cuzco and the coast (Betanzos 2015 [1551]: 203-205;
Cieza de Leòn 1984 [1553]: 113-114; Meddens and Schreiber 2010). Speaking specifically about
the Soras people, Cieza de Leòn affirmed that “[the soras and lucanas] bring their signs in order
to be recognized and, like the way they used to do in their past, they were given to looking at
signs and they were great diviners, priding themselves on telling what was going to happen in the
future, in which they raved (as they still do) when they wanted to predict things no creature could
know, because only God knows what is going to happen” (Cieza de Leòn 1984 [1553]: 114).59
Here, Cieza affirms the Soras propensity to utilize divination in daily life, a practice seen as
idolatrous by Spanish authorities. Cieza’s account also describes the land and people of Soras,
offering that “in the headwaters of this river [Rio Pampas] is the province of Soras, a very fertile
and abundant land, populated by bellicose people” (Cieza de Leòn 1985 [1554]: 253). Here,
Cieza’s account supports the archaeological evidence that the Chicha-Soras Valley was desirable
for agricultural activities, and categorizes the Soras as a warlike people.
The Spanish accounts of the Inka conquest of the Soras emphasize a central framework:
after conquering the Chanka, Pachakuti Yupanki decided to militarily expand his empire,
deciding whether to move to the west and attack Soras and Lucanas, or to the north to attack
Huamanga and Xauxa (Cieza de Leòn 1985: 132 [1553]).
Having ordered this with great prudence, the King left there and went to the
province of Andahuaylas, where they gave him solemn reception and he was there
a few days / determining if he would go to conquer the natives of Huamanga and
Xauxa or the Soras and Lucanas. And leaving from there, he went through a
dispopulated area that was going to lead to Soras, who knew of his arrival and
Translated from “[los soras y lucanes] traen sus señales para ser reconocidos y como lo usaron sus pasados y
algunos hubo que se dieron mucho en mirar señales y que fueron grandes agoreros, preciándose de contar lo que
había de suceder de futuro, en lo cual desvariaron, como ahora desvarían cuando quieren decir o pronosticar lo que
criatura ninguna sabe ni alcanza, pues lo que esta por venir solo Dios lo sabe” (Cieza de Leon 1984 [1553: 114),
translation mine.
59
192
united to defend themselves. Inka Yupanki had sent captains with people from
many other places to ally people to his service as discretely as he could and the
Soras sent messengers about not taking arms against him, promising to meet
without causing them injury or harm, but they did not want peace with servitude,
but to fight without losing their freedom. And so, together with others they fought
the battle, which –say those who remember it—was very close and that many died
from both parties, but the field remained for those from Cuzco. (Cieza de León
1985[1553]:138)60
Cieza de Leòn is relating the story of Inka Yupanki fighting with and eventually overtaking the
Soras. He writes that after Pachakuti’s triumph over the Chanka, the Inka ruler sets out to
conquer either the residents of Huamanga and Xauxa, or those of Soras and Lucanas. He decides
to go to Soras, but sends out messengers first in order to tell the Soras that if they do not want to
fight, they could be peacefully subsumed into the Inka Empire. However, the Soras decide to join
with other groups and resist—the battle was “well-fought” and many died on both sides, with the
Inka eventually triumphant.
Juan de Betanzos’ account is similar to that of Cieza’s, and he writes that the Inka found
out that Soras and Lucanas, two very populous provinces, had heard talk that the Inka would be
headed to take the lands, and thus were stockpiling weapons and warriors (Betanzos
1996[1557]:81-83). Betanzos and Cieza both describe the battle between Pachakuti and Soras,
with Betanzos affirming that Pachakuti divided his forces and quickly defeated Soras, while
Cieza writes that the Soras were joined by people from Vilcas and Huamanga, were “well
Translated from: “Ordenado esto con gran prudencia, el rey salió de allí y anduvo hasta la provincial de
Andahuaylas, adonde le fue hecho solene recibimiento y estuvo allí algunos días / determinando si iría a conquistar a
los naturales de Huamanga e Xauxa o a los Soras y Lucanas. Y saliendo de allí, anduvo por un despoblado que iba a
salir a los Soras, los cuales supieron su venida y se juntaron para se defender. Había enviado Inka Yupanki capitanes
con gente por otras partes muchas a que allegasen las gentes a su servicio con la más blandura que pudiesen y a los
soras envió mensajeros sobre que no tomasen armas contra él, prometiendo de los tener en mucho sin les hacer
agravio ni daño, mas no quisieron paz con servidumbre sino guerrear por no perder la libertad. Y así, juntos unos
con otros tuvieron la batalla, la cual – dicen los que tienen delo memoria—que fue muy reñida y que murieron
muchos de ambas partes, mas quedando el campo por los del Cuzco. (Cieza de Leon, Segunda Parte
1985[1553]:138),” translation mine.
60
193
organized, wealthy, and proudly independent” (Stern 1993:20), and that they were able to hold
off the Inka in a high altitude fortress, but still were defeated. From the works of both Cieza and
Betanzos, we see that the Soras were regarded as a fierce peoples willing to stand up against the
Inka. While their accounts differ slightly as to the exact details of the battle, we learn that the
Soras may have allied with other groups and that the Inka were eventually victorious.
Later accounts of the Inka/Soras battle follow a similar chronology as that of Cieza de
León and Betanzos. Sarmiento de Gamboa was the official recorder for Francisco de Toledo in
the 1570s and was sent by royal officials throughout the former Inka Empire with the explicit
goal of portraying the Inka as “unjust” and “cruel” leaders in order to justify the Toledan
reducciones. Nevertheless, we can still use the information given by Sarmiento while keeping in
mind that he had a strong bias and was writing through a lens refracted through several sources.
Like many of the other sources, Sarmiento spends several chapters discussing the Chanka-Inka
war, and then goes on to relate the subsequent conquests of Pachakuti. Discussing Pachakuti’s
conquests into other regions, Sarmiento writes:
Moving toward the Soras, forty leagues from Cuzco, the natives came forth to
resist them, asking [Inka Yupanki] what he was searching for in their lands, [and
telling him] that he should immediately leave them or they would expel him by
force. And a battle was fought over this, and at this time, Inka Yupanki
subjugated two towns of Soras. The first was called Chalco and the other one
Soras. The cinchi of Chalco was named Puxayco, and that of Soras was Guacralla.
Inka Yupanki brought them captive to Cuzco and celebrated his victory over
them. (Sarmiento et al. 2007:126-127)
Here, we can see that Sarmiento’s account mirrors the accounts of both Cieza de Leon and Juan
de Betanzos. Sarmiento’s details are vague, but he confirms that Pachakuti Inka Yupanki defeats
the towns of Soras in a large battle, and then brings back the leaders of the Soras to Cuzco in
order to celebrate his victory.
194
The final group of sources which contribute to the historical and documentary
background of the Chicha-Soras region and the Soras peoples were written by native authors
Garcilaso de la Vega and Guaman Poma in the early seventeenth century. “El Inca” Garcilaso de
la Vega was a writer born in Peru after the Spanish conquest, and was educated in Spain at the
age of 21 (Mazzotti 2008). Garcilaso de la Vega was the son of a Spanish soldier and an Inka
elite woman, born in the first generation after the conquest. Garcilaso provides a unique
perspective on prehispanic history, having been raised in his maternal family as a direct
descendant of the Inka ruler Inka Yupanki. Like Garcilaso de la Vega, Felipe Guaman Poma de
Ayala was also born in Peru after the conquest, the son to a noble family in Lucanas, the
province just south of Soras. Guaman Poma’s famous work is a 1,189-page document entitled El
primer nueva coronica y buen gobierno (1600-1615), which was a scathing critique of Spanish
treatment of the Andean people throughout the Early Colonial Period. Guaman Poma’s work is
essential to our understanding of some of the horrible crimes committed by the ruling Spaniards
in the sixteenth and seventeenth centuries. Guaman Poma also provides nearly 400 drawings
depicting local life and rulers of the Inka period. These drawings are invaluable in their
contributions to knowledge of Spanish and Inka dress, religious ceremonies, and architectural
styles. Furthermore, given his childhood experience in Huamanga, his knowledge of the region is
useful to this project. Finally, Guaman Poma served as the Quechua translator for Cristóbal de
Albornoz during his extirpation of idolatry campaigns, and provides first-hand accounts of this
spiritual crusade.
Garcilaso de la Vega does not mention the Soras explicitly, but instead vaguely suggests
that Inka Yupanki peacefully incorporated the Aymara (east of Soras) (Garcilaso de la Vega
2000; Meddens and Schreiber 2010). Contrary to most other sources of the time, Garcilaso de la
195
Vega writes that Inka Roca, the sixth Inka, was the ruler who conquered the Chankas, rather than
Inka Yupanki, the fifth Inka, to whom this feat is normally attributed. This discrepancy may be
due to Garcilaso’s personal bias due to his Inka descent.
While Garcilaso de la Vega attributes the Soras conquest to the sixth Inka Roka, Guaman
Poma suggests that the Soras were conquered with the Chankas by the seventh Inka, Yahuar
Huacac Inka (Guaman Poma de Ayala 2015: ff.104, 105, 155, 156). Of the Soras region
specifically, Guaman Poma writes that “ in order to sustain the Inka, the deposits of this kingdom
were called colkas, of chuño, muraya, caya, charque, and lana, located in Apcara, Callco, Soras,
and Huánuco Pampa” (Guaman Poma de Ayala 2015: ff.336[338]).61 Thus, upon conquering the
Soras of the region, the Inka invested resources to construct colkas in which to store food and
wool for future use. Furthermore, part of the Qhapac Ñan (Inka royal road) passed through Soras,
and there was a tambo located in the pueblo of Soras (f. 1089 [1099] [1615]). Guaman Poma
suggests that there was a major suspension bridge in the region as well (f. 357 [359] [1615]), the
remains of which might be the colonial suspension bridge located between the towns of Chicha
and Larcay. Additionally, Guaman Poma affirms the elevated status of the Soras in the Inka
Empire, suggesting that they were converted into “honored carriers” of high-status Inka leaders
and their litters, as well as exempted from other types of more manual labor which was normally
levied on conquered groups (Stern 1993: 23).
From the passages cited above of the Inka conquest, it is clear that the descriptions of the
Soras are rather vague and are taken from an imperial perspective. Yet, there are several critical
pieces of evidence which are pertinent to my study of Taki Onqoy. First, the general descriptions
Translated from: “como sustentuaba el Inca, los depositos de este reino llamado colkas…de chuo,
muraya, caya, charque, y lana…en Apcara, Challco, Soras, Huanuco Pampa” (1980 [1616] : f. 336 [338])
61
196
of the Soras region put forth by Cieza de Leon confirm the archaeological evidence of heavy
terracing and Late Intermediate Period/Late Horizon settlement in the Chicha-Soras Valley.
Second, that the Soras were accustomed to divination is an important observation—in the later
extirpation campaigns, the Soras would be punished for these sorts of activities deemed
idolatrous by Spanish priests. Third, most of the Spanish authors verify that the Soras were
conquered after the Chanka, and that they were a population of “bellicose” warriors. They also
all suggest that the Soras often periodically allied or fought against neighboring groups. From
these overlapping descriptions, it is likely that the Soras were part of a Late Intermediate Period
ethnic group which aligns with LIP groups elsewhere who had shifting and strategic
relationships with their neighboring groups (Arkush 2011; Kellett 2010; Kohut 2015; Kurin
2012).
The Soras after Spanish Conquest
Understanding of the Chicha-Soras Valley after Spanish conquest is only known through
Spanish ecclesiastical and administrative records—there has never been a thorough survey
identifying Colonial Era sites in the region, and my research is the first project to investigate
post-conquest themes at the site level. Nevertheless, the detailed accounts of idolatry in the Soras
region provided by Cristóbal de Albornoz and the administrative visita completed by Monzón in
1586 shed light on the transformations in the Chicha-Soras Valley in the second half of the
sixteenth century. In Chapter 2, I outlined the major contributions of Albornoz as his accounts
specifically related to Taki Onqoy. In this section, I focus on Albornoz’s documentation on the
Chicha-Soras region, and especially his work at Iglesiachayoq. At the end of this section, I will
197
overview the limited information derived from Monzón’s visita in order to create a framework
for understanding Taki Onqoy in the region.
Albornoz’s letters to the King discuss his extirpation activities specifically related to Taki
Onqoy, but it is his 1584 Anexos de la informaciòn: Relaciòn de la visita de extirpaciòn de
idolatrias which is the most useful in considering specific locations, people, and huacas related
to Taki Onqoy. This visita, though sent to the King at the relatively late date of 1584,
painstakingly details Albornoz’s activities during his time as a visitador in Huamanga in the
early 1570s. There is some overlap between Albornoz’s ~1570s ecclesiastical visita and
Monzón’s 1584 administrative visita (Tables 4.1 and 4.2). Namely, the towns of Matara, Soras,
Chiclayoc, and Quiji are found listed in both documents. More work must be completed in order
to identify those towns listed as seats of the Taki Onqoy movement and eventually assess how
people were relocated to planned reduction towns in the 1570s and 1580s.
198
1570s Town
Modern
Town
Valley
Chinchera
Chicha
-Soras
Chicha
-Soras
Matara
Chincher
os
Archaeol
ogical
site of
Iglesiach
ayoq
Matara
Marquilla
?
Anansora
Soras
Lurinsora
Soras
Naupa
1586
repartim
iento?
No
Yndios
punished
Destroyed
huacas
17
160
No
12
32
Chicha
-Soras
?
Yes
22
20
No
3
36
Chicha
-Soras
Chicha
-Soras
Yes
12
20
No
57
110
?
?
No
40
112
Llamoca
?
?
No
Yanapillo
?
?
No
15
40
Collata
?
?
No
24
60
Chillcayoc
Chilcayo
c
Chicha
-Soras
Yes
44
22
Quisi
Quiji
Chicha
-Soras
Yes
32
Caiba
?
No
60
Chicha
31
Table 4.1. Accounts of Albornoz's activities in the Soras region.
199
Comments
further east from the valley,
Apurímac
North of Soras, along Rio
Chicha
possibly Morocalla, in Lucanas
Valley
Albornoz got a little lazy here
and started to estimate, seems
like Anansoras was less
numerous than Lurinsoras in
terms of idolatrous practices
Not sure about the number of
people punished
Albornoz differentiates
between maestros and regular
takiongos, with 19 and 25
respectively, annex of Quisi
1586 Town
Atunsoras
Modern Town
Soras
Valley
Chicha-Soras
San Pedro de Larcaya
Larcay
Chicha-Soras
Santa Maria de Matara
Matara
Chicha-Soras
Santiago de Paucara
Paucaray
Chicha-Soras
Anexo of Matara
San Juan de Payco
Palco
Chicha-Soras
Anexo of Matara
San Pedro de Guacana
Huacana
Lucanas
name also appears as Guacatra
San Francisco de
Morocolla
San Salvador de Quiji
Morocolla
Lucanas
Anexo of Guacana
San Salvador de
Quije
Chilcayoc
Chicha-Soras
Corimayo tributary
Chicha-Soras
Anexo of Quiji
Santo Domingo de
Queropampa
Santa Maria Magdalena de
Poma
San Juan de Guayguapata
Querobamba
Lucanas
Chanta tributary; anexo of Quiji
Poma
Lucanas
Calcos/Belen?
San Andres de Ocopa
Ocopa
Between Pampas and
Chicha-Soras
Pampas
Chonta tributary; Anexo of
Guayguapata
No modern town with the old
name
Anexo of Guayguapata
San Pedro de Carguanga
Carhuanca
Pampas
Anexo of Guayguapata
Santa Maria de Chilcayo
Comments
Administrative center of
repartimiento
Anexo of Hatun Soras
Table 4.2. Towns of the 1586 Soras repartimiento; adapted from Meddens and Schreiber 2010.
At the regional level, Albornoz names six towns whose inhabitants were punished for
acting as takiongos in the Soras region whose modern locations have not been identified—
Marquilla, Naupa, Llamoca, Yanapillo, Collata, and Caiba. Conversely, there are seven places
listed in the 1570s ecclesiastical visita which I have connected to modern locations. As discussed
in the Introduction, the location where I completed my field research, Iglesiachayoq, was known
to Albornoz as Chicha. In his visitation of this town, Albornoz writes that he 12 punished Indians
here:
They were punished as ayras taquiongos Hernando Marca Susuma, Mateo Supa,
and Joan Mollo, and another nine native indians of the town of Chicha, subjects of
Joan Hachi, were bound and shorn, and ordered to serve in the construction of the
major church of this town and province of Atunsora, serve in the church as it was
200
necessary, and that three days a week, meet with the priest to be taught in the
rules of our sacred Catholic faith (Albornoz 1990[1584]: 270-271).62
Here, it is clear that Albornoz is relating his direct role in punishing the taquiongos he finds in
Chicha. While he can only name a few of the Indians, he writes that they were forced to work in
the construction of the major church of Chicha, and that three days a week they had to meet with
the priest in order to be taught the rules of Catholicism. Albornoz goes on to name the idols he
destroyed while he was in Chicha:
32 idols of the caciques of the town of Chicha. The guacas of Joan Hacha, leader
of Chicha, were found and discovered and were named Supayco, Guamancapcha,
Quilca, their pacarinas Coalla visa, Malloalca, Tuputa Anpalla, Vilucha, Guaova
Supaico, Guman Capcha, Guamanilea, Anra, Yachi, Guachucata, Ronto, Angas
Guasca, Mollomolla, Qupchiparca, Taxcarima, Apaxallaquio, Caxapi Guazmi,
Ochuylla Poboznos, seven lamayllas, three mamasaras, one Vilca; almost all were
sent by the Visitador [Albornoz] were publicly burned like the others, and those
which were too big to be brought were burned by Juan Cocha Quispe. (Albornoz
1990[1584]: 275).63
Here, Albornoz affirms that he found and destroyed 32 idols, or huacas, from Chicha. He says
that he brought them to Chicha and burned them publicly, as punishment in order to eradicate, or
extirpate, the worship of the huacas in the face of Catholic beliefs.
Both of these passages by Albornoz are useful for this research. In the first passage, we
are given the names of the specific Indians who were punished for their participation in Taki
Onqoy. Most likely with the kuraka, Joan Hachi, and potentially with other names, we have the
Translated from: “Fueron castigados por ayras taguiongos Hernando Marca Susuma, Mateo Supa, y Joan Mollo y
otros nueve yndios mas naturales del pueblo de Chicha, subjetos a don Joan Hachi; fueron acotados y trasquilados, y
que mientras durase la fabricacion de la yglesia mayor de este pueblo y provincial de atunsora, sirviesen en ella en lo
que fue se necesario, y que tres dias en la semana se presentasen ante su cura para que fuesen ensenados en las cosas
de nuestra santa fe católica,” (Albornoz 1990 [1584]: 270-271), translation mine.
63
Translated from: “32 ydolos de los caciques del pueblo de Chicha. Fueron halladas y descubiertas las guacas de
don Juan Hacha, principal del pueblo de Chicha, las quales se llamaban Supayco, Guamancapcha, Quilca, sus
pacarinas Coalla visa, Malloalca, Tuputa Anpalla, Vilucha, Guaoya Supaico, Guman Capcha, Guamanilea, Anra,
Yachi, Guachucata, Ronto, Angas Guacsa, Mollomolla, Qupchiparca, Taxcarima, Apaxallaquio, Caxapi Guazmi,
Ochuylla Poboznos siete llamayllas, tres mamasaras, una Vilca; casi todas ellas por mandado del dicho señor
Visitador fueron traídas y se quemaron públicamente según que las demás, y las que no se truxeron por ser grandes
las fueron a quemar don Joan Cocha Quispe” (Albornoz 1990 [1584]: 275), translation mine.
62
201
possibility of connecting individuals with structural remains. Furthermore, with the names of the
huacas, many are still features on the landscape today, known by the local people in the region.
While some of the huaca names were surely idols or portable objects, many also are toponyms
for specific landscape features, and will be identifiable through survey and conversing with local
elders. Albornoz specifically mentions the pacarinas of the people of Chicha, or the places from
which they descended. These should also be notable places within the landscape, which future
research will identify. Albornoz’s account at Chicha provides a small section of the history of
Taki Onqoy, and gives us invaluable evidence in the long-term history of the Soras people in the
Chicha-Soras.
The final Spanish documentary source was transcribed in 1586 by the corregidor of the
repartimiento of Atunsoras, Luis de Monzòn, and suggests that the Soras were divided into three
parts: Anansoras, Lurinsoras, and Chalcos (Meddens and Schreiber 2010: 132; Monzòn 1586).
Monzòn also documents the 14 towns which made up the Soras repartimiento (Table 4.2). All of
the modern locations of these towns are known today (see Meddens and Schreiber 2010), and so
we are able to derive understanding of the Soras territory after Spanish reducción. Though the
towns listed in the 1586 visita differ from those visited by Albornoz, reduction settlements were
usually near their prehispanic precursors. Utilizing the Albornoz and Monzòn visitas together, it
seems as though Andeans were removed from the locations where Taki Onqoy was taking place,
surely a strategy during Toledo’s reducciόn resettlement program in order to promote religious
faithfulness to Catholicism.
Albornoz and Monzòn offer specific details about Taki Onqoy towns, people, practices,
and punishments, and the resulting post-reducciόn Soras landscape, respectively. In the case of
the Chicha-Soras, the majority of Andean local towns and people were resettled at the
202
reducciones of Atunsoras (today Soras) and San Pedro de Larcaya (today San Pedro de Larcay,
Figure 4.24). If Albornoz’s account is both truthful and accurate, then Albornoz visited the towns
of Chinchera, Chicha, Matara, Marquilla, Anansora and Lurinsora, Yanapillo, Collata,
Chillcayoq, Llamoca, Naupa, Yanapillo, Quisi, Acollacta, and Cayba. This list is noteworthy
because it separates “Anansora” and “Lurinsora” as two towns, possibly indicating that the
traditional Andean ayllu dualism was still vital and present in the 1560s, after 30 years of
conquest. Since many of the punishments for the participating Andean takiongos included
helping to construct churches either in their town or in the regional capital at Soras, it seems that
Albornoz and his campaign, as well as other Spanish groups, had a presence in the region, or at
least were able to coerce church construction projects. Archaeologically, Spanish presence is
difficult to define, since Spanish-occupied structures were often built by local groups, thus
leading to identical architectural masonry styles. Furthermore, churches like the one at
Iglesiachayoq, have been identified as kallankas, or long, rectangular Inka halls (see Mallco
2013; Meddens 1985, Meddens and Schreiber 2010). This misidentification is likely due to the
overinvestment in “prehispanic” occupations at the expense of investigations into the Early
Colonial Period. More in-depth study of specific sites in Chicha-Soras is thus needed to confirm
Spanish presence on-the-ground, in comparison to the extensive documentary record.
203
Figure 4.24. Area of research with modern towns, tourist attractions, and Iglesiachayoq. See
Soras and San Pedro de Larcay in the northern part of the valley.
Recent History of the Chicha-Soras Valley
Although the focus of this dissertation is the revitalization movement Taki Onqoy,
present-day Peru has experienced its fair share of revitalization movements in the last 50 years.
While not directly pertinent to my archaeological data and findings, I find it necessary to discuss
Shining Path and its impact in the Chicha-Soras region—the violence of the movement
threatened all inhabitants of the Chicha-Soras and prevented archaeological research in the here
204
during the 1980s and 1990s. Modern inhabitants of the Chicha-Soras region who participated in
my project speak of both Taki Onqoy and Shining Path movements as markers of identity, often
relating the two movements. In this section, I will first give a brief history of Shining Path’s
ideology and the Civil War in general. Next, I will discuss Shining Path’s indirect influence on
archaeology and my project in particular. Finally, I will discuss the personal narratives of close
friends in the region, in order to give a modern account of how everyday Andeans were
traumatized by the violence of the 1980s and 1990s.
The Partido Comunista del Peru Sendero Luminoso, or Shining Path, was an extreme
left-wing Maoist sect which perceived the socioeconomic situation in the 1970s and 80s as a
“semi-colonial country” which was dominated by imperialism (Gorriti 1999; Taylor 2006).
Influenced by the work of Jose Carlos Mariategui (1928), Shining Path leaders believed that
Peru’s extensive natural resources (mercury and silver in the Early Colonial Period, guano and
nitrate in the second half of the nineteenth century, and agricultural exports in the 1920s) had
created a modern day extractive economy. This extractive economy led to an overdependence on
the state, which controlled capitalist development and bureaucracy, thus alienating peasant
classes and rural Andeans.
Shining Path members paired the writings of Mariategui with the work of Chinese
Communist Party member Mao Zedong to develop their overall ideological viewpoint in the late
1970s (Taylor 2006). Mao was a founder of the Communist Party of China (1927), and is
credited with taking China from imperialism into a modern world power. In particular, Shining
Path leaders were interested in utilizing Mao’s strategies for their own movement in Peru, which
included both a major focus on the peasants and the theory of guerrilla warfare. Specifically, a
major tenet in guerrilla warfare is the use of small, surprise attacks in lieu of the fixed battle lines
205
historically used in warfare. By marrying Mariategui’s dystopian vision with Mao’s strategy for
a communist overthrow of imperial government, Shining Path officials developed their ideology
in the late 1970s, based in the Universidad Nacional de San Cristóbal de Huamanga (UNSCH),
located in Ayacucho.
Led by Abimael Guzmán, a philosophy lecturer at UNSCH, Shining Path began modestly
in the late 1970s, strategically gathering members through Guzman’s lectures and recruiting, and
refining their ideological goals (Taylor 2006). For the first few years after the party’s founding,
they were ignored as a real threat by the Peruvian government and able to remain under the radar
due to their “remote” Ayacucho location, and thus were able to solidify their infrastructure
during this time. Early provincial support of Shining Path tactics was passive—local Andeans
had an ambivalent attitude toward the Peruvian state and generally felt ignored or patronized by
state programs. Since many Shining Path members grew up in these peasant communities before
attending UNSCH or moving to Ayacucho, they were able to utilize their kinship ties to develop
alliances in some of these highland communities. From 1980 through 1982, Shining Path was
more of a utopian group which appeared to be interested in assisting peasant communities in and
around Ayacucho (Taylor 2006).
Between 1982 and 1983, the apparent benevolence of Shining Path in highland
communities gave way to a more direct control and abuse of power, with Shining Path party
members attempting to oust established village leaders and replace them with Party candidates.
Furthermore, Shining Path began to take an active role in community-based activities and
traditions, such as religious practices and civil disagreements. At the same time, increasingly
learning of Shining Path goals and actions in the highlands, the Peruvian military ramped up
occupation of the highlands. While theoretically a strategic move to suppress the Shining Path,
206
the Peruvian national forces were woefully ignorant of Shining Path organizational structures
and membership. Furthermore, the majority of Peruvian national forces who were sent into
Ayacucho were not from the highlands—they did not speak Quechua, could not speak with the
peasants in the villages, and did not understand the cultural practices of the people (Taylor
2006). Since the military could not parse Shining Path forces from local peasants—due to this
lack of knowledge of Shining Path structures and Quechua—their strategy was to wipe out any
potential support of Shining Path. This horrendous tactic resulted in enormous human rights
abuses, the massacres of Andean peasants in numerous highland villages, and a general mistrust
of governmental policy in the highlands.
In this toxic environment of the mid 1980s, peasants in the Chicha-Soras Valley were
caught between Shining Path insurgents and Peruvian national forces. With reason, people in the
Chicha-Soras today have a deep mistrust of outsiders. When speaking with the comuneros there,
many would tell me that during the Shining Path reign, daily activities included posting a young
child at either entrance to the village to watch out for people “foreign” to the village. Upon
sighting a potential threat—Shining Path or Peruvian national force member—the child would
alert the rest of the village, who would hurriedly retreat to nearby caves in order to wait out the
passing of the strangers. Several community members told me that they often waited in the caves
days at a time. One member of my excavation team told me that Shining Path would pass
through the town one day and force town members to give them food and water at gunpoint. A
few days later, Peruvian National forces would enter the town, and beat up or kill those who had
given the Shining Path food. The peasants of the Chicha-Soras region were thus caught in the
middle of a horrible civil war, where the very real threat of death was present on both sides and
local people had no choice but to hide. The realities of the 1980s were traumatic and tragic for
207
the people in the Chicha-Soras—most individuals in the older generations there today lost at
least two or three friends and family members to Shining Path and Peruvian military abuses.
Abimael Guzmán’s capture in 1992 initiated a strong decline in Shining Path activity and
Peruvian National Government retaliation in the highlands. All in all, some 69,000 people were
murdered between 1980 and 2000, with over 75% of those killed speaking only Quechua
(Comision de la Verdad y Reconciliación, 17). The result of the civil war not only caused a deep
mistrust in foreigners and the government in the Chicha-Soras, but also prevented any sort of
intensive archaeological investigation for nearly 30 years. My excavation was the first long-term
project since the 1980s, and there is still much to do in the region in the future. When I arrived in
the Chicha-Soras, I experienced threats and suspicion, for good reason. People were not
comfortable with my presence in the town, nor were they supportive of a government-issued
permit to study one of their most precious archaeological sites. The reality is that the civil war in
the Chicha-Soras Valley created deep scars and greatly influences identity today. The people do
not see themselves as victims, but rather look back both to Taki Onqoy and the Shining Path
years as times of resistance and survival—the people in the Chicha-Soras are fighters. At the end
of my time there in 2015, I had some incredible friends—the generosity and strength of those I
worked with was inspiring.
Perhaps a discussion of modern terrorism is somewhat inappropriate in an archaeology
dissertation; however, I find it difficult to talk about my findings without also discussing the
region today. People in the Chicha-Soras link Taki Onqoy and resistance to Shining Path. They
identify as fierce and independent, and often bring up the two movements in the same
conversations. I will discuss identity more thoroughly in my conclusion, but here I want to end
by saying that a discussion of archaeology and history in the region cannot be completed without
208
a discussion of Shining Path. The archaeological background cannot be divorced from its modern
history, and I wanted to give a history of occupation of the valley, modern movements in the
region, and modern day life in the Chicha-Soras.
Discussion: Patterns in the History and Archaeology of the Chicha-Soras Valley
The Chicha-Soras Valley is similar to other highland Andean valleys, in its ecological
zones, growing seasons, and extensive terracing networks. Habitation in the region began in the
Preceramic or Initial Periods, but increased dramatically in the Middle Horizon, as Wari settled
in the Valley. Drawn perhaps by the agricultural and herding potential of the region, Wari
invested heavily at several sites, the largest being Chiqnajota. During the Late Intermediate
Period, the influence of Wari declined, and was replaced by the local ethnic group Soras,
characterized by their own ceramic style and the construction of rough fieldstone circular
structures. Finally, the Inka conquered the Soras and also heavily invested in the Chicha-Soras,
as demarcated by the Inka occupation at several LIP sites, the construction of the new regional
administration center Iglesiachayoq, and evidence for imperial investment in the ashlar
polygonal masonry at multiple sites. The Chicha-Soras—either because of its economic potential
or its strategic location between Vilcashuaman and Cuzco—thus was a desirable area for
occupation throughout Andean prehistory.
Beyond the archaeological history of the region, however, the broad theme which
characterizes the history of the Chicha-Soras region is that of revitalization. Marked by Soras
and Inka participation in Taki Onqoy in the 1560s, and modern-day Shining Path presence in the
1980s, the people who live in Chicha-Soras today identify with both of these movements and
consider themselves fighters, resistors, and survivors. Andean groups in the Chicha-Soras
209
witnessed countless horrors beginning with the Inka conquest in the 1400s. After this conquest,
the Inka heavily inhabited the Chicha-Soras, likely changing local traditions and basic lifeways.
Following the Inka conquest was the Spanish conquest, which brought massive restructuration of
local religious beliefs and cultural practices, along with disease, death, and onerous tribute
demands. After some 30 years of Spanish presence, the inhabitants of the Chicha-Soras began to
reject all Spanish lifeways; they reverted to worshipping their local huacas and rejected Spanish
food and dress through Taki Onqoy. While this covert movement was developed in order to
intentionally subvert Spanish rule and Catholicism, its discovery by Albornoz and other Spanish
authorities in the late 1560s prompted swift and severe retribution and punishment by Spanish
forces. This included the burning of huacas and forcing local Andeans to construct churches,
among other forms of physical penance.
Partially influenced by the anti-idolatry campaigns, Francisco de Toledo resettled
Andeans in the 1570s, relocating them from their dispersed highland settlements into gridded,
planed towns where they could be more easily supervised (at least in theory). For the inhabitants
of Chicha, this meant moving to either San Pedro de Larcay or Soras. Finally, centuries later,
those who live in the Chicha-Soras were victims of Shining Path massacres and governmental
retaliation, resulting in the area being deemed a red zone for some 25 years. Likewise, in this
time, people in the Chicha-Soras developed a healthy weariness of outsiders and government
mandates alike, thus preventing academic research in the region for some time.
Although the archaeological and historical description I have supplied here largely casts
the people of the Chicha-Soras as victims or passive receivers of horrible policies and various
waves of violence, the story is not entirely one-sided. Local people found ways to manipulate
policy and participate in revitalist movements in order to covertly subvert and influence those
210
who held control. Documentary bias written by “the victors” has dictated the history of the
region from the perspective of Spanish forces (as in the case of Taki Onqoy) or government
officials (as in the case of Shining Path). However, the archaeological research presented here
can provide an intricate slice of material evidence for local practices in the sixteenth century,
supplying an alternate history of colonialism and conquest than those found in the documents. In
the next four chapters, I will present my methodology and archaeological findings regarding
local practices, Taki Onqoy, and Spanish retribution in the sixteenth century.
211
CHAPTER 5
RESEARCH QUESTIONS AND METHODOLOGY
The Taki Onqoy revitalization movement has long been considered a marker of
indigeneity and the subversion of colonial norms, and has heretofore only been studied through
documentary sources. Scholars have predominantly utilized these sources to discuss immaterial
aspects of the movement—what it meant to the people who practiced it, its connection with
broader rebellions in the sixteenth century, and the extent to which it was an exaggeration put
forth by Spanish priests. Though these debates have been critical in understanding the nuances of
both the movement and the documents which describe the movement, they have not attempted to
assess how the movement was performed, and how material media structured the movement.
Thus, although Taki Onqoy was undoubtedly an Andean or Inka bottom-up revitalization, the
previous methods in studying this movement have been grounded in biased documents by
Spanish officials. Though this dissertation is firmly building off of the nuanced work completed
by secondary scholars, a materials-based perspective can offer a new type of evidence which
allows insight into how the movement was practiced at the local level.
In this chapter, I first discuss my research questions and hypotheses, and outline the
archaeological correlates for each scenario. In this section, I draw heavily on my explication of
materiality and Taki Onqoy in Chapter 2. Next, I describe my multi-phase research design and
argue that this design was essential to answering my research questions. Lastly, I relate the
methods and protocols for each phase of my research design.
Research Questions and Hypotheses
212
My dissertation research addressed several questions related to Taki Onqoy and the
material manifestations of its performance at Iglesiachayoq. What was the relationship between
those who practiced Taki Onqoy and their interaction with Spanish cultural and Catholic
materials? Were Spanish responses to Taki Onqoy overtly violent? How did Taki Onqoy affect
mortuary practices? Was Taki Onqoy performed in specific areas (i.e. public or private spaces,
Catholic or local spaces)?
Materiality of Taki Onqoy in the Non-Religious Sphere
One goal of my investigation was to understand the materiality of Taki Onqoy in nonreligious contexts and daily life. By studying how Andeans used Taki Onqoy materials and in
turn, how characteristics of these materials reciprocally influenced people, we can develop a
fuller picture of the movement itself and ask more specific questions than has previous
scholarship. By “non-religious” sphere, I refer to activities occurring primarily within
households in private settings, or public activities taking place outside of households within
plazas. My hypotheses are as follows:
1) If Taki Onqoy practices did not entirely proscribe the use of Spanish things, they would likely
be characterized by the repurposing of Catholic or Spanish objects in new ways or entangled
forms of material culture.
All of the documents which discuss Taki Onqoy practices (Chapter 2) aver that takiongos
rejected Spanish material culture and Catholic religion. Materially, this culture manifested in two
separate realms: the profane and the sacred. At the irreligious level, the material practices would
include avoidance of Spanish agricultural crops (wheat), Old World fauna (cows, horses, sheep,
goats, chickens), and Spanish styles of dress (Albornoz 1990; Álvarez 1998; Molina 2010). In
213
the religious sphere, Spanish culture included rosary beads or crosses, or other personal items
indicative of Catholic religious practice. Archaeologically, if takiongos rejected daily Spanish
cultural practices, I would expect an absence of Old World flora and fauna, no evidence of
Spanish dress, and even the rejection of Spanish personal decorative items such as Nueva Cadiz
beads, glazed or wheel-thrown ceramics, and a lack of Catholic imagery in households. There
could also be evidence of overt destruction of Catholic religious objects such as crosses. Paired
with this absence of Spanish goods, there should be evidence of continued Andean cultural
practices, including the presence of camelid and cuy fauna, maize, and the continued worship of
Andean huacas, as demonstrated through bulk chicha production, animal sacrifices, and
indications of Taki Onqoy ceremonial goods including red pigments. Conversely, if the
inhabitants of Iglesiachayoq did not reject Spanish goods, I would expect to find unbiased
incorporation of Spanish flora, fauna, and religious items throughout domestic units.
In previous archaeological investigations into revitalization movements (namely, I am
drawing on Liebmann’s 2008 work with the seventeenth-century Pueblo Revolt), evidence
suggests that not only do indigenous participants reject colonial cultural materials, but in some
cases they repurposed them in order to transform their power and channel that significance into
local belief systems. By extension, the use of Spanish goods in novel contexts could also indicate
that takiongos were manipulating Spanish items to support huaca worship. Archaeologically,
these practices may include new ceramic styles, the use of Spanish fauna in ritual contexts, or
Catholic objects and imagery in spaces of traditional Andean worship (i.e. ushnus, huacas, etc.).
2) If Spanish responses to Taki Onqoy at Iglesiachayoq were violent, they would be marked by
destruction of huacas in public arenas or physical punishment of takiongos in public spaces.
214
Albornoz affirms that he extirpated idolatry at Iglesiachayoq through the burning of
huacas in the central plaza, the whipping and shaving of individual takiongos in the central plaza,
and by forcing those caught practicing Taki Onqoy to labor at the central church of the site
(Albornoz 1990). Archaeologically, correlates of this hypothesis would include evidence of
burning in the central plaza along with elevated amounts of fragmented stone and possible
indications of burned mummies or other exotic objects. Evidence of physical punishment could
include embodied defense wounds or broken bones in burials recovered at Iglesiachayoq. If there
were no violent responses to Taki Onqoy levied by Spanish officials in the central plaza, I would
expect an archaeological assemblage which resembled other exterior spaces at Iglesiachayoq.
Taki Onqoy in the Religious Sphere
Taki Onqoy practices also had implications for the Catholic and Andean religious
spheres. These spheres included the mortuary treatment of Andean individuals at Iglesiachayoq
and the spatial areas and materiality of Taki Onqoy performances. Spanish Catholic practices at
this time required baptized individuals to be buried in the church, while unbaptized individuals
were to be kept separate from the church (Estenssoro 2003, see Chapter 2). However, Taki
Onqoy leaders demanded religious exclusivity and preached for their followers to avoid churches
and by extension, to inter their dead in traditional places of Andean importance.
3) If the inhabitants at Iglesiachayoq were challenging Catholic traditions in burial practices,
then interments would reflect Andean traditions rather than standardized Catholic norms in terms
of position and burial goods.
If inhabitants of Iglesiachayoq were subverting Catholic burial traditions, then I would
expect evidence of a variety of burial positions in relation to the altar. There could also be
215
evidence of grave goods in interments, and interments could be either primary or secondary.
There may have been post-death interaction with human remains (like in prehispanic Andean or
Inkan traditions), and in the most risky behavior, we may find evidence of burial removal.64
Since there has never been an archaeological investigation into Taki Onqoy, I would be prepared
for a variety of diverse practices in relation to mortuary styles apart from those I have listed here.
Conversely, if Andeans were following Catholic mortuary dictates, then I would expect
individuals interred in extended position, facing the altar, and without grave goods. All
interments would be primary and undisturbed—unbaptized individuals were not to be buried in
the church and baptized individuals had to be buried in the church. Thus, there is no scenario in
which a person originally interred elsewhere should be reburied in the church. Moreover,
Catholic officials were very concerned with Andean ancestor worship. If the inhabitants of
Iglesiachayoq were following Catholic practices, then there should be no evidence of
posthumous interaction with human remains.
4) If Taki Onqoy performances were an explicit revitalization of prehispanic cultural traditions,
then their associated materials would reflect practices common in the region prior to Spanish
invasion.
In Taki Onqoy religious rituals, participants would gather in sanctioned spaces or private
households which had been deliberately cleaned. They would drink lots of chicha, dance, paint
their faces with red pigments, sacrifice rams or cuy, and chant, all practices which were common
prior to Inka and Spanish invasion. The huacas were said to have inhabited individuals, who
would tremble as they spoke through the huaca. Archaeologically, if Taki Onqoy performances
64
As has been documented in later Spanish accounts, for example see Arriaga 1999, Salomon et al. 1991
216
mirrored prehispanic performances, I would expect to find evidence for chicha production and
consumption (large ceramic vessels, feasting wares, and evidence for maize cultivation), interior
or exterior spaces which had been deliberately cleaned, red pigment (either achiote or cinnabar),
and sacrificed camelids and cuyes. In these ritual spaces, there would be no evidence of Spanish
objects. Some of the central practices of the movement—dancing, chanting, and huaca
possession—would leave no archaeological impact, and so these cannot be assessed materially.
Taki Onqoy and Catholic Spaces
My final questions and hypotheses investigate where Taki Onqoy practices were taking
place. While documents name the people and practices of Taki Onqoy, there is little discussion
about where these performances were invoked at the site level. Though Albornoz identifies
different towns and toponyms, he omits critical information about where Taki Onqoy was
practiced at each site. Moreover, Taki Onqoy practices were not universal at any one site at any
one time, and specific site-level locations for these practices provided by witnesses are
contradictory. While some affirm Andeans practiced Taki Onqoy at local huacas such as
mountain peaks, lakes, or rivers, others suggest that these ritual activities took place covertly in
domestic households. Still others affirm that takiongos constructed special outdoor, circular
structures that were specifically designed for these dancing and chanting ceremonies. My final
hypothesis is as follows:
5) If Taki Onqoy was a clandestine movement, its material and spatial signatures will be
uncovered in private areas away from central religious spaces. If there are public spaces of Taki
Onqoy performances, these will be located away from places of Spanish administration and
Catholic practices.
217
This last hypothesis accounts for the possibility of both private and public areas of Taki
Onqoy. Since punishments for Taki Onqoy practices were severe during Albornoz’s
ecclesiastical visita, it is likely that these performances would be enacted in places away from the
direct sight of Catholic authorities. Additionally, since the movement advocated for the complete
rejection of Catholic religion and Spanish practices, the spaces in which the movement were
performed would logically be away from places of Catholic importance. From an archaeological
perspective, if this hypothesis were true, then there would be evidence for Taki Onqoy
performances in areas which were not visible from the site center (the location of the Catholic
Church). Areas near the Catholic Church would be less likely to be associated with Taki Onqoy
and by extension, show no evidence of Taki Onqoy practices.
Research Design
To address these research questions and hypotheses, I implemented a four-phase research
plan which included architectural survey and site mapping, intensive excavation of 19 units
throughout Iglesiachayoq, laboratory analysis of recovered materials (especially ceramic, faunal,
and bioarchaeological study), and spatial analysis of walking paths and visibility. Each of these
phases was instrumental in answering my research questions, and was also critical in learning
more about Iglesiachayoq and the Soras, particularly since most previous excavation in the
Chicha-Soras Valley was at Middle Horizon sites. First, the Google Earth reconnaissance and
on-the-ground architectural survey were both parts of Phase One, and were completed during the
summers of 2013 and 2014. The goal of Phase One was to understand the site layout, site sectors,
and general architectural trends at Iglesiachayoq. Additionally, Phase One mapping helped to
assess occupations at the site and was critical to selecting regions or structures for further
218
excavation. During Phase One, my team and I found evidence—topographic differences,
variations in quality and form of structures—of three separate sectors (Figure 5.1.): Sector 1 (in
red) included the central church and plaza, and a large quadrangular structure with rounded
corners that may have been the kuraka house. Sector 2 was located Northeast of Sector 1 (in
yellow), and Sector 3 was located southwest of Sector 1, and appeared to be the more “local”
area of the site. My team and I also documented 91 structures of three different types: ovoid,
quadrangular with squared corners, and quadrangular with rounded corners. Ovoid structures (n
= 23) were circular or ovoid in shape, with accesses located at various angles across the site.
Quadrangular structures with squared corners (n = 10) were square or rectangular and contained
sharp edges on the exterior corners of the structures. Like the ovoid structures, the accesses for
the quadrangular structures varied in terms of width and angle. The last type of structure was
quadrangular with rounded corners (n = 58), and this type made up the majority of the structures
(Table 5.1).
219
Figure 5.1. Sectorization of Iglesiachayoq.
220
Sector Number
Ovoid
Quadrangular
with Rounded
Corners
16
Total Number of
Structures
5
Quadrangular
with Squared
Corners
9
1
2
2
1
24
27
3
16
0
18
34
Total
23
10
58
91
30
Table 5.1. Distribution of structure forms by Sector.
Drawing on the results from Phase One of this investigation, Phase Two was completed
in 2015 by the Proyecto Arqueológico de Taki Onqoy (PATO) and was marked by large-scale,
targeted excavation at Iglesiachayoq. Unit selection was a mix of judgmental and systematic
sampling. The 19 units were strategically placed to maximize the variety of structures and areas
sampled in order to collect evidence on the range of past activities at Iglesiachayoq within the
different structure forms and different sectors (Figure 5.2). The kuraka household in Sector 1 is
of a different category in scale an elaboration compared to the other domestic structures at the
site. Intensive excavations in this building provided a unique window into the intimacies of
domestic practice, patterns of consumption, and covert ritual of a Taki Onqoy leader. In addition
to this judgmentally-selected location, we followed a stratified, systematic sampling method. The
three sampling strata were defined by formal differences in domestic structures: 1) ovoid, 2)
quadrangular with squared corners, and 3) quadrangular with rounded corners. We excavated a
total of nine structures other than the kuraka house (11.25% of the total count)—which
proportionally represented the percentage of each type of structure at Iglesiachayoq. This
included excavations in two ovoid structures, two quadrangular with squared corner structures,
221
and five quadrangular with rounded corner structures. The selected households were evenly
distributed throughout all three sectors of the site in order to assess and interpret differences
between and within these sectors. In addition to the domestic households, we selected five 2m x
2m units in the patios of a subsample of structures to discern activities occurring outside of the
houses. We also excavated a large 5m x 5m unit in the plaza in order to target practices in the
religious center of the site. Finally, we excavated three 5.5m x 2m units in the church, one at the
west end, one at the east end near the altar, and one in the center of the church.
The selection of unit placement was designed to answer my research questions (Table
5.2). Hypothesis One, which proposed that “if Taki Onqoy practices did not entirely proscribe
the use of the use of Spanish material culture, they would likely be characterized by the
repurposing of Catholic or Spanish objects in new ways or hybrid forms of material culture,” was
aimed at addressing Taki Onqoy practices in domestic spheres. Excavations in the ten domestic
structures of different forms and in different sectors in combination with the five patio units
sought to address Hypothesis One. Hypothesis Two proposed that “if Spanish responses to Taki
Onqoy were overtly violent, they would be marked by destruction of huacas in public arenas or
physical punishment of takiongos in public spaces,” and the central plaza unit was placed to
address this hypothesis. Hypothesis Three posited that “if the inhabitants at Iglesiachayoq were
challenging Catholic traditions in burial practices, then interments would reflect Andean
traditions rather than standardized Catholic norms” and was addressed through excavation in the
central church. Hypothesis Four proposed that “if Taki Onqoy performances were an explicit
revitalization of prehispanic cultural traditions, then their associated materials would reflect
practices common in the region prior to Spanish invasion” can be considered in conjunction with
Hypothesis 5, which is related to the locations of Taki Onqoy performances. Since the site-level
222
locations of Taki Onqoy had never been investigated, all excavation units address both
hypotheses.
Unit
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Sector
1
1
1
1
2
2
2
3
3
3
1
1
1
2
3
1
1
1
1
Size (m)
9.6 x 12
8x4
4.5 x 4
5.5 x 3
4.5 x 3
4.5 x 3
4.5 x 3
4 x 3.5
4 x 2.5
5 x 3.5
2x2
2x2
2x2
2x2
2x2
5x5
5.5 x 2
5.5 x 2
5.5 x 2
Description
Casa de kuraka
Quad w/ squared corners
Ovoid
Quad w/ squared corners
Quad w/ rounded corners
Quad w/ rounded corners
Quad w/ rounded corners
Ovoid
Quad w/ rounded corners
Quad w/ rounded corners
Patio south of Unit 1
Patio west of unit 2
Patio east of Unit 3
Patio north of Unit 6
Patio north of Unit 8
Central plaza
Western church unit
Central church unit
Eastern church altar unit
Table 5.2. Excavated Units at Iglesiachayoq.
223
Hypotheses
addressed
1, 4, 5
1, 4, 5
1, 4, 5
1, 4, 5
1, 4, 5
1, 4, 5
1, 4, 5
1, 4, 5
1, 4, 5
1, 4, 5
1, 2, 4, 5
1, 2, 4, 5
1, 2, 4, 5
1, 2, 4, 5
1, 2, 4, 5
2, 4, 5
3, 4, 5
3, 4, 5
3, 4, 5
Figure 5.2. Excavation Units at Iglesiachayoq.
Phase Three of this investigation consisted of macroanalysis of ceramic, lithics, faunal
remains, and bioarchaeological specimens in Peru, as well as the exportation of ceramic and
224
biological remains for microscopic and isotopic analyses. Phase Three thus consists of all
completed in-field and laboratory analysis. Finally, Phase Four utilized GIS and spatial analysis
to model walking times and intervisibility at Iglesiachayoq. Phase Four was also aimed at
addressing Hypotheses Four and Five, as it explicitly considered spaces of performance at
Iglesiachayoq. This mixed-methods approach which included site-level survey, site-wide
excavation, and contextual analysis, worked together in order to provide a comprehensive
understanding of daily life and mortuary practices at Iglesiachayoq in the mid-sixteenth century.
The methodology used in this research was also highly flexible, and timelines were modified or
postponed in order to work in accordance with the needs and demands of the surrounding
communities, Chicha and Larcay. By excavating in all three sectors of the site, the investigation
questions how space was utilized by different groups, and how the production or altering of
space produced and structured hegemonic Spanish control. Specifically, the mixed-methods
approach was designed to consider possible locations for Taki Onqoy activities.
In general, Iglesiachayoq is an ideal site for excavation. There is only one major
occupation at the site, which dates to the Inka and Early Colonial Period (1450-1572). Since
there is a maximum of 150 years of occupation at Iglesiachayoq, unit deposits were shallow,
averaging 40cm of cultural material before arriving at sterile soil. The shallow depth of cultural
deposits was affected by modern farming practices across the site, however. The present-day
town of Chicha is located a mere 1.5 km from Iglesiachayoq, and many of the people who live in
Chicha today use structures and plaza spaces as chakras, or agricultural fields (Figure 5.3). This
modern-day disruption of soil stratigraphy may have affected findings at specific units during the
excavation.
225
Figure 5.3. View of Iglesiachayoq looking north. Notice the tilled fields in the foreground south
of the church. Photo by Monica Barnes.
Methodology
Phase One: Preliminary Mapping and Architectural Survey
As the first archaeological project to undertake extensive excavation at Iglesiachayoq,
and the first excavation project located in the Chicha-Soras Valley in over thirty years, my team
and I first generated a detailed, accurate basemap before creating an excavation plan. In 2013, I
first visited Iglesiachayoq, and returned that summer in order to become familiar with the site. In
between trips, I located the site on Google Earth and saved satellite imagery of the region and the
structures (Figure 5.4).
226
Figure 5.4. Example of Google Earth imagery of Iglesiachayoq. Church and kuraka house in the
center, surrounded by heavily-terraced hillsides.
Fortunately, the imagery in this region is very clear, and we were able to make out most of the
structures from the aerial photos. I saved screen shots of these aerial photos and imported them
into ArcGIS 10.1, where I then georeferenced them using visible places on the landscape in order
to anchor the structures in real space. These technological advances have made vast
improvements on mapping capabilities from the 1980s, as indicated by the sketch map drafted by
Barnes (Figure 5.5). In the rest of the 2013 season, I visited Iglesiachayoq two other times—I
used these visits to take GPS points using a Garmin Trimble, so that I could ground-truth the
photo imagery from Google Earth. During these other visits, I took notes about the types of
architecture and masonry, and I recorded potential variables to measure in the next phase of
fieldwork. Iglesiachayoq is an ideal location for architectural survey using both aerial imagery
and on-the-ground prospection methods. For the majority of the structures, at least a portion of
the walls are fully intact, allowing us to intuit the original height of the structures. For all
227
structures, the base architecture was intact which provided the original size and shape of the
structures. Additionally, many of the structures lacked only their thatched roofs, such that
Iglesiachayoq likely looks similar to its original appearance during the fifteenth and sixteenth
centuries of occupation.
Figure 5.5. Georeferenced Barnes map--while spatial relationships are more or less rendered
accurately, the map itself is skewed. Structures mapped by PATO are in red.
In 2014, I began an in-depth architectural survey of Iglesiachayoq. I imported the
georeferenced imagery onto an Apple iPad using the program GISPro, made by Garafa. GISPro
is a tablet-based mapping program in which one can import and export shapefiles and imagery,
as well as record points, lines, or polygons in the actual environment. After importing the Google
Earth imagery into GISPro, I then visited Iglesiachayoq in 2014 and completed a detailed map
which recorded 53 quantitative and qualitative attributes for every structure, 12 attributes for
228
every wall, and 6 attributes for every accessway. By far, the most important of these feature
classes was the structure feature class, which recorded attributes for 91 structures at
Iglesiachayoq. Each structure was given an EA number to differentiate it, followed by the
recording of attributes, a subset which included a general description, preservation percentages
and descriptions, form of structure, form of interior and exterior corners, special construction on
the floor, room divisions, estimated time period, function and design, modern use, state of the
surface, masonry style, length, width, diameter, height of walls, width and height of access,
masonry of the access, orientation of the structure, number of niches, presence or absence of
stucco, number of windows and description, other details, and photos for every structure. While
this is not the full list of attributes recorded, it does provide a feel for the sort of information we
were collecting for each structure (Figure 5.6).
229
Figure 5.6. Work space of map in GIS Pro, attributes on the left.
For wall attributes, each wall segment was also given an EA number to differentiate it,
followed by the collecting of various attributes including a general description, type of wall,
associated structures or other walls, level of preservation, width and height of the wall, and the
type of masonry. Photos were taken for all walls as well. Finally, the architectural survey
assigned EA numbers and collected attribute data for accesses in walls, structures, or patios.
Some of these attributes include type of access, description, width, and photos. Iglesiachayoq
does not follow a site plan, but rather was constructed following the topographic contours of the
230
valley. As such, accessways identified in walls or between structures were less definitive than
those at other planned sites.65 The architectural survey phase of this project was designed to
gather the most data in an efficient manner, and specifically to help select structures for
excavation during Phase Two. By collecting architectural data on all structures, walls, and
accesses at Iglesiachayoq, this information facilitates the potential for advanced spatial analysis
including Spatial Network Analysis, Least Cost Paths, and Viewshed Analyses at the site-level.
Furthermore, with quantitative data on all structures at the site, future statistical analysis is
possible in order to statistically differentiate size and form between and among sectors of the
site. Not only does this data facilitate more advanced spatial analyses and allow for strategic
excavation design, but it also allows me to make hypotheses about wealth and status differences,
cultural distinctions, and temporal use at Iglesiachayoq.
All data and photos were compiled into a database using Microsoft Access, and were
later joined with excavation and ceramic analysis data. Shapefiles for structures, walls, and
accesses were imported into ArcGIS 10.3.1, where building polygons were corrected to account
for a point-snapping flaw in GISPro. The corrected map was then imported back into GISPro for
use in the field during excavation. Masonry styles were more or less uniform throughout the site.
In general, all structures were built out of roughly-dressed fieldstone with a loose mortar holding
the stones together, and three of these structures had stucco on both the exterior and interior. Few
structures at the site (n = 9) contain fine Inka cut-masonry, with some structures (n = 8)
containing niches. Doorways were uniformly trapezoidal, wider at the base and narrower at the
65
For example, While apprenticing with my adviser, Dr. Steve Wernke with mapping and architectural survey at the
reduccion town of Mawchu Llacta, the high pirca walls and gridded layout created the need for structured
accessways and rigid pathways for movement as compared to the agglutinated patio layout seen at Iglesiachayoq
(Wernke 2017).
231
top. While this will be discussed in more detail in Chapter 6, we differentiated between three
types of structure forms.
In 2015, PATO was able to borrow and utilize a Sokkia SET630R Total Station to
ground-truth images in ArcGIS 10.3.1. Since the basemap was generated using aerial imagery,
we took anchor points using the Total Station in order to ground the imagery in actual space.
Using a Garmin Trimble, three site datum were placed, one in each sector of the site. A backsight
was recorded at a distance of 500m away from the overall site datum, which is located at
656928E and 8427826S. Using these three site datum, we mapped the corners and centers of all
structures on site which were visible from these datum. We also mapped the four corners of
every unit excavated, as well as the center points of these units. Finally, we imported the
coordinate data into ArcGIS 10.3.1 and corrected the map, with structures moving an average of
3m between the aerial photography and the Total Station points. Additionally, we plotted the
limits of the site using the Trimble in order to generate an accurate polygon of the extents of
Iglesiachayoq. The extents of this polygon were defined by a lack of structures and surface
ceramic.
Phase Two: Excavation at Iglesiachayoq
The excavation strategy, which took place over three months of fieldwork, was guided by
a mix of judgmental and systematic sampling. Units were strategically placed to maximize the
variety of structures and areas sampled in order to collect evidence on a maximum range of past
activities at Iglesiachayoq. Unit locations were selected after considering visible surface features
and architectural patterns, activity areas, aerial imagery, and practical considerations like
location and degree of brush coverage in specific areas. A total of 19 units were excavated,
232
ranging from 2m x 2m patio units to an excavation unit of 12m x 9.6m, with cultural levels
reaching an average depth of 40cm below the surface. In total, we excavated some 284.1 m2
surface area, and the excavations confirmed very shallow occupation contexts indicative of the
only ~150 year occupation of the site.
A local grid (anchored to local datum) divided each unit in order to provide more control
and horizontal spatial reference for collected archaeological materials. In most cases, units in
structures were divided into four quads, but larger units were subdivided more times in order to
create quads of smaller sizes to maintain control. All excavation contexts were conducted in
accordance with known, established techniques and methods for standard archaeological
excavation (Barker 1993; Collis 2004). All excavations followed the Harris method, utilizing the
“locus” as the unit of analysis within each excavated context. The locus system provides more
flexibility for the excavator—any context, special find, natural level, arbitrary level, burial, etc.
can be designated as a new locus, thus definitively separating findings. As Tripcevich and
Wernke (2010) write, the locus system provides “greater analytical flexibility in the field than a
level and feature system; it enables excavation of culturally-defined contexts without forcing
them into a binary ‘level’ or ‘feature’ classes as minimal units of provenience; and it facilitates
database design by providing a single minimal unit of provenience with a unique identification
number” (Tripcevich and Wernke 2010: 386). I assigned each unit leader 50 locus numbers
which they used in consecutive order for different loci. These locus numbers were never
duplicated across units, so each locus has only one possible context. For example, locus 207 was
an area of intense burning filled with bone fragments and carbon; locus 207 only corresponds to
this soil matrix and these artifacts. Locus numbers began with 001 and the last recorded locus
233
was 1012, but not all numbers in between were utilized. In total, 141 separate contexts were
excavated at Iglesiachayoq.
Contexts were excavated by pickaxe, trowel, and brush. In general, the surface areas of
each unit were covered with extremely thick vegetation and grass, so machetes and pickaxes
were need to clear these top levels. In Unit 2 and Unit 10, a large, thick tree had grown in the
periphery of either unit. In both cases, the tree was too difficult to remove so we left it in the grid
and excavated around it to the best of our abilities. Small, delicate, or fragile materials were
excavated using wooden sticks, metal dental tools, small brushes, and tongs so as not to damage
or disturb the data. All soil, with the exception of one 2.5 liter sample per locus, was sieved
through ¼ inch screens. The 2.5 liter sample per locus was collected as a matrix sample;
however, these soil samples were not processed due to a lack of time. All recovered materials
were separately bagged and labeled with the provenience string (site, unit, quad, locus), date, and
type of material (e.g. ceramic, lithic, bone). Artifact bags were catalogued every night in the lab
and placed in boxes which were separated by unit. In general, preservation at the site is rather
poor, with much of the collected bone in very bad shape. Organic materials were collected in situ
for potential radiocarbon dating, carefully wrapped in aluminum foil to prevent contamination.
Bones which were particularly fragile were also wrapped in aluminum foil in order to preserve
the integrity and shape of the organic material. After full excavation of each unit, units were
refilled with excavated soil in a timely manner. Metals were wrapped in paper and stored
separately, as instructed by the project supervisor from the Ministry of Culture in Ayacucho.
Sector 1 Units
234
Household units were placed to encompass the majority of the interior space of each
household, and small extensions were added to some units in order to define base architecture of
the structures. The rest of this section will describe each unit’s placement, size, and extensions in
order to justify my excavation methodology (Figure 5.6, and see Appendix E for detailed unit
summaries).
Figure 5.7. Units, sizes, and UTM coordinates for Sector 1 excavations.
235
Located in the kuraka house in the religious center of the site in Sector 1, Unit 1 was 12m
x 9.6m placed in the center of the structure and then divided into 12 3m x 3.2m quads.66 For all
units, quads were numbered from northwest to southeast in a linear fashion, mirrored on how
English is read. Prior to excavation of each unit, the team and workers cleared the shrubs and
trees from the surface, and collected any ceramic which was found on the surface. Because of the
massive size of Unit 1, we made a strategic decision to excavate quads one, three, five, seven,
nine and twelve (see Figure 5.8). In quads one, three, seven, and ten, we extended units to the
structure wall to define the base architecture. The goal of this unit was to see differences in
artifact type and overall quality, as this structures is much larger and has evidence of more
investment than most other structures at the site. Some have even argued that this structure was
an Inka palace (Mallco, pers comm. 2015), although the artifacts found in this unit do not
support this hypothesis. Unit 1 is located in the religious center of the site, so its placement was
also designed to reflect occupation changes, or evidence of Spanish presence in this area.
Additionally, since the kuraka of Iglesiachayoq was implicated in the Taki Onqoy movement,
this unit was selected in order to study the intricacies of the local leader’s daily practices.
66
Unit 1 was placed in EA 2, the identifying structure number for the kuraka house.
236
1
2
3
2
4
7
1
0
11
5
6
8
9
12
Figure 5.8. Unit 1 with quad subdivisions.
Unit 2 was an 8m x 4m unit located in Sector 1, and placed to cover the maximum
amount of domestic space in structure EA 30. Along with two other quadrangular structures with
squared corners (EA 29 and EA 32), these three structures make up a small patio group which is
separated from other patio groups at Iglesiachayoq. All are quadrangular with squared corners,
and of a much greater size and quality than most other structures at the site. Furthermore, all
three structures are surrounded by a small restraining wall, indicating that access may have been
restricted into this patio group. The object of this unit was to understand who was living in this
area, if it was separated for a particular reason, and if there was a noticeable difference in the
237
function of the houses in this area. In addition to the 8m x 4m unit, divided into eight 2m x 1m
quads, we extended the unit in quad 2 by an average of 52cm from the northern limit of the unit
to the wall in order to define base architecture in the structure.
Unit 3 was also located in Sector 1, and was a 5m x 4.5m unit placed to maximize the
space in the circular structure number EA 47. The goal of this unit was to identify if activities
performed in circular structures were qualitatively different than those in other types of
structures. Cristóbal de Albornoz (1990 [1577]:191) writes that during Taki Onqoy, takiongos
constructed circular enclosures to perform the dances associated with the movement. This unit
was subdivided into four quads, 2.5 m x 2.25m in size. South of quad 4, we extended the unit
50cm to the southern wall of structure EA 47 to identify base architecture.
Unit 4 was also located in Sector 1, and was a 3m x 2.75m unit placed in structure EA 4.
Structure 4 is located alongside of the church and the enormous kuraka house, so I hypothesized
that structure 4 would contain materials to support the functioning of either or both buildings, or
could have been used to support activities in the central plaza. This unit was placed here in order
to investigate possible feasting or other food preparation tasks in a house alongside the religious
center of the site. Extensions in this unit include an average 40cm extension from the western
limit of quads one and three to the wall to define architecture. Unit 4 suffered from extensive
wall collapse to the interior of the structure; however, these large wall blocks actually protected
the archaeological contexts and we had much better preservation in this unit than in most other
units.
Unit 11 was one of the 2m x 2m patio units, this one located south of the accessway of
the kuraka house by one meter, placed to observe any activities that were occurring in the
exterior spaces of houses at Iglesiachayoq. In his survey, Meddens put in a test unit in the
238
entrance of the structure (Figure 5.9). Patio units were not divided into grids, nor did they
contain any extensions. Unfortunately, in many of the patio spaces, including unit 11, the unit
was placed in an area where cultivation or agricultural production had occurred in the past—the
frequency of this problem was astounding at Iglesiachayoq. The majority of spaces in between
structures have been used for planting and growing crops or as pastureland for animal husbandry
at some point in the recent past.
Figure 5.9. Image of the test pit placed at the entrance to the kuraka house facing north. Photo by
Frank Meddens.
Unit 12 was also one of the 2m x 2m patio units, this one located one meter west of the
access to structure 30 and Unit 2. This unit was also placed in order to understand activities
occurring outside of the houses. Unit 2 was intentionally placed within the patio group located
here, to see if any special activities or practices were taking place in the enclosed, yet shared
external space associated with these structures.
239
Unit 13 was the last of the 2m x 2m patio units in Sector 1, this one located northwest of
the access, slightly offset one meter from the doorway in structure 47. This unit was placed in the
exterior space of an ovoid structure, again to see if there were differences in the outdoor
activities between sectors or types of buildings.
Unit 16 was a 5m x 5m large-scale excavation unit located in the central plaza of
Iglesiachayoq. This unit was in the center of the plaza, north of the church and east of the kuraka
house. According to Albornoz, punishments for those who practiced Taki Onqoy included
destroying and burning portable huacas and idols in the central plaza of several of the towns he
visited. Since Albornoz implicates himself in this activity at Iglesiachayoq, this unit was placed
here in order to test Albornoz’ claim. This is the only large-scale unit in an open space at
Iglesiachayoq.
Units 17, 18, and 19 were 2m x 5.5 units located in three places in the church. Unit 17
was located at the western limit of the church, with its western boundary abutting the western
wall. A 0.50m x 1m extension was added to the northeast quad of Unit 17 in order to follow a
burial that was uncovered. Unit 18 was located in the center of the church, approximately nine
meters east of Unit 17 and nine meters west of Unit 19. Unit 18 was placed to encompass the
location of the benediction bowl found 3.5m east of the doorway in the church. Finally, Unit 19
was located abutting the three-stepped altar at the eastern limit of the church. Taken together,
these three units were strategically placed in order to understand potential Taki Onqoy practices
related to burial, as well as status and health differences between people buried in all areas of the
church. Furthermore, the units in the various places in the church facilitated analysis between
different types of burials (e.g. flexed vs. extended, primary vs. secondary). Above all, going into
the excavation of the church, we were not sure that the structure could be positively identified as
240
a church. As discussed elsewhere in this dissertation, several investigators have identified the
structure as a kallanka, or Inka long hall, in previous publications (Mallco 2013, Meddens and
Schreiber 2010). Thus, an overarching goal when excavating in this structure was to confirm or
deny the hypothesis that the structure was indeed the early church mentioned by Albornoz in his
description of his activities at Iglesiachayoq (1584).
Church units were excavated in a very precise manner. Every individual burial was
separated as a unique locus in order to preserve all of the bones and materials associated with
that burial together under one context. Thus, while a natural level may have contained several
burials, instead of grouping these burials under one locus, they were each assigned an individual
locus. Here, then, the locus was more of a special feature rather than a natural or arbitrary level.
The bones found in these burials were generally in very poor shape as a result of soil acidity—
bones which were actively disintegrating were wrapped in aluminum foil to preserve the
integrity of the bone. These burials were excavated with wooden dowels, some dental picks, and
very small brushes to expose the extents of the burials before removal. In several of the burials,
humans were interred with one or more metal artifacts. These metals were carefully
photographed and removed, and then placed in a sheet of clean paper and bagged so as to prevent
rust and other damage.
Because Sector 1 was the religious and administrative center of the site, we chose to
excavate the most units here. Sectors 2 and 3 have less units, but are equally important in
understanding differences between sectors and among sectors.
Sector 2 Units
241
There were four units excavated in Sector 2, which is the northeastern part of the site
located on a hill (Figure 5.10). This sector of the site is locally referred to as “Wallpa Wiri,” and
is occasionally considered as a separate site from Iglesiachayoq (Mallco 2013). However, given
the proximity of this sector to the other sectors of the site as well as the similarities in ceramic
styles and architectural styles as other parts of Iglesiachayoq, I affirm that it is simply a separate
sector of the same site.
Figure 5.10. Units, sizes, and UTM coordinates for Sector 2 excavations.
242
Unit 5 was located in the uppermost tier to the southeast of Units 6 and 7, and was 4.5m x
3m. Overall, the structures in Sector 2 seem to be much more standardized in terms of size and
form of the house. Perhaps they were all constructed over the same period of time. Unit 5 was
placed to encompass the domestic interior of structure EA 175. Also quadrangular with rounded
corners, this unit was randomly placed in Sector 2 to gain knowledge as to the practices in this
portion of the site.
The structures in Sector 2 were dispersed among three tiers or terraces of land. The
second unit in Sector 2, Unit 6, was located in the middle of one of these tiers, and was also 4.5m
x 3m, designed to encompass the domestic interior of structure EA 168. This structure was also
quadrangular with rounded corners, and was placed in order to gain a better understanding of
practices happening in Sector 2.
Unit 7 was located at the northern end of Sector 2, and had a commanding view of the
rest of the valley and potential movement in between the valleys. This unit was 4.5m x 3m and
located to encompass the maximum domestic space in Structure EA 127 (quadrangular with
rounded corners). The goal of this unit was to understand differences between use of domestic
spaces in Sector 2 as compared to Sectors 1 and 3. Furthermore, given the distinctive location of
Structure EA 127 and its strategic construction as a possible “lookout” house or guard post, I
hypothesized that the artifacts found in this house might reflect different practices occurring here
than in other houses at the site.
The last unit in Sector 2 was Unit 14, which was located slightly northwest (one meter) of
structure EA 168, and was a 2m x 2m patio unit. Unfortunately, like many of the patio units at
Iglesiachayoq, this unit was directly located on an agricultural field, so much of the data may
243
have been disrupted or disturbed. Unit 14 was placed here to gain an understanding of activities
in outside spaces in Sector 2.
Sector 3 Units
Sector 3 is located in the southernmost portion of Iglesiachayoq, on a hill located
southwest of Sectors 1 and 2. After the architectural survey completed in 2013 and 2014, we
recorded a higher percentage of ovoid structures (n=16/23 total ovoid structures site-wide, or
69.6%, and 16 out of 34 total structures, or 47.1% of all structures in Sector 3) in this region as
compared to the other site sectors. Since pre-Inka antecedents typically constructed ovoid
structures similar to those found in Sector 3, I hypothesized that Sector 3 was the older, firstoccupied area of the site. Like Sector 2, we placed four units in Sector 3, largely to test
hypotheses regarding the potentiality of Taki Onqoy rituals in this area (Figure 5.11).
244
Figure 5.11. Units, sizes, and UTM coordinates for Sector 3 excavations.
Unit 8 was a 4m x 3.5m unit excavated in the ovoid structure EA 292. Structure EA 292
(see map) is located in a concealed, low-visibility part of Iglesiachayoq, sheltered by a large rock
outcrop and hill, and hidden from view from individuals in Sectors 1 and 2 (Figure 5.12).
Structure EA 292’s access faces a small plaza divided into three sections by small restraining
walls, and this plaza abuts massive rock faces which could easily have been worshipped as
huacas. This unit was placed in Structure 292 in order to test the hypothesis that Taki Onqoy was
245
being practiced in hidden, covert areas away from the center of the site. Given the layout of this
structure and the small plaza, I wanted to see if the practices visible in the archaeological record
here would reflect Andean ritual practices.
Figure 5.12. Prehispanic or “local” plaza in Sector 3 facing north.
Unit 9 was a 4m x 2.5m unit placed to encompass the domestic interior of structure EA
247. Located southwest of Unit 8, Unit 9 was in a rectangular structure with rounded corners that
was very modest in size. Like Unit 4 in Sector 1, parts of the walls of Structure 247 were
collapsed, ironically allowing for better overall preservation of the surface materials in this
structure. The object of this unit was to gain an understanding of a more modest household, or
how the common residents of Iglesiachayoq lived at the time. The placement of Unit 9 was
designed to investigate lifeways of a local Andean in the sixteenth century.
246
The last domestic unit was unit 10, which was located between Sectors 1 and 3 and had
an excellent view of Sector 1 and the center of the site. As another potential “lookout” or
“outpost” (like Unit 7), Unit 10 was placed in Structure EA 212, which was quadrangular with
rounded corners. The access of this structure faces to the west, and affords vision of anyone
moving between sectors of the site, as well as a perfect view of church activities. There was an
enormous tree which had disrupted the wall at the eastern end of the structure, which we left in
situ and worked around. The unit was 5m x 3.5m, and placed to encompass the interior space of
the structure, and had an extension of 1m x 0.55m to the southeast in order to define base
architecture and the potential presence of a hearth in this corner. The goal of this unit was to lend
insight into who was living here and what sorts of practices were occurring—if the resident of
this structure was some sort of Spanish ally or someone who was monitoring activities in Sector
3 of the site, we might expect a higher density of Spanish artifacts.
The last unit excavated was Unit 15, which was a 2m x 2m patio unit placed one meter
north of the access of structure EA 292, where Unit 8 was located. As mentioned above,
Structure 292 is located in a secluded, nonvisible location and as such may have been an ideal
location for Taki Onqoy activities. Unit 15 was placed with the associated Structure 292 in order
to collect evidence for activities occurring outside of enclosed spaces. Units 8 and 15 were the
first two structures excavated, so the depth of these units is much greater—I advised the crew
chief to continue excavating in portions of these units in order to understand the full stratigraphy
of the site. In all other units, we excavated until sterile, but in Units 8 and 15, we excavated well
below sterile in order to insure that there was no other evidence of cultural occupations.
Unit Documentation Procedures
247
Peruvian students from the Universidad Nacional de San Cristóbal Huamanga (UNSCH)
directed the excavations of all units and were responsible for documentation and collection of
data. All excavation data was recorded on Locus forms specially designed for PATO. Prior to
beginning excavation, a starting photo was taken in order to document the current condition of
the unit and structure. During excavation, sketch maps were drawn on forms for each locus, and
photographs were taken at the start, end, and throughout excavation of loci. Upon finishing
excavation to the sterile level, aerial photos were taken which included the full top-down view of
the structure as well as all corners of the unit so that these photos could be georeferenced in
ArcMap. Profiles, cuts, and plan views were also drawn for each unit. After the final visitation
from our regional archaeological supervisor in Ayacucho, units were backfilled and
photographed.
Phase Three: Artifact and Materials Analysis
Phase Three of the Proyecto Arqueolόgico de Taki Onqoy consisted of several months of
laboratory analysis in Peru, followed by the exportation of 152 samples67 for isotopic and
compositional analysis. While broad architectural survey and in-depth excavation allow for basic
interpretations of daily life at Iglesiachayoq, it is the small-scale analysis of contexts and artifacts
which allows for nuanced analysis between practices and activities in different regions of the
site. The lab analysis portion of this investigation consisted of both macro and micro analyses
and spanned ceramic, lithic, biological, and faunal data. All analyses which were feasible to
complete in Peru were completed in the field, while 152 samples were exported from
67
These included bone and teeth samples for stable isotope analysis and radiocarbon dating, as well as a subset of
human remains for DNA analysis and 126 samples of ceramic fragments for LA-ICP-MS analysis.
248
Iglesiachayoq with permit Number 127-2016-VMPCIC-MC. One-hundred and twenty-six of
these samples were ceramic, and were analyzed at the Elemental Analysis Facility in Chicago,
Illinois at the Field Museum. Six of these samples were bone and teeth, and will be analyzed at
the Genetic Anthropology and Bio-Cultural Studies Lab at Vanderbilt University, in order to
achieve DNA samples from the individuals in the church.68 The last 20 samples were human
bone and human teeth. Eight human bone samples (the biggest pieces which likely contained the
most collagen), 23 plant samples, and four water samples were chemically prepared at the
Bioarchaeology and Stable Isotope Research Lab (BSIRL) at Vanderbilt University in March of
2017 and were subsequently analyzed on a Thermo Delt Plus Advantage with a Costech ECS
4010 Elemental Analyzer with Conflo III interface at Yale Analytic Stable Isotope Center in
April 2017. Eleven apatite samples were prepped at the Paleodiet Laboratory at Northern
Arizona University and then processed at the Colorado Plateau Stable Isotope Laboratory at
Northern Arizona University. The collagen remaining from the eight bone samples was then
processed at the Keck Carbon Cycle AMS Facility at the University of California, Irvine in order
to obtain radiocarbon dates; only seven were returned because there was not enough collagen in
the remaining in the eighth sample to process it. These bone samples were tested for comparative
stable isotope levels including Strontium, Nitrogen, Oxygen, and Carbon, and also for Carbon 14
dating if enough material remained (Appendix B).
Ceramic Macroscopic Analysis
68
Currently in contact with Dr. Jada Benn Torres, and I have received $1100 to fund the first phases of this project
from the Department of Anthropology at Vanderbilt University.
249
The majority of my post-excavation fieldwork was spent on ceramic analysis. From
September through December of 2015, I completed ceramic analysis in Andahuaylas (district of
Apurímac). Permission was granted from the Ayacucho Ministry of Culture to complete this
analysis in a separate department, as Andahuaylas is much closer to Chicha-Soras and has been
my home base in Peru for several years. Conforming to ceramic analysis standards I have
completed in other regions, my team first washed and dried all sherds we collected, and rebagged/tagged them if necessary. Next, each bag was primarily sorted into diagnostic and nondiagnostic sherds. Diagnostic sherds included rims (borders), bases, and anything with a design
or unique external aesthetic treatment. Diagnostic sherds also included those which had a novel
paste type. Each diagnostic sherd was photographed, drawn, and analyzed separately according
to individual sherd (rather than as a group of sherds with similar characteristics). The 2015
Reglamiento de Intervenciones Arqueológicas (RIA) suggests that nondiagnostic sherds be
reburied after being separated from diagnostic sherds. Thus, in preparation for this reburial, I
also separated nondiagnostic sherds into groups, weighed and photographed them, and analyzed
them according to broad categorical trends. In this way, I was attempting to prevent a loss of data
when the sherds were reburied.
Diagnostic sherds were numbered according to the following provenience string:
(site.locus.quad.ceramicnumber). For example, a ceramic sherd coming from Locus 104, quad 2
would read as: IC.0104.2.001. In this way, all ceramic sherds had distinct numbers from one
another and could not be comingled. After sherds were separated and numbered, the diagnostic
fragments were individually weighed, photographed, and drawn on millimeter paper. Those
which were rims were measured as to the overall size and angle of the mouths of the vessels,
which was reflected in the drawings. For all diagnostic sherds, macroscopic analysis was
250
completed for the following attributes: weight, dimensions, culture/style, form, part of the
ceramic, type of border/handle/base/mango, diameter, color (Munsell) of exterior/interior/paste,
decoration description, exterior and interior treatment, thickness, presence of firing scars, paste
texture, paste composition, paste color, degree of thorough firing, presence of tempers,
consistency, conservation status, and any other general comments. Appendix C provides a
description of the characteristics of each paste type and their frequencies site-wide, as well as the
raw data from the project Access database.
Lithic Macroscopic Analysis
I also completed the lithic macroanalysis from our excavations at Iglesiachayoq. I am not
a lithics specialist, but for the purposes of this project, I was primarily assessing lithic fragments
in several basic descriptive, categorical, and quantitative traits. For each diagnostic lithic artifact,
we recorded the weight and dimensions, primary material, type of artifact, condition (complete
or fragmented), if there was a cortex or patina, the presence of retouching, and several other
factors related to the bulb of percussion and possible function. Like the diagnostic ceramic,
diagnostic lithic fragments were given a unique code (site.locus.quad.Lnumber).
Metal Macroscopic Analysis
The majority of metal artifacts recovered at Iglesiachayoq were tupus from the church
burials. However, we did recover a few other metal artifacts from various household units. I
completed all metals analysis, primarily through macroscopic and descriptive means (Appendix
E. Each metal was weighed and its dimensions were recovered, along with an educated guess as
to the type of metal (the majority of these artifacts were of copper or iron, as designated by the
251
greenish or orange hues in the rust patterns). We recorded the form of the metal, the estimated
period of creation, the colors, decorations, evidence of use, inscriptions, and any degradation the
metal may have experienced over the centuries underground. Each metal was extensively
photographed and drawn by hand and in Adobe Illustrator (Figure 5.13).
Figure 5.13. Example of a metal artifact drawn in Adobe Illustrator.
In compliance with the instructions from our supervisor from the Ministry of Culture in
Ayacucho, each metal was separated and wrapped in acid-free paper, and eventually submitted to
the ministry for storage. While no chemical or microscopic analyses were completed on the
metals for this dissertation, this could be an avenue for future research.
252
Faunal Analysis
Likely due to the short occupation and poor preservation at Iglesiachayoq, we recovered a
very small sample of faunal remains, which totaled 673 NISP (number of identifiable
specimens). Total bone weight of the assemblage was 2,196.71 grams. All faunal analysis was
completed by Sarah Kennedy (Ph.D. Candidate at the University of Pittsburgh, Appendix A).
Identification of vertebrate remains was aided by a collection of identification guides (Pacheco
Torres 1986) and remains were identified to the lowest taxonomic level possible. Further
analysis included the identification of elements represented, specimen counts (number of
identifiable specimens, NISP), bone surface modifications (weather, burning, butchery), age
estimations, and bone weight (Reitz and Wing 2008). For the purposes of this study, the
minimum number of individuals (MNI) was calculated using conservative criteria such as
symmetry, size, and fusion (Reitz and Wing 2008). Age estimations for mammals combined data
on fusion, dentition, and wear patterns (Sandefur 1988; Sisson, Grossman, and Getty 1975;
Wheeler 1982). Three mammal age categories were used: juvenile, sub-adult, and adult
(deFrance and Hanson 2008). Burning was scaled by intensity and included four categories:
partially black, black, black-white, and white. Finally, some pathologies were recorded—
pathologies on animal bone can lend insight into what the animal did during its lifetime.
Kennedy next took photos of all bones, and took selected photos of specific evidence of human
effects on bones (cutting, chopping, etc.) under magnification. Kennedy recorded all results into
an Excel spreadsheet and created different charts by Unit or Sector. These results were then
entered into the PATO Access database.
253
Bioarchaeological Data Collection
The only human bone remains we recovered were from the central church at
Iglesiachayoq—all remains were in discrete loci, so there was no comingling of bones. Anna
Gurevitz (Ph.D. candidate, University of California, Merced), completed all analysis of
osteological characteristics for this study. Standard bioarchaeological techniques were used to
estimate age from the pattern of dental eruption. Since the preservation was poor at
Iglesiachayoq, we often were left with only teeth and long bones from which to estimate age.
Age estimates were divided into basic categories (see Table 5.3).
Age Category
Juvenile
Adult
Age Range
Fetus
<0
Infant
0-3
Child
3-12
Adolescent
12-20
Adult
20+
Table 5.3. Breakdown of age categories used by Gurevitz in osteological analysis.
Similarly, we were not able to assess biological sex from the remains—when differentiating
between sexes, we took the relative size of osteological elements into account. We also used
cultural markers such as the presence of tupu pins to tentatively suggest that these individuals
may have been female (although we understand that those interred without tupus may also have
been female, or those with tupus may have been male).69 For the purposes of classifying burials
69
In the case that we find biologically male individuals interred with tupu pins, then we can consider interpretations
of nonbinary genders, as have been documented previously in the Andes (Horswell 2005)
254
as either primary or secondary, we developed several criteria informed by deposition processes
within the church and anatomical positioning of skeletal remains. In order for a burial to be
categorized as primary, each set of remains was judged according to the following:
1. The remains had to be in proper anatomical position in relation to decomposition
processes.
2. The amount of teeth recovered had to be consistent with the condition of the cranium
and postcranial preservation.
All other burials which did not meet the above criteria were considered to be disturbed,
secondary, or not identifiable.
In the process of excavating the human remains from the church, we found a spectrum of
interment patterns. We developed an organizational strategy in order to better categorize the
variance in these burials, resulting in six distinctive interment types. These types were designed
for ease of classification and explanation, rather than any restrictive, non-flexible categorization
of individual burials. Two of the burials in Unit 18 were indeterminate in terms of
primary/secondary status, and do not align with any of the types below:
Type 1: Extended and primary (n=3)
Type 2: Flexed and primary (n=2)
Type 3: Extended and secondary (relocated)—these were interred with an obvious
attempt at anatomical accuracy, with mixed results (n=1)
Type 4: Flexed and secondary—these generally appear to have been relocated bundles
which were reinterred below the church floor (n=4)
255
Type 5: Removed—identified by the presence of only distal bones, stray bones or teeth,
soft soil, and possibly evidence of rock fill (n=4)
Type 6: Disturbed—disarticulated and arbitrarily strewn (n=5)
During excavation, all burials were fully exposed and photographed, and then collected as an
entire locus. In this way, all of the elements which pertained to one individual were kept separate
from all other individuals. In the lab, bones were cleaned and analyzed by Gurevitz, who then
inventoried each individual bone fragment and recorded weights and measurements, as well as
took photos (see Chapter 7 for a more detailed discussion on this schema).
Stable Isotope Analysis: Osteological Sample Protocol
Due to the poor preservation of the human skeletal remains at Iglesiachayoq, little
observational analysis could be performed. Stable isotope analysis provides a way to gain
invaluable information on population diet, migration, and mobility, allowing a large amount of
data with a minute amount of destruction. Samples were chosen based on the criteria of
preservation, and the educated assessments of which samples would be able to provide the most
information. Enamel samples were taken from most individuals when available as tooth enamel
is typically the best preserved tissue of the human skeleton. When possible, samples were taken
with an intact and relatively stable root in hopes that dentin was preserved. In one case, an
unerupted first molar was taken just in case other samples from the same unit would not provide
sufficient material. The goal was to get at least two samples from each unit for comparison of
both apatite and collagen. Each bone and tooth sample were packed in separate Ziploc bags with
their specimen codes written in sharpie on the front.
256
Sample Protocol: Plants, Animals, and Water
As part of the stable isotope analysis, we collected locally grown food in order to
compare the local results with our archaeological results. Depending on the plant, we dried plants
on a covered surface in the sun for at least 24 hours. For maize, we removed the kernels from the
cob and laid them on to the surface. For plants that retained water (cacti, potatoes, yams, tubers,
etc.), we cut them into smaller wedges and laid them flat on the surface to dry out. Once dried,
we collected each plant individually and placed it in its respective plastic bag. The plant names
were recorded on the bag and in a separate list, which documented the plant species, location of
collection, the start time of drying, the end time of drying, the total dry time, the date collected,
and the weight of the sample (see Appendix D for full list).
During the course of fieldwork, we also attempted to collect several animal species in
order to have the comparative isotope values from modern animals. While we had planned to
obtain cuy or other rodent, this proved extremely difficult and we were only able to collect local
trout. We heated the trout in distilled water to no more than 40 degrees Celsius (104 degrees
Fahrenheit) and let them boil for two hours. Next, we carefully removed the flesh with a knife,
scalpel, and tweezers. We cleaned the bones thoroughly with distilled water and let them dry,
and then put the bones in a plastic bag with the species and date collected.
Finally, we also collected water from two local sources, a lake near the site and the
Chicha-Soras River which is the current dividing line between Apurímac and Huamanga. We
rinsed two 30mL glass bottles in each water source, then placed each bottle fully submerged in
the water until the bottle was fully filled. While the bottle was submerged in the water, we
screwed on the cone-capped lid (Figure 5.14).
257
Figure 5.14. Map of structures and water sources at Iglesiachayoq.
Stable Isotope Lab Protocol
Once all of our plant, animal, water, and human samples were transported to the United
States, we further dried all samples and eventually crushed them for stable isotopic analysis. The
samples prepared for collagen extraction (n=8) were prepped with minor modifications as
described in Ambrose (1990). All prep work for plants and human bone collagen extraction was
completed at the Bioarchaeology and Stable Isotope Research Laboratory at Vanderbilt
University under the direction of Tiffiny Tung. Plant and human samples were analyzed at the
Yale Analytic Stable Isotope Center on a Thermo Delta Plus Advantage with a Costech ECS
4010 Elemental Analyzer with Conflo III interface. The C:N ratio for the human bone samples
258
were 2.7, which is in the acceptable range when evaluating for diagenetic change (Schoeninger et
al. 1989).
The human bone and enamel samples (n=11) were prepped for carbonate analysis at the
Anthropology Paleodiet Laboratory at Northern Arizona University. Samples were prepped
using a protocol adapted from Schoeninger (2007). These samples were then analyzed at the
Northern Arizona University Colorado Plateau Stable Isotope Laboratory (see Appendix D for
the raw data from isotopic and carbon 14 dating results).
Sample Protocol for DNA Samples
DNA extraction and analysis obtains DNA from archaeological samples in order to
obtain DNA sequences or complete genomes. Through DNA extraction, researchers can ask
questions related to how culture and biology shape human diversity, demography, and/or health
of a population or populations. This method is completely destructive due to the possibility of
contamination of DNA sequences through environmental and human interactions. For this study,
after conferring with Dr. Lars Fehren-Schmitz, we selected samples of a human molar and/or the
petrous portion of the temporal bone. For the sampling procedure, we used a 10% bleach
solution, detergent, a Dremel Drill and diamond saw, paper pads, and face masks. Following a
contamination protocol from Yang and Watt 2005, we first cleaned the Dremel bits with
detergent and wiped them with the bleach solution. They were allowed to dry completely. Next,
we used the dremel drill with the diamond saw at low speed to prevent the bone dust from
spreading as much as possible. After each sample was taken, we wiped down the bench, table,
and scale with the bleach solution and replaced the pad. We let everything dry completely after
each individual. We stored each individual sample in an unused, sealable plastic bag once the
259
samples were dry. Finally, we recorded a brief history of who handled and conducted research on
the skeletal material and recorded the environment from which the material came from and the
condition of each sample when it was obtained (see Appendix D for raw data).
LA-ICP-MS Compositional Analysis Methods
I obtained an NSF subsidized grant to analyze 126 ceramic samples at the Field Museum
in Chicago, Illinois. In July and August of 2017, I stayed in Chicago to complete this analysis.
Each day, the lab manager, Laure Dussubieux, would run the machine in order to ensure full
functionality. I broke very small fragments off of my larger ceramics and placed five of them on
the tray. Once in the machine, I selected ten locations on each ceramic fragment from which to
analyze the composition. I then selected five locations on each of three standards which were
used to determine the machine’s functionality during each run. This procedure was repeated for
all 126 samples. After each ceramic sample had been processed, the data was collected by date
into excel files. This data was then cleaned and combined into one large spreadsheet for
statistical analysis.70
Post-Analysis Procedure and Methods
After completing all analysis and selecting samples for export, my team and I created a
final inventory of all diagnostic artifacts and then returned these to the Ministry of Culture in
Ayacucho. Here, we were advised that we should reinter nondiagnostic artifacts back at
Iglesiachayoq, in accordance with the new RIA. My co-director and I boxed up all of the
70
I am in the process of analyzing these results, and they will be prepared for publication at a later date.
260
nondiagnostic artifacts, inventoried them, and placed them into plastic containers. Each container
held the inventory for its respective contents. With the assistance of two local workers, we
returned the remaining artifacts to the site and reburied them in Unit 3.
Phase Four: Spatial Analysis
The last phase of research for this dissertation utilized spatial analysis to test Hypotheses
Four and Five, and was undertaken after all fieldwork was completed. All spatial analysis was
performed at the site level, and so was based on initial Phase One mapping and architectural
survey. Here, I briefly discuss the key data sources utilized in spatial analysis, and I describe
these analyses in more depth in later chapters.
Since I was looking only at site-level practices, I used one background Digital Elevation
Model (DEM), SRTM DEM v2 with 1 arc second (~30m resolution). Satellite imagery was
provided from CNES, Digital Globe, and Landsat via Google Earth. Networks were developed
from my Phase One architectural survey and mapping. All local source data was transformed and
projected to the Universal Transverse Mercator projection (UTM Zone 19 South), using the
WGS 1984 datum.
Summary of Methods
In short, my research at Iglesiachayoq was completed over four years and several phases
of work. Initially, I visited the site and completed a brief archaeological survey and basemap of
Iglesiachayoq. During this time, I also visited Ayacucho archives in order to photograph any
potentially useful documents. Next, after securing funding for a more in-depth study at
Iglesiachayoq, I utilized the survey data and basemap to select 19 units for excavation. These
261
units were a mixture of strategic selections and random sampling, based on the excavation
questions. Excavation and collection techniques followed standardized methods and were
documented according to the locus system in order to keep all contexts fully separate from one
another. Ceramic and lithic artifacts were washed and dried in the field, and then separated
according to diagnostic and nondiagnostic categories in the lab. I completed the macroscopic lab
analysis for ceramic and lithic data. Sarah Kennedy completed the faunal analysis and Anna
Gurevitz completed the osteological analysis, both according to standardized analysis techniques.
The methodology was thus designed both to answer specific questions regarding Taki
Onqoy practices and also to gather more fundamental data about the site’s occupation, the
individuals living at the site, and the material culture compositions. Excavation in both domestic
and ritual contexts in all three sectors of the site were designed to evaluate differences between
sectors and highlight locations where Taki Onqoy practices may have been performed.
Macroscopic analysis was performed on all artifact assemblages from all units, with special
consideration for ceramic, lithic, faunal, and bioarchaeological data.
Because of the poor preservation of the soil, we decided to complete further microscopic
or elemental analyses on a subsection of ceramic, human bone, and flora. These analyses not
only added to the scientific training of the investigators, but also provided more nuanced data as
to diet, migration, and ceramic sourcing at Iglesiachayoq. While stable isotope analyses allowed
us to generate a diet profile for individuals interred in the church, compositional analysis of
ceramic had provided the first elemental study of sources from the Chicha-Soras region, and will
be an invaluable contribution to future archaeological studies.
262
CHAPTER 6
HOUSEHOLD ARCHAEOLOGY AT IGLESIACHAYOQ
Taki Onqoy was a critical movement which reportedly infiltrated daily practices in
addition to religious worship and ritual. Its ability to affect production, consumption, and dress,
and its potentiality to influence any demographic of Andean populations (elderly, young, men,
women, etc.) alarmed Spanish authorities, who could not predict the spread of the movement
(Stern 1993). Though primary accounts aver that Taki Onqoy inspired a reaffirmation of
prehispanic daily practices in combination with a new emphasis on rejection of Spanish cultural
materials, the implementation of these new practices has not been considered at a household
level. A goal of this investigation was thus to begin to identify how individual households at a
known Taki Onqoy settlement rejected or incorporated Spanish goods into daily practices. This
study also questioned how Spanish authorities punished those caught participating in Taki
Onqoy. Specifically, archaeological excavation of households and open spaces at Iglesiachayoq
addressed the following hypotheses: 1) If Taki Onqoy practices did not entirely proscribe the use
of Spanish things, they would likely be characterized by the repurposing of Catholic or Spanish
objects in new ways or hybrid forms of material culture, and 2) If Spanish responses to Taki
Onqoy at Iglesiachayoq were violent, they would be marked by destruction of huacas in public
arenas or physical punishment of takiongos in public spaces. To test these hypotheses, my team
and I selected excavation units which would maximize the range of household forms throughout
the site, in addition to a subsection of their associated patios (see Chapter 5 for research design
and methodology).
In this chapter, I present the results of household and domestic archaeology throughout
Iglesiachayoq. In addressing Hypothesis One, I argue that prehispanic practices related to food
263
consumption, ceramic styles, and stone technologies were ubiquitous and unchanged throughout
Iglesiachayoq. In fact, the material presence of Spanish cultural items—Old World animals, new
ceramic technologies, imported personal goods, etc.—was faint throughout household units. I
suggest that the lack of European goods at Iglesiachayoq can be interpreted in two ways, since
the assemblage is defined by an absence of evidence: first, it is possible that those households
without Spanish goods were actively rejecting Spanish cultural traditions. Second, the absence of
Spanish goods could be the result of a lack of access to these goods—Taki Onqoy was taking
place in the 1560s, a mere thirty years after the Spanish invaded Peru. The absence of Spanish
cultural material at Iglesiachayoq could thus also demonstrate that associated colonial processes
related to material culture had not yet reached the Chicha-Soras Valley at this time. I also argue
that while there were no clear patterns in incorporation or rejection of Spanish material culture,
my team and I did find evidence of status differences and types of activities occurring in
different sectors.
In addressing Hypothesis Two, my team and I found limited evidence to support the idea
that Spanish authorities were burning huacas in the central plaza. Though we collected tentative
evidence for burning events and documented the fragmentation of a specific type of lithic
material in elevated levels in the plaza, more research must be completed to fully prove or
disprove this hypothesis.
In this chapter, I first discuss potential Early Colonial Period material assemblages,
especially the processes by which Old World goods were incorporated into New World cities.
Next, I describe the site layout and site chronology in detail in order to give the reader a
thorough background in understanding where “more Catholic” or “more Andean” spaces were
located. Then, I discuss the findings of all household, patio, and plaza units, highlighting
264
important contexts. Finally, I compare general findings according to artifact assemblages and
return to my hypotheses to end the chapter.
Early Colonial Period Materials
Animals and Plants
Though the Columbian Exchange has been heralded as a major turning point in world
history, the timing of the arrival of European plants and animals to the New World is still
understudied (Crosby 1972, 1986; Gade 2015). For the study of Taki Onqoy and the practices of
rejection of Spanish foodstuffs, understanding this timing is critical to deciphering whether
takiongos were rejecting these materials, or if they lacked access to them. Archaeological studies
of the incorporation of Spanish foodways have “illustrated how changes in household economies
associated with imperial incorporation and resistance varied greatly based on region and
settlement, if not also across domestic units within the same communities” (Kennedy and
VanValkenburgh 2016: 74). That is, archaeological studies have not recorded any one particular
pattern of incorporation or rejection of Spanish goods throughout the Colonial Period Andes.
Indeed, variation is even present at the site level, which could be indicative of differential access
to these goods, personal choice, or even overt rejection of Spanish goods.
When the Spanish invaded the Andes, they did not arrive equipped with livestock or
seeds for production and cultivation. In fact, most of their goods consisted of horses and dogs
used in conquest or personal items which could be easily transported (Covey in press, Gade
2015). These conquistadores were accustomed to consuming native foods from their subject
communities in the initial years after conquest. Recent published research by Daniel W. Gade in
2015 investigates the journey of Old World biota to Peru more thoroughly, and elucidates the
265
processes by which physical flora and fauna were transported to the Andes via Panama. Gade
argues that animals and plants were brought from Spain to the Panamanian Isthmus, then were
taken to the Pacific port of Panama, and then sailed to the port of Callao to the south (Gade
2015). These organisms thus had to survive an incredibly long journey in less-than-ideal
conditions, and their incorporation into Peruvian daily life could not be implemented until well
after the infighting between conquistador factions had ceased.
Scholarly research has suggested, however, that the importation and commodification of
European foodstuffs became cultural identifiers used to separate “Spaniard” from “indio” (Covey
in press). By the late 1530s, the Spanish started to import, introduce, and integrate Old World
animals to Andean cities in the highlands, likely from their previous exploits in the Caribbean
(Earle 2012a, 2012b; Gade 2015). Initially, it was unknown how high altitudes would affect the
success of these plants and animals in the Andes. Those species which reproduce at a fast rate
(pigs and chickens) became widespread rather quickly, while those animals which reproduce
more slowly (horses and cows) were rarer into the 1600s (deFrance 1996, 2003).
Documentary sources provide some information about the incorporation of specific plants
and animals in the highland Andes. The Spanish placed a high importance on wheat and bread,
insuring its success in Peru as a priority in the years immediately following conquest. In 1541,
the first major wheat harvest was documented, and in 1543, the price of wheat dropped,
indicating that its production had become successful (Cobo 1956 [1643]: 408; Covey in press,
Gade 2015). From the repartimiento of Soras specifically, Pedro de la Gasca’s tasa of 1549
provides clues as to the types of foods and products available prior to Taki Onqoy’s fluorescence
in the region. La Gasca records that the Spanish Crown demanded tribute in the form of “4,500
pesos in gold or silver, 120 tunics of cumpi cloth…600 fanegas of maize, 60 fanegas of potatoes,
266
156 camelids, 36 pigs, 500 birds, 150 eggs per week, and fresh fish” (Rostworowski 2005
[1975]: 265).71 From this record, it is clear that in 1549, the abundant foodstuffs were still
primarily Andean products, with the exception of pigs. In the nearby Department of Cuzco,
Covey argues that horse, donkey, and cow were all plentiful by the early 1600s (Covey in press,
Garcilaso de la Vega 2000: Book 9). Though we have no exact records for the Chicha-Soras
region, its proximity to Cuzco and the documentation of pigs in the repartimiento by 1549
suggests that European animals were available in the region, though they were likely
incorporated at differential rates in different towns (as supported by documentary and
archaeological studies Cook 1981; Deagan 1996; deFrance 1996; deFrance et al. 2016; Wernke
2013).
At the time of Taki Onqoy at Iglesiachayoq, it is thus plausible that both European and
native goods were available and attainable. From the documentary evidence discussed above, a
potential mixed assemblage would include both Andean and European plants and animals (Table
6.1).
Translated from “Primeramente da en cada un ano quarto mil y quinientos pesos de ley perfeta en oro o en plata
puesto en casa del encomendero, Y ciento y veynte vestidos de lana de cunbi…Y seiscientas fanegas de mayz las
quatrocientas fanegas puestas en casa del encomendero y las demás en sus tierras, Y sesenta fanegas de papas, Y
ciento y cinquenta y seis ovejas de la tierra…Y treinta y seis puercos de ano y medio, Y quinientas aves la mitad
henbras en casa del encomendero, Y todos los viernes y días de pescado fuera de quaresma treinta huevos y en la
quaresma cada semana ciento e cinquenta huevos y en los dichos días algún pescado fresco del rio” (Rostworowski
2005 [1975]: 265), translation mine.
71
267
Common Staple Goods in the Early Colonial Period in the Chicha-Soras Valley
Andean
European
Flora
Maize
Wheat
Potatoes/tubers
Quinoa
Fauna
Camelids
Horse
Cuy
Cow
72
Chicken
Pig
Sheep/Goat
Chicken
Table 6.1. Flora and fauna in the Chicha-Soras Valley in the Early Colonial Period.
Ceramic and Lithic Technologies
Much like the introduction of Old World biota to Peru, the introduction of European
technologies in ceramic production was implemented at variable rates throughout the Andes. In
the Early Colonial Period, Spanish conquistadors were importing not only their foodstuffs, but
also new technologies such as glazed and wheel-thrown ceramic wares. In particular, tin-glazed
earthenwares known as majolicas are a key indicator of Early Colonial Period wares, and their
production has been traced to both Panama and places in Peru and Ecuador (Jamieson 2002;
Kelloway et al. 2018; Weaver 2015; Figure 6.1). Lead-glazed wares are also common in Early
Colonial Period assemblages, and are characterized by their green to yellow-green lead glaze
(VanValkenburgh et al. 2017; Weaver 2015; Figure 6.2). Finally, botijas were large, coarse,
earthenware amphorae which were utilized in the transportation and storage of liquids (Weaver
2015). When considering the ceramic assemblages during the Early Colonial Period in Peru,
72
There is some debate on whether the chicken was native to South America, or whether it was brought from Spain,
see Gade 2015 and Seligman 1987.
268
these three types of vessels are key indicators for transmission of Spanish goods and
technologies.
Figure 6.1. Majolica sherds produced in Peru. Picture from Kelloway et al. 2018.
Figure 6.2. Early green glaze wares from Peru. Photo from VanValkenburgh et al. 2017.
269
In the Chicha-Soras Valley, prior to my study there were no documented Early Colonial
Period occupations. In all previous work completed in the valley, reports of European wares are
nonexistent; however, most previous archaeological work in the region was more interested in
earlier occupations (Wari, LIP, Inka), and so the lack of identification of these vessels and
techniques may be a function of scholarly bias rather than actual absence of goods. Once again,
following the hypothesis that takiongos rejected or repurposed Spanish goods, material correlates
for rejection of these goods would be in their absence in households, or in their utilization and
subsequent rejection or intentional destruction in household contexts. Despite the lack of
identification of European-style technologies and vessels in the Chicha-Soras Valley, I suggest
that a possible Early Colonial Period ceramic assemblage may include both European and native
technologies (Table 6.2).
Possible Ceramic Types in the Early Colonial Period in the Chicha-Soras Valley
Andean
European
Ceramic Styles
Coarse brown wares (both
Majolicas
Chicha and Soras style)
Lead-glazed wares
Provincial Inka wares
Botijas
Wheel-thrown wares
Table 6.2. Possible ceramic types in the Early Colonial Period in the Chicha-Soras Valley.
While ceramic technologies brought by Europeans created an entirely new material
signature, lithic technologies would have remained largely unaltered by Spanish invasion.
Andeans continued to use specific stone materials for specific activities, while the Spanish could
rely on metals and firearms. In the Chicha-Soras Valley, common stone types include basalt,
obsidian, quartz, chert, andesite, and tuff.
270
Personal Goods and Portable Items
The last type of material artifact on which could have been relevant to Taki Onqoy
practices in the Chicha-Soras Valley is that of personal goods or portable items. These artifacts
include the objects that the Spanish could have brought with them on their journeys to Peru, and
are predominantly small religious items (crosses, rosaries, and amulets), personal adornments
and clothing (jewelry, beads, ornaments, buckles, hooks, sewing goods), or personal items
(combs, coins, weapons, etc.) (Deagan 2002). Additionally, portable objects could consist of iron
objects (nails, door latches, bells, etc.). Of these objects, Nueva Cadiz beads (turquoise straight
or twisted cylindrical beads) are often used to date Early Colonial Period assemblages to the first
half of the sixteenth century (Little 2010; Wernke 2013). Though their trajectories are not fully
understood, iron caret-head nails found in Spanish sites in the Americas also suggest occupations
which date to the first half of the sixteenth century (Deagan 2002; Flint and Flint 2003). If Taki
Onqoy practices included the rejection of all Spanish goods, then I would expect to see an
absence of these goods in some households at Iglesiachayoq.
Personal goods for Andean populations include much of the same categories, but made of
different materials. Small religious items could have been amulets, figurines, illas, and pieces of
huacas. Personal adornments and clothing could have consisted of metal jewelry, shell beads,
tupu pins, needles, or other weaving tools. Additionally, portable objects may have consisted of
musical instruments, ceramic miniatures, or small weapons such as axe heads or bola stones.
Early Colonial Period personal object assemblages may thus contain any mixture of European
and Andean items (Table 6.3).
271
Personal Items in Early Colonial Period Assemblages
Andean
European
Religious items
Amulets
Crosses
Figurines
Rosaries
Illas
Amulets
Huacas
Personal adornments and
Jewelry
Jewelry
clothing
Shell beads
Beads
Tupu pins
Ornaments
Weaving tools
Buckles
Hooks
Sewing goods
Portable objects
Ceramic miniatures
Combs
Axe heads
Coins
Bola stones
Weapons
Nails
Door latches
Bells
Table 6.3. Possible personal items in Early Colonial Period assemblages in the Chicha-Soras
Valley.
I have outlined and defined these artifact classes and the potential archaeological
correlates of Early Colonial Period assemblages generally in order to provide a background for
the excavations contexts recovered from my work at Iglesiachayoq. It is critical to understand
what materials could reasonably be found at Iglesiachayoq before we are able to infer which
things were associated with Taki Onqoy practices, and which things could have been rejected,
repurposed, or entangled with performances. In the next sections, I describe the sectorization,
architectural styles, and archaeological occupations of Iglesiachayoq, and then present
excavation results from household archaeology.
Sectorization of Iglesiachayoq
272
Iglesiachayoq contains three distinct sectors which are in discrete geographical locations
but still closely interrelated as a whole site (Figure 6.3). My strategy in separating the site into
these sectors was to provide a heuristic for analysis of the site as a whole, based on informed
hypotheses of occupation periods and usages of different sectors of the site. Furthermore, by
analyzing artifacts both by sector and by type of structure (ovoid vs. quadrangular with squared
corners vs. quadrangular with rounded corners), I am considering overlapping axes spatially and
materially in order to provide a better understanding of the activities at Iglesiachayoq.
273
Figure 6.3. Sectorization of Iglesiachayoq.
274
Sector 1
Sector 1 is the obvious site center at Iglesiachayoq. It is roughly located at 656769E and
8427951N (UTM 18S) and makes up the northwestern portion of the site. Sector 1 is essentially
nestled in a small valley, with Sectors 2 and 3 located on low hills to the east and south of Sector
1. Overall, there are 30 structures in Sector 1, which can be broken down into 5 ovoid structures,
9 quadrangular with squared corners, and 16 quadrangular with rounded corners. Most
significantly, there are two massive structures in Sector 1 which clearly make up the site center.
First, there is a 14m x 11m quadrangular structure with rounded corners, with its accessway
facing south (Figure 6.4). Indicative of typical Inka archaeological signatures, the entryway to
this structure is lined with polygonal ashlar cut stone masonry and the interior includes 16 even
niches located along the same horizontal plane.73 There are three rectangular windows at the top
of the walls, one each on the west, north, and east walls. This structure was likely a local leader’s
household, perhaps the kuraka, Joan Hachi, and I will refer to it as the “kuraka house” for the
rest of this dissertation.
“Typical” Inka masonry consists of polygonal ashlar stones which fit snugly together, leaving no gaps between
them. For examples and greater description see Dean 2010; D’Altroy 1992, 2015; Hyslop 1990).
73
275
Figure 6.4. Some of the 16 aligned niches, looking northwest.
The second central structure is the Early Colonial Period church, rectangular in shape,
and measuring 35m east/west by 9m north/south. Opposite of the kuraka house, the church has
two possible entryways in the northern wall, and a three-step altar leading to the eastern wall
(Figure 6.5 and 6.6).74 On the western wall, there is evidence of red, black, and cream paint,
which probably are the remnants of a mural. There are also three rectangular windows in the
upper portion of the southern wall of the church. The east and west walls of this structure have
74
One of these possible accessways looks formal, and the other appears to have been more related to a collapse
rather than an intended entrance.
276
very large gables, 7.7m in height. In each of these gables, there is a small circular opening at the
top. Both south of the church and north of the church, there are large, open spaces which likely
were used as plazas, or central ritual locations. Both plazas are semi-circular and make use of the
existing space. The plaza to the north was the central plaza, as it is encompassed by a low wall
(which also divides the kuraka house and the church), and contains a small, rectangular structure
on the western side which was likely an Early Colonial Period chapel.
Figure 6.5. Exterior of the northern wall of the church, photo looking south.
277
Figure 6.6. Three-step altar facing east, with window at the top of the gable.
Since Frank Meddens first published on the site of Iglesiachayoq, a number of scholars
have characterized the central structure (the church or “iglesia”) as a kallanka. A kallanka is a
278
standard Inka structure denoted by a “large rectangular building with an undivided interior
space” (D’Altroy 1992:100; Hyslop 1990) which was primarily used for housing traveling
groups such as militaries, a shelter for visitors during bad weather, or a large area utilized for
ritual and feasting (McEwan 2006:176). Kallankas often had gabled ends and several trapezoidal
doorways, and their interior walls contained trapezoidal niches for storage. They often opened
out onto a central plaza where religious ceremonies could be performed.
The central structure at Iglesiachayoq is a long, rectangular structure with gables which
opens out into a central plaza. With these architectural features, it certainly resembles an Inka
kallanka. There are remains of three rectangular windows and one entrance on the northern
access. As Meddens and Schreiber argue in their 2010 article, as well as Mallco in his 2013
survey report, the long structure is comparable to an Inka niched hall. However, there are several
reasons to suggest that this structure was constructed and utilized as a church. First, Albornoz
emphasized that the people who inhabited Chicha were punished by being forced to assist with
the construction of the church. The construction style of the church is of a roughly-dressed field
stone, with mud mortar on both the interior and exterior walls. This construction style is identical
to other structures of importance at the site, including the kuraka house. It thus appears that the
same population which constructed the rest of the structures at Iglesiachayoq also built the
church. Second, there are no trapezoidal niches in the walls of the church, and there is likely only
one accessway.75 Unlike typical kallankas at other sites, obvious Inka features such as polygonal
ashlar masonry, trapezoidal niches, and multiple trapezoidal doorways do not exist in the
structure at Iglesiachayoq. Thirdly, and perhaps most importantly, there is a three-step altar at the
75
The area of collapse in the eastern part of the northern wall has been considered a second accessway, and future
archaeological evidence can test this. For now, it looks like basic wall collapse to me.
279
eastern end of the church, with a possible altar niche (arched opening) in the short access. We
also found possible remains of the pedestal and bowl used as the baptismal font (Figure 6.7 and
6.8) directly west of the accessway. Finally, the burials located beneath the church floor in both
flexed and extended styles assert that the structure was not intended as a great hall or kallanka,
but rather as the Spanish religious center (Chapter 7). Excavations supported this conclusion, as
very few ceramics were recovered and those that were seemed to be associated with fill rather
than usage.
Figure 6.7. Possible remains of the bowl which contained holy water at Iglesiachayoq.
280
Figure 6.8. Font at Iglesia de Pampachiri. Notice the similarities in shape and size with the bowl
in Figure 6.7.
It is possible that the structure was initially constructed as a kallanka and then
rehabilitated as a church—indeed there are examples of this type of transformation at several
sites across the Andes.76 However, in comparing artifact assemblages at Iglesiachayoq with
assemblages in kallankas at other sites, it is apparent that the structure at Iglesiachayoq differs
greatly from those in other places across the Andes. For example, at Huánuco Pampa,
excavations in an Inka kallanka revealed a massive food-preparation context and a dense artifact
assemblage with more than 12,000 pottery fragments (Morris et al. 2011:42-43). In a second
kallanka excavated at Huánuco Pampa, archaeologists uncovered another dense artifact
assemblage with pottery, animal bone, and weaving tools. At the Inka complex of Oroncota (in
what is today Bolivia), Alconini’s excavations in a kallanka demonstrated the densest
76
For examples see the Qoricancha in Cuzco or the converted kallanka at Vilcashuaman, or the site of Huaytara
281
assemblage of ceramic findings consistent with food preparation for feasting (2018). In contrast
to the recovered artifact assemblages at Huánuco Pampa and Oroncota, similar sites in terms of
time period and occupations, the assemblage at Iglesiachayoq was incredibly sparse. All metal
artifacts were associated with burials, the recovered ceramic totals were meager (n=308
fragments out of 8720 site-wide, or 3.5%, compared with church units making up 11.6% of the
surface area during excavations), and the only recovered faunal remain was classified as canine
and probably modern, totaling a mere 0.022kg in weight.
Archaeological and architectural evidence thus suggests that the long rectangular
structure at Iglesiachayoq was indeed constructed and utilized as a church, in contrast to recent
arguments (Mallco 2013: Meddens and Schreiber 2010). The unwillingness to consider the
structure as a church could be due to an overemphasis in study of the Inka period, and initial
disinterest by scholars in the Early Colonial Period. Similarities in architectural style between the
church and kallankas in other parts of the Andes can be attributed to those Andeans who
constructed the church. Spaniards, disadvantaged by a very small Spanish workforce, had to rely
on native peoples to complete building projects—thus, it is no surprise that the local people who
inhabited Iglesiachayoq constructed the church in the way they were accustomed. Furthermore,
the sparse assemblage inside the structure indicates that it had no previous existence in which it
was utilized as a kallanka or communal space. If the structure had initially been constructed as a
kallanka, then I would expect to find several trapezoidal accessways and niches, as well as
archaeological evidence for feasting or food preparation instead of funerary activities.
Beyond these two central structures and the central plaza, there are several other notable
features in Sector 1. First, east of the church, there is a small, three-structure complex which is
very distinct from other areas of the site. Here, there are three large (average 10m x 5m)
282
structures which are all quadrangular with squared corners. The entryways of these structures
face each other, and they all share a patio or common area. In addition, the whole complex is
surrounded by a low wall with a defined accessway which sets this complex apart from others at
Iglesiachayoq. My initial hypothesis regarding this complex was that it was a later addition to the
site, possibly of the Early Colonial Period, and that those who lived in this region were likely
elites associated with church activities, given the proximity of their houses to the central plaza
and church. This hypothesis is bolstered by the division of structure forms at the site—while the
most common type of structure site-wide was quadrangular with rounded corners, there were
only 10 structures which were quadrangular with squared corners, and nine of these are located
within 100 meters of the church in Sector 1. It is likely that these squared-corner structures were
constructed at different time than the common domestic structure form (quadrangular with
rounded corners).
The kuraka house, church, and large three-structure complex are all grouped together in
the center of the site, and there are several other smaller structures which were likely
administrative in nature. These structures would have been used as places to prepare food and
drinks for activities in the central portion of the site. Beyond this nucleus of ritual activity, the
rest of the structures in Sector 1 are much more dispersed. Sector 1 clearly has the most
quadrangular with squared corners, containing 9 out of 10 of these structures, which are all
clustered within a 100m radius of the central church (see Figure 6.9).
283
Figure 6.9. The 100m radius is in green. The structures in red are the ones with squared corners
distributed throughout Iglesiachayoq.
In general, the rest of the structures in Sector 1 are distributed evenly across the area,
with each structure containing its own patio or small plaza which is encompassed by a low wall.
Unlike the rectangular complex in Sector 1, the rest of these structures could have been
284
individual family units, given their separation from other structures and the amount of land
designated for each structure.
Sector 2
Sector 2 is located northeast of Sector 1, and was constructed on a ridgeline from which it
is difficult to view activities in Sector 1. Sector 2 is known as “Wallpa Wiri” locally, and while
its view of Sector 1 is lacking, it has a much better view of the river and road going from Chicha
to Larcay.77 From the northernmost structures in Sector 2, it is possible to see the entire valley,
and would have been a useful location for monitoring traffic in and out of the valley. There is
one large structure which stands out in Sector 2, which parallels the kuraka house in Sector 1.
This structure is also large, approximately 14m x 8.5m in size, and contains two entryways, one
in the north and one in the south. This structure is quadrangular with squared corners, like the
church, yet also contains Inka cut-stone masonry. This structure also has 16 niches, mirroring the
kuraka house in Sector 1 (Figure 6.10). While I did not place any excavation units in this
structure, I suggest that it could be a second type of kuraka house, perhaps indicating a
hanan/hurin dual organization at Iglesiachayoq, pending excavation in the structure. The north
and south walls of this structure are partially collapsed, and there are several deep pits in the
structure, suggesting that it may have been looted at some point in the past.
Though I have not been able to find a good translation for Wallpa Wiri, I can confirm that “Wallpa” means
“chicken” or “hen.” In her 1987 article “The Chicken in Andean History and Myth,” Linda Seligmann suggests that
“wallpa,” though referring to a European livestock animal, could also have been associated with the Inka king
Atahualpa (142). She also writes that “while some of the original meanings of wallpa remained intact, others were
transformed by unforeseen historical events…the indigenous category of Creator God, one of the primary meanings
of wallpa, became altered by the presence of other viracochas on the horizon, the Spaniards” (Seligmann 1987: 143).
“Wallpa” could also have been altered by the native experience of conquest. Through personal communication with
Linda Seligmann, she offered that “wiri” could have meant “part of a plow, a wide belt, and in the case of aya wiri, a
belt with which to tightly wrap cadavers” (Seligmann pers. Comm 2018).
77
285
Figure 6.10. 16-niched hall in Sector 2, facing west.
There are 27 structures in Sector 2, which are divided into two ovoid structures, 24
quadrangular with rounded corners, and one quadrangular with squared corners. Like Sector 1,
these structures are dispersed across the wide expanse, each structure containing its own
respective patio with enclosing wall. Much of the structures in Sector 2 are more poorly
preserved than in Sector 1, having been used either for growing crops or animal husbandry in
modern times. Prior to beginning excavation at Iglesiachayoq, my hypothesis for Sector 2 was
that it was built contemporaneously with Sector 1, after construction of Sector 3. Excavation
supported this hypothesis, as occupation levels in Sector 2 were extremely short, with only one
occupation level ~30cm below the surface. For whatever reason, the structures in Sector 2 were
constructed rapidly and only briefly occupied before the site was abandoned.
286
Sector 3
Sector 3 is located directly south of Sector 1 and southwest of Sector 2. From Sector 1
moving south, there is a large, steep hill, which flattens out into a plateau. The majority of the
structures in Sector 3 are located on this raised plateau above Sector 1, with some located even
farther south outside the core area of Iglesiachayoq. Sector 3 contains 34 of the registered
structures, which can be broken down into 16 ovoid structures, no quadrangular with squared
corners, and 18 quadrangular with rounded corners. The locations of structures in Sector 3 are
more restricted in nature—that is, they are hidden and blocked from view by various rock
outcrops or natural geologic features such as small hills and peaks. Before excavation, I
hypothesized that this sector of the site was older, and likely contained a segment of the
population with less wealth or resources, given the average size of the structures and their
construction quality. Sector 3 is contains the most ovoid structures—the local Soras style of
building—thus suggesting that this portion of the site was either constructed at an earlier time or
that building forms followed local Soras traditions. In addition to the structures, there are several
small, walled plazas with communal access. There is a waterfall a short hike south of Sector 3,
which could have been utilized by the people in this sector as they needed it. There is also a
small modern lake which is used for animal grazing today—while I don’t know if this lake
existed in the fifteenth and sixteenth centuries, the depression of the landscape in this area
suggests that it could have filled with water during times of heavy rain.
A dominant feature of Sector 3 contains two circular structures whose entryways face
each other. These structures are an average of 8m in diameter, and are encompassed by a large
wall and two natural rock outcrops. One of them has Inka cut stone architecture lining the
entryway, suggesting that it may have been occupied by a member of the Inka elite. Toward the
287
northernmost portion of Sector 3, there is an enclosed semi-circular patio which is defined by the
high-quality stone wall which was built into the hillside (Figure 6.11 and 6.12).
Figure 6.11. Circular structure surrounded by semi-circular prehispanic/local plaza.
Figure 6.12. View of the northern retaining wall of the prehispanic/local plaza. Notice
incorporation of large stones and inability to see beyond hill.
288
Archaeological Occupations at Iglesiachayoq
Iglesiachayoq has only one clear occupation level—deposits were shallow throughout the
site, with an average unit ending some 40cm below the surface (although church units were
clearly intentionally deposited below sterile soil, with the average church unit terminating at
~92cm below the surface). With the exception of church units, all other units contained a surface
level, one or two cultural levels with similar assemblages,78 and finally, sterile soil. With this
abbreviated vertical occupation, it is difficult to parse levels of occupation which are postSpanish presence, and those that are prehispanic in nature—it is likely that these two occupations
would contain identical assemblages. I propose that in the case of Iglesiachayoq, occupation
would be better understood in terms of horizontal stratigraphy. Horizontal stratification is widely
used in historical archaeology and has been implemented recently in other studies with shallow
and short occupations (Hernandez and Osores 2019). For example, in the Lurín Valley, Carla
Hernández utilizes horizontal strategies of site analysis in lieu of shallow deposits and
overlapping phase-based occupations, albeit during the Late Intermediate Period to Late Horizon
transition. In the case of Iglesiachayoq, Inka and local actors erected the site in the late 1400s
and lived there for around 50 years until the Spanish likely arrived. A known site of Taki Onqoy,
Iglesiachayoq was inhabited into the 1570s (as discussed by Albornoz 1990 [1574]), and its
population was relocated to one of three reducciónes in the region: Pampachiri (1.2km east),79
San Pedro de Larcay (~6km northwest), or Soras (~13km northwest) during Viceroy Toledo’s
reform and resettlement campaigns beginning in the 1570s. With such concrete dates of
78
The first cultural level usually contained a much higher density of artifacts, while the second cultural level had
much less, although all cultural levels appear to have been deposited in a continuous occupation at Iglesiachayoq.
79
Soras or San Pedro de Larcay are the most likely locations since they were both a part of the Soras repartimiento.
This repartimiento had a natural border at the Chicha-Soras River. Although Pampachiri is the closest town by
distance, the river serves as a natural frontier.
289
settlement and abandonment as conferred through ethnohistoric sources (and confirmed through
my excavations and church radiocarbon dates), it is more useful to consider the site as it
expanded rather than its local reutilization of the same spaces over time.
Architectural Styles and Forms
There were three distinctive structure forms at Iglesiachayoq, including ovoid (n=23),
quadrangular (n=10), and quadrangular with rounded corners (n=58) (Table 6.4). All structures
at Iglesiachayoq consisted of uncoursed, roughly dressed fieldstone which was locally obtained
in the valley (Figure 6.13).80 Occasionally, Inka polygonal ashlar stone masonry was utilized in
construction of the accessways for several large structures (n=9, or ~10% of structures), although
these specialized stones were typically stacked vertically rather than in courses (Figure 6.14).
During the architectural survey I completed in 2013 and 2014 (Chapter 5), my team and I
recorded quantitative measurements for every structure, including both interior and exterior
dimensions. Ovoid structures were more or less circular, and lacked a uniform directionality for
the entrance (Figure 6.15). The average diameter of the ovoid structures was 6.38m, with a
standard deviation of 2.34m, and the interior dwelling space averaged 37.73m². Ovoid structures
tended to be more poorly preserved than other types of structures, and often only the foundations
remained of this structure form. Of the intact walls, the average height was 2.24m and of the 23
ovoid structures, five had niches located in interior spaces. All ovoid structures lacked stucco
either on the inside or the outside.
80
This masonry style is consistent with other late prehispanic masonry styles in other regions, see Type 2 Masonry
from Kohut 2016, described as “some worked/selected fieldstone, no coursing, dressed corners and heads”:187.
290
Sector Number
Ovoid
Quadrangular
Total Number of
Structures
9
Quadrangular
with Rounded
Corners
16
1
5
2
2
1
24
27
3
16
0
18
34
30
Table 6.4. Structure forms by sector at Iglesiachayoq.
Figure 6.13. Typical uncoursed masonry style with faced stones lining the entrance, photo facing
north.
291
Figure 6.14. Inka cut-stone architecture stacked vertically lining the entrance of an important
household, photo facing north.
Figure 6.15. Example of a well-preserved ovoid structure with access to the northeast.
292
In total, there were only ten quadrangular structures, and nine out of these structures were
clustered in the Spanish center of the site. Quadrangular structures were generally of a larger size
(average size 11.71m x 5.81m) and housed a larger average dwelling area than ovoid structures,
76m² (Welch two sample t-test t=-1.7039, df=9.7678, p-value=0.1199).81 These structures were
better preserved and their wall heights indicate more intensive investment in structure
construction, with average wall height at 2.67m. Notably, the only structure with gables at
Iglesiachayoq is the central church, and the gables measure 7.7m from ground level (Figure
6.16).
81
Although this t-test was not significant, this is likely due to a combination of small sample size and the outlier of
the church dwelling area. I ran the same test with a trimmed mean, with similar results.
293
Figure 6.16. Church with visible gables, photo facing north.
The most ubiquitous structure form recorded at Iglesiachayoq was quadrangular with
rounded corners, and this form seems to have been the standard domestic structure form. The
average wall height of quadrangular structures with rounded corners was 2.41m, while the
average size overall was 6.88m x 5.15m with an average interior dwelling space of 37.14m².
Quadrangular structures with rounded corners are thus more similar in size and dwelling area to
ovoid structures, while quadrangular structures with squared corners indicate a greater size,
dwelling area, and degree of investment in construction than the other two structure forms (Table
6.5, Figure 6.17 and 6.18).
294
Structure Form
Average Wall Height (m)
Average Dwelling Area (m2)
Ovoid
2.24
37.73
Quadrangular
2.67
76
Quad. With Rounded Corners
2.41
37.14
Table 6.5. Comparison of average wall heights and dwelling areas by structure form, n=91
structures.
Figure 6.17. Boxplot comparing the three structure forms by dwelling area, ovoid n=23;
quadrangular with rounded corners n=58; quadrangular n=10; total population with 100%
coverage n=91.
295
Figure 6.18. Boxplot comparing the three structure forms by wall height, ovoid n=23;
quadrangular with rounded corners n=58; quadrangular n=10; total population with 100%
coverage n=91.
Ceramic Vessel Types and Pastes
Diagnostic sherds were sorted into vessel form categories of unrestricted and restricted
forms, and where possible, broken down even further into categories including ceremonial wares
(aribalos), storage wares (cantaros, jars, bottles), cooking wares (ollas), feasting/serving wares
(cups, bowls, plates, escudillas, tazones), figurines, and pirurus.
General paste types were identified while analyzing all diagnostic sherds after a fresh
fracture using a hand lens. Each diagnostic ceramic was evaluated according to texture of paste,
composition or inclusions of paste, color and Munsell value, oxidation, size of temper, and the
consistency of the fracture pattern. There were some shared commonalities amongst all sherds—
the vast majority of diagnostic ceramic had inclusions of mica and quartz, with subsidiary
296
inclusions of basalt or feldspar. There were three main paste types which accounted for 91.5% of
the diagnostic sherds, though I also identified four subsidiary paste types which were much less
common. One-hundred and twenty-two sherds were exported for analysis at the Elemental
Analysis Facility at the Field Museum, and statistical analysis is ongoing.
Type A: Paste Type A is the most common ceramic type, making up 64.4% of the
diagnostic sherds (n=367/570) and 51.2% of the overall sample (n=4461/8720). Paste
Type A is medium or fine-grained in texture, with medium-fine or fine inclusions. The
paste is generally of high quality, and inclusions consist of mica, basalt, and quartz. In
general, the paste colors range from reddish-yellow to reddish-brown to brown. Paste
Type A was the typical paste type for locally-made ceramics.
Type B: Paste Type B is a variant of Paste Type A, but a more “rustic” type. Paste Type
B is the second most common paste type, making up 14.6% of the diagnostic sherds
(n=83/570) and 46.7% of the overall sample (n=4074/8720). Paste Type B has the same
range of colors and types of inclusions as Paste Type A, but it has a much cruder texture
with larger grains and inclusions.
Type C: Paste Type C is the third most common paste type recovered at Iglesiachayoq,
making up 12.5% of diagnostic sherds (n=71/570) and 1.4% of all total sherds
(n=123/8720). Paste type C was either fine or very fine in both texture and inclusion size,
and inclusions consisted of mica, quartz, and basalt. The main identifying feature of this
paste was the pink or cream color of the paste. Fragments with Paste Type C were more
likely to have both slip and decoration, indicating their elevated proportion in the
diagnostic wares.
297
Type D: Paste Type D was only identified in diagnostic sherds at 1.4% of the diagnostic
sample (n=8/570), and made up a very small subsection of the assemblage. Type D has a
fine texture and fine inclusions with a pale red color, but has large and numerous mica
inclusions. Compared to Types A, B, and C, Type D is very distinctive in its color and
the size and density of these mica inclusions. Type D ceramics were only found in Sector
1, in the kuraka house and in the church units.
Type E: Paste Type E was also only represented in diagnostic ceramics (n=19/570 or
3.3% of the diagnostic sample) and was characterized by a very fine texture almost
devoid of inclusions. The main identifying characteristic of Type E is that these ceramics
took on a laminate quality when fractured, and appeared more like rock than ceramic.
They were overfired, and the interior and exterior surfaces often took on a vitrified
appearance with bubbles or a metallic sheen. The sherds which have this paste are very
rustic and roughly-hewn, sometimes with added plastic applique or etched design. Type E
ceramics may correspond with Soras-style overfired pottery as defined by Meddens and
Vivanco Pomacanchari (2018).
Type F: Paste Type F was also only recovered as a portion of the diagnostic Ceramic
assemblage (n=15/570, or 2.6%). Paste Type F is similar to Type E in that the paste is
almost devoid of inclusions and appears laminate when broken, but these ceramics were
well-crafted and almost exclusively found in Sector 1. All of the Type F sherds had either
slip, decoration, or both, and 33% were recovered in church excavations. The color of
Paste Type F was generally very dark gray.
Type G: Type G was the last identified paste type, and was characterized by mediumtextured grains with large inclusions of mica, basalt, quartz, and feldspar. The paste itself
298
was bright red in color, and set apart by the high concentration of large feldspar
inclusions. Paste Type G was the least common ceramic type recovered from excavations
(n=7/570 or 1.2%) and was only recovered in diagnostic ceramic sherds.
Excavation Results
In the following section I will broadly present the excavation results from each unit (see
Appendix F for unit summaries). In general, ceramic found in Sectors 1 and 2 were of a higher
quality than artifacts found in Sector 3.82 Along with the differences in building construction,
there were also major differences between sectors in activities taking place and status of people
occupying the structures. After the sections regarding general excavation and sector-wide
contrasts, I will also give a comparative discussion of site-wide differences between the three
types of structures, and how these differences may reflect the uses of each type of structure.
Sector 1 Domestic Structure Excavations
As mentioned above, the majority of excavation units were placed in Sector 1. The
reasoning behind this is that Sector 1 was much more varied in types of structures and the
functions of those structures. Furthermore, as a major area of site-wide activity in which people
from all sectors would have participated in events in Sector 1, it is crucial to understand the
nature of those activities and how they may have reflected changing religious contexts.
Additionally, the kuraka house is located in Sector 1. Since the kuraka, Joan Hachi, was
82
Finer paste, more decoration, etc.
299
implicated in Albornoz’s list of takiongos, I had planned to complete a full excavation of his
residence at Iglesiachayoq.
Unit 1 Excavations: Intermixing and Interconnection between Inka and Spanish
Unit 1 was the unit located in the kuraka house, and it was the domestic unit with the
greatest excavated volume (57.6 cubic meters and extensions). The excavation unit was
strategically placed to encompass the maximum amount of space in this structure in order to
learn about the activities practiced by the kuraka and/or his family. Given the size of the
structure (14m x 11m), we were able to place a 12m x 9.6m unit, which we then divided into 12
separate 3m x 3.2m quads. Because Unit 1 had such a massive size, we were unable to excavate
all of these quads—to control for the large size, each quad was excavated as a separate individual
unit with its own set of loci. The quads were numbered from 1 to 12 starting with the northwest
quad and moving across the unit and down, as if reading a text (Figure 6.19). In general, Unit 1
was covered in icchu grass and had surface damage from the presence of cows and other herding
animals. However, there were no pits or below-surface areas which appeared to be damaged or
looted. In his dissertation research, Meddens (1985) excavated a unit abutting the entrance to the
kuraka house and identified Inka cut-stone masonry wall foundations and ceramic which was
either Late Horizon or Early Colonial Period.83
Meddens writes: “Surface collection at this site produced some lithics and very little pottery. A few test pits were
excavated here, but the material recovered was not stratified. All was of Late Horizon, and some possibly even of
early colonial date. This dating appears accurate, considering the structural remains. The largest rectangular building
may be of early colonial date (pers. comm. J. H. Rowe), whereas some of the others, like the ones with polygonal
ashlar masonry, date to the Late Horizon” (Meddens 1985:103).
83
300
1
2
3
4
5
7
8
6
9
10
11
12
Figure 6.19. Layout of quads in Unit 1. Fully excavated quads included 1, 3, 5, 7, 8, 10, and 12.
Quad 1 was located in the northwestern corner of the unit and Quad 3 was located in the
northeastern corner of the unit. I will discuss these two quads first because they had a similar
301
stratigraphic pattern which was different from the rest of the unit. In general, there were two
cultural layers. The first cultural layer was located just below the surface and contained the last
occupation of the structure. In Quad 1, we found a caret-head iron nail associated with an even
mix of possible Early Colonial Period wares and local Inka (red-slipped) wares.84 Caret-head
nails are a Spanish artifact which date to the first half of the sixteenth century (Flint and Flint
2003:153; Wernke 2013)(Figure 6.20). The presence of the caret-head nail along with the
prevalence of Inka-period ceramics indicates the continued use of this structure through the Early
Colonial Period. Additionally, the nail suggests that the occupant of this structure had access not
only to Inka goods, but also to Spanish goods—as mentioned earlier in this dissertation, this
person was likely Joan Hachi, the kuraka of Chicha. Near the caret-head nail was an animal
effigy handle, which is a characteristic of both Inka and Late Intermediate Period (Chicha-Style)
plates (Figure 6.21, Meddens and Vivanco Pomacanchari 2018: 15). In this case, the handle was
either a representation of a fish or a bird.
84
I was unable to distinctively identify Early Colonial Period ceramic artifacts—there was neither glazed nor wheelthrown ceramics. Some of the ceramic recovered in Sector 1 appeared to be of a different form than was typical for
the Inka or the Soras, so we tentatively classified these as Early Colonial Period.
302
Figure 6.20. Caret-head nail found in the northwest corner of the kuraka house.
Figure 6.21. Anthropomorphic mango handle from a plate.
In Quad 3, there was evidence of both camelid, and other unidentified artiodactyla (pig,
sheep, cow, goat, or camelid), also associated with local Inka ceramics. Below this first level, we
found another cultural level which was marked by less ceramic and cultural artifacts, and ended
when we began to hit sterile soil. We did not recover any evidence of a prepared floor, which
303
could have been caused by the acidic soil and poor preservation at Iglesiachayoq. In this northern
portion of Unit 1, sterile soil was reached quickly, and the soil matrix of this sterile earth was
different here than in the southern portion of the unit. Namely, in quads 1 and 3, we found a
dense clay matrix completely absent of evidence of cultural activity. These clays were brown,
yellow, and red in color and the level was nearly impossible to dig through and fairly uneven.
The various colors of clay matrix were not present in the rest of the unit, where sterile soil
appeared to be linked to an absence of materials and dense soil rather than multi-colored clays.
Moving south through Unit 1, the next quad we excavated was quad 5. Quad 5 was
located in the center of Unit 1 and had one of the most remarkable finds in our entire excavation.
Just below the surface level in the first locus, we uncovered a beautiful, fragile cross which
contained a date of 1892 embossed on it (Figure 6.22, 6.23, and 6.24).
Figure 6.22. Front side of 1892 cross.
304
Figure 6.23. Reverse side of the 1892 cross.
Figure 6.24. Drawing of 1892 cross.
305
There are several intriguing details related to this cross: first, it was found in the center
quad of our unit, more or less in the center of the structure and near the surface. This placement
indicates that the cross may have been intentionally deposited at a later date in the structure,
suggesting that the site held continued importance or sacredness for the region in later centuries.
This cross contains text suggesting that the saint is Saint Anthony of Padua holding the Christ
Child, and says “100 days of indulgence,” made during the reign of Pope Leo XIII in 1892. It
appears to be made of copper or aluminum, and the text is in Latin. While I have completed
some background research on this cross, the information I found is located on internet antiquities
sites, which are selling various versions of this cross. It appears to have been a limited edition
cross or medallion which was created during Pope Leo XIII’s reign, and originates from the
Vatican in Italy. Given the date and the limited and foreign production points of the cross, it
seems clear that an important representative of the church may have visited Iglesiachayoq
sometime in the early twentieth century or late nineteenth century. The placement of the cross in
the structure’s center suggests it was deliberately deposited, as does its location near the
sixteenth-century Catholic church.85
The ceramic found in quad 5 contained predominantly Inka-style ceramics. Namely, there
were two rim fragments from Inka aribalos, both which had polychrome red-slipped exteriors
and borders. The Inka aribalo is one of the most representative Inka wares, and was used for a
variety of purposes, often associated with chicha—it is identified by its long neck and flared
opening, along with a globular body and conical base (Bray 2003; D’Altroy 2015). The rest of
85
Further research on this cross demonstrated that it was founded as an intentional religious token in honor of
Columbus’s quadricentennial. I am working on a forthcoming article about this cross and what its significance was
both at Iglesiachayoq and globally.
306
the ceramic from this quad was identified as either ollas (large cooking pots), serving pots,
bottles, or cups.
We next excavated quads 7 and 10, which I will discuss together given their contextual
associations with one another. Quads 7 and 10 were located in the southwestern corner of Unit 1,
and had a much higher density of ceramic, lithic, and animal bone than the rest of the quads in
Unit 1. Both of these quads contained evidence of both artiodactyla and perissodactyla—camelid
and cow, horse, or donkey, which were intermingled. One of these bone fragments which was
identified as either camelid, horse, or donkey, was part of a heel bone which showed two distinct
cut marks, indicating the bone had likely been butchered. The ceramics found in quads 7 and 10
were a combination of local Inka and Early Colonial Period wares, and represented mostly
serving or feasting vessels. Like quad 1, we recovered another effigy handle, which was also in
the shape of a fish or a bird (Figure 6.25).
Figure 6.25. Fish head mango characteristic of Inka or Soras plates.
Like the figurine handle found in quad 1, this ceramic in relation with the Old World
animal remains found in the unit suggests a continued usage of the structure in the Early Colonial
period. Furthermore, the slipped handles show a degree of fineline painting and polychrome
technology characteristic of more elite households during the Inka period.
307
The last quad excavated in Unit 1 was quad 12, which was in the southeastern most
corner of the unit. Here, we collected the highest density of diagnostic ceramics from this unit,
which together represented a mixture of large pots, aribalos, and cups in local Inka styles.
Taken together, the artifact and context findings from Unit 1 show a period of mixed
usage—Inka artifacts (aribalos, polychrome ceramic, animal effigy handles) and Spanish
artifacts (iron caret-head nail, Old World animal bones, and possible Early Colonial Period
ceramic fragments) were found in the same occupation levels. The most common ceramic forms
were cooking wares (n=36/141, or 25.5% of the vessels found), feasting wares (bowls and plates,
n=32/141, or 22.7% of the vessels found), and restricted storage wares (n=22/141, or 15.6% of
the vessels found) (Table 6.6). The majority (n=110/141 or 78%) of ceramic paste types were
Type A (Table 6.7). In Unit 1, we recovered 20% of the MNI amongst all excavation units
(n=6/30). Of these fauna, two could be confidently identified as camelidae (Andean, n=2/13
individuals site-wide), two as cow or horse/donkey (n=2/2 individuals site-wide), and two
artiodactyla or perissodactyla (n=1/1 individuals site-wide) (Table 6.8). Notably, all examples of
horse or donkey were found in Unit 1. The combination of the artifacts from these two periods
suggests a pattern of continued usage from prehispanic to Early Colonial periods.
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Table 6.6. Vessel types in Unit 1.
308
Number of Vessels Percentage
10
7%
22
15.60%
36
25.50%
32
22.70%
2
1.40%
2
1.40%
37
26.20%
141
100%
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
110
78%
3
2.10%
14
10%
6
4.30%
6
4.30%
2
1.40%
0
0%
141
100%
Table 6.7. Paste types in Unit 1.
Taxon
Common Name
Cavia porcellus
Canidae
Artiodactyla
Camelidae
Bos taurus
Perissodactyla
Bos taurus or
Perissodactyla
Artiodactyla or
Perissodactyla
Domestic guinea pig
Canines
Even-toed ungulates
Camelids
Cow
Horse/donkey
Cow or horse/donkey
Even or Odd-toed
ungulates
Number of
Individuals
0
0
1
2
0
1
1
Percentage (sitewide)
0%
0%
10%
15.40%
0%
100%
100%
1
100%
Table 6.8. Unit 1 recovered fauna by Minimum Number of Individuals (MNI) and site-wide
percentages.
Unit 2 Excavations: Elite Domestic Household
Unit 2 was located in the easternmost structure of the small, squared-corners complex in
Sector 1. As briefly mentioned above, these three structures were all roughly the same size, some
10m in length and 5m in width, making them the largest residential structures in Sector 1 other
than the kuraka house. The three structures each had accessways which faced inward and thus
one another, and the entire complex of the three structures was encircled by a divisive wall
309
which contained only one restricted accessway. The three structures appear to have been of later
construction than those in Sector 3, given the difference in their construction form and confirmed
by the shallow occupation unearthed during excavation.
Unit 2 was an 8m x 4m unit which we divided into eight separate quads in order to
maintain a degree of control in excavation, each quad measuring 2m x 1m. We numbered the
quads from the Northwest quad to the Northeast quad (1-2), and then moved south. Unlike Unit
1, Unit 2 utilized the same set of loci for all quads, and all quads were excavated fully to the best
of our abilities. Unit 2 was excavated to a depth of 32cm below the surface, upon which we hit
sterile soil. This unit contained few notable artifacts or contexts.
Unit 2 seems to have been part of a domestic complex. While we did not find clear
preserved evidence of a hearth, we did find many small lithic tools including four basalt knives
and an andesite grindstone. The grindstone and knives suggest that food preparation may have
happened in this household. As in the large “casa de kuraka,” we found the third and final animal
effigy plate handle in Unit 2. Like the other two, this was also in the shape of a fish or a bird,
depicted below (Figure 6.26).
Figure 6.26. Example of mango from Unit 2.
310
The rest of the diagnostic ceramic in Unit 2 contained several fragments which were
characterized as Type E (Table 6.10), and were overfired to the appearance of having metallic or
blistered glassy surfaces on the exterior, interior, or both sides and came in a variety of colors
(Figure 6.27). Meddens and Vivanco Pomancanchari’s 2018 article suggests that similar styles of
overfired or cindered surfaces are known as a variant of the “Soras style” in the Chicha Soras
Valley which “appears to be a late phenomenon” (2018: 18).86 The most common vessel forms
were feasting wares (n=12/45, or 26.7%), restricted storage wares (n=9/45, or 20%), and cooking
wares (n=5/45, or 11%), thus suggesting that the primary activities taking place in this structure
were related to drinking and eating (Table 6.9). While the most common paste type was Type A,
Unit 2 diagnostic ceramics had elevated proportions of Type C (n=12/45, or 27%, pink or cream
fine-grained paste with inclusions of mica, quartz, and basalt) and Type E (n=5/45, or 11%,
overfired, paste devoid of inclusions with laminate quality).
Figure 6.27. Blistered/vitrified ceramic fragments with Type E paste from Unit 2.
86
No years or time periods were given for this type of ceramic style, and more research needs to be completed to
create a more detailed ceramic sequence.
311
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels Percentage
0%
0
20%
9
11.1%
5
26.7%
12
2.2%
1
0%
0
40%
18
100%
45
Table 6.9. Vessel types in Unit 2.
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
25
0
12
0
5
2
1
45
55.6%
0%
26.7%
0%
11.1%
4.4%
2.2%
100%
Table 6.10. Paste types in Unit 2.
Finally, of all of the animal bone fragments found (n=13, weight = 40.43g), Kennedy was
able to determine an MNI of two, with one camelid and one specimen which could only be
identified to the order of artiodactyla (Table 6.11). With such a small sample of faunal remains, I
tentatively suggest that those individuals residing in Unit 2 were likely continuing prehispanic
diet standards (camelid). In sum, Unit 2 was located in a large rectangular structure with squared
corners which had a very short vertical provenience, suggesting a brief occupation. The artifacts
recovered in Unit 2 indicate that the structure was primarily used for domestic purposes, while
the location and restricted access to the structure suggest that a family of higher status may have
resided here.
312
Taxon
Common Name
Cavia porcellus
Canidae
Artiodactyla
Camelidae
Bos taurus
Perissodactyla
Bos taurus or
Perissodactyla
Artiodactyla or
Perissodactyla
Domestic guniea pig
Canines
Even-toed ungulates
Camelids
Cow
Horse/donkey
Cow or horse/donkey
Even or Odd-toed
ungulates
Number of
Individuals
0
0
1
1
0
0
0
Percentage (sitewide)
0%
0%
10%
7.7 %
0%
0%
0%
0
0%
Table 6.11. MNI of fauna and site-wide percentage for Unit 2.
Unit 3: Local Ritual?
Unit 3 was a 4.5m x 4m unit placed in an ovoid structure in which the access faced north
and away from the church. This structure was relatively isolated and situated within its own
terrace. Unit 3 presented a different collection of artifacts than either of the previous two units.
First, there was no evidence of Early Colonial Period artifacts, with the exception of two
fragments of cow bone (Kennedy 2016). However, these fragments were found in the surface
level of the unit and were less weathered than the rest of the animal bone found during
excavations. Thus, I suspect that these two bones are an example of modern influence (especially
given the large amount of cow present at the site, and the open nature of where this unit was
placed).
In fact, Unit 3 presented less diagnostic ceramic than any other household unit, with only
19 diagnostic sherds out of 502 total sherds (Figure 6.28). Units 2 and 3 had the least amount of
ceramic by weight than any other household unit (Figure 6.29). Diagnostic sherds consisted
primarily of feasting wares (n=10/19, or 52.6%) and cooking wares (ollas, n=5/19, or 26.3%),
313
along with one figurine (Table 6.12). The figurine found appears to be a set of legs crossed over
one another, a braid, or another unidentified form (Figure 6.30). It is only in Unit 3 that we have
any examples of ceramic in the shape of a figurine that was not originally part of a vessel. Paste
types were almost all A, B, or C, with the exception of two Type G sherds (Table 6.13). Of all of
the fragments in Unit 3, only one had decoration, which was an orange stripe on the exterior.
Figure 6.28. Sherds per excavated cubic meter in ten household units. Notice Units 2 and 3 with
the least material.
314
Figure 6.29. Weight in grams of ceramic per excavated cubic meter in household units.
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels
0
2
5
10
1
0
1
19
Percentage
0%
10.5%
26.3%
52.6%
5.3%
0%
5.3%
100%
Table 6.12. Vessel types, counts, and percentage of unit assemblage in Unit 3.
315
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
14
2
1
0
0
0
2
45
73.7%
10.5%
5.3%
0%
0%
0%
10.5%
100%
Table 6.13. Paste types from Unit 3.
Figure 6.30. Figurine from Unit 3.
Unit 3 excavations recovered lithic scrapers, points, and debitage fragments made of
andesite, obsidian, and basalt. We also recovered one unidentifiable lithic fragment which was
slightly retouched and of a different color (reddish brown) and texture (waxy) than any of the
other lithics found at Iglesiachayoq (Figure 6.31). I have tentatively identified this lithic as silex,
but further research must be completed to confirm it.
316
Figure 6.31. Lithic possibly made of silex.
Unit 3 was difficult to interpret—while the ceramic assemblage seems typical for
residential structures, the presence of the figurine and rare lithic materials could complicate this
interpretation.
Unit 4 Excavations: Food Preparation for Church Activities
Unit 4 was a 5.5m x 3m unit placed in EST-004. This structure was quadrangular with
squared corners, and was likely associated with food preparation for ritual activities occurring at
the Spanish religious center of the site (near the church). Unit 4 was located due west of the
church, south of the kuraka house, and had its accessway facing the church and central plaza. We
collected a very high density of ceramic from Unit 4, including 254 diagnostic sherds (separated
into 75 separate vessels) out of a total of 1,058 sherds, and a total weight of collected ceramic of
16.69 kg (Figure 6.29). This unit had the most ceramic sherds per cubic meter, at 178.1
sherds/m3.
317
The surface of Unit 4 was covered by wall collapse from the structure itself. Due to this
wallfall, the preservation in Unit 4 was much better than in many other units—we recovered
evidence of a pressed mud floor in this unit. It was also one of the only units which had a
definitive hearth with in situ evidence of burned wood, bone (camelid), and corn (Figure 6.32).
Figure 6.32. Carbonized maize and camelid vertebrae in situ.
The hearth lens was 1.85m x 1.50m in size and was over 10cm in depth. Ceramic
fragments were not collected from the hearth itself, but were scattered throughout the rest of the
unit. Not only was this unit dense in ceramic, but also in animal bone. We collected the highest
amount (n = 359 fragments, weight = 2.83 kg) of animal bone from any of the units in Unit 4,
which had an MNI of two camelids and two artiodactyla (even-toed ungulates) (Table 6.14). The
predominance of prehispanic foodstuffs in cooking contexts is not unusual. As deFrance et al.
2016 write regarding faunal consumption at Malata (a sixteenth-century doctrina in the Colca
Valley), “the zooarchaeological evidence from a variety of contexts in this doctrina is
unambiguously consistent with the persistence of prehispanic indigenous domestic productive
practices similar to those reported for other mid-elevation and highland sites dating to the Middle
Horizon and later periods” (deFrance et al. 2016: 301). That is, Spanish authorities and Spanish
presence were known to have a variable effect on indigenous and Inkan and indigenous
318
foodways: in some cases exotic or Old World items were imported, while in others, prehispanic
traditions continued to flourish (see Kennedy and Van Valkenburgh 2016, deFrance 1996, 2003,
2012; see also Norman and Kennedy 2019 edited volume for more discussion of changes in
foodways after Spanish arrival). All of the bone collected from Unit 4 was either juvenile or subadult, indicating that the inhabitants of Iglesiachayoq were consuming food from animals which
had not reached their full caloric potential. These young animals would have been choice cuts of
meat, usually reserved for high-status individuals.87
Taxon
Common Name
Cavia porcellus
Canidae
Artiodactyla
Camelidae
Bos taurus
Perissodactyla
Bos taurus or
Perissodactyla
Artiodactyla or
Perissodactyla
Domestic guniea pig
Canines
Even-toed ungulates
Camelids
Cow
Horse/donkey
Cow or horse/donkey
Even or Odd-toed
ungulates
Number of
Individuals
0
0
2
2
0
0
0
Percentage (sitewide)
0%
0%
20%
15%
0%
0%
0%
0
0%
Table 6.14. Faunal MNI recovered from Unit 4.
The results of the lithic recovered in Unit 4 were also distinctive—artifacts included three
basalt knife blades, three obsidian biface fragments, and three pieces of white/tan chert. Two of
these chert pieces were beautiful knife blades, and were unique considering the preponderances
of basalt, andesite, and obsidian across the rest of the site (Figure 6.33). The overall lack of chert
87
Though there are no universal correlates for age of animal at death and cut of meat, some have suggested that
younger animal at death would create more tender meats and perhaps indicate higher status and/or more wealth
(deFrance 2009; Twiss 2007).
319
at Iglesiachayoq suggests that this material may have been imported, and likely represented a
higher-status (less access) good. Notably, we did not find any grindstones in Unit 4—this could
indicate that while food was definitely being cooked and served from this structure, the difficult
grinding of grains, spices, etc. could have taken place elsewhere.
Figure 6.33. Chert blades from Unit 4.
As mentioned briefly above, Unit 4 had the most ceramic out of all of the units. Of the
diagnostic ceramic in Unit 4, we identified pieces of at least eleven separate aribalos, two which
were particularly notable (Figure 6.34).
Figure 6.34. Red and cream aribalo (left), red and black aribalo (right)
320
The remains of these two aribalos were some of the finest ceramics recovered during
excavations, and indicate the importance of the activities in which they were utilized. These
aribalos likely would have been used to transport or store liquids, and may have been serving
vessels for activities near the church. We also recovered 13 ollas and 21 storage vessels (Table
6.15). Although these vessels had openings of various sizes, the largest of them were over 35cm
in diameter with a brown, burnished exterior (Figure 6.35). Paste types consisted of mostly Type
A (n=58) and Type C (n=11) (Table 6.16).
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels
11
21
13
16
1
3
10
75
Table 6.15. Numbers and Percentages of Vessels from Unit 4.
Figure 6.35. Two ollas with diameters over 35cm.
321
Percentage
14.7%
28%
17.3%
21.3%
1.3%
4%
13.3%
100%
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
58
2
11
0
1
2
1
75
77.3%
2.7%
14.7%
0%
1.3%
2.7%
1.3%
100%
Table 6.16. Paste Types from Unit 4.
Taken together, the ceramic data from Unit 4 suggest that this space was used for largescale food preparation. The hearth full of camelid bone, the large size and volumes of the pots
and cooking vessels, and the structure’s proximity to the ceremonial Spanish center indicate that
those who utilized this structure were preparing massive quantities of food and drink for feasting
and group consumption.
Sector 2 Domestic Structure Excavations
The excavations from Sector 2 demonstrate a different pattern of usage than those from
Sector 1. Removed from the site center, the structures in Sector 2 were primarily residential, and
represent a portion of the site which was occupied for less time than the structures in Sector 3.
Three structures were excavated in Sector 2: all were rectangular with rounded corners, and each
unit was strategically placed to encompass the altitudinal variations in this Sector. In general, the
occupation levels in Sector 2 were very brief—all of these structures had one occupation level
which contained high densities of ceramic, with a second level that was nearly devoid of
artifacts. All units in Sector 2 were 4.5m x 3m and placed to encompass the largest surface area
of each structure—each of these structures was approximately 6.5m x 5m in size. The average
322
total depth of excavation levels from Sector 2 was 40cm from surface to sterile, indicating the
brief occupation period in this sector of the site.
Unit 5 Excavations: Domestic Life
Unit 5 was excavated in the southeasternmost structure of Sector 2, and was an isolated
quadrangular structure with rounded corners that had its own terrace, with its accessway facing
northwest (Figure 6.36).
Figure 6.36. Unit 5 after excavations, showing the shape of the structure.
There was remarkably few fragments of diagnostic ceramic in Unit 5, only 12 out of a
total of 455 collected sherds (see Figure 6.28). Of the diagnostic sherds, the assemblage
consisted of mostly feasting wares and ollas (although with smaller openings, average of 16cm in
diameter, a great distinction from the large size of the openings in Unit 4, Table 6.17). The
323
majority of the nondiagnostic sherds were roughly-hewn brown fragments which had incomplete
oxidation, soot (hollin) or burn marks on the outside or both surface, and treated with a rough
finish in which striation marks were visible, of Type A (n=6) or Type C (n=3) (Table 6.18). All
ceramics appear to have been in local Soras style. The ceramic assemblage from Unit 5 presents
a very different activity profile than that of Unit 4—rather than cooking large amounts of food or
preparing lots of chicha for activities on a group level, the assemblage from Unit 5 instead
suggests that a small family lived here, preparing and consuming food within the household.
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels
0
1
4
7
0
0
0
12
Table 6.17. Vessel Types and Counts from Unit 5.
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels
Percentage
50%
8.3%
25%
0%
8.3%
0%
8.3%
100%
6
1
3
0
1
0
1
12
Table 6.18. Paste types from Unit 5.
324
Percentage
0%
8.3%
33.3%
58.3%
0%
0%
0%
100%
The animal bone recovered from Unit 5 included a minimum number of one even-toed
ungulate and one camelid. All of the fragments recovered were from adult animals, again
demonstrating the difference between the faunal assemblage from Unit 4. Kennedy documented
15 separate cut or saw marks from the animal bone in this unit—the most of any unit—
suggesting that those individuals who resided here were involved with animal butchery and
consumption.
We recovered the head of a metal tupu in the second level of Unit 5. This tupu head
corresponds to similar styles found in the church excavations (Chapter 7). These types of tupus
were traditionally used by Andean women to secure shawls or other clothing. The tupu recovered
in Unit 5 was made of thin copper (Figure 6.37).
Figure 6.37. Metal tupu head recovered in Unit 5.
Finally, I will briefly discuss the lithic assemblage from Unit 5. No diagnostic lithic
pieces were recovered from this unit, and the variety of materials included obsidian, chert, basalt,
and several heavy pebbles which we tentatively identified as hematite (Figure 6.38). Recently, a
hematite mine has been discovered in the Nasca region, and the smooth, polished hematite stones
have been documented in metalworking in the Andes (Vaughn 2013), as only the finest materials
325
were used to process metals and shape them into desired forms. Hematite, which manifests in
both shiny black stones which—in archaeological contexts—are polished and used as mirrors
(for example in Mesoamerica Gazzola 2016) as well as red ore which was utilized for face
painting and ceramic red slips (Gallaga and Blainery 2016).
Figure 6.38. Possible hematite.
In addition to the hematite stones and metal tupu, we also recovered a ceramic fragment
(IC.0602.2.002) which had evidence of metal slag and oxidation on both the interior and exterior
surfaces (Figure 6.39).
326
Figure 6.39. Ceramic fragment with oxidation.
The occupational profile of Unit 5 suggests that the structure was utilized as a private
residence, and that some metalwork may have taken place. The ceramic assemblage is domestic
in nature, with small cooking vessels, while the presence of the pendant and array of hematite
stones suggests that those who resided in the structure may have participated in metallurgy.
Unit 6 Excavations-Domestic Life
Like Unit 5, the artifact assemblage in Unit 6 was rustic in nature, and indicated
occupation by a family unit. Similarly, the amount of diagnostic ceramic in Unit 6 was very low,
with only 5% of the recovered ceramic (n = 29/610) selected for further analysis and drawing.
Vessel forms in Unit 6 also suggest a domestic usage for cooking and storage. The most common
vessel forms recovered were ollas (n=9), and restricted storage wares (n=9) (Table 6.19). The
mouths of the ollas had an average diameter of 14 cm, indicating the smaller scale of food
preparation in this unit. Of the diagnostic sherds collected, only two have decoration or slip, and
327
nearly all of the paste types are Type A (n=10) or Type B (n=16) (Table 6.20). The recovered
diagnostic ceramic, then, conforms to the standard paste found at Iglesiachayoq.
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels
2
9
9
7
0
1
1
29
Percentage
6.9%
31%
31%
24.1%
0%
3.4%
3.4%
100%
Table 6.19. Counts of vessels from Unit 6.
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
10
16
1
0
0
1
1
29
34.5%
55.2%
3.4%
0%
0%
3.4%
3.4%
100%
Table 6.20. Paste Types from Unit 6.
The lithic assemblage recovered from Unit 6 was relatively standard, with the exception
of one quartz knife (Figure 6.40). While small quartz inclusions were common in the pastes of
the standard ceramic style at Iglesiachayoq, the existence of quartz artifacts was relatively rare—
we recovered only 5 lithic fragments which we identified as quartz.
328
Figure 6.40. Quartz knife from Unit 6.
We recovered one small metal ornament from the top level in Unit 6. This piece of metal
appears to have been part of a necklace or piece of jewelry, and is shaped as a small, thin flat
disc with a hole in one end (Figure 6.41). This piece of metal was likely made of copper, as
indicated by the green/blue oxidation present on both sides. The presence of a small metal
artifact in a domestic household suggests that the people that lived here had at least partial access
to luxury items.
Figure 6.41. Small metal pendant from Unit 6.
329
The faunal assemblage in Unit 6 contained an MNI of one Bos Taurus (cow), and one
camelid. All of the fragments of cow bone were found in the first level of excavation and were
heavily weathered. In contrast, the camelid bone fragment was a metacarpal found in the second
level of the unit, with evidence of slicing down the middle. Camelid could have been consumed
in this unit, suggesting a pattern of continuity in New World consumption patterns for the
individuals living in this structure.
The artifact assemblage from Unit 6 corresponds with that of Unit 5—both units
contained basic cooking and storage vessels which lacked decoration, as well as small metal
pieces used for bodily ornamentation. The overall picture of this unit suggests that a family
briefly resided here, and was engaged in domestic activities and daily life. The prehispanic
practices of consuming camelid and making standard cooking vessels was unaltered by Spanish
presence in this portion of the site—instead, the practices demonstrated continuity until site
abandonment.
Unit 7: Intermixing of Spanish and Local/Inka peoples
While Units 5 and 6 show similar patterns of ceramic, metal, faunal, and lithic
assemblages, Unit 7 presents a very different type of assemblage. Unit 7’s location was the
furthest north and closest to the edge of the ridge, and would have had the most visual
precedence over the rest of the valley and Sector 1. Unit 7 also had significantly more diagnostic
ceramics (at 10% of the sample, n = 48) than the other two domestic units excavated in this
sector. Of the diagnostic ceramic, 30% were decorated or painted with red or orange slip,
suggesting that those who lived in this structure may have had higher status or more access to
finer items. The most common vessel in this unit was ollas (n = 22), followed by feasting wares
330
(n = 16) (Table 6.21). Ceramic paste types followed the standards at the site, with most vessels
consisting of Types A and B (Table 6.22). Three of the slipped diagnostic ceramics appear to
have been from the same vessel, which differed in form from the other vessels found at the site.
This vessel seems to have been a cylindrical, straight-walled vase or kero, which was unique to
this unit (Figure 6.42).
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels
1
5
22
16
0
0
4
48
Table 6.21. Vessel types and counts from Unit 7.
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
35
8
3
0
0
2
0
48
Table 6.22. Paste Types from Unit 7.
331
72.9%
16.7%
6.3%
0%
0%
4.2%
0%
100%
Percentage
2.1%
10.4%
45.8%
33.3%
0%
0%
8.3%
100%
Figure 6.42. Straight-walled vessel, perhaps a kero.
Since this form is not seen in the rest of the Late Horizon ceramic assemblage recovered
at Iglesiachayoq, it may have been a hybrid Late Horizon/Early Colonial Period vessel.
Additionally, we recovered one of the most decorated ceramic fragments from the entire site in
the last level excavated in Unit 7. This fragment was classified as a provincial Wari-style cup,
containing black and dark red slip line paintings (Figure 6.43). This style was identified by
Meddens in his 1985 dissertation as a Middle Horizon 2 (750-900 CE) style, based on Menzel’s
1964 study.
Figure 6.43. Middle Horizon 2 fragment.88
88
See Meddens 1985: 363.
332
Additionally, the paste of this artifact was much finer than the typical Iglesiachayoq
paste, with a laminated appearance and little to no inclusions. The ceramic itself was much
thinner than typical ceramics at the site. Taken together, the characteristics of this fragment
suggests that it was imported or brought from another location (perhaps the nearby site of Chiqna
Jota). The rest of the artifact assemblage supports this interpretation, as we recovered a mixture
of Early Colonial Period and Late Horizon artifacts.
Namely, we recovered an iron piece of hardware that may have been used in the
doorway, typical of Early Colonial Period assemblages (see Bauer 2015 for similar latches at
Vilcabamba, Wernke 2013: 182 for the Colca Canyon, Figure 6.44). Versions of door hardware
have been found across the Andes during this time period. The iron door latch suggests that those
residing in this structure may have had contact with Spanish authorities at the site, or at least the
ability to procure such luxury items.
Figure 6.44. Iron door hardware.
While the hybrid ceramic forms and the iron door hardware suggest at least some contact
with Early Colonial Period Spaniards, the faunal assemblage from Unit 7 reiterates the same
continuity discussed earlier. Within this unit, we recovered an MNI of three individuals, at least
one even-toed ungulate and two camelids. Kennedy recovered evidence of seven separate cut
333
marks on these faunal remains, suggesting that the people here were processing and consuming
meat.
The lithic assemblage of Unit 7 differed greatly from others at the site. While basalt and
obsidian were common in other units, Unit 7 contained a majority of roughly-hewn andesite
flakes and bifaces. Unit 7 paints a fascinating picture of occupation at Iglesiachayoq—I suggest
that those residing here likely had connections with some Spanish individuals, or were non-local
people themselves. The variance in ceramic styles and pastes found in this unit, coupled with the
iron door hardware affirms more access to non-local goods. Similarly, the preponderance of
andesite lithic fragments may suggest that those who lived in this structure did not have the
knowledge or cultural capital to locate sources of obsidian or basalt from which to make better
tools. Rather, they used the andesite extremely common across Iglesiachayoq.
Sector 3 Domestic Structure Excavations
Excavations and recovered artifacts in Sector 3 presented a different assemblage from
Sectors 1 and 2. Namely, Sector 3 is the more “local” Sector of Iglesiachayoq, contained the
most circular structures, and demonstrated fewer high status goods or ceramic sherds than the
other two sectors. We fully excavated the interiors of three structures, including one ovoid
structure and two quadrangular structures with rounded corners.
Unit 8: Ritual Space?
Unit 8 was placed in a circular structure which may have been periodically cleaned—we
found no in situ contexts, and all materials recovered were extremely fragmentary in nature. For
example, with the exception of one ceramic fragment, all other fragments were less than 5cm in
334
size and were very eroded (average weight 5g per fragment). No animal bone fragments were
recovered, and lithic fragments were also made up of only very small flakes.
While we recovered 39 diagnostic sherds, 15 of these were too small to identify the form
of the vessel, and the vessels that we were able to identify are all tentative classifications due to
the small size of the ceramic. The most common vessel type in Unit 8 was feasting wares (n=11),
followed by ollas (n=7), and finally storage wares (n=5) (Table 6.23). Most of the vessels were
rustic domestic wares of paste Type A, B, or C (Table 6.24), but we did recover two fragments
identified as Wari. Since Iglesiachayoq is located near several other Wari sites, it is possible that
inhabitants imported these sherds from other locations.
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels
1
5
7
11
0
0
15
39
Table 6.23. Proportions of Vessels Recovered from Unit 8.
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
20
6
9
0
3
0
1
39
Table 6.24. Vessel Pastes from Unit 8.
335
51.3%
15.4%
23.1%
0%
7.7%
0%
2.6%
100%
Percentage
2.6%
12.8%
17.9%
28.2%
0%
0%
38.5%
100%
No fragments of animal bone were collected from this unit, and only few lithic flakes
were recovered. The lithic consisted of small (<3cm) debitage flakes of obsidian, two small chert
flakes, a knife fragment of basalt, and one andesite grindstone of large size. We believe this
grindstone is not in situ and was likely moved to this structure given its location on the surface of
the unit. In fact, all except three lithic fragments were recovered in the surface level.
Taken together, I suggest that those who utilized this structure were keeping it relatively
clean, and not performing traditional domestic activities such as cooking or weaving here. We
found no evidence of faunal or botanical remains, no hearth or ash lenses, and all diagnostic
ceramic recovered was less than 5x5 centimeters. Unlike other units where we found large, intact
remains of cooking vessels, Unit 8 lacked these contextual markers of various practices.
Additionally, the accessway of Unit 8 faces north, opening up into a semi-restricted semi-circular
upper plaza—this plaza conforms to standards put forth by Albornoz and Molina about
enclosures utilized for Taki Onqoy dancing. In Chapter 8, I will provide spatial analysis to
support this claim. Furthermore, ritual cleaning of houses was identified by Molina as a Taki
Onqoy practice. Although cleaning of houses is demonstrated in Andean rituals in later times
(Lara 2016) and thus may not be unique to Taki Onqoy, I will tentatively suggest here that this
structure may have been a space where inhabitants of Iglesiachayoq performed the dance, songs,
and cleaning of the revitalization movement.
Unit 9: Local Domestic Life
In contrast with Unit 8, Unit 9 provided larger ceramic fragments, lithics, and faunal
remains from which to hypothesize its function. Upon removing the wallfall in the southern
336
portion of the unit, we recovered one whole vessel and another crushed, but almost complete
vessel (Figure 6.45).
Figure 6.45. Two complete vessels found in Unit 9, left: olla, right: jar.
As can be seen in the two nearly-complete vessels, the ceramic in this unit was much
more rustic and utilitarian than in Sectors 1 and 2. The jar is incredibly similar to the Soras-style
jars depicted and identified by Meddens’ in his dissertation (Meddens 1985: 396). Thus, it is
perhaps an example of Late Horizon regional wares used by non-elite people. The olla (left), has
a small mouth and rounded base, and was located slightly west of the whole jar. Looking at these
two vessels as indicative of the activities in this unit, two major characteristics are pertinent:
first, both vessels are clearly utilitarian rather than decorative—they both lack slip or any
exterior decoration, and fingerprints of use can be seen in the surface of the jar. Second, both
vessels have at least some evidence of soot or heat exposure, indicated by the presence of black
residue on the external portions. Thus, these vessels were likely used for domestic activities such
as cooking and storage.
337
Due to the wall collapse at the south of Unit 9, the contexts were relatively intact
compared to many of the units in other structures. We uncovered parts of a prepared mud floor in
this southern end of the unit, upon which the crushed olla and jar were found (Figure 6.46). The
rest of the ceramic in this unit consisted mainly of ollas (n=8) and storage wares (n =6) (Table
6.25), and paste types were mostly A and B (Table 6.26). Notably, we recovered 3 pirurus, or
spindle whorls, from Unit 9, the highest concentration of all units. The presence of the three
pirurus in various stages of use or construction adds credence to the interpretation of this
structure as a local domestic household.
Figure 6.46. Prepared floor in southern half of Unit 9 facing south.
338
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels
5
6
8
4
0
3
12
38
Percentage
13.2%
15.8%
21.1%
10.5%
0%
7.9%
31.6%
100%
Table 6.25. Vessel types and proportions from Unit 9.
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
18
15
4
0
0
1
0
38
47.4%
39.5%
10.5%
0%
0%
2.6%
0%
100%
Table 6.26. Paste types of vessels from Unit 9.
Loci 103 and 104, excavated in the southern portion of the Unit were both identifiable
contexts, containing the paved floor (Locus 103) and a prepared hearth (Locus 104) lined with
stones, containing several piece of carbon as well as camelid bone. Both fragments of bone
collected in Locus 104 were identified by Kennedy as camelid. Although with such a small
faunal sample we cannot make definitive conclusions about diet at Iglesiachayoq, it is likely that
those who lived in this structure were consuming a diet of local animals and crops. Since the
bone was found in the context of a hearth, I suggest that the inhabitants of Unit 9 regularly
prepared and consumed camelid.
339
The lithic assemblage of Unit 9 included primarily obsidian and basalt fragmentary
flakes, one large andesite grindstone associated with both of the complete vessels, and one rare,
polished quartz cone which looks more like some sort of amulet rather than any type of
utilitarian tool. This lithic was polished on three sides, and was the color of purple and cream,
with small reddish inclusions. It was found in the eastern border of the unit, and seems to have
been a rare object to be included in the artifact assemblage of Unit 9 (Figure 6.47).
Figure 6.47. Two views of the polished quartz.
This piece of quartz is large and lots of time must have been invested in creating such a
smooth, shiny surface. The purple/blue color of the quartz was unique to this lithic item—we did
not recover any other fragments of quartz in this color, and no other stone item was as heavily
worked. The presence of this fine quartz object in an otherwise rustic or domestic household
seems to be an anomaly.
The artifact assemblage of Unit 9 was thus mixed—the large, nearly complete ceramic
vessels conform to local Soras vessel forms, while the faunal data represents a continuation of
prehispanic consumption patterns. The pirurus suggest weaving was taking place in this
household, indicating another domestic activity for people at this time. The quartz object is a
340
complicating find vis a vis the rest of the artifact assemblage. Was this a prized object passed
down by family members? Was it an amulet used in ritual? Was it merely a piece of quartz that
became worn down simply by handling the fragment over time? Was it a portable huaca typical
of the Taki Onqoy movement, indicating private, covert worship in this particular household?
Unit 10: Cultural Connection and Intermixing
The last domestic unit excavated in Sector 3 was Unit 10. While the structure associated
with unit 10 was grouped with the other Sector 3 structures, it occupied a unique place on the hill
south of the church, with visual access to the church and central plaza. Those who lived in this
structure held somewhat of a “lookout” position in which their door faced the main route
between Sectors 1 and 3. Thus, those who lived in this structure would have been able to monitor
travel between both sectors, spatially indicating a position of ambiguity between the Spanish
center and the local center.
The artifact assemblage from Unit 10 also reflects this ambiguity, containing both Early
Colonial Period and Soras or provincial Inka items. Like Unit 9, the preservation in Unit 10 was
better than in other units, likely due to the wallfall covering most of the eastern portion of the
unit. The first two loci (151 and 152) were surface levels which contained many ceramic and
lithic fragments. In Locus 152, concentrated in the northwestern quad 2, we found a dense
concentration of obsidian debitage, suggesting the individuals who resided here may have been
making lithic obsidian tools here.
Similar to Unit 9, we found a prepared hearth (Locus 153) in the southeast corner of the
unit. Perhaps the identical location of the two hearths suggests this particular corner was better
suited to cooking activities for those living in this upper sector of the site. In this locus, we
341
collected carbon, a camelid mandible with evidence of burning, and several fragments of ceramic
(Figure 6.48).
Figure 6.48. Locus 153 hearth with stained soil and containing a camelid mandible and carbon.
North of this hearth was another example of a prepared floor, with a pressed clay matrix.
In this corner, we recovered a caret-head iron nail, indicative of Early Colonial Period activities
(Figure 6.49). Since this artifact was recovered resting on the floor of the structure, below a
surface level, I argue that it would have been an artifact present in this household during its use.
This nail was found associated with Soras ceramic, suggesting the people that lived here were in
contact with both Spanish and local populations.
342
Figure 6.49. Caret-head iron nail from the first half of the sixteenth century.
The most common vessel forms recovered from Unit 10 were feasting wares (n=15) and
cooking wares (n=14) (Table 6.27). Most ceramic recovered from this Unit were Type A or Type
B (6.28). The recovered ceramic was relatively rustic, some with a red slip, but mostly lacking
decoration.
Vessel Type
Aribalo
Restricted Storage Wares (Cantaro, Jar, Bottle)
Cooking Wares (Olla)
Feasting Wares (Bowls/Plates)
Figurines
Pirurus
Unidentified
Total
Number of Vessels
0
10
14
15
0
0
2
41
Table 6.27. Vessel types and proportions from Unit 10.
343
Percentage
0%
24.4%
34.1%
36.6%
0%
0%
4.9%
100%
Vessel Paste
Type A
Type B
Type C
Type D
Type E
Type F
Type G
Total
Number of Vessels Percentage
19
21
1
0
0
0
0
41
46.3%
51.2%
2.4%
0%
0%
0%
0%
100%
Table 6.28. Vessel paste types from Unit 10.
As briefly mentioned above, we recovered a dense amount of obsidian debitage flakes
(n=34) in the northwest quad of the unit. The concentration of these flakes in this one areas
suggests that someone in the structure may have been involved in the creation or shaping of
obsidian tools. We also recovered a small grindstone associated with the hearth, suggesting
grinding and food preparation was also occurring in this area.
The faunal assemblage from Unit 10 contained an MNI of two, with one cuy and one
camelid. Notably, this is the only structure in which we found evidence of cuy. Taken together,
the presence of these two animals suggests a continuation of local foodways and consumption.
According to Molina and Albornoz, Taki Onqoy practitioners often sacrificed cuy as part of their
associated practices. Since we only recovered the femur and a phalanx of the cuy, we cannot
argue for the sacrifice of this animal. However, it is notable that the cuy and camelid were found
in conjunction with a structure which also had evidence of the incorporation of Spanish goods.
The materials recovered from Unit 10 seem to mirror the structure’s spatial location at
Iglesiachayoq. As discussed above, this structure is situated between Sectors 1 and 3, and would
have been the ideal location from which to monitor foot traffic between these two Sectors. The
people who inhabited this structure would have had a clear view of the church and activities
344
taking place in the church from this high vantage point, but would have had to walk to the south,
up the hill, in order to participate in local activities in Sector 3. Like its spatial connection with
both sectors, the material remains from Unit 10 also demonstrate connection between
prehispanic and new traditions.
Units 11-15: Empty Patio Spaces
I will now discuss the five patio units in one section. In general, Iglesiachayoq is a very
low-density artifact site—we recovered 8,720 ceramic fragments, of which 843 were diagnostic.
With the total excavated volume at 135.53 cubic meters, the average amount of sherds per cubic
meter was 64, and the average amount of diagnostic sherds per cubic meter was 6. The site’s
short occupation in conjunction with highly acidic soil, as well as the modern usage of the site
for animal husbandry, have created an atmosphere in which most primary contexts are only wellpreserved if they were protected by wallfall from a relatively early time. In most domestic units,
we recovered primary contexts near walls and corners. However, all patio units were 2m x 2m
and located in open spaces outside of five structures. These areas have been consistently used for
grazing or growing crops by farmers from the town of Chicha. None of the patio units provided
any sort of primary context, and all patio units had very low percentages of ceramic and other
artifacts.
Unit 11 was located south of Unit 1, occupying the space outside the accessway. Only
three diagnostic sherds were collected out of a total of 59 ceramic fragments. Of these three, one
vessel was categorized as a vase, one was a worn pedestal base, and one was a Late Horizon
sherd with some black slipped designs. No diagnostic lithic was recovered, and undiagnostic
pieces consisted of obsidian and chert debitage flakes. We did not recover any faunal remains.
345
Unit 12 was located west of Unit 2, in the central shared patio of the rectangular structure
complex in Sector 1. This unit was placed here because it was in a more restricted area, and we
hoped to define what sorts of activities were happening in the private outdoor space of an elite
complex. While none of the ceramic recovered (n = 7) was considered diagnostic, all paste
textures were classified as “fine” or “moderately fine” and three fragments were created from a
cream-colored paste. Several of these ceramic fragments also had red slip or evidence of
oxidation on them. Thus, although we did not recover any rim or base sherds, the ceramic we did
collect suggests that it was of a finer quality than other patio units. We did not recover any lithic
remains or faunal remains from this unit.
Although Unit 12 was unremarkable in terms of artifacts present, we were able to identify
one specific cultural context which was designated by a scattered carbon lens which was circular
in nature. We were not able to determine if the lens was the result of modern activities in this
area or if it represented a past burning event, as there were no associated ceramic. It is possible
that this area demonstrates a past collective activity which involved a small fire.
Unit 13 was located east of Unit 3, outside of the circular structure excavated in Sector 1.
As discussed above, we did not recover many diagnostic ceramic fragments in Unit 3, and the
ceramic from the patio presented a similar scenario. Out of a total of 53 sherds, only one was
considered to be diagnostic—a light brown, Type C paste tazon with faint evidence of orange
and white slip. None of the lithics recovered were diagnostic, and consisted of fragments of
basalt, andesite, and obsidian. No faunal remains were recovered.
Unit 14 was located in Sector 2, and was the patio unit associated with Unit 5. Out of a
total of 40 recovered ceramic fragments, only two were considered diagnostic, and these
consisted of an olla and a cantaro. Both of these fragments were Type A paste type, and neither
346
had evidence of slip or decoration. Only one small lithic was recovered, consisting of an obsidian
flake. No faunal remains were recovered. Unit 14 was particularly disturbed—upon arriving at
the site, we discovered that a local farmer had recently tilled the field where Unit 14 was located
in order to prepare for the planting season.
The last patio unit excavated was Unit 15, which was located in the semi-circular upper
plaza in Sector 3, north of Unit 8. The occupation levels in Sector 3 were comparatively deeper
than in Sectors 1 and 2, and the patio unit located in this sector was no exception. We recovered
191 sherds from Unit 15, of which 17 were considered to be diagnostic. Like the ceramics
recovered from the associated Unit 8, all of the sherds were under 5cm in size, suggesting that
the vessels may have been destroyed either during the Late Horizon, or by cow trampling in the
present. Because of their small size, the majority of the vessel forms of the diagnostic ceramic
could not be determined—of the 16 diagnostic sherds, we were able to identify one olla, and five
feasting vessels. We recovered three diagnostic pieces of lithic, which consisted of two small
basalt knives, and one obsidian blade. No faunal remains were recovered from Unit 15. The
comparatively more numerous assemblage from Unit 15 suggests that there was more activity
occurring in this region, specifically in regards to feasting. I propose that this area would have
been ideal for performance of Taki Onqoy practices, since it was located in a local plaza and
away from view of the site center.
Taken together, the patio units do not provide much positive information for activities
taking place in outdoor spaces. Overall, we recovered very little ceramic and lithic from the patio
units, and we found no evidence of faunal remains. As discussed above, Iglesiachayoq is utilized
by various individuals to keep herds of cows and sometimes horses, who are given free rein to
wander the site. The lack of artifacts in these spaces could be indicative of modern disturbances
347
by farmers and animals. Alternatively, the lack of both artifacts and primary contexts also could
indicate that there were no specific activities occurring in these spaces in the Late Horizon and
Early Colonial Period. The number of ceramic fragments recovered from Unit 15 was
significantly greater than any of the other patio units, at almost three times the amount recovered
from Sectors 1 and 2 patios. I hypothesize that this difference can be explained by two potential
processes. First, the units in Sector 3 demonstrated a deeper occupation stratigraphic pattern,
suggesting that this portion of the site was utilized for a longer time than the other two sectors.
The Unit 15 ceramic assemblage, then, could reflect this generalized trend in site usage.
Alternatively, the greater density of ceramic recovered in Unit 15 could also indicate a pattern of
more intensive use of this particular space at Iglesiachayoq. As mentioned above, Unit 15 was
located in the semi-circular plaza in Sector 3, and the elevated amount of ceramic could indicate
that people were occupying and interacting in this space more frequently than in other outdoor
spaces.
Unit 16—Stone Fragmentation and Burned Soil in the Plaza
The last non-mortuary unit excavated was Unit 16, which was a 5m x 5m unit placed in
the center of the central plaza. Like many of the other patio spaces, this area is used in modern
times for animal grazing, so the lack of ceramic and other artifacts may reflect modern-day
interference. Unit 16 contained four diagnostic sherds out of a total sample of 83 sherds, of
which we were only able to classify one bowl. The relative lack of ceramic in the central plaza
could also suggest that this space was regularly cleaned, and it affirms that domestic activities
were not occurring here.
348
In my hypotheses, I proposed that there would be evidence of burning or destruction in
the central plaza. This hypothesis was formulated based on Albornoz’s reports of the
punishments he inflicted on takiongos in this plaza, and also his affirmation that he burned or
destroyed several huacas here. During Taki Onqoy, huacas were not only able to inhabit various
individuals, but they also were considered portable in some cases. Thus, the destruction of
portable stone huacas could certainly have been possible in this area of the site. In Unit 16, we
recovered 152 fragments of a brown, waxy stone which ruptured in conical or spherical patterns.
The total weight of this type of stone recovered from this unit was 1,855g, representing the
densest sample of any lithic collected at Iglesiachayoq. Geologists from the Vanderbilt
University Geology department confirm that this stone is a “welded tuff,” an igneous rock
generated from explosive volcanic eruptions. “Welded” tuff means that the material was hot
enough to compress very quickly into a stone (Figure 6.50).
Figure 6.50. Example of welded tuff collected in high density from Unit 16.
If this stone were present in such high quantities across all patio units, then I would expect it to
compose a natural part of the site’s geology. However, this stone’s elevated presence is
349
significant in the plaza unit in comparison to other patio units, even if we account for the size of
the unit. Furthermore, welded tuff is not present in the interior spaces of other domestic units,
suggesting its presence in the plaza is unique and intentional.
In addition to the elevated amount of welded tuff in Unit 16, we also recovered a large
sandstone slab with evidence of divets, cut marks, and what appear to be rust spots (Figure 6.51).
This slab is different from all other grindstones recovered during excavation at Iglesiachayoq—
other grindstones are of basalt or andesite, and contain the expected, smoothed surface indicative
of food preparation activities. The harder stone types are better suited for grinding surfaces
because they provide more resistance and are thus more effective, and also because the stone
material itself was less likely to break down into the food. In contrast, this slab was malleable,
cream/gray in color, and shows clear evidence of human activities other than grinding.
Figure 6.51. Sandstone slab with divets.
While I cannot affirmatively identify these artifacts as remnants of huacas, I will suggest
that they are indicative of different practices than those lithics associated with domestic
households. In the Early Colonial Period, huacas could take on infinite forms, making them
350
difficult to eradicate. Furthermore, when stone huacas were smashed or destroyed, takiongos
believed that each resultant fragment of the original stone became a new huaca, perpetuating the
cycle. It is possible that the tuff and sandstone lithic materials collected in Unit 16 were portable
huacas which were smashed in the central plaza. Alternatively, the concentration of rare (to
Iglesiachayoq) stone in this unit could also be a coincidental occurrence.
No faunal remains were recovered from Unit 16, again suggesting that the area was
regularly cleaned. As mentioned above, if Albornoz did burn huacas in the central plaza, then we
should see a contextual lens indicative of this practice slightly below the surface. Our subsurface
soil matrices were both darker (7.5YR 3/2) than the surface level, and also darker in color than
the common soil type at Iglesiachayoq. This could indicate burned soil; however, since we did
not recover any carbon or ash, I am hesitant to definitively suggest there was a burning event in
this location.
In sum, the excavation results from Unit 16 presented distinctive findings from all other
units at Iglesiachayoq. First, there was a complete lack of diagnostic ceramic, and the amount of
total sherds recovered (n=82) was low compared to other units of relatively the same size
(avg=690 fragments total). In terms of ceramic presence, the plaza area resembles that of the
other patios, which also lacked ceramic in general, and especially diagnostic ceramic. Second,
Unit 16 differed greatly from all other units in its lithic assemblage. While domestic and patio
units contained grindstones and small handheld stone tools, the lithic recovered from Unit 16
was denser and of a different type. The high number of welded tuff fragments suggests that there
may have been some sort of fracturing event here. Similarly, the large sandstone slab was
distinct from all other stones recovered at the site. In order to make definitive conclusions about
the activities occurring in the central plaza at Iglesiachayoq, much more work must be
351
undertaken. Future excavations will target other spaces in the central plaza in order to compare
our own findings. These excavations should also be scattered in open spaces site-wide in order to
determine if fractured tuff is present in other open areas or if it is indeed a specific concentration
indicative of unique practices in the central plaza.
Comparative Assemblage Analysis by Sector
Comparative Ceramic Data: Differential Status and Activities by Sector
The overall amount of ceramic recovered from Iglesiachayoq was rather low for the
amount of excavation completed.89 In general, there are very few ceramic fragments on the
surface of the site, and occupation levels were very short—no more than 50cm in depth in most
units. The ceramic assemblage at Iglesiachayoq is in contrast with other sites in the region, such
as Chiqna Jota or Chichaqasa, Middle Horizon sites which are covered in surface ceramic. In his
1985 dissertation which surveyed the region, Frank Meddens discusses the ceramic assemblage
at Iglesiachayoq and asserts that all recovered ceramic dated to the Late Horizon and perhaps the
Early Colonial Period. The majority of diagnostic ceramic recovered during excavation were
Type A or B, brown, medium-textured pastes with inclusions of quartz, mica, basalt, and
sometimes feldspar, typical of Late Intermediate Period and Late Horizon local styles throughout
the Andes. Another common type of ceramic was Late Horizon/provincial Inka style, which
included red and orange slipped ceramic, some with black decorations. Next were sherds which
were identified as Wari, and likely came from other nearby sites. Finally, we had some fragments
89
This is more my impression from working at dozens of sites throughout the Andes, and this perspective was
echoed by PATO team members and workers from Chicha. Moreover, in his thesis, Meddens remarks several times
that his experience at Iglesiachayoq supported my own, with mixed stratigraphy and low ceramic density.
352
which may have been from the Early Colonial period, although I am hesitant to definitively
characterize them as such because they lack the characteristic glazes typical of this time. We also
recovered several sherds which had either sculptural characteristics like fish heads, or rustic
protruding faces, which appear to have been either Wari or Soras in style. These “fancier”
ceramic findings were all recovered from Sector 1 excavations.
Ceramic studies from the Chicha-Soras region are still in nascent stages. The dearth of
archaeology in the region from the 1980s through today has created a scenario in which
understanding of ceramic pastes and styles is lacking. There are no geologic or clay samples
from this valley, and my ongoing LA-ICP-MS project represents the first geochemical analysis
of ceramics from the region. More work must be completed in order to understand processes of
trade or imported goods between this region and other nearby areas.
Sector 1 had the highest percentage of decorated and/or slipped ceramic sherds,
representing 18% of the total amount of recovered ceramic in this sector (n=717 decorated sherds
out of 3922 total). For purposes of this analysis, I removed church units from the Sector-based
analysis, as the ceramic recovered in the church was generally of a much finer quality, and
slipped and/or decorated sherds made up 33% of the total amount recovered (n=102/308).
Comparatively, Sector 2 had 7% of all recovered sherds with Inka-style slip and/or decoration
(n=119/1678), while Sector 3 had 14% of all recovered sherds with Inka-style slip and/or
decoration (n=392/2813).
There were two units—one in Sector 2 and one in Sector 3—which raised the overall
mean of decorated ceramic fragments. In Sector 2, Unit 7 had 14% of recovered fragments with
decoration or slip (n=82/582), and in Sector 3, Unit 10 had 19% of all sherds recovered with
decoration or slip (n=267/1434). Both of these units are anomalies in their respective sectors in
353
terms of the amount of decorated wares, the presence of Spanish artifacts, and their location
within the sector itself. For example, the structures in which Units 7 and 10 were placed
occupied liminal, or transitional spaces in relation to the other structures in their respective
sectors. Unit 7 was the structure located the furthest North in Sector 2, and it also had the
greatest visual accessibility of both the Chicha-Soras Valley, but also those activities in Sector 1.
Similarly, Unit 10 was located in a structure which was mid-way between Sectors 1 and 3, and
had visual access to activities in Sector 1 associated with the church. I will elaborate on the
potential significance of the spatial location of these two structures in the following chapter.
In addition to their spatial location “in-between” two separate areas, those who inhabited
the structures in Unit 7 and Unit 10 also contained the only clear evidence recovered of Spanish
artifacts outside of Sector 1. As mentioned above, in Unit 7 we recovered an iron door latch and
in Unit 10 we recovered an iron caret-head nail. These two structures 1) occupied spatial
locations with visibility both to the central church and their respective sectors, 2) contained
elevated totals of decorated or slipped sherds, and 3) were the only two structures outside of the
center which had direct evidence of Spanish goods. It seems more than coincidental that both
structures had similar artifact patterns. Perhaps those who lived here were sympathetic or
actively aiding Spanish authorities, watching for idolatrous practices in return for access to rare
goods. The logical conclusion from this data is that those households which contained Spanish
goods and higher-quality ceramic likely held a higher status in the community. However, this
conclusion is complicated by the documented Taki Onqoy activities at Iglesiachayoq. The
absence of Spanish goods in indigenous households may reflect an active choice to avoid these
items, rather than an inability to access or afford such goods. In any case, the blurring of spatial
boundaries of these two structures is mirrored by the intermixing of artifacts.
354
Vessel forms also varied by sector (Figure 6.52). All three sectors had even proportions
of feasting wares, but Sector 2 had a higher proportion of identified cooking wares (41.7% in
Sector 2 as opposed to 27.6% in Sector 1 and 32.6% in Sector 3). Sector 2 also had the lowest
proportion of Aribalos, at 3.6% of identified vessels for the sector, as compared to 9.8% in
Sector 1 and 6.7% in Sector 3. In general, the data from ceramic vessel forms suggests that
domestic activities such as cooking and food storage were the main activities occurring in both
Sectors 2 and 3, while there was a higher preponderance of ceremonial wares (aribalos) in Sector
1. We also recovered all figurines and fish-head appendices in Sector 1, indicating the more
decorative nature of the ceramic in this area. All ceramics tentatively identified as “Early
Colonial Period” were recovered in Sector 1.
355
Percentage of Identified Vessel Types by Sector
2.3%
100%
2.3%
1.2%
0.0%
3.4%
0.0%
90%
80%
32.7%
35.7%
33.7%
70%
60%
50%
27.6%
32.6%
41.7%
40%
30%
25.2%
20%
23.6%
17.9%
10%
9.8%
6.7%
3.6%
0%
1
Aribalos
Restricted Storage Wares
2
Cooking Wares
3
Feasting Wares
Figurines
Pirurus
Figure 6.52. Percentage of identified vessel types by sector.
When I assessed the differences between vessel types according to structure form, a
different pattern emerged (Figure 6.53). As can be seen by the graph below, there appear to be
basic differences in vessel function between the three types of structures. I have used proportions
of identified vessels in order to account for the differences in amount of each structure type
excavated. The quadrangular structures with rounded corners were the typical domestic
household at Iglesiachayoq. As can be seen from the chart below, the assemblage in these
structures (n=6) had even numbers of cooking wares and feasting wares, and demonstrated rather
356
standard domestic practices. Structures which were quadrangular with squared corners (listed as
“squared” below) showed the greatest amount of storage wares and aribalos of the three structure
types, as well as the least amount of cooking wares. This could indicate that domestic activities
were occurring less frequently in the squared-structure residences. I have hypothesized that those
individuals who resided in squared structures may have been associated with religious activities
given their proximity to the church, or that they held a higher societal status. This hypothesis is
tentatively supported given the lower number of cooking wares recovered.
Percentage of Identified Vessel Types by
Structure Form
Aribalos
Restricted Storage Wares
Cooking Wares
100%
2.4%
1.0%
3.3%
2.2%
80%
32.0%
30.4%
Feasting Wares
Figurines
Pirurus
2.4%
50.0%
60%
19.6%
36.8%
40%
28.6%
32.6%
20%
20.9%
16.7%
0%
7.1%
Quadrangular with Rounded
Corners
12.0%
2.4%
Squared
Ovoid
Figure 6.53. Percentage of identified vessel types by structure form.
Finally, 50% of the vessels recovered from ovoid structures were feasting wares,
including bowls, plates, or serving vessels. If circular enclosures were spaces utilized for Taki
357
Onqoy activities, then this proportion of feasting wares certainly supports documented practices
associated with the movement including heavy drinking and celebrating.
Faunal Data
The 19 units excavated at Iglesiachayoq produced a total of 673 identifiable vertebrate
specimens, weighing a total of 2,196g and representing a minimum of 30 individuals (completed
by Sarah Kennedy, Ph.D. candidate, University of Pittsburgh). Perhaps due to the poor
preservation at Iglesiachayoq, we recovered only mammalian faunal remains, and were able to
only positively identify 20% of the fragments to the family level. These included domestic
guinea pig, dog, camelid, cow, and horse/donkey. In general, patterns in faunal data follow
trends in differences between sectors in terms of change and continuity. Cow and horse/donkey
made up 5.05% of the identified assemblage, and were almost exclusively found in Sector 1,
with the exception of one cow bone in Sector 2. This affirms that remains of these large, Old
World animals were found only where Spanish presence would have been more common. All of
the cow and horse/donkey were adults at the time of death, suggesting they may have been
utilized as pack animals and for transportation.
In contrast, camelid bone was the most common faunal remain recovered in excavation,
making up 15% of the assemblage found site-wide. Camelids were recovered in ages ranging
from juvenile to sub-adult to adult, indicating a more generalized and renewable pattern of use.
The canine bone was found in an upper level in the church and was likely modern, while the
guinea pig remains were found in Sector 3, adding credence to the continuity in Andean foods in
this area of the site.
358
From the faunal data, I suggest that people of either higher status or those who embraced
Spanish traditions were clustered in Sector 1, the Spanish religious center of Iglesiachayoq. Old
World foodstuffs were more densely recovered in this region. In Sectors 2 and 3, a pattern of
continuity emerges throughout the Late Horizon and Early Colonial Period, in which local fauna
such as camelid and cuy were recovered. This continuity can be interpreted in two ways: first, as
in other sites across the Andes, continuity between prehispanic and post-conquest occupations
can be explained as the result of temporality. There simply was not enough time for the Spanish
to arrive at Iglesiachayoq, import their ceramic vessels, animals, and religious practices and
spread these traditions throughout all areas of the site. Indeed, this is the case at Malata, where
deFrance and Wernke find no evidence of Old World fauna in the early Spanish doctrina
(deFrance et al. 2016).In contrast, since there are broad differences between sectors which are
mirrored in faunal, ceramic, and spatial data, it is also possible that the people living in certain
areas of the site were actively avoiding Spanish goods and traditions, as predicted by Molina and
Albornoz.
Comparative Lithic Data by Sector
Finally, I will briefly discuss overall trends in lithic distribution, as well as consider the
differences between sectors in terms of lithic material. Overall, we recovered the most welded
tuff from excavation units (n=230/643 total fragments), if the site is taken as a whole. However,
with the exception of three fragments, all other tuff was recovered from Sector 1, with the
highest concentration (78% of all recovered fragments) found in the plaza unit. Units 17, 18, and
19, the church excavation units, contained another 26 fragments of tuff. Taken together, then,
between the church and the plaza, 90% of the welded tuff was recovered. As hypothesized
359
above, it could be that these tuff fragments were considered portable huacas—this would explain
its presence in both the plaza and the church units. The presence of this type of stone in the plaza
could correlate with Albornoz’s overt destruction of huacas in this location, and its density in the
church units could represent local takiongos bringing fragments of this stone into the new sacred
center of Iglesiachayoq. No fragments of welded tuff were recovered in outdoor patio units,
suggesting that perhaps its local distribution was not as ubiquitous as the plaza unit would
suggest. If the welded tuff had a normal distribution in external spaces at Iglesiachayoq, then it
could be explained as part of the general geology of the site. However, its concentration in the
plaza unit and church suggests the fragments may have been gathered from elsewhere and
intentionally deposited in these two areas.
The second most numerous type of lithic recovered in excavation units was obsidian
(n=182/643 total fragments), found in points, flakes, and cores. Obsidian was most often
recovered in household domestic units (n=160/182), with an average of 16 fragments per
household. There were three exceptions to this general pattern, and those were Units 3, 5, and 8.
Unit 3, an ovoid structure in Sector 1, had only 5 fragments of obsidian, and was generally rather
bare in terms of artifact assemblage in general. Similarly, Unit 8, an ovoid structure in Sector 3
had only 3 fragments of obsidian. Unit 5, a quadrangular unit with rounded corners, also had
only 3 fragments of obsidian. However, In Unit 5, we recovered the entire amount of rounded
hematite stones (n=9, site wide) found during excavation. I suggest that Units 3, 5, and 8 were
thus used for purposes other than general domestic activities, as suggested by the sparse amount
of obsidian and lithic in general in these units. Since Units 3 and 8 were ovoid, the lack of
obsidian supports my claim that this type of structure likely had a separate function at
Iglesiachayoq, and may have been a type of enclosure constructed or utilized for Taki Onqoy
360
dancing. Similarly, the presence of hematite stones and a small metal artifact suggests that this
structure may have been utilized for metal production rather than a typical domestic occupation.
The next two most numerous types of lithic were basalt (n=82/643) and andesite
(n=75/643) respectively. Both basalt and andesite were utilized as grindstones or lightly
processed stone tools, and were found relatively evenly site-wide. There were no statistically
significant concentrations of basalt nor andesite. There were four other types of lithic which were
found in much smaller assemblages site-wide. Pumice (n=17/643) was recovered almost
exclusively in Sector 1, and was generally reddish-brown in color. Chert (n=15/643) was creamy
beige in color, and much of the recovered chert was unworked or still had core attached.
Although fragments of chert were found in all three sectors, the three finely-honed chert blades
found in Unit 4 suggests that they may have been used in preparation of food related to church
activities. The relative lack of chert site-wide may indicate that the material was difficult to
obtain, and therefore was worth more at Iglesiachayoq.
Finally, the rarest lithic materials recovered at the site were hematite (n=9/643), and
quartz (n=7/643). The hematite stones recovered were rounded and very heavy (66.76g), and
were only recovered in Unit 5. There are several possible explanations for the presence of
hematite in only one structure: first, the hematite could have been used in metal working. We did
recover a small, finely-made copper pendant from Unit 5, suggesting that those who lived in this
structure may have been engaged in metal working activities has described. Second, small, heavy
pebbles were sometimes used in food storage. This second interpretation is less likely since we
found these stones only in on structure, whereas food preparation and storage was occurring
sitewide. Finally, the hematite may have simply been collected by an individual in this household
for indeterminate reasons. Although there are multiple explanations for the small hematite
361
stones, they were not randomly located in this structure. Unit 5 was placed in a structure which
was similar in overall assemblage to several of the other quadrangular structures with rounded
corners. Thus, the hematite concentration was an anomaly amongst the site-wide assemblages.
The least common type of lithic material was quartz, with only seven fragments
recovered at Iglesiachayoq. While most of these were small fragments, the conical, polished
purple quartz lithic recovered from Unit 9 (Sector 3) was unique. Heavily polished on three sides
so that they were almost glassy on the surface, the purple quartz resembled a deliberately-hewn
object. There were no divets or pock-marks on the outside, leading me to believe that it probably
was not utilized for food preparation, but rather that the aesthetic qualities of the lithic were the
reason for its presence in the structure. It may have been some sort of amulet or prized
possession.
In sum, the general sample of lithic at Iglesiachayoq follows traditional Andean
assemblages, particularly with the amounts of obsidian, andesite, and basalt. As discussed above,
the welded tuff was statistically significant in its density in the plaza unit, as was the hematite in
Unit 5. These results suggest that the densities of these types of lithic in these two units was
nonrandom and therefore intentional.
Discussion and Conclusion
Iglesiachayoq’s relatively short occupation period is reflected in the stratigraphy at the
site in that temporal differences cannot be clearly assessed through vertical levels. The ceramic,
lithic, and faunal assemblages recovered generally show a pattern of endurance, including a
continuation of Late Horizon styles (both provincial Inka and local Soras) in ceramic
technologies and dietary preferences, which demonstrated ongoing consumption of camelids. In
362
fact, temporal markers of Spanish contact were recovered only in the central church (Chapter 7)
and in the discovery of a few Old World fauna, as well as the presence of iron artifacts.
When revisiting my hypotheses regarding the material imprints of both Taki Onqoy
practices and evidence of Spanish colonialism, the findings from my work in domestic
households at Iglesiachayoq did not conclusively affirm that either religious sect had a dramatic
effect on daily life in the sixteenth century. We did not find overt examples of rejection of
Spanish practices or religious beliefs, nor did we find clear evidence of Taki Onqoy
performances. Ceramic styles, lithic technologies, and overall domestic practices were
indistinguishable from those at other contemporaneous Late Horizon or Early Colonial Period
highland Andes sites. I suggest that neither Takiongos nor Catholic church authorities
emphasized changes in household practices, and instead were focused on outward ephemeral
participation in religious performances or mortuary rituals—whether overt or covert—as a means
of supporting new religious practices.
363
CHAPTER 7
MORTUARY POLITICS AT IGLESIACHAYOQ
In the first century after Spanish invasion, mortuary practices were structured by the
disjuncture between prehispanic Andean and Catholic eschatology. Though both Andeans and
Catholics participated in a “cult of the dead,” the two religious traditions differed in
interpretations of which dead. The Andean practice of ancestor veneration was interpreted as an
idolatrous practice by Catholic authorities—the reciprocal relationship between living Andeans
who provided gifts of food, drink, and clothing to their deceased ancestors and, in turn, the
ancestors retributive ability to ensure good fortune, alarmed Catholic priests. These priests
perceived these venerative practices as indicative of the failures of Catholic evangelization—that
the Andeans were venerating or looking to guidance from a force other than the universal creator
was distressing and part of the justification for the mass resettlement program in the 1570s (Gose
2008; Toledo 1986). From as early as 1541, a mere decade after conquest, Spanish priests
reported that Andeans were reentering the churches where their baptized kin were interred,
removing the bodies, and reinterring them at the locations of their huacas, often in mountaintops
(Gose 2008). In the sixteenth century, the fraught struggle between Catholic mortuary traditions
and Andean mortuary practices manifested in a wide range of interment rituals, styles, and
locations (or disinterments), paired with shifting Spanish policies on how to counter antiCatholic practices.
Scholarly research on Early Colonial Period Catholic and Andean mortuary practices is
grounded in 1) the primary accounts of Spanish priests who were actively trying to extirpate
Andean idolatries, and 2) official edicts put forth by the Councils of Lima, which dictated best
364
practices for evangelization, but were ultimately not fully implemented (Estenssoro 2002; Gose
2008; Ramirez 2005). Ecclesiastical accounts were recorded as a part of extirpation campaigns,
and the priests writing them had personal interests in achieving recognition from the Spanish
Crown, partially through these extirpation practices. More broadly, Spanish authorities and the
overall Spanish agenda were interested in maintaining the dichotomy between “European” and
“indio” so that even as Andean practices began to more closely resemble Catholic ritual, they
were still seen as idolatrous and influenced by the Devil (Gose 2008; MacCormack 1991). The
ecclesiastical accounts of this period are thus framed with an anti-Andean bias, one which
identified Andean mortuary practices as intentionally anti-Catholic and idolatrous. Official edicts
put forth by the Councils of Lima outlined specific guidelines directing both baptism and burial
practices, but these rules were not able to be enforced until after Viceroy Toledo’s intensive
resettlement program.
Catholic church burial practices are not explicitly discussed in the Taki Onqoy literature.
Since there are few clear guidelines in Taki Onqoy primary accounts regarding mortuary
practices, in order to generate my hypotheses regarding Taki Onqoy and Catholic church burials,
I considered both Spanish religious mandates at this time, and the possible continuation or
renewal of prehispanic Andean practices. First, I reviewed the accounts of Spanish priests in the
first century after conquest. Though they do not explicitly address Taki Onqoy as a unique
movement, these accounts can provide clues as to which “idolatrous” practices were common
even after stricter mandates were implemented by the Spanish Church in the New World. I also
projected Taki Onqoy practice frameworks to their logical purviews, and considered the
“rejection of Spanish cultural practices” as potentially applicable to Catholic burial traditions.
365
Finally, I studied burial practices in Spanish Catholic churches throughout the Andes and the
broader Americas.
In this chapter, I first give a brief presentation about mortuary theory and the ways by
which practice related to death structured cultural frameworks of the living. Next, I discuss
Andean and Catholic eschatology and its relationship to mortuary ritual, drawing on the work of
historians and ethnographers. Then, I restate my hypotheses regarding mortuary practice and
provide evidence for the material correlates of these practices, indicating how takiongos and
Catholic converts could have implemented these practices. In the bulk of the chapter, I present
the data recovered from my work at Iglesiachayoq to argue that although the space of the church
itself was where Spanish religious authorities nominally had the most control and oversight, it
was also where the most “resistance” or heterodox practices by indigenous actors are in evidence
at Iglesiachayoq.
The Living Bury the Dead, the Dead Transform the Living
The ritual commemoration, manipulation, and burial of the dead are laden with politics,
social memory, and the reproduction and transformation of society. Postmortem treatment,
though nominally for the dead, is enacted by and impacts the living. The treatment of deceased
bodies reflects symbolic practices which can perform and enact social order (Chesson 2001,
Salomon 1995). The treatment and interment of the dead not only conveys information about the
particular people who were interred, but also can reaffirm social alliances or attest to social and
symbolic differences between groups of people (Joyce 1999). The dead are thus not mere
representations of social status and identity at an individual level, but they are also important
political tools which play symbolic roles in negotiating cultural change.
366
The criteria through which archaeologists interpret the significance of mortuary
treatments has changed over time and generally, has fallen into two schools of thought since the
1970s (for a more thorough debate on these approaches, see Klaus 2008 and Rakita et al. 2005).
The first school of thought follows from first principles that archaeological remains are outcomes
of group adaptation to certain environments. For processual archaeologists, mortuary
treatment—burial type, location, grave goods, etc.—represented the identity of the dead during
his/her life (Binford 1971, Klaus 2008, Rakita et al. 2005). Mortuary treatment was
epiphenomenal to social status, which in turn was epiphenomenal to wealth (Brown 1995).
Processualist archaeologists, then, viewed mortuary treatments as static and reflective of cultural
processes rather than constituting these cultural aspects. Processualist approaches to mortuary
treatment are thus difficult to reconcile with the wide array of burial practices seen in Early
Colonial Period churches. A church is a definitive place of religious importance, rather than a
critical resource or overt indication of geographical territoriality.
Post-processualists argued that mortuary practices are inherently ideological, and that
ideological freighting is consequential—not epiphenomenal—to the (re)production of society
(Parker Pearson 1982; Rakita et al. 2005; Velasco 2016). The living bury the dead—mortuary
practices are thus not mere indicators of status, but also are important ideological statements
(Parker Pearson 1982; Velasco 2016). Post-processualists use practice approaches which
consider mortuary practices of indicative of and constituting cultural frameworks which are
historically contingent and contextual. Following post-processual models, I take a practice-based
approach to the mortuary treatments recovered at Iglesiachayoq. The dead and their places and
styles of interment can actively influence social memory and behavior—indeed, in recent years,
bioarchaeologists have considered the agency of corpses, and how deceased individuals can
367
actually influence human action (Crandall and Martin 2014; Tung 2014). As Velasco writes,
“corpses and body parts are seen not only as repositories of social memory or symbols of
political legitimacy, but as politically efficacious social agents that instantiate relationships of
power, including domination and subjugation” (2016: 63, Tung 2014).
At Iglesiachayoq, the church is the most visible structure at the settlement, and the sacred
space inside of the church played an active role in shaping relationships between Andeans and
Spaniards. The diversity in burial treatments (discussed below) reflects the political and
ideological orientations and aspirations of those who interred (or disinterred) their dead in the
Catholic space. The distribution of burials within the church space can indicate broader political
goals such as alliance with Spaniards, while removal of burials from the church space suggests
resisting native peoples. In the analysis presented here, I view the treatment of the dead at
Iglesiachayoq as produced by and constitutive of the cultural frameworks of the ancestors who
interred them.90
The Andean Animate Dead and the Spanish Corruptible Body
Andean treatment of the dead is variable across time and space, but is very welldocumented for late prehistory, given the Spanish obsession with Andean “idolatrous” burial
practices. Several Spanish chroniclers have described the treatment of the dead, specifically in
regard to Inka rulers. They affirm that there was an “integral role and critical significance of the
deceased Inka king (including his carefully preserved and mummified remains) in maintaining
Though a caveat here for “agentive” entitities would be that I align with the views of Tung 2014 in her “Agency,
‘Til Death Do Us Part? Inquiring about the Agency of Dead Bodies from the Ancient Andes” in which she argues
that scholars must consider how objects influence other things and how the relationship between objects and humans
can extend and cement social relations.
90
368
order and perpetuating the world in conjunction with the reigning king” (Shimada 2015: 9).91
Inka kings, even after their death, were cared for and venerated, and continued to yield power.
Dead Inka kings were sentient beings who participated in rituals and feasts (Figure 7.1)—they
were paraded during festivals, and venerated as ancestors.
Figure 7.1. Mummified Inka king being carried on a litter by the Inka. Source: Guaman Poma de
Ayala 1615: 258v.
Beyond the active role of Inka rulers, Andean peoples in general believed that there was
life after death, and that this life was very similar to “earthly” life (Cieza de Leon 1984 [1553],
Ramos 2010: 11). It would be false to consider Andean ideological beliefs of the dead in the
same life/death, animate/inanimate binaries common in Western societies. Instead, “death was
neither conceptually separated from life, nor viewed as an instantaneous event or permanent
91
For example see Guaman Poma de Ayala [1615], Juan Diaz de Betanzos (1996 [1551]), for critiques of
documentary sources see Kaulicke 2003, Pillsbury 2008, Salomon 1995)
369
state” (Shimada 2015: 11, Saignes 1999: 117). The social life of people and things extended
beyond death—the body and the soul were not separate as they are in Christianity, but rather all
things were “alive” including that of the physical deceased body (Allen 2002: 44). Like any
living, breathing human, the dead required sustenance and care—food, drink, and participation in
Andean rituals. Through this care of the dead, living Andean peoples were able to converse,
consult with, and gain assistance from their ancestors. As Salomon affirms, “A code of
reciprocity between dead and living was the simple central rule of ancestor cult: dispatching
goods by destroying them from among the living, one compensated the dead” (Salomon
1995:324).
Not only did the Andean dead need care from the living, but these dead ancestors in turn
had the ability to sway the fortunes of their living kin. As Ramirez affirms, “ancestors were said
to exert influence over the health of the group as a whole, over the abundance of crops, and over
life and death” (Ramirez 2005: 143). Ancestors also had the power to promote fertility, heal
plagues and illnesses, and generally provide good fortune over the well-being of their kin.
Because of their power to exert influence on future events and health, these ancestors had to be
cared for and venerated—indeed, a major tenet of the Taki Onqoy movement was that Andeans
had neglected both huacas and ancestors, and thus had suffered extreme misfortune. This
foundational reciprocal relationship (ayni) characterized material practices related to death
throughout the Andes: scalar collectives of kin relationships ranging from a few families to
extensive regions formed local religious suites dedicated to maintaining kin ties with their
ancestors, thus ensuring the continuation of cultural traditions (Salomon 1995).
Conversely, Catholic and Christian burial traditions centered on a very different concept
of body and soul: that the physical body and intangible soul were separate (Eire 1995:87). The
370
body could deteriorate and decay without damaging the prolonged, infinite life of the soul.
Catholic burial traditions reflected this belief, such that Catholic bodies were laid to rest in an
extended manner in a church facing the altar. The Catholic dead ceased to exist in an animate
physical form after the last breath—instead, the physical remains of their bodies were buried
beneath church floors to await the Resurrection and be reunited with the soul (Jacobi 2000). By
placing importance on the physical body itself, Andean peoples demonstrated their lack of
understanding for a single God: “the explanation of the Christian concept of the individual,
understood as a man endowed with a corruptible body and an immortal soul, had to be preceded,
accompanied, and reinforced by the idea of a single God who created the universe” (Ramos
2010: 73, emphasis mine). From a Catholic perspective, Andean preoccupation with the physical
body indicated that indigenous actors did not understand the concept of the soul, nor of
resurrection in an afterlife.
The emphasis on bodily, physical care of the dead in Andean belief systems is critical to
understanding the struggle over bodies and burials during the Early Colonial Period. Western
notions of an eternal soul removed from a physical body contrasted with Andean beliefs in the
interconnectedness of the physical and the ideological (Ramos 2010). Although conversion
efforts attempted to instruct and emphasize these Catholic concepts, Andean actors retained their
beliefs that the physical body was one in the same with the soul. These beliefs were somewhat
reinforced by the disconnect between Catholic doctrine on death and the actual mortuary
practices of Spaniards (Saignes 1999:117). For example, funerary treatment—the honors given
to the dead, the altars constructed with saintly relics from Europe, the pomp and circumstance of
funerals—contrasted with actual Catholic preaching about the differences between corruptible
bodies and eternal souls (Ramos 2010: 76-77, Eire 1995). It is not surprising, then, that Andean
371
actors retained belief in the importance of physical bodies, when they saw these rituals
seemingly reenacted by Spaniards in their own treatment of the dead.
In practice, Andeans continued to worship physical bodies of their ancestors, particularly
in the first few generations after Spanish conquest (Saignes 1999). These practices manifested in
the removal of Andean bodies from Catholic spaces, which plagued anti-idolatry officials: for
example, as Father Pablo Jose de Arriaga writes,
Their greatest abuse is to disinter the dead and remove their bodies to their
machays, or burial places of their ancestors in their fields…On being asked why
they do this, they say that this is cuyaspa, for the love they bear them. They say
that the dead lying in the church are in great torment and bound to the earth,
whereas in the fields, because they are in the open air and not buried, they have
more rest” (1968[1621]).92
For all forms of idolatry practiced by Andeans in the first generations after conquest, “the theft
of Indian interments from Christian cemeteries was particularly anathema to officials—an affront
to but also a blunt sign of the failure of Christianization efforts” (Lau 2015: 203). Reports of
clandestine removal of Catholic Andeans from church sacred spaces is documented as early as
1541, less than ten years after the conquest (Gose 2008: 139). The Andean emphasis on physical
bodies was at odds and indicated worship of false idols to an extent—Andean worship of things
(the dead) precluded their ability to have an exclusionary relationship with God and was thus
heretical.
The cognitive relationships between the living and the dead in the prehispanic Andean
world are derived from documentary sources and ethnographic work, both of which can have
biases and errors, or can tend to flatten the variety of attitudes toward treatment of the dead. It is
See also Mary Doyle’s 1988 dissertation, pp. 203-204 for a greater explanation of burial removal in the
seventeenth and eighteenth centuries.
92
372
impossible to understand what it means to have a relationship with the dead in all cases—these
relationships are not universal, and they are certainly contextual and shifting over time. Thus,
archaeology can provide evidence of context-specific mortuary practices, and is uniquely
positioned to take a materials-based and practice-based approach to the understanding of these
funerary policies. It is only recently that archaeologists have taken a broader approach to the
study of mortuary sites—the first archaeological “studies” of burials initially were focused on
objects and grave goods such that the actual human bones were often ignored or discarded
(Dillehay et al. 1995, Isbell 1997, Klaus 2008).
Andean Burial Practices in the Early Colonial Period
The Jesuit priest Bernabé Cobo, who lived in Peru in the early 1600s, compiled
descriptions of the various types of burial places he encountered in his time in Peru. He
differentiates between two main types of tombs, subterranean chambers and surface structures
(Cobo 1956 [1652], Ramos 2010). It is important to note that there are broad differences between
coastal burials and highland burials, in that while those on the coast used pyramids or mounds,
highland Andean societies often interred their dead in above-ground chullpas or mountain caves
(machays) (Isbell 1997). The actual location of these tombs varied according to the relationships
between those who cared for the dead and continually visited them. Mortuary places were
“intimately linked to the subsistence activities and political life of their inhabitants: they were
indicators of the control of a territory and its resources, of the ties that united society, of internal
hierarchies, and of the relations among neighboring groups” (Ramos 2010: 20). Furthermore,
there was not always continuity in who utilized tombs, as different ethnic groups would reuse or
373
transform existing burial spaces, thereby shifting the sacred nature of these places to suit their
own needs.
The death of Andean rulers or powerful kin could cause social and cosmic disequilibrium
(MacCormack 1991), such that the treatment of the death, and the rituals associated with the
dead could mitigate the problems in exchange of power (Ramos 2010; Salomon 1995).
Specifically, Andeans believed that these deceased ancestors maintained their claims to lands
and goods, and their living kin regularly and ritually upheld these claims through interaction with
the dead (Ramirez 2005; Salomon 1995). Kurakas and other high-ranking Andean officials were
in charge of mediating these interactions, becoming the embodied links to the deceased
ancestors—kurakas were thus celebrated for good fortunes for their communities, and also
blamed for inadequate veneration for the dead during catastrophic events.
During the Early Colonial Period, Spanish religious officials often lamented the Andean
practice of removing the dead from their Catholic burial places inside churches, and the
reinterment of these ancestors in places of Andean religious significance, often near mountain
huacas (Gose 2006; 2008). Peter Gose’s meticulous research on this practice has recovered
accounts of burial removal from churches as early as 1541, continuing throughout the
seventeenth century (Gose 2008). The First Council of Lima (1551-1553) addressed this practice,
mandating that punishments for burial removal should include corporal punishment such as
whipping, and even jail time (Gose 2006: 157). Even Cristóbal de Albornoz, who almost never
explicitly addressed Andean burial practices, recorded a note in his 1584 letter, affirming that he
punished a cacique from Lurin Lucanas for “disinterring the body of his father from the church”
374
(Albornoz 1990: 290).93 The removal of the dead from their church burials was performed in
secret, away from the eyes of the Spanish priests (Álvarez 1998 [1588]; Gose 2008).
The removal of body parts from burials has been documented throughout Andean
prehistory—because of the common practice of interacting with the dead, it was not uncommon
for Andean individuals to disinter bodies or parts of bodies. There are several possible
explanations for this disinterment: the disturbance or removal of burials may have been related to
the negotiation of space in burial locations which were laden with meaning. Some prehispanic
burial spaces were considered to be sacred, and thus many interments were disrupted, removed,
or destroyed in order to accommodate later burial intrusions. Second, the removal of partial body
parts—and especially the cranium—could have served an array of functions. Skulls were known
to be kept in households as a means of ensuring good luck for the remaining kin. In other cases,
large, kin-based burial locations were constantly being restructured in order to combine longdead and more recent ancestors. The layout of these tombs were continuously edited in the
process of incorporating parts of various ancestors, which were thought to have represented the
entire being (Shimada et al. 2015).
In the Early Colonial Period, the removal of the dead or parts of the dead from their
supposedly-final resting places beneath the church floors took on a different significance. In
1551, the First Council of Lima passed an ordinance that all Andean dead be interred in
churches—while this mandate was haphazardly enforced, it was still critical to the process of
evangelization. Catholic church authorities understood the significance not only as a crucial part
of conversion, but also as a way of differentiating Catholicism from Judaism and Muslim
93
Translated from “Porque desenterro el cuerpo de su padre de la yglesia” (Albornoz 1990: 290).
375
practices, the latter two which typically interred their dead outside of the church in graveyards in
Spain (Gose 2008). In 1580, Viceroy Toledo specifically denounced Andean idolatrous practices
of removing the dead, arguing that:
I order and command that each magistrate ensure that in his district all the tower
tombs be knocked down, and that a large pit be dug into which all of the bones of
those who died as pagans be mixed together, and that special care be taken
henceforth to gather the intelligence necessary to discover whether any of the
baptized are buried outside of the church, with the priest and the judge helping
each other in such an important matter, and that they have great caution in the
doors of the temples since they remove and take them [the baptized dead] from
the sepulchers at night when they are authorities and important people for said
effect, and they kill some women and Indian men saying they will serve them in
the other life (Toledo, Lohmann Villena, and Sarabia Viejo 1986:413-414; Gose
2008:124).
From this account, it is clear that the most powerful Spanish authorities were greatly concerned
with the removal of dead ancestors from the church. Moreover, the act of removing the dead
from churches and relocating them to places of indigenous importance was very dangerous. The
kin of dead individuals would have had to somehow break into the church, locate the precise spot
of their deceased relative, excavate down to the remains (in the case of Iglesiachayoq, over one
meter deep), and then refill the void left by the removal of the body. Removing the dead from
churches in the Early Colonial Period would not have been a celebrated form of interacting with
the ancestors, but rather a covert action which had very serious ramifications and punishments
for those who were caught (Gose 2008; Ramos 2010; Vargas Ugarte 1951).
Catholic Treatment of the Dead
In the Early Colonial Period, Catholic treatment of the dead was distinctive from Andean
mortuary practices, though there were several similarities in both Catholic and Andean
eschatology. For example, both religious traditions participated in a “cult of the dead”: for
376
Catholics, this manifested in the veneration of deceased martyrs which arose in late antiquity
(Brown 1981), as differentiated from Andean veneration of local proximate ancestors. Yet, the
adoration of these deceased individuals by the living functioned similarly to promote security
and maintenance of societal norms. Though nominally similar, Catholic and Andean traditions of
venerating the dead manifested in distinct variations in terms of practice—beliefs diverged in
terms of which dead should be venerated, where these dead should be venerated, and how the
dead should be venerated. For Catholic authorities, veneration of the Andean ancestors signified
misinterpretation of Catholic doctrine, wherein Andeans focused on their pagan ancestors instead
of embracing representations of God (in saints or martyrs). As new Catholics, these Andeans
received idolatry and error from their ancestors, who were not practitioners of the universalizing
faith (Gose 2008).
In late antiquity, Catholics began to inter their dead in churches, partially as a function of
creating a separation between Christian and “other” (Gose 2008). Jewish and Moorish
populations interred their dead outside of town centers, and by interring the Catholic dead within
churches, practicing Catholics separated their practices and beliefs from others coexisting at the
same time. At this time, saintly relics became associated with the church altars—these parts of
deceased saints served to bless and protect those interred below the church floor (Brown 1981;
Gose 2008). Thus, burial location within Catholic churches took on new significance, as the
physical proximity to the altar and by extension, the saintly relics often housed there, was
desirable so as to receive the grace produced by these relics (Eire 1995; Gose 2008). Catholic
parishioners desired specific church locations for their physical remains to rest after death, and
described these places in their wills. Specifically, the most valuable place of interment was the
main altar, “where the expiatory sacrifice of the mass was offered regularly” (Eire 1995: 99). A
377
second place of sacredness was “next to the holy water font, because so many worshipers would
necessarily tread on the grave and cross themselves over it, its occupant could gain great
recognition and some unexpected prayers, along with a generous and constant sprinkling of
spilled holy water” (Eire 1995:100). That is, beneath the church floor, sacred places were located
where masses were performed or where there was the opportunity of “extra” prayer and holy
water (Eire 1995:99; Jacobi 2000; Ramos 2010).
Deceased Catholic individuals were buried in extended positions oriented on the long
axis of the church, with their feet toward the main altar. This altar was ideally though not
consistently situated toward the East: “Ideally, bodies were oriented with heads to the west and
feet to the east…‘To the Christian the burial of bodies with their faces to the East is the outcome
of the belief not only of the resurrection of the body, but also that from the East shall come the
final summons to Judgement’ ” (Jacobi 2000:26-27, also quoting Puckle 1926:149). More
broadly, burial customs in Spain dictated that the feet of the deceased parishioners should be
oriented toward the altar (and thus facing where redemption should take place), while priests
were interred with their feet toward their flock and their head nearest to the altar (Jacobi 2000).
Catholic Church Mandates During the Early Colonial Period
In order to streamline the variety of indigenous positions on conversion, the various
Councils of Lima put forth edicts to control mortuary practices in the New World, and the
Spanish goal was to ultimately “Catholicize” these practices through appropriations of space,
style, and ritual. Although it varied greatly in practice, Spanish authorities developed an
idealized set of rules for the dying and dead in sixteenth-century Peru. The First Council of Lima
(1551-1552) initiated several rules for interment of Catholic (baptized) Andeans. First, Andean
378
peoples were to be buried beneath church floors immediately after their death: “We demand that
all who were Christians should be brought to inter in the church and cemetery, and that they are
interred with the cross and Holy water, and buried with the rites that other Christians usually
have” (Vargas Ugarte 1951: 20-21).94 This law was instituted to prevent the ongoing idolatrous
mortuary practices in Andean pueblos, which included neglecting to bury deceased bodies for
several days, interacting with these bodies, and interring them in places away from the church
(Vargas Ugarte 1951). Second, Spanish priests were to delineate a specific public place outside
of the church for those Andeans who had not converted to Catholicism nor received baptism to
be interred. This interment of non-Catholic Andeans included the “bodies of their deceased that
they have in their houses or in other mass graves, to prevent the idolatries that they practice in
their houses” (Vargas Ugarte 1951: 21).95 Third, during church interment, Spanish authorities
were to uncover the faces of the deceased to ensure that the body of the individual being interred
was the correct one. Fourth, during church interment, authorities could “neither permit that they
provide more clothing than what is necessary to encompass the body; nor after the interment
could they permit the inclusion of food nor drink nor any other thing” (Vargas Ugarte 1951:
21).96 Finally, the First Council warned that if a kuraka or principal Andean authority died, the
person succeeding them should be made aware of the previous kuraka’s servants and women so
that they would not kill themselves or anyone else.
Translated from” mandamus que de aqui Adelante todos los que fueren cristianos sean traidos a enterrar a la
iglesia y cementerio, y se lleven con cruz y agua bendita, y con los oficios que a los demas critianos les suelen
hacer” (Vargas Ugarte 1951: 20-21), translation mine.
95
Translated from “Y para los que no son cristianos, tenga a vista del pueblo o tambo un lugar public donde todos
los cuerpos de los difuntos sean enterrados y hagan a todos los indios cristianos que traigan los cuerpos de sus
difuntos, que tienen in sus casas y en otros sepulcros grandes, a enterrar en el dicho lugar, por quitar los
inconvenientes que de tenellos en sus casas se siguen” (Vargas Ugarte 1951: 21), translation mine.
96
Translated from “Y no permitan que le echen mas ropa de la necesaria para envolver el cuerpo; ni despues de
enterrado permita echar sobre el cuerpo comida ni bebida ni otra cosa alguna” (Vargas Ugarte 1951: 21), translation
mine.
94
379
The First Council of Lima thus specified the proper practices for Catholic interment
during the Early Colonial Period. Additionally, in the same edict (Constitución 25, Vargas
Ugarte 1951), officials also outlined the proper punishments for any who subverted the
guidelines above. First, the Council affirmed that “if any Christian indian buried anybody outside
of these said places, for the first offense they should spend three days in jail and should be given
fifty lashes in public, and for the second offense the penalty should be worse” (Vargas Ugarte
1951: 21).97 If caught, those interring their deceased outside of the designated places (the church
for baptized Catholics and the public area for unbaptized Andeans), would receive both corporal
punishment and jail time. In the final part of this edict, the First Council directed that the body in
question with whom these criminals were interacting should be removed and publicly burned.
Hypotheses for Mortuary Practice at Iglesiachayoq
Documentary evidence about Early Colonial Period Andean and Catholic mortuary
practices, as well as Spanish mandates on proper mortuary practices (discussed above) allows for
educated hypotheses about possible findings in the church at Iglesiachayoq. As outlined in
Chapter 5, I generated the following hypothesis about Andean interment in the Catholic church:
If the inhabitants at Iglesiachayoq were challenging Catholic mandates in burial practices, then
interments would reflect Andean traditions rather than standardized Catholic norms in terms of
burial position, burial goods, and disinterment or interaction with interred bodies.
1. Burial position: If inhabitants of Iglesiachayoq were subverting Catholic burial traditions,
then I would expect to find evidence of a variety of burial positions in relation to the
97
Translated from: Y si algun indio Cristiano enterrare o hiciere enterrar algun cuerpo fuera de los dichos lugares,
por la primera vez este tres dias en la carcel y le sean dados cincuenta azotes publicamente, y por la seguna vez le
agraven la pena” (Vargas Ugarte 1951: 21), translation mine.
380
altar, and burials could be secondary. Materially, this would include individuals interred
at various angles relative to the altar, evidence of secondary burials, or flexed burial
styles.
2. Burial goods: If inhabitants of Iglesiachayoq were not following Catholic burial
traditions, then I would expect evidence of grave goods in interments.
3. Interaction with interments: If inhabitants of Iglesiachayoq were not following the edicts
of the First Council of Lima, then I would expect to see evidence of continued interaction
with burials or disinterment from the Catholic church. Materially, this would include
missing parts of skeletons, parts of skeletons in the wrong anatomical locations, or voids
within the soil matrices where individuals were interred.
In practice, few examples of variations on the ideal Catholicized funerary treatment have
been recovered archaeologically in Peru in the Early Colonial Period. The lack of intensive
studies could be due to a lack of scholarly interest in excavating these sites during the Early
Colonial Period, the destruction of these church sites, and conversely, the prolonged use of these
sites into contemporary times (Van Buren 2010). However, some mortuary archaeological
investigations have been undertaken in Spanish churches. In the North Coast Lambayeque
Valley at the site of San Pedro de Mórrope, a sample of 862 individuals was excavated, spanning
the years 1536-1751 (Klaus 2008). This massive project is the largest mortuary sample after the
Spanish conquest, but the broad chronological usage of the church may have obscured practices
which specifically date to the pre-reducción era. In his funerary analysis, Klaus determined that
many of the dead were dressed in European garments, and that there were no grave goods
recovered from the distinct burials. The burials at Mórrope were aligned on a north-south axis,
and Klaus recovered evidence of continued and repetitive ritualized burial alteration wherein
381
local individuals were revisiting graves within the church and interacting with the dead. Klaus
interpreted the retention of prehispanic practices as a form of “symbolic resistance and reaction
against an invading foreign power’s attempt to transform a long-lived and pervasive Mochica
cultural identity” (Klaus 2008: 9). Klaus also identified several areas where the dead were
removed, relocated, or reburied.
Other examples of mortuary archaeology of church burials in the Early Colonial Period is
sparse in Peru. At Torata Alta (Moquegua Valley, Peru), limited excavation has been completed
at the central church, resulting in the discovery of only two extended primary interments in what
Rice calls “Christian-style” (Rice 2012: 13). Finally, Jeffrey Quilter’s research at the Colonial
Period site of Magdalena de Cao Viejo compares archaeological and biodistance data to affirm
that there was no evidence of a kin-based burial pattern, as depicted in documentary sources
(Ortiz et al. 2017; Ramos 2010).98 Beyond this handful of published studies, excavations in
churches in the Andes have mostly targeted salvaging burials during renovation, sometimes at
the expense of detailed on-site mapping and analysis (Talavera de la Vega and Benoki 2008).
Outside of Peru, studies of mortuary practices in early Spanish Catholic churches have
been completed in greater South America (Ecuador), Central America, and Spanish Florida and
Spanish California. In Ecuador, Douglas Ubelaker has investigated human remains from
ossuaries and individual burials from the Hospital San Juan de Dios, the Convento de Santo
Domingo, and the Convento de San Francisco. These burials range in date from 1500 through the
eighteenth century (Ubelaker 1999, 2000). Ubelaker’s data is primarily recovered from
secondary burials, and focuses on bioarchaeological data (sex, age, pathologies, etc.) and
98
Though there is evidence of secondary burial of some body parts, Quilter 2011.
382
concludes that health problems in the Early Colonial Period did not drastically differ from those
which ailed prehispanic populations (Ubelaker 1999: 31). Jacobi’s bioarchaeological study of the
Maya interred in the Catholic church at the site of Tipu, Belize (1568-1638) recovered 266
individuals (Jacobi 2000). All individuals were interred in a supine extended position with their
feet toward the altar. Similarly, of the excavations in Spanish Florida and California, nearly all
church burials were interred in the supine extended form, with occasional multiple interments
and bundle burials (Table 7.1). In all of these studies, burial style was almost uniform, there were
very few grave goods, and while there was evidence of interacting with the burials, most
investigators interpreted these interactions as a result of overcrowding in the church and
restructuring the space within, rather than deliberately removing interments for burial elsewhere
(Jacobi 2000; Larsen 1993).
383
Mission/
Date
Santa
Catalina
de Guale
(15971680)
# of
Burials
432
excavated
Context
Church
Head
Orient.
SW
Mode of
Interment
Supine
extended
Body
Orient.
NE/SW
Coffin/
Shroud
1 coffin
Grave
Goods
1 indiv.
Near
altar
Santa
Catalina
de Guale
de Santa
Maria
(16881702)
121
excavated
Church
SE
Supine
extended
SE/NW
1 coffin
Few
Santa
Maria de
los
Yamassee
<1675
124
excavated
Church
NE/SW
Supine
extended
San
Martin de
Timucua
(13341616)
65
excavated
Church
E
Supine
extended
E/W
No
3 indiv.
San Pedro
y San
Pablo de
Patale
abandoned
by the
mid-17th
c.
67
excavated
Church
NE
Supine
extended
SW/NE
384
14 indiv.
Notes
Hands folded
on torso from
chin to midabdomen,
heads away
from altar,
one multiple
interment,
subfloor
burials
Altar at
north, Wood
sanctuary
floor with
burials
below, few
disturbed,
two double
interments,
one bundle
burial (adult
50+)
1 Double
interment (2
juveniles), 2
flexed, 2
perpendicular
to NE/SW
direction of
most
1 burial had
metal
crucifix,
hands
beneath
mandible or
folded on
chest or
abdomen,
multiple
interments
common, 7
orderly rows
of burials
1 indiv. Had
head to SW,
1 had
semiflexed
legs, church
with subfloor
burials,
subadults
more
numerous at
San Luis
de
Talimali
(16331704)
75
excavated;
estimated
600-700
burials
Church
SE
Supine
extended
NW/SE
7
wooden
coffins
Few
Nuestra
Senora de
la Soledad
(15971784)
26-30
excavated
Church
and
cemetery
E
Supine
extended
E/W
Shrouds,
few
coffins
Religious
jewelry;
rare
SW end of
nave
Children
segregated at
west end,
adults in
coffins near
altar, 3 at
west end had
heads to west
opposite all
others (2 in
coffins),
hands folded
on chest
Arms crossed
on chest or
abdomen,
coffins
decorated
with nails
and tacks,
multiple and
intruding
burials
Table 7.1. Burials and mortuary contexts in churches from the Colonial Period Spanish
Southeast, North America. Data adapted from Larsen 1993: 336.
While evidence of secondary burial and burial removal can be seen in Early Colonial
Period churches (although much more archaeology must be completed), some scholars have also
found evidence of reburial of Catholicized Andeans outside of church spaces. For example, in
their work at Puruchuco-Huaquerones, a central coast site a mere 12 km away from Lima,
Murphy et al. excavated several cemeteries which appear to have held both Late Horizon and
Early Colonial Period burials. In one of these cemeteries primarily utilized during the Late
Horizon, the authors uncovered an adult female in an extended position accompanied by an Early
Colonial Period ceramic vessel (Murphy et al. 2011: 71). This burial was an intrusion into the
Late Horizon occupation of the cemetery, suggesting that the burial itself may have been an
example of interment removal from a church—its extended position and Early Colonial vessel in
385
combination with the intrusive nature of the burial are consistent with burial removal, and
reburial in an important Inka cemetery.
The differences between extended burials and flexed burials was a reflection of varying
ideologies on the afterlife of the dead (Saignes 1999). During extirpation campaigns in the
seventeenth and eighteenth centuries, Andean people complained that interment in the extended
position underground was problematic because “the earth on top of him [the deceased] prevented
the deceased from ‘breathing and moving;’ thus he or she could not rest and could not reach the
next world” (Doyle 1988: 205). As late as the seventeenth and eighteenth centuries, Andean
groups still desired their dead to be relocated to cave burial sites so that nothing would be on top
of the corpse, and so that they could be interred with their relatives from long ago.
It is clear, then, that although Spanish religious authorities attempted to institute orthodox
interment practices, in reality, these practices were variably implemented by Andean peoples in a
host of different ways. With a small clerical corps dispersed thinly through the central Andes in
the mid-sixteenth century, it would have been difficult or impossible for clergy to insure the
orthodox preparation, interment of, and maintenance of burials within churches.99 It is in these
very contested spaces of religious practice, however, where clergy would have focused their time
and resources. In fact, in sixteenth-century Madrid, Spaniards often requested religious clergy to
be present from the entire temporal space between death and interment (Eire 1995). In the
following sections, I will discuss the excavated burials at the site. These sections will not focus
on discrete aspects of burial such as skeletons, pathologies, or spatial location, but rather will
Both Saignes 1999 and Doyle 1988 discuss burial removal as a function of “saving” those interred in the
churches.
99
386
combine all potential aspects of the act of interment to reach a greater understanding about the
practices associated with death at Iglesiachayoq.
Mortuary Practices at Iglesiachayoq
In 2015, the Proyecto Arqueológico de Taki Onqoy excavated 21 distinct burials or burial
removals from three separate units in the church at Iglesiachayoq. The three units were placed to
recover individuals from the western, central, and easternmost (altar) portions of the church
(Figure 7.2). The church is 33m x 8m, making the church floor area 264m². Of this area, we
excavated a surface area of 33m², or 12.5% of the total surface area of the church. Although the
excavation units did not cover 100% of the church interior, burials allow us to make cautious
statements about those interred in the church, their burial treatments, their status as local or nonlocal individuals, and their ideological positioning compared to Spanish norms.
387
Figure 7.2. Locations and burials in church units.
General Findings and Burial Descriptions
My team and I excavated a total of 21 burial loci. Each interment was excavated as its
own locus, and burial soil matrices were markedly different from the surrounding strata. Each
locus which contained a burial was made up of a loose sand, and were clearly visible amongst
the strata in which they existed, a fine silt that was difficult to excavate. While the average
occupation strata in domestic units reached only 30-50cm, burials were interred an average of 11.5m below the surface. Burials were generally not comingled and there is little intrusion
between burials. That is, nearly all of the burials excavated were located in their own proper
388
space within the same horizontal plane, suggesting that the church was only used for a brief
period of time.100 The recovered individuals were interred in highly acidic soil, resulting in the
destruction and breakdown of the majority of spongy bone. During excavation, the extents of the
burials were exposed and assessed by Lic. Guni Baslut Monteagudo Espinoza. Due to the poor
nature of preservation, sex and pathologies could not be determined in situ. Anna Gurevitz, a
bioarchaeologist from the University of California, Merced, acted as the project bioarchaeologist
and completed the inventory of recovered skeletal elements in a lab setting. 101 Skeletal data
primarily consisted of cranial fragments, teeth, long bones, and tarsals and carpals. Following
White et al. 2012 and Hass et al. 1994 we assigned the age of individuals in seven classes: fetus
(before birth), infant (0-3 years), child (3-12 years), adolescent (12-20 years), young adult (20-35
years), middle adult (35-50 years), and old adult (50+ years).
For the purposes of classifying burials as either primary or secondary, we developed
several criteria informed by deposition processes within the church and anatomical positioning of
skeletal remains. These criteria follow general taphonomic principles (Haglund and Sorg 2001)
and are modeled on principles discussed by Klaus in his 2008 dissertation, but are modified
according to the preservation limitations at Iglesiachayoq. In order for a burial to be categorized
as primary, each set of remains was judged according to the following:
1. The remains are in anatomical position (allowing for shifting due to taphonomic
processes.
100
For example, in contrast with studies at other Spanish mortuary sites such as Morrope (Klaus 2008) or La Florida
(McEwan et al. 1993; Larsen 1993) which had evidence of long-term continued use of relocating earlier interments
to create more space, the individuals interred at Iglesiachayoq were not intrusive on one another.
101
Anna Gurevitz used the data from this project to write her Master’s Thesis, entitled “Catholicism, Negotiation
and Taki Onqoy: How Mortuary Ritual and Biogeochemistry Re-members Identity During the Early Colonial Period
at Iglesiachayoq, Chicha, Peru” (2017).
389
2. The dental remains are consistent with cranial and postcranial preservation.
All other burials which did not meet the above criteria were considered to be disturbed and/or
secondary.
Burial Types and Spatial Location
In the process of excavating the human remains from the church, we found a spectrum of
interment patterns, including both primary and secondary burials, extended and flexed, and
disturbed or removed burials. In this section, I discuss burial type and spatial location together.
Extended Primary Burials
Extended primary burials (n=3) were those which were interred in a supine extended
position, akin to those interments excavated in the broader Spanish Americas. We recovered
three primary extended burials with no evidence of disturbance or interaction post-interment.
These three individuals were oriented with their feet closest to the altar (East) and their crania
toward the west (Figure 7.3). Two of the extended individuals were located parallel to one
another beneath the western side of the Northern accessway, directly below the space where we
recovered the pillar which held the bowl of Holy water. Their proximity to one another and
identical burial style suggests they could have shared some sort of kin-based or marital
relationship. One of these individuals (Locus 912) was interred with a small metal pendant
located near the neck. This pendant is likely a copper alloy or a crude bronze (Figure 7.4.). This
individual was a young adult (20-35 years of age). We recovered phalanges at the shoulder level
of the corresponding arm, indicating that the arms were not crossed over the chest, but rather
folded so that each hand would have rested on the shoulder of the same side. The teeth were
390
worn, indicating his/her slightly more advanced age, and the incisors have evidence of shoveling,
which is usually associated with indigenous peoples. The bioarchaeological data from this
individual thus suggests he/she was of Andean descent, while the burial positioning suggests
he/she was interred in order to correspond with the proper Catholic style.
The skeleton of the other of these individuals (Locus 911) was very damaged due to a large
rock atop the upper half of the remains: we thus were only able to recover long bone fragments.
However, at the western end of the burial locus, we recovered four Nueva Cadiz beads which
date to the first half of the sixteenth century (Bray 2019; Smith and Good 1982, Figure 7.5.). The
Nueva Cadiz beads in conjunction with this burial indicates that these items were brought from
Spain sometime in the sixteenth century, and interred with the individual sometime after that.
These beads were likely part of aesthetic adornment, perhaps a necklace. In addition to the four
beads recovered with this individual, we also collected a large copper tupu and a conical piece of
copper which was associated with the tupu (Figure 7.5.). This individual has a combination of
both Andean and Spanish grave goods. The extended positioning of this individual suggests that
he/she was buried in order to conform to the idealized Spanish style. However, the grave goods
and copper tupu suggest that this individual may have been of Andean descent.
391
Figure 7.3. Primary extended burials (in blue) recovered at Iglesiachayoq.
392
Figure 7.4. Primary extended burial and metal pendant.
Figure 7.5. Burial goods interred with extended burial.
The third extended burial (Locus 1010) was uniquely interred below the center of the
altar in a stone sarcophagus. The extended burial was surrounded by two courses of roughly
dressed fieldstone in an oval shape, and topped with large, flat slabs of stone in order to protect
the body (Figure 7.6). The feet of the individual were oriented toward the altar with the head to
the west. Although we found the tomb seemingly undisturbed, the preservation of the bones was
393
very poor, not unlike the majority of the other burials in the church. The spatial location of this
interment, in combination with the mortuary treatment, suggests that this individual held the
most prominent status in the community at Iglesiachayoq.102 He was likely the priest which
tended to the church and/or was a high-status Andean individual. This burial was the only one in
which we found neither biological (shoveled incisors) or cultural (flexed positioning, orientation,
grave goods) markers of indigenous identity.
Figure 7.6. Sarcophagus with leg bones showing, photo looking west.
The bioarchaeological analysis of Individual 1010 indicates that the individual was a young
adult (20-35 years of age). The three teeth recovered from this individual are relatively clean,
The stone sarcophagus also echoes findings of Larsen’s 1993 chapter on Mission Bioarchaeology in La Florida,
where he found one single instance of coffin burial directly associated with the altar (325).
102
394
and have almost a polished sheen to them—this is different from the rest of the burials, whose
teeth often had black spots of discoloration or a soot-like residue on the tooth surface. The lack
of discoloration is likely due to the protection offered by the stone sarcophagus—the teeth would
have spent less time in direct contact with the surrounding soil. No additional information could
be retrieved from the postcrania given the level of degradation.
The pattern demonstrated by the primary extended burials mimics documentary evidence
about these types of interments in the Spanish Americas. All three individuals were interred
along the long axis of the church, with their feet oriented toward the altar. Moreover, all three
individuals were spatially located in places of religious importance—either below the Holy water
or directly below the altar. These interments thus likely represent “proper Catholics” in that they
followed the dictates put forth in the Early Colonial Period by the First Council of Lima (see
above).
Flexed Primary Burials
Flexed primary burials (n=2) were those individuals interred in the church as their
original resting place, but in typical Andean style in the fetal position. We could only definitively
identify two burials as primary flexed burials, and both were located in the westernmost portion
of the church, abutting the western wall (Figure 7.7.).
395
Figure 7.7. Location of primary flexed burials in the church.
The first of these primary flexed individuals (Locus 906) consisted of long bones
(ulna/radii, tibia/fibula), crania and mandible fragments, and four teeth (Figure 7.8.). There is a
possible lesion on the distal portion or patellar surface of the right femur. The cranial fragments
have no visible sutures and are also very worn, with evidence of cracking. The mandible is small,
with a width of ~89.31mm at the ramus, and there are strong indentations to the chin. The right
side of the mandible is almost closed over, indicating teeth loss. On the left side of the mandible,
the alveoli are still intact, confirming that the teeth were still present antemortem. We recovered
four teeth from this individual, two of which had dental caries. The bioarchaeological data from
this individual asserts that he/she was at least a middle adult (35-50 years)—the fully sutured and
396
worn crania in combination with the worn mandible indicates that the individual had lost teeth
prior to death, and likely would have needed assistance from other members of the community.
The burial positioning was in a typical Andean style, flexed position, interred along a north/south
axis as opposed to an east/west axis.
Figure 7.8. Primary flexed interment abutting western wall of church.
The second primary flexed individual (Locus 907) was interred with his/her head directly
in the northwest corner of the unit, with the long bones extending to the southeast (Figure 7.9.).
The burial area measured 110cm x 80cm. The bone was extremely degraded in this individual,
such that the cranium was completely disintegrated, with only teeth and long bones intact. There
were a few anomalous aspects of this burial: first, there appeared to be two distinct bone
groupings, which we divided into two features (hallazgos). Hallazgo 2 (as seen in Figure 7.9)
was determined to be a separate individual during analysis. Second, this was the only burial in
which we recovered grave goods other than metal artifacts. While we did recover two identical
metal tupus from the neck area of this individual, we also collected a miniature ceramic teacup
buried near the mandible of the individual (Figure 7.10). The tupus have a greenish hue in their
oxidation, suggesting that they were made of copper. They both have a half-moon at one side
397
with a small hole near the center, a shape very typical in the later periods of the Andes (Tiballi
2010). Although we were not able to gather much data from either the long bones or the cranium
in this individual, we did recover the entire set of teeth. Nearly all of the teeth had black and blue
discoloration—in other contexts, this pattern of discoloration may indicate burning. However,
since the soil degraded the bones in each burial, we suggest the black/blue color is more likely a
result of the soil properties. Individual 907 had between 4 and 10 dental caries. The third molars
of this individual had not erupted yet, indicating that she/he was an adolescent (12-20 years).
Figure 7.9. Primary flexed interment with tupus and teacup.
Figure 7.10. Left: matching tupu pins. Right: miniature ceramic teacup.
398
The bioarchaeological and archaeological data from this individual suggests that she/he
was an Andean teenager who was interred with some form of prehispanic traditions by her
family. The tupu pins and miniature teacup contrast with the idealized Catholic Church burial
style. Specifically, the miniature cup suggests that those who interred this individual were
deliberately ignoring or subverting the edicts put forth by the First Council of Lima, which
affirmed that dead Catholic converts were not to be interred with any grave goods, especially
those related to consumption of food or drink (Vargas Ugarte 1951).
The two primary flexed burials were both located in the westernmost portion of the
church, furthest from the altar. Their flexed positioning and variable angles in relation to the
altar, in conjunction with the grave goods recovered with Individual 907, suggest that these
interments were Andean individuals. Additionally, the presence of the miniature teacup suggests
a retention or renewal of prehispanic Andean practices perhaps associated with Taki Onqoy—
this cup could have been intended for use by the dead kin.
Extended Secondary Burial
We recovered one extended secondary burial (Locus 1008)—this interment was the bestpreserved, and was abnormal in its positioning. Several aspects of the burial suggest that those
who interred this individual attempted to arrange the bones in an anatomically-extended position
with the head to the west and the legs toward the east (Figure 7.11).
399
Figure 7.11. Secondary extended burial beneath the altar.
First, the cranium of 1008 was propped up on the surface so that it was directly looking at the
altar. If this were a primary burial, the cranium would have been laid to rest such that the
individual would have been looking up at the sky. Second, the arm bones are shifted toward the
cranium such that the individual is resting with perpetually shrugged shoulders. Finally, the leg
bones were also moved toward the cranium and were spread apart (Figure 7.12). No tarsals or
carpals were found, suggesting that only the most visible parts of the skeleton—the cranium, arm
bones, and leg bones—were relocated to the church.
400
Figure 7.12. Secondary extended individual.
The interment of Individual 1008 indicates yet another type of burial practice. In this
case, descendants of 1008 attempted to inter him/her facing the front altar, and in traditional
extended positioning. This individual was categorized as a young adult, and its secondary burial
status could have several explanations. First, this individual could have been a case of “baptizing
the ancestors,” in which the kin interred him/her in the church in hopes of Catholic redemption.
Second, Spanish authorities could have demanded that this person was interred in the church, and
forced the kin of the deceased person to inter him/her in this place in this style. Third, there is the
possibility that this individual was originally interred in the church, then removed, and then
reinterred in this location. The spatial location of this individual beneath the altar suggests that it
may have been a person with high status, or someone with importance to the community of
Iglesiachayoq.
401
Flexed Secondary Burials
Flexed secondary burials (n=2) were those which had originally been interred elsewhere, and
then were relocated to their final resting place in the church. Both flexed secondary interments
were recovered from the central unit of the church (Figure 7.13).
Figure 7.13. Secondary flexed individuals in the center of the church.
The first of these interments was buried with its feet toward the altar, corresponding with
the Catholic orientation. That is, although the burial style reflects Andean traditions, the burial
position (head to the west, feet to the east) indicates Catholic idealized practice. In the photo and
drawing, it is possible to identify the “bundled” nature of the postcranial bones (Figure 7.14).
402
The bones were interred in a group, with femur, humeri, radii, tibiae, and even tarsals mixed
together. The fact that the tarsals were still present, albeit distributed throughout the bundle,
suggests that this individual was previously interred in another location, and was then brought to
the church and reinterred. The anatomical positions of the bones cannot be due to natural
decomposition processes. Additionally, there were no teeth recovered with this individual,
confirming its secondary status. Because teeth preserve much better than other bone, if this were
a primary burial, we would expect to find teeth present along with the other bone. I therefore
suggest that Individual 958 was previously interred in Andean tradition in a machay or other
local burial location, and at a later time was brought to the church and intentionally reburied in
Catholic positioning. This individual was categorized as a young adult (20-35 years of age).
Figure 7.14. Left: Photo of interment, right: drawing of interment.
The second of the flexed secondary burials (Locus 960) was located very near the first,
and had the same orientation. Like the other flexed secondary burial, this interment also
contained a mixed bundle of postcranial long bones, with some degraded cranial fragments
oriented to the western side of the burial (Figure 7.15.).
403
Figure 7.15. Secondary flexed burial.
As can be seen in the photo and drawing, we recovered four metal fragments of tupus
positioned approximately where the neck of the individual would have been located. One of the
tupus is in the classic half-moon style of the Inka, while the other one has a pin with a small, flat
circle at the top. In sum, both of these secondary flexed interments were located in the center of
the church, and both were young adults (20-35 years of age). They each had the correct
orientation for Catholic edicts, but retained positioning of Andean traditions. The fact that these
burials were secondary contradicts the edicts of the First Lima Council, and suggests that these
individuals were unbaptized and relocated to the church.
Removed Burials
We recovered evidence of four removed burials from the westernmost portion of the
church at Iglesiachayoq. I define removed burials as those which maintained the same loose soil
matrix as the rest of the interments, and which only contained distal bones such as tarsals or
carpals, or had evidence of intentional refilling of the void left by the skeleton. Removed burials
404
made up 20% of the sample and all were in the unit closest to the western wall. Since all
evidence of removed burials comes from the unit furthest from the church, it appears that burial
practices were more variable in the location which would have held the least religious prestige in
that it was the furthest away from the altar (Figure 7.16.).
Figure 7.16. Locations of removed burials in the western portion of the church furthest from the
altar.
The first removed burial (Locus 909) presented an individual which was interred in the
church and may have been removed at a later date. This locus was located in the southeastern
corner of the unit. We only recovered seven bone fragments and one tooth from this burial area.
Of the bone, six out of the seven bone fragments were short bones, likely carpals or tarsals.
These bones are found at the extremities of the human body, and are usually the first to detach in
405
normal processes of degradation (Pankowska 2018; Schroeder 2001). The tooth was an
unerupted lower first molar, and thus we categorized the likely previous inhabitant of this locus
as a child (3-12 years of age). The fact that we found only short bones and one tooth makes a
strong case for burial removal—the head and the long bones were completely absent. In the
process of disinterring and relocating individuals in a rushed, perhaps clandestine manner, the
cranium and the long bones would be collected first, leaving the smaller bones to detach, fall,
and be left behind (Schroeder 2001).
We also recovered two tupus from this locus, although they did not appear to be in situ
and could have also been forgotten when the body was moved. These two tupus were not of the
traditional Inka style (see Figure 7.17). The conical tupu (IC.0909.2.M002) has two tufts of
brown fibers still attached, affirming its attachment to the clothing of the interred individual.
Figure 7.17. Tupus associated with the disinterment of Individual 909.
The second individual disinterred from the church (Locus 910) was located on the eastern
portion of the unit, and the soil matrix was much like those of the other burials—a loose area of
406
medium-grained sand disrupting a much more compact level of soil. This locus contains the most
convincing evidence we have for a removed burial. We recovered only a few small distal bone
fragments, and the pit had been loosely filled with large stones (see Figure 7.18). Locus 910
contained the only soil matrix which was made up of larger stones, thus affirming that the stones
were deliberately placed in this locus. The refilling of burial areas within churches has been
documented in archival records and was a practice which plagued church authorities (Gose
2008).
Figure 7.18. Left: indented area of loose sand among the compact soil matrix. Right: excavated
pit filled with larger stones.
The last two removed burials (Loci 913 and 914) were both areas of loose, mediumgrained sand matrix within the more compact surrounding matrix. Locus 913 contained three
small distal bones, while Locus 914 had no evidence of bone fragments or any other goods, and
was just an area of loose soil.
Burial removal has been explicitly linked to both Taki Onqoy and broader idolatrous
practices (Albornoz 1990; Doyle 1988; Gose 2008; Saignes 1999). Andean peoples were
documented interring their dead in the church by day, and disinterring and relocating these same
407
relatives under the cover of night, in a covert fashion. I argue that burial removal may have been
a strategic component of Taki Onqoy—by outwardly conforming to Spanish mandates, Andean
peoples secured favorable social and political standing in the eyes of the new authority.
However, by removing and reburying their dead in secret, takiongos would have demonstrated
their respect for and belief in huacas and prehispanic Andean traditions. It is impossible to
overstress the importance of removed burials in the church. In the center of the site, the space in
which Spanish authorities had the most control and oversight, Andean peoples were so
appreciative of their prehispanic beliefs that they were willing to risk severe punishment in order
to make sure their dead had an “Andean” afterlife. Moreover, not only would the Andean agents
removing their dead be punished, but the physical bodies of the ancestors they were removing
would have been publicly burned, thus preventing a smooth transition into the afterlife (Ramos
2010).
Disturbed and Indeterminate Burials
The last two types of burials were disturbed (n=5) or indeterminate (n=4). Disturbed
burials were defined as disarticulated and arbitrarily strewn skeletal elements, or evidence of
interaction with specific body parts post-interment (Figure 7.19.).
408
Figure 7.19. Disturbed burials in clue and indeterminate burials in red.
We recovered disturbed interments in every spatial location in the church. In the
westernmost location, one individual (Locus 915) appeared to have originally been interred in
extended supine style oriented with feet toward the altar. However, at a later date, the cranium
was removed and subject to interaction with the living. We first encountered the cranium, which
was on a shallower plane than the rest of the skeleton (only the lower extremities). The cranium
was at a depth of 103cm below the surface, while the rest of the lower extremities were at a
depth of 126cm below the surface. The cranium was inverted, with the foramen magnum in a
superior position, meaning that the individual’s face would have been upside-down and looking
at the western wall of the church (see Figure 7.20). If the positioning of the cranium were
409
anatomically associated with the rest of the skeleton, then the individual would have been buried
upside down and backwards. Furthermore, the teeth and mandible associated with the cranium
demonstrate evidence of at least one indirect burning event—they are caked with a black sootlike substance, but lack the distinctive blue/purple coloration of direct burning. The teeth for this
individual were also exceptionally worn, allowing us to categorize this person as at least a
middle adult (35-50 years of age).
While the cranium had an inverted and backwards positioning, the associated lower
extremities were interred in a primary extended position, with the feet oriented toward the
eastern altar. The fibulae, tibiae, and tarsals were laid in an anatomically accurate position, and
the spacing between these bones indicates normal decomposition processes (see Figure 7.20).
The lower extremities do not show the same pattern of burning that is present on the teeth and
mandible in the cranium.
Figure 7.20. Left: inverted and backwards cranium. Right: articulated, extended lower
extremities.
In sum, it appears that the cranium associated with the burial was repeatedly removed and
replaced in the burial locus, while the lower extremities remained in place. The disjuncture
410
between the cranium and the long bones could indicate several possible scenarios. The most
plausible explanation is that the cranium was removed and reinterred at least once. During
disinterment, the cranium could have been located near an open flame, which caused the indirect
burning pattern on the front teeth. Since we did not find any evidence of additional, primaryinterred crania in this burial locus, it seems as though those who were removing the cranium
remembered its location within the church, and were able to return it to its approximate original
positioning. Alternatively, the cranium we recovered could have been from a separate individual
altogether, completely distinct from the lower extremities.
A second disturbed burial (Locus 962) was located in the central portion of the church
and has evidence of post-interment disturbance. We collected primarily long bones and distal
limb fragments from this bundle, but the bones were strewn about such that there was no real
anatomical association between all of the postcrania (Figure 7.21). The dispersed nature of the
bones suggests that they were tousled at some point after their interment. No tupus or other
artifacts were recovered from the burial. Individual 962 was classified as a young adult (20-35
years of age).
411
Figure 7.21. Left: photo of Individual 962. Right: drawing of remaining disturbed bones from
962.
The remaining disturbed burials were located nearest to the altar. The first of these
(Locus 1005) was likely a secondary burial. Individual 1005 was a secondary burial in a very
unique position—the interment abutted the third and final step of the altar, and the individual
appears to have been arranged on his/her left side facing the altar in a north/south orientation (see
Figure 7.22).
412
Figure 7.22. Left: Photo of Individual 1005. Right: Drawing of 1005.
The anatomical association of the humerus with the femur is spaced too far apart to have
been anatomically possible. Furthermore, the relation of the femur to the lower leg portions leave
too great a space where the patella should have been. We did not recover a cranium nor any
teeth, which could be due to the looting of the northern portion of the unit. The postcranial bones
we did recover were of a relatively small size, and had evidence of weaving. Weaving in long
bones indicates that the bones are not fully hardened, and suggests that this individual was a
child or adolescent. Just west of the eastern altar and abutting the lowest step, then, a child was
interred on his/her side facing the altar. Those who interred this individual likely removed
him/her from a primary burial location and attempted to reinter in what they thought was the
413
correct anatomical positioning. The individual’s orientation adjacent to the altar suggests that
he/she was of high status within the community.
Finally, the last disturbed burial (Locus 1009) contained the cranium of a child and the
postcranial skeleton of an adult. This individual was also centrally located below the altar, and
was vertically interred such that we first found the cranium, with the leg bundle underneath (see
figure 7.23). I believe that Individuals 1009a and 1009b were originally interred in a group tomb
in another location. When the descendants of this individual attempted to relocate the burial, they
likely selected a skull and bundled legs which were spatially near each other, but not from the
same individual. The cranial bones were both small in size and contained the weaving pattern
present in other juvenile bones, while the femurs indicated normal adult stature and mature bone
development.
414
Figure 7.23. Vertical orientation of juvenile cranium, adult leg bones, and tupus.
Individual 1009 actually consisted of two individuals. Directly south of the bundle of
legs, we recovered an unorthodox tupu and a sewing needle, suggesting that at least part of this
burial consisted of an individual who was female (see Figure 7.24).
415
Figure 7.24. Top: sewing needle. Bottom: unorthodox tupu.
The interment of juvenile individuals in Early Colonial Period churches is well
documented, including the intentional interment of juveniles below the church altar. In his 2013
book on the bioarchaeology of mission cemeteries in La Florida, Stojinowski asserts that
subadults occupied a prominent position near the altar. In the cemetery of Guale at Santa Maria,
Stojinowski documents a trend in which “the very young (less than five years of age) were
buried only in the front row of the church” (2013: 152). Stajinowski suggests that this pattern
may hint that subadults occupied a privileged position in death, although he does not provide any
further elucidation on the meaning of this trend.
The disturbed burials are indicative of post-interment interaction between the
deceased/interred and the living. This practice would have been forbidden by Spanish church
authorities, and simultaneously was indicative of Taki Onqoy veneration of ancestors. The final
four interments consisted mostly of fragments of individuals, and we were unable to determine
416
mortuary treatment due to preservation concerns. Disturbed individuals were located throughout
the church’s interior space.
The first indeterminate individual (907) consisted of only a scapula isolated in the
westernmost part of the church. The second indeterminate individual (908) was also located in
the western portion of the church with its head to the north and legs to the south. Individual 908
suffered from extremely bad erosion—diagnostic fragments included 22 teeth and small crania
fragments from this interment, and we recovered no grave goods. Of the 22 teeth recovered, 6 of
the molars were deciduous (baby teeth), affirming that this individual was a child (White et al.
categorize child as 3-12 years old, and Gurevitz narrowed this down to between 5-6 years old,
plus or minus two years). The young age of the individual may explain the extreme degradation
experienced by the bones. We categorized this individual as indeterminate given the presence of
only the cranium and teeth, without evidence of long bones.
The next two indeterminate interments were located in the center of the church. The first
(Locus 955) contained only a very poorly-preserved cranium. It is likely that this individual was
a secondary burial, since we have only a cranium and no other connected human bone. The
cranium was situated upside-down, affirming that this interment was either secondary, disturbed,
or both. We also did not recover any teeth with this cranium, suggesting that it may have been
moved from another location. Due to the small size of the cranium, taken together with the
fragility of the remaining bone, we believe this individual was likely a child (3-12 years of age).
The fourth indeterminate burial (Locus 956) shared the same horizontal plane as
Individual 955, and was also located in the center of the church. Here, we recovered the remains
of a cranium, this time with assorted teeth. The teeth consisted of a mixture of unerupted canines,
incisors, and first molars, indicating that this individual was approximately 4-5 years of age plus
417
or minus one year and thus categorized as a child. We did not recover any postcranial bones for
this individual—since in this case we had cranial fragments with an absence of long bone
fragments, it is likely that Individual 956 was also secondary. In addition to the cranium and
teeth, we also collected a tupu in the same spatial proximity of the bones (Figure 7.25). The tupu,
cranium, and teeth seem to be intermixed, however, suggesting that not only was this individual
a secondary burial, but also that it was disturbed at a later date.
Figure 7.25. Mixed cranium and dental fragments, with tupu pin in the center.
Disturbed and indeterminate burials are grouped together here because they are similar in
the unorthodoxy of their interments. Disturbed burials were recovered throughout all units,
suggesting that there was no spatial patterning to those interments which were the foci of kin
interaction and repeated post-interment contact. Indeterminate burials were located in the
western and central portions of the church, and primarily consisted of body parts, including a
418
scapula and three crania. The indeterminate burials were also children. I suggest there are two
possible explanations for the indeterminate burials. First, these burials could have been primary
interments which degraded at a rapid rate due to the delicate nature of juvenile bone. Second,
these burials could have represented the secondary interments of parts of individuals—since
there was a prehispanic Andean focus on saving crania, it is possible that three of these
indeterminate burials were unsanctioned intrusions by living Andean kin.
Discussion of Spatial Location and Burial Styles
The Spanish religious center of Iglesiachayoq was a place full of contested practices and
competing notions of orthodoxy and heterodoxy. While the recovered mortuary assemblage
demonstrated six distinctive types of interment or disinterment (and an indeterminate category),
it is critical to highlight that we only excavated 12.5% of the surface area of the church. Our
sample is thus not representative of the overall church population, nor of the broader population
of Iglesiachayoq as a whole. With this caveat aside, we can still interpret the array of burial
practices as part of a spectrum of mortuary treatments in this Taki Onqoy center (Table 7.2.).
Ind.
No.
906
Location Style
Orientation Age
Sex
Western
Flexed
Primary
S-N
Middle
Adult
~35+
?
907a
Western
Flexed
Primary
NW-SE
907b
Western
Indeterminate --
Adolescent F?
~15+/-36
mo
-?
908
Western
Flexed
Secondary
N-S
Child
~5+/36mo
419
?
Grave
Goods?
--
Notes
Mandible
closed,
perhaps
elderly
Ceramic
teacup, 2
tupu pins
Jumble of
arm bones
Only crania
and teeth
909
910
Western
Western
Removed
Removed
---
Child ~4
--
?
?
2 tupu pins
911
Western
Extended
Primary
E-W
--
F?
912
Western
E-W
Western
Western
Western
Young
Adult ~25
--Middle
Adult ~35
?
913
914
915
Extended
Primary
Removed
Removed
Disturbed
Tupu, 4
Nueva
Cadiz
beads
Metal
pendant
955
Center
Indeterminate
--
Child ~10 ?
956
Center
Indeterminate
--
958
Center
E-W
960
Center
962
Center
Flexed
Secondary
Flexed
Secondary
Disturbed
1005
Altar
Disturbed
N-S
Child ~5+/ F?
36mo
Young
?
Adult
Young
F?
Adult
Young
?
Adult
Child
?
1008
Altar
E-W
1009a
Altar
Extended
Secondary
Disturbed
1009b
Altar
Disturbed
--
1010
Altar
Extended
Primary
E-W
--E-W
E-W
--
?
?
?
Young Adult?
~25
Child
F?
Crania
juvenile
Young
?
Adult
postcrania
Young AdultM
~25
--
Table 7.2. Summary of data from burial loci at Iglesiachayoq.
420
Locus
refilled with
large stones
Very
disturbed,
but likely
extended
Lower
limbs
primary,
head
secondary
Very disturbed, no
teeth
1 tupu
Possible bundle
4 tupu
pieces
Tupu and
needle
Possible bundle
Interred partially
extended on left
side facing the alt
Head propped up
facing altar
Juvenile cranium
Associated with
1009a
Stone coffin
There are several notable trends in the burials we recovered, which are distinctive on
different levels. In the following discussion, I will hypothesize as to explanations for some of
these trends, and discuss select individuals or groups of individuals to highlight my
interpretations. There are many ways in which the data from the burials at Iglesiachayoq can be
summarized, particularly in regards to the relationships between burial types, sex of burials,
burial positioning, ages of burials, and the spatial locations of these interments in the church.
Burial Type and Spatial Location
Overall, the 21 burial loci we excavated broke down into seven basic types of burials
uncovered at Iglesiachayoq, shown below in a bar chart (Figure 7.26). I have further broken
down the types of burials found in each spatial location (unit) in the church, as can be seen in
Figure 7.27.
Number of Burials by Type
6
5
4
3
2
1
0
Figure 7.26. Burial type distributions in the church.
421
Distribution of Burial Types by Spatial Location
5
4
3
2
1
0
West (17)
Center (18)
Altar (19)
Figure 7.27. Burial types by location.
As is seen in the charts, the most common burial type disturbed (n=5). The justification
for the overrepresentation of disturbed burials likely is due to the categorization strategy, which
included burials with missing bones, areas where there may have been looting (below the altar),
or loci in which the bones did not contain anatomical associations.
The second most common burial type was removed burials (n=4). This number may seem
high in relation to the array of other burial types--it makes up 20% of the sample. However, all of
the disinterred burials we found were in the unit closest to the western wall. Burial removal has
been explicitly linked to both Taki Onqoy and broader idolatrous practices (Doyle 1988; Gose
2008; Saignes 1999). Andean peoples were documented interring their dead in the church by
day, and disinterring and relocating these same relatives under the cover of night, in a covert
fashion. I argue that burial removal may have been a strategic component of Taki Onqoy—by
outwardly conforming to Spanish mandates, Andean peoples secured favorable social and
political standing in the eyes of the new authority. However, by removing and reburying their
422
dead in secret, takiongos would have demonstrated their respect for and belief in huacas and
prehispanic Andean traditions. The importance of removed burials in the church is hard to
overstate. In the center of the site, the space in which Spanish authorities had the most control
and oversight, Iglesiachayoq community members risked severe punishment to make sure their
dead had an “Andean” afterlife. Moreover, not only would the Andean agents removing their
dead be punished, but the physical bodies of the ancestors they were removing would have been
publicly burned, thus preventing a smooth transition into the afterlife (Ramos 2010).
My team and I recovered three examples of primary extended burials, and all were
located in sacred spaces beneath the church floor. Since this was the idealized type of burial for
the Catholic Church, it would make sense if the majority of burials were of this type. However,
because of the early date of the usage of Iglesiachayoq, I suggest that the array of burial types
predates Spanish Catholic insistence on extended style in the New World.103 There were
anomalies in the extended primary burials, however. First, Individual 911 and Individual 912
were buried in the northern portion of the westernmost unit, below the pillar which held the
baptismal font. This space would have been sacred for devout Catholics, and thus it is reasonable
that two extended burials were found here. However, both individuals were interred with luxury
items, including a metal pendant, Nueva Cadiz beads, and a tupu. Second, the extended burial
beneath the altar was the only individual interred in a stone sarcophagus. This individual was
also the only one which lacked any physical or cultural evidence for indigeneity—there were no
shovel incisors or grave goods.
Secondary flexed burials (n=2) were those which had likely been relocated from
prehispanic burial places. Both of these interments were oriented toward the altar, suggesting
103
For example, in this early phase, the variable nature of the burials is very different from the record at Morrope
(Klaus 2008) or in La Florida (Larsen 1993), where most burials were in extended form.
423
that these individuals had been relocated to the church and interred in Catholic orientation. The
intentional removal and relocation of Andean bodies potentially demonstrates a documented
tradition of “baptizing the dead,” or, interring previously deceased ancestors in Catholic spaces.
Andean individuals could have regularly practiced this in order to ally with the new dominant
power. By conforming to Spanish burial norms, Andean peoples demonstrated their support of
the church, and would have received social and/or political benefits from this association.
However, one of these burials included tupu pins—in this case, Andeans were subtly thwarting
strict Catholic laws by providing their ancestors with the clothing and drinking vessels necessary
for bodily fulfillment in the afterlife. Alternatively, those Andeans who reinterred deceased
ancestors could have been coerced or forced by Spanish authorities to do so. However, church
authorities would thus have been in violation of church edicts promoting separation of baptized
and unbaptized individuals. A third scenario is also possible in which Catholic church authorities
and Andeans were both willing to defer to the others’ traditions, thus creating entangled spaces
of interment. Without further investigation (not just at Iglesiachayoq but also at other Early
Colonial Period sites and in archival research), I am unable to conclude the intent behind the
secondary burials in the central church.
The primary flexed individuals were both recovered in the westernmost unit, though there
could have been more in the unexcavated portions of the church. Interestingly, Individual 906
was also the first burial we excavated, the closest to the church floor, had better preservation than
most of the burials, and was thus likely one of the last individuals interred given his/her location
abutting the western wall of the church. If Individual 906 was indeed one of the last burials
interred in Iglesiachayoq, his/her mortuary treatment could mark a return to indigenous burial
styles in a somewhat syncretic fashion, and is perhaps indicative of Taki Onqoy. It is important
424
to note that the Type 2 burials were found in the westernmost portion of the church, the location
farthest away from the altar and by proxy, farthest from the area of Catholic importance while
still remaining in the church.
In contrast, while primary flexed burials could possibly indicate a form of religious
resistance in the face of Spanish authorities, the secondary extended burial constituted the other
end of the spectrum. The secondary extended burial was found below the church altar and may
have signified a completely different attitude toward Spanish religious beliefs. This individual
was brought from a previous resting place and reinterred in the Catholic extended orientation.
His/her position in the front of the church below the altar suggests that the descendants of these
individuals would have fought for their place of spatial importance.
Age, Gender, and Spatial Context
Although we uncovered 17 individuals at Iglesiachayoq, the disintegrated nature of the
skeletal material made identification of biological sex impossible (Agarwal and Wesp 2017).104
For all individuals, we thus suggest that those interred with tupus may have been female, but we
also cannot rule out the possibility that other individuals without tupus were female, nor that the
absence of tupus affirmed the individual was male.105 There were two individuals in the
westernmost region and two individuals in the center region who were interred with tupus and
thus were likely female. I am not comfortable making an overarching claim about gender
Though bioarchaeologists actually work with “gray areas” more than the biological sex dichotomy would
suggest, it is only in recent literature that these scientists have begun to acknowledge the “actual existence of less
certainty and less dualism than the folk model that grounds the western dichotomous gender system in two natural
sexes” (Joyce 2017: 2).
105
This separation between male and female based on both biological determination and cultural markers of gender
identity are both false constructs, and I am currently thinking through ways to step outside of the normalized
dichotomy. aDNA and mtDNA testing is ongoing—if we find evidence of biologically male individuals interred
with tupus, then this suggests the interment of nonbinary genders, a phenomenon which has been documented
previously in the Colonial Period Andes (Horswell 2005).
104
425
patterning in church burials, however, given our small sample size and inability to confirm
gender hypotheses with skeletal data.
We have a much better data set for the age of individuals interred in the church. For most
individuals, we were able to either collect teeth, or judge the age at the time of death through the
quality of the remaining bone. In order to understand the age distribution of the recovered
individuals, Gurevitz followed the models of Buikstra and Ubelaker (1994) and White et al.
(2013) (Table 7.3). Of the 17 individuals, we were only able to determine age estimates for 16.
Age Classification (Buikstra and Ubelaker 1994; White
et al. 2012)
Fetus (before birth)
Infant (0-3 years)
Child (3-12 years)
Adolescent (12-20 years)
Young Adult (20-35 years)
Middle Adult (35-50 years)
Old Adult (50+ years)
Totals
Unit 17
Unit 18
Unit 19
0
0
2
1
1
2
0
6
0
0
2
0
3
0
0
5
0
0
2
0
3
0
0
4
Table 7.3. Age breakdown of individuals at Iglesiachayoq.
When disaggregating the burials by age and compare them with burial location, the following
pattern emerges (Figure 7.28):
426
Total
0
0
6
1
7
2
0
16
Number of Burials by Age Class and Unit
4
3
2
1
0
Fetus
(before
birth)
Old Adult
Infant (0-3 Child (3-12 Adolescent Young Adult Middle
(20-35 Adult (35-50 (50+ years)
years)
years)
(12-20
years)
years)
years)
Wetern Unit (17)
Center Unit (18)
Altar (19)
Figure 7.28. Number of Burials by age class and unit.
As one can be seen from the chart, the altar area (in gray), which would have been
reserved for important individuals, had two juveniles and three young adults interred below it.
There were two middle adults in our sample, and both were interred in the westernmost portion
of the church away from the altar. There were seven young adults in the sample, and these were
distributed throughout all of the units. The lack of “old adult” individuals indicates several
possible scenarios: first, the absence of old adults could signify broader health problems in the
overall population of Iglesiachayoq such that individuals were not living to old ages in general.
Second, the lack of older individuals could also demonstrate a difference in burial treatment;
older individuals may have lived both prior to and throughout the conquest and sought burial
treatment elsewhere. Third, the sample may have simply lacked old adults because of the
locations of units, or the remains of these individuals did not preserve well.106
106
Indeed, the poor preservation of very young and very old individuals can be greatly affected by pH levels in soil.
See Walker, Johnson, and Lambert 1988.
427
Excavations in the central church at Iglesiachayoq recovered a total of six children (3-12
years of age) and one adolescent (12-20 years of age). These seven individuals were relatively
evenly distributed throughout the church. Of the seven burials, three were indeterminate, two
were disturbed, one was removed, and one was primary flexed. I hypothesize that some of the
burials which were deemed as indeterminate or disturbed may have been due to the
compounding factors of general poor site preservation, and the possibility of greater degradation
associated with juvenile remains (Beauchesne and Agarwal 2018; Gordon and Buikstra 1981;
Walker, Johnson, and Lambert 1988,). We believe this hypothesis is supported by the
taphonomy of juvenile burials in general and the fact that this site was only briefly occupied after
Spanish arrival. It is logical that juvenile individuals would be present in the church at elevated
levels since survival rates for juveniles in history are low and children were often primary targets
of evangelization. Thus, we would expect children to be interred underneath the church. We do
not know if the children were buried coercively, willingly, or some combination of the two, but
we can suggest that children were central to the Spanish’s introduction and indoctrination into
the Catholic faith and overall acceptance of Spanish authority (Stojanowski 2013).
Secondary Burials
In all, we confidently identified three secondary burials, two secondary flexed individuals
in the center, and one secondary extended individual beneath the altar. These interments prompt
several questions, particularly regarding the negotiation of these individuals in the church. In the
eyes of the Catholic Church, unbaptized individuals were not to be interred in the church with
baptized individuals, while baptized individuals were supposed to be interred in the church
(Vargas Ugarte 1951). If all baptized individuals were to be interred in the church at this time,
428
then the secondary interment of these three individuals suggests that they were unbaptized. Why
would the living reinter their ancestors from their original resting places into the church? This
practice would not only make it more difficult for the ancestor population to interact with this
kin, but it would also deny kinsmen a physical locator of social memory of identity. In
reinterring these individuals into the new Catholic space, familial members would have been
forced to rupture portions of local memory and identity. The reinternment of skeletal bundles
beneath the church can be interpreted in two possible scenarios. First, living ancestors actively
chose to reinter their ancestors beneath the church in order to gain prestige or obey Spanish
religious authorities and/or to claim a space for their lineage within the new religion. Conversely,
the bundles may have been “captured” by the Spanish and were reinterred as part of a “forced
baptism” of the ancestors. This second interpretation seems unlikely, however, as baptizing a
dead body would have meant performing the very living/dead interactions which Spanish
authorities viewed as idolatrous.
Individual 1010
Individual 1010 was a primary extended burial found in a stone sarcophagus which
consisted of two courses of stones and sealed with five flat stone slabs on top, forming a stone
coffin or tomb in front of the altar. This burial is notable because it is the only individual with no
indigenous biological (shovel incisors) or cultural (grave goods, tupus, positioning) markers. Due
to the elaborate mortuary treatment of this individual, we hypothesize that he/she was an
attendant priest of Iglesiachayoq, or someone of high church standing. We are currently
performing aDNA and isotopic analyses in order to test this more conclusively.
429
Transforming Identities: Extended Burials in the Western Area of the Church
Individuals 911 and 912 were primary extended burials following Catholic tradition
which were recovered with the richest grave goods of any of the burials. Individual 911 was
recovered with both a tupu and colonial era Nueva Cadiz beads (Smith and Good 1982; Wernke
2013). The terminal production of these beads dates to the 1550s, and the beads have been shown
to be some of the earliest artifacts brought from Europe. Individual 912, located just south of
911, was interred with a metal pendant, and had most of his/her teeth present giving us an age of
~25 and demonstrating shoveled incisors. We conclude that individuals 911 and 912 were both
likely indigenous, as demonstrated by biological and cultural markers, but buried in Catholic
tradition and containing evidence of colonial contact and connection. The shoveled incisors
suggest Individual 912 was indigenous, while the combination of the tupu and Nueva Cadiz
beads affirms an intermixing of cultural artifacts.
Readdressing Hypotheses
To close this chapter, I revisit the hypotheses I generated in earlier sections. My general
hypothesis regarding mortuary practice at Iglesiachayoq was as follows:
If the inhabitants at Iglesiachayoq were challenging Catholic mandates in burial practices, then
interments would reflect Andean traditions rather than standardized Catholic norms in terms of
burial position, burial goods, and disinterment or interaction with interred bodies.
My archaeological excavations in the church at Iglesiachayoq demonstrated a wide array
of burial practices which were in contrast to the idealized mortuary treatments put forth by the
First Council of Lima. We only recovered three primary extended individuals—these individuals
were located in sacred spaces within the church (below the Holy water and the altar), and I
430
suggest these were examples of Andean converts.107 Beyond the three primary extended burials,
we recovered one other extended individual below the altar, although this interment was
secondary. These extended burials all signified orthodox Catholic practice vis a vis mortuary
treatment. However, the vast majority of interments beneath the church floor at Iglesiachayoq
were unorthodox or heterodox. I argue that inhabitants of Iglesiachayoq were subverting or
ignoring Catholic burial traditions, or that these Andeans were attempting to adapt to Catholic
mandates and were acting independently of a Spanish authority. A third possibility could be that
the Spanish priest which attended Iglesiachayoq did not demand uniformity in burial practices.
Materially, this divergence from Catholic traditions manifested materially through the flexed
primary and secondary burials, the evidence of interactions with the deceased, the inclusions of
grave goods, and most significantly, burial removal.
Since Albornoz listed burial removal as a practice indicative of Taki Onqoy, I support the
hypothesis that removed burials are clear evidence of the religious revitalization movement. This
conclusion is far more compelling when paired with the findings of household excavations,
which recovered no clear evidence for rejection of Spanish goods. Taking the two chapters
together, then, my findings at Iglesiachayoq create a paradox: in private areas away from the
religious center in Sector 1, there was no clear indication of Taki Onqoy, as manifested through
the rejection of Spanish goods. However, it is in the area with (nominally) the most Spanish
oversight—the church—in which we see the greatest variety of unorthodox practices which
directly relate to how Taki Onqoy was performed at Iglesiachayoq.
107
See Appendix D for isotope and carbón dating—Isotope data confirmed that all of those interred beneath the
church were consuming the same diet prior to death, so I hypothesize that all were local individuals.
431
Conclusion
Although the sample of burial loci was relatively small, my excavations at Iglesiachayoq
can provide preliminary data regarding the array of burial treatments—and by extension, a
diversity of intra-community orientations regarding mortuary practice—in the Early Colonial
Period. The Spanish Catholic church at Iglesiachayoq visually dominates the settlement, and
would have been the locus in which Spanish religious authorities had the most control and
oversight. Nonetheless, in this place of ideological importance, we find the most evidence for a
spectrum of resistive practices in burial style and positioning. Because the dead and the treatment
of the dead were ideological dimensions so critical to Andean peoples, many went to great
lengths in order to insure the protection of their ancestors after death. By completing practices
which reflected Andean traditions (flexed positioning, burial goods, burial direction, burial
removal), local peoples were demonstrating their respect for their native traditions.
Taki Onqoy was a movement in which local preachers advocated for the rejection of
Spanish tradition in favor of a return to prehispanic huaca worship. I suggest that Andean
peoples caught between these two extremes could not simply choose to follow one or the other
and hope that they selected the “winning” side. Instead, I believe Andean peoples took a variety
of stances in their negotiation of indigenous and Spanish religious traditions. As we can see from
the burial data, some individuals interred their dead in deliberate Catholic extended style, with
heads to the west and feet to the east. On the other side of the spectrum, some individuals
interred their ancestors, only to remove them at a later point in time. Andean peoples likely
practiced aspects of both Taki Onqoy and Catholicism—in doing so, they could navigate both
poles of control. The diverse array of burial practices asserts that while bodies may have been
432
spatially interred in a Catholic center, the physical placement of the body often conformed to
Andean beliefs.
As discussed in previous chapters, I do not find evidence of ruptures in daily life which
suggests Taki Onqoy had a massive effect on sixteenth century Andeans. Indeed, the sparse
quantity of Spanish authorities led to a material imprint that was very faint in the first decades
after the conquest. That is, Old World pottery, animals, crops, and material traditions did not
become commonplace in the archaeological record until their presence was established. Thus, the
continuity of prehispanic daily practices in the immediate aftermath of the conquest is expected.
However, it is within the critical religious and ideological place—the church—where I find
variance consistent with Taki Onqoy. While Andean peoples may have attended church services,
they still interred or disinterred their dead in heterodox ways which were distinctly Andean.
433
CHAPTER 8
SPATIAL ANALYSIS OF VISIBILITY AND WALKING PATHS AT IGLESIACHAYOQ
The previous two chapters presented aspects of the recovered data which demonstrate
clear variations in household and burial practices at Iglesiachayoq. Domestic assemblages
showed inconsistent incorporation or rejection of Spanish cultural practices and materials,
though no clear discrepancy could be discerned between Late Horizon and Early Colonial Period
assemblages. There were few definite or obvious changes in consumption, ceramic traditions,
and daily practices during the Late Horizon and throughout the Early Colonial Period at
Iglesiachayoq. The exceptions to this trend were located in household excavations in Sector 1
(the Spanish ritual center), which had a higher proportion of Old World foodstuffs and artifacts
which are contextually consistent with the sixteenth century. Similarly, mortuary remains
excavated from beneath the central church also demonstrated variations in practices in mortuary
treatments, reflecting a range of interments—traditional Catholic-style burial (primary extended),
traditional Andean highland-style (primary flexed with a range of orientations in relation to the
altar), secondary burials in both Spanish and Andean styles, interaction with burials, disturbed
burials, and finally, removed burials. What could account for the continuity in domestic
assemblages and the proliferation of change in mortuary contexts?
At its core, Taki Onqoy was a movement fraught with ambiguity and contradiction—it
was both anti-Catholic, yet its framework was influenced by Catholic tenets. It preached the
rejection of Catholic and Spanish goods and traditions, yet many of the female leaders of Taki
Onqoy took Catholic names such as Mary and Magdalena. While Molina and Albornoz
described Taki Onqoy as a covert movement, its main practice—dancing—was performative and
required several days of drinking and fasting. How can the archaeological record provide
434
material insight into a movement which was laden with these contradictions, and whose main
performative practice was ephemeral? Specifically, this chapter addresses the final hypothesis to
be examined in this dissertation:
If Taki Onqoy was a clandestine movement, evidence of the materials associated with its
performance will be uncovered in private areas away from central religious spaces. If there are
public spaces of Taki Onqoy performances, these will be located away from places of Spanish
administration and Catholic practices.
In this chapter, I consider how spatial structures—visibility and invisibility, and
proximity to sacred landscape features or distance from the Spanish church—can aide inference
of Taki Onqoy and Catholic spatial practices. I will first discuss visibility and sacred landscape,
and then briefly summarize the distribution of structures and sectors, denoting areas which were
more associated with local ceramics and locations. Next, I will show how spatial analysis
methods (Viewsheds and Spatial Network Analysis) can model places of invisibility and local
and Spanish movement through the various spaces at Iglesiachayoq. By combining these two
methods, I aim to understand what daily life was like for local people at Iglesiachayoq in two
contrasting ways—a static approach which analyzes specific visibilities from fixed points at
Iglesiachayoq, and a dynamic approach which explores travel through a landscape.
For both of these analyses, I consider variation in the spatial organization not only among
the different sectors of the settlement, but also specific buildings (determined as anomalous in
their domestic assemblages), open spaces (central plaza, upper plaza), areas of sacred importance
(the church or proposed Taki Onqoy spaces), and, more broadly, potential differences between
structure types. In this way, I hope to gain an understanding not only of how Spanish authorities
435
may have manipulated the spaces of daily interaction and ritual, and also how local peoples
resisted these transformations during Taki Onqoy.
Taki Onqoy Spaces and Catholic Spaces
The primary source documents which discuss Taki Onqoy provide little information as to
where the ritual drinking and dancing occurred at the site-level, and the witnesses who describe
these rituals often provide contrasting accounts. Moreover, the physical spaces where Taki
Onqoy was performed changed over its lifetime and were variable between different settlements.
In the earliest account from 1564, the letter from Guerrero to Luis de Olvera (Chapter 2),
takiongos are described as climbing to mountain peaks and other places of importance in the
landscape which were venerated as huacas. It was at these remote locations where Taki Onqoy
rituals were performed, thus allowing the movement to remain underground for the initial years
of its existence. During the 1570s, when Taki Onqoy was understood as a distinctive idolatrous
phenomenon by Spanish extirpation priests, the documents affirm that these dances were
undertaken in circular outdoor enclosures which were explicitly constructed as Taki Onqoy
spaces (Albornoz 1990: 191, Molina 2010). Finally, in the 1580s and 1590s, after the active
campaigns of Albornoz, the movement is described as being driven underground, and accounts
suggest that the movement’s performative aspects were enacted within domestic households
(Álvarez 1998: 126-127).
With these contradictions in descriptions of the physical spaces of Taki Onqoy
performances, I suggest it is useful to broaden the frameworks within which we interpret these
places. It would be an impossible task to excavate every household, outdoor space, and sacred
mountain top at or around Iglesiachayoq. However, advances in GIS and spatial analysis allow
for the modeling of potential spaces of Taki Onqoy performances. Specifically, I utilize two
436
analytical concepts to model where these performances may have occurred: visibility and
invisibility, and proximity to Spanish and Taki Onqoy sacred landscapes, as considered through
pathways and walking time.
Visibility and Invisibility
Taki Onqoy was described by multiple priests as a covert movement (Albornoz 1990;
Álvarez 1998; Gose 2008). These priests and witnesses were alarmed at their collective
impression that the movement had arisen—fully formed—so rapidly, thus leading them to
conclude that Taki Onqoy had been practiced in hidden places at carefully-planed times when
Spanish authorities were absent. Punishments for Taki Onqoy practices were harsh and swift
(Chapter 2), and involved physical violence (whipping), symbolic embarrassment (shaving of
heads, hats), and forced attendance at Catholic rites (Albornoz 1990). Yet, Taki Onqoy rituals
often lasted several days—they were not ephemeral, rapid performances which were complete in
the span of an afternoon. They were likely important community events. Since these
performances were lengthy and yet, seemingly clandestine, I suggest that they were enacted in
spaces which were less visible to Spanish authorities. In this chapter, I use notions of visibility
and invisibility to model spaces where Taki Onqoy performances may have occurred. To address
this spectrum of visibility, I utilize viewshed analyses at Iglesiachayoq.
Measures of visibility as assessed through viewshed analyses have become a common
method in the archaeological toolkit, and have been used to address a number of theoretical
questions such as visual connectivity (Lake and Ortega 2013), visibility in the structuration of
sacred landscapes (Bongers et al. 2012), or even phenomenological understandings of visual
perception (Tilley 2010). Viewshed analyses display a rather simple visible/not visible map
437
which has been refined according to “manipulation of view angles and parameters, fuzziness,
visual acuity, visual prominence, horizon delineation, and 3D visibility modeling” (Gilling 2015:
1).108 Yet, the rise of viewshed model modification has not concomitantly inspired theoretical
diversification (Gillings 2017). Archaeologists still use visibility as a dominant characteristic for
assessing the importance of sites, features, and interconnectivity.
Notably under-utilized is the application of viewshed analyses to address invisibility.
With an emphasis on seeing and being seen, viewshed studies have largely neglected desires for
seclusion, privacy, or the necessity for being unseen (Gillings 2015). In his 2015 article, Gillings
applies viewshed analysis models to assess non-intrusive prehistoric megalithic monuments,
concluding that the monuments were “neither visually prominent or show any evidence of being
concealed, hidden, or deliberately tucked out of view” (Gillings 2015:13). Gillings conducts
these analyses largely as an example of how they could be applied, and he takes a regional view
of the monuments. Building off of his study, I use visibility and invisibility to model spaces for
future investigation, and as an analytical tool to explore which areas of Iglesiachayoq might have
been more beneficial to takiongos who sought privacy and seclusion in order to avoid potential
punishment. In addressing the final research question regarding where Taki Onqoy was
performed, I hypothesized that if Taki Onqoy was clandestine (as suggested in documents), then
evidence of the materials associated with its performance would be uncovered in spaces which
were less visible to Spanish authorities.
Proximity to Sacred Landscapes
108
See Bernardini et al. 2013, De Reu et al. 2011, and Ogburn 2006 for examples of these.
438
In addition to the notion of covertness or invisibility, Taki Onqoy performances were also
intended to reestablish connections to a prehispanic pantheon of huacas, while simultaneously
explicitly avoiding Catholic spaces (Chapter 2). As discussed in Chapter 2, initial Taki Onqoy
rituals were performed at the sacred places of landscape huacas, often at mountain peaks,
streams, lakes, rock outcrops, etc. It is thus likely that in addition to being performed in places
invisible to Spanish religious authorities, Taki Onqoy rituals also were enacted in places
associated with prehispanic landscapes, and further from Catholic spaces. I thus suggest that
Taki Onqoy dances would be located in places which were clearly connected with local
traditions and simultaneously explicitly detached from places of Catholic or Spanish importance.
In order to model where these places might be located, I utilize a Spatial Network Analysis,
which measures movement throughout a defined network of pathways.
Spatial Network Analysis (SNA) allows modeling of movement through the built
environment of any landscape (Cooper 2017; Entwisle 1997; Wheatley and Gillings 2002;
Wernke 2012, 2013, and 2017). Although SNA is most commonly used in modern urban
planning, the tool has more recently been employed by archaeologists to map past movement and
changes in access to various areas of sites. For example, in the Peruvian Andes, Wernke has
demonstrated that Spanish restructuration of the built spaces at the site of Malata (Colca Valley,
Peru) routed traffic away from the Inka ceremonial center of the site and toward the Spanish
Catholic center of the site, despite the two regions being spatially near one another (Wernke
2012). The site of Malata is directly comparable to Iglesiachayoq—both are small (80-90
structures) settlements with Inka and Catholic public and ritual spaces which are near in
proximity to one another. Additionally, Iglesiachayoq and Malata both have standing
architecture, allowing for the clear discernment of accessways, pathways, and walls at both sites.
439
Building on Wernke’s work at Malata, my own work at Iglesiachayoq considers past pathways
and accessways as potential areas of movement—spaces where individuals could move through
an environment. In contrast, walls, other structures, and environmental barriers such as steep
rock faces or impassable boulders were considered to be impedances which helped shape
movement by leading people to move around these obstacles. Doorways and walls were assumed
to be guiding factors in shaping movement—that is, people likely moved through doorways
instead of climbing over walls.
Built Environment and Architectural Styles at Iglesiachayoq
Since I have already provided a full description of sectors and structure types in Chapter
6, I will provide a spatial history and discussion of architectural masonry styles here. As
discussed previously, the occupation at Iglesiachayoq spans from the Late Horizon through the
Early Colonial Period, until the people were relocated to the reducciones of either Soras or
Larcay. Within the site layout, it is possible to discern broad differences between construction
periods, with Sectors 1 and 2 containing the shortest occupations and most recent built
constructions, and Sector 3 marking the oldest portion of the site.
The Central Church Complex (Sector 1)
Although there has been little work completed at Iglesiachayoq, the two scholars who
have invested time surveying the site—Meddens (1985) and Mallco (2013) both categorize the
site as being constructed during the Late Horizon, as suggested by the Inka cut-stone masonry in
several structures. For example, in his 2013 article, Mallco suggests that the large central
rectangular structure “because of its dimensions and the form it has, the population have deemed
440
it a church, from which the name Iglesiachayoq is derived, but in reality the residence is a
kallanka which pertains to the Inka epoch” (Mallco 2013: 227).109 Although Mallco is correct in
that the church does share the proper dimensions as a typical kallanka, I suggest that he is
focused on categorizing sites according to architectural canons which have been used as markers
of “Inka.” He thus does not consider the possibilities of a colonial occupation at Iglesiachayoq.
Meddens is slightly more cautious in his description of Iglesiachayoq, despite excavating a “few”
test pits at the site. Like my own experience at the site, Meddens found a low density of surface
ceramic, and that which he did recover he considers Late Horizon or Early Colonial. Meddens
goes on to say that “the largest rectangular building may be of early colonial date (pers. Comm.
J.H. Rowe), whereas some of the others, like the ones with polygonal ashlar masonry, date to the
Late Horizon” (Meddens 1985: 103).
Albornoz’s documentary account and my own archaeological research at Iglesiachayoq
support Meddens’ proposal that the site consists of a mixture of Late Horizon and Early Colonial
Period structures. Albornoz attests that those takiongos punished for their idolatrous practices
were put to work constructing the church here. There are two significant conclusions which can
be drawn from Albornoz’s account regarding church construction: first, this suggests that the
church was constructed as a church, and was not a kallanka which was later converted into a
church. Excavation results confirm this conclusion, as my team and I encountered a stratigraphy
which consisted of fill with few ceramic fragments in the upper levels of excavation, then
burials, and finally, sterile soil which was nearly impossible to remove. Excavation results from
the church, then, contrast with excavations of other kallankas at contemporary sites. For
Translated from: “Por la dimension y la forma que tiene los pobladores lo denominan como iglesia, de alli
proviene el nombre de Iglesiachayoq; en realidad el recinto es una kallanka que pertenece a la época Inka y se
encuentra en buen estado de conservación, solo las paredes cerca de los accesos presentan derrumbes” (Mallco
2013: 227), translation mine.
109
441
example, at Huánuco Pampa, Morris et al. report that in their excavations of kallankas, they
recovered intense food preparation areas, ash lenses, secondary activities indicated by spindle
whorls and needles, musical instruments, and abundant ceramic material (2011: 43-45). Had the
church at Iglesiachayoq been originally conceived of as a kallanka, I would have expected to find
evidence of diverse activities ranging from food preparation to ceremonial events and other
assorted contexts. Notably, in my research on kallanka excavations, I have not found any
examples of multiple burials beneath the surface.
Second, Albornoz suggests that as part of the takiongos’ punishment, they were
conscripted to assist in construction on the local church. The church masonry is clearly in the
style of other local structures at the site, with large base stones at the bottom supporting roughly
dressed fieldstone, which is finally covered with stucco. On the western wall of the church, this
stucco was decorated with black, cream, and red paints and likely once depicted a mural. In the
eastern altar, there are three low steps leading up to a flat space, and the eastern wall contains an
altar niche as well—these features are characteristic of church construction at this time period.
The four interior corners each contain two diagonal wooden braces, which likely were placed to
enhance the stability of the structure. The construction of the outside corners of the church are
modified angular blocks, again very similar to the other structures with angular corners at the
site. Finally, although the church contains no niches consistent with Inka construction styles, it
does have three rectangular windows on each long axis, as well as two circular apertures, one in
each of the short axis gables. When considering all of the characteristics of the church and
comparing them with other structures at the site, as well as other contemporaneous local or Inka
442
structures, it is likely that local individuals were heavily involved in both the design and
realization of this structure.110
It is reasonable to conclude that community individuals constructed the central church at
Iglesiachayoq—there would not have been enough of a permanent Spanish presence at the site to
completely oversee the project. Instead, Spanish authorities likely tasked the local population
with constructing the church, and then this population was given leeway to erect the building
utilizing local traditions of construction. Archaeological investigation and architectural analysis
thus are consistent with Albornoz’s account which suggests that local individuals were actively
invested in constructing the new church.
The church’s construction and placement in the built environment are both critical to
examine in order to understand its desired function within the site, as well as how the structure
incorporated Spanish and local traditions. As Moore writes, “Architecture is more than a passive
product of potential later investment; it reflects other dimensions of public life and in turn, helps
shape the nature of social interaction” (1996: 3). The architectural styles influence the overall
ritual experience—accessways can funnel individuals to certain spaces and the altar can impart a
feeling of sanctity given its elevated location within the church. Similarly, the church’s overall
location within the built environment can provide clues as to prehispanic practices and Spanish
attempts at physical dominance of local places on the landscape. In some cases, colonial
religious structures were deliberately located atop spaces of indigenous religious importance, or
incorporated aspects of the indigenous sacred landscape. As Lara writes, “keeping existing
shrines intact allowed the Spanish to bring people to God in a familiar place” (Lara 2008: 20).
110
So much so that my own excavation team thought that the structure could have dated to the Late Horizon initially
because the construction styles were identical.
443
There are several examples of Spanish church construction deliberately placed atop Inka
structures, a literal physical dominance of the previous religious order.111 As Abraham points
out, however, placing a church on top of an Inka or local structure is not the same as constructing
a church which includes aspects of local religious architecture (Abraham 2017:220; Rosenfeld
and Bautista et al. 2017). Moreover, indigenous religious practices did not always require a
specific sacred structure, but rather the spaces where people worshiped were imbued with
meaning through dance, chanting, and music.
The church at Iglesiachayoq differs from other churches at this time in greater Peru. For
example, in the Colca Canyon, Wernke documents several churches and chapels at doctrinas
which were of a much smaller scale than at Iglesiachayoq, and which were constructed with the
access on the short axis, and a hexagonal altar area (Wernke 2013:168-175). The idealized
church construction plan imported from Europe, however, included a cruciform floor plan with
both an apse and a nave, with the altar situated toward the east, where Jesus would rise
(Abraham 2017: 231). In contrast to either of these floorplans, the church at Iglesiachayoq is
fully rectangular, with no apse and only one long solo nave, 33m x 9m. Although the church at
this site does not follow typical canons, it is remarkably similar to other churches in the
Huamanga and Apurímac regions.112 For example, the church of San Cristóbal de Pampachiri
has only one solo nave and measures 35m x 8m. In addition, the church of San Juan Bautista
constructed above the Inka ruins at Vilcashuaman shares these same dimensions. All three
churches are oriented with the long axis running east/northeast to south/southwest (Barnes 1993:
7). Perhaps, then, in this region rife with Taki Onqoy practitioners, the church construction style
111
For examples see Huaytara, Vilcashuaman, and the Santo Domingo convent in Cuzco.
I would also argue that given the paucity of archaeological investigation into the Early Colonial Period, there is
no cohesive “standard” construction style for these very early churches.
112
444
took on a distinct local form, while still conforming to the proper orientations dictated by the
Spanish.
The church at Iglesiachayoq seems to have been inserted into the sacred built
environment at the center of the site. The church dominates this sector of the site, and has both a
circular plaza upon which it opens, as well as an anterior plaza behind the church. These spaces
located adjacent to church construction were often used as overflow burial spaces once the
interior of the church was full (Ramos 2010). Beyond utilizing this large space, the construction
of the church effectively blocked off the kuraka house (Figure 8.1). The entrance to the church is
located to the north, while the entrance to the kuraka house is to the south, and they are situated
adjacent to one another, such that the kuraka would not have been able to see church activities
from his home. A small wall was also assembled physically separating the kuraka house from the
church, clearly dividing the space to create two distinct areas of use.
445
Figure 8.1. Map of central church complex with structures and plaza mentioned in the text.
This low wall continues mostly uninterrupted, encircling the entire plaza with the
exception of two accessways (one to the east and one to the west), and a small 5m x 5m
rectangular structure with squared corners (Structure 003). This small, symmetrical structure
appears to have been an associated chapel—its entrance opens onto the church plaza, and its
form is consistent with chapels in other parts of the Andes (Wernke 2013).
Lastly, the church complex was likely associated with eight out of the nine rectangular
structures with squared corners other than the church itself. To the northeast of the church and
central plaza, there is a small complex of three structures surrounded by a privacy wall (see
446
Chapter 6 Unit 2, Structures 029, 030, 032). These structures were constructed such that their
entrances faced one another and opened up onto a shared common space. The privacy wall has
only one access, which is located in the western portion of the complex, facing the central
church. These three structures were the closest residential structures, and would have had
visibility and easy access to all church activities. Excavations in one of these structures (Chapter
6) revealed two very shallow occupation levels, and recovered a mixture of domestic artifacts
such as lithic tools, grindstones, and ceramic from both the Late Horizon and Early Colonial
Period. The size of the structures within this complex, the short occupation period, and the
restricted access to the complex suggest that these structures were built at a later date than most
areas of the rest of the site, and that they may have been occupied, at least periodically, by elite
individuals, or Catholic authorities or practitioners.
One last notable feature of the central church complex was Structure 004, a small
rectangular structure with squared corners which was used for extensive food and drink
preparation. Because of the comparative well-preserved contexts of this structure, it likely
suffered much wall collapse at a relatively early date. Structure 004 had several enormous
rimmed-vessels, likely used for either food prep or storage. There was a concentrated burning
lens which contained the remains of a juvenile camelid, as well as burned maize (Chapter 6 Unit
4). Given the large size of the vessels in conjunction with the broad hearth, it is apparent that
those who were utilizing this structure were preparing food and drink on a scale beyond the
nuclear family. I suggest that this structure, then, was associated with activities in the central
plaza and church.
In sum, the built environment of this central church complex extended beyond the church
and central plaza. The architectural form of the church corresponds to other churches in this
447
region, while the masonry of the church aligns with local construction styles. This suggests that
those who constructed the church were local, but also mapping onto church forms in the greater
region. Finally, the spatial layout of the church complex can give clues into the rearrangement of
space post-conquest. The church and many of the rectangular structures were likely additions to
this central sector, and they effectively shifted the layout of the built environment to detract from
the spatial dominance of the kuraka house. By constructing a low plaza wall and dividing the
kuraka house from the church, the postconquest spatial layout shifted the focus of the site to the
church, and set this complex apart from the rest of the sectors.
The Upper Prehispanic Plaza Complex (Sector 3)
The southwestern sector of the site (Sector 3) contains the deepest occupation levels, and
artifact assemblages confirm the more “localized” nature of deposits in this area. Most strikingly,
there is a semi-circular plaza in this area of Iglesiachayoq, which was constructed by
incorporating boulders and large fieldstones into the natural hillside (Figure 8.2). The
“prehispanic plaza,” as I am calling it, resembles an amphitheater with its inset construction
style. This plaza is walled on all sides and can be accessed only by specific entrances toward the
east, west, and south. Furthermore, it occupies the highest point in the southwestern sector of
Iglesiachayoq, and so foot traffic from the church and central plaza could be easily monitored by
those in the local plaza. Supported by archaeological evidence from domestic excavations in this
area of the site, I suggest that this prehispanic plaza corresponds with descriptions of the outdoor
enclosures constructed for Taki Onqoy dancing, as affirmed by Albornoz. Other sites such as
Torata Alta have similar built environment features: for example, Rice writes that “the walls of
structures at Torata Alta frequently incorporated sizable boulders, which calls to mind the well-
448
known Inka veneration of rocks and stone outcrops, many of which are large and elaborately
carved” (Rice 2012: 22). Similarly, the upper plaza walls integrate carved boulders within the
built space.
Figure 8.2. Prehispanic plaza in southern sector (Sector 3). Notice the large boulders
incorporated into the built environment, photo looking north.
There is only one structure associated with this upper plaza, which is a 6.45m diameter
ovoid structure, whose only accessway opens northward into the upper plaza. Excavation units
placed in both the plaza and the circular structure did not recover the typical indicators of
household activities (Chapter 6, Units 8 and 15). In the circular structure, there was very little
ceramic, and that which was recovered was incredibly fragmentary and difficult to identify.
Some of the larger boulders incorporated into the wall construction of this structure had dark
discoloration, suggesting they had been exposed to multiple burning events. However, no
concentrated hearth or faunal remains were recovered in this structure. It may be that these large
boulders were reminiscent of the burning events described by Rice 2012 at Torata Alta, in which
she argues that “what is particularly striking is that in six residences, hearths were positioned at
449
the base of these boulders (or against the bedrock). This suggests the continuation of longstanding indigenous practices of burning offerings to stones, use of fire in divination, and
Aymara beliefs in ‘house spirits’” (Rice 2012: 22). Since excavation results demonstrated a lack
of clear evidence of domestic activities in the circular structure and the semi-circular plaza, it is
plausible that these spaces were used for more ephemeral activities such as performance or
dancing. The structures in this broader “local sector” share attributes suggesting they were both
older, and could be considered “Soras” or “local.” The units excavated here had more extensive
occupation levels, the ceramic was much more “rustic” in nature and largely undecorated, and
the sector is the nearest to Soras households located to the southwest on a related hillside. If Taki
Onqoy was preferentially practiced in areas of local importance, then the local sector would have
been an ideal location for the ritual dancing.
Sector 2 Layout: A Haphazard Grid?
The spatial layout of Sector 2 (northeastern) was distinct from Sectors 1 and 3, in that it
was plausibly intentionally gridded. In fact, Sector 2 is known as an entirely different site to
some local people, called Wallpa Wiri. For example, in his 2013 survey, Mallco separates this
sector as its own separate site apart from Iglesiachayoq. In his dissertation, Meddens does not
acknowledge the local denomination “Wallpa Wiri,” but in the 2010 article with Schreiber, the
two authors also affirm Wallpa Wiri as a separate sector rather than a separate site. From my
own work at Iglesiachayoq, the Wallpa Wiri sector has masonry styles and architectural types
which are consistent with those of the rest of Iglesiachayoq, suggesting that the same
construction tactics were used in all sectors. Wallpa Wiri consists of quadrangular structures with
either rounded or squared corners. Like Sector 1, there is one massive structure (Structure 107)
450
which dominates the landscape—this 13.7 x 8.5m rectangular structure has a north/south
orientation, with entrances on both short axes. Although Meddens documents that there is only
one entrance in the southern short wall, my team and I confirmed the presence of a northern
accessway as well, lined with Inka polygonal cut-stone masonry. Like the kuraka house in Sector
1, this structure was constructed of roughly-dressed field stone with a mud mortar. Most
strikingly, there are also 18 niches decorating the inside of this rectangular structure, matching
the kuraka house from Sector 1. According to Meddens and Schreiber’s 2010 article, there were
wooden braces in the corners of this structure, much like in the central church; however, my
survey at the site did not document these wooden braces, so they may have been removed in the
last several years.
Thus, the rectangular church and Structure 107 in Wallpa Wiri share many architectural
features. First, the large Wallpa Wiri structure is the only structure of this form in Sector 2—all
of the other rectangular structures with squared corners are located within 100m of the central
church. Second, the wooden braces documented in the corners of both structures suggest that
either the same populations built both structures, the structures may be contemporaneous, or the
structures may have utilized existing knowledge in the region. Third, Structure 107 also contains
at least four floor-level arched niches (a colonial feature, see Wernke 2013: 194; Figure 8.3).
Given the presence of these altar niches, it is possible that this structure has an Early Colonial
Period construction date, consistent with the church. That this structure mimics not only the form
and construction style of the church, but also borrows traits from the kuraka house in Sector 1
(same amount of niches, Inka cut-stone masonry) is intriguing. Could this structure have been an
Inka construction in the face of Spanish conquest? More research is necessary to identify the
451
function and construction date of this structure, and I will look into excavation here in future
years.
Figure 8.3. Floor-level arched niches and rectangular niches in Structure 107, photo looking
west.
The rest of the structures in Wallpa Wiri also seem to have been constructed at a later
date than the rest of Iglesiachayoq, indicated by the spatial layout of this sector and excavation
findings. The vast majority of the structure forms here were quadrangular with rounded corners
(n=24/27 total structures), and most were relatively the same size, average 6.45m x 4.87m
(standard deviation 0.74m x 0.87m). Unlike the distribution of ovoid structures in Sectors 1 and
3, there were only two ovoid structures in Sector 2, both of small size and which my team
considered colqas or small storage areas. These two structures were associated with larger
quadrangular structures. Furthermore, the layout of the structures in Wallpa Wiri seems to have
been more clearly planned—although the accesses to the structures vary in direction, the
structures themselves are laid out in several rows which follow the natural landscape. There are
systematic walls which define the property lines, each with clear accessways for people passing
452
through. The partial grid in Wallpa Wiri is not consistent with the sprawling nature of the
structures in Sectors 1 and 3, and thus may have been planned beforehand or constructed at a
later date (Figure 8.4).
Figure 8.4. Sector 2 layout with Structure 107 and units.
As mentioned briefly above, the excavation findings from Wallpa Wiri support the
hypothesis that this sector of the site may have been constructed post-conquest. My team and I
excavated three units in three separate households, finding occupation levels well under 50cm in
depth, some Old World faunal remains, and what appears to have been an iron doorlatch, dating
to the sixteenth century. All ceramics date from the Late Horizon, and possibly some from the
Early Colonial Period.
453
Synthesis of Iglesiachayoq building phases and sectorization
Although the three sectors of Iglesiachayoq are clearly related and contemporaneous,
each sector also has its own distinctive “identity” which separates it from the other two. Each
sector has a central focus—whether it be the central church and plaza, the prehispanic plaza, or
the large rectangular structure in Wallpa Wiri, those who resided at Iglesiachayoq in the different
areas had clear spaces in which they could interact publicly. Similarly, the masonry techniques in
all three sectors consisted of roughly-dressed fieldstone with mud mortar, and rectangular or
trapezoidal doors and niches in some structures. Finally, ceramic findings were relatively
consistent site-wide with nearly all dating to the Late Horizon, with the possibility of some Early
Colonial Period sherds.
Conversely, there are broad differences between each sector as well—these differences
can be categorized according to common structure form, overall layout, and occupational
deposits. The three sectors have significant differences in the amounts of each structure form:
while Sector 1 contains all but one of the rectilinear structures (quadrangular with squared
corners), Sector 2 is dominated by quadrangular structures with rounded corners, and Sector 3
has an even mix of ovoid structures and the rounded-corner square structures. The overall layout
of each sector differs greatly from one another as well. While the sectors 1 and 3 both appear to
have been unplanned, with urban sprawl organically radiating from site centers, Wallpa Wiri
seems to have been planned on a haphazard grid. In this sense, the built environment of Sectors 1
and 3 corresponds to some areas of Huánuco Pampa. As Morris et al. attest in their Huánuco
Pampa 2011 monograph:
The encroachment of residential areas onto open spaces and accessways suggests
that the Inka state took a fairly hands-off approach to local residence at the
provincial capital—local groups might be required to be in the plaza on a given
454
day for a festival or administrative proceeding, but state administrators were not
necessarily concerned with how they arrived at the city or the nature of their
accommodations…there is little evidence of central planning or layout, and
houses are fairly undifferentiated and laid out opportunistically, leaving open
spaces for accessways and modest open spaces for communal activities. Morris et
al. 2011: 31.
Although Iglesiachayoq was founded by the Inka in the fifteenth century, the site grew through
unplanned urban sprawl. The built environment and spatial layout of Iglesiachayoq consists of
residential structures in their own private areas, often with an associated patio or low-walled
open space. Most structures stood apart from other structures, with the exceptions of a few small
complexes—in these complexes, multiple houses faced one another and shared a communal area
in which group activities could have taken place. These complexes were often located so as to
develop their own group privacy, and were surrounded by large natural rock faces or hills, or
walled enclosures.
The residents of Iglesiachayoq manipulated existing landscape features to suit their own
needs. The site is decorated with large boulders of andesite and other stones, which were often
incorporated into domestic structures or restraining walls. Smaller fieldstones were added to
buttress the large boulders, such that most walls at Iglesiachayoq consist of a combination of
both types of building material. The built environment of the site thus assumes the appearance of
being heavily manipulated by human actors, who transformed the built environment to suit their
own needs.
Modeling Visibility at Iglesiachayoq—Viewsheds as Indices of Privacy
Visibility of specific features in built environments of landscape are thought to have
indicated overall “importance” or meaning to various cultures. The construction of monuments,
placement of cemeteries, or “capturing” of natural features are often intended to create a
455
significant impact on individual senses. Visibility within archaeological landscapes are of
interest to researchers because they can lend insight into location choices for particular
settlements, the potential for intervisibility between sites and features, and the ability to model
past cognitive decision-making using visual acuity as a clue. In their 2002 spatial technology
volume, Wheatley and Gillings argue that visual aspects of places are “the most significant
impact a location has upon any individual’s many senses…in some instances, visibility can be
regarded as a key factor in attempting to answer the question as to why a particular site is in a
particular place, rather than all the other places it might have been located” (180). That is, the
sense of sight is considered to be the most dominant of all human senses, largely because of its
permanence—sound, smell, and taste are all ephemeral sensations, while touch requires close
proximity to be relevant. Thus, visibility and intervisibility analyses are critical to understanding
broader archaeological landscapes (Llobera 2003, 2007).
As discussed briefly above, though visibility and intervisibility have been common
measures for “importance” or “interaction” between distinct places within the landscape, notions
of invisibility have been far less theorized. The importance of seclusion in past communities is
difficult to assess and measure, largely because the intentionality motivating hiding of places in
the landscape is impossible to quantify (Gillings 2015). In this respect, Taki Onqoy is a unique
case study—takiongos often purposefully hid their performances from Spanish authorities and
thus, modeling spaces which were hidden to Catholic areas of Iglesiachayoq can shed light on
where these performances likely occurred. In this analysis, I utilize the spatial modeling tool
known as Viewshed, which analyzes raster digital elevation models to theoretically calculate
which areas could be seen from any one point. It is important to note several caveats of viewshed
analyses--they are positivistic in nature and fail to account for the degrees of variation in human
456
eyesight, the angles of sight, and the potentials for vegetation and other interference. Other
scholars in the Andes have used visibility analyses for other projects. For example, GIS analysis
demonstrates that apus are visible from important architectural complexes in the Moquegua
Valley (Williams and Nash 2006). Additionally, Arkush (2011) found that pukaras were located
in visible proximity to one another and Bongers and Arkush found that both visibility and
elevation were mitigating factors in the placement of chullpas in the Titicaca Basin (2012).
The common, guiding factor in all of these studies is that they take a regional approach to
studies of visibility. In contrast, for this chapter I will utilize a site-specific approach in order to
model invisibility between sectors and plazas rather than sites or landscape features at the
regional scale. If, as Wheatley and Gillings argue, visibility is the dominant sensory input for an
individual, then it is also important to consider visibility at the site level. I propose that there are
two contrasting aspects of visibility at Iglesiachayoq, which broadly align with the two
competing religious movements, Catholicism and Taki Onqoy. As discussed above, Spanish
authorities often constructed their churches atop or near places of Andean importance. In this
way, religious officials could draw analogies based on sacred space—Andean peoples would not
have to completely shift their worldviews, but only alter them to account for the new religion.
The church at Iglesiachayoq is an example of this phenomenon, dominating the primary site
entrance by its presence and visibility erupting out of the flat plaza. Conversely, it is likely that
takiongos sought a completely different type of space for their activities—a space which had a
high degree of invisibility. By selecting locations for performance which both had visibility of
approaching individuals, but also were protected from being seen by these individuals, takiongos
could practice their huaca possessions in private, covert ways and thus avoid punishment.
457
Viewshed Modeling at Iglesiachayoq
The order to test my visibility/intervisibility hypotheses, I first had to manipulate my
DEM in order to include the visual impedances of the structures at Iglesiachayoq. During my
architectural survey (Chapter 5), I recorded the heights of the remaining walls of the structures
and walls at the site. I then converted this polygon shapefile of structures into a raster whose
values ranged from 0 (the NullData value) to 4.1m (the height of the walls of the tallest
building). Next, I resampled a 30m resolution DEM from NASA JPL. Ideally, in a site-level
viewshed analysis, I would first generate my own DEM utilizing elevation and location indices
from Total Station points. In creating this fine-resolution DEM, the models for determining
visibility between different points at Iglesiachayoq would then be able to take building heights
into account and produce a more accurate viewshed map. Unfortunately, during my fieldwork in
2015, we were utilizing a Total Station which lacked the software and cables necessary to
download the data points. All were hand written in a small journal, but will not be used here in
order to account for human error. Instead, I have utilized the resampled 30m resolution DEM
(resampled to 1.7 x 1.7 cell size using bilinear interpolation) as a base raster at Iglesiachayoq. I
then applied the Raster Calculator to add the structures raster and the resampled DEM, thus
resulting in a cumulative raster which contained both topography and building heights. For all
viewshed models, I assumed a viewer of height 1.5m.
Model 1: Sector 1 and Church Complex
First, I ran a multiple viewshed analysis from all accessways from all structures in Sector
1. As can be seen in Figure 8.5, the resulting viewshed demonstrated that the Sector 1 structures
predominantly had visibility of the northern portion of the quebrada, including where the
458
probable site entrance was located, with visibility to the Inka road. The orientation of the central
plaza promoted visibility to this site entrance. Notably, the viewshed demonstrated that Sector
3—and explicitly the prehispanic plaza—was not visible from Sector 1. This viewshed thus
indicates that activities or performances occurring here would not have been visible to the central
cluster of residences in Sector 1. The Sector 1 cumulative viewshed also suggests that much of
the southeastern portion of Sector 2 would have been hidden from the eyes of those residing in
Sector 1.
459
Figure 8.5. Multiple viewshed from all structures in Sector 1. Red circle: local plaza and location
of proposed Taki Onqoy performances.
Model 2: Local Plaza
Next, I ran a multiple viewshed analysis all of the structures in Sector 3. As can be seen
in Figure 8.6, the viewshed output raster only had visibility to the eastern and northwestern areas
460
of the valley. All religious performances which were related to the Catholic church in Sector 1
would not have been visible to those residing in the more local Sector of the site. Similar to the
cumulative viewshed for Sector 1, the southeastern portion of Sector 2 was also not visible to
those who lived in Sector 3.
461
Figure 8.6. Multiple Viewshed results from Sector 3.
Model 3: Wallpa Wiri
462
Finally, I ran a third multiple viewshed analyses, this time from all of the structures in the
Wallpa Wiri Sector (Sector 2). As can be seen in Figure 8.7, the cumulative viewshed modeled
visibility of the majority of Sector 2, and no visibility of the Catholic church complex in Sector 1
and all of Sector 3.
Figure 8.7. Sector 2 multiple viewshed.
463
Discussion of Viewsheds and Visibility
The three multiple viewsheds demonstrate that the three distinct sectors of Iglesiachayoq
were relatively isolated from one another, at least in terms of visual interconnectivity. The
inability to see other sectors of Iglesiachayoq from any one particular vantage point contrasts
with the relative nearness in spatial proximity of all sectors—for example, the Spanish plaza and
the local plaza are 110m from one another as the crow flies, yet neither would have been able to
see the activities occurring in the either space (Figure 8.8). The local plaza’s relative invisibility
to the Spanish sector of Iglesiachayoq mirrors some of the contradictions in the movement itself:
while Taki Onqoy activities were explicitly anti-Spanish, yet shaped by Spanish cultural
frameworks, similarly the spatial locations of Taki Onqoy performances were positioned such
that they were invisible to the Spanish ritual center of Iglesiachayoq, yet spatially near this center
as well.
Viewshed analyses are useful for modeling places which could have been visible or
invisible—the results of the multiple viewsheds presented here are intended to serve as
possibilities for understanding why performances could have occurred in different areas. In
emphasizing the notion of invisibility, I suggest that Taki Onqoy rituals were performed in
places which were deliberately hidden from places where Spanish authorities would have
resided. This claim is bolstered through the accounts of Spanish priests (Chapter 2) and the
results of excavations in the local plaza in Sector 3.
464
Figure 8.8. Cumulative viewshed from all structure accessways to all portions of the site. Those
regions which were visible from the most other observer points are in red, and those which were
visible from the least observer points are in red.
465
Modeling Movement through Iglesiachayoq—Walking Times as Indices of Proximities to Sacred
Landscapes
The second spatial framework I use to assess Taki Onqoy and Catholic spaces considers
walking times as indices of proximity to sacred landscapes. Within landscapes, the structuration
of mobility allows inferences into power relationships (Parno and Beaudry 2014). For example,
Wernke’s Spatial Network Analysis at the site of Malata, demonstrated that Spanish authorities
restructured mobility away from the Inka center of the site and toward the Spanish center, thus
shifting the site’s focus and the resulting views of the people walking these pathways (2012).
Similar to the site of Malata, the site of Iglesiachayoq is a strong location for running SNA. The
standing architecture at Iglesiachayoq has preserved structure walls of original height for 30% of
all structures (n=27/91, or 29.7%), and all other structures consist of preserved foundations and
partial walls. Similarly, the walls (retaining walls, designations for property lines, terraces, etc.)
which structured and guided movement throughout Iglesiachayoq are also still standing, allowing
for identification of accessways throughout both types of architecture.
Mirroring the viewshed analyses of the previous section, I next utilized SNA to trace
movement through Iglesiachayoq. Archaeologically, movement can be simulated by creating
pathways which conform to the built environment of a landscape or site, but movement such as
Taki Onqoy dancing is more difficult to infer because of its place-bound locations. However, it is
more feasible to trace movement through a site or a landscape, utilizing the built environment
and natural landscape as barriers or thoroughfares. At Iglesiachayoq, this modeling takes on a
great importance, linking the previous two data chapters on archaeological results in domestic
households which largely demonstrated continuity over time, and the array of burial practices in
the central church. Although its main practice is characterized by dancing, we can begin to
466
hypothesize areas where Taki Onqoy may have been practiced by modeling movement between
areas of Catholic sacred importance and local, private areas of Iglesiachayoq.
At the site of Malata, creating a network was much simpler because pathways and routes
to different areas of the site were clear, took place over a relatively flat area with little change in
slope, and well-preserved. Conversely, Iglesiachayoq is draped between two steep hills and a
central flat area, is covered with natural immovable boulders which were worked into the built
environment, and lacks the convenience of a grid system from which to work. Since SNA was
developed for optimization in travel time according to vehicular travel, it does not take slope into
account. However, assessing the walking times between sectors at Iglesiachayoq is still
worthwhile, as it can provide a model other than visibility or spatial proximity for assessing
interconnectivity and the construction of movement throughout the site. In testing the final
hypothesis, I propose that the accessways and pathways at Iglesiachayoq structured movement in
meaningful ways to both the Spanish center and the local center. Specifically, I suggest that
those living in Sector 3, the local sector of Iglesiachayoq, constructed their plaza in a restricted
manner such that Spanish religious authorities would not have been able to unwittingly surprise
Taki Onqoy performances. Similarly, I suggest that Spanish religious authorities structured
movement away from the kuraka house and Inka center of the site, moved along the long axis of
the church, and into the central Spanish plaza through a restricted access point. In this way, those
individuals traveling from the local sector would have been forced to pass by the immense
presence of the church on their way to Catholic masses. This hypothesis also has implications for
burial removal from the church—the walled plaza in Sector 1 with its one public accessway
would still have forced takiongos moving to covertly remove their dead along the length of the
church, thereby increasing the chances of being caught and punished.
467
Spatial Network Analysis Modeling at Iglesiachayoq
In my architectural survey of Iglesiachayoq beginning in 2013, I traced all walls and
accessways so that I would be able to create pathways of movement between different structures
and sectors. By mapping every accessway to every large open space, structure, and wall, I was
able to develop a complete map of the site which has the possibility of accounting for movement
between all structures and spaces, giving 100% coverage. I must reiterate a central caveat here,
that while traffic can be taken into account using SNA, slope cannot. Thus, traveling between
two places which are at different elevations will generate the same time of travel, rather than
accounting for a slightly elevated pace due to walking downhill or a slightly slower pace from
traversing uphill paths.
The spatial modeling presented here considers past pathways and accessways as potential
areas of movement—spaces where individuals could move through an environment. In contrast,
walls, other structures, and environmental barriers such as steep rock faces or impassable
boulders were considered to be impedances which helped shape movement by leading people to
move around these obstacles. Doorways and walls were assumed to be guiding factors in
shaping movement—that is, people likely moved through doorways instead of climbing over
walls. Employing these pathways and impedances at Iglesiachayoq, I constructed a network
dataset and ran the closest facility model in order to trace pathways of movement at the site. The
construction of a network dataset involves demarcating endpoints (a point feature which denotes
entrances or plaza centers) and paths (a line feature which connects endpoints through linear
routes across the site). SNA is designed to optimize movement through a build environment, and
so it works best when archaeological sites are well-preserved and can be considered as a whole.
The closest facility function ranks endpoints in terms of walking time to their final destinations
468
(facilities) and allows a comparison of pathways between different sets of structures. Walking
time was based on Wernke’s 2012 model of 68.9 meters per minute (Wernke 2012).
For the purposes of this chapter, I employed spatial analysis to consider interactions
between two basic independent variables: population proxy and plaza. More specifically, using
the rectilinear structures as a representation of those who were likely involved with Catholic
services and the households in the local (Sector 3) area as those who were participating in Taki
Onqoy, I modeled travel time and routes to both the central plaza and the prehispanic plaza
(Table 8.1).
Model 1
Model 2
Model 3
Model 4
Start Point
Rectilinear Structures
Sector 3 (local) Structures
Rectilinear Structures
Sector 3 (local) Structures
End Point
Church Plaza
Church Plaza
Prehispanic Plaza
Prehispanic Plaza
Table 8.1. Four walking models between starting and ending points.
For the central plaza, there were three basic accessways, one at the eastern side of the
plaza, one at the western side of the plaza, and one toward the north of the plaza (Figure 8.9).
The western accessway was associated with preparation areas for church activities, and
prevented clear movement from other sectors of the site. For the prehispanic plaza, there were
three accessways, one at the east, west, and south ends of the plaza (Figure 8.10). The northern
portion of the local plaza was hidden from view to those in the other two sectors by an immense
rock face.
469
Figure 8.9. Central plaza with three accessways designated by stars.
470
Figure 8.10. Prehispanic plaza with entrances designated by stars.
Model 1 Results: Rectilinear Structures to Church Plaza
For the rectilinear structures, I ran the model from the entry of each of the nine structures
to the doorway of the church. As discussed above, eight of these structures are located within
100m of the church and have entryways which face the church, while the ninth rectilinear
structure is located in the northeast sector, apparently removed from the center of the site. The
overall average walking time from the rectilinear structures to the church was 73.43 seconds, and
471
seven out of the nine structures utilized the eastern accessway to the central plaza (Figure 8.11).
If the rectilinear structure from the northeast sector is removed from the analysis, walking time to
the church drops to 47.9 seconds (Figure 8.12). Thus, the majority of rectilinear structures are
located within a minute’s walking distance to the church. The proximity to the church, the trend
for entryways to face the central plaza and church, and the fact that 90% of the rectilinear
structures are located within 100m of the church plausibly indicates that the inhabitants of the
structures were associated with church activities. The individuals living in these structures likely
publicly participated in Catholic rites, and were highly visible to the rest of the community at
Iglesiachayoq. Further research is needed to determine the function of the rectilinear structure in
the northeastern sector.
472
ObjectID
1
2
3
4
5
6
7
8
9
Name
3 - Church
4 - Church
5 - Church
27 - Church
29 - Church
30 - Church
32 - Church
36 - Church
107 - Church
Total_Time
16.632012
43.455039
39.322348
53.281026
53.967212
52.310978
53.720857
70.533795
277.659094
Figure 8.11. Left: Map of routes used by rectilinear structures to the central plaza, n=9/9. Right:
walking times from structures to plaza.
473
Figure 8.12. Boxplot distribution of walking times from rectilinear structures to the central
church, one outlier removed, structure 107 with a time of 278 seconds; n=8.
Model 2 Results: Sector 3 Structures to Central Plaza
Conversely, I ran the closest facility model from households in Sector 3 to the church
plaza. Those moving between these areas were routed to the eastern accessway of the central
plaza, and their pathway took about four minutes and twenty seconds to traverse (Figure 8.13).
Notably, the majority of these local structures are located southwest of the church itself. Thus,
the individuals traveling from the local sector would have had to walk the extended length of the
church in order to reach the eastern access, rather than being able to use the closer western
access. The forced path emphasizes the grandeur of the central church and allowed visibility into
the activities occurring in the central plaza. The difference in walking times between the squared
structures and the local structures is seen in the side-by-side boxplot below (Figure 8.14). A
474
paired-samples t-test was conducted to compare walking time to the central plaza from
rectilinear structures and Sector 3 structures. There was a significant difference in the times for
rectilinear structures (mean = 73.43s) and Sector 3 structures (mean = 257.23s); t=-7.65021, pvalue is <.00001.
ObjectID
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
199
202
205
209
212
215
218
219
227
228
230
235
238
239
244
246
247
248
250
251
254
255
257
259
260
261
262
263
264
266
269
272
274
292
Name
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
- Church
Total_Time
124.609676
120.931992
85.69066
175.453447
226.110855
262.418105
270.222901
308.291661
311.121604
314.281789
323.750898
203.945868
230.201029
234.597678
250.098138
243.755052
239.570068
247.291078
345.618229
328.920677
305.732929
282.280165
271.766707
254.197591
253.871462
293.389776
295.823806
298.332956
323.641721
225.314416
261.293738
282.149222
298.433854
252.780541
Figure 8.13. Left: Routes from the southwest sector to the church, all utilizing the eastern
accessway and being directed to walk the length of the church. Right: walking times from Sector
3 structures to the church, n=34/34 structures.
475
Figure 8.14. Boxplot distributions comparing walking times to the central church between
rectilinear structures and Sector 3 structures, rectilinear structures n=9; sector 3 structures n=34.
There was a significant difference in the times for rectilinear structures (mean = 73.43s) and
Sector 3 structures (mean = 257.23s); t=-7.65021, p-value is <.00001.
Model 3 Results: Rectilinear Structures to Prehispanic Plaza
Next, I ran the closest facility model from the rectilinear structures to the prehispanic
plaza in Sector 3. As seen on the map, the pathways from rectilinear structures to the prehispanic
plaza were all routed to either an eastern or western accessway in the prehispanic plaza. Access
would have been restricted to this area, funneled to either side of the plaza due to the large rock
face blocking direct access to the northern portion of the plaza. Those approaching the upper
plaza from the lower sector could have been seen from afar. On average, it took over four
minutes for those traveling from the rectilinear structures to arrive at the local plaza (Figure 8.15
476
and 8.16). However, because SNA does not take slope into account, the travel time of four
minutes is likely a conservative estimate for those traveling from the rectilinear structures toward
the upper plaza because they would have been forced to traverse a steep uphill slope.
ObjectID
1
2
3
4
5
6
7
8
9
10
Name
1 - Circular Plaza
3 - Circular Plaza
4 - Circular Plaza
5 - Circular Plaza
27 - Circular Plaza
29 - Circular Plaza
30 - Circular Plaza
32 - Circular Plaza
36 - Circular Plaza
107 - Circular Plaza
Total_Time
252.780541
254.310532
253.445468
249.312777
235.247999
235.934185
234.277951
235.68783
252.500768
393.238642
Figure 8.15. Left: Map of routes from rectilinear structures to prehispanic plaza. Right: travel
times from rectilinear structures to local plaza, n =10/10.
477
Figure 8.16. Boxplot distribution of walking times from rectilinear structures to prehispanic
plaza, n=9, one outlier removed, structure 107 with a walking time of 393 seconds.
Model 4 Results: Sector 3 Structures to Prehispanic Plaza
Finally, I ran the closest facility model from the Sector 3 structures to the local plaza.
Unsurprisingly, the walking time between these areas was two minutes and twenty seconds, and
was shorter compared to both 1) the travel time from Sector 3 structures to the central plaza, and
2) the travel time from squared structures to the local plaza (Table 8.2). Notably, the model
generated routes which utilized all three access points, including the southern access which was
more conducive for those living in this local sector (Figure 8.17). In visually comparing the
differences in walking time to the prehispanic plaza from rectilinear structures and Sector 3
structures, it appears as though the two distributions were different (Figure 8.18). A Mann
Whitney U Test was conducted to compare walking times to the prehispanic plaza from
478
rectilinear structures and Sector 3 structures. There was a significant difference in the scores,
with a z-score of -4.38288, and thus the difference in walking times was significant at p<.05.
ObjectID
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
Name
199 - Circular Pl
202 - Circular Pl
205 - Circular Pl
209 - Circular Pl
212 - Circular Pl
215 - Circular Pl
218 - Circular Pl
219 - Circular Pl
227 - Circular Pl
228 - Circular Pl
230 - Circular Pl
235 - Circular Pl
238 - Circular Pl
239 - Circular Pl
244 - Circular Pl
246 - Circular Pl
247 - Circular Pl
248 - Circular Pl
250 - Circular Pl
251 - Circular Pl
254 - Circular Pl
255 - Circular Pl
257 - Circular Pl
259 - Circular Pl
260 - Circular Pl
261 - Circular Pl
262 - Circular Pl
263 - Circular Pl
264 - Circular Pl
266 - Circular Pl
269 - Circular Pl
272 - Circular Pl
274 - Circular Pl
292 - Circular Pl
Total_Time
226.607366
222.929681
252.305507
144.631879
80.487239
180.441102
188.245897
226.314658
229.144601
232.304785
241.773894
94.729669
120.98483
125.381478
86.60337
80.260284
76.0753
83.79631
239.165387
222.467835
199.280087
175.827323
165.313865
147.744749
147.41862
186.936934
189.370964
191.880114
217.188879
61.819648
187.591214
208.446698
224.73133
0
Figure 8.17. Left: Map of routes from Sector 3 structures to prehispanic plaza. Right: travel times
from Sector 3 structures to prehispanic plaza, n =34/34.
Model 1
Model 2
Model 3
Model 4
Start Point
Rectangular Structures
Local Sector Structures
Rectangular Structures
Local Sector Structures
End Point
Church Plaza
Church Plaza
Local Plaza
Local Plaza
Table 8.2 Summary of walking times for each model.
479
Travel Time
47.9 sec
4 min, 20 sec
4 min, 5 sec
2 min, 20 sec
Figure 8.18. Boxplot distributions comparing walking times to the prehispanic plaza from
rectilinear structures and Sector 3 structures, rectilinear structures n=10, sector 3 structures n=34.
A Mann Whitney U Test was conducted to compare walking times to the prehispanic plaza from
rectilinear structures and Sector 3 structures. There was a significant difference in the scores,
with a z-score of -4.38288, and thus the difference in walking times was significant at p<.05.
In sum, there is a pattern of differential access to either plaza, in addition to variances in
the travel time walking between the two sectors and areas of ceremonial importance. In general,
rectilinear structures were located a very short distance from the church and could access the
central plaza from both the western and eastern accessways. Conversely, those residing in the
Sector 3 had to travel four times as long to reach the central church, and were funneled through
the eastern entryway. The fact that some rectilinear structures (as opposed to all other structures
from all other sectors) had access to the western entrance in the central plaza suggests that this
entrance was utilized by those assisting in church activities. In terms of access to the prehispanic
480
plaza, those traveling from squared structures were forced to travel uphill through narrow
accessways to enter the upper plaza, and activities occurring in this plaza would have been
obstructed from view. Those inhabitants of Sector 3 traveling to the prehispanic plaza could
access this region from the southern direction, unlike those who inhabited the squared structures
in the central sector.
Discussion: Taki Onqoy and movement through Iglesiachayoq
Because Taki Onqoy was a movement in which the main practice was performative and
involved movement, I have utilized SNA to hypothesize an area in which Taki Onqoy may have
been enacted. I suggest that Taki Onqoy activities likely were performed away from Catholic
ritual spaces, out of sight of church authorities, and associated with the “local,” or prehispanic
footprints of the site. The prehispanic plaza meets all three criteria, and the closest facility
models generated support the removed nature of this plaza. That is, not only were those living in
Sector 3 closer to the prehispanic plaza, but those living in rectilinear structures were more
distant (by time) and would have been visible in their journeys to the prehispanic plaza.
Excavation results from this plaza and the associated ovoid structure affirm that neither space
was used for domestic activities—we recovered no evidence of food preparation, and the ceramic
assemblages were sparse in comparison to those recovered across the other units. The results of
these two excavation areas thus conform to assertions made by Albornoz and Molina that Taki
Onqoy spaces were ritually cleaned. Furthermore, the prehispanic plaza is located in a rather
private, protected area of the site—it was intentionally constructed with large boulders and it
incorporates various rock outcrops. Since Taki Onqoy dance performances were ephemeral, it is
difficult to definitively suggest that takiongos practiced their rituals in this space. However, SNA
allows us to generate models of movement so that we can begin to see how local people may
481
have moved throughout their built environment, thus providing a local voice not present in
primary documents.
Discussion and Conclusions
From both the SNA and the viewshed analyses, I suggest that those residing in the three
discrete sectors of Iglesiachayoq were able to exist and practice in relative privacy. Both the
SNA and the viewshed mimic my own phenomenological experience at the site, as well as the
excavation results documented in Chapter 6. Although the artifact assemblages indicate that the
three sectors were occupied more or less contemporaneously, SNA and viewshed analyses affirm
the spatial separations between these three sectors.
Although the Catholic Church is imposing within the landscape of Iglesiachayoq, it is
situated in the lowest area of the three sectors. Functionally, this could have been the only
location with enough open space for the required church and central plaza. However, the church
may also have been constructed here in order to challenge the large, preexisting Inka casa de
kuraka. Moreover, the central Catholic Church is near the main valley road which was part of the
Qhapaq Ñan, which both provides access to Cotahuasi and Yauca, as well as the most direct
route to the mountain apu Carahuarazu. Furthermore, Iglesiachayoq’s location borders the road
from Pampachiri and north toward Soras. The location of the central church at Iglesiachayoq,
then, was situated at the nexus of several major roads, and remains an imposing structure in the
center of the site today.
While visibility may have been a concern for Spanish authorities in directing the
construction of the Catholic Church, I suggest that invisibility was critical to those practicing
Taki Onqoy in outdoor enclosures. While these performances could last for several days, their
482
frequency and intensity was likely dependent on multiple factors which varied according to the
different settlements. For example, Albornoz suggests that kurakas were differentially invested
in the Taki Onqoy movement. While some turned a blind eye, others participated or led this
movement, and some instead allied with Spaniards to eradicate it. Another factor which
influenced Taki Onqoy performances was the ability to practice these rituals in areas which were
more “local” (i.e. closer to landscape huacas or apus, incorporating rock outcrops, etc.) and
further from the vigilance of Catholic authorities. At Iglesiachayoq, both Sectors 2 and 3 would
have been ideal for covert practices since these sectors are both located on high ridges above the
central church.
483
CHAPTER 9
SUMMARY AND CONCLUSIONS
Taki Onqoy was a revitalization movement whose leaders attempted to unify Andean
populations across the dimensions of ethnicity, status, or gender with the avowed goal of
rejecting Spanish culture and Catholicism and promote a return to prehispanic huaca veneration.
By overlooking these internal differences in Andean groups, Taki Onqoy leaders were either
explicitly or inadvertently mirroring Spanish essential divisions of “Spaniard” and “Andean” in
order to create a unified force and resist Spanish colonialism. The movement was thus antiCatholic and anti-Spanish, yet shaped by Catholic tenets of religious exclusion and Spanish
social and racial divisions. In their attempts to gloss over critical social differences between and
within Andean groups, Taki Onqoy leaders could not account for the array of indigenous
positionalities toward both Catholicism and the movement itself, and thus Taki Onqoy was
fraught with contradictions, partial participants, and a population of Andeans who practiced both
religious traditions.
As Taki Onqoy was indelibly structured through the entanglement of both Spanish
colonial frameworks and Andean notions of a prehispanic utopia, so too were the movement’s
central practices a product of both cultural systems. Taki Onqoy was enacted through the
interaction of people, places, and things in assemblages which were themselves already
entangled through the colonial encounter. European objects and Andean objects interacted with
one another, and their identities as “European” or “Andean” shifted as they were reused and
repurposed in different cultural spaces. Places of prehispanic sacrality became imbued with
Catholic importance as the Spanish constructed churches atop or near sacred landscapes. The
Catholic god—who had demonstrated his power throughout the Spanish conquest—was
484
considered as an addition to the Andean pantheon, even as Spanish authorities emphasized
religious exclusion.
Taki Onqoy has heretofore been studied through the primary accounts of Spanish priests
who were attempting to extirpate the movement, yet I argue that these documents demonstrated
uniform conceptions of “Spanish” and “Andean.” Taki Onqoy was enacted and structured in situ
through entangled practices with entangled objects in places which held diverse meanings for
different people, and thus an archaeological study of how these new colonial constructions
interacted is a critical line of inquiry. While Taki Onqoy was powerful in the short term in its
ability to create rapid cultural change and to frighten Spanish religious authorities, its promises
to return the Andes to a prehispanic utopia was ultimately never fulfilled. However, the practices
indicative of Taki Onqoy served as a blueprint for future Andean revitalization and millenarian
movements (Dean 1999; Gose 2008), and some of its key practices are thought to still be
practiced in the highlands of Huamanga, Huancavelica, and Apurimac today (Arguedas 1968;
Bush 2012). Taki Onqoy’s legacies also endure in its central role in historical debate as a
potential marker of Andean nationalism (Millones et al. 1990). In the remainder of this
concluding chapter, I summarize the movement’s key practices, my hypotheses, and the results
of my archaeological investigation into Taki Onqoy.
Taki Onqoy Sources and Practices
To date, there are four known primary sources which discuss Taki Onqoy by name—each
of these sources was recorded at variable times and in different places, and they each provide
unique accounts of the movement’s practices (Chapter 2). Of these accounts, the work of the
Spanish priests Cristóbal de Molina and Cristóbal de Albornoz are considered the only lengthy
485
documents about Taki Onqoy: while Molina described the movement’s origins, practices, and
beliefs, Albornoz chronicled accounts of his work extirpating idolatries in the Central Highlands
of Peru. Both Molina and Albornoz suggested that the movement originated in the last Inka
holdout of Vilcabamba and was spread by traveling Inka priests who wandered throughout the
highlands (Chapter 2). Both authors also affirmed that takiongos believed that the huacas—who
had been neglected by Andeans or destroyed by Spaniards—were resurrecting into two armies,
one following Pachacamac and the other supporting Titicaca. These two armies were forming in
order to battle the Catholic God and return Andean peoples to a perceived revitalized utopia in
the mold of prehispanic Andean belief systems. The reanimated huacas were said to fly through
the air in order to inhabit various individuals, who claimed that they had been purified by huaca
possession. Possessed takiongos would drink heavily and dance for days, often in outdoor
enclosures specifically designated for Taki Onqoy practices.
Molina and Albornoz identified several Taki Onqoy practices related to the rejection of
Spanish traditions, all which would indicate an absence of archaeological materials. This
rejection included avoidance of Spanish foods, clothing, churches, and prayer—archaeologically,
these practices would manifest in the absence of non-local (Spanish) artifacts in households. The
two priests also identified Taki Onqoy practices which provide positive archaeological material
correlates. Takiongos repurposed traditional Andean practices such as use of red body paints,
sacrifice of rams and cuy, and ancestor worship. Some individuals even practiced outright
rejection of Catholicism, manifested through the destruction of Catholic symbols and materials.
In response, Spanish authorities who punished takiongos often whipped the preachers, shaved
their heads, forced them to help construct churches, mandated ecclesiastical teachings several
days a week, and burned or destroyed the huacas associated with each Taki Onqoy town.
486
With the introduction of monotheistic, exclusivist religion at the cost of the destruction of
local huacas, Spanish Catholics thrust local religious practices into turmoil. Andeans interpreted
the destruction of their huacas as a bodily sickness. This interpretation was not metaphorical—
Andean groups perceived the destruction of their huacas as the cause of the ailments in the
second half of the sixteenth century. It was in this antagonistic milieu that Taki Onqoy
developed, and it is why its central tenet was to worship those huacas who had been destroyed.
In the Andean religious framework, individuals could worship both their traditional idols while
simultaneously acknowledging the power of the Catholic God. Yet, these takiongos also must
have been influenced by the very teachings of the Catholic Church in that the most novel practice
associated with Taki Onqoy was the rejection of the Spanish God. Taki Onqoy preachers were
insistent that those Andean peoples who had actively embraced “Hispanic” orientations were
thus enemies of the native communities. In order to regain the alliances and trust of their fellow
Andeans, those who practiced Catholicism had to renounce these beliefs and instead join the
Taki Onqoy movement.
However, disbelieving in the Catholic God was difficult for Andean individuals—God
had demonstrated his power in helping the Spaniards triumph over the Inka, and thus was a deity
to be feared and admired. Taki Onqoy preachers creatively did not denounce the power of this
God; instead, they affirmed the power, but preached that it did not apply to Andeans. While God
had given Spaniards their successes in Spain and the New World, it was the huacas which
supported the Andeans—Andean peoples had to both recommit to worshipping the huacas,
providing them with gifts and food, and simultaneously exclude the Catholic God.
487
Utilizing the accounts from Molina and Albornoz in conjunction with other
archaeological studies of the Early Colonial Period, I generated the following hypotheses to
guide my investigation at Iglesiachayoq (Chapters 2 and 5).
1. If Taki Onqoy practices did not entirely proscribe the use of Spanish material culture,
they would likely be characterized by the repurposing of Catholic or Spanish objects in
new ways or hybrid forms of material culture.
2. If Spanish responses to Taki Onqoy at Iglesiachayoq were overtly violent, they would be
marked by destruction of huacas in public arenas or physical punishment of takiongos in
public spaces.
3. If the inhabitants at Iglesiachayoq were challenging Catholic traditions in burial practices,
then interments would reflect Andean traditions rather than standardized Catholic norms
in terms of position and burial goods.
4. If Taki Onqoy performances were an explicit revitalization of prehispanic cultural
traditions, then their associated materials would reflect practices common in the region
prior to Spanish invasion.
5. If Taki Onqoy was a clandestine movement, its material and spatial signatures will be
uncovered in private areas away from central religious spaces. If there are public spaces
of Taki Onqoy performances, these will be located away from places of Spanish
administration and Catholic practices.
In the rest of this conclusion, I will briefly summarize the results of my archaeological
investigations in the domestic, mortuary, and spatial spheres, and then address each hypothesis
individually.
488
Results: Household Archaeology
In our excavations of ten households, Late Horizon and Early Colonial Period
occupations were indistinguishable, and there was broad continuity within sectors and structure
types. The only evidence of non-local artifacts in these realms was indicated through two carethead nails, an iron doorlatch, and the presence of Old World fauna such as cow or horse/donkey.
Of the three iron artifacts, one was found in each sector, in three separate houses.
Excavation results of fauna produced a minimum of thirty individuals (Chapter 6). In
general, patterns in faunal data follow trends in differences between sectors. Cow and equid
made up 5.05% of the assemblage, and were almost exclusively found in Sector 1, with the
exception of one modern cow bone in Sector 2. This affirms that these large Old World animals
were most commonly associated with Sector 1, the Spanish ritual center. All of the cow and
equid were adults at the time of death, suggesting they may have been utilized as pack animals or
for transportation. In contrast, camelid bone was the most common faunal remain, making up
15% of the assemblage found site-wide. Camelids were recovered in ages ranging from juvenile
to sub-adult to adult, indicating a more generalized and renewable pattern of use.
We found no ceramics that we could confidently categorize as colonial period (glazed or
wheel-thrown), and it appears as though site-wide ceramic traditions remained constant through
the Late Horizon and Early Colonial Period (Chapter 6). With such a sparse assemblage of
European foods, goods, and personal items, and the documentary record of Taki Onqoy at
Iglesiachayoq, there are two possible interpretations for the data: first, as in other sites across the
Andes, continuity between prehispanic and post-conquest occupations can be explained as the
result of temporality. There was not enough time for imported vessels, animals, and religious
practices to arrive and infiltrate Iglesiachayoq. In contrast, since there was evidence of European
489
goods, albeit in incredibly small amounts, it is also possible that the people living in certain areas
of the site were actively avoiding Spanish goods and traditions, as affirmed by Molina and
Albornoz.
Now I will turn to the material correlates of Taki Onqoy. In Sectors 2 and 3, those which
did not contain the central Catholic Church, we did find a relatively normal assemblage of
ceramic vessels for cooking, drinking, and storage. In Sector 3, there were two units excavated
which were nearly devoid of artifacts—the unit in the semi-circular plaza and the associated
ovoid structure (Chapter 6). Both of these correspond with descriptions of Taki Onqoy rituals
and the architectural enclosures where these rituals would have been performed. Unfortunately,
the assemblage I found, while consistent with Taki Onqoy practices, is also consistent with
general Late Horizon or Early Colonial Period settlements across the Andes. We recovered no
evidence of red pigments, sacrificed animals, or spondylus shell, nor were there any overt
examples of destroyed Catholic religious items or iconography related to Taki Onqoy (Chapter 2
and 6).
One other way in which Taki Onqoy performances can be inferred is through the
punishments given to takiongos as documented by Albornoz (Chapter 2). In the central plaza unit
in Sector 1, my team and I did not recover any diagnostic ceramic, nor were there any faunal
remains. During Taki Onqoy, huacas were not only able to inhabit various individuals, but they
also were considered portable in some cases. Thus, the destruction of portable stone huacas could
have been possible in this area of the site, and would be consistent with the documented Taki
Onqoy punishments and huaca destruction noted by Albornoz (1990:275). In this unit, we
recovered 152 fragments of a brown, waxy stone known as “welded tuff,” an igneous rock
generated from explosive volcanic eruptions. If this stone were present in such high quantities
490
across all patio units, then I would expect it to be a natural part of the site’s geology. However,
no other patio unit contained welded tuff. Moreover, the only places we also recovered this
material was in the church excavation units. It could be possible that these tuff fragments were
considered portable huacas—this would explain their presence exclusively in the plaza and
church units. The presence of this type of stone in the plaza could correlate with Albornoz’s
overt destruction of huacas here, and its density in the church units could represent local
takiongos bringing fragments of this stone into the new sacred center of Iglesiachayoq.
Besides differences between sectors, we also recovered differences between structure
forms. Quadrangular structures with squared corners (rectilinear structures) were clustered
within one hundred meters of the central Spanish church, and seem to have been associated with
the activities which occurred in Sector 1. Quadrangular structures with rounded corners were the
typical domestic structure form at the site, and demonstrated typical domestic assemblages with
similar faunal, ceramic, and lithic remains. Finally, in ovoid structures we recovered distinctive
assemblages which lacked faunal remains, diagnostic ceramics, and prepared hearths, and in
which most ceramic sherds recovered were under 5cm in size. I suggest that these ovoid
structures were thus neither used for storage, nor for domestic activities, and instead were
regularly cleaned and utilized as Taki Onqoy enclosures for dancing and chanting.
Results: Mortuary Archaeology
In domestic contexts, my team and I recovered evidence of general continuity. In
contrast, church excavations demonstrated extreme variability in mortuary treatments. Three
units were excavated within the church structure. The excavated human remains were very
fragile and could not be biologically sexed with confidence. Burial treatment of individuals
491
interred (or disinterred) from this space indicates that postmortem practices were highly
contested. In fact, although the central Catholic Church is the space in which Catholic authorities
would have had the most oversight, it is also the space in which we found the most evidence for
heterodox practices vis a vis Catholicism and Taki Onqoy. We recovered several different types
of burials at Iglesiachayoq including primary extended, primary flexed, secondary extended,
secondary flexed, disturbed burials, removed burials, and burials which we were unable to
categorize (indeterminate).
The array of burial practices in such a symbolic space suggests that Andeans were either
embracing new Catholic traditions or retaining aspects of their native identity in the most critical
context possible. Primary extended burials affirmed that some Andean individuals were publicly
embracing Catholic tenets—these individuals would have been interred in public ceremonies.
Primary flexed burials were often interred with tupus or other grave goods, and at various angles
in relation to the eastern altar. While the position and presence of grave goods suggests a
retention or revitalization of Andean practices, the primary interment in the church suggests that
the kin of these deceased individuals were willing to negotiate with Catholic mandates and that
Catholic authorities allowed leeway in interment.
We recovered one example of a secondary extended individual. This individual was
originally buried in a different location, and then his/her remains were relocated into the church
near the altar. Those who reinterred this individual attempted to situate the body in the extended
form without grave goods. Perhaps, then, some Andean peoples were actively attempting to inter
their relatives in ways that were overtly Catholic. The secondary flexed burials similarly were
relocated after death, but haphazardly such that the bones were more or less “bundled” and
occasionally contained parts of different individuals. Finally, in what I consider to be the most
492
overt example of Taki Onqoy tenets, four individuals were originally interred beneath the church
floor and were removed at a later date. Although the punishment for burial removal at this time
was severe, the removed burials suggest that Andean removing their dead and reinterring them
elsewhere.
Results: Spatial Analysis
I tested both walking times to different locations and intrasite visibility of these same
places through spatial modeling (Spatial Network Analysis and Viewshed Analysis, respectively)
in ArcGIS 10.6. The basic premise of my spatial analysis models was that there were two
distinctive areas of Iglesiachayoq which were associated with religious ritual practices: the
Spanish Catholic church in Sector 1, and the prehispanic circular structure and plaza in Sector 3.
In addition to the differences between these two spaces, there were also variations in function of
different structure forms, as supported by archaeological artifact assemblages. For the purposes
of the Spatial Network Analysis, I used rectilinear structures and the central church plaza as
proxies for Catholic religious practices, and I used Sector 3 structures and the prehispanic plaza
as proxies for Taki Onqoy practices.
I ran the SNA model four times, tracing walking times between rectilinear structures to
the church plaza, Sector 3 structures to the church plaza, squared structures to the prehispanic
plaza, and Sector 3 structures to the prehispanic plaza. Results indicated differential access to
both plazas. Those residing in rectilinear structures had the shortest walk to the church plaza, and
two of these structures were able to utilize the western access of this plaza, which was restricted
for Sector 3 travelers. In contrast, those residing in Sector 3 had a much longer walk to the
church plaza, and were routed through the eastern plaza access. This route meant that those
493
walking from Sector 3 would have been forced to travel the length of the church before entering
the church plaza. Conversely, travel times were much longer for those living in the rectilinear
structures to arrive at the local plaza than for those in Sector 3. Additionally, those authorities
who were likely residing in Sector 1 would have been visible in their movement toward the local
sector since they were traveling uphill and the local plaza was hidden from view.
While the SNA was useful for envisioning and modeling movement through
Iglesiachayoq, I found the viewshed analyses to be even more compelling in terms of
reconstructing lived experience at the site. I ran cumulative viewshed analyses from the three
Sectors as an observer of 1.5m height. Intriguingly, each sector lacked visibility to the other two
sectors. Visually, then, those existing and performing in each sector would have had relative
privacy and would not have been able to see or be seen by people in the other two sectors. In
considering Taki Onqoy, practitioners of the movement would have clearly been able to maintain
the clandestine nature of these rituals from Catholic authorities, even if they were dancing in
open spaces in different parts of the site.
Re-evaluation of Hypotheses
Now that I have presented the general overall findings of my investigations, I will
explicitly address each hypothesis in order to reassess my research questions.
Hypothesis 1: If Taki Onqoy practices did not entirely proscribe the use of Spanish material
culture, they would likely be characterized by the repurposing of Catholic objects in new ways or
hybrid forms of material culture.
Multiple practices can explain the material assemblage we recovered at Iglesiachayoq. In
all units except the kuraka house, Unit 7, and Unit 10, we found no clear evidence of “Spanish”
494
material culture. The maintenance of local traditions in consumption and artifact production
suggests that either:
1) Hypothesis 1 is supported. Taki Onqoy practices did entirely proscribe the use of
Spanish material culture.
2) Spanish material culture was not accessible at Iglesiachayoq to begin with, and
therefore there is no evidence for rejection of this material culture.
My excavations found no evidence of Catholic objects, with the exception of the 1892 cross
which postdates the site occupation. Following this lack of data, we also lacked evidence for
repurposing of Catholic objects or hybrid forms of material culture. The material culture
recovered all supports the continuation of local traditions of Soras and Inka peoples at
Iglesiachayoq.
Hypothesis 2: If Spanish responses to Taki Onqoy at Iglesiachayoq were overtly violent, they will
be marked by destruction of places of indigenous importance. These practices would manifest in
public arenas in efforts to quell the movement through public punishment.
I also am reluctant to provide a definitive “support” or “reject” opinion on Hypothesis 2.
While we did find anomalous lithic materials in the central plaza which could be evidence of
huaca destruction, more work must be completed in order to test this hypothesis. In the future, I
will expand excavations in the central plaza and investigate local landscape huacas. If there is
evidence of intentional destruction of landscape huacas, then Hypothesis 2 would be supported.
If continued excavation in the central plaza affirms clear evidence of burning or smashing of
stones and vessels, this would also support Hypothesis 2.
495
Hypothesis 3: If the inhabitants at Iglesiachayoq were challenging Catholic traditions in burial
practices, then interments would reflect Andean traditions rather than standardized Catholic
norms in terms of position and burial goods.
As discussed in Chapter 7, I found evidence of a mixture of burial treatments beneath the
church at Iglesiachayoq. While some of these interments reflected Andean traditions (flexed
positioning, grave goods), other individuals were interred in the preferred Catholic style. I also
suggest that the removed burials are clear evidence of subversion of Catholicism and Spanish
culture at Iglesiachayoq. It is highly unlikely that Andean peoples would have been unaware of
the meaning of their burial removals in the eyes of Spanish ecclesiastical authorities.
Hypothesis 4: If Taki Onqoy performances were an explicit revitalization of prehispanic cultural
traditions, then their associated materials would reflect practices common in the region prior to
Spanish invasion.
I did not find any clear evidence of Taki Onqoy performative materials—we did not
recover any evidence for animal sacrifice, no red pigments, and no overtly destroyed Catholic
objects. However, household excavations and investigation of the local plaza and circular
structure in Sector 3 suggest that performances may have taken place here, out of the direct sight
of those associated with church activities. The local plaza and associated ovoid structure both
had few artifacts, and it is plausible that they were intentionally cleaned in preparation for these
performances.
Hypothesis 5: If Taki Onqoy was a clandestine movement, its material and spatial signatures will
be uncovered in private areas away from central religious spaces. If there are public spaces of
496
Taki Onqoy these will be located away from places of Spanish administration and Catholic
practices.
Though Taki Onqoy has been described as a covert movement, some of its documented
practices seemingly contradict this notion. For example, excessive drinking, singing, and
chanting for days at a time would certainly be considered noticeable, or even openly
conspicuous. I suggest that the covert nature of Taki Onqoy practices should be considered on a
spectrum of risky behaviors, each which would incur consequences or punishments at different
scales. The majority of structures in Sector 3, the “local” portion of Iglesiachayoq, showed no
evidence of Early Colonial Period artifacts, including iron nails, Old World food items, or
Catholic religious items. At the domestic level, rejection of Spanish practices would have been
both covert and rather harmless ventures—Spanish authorities could not be present at every
settlement all of the time, and were more invested in Catholic rites rather than the day-to-day life
in the home.
Public displays of dancing and heavy drinking were more unsafe than were everyday
choices regarding food and ceramic traditions. As discussed in Chapter 2, these displays of
religious fervor were loud, long, and could become so intense so as to cause illness or death in
those possessed by huacas. If these performances were located in public spaces and associated
with more “local” parts of the site or circular structures, then these are definitively located away
from places of Spanish administration and Catholic practices. As the viewshed analysis
demonstrates, each sector was invisible to the other two sectors. Additionally, the ovoid
structures in Sector 1 are all situated well out of the site nucleus containing the church, chapel,
and central plaza. It is therefore feasible that public performances of Taki Onqoy were still held
in covert areas away from central authorities, even though they were still conspicuous events.
497
The riskiest practices associated with Taki Onqoy are those which occurred in the
mortuary realm in the central church. This was the space which theoretically had the most
Spanish oversight, and still held the most importance for both Catholicism and Taki Onqoy. The
church space and burial were also two aspects of Catholic practices which were heavily regulated
through the Councils of Lima. If all residents at Iglesiachayoq were publicly conforming to
Spanish Catholic mandates, then I would expect a burial population of extended individuals with
east/west orientations. Yet, it is in the church where we find the most convincing evidence of
heterodox practices which challenge Catholic mandates. Alternative interment styles of the dead
would have been visible to Catholic authorities who were in charge of burial ceremonies and
perhaps allows space for considering the leniency of these authorities. However, burial removal
was a problem which plagued Catholic authorities for centuries after conquest, and would have
been a definitively covert activity.
In considering Hypothesis Number 5, I suggest that the archaeological results present
mixed support for this statement. Taki Onqoy was a clandestine movement in that its spatial
signatures were uncovered away from central religious spaces to an extent. However, in the
critical realm of death, those sixteenth-century inhabitants of Iglesiachayoq were directly
interacting with and challenging the central Catholic religious space.
Was Taki Onqoy a Unique Movement, or a Reflection of Pan-Andean Sixteenth-Century
Practices?
Historians of the Early Colonial Period have taken a wide range of perspectives on Taki
Onqoy, such that conception of the movement has been cast at one extreme as a marker of
indigenous identity and rebellion, or at the other extreme, as the artificial creation of Spanish
priests. My interpretation of the movement is that it was a unique movement which was defined
498
by its contradictions. While Andean archaeology has shifted from accepting chroniclers’
accounts as truth to questioning these same accounts, in the case of Taki Onqoy I suggest that
denying the movement’s existence erases the native practices and voices associated with the
movement. Though Molina and Albornoz are writing for a royal audience, other letters between
Spanish authorities in the New World suggest that Taki Onqoy was very real and that it
threatened effective evangelization.
It is logical that Taki Onqoy practitioners utilized and built upon prehispanic practices.
The movement was an explicit attempt to revitalize these practices in order to overthrow Spanish
and Catholic rule. While ultimately unsuccessful in that it did not produce a change in control,
the Taki Onqoy movement and the extirpation campaign undertaken by Albornoz to stomp out
these practices served as models for future revitalization processes and extirpation campaigns in
the Andes in the seventeenth and eighteenth centuries. The relevance of Taki Onqoy persists in
the regions of Huamanga, Apurímac, and Huancavelica today (Figure 9.1). As a movement, Taki
Onqoy is regularly discussed by people who live in these regions, and many say that the
“danzantes de las Tijeras” competitions are a direct relative of the Taki Onqoy dances.113
While debating the existence of Taki Onqoy in the sixteenth century is a worthwhile
pursuit, and indeed, one that needs much more study, Taki Onqoy’s position as an indigenous
rallying cry in present-day highland Peru legitimizes the movement and its meaning. I suggest
that scholars shift their considerations of Taki Onqoy not as a bounded “thing” which did or did
not exist, but rather as a suite of heterodox (idolatrous according to the Spanish) practices which
113
See José María Arguedas, who uses these dances often in his work (1975). See also Samuel Villegas Paucar
(2015), Bush (2012), Gose (2008). For the links between spirit possession and trancelike dancing, see Roel Pineda
(1976), Vivanco (1976), Barrionuevo (1988), Castro-Klarén (1990), and Cavero (2001).
499
were a continuation or revival of native practices but were entangled with Spanish Catholic
practices.
Figure 9.1. Sign in Pampachiri advertising “las rutas del Taki Unquy.”
Contributions and Future Directions
Prior to this dissertation, debate over the various facets of Taki Onqoy was derived from
few documents by Spanish authors produced in the second half of the sixteenth century.
Secondary literature on the movement arrived at various conclusions regarding the geographic
expanse of the movement, its practices, its threat to Spanish control, and its tendency to inflict
local Andeans throughout the Central Highlands of Peru. This dissertation makes important
contributions to the study of Taki Onqoy, the Early Colonial Period, Spanish conquest, local
ritual, and revitalization movements. It provides a new methodology for how to study religious
500
movements by arguing that cultural change is generated through the repeated practices of
communities and thus like any other facet of life, religious practices, performance, and ritual
should be visible archaeologically. Perhaps more importantly, this research re-situates what has
been considered a “local rebellion” back to the material remains of those who practiced it rather
than conceiving of the movement solely in how it related to Spanish priests. Theoretically, I have
tried to account for the diversity and ambiguity in human behavior—those who practiced Taki
Onqoy could also have publicly consented to Spanish ritual and simultaneously, Taki Onqoy
preachers who were emphatic in their huaca support could also have believed in the Spanish
God. Despite the fact that both Catholic authorities and Taki Onqoy preachers demanded
exclusivity in worship, documents and material assemblages instead demonstrate a much more
complicated picture of the negotiation of religion and religious performance at Iglesiachayoq.
This research has made an initial step into the materials-based study of Taki Onqoy. The
questions generated in investigating this new frontier cannot be solved by one season of field
excavation at one Taki Onqoy center. The movement was not practiced by every person in every
highland town, and was deliberately hidden from view from Spanish authorities. In fact, in my
development of this project, many colleagues suggested that the project was impossible, given
that I was proposing to study a suite of largely ephemeral practices which were deliberately
concealed. However, as Liebmann writes, “the lack of identified instances of revitalization in the
archaeological record is fueled in part by the assumption that the core characteristics of these
phenomena—charisma, revelation, and prophecy—are immaterial and therefore archaeologically
invisible” (2012:16). Yet, cultural change is manifested through practices, and behaviors which
have material remains are associated with Taki Onqoy, so these remains should be
archaeologically discernable. The results of my 2015 excavations—like all archaeological
501
pursuits—offer intriguing data which must be challenged through more work on this subject. To
that end, I have identified several future directions of research which will extend Taki Onqoy
studies beyond the single-site level.
First, and perhaps most importantly, it is critical to utilize Albornoz’s visita accounts as a
way to identify other archaeological sites for comparative studies of Taki Onqoy. The documents
suggest that participation in Taki Onqoy varied between different towns, and so testing the suite
of practices at other sites is critical for understanding how the movement may have looked
throughout the highlands and not just at Iglesiachayoq. I have already linked several of
Albornoz’s settlements with their physical remains or present-day towns. Second, more
excavation is necessary at Iglesiachayoq. Since Taki Onqoy was not practiced by all individuals
at the site level, it is impossible to discern if the structures I excavated were the seats of Taki
Onqoy ritual performances.
Like the identification of other Taki Onqoy settlements, I suggest that connecting the
names of huacas listed by Albornoz with their modern toponyms is another critical venture into
Taki Onqoy studies. Several of Albornoz’s witness testimonies affirm that takiongos would also
gather together and visit prominent landscape huacas. By identifying and visiting these huacas,
we can assess how Andeans and huacas interacted during the broader Colonial Period. Similarly,
by studying these landscape huacas, we can also investigate Spanish destruction of these deities.
I have begun this process and also started to trace Albornoz’s journey between the different
settlements as a side project.
In addition to these ideas of how to investigate Taki Onqoy archaeologically at the local
level, I suggest that scholars should also more broadly study the Early Colonial Period. Since
there have only been limited projects which have directly explored the rapidly shifting
502
“transconquest” environment—and even fewer projects which are situated in the highlands rather
than the coast—there is not enough of a baseline level of understanding of this period. It is
impossible to understand which local practices are specific to Taki Onqoy and which were
generalized sixteenth-century practices without much more extensive investigation into the
period as a whole. Specifically, scholars should examine the material assemblages of local, Inka,
and Spanish settlements and move beyond a “presence/absence” understanding of non-local
artifacts in favor of more precise questions regarding the entanglements of the two cultures and
how local actors utilized European items (Bray 2019; VanValkenburgh 2019). In the same way,
church archaeology can provide perspectives on how local and Spanish interests intertwined in
such a critical space—while there are several examples of salvage excavations during church
remodeling, the positioning and interment patterns give clues as to how the living were
conceiving of their dead. Moreover, studying the physical remains of these interments can shed
light on how health and activities changed throughout the Spanish conquest.
Returning to Taki Onqoy, one last major worthwhile archaeological pursuit should test
the documents at face value through evaluating the connections between Taki Onqoy and
Vilcabamba. Though both Molina and Albornoz definitively link the two events, secondary
literature has largely critiqued this connection based on the arguments by the two priests. Gose
(2008) presents the only recent opinion that Taki Onqoy was the religious counterpart to the
violent planned uprising in Vilcabamba, and he even brings in new data (cited in Lohman
Villena 1941) in the form of a letter from Gaspar de Sotelo discussing weapons amassing in
Xauxa. The connections between Vilcabamba and Taki Onqoy can and should also be considered
archaeologically, in addition to the continued work of historians researching this time period.
503
In conclusion, Taki Onqoy could have functioned as a covert movement. Taki Onqoy
preachers demanded their adherents to reject Spanish traditions, religion, and everyday practices,
and excavation results are largely consistent with this demand in the domestic sector. Inhabitants
of Iglesiachayoq were able to continue daily life with little interruption from either Spaniards or
Takiongos. However, mortuary remains of those interred in the church suggests that those who
lived at Iglesiachayoq took a wide array of positions in relation to either Catholicism or Taki
Onqoy. While some directly sought to conform to new Spanish mandates regarding death and
burial, others maintained or revitalized prehispanic practices when interring their dead in the
church. Still others likely followed Spanish rules in public interment of individuals, while
secretly resisting these rules through the covert disinterment and relocation of their kin from
beneath the church.
Taki Onqoy’s internal contradictions were mirrored by the remains of its entangled
practices. While explicitly indicative of the rejection of Spanish culture and religion, Taki Onqoy
was also at least partially structured by colonial cultural milieu in which it was enacted. It was
produced through its performances at specific places by individuals who interacted with
particular objects, and exploring these entanglements is key to understanding Taki Onqoy. Taki
Onqoy also serves as a bounded example of how religious practices, landscapes, and objects took
on new meanings in the Early Colonial Period, and its contradictions should be embraced as
indicative of a broader framework of rapid and intentional cultural revitalization.
504
APPENDIX A: Faunal Analysis
Faunal Analysis – Unit Descriptions/Summaries
September 30, 2016: Sarah Kennedy
General Findings
The faunal assemblage from 2016 excavations at PATO totaled 673 NISP (number of
identifiable specimens), indicating a relatively small sample. Total bone weight of the
assemblage was 2,196.71 grams. The minimum number of individual (MNI) animals in the
assemblage was 30, calculated by excavation unit. The assemblage consisted entirely of mammal
remains, with no other classes present. Within mammals, 8 different taxa categories were
identified. Native Andean taxa included guinea pig (Cavia porcellus), camelid (Camelidae) such
as llama or alpaca, and even-toed ungulates (Artiodactyla), which include camelids and possible
deer. Introduced Eurasian domesticates included cattle (Bos taurus) and horse and donkey
(Perissodactyla). One canine (Canidae) fragment was also identified.
While newly introduced Old World mammals are present in the sample, camelids tend to
dominate. Camelid specimens are the most frequent besides unidentifiable mammal remains,
with NISP = 73 (11% of the sample) and MNI = 13 (43% of the sample). Eurasian animals (cow,
horse/donkey) were found in three different units (2 total sectors). In Sector I, Eurasian mammals
are found in UE’s 1 and 3. In Sector II, Eurasian domesticates are found in UE 6. There was only
one unit (UE 10) to have guinea pig remains (NISP = 2, from a juvenile).
Overall, there was a relatively high percentage of burned specimens in the assemblage,
with 59% (NISP = 398) exhibiting burning. UE 4 material almost entirely burned (97%). Units
17-19 were the only units that did not include a burned specimen. Degree of burning revealed
variation in burning temperature, with partially black (NISP = 7), black (NISP = 28), and
black/white (NISP = 363) specimens identified. The majority of specimens were burned at higher
temperature levels (black/white), although the majority of the 363 black/white specimens came
from one specific context (UE 4, 92% of black/white total). This could indicate a time-specific
burn event, or possible one type of trash removal.
Butchery was only observed on 23 specimens in the assemblage (3.42%). Types of
butchering included saw marks (n=8), hacks (n=6), and cuts (n=39). Age profiles for the
assemblage indicate a majority of adult animals, with some juvenile camelid and guinea pigs
identified. One sub-adult camelid was identified in UE 4, along with many specimens of a
juvenile camelid (NISP = 16, MNI =1). All Eurasian domesticate remains were from adult
individuals.
Skeletal frequencies determined by minimum anatomical units (MAU) derived from
NISP values were calculated for camelid specimens only. A majority of meat cuts came from
hindquarter, hind foot, and foot elements. Foot and hind foot elements are less-meaty, and may
indicate less nutrient cuts of meat. However, these elements are also usually left behind after
butchery, and thus may indicate areas of local slaughter and butchery.
Finally, some pathologies and taphonomical effects were recorded for this assemblage.
Mild exotosis was identified on 2 camelid elements in the assemblage, and some form of
weathering did occur on 7% of the sample, with some unidentifiable black shiny residue, as well
as root etching, possible canine gnawing, and general weathering due to climate.
505
UE 1
UE1 was one of the larger contexts, with 188 NISP. Of these, 5.8 % were burned (black/white)
and 4.8 % exhibited some form of butchery. There were at least two camelids (one adult and one
juvenile), as well as at least one adult horse or donkey. Additionally, one artiodactyl (camelid or
deer) and one possible cow or horse was identified. UE 1 was one of only three units to have
both Eurasian (NISP = 29) and native (NISP = 22) taxa identified.
When examining frequency of camelid meat cuts, we see a higher number of hindquarter
specimens (NISP = 8), and MAU (minimum anatomical unit) proportions indicate higher
concentration of hindquarter elements (MAU % = 42.8).
UE 2
UE 2 consisted of 13 NISP. 6 specimens exhibited burning (46% of unit total), and were all
burned black. There was no observed butchery. One camelid individual and one artiodactyl
individual were identified, although age was not possible to assess. The camelid element came
from the forefoot of the animal.
UE 3
UE 3 included 8 NISP. Of these, 3 were burned varying degrees of black and black/white, and 2
exhibited signs of butchery (cut and hacks). One camelid and one cow individual were identified,
making UE 3 one of the three units to include both Eurasian and native taxa. Camelid meat cuts
indicate elements from the hindfoot and foot.
UE 4
UE 4 was the largest contexts, with 359 NISP. Almost the entire context exhibited burning
(NISP=347, 96.7%) and were burned varying degrees of partially black, black, and black/white.
Interestingly, none of the specimens exhibited observable signs of butchery. One juvenile
camelid and one sub-adult camelid were identified, as well as at least two artiodactyls. Meat cuts
for the juvenile and sub-adult camelid indicate a range of cuts, with elements from the head,
hindquarter, forequarter, and foot. UE 4 was the only unit where a sub-adult specimen was
identified.
UE 5
UE 5 consisted of 32 NISP. Of these, 12 fragments were burned (37.5 %) varying degrees of
partially black, black, and black/white, and 6 fragments (18.8%) exhibited signs of butchery such
as cuts and sawing. One adult camelid and one artiodactyl were identified in the sample. The
adult camelid in question exhibited mild exotosis (or ringbone) on two 1st phalanges (a pathology
related to over-work, shock, or prolonged time on uneven surfaces) (see deFrance 2010). This
pathology could eventually lead to lameness in animals, although the very mild condition of
exotosis on the camelid phalanges in question do not suggest considerable lameness. Skeletal
frequencies calculated for the camelid indicate mostly lower limb elements (hindfoot and foot).
UE 6
UE 6 had a total of 6 NISP. One of the specimens was burned (7.7% of context) partially black,
and two exhibited butchering with both cut and saw marks (15.4%). This was the final unit that
506
had both native and Eurasian animals identified, with one adult cow and one adult camelid
individual were identified.
UE 7
UE 7 included 25 NISP. Of these, 9 fragments were burned (36%) varying degrees of partially
black, black, and black/white, and 3 fragments exhibited signs of butchery (cut marks) (12% of
sample). Two camelid individuals and one artiodactyl individual were identified in the unit. Age
profiles indicate at least one juvenile and one adult camelid. Skeletal frequencies calculated for
camelids in UE 7 indicate higher proportions of hindquarter elements (MAU % = 50).
UE 9
UE 9 consisted of 16 NISP. 7 of the specimens were burned (43.7%) black or black/white and
there was no observable butchering. One adult camelid and one artiodactyl were identified.
Camelid elements consisted of forefoot and hindfoot meat cuts.
UE 10
UE 10 was one of the smaller contexts, with a total of 7 NISP. Two of the specimens were
burned (28.6 %) black/white and no butchery was observed. This was the only unit where guinea
pig was identified (NISP = 2). In total, one juvenile guinea pig was identified, as well as one
camelid and one artiodactyl specimen.
UE 17
UE 17 was also a smaller context, with 4 NISP. None of the remains were burned and none were
butchered. There were no identifiable taxa in this context besides unidentifiable mammal
remains.
UE 18
UE 18 was a small context with only 6 NISP. None of the remains were burned nor did they
show signs of butchery. The one canine specimen identified from this assemblage was found in
this unit (NISP=1), and was most likely a local fox, though could have possibly been a domestic
dog (unable to determine with present element).
UE 19
UE 19 was the smallest context, with a NISP of 2. Artiodactyl (NISP=1) and camelid (NISP=1)
were identified in this unit. None of the remains were burned. The camelid specimen was from
an adult camelid, and exhibited evidence of heavy butchering (5 cut marks). This was an
astragalus (foot).
References:
deFrance, Susan D. (2010). Paleopathology and Health of Native and Introduced Animals on
Southern Peruvian and Bolivian Spanish Colonial Sites. International Journal of
Osteoarchaeology, 20(5), 508-524.
507
APPENDIX B: Stable Isotope Analysis Report
Prepared by Anna Gurevitz
STABLE ISOTOPE ANALYSIS – Results Collagen
The isotopic values for the 8 individuals excavated and sampled are listed in Table 4. The range
of δ13Ccollagen is -13.4‰ to -11.1‰ (average= -12.5‰; sd=0.79‰). The range of δ15Ncollagen is
10.1‰ to 11.8‰ (average=11.26‰, sd=0.70‰). This falls into the range of individuals eating
C4 plants, particularly maize in the Andes and animals that fed on C4 plants as well, indicating a
diet that was common, especially in Andahuaylas, a settlement close-by, during the Late
Intermediate Period (Kurin). All eight individuals cluster together in a tight range for δ13Ccollagen
values, indicating that these individuals, who relatively represent each unit, ate similar diets,
although those excavated from Unit 17 (n=4) have the largest range, but this is not significant.
When looking at the δ15Ncollagen, we see a similar tight cluster, with Unit 19 (n=2) having the
largest range, although again, the range is not significant. Overall, individuals seem to have a
similar diet in the last few years before their death, even individual 1010 from Unit 19 whom we
hypothesize to be the Spanish priest.
Table 4. Results of the for δ13Ccollagen, δ15Ncollagen, and δ13Ccarbonate analysis for the individuals
buried beneath the church at Iglesiachayoq.
δ15N
(‰)
Collagen
δ13C
(‰)
Enamel/Bone
Carbonate
C:N
Collagen
Locus
Unit
Sex
Age
(years)
Bone/Tooth
δ13C
(‰)
Collagen
906
17
U
35+
Premolar/Lon
g Bone
-12.9
11.8
-6.39
2.7
907*
17
F?
15+/-36
months
Long bone
-
-
-7.58
-
907
*
*
*
3rd molar
-
-
-6.17
-
508
907-2
17
U
U
Long Bone
-
-
-7.85
-
908
17
U
5+/- 1.5
years
1st Molar
-
-
-5.67
-
912
17
U
20s
-13.4
10.3
-3.83
2.7
915
17
U
20-30
-11.6
11.8
-4.64
2.7
916
17
U
U
Long Bone
-13
11.1
-
2.7
956
18
F?
5+/- 1.5
years
1st Molar
-
-
-5.80
-
958
18
U
Adult
Long Bone
-11.1
11.6
-
2.7
960
18
F?
Adult
Long Bone
-12.8
11.6
-
2.7
1008*
19
U
20s
Long Bone
-
-
-8.69
-
1008
*
*
*
3rd Molar
-
-
-6.56
-
-13
10.1
-3.23
2.7
-12.2
11.8
-
2.7
3rd Molar/
Long Bone
3rd
Molar/Long
Bone
1009
19
U
Mix
3rd
Molar/Long
Bone
1010
19
U
20s
Long Bone
Table B.1. Results of the for δ13Ccarbonate, δ13Ccollagen and the δ15Ncollagen analysis for the individuals
buried beneath the church at Iglesiachayoq.
509
Figure B.1. Results of the stable isotopic analysis of the plant material collected from the area of
Iglesiachayoq. Plant samples included maize, oculla, quinoa and potatoes. Data of the camelids
and guinea pigs comes from published data out of Conchopata in Ayacucho (Finucane et al.
2006).
STABLE ISOTOPE ANALYSIS – Results Carbonate
The isotopic values for the 9 individuals excavated and sampled are listed in Table 5. For
individuals 907 and 1008, bone and enamel carbonate samples were taken. The range of
δ13Ccarbonate for the enamel is -6.56‰ to -3.23‰ (n=8; average= -5.29‰, sd=1.24‰). The range
of δ18Ocarbonate for the enamel is -11.41‰ to -7.66‰ (n=8; average= -9.79‰; sd=1.55‰).
However, the premolar and the first molars sampled have not been corrected for, therefore, when
taken out of the sample, the range of δ13Ccarbonate is -6.17‰ to -3.32‰ (n=5; average= -4.89‰,
510
sd=1.45‰). The range for δ18Ocarbonate when the premolar and first molars are removed is the
same (n=5; average= -10.03‰, sd=1.40‰).
Overall, this data suggests that there may be some migration or differentiation happening
within Iglesiachayoq’s church population. Unit 19, individuals 1008 and 1009, show the largest
range of δ13Ccarbonate values, with a difference of just over 3‰, which is usually considered
significant. However, these individuals have similar δ18Oapatite values, possibly suggesting access
to a similar elevation (ecotone), although in the Andes, δ18Ocarbonate values are difficult to
interpret (Knudsen 2009; Knudson and Torres-Rouff). Additionally, the difference of Individual
1008’s δ13Ccarbonate values between the third molar enamel and their long bones, suggesting there
may have been a slight change in diet from childhood to adulthood, although the difference is not
significant.
δ13C
Bone/Tooth
(‰)
Carbonate
δ18O
(‰)
Carbonate
Locus
Unit
Sex
Age
906
17
U
35+
Premolar
-6.39
-9.95
907*
17
F?
12 – 18
years
Long Bone
-7.58
-12.79
907
*
*
*
3rd Molar
-6.17
-11.41
907-2
17
U
U
Long Bone
-7.85
-12.64
908
17
U
3 – 7 years
1st Molar
-5.67
-8.81
912
17
U
20s
3rd Molar
-3.83
-7.66
915
17
U
3rd Molar
-4.64
-10.41
956
18
F?
1st Molar
-5.80
-9.36
1008*
19
U
20 - 30
years
3 – 7 years
20 – 30
years
Long Bone
-8.69
-11.30
1008
*
*
3rd Molar
-6.56
-10.41
*
511
1009
19
U
Mix
3rd Molar
-3.23
-10.27
Table B.2. Results of the for δ13Ccarbonate and the δ18Ocarbonate analysis for the individuals buried
beneath the church at Iglesiachayoq.
Figure B.2. Results of the stable isotopic analysis of the skeletal remains from Iglesiachayoq.
DISCUSSION: Highlighted Individuals
Individual 912 and locus 1009
Individual 912 and locus 1009 comes up again as a highlighted individual with their
isotope data. Individual 912 had the highest δ18Ocarbonate value out of all the samples, indicating
that this individual may have grown up in a lower altitude than where they died. However, as
mentioned earlier, δ18Ocarbonate values are difficult to interpret in the Andes, so this could be
misleading. However, the δ13Ccarbonate, δ13Ccollagen, and δ15Ncollagen values also separate this
individual as having a unique signature compared to the other individuals sampled. Their
512
δ13Ccarbonate was high, pairing closely with locus 1009, which represents the locus that had the
adult bundle and the juvenile skull (the sample was taken from the adult long bones and a 3rd
molar found with the bundle). Additionally, their δ13Ccollagen and δ15Ncollagen values were lower
than the rest of the group, again pairing closely with locus 1009, but the difference from the rest
of the samples is not significant. Finally, both of these loci together form an interesting pair in
that both present arguments for being individual(s) of prestige and showcasing the range of
negotiation within the church space. Locus 1009 was found at the foot of the altar, a traditionally
prestigious space within the church and was found with both a tupu and a needle. However, this
locus was found as a bundle with adult long bones and a juvenile cranium. Individual 912, on the
other hand, was found extended in the back of the church with a tupu and Nueva Cadiz turquoise
beads. The fact that this individual was found with these grave goods suggests they may have
had some prestige in life, even though they were in the back of the church. The only value that
differed within their isotopic values were the δ18Ocarbonate, as individual 912’s value indicated
they grew up at a lower altitude while locus 1009’s δ18Ocarbonate value falls within the middle of
the group, suggesting this individual grew up in the Chicha-Soras Valley or an area in a similar
altitude. Overall, these two groups of samples give arguments for migration and possible
individuals (or their families) with more prestige than others within the negotiation of their final
resting place in the church. With a larger sample size, they may be within the range of expected
values, but for this sample size, they represent one of the ends of the range.
513
APPENDIX C: Ceramic Analysis
Ceramic Artifacts at Iglesiachayoq: Results of the 2015 Excavations
Iglesiachayoq was founded by the Inka, but inhabited by local Soras peoples and
eventually, some Spanish authorities.114 Given Iglesiachayoq’s location outside of Cuzco and
date of founding ~mid fifteenth century, we hypothesized that the site would contain mostly
provincial Inka and local (Chicha/Soras) utilitarian wares. However, since Iglesiachayoq is
located near several other Wari sites (see Meddens 1986), we also considered the possibility of
Wari fragments which had been brought from nearby Wari-era settlements. Finally, given the
intermittent presence of Spanish authorities (as demonstrated from metal artifacts and church
burials), we also predicted that a conservative amount of Early Colonial Period ceramic
fragments would be found. Goals of analysis focused on establishing basic vessel forms,
decorations, and chronological typologies for the Iglesiachayoq. In the next sections, I will detail
basic characteristics of Wari, Soras, Inka, and Spanish ceramic assemblages and then present the
analysis data from ceramic recovered during excavation at Iglesiachayoq.
Wari Ceramic Assemblages
Wari ceramic assemblages contain regional variations—Meddens’ 1986 dissertation
focuses on the Wari occupation of the Chicha-Soras Valley, and thus I will refer to Meddens’
findings in order to describe Wari ceramic traditions in this area. Meddens divides the Middle
Horizon ceramic assemblage in the Chicha-Soras Valley into four main phases, and goes into
great detail about the specific vessel types and decorations found during each phase.
114
Although this population of Spanish authorities may not have been permanent—priests could have traveled
between various towns in the region.
514
Middle Horizon Epoch 1 (All recovered from the site of Yako)
Vessel Forms: round based open bowls, flat based open bowls, and short neck jars.
Decoration: Acuchimay polychrome zigzag (Figure C.1. from Knobloch 1983)115, Black
Decorated designs, absence of mythical creature designs, matte finishes
Figure C.1. Acuchimay polychrome zigzag from Knoblock 1983.
Middle Horizon Epoch 2 (Excavated at Chiqna Jota)
115
Bennett 1953, Knobloch 1983
515
Vessel Forms: round based open bowl, flat based open bowl, cup, incurving bowls,
cantaro, short neck jars
Decoration: Vinaque, Black Decorated, and Curawasi, largely geometric designs, Chicha
style
o Chicha style: lines, bands of the designs are thinner, more regular, and colors are
paler, most common designs are variations on ovals and rectangles
Middle Horizon Epochs 3 and 4
Vessel Forms: round based open bowl, simple incurving bowl, animal lug bowl, simple
face neck jar, simple short neck jar
Decoration: dominated by Chicha style, designs derived from Vinaque types, maybe
derived from wing panel
Late Intermediate Period Local Ceramic Styles
The Late Intermediate Period was characterized by a breakdown of the Wari state such
that smaller, local alliances and rivalries developed. There are a number of regional local ceramic
styles during this time, which vary according to area. Again, I borrow from Meddens’ 1986
dissertation here to describe the Late Intermediate Period ceramic styles in the Chicha-Soras
Valley. Meddens divides the LIP into two phases, LIP 1 and LIP 2.
Late Intermediate Period Phase 1
Decoration: Chicha style loses last vestiges of influence from earlier Wari styles, Toqsa
style (Sondondo) present
Late Intermediate Period Phase 2
516
Decoration: Chicha style is replaced by Soras style, in which none of the earlier designs
are retained, many vessel shapes disappear.
o Soras: smudged yellow brown wares, mostly undecorated, rough plastic
decoration. Modeled faces with large features and pronounced bottom lips or
chins. Badly made, overfired, sometimes has vitrified surfaces like a natural
glaze. (see image below, from Meddens 1985:396)
Figure C.2. Soras-style ware, from Meddens 1985.
Late Horizon Ceramic Assemblages
The types and distributions of Inka state pottery across the Empire varies greatly
according to proximity to the capitol of Cuzco, investment in particular regions, and
517
administrative centers vs. local settlements. Cuzco-style Inka wares are generally characterized
by standardized vessel forms and geometric designs, and are usually related to serving and
consumption functions (Julien 2004). Outside of Cuzco, the most common Inka forms are
aribalos, shallow plates, and pedestal cooking pots (Bray 2003: 125). The basic decorative styles
of Inka state pottery includes Cusco Polychrome A, the fern motif, and Cusco Polychrome B,
which consists of black triangles in horizontal rows and/or concentric black diamonds (Rowe
1944:47).
In provincial settlements outside of the Inka heartland, local potters often imitated Inka
state vessel forms or designs (Abraham 2010: 190). The adoption of these forms or designs was
often a response to incorporation into the Inka Empire, although local wares also continued into
the Late Horizon.
Vessel Forms: aribalos, shallow plates, pedestal cooking pots
Decoration: fern motif, diamond motif, polished wares, coarser wares, slipped wares,
burnished wares
Early Colonial Period Ceramic Assemblage
The Early Colonial Period ceramic assemblage in the New World consisted of new
ceramic technologies: wheel-thrown ceramic forms and new types of glazes. One type of
commodity circulating in the Early Colonial Period were majolica wares, “relatively high-fired
earthenware ceramic covered with an opaque, tin-lead glaze or enamel and brightly painted
decoration” (Rice 2013: 651-52). Majolica was used in all types of contexts, and was produced
518
both in Spain and in New World manufacturing centers116. Majolica produced in Cuzco was
decorated in green and purple-black-brown, and Cajamarca pottery was blue-and-white (Rice
2013). A typical vessel form in the Early Colonial Period was the botija, or coarse earthenware
ceramic amphora storage jar often used for wine or brandy storage (Weaver 2015).
Given the early date of occupation at Iglesiachayoq, we expected very little Early
Colonial Period ceramic—perhaps some glazed sherds or novel vessel forms.
Vessel Forms: All types
Decoration: tin or led enameled glaze, wheel-thrown ceramics, imported wares.
Ceramics at Iglesiachayoq
Ceramic artifacts were the most ubiquitous type of artifact recovered at Iglesiachayoq,
consisting of 8,720 total recovered sherds from 19 excavated units. In general, the amount of
ceramic sherds recovered is very low for the area of excavation—this is because Iglesiachayoq
was only occupied for 100 years maximum, and its inhabitants were likely resettled to the
reduccion of Pampachiri in the 1570s. We divided the assemblage into diagnostic (n=570 sherds)
and non-diagnostic (n=8150 sherds). Diagnostic sherds consisted of rims, bases, necks, and
decorated body sherds. All diagnostic sherds were analyzed for vessel forms, types, decoration
styles, and pastes. For non-diagnostic sherds, only count and weight were recorded.
There were several difficult aspects to studying the ceramic at Iglesiachayoq. First,
although Frank Meddens completed a ceramic typology for the Middle Horizon in the Chicha-
116
Including Cuzco, established in 1588 (Oberti Rodriguez 1999), Cajamarca in the north (Stastny 1981), and the
northern Lake Titicaca basin (Tschopik 1950).
519
Soras Valley, he did not provide a detailed typology for the Late Intermediate Period or Late
Horizon. Local artifacts collected by Meddens are lost at this time, and thus there was no
reference collection with which to compare the sample from Iglesiachayoq. Second, there was
only one whole vessel collected at Iglesiachayoq, although we were able to partially reconstruct
several more. We thus utilized the rim sherd as the primary unit of analysis, and because they
were often so fragmented, we could not always determine vessel shape or the design motif.
Vessel form
Diagnostic sherds were sorted into vessel form categories of unrestricted and restricted
forms, and where possible, broken down even further into categories including restricted
ceremonial wares (aribalos), restricted storage wares (jarras, botellas), restricted cooking wares
(ollas) unrestricted feasting wares (tazas, vasos, copas, curved bowls, straight bowls, platos),
unrestricted serving wares (escudillas, tazones), figurines, pirurus, and unidentifiable.
Restricted Vessels
Restricted vessels were defined as any vessel in which the diameter of the opening was
less than the maximum vessel diameter (Shepard 1956: 228). Restricted vessels included
aribalos, jarras, botellas, cantaros, and ollas. The differences between the subtypes were
identified and differentiated by necks, shoulders, and body forms, as well as the diameters of
their openings. Overall, restricted vessels composed 46.8% of the identifiable diagnostic ceramic
assemblage, broken down into aribalos (4.4%), storage wares (10.2%), and ollas (32.3%). The
morphological characteristics of the restricted forms suggests that they were used for storage,
cooking, fermentation, serving, and transport of food and drink.
520
521
Figure C.3. Examples of Aribalos.
522
523
Figure C.4. Examples of Jarras.
Figure C.5. Examples of Botellas (smaller diameter, flared opening)
524
525
Figure C.6. Examples of flared necked ollas.
526
Figure C.7. Example of a straight-necked olla. IC.0751.0.010
527
Aribalos are Inka flared-rim jars which usually contain tall, flared necks,
anthropomorphic “ears” on the underside of the rim, zoomorphic lugs on the center of the body,
two handles, and a pointed base. (Bray 2009: 114-115). These vessels were largely ceremonial
for the Inka Empire, and were used for producing, storing, and serving chicha. Utilizing museum
studies and previously recovered collections, Bray’s 2003 article argues that the aribalo was the
most common Inka state vessel form outside of Cuzco.
Despite Bray’s claims, we recovered only a low frequency of independent aribalos at
Iglesiachayoq, consisting of some 4.4% of the assemblage and clustered in two locations in
Sector 1. The two clusters of aribalos fragments in Sector 1 (some 148 fragments which could be
distinctly separated into 25 different individual vessels) were located 1) in the kuraka house and
2) in Structure 004, a rectangular structure adjacent to the kuraka house. The locations of the
aribalos are indicative of their main usage at Iglesiachayoq—in areas of importance or areas used
to prepare and serve food for church activities. All were classified as Inka style, often have a red
slip, and some had the “pierced ears” under the rim or nubbed handles. The rim diameters of the
aribalos in the assemblage range from 14cm to 40cm. Although the proportion of aribalos is low
compared to other vessel forms at Iglesiachayoq, the numbers are consistent with other sites
excavated which date to this time period117.
Jarras and botellas made up 10.2% of the diagnostic ceramic forms, and are vessels of
varying sizes with rounded bodies and restricted necks. The majority of the recovered sherds
contained either rims/necks or body sherds, so I have lumped these forms together into the
category of “restricted storage wares” since the differences between them could not be easily
117
See Abraham 2010
528
parsed. Only 36% of these vessel types had evidence of soot, suggesting that these vessels were
not often used for cooking. The average diameter for these types of vessels was 13cm, and the
diameters ranged from 5-5cm.
Figure C.8. Example of Soras-style rustic jar.
The final type of restricted vessel was ollas, which made up 32.3% of the overall
assemblage. Compared to jars, ollas have wider mouths and rounded bodies, and many are of a
much larger size in order to accommodate cooking of various foodstuffs. Wider-mouthed vessels
allow individuals to access the interior contents of the vessels, thus making them useful for food
preparation. Ollas had the highest percentage of soot-covered sherds (60%), supporting the
function of the vessel. The predominant olla form in this assemblage was flared-neck ollas,
although there were some straight-necked and neckless ollas as well. The diameters for ollas
ranged from 7-46 cm with a mean diameter of 19cm.
529
530
Figure C.9. Examples of Rounded Bowls.
Figure C.10. Examples of Straight-sided Bowls.
531
Figure C.11. Example of Curved Taza.
Figure C.12. Example of Straight-sided Vaso.
Figure C.13. Example of plato.
532
Figure C.14. Example of escudilla
Figure C.15. Example of tazon.
533
Other ceramic artifacts: Pirurus and Figurines
Finally, the last two categories of ceramic forms were pirurus (1.4% of diagnostic sherds)
and figurines (1.2% of diagnostic sherds). Piruru is the quechua word for spindle whorl, and
these are found throughout Andean history and typically made of stone, ceramic, or wood. All of
the pirurus recovered at Iglesiachayoq were made of ceramic and represented various forms of
completion.
Figure C.16. Example of a piruru recovered from Unit 9.
We recovered nine pirurus from various units, with three coming from Unit 9 and three
coming from Unit 4. Unit 9 was a residential domestic household, and the presence of spindle
whorls is logical and corresponds with the rest of the findings. Unit 4 seems to have been a
structure used for preparation of large quantities of food and drink, so the presence of the three
pirurus recovered here is a bit more complicated. It is possible this household was inhabited
regularly by a family, which also prepared the large quantities of food and drink for church
ceremonies.
The figurines recovered are almost exclusively mangoes, or the small handles generally
found on plates. All recovered mangoes appear to have been made in the shape of fish, as
designated by their large eyes and gaping mouths.
535
Figure C.17. Examples of fish effigy mangoes.
Beyond these three mangoes, we also recovered what appears to be a “proper” figurine,
which we are hypothesizing could have been crossed legs or some sort of braid representation.
Figure C.18. Crossed leg figurine.
In some, the ceramic vessel assemblage from Iglesiachayoq is made up of a number of
restricted and unrestricted vessel forms, whose proportions varied across the site. We did not
recover any obviously new vessel forms which may have indicated Spanish presence, although
we did find a typical mixture of Inka and Soras wares.
536
Paste Composition
General paste types were identified after analyzing all diagnostic sherds after a fresh
fracture and with a hand lens. Each diagnostic ceramic was evaluated according to texture of
paste (very rough, rough, medium, fine, or very fine), composition or inclusions of paste, color
(and Munsell value), oxidation (complete, incomplete, or reduced), size of tempers (largegrained and rough, moderately rough, moderately fine, and very fine), and the consistency of the
fracture pattern (sandy and irregular, semi-compact consistency with irregular fracture, compact
consistency with regular fracture, and laminate appearance with regular fracture).
There were some general similarities amongst all sherds—the vast majority of sherds had
inclusions of mica and quartz, with subsidiary inclusions of basalt, feldspar, and a dark red grain
which we tentatively identified as iron. There were three main paste types which accounted for
the majority of sherds, and we also identified four subsidiary paste types which were much less
common. 122 sherds were exported for analysis at the Elemental Analysis Facility at the Field
Museum. The raw results from the sherds are included in the appendix, but statistical analysis is
still ongoing.
Type A: Paste Type A is the most common ceramic type (n=4461/8720 sherds, or 51.2%
of the overall sample). Of the diagnostic sherds, Paste Type A is also the most common
(n=367/570 or 64.4% of the diagnostic sherds). Paste Type A is medium or fine-grained
in texture, with medium fine or fine inclusions. The paste is generally of high quality, and
inclusions consist of mica, basalt, and quart. In general, the paste colors range from
reddish yellow to reddish brown to brown. Paste Type A was the typical paste type for
locally made ceramic.
537
Type B: Paste Type B is more of a variant of Paste Type A, but a more “rustic” type.
Paste Type B is the second most common paste type (n=4074/8720, or 46.7% of the
overall amount of sherds), and also the second most common paste type for diagnostic
sherds (n=83/570, or 14.6% of the diagnostic sherds). Paste Type B has the same range of
colors and types of inclusions as Paste Type A, but it ss a much cruder texture with larger
grains and larger inclusions. Combining Paste Types A and B, these two are versions of
the most common ceramic recovered at Iglesiachayoq.
Type C: Paste Type C is the third most common paste type recovered at Iglesiachayoq,
although in comparison to the previous two paste types, it is very rare (n=123/8720 or
1.4% of all total sherds, and n=71/570 or 12.5% of diagnostic sherds). Paste Type C was
either fine or very fine in both texture and inclusion size, and inclusions consisted of
mica, quartz, basalt, and iron. The main identifying feature of this paste was the pink or
cream color of the paste. Fragments with Paste Type C were more likely to have both slip
and decorations, indicating their elevated proportion in the diagnostic wares.
Type D: Paste Type D was only identified in diagnostic sherds (n=8/570 or 1.4% of the
diagnostic sample), and made up a very small subsection of the ceramic assemblage at
Iglesiachayoq. Type D has a fine texture and fine inclusions with a pale red color, but has
large and numerous biotite mica inclusions. Compared to Types A, B, and C, Type D is
very distinctive in its color and the size and density of these biotite mica inclusions.
Type E: Paste Type E was also only represented in diagnostic ceramics (n=19/570 or
3.3% of the diagnostic sample) and was characterized by a very fine texture and almost
devoid of inclusions. The main identifying characteristic of Type E is that these ceramics
took on a laminate quality when fractured, and appeared more like rock than ceramic.
538
Type E ceramics were overfired, and the interior and exterior surfaces often took on a
vitrified appearance with bubbles or metallic surfaces. The sherds which are made of this
paste are very rustic and roughly-hewn, sometimes with added plastic applique or etched
designs.
Type F: Paste Type F was also only recovered as a portion of the diagnostic ceramic
assemblage (n=15/570, or 2.6%). Paste Type F is similar to Type E in that the paste is
almost devoid of inclusions and appears laminate when broken, but these ceramics are
well-crafted and almost exclusively found in Sector 1. All of the Type F sherds had either
slip, decoration, or both, and many (n=5 or 33%) were recovered in church excavations.
None of the Type F sherds had evidence of soot, thus indicating their usage was for
purposes other than cooking.
Type G: Paste Type G was the last identified paste type, and was characterized by
medium-textured grains with large inclusions of mica, basalt, quartz, and feldspar. The
paste itself was generally bright red in color, and set apart by the high concentration of
large feldspar inclusions. Paste Type G was the least common ceramic type recovered
from excavations (n=7/570 or 1.2%) and was only recovered in diagnostic ceramic
sherds.
Decorative Techniques
A portion of the pottery at Iglesiachayoq was decorated with paint, slip, plastic
applications, or incisions. Due to the fragmented nature of the assemblage, entire design motifs
were difficult to discern. This section describes each decoration type and its associated vessel
forms. In general, restricted vessels had decoration only on the exterior surface or rim, while
unrestricted vessels usually had decoration on the interior surface.
539
Painted Decoration
Sherds with painted decoration make up N of the decorated assemblage. Below I provide
pictures of common design motifs and their proposed chronological date.
Monochrome Decoration/Slip
Buff/Yellow slip: Pale yellow in color, external slip only, recovered only in unit 1. Late
Horizon/Early Colonial
Figure C.19. Example of Buff/yellow slip
Red slip: red or orange red, found on external and internal surfaces, recovered site-wide.
Late Horizon, likely provincial Inka.
540
Figure C.20. Provincial Inka ware.
Burgundy Slip: Reddish-brown slip very distinct from the Red slip
Figure C.21. Burgundy slip.
Polychrome Decoration and Designs
Although a large proportion of the ceramic assemblage had some form of slip, we had a
much smaller proportion which had painted polychrome designs. Of the design types, I have
divided the assemblage into five basic categories, of which break down into provincial Inka
or Wari designs.
Red slip with black line decoration: These designs were predominately black lines atop a
red slip. They had various patterns, including two which appeared to be versions of Inka
motifs.
541
Figure C.22. Versions of Inka motifs recovered from the kuraka house at Iglesiachayoq.
Figure C.23. Examples of Black line on red slip.
542
Red slip with multicolor designs: These decorated sherds had various colors including
black, cream, red, and maroon, usually atop a red slip. Some of these fragments may date
to the Middle Horizon, given their specific design elements.
Figure C.24. Provincial Inka fragments.
543
Figure C.25. Possible Middle Horizon Epoch 2 Design, based on similarities between the
images on the right (Meddens 1985: 363).
544
Figure C.26. Possible Wari sherd.
Gray with brown or reddish brown. There were very few fragments with this type of
design schema.
Figure C.27. Gray with Brown and Black. Late Horizon/Early Colonial Period
Peach with Reddish Brown: The majority of these ceramics were recovered in Sector 1
and consist of a peach/light red slip ith burgundy or reddish brown designs atop.
545
Figure C.28. Reddish Brown on cream.
Plastic Decoration.
Vessels with applique are those with molded clay added to the vessel. These additions are
not structural and are purely decorative on purpose. At Iglesiachayoq, we recovered applique in
either raised horizontal bands with small vertical impressions, a raised oval mound with deep
impressions, a chain-link pattern that perhaps dates to the formative, and two examples of faceneck jars.
Figure C.29. Possible Formative fragment with chain motif applique.
546
Figure C.30. Applique with horizontal bands.
Figure C.31. Two faces from vessels with raised applique.
Surface Treatment
Vessels in the Iglesiachayoq assemblage had a variety of surface finishes, including
burnishing and smoothing. Burnishing involves rubbing the vessel with a hard, smooth object to
create a shiny, more uniform surface and is the most labor-intensive surface treatment.
Smoothing implies that the vessel was wiped with some material to create a smoothed, uniform,
but matte finish. Surface treatment is correlated with decorative technique. Most painted and
547
slipped vessels are burnished. However, vessels with applique almost always have a more rustic,
likely wiped surface. Non-diagnostic sherds usually had no surface treatment.
Spatial Patterning
Attributes like vessel form and design were also evaluated in terms of where they were
found at the site and in what frequencies they were recovered. In this way, we could determine
patterns in the spatial distribution of key attributes and test for correlations between specific
attributes and areas, sectors, or structures. The units were not uniform in size or depth, making it
difficult to compare unit assemblages. For each unit, assessing the form and styles in order to
make comparisons were reduced to the individual vessel so that multiple fragments from the
same vessel were not counted as separate sherds and thus influencing the density.
Type A Type B Type C Type D Type E Type F Type G Totals
Unit 1
110
3
14
6
6
2
0
141
Unit 2
25
0
12
0
5
2
1
45
Unit 3
14
2
1
0
0
0
2
19
Unit 4
58
2
11
0
1
2
1
75
Unit 11
3
0
0
0
0
0
0
3
Unit 12
0
0
0
0
0
0
0
0
Unit 13
0
0
1
0
0
0
0
1
Unit 5
6
1
3
0
1
0
1
12
Unit 6
10
16
1
0
0
1
1
29
Unit 7
36
8
3
0
0
2
0
49
Unit 14
2
0
0
0
0
0
0
2
Unit 8
20
6
9
0
3
0
1
39
Unit 9
18
15
4
0
0
1
0
38
Unit 10
19
21
1
0
0
0
0
41
Unit 15
7
8
1
0
0
0
0
16
Unit 16
3
0
1
0
0
0
0
4
Unit 17
7
1
1
1
1
1
0
12
Unit 18
8
0
1
0
0
2
0
11
Unit 19
21
0
7
1
2
2
0
33
Totals
367
83
71
8
19
15
7
570
Figure C.32. Distribution of Paste Types by Unit and Sector.
548
Distribution of Ceramic Types by Sector
400
350
300
250
200
150
100
50
0
Type A
Type B
Type C
Type D
Sector 1
Sector 2
Sector 3
Type E
Type F
Type G
Church and Plaza
Figure C.33. Distribution of Ceramic Types by Sector
Proportion of Ceramic Type by Sector
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Type A
Type B
Type C
Type D
Type E
Type F
Sector 1
Sector 2
Sector 3
Church and Plaza
Type G
Figure C.34. Proportion of Ceramic Type by Sector.
As is apparent in the last chart, there are some obvious discrepancies in ceramic paste
type by sector. The church and plaza as well as all of Sector 1, considered to be some of the most
important spaces at Iglesiachayoq, had the least proportion of Type B pastes, indicating that the
fragments recovered here were of a finer quality than in all other sectors. The more rustic Type B
549
ceramics were most common in Sectors 2 and 3, which is where local residents of Iglesiachayoq
would have resided. Type A and Type C ceramic were distributed relatively evenly amongst all
sectors, which is reasonable since Type A is the most common ceramic type and Type C appears
to be consistent with Wari sherds, which could have been brought from any number of Wari sites
nearby. Types D, E, and F were found in high proportions in the church—this pattern can also be
explained by the tendency to use “fancier” or exotic wares in highly visual and important spaces.
Ceramic paste types did not typically correlate with their form, but Type A and Type C
ceramics were more often associated with exterior or interior design motifs. Because
Iglesiachayoq only had one definitive occupation strata, ceramic types only tangentially
correspond with chronological time periods. The majority of the recovered ceramic was
produced in the Late Horizon or Early Colonial Period, and those which are Wari in style likely
were brought from other nearby sites.
There were no significant differences in spatial patterning for vessel types between the
three sectors. In general, those assemblages recovered in sectors 2 and 3 (away from the church
and site center) represented household cooking and storage materials, and were more rustic in
design and decoration. In Sector 1, we recovered the most decorated sherds, and in the church,
we recovered the greatest variety of ceramic types and styles, perhaps indicating that the area
was recognized as a sacred space. Almost no ceramic fragments were recovered in any external
spaces including the plaza—those fragments which were recovered were incredibly small and
difficult to determine vessel form.
Unit 4 presented a unique ceramic assemblage, consisting of large aribalos sherds and
massive ollas, whose mean diameter was 28.5 cm. The large diameters of these ollas suggests
that they were used to prepare food and drink for large numbers of people, rather than for the
550
family household unit. It thus confirms the hypothesis that those who utilized the space in Unit 4
were responsible for food preparation for church activities.
Discussion
Based on the various lines of evidence presented above, it is possible to identify a few
ceramic traditions at Iglesiachayoq and comment on their function at the site.
Inka Pottery
Iglesiachayoq contained little to know Cuzco-style Inka sherds, but most sherds with
black line decoration were classified as “provincial Inka.” 345 out of the 570 diagnostic sherds
were classified as “provincial Inka,” and of these, 55 sherds were decorated with patterns on
either the exterior or interior surface. For the Provincial Inka forms, see the table below:
Aribalo
Sector 1
Sector 2
Sector 3
Total
Olla
20
3
3
26
40
40
41
121
Jarra/Botella
Escudilla/Taz
Taza/Vaso/CStraight-sideCurved Bowl
Plato
19
2
13
10
14
5
2
3
3
8
7
3
5
0
10
31
7
21
13
32
Piruru
6
3
7
16
Figurine
4
1
1
6
5
0
0
5
Not observaTotals
45
178
5
73
17
94
67
345
Table C.1. Raw data for numbers of Provincial Inka vessels.
As can be seen in the table, the most common provincial Inka form was ollas, which were
evenly distributed amongst all three sectors. Next were curved bowls, also relatively evenly
distributed, followed by aribalos which were predominately found in Sector 1. This is reasonable
given the more ceremonial nature of the spaces and places within Sector 1. All figurines were
found in Sector 1, which had the most Inka ceramic overall. Inka sherds were most commonly
found in the core—elite households and ceremonial contexts had greater access to aribalos than
non-elite households, and given the role of these vessels, the data suggest that the elite
participated in preparing chicha for feasts.
551
Inka pottery could have been produced at Iglesiachayoq, as there is evidence of local
imitations of state wares in the ceramic assemblage. Few of the sherds appear to copy or modify
some of the distinctive Inka designs, and usage of red slip was very common in the assemblage.
Soras Pottery
Based on the ceramic assemblage from Iglesiachayoq, Soras style pottery is comprised
mainly of utilitarian plainware. A basic inventory of food preparation, service, and storage
vessels was needed to fulfill everyday household tasks and little attention or effort went into
surface finish or decoration. With little standardization in rim shape, size, or paste, it is likely
that there was no controlled or specialized ceramic production at Iglesiachayoq. These more
rustic wares were distributed site-wide, but also clustered in Sector 3, the local sector.
Colonial Pottery
We recovered no clear evidence of colonial pottery at Iglesiachayoq. Although we have
evidence of the colonial presence in iron nails, Nueva Cadiz beads, and Old World animals, we
recovered no sherds with clear evidence of glaze or wheel-spinning. Undoubtedly, ceramic
vessels were produced during the Early Colonial Period at Iglesiachayoq, but these vessels
marked a continuation in local and provincial Inka ceramic traditions and were relatively
unaffected by new technologies. It is likely that the Spanish inhabitants of Iglesiachayoq were
sparse and only residing at the site periodically, such that the brief period of occupancy did not
affect many local lifestyles in terms of material culture.
Wari Pottery
While we did recover ceramic fragments which we tentatively categorized as Wari, these
did not bear any chronological significance and instead were likely brought from other sites in
552
the region. Wari sherds may have been valued at Iglesiachayoq, especially during the nativist
revivalism movement, Taki Onqoy.
Chronology
Although a major objective of ceramic analysis, organizing the ceramic assemblage from
Iglesiachayoq in any kind of chronological order was impossible. As discussed above, this
endeavor is complicated by the highly fragmented nature of the assemblage, the small percentage
of diagnostic sherds, and the lack of complete vessels. Adding to these problems is the continuity
in form, style, and surface treatment of most of the wares. All of the sherds recovered could have
been produced in the Late Horizon and Early Colonial Period, and thus correspond to the rest of
the artifact assemblage.
553
APPENDIX D: Exported Samples and Isotope Analysis
Muestra
IC-021H
IC-022H
IC-023H
IC-024H
IC-025H
IC-026H
Codigo
IC.906.17.2.005H
Peso
Description
Condition/Handlers
13.116g Craneo – Inner
Taken from crania, Crania as
meatus izqueirda whole very
fragemented/Handled without
gloves by excavators and
researchers.
IC.907.17.1.018- 2.136g Primer molar –
All teeth were loose – crania was
H
inferior derecha pulverized and was collected
with the soil/Handled without
gloves by excavators and
researchers.
IC.912.17.1.014- 1.684g Segundo molar – Tooth was taken from maxilla –
H
superior derecha crania collapsed/Handled
without gloves by excavators
and researchers.
IC.915.17.2.016- 1.747g Tercera molar – Tooth was taken from mandible
H
inferior
– crania collapsed/Handled
izqueirda
without gloves by excavators
and researchers.
IC.1008.19.2.033- 7.048g Craneo – Inner
Taken from crania, Crania as
H
meatus izqueirda whole very
fragemented/Handled without
gloves by excavators and
researchers.
IC.1010.19.2.002- 2.153g Primer molar –
Tooth was loose, very poorly
H
superior
preserved, but taken because
izqueirda
particular individual is of interest
(found in stone coffin)/Handled
without gloves by excavators
and researchers.
Table D.1. Exported bioarchaeological samples for isotope analysis.
554
#
Type
1
Grain
2
Grain
3
Grain
4
Grain
5
Grain
6
Grain
7
Grain
8
Grain
9
Grain
10 Grain
11 Grain
12 Grain
13 Grain
14 Tuber
15 Tuber
16 Tuber
17 Tuber
18 Tuber
19 Tuber
20 Cactus
Common Species
Name
Quinoa
Chenopodium
quinoa
Quinoa
Chenopodium
quinoa
Quinoa
Chenopodium
quinoa
Quinoa
Chenopodium
quinoa
Chicha
Zea mays L.
Morada
Chicha
Zea mays L.
Morada
Chicha
Zea mays L.
Morada
Maize
Zea mays
indurata
Maize
Zea mays
indurata
Maize
Zea mays
indurata
Maize
Zea mays
indurata
Maize
Zea mays
indurata
Maize
Zea mays
indurata
Olluco
Ullucus
tuberosus
Olluco
Ullucus
tuberosus
Papa
Solanum
Natural
brevicaule
Papa
Solanum
Natural
brevicaule
Camote
Ipomoea
batatas
Papa
Solanum
Regular brevicaulte
Cactus
Echinopsis
peruviana
Photo Date
Start
Path
Collected Drying
C3
7/21/16
7/21/16
End
Drying
7/21/16
Weight From
26g
Chicha/Pampachiri
C3
7/21/16
7/21/16
7/21/16
23g
Chicha/Pampachiri
C3
7/21/16
7/21/16
7/21/16
24g
Chicha/Pampachiri
C3
7/21/16
7/21/16
7/21/16
26g
Chicha/Pampachiri
C4
7/18/16
Andahuaylas
7/18/16
14.69g
Andahuaylas
C4
7/18/16
11.86g
Andahuaylas
C4
7/19/16
11.03g
C4
7/19/16
C4
7/19/16
C4
7/19/16
C4
7/19/16
C4
7/19/16
C3
7/18/16
2.21g
Chicha/Pampachiri
Pink
Chicha/Pampachiri
Purple and yellow
Chicha/Pampachiri
Red
Chicha/Pampachiri
Yellow
Chicha/Pampachiri
Pale yellow
Chicha/Pampachiri
Golden
Chicha/Pampachiri
C3
7/18/16
2.35g
Chicha/Pampachiri
C3
7/18/16
9.02g
Chicha/Pampachiri
C3
7/18/16
9.40g
Chicha/Pampachiri
C3
7/18/16
8.17g
Andahuaylas
C3
7/18/16
7/22/16
2:30PM
7/22/16
2:30PM
7/22/16
2:30PM
7/22/16
2:30PM
7/22/16
2:30PM
7/22/16
2:30PM
7/22/16
2:30PM
7/22/16
2:30PM
7/22/16
2:30PM
7/25/16
12PM
7/25/16
12PM
7/25/16
12PM
7/25/16
12PM
7/25/16
12PM
7/25/16
10.91g
C4
7/20/16
2:30PM
7/20/16
2:30PM
7/20/16
2:30PM
7/20/16
2:30PM
7/20/16
2:30PM
7/20/16
2:30PM
7/20/16
2:30PM
7/20/16
2:30PM
7/20/16
2:30PM
7/23/16
12PM
7/23/16
12PM
7/23/16
12PM
7/23/16
12PM
7/23/16
12PM
7/23/16
9.87g
Andahuaylas
CAM
7/19/16
7/19/16
1PM
7/23/16
12PM
3.65g
Chicha/Pampachiri
555
13.91g
11.17g
13.61g
10.71g
15.17g
21 Shrub
Shrub
22 Shrub
Shrub
23 Shrub
Shrub
Baccharis
?
genera??
Margyricarpus ?
genera?????
Tetraglochin
?
Genera???
7/19/16
7/19/16
7/19/16
7/19/16
1PM
7/19/16
1PM
7/19/16
1PM
7/22/16 2.22g
2:30PM
7/22/16 1.85g
2:30PM
7/22/16 1.37g
2:30PM
Chicha/Pampachiri
Chicha/Pampachiri
Chicha/Pampachiri
Table D.2. Exported flora samples for isotope analysis.
Radiocarbon Dating Results
With the remaining bone used for stable isotopic analysis, I submitted eight samples to
the Keck Carbon Cycle AMS Facility at the Earth System Science Department at the University
of California, Irvine. All results were corrected for isotopic fractionation according to the
conventions of Stuiver and Polach (1977), and results were calibrated using the Southern
Hemisphere curve (Table D.3, Figure D.1.).
Site
Sample
ID
Lab #
Iglesiachayoq
Iglesiachayoq
Iglesiachayoq
Iglesiachayoq
Iglesiachayoq
Iglesiachayoq
Iglesiachayoq
IC.001.H
IC.007.H
IC.011.H
IC.014.H
IC.015.H
IC.018.H
IC.020.H
191876
191877
191878
191879
191880
191881
191882
Radiocarbon
Age
340 ± 20
340 ± 20
350 ± 20
330 ± 20
315 ± 20
310 ± 20
360 ± 20
δ13C
-13.3
-13.7
-13.4
-11.6
-13.6
-13.7
-12.4
Collagen
Yield
1.9
2.0
12.2
2.0
1.1
1.7
4.6
Calibrated
Radiocarbon
Age
1503-1645
1503-1645
1500-1641
1505-1650
1508-1655
1509-1659
1496-1636
Table D.3. Calibrated radiocarbon dates from remains excavated under the church (n=7). All
radiocarbon ages calibrated with the ShCal13 curve using OxCal 4.3. Calibrated Radiocarbon
Age, A.D.; 95.4% probability.
556
Figure D.1. Multiple plot graph of calibrated radiocarbon age for human bone samples taken. All
radiocarbon age calibration was done on the ShCal13 curve through OxCal 4.3. Figure by
Gurevitz 2017.
All samples fell in an acceptable range and are tightly clustered, indicating the general
use period of the church at Iglesiachayoq.
557
APPENDIX E: Tupus and Metal Artifacts
Tupus and metal artifacts from interred individuals
Metalwork is a time-honored tradition in the Andes, which has a rich history of
exploitation of copper, tin, arsenic, lead, gold, and silver in order to create an array of metal tools
and ornaments (Tiballi 2010).118 Metallurgy in the Andes was used primarily for symbolic
aspects of life instead of warfare or technology, as in the Old World (Lechtman 1984). Rather
than valuing metal qualities as a function of tools (hardness, durability, etc.), metallurgy in the
Andes often valued “aesthetic” qualities such as color, malleability, symbolic meaning, or
luminosity (Lechtman 1993). Thus, gold and silver held ritual and political significance in
Andean prehistory, particularly in relation to the Inka king and the cult of the sun. During the
Late Horizon, the production and use of metal artifacts increased, with Andean metallurgists
trying to reproduce surfaces of gold or silver on artifacts which were actually made of neither
(Lechtman 1984).
Despite the chroniclers’ focus on gold and silver in the Inka Empire, a more commonly
used metal was copper (Lechtman 1984; Tiballi 2010). Copper could be used in alloys with both
tin and arsenic. At Iglesiachayoq, all metal artifacts (other than Early Colonial Period iron nails)
were derived from either copper-tin alloys or copper-arsenic alloys. The turquoise oxidation
pattern of copper artifacts is present on nearly every recovered metal artifact within the church.
In total, the metal sample from the central church at Iglesiachayoq contains 16 artifacts: 12
tupus, one needle, one circular pendant, and two unidentified pieces of metal. Five of the
recovered tupus conform to the most basic Inka form, which is characterized by a long, narrow
118
The use of metal artifacts in religious rituals is first convincingly documented in the Chavin Cult (800 B.C.), and
was continuously practiced through the Late Horizon (Lechtman 1993, 2007).
558
pin on one end with a flattened, cresent-shaped head which contains a small hole near the place
where the pin attaches (see Figure E.1.). The remaining seven tupu pins are in a variety of forms
ranging from having a circular head, flattened, horizontal head, or even a small conical bulb on
the end (see Figure E.2.). The variety in tupu styles may indicate changing traditions after
Spanish conquest.
Figure E.1. Standard tupus recovered in the church.
559
Figure E.2. Non-standard tupus recovered in the church.
Tupu pins were very common in Andean prehispanic times (and through today) in female
styles of dress. The chronicler Guaman Poma demonstrates the use of tupu pins in two distinctive
ways in his illustrated letter to the king: in one illustration, a set of paired tupu pins are used
vertically to secure the main dress garment (Figure E.3.).119 In a second illustration, Poma
depicts the horizontal pin (t’ipki) which holds the women’s shawl closed around her shoulders
(Figure E.3.). Cobo writes that tupu pins were normally between 10 and 28cm in length (Cobo
and Hamilton 1990). For the tupus recovered at Iglesiachayoq, the length varies from the
smallest tupu at 4.21cm, and the largest measuring 20.92cm.
See Guaman Poma’s complete works, and specifically drawings 136 and 175. Also see Rowe 1995-96 and Cobo
and Hamilton 1990.
119
560
Figure E.3. Left: queen depicted wearing two vertical tupu pins. Right: lady of high status depicted
with one horizontal pin. Drawings by Guaman Poma v. 136 and v. 175.
561
APPENDIX F Unit Summaries
Unit 1, Sector 1, EST-002
Location (WGS 1984): 656773E 8427949N
Dimensions:
12m x 9.6m (divided into 12 3m x 3.2m quads)
Declared Loci: 751, 752, 753, 761, 762, 763, 764, 801, 802, 803, 791, 792, 793, 794, 771, 772,
773, 774, 775, 776, 781, 782, 783, 784
Unit 1 was located in Sector 1, the religious core of the site at Iglesiachayoq, the interior
of EST-002, also known as the “casa de kuraka.” The excavation unit was strategically placed to
encompass the maximum amount of space in EST-002, in order to learn about the domestic
activities that were occurring inside of this structure.
Summary of Quads and Loci
In total, we divided Unit 1 into twelve separate quads in order to maintain a degree of
control in excavation, each quad measuring 3m x 3.2m. We numbered the quads from the
Northwest quad to the Northeast quads (1-3), and then moved south according to the way
language is read. Each quad was also assigned a separate set of loci, and only quads 1, 3, 5, 7,
10, 12 were excavated due to time constraints. I will go over each quad locus by locus.
Quad 1
Quad 1 was located in the Northeast corner of Unit 1, and had three contexts among
natural strata and arbitrary levels, which were identified as loci. The excavation team tried to
follow each natural level as completely and faithfully as possible. Locus 751 represented the
unit’s natural soil and consisted of a moderately homogenous sand with roots, poorly
consolidated, and brown (10YR 4/3) in color. In the surface, there was a large quantity of roots,
with little evidence of cultural material. In this locus, we collected four bags of ceramic, one bag
562
of lithic, one bag of animal bone, one soil sample, and one metal artifact. The metal artifact was
an iron carret-head nail of Spanish origin, which was produced during the first half of the 16th
century (see drawing). This level was an average of 14 cm thick, ending at an average of 125 cm
BD throughout the unit.
Directly below locus 751 was locus 752, which was a natural level defined by a soil
change. Locus 752 consisted of a poorly consolidated medium-grained sand of moderate
homogeneity, very dark brown (10YR 2/2) in color. There was significantly less roots in this
level, as well as less cultural material—we only collected one bag of ceramic and one soil
sample. Locus 752 was an average of 10 cm in depth, and ended at an average of 135 cm BD
throughout the quad. At the bottom of this locus, sterile earth began to appear, marked by a very
hard, clay matrix.
Locus 753 was uncovered directly below locus 752, and was also uniformly
excavated throughout the quad. Locus 753 was a natural level, marked by a soil matrix of dark
brown sandy clay (7.5YR 3/2), which was uniformly compact. This locus was characterized by
large inclusions of pieces of clay of different colors such as yellowish red (5YR 3/8), strong
brown (7.5YR 5/6), and brownish yellow (10YR 6/8). There were no cultural materials in this
locus, and we collected only one soil sample. The average depth of this locus was 8 cm, ending
at an average of 143 cm BD across the quad.
Based on the lack of material evidence found in the preceding levels and the extremely
dense sandy clay in this locus, we decided to stop excavating this quad.
Quad 3
563
Unit 1, Quad 3 was located in the Northeast corner of EST-002. In total, Unit 1, quad 3
had four contexts among natural strata and arbitrary levels, which were identified as loci. The
excavation team tried to follow each natural level as completely and faithfully as possible. Locus
761 represented the quad’s natural topsoil and consisted of a brown (7.5YR 4/3), mediumgrained sand uniformly distributed throughout the quad. From the start of excavating this locus,
we made two amplifications to the north and east, going from the border of the quad to the north
and east walls (see drawings). From the northwest corner of this quad to the north wall, the
distance was 1.22m. From the northeast corner of this quad to the north wall, the distance was
0.20m. From the northeast corner of this quad to the east wall, the extension was 0.10m, and
from the southeast corner of this quad to the east wall, the extension was 0.54m. This locus was
filled with roots and plant material. We collected three bags of ceramic, one bag of lithic, one
bag of animal bone (located in the northern extension, 116 cm BD), and one soil sample. The
average depth of this locus was 18cm, ending at an average of 129cm BD.
Directly below locus 761 was locus 762, which we excavated fully, including the
extensions to the northern and eastern walls. Locus 762 was a dark brown (7.5YR 3/3) semicompact medium-grained sand which was uniformly distributed throughout the locus. There
were less roots in this locus, with a matrix that contained more clay and some fragments of
ceramic. In total, we collected three bags of ceramic, one bag of animal bone, and one soil
sample. Locus 762 was an average of 7cm in depth and ended at an average of 137cm BD.
Locus 763 was directly below locus 762, and was changed due to a natural change in the
soil, brown (7.5YR 4/4) semi-compact sandy clay, spread uniformly throughout the locus. Like
locus 753 in quad 1, the matrix in this quad had significantly more hard, clay rocks and
contained the beginning of sterile soil. No cultural artifacts were recovered from this locus,
564
although we collected one soil sample. On average, this locus was an average of 15cm deep,
ending at an average of 146cm BD.
In order to expose and define the various colors of clay we encountered in quad 3, we
excavated one more locus—764—only in the northern amplification of the unit. Locus 764
contained brown (7.5YR 4/4) sandy clay which was uniform and compact throughout the quad.
This level also had several colors of clay including 7.5YR 2.5/1 (black), and 5YR 3/8 (yellowish
red).
Based on the lack of material evidence found in the preceding levels, we decided to end
the excavation of Quad 3 at this point.
Quad 5
Quad 5 was located in the center of the unit, southeast of Quad 1 and southwest of Quad
3. We began with locus 801, which was characterized as a dark brown (7.5YR 3/2) mediumgrained sand, which was loose and very poorly distributed throughout quad 5. Locus 801had an
abundance of roots, and some cultural material. We collected one bag of ceramic, one soil
sample, and one special find, which we called “hallazgo 1.” This special find was a metal cross
(see drawing) found in the southeast corner of the quad at 122cm BD. On average, locus 801 was
15cm thickness, and ended at a depth of 132cm BD.
Locus 802 was located directly below locus 801, and was marked by a natural change in
soil matrix. Locus 802 was a uniform, semi-compact brownish-black (7.5YR 3/2) mediumgrained sand. This locus had less cultural material, and we collected only one bag of ceramic and
one bag of lithic, along with one soil sample. The average depth of this locus was 9cm, ending at
an average of 141cm BD.
565
Locus 803 was directly below locus 802, and was changed due to a natural soil change.
The matrix in locus 803 was similar to the matrices in 753 and 763, characterized by large
chunks of clay and the appearance of sterile earth. Locus 803 was a compact, uniform sandy clay
which was brownish black (7.5YR 3/2) in color. The presence of cultural material here was
much less than in locus 802, and we recovered only one fragment of ceramic and one soil
sample. The average depth of locus 803 was 5cm, ending at an average depth of 147cm BD
across the quad. Since we recovered only a small amount of cultural material from locus 803, we
decided to end excavation of this quad with this locus.
Quad 7
Quad 7 is located in the southwest portion of Unit 1, sharing its southern border with the
northern border of Quad 10 (see drawings). In total, quad 7 had four contexts among natural
strata and arbitrary levels, which were identified as loci. The excavation team tried to follow
each natural level as completely and faithfully as possible. Locus 791 represented the quad’s
natural topsoil and consisted of a loose medium-grained sand, distributed throughout the unit and
brown (7.5YR 4/3) in color. We extended the western border of this unit to the western wall in
order to define the architecture of the base of the structure. From the northwest corner of the unit
to the western wall measured 0.76m, and from the southwest corner of the unit to the western
wall measured 0.77m. This locus had dispersed fragments of ceramic in the loose soil, with many
roots and some gravel. We also recovered a small amount of lithic material. In all, we recovered
two bags of ceramic, two bags of lithic, one bag of animal bone, and one soil sample. The
average depth of locus 791 was 11cm, and the locus ended at a depth of average 128cm across
the quad.
566
Locus 792 was located directly below locus 791, and was a natural level, dull reddish
brown (5YR 4/3) in color, with the soil a more uniform, compact medium-grained sand. At the
bottom of this locus, the sterile earth began to appear. In the western amplification, we defined a
special “hallazgo 1,” defined by fragments of ceramic associated with bone and carbon very
close to the wall. In total, we collected three bags of ceramic, one bag of lithic, two bags of
animal bone, one sample of carbon, and two soil samples. The elevated amount of cultural
material in locus 792 suggests that it was some sort of activity area, indicated by the ceramic
associated with carbon and animal bone. The average depth of this locus was 8cm, and the locus
ended at an average of 136cm across the quad and the amplification.
Locus 793 was excavated directly below locus 792 and was a natural level. Here, the soil
matrix was also more compact, brown (10YR 4/3) in color, and a uniform sand of medium grain.
Here, there was much less cultural material, with sterile soil coming through at the end marked
by large formations of colorful clay and very hard soil. In total, we collected two bags of
ceramic, one bag of lithic, one bag of animal bone, and one soil sample. On average, this locus
was 13cm, and the locus ended at an average depth of 148cm across the quad.
Locus 794 was the final locus in this quad, located directly below locus 793, and was a
natural level. The soil here was a sandy clay, brown (7.5YR 4/3) in color, distributed uniformly
throughout the quad. We only found one fragment of ceramic in this locus, and the tooth of one
animal, and we also took a soil sample. At the end of this level, we reached sterile soil, which
was nearly impossible to excavate, so we stopped the excavation of the quad at this point. The
average depth of locus 794 was 7cm, and the locus ended at an average of 156cm distributed
across the quad. Because of the extremely compact sterile soil and lack of cultural artifacts, we
decided to stop excavation of this quad with locus 794.
567
Quad 10
Quad 10 was located in the southwestern corner of the grid in EST-002, directly south of
Quad 7 and west of the accessway. In total, Quad 10 had six contexts among natural strata and
arbitrary levels, which were identified as loci. The excavation team tried to follow each natural
level as completely and faithfully as possible. Locus 771 represented the quad’s natural topsoil
and consisted of a loose, dark reddish brown (5YR 3/3) medium-grained sand, distributed
uniformly across the quad. The soil matrix in this locus had some roots and cultural evidence. In
order to define the masonry at the base of the structure and to follow a higher concentration of
cultural evidence found on the surface of the same corner, we amplified this quad to the west and
to the south. From the northwest corner of the unit to the western wall, the distance was 0.73m.
From the southwest corner of the unit to the western wall, the distance was 0.40m. From the
southwest corner of the unit to the southern wall, the distance was 0.60m. From the southeastern
corner of the unit to the southern wall, the distance was 1.10m. In total, we recovered three bags
of ceramic, two bags of lithic, one bag of animal bone, and one soil sample from locus 771. The
average depth of this locus was 15cm, and the average ending depth of locus 771 was 128cm
BD.
Locus 772 was excavated directly below locus 771, and represented a natural level and
soil change. The texture of the soil was semi-compact and dark reddish brown in color (5YR
3/2), medium-grained sand uniformly distributed. In the northern portion of the western
amplification, we found a special hallazgo—here, there were fragments of ceramic and
intentional depositions of ceramic, lithic, and animal bone. In total, we collected two bags of
ceramic, one bag of lithic, one bag of animal bone, and one soil sample from locus 772. The
average depth from locus 772 was 7cm, and the average ending depth was 134cm BD.
568
Locus 773 was located directly below locus 772, and represented a natural level change
with soil more compact, dark reddish brown (5YR 3/3) in color, and was a medium-grained sand
with uniform distribution across the quad. In this level, we recovered three bags of ceramic, three
bags of lithic, two bags of animal bone, two carbon samples, and one soil sample. In the southern
amplification, the beginning of an activity area was present, including a higher concentration of
ceramic and bone, which appear to have been deliberately placed. The average depth of locus
773 was 13cm, and the average ending depth below datum was 147cm.
Locus 774 was located below 773, but only in the southern amplification of the
quad. This locus represented a special context, or “hallazgo especial,” in which we found large
fragments of animal bone in the same level of fragments of ceramic, with a small piece of
eggshell. This locus indicates an activity region, perhaps indicating food processing or
butchering. The soil matrix in this locus was compact, medium-grained and uniform sand which
was dark reddish brown (5YR 3/2). In locus 774, we collected one bag of ceramic, one bag of
animal bone, one soil sample, and one bag which contained the eggshell. This locus was an
average of 24cm in depth and ended at an average depth of 137cm BD.
Locus 775 was located below loci 773 and 774, and was excavated uniformly below
these two loci. Locus 775 was a natural level, with a soil matrix more compact and dark reddish
brown in color (5YR 3/3), a sandy clay uniformly distributed throughout the quad. In locus 775,
we collected three bags of ceramic, one bag of lithic, and two soil samples. The locus was an
average depth of 11cm, ending at an average of 158cm BD.
The final locus in Quad 10 was locus 776, which we excavated only in the southern
portion of the western amplification. This small cut was only 50cm x 65cm and was placed to
investigate if the cultural material continued below the cultural levels. In this calicata, we found
569
only one fragment of worked stone, and the soil was much more compact and dull reddish brown
(5YR 4/3), with more clay inclusions. We were not able to arrive at the base of the wall of
Structure 002, indicating that those who constructed the base of the building must have dug at
least one meter below the surface in order to place the massive stones. In locus 776, we collected
only one bag of lithic and one soil sample. The calicata was an average depth of 52cm and ended
at an average depth of 205cm BD. Locus 775 was the final locus in this quad, and the hard,
sterile soil began to emerge at the bottom of the locus.
Quad 12
The last quad excavated was quad 12, located in the southeastern corner of EST-002. In
total, Quad 12 had four contexts among natural strata and arbitrary levels, which were identified
as loci. The excavation team tried to follow each natural level as completely and faithfully as
possible. Locus 781 represented the quad’s natural topsoil and consisted of a loose, poorly
distributed medium-grained sand, dark reddish brown (5YR 3/3) in color. Locus 781 had
numerous plant roots, and cultural evidence including ceramic and lithic. In all, we collected one
bag of ceramic, one bag of lithic, and one soil sample. The locus was an average depth of 20cm,
ending at an average level of 131cm BD.
Locus 782 was excavated directly below locus 781, and was marked by a natural change
in the soil to a more compact, uniform, sand of medium grain, brown in color (7.5YR 4/3). This
locus had less roots and less cultural artifacts, including one bag of ceramic, one bag of lithic,
and one soil sample. On average, locus 782 had a depth of 13cm and ended at an average depth
of 144cm BD.
570
Locus 783 was excavated directly below locus 782, and represented a natural level
marked by a soil change to a more compact, uniform medium-grained sand, brown (7.5YR 4/2).
Locus 783 had less evidence of cultural materials, with one bag of ceramic and one soil sample.
The locus was an average of 8cm in depth, ending at an average depth of 151cm BD.
The final locus of Quad 12 was 784, which was excavated directly below locus 783 and
represented a natural level. It consisted of a soil matrix which was a medium-grained, very
compact and uniform sand, with clay inclusions and brownish black (7.5YR 3/2) in color. The
average depth of locus 784 was 7cm in depth and ended at an average depth of 159cm BD. There
were few cultural artifacts in this level, and we collected only one bag of ceramic and one soil
sample.
After excavating quads 1, 3, 5, 7, 10, and 12, we terminated the excavation of Unit 1.
After drawing profiles, cuts, and plans, the unit was closed and backfilled with the sieved soil.
Unit 1 Level Groupings
Level 1: Locus 751, 761, 801, 791, 771, 781
Level 2: Locus 752, 762, 802, 792, 772, 782
Level 3: Locus 753, 763, 803, 793, 773, 774, 783
Level 4: Locus 764, 794, 775, 776, 784
571
Unit 2, Sector 1, EST-030
Location (WGS 1984): 656850E 8427955N
Dimensions: 8m x 4m (divided into 8 2m x 1m quads)
Declared Loci: 401, 402, 403
Unit 2 was located in Sector 1, the religious core of the site at Iglesiachayoq, the interior
of EST-030. EST-030 was part of a small patio group which included three rectangular structures
whose doors faced one another. This patio group was encircled by a low wall with an accessway
in the western portion. The excavation unit was strategically placed to encompass the maximum
amount of space in EST-030, in order to learn about the domestic activities that were occurring
inside of this structure.
Summary of Quads and Loci
In total, we divided Unit 2 into eight separate quads in order to maintain a degree of
control in excavation, each quad measuring 2m x 1m. We numbered the quads from the
Northwest quad to the Northeast quad (1-2), and then moved south according to the way
language is read. Unlike Unit 1, Unit 2 utilized the same set of loci for all quads, and all quads
were excavated according to natural and arbitrary levels throughout the unit. Unit 2 had three
contexts among natural strata and arbitrary levels, which were identified as loci. The excavation
team tried to follow each natural level as completely and faithfully as possible.
Locus 401
Locus 401 represented the unit’s natural soil and consisted of a loose, poorly
homogeneous medium-grained sand which was dark reddish brown (5YR 2.5/2) in color. The
sand was mixed with lots of roots and some archaeological evidence including ceramic
fragments, obsidian, and some animal bone. In order to define the base architecture of EST-030,
572
we extended the northeast quad (Quad 2) to the wall, which was 0.57m from the northwest
corner of quad 2 to the wall, and 0.40m from the northeast corner of quad 2 to the wall. In the
center of the extension, a mano was uncovered at 98cm BD. In this locus, we collected 26 total
bags: 11 bags of ceramic, 10 bags of lithic, one bag of animal bone, one bag of carbon, one soil
sample, and one clay sample. This level was an average of 13 cm thick, ending at an average
depth of 1.18m BD.
Locus 402
Directly below Locus 401 was Locus 402, which was a natural level. We started Locus
402 because of a change in soil matrix, such that Locus 402 consisted of a medium-grained sand
which was more uniform and compact than Locus 401. Locus 402 was dark reddish brown in
color (5YR 3/2) and contained less roots, ceramic, and lithic than in Locus 401. In Quads 1 and
2, there were many more large rocks which were clay-like. There was a large tree in quads 5 and
7, which prevented thorough excavation here. There was one special find in Locus 402, which
was a large grindstone 1.30m BD. In total, we collected 16 artifact samples—nine bags of
ceramic, three bags of lithic, two bags of animal bone, one soil sample, and the grindstone.
Locus 402 was an average of 19cm in depth, ending at an average of 1.34m BD throughout the
unit. We stopped excavating Locus 402 when we reached a very hard, thick clay level which
appeared to be sterile soil.
Locus 403
Locus 403 was a small, 50cm x 50cm cut which we placed in the southeastern corner of
the unit in order to make sure there was no more evidence of cultural material and that we had
indeed reached sterile soil. In Locus 403, there was no evidence of cultural material, the soil was
573
much more granular and compact, and very difficult to dig through, and there was a large
distribution of rocks of different sizes which appeared to correspond to sterile soil. The color of
the soil was dark reddish brown (5YR 3/2), and we collected only one soil sample from this
Locus. Locus 403 was an average of 6cm in depth and ended at an average of 1.56m BD across
the cut. We terminated the excavation of Unit 2 at this point since we had found no cultural
material and the soil was very difficult to excavate. After drawing the southern profile and a cut
and taking photos, the unit was closed and backfilled with the sieved soil.
Unit 2 Level Groupings
Level 1: Locus 401
Level 2: Locus 402
Level 3: Locus 403
574
Unit 3, Sector 1, EST-047
Location (WGS 1984): 656846E 8428016N
Dimensions: 4.5m x 4 m (divided into 4 2.25m x 2m quads)
Declared Loci: 251, 252, 253, 254, 255, 256
Unit 3 was located in Sector 1, the religious core of the site at Iglesiachayoq, the interior
of EST-047. EST-047 was an isolated circular structure situated within its own terrace, with its
accessway facing to the northeast. The excavation unit was strategically placed to encompass the
maximum amount of space in EST-047, in order to learn about the domestic activities that were
occurring inside of this structure.
Summary of Quads and Loci
In total, we divided Unit 3 into four separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.25m x 2m. We numbered the quads from the
Northwest quad to the Northeast quad (1-2), and from the Southwest quad to the Southeast quad
(3-4). Unit 3 utilized the same set of loci for all quads. Unit 3 had six contexts among natural
strata and arbitrary levels, which were identified as loci. The excavation team tried to follow
each natural level as completely and faithfully as possible.
Locus 251
Locus 251 was the first level, and represented the unit’s natural soil. Locus 251 was
characterized by the presence of roots, and a very loose, poorly homogeneous medium-grained
sand which was dark brown (7.5YR 3/2) in color. There were few cultural materials associated
with this locus, and most of the soil was affected by wall fall throughout the unit. In total, 9
artifact bags were collected: four ceramic bags, three lithic bags, one sample of soil, and one
sample of burned soil. To the south of quad four (the southeast corner), we placed an extension
575
of the unit in order to define the base of the wall. From the southwest corner of quad 4, this
measured 70cm to the southern wall. From the center of quad 4 to the southern wall, the distance
was 50cm. The southeastern corner of quad four (and the unit) abutted the southern wall. On
average, this locus was 6cm thick, ending at an average depth of 61cm BD throughout the unit.
Locus 252
We closed Locus 251 and opened the new Locus 252 when we reached a soil change.
The soil matrix in Locus 252 was a medium-grained sand that was moderately homogeneous and
semi-compact, and dark brown (7.5YR 3/2) in color. In this locus, we found some fragments of
ceramic and lithic, but the fragments are very small and there is almost no diagnostic material.
The soil is more compact and granular, and in quad 1 we uncovered small pieces of distributed
carbon. In quad 3, we encountered some fragments of animal bone. In total, we recovered 15
artifact bags from Locus 252: six bags of ceramic, 5 bags of lithic, 3 bags of animal bone, and
one soil sample. Locus 252 was an average of 10 cm in depth, and ended at an average depth of
71cm BD.
Locus 253
We closed Locus 252 and opened the new Locus 253 when we reached a soil change.
The soil matrix in Locus 253 was a medium-grained sand that was moderately uniform and more
compact than locus 252, of the color very dark grayish brown (10YR 3/2). Locus 253 was only
excavated in quads 1, 3, and 4 because quad 2 presented a different context in the same
horizontal level. In general, Locus 253 had less fragments of ceramic than locus 252, and small
nuclei of stone and carbon appeared. In quad 3, there was a slightly denser concentration of
fragments of ceramic and obsidian. We collected a total of 7 bags from Locus 253: four of
576
ceramic, two of lithic, and one soil sample. The locus ended at an average of 82 cm below datum
and was an average of 11cm total depth throughout the unit.
Locus 254
Locus 254 was in the same horizontal level as locus 253, but presented as a context
which was distinct between both Locus 252 and 253. Locus 254 was excavated only in quad 2
and measured 2.80m x 2.19m in size. This locus was characterized by the presence of carbon,
appearing sparsely in a small space. The soil was very wet (since there is not much sun that
reaches this quad), and as a result the carbon was in very poor condition. There was very little
ceramic. Our guess is that this context was the product of a hearth or a fire, since the cold is very
strong during certain seasons, and clearly affected this house. The soil was a medium-grained,
compact sand moderately uniform throughout the unit and was brown (10YR 4/3) in color. In
total, we collected 5 bags of artifacts: two of ceramic, one of animal one, one of carbon, and one
soil sample. Locus 254 ended at an average depth of 82cm below datum, and was an average of
10cm in depth.
Locus 255 and 256
Since we had reached a level where no cultural artifacts were being removed from the
soil, we decided to continue excavating only in quad 4, and so loci 255 and 256 were excavated
from this quad. Locus 255 was a natural level, with soil nearly the same color as locus 253 (very
dark grayish brown (10YR 3/2)). After excavating only two buckets of material, the color of the
soil began to change rapidly, and we changed to locus 256 to account for the change. Locus 255
was thus likely a part of Locus 253, located above. Locus 256 is thus the last locus in this unit.
The soil was a clay/sand that was moderately distributed and compact throughout the unit, of
577
dark yellowish brown color (10YR 4/4). Along with the color change, we encountered several
large stones which delimited the space where the clay was located. With the presence of the
delimited area, we hypothesized that this area was dedicated to artisanal usage of clay sources. In
total, we collected four bags from loci 255 and 256: one of ceramic, one of lithic, one of clay,
and one of soil. The locus ended at an average of 1.08m below datum, and was an average of
23cm in depth.
We terminated the excavation of Unit 3 at this point since we had found only one ceramic
sherd and the soil was filled with clay and hard to excavate. After drawing the southern wall of
the structure, the eastern profile, the western cut, and a plan view, and taking photos, the unit was
closed and backfilled with the sieved soil.
Unit 3 Level Groupings
Level 1: Locus 251
Level 2: Locus 252
Level 3: Locus 253 and 254
Level 4: Locus 255 and 256
578
Unit 4, Sector 1, EST-004
Location (WGS 1984): 656751E 8427938N
Dimensions: 5.5m x 3m (divided into 4 2.75m x 1.5m quads)
Declared Loci: 201, 202, 203, 204, 205, 206, 207, 208, 209, 210
Unit 4 was located in Sector 1, the religious core of the site at Iglesiachayoq, the interior
of EST-004. EST-004 was a quadrangular structure with squared edges located very near the
kuraka house and the church, with its entryway facing both structures (southeastern side of the
long axis of the structure). The excavation unit was strategically placed to encompass the
maximum amount of space in EST-004, in order to learn about the domestic activities that were
occurring inside of this structure.
Summary of Quads and Loci
In total, we divided Unit 4 into four separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.75m x 1m. We numbered the quads from the
Northwest quad to the Northeast quad (1-2) and from the Southwest quad to the Southeast quad
(3-4). Unit 4 utilized the same set of loci for all quads. Unit 4 had 10 contexts among natural
strata and arbitrary levels, which were identified as loci. The excavation team tried to follow
each natural level as completely and faithfully as possible.
Locus 201
Locus 201 was the first level, and represented the unit’s natural topsoil. However, we
soon encountered a change between quads 1 and 2 (north half of the unit), and so excavated
quads 3 and 4 separately (south side of the unit) as Locus 203. Locus 201 was characterized by
the presence of roots and was a uniform, loose medium-grained sand that was dark reddish
brown (5YR 3/2) in color. There were few cultural materials in this locus, and we collected three
579
bags of artifacts, two of ceramic and one of lithic. Locus 201 ended at an average of 134cm
below datum, and was an average of 12cm deep in the excavated quads.
Locus 252
While excavating Locus 251, we encountered a difference in soil composition and
cultural materials in quad 2. We designated this area as a special locus, defined as an area with
more concentration of carbon, bone, and ceramic. We considered this an activity area because of
the higher level of cultural material. Locus 202 was 80cm x 69cm in size, was more compact
than locus 201, and was a medium-grained sand that was dark reddish brown in color (5YR 3/2).
We collected a total of four artifact bags in this locus, including two bags of ceramic, one bag of
animal bone, and one soil sample. This lens of soil was an average of 6cm in depth and ended at
an average depth of 140cm below datum throughout the locus.
Locus 203
We closed Locus 202 and opened the new Locus 203 when we encountered a soil change.
Locus 203 was excavated in the northern portion of quads 3 and 4 and was distinctive due to a
large quantity of ceramic and artifacts. The color of the soil an the consistency also changed
from Locus 201. Locus 203 encompassed the entire width of the unit (3m) but was only 1.62m in
length, as we encountered another horizontal soil change further south in the unit. Locus 203
consisted of a loose, uniformly distributed medium-grained sand that was dark brown (7.5YR
3/2) in color. We collected 11 artifact bags from Locus 203, including 7 bags of ceramic, one bag
of lithic, one carbon sample, one soil sample, and a “special find” which was a button. The
button looked like a piruru but without the hole in the middle. In general, this locus was
characterized by a dense proportion of ceramic and was clearly an area used for food preparation.
580
Locus 203 was terminated at an average depth of 143cm below datum, and was an average depth
of 21cm in thickness.
Locus 204
We closed locus 203 and moved southward in Unit 4, where we soon discovered a floor
with artifacts on top which we designated as Locus 204. This locus took up the entire width of
quads 3 and 4, but measured 113 cm of length in quad 3 and 85cm of length in quad 4. We found
a huge amount of ceramic fragments in quad 4, especially concentrated in the southeast corner.
We also found carbon, a piruru, animal bone, and a sewing tool. In quad 4, we extended the unit
east until it reached the eastern wall in order to define the base architecture of the structure and
follow the locus outward. The doorway was located east of quad 4, so we extended the quad until
it reached the door. The extension was the length of quad 4 (2.75m) and was extended 16cm to
hit the wall. The soil matrix for Locus 204 was a compact, uniform medium-grained sand which
was dark brown (7.5YR 3/2) in color and likely represented a floor. We collected 10 total artifact
bags from Locus 204: four of ceramic, one of lithic, one of animal bone, one carbon sample, one
soil sample, and two special finds which included the piruru (hallazgo 1) and the tool to sew
(hallazgo 2). Locus 204 ended at an average of 138cm below datum and was an average of 17cm
in depth.
Locus 205
We closed locus 204 and opened the new locus 205 when we encountered a soil change,
such that the soil matrix in Locus 205 was a uniform, compact, fine ash which was black in color
(5YR 2.5/1). This locus was a special burned context which indicated a lens of intense burning
which contained several types of carbon, ceramic, and lithic. Locus 205 was located only in
581
quads 3 and 4, directly below locus 204, and was 3.16m in width and 2.75m in length. We
collected 10 artifact bags from Locus 205, 2 bags of ceramic, 1 bag of lithic, 6 bags of carbon (2
unidentified carbon, one of bone, one of corn, one of cloth, and one of wood), and one soil
sample. In general, this locus ended at an average depth of 148cm below datum, and was an
average of 15cm in depth throughout the southern portion of the unit.
Locus 206
We closed Locus 205 and opened the new Locus 206 when we reached a soil change.
The soil matrix in Locus 206 was a uniform, compact medium-grained sand with ash mixed in
that was a dark reddish brown (5YR 3/3) in color. We found a large concentration of carbon
here, but this locus is a natural level. We did not find a single ceramic or other artifact, but we
found plenty of carbon. This locus is located below the base of the wall. It is possible that those
who utilized this structure excavated below the base of the wall in order to place a more
concentrated hearth. Locus 206 was 1.85m by 1.50m in size. This locus ended at an average of
159cm below datum and was an average of 9cm in depth. In total, we collected two bags from
Locus 206—one carbon sample and one soil sample.
Locus 207
Locus 207 was an intruding locus which shared the same horizontal plane as locus 206.
Locus 207 was likely a hearth or area of intense burning with a high concentration of animal
bone. Locus 207 was located in the southwest corner of quad 3. We have a much higher
proportion of animal bone than in any other locus at the site. The animal bone appears to be from
a juvenile camelid because the joints are not fused. Locus 207 was 90cm by 60cm and consisted
of a loose, uniform, ash layer which was black (5YR 2.5/1) in color. We collected only one bag
582
from this locus, which consisted of animal bone. Locus 207 ended at an average of 158cm below
datum and was an average of 18cm in depth.
Locus 208
Locus 208 was excavated in quads 1 and 2 exclusively, and was the locus below Locus
201 and 202 (surface). Locus 208 was a natural level, and we found fragments of ceramic and
lithic, as well as animal bone. The soil matrix was compact, uniformly distributed and a mediumgrained sand which was brown (7.5YR 4/2) in color. The lithic fragments are of a type of stone
very rare in this area and seem to be knives used for cooking, and made of chert. I believe that
this entire house was used to prepare food on a large scale because the ceramic fragments are
enormous. We collected 6 bags of artifacts from this locus: 2 of ceramic, 2 of lithic, 1 of animal
bone, and 1 soil sample. In general, this locus ended at an average depth of 144cm below datum
and was an average of 7cm in depth.
Locus 209
We closed Locus 208 and opened Locus 209 upon reaching a soil change and a change in
artifact density, such that the soil matrix in 209 consisted of a compact, uniform medium-grained
sand that was dark brown (7.5YR 3/2) in color. Because we found ceramic and lithic in the locus
above, we excavated quads 1 and 2. But, upon finishing locus 209, we reached the base of the
wall and there were very few ceramic fragments. We collected three bags from this locus,
including 2 ceramic bags and one soil sample. In general, this locus ended at an average depth of
159cm below datum, and was an average of 15cm in depth.
Locus 210
583
At this point in the excavation of Unit 4, we had reached sterile soil in most of the unit. In
order to ensure that there were no more cultural levels, we placed a small 60cm by 70cm test pit
in the center of the unit, which had one corner in all four quads. We only collected one ceramic
fragment and one soil sample from Locus 210, and we found some evidence of carbon which
was likely residual from the loci above. The soil matrix of 210 was a compact, uniform, mediumgrained sand which was very dark brown (7.5YR 2.5/2) in color. This locus ended at an average
of 169cm below datum and was an average depth of 15cm.
We terminated the excavation of Unit 4 at this point since we had only found one ceramic
sherd and the soil matrix was difficult to excavate. After drawing the northern cut, western cut,
western profile, and eastern accessway, and taking photos, the unit was closed and backfilled
with sieved soil.
Unit 4 Level Groupings
Level 1: Locus 201
Level 2: Locus 202, 203, 204
Level 3: Locus 205, 208
Level 4: Locus 206, 207, 209
Level 5: Locus 210
584
Unit 5, Sector 2, EST-175
Location (WGS 1984): 657031E 8428069N
Dimensions: 4.5m x 3 m (divided into 4 2.25 x 1.5m quads)
Declared Loci: 601, 602, 603
Unit 5 was located in Sector 2, located northeast of Sector 1, and was the interior of EST175. EST-175 was an isolated quadrangular structure with rounded corners that had its own
terrace, with its accessway facing to the northwest. The excavation unit was strategically placed
to encompass the maximum amount of space in EST-175, in order to learn about the domestic
activities that were occurring inside of this structure.
Summary of Quads and Loci
In total, we divided Unit 5 into four separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.25m x 1.5m. We numbered the quads from the
Northeast quad to the Southeast quad (1-2), and from the Northwest quad to the Southwest quad
(3-4). Unit 5 utilized the same set of loci for all quads. Unit 5 had three contexts among natural
strata and arbitrary levels, which were identified as loci. The excavation team tried to follow
each natural level as completely and as faithfully as possible.
Locus 601
Locus 601 was the first level, and represented the unit’s natural soil. Locus 601 was
characterized by the presence of roots, and a very loose, poorly homogeneous medium-grained
sand that was dark reddish brown (5YR 3/3) in color. The archaeological evidence was
fragments of ceramic and lithic and the presence of dispersed carbon in the four quads. In total,
we collected 9 bags of artifacts from locus 601 including 6 bags of ceramic, one bag of lithic,
one metal artifact, and one soil sample. The metal artifact was a flattened copper rectangle with a
585
small hook at one end, which could have been used as a pendant or part of a necklace. In general,
the locus ended at a depth of 100cm below datum and was an average depth of 12 cm throughout
the unit.
Locus 602
We closed Locus 601 and opened the new Locus 602 when we reached a soil change.
The soil matrix in Locus 602 was a poorly consolidated and moderately homogeneous finegrained sand which was dark brown (7.5YR 3/2) in color. In this locus, the archaeological
evidence was fragments of ceramic, lithic, bone, and small pieces of lithic located in quads 1 and
2 in a dispersed manner. These stones were distributed almost like a layer of gravel, and were
approximately golf-ball sized. We placed an extension to the north of quad 1 which measured
1.5m in width and 65cm from the northwest corner of quad 1 to the northern wall and 20cm from
the northeast corner of quad 2 to the northern wall. In total, we collected 16 bags from Locus
602: 6 bags of ceramic, 1 bag of lithic, 3 bags of metal (likely natural stone or hematite), 5 bags
of animal bone, and one soil sample. In general, Locus 602 ended at an average depth of 113cm
below datum throughout the unit and was an average depth of 13cm throughout the unit.
Locus 603
We closed Locus 602 and opened the new Locus 603 when we reached a soil change.
This natural level was characterized by a loose sand with minimal quantity of cultural material
like ceramic and lithic. In quads 1 and 2, there was a large amount of loose stones. In general, the
soil matrix in Locus 203 was poorly consolidated and moderately homogeneous, and was brown
in color (7.5YR 4/3). In total, we collected four bags from Locus 603, including 2 bags of
586
ceramic, one bag of lithic, and one soil sample. The average closing depth was 125cm below
datum, and the average depth of the locus was 12cm.
We terminated the excavation of Unit 5 at this point since we had found only one ceramic
sherd and the soil was filled with a matrix that was hard to excavate. After drawing the plan
view, northern profile, two cuts, and taking photos, the unit was closed and backfilled with the
sieved soil. In general, this unit only had three basic levels—the first was the surface, the second
was the cultural level and possible floor (rocks), and the third was below the floor and possibly
sterile soil. There were no distinctive contexts such as hearths or activity areas, and in general it
appears that this unit was only occupied for a very short time.
Unit 5 Level Groupings
Level 1: Locus 601
Level 2: Locus 602
Level 3: Locus 603
587
Unit 6, Sector 2, EST-168
Location (WGS 1984): 657006E 8428110N
Dimensions: 4.5m x 3 m (divided into 4 2.25 x 1.5m quads)
Declared Loci: 501, 502, 503, 504
Unit 6 was located in Sector 2, located northeast of Sector 1, and was the inerior of EST168. EST-168 was an isolated quadrangular structure with rounded corners that had its own
terrace, with its entryway facing to the northwest. The excavation unit was strategically placed to
encompass the maximum amount of space in EST-168, in order to learn about the domestic
activities that were occurring inside of this structure.
Summary of Quads and Loci
In total, we divided Unit 6 into four separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.25m x 1.5m. We numbered the quads from the
Northeast quad to the Southeast quad (1-2), and from the Northwest quad to the Southwest quad
(3-4). Unit 6 utilized the same set of loci for all quads. Unit 6 had four contexts among natural
strata and arbitrary levels, which were identified as loci. The excavation team tried to follow
each natural level as completely and as faithfully as possible.
Locus 501
Locus 501 was the first level, and represented the unit’s natural soil. Locus 501 was
characterized by the presence of abundant roots. There was regular ceramic throughout Locus
501, resulting in lots of borders with various sizes. There was also lithic, one fragment of metal
and human bone, probably of camelid. Along the southern side of the unit where quads 2 and 4
are located, we placed an amplification to be able to define the base of the wall, and in this
extension we collected ceramic, carbon, and part of a fragmented batan. The amplification was 3
588
meters in width and 50cm in length to reach the southern wall. In general, the soil matrix was a
medium-grained, loose sand that was moderately homogeneous and dark brown in color (7.5YR
3/2). We collected 17 artifact bags, including 7 bags of ceramic, 5 bags of lithic, 1 metal artifact
(a small pendant), 3 bags of animal bone, and one soil sample. In general, Locus 501 ended at a
depth of 112cm below datum and was an average depth of 6cm.
Locus 502
We closed Locus 501 and opened Locus 502 when we arrived at a natural soil change,
such that the soil in 502 was more compact, but still a moderately distributed medium-grained
sand that was dark brown (7.5YR 3/2) in color. The associated materials were fragments of
ceramic, obsidian, and animal bone. In quads 3 and 4, we recovered large pieces of dispersed
carbon. In total, we collected 12 artifact bags, including 5 bags of ceramic, 3 bags of lithic, 2
bags of animal bone, one soil sample, and one piece of wood. This level ended an average depth
of 118cm below datum, and was an average of 6cm in depth.
Locus 503
We closed locus 502 and opened Locus 503 when we encountered a soil change, such
that the frequency of cultural material decreased and the soil became more uniform in
consistency. The soil matrix of Locus 503 was a uniformly distributed medium-grained sand that
was compact and brown (7.5YR 4/3) in color. Besides the ceramic, we didn’t find many other
cultural materials with the exception of carbon. In quads 2 and 4, irregular-sized stones began to
appear which were likely representative of sterile soil. In total, we collected 8 bags from this
locus, including 5 bags of ceramic, 2 bags of lithic, and 1 soil sample. Locus 503 ended at an
average depth of 126cm below datum and was an average of 8cm in depth. We decided to only
589
continue the excavation in quad 3, since the most material was emerging from this quad and
since we had hit sterile soil in quads 1, 2, and 4.
Locus 504
Locus 504 was the last locus excavated in Unit 6, and it was only located in quad 3. The
soil matrix in locus 504 was clearly sterile soil, a very compact, moderately-distributed mediumgrained sand that was dark reddish brown (5YR 3/2) in color. There were large hunks of clay and
thick soil, but we did not collect a single cultural artifact. In total, we only collected one bag from
this Locus, and it was a soil sample. Locus 504 ended at an average depth of 148cm below datum
and was an average of 19 cm in depth.
We terminated the excavation of Unit 5 at this point since we had found no ceramic sherds
and the soil was filled with a matrix that was hard to excavate. After drawing a plan view, the
southern profile, the eastern cut, the northern cut, and the masonry which incorporated large
boulders, we took photos and terminated the unit. The unit was closed and backfilled with sieved
soil. In general, this unit only had three basic levels, with the fourth level consisting purely of
sterile soil. There were no distinctive contexts such as hearths or activity areas, and in general it
appears that this unit was only occupied for a very short time.
Unit 6 Level Groupings
Level 1: Locus 501
Level 2: Locus 502
Level 3: Locus 503 and 504
590
Unit 7, Sector 2, EST-147
Location (WGS 1984): 656957E 8428162N
Dimensions: 4.5m x 3m (divided into 4 2.25m x 1.5m quads)
Declared Loci: 451, 452, 453
Unit 7 was located in Sector 2, located northeast of Sector 1, and was the interior of EST147. EST-147 was an isolated quadrangular structure with rounded corners that had its own
terrace, with is accessway facing to the east. EST-147 is the structure with the best view of the
valley and roadways to the north, and likely had the greatest viewshed of any of the structures at
Iglesiachayoq. The excavation unit was strategically placed to encompass the maximum amount
of space in EST-147, in order to learn about the domestic activities that were occurring inside of
this structure.
Summary of Quads and Loci
In total, we divided Unit 7 into four separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.25m x 1.5m. We numbered the quads from the
northwest to the northeast (1-2) and from the southwest to the southeast (3-4). Unit 7 utilized the
same set of loci for all quads. Unit 7 had three contexts among natural strata and arbitrary levels,
which were identified as loci. The excavation team tried to follow each natural level as
completely and as faithfully as possible.
Locus 451
Locus 451 was the first level, and represented the unit’s natural soil. Locus 451 was
characterized by loose, uniformly distributed medium-grained sand that was dark brown (7.5YR
3/2) in color. The archaeological material was ceramic, obsidian, burned bone, and a piece of
metal which likely was used as a knife. There are many types of borders and bases in this
591
primary level. Since we had such a dense array of ceramic in the top level, we added two very
large extensions to define the base of the wall and encompass the volume of the unit. To the east,
we extended the unit by 3m in width and 22cm from the northeast corner to the wall, 38cm from
the center of the unit to the entryway, and 16cm from the southeast corner to the wall. In the
western side of the unit, we extended 3m in width and 58cm from the northwest corner to the
wall, 85cm from the center to the wall, and 17 cm from the southwest corner to the western wall.
In total, we collected 26 bags of artifacts from Locus 451, including 13 bags of ceramic, 6 bags
of lithic, 1 metal artifact (knife), 4 bags of animal bone, one bag of carbon, and one soil sample.
Locus 451 ended at an average of 90cm below datum and was an average of 10cm in depth.
Locus 452
We closed Locus 451 when we encountered a soil change such that the soil matrix for
Locus 452 was a loose, medium-grained sand which was dark brown (7.5YR 3/2) in color and
uniformly distributed. We found much less fragments of ceramic but we also found animal
bones, likely from a cow or a horse. In total, we collected 11 samples from this locus, 5 bags of
ceramic, 3 bags of lithic, 2 bags of animal bone, and one soil sample. In general, the locus ended
at a depth of 100cm below datum, and was an average of 10cm in depth.
Locus 453
Locus 453 was the last locus excavated in Unit 7. We only excavated quads 1 and 4 from
this locus because they presented the most material in previous loci. There was a soil change for
Locus 453, in which the soil was a more compact, medium-grained sand which was moderately
distributed and very dark brown in color (7.5YR 2.5/2). We found almost no cultural materials,
only a few fragments of ceramic and lithic. In general, we collected two bags from this locus, and
592
they were both ceramic. Locus 453 ended at an average of 124cm below datum, and was an average
depth of 24cm throughout quads 1 and 4.
We terminated the excavation of Unit 7 at this point since we had found very little ceramic
sherds and the soil matrix was dense and difficult to excavate. After drawing the eastern profile
and the northern cut, we took photos and closed and backfilled the unit with sieved soil. In general,
this unit only had three basic levels—the first was the surface, the second was the cultural level,
and the third was nearly sterile soil. There were no distinctive contexts such as hearths or activity
areas, and in general it appears that this unit was only occupied for a very short time.
Unit 7 Level Groupings
Level 1: Locus 451
Level 2: Locus 452
Level 3: Locus 453
593
Unit 8, Sector 3, EST-292
Location (WGS 1984): 656826E 8427836N
Dimensions: 4m x 3.5m (divided into 4 2 x 1.75m quads)
Declared Loci: 51, 52, 53, 54, 55, 56, 57
Unit 8 was located in Sector 3, located south of Sector 1, and was the interior of EST292. EST-292 was an isolated ovoid structure which faced a small, semi-circular plaza and a
steep rock face, with its entryway facing to the northeast. The excavation unit was strategically
placed to encompass the maximum amount of space in EST-292, in order to learn about the
domestic activities that were occurring inside of this structure.
Summary of Quads and Loci
Unit 8 was the first unit excavated in the 2015 field season. As such, we dug much deeper
than in the rest of the units so that we would be sure we reached sterile soil and excavated all
cultural contexts. In total, we divided Unit 8 into four separate quads in order to maintain a
degree of control in excavation, each quad measuring 2m x 1.75m. We numbered the quads from
the Northwest quad to the Southwest quad (1-2), and from the Southwest quad to the Southeast
quad (3-4). Unit 8 utilized the same set of loci for all quads. Unit 8 had seven contexts among
natural strata and arbitrary levels, which were identified as loci. The excavation team tried to
follow each natural level as completely and as faithfully as possible.
Locus 51
Locus 51 was the first level, and represented the unit’s natural soil. Locus 51 was
characterized by the presence of roots, and was a uniformly-distributed fine-grained sand that
was loosely compacted and dark reddish brown (5YR 2.5/2) in color. The cultural materials
included ceramic and lithic, and what we thought initially was metal—what we thought was
594
metal was characterized by smoothed, polished black stones which are heavy and likely of
hematite. In actuality, these are not metal, but rather a polished natural substance. In total, we
collected 10 bags from this locus—5 of ceramic, 2 of lithic, 1 of metal, 1 soil sample, and 1
“special find” which was a batan with evidence of use. Locus 51 ended at an average depth of
69cm below datum and was an average of 16cm in depth.
Locus 52
Locus 51 was closed and locus 52 was opened due to a natural change in soil matrix, such
that Locus 52 was characterized by a moderately homogeneous medium-grained sand which was
more compact and dark reddish brown (5YR 3/3) in color. Locus 52 contained less roots and
organic material, as well as less presence of cultural material. There were several perfectly round
stones which could have been used as weapons or sling stones. With the low density of material,
this structure likely was not used as a domestic residence, or was only occupied for a short period
of time. We collected 6 artifact bags from this unit, including 4 bags of ceramic, 1 bag of lithic,
and one soil sample. In general, Locus 52 ended at an average depth of 77cm below datum, and
was an average of 9cm in depth throughout the unit.
Locus 53
Locus 53 was a natural level distinguished by a higher degree of compaction than Locus
52. After excavating all four quads in Loci 51, and 52, we decided to proceed excavation only in
Quad 2, which contained the most material. In Locus 53, there was almost no cultural material—
we only recovered 3 ceramic sherds. Locus 53 represented the beginning of sterile soil, and the
matrix was more compact and granular, a thick-grained sand that was poorly homogeneous
throughout the quad and dark reddish brown (5YR 3/3) in color. We only collected one bag from
595
this locus, which was of ceramic. In general, Locus 53 ended at an average depth of 86cm below
datum, and was an average of 8cm in depth.
Locus 54
Although we reached nearly-sterile soil in Locus 53, we decided to excavate at least
50cm more in depth to ensure that we had reached and excavated all cultural levels. Locus 54
was a natural level without artifacts, and was a compact moderately-homogeneous fine-grained
sand, which was dark brown in color (7.5YR 3/2). We collected two artifact bags from Locus 54,
including 1 bag of ceramic (only one fragment) and one soil sample. Locus 54 ended at an
average depth of 93cm below datum, and was an average of 7cm in depth.
Locus 55
Locus 55 was a natural level without any evidenced of cultural material. Locus 55
represents the definitive sterile soil, and was a uniformly distributed fine-grained sand, which
was compact and granular and dark brown (7.5YR 3/2) in color. We collected no artifact bags
from this locus. Locus 55 ended at an average depth of 101cm below datum, and was an average
of 8cm in depth.
Locus 56
Because Unit 8 was the first excavated at Iglesiachayoq, we dug much deeper than
normal in order to ensure there were no more cultural levels. Like Loci 53, 54, and 55, we
continued down only in quad 2. There were no cultural artifacts recovered from this locus, and
we only collected one bag which was a soil sample. Locus 56 ended at an average depth of
116cm below datum, and was an average depth of 12cm throughout the quad.
596
Locus 57
Locus 57 was the final locus excavated in Unit 8, and like the loci above, we only
excavated in quad 2 and found no cultural evidence. The soil matrix in Locus 57 was a uniformly
distributed fine-grained sand, which was compact and granular in nature and dark brown (7.5YR
3/2) in color. The locus ended at an average of 138cm below datum, and was an average of 23cm
in depth.
At this point in the excavation, we had not encountered a single fragment of cultural
material in over 50cm, so we decided to close the unit. After drawing the southern cut, eastern
cut, and eastern profile, we took photos and closed and backfilled Unit 8 with sieved soil. In
general, this unit only had three basic levels—the first was the surface, the second was a cultural
level, and the third was the end of that cultural level, heading toward sterile soil. There were no
distinctive contexts such as hearths or activity areas. Given the relative low percentage of
cultural materials, we hypothesize that this unit likely was in a structure that was not utilized for
domestic activities.
Unit 8 Level Groupings
Level 1: 51
Level 2: 52
Level 3: 53
Level 4 (sterile soil): 54, 55, 56, 57
597
Unit 9, Sector 3, EST-247
Location (WGS 1984): 656799E 8427790N
Dimensions: 4m x 2.5m (divided into 4 2m x 1.25m quads)
Declared Loci: 101, 102, 103, 104, 105, 106
Unit 9 was located in Sector 3, located south of Sector 1, and was the interior of EST247. EST-247 was an isolated quadrangular structure with rounded corners located on its own
terrace, with its accessway facing north. The excavation unit was strategically placed to
encompass the maximum amount of space in EST-247, in order to learn about the domestic
activities that were occurring inside of this structure.
Summary of Quads and Loci
In total, we divided Unit 9 into four separate quads in order to maintain a degree of
control in excavation, each quad measuring 2m x 1.25m. We numbered the quads from
Northwest to Northeast (1-2) and from the Southwest quad to the Southeast quad (3-4). Unit 9
utilized the same set of loci for all quads. Unit 9 had 6 contexts among natural strata and
arbitrary levels, which were identified as loci. The excavation team tried to follow each natural
level as completely and as faithfully as possible.
Locus 101
Locus 101 was the first level, and included the surface of the unit. On the surface after
clearing some of the vegetation, we encountered several in situ finds. First, we found two
vessels—one was complete and intact and appeared to be a bottle or vessel for liquid storage,
and the other vessel, while in pieces, was also nearly complete. These were both located in the
southern portion of the unit and the structure. WE also collected two grindstones likely used in
preparing food and cooking. Near the whole vessel, we found a piece of carbon which we
598
collected for possible radiocarbon dating in the future. The soil matrix in Locus 101 consisted of
coarse, small stones which were poorly distributed throughout the unit and were mediumcompaction and a very dark grayish brown (10YR 3/2) in color. In total, we collected 15 bags
from Locus 101, including 6 bags of ceramic (not including the whole vessels), 2 bags of lithic,
one bag of animal bone, one carbon sample, one soil sample, and four special finds (the two
whole vessels and the two grindstones). In general, Locus 101 ended at an average depth of
163cm below datum, and was an average depth of 8cm throughout the unit.
Locus 102
Locus 102 was a natural level defined by a change in soil matrix, such that Locus 102
was a semicompact, moderately homogeneous medium-grained sand which was very dark
grayish brown (10YR 3/2) in color. Locus 102 was more compact than locus 101, and more
granular. Some sections of Locus 102 seemed to be part of a floor, which is reasonable
considering all of the domestic artifacts we found in Locus 101. In Locus 102, we collected three
special finds. First, was a large, flat grindstone located in the Northeast of quad 2. The second
special find was another grindstone, but it was smaller and round and located in the south of
quad 2. The third special find was a small ceramic vessel, broken but located between the two
other special finds. In total, we collected 14 bags from Locus 102, including 5 bags of ceramic, 2
bags of lithic, 2 bags of animal bone, 1 carbon sample, 1 soil sample, and 3 special finds (the two
grindstones and the ceramic vessel). We also found one piruru in this locus. In general, Locus
102 ended at an average depth of 175cm below datum and was an average depth of 12cm
throughout the unit.
Locus 103
599
Locus 103 was only excavated in the southernmost quads, 3 and 4, and was a floor in the
south of the unit. There were places that were more compact and almost smoothed which appear
to be parts of a floor, although these were not uniform throughout the unit. There is a possible
hearth associated and some animal bone located in these quads. In general, the soil matrix in
Locus 103 was a compact and semicompact medium-grained sand which was poorly distributed
throughout the unit and brown (7.5YR 4/4) in color. We collected 6 bags of artifacts in this unit,
including 3 bags of ceramic, 1 bag of animal bone, 1 carbon sample, and 1 soil sample. The
carbon was associated to the hearth in quad 4. Locus 103 ended at an average depth of 176cm
below datum and was an average depth of 12cm throughout quads 3 and 4.
Locus 104
Locus 104 was excavated only in quad 4 and was designated by the presence of abundant
carbon located within a ring of irregular stones which possibly were part of the hearth. We also
found a fragment of camelid bone and fragments of ceramic. Additionally, the stones in the
southeast corner of the structure are black with evidence of smoke damage or cooking. Locus
104 was semicompact and uniformly homogeneous, with a medium-grained sand filled with
small stones, and was a very dark grayish brown (10YR 3/2) in color. In total, we collected 5
bags from Locus 104, including 1 bag of ceramic, 1 bag of animal bone, 2 carbon samples, and 1
soil sample. On average, Locus 104 ended at a depth of 178cm below datum, and was an average
of 9cm in depth.
Locus 105
At this point in the excavation, we were nearing sterile soil, so Locus 105 was excavated
only in quads 1, 3, and 4. Locus 105 was defined by a sparse quantity of ceramic and in some
600
quads, almost no evidence of cultural material. There are obsidian flakes and another piece that
looks likes quartz. To the south of quad 3, we placed an amplification of 8cm in order to define
the base of the wall. Locus 105 consisted of a very dark grayish brown (10YR 3/2) compact,
uniformly distributed medium-grained sand with inclusions of small pebbles. In total, we
collected 7 artifact bags, including 4 bags of ceramic, 2 bags of lithic, and one soil sample. Locus
105 ended at an average depth of 189cm below datum and was an average of 17cm in depth.
Locus 106
Locus 106 was the last locus excavated in Unit 9. This locus was excavated only in quad
3 and contained no archaeological material. Thus, in this locus, our main goal was to define the
base of the wall, which reached a depth of 30cm under the surface layer and was constructed
with large stones. After we excavated 25cm, the color of the soil changed to a black color, and
there were no more cultural artifacts. The soil matrix in Locus 106 was a very dark brown (10YR
2/2) and compact medium-grained sand which was uniformly distributed throughout quad 3. In
total, we collected 2 bags from this locus, including one bag of ceramic and one soil sample. The
locus ended at an average depth of 212cm below datum and was an average depth of 24cm
throughout the quad.
We terminated the excavation of Unit 9 at this point since we had found only one ceramic
sherd and the soil was filled with a dense, compact matrix. Throughout the unit, we drew a plan
view of Locus 101 and 102, a plan view of Locus 103, a plan view of Locus 104, a cut of the
eastern wall, and the western profile. After completing all drawings and taking photos, the unit
was closed and backfilled with the sieved soil.
Unit 9 Level Groupings
601
Level 1: Locus 101
Level 2: Locus 102
Level 3: Locus 103, 104
Level 4: Locus 105, 106
602
Unit 10, Sector 3, EST-212
Location (WGS 1984): 656756E 8427856N
Dimensions: 5m x 3.5m (divided into 4 2.5m x 1.75m quads)
Declared Loci: 151, 152, 153, 154, 155, 156, 157, 158, 159
Unit 10 was located in Sector 3, located south of Sector 1, and was the interior of EST212. EST-212 was an isolated quadrangular structure with rounded corners, located on the
hillside behind the church and kuraka house. EST-212 has an entryway which is almost complete
and facing west, and has a clear view of the church, plaza, and the activities which would have
happened there. The excavation unit was strategically placed t encompass the maximum amount
of space in EST-212, in order to learn about the domestic activities that were occurring inside of
this structure.
Summary of Quads and Loci
In total, we divided Unit 10 into four separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.5m x 1.75m. We numbered the quads from
Northwest to Northeast (1-2) and from the Southwest quad to the Southeast quad (3-4). Unit 10
utilized the same set of loci for all quads. Unit 10 had 9 contexts among natural strata and
arbitrary levels, which were identified as loci. The excavation team tried to follow each natural
level as completely and as faithfully as possible.
Locus 151
Locus 151 was the first level, and included the surface of the unit. We immediately added
an amplification to the southeast corner of the unit. To the south, we increased the unit 1m x
0.55m until it reached the wall. To the east, we increased the unit 1m x 0.55m so that it also
reached the wall. We placed this amplification in order to be able to define the bottom of the wall
603
and because we immediately encountered an activity area which had bone, carbon, and ceramic,
which we excavated as locus 153. The majority of Locus 151 was the surface, but the activity
area in quad 4 may have been used as a hearth. In general, the soil matrix in Locus 151 was a
loose, medium-grained sand which was moderately homogeneous throughout the locus and very
dark brown (7.5YR 2.5/3) in color. In total, we collected 15 bags from Locus 151, including 9
bags of ceramic, 4 bags of lithic, 1 soil sample, and one “special find,” which was a colonialperiod caret head nail in the northeast portion of quad 2. In general, Locus 151 ended at an
average depth of 54cm below datum, and was an average of 7cm in depth.
Locus 152
Locus 152 was located just below locus 151, and was marked by a change in soil matrix
such that the soil in this locus was much more compact and uniformly distributed, a mediumgrained sand that was dark brown (7.5YR ¾) in color. There were many fragments of ceramic
and obsidian, and the obsidian was especially concentrated in quad 1. In total, we collected 11
bags of artifacts from Locus 152, including 5 bags of ceramic, 4 bags of lithic, one bag of animal
bone, and one clay sample. Locus 152 ended at an average depth of 62cm below datum, and was
an average depth of 15cm throughout the unit.
Locus 153
Locus 153 was a special context, which appeared to be an area of activity denoted by the
presence of carbon associated with a camelid mandible and fragments of ceramic. Given the
carbon, animal remains, and the darkened stones in this corner (southeast), we believe that this
area was a hearth or food preparation area. Locus 153 was excavated only in quad 3 and the
amplifications, and was 1.36m x 1.46m in size. In general, the soil matrix in this locus was a
604
compact, granular, homogeneous medium-grained sand with carbon inclusions that was dark
brown (7.5YR ¾) in color. We collected 5 artifact bags from this locus, including 2 bags of
ceramic, one animal bone, one carbon sample, and one soil sample. In general, Locus 153 ended
at an average depth of 57cm below datum, and was an average of 11cm in depth.
Locus 154
Locus 154 was fully present in quads 1 and 3, but shared its horizontal provenance with
Locus 155 in quads 2 and 4. Locus 154 had less ceramic and obsidian material than the locus
above (152), and the consistency of the soil was compact with areas that were looser. The Locus
154 soil matrix was a medium-grained sand which was dark brown (7.5YR 3/2) in color. We
collected a total of 7 artifact bags from this locus, including 4 ceramic bags, 2 lithic bags, and
one soil sample. In general, Locus 154 ended at an average depth of 74cm below datum, and was
an average thickness of 13cm.
Locus 155
Locus 155 was at the same horizontal level as Locus 154, but was located only in quads 2
and 4 in the eastern side of the unit. Locus 155 appeared to be a floor or area of compaction, with
wallfall on top of it. The soil matrix in 155 was very compact and a medium-grained, uniformly
distributed brown (7.5YR 4/3) sand. We collected 7 bags from this locus, including 3 bags of
ceramic, 2 bags of lithic, 1 soil sample, and one “special find,” which was a caret-head early
colonial period iron nail. The presence of this nail in the lower levels of the unit suggests that
this structure was occupied into the Early Colonial Period, particularly since it was associated
with fragments of local ceramic. Locus 155 ended at an average depth of 64cm below datum and
was an average thickness of 8cm.
605
Locus 156
Locus 156 was a natural level that was excavated through all four quads. There was much
less cultural material, only some ceramic and lithic, and the soil matrix was a compact, uniformly
distributed medium-grained sand with small pebble inclusions which was dark brown (7.5YR
3/2) in color. In total, we collected 9 bags of material from this locus, including 4 bags of
ceramic, 4 bags of lithic, and one soil sample. Locus 156 ended at an average depth of 83cm
below datum and was an average thickness of 12cm
Locus 157
Because we had a dearth of materials in Locus 156, we decided to continue excavating
only in quads 3 and 4, where most of the material was coming from. We found no cultural
materials in this locus, but there was a soil change such that Locus 157 was compact and dark
reddish brown (5YR 3/5) in color, and was a uniformly distributed medium-grained sand. We
collected only one soil sample from Locus 157. Locus 157 ended at an average of 84cm below
datum and was an average thickness of 8cm.
Locus 158
Locus 158 was excavated only in quad 4, to ensure that we recovered all cultural
materials. The soil matrix in Locus 158 was a very compact and uniform granular mediumgrained sand that was dark reddish gray (5YR 4/2) in color. We only collected one soil sample
from this locus, and no cultural materials, In general, this locus ended at an average depth of
99cm below datum, and was an average thickness of 12cm.
Locus 159
606
Locus 159 was the last locus excavated in Unit 10, and was marked by a soil change such
that Locus 159 was a very compact, very uniformly distributed medium-grained sand which was
dark reddish brown (5YR 3/3) in color. There was no material culture here, and the matrix was
very difficult to excavate, leading us to conclude that we had reached sterile soil. Locus 159 was
excavated only in quad 4 and ended at an average depth of 108cm below datum and was an
average thickness of 10cm.
We terminated the excavation of Unit 10 at this point since we had found no ceramic
sherds in the previous two loci and the soil was a dense, compact matrix. Throughout the unit,
we drew the floor in locus 155, the southern profile, the eastern cut, the southern cut, a final
excavation plan view, the front view of the masonry style, the overall plan view, and the northern
cut. Finally, after taking lots of photos, we backfilled the unit with sieved soil.
Unit 10 Level Groupings
Level 1: Locus 151
Level 2: Locus 152, 153
Level 3: Locus 154, 155
Level 4: Locus 156
Level 5 (sterile soil): Locus 157, 158, 159
607
Unit 11, Sector 1, PAT-002
Location (WGS 1984): 656769E 8427938N
Dimensions: 2 m x 2 m
Declared Loci: 851, 852, 853, 854
Unit 11 was located in Sector 1, the religious core of the site at Iglesiachayoq, south of
the entrance to EST-002, the “casa de kuraka.” The excavation unit was strategically placed in
the patio of EST-002, in order to learn about domestic or non-domestic activities that were
occurring in the patio outside of this structure. The unit was located one meter southeast of the
entrance.
Summary of Loci
In total, Unit 11 had four contexts among natural strata and arbitrary levels, which were
identified as loci. The excavation team tried to follow each natural level as completely and
faithfully as possible. Locus 851 represented the unit’s natural topsoil and consisted of a compact
brown (10YR 4/3) medium grained sand, with gravel inclusions. In the surface, there were
ceramic fragments, one fragment of bone and one of obsidian. The topsoil of this zone may have
been used for cultivation, which indicates that the cultural material found in this level is likely
out of context. The soil in this locus was uniform throughout the unit, without any other specific,
unique contexts. This level was an average of 12 cm thick, ending at 129 cm BD throughout the
unit. Four artifact bags were collected—one of ceramic, one of lithic, one of animal bone, and
one soil sample.
Directly below locus 851 was locus 852, which was a natural level defined by a soil
change. Locus 852 consisted of compact, dark brown (7.5YR 4/3) medium grained sand of
608
uniform homogeneity. There was also evidence of gravel and small stones mixed with this level,
with significantly less plant material than locus 851. There was less evidence of cultural material,
especially marked by a decrease in density of both ceramic and lithic. Locus 852 was an average
of 13 cm in depth, and ended at average 142 cm BD throughout the unit. There were three
artifact bags initially collected from Locus 852, one of ceramic, one of lithic, and one soil
sample.
Locus 853 was uncovered directly below locus 852, and was also uniformly excavated
throughout the unit. Locus 853 was a natural level, marked by a soil matrix of a compacted dark
reddish brown (5YR 3/3) sand of medium grain and a much lower density of cultural material.
There were small inclusions of gravel, as in the previous two loci, but in 853, we collected only
one fragment of ceramic and one of lithic—along with the soil sample there were three artifact
bags total. This locus was an average of 11 cm deep, ending at 153 cm BD across the unit.
Based on the lack of material evidence found in the preceding levels, we decided to
continue excavating in only the southeast 1m x 1m square of the unit, in order to save time. This
1m x 1m was excavated as arbitrary locus 854, and the soil matrix was a very compact, very dark
brown (7.5YR 2.5/3) medium-grained sand. We found no artifacts or evidence of human
occupation, and we guess that this locus is the natural soil. We collected one soil sample, and the
locus was an average depth of 33 cm. We terminated the excavation of Unit 11 at this point, 186
cm BD in the southeast corner, and 153 cm BD across the other three quadrants. After drawing
the eastern profile and two cuts and taking photos, the unit was closed and backfilled with the
sieved soil.
Unit 11 Level Groupings
609
Level 1: Locus 851
Level 2: Locus 852
Level 3: Locus 853 and 854
610
Unit 12, Sector 1, PAT-030
Location (WGS 1984): 656843E 8427955N
Dimensions: 2 m x 2 m
Declared Loci: 651, 652, 653, 654
Unit 12 was located in Sector 1, the religious core of the site at Iglesiachayoq, west of the
entrance to EST-030, one of three quadrangular structures with squared corners which shared an
enclosed patio area. The three structures in this enclosed area are of a greater size and masonry
quality than those in other sections of the site, and were clearly separated in a manner which
suggests that entrance was restricted. The excavation unit was strategically placed in the patio of
EST-030, in order to learn about domestic or non-domestic activities that were occurring in the
shared patio outside of these three structures. The unit was located one meter west of the
entrance.
Summary of Loci
In total, Unit 12 had four contexts among natural strata and arbitrary levels, which were
identified as loci. The excavation team tried to follow each natural level as completely and
faithfully as possible. Locus 651 represented the unit’s natural topsoil and consisted of a loose,
moderately homogeneous brown (7.5YR 4/4) medium grained sand. This initial level contained
some ceramic, but no evidence of lithics, and contained plant material. In a small portion of the
eastern side of the unit, small fragments of carbon appeared. The soil in this locus was uniform
throughout the unit, without any other specific, unique contexts. This level was an average of 9
cm thick, beginning at an average of 58 cm below datum, and ending at an average of 67 cm
below datum. Two artifact bags were collected—one of ceramic and one soil sample.
611
Directly below locus 651 was locus 652, which was a natural level defined by a soil
change. Locus 652 consisted of compact, uniform dark brown (7.5YR 3/2) medium grained sand.
There was almost no archaeological material in this locus, with the exception of the intrusive
locus 653 (described below). There was significantly less plant material than locus 651. Locus
651 was an average of 17 cm in depth, beginning at an average of 67 cm below datum and
ending at an average of 84 cm below datum. Two artifact bags were initially collected from
Locus 652, one of ceramic and one soil sample.
Locus 653 was an intrusive locus which began slightly deeper than locus 652, but was
only located in the eastern portion of the unit. Locus 653 was separated as a separate locus
representing a feature, because it contained an abundance of carbon in a circular form. There are
two possible interpretations for this feature—it could have been a recent burning event since it
was rather close to the surface. Or, it could be the product of the extended burning of the huacas
described by Albornoz. The soil matrix in this locus was a loose, poor homogeneity black (10YR
2.5/1) fine-grained sand. This locus began at an average depth of 75 cm below datum, and ended
at an average depth of 82 cm below datum, for an overall depth of 7 cm. There were two artifact
bags collected from this locus, one of carbon and one soil sample.
Locus 654 was located below loci 652 and 653 and was a natural level. No archaeological
material was recovered from this locus, and sterile soil matrix marked by dense rock quickly
appeared in the unit. This locus was a uniformly compact very dark grayish brown (10YR 3/2)
medium-grained sand. Locus 654 began at an average depth of 84 cm below datum and ended at
an average depth of 95 cm, for an average total depth of 11 cm. Only one artifact bag was
collected, and this was a soil sample. We terminated the excavation of Unit 12 at this point. After
612
drawing the eastern profile and two cuts and taking aerial and normal photos, the unit was closed
and backfilled with the sieved soil.
Unit 12 Level Groupings
Level 1: Locus 651
Level 2: Locus 652 and 653
Level 3: Locus 654
613
Unit 13, Sector 1, PAT-047
Location (WGS 1984): 656853E 8428014N
Dimensions: 2m x 2m
Declared Loci: 351, 352, 353
Unit 13 was located in Sector 1, the religious core of the site at Iglesiachayoq, northeast
of the accessway of the circular structure EST-047. This circular structure was located somewhat
isolated, with its own patio and surrounding border walls. The excavation unit was strategically
placed in the patio of EST-047 in order to learn about domestic or non-domestic activities that
were occurring in the patio outside of this structure. The unit was placed offset from the entrance
of the structure, in an attempt to discover any sort of trash midden that would have been dumped
outside the structure.
Summary of Loci
In total, Unit 13 had three contexts among natural strata and arbitrary levels, which were
identified as loci. The excavation team tried to follow each natural level as completely and as
faithfully as possible.
Locus 351
Locus 351 represented the unit’s natural topsoil and consisted of a loose, poorlyconsolidated medium-grained sand that was brown (7.5YR 4/4) in color. This initial level
contained almost no cultural material, possibly because the area may have been used for
agricultural activities in the past. In total, we collected three bags from Locus 351, including one
bag of ceramic, one of lithic, and one soil sample. The locus ended at an average depth of 67cm
below datum and was an average thickness of 19cm throughout the unit.
Locus 352
614
Locus 352 was a natural level just below Locus 351, and was characterized by a soil
change, such that the matrix was a semicompact, moderately distributed medium-grained sand
which was brown (7.5YR 4/3) in color. Like the previous locus, we collected very little
archaeological material from this locus. We found one small piece of carbon, but did not collect
it since it was not associated with any context. We collected three bags in total, including one
bag of ceramic, one bag of lithic, and one soil sample. In general, Locus 352 ended at an average
depth of 82cm below datum, and was an average thickness of 15cm throughout the unit.
Locus 353
Locus 353 was a natural level below Locus 352, and was characterized by less cultural
material and was a uniformly distributed, semicompact medium-grained sand that was brown
(7.5YR 4/4) in color. In total, we collected only two bags from this locus, including one bag of
ceramic and one soil sample. We reached sterile soil at the end of the unit. In general, Locus 353
ended at an average depth of 92cm blow datum and was an average thickness of 10cm
throughout the unit.
We terminated the excavation of Unit 13 at this point, because we had found almost no
cultural materials and had reached sterile soil. After drawing the southern profile and cut, and
taking lots of photos, we refilled the unit with sieved soil and closed it.
Unit 13 Level Groupings
Level 1: Locus 351
Level 2: Locus 352
Level 3: Locus 353
615
Unit 14, Sector 2, PAT-168
Location (WGS 1984): 657001E 8428114N
Dimensions: 2m x 2m
Declared Loci: 551, 552, 553
Unit 14 was located in Sector 2, situated geographically northeast of the core of the site at
Iglesiachayoq, slightly northwest of EST-168 but aligned with the entry of EST-168. This
quadrangular structure with rounded corners was located in a somewhat isolated space, with its
own patio and surrounding border walls. The excavation unit was strategically placed in the patio
of EST-168 in order to learn about domestic or non-domestic activities that were occurring in the
patio outside of this structure.
Summary of Loci
In total, Unit 14 had three contexts among natural strata and arbitrary levels, which were
identified as loci. The excavation team tried to follow each natural level as completely and as
faithfully as possible.
Locus 551
Locus 551 represented the unit’s natural topsoil and consisted of a loose, medium-grained
sand that was poorly homogeneous throughout the unit and brown (7.5YR 4/3) in color. This unit
was placed in what we now know was an agricultural field, so all contexts were likely disturbed.
We have found fragments of ceramic strewn with lots of roots. In total, we collected three bags
from Locus 551, including two bags of ceramic and one bag of lithic. In general, Locus 551
ended at an average depth of 51cm below datum and was an average thickness of 12cm
throughout the unit.
Locus 552
616
Locus 552 was a natural level below Locus 551 and was characterized by a soil matrix
that was compact, uniformly distributed, and a medium-grained sand that was brown (7.5YR 4/3)
in color. There was very little cultural material, and we collected only four bags of artifacts,
including 2 bags of ceramic, one bag of lithic, and one soil sample. Locus 552 ended at an
average depth of 64cm below datum and was an average thickness of 13cm throughout the unit.
Locus 553
Locus 553 was the last natural level in Unit 14, and was made up of the sterile soil. The
soil matrix was a very compact, uniformly distributed medium-grained sand which was brown
(7.5YR 4/3) in color. There was no cultural material in this locus, and we did not collect any
bags rom Locus 553. In general, Locus 553 ended at an average depth of 17cm below datum, and
was an average thickness of 10cm throughout the unit.
We decided to terminate the excavation at this point since we had reached sterile soil and
there were no archaeological materials. After drawing the northern cut, western cut, and northern
profile, we took lots of photos and backfilled the unit with sieved soil.
Unit 14 Level Groupings
Level 1: Locus 551
Level 2: Locus 552
Level 3: Locus 553
617
Unit 15, Sector 3, PAT-292
Location (WGS 1984): 656826E 8427843N
Dimensions: 2m x 2m
Declared Loci: 001, 002, 003, 004, 005, 006, 007
Unit 15 was located in Sector 3, situated geographically south of the core of the site of
Iglesiachayoq. Unit 15 was located north of EST-292, and aligned with the entry of EST-292.
EST-292 was an ovoid structure located facing an upper plaza. The excavation unit was
strategically placed in the patio of EST-92 in order to learn about domestic or non-domestic
activities that were occurring in the patio outside of this structure.
Summary of Loci.
In total, Unit 15 had seven contexts among natural strata and arbitrary levels, which were
identified as loci. The excavation team tried to follow each natural level as completely and as
faithfully as possible.
Locus 001
Locus 001 represented the unit’s natural topsoil, and was characterized as a semicompact, moderately homogeneous medium-grained sand which was dark reddish gray (5YR
4/2) in color. There was very little material, and we collected only two bags from this locus
including one of ceramic and one of lithic. In general, Locus 001 ended at an average depth of
60cm below datum, and was an average thickness of 13cm throughout the unit.
Locus 002
Locus 002 was the second natural level, and was characterized by a soil matrix that was a
semi-compact, moderately homogeneous medium-grained sand which was brown (5YR 4/3) in
color. The presence of cultural material diminished in this Locus compared to the preceding
618
locus, and we collected only one bag of ceramic from Locus 002. In general, Locus 002 ended at
an average depth of 72cm below datum and was an average thickness of 13cm throughout the
unit.
Locus 003
Locus 003 was the third natural level and was characterized by a compact, granular
medium-grained sand which was a dark yellowish brown (10YR ¾) in color. There was even
less cultural material in Locus 003, and we collected only one bag of ceramic and one bag of
carbon, which was found without an associated context. In general, Locus 003 ended at an
average depth of 86cm below datum and was an average thickness of 13cm throughout the unit.
Locus 004
For the rest of the excavation of Unit 15, we only dug in Quad 4 (the southeast quad)
since we had found so little cultural material. We decided to go down at least 50cm without
cultural material to confirm that there were no more cultural levels, as this was one of the first
units we excavated. Locus 004 represented a natural, sterile level which was characterized by a
soil matrix that was a medium-grained sand that was uniformly distributed throughout the quad,
compact, and black (5YR 2.5/1) in color. We found no cultural materials in Locus 004 and
collected no bags. In general, Locus 004 ended at an average depth of 100cm below datum and
was an average thickness of 14cm throughout the quad.
Locus 005
Locus 005 continued only in quad 4 to make sure there were no more cultural levels in
Unit 15. This locus was an arbitrary locus to maintain control, and was characterized by a
compact, uniformly distributed medium-grained sand that was black (5YR 2.5/1) in color. There
619
were no associated artifacts and we collected no bags. In general, Locus 005 ended at an average
depth of 123cm below datum and was an average thickness of 22cm throughout the quad.
Locus 006
Locus 006 was an arbitrary level excavated in only quad 4, in order to make sure there
were no more cultural levels. The soil matrix was a compact, medium-grained sand which was
uniformly distributed throughout quad 4 and was black (10YR 2/1) in color. We collected no
artifacts from this locus, and the locus ended at an average depth of 150cm below datum and was
an average thickness of 27cm throughout the quad.
Locus 007
Locus 007 was the last locus excavated in Unit 15, and was an arbitrary level to ensure
we had arrived at sterile soil. The soil matrix in Locus 007 was a compact, uniformly distributed
medium-grained sandy-clay which was black (5Y 2.5/1) in color. There were no artifacts in this
Locus, and the sand was very difficult to excavate. We collected one sample from this locus,
which was clay. In general, Locus 007 ended at an average of 171cm below datum and was an
average thickness of 21cm throughout the quad.
We terminated the excavation of Unit 15 at this point, after having excavated more than
50cm without a single cultural artifact. Loci 001, 002, and 003 represented natural levels, and
loci 004, 005, 006, and 007 were arbitrary loci which we dug to ensure we had reached sterile
soil. After drawing the southern cut and eastern profile and taking lots of photos, we backfilled
Unit 15 with sieved soil and closed the unit.
Unit 15 Level Groupings
620
Level 1: Locus 001
Level 2: Locus 002
Level 3: Locus 003
Level 4: Locus 004, 005, 006, and 007
621
Unit 16, Sector 1, PAT-001
Location (WGS 1984): 656796E 8427953N
Dimensions: 5m x 5m (divided into 4 2.5m x 2.5m quads)
Declared Loci: 301, 302, 303
Unit 16 was located in the central plaza due North of the church in Sector 1, the religious
core of Iglesiachayoq. Unit 16 was placed in the center of the plaza and was a 5m x 5m unit
designed to learn about the church ritual activities that were occurring in the central space at
Iglesiachayoq.
Summary of Loci
In total, Unit 16 had three contexts among natural strata and arbitrary levels, which were
identified as loci. The excavation team tried to follow each natural level as completely and as
faithfully as possible.
Locus 301
Locus 302 represented the unit’s natural topsoil and was characterized by a semi-loose,
moderately homogeneous medium-grained sand which was brown (7.5YR 4/3) in color. We
collected very little archaeological material, but found lots of roots and small, granular stones. In
total, we collected 8 bags of artifacts from this level, including 4 bags of ceramic (one for each
quad), 3 bags of lithic, and one soil sample. Locus 301 ended at an average of 26cm below
datum and was an average thickness of 10cm throughout the unit. There were no associated
contexts, and the lack of material may have been caused by disruption due to agricultural activity
in the recent past.
Locus 302
622
Locus 302 was excavated only in quads 1 and 4 due to the lack of material found in
Locus 301 and the great size of the quads. Locus 302 was a natural level characterized by a semiloose, moderately homogeneous medium-grained sand which was brown (7.5Locus 302 was a
natural level characterized by a semi-loose, moderately homogeneous medium-grained sand
which was dark brown (7.5R 3/2) in color. The darker color of this soil, combined with dispersed
flecks of carbon suggests a burning event could have taken place here. We found very little
cultural material in this locus, collecting six bags total: one bag of ceramic and three bags of
lithic, and two soil samples. The lithic recovered was in huge quantities, and appeared to be
shattered pieces of slate, basalt, and hornfel. The greater degree of lithic here as opposed to other
areas of the site suggests that certain stones could have been intentionally destroyed here,
possibly remnants of huacas. Locus 302 ended at an average depth of 55cm below datum and
was an average thickness of 29cm in both quads 1 and 4.
Locus 303
The last locus excavated was locus 303, which was dug in all of quad 4 and a 50cm x
50cm cut in quad 1. We decided to fully excavate quad 4 because this is where the majority of
lithic was found in the previous locus. The soil matrix in Locus 303 was characterized by a dark
brown (7.5YR 3/2) compact, uniformly homogeneous medium-grained sand with an absence of
cultural materials. We collected three bags from this locus, including 1 bag of ceramic, one bag
of lithic, and one soil sample. Like the previous locus, Locus 303 was characterized by a darker
soil color perhaps indicating a burning event, as well as the presence of lots of fractured stone,
which could have been huacas destroyed by Albornoz. Locus 303 ended at an average depth of
66cm below datum and was an average thickness of 10cm.
623
At this point, since we had found scarce cultural material, we decided to terminate
excavation of Unit 16. We drew the southern cut, the eastern cut, and the southern profile, took
lots of photos, and then backfilled Unit 16 with sieved soil.
Unit 16 Level Groupings
Level 1: Locus 301
Level 2: Locus 302
Level 3: Locus 303
624
Unit 17, Sector 1, EST-001
Location (WGS 1984): 656786E 8427936N
Dimensions: 5.5m x 2m (divided into two 2.75 x 2 quads)
Declared Loci: 901, 902, 903, 904, 905, 906, 907, 908, 909, 910, 911, 912, 913, 914, 915, 916
Unit 17 was located inside of the church in Sector 1, and was one of three large trenches
placed in EST-001 in order to define whether or not the structure was a church and to gain
bioarchaeological data from the burials located in the church. Unit 17 was located against the
western wall of the church, abutting this wall where the mural was located.
Summary of Quads and Loci
In total, we divided Unit 17 into two separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.75m x 1m. We numbered the quads from North to
South (1-2). Unit 17 used the same set of loci for all quads. Unit 17 had 16 contexts among
natural strata and arbitrary levels, which were identified as loci. The excavation team tried to
follow each natural level as completely and as faithfully as possible.
Locus 901
Locus 901was the surface of Unit 17. The soil matrix was a loose, fine-grained sand
which was uniformly distributed throughout the locus and brown (7.5YR 4/3) in color. There
were very little cultural materials recovered in this locus—this could be because this area was
used as an area for cultivation. In total, we collected 8 bags from Locus 901, including 3 bags of
ceramic, 3 bags of lithic, 1 bag of animal bone (likely from present day), and one soil sample. In
general, Locus 901 ended at an average depth of 50cm below datum and was an average
thickness of 16cm.
Locus 902
625
Locus 902 was a natural level below locus 901, and the soil matrix was a semi-compact,
medium-grained sand which was uniformly distributed throughout the unit and was dark brown
(7.5YR 3/2) in color. There was little evidence of cultural activity, only a few fragments of
ceramic. In total, we collected 5 bags from Locus 902, including 2 bags of ceramic, 1 bag of
lithic, 1 carbon sample, and 1 soil sample. Locus 902 ended at an average depth of 61cm below
datum and was an average thickness of 11cm.
Locus 903
Locus 903 was a natural level in Unit 17, designated by a soil change such that the matrix
in 903 was a semicompact, uniformly distributed medium-grained sand that was brown (10YR
4/3) in color. In the southeast of quad one, we recovered a special find which was a copper tupu
at 71cm below datum. In total, we collected 6 artifact bags from Locus 903, 2 of ceramic, 2 of
lithic, 1 of metal, and 1 soil sample. In general, the locus ended at 71cm below datum, and the
thickness of the locus was an average of 10cm.
Locus 904
Although we were not finding cultural material, we decided to continue excavating until
we were absolutely sure we had hit sterile soil, because burials likely would be located deeper in
the unit. Locus 904 was excavated in both quads 1 and 2 and was a natural level defined by a
compact, dark brown (7.5YR 3/2) medium-grained sand that was uniformly distributed
throughout the unit. There was very little cultural material, and we collected 5 bags including 2
of ceramic, 1 of lithic, 1 carbon sample (without associated context), and one soil sample. In
general, Locus 904 ended at an average depth of 83cm below datum, and was an average
thickness of 12cm throughout the unit.
626
Locus 905
Locus 905 was located just under Locus 904, and shared the horizontal level with Loci
906, 907, 908, 909, 910, 911, 912, 913, 914, and 915, all which contained burials or areas where
burials had been removed. In general, the soil in Locus 905 was poorly homogeneous—there
were areas which were very compact, and there were intrusions (burials) where the soil was
looser and less compact. Locus 905 was located below the church floor and the soil matrix was a
dark grayish brown (10YR 4/2) medium-grained sand which was poorly homogeneous. We
collected only here artifact bags from this locus, including 1 of ceramic and 1 of lithic, and one
soil sample. Locus 905 ended at an average depth of 91cm below datum, and had an average
thickness of 8cm.
Locus 906
Locus 906 was the first burial we encountered, and the area we defined for this locus was
90cm x 36cm. The soil in this area is extremely acidic, so we only have crania, teeth, and some
long bone for each of the burials. Locus 906 was a burial interred in a flexed position with its
head to the south and feet to the north. It was located in quad 2 abutting the western wall of the
church. The cranium was 95cm BD, the mandible was 107cm BD, the superior extremities were
105cm BD, and the interior extremities were 105-106cm BD. The soil in Locus 906 was very
loose, and no cultural artifacts were associated with the individual. After preliminary
bioarchaeological analysis, it appears that this individual was elderly. The soil matrix was a
loose, poorly homogeneous fine-grained sand that was very dark grayish brown (10YR 3/2) in
color. In total, we collected three bags from Locus 906, including 2 bags of human bone (one
cranium and the other long bones) and one soil sample. Locus 906 ended at an average of 115cm
below datum and was an average thickness of 17cm throughout the locus.
627
Locus 907
Locus 907 was in the same horizontal level as 905, but was another burial designated by
a loose soil matrix. In this case, the individual had his head directly in the northwest corner of
quad 1 and the unit, with the long bones at an angle to the southeast. This locus measured 110cm
x 80cm. The cranium, while visible, was completely disintegrated, but we were able to recover
nearly all the teeth. We also recovered two identical tupus located at the neck of the individual,
which were likely part of the burial attire. We divided this locus into two hallazgos which
appeared separated but associated. One of these was the main burial, and hallazgo 2 was a
collection of long bones located in the very north of the unit. We recovered one grave good with
this burial which was a small, miniature teacup that was gray in color. In total, we recovered 11
bags from Locus 907, including 2 of ceramic, 2 of lithic, 1 of metal (the two tupus), 4 of human
bone, and 2 soil samples. Locus 907 ended at an average depth of 110cm below datum, and was
an average thickness of 21cm. The soil matrix was a very loose, moderately homogeneous finegrained sand that was dark grayish brown (10YR 4/2) in color.
Locus 908
Locus 908 was located in the same horizontal level as 905, and was another burial located
halfway down quad 1, pegged against the western wall of the unit. Locus 908 was 85cm x 50cm
in size, and the bones were in extremely bad shape. All we recovered from Locus 908 was some
fragments of human bone, and more specifically, the cranium. The cranium was located to the
northern side of this locus, with its head facing the south. Apart from the cranium, we recovered
only a few small pieces of long bone. The soil matrix in Locus 908 was a loose, moderately
uniform fine-grained sand which was dark grayish brown (10YR 4/2) in color. We recovered 5
bags from this locus, including 1 bag of ceramic, 2 of human bone, 1 soil sample, and one
628
sample of what appears to have been ocher. The locus ended at an average depth of 106cm below
datum, and was an average thickness of 14cm throughout the locus.
Locus 909
Locus 909 was located in the same horizontal level as Locus 905 and presented as an area
with looser soil. The Locus measured 165cm x 75cm in size and was located in the southeastern
corner of quad 2. The soil matrix was a loose, fine-grained sand that was poorly homogeneous
and a dark reddish brown (5YR 3/2) in color. We recovered only a few very small fragments of
human bone, one small fragment of obsidian, and two different tupus—one of these tupus has a
fiber stuck through the hole. It is possible that the burial was removed from this locus since we
only had distal bone fragments and the density of the soil was looser than the rest of Locus 905.
In all, we recovered 4 bags from this locus including one of lithic, one of metal (the tupus), one
of human bone (the fragments), and one soil sample. The unit ended at an average depth of
121cm below datum and was an average thickness of 17cm throughout the locus.
Locus 910
Locus 910 was yet another area of loose soil in the same horizontal level as 905. Locus
910 was located in the northeast portion of quad 2 and measured 84cm x 53cm in size. In this
locus, we did not recover any evidence of cultural activity. However, it appears this was once a
burial pit that had the skeleton removed and was refilled. Evidence of this can be found in the
several loose stones which were in the pit, and the looseness of the soil. In all, we recovered only
one soil sample in this locus, and the soil matrix was a loose, poorly consolidated fine-grained
sand which was dark brown (7.5YR 3/2) in color. This locus ended at an average depth of 122cm
below datum, and was an average thickness of 31cm.
629
Locus 911
Locus 911 represented another area of loose soil in the same horizontal level as Locus
905. Locus 911 was an extended burial located in the northeast corner of quad 1. We placed an
extension of 15cm x 85cm in order to account for the length of this burial. While the human bone
in this locus was in very bad condition (some long bone fragments, we also recovered four
turquoise Nueva Cadiz beads which date to the Early Colonial Period, and one tupu with fibers
attached. There was also a small piece of carbon which we recovered. While we did not recover
the cranium, the location of the necklace was at the west, and the feet of the individual were
located to the east, facing the altar. The soil matrix was a loose, poorly consolidated fine-grained
sand which was brown (7.5YR 4/3) in color. We recovered 7 total bags from this locus,
including 2 bags of metal, one bag of human bone, one macrobotanical, two carbon samples, and
one soil sample. Locus 911 ended at an average of 106cm below datum, and was an average
thickness of 16cm.
Locus 912
Locus 912 was located directly south of Locus 911, and measured 116cm x 62cm in size,
utilizing the same extension we placed for Locus 911. Locus 912 was a definitive extended
burial, with the head of the individual in the west and the feet in the east, facing east. The arms
were flexed with the hands near the cranium. We recovered a small metal disc near the
neck/chest, which could have been a medallion. The soil matrix was a loose, poorly
homogeneous fine-grained sand which was brown (7.5YR 4/3) in color. We recovered 7 total
artifact bags from Locus 912, including 1 bag of metal (the medallion), 4 bags of human bone, 1
soil sample, and the cranium. Locus 912 ended at an average depth of 112cm below datum and
was an average thickness of 22cm throughout the locus.
630
Locus 913
Locus 913 was another area of loose soil located in the south part of quad 1, 110cm x
75cm in size and in the same horizontal level as Locus 905. We encountered almost no cultural
material, only a fragment of ceramic, and only one long leg. This locus also may have been an
area where a burial was removed, since we only have evidence of one bone and no teeth or
evidence of a cranium. The soil matrix was a loose, uniformly homogeneous fine-grained sand
that was brown (7.5YR 4/3) in color. In total, we recovered three bags from this locus, including
1 of ceramic, 1 of human bone, and 1 soil sample. Locus 913 ended at an average depth of
120cm below datum and was an average thickness of 27cm throughout the locus.
Locus 914
Locus 914 was an area of loose soil matrix in the same horizontal level as Locus 905. It
was located in the southwestern corner of quad 2 and measured 80cm by 60cm in size. There was
no evidence of archaeological material or bone, and could have been where a burial was removed
in the past. The soil matrix was a loose, fine-grained sand which was moderately homogeneous
throughout the unit and dark grayish brown (10YR 4/2) in color. We recovered only one soil
sample from Locus 914. In general, Locus 914 ended at an average depth of 112cm below datum
and was an average thickness of 17cm throughout the locus.
Locus 915
Locus 915 was the last burial excavated in Unit 17, and was an extended burial in the
north of quad 2. The size measured 158cm x 55cm, and the individual had its head to the west
and feet to the east, facing the altar. The soil was very loose and was a dark reddish brown (5YR
3/3) poorly homogeneous fine-grained sand. We recovered the cranium and inferior extremities
631
from this locus, without any evidence of cultural materials. In all, we recovered 4 artifact bags
from Locus 915, including 2 of human bone, 1 cranium, and 1 soil sample. Locus 915 ended at
an average depth of 129cm below datum and was an average thickness of 27cm throughout the
locus.
Locus 916
Locus 916 was located below all of the burials, and was a very compact medium-grained
sand which was uniformly distributed throughout the unit. The soil matrix was very dark brown
in color (7.5YR 2.5/3) and represented the sterile soil. There were no cultural artifacts, but we
did recover two fragments of human bone which were likely remnants of the burials. The soil
matrix was very difficult to excavate, so we terminated the excavation with this locus. In total,
we recovered three artifact bags from this locus, including 2 bags of human bone and one soil
sample. Locus 916 ended at an average depth of 130cm below datum and was an average
thickness of 36cm throughout the unit.
We terminated the excavation of Unit 17 at this point since the soil was extremely
difficult to excavate and we had found no cultural materials in nearly 40cm. In total, we
excavated 6 definitive burials in this unit, 3 which were flexed and 3 which were extended with
their feet pointed toward the altar. There were four more areas which had one or two bone
fragments but no other evidence of human bone, and these were either disintegrated burials or
areas where burials had been removed. In general, the burials presented as areas of loose soil
amongst a more compact soil matrix (Locus 905). We recovered several associated tupus from
this unit, which will be useful for a study in the future. We drew sketches of each burial and took
lots of photos (including aerial photos), and we drew the northern cut and the western profile.
632
Finally, we took lots of photos, backfilled the unit with sieved soil, and terminated the
excavation of Unit 17.
Unit 17 Level Groupings
Level 1: Locus 901
Level 2: Locus 902
Level 3: Locus 903
Level 4: Locus 904
Level 5: Locus 905, 906, 907 908, 909, 910, 911, 912, 913, 914, 915
Level 6 (sterile soil): Locus 916
633
Unit 18, Sector 2, EST-001
Location (WGS 1984): 656796E 8427938N
Dimensions: 5.5m x 2m (divided into two 2.75m x 2m quads)
Declared Loci: 951, 952, 953, 954, 955, 956, 957, 958, 959, 960, 961, 962, 963, 964, 965
Unit 18 was located inside of the church in Sector 1, and was one of three large trenches
placed in EST-001 in order to define whether or not the structure was a church and to collect
bioarchaeological data from the burials located in the church. Unit 18 was located just east of
western door, encompassing where we found the benediction water bowl. Unit 18 extended from
the northern wall of the church to the southern wall of the church.
Summary of Quads and Loci
In total, we divided Unit 18 into two separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.75m x 1m. We numbered the quads from North to
South (1-2). Unit 18 used the same set of loci for all quads. Unit 18 had 15 contexts among
natural strata and arbitrary levels, which were identified as loci. The excavation team tried to
follow each natural level as completely and as faithfully as possible.
Locus 951
Locus 951 was the surface of Unit 18. The soil matrix was a loose, medium-grained sand
that was uniformly distributed throughout the unit and dark reddish brown (5YR 3/3) in color.
There were very little cultural materials collected in this locus, including 7 total bags: 2 bags of
ceramic, 2 of lithic, 1 of animal bone, one soil sample, and one sample of clay tiles from the
northern end of the unit. At the north of the unit abutting the wall, we recovered three pieces of a
round, stone bowl which was likely used to hold the holy water near the door of the church. Like
Unit 17, this area was also used for agriculture, so the surface could have been disturbed. In
634
general, Locus 951 ended at an average of 65cm below datum and was an average thickness of
11cm throughout the unit.
Locus 952
Locus 952 was a natural level below locus 951, and the soil matrix was a loose,
uniformly distributed medium-grained sand which was dark reddish brown (5YR 3/3) in color.
There was less organic material than in Locus 951, as well as less cultural material. In total, we
collected 5 artifact bags from Locus 952, including 1 bag of ceramic, two bags of lithic, a soil
sample, and a second sample from the burned, thick clay pieces at the north of the unit. Locus
952 ended at an average depth of 76cm below datum, and was an average thickness of 11cm
throughout the unit.
Locus 953
Locus 953 was the third natural level in Unit 18, designated by a soil change such that the
soil in Locus 953 was a compact, uniformly distributed medium-grained sand that was dark
reddish brown (5YR 3/3) in color. Like the previous two loci, we recovered flat, hard clay pieces
which were red in color, which could have been part of the roof tiles or the floor. They also
could have been fragments of mortar from the walls. In total, we recovered 5 artifact bags from
Locus 953, including 2 bags of ceramic, 1 bag of lithic, one soil sample, and one sample of the
flat clay fragments from the north of the unit. In general, Locus 953 ended at an average depth of
85cm below datum and was an average depth of 9cm throughout the unit.
Locus 954
We began excavating Locus 954 uniformly throughout Unit 18, but then encountered two
distinctive loci at the south of the unit, which we designated 955 and 956. Locus 954 was a
635
compact, uniformly homogeneous medium-grained sand that was brown (7.5YR 4/3) in color. In
quad 1, we recovered one metal artifact which was a tupu, but it lacked association with any
feature. In total, we collected 4 artifact bags from Locus 954, including 1 bag of ceramic, 1 bag
of lithic, 1 metal artifact (the tupu) and a soil sample. Locus 954 ended at an average depth of
94cm below datum and was an average thickness of 9cm in quads 1 and 2.
Locus 955
Locus 955 shared the same horizontal provenience with Locus 954, and was a 1m x 0.5m
sized area in the southeast corner of quad 2. Locus 955 was designated by the presence of a
poorly-conserved human cranium, possibly of a minor individual. We recovered the bone with
the soil matrix to maintain the integrity of the remaining cranial bone. It is possible that this
individual was a secondary burial, since we have only a cranium and no other connected human
bone. The soil matrix was a compact, uniformly-distributed fine-grained sand which was brown
(7.5YR 4/3) in color. In all, we collected only two bags from this locus, including the cranium
and a soil sample. In general, Locus 955 ended at an average depth of 102cm below datum, and
was an average thickness of 17cm throughout the locus.
Locus 956
Locus 956 shared the same horizontal provenience as Locus 954 and 955, and was a 1m
x 0.5m sized area in the southwest corner of quad 2. We found the remains of a cranium and a
tupu here, as well as assorted teeth. The bones were very fragile and difficult to remove. It is
possible that the bones in Locus 955 are associated with the cranium in Locus 956, as they were
recovered at the same depth; however, because there are two separate crania (one in each locus),
we separated the two regions into two distinct loci. The cranium in Locus 956 was also likely
636
from a minor individual. The soil matrix was a compact, uniformly-distributed fine-grained sand
which was brown (7.5YR 4/3) in color. We collected only two artifact bags from Locus 956,
including one metal artifact (the tupu) and one bag of human bone. In general, Locus 956 ended
at an average depth of 121cm below datum and was an average thickness of 32cm throughout the
locus.
Locus 957
Locus 957 was a natural level below Locus 954, which we excavated in order to level out
the burial depths with the rest of the unit. Locus 957 was a compact, uniformly distributed
medium-grained sand which was brownish black (7.5YR 3/2) in color. We collected one carbon
sample which lacked association from the northwest corner of quad 1. In general, there was no
cultural material from Locus 957, including only 1 carbon sample, and 1 soil sample. In general,
Locus 957 ended at an average depth of 109cm below datum and was an average thickness of
15cm throughout the unit.
Locus 958
Locus 958 was a 1m x 0.9m-sized area which contained a burial located in the western
side of quad 1, approximately 25cm south of the northern wall. Locus 958 shared part of the
same horizontal stratigraphy as Locus 957 and was a loose, uniformly distributed fine-grained
sand which was brownish black (7.5YR 3/2) in color. We recovered long bones and the occipital
part of the cranium. The long bones are somewhat disarticulated, and this burial appears to be
secondary, as if the bones were intentionally placed on top of the other. The burial was interred
with the head at the west and the long bones to the east, and was in a flexed position, as though it
were a bundle. We collected 4 bags from Locus 958, including 1 bag of ceramic, 2 of human
637
bone, and one soil sample. In general, Locus 958 ended at a depth of 118cm below datum, and
was an average thickness of 20cm throughout the locus.
Locus 959
Locus 959 was another natural level excavated below Locus 957/958. In quad 2, the soil
was very compact without any cultural material. In quad 1, the soil was a little looser with small
round stones. There were no archaeological materials in either quad, so we only collected one
soil sample from Locus 959. In general, Locus 959 ended at an average depth of 117cm below
datum and was an average thickness of 7cm throughout the unit.
Locus 960
Locus 960 shared the same horizontal strata as Locus 959 and was a 0.65m x 0.6m sized
area in the western side of quad 1. Locus 960 was separated because of the presence of human
bone, including long bones and fragments of a cranium. We found three associated metals—two
whole tupus, and a third fragment of a tupu. Given the bones present in Locus 960, we believe
that this possibly was a secondary burial since there is no anatomical connection between the
bones. The head was situated to the west of the long bones, and was overall oriented to the altar
in a flexed position. Locus 960 was characterized by a loose, uniformly homogeneous finegrained sand which was brownish black (7.5YR 3/2) in color. In total, we collected 6 bags from
Locus 960, including 1 bag of ceramic, 1 bag of lithic, 1 bag of metals (the two tupus and a third
fragment of a tupu), 2 bags of human bone, and one soil sample. In general, Locus 960 ended at
an average depth of 121cm below datum and was an average thickness of 8cm throughout the
locus.
Locus 961
638
Locus 961 was a natural level which we excavated throughout quads 1 and 2. It was
characterized by a compact, medium-grained sand which was uniformly homogeneous and
brownish black (7.5YR 3/2) in color. We collected no ceramic or lithic from this level; however,
there was one metal tupu in quad 2 at a depth of 1.23m. In all, we collected 2 bags from Locus
961, including 1 metal artifact (the tupu) and one soil sample. In general, Locus 961 ended at an
average depth of 125cm below datum and was an average thickness of 8cm throughout the unit.
Locus 962
Locus 962 was another burial located in the northwest corner of quad 1 and measuring
0.9m x 0.45m in size. Here, we recovered long bones and phalanges. Part of the burial was
located outside of the western boundary of the unit, so we added an internal amplification to
remove the human bones. This burial was also oriented toward the altar, with the head to the
west and the legs to the east. There was no anatomical connection between the bones, so we
think this burial could have been secondary. The soil matrix in Locus 962 was a loose, uniformly
distributed fine-grained sand which was brownish black (7.5YR 3/2) in color. In all, we
recovered 5 artifact bags from Locus 962, including one bag of ceramic, 2 bags of human bone,
and two soil samples. Locus 962 ended at an average depth of 127cm below datum and was an
average thickness of 14cm throughout the locus.
Locus 963
Locus 963 was a natural level which shared the same horizontal strata as Locus 962.
Locus 963 was excavated in both quads and was a compact, medium-grained sand which was
uniformly homogeneous and brownish black (7.5YR 3/2) in color. There were no cultural
639
materials in this locus, and we collected only one soil sample. In general, Locus 963 ended at an
average depth of 132cm below datum and was an average thickness of 7cm throughout the unit.
Locus 964
Locus 964 was an arbitrary level located below Locus 963. Although we hadn’t found
any cultural materials in some time, we wanted to continue excavating until sterile soil to ensure
that all burials and evidence of cultural occupations had been recovered. The soil matrix in Locus
964 was very compact with small pebbles, and brownish black (7.5YR 3/2) in color, but lacking
any cultural evidence. We collected only one soil sample. In general, Locus 964 ended at an
average depth of 140cm below datum and was an average thickness of 9cm throughout the unit.
Locus 965
Locus 965 was excavated only in quad 1, because the soil matrix in quad 2 was extremely
compact and nearly impossible to dig through. Locus 965 was an arbitrary level to ensure that we
had collected all possible cultural remains from the preceding loci. The soil matrix in Locus 965
was a compact, thick-grained sand which was uniformly homogeneous and brownish black
(7.5YR 3/2) in color. There was no cultural evidence in this locus, and we believed that we had
reached the sterile soil so we terminated excavation of Unit 18 at this point. On average, Locus
965 ended at a depth of 149cm below datum and was an average thickness of 8cm throughout the
unit.
We terminated the excavation of Unit 18 at this point since the soil was extremely
difficult to excavate and we had found no cultural materials in the previous two loci. In total, we
excavated 5 definitive burials in this unit, the majority of which were likely secondary and in
flexed position. Three out of the five burials were oriented toward the altar, with the cranium to
640
the west and the long bones to the east, and the orientation of the other two burials could not be
determined. In general, the burials presented as areas of loose soil amongst a more compact soil
matrix, and they did not share a horizontal provenience but rather were located at different
depths throughout the unit. We recovered several associated and unassociated tupus from Unit
18, which will be useful for a study in the future. We drew sketches of each burial and took lots
of photos (including aerial photos). We drew the western cut and the western profile, took lots of
photos, and backfilled the unit with sieved soil.
Unit 18 Level Groupings
Level 1: Locus 951
Level 2: Locus 952
Level 3: Locus 953
Level 4: Locus 954, 955, 956
Level 5: Locus 957, 958
Level 6: Locus 959, 960
Level 7: Locus 961
Level 8: Locus 962, 963
Level 9: Locus 964, 965
641
Unit 19, Sector 1, EST-001
Location (WGS 1984): 656808E 8427941N
Dimensions: 5.5m x 2m (divided into 2 2.75m x 1m quads)
Declared Loci: 1001, 1002, 1003, 1004, 1005, 1006, 1007, 1008, 1009, 1010, 1012
Unit 19 was located inside of the church in Sector 1, and was one of three large trenches
placed in EST-001 in order to define whether or not the structure was a church and to collect
bioarchaeological data from the burials located in the church. Unit 19 was located abutting the
eastern altar steps and extended from the northern wall of the church to the southern wall of the
church. The northern part of the unit was sunken into the ground, and may have been looted or
disturbed at an earlier time.
Summary of Quads and Loci
In total, we divided Unit 19 into two separate quads in order to maintain a degree of
control in excavation, each quad measuring 2.75 x 1m. We numbered the quads from North to
South (1-2). Unit 19 used the same set of loci for all quads. Unit 19 had 11 contexts among
natural strata and arbitrary levels, which were identified as loci. The excavation team tried to
follow each natural level as completely and as faithfully as possible.
Locus 1001
Locus 1001 was the surface level of Unit 19, and was characterized by a loose,
moderately homogeneous medium-grained sand which was brown (10YR 4/3) in color. In the
eastern side of the unit, we defined the third and lowest step of the altar staircase. In the northern
side of the unit, there was a looter’s pit. In general, there was not much cultural material in Locus
1001, and we collected 6 total bags including 4 bags of ceramic, 1 bag of lithic, and one soil
sample. Locus 1001 ended at an average depth of 85cm below datum and was an average
642
thickness of 12cm throughout the unit. Eventually, we placed a 1m by 1m extension to the west
of quad 2, starting at the northwestern corner and moving 1m down the quad. I will describe this
extension in more detail later on.
Locus 1002
Locus 1002 was a natural level excavated in quads 1 and 2. In quad 2 just in front of the
third step of the altar, we recovered several pieces of a vessel which had been placed
deliberately. We defined this as “hallazgo 1” and collected all of the pieces together. The matrix
in Locus 1002 was a semicompact medium-grained sand which was moderately homogeneous
and a dark grayish brown (10YR 4/2) in color. The soil distribution was only moderately
homogeneous because there were some looser parts and more compact parts throughout the unit,
likely caused by the intrusions in quad 1. In total, we collected 3 bags of artifacts including 2
bags of ceramic and one soil sample. Locus 1002 ended at an average depth of 93cm below
datum and was an average thickness of 8cm throughout the unit.
Locus 1003
Locus 1003 was a natural locus defined by a soil change, such that the soil matrix was a
loose clay/sand which was uniformly distributed throughout the unit and brown (10YR 4/3) in
color. There was very little cultural materials, and we collected only 3 artifact bags, including 2
bags of ceramic and one soil sample. Locus 1003 ended at an average depth of 108cm below
datum and was an average thickness of 9cm throughout the unit.
Locus 1004
Locus 1004 was a natural level located below Locus 1003 which was excavated
throughout the unit. In this locus, there was a very low density of cultural materials, just a few
643
fragments of ceramic randomly distributed throughout quads 1 and 2. There were several rocks
in quad one which could have been used in church activities, since they were finely hewn and
polished. The soil matrix in Locus 1004 was a semi-loose medium-grained sand which was
uniformly distributed throughout the unit and dark brown (7.5YR 3/2) in color. In total, we
collected only three bags from this unit, including 2 bags of ceramic and 1 soil sample. Locus
1004 ended at an average depth of 110cm below datum and was an average thickness of 8cm
throughout the unit.
Locus 1005
Locus 1005 shared the same horizontal strata as Locus 1004, and was located directly
west of the last altar step in quad 2. Locus 1005 was 1.33m x 0.33m, and we recovered 5 long
bones without anatomical connection. We also recovered a few ceramic fragments. The soil
matrix was a loose sandy clay which was uniformly distributed throughout the locus and very
dark grayish brown (10YR 3/2) in color. In total, we collected 3 bags from Locus 1005,
including 1 of ceramic, 1 of human bone, and one soil sample. Locus 1005 ended at an average
depth of 129cm below datum and was an average thickness of 25cm throughout the locus.
Locus 1006
Locus 1006 was a natural level below loci 1004 and 1005, and was characterized by a
loose, uniformly distributed medium-grained sand which was very dark brown (7.5YR 2.5/2) in
color. In quad 1 there were several cut stones which were likely removed from their original
context. The ceramic recovered from quad 1 was likely disturbed and does not reflect original
contexts. There were 3 bags recovered from Locus 1006, including 2 bags of ceramic and one
644
soil sample. Locus 1006 ended at an average depth of 128cm below datum and was an average
thickness of 18cm throughout the unit.
Locus 1007
Locus 1007 was a loose, medium-grained sand which was uniformly distributed
throughout the unit and brown (10YR 4/3) in color. The density of cultural artifacts was
minimal, only fragments of ceramic in quad 1 and almost nothing in quad 2. In total, we
recovered three bags from Locus 1007, including 2 bags of ceramic and one soil sample. Locus
1007 ended at an average depth of 155cm below datum and was an average thickness of 27cm
throughout the unit.
Locus 1008
Locus 1008 was a burial located in the northwest corner of quad 2, sharing the same
horizontal strata as Locus 1007. Locus 1008 was differentiated by a looser soil matrix, which
was a fine-grained sand that was very dark brown (7.5YR 2.5/2) in color and was
homogeneously distributed throughout the locus. The individual we recovered was flexed, with
the cranium situated to the west and the long bones to the east, facing the eastern altar. We
recovered no associated artifacts, and we believe that the individual was a secondary interment.
The cranium was resting on top of the soil, literally looking at the altar, and the bones were in a
much better conservation state than the rest of the bones recovered in the church. This led us to
believe that this individual had been interred first in another area and then deliberately moved to
the church. We found a carbon sample beneath the inferior extremities at a depth of 1.52m below
datum. In all, we recovered 7 bags from this locus, including 1 bag of lithic, 3 bags of human
645
bone, 2 carbon samples, and one soil sample. Locus 1008 ended at an average depth of 142cm
below datum and was an average thickness of 3cm throughout the locus.
Locus 1009
Locus 1009 was another human burial which was located in the eastern portion of quad 2,
just below the altar stairs. The soil matrix in Locus 1009 was a loose, fine-grained sand which
was uniformly distributed throughout the unit and brown (10YR 4/3) in color. We recovered the
cranium of an individual which was likely less than 15 years of age. Below the stairs, we also
recovered a tupu and a needle. Below the cranium, we found long bones which we considered
associated with this locus. However, upon post-field bioarchaeological analysis, it appears that
these long bones likely belonged to an older individual. Taken together, it seems as though the
bones in Locus 1009 represent two separate individuals in secondary burials under the altar
steps. In total, we recovered 5 bags from Locus 1009 including 1 bag of metal (the tupu and the
needle), 3 bags of human bone, and one soil sample. Locus 1009 ended at an average depth of
192cm below datum and was an average thickness of 44cm throughout the locus. This locus was
much thicker than the other loci throughout excavations due to the vertical nature of the burial.
Locus 1010
Locus 1010 was one of the most important finds during excavations at Iglesiachayoq.
Locus 1010 was an extended individual interred with its head to the west and its feet to the east,
directly west of the altar and thus facing the altar. Locus 1010 was an undisturbed tomb with a
stone sarcophagus—there were two courses of stones forming an oval frame for the burial. Atop
of these courses, there were 6 wide stone slabs which enclosed and protected the burial. The
sarcophagus extended west beyond the limits of the unit, so we placed a 1m x 1m extension to
646
the west of the unit in order to define the western edge of the burial. We believe that given the
interment style and position/location of the individual, that this could have been the priest
serving Iglesiachayoq during the Early Colonial Period. The soil matrix in Locus 1010 was a
loose, finely-grained sand which was homogeneous and dark brown (7.5YR 3/2) in color. In
total, we recovered 11 bags from Locus 1010, including 1 bag of ceramic, 2 bag of lithic, 6 bags
of human bone, 2 carbon samples, and one soil sample. Locus 1010 ended at an average depth of
189cm below datum and was an average thickness of 34cm throughout the locus.
Locus 1012
We had declared a Locus 1011 in quad 1, but ended up not excavating any material from
this level, since Locus 1012 extended throughout the unit. Locus 1012 thus was excavated in
quad 1. The ceramic density was very small, and the area was likely disturbed in the recent past.
In quad 1, recovered a carved, square, massive stone which could have been used as the altar
table. The soil matrix was a loose, medium-grained sand which was homogeneous and dark
brown (7.5YR 3/2) in color. In total, we recovered three bags of artifacts from Locus 1012,
including 1 bag of ceramic, one bag of nonassociated human bone, and one soil sample. Locus
1012 ended at an average depth of 176cm below datum and was an average thickness of 22cm
throughout the unit.
At this point, the rainy season had started and we had reached what appeared to be sterile
soil, so we terminated the excavation of Unit 19. In total, we excavated 3 definitive burials in this
unit, two of which were secondary flexed burials and one which was the extended burial in the
stone sarcophagus. In general, the burials presented as areas of loose soil amongst a more
compact soil matrix and were located at different depths throughout the unit. We drew sketches
of each burial and took lots of photos (including aerial photos). Throughout the unit, we drew the
647
northern cut, the southern and eastern profiles, a plan view of the stone sarcophagus (quad 2), a
plan view of the mass table and disrupted context in quad 1, the western cut, and the altar steps
in quad 2. We took lots of photos, and backfilled the unit with sieved soil.
Unit 19 Level Groupings
Level 1: Locus 1001
Level 2: Locus 1002
Level 3: Locus 1003
Level 4: Locus 1004, 1005
Level 5: Locus 1006
Level 6: Locus 1007, 1008, 1009, 1010
Level 7: Locus 1012
648
REFERENCES
Abercrombie, Thomas A.
1998 Pathways of memory and power: ethnography and history among an Andean
people. University of Wisconsin Press, Madison.
Abraham, Sarah
2017 Reconstructing Early Colonial Andean ritual practice at Pukara, Peru: an
architectural approach. In Rituals of the past: Prehispanic and Colonial case studies in
Andean archaeology, edited by S. A. Rosenfeld and S. L. Bautista, pp. 217–240.
University Press of Colorado, Boulder.
Adas, Michael
1979 Prophets of rebellion: millenarian protest movements against the European
colonial order. University of North Carolina Press, Chapel Hill.
Agarwal, Sabrina C., and Julie K. Wesp
2017 Exploring sex and gender in bioarchaeology. University of New Mexico Press,
Albuquerque.
Albornoz, Cristóbal de
1989 Instrucción para descubrir todas las guacas del Pirú con sus camayos y haziendas.
In Fabulas y mitos de los Incas, edited by H. Urbano and P. Duviols, pp. 161-199.
Cronicas de America, Madrid.
1990 Informaciones de servicios. In El retorno de las huacas: estudios y documentos
sobre el Taki Onqoy, edited by L. Millones and P. Guibovich, pp. 43–308. Instituto de
Estudios Peruanos, Lima, Peru.
Albornoz, Cristóbal de, and Pierre Duviols
1967 La instrucción para descubrir todas las guacas del Pirú y sus camayos y haziendas.
Musée de l’homme, Paris.
Alconini, Sonia, and Alan Covey, editors
2018 The Oxford handbook of the Incas. 1st edition. Oxford University Press, Oxford.
Aldenderfer, Mark S.
1993 Domestic architecture, ethnicity, and complementarity in the South-Central
Andes. University of Iowa Press, Iowa City.
Allen, Catherine J.
2002 The hold life has: coca and cultural identity in an Andean community. 2nd ed.
Smithsonian Institution Press, Washington D.C.
649
2015 The whole world is watching: new perspectives on Andean animism. In The
archaeology of wak’as: explorations of the sacred in the Pre-Columbian Andes, edited by
T. L. Bray, pp. 23–46. University Press of Colorado, Boulder.
Álvarez, Bartolomé
1998 De las costumbres y conversión de los indios del Perú: memorial a Felipe II
(1588). Crónicas y memorias v. 4, Ediciones Polifemo, Madrid.
Ambrose, Stanley H.
1990 Preparation and characterization of bone and tooth collagen for isotopic analysis.
Journal of Archaeological Science 17(4): 431–451.
Ambrose, Stanley H., Brian M. Butler, Douglas B. Hanson, Rosalind L. Hunter‐Anderson, and
Harold W. Krueger
1997 Stable isotopic analysis of human diet in the Marianas Archipelago, Western
Pacific. American Journal of Physical Anthropology 104(3): 343–361.
Ambrose, Stanley H., and Lynette Norr
1992 On stable isotopic data and prehistoric subsistence in the Soconusco region.
Current Anthropology 33(4): 401–404.
Andersson, Rani-Henrik
2008 The Lakota Ghost Dance of 1890. University of Nebraska Press, Lincoln.
Andrien, Kenneth J.
2001 Andean worlds: indigenous history, culture, and consciousness under Spanish
rule, 1532-1825. University of New Mexico Press, Albuquerque.
Archivo General de Indias.
1943 La iglesia de España en el Perú. Sección Primera, Archivo General de Indias.
Arguedas, José María
1962 La agonía de Rasu Ñiti. Icaro, Lima.
Arkush, Elizabeth N.
2011 Hillforts of the ancient Andes: Colla warfare, society, and landscape. University
Press of Florida, Gainesville.
Arriaga, Pablo José de
1999 La extirpación de la idolatría en el Pirú (1621). Cuadernos para la historia de la
evangelización en América Latina ; [no.] 13. Centro de Estudios Regionales Andinos
Bartolomé de Las Casas, Cuzco, Peru.
650
Arriaga, Pablo José de, and L. Clark Keating
1968 The extirpation of idolatry in Peru. University of Kentucky Press, Lexington.
Asad, Talal
1973
Anthropology and the colonial encounter. Humanities Press, New York.
Auer, Peter
1998 Code-switching in conversation: language, interaction and identity. Routledge,
New York.
Barker, Philip
1993 Techniques of archaeological Excavation. Psychology Press, New York.
Barnes, Monica
1985 Taki Oncoy in Soras. Paper on file, Craig Morris Archive, Cornell, students,
Monica Barnes, Division of Anthropology, American Museum of Natural History.
1993 Un analisis de la iconografía y la arquitectura Andinas en una iglesia colonial:
San Cristóbal de Pampachiri (Apurímac, Peru). In Mito y simbolismo en los Andes: la
figura y la palabra, edited by H. Urbano, pp. 183–211. Centro de Estudios Rurales
Andinos Bartolomé de las Casas, Cusco, Peru.
Barrenechea, Raúl Porras, Graciela Sánchez Cerro, and Oswaldo Holguín Callo
1986 Los cronistas del Perú y otros ensayos: 1528-1650. Ediciónes del Centenario,
Banco de Crédito del Perú, Lima.
Barrionuevo, Alfonsina
1988 Ayacucho, la comarca del pulsa amaru. Concytec, Lima, Perú.
Bauer, Brian S.
1998 The sacred landscape of the Inca : the Cusco ceque system. University of Texas
Press, Austin.
2011 Introduction. In Account of the Fables and Rites of the Incas by Cristobal de
Molina, translated and edited by B. S. Bauer, V. Smith, and G. E. Cantarutti. University
of Texas Press, Austin.
Bauer, Brian S., Lucas C. Kellett, and Miriam Aráoz Silva
2010 The Chanka: archaeological research in Andahuaylas (Apurímac), Peru. Cotsen
Institute of Archaeology Press, Los Angeles.
Bauer, Brian S., Javier Fonseca Santa Cruz, and Miriam Aráoz Silva
651
2015 Vilcabamba and the archaeology of Inca resistance. Cotsen Institute of
Archaeology Press, Los Angeles.
Bauer, Brian S., and Lucas C. Kellett
2010 Cultural transformations of the Chanka homeland (Andahuaylas, Peru) during the
Late Intermediate Period (A.D. 1000-1400). Latin American Antiquity 21(1): 87-111.
Bauer, Brian S, and Charles Stanish
1990 Killke and Killke-related Pottery from Cuzco, Peru, in the Field Museum of
Natural History. Field Museum of Natural History, Chicago.
Beauchesne, Patrick and Sabrina C. Agarwal
2018 Children and childhood in bioarchaeology. University Press of Florida,
Gainesville.
Beck Jr, Robin A, Douglas J Bolender, James A Brown, and Timothy K Earle
2007 Eventful archaeology: the place of space in structural transformation. Current
Anthropology. 48(6): 833-860.
Bernadini, W., A. Bamash, M. Kumler, and M. Wong.
2013 Quantifying visual prominence in social landscapes. Journal of Archaeological
Science 40(11) 3946-3954.
Betanzos, Juan de, Maria del Carmen Martin Rubio, Horacio Villanueva Urteaga, and Demetrio
Ramos Pérez
1987 Suma y narración de los Incas. Atlas, Madrid.
Binford, Lewis R.
1971 Mortuary practices: their study and their potential. In Approaches to the social
dimensions of mortuary practices, edited by J. A. Brown, pp. 6–29. Society for American
Archaeology, Washington.
Bogaard, Amy, and Alan K. Outram
2013 Palaeodiet and beyond: stable isotopes in bioarchaeology. World Archaeology
45(3): 333–337.
Bongers, Jacob, Elizabeth Arkush, and Michael Harrower
2012 Landscapes of death: GIS-based analyses of chullpas in the western Lake Titicaca
Basin. Journal of Archaeological Science 39(6): 1687–1693.
Branch, Nicholas P.
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): 19.
652
Branch, Nicholas P., Rob A. Kemp, Barbara Silva, et al.
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):
1–9.
Braudel, Fernand
1980 On history. University of Chicago Press, Chicago.
Bray, Tamara L.
2003 Inka pottery as culinary equipment: food, feasting, and gender in imperial state
design. Latin American Antiquity 14(1): 3-28.
2009 An archaeological perspective on the Andean concept of camaquen: thinking
through late Pre-Columbian ofrendas and huacas. Cambridge Archaeological Journal
19(3): 357–366.
2012 From rational to relational: re-configuring value. In The construction of value in
the ancient world, edited by G. Urton and J. Papadoupoulus, pp. 344-357. Cotsen
Institute of Archaeology, UCLA, Los Angeles.
2015 Andean wak’as and alternative configurations of persons, power, and things. In
The archaeology of wak’as: explorations of the sacred in the Pre-Columbian Andes,
edited by T. L. Bray, pp. 3–22. University Press of Colorado, Boulder.
2019 Temporal plurality and temporal transgressions: time and things in an Early
Colonial Period mortuary assemblage from Northern Highland Ecuador. International
Journal of Historical Archaeology.
Brosseder, Claudia
2014 The power of huacas: change and resistance in the Andean world of Colonial
Peru. University of Texas Press, Austin.
Brown, Jennifer
2004 The Wasitay religion: prophecy, oral literacy, and belief on Hudson Bay. In
Reassessing revitalization movements: perspectives from North America and the Pacific
Islands, edited by M. E. Harkin, pp. 104–123. University of Nebraska Press, Lincoln.
Brown, Michael F.
1996 On resisting resistance. American Anthropologist 98: 729–735.
Brown, Michael F, and Eduardo Fernández
1991 War of shadows: the struggle for utopia in the Peruvian Amazon. University of
California Press, Berkeley.
Brown, Peter
653
1982 The cult of the saints: its rise and function in Latin Christianity. University of
Chicago Press, Chicago.
Brown, Vincent
2008 The reaper’s garden: death and power in the world of Atlantic slavery. Harvard
University Press, Cambridge.
Burga, Manuel
1988 Nacimiento de una utopia: muerte y resureccion de los incas. Instituto de Apoyo
Agrario, Lima.
Burger, Richard L, Fidel A. Fajardo Rios, and Michael D Glascock
2006 Potreropampa and Lisahuacho obsidian sources: geological origins of
Andahuaylas A and B type obsidians in the province of Aymaraes, Department of
Apurímac, Peru. Ñawpa Pacha (28): 109–127.
Bush, Jason A.
2012 Staging Lo Andino: the Scissors Dance, spectacle, and indigenous citizenship in
the new Peru. Ph.D. Thesis, The Ohio State University.
Cabello Balboa, Miguel
1951 Miscelánea antártica: una historia del Perú antiguo. Universidad Nacionál
Mayor de San Marcos, Facultad de Letras, Instituto de Etnología, Lima.
Carré, Enrique González
1979 El estudio de los Chankas. Investigaciones 2(2): 55–76.
1992 Los señoríos Chankas. Universidad Nacional de San Cristóbal de Huamanga,
Ayacucho.
Casas, Bartolomé de las
1986 Historia de las indias. Caracas: Ayacucho.
Castro de Trelles, Lucila
1992 Relación de la religión y ritos del Perú hecha por los padres agustinos. Perú:
Pontificia Universidad Católica del Perú, Lima.
Castro-Klarén, Sara
1990 Discurso y transformación de los dioses en los Andes. In El retorno de las
huacas: estudios y documentos sobre el Taki Onqoy siglo XVI, edited by Luis Millones,
pp. 407–424. Instituto de Estudios Peruanos, Lima.
Cavero Carrasco, Ranulfo
654
2001 Los dioses vencidos: una lectura antropológica del Taki Onqoy. Universidad
Nacional de San Cristóbal de Humanga, Ayacucho.
Césaire, Aimé
1972 Discourse on colonialism. Monthly Review Press, New York.
Charles, John
2010 Allies at odds: the Andean church and its indigenous agents, 1583-1671.
University of New Mexico Press, Albuquerque.
Chase, Zachary J.
2015 What is a wak’a? When is a wak’a? In The archaeology of wak’as: explorations
of the sacred in the Pre-Columbian Andes, edited by T. L. Bray. University Press of
Colorado, Boulder.
2016 Performing the past in the historical, ritual, and mythological landscapes of
Huarochirí, Peru (ca. ad1400-1700). Ph.D. Thesis, University of Chicago, Illinois.
Chesson, Meredith, Rosemary Joyce, Lynn Meskell, et al.
2001 Social memory, identity, and death: anthropological perspectives on mortuary
rituals. Archaeological Papers of the American Anthropological Association 10: 1–131.
Christiansen, Palle
1969 The Melanesian cargo cult. Millenariansim as a factor in cultural change.
Akademisk Forlag DBK, Copenhagen.
Clendinnen, Inga
1987 Ambivalent conquests: Maya and Spaniard in Yucatan, 1517-1570. Cambridge
University Press, Cambridge.
Cobo, Bernabé
1956 Obras del P. Bernabé Cobo de la Compañía de Jesús estudio preliminar y edición
del P. Francisco Mateos de la misma compañia. Biblioteca de autores españoles ; t. 9192, Madrid.
1990
Inca religion and customs. University of Texas Press, Austin.
Cock, Guillermo Alberto Carrasco
1980 El sacerdote andino y los bienes de las divinidades en los siglos XVII y XVIII.
Pontificia Universidad Católica del Perú, Lima.
Cock, Guillermo, and Mary Eileen Doyle
1979 Del culto solar a la clandestinidad de Inti y Punchao. Historia y Cultura 12: 51–
73.
655
Collis, John
2004 Digging Up the Past: An Introduction to Archaeological Excavation. Sutton
Press, London.
Comision de la verdad y reconciliacion
2003 Informe Final.
Conklin, Beth A.
2001 Consuming grief : compassionate cannibalism in an Amazonian society.
University of Texas Press, Austin.
Cook, Anita G.
2001 Huari D-shaped structures, sacrificial offerings, and divine rulership. In Ritual
sacrifice in ancient Peru, edited by E. P. Benson and A. G. Cook, pp. 137–164.
University of Texas Press, Austin.
2015 The shape of things to come: the genesis of Wari wak’as. In The archaeology of
wak’as: explorations of the sacred in the Pre-Columbian Andes, edited by T. L. Bray, pp.
295–334. University Press of Colorado, Boulder.
Cook, Noble David
1981 Demographic collapse :indian Peru, 1520-1620. Cambridge University Press,
New York.
Cooper, Crispin H. V.
2017 Using Spatial Network Analysis to model pedal cycle flows, risk and mode
choice. Journal of Transport Geography 58(C): 157–165.
Cordy-Collins, Alana
2001 Decapitation in Cupisnique and early Moche societies. In Ritual sacrifice in
ancient Peru, edited by E. P. Benson and A. G. Cook, pp 21-34. University of Texas
Press, Austin.
Covey, Alan
N.d. Imperial appetites and altered states: the Spanish transformation of the Inca
heartland. In Andean foodways, edited by J. Staller and S. deFrance. University of
Alabama Press, Birmingham.
Crandall, Jj, and D. L. Martin
656
2014 Special section: the bioarchaeology of postmortem agency: integrating
archaeological theory with human skeletal remains. Cambridge Archaeological Journal
24(3): 429–435.
Crosby, Alfred W.
1972 The Columbian exchange: biological and cultural consequences of 1492.
Greenwood PubCo, Westport.
1986 Ecological imperialism: the biological expansion of Europe, 900-1900.
Cambridge University Press, Cambridge.
Curatola, Marco
1976 Mito y milenarismo en los Andes: del Taki Onqoy a Inkarri. Allpanchis
Phuturinqa 9: 65–92.
D’Altroy, Terence N.
1992 Provincial power in the Inka Empire. Smithsonian Institution Press, Washington.
2015
The Incas. Wiley Blackwell, Malden.
De Reu, J., J. Bourgeois, P. De Smedt, A. Zwertvaegher et al.
2011 Measuring the relative topographic position of archaeological sites in the
landscape, a case study on the Bronze Age barrows in northwest Belgium. Journal of
Archaeological Science 38: 3435-3446.
Deagan, Kathleen
1996 Colonial transformation: Euro-American cultural genesis in the early SpanishAmerican colonies. Journal of Anthropological Research 52:135-160.
2002 Artifacts of the Spanish colonies of Florida and the Caribbean, 1500-1800:
Volume 2: Portable Personal Possessions. Smithsonian Books, Washington D.C.
2004 Reconsidering Taíno social dynamics after Spanish conquest: gender and class in
culture contact studies. American Antiquity 69(4): 597–626.
Dean, Carolyn J.
1999 Inka bodies and the body of Christ: Corpus Christi in Colonial Cuzco, Peru.
Duke University Press Books, Durham.
2010 A culture of stone: Inka perspectives on rock. Duke University Press Books,
Durham.
657
2015 Men who would be rocks: the Inka wank’a. In The archaeology of wak’as:
explorations of the sacred in the Pre-Columbian Andes, edited by Tamara L. Bray, pp.
213–238. University Press of Colorado, Boulder.
DeFrance, Susan D.
1996 Iberian foodways in the Moquegua and Torata Valleys of Southern Peru.
Historical Archaeology 30(3): 20–48.
2003 Diet and provisioning in the high Andes: a Spanish colonial settlement on the
outskirts of Potosí, Bolivia. International Journal of Historical Archaeology 7(2): 99–
125.
DeFrance, Susan D. and Craig A. Hanson
2008 Labor, population movement, and food in sixteenth-century Ek Balam, Yucatán.
Latin American Antiquity 19(3): 299–316.
Defrance, Susan D., Steven A. Wernke, and Ashley E. Sharpe
2016 Conversion and persistence: analysis of faunal remains from an Early Spanish
Colonial doctrinal settlement in highland Peru. Latin American Antiquity 27(3): 300–317.
DeMarrais, Elizabeth
2014 Introduction: the archaeology of performance. World Archaeology 46(2): 155–
163.
2017 Animacy, abstraction, and affect in the Andean past: toward a relational approach
to art. Cambridge Archaeological Journal 27(4): 655-669.
Der, Lindsay and Francesca Fernandini
2016 Introduction. In Archaeology of entanglement, edited by L. Der and F. Fernandini,
pp 11-30. Left Coast Press, Walnut Creek.
Dietler, Michael
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 Gil Stein, pp. 33–68. School of
American Research Press, Santa Fe.
2010 Archaeologies of colonialism: consumption, entanglement, and violence in
ancient Mediterranean France. University of California Press, Berkeley.
Dillehay, Tom D., editor.
1995 Tombs for the living: Andean mortuary practices. Dumbarton Oaks Research
Library and Collection, Washington D.C.
658
Dobyns, Henry F.
1963 An outline of Andean epidemic history to 1720. Bulletin of the History of
Medicine 37(6): 493-515.
Donnan, Christopher B.
1978 Ancient burial patterns of the Moche Valley, Peru. University of Texas Press,
Austin.
Doyle, Mary Eileen
1988 The ancestor cult and burial ritual in seventeenth and eighteenth century Central
Peru. Ph.D. Thesis, University of California, Los Angeles.
Du Bois, Cora Alice
2007 The 1870 Ghost Dance. University of Nebraska Press, Lincoln.
Du Bois, W. E. B
1945 Color and democracy: colonies and peace. Harcourt, Brace and Co, New York.
Durston, Alan
2007 Pastoral Quechua: the history of Christian translation in Colonial Peru, 15501650. University of Notre Dame Press, Notre Dame.
Duviols, Pierre
1971 La lutte contre les religions autochtones dans Le Perou colonial: l’extirpation de
l’idolatrie, entre 1532 et 1660. Institut Francais d Etudes Andines, Paris.
1977 La destrucción de las religiones andinas: conquista y colonia. Universidad
Nacional Autónoma de México, Mexico.
1986 Cultura andina y represión: procesos y visitas de idolatrías y hechicerías
Cajatambo, siglo XVII. Centro de Estudios Rurales Andinos Bartolomé de las Casas,
Cusco, Peru.
Earle, Rebecca
2012a The Columbian exchange. In The Oxford handbook of food history, edited by J.
M. Pilcher, pp. 341–357. Oxford University Press, Oxford.
2012b The body of the conquistador: food, race and the colonial experience in Spanish
America, 1492-1700. Cambridge University Press, Cambridge.
Eastman, J. R.
1980 The Chicha archaeological project, geomorphological and geological report.
Unpublished manuscript.
659
Eire, Carlos M. N.
1995 From Madrid to Purgatory: the art and craft of dying in sixteenth-century Spain.
Cambridge University Press, Cambridge.
Engel, Zbyněk, Grzegorz Skrzypek, Tomáš Chuman, Luděk Šefrna, and Martin Mihaljevič
2014 Climate in the western cordillera of the central Andes over the last 4300 years.
Quaternary Science Reviews 99: 60–77.
Entwisle, Barbara
1997 Geographic Information Systems, Spatial Network Analysis, and contraceptive
choice. Demography 34(2): 171-187.
Espinoza Martinez, Hector
1984 Asentamientos Prehispanicos en la cuenca del Rio Chicha: hacia una
etnoarqueologia de los Soras, Ayacucho. Tesis para optar el Titulo de Licenciado en
Antropologia en la UNSAAC, Universidad Nacional de San Antonio Abad del Cuzco.
1995 Asentamientos Prehispanicos de la provincia Sucre-Ayacucho. Revista
Arqueologica Saqsaywaman 4.
Estenssoro, Juan Carlos
1992 Los bailes de los indios y el proyecto colonial. Revista Andina 10(2): 353-394.
2001 El simio de Dios: los indígenas y la iglesia frente a la evangelización del Perú,
siglos XVI-XVII. Bulletin de l’Institut Français d’Études Andines 30(3): 455–474.
2003 Del paganismo a la santidad: la incorporación de los indios del Perú al
catolicismo, 1532-1750. IFEA, Instituto Francés de Estudios Andinos : Pontificia
Universidad Católica del Perú, Instituto Riva-Agüero, Lima.
Fanon, Frantz
1952 Black Skin, White Masks. Paladin, London.
Finucane, Brian, Patricia Maita 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(12): 1766–1776.
Flint, Richard and Shirley Cushing Flint
2003 The Coronado expedition: from the distance of 460 years. University of New
Mexico Press, Albuquerque.
Foucault, Michel
660
1977
Discipline and punish: the birth of the prison. Pantheon Books, New York.
Fry, Robert E.
1985 Revitalization movements among the Postclassic lowland Maya. In The lowland
Maya Postclassic, edited by P. M. Rice and A. F. Chase, pp. 19–41. University of Texas
Press, Austin.
Gade, Daniel W.
2015 Particularizing the Columbian Exchange: Old World biota to Peru. Journal of
Historical Geography 48: 26–35.
Gagnon, Celeste Marie, C. Fred T. Andrus, Jennifer Ida, and Nicholas Richardson
2015 Local water source variation and experimental chicha de maíz brewing:
implications for interpreting human hydroxyapatite Δ18O values in the Andes. Journal of
Archaeological Science 4: 174–181.
Gallaga, Emiliano, Marc Gordon Blainey, and Karl A. Taube
2016 Manufactured light: mirrors in the Mesoamerican realm. Boulder: University
Press of Colorado, Boulder.
Garcilaso de la Vega
2000 Comentarios reales. Editorial Castalia, Madrid.
Gaughen, Shasta
2016 (R)Anthropology class: revitalization movements. Shasta Gaughen.
https://medium.com/@shasta71/r-anthropology-class-revitalization-movementsafabf800c30d.
Gazzola, Julie
2016 Identification and use of pyrite and hematite at Teotihuacan. In Manufactured
light: mirrors in the Mesoamerican Realm, edited by E. Gallaga and M. G. Blainey, pp.
107-124.
Gillings, Mark
2015 Mapping invisibility: GIS approaches to the analysis of hiding and seclusion.
Journal of Archaeological Science 62: 1-14.
2017 Mapping liminality: critical frameworks for the GIS-based modelling of visibility.
Journal of Archaeological Science 84: 121-128.
Given, Michael
2004 The archaeology of the colonized. Routledge: London.
661
Glowacki, Donna M.
2012 The role of religion in the depopulation of the central Mesa Verde region. In
Religious transformation in the Late Pre-Hispanic Pueblo world, edited by D. M.
Glowacki and S. Van Keuren, pp. 66–83. University of Arizona Press, Tucson.
Goffman, Erving
1959 The presentation of self in everyday life. Anchor Books, Doubleday, New York.
González Holguín, Diego
1989 Vocabulario de la lengua general de todo el Perú, llamada lengua qquichua o del
Inca. Universidad Nacional Mayor de San Marcos, Editorial de la Universidad, Lima.
Gordon, Claire C., and Jane E. Buikstra
1981 Soil PH, bone preservation, and sampling bias at mortuary sites. American
Antiquity 46(3): 566–571.
Gorriti, Gustavo
1999 The Shining Path: a history of the millenarian war in Peru. The University of
North Carolina Press, Chapel Hill.
Gosden, Chris
2004 Archaeology and colonialism : cultural contact from 5000 B.C. to the present.
Cambridge University Press, Cambridge.
Gosden, Chris, and Chantal Knowles
2001 Collecting colonialism: material culture and colonial change. Berg Publishers,
London.
Gose, Peter
2006 Converting the ancestors: indirect rule, settlement consolidation, and the struggle
over burial in Colonial Peru, 1532-1614. In Conversion: Old Worlds and New, edited by
K. Mills and A. Grafton, pp. 140–174. University Rochester Press, New York.
2008 Invaders as ancestors : on the intercultural making and unmaking of Spanish
colonialism in the Andes. University of Toronto Press, Toronto.
Grossman, Joel Warren
1972 Early ceramic cultures of Andahuaylas, Apurímac, Peru. Doctoral Thesis,
University of California Berkeley.
Guamán Poma de Ayala, Felipe
2015 Nueva coronica y buen gobierno. Biblioteca Nacional del Perú, Lima.
Guibovich Perez, Pedro
662
1990 Nota preliminar al personaje historico y los documentos. In El retorno de las
huacas: estudios y documentos sobre el Taki Onqoy, edited by L. Millones, pp. 23-40.
Instituto de Estudios Peruanos, Lima.
1991
Cristóbal de Albornoz y el Taki Onqoy. Historica 15: 205–36.
Gurevitz, Anna Inez
2017 Catholicism, negotiation and Taki Onqoy: how mortuary ritual and
biogeochemistry re-Members identity during the Early Colonial Period at Iglesiachayoq,
Chicha, Peru. Master’s Thesis, University of California, Merced.
Haas, Jonathan, Jane E. Buikstra, Douglas H. Ubelaker, David Aftandilian, and Field Museum of
Natural History
1994 Standards for data collection from human skeletal remains: proceedings of a
seminar at the Field Museum of Natural History, organized by Jonathan Haas. Arkansas
Archeological Survey Research Series ; No. 44. Arkansas Archeological Survey,
Fayetteville.
Haglund, William D., and Marcella H. Sorg
2001 Advances in forensic taphonomy: method, theory, and archaeological
perspectives. CRC Press
Harkin, Michael Eugene
2004 Reassessing revitalization movements: perspectives from North America and the
Pacific Islands. University of Nebraska Press, Lincoln.
Harrison, Regina
2014 Sin and confession in colonial Peru: Spanish-Quechua Penitential texts, 15601650. University of Texas Press, Austin.
Heaney, Christopher
2018 How to make an Inca mummy: Andean embalming, Peruvian science, and the
collection of empire. Isis 109(1): 1–27.
Heidegger, Martin
1971 Poetry, language, thought. Translated by A. Hofstadter. Harper, London.
1973
Being and Time. Blackwell, Oxford.
2002. The origin of the work of art. In Heidegger: off the beaten track, edited and
translated by J. Young and K. Haynes, pp. 1-56. Cambridge University Press, Cambridge.
Heilman, Jaymie
663
2002 A movement misconstrued? A response to Gabriela Ramos’s interpretation of
Taki Onqoy. Colonial Latin American Review 11(1): 123–138.
Hemming, John
1970 The conquest of the Incas. Harcourt, Brace, Jovanovich, New York.
Henare, Amiria, editor.
2006 Thinking through things. Routledge, New York.
Hernández Garavito, Carla, and Carlos Osores Mendives
2019a Colonialism and domestic life: identities and foodways in Huarochirí during the
Inka Empire. International Journal of Historical Archaeology.
Hodder, Ian
2000 Towards reflexive method in archaeology : the example at Çatalhöyük. McDonald
Institute for Archaeological Research, University of Cambridge ; Distributed by Oxbow
Books, Cambridge.
2012 Entangled: an archaeology of the relationships between humans and things.
Wiley-Blackwell, Malden.
Horswell, Michael J.
2006 Decolonizing the Sodomite: Queer Tropes of Sexuality in Colonial Andean
Culture. annotated edition edition. Austin: University of Texas Press.
Howey, Meghan C. L.
2011 Multiple pathways across past landscapes: circuit theory as a complementary
geospatial method to Least Cost Path for modeling past movement. Journal of
Archaeological Science 38(10): 2523–2535.
Hyslop, John
1990
Inka settlement planning. University of Texas Press, Austin.
Inomata, Takeshi, and Lawrence A. Coben
2006 Archaeology of performance: theaters of power, community, and politics.
Archaeology in Society Series, AltaMira Press, Lanham.
Insoll, Timothy
1999 The archaeology of Islam: social archaeology. Blackwell Publishers, Oxford.
2001
Archaeology of world religion. Routledge, London.
2004
Archaeology, ritual, religion. Routledge, London.
664
Instituto de Geología y Minería
1975 Mapa geológico del Perú. El Instituto, Lima.
Isbell, William Harris
1977 The rural foundation for urbanism: economic and stylistic interaction between
rural and urban communities in eighth-century Peru. University of Illinois Press, Urbana.
1997 Mummies and mortuary monuments : a postprocessual prehistory of Central
Andean social organization. University of Texas Press, Austin.
Isbell, William H., and Antti Korpisaari
2012 Burial in the Wari and the Tiwanaku heartlands: similarities, differences, and
meanings. Diálogo Andino (39): 91–122.
Isbell, William Harris, Gordon Francis McEwan, and Dumbarton Oaks, editors.
1991 Huari administrative structure: prehistoric monumental architecture and state
government. Dumbarton Oaks Research Library and Collection, Washington D.C.
Itier, Cesar
1991 Lengua general y comunicacion escrita: cinco cartas en Quechua de Cotahuasi,
1616. Revista Andina 9(1): 1–21.
Jacobi, Keith P.
2000 Last rites for the Tipu Maya: genetic structuring in a colonial cemetery.
University of Alabama Press Tuscaloosa.
Jamieson, Ross W.
2002 Domestic architecture and power the historical archaeology of Colonial Ecuador.
Kluwer Academic Publishers, New York.
Janusek, John W.
2015 Of monoliths and men: human-lithic encounters and the production of an
animistic ecology at Khonkho Wankane. In The archaeology of wak’as: explorations of
the sacred in the Pre-Columbian Andes, edited by T. L. Bray, pp. 335–368. University
Press of Colorado, Boulder.
Jennings, Justin
2011 Globalizations and the ancient world. Cambridge University Press, Cambridge.
Jordan, Kurt A.
2014 Prunin colonialism: vantage point, local political economy, and cultural
entanglement in the archaeology of post-1415 indigenous peoples. In Rethinking colonial
pasts through archaeology, edited by N. Ferris, R. Harrison, and M. Wilcox, pp. 103-120.
Oxford University Press, Oxford.
665
Joyce, Rosemary A.
1999 Social dimensions of Pre-Classic burials. In Social patterns in Pre-Classic
Mesoamerica: a symposium at Dumbarton Oaks, edited by D. C. Grove and R. A. Joyce,
pp. 15–47. Dumbarton Oaks Research Library and Collection, Washington D. C.
2017 Sex, gender, and anthropology: moving bioarchaeology outside the subdiscipline,
In Exploring sex and gender in bioarchaeology, edited by S. C. Agarwal and J. K. Wesp,
pp. 1–14. University of New Mexico Press, Albuquerque.
Kaulicke, Peter
2003 Memoria historiografiada y memoria materializada: problemas en la percepción
del pasado andino preeuropeo. Estudios atacameños (26): 17–34.
Keeley, H., and F. Meddens
1992 Prehispanic agricultural terrace systems in the Chica-Soras Valley, Peru. Bulletin
of the Institute of Archaeology 29: 121.
Kehoe, Alice Beck
1989 The Ghost Dance: ethnohistory and revitalization. Holt, Rinehart, and Winston,
New York.
Kellett, Lucas
2010 Chanka settlement ecology: hilltop sites, land use and warfare in Prehispanic
Andahuaylas, Peru. Ph.D. Thesis, University of Illinois, Chicago.
Kellett, Lucas C., Mark Golitko and Brian S. Bauer
2013 Prehistoric obsidian procurement strategies in the Andahuaylas region of
Southern Peru. Journal of Archaeological Science 40: 1890-1902.
Kelloway, Sarah J., Parker VanValkenburgh, Javier G. Iñañez, et al.
2018 Identifying New World majolica from 16th–18th century Sites on Peru’s North
Coast. Journal of Archaeological Science: Reports 17: 311–324.
Kemp, Rob, Nick Branch, Barbara Silva, et al.
2006 Pedosedimentary, cultural and environmental significance of paleosols within
Pre-Hispanic agricultural terraces in the southern Peruvian Andes. Quaternary
International 158(1): 13–22.
Kennedy, Sarah A., and Parker VanValkenburgh
2016 Zooarchaeology and changing food practices at Carrizales, Peru following the
Spanish invasion. International Journal of Historical Archaeology 20(1): 73–104.
Klaus, Haagen D.
666
2008 Out of light came darkness: bioarchaeology of mortuary ritual, health, and
ethnogenesis in the Lambayeque Valley Complex, North Coast Peru (AD 900-1750).
Ph.D. Thesis, The Ohio State University.
2009 Contact in the Andes: bioarchaeology of systemic stress in Colonial Mórrope,
Peru. American Journal of Physical Anthropology 138(3): 356–68.
2013 Hybrid cultures…and hybrid peoples: bioarchaeology of genetic change, religous
architecture, and burial ritual in the Colonial Andes. In The archaeology of hybrid
material culture, edited by J. J. Card, pp. 207–238. Southern Illinois Univdersity Press,
Carbondale.
Klepinger, L. L., R. W. Mintel, J. S. Olin, and M. J. Blackman
1984 Metabolic considerations in reconstructing past diet from stable carbon isotope
ratios of bone collagen. Smithsonian Institution, Washington D.C.
Knudson, Kelly J.
2009 Oxygen isotope analysis in a land of environmental extremes: the complexities of
isotopic work in the Andes. International Journal of Osteoarchaeology 19(2): 171–191.
Knudson, Kelly J., and Christina Torres‐Rouff
2009 Investigating cultural heterogeneity in San Pedro de Atacama, Northern Chile,
through biogeochemistry and bioarchaeology. American Journal of Physical
Anthropology 138(4): 473–485.
Knudson, Kelly J., W. J. Pestle, C. Torres‐Rouff, and G. Pimentel
2012 Assessing the life history of an Andean traveller through biogeochemistry: stable
and radiogenic isotope analyses of archaeological human remains from Northern Chile.
International Journal of Osteoarchaeology 22(4): 435–451.
Kohut, Lauren
2016 Cooperation through war: Late Intermediate Period warfare and community
formation in the South-Central Andes. Ph.D. Thesis, Vanderbilt University.
Kosiba, Steve
2012 Emplacing value, cultivating order: places of conversion and practices of
subordination throughout early Inka state formation (Cusco, Peru). In The construction of
value in the ancient world, edited by J. K. Papadopoulos and G. Urton, pp. 99–127.
Cotsen Institute of Archaeology Press, Los Angeles.
2015 Of blood and soil: tombs, wak’as, and the naturalization of social difference in the
Inka heartland. In The archaeology of wak’as: explorations of the sacred in the Pre-
667
Columbian Andes, edited by T. L. Bray, pp. 167–212. University Press of Colorado,
Boulder.
Kubler, George
1946 The Quechua in the colonial world. Smithsonian Institution, Washington, D.C.
Kurin, Danielle Shawn
2012 The bioarchaeology of collapse: ethnogenesis and ethnocide in Post-Imperial
Andahuaylas, Peru (AD 900-1250). Ph.D. Thesis, Vanderbilt University.
2016 The bioarchaeology of societal collapse and regeneration in ancient Peru.
Springer, Switzerland.
Lake, M. and D. Ortega
2013 Compute-intensive GIS visibility analysis of the settings of prehistoric stone
circles. In Computational approaches to archaeological space, edited by A. Bevan and
M. Lake, pp. 213-242. Left Coast Press, Walnut Creek.
Lara, Jaime
2008 Christian texts for Aztecs: art and liturgy in Colonial Mexico. University of Notre
Dame Press, Notre Dame.
2016 Birdman of Assisi: art and the apocalyptic in the Colonial Andes. ACMRS
Arizona Center for Medieval and Renaissance Studies and Bilingual Press/Editorial
Bilingüe, Tempe.
Larsen, Clark Spencer
1993 On the frontier of contact: mission bioarchaeology in La Florida. In The Spanish
missions of La Florida, edited by Bonnie G. McEwan, pp. 322–356. University Press of
Florida, Gainesville.
Laour, Bruno
1999 On recalling ANT. The sociological review 47(S1): 15-25.
2005 Reassembling the social: an introduction to Actor-Network-Theory. Oxford
University Press, Oxford.
Lau, George
2015 The dead and the longue duree in Peru’s North Highlands. In Living with the dead
in the Andes, edited by I. Shimada and J. L. Fitzsimmons, pp. 200–244. University of
Arizona Press: Tucson.
Lechtman, Heather
668
1984 Andean value systems and the development of prehistoric metallurgy. Technology
and Culture 25(1): 1.
1993 Technologies of power: the Andean case. In Configurations of power: holistic
anthropology in theory and practice, edited by J. S. Henderson, P. Netherly, and J. V.
Murra, pp. 244–280. Cornell University Press, Ithaca.
Lee-Thorp, Julia A., Judith C. Sealy, and Nikolaas J. van der Merwe
1989 Stable carbon isotope ratio differences between bone collagen and bone apatite,
and their relationship to diet. Journal of Archaeological Science 16(6): 585–599.
Lemlij, Moises, and Cristóbal de Albornoz
1990 El Taki Onqoy: reflexiones psicoanaliticas. In El Retorno de las huacas: estudios
y documentos sobre el Taki Onqoy, siglo XVI, edited by Luis Millones, pp. 425–34.
Instituto de Estudios Peruanos : Sociedad Peruana de Psicoanálisis, Lima.
León, Pedro de Cieza de, and Carmelo Sáenz de Santa María
1985 Obras completas. Consejo Superior de Investigaciones Cientificas, Lima.
Liebmann, Matthew
2007 The archaeology of the Pueblo Revolt and the formation of the modern Pueblo
world. The Kiva. 73(2): 195–218.
2008 The innovative materiality of revitalization movements: lessons from the Pueblo
Revolt of 1680. American Anthropologist 110(3): 360–372.
2012 Revolt: an archaeological history of Pueblo resistance and revitalization in 17th
century New Mexico. University of Arizona Press: Tucson.
Liebmann, Matthew, and Melissa Scott Murphy
2011 Rethinking the archaeology of “Rebels, Backsliders, and Idolaters.” In Enduring
conquests : rethinking the archaeology of resistance to Spanish colonialism in the
Americas, edited by M. Liebmann and M. S. Murphy, pp. 3–18. School for Advanced
Research Press, Santa Fe.
Linton, Ralph
1943 Nativistic movements. American Anthropologist 45: 230.
Little, Keith J.
2010 Sixteenth-century glass bead chronology in southeastern North America.
Southeastern Archaeology 29(1): 222–232.
669
Llobera, Marcos
2003 Extending GIS-based visual analysis: the concept of Visualscapes. International
Journal of Geographical Information Science 17(1): 25–48.
2007
Reconstructing visual landscapes. World Archaeology 39(1): 51–69.
Lohmann Villena, G.
1941 El Inca Titu Cusi Yupanqui y su entrevista con el Oidor Matienzo (1565).
Mercurio Peruano 166: 3–18.
Loomba, Ania
1998 Colonialism-postcolonialism. Routledge, New York.
Lumbreras, Luis
1959
Sobre los Chancas. Congreso Nacional de Historia Del Peru 1: 211–242.
2000
El imperio Wari. IFEA-Inst.Francés de Estudios Andinos, Paris.
Maccormack, Sabine
1985 “The heart has its reasons”: predicaments of missionary Christianity in early
colonial Peru. Hispanic American Historical Review 65(3): 443–466.
1988 Pachacuti: miracles, punishments, and last judgment: visionary past and prophetic
future in early colonial Peru. The American Historical Review 93(4): 960–1006.
1991 Religion in the Andes: vision and imagination in early colonial Peru. Princeton
University Press, Princeton.
Makowski, Krzysztof
2015 Pachacamac--old wak’a or Inka syncretic deity? Imperial transformation of the
sacred landscape in the lower Ychsma (Lurin) valley. In The archaeology of wak’as:
explorations of the sacred in the Pre-Columbian Andes, edited by T. L. Bray, pp. 127–
166. University Press of Colorado, Boulder.
Mallco, Rafael
2013 Ocupación Prehispanica en el area del distrito de San Pedro de Larcay, Sucre Ayacucho. Arqueología y Sociedad 25: 215–258.
Mannheim, Bruce
1986 The language of reciprocity in Southern Peruvian Quechua. Anthropological
Linguistics, 28(3): 267-273.
1991 The language of the Inka since the European invasion. University of Texas Press,
Austin.
670
1992 A nation surrounded. In Native traditions in the postconquest world : a
symposium at Dumbarton Oaks, 2nd through 4th October 1992, pp. 383–420. Dumbarton
Oaks Research Library and Collection, Washington D.C.
Mannheim, Bruce, and Carreno, Guillermo Salas
2015 Wak’as: entifications of the Andean sacred. In The archaeology of wak’as:
explorations of the sacred in the Pre-Columbian Andes, edited by T. L. Bray, pp. 47–74.
University Press of Colorado, Boulder.
Martin, Joel
2004 Visions of revitalization in the eastern Woodlands. In Reassessing revitalization
movements: perspectives from North America and the Pacific Islands, edited by M. E.
Harkin, pp. 61–87. University of Nebraska Press, Lincoln.
Mazzotti, José A.
2008 Garcilaso de La Vega, El Inca (1539-1616). In Guide to Documentary Dources
for Andean Studies 1530-1900, edited by J. Pillsbury, pp. 229–241. University of
Oklahoma Press, Norman.
McEwan, Bonnie G.
1993 The Spanish missions of La Florida. University Press of Florida, Gainseville.
McEwan, Gordon F.
1984 The Middle Horizon in the valley of Cuzco, Peru: the impact of the Wari
occupation of Pikillacta in the Lucre Basin (Andes). Ph.D. Thesis, University of Texas at
Austin.
2006
The Incas: new perspectives. W. W. Norton & Company, New York.
McMullen, Ann
2004 “Canny about conflict”: nativism, revitalization, and the invention of tradition in
native southeastern New England. In Reassessing revitalization movements, edited by M.
E. Harkin, pp. 267-278. University of Nebraska Press, Lincoln.
Meddens, Frank
1985 The Chicha/Soras Valley during the Middle Horizon: provincial aspects of Huari.
Ph.D. Thesis, Institute of Archaeology of the University of London.
1994 Mountains, miniatures, ancestors, and fertility: the meaning of a Late Horizon
offering in a Middle Horizon structure in Peru. Institute of Archaeology Bulletin 31: 127.
2001 La sequencia cultural en la cuenca de Chicha-Soras. In XII Congreso Peruano,
“Luis Lumbreras”, edited by I. Perez, C. W. Aguilar, and M. Purizaga. Universidad de
San Cristobal de Huamanga, Vol 2, pp. 200-210.
671
2006 Rocks in the landscape: managing the Inka agricultural cycle. The Antiquaries
Journal 86: 36–65.
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 Springer New York. In Pre-Columbian landscapes of
creation and origin, edited by J. E. Staller, pp. 315-355.
Meddens, Frank M., and Cirilo Vivanco Pomacanchari
2002 The Chanca confederation; political myth and archaeological reality. Xama 15–
18: 72–86.
2018 The Late Intermediate Period ceramic traditions of Ayacucho, Apurímac, and
Huancavelica: current thoughts on the Chanca and other regional polities. Ñawpa Pacha
38(1): 3–56.
Meddens, Frank. M, and Katharina Schreiber
2010 Inca strategies of control: a comparison of the Inca occupations of Soras and
Andamarca Lucanas. Ñawpa Pacha 30(2).
Memmi, Albert
1965 The colonizer and the colonized. Orion Press, New York.
Menzel, Dorothy
1968 La cultura Huari. Peruano-Suiza, Lima.
Meskell, Lynn, editor.
2005 Archaeologies of materiality. Wiley-Blackwell, Malden.
Millones, Luis
1964 Un movimiento nativista del Siglo XVI: el Taki Onqoy. Revista Peruana de
Cultura 3: 134–140.
1973 Un movimiento nativista del Siglo XVI: el Taki Onqoy. In Ideologia mesianica
del mundo Andino, edited by J. Ossio, pp. 83–94. Ignacio Prado Pastor, Lima.
2007
39.
Mesianismo en América Hispana: el Taki Onqoy. Memoria Americana (15): 7–
2008 Cristóbal de Albornoz. In Guide to documentary sources for Andean studies,
1530-1900 Vol. II, A-L, edited by J. Pillsbury. University of Oklahoma Press, Norman.
Millones, Luis, Sara Castro-Klarén, and Cristóbal de Albornoz
672
1990 El retorno de las huacas: estudios y documentos sobre el Taki Onqoy, siglo XVI.
Instituto de Estudios Peruanos : Sociedad Peruana de Psicoanálisis, Lima.
Millones, Luis, and Hiroyasu Tomeda
1998 El mundo del color y el movimiento: de los takis Precolombinos a los Danzantes
de Tijeras. In Historia, religion, y ritual de los pueblos Ayacuchanos. National Museum
of Ethnology, Osaka.
Mills, Kenneth
1997 Idolatry and its enemies: colonial Andean religion and extirpation, 1640-1750.
Princeton University Press, Princeton.
Molina, Cristóbal de
2010 Relación de las fábulas y ritos de los incas. Parecos y australes no. 7. Madrid :
Frankfurt am Main: Iberoamericana, Madrid.
Molina, Cristóbal de, and Brian Bauer
2011 Account of the fables and rites of the Incas. University of Texas Press, Austin.
Monzón, Luis de
1965 Descripción de la tierra del repartimiento de Atunsora, encomendado en
Hernando Palomino, Jurisdición de la ciudad de Guamanga, año de 1586. In Relaciones
geograficas de Indias – Peru, edited by M. Jimenez de la Espada, pp. 220–225.
Biblioteca de autores espanoles, Madrid.
Moore, Jerry D.
1996 Architecture and power in the ancient Andes: the archaeology of public buildings.
Cambridge University Press, Cambridge.
Morris, Craig
1985 Huánuco Pampa: an Inca city and its hinterland. Thames and Hudson, London.
Morris, Craig, R. Alan Covey, and Pat H Stein
2011 The Huánuco Pampa archaeological project. Volume 1, the plaza and palace
complex. American Museum of Natural History, New York.
Mumford, Jeremy
1998 The Taki Onqoy and the Andean nation: sources and interpretations. Latin
American Research Review 33(1): 150–165.
2012 Vertical empire: the general resettlement of Indians in the colonial Andes. Duke
University Press, Durham.
Murphy, Melissa Scott, Elena Goycochea, and Guillermo Cock
673
2011 Resistance, persistence, and accommodation at Puruchuco-Huaquerones, Peru. In
Enduring conquests : rethinking the archaeology of resistance to Spanish colonialism in
the Americas, edited by M. Liebmann and M. Murphy, pp. 57–76. School for Advanced
Research Press, Santa Fe.
Niles, Susan A
1999 The shape of Inca history : narrative and architecture in an Andean empire.
University of Iowa Press, Iowa City.
Norman, Scotti M. and Sarah A. Kennedy
2019 Status and Identity in the Imperial Andes. In International Journal for Historical
Archaeology, edited volume.
Obeyesekere, Gananath
1992 The apotheosis of Captain Cook : European mythmaking in the Pacific. Princeton
University Press, Princeton.
Ogburn, Dennis E.
2006 Assessing the level of visibility of cultural objects in past landscapes. Journal of
Archaeological Science 33(3): 405–413.
Ohnuki-Tierney, Emiko
1990 Culture through time : anthropological approaches. Stanford University Press,
Stanford.
Ortiz, Alejandra, Melissa S. Murphy, Jason Toohey, and Catherine Gaither
2017 Hybridity? Change? Continuity? Survival? Biodistance and the identity of
colonial burials from Magdalena de Cao Viejo, Chicama Valley, Peru. In Colonized
bodies, worlds transformed: toward a global bioarchaeology of contact and colonialism,
edited by M. S. Murphy and H. D. Klaus, pp. 375-410. University Press of Florida,
Gainseville.
Ortner, Sherry B
1995 Resistance and the problem of ethnographic refusal. Comparative Studies in
Society and History : An International Quarterly 37: 173–193.
Pacheco Torres, Víctor R.
1986 The osteology of South American camelids. Cotsen Institute of Archaeology,
University of California, Los Angeles.
Pankowská, Anna
2018 Diachronic variation in secondary burial practices in Bronze and Iron Age
Moravia. Journal of Archaeological Science: Reports 21: 460–471.
674
Parker Pearson, Michael
1982 Mortuary practices, society and ideology: an ethnoarchaeological study. In
Symbolic and structural archaeology, edited by I. Hodder, pp. 99–113. Cambridge
University Press, Cambridge.
Parno, Travis G., and Mary C. Beaudry
2014 Archaeologies of mobility and movement. Springer, New York.
Patiño Paúl, Mariano
2001 Huancavelica colonial: apuntes históricos de la ciudad minera más importante
del Virreynato peruano. 1. ed. Huancavelica: El Autor.
Pease, Franklin
1973 El dios creador andino. Lima: Mosca Azul Editores.
Pillsbury, Joanne
2008 Guide to documentary sources for Andean studies, 1530-1900 Vol. 1 Norman:
Univ. of Oklahoma Press.
Price, T. Douglas, Clark M. Johnson, Joseph A. Ezzo, Jonathan Ericson, and James H. Burton
1994 Residential mobility in the prehistoric southwest United States: a preliminary
study using strontium isotope analysis. Journal of Archaeological Science 21(3): 315–
330.
Protzen, Jean-Pierre
1993 Inca architecture and construction at Ollantaytambo. Oxford University Press,
Oxford.
Puckle, Bertram S.
1926 Funeral customs; their origin and development. TWLaurie, ltd., London.
Quilter, Jeffrey
2011 Cultural encounters at Magdalena de Cao Viejo in the Early Colonial Period. In
Enduring conquests : rethinking the archaeology of resistance to Spanish colonialism
in the Americas, edited by M. Liebmann and M. Murphy, pp. 103–126. School for
Advanced Research Press, Santa Fe, New Mexico.
Rakita, Gordon F. M.
2005 Interacting with the dead: perspectives on mortuary archaeology for the new
millennium. University Press of Florida, Gainseville.
Ramírez, Susan Elizabeth
675
2005 To feed and be fed: the cosmological bases of authority and identity in the Andes.
Stanford University Press, Stanford.
Ramos, Gabriela
1992 Política eclesiástica y extirpación de la idolatría: discursos y silencios en torno al
Taqui Onqoy. Revista Andina 10: 147–69.
2002 Política eclesiástica, cultura e historia: Cristóbal de Albornoz y el Taqui Onqoy,
otra vez. Colonial Latin American Review 11(1): 139–145.
2010 Death and conversion in the Andes : Lima and Cuzco, 1532-1670. University of
Notre Dame Press, Notre Dame.
Regalado de Hurtado, Liliana
1993 La sucesión incaica : aproximación al mando y poder entre los incas a partir de
la crónica de Betanzos. Pontif. Univ. Católica del Perú, Lima.
Reitz, Elizabeth J., and Elizabeth S. Wing
2008 Zooarchaeology. 2 edition. Cambridge University Press, New York.
Renfrew, Colin, Steven Mithen, Mary Weismantel, and Rosemary Joyce
2011 Religion in the emergence of civilization: Catalhoyuk as a case study. Current
Anthropology 52(6): 915–920.
Restall, Matthew
2003 Seven myths of the Spanish conquest. Oxford University Press, New York.
Rice, Prudence M
1994 The kilns of Moquegua, Peru: technology, excavations, and functions. Journal
of Field Archaeology 21(3): 325–344.
1996 The archaeology of wine: the wine and brandy haciendas of Moquegua, Peru.
Journal of Field Archaeology. 23(2): 187.
2012a Torata Alta : an Inka administrative center and Spanish colonial reducción in
Moquegua, Peru. Latin American Antiquity 23(1): 3–28.
2012b Vintage Moquegua : history, wine, and archaeology on a colonial Peruvian
periphery. University of Texas Press, Austin.
Rice, Prudence M, and Sara L Van Beck
1993 The Spanish colonial kiln tradition of Moquegua, Peru. Historical Archaeology
27(4): 65–81.
676
Rosenfeld, Silvana A., and Stefanie L. Bautista
2017 Rituals of the past: prehispanic and colonial case studies in Andean
archaeology. University Press of Colorado, Boulder.
Rostworowski de Diez Canseco, María
2005 Ensayos de historia andina. 2. ed. Obras completas de María Rostworowski ;
vols. 5-6. Instituto de Estudios Peruanos, Lima.
Rowe, John Howland
1944 An introduction to the archaeology of Cuzco. Peabody Museum, Harvard
University, Cambridge.
1979
An account of the shrines of ancient Cuzco. Ñawpa Pacha 17: 1–80.
2006 The Inca civil war and the establishment of Spanish power in Peru. Ñawpa
Pacha 28: 1–9.
Sahlins, Marshall David
1981 Historical metaphors and mythical realities: structure in the early history of the
Sandwich Islands kingdom. University of Michigan Press, Ann Arbor.
1985
Islands of history. University of Chicago Press, Chicago.
2004 Apologies to Thucydides : understanding history as culture and vice versa.
University of Chicago Press, Chicago.
Said, Edward W
1978 Orientalism. Vintage Books, New York.
Saignes, Thierry
1999 The colonial condition in the Quechua-Aymara heartland (1570-1780). In South
America. Vol. 3, Pt. 2, edited by F. Salomon, pp. 59–137. Cambridge University Press,
Cambridge.
Salomon, Frank
1982 Chronicles of the impossible: notes on three Peruvian indigenous historians.
Syracuse University, New York.
1991 Introductory essay: the Huarochirí manuscript. In The Huarochirí manuscript : a
testament of ancient and colonial Andean religion, pp. 1–38. University of Texas Press,
Austin.
677
1995 “The beautiful grandparents”: Andean ancestor shrines and mortuary ritual as
seen through colonial records. In Tombs for the living: Andean mortuary practices, edited
by T. D. Dillehay, pp. 315–353. Dumbarton Oaks, Washington, D.C.
Salomon, Frank and Sabine Hyland
2010 Graphic pluralism : Native American systems of inscription and the colonial
situation. Duke University Press, Durham.
Salomon, Frank, and Mercedes Niño-Murcia
2011 The lettered mountain : a Peruvian village’s way with writing. Duke University
Press, Durham.
Salomon, Frank, Jorge Urioste, and Francisco de Avila
1991 The Huarochirí manuscript : a testament of ancient and colonial Andean religion.
University of Texas Press, Austin.
Sancho, Pedro
2012 An account of the conquest of Peru. Forgotten Books, Charleston.
Sandefur, Elsie Catherine
1988 Andean zooarchaeology: animal use and the Inka conquest of the upper Mantaro
Valley. University of California, Los Angeles.
Sarmiento de Gamboa, Pedro de, Brian S. Bauer, and Vania Smith
2007 The history of the Incas. University of Texas Press, Austin.
Sarmiento de Gamboa, Pedro, and Angel Rosenblat
1952 Historia de los incas. Emecé editores, Buenos Aires.
Schechner, Richard
2003 Performance theory. 1 edition. Routledge, London.
Schoeninger, Margaret J., Michael J. DeNiro, and Henrik Tauber
1983 Stable Nitrogen isotope ratios of bone collagen reflect marine and terrestrial
components of prehistoric human diet. Science 220(4604): 1381–1383.
Schoeninger, Margaret J., Katherine M. Moore, Matthew L. Murray, and John D. Kingston
1989 Detection of bone preservation in archaeological and fossil samples. Applied
Geochemistry 4(3): 281–292.
Schroeder, Sissel
2001 Secondary disposal of the dead: cross-cultural codes. World Cultures 12(1): 7793.
678
Scott, James C
1985 Weapons of the weak : everyday forms of peasant resistance. Yale University
Press, New Haven.
1990 Domination and the arts of resistance: hidden transcripts. Yale University Press,
New Haven.
Seed, Patricia
1991 “Failing to marvel”: Atahualpa’s encounter with the word. Latin American
Research Review 26(1): 7–32.
Seligmann, Linda
1987 The chicken in Andean history and myth: the Quechua concept of “Wallpa.”
Ethnohistory 34(2): 139–170.
Sempat Assadourian, Carlos
1992 The colonial economy: the transfer of the European system of production to New
Spain and Peru. Journal of Latin American Studies 24: 55–55.
Sewell, William H.
2005 Logics of history: social theory and social transformation. University of Chicago
Press, Chicago.
Shimada, Izumi, James L. Fitzsimmons, Nicole Hayward, and Catherine J. Allen
2015 Living with the dead in the Andes. The University of Arizona Press, Tuscon.
Siikala, Jukka
2004 Priests and prophets: the politics of voice in the Pacific. In Reassessing
revitalization movements: perspectives from North America and the Pacific Islands,
edited by M. E. Harkin, pp. 88–103. University of Nebraska Press, Lincoln.
Silliman, Stephen
2001 Agency, practical politics and the archaeology of culture contact. Journal of
Social Archaeology 1(2): 190–209.
2005 Culture contact or colonialism? Challenges in the archaeology of Native North
America. American Antiquity 30(1): 75-94.
2010
Indigenous traces in colonial spaces. Journal of Social Archaeology 10(1): 28–58.
679
2016 Disentangling the archaeology of colonialism and indigeneity. In The archaeology
of entanglement, edited by F. Fernandini and L. Der, pp.31-38. Left Coast Press, Walnut
Creek.
Silverblatt, Irene
1987 Moon, sun, and witches: gender ideologies and class in Inca and Colonial Peru.
Princeton University Press, Princeton.
Silverblatt, Irene, Ana Sanchez, and Sabine Maccormack
1993 Amancebados, hechiceros, y rebeldes. Chancay, Siglo XVII. The Hispanic
American Historical Review 73(1): 157.
Sisson, Septimus, James Daniels Grossman, and Robert Getty
1975 Sisson and Grossman’s the anatomy of the domestic animals. W B Saunders Co,
Philadelphia.
Smith, Marvin T.
1982 Early sixteenth century glass beads in the Spanish colonial trade. Cottonlandia
Museum Publications, Greenwood.
Smit, Douglas Karel
2018 Mercury and the making of the Andean market: an archaeological study of
indigenous labor in colonial Peru. Ph.D. Dissertation, University of Illinois, Chicago.
Spalding, Karen
1984 Huarochirí, an Andean Society under Inca and Spanish rule. Stanford University
Press, Stanford.
Spicer, E. H.
1961 Perspectives in American Indian culture change. University of Chicago Press,
Chicago.
Spivak, Gayatri Chakravorty, Donna Landry, and Gerald M MacLean
1996 The Spivak reader: selected works of Gayatri Chakravorty Spivak. Routledge,
New York.
Stein, Gil
2005 The archaeology of colonial encounters : comparative perspectives. School of
American Research Press, Santa Fe.
Stern, Steve J.
1993 Peru’s Indian peoples and the challenge of Spanish conquest: Huamanga to 1640.
University of Wisconsin Press, Madison.
680
Stojanowski, Christopher M.
2013 Mission cemeteries, mission peoples: historical and evolutionary dimensions of
intracemetery bioarchaeology in Spanish Florida. University Press of Florida,
Gainesville.
Swenson, Edward R.
2014 The materialities of place-making in the ancient Andes: a critical appraisal of the
ontological turn in archaeological interpretation. Journal of Archaeological Method and
Theory, 22(3): 677-712.
Talavera de la Vega, C, and C. B. Benoki
2008 Informe final del Proyecto de Investigación Arqueológica con excavaciónes en la
Iglesia de la Companía de Jesus de Huamanga, dentro del marco del Proyecto Integral de
Restauración de la Iglesia de la Companía de Jesus de Ayacucho. Unpublished site
report, National Ministry of Culture, Lima, Peru.
Taliaferro, Matthew S., Bernard A. Schriever, and M. Steven Shackley
2010 Obsidian procurement, Least Cost Path analysis, and social interaction in the
Mimbres area of southwestern New Mexico. Journal of Archaeological Science 37(3):
536–548.
Taylor, Anne Christine
1996 The soul’s body and its states: an Amazonian perspective on the nature of being
human. The Journal of the Royal Anthropological Institute 2(2): 201.
Taylor, Lewis
2006 Shining Path: guerrilla war in Peru’s northern highlands. Liverpool University
Press, Liverpool.
Terrell, Author John
N.d. America’s Latest “Cargo Cult”? Science Dialogues.
http://sciencedialogues.com/articles/social/americas-latest-cargo-cult/
Thornton, Russell
1981 Demographic antecedents of a revitalization movement: population change,
population size, and the 1890 Ghost Dance. American Sociological Review 46(1): 88–96.
1982 Demographic antecedents of tribal participation in the 1870 Ghost Dance
Movement. American Indian Culture and Research Journal 6: 79–91.
1986 We shall live again: the 1870 and 1890 Ghost Dance Movements as demographic
revitalization. Cambridge University Press, New York.
681
1990 Boundary dissolution and revitalization movements: the case of the nineteenthcentury Cherokees. Ethnohistory 40(3): 359–383.
Tiballi, Anne
2010 Gender, bodies, and technologies in the Inka Empire: the archaeological materials
from the Cemetery of the Sacrificed Women, Pachacamac, Peru. Ph.D. Dissertation,
Binghamton University.
Tieszen, Larry L., and Tim Fagre
1993 Carbon isotopic variability in modern and archaeological maize. Journal of
Archaeological Science 20(1): 25–40.
Tilley, Cristopher
2010 Interpreting landscapes: geologies, topographies, identities; explorations in
landscape phenomenology 3. Left Coast Press, Walnut Creek.
Toledo, Francisco de, Guillermo Lohmann Villena, and María Justina Sarabia Viejo
1986 Francisco de Toledo: disposiciones gubernativas para el Virreinato del Perú.
Publicaciones de la Escuela de Estudios Hispano-Americanos de la Universidad de
Sevilla, Sevilla.
Torres‐Rouff, Christina, Kelly J. Knudson, William J. Pestle, and Emily M. Stovel
2015 Tiwanaku influence and social inequality: a bioarchaeological, biogeochemical,
and contextual analysis of the Larache Cemetery, San Pedro de Atacama, Northern Chile.
American Journal of Physical Anthropology 158(4): 592–606.
Tripcevich N, and Steven Wernke
2010 On-site recording of excavation data using mobile GIS. Journal of Field
Archaeology 35(4): 380–397.
Tung, Tiffiny 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(3):
294–308.
Tung, Tiffiny, John Crandall, and Debra Martin
2014 Agency, ’til death do us part? Inquiring about the agency of dead bodies from the
ancient Andes. Cambridge Archaeological Journal 24(3): 437–452.
Turner, Bethany L., George D. Kamenov, John D. Kingston, and George J. Armelagos
682
2009 Insights into immigration and social class at Machu Picchu, Peru based on
Oxygen, Strontium, and Lead isotopic analysis. Journal of Archaeological Science 36(2):
317–332.
Turner, Victor Witter
1974 The ritual process: structure and anti-structure. Aldine Publishing Co, Chicago.
Twiss, Katheryn
2007 The archaeology of food and identity. Southern Illinois University, Carbondale.
Ubelaker, Douglas H.
1994 Standards for data collection from human skeletal remains: proceedings of a
seminar at the Field Museum of Natural History, edited by Jane E. Buikstra. Arkansas
Archeological Survey, Fayetteville.
2000 Human remains from La Florida, Quito, Ecuador. Smithsonian Institution Press,
Washington, D.C.
Ubelaker, Douglas H, and Catherine E Ripley
1999 The ossuary of San Francisco Church, Quito, Ecuador: human skeletal biology.
Smithsonian Institution Press, Washington D.C.
Urbano, Henrique
2008a Cristobal de Molina. In Guide to Documentary Sources for Andean Studies, 15301900, edited by J. Pillsbury, pp. 427–430. University of Oklahoma Press, Norman.
2008b Introducción. In Relación de las fábulas y ritos de los incas, pp. 9–46. Estudios
coloniales / Instituto de Investigaciones. Lima.
Urton, Gary
1999
Inca myths. University of Texas Press, Austin.
Van Buren, M
1999 Tarapaya: an elite Spanish residence near colonial Potosi in comparative
perspective. Historical Archaeology 33(2): 108–122.
2010 The archaeological study of Spanish colonialism in the Americas. Journal of
Archaeological Research 18(2): 151–201.
Vanvalkenburgh, Nathaniel
2012 Building subjects: landscapes of forced resettlement in the Zaña and Chamán
Valleys, Peru, 16th and 17th centuries C.E. Ph.D. Dissertation, Harvard University.
683
2017 Unsettling time: persistence and memory in Spanish colonial Peru. Journal of
Archaeological Method and Theory 24(1): 117–148.
2019 The past, present, and future of transconquest archaeologies in the Andes.
International Journal of Historical Archaeology.
Vanvalkenburgh, Parker, Sarah J. Kelloway, Laure Dussubieux, Jeffrey Quilter, and Michael D.
Glascock
2015 The production and circulation of indigenous lead-glazed ceramics in northern
Peru during Spanish colonial times. Journal of Archaeological Science 61(C): 172–185.
Vanvalkenburgh, Parker, Sarah J. Kelloway, Karen L. Privat, Bill Sillar, and Jeffrey Quilter
2017 Rethinking cultural hybridity and technology transfer: SEM microstructural
analysis of lead glazed ceramics from early colonial Peru. Journal of Archaeological
Science 82(C): 17–30.
Vargas, José María
1948 La conquista espiritual del imperio de los incas. Unión de Intelectuales Católicos
Ecuatorianos, Quito.
Vargas Ugarte, Reuben
1951 Concilios limenses (1551-1772). Ecclesiastical Province, Lima.
Varon, Rafael
1990 El Taki Onqoy: las raices andinas de un fen6meno colonial. In El Retorno de las
huacas: estudios y documentos sobre el Taki Onqoy, siglo XVI, edited by L. Millones and
S. Castro-Klarén, pp. 331–405. Instituto de Estudios Peruanos, Lima.
Velasco, Matthew Carlos
2016 Mortuary tradition and social transformation during the Late Intermediate Period
(A.D. 1100-1450): a bioarchaeological analysis of above-ground burials in the Colca
Valley, Peru. Ph.D. Dissertation, Vanderbilt University.
Verano, John
1997 Human skeletal remains from tomb 1, Sipán (Lambayeque River Valley, Peru);
and their social implications. Antiquity 71(273): 670–682.
Villarías Robles, Juan José R., and María del Carmen Martín Rubio
1998 Sobre el autor. In De las costumbres y conversión de los indios del Perú:
memorial a Felipe II (1588) Crónicas y memorias. Ediciones Polifemo, Madrid.
Vivanco, Alejandro
684
1973 El migrante de provincias como interprete del folklore Andino en Lima. B.A.
Thesis, Universidad Nacional Mayor de San Marcos, Lima.
Voss, Barbara L
2008 Domesticating imperialism: sexual politics and the archaeology of empire.
American Anthropologist 110(2): 191–203.
Wachtel, Nathan
1977 The vision of the vanquished: the Spanish conquest of Peru through Indian eyes,
1530-1570. Harvester Press, Hassock.
Walker, P. L., J. R. Johnson, and P. M. Lambert
1988 Age and sex biases in the preservation of human skeletal remains. American
Journal of Physical Anthropology 76(2): 183–8.
Wallace, Anthony
1956 Revitalization movements. American Anthropologist 58(2): 264–281.
Weaver, Brendan J. M.
2015 “Fruit of the vine, work of human hands”: an archaeology and ethnohistory of
slavery on the Jesuit wine haciendas of Nasca, Peru. Ph.D. Dissertation, Vanderbilt
University.
Wernke, Steven A.
2008 Negotiating community and landscape in the Peruvian Andes: a transconquest
view. American Anthropologist 109(1):130-152.
2011 Convergences: producing early colonial hybridity at a Doctrina in Highland Peru,
in Enduring conquests: rethinking the archaeology of resistance to Spanish colonialism
in the Americas, edited by M. Liebmann and M. S. Murphy, pp77-102. School for
Advanced Research Press, Santa Fe.
2012 Spatial Network Analysis of a terminal prehispanic and early colonial settlement
in highland Peru. Journal of Archaeological Science 39(4): 1111–1122.
2013 Negotiated settlements : Andean communities and landscapes under Inka and
Spanish colonialism. University Press of Florida, Gainseville.
2018 Transformations: evangelization, resettlement, and community organization in the
early Viceroyalty of Peru. In The Oxford Handbook of the Incas, edited by S. Alconini
Mujica and R. Alan Covey, Oxford University Press, New York.
Wernke, Steven A., Lauren E. Kohut, and Abel Traslaviña
685
2017 A GIS of affordances: movement and visibility at a planned colonial town in
highland Peru. Journal of Archaeological Science 84: 22–39.
Wheatley, David, and Mark Gillings
2002 Spatial technology and archaeology. Taylor & Francis, New York.
Wheeler, Jane
1982 Aging llamas and alpacas by their teeth. Llama World 1: 12–17.
White, T. D.
2012 Human osteology. 3rd ed. Academic Press, San Diego.
Whitehead, Neil L
1995 The historical anthropology of text : the interpretation of Ralegh’s “Discoverie of
Guiana.” Current Anthropology, 36(1): 53–74.
Willey, Gordon R.
1971 An introduction to American Archaeology, Vol. 2: South America. Second
Printing edition. Prentice-Hall, Englewood Hall.
Williams, Patrick Ryan, and Donna J. Nash
2006 Sighting the Apu: A GIS analysis of Wari imperialism and the worship of
mountain peaks. World Archaeology 38(3): 455–468.
Worsley, Peter
1957 The trumpet shall sound: a study of “cargo” cults in Melanesia. Schocken Books,
New York.
Xerez, Francisco de, and Miguel de Estete
1891 Verdadera relación de la conquista del Perú, Madrid.
Yang, Dongya Y., and Kathy Watt
2005 Contamination controls when preparing archaeological remains for ancient DNA
analysis. Journal of Archaeological Science 32(3): 331–336.
Yaranga, A.
1978 Taki Onqo ou la vision des vaincus au XVIe siecle. In Les mentalites dans la
peninsule Iberique et en Amerique Latine aux XVIe et XVIIe siecle, pp. 119–79.
Publications de l’Universite de Tours, Tours.
Zuidema, R. Tom
1964 The ceque system of Cuzco : the social organization of the capital of the Inca. E.J.
Brill, Leiden.
686