What Ballcourts Tell Us About Sociopolitical Negotiation in Nejapa and the Eastern Sierra Sur, Oaxaca, Mexico

The Seriousness of Play: What Ballcourts Tell Us about Sociopolitical Negotiation in Nejapa and the Eastern Sierra Sur, Oaxaca, Mexico

Item Type text; Electronic Dissertation

Authors Stoll, Marijke Maurine

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 04/10/2021 03:39:58

Link to Item http://hdl.handle.net/10150/630223

THE SERIOUSNESS OF PLAY: WHAT BALLCOURTS TELL US ABOUT SOCIOPOLITICAL NEGOTIATION IN NEJAPA AND THE EASTERN SIERRA SUR, OAXACA, MEXICO

Marijke M. Stoll

A Dissertation Submitted to the Faculty of the

SCHOOL OF ANTHROPOLOGY

In Partial Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2018

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of the requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that an accurate acknowledgement of the source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the copyright holder.

SIGNED: Marijke M. Stoll

ACKNOWLEDGMENTS

They say that writing a dissertation is one of the hardest things that you will ever have to do in your professional academic career. Never has an adage been more true, and never has one set out to prove itself so much, despite all of my best-laid plans and intentions: the financial collapse and Great Recession of 2008 that happened as I entered graduate school, making it hard to find funding for a project about ancient sports arenas; losing access to my first planned dissertation project and losing two years of research and writing grant applications; a string of broken laptops and dead hard drives that also always managed to lose important documents no matter how often I backed up them up…the list goes on. To say that this dissertation is a result of blood, sweat, and tears is a vast understatement, not only because I have cried rivers of tears and bled and sweated buckets while working on this, but also because without the dedication, love, and the blood, sweat, and tears of the people in my life – from friends and colleagues to family – I would never have been able to do any of this.

My love of anthropology, archaeology, and the deep past began very early in life, and for that I owe much to my parents who always supported me in my decision to follow this career path, even when at 10 years old I proudly announced to my mom one day, “Mom, I want to be a paleoanthropologist”, and she wasn’t quite sure what I was babbling on about. Through all the successes, failures, hardships, and joy, my parents have stood by my side, always ready with words of encouragement, a good hug, or even a little financial help when I needed it. One of the best things they ever did was our big family vacation to Cancun, Mexico. It was during our day

4 trip Chichen Itza when, while gazing at the depiction of a human skull surround by flowery designs carved on the outside wall of the Great Ballcourt, that I fell in love with the ancient peoples of Mesoamerica, los antepasados. I will always remain forever grateful to my parents for always encouraging my passions for knowledge and exploration, for the many visits to our local library and museums, and for sharing my love of history and old things. And while he may only listen with half an ear whenever I talk about archaeology, I give thanks to Travis for being an awesome brother and a wonderful, caring person.

My intellectual development was guided and nurtured by many wonderful professors I have had the pleasure of meeting throughout my life, beginning with my undergraduate experience at the University of Chicago. From the Uncommon Core where I first took a serious dive into the history of social theory and political thought, to the intense archaeology courses I took in the Anthropology Department, it was here at U of C where I first began to discover myself as a scholar. For that I have to give many thanks to two professors who had the most impact. First, Mark Lycett, a man who I always found to be a little intimidating but who always pointed me in the right direction. His grounded and common sense approach made the most dense theory relatable and interesting, which is why I tried to get into every class he offered.

Second, Edward Swenson, whose class on Ideology, Ritual and the Moche really sparked my interest in these those and sent me on the road to the present research and dissertation. Since that first class I have had the fortunate to get to know Ed as a colleague and a good friend.

Our undergraduate courses of course are only the beginning our development as academics and scholars; it is graduate school where we really come into our own. Although I had never imagined myself at the University of Arizona when I was younger, I am so thankful that this is where I ended up. I found myself welcomed into the fold of a warm, close-knit, and

5 intellectually stimulating department – even before I had officially entered the program! Even when there weren’t many points in common between our research interests, every faculty member I interacted always took the time to listen and offer advice. From the professors I took classes with to those who showed me how to teach as newbie Teaching Assistant, to the faculty members I only ran into in the hall or chatted with over beers at Happy Hour, you all have touched my life and helped me in uncountable ways, even at those times when I wasn’t my best.

I am most especially thankful to (in no particular order) Dave Killick, Steve Kuhn, Lars Fogelin,

Barbara Mills, Tom Sheridan, Daniela Triadan, Diane Austin, E. Charles Adams, T.J. Ferguson,

James Watson, Michael Schiffer, and J.J. Reid. I will forever be grateful to John Olsen, a dear friend who helped me out when I was on my first field project in Mexico after becoming ABD and had had my funding taken away. Finally, my dearest thanks to Mark Nichter – you weren’t sure how I would be as your TA so I am glad that I could prove that archaeologists can hang! I enjoyed our before and after class chats, and I am glad to count you as a colleague and friend.

As a graduate student, your advisor is one of the most important people in your life. I still remember my first lunch meeting with Takeshi Inomata when I told him, rather bluntly, that

“I have no interest in the Maya and plan to work in Oaxaca, but I like your ideas and I think you can help me a lot.” I always wondered what he had thought about that! At times I felt like the red-headed stepchild, the one person not studying the Maya or working at Ceibel like his other students; and like that dreaded stepchild I know I have disappointed him at times and made him smile despite himself at others. Yet Takeshi has always been my advocate and has helped me find my footing intellectually, always reminding me to let the data speak for itself. I am forever grateful that Takeshi took a chance and took me on as a student. Through my future projects and contributions to archaeology, I hope to make him feel as proud of me as I am to be his student.

I also have to give many hearty and profound thanks to the other members of my committee who have also helped me so much over the course of the dissertation research and writing. I met María Nieves Zedeño my first year at Arizona and instantly liked her. My admiration and respect were further cemented when I ran into her wearing a teddy bear costume at a graduate student Halloween party that same year. Over the course of that first year I knew that she had to be a part of my committee. Nieves was key in helping me design, get permission for, and implement my ethnography of the pelota Mixteca players in Oaxaca. This project was a pivotal moment in my life. It shocked me out of the many romantic notions I had about the ballgame, but also showed me other critical insights that have had a great impact on my research and interpretations of the game. I would not have been so successful with the ethnographic project – really my first self-directed research study – without her guidance and wisdom.

I had never planned to take GIS courses initially, but found myself doing so in the Fall

2010 when it appeared that I was to be in charge of GIS data processing and analysis for a big research project. Although that is not how life turned out, taking this course was another game changer for me in my graduate career and for my research. Gary Christopherson co-taught that course and won me over with his dry sense of humor. When I returned to Tucson in 2014 after a three-year hiatus, knowing that my dissertation was heading in a GIS direction, I took another class with Gary and quickly learned that along with his dry wit was a warmth and care for his students no matter what department they came from. Over the years I often turned to Gary for help and advice, pestering him with GIS questions – he even helped me in a critical moment when I had trouble downloading the mapping data from the total station to my computer. When

I truly found myself in a crisis, Gary came through with no hesitation, stepping up to serve on my committee when I needed it most. I know the dissertation is much better for it.

Of course, I cannot forget one of my most important committee members, Stacie M.

King. When I was most adrift at sea, having lost almost two years of research and in desperate need of a project somewhere in Oaxaca, Stacie invited me to join her and her team on their project in Nejapa and again the direction of my life and research changed, and so much for the better. Through the PANT study I finally found a perfect case study for looking at the ballgame at a regional scale through a communal lens. Working on the PANT study was one of the best field seasons I have been on so far – Stacie is an amazing project director and I learned so much from her. Since that field season, and despite her own (numerous) duties, responsibilities, and many students, she always took the time to give advice, constructive criticism, and read everything I sent her. In many ways, she was a second advisor to me. My eternal gratitude to you Stacie for giving me so much.

Outside of my graduate program I have met many other kind scholars, academics, and archaeologists who have given advice, helped me out in big ways, and in general gave me opportunities to pursue my interests. Many thanks to the following, again in no particular order.

Gary Feinman and Linda Nicholas, for inviting onto my very first field project in Oaxaca. It was

Gary who said, “Well, no one’s really looked at the ballcourts…”, starting me on the journey whose eventual end is this present research. Ronald Spores, el Tigre de Tlaxiaco, who took me under his wing for a while and generously hosted me during a visit to the Oregon coast. Arthur

Joyce, who was kind enough to give me great advice on my NSF proposal so that I could be successful the second time around. Stephen Kowalewski, for a wonderful opportunity to do survey in the Mixteca Alta. While things didn’t work out as I had hoped, I am still ever grateful.

Marcus Winter, for first introducing me to Oaxacan archaeology all the way back in 2004 and for always being willing to listen and give advice over the years. Finally, for all the people who

8 work in Oaxaca that have become dear friends and colleagues to me over the years: Laura Stiver-

Walsh, Veronica Perez Rodriguez, Ronald Faulseit, Jaime Ford, Guy Hepp, Andrew Workinger,

Elizabeth Konwest, Alex Badillo, Nelly Robles Garcia, Mark Levine, Juan Jarqin, Hilary M., and so many others – thank you!

Life is empty without friends, and even more so in graduate school. The people I have met and became friends with as we struggled, cried, fought, and celebrated our way through the ups and downs of being graduates students together have made such a difference in my life. I have learned so much from all them, and they have helped me to grow and stay sane when the odds just seemed to be too great. To my friends I made at the U of A, I love you all so much, thank you for everything. Randy Haas, whose kindness and warmth made U of A an easy choice to make. Susan Ryan and Katie MacFarland, my wonderful cohort – I am so happy to finally join you guys on the other side! Saul, Lewis, Taylor and Ben, you guys are simply the best.

Shane Miller, I still miss hanging out with you in the lab. I’m glad I can still get a good Shane- hug at the SAAs. Natalia Martínez, mi reina, me haces mucha falta y espero verte otra vez.

Kenichiro Tsukamoto, mi hermano! Thank you so much for being such a wonderful older brother and mentor to me, I miss you greatly. Jessica C. and Tom Fenn, you two took me in as a little sister. I will always cherish the memories of hanging out at El Ferro Carríl in Trincheras and all those fun nights in Tucson. Bryan James Gordon, who is so awesome all three of his names must be said: I love you and can’t wait to sing karaoke together again as Zeppels and Led.

Melissa Burnham, it’s been fun watching you come into your own as an archaeologist, all while being a super mom. Thanks for all the rides home that first year and all the girlie hangouts after that. Jessica MacLellan, my mentee and dearest friend. You are simply just too awesome and I admire you so much – I can’t wait to see where your life takes you! Finally to my Oaxacan

9 colleagues and friends, most especially mis heramanitos Pedro Guillermo Ramon Celís and

Dante Rios Garíca, thank you for making this güera feel like a chapulíneca.

This dissertation would never have happened without the kindness, warmth, and friendship of the Oaxacan people. Since I first came down to Oaxaca in 2004, I have been swept up in the magic of this amazing place. To live and work in Oaxaca is to experience color, food, and music in whole new ways. Those times I have been frustrated to the point of tears have been more than balanced by the sheer joy I have felt whether tramping through the mountains or celebrating a patron saint feast day dancing in a calenda. Thank you to all my friends: Done

Pepe y familia, the Xochimilco corner crew, Don Agustín y Eleni, Teri and Allan, and everyone else who has touched my in different ways, too many to name here. Nearest and dearest to my heart of course is Nejapa, a place that still awes me with its beautiful landscape no matter how many times I drive through. I want to express my sincere gratitude to the people of Nejapa de

Madero, San Carlos La Baeza, Las Animas, and most especially Santa Ana Tavela. They not only gave their time and labor to help me out with my research, but also welcomed me in, sheltered, and fed me. Doña Roma’s cooking in particular nourished both my body and my soul through the field season. Thank you too to Don Matús and Doña Mele, who always have a place for me to rest my head and some delicious tamales to fill my belly.

More than anyone this dissertation is for my grandfather, who sadly passed away while I was in the field in 2013. When he was in his 50s, Grandpa Stoll packed up and moved with his wife and youngest son to Cholula, Puebla, Mexico, to get his Master’s of the Arts in anthropology at the University of the Americas and fulfill his dream of becoming an archaeologist. Grandpa generously shared his books on archaeology with me and regaled me with stories about his adventures in Campeche and Belize, including the time he and a friend

10 were canoeing up Jenny Creek and had to fight off aggressive Flor-du-lance snakes. For my first field school in Peru he sent me his archaeology toolkit, which I still carry with me and use to this day. Moving to Tucson gave me the chance to get to know him as an adult. On my visits to

Green Valley, we would talk about archaeology, watch British comedy shows, and sip whiskey on the rocks while he worked on his memoirs. I truly miss those moments of just hanging out and talking shop, even when we sometimes just sat in silence enjoying each other’s company. I will think of him every time I hold his Marshalltown trowel with our last name carved on the handle.

The journey to a PhD is a long and arduous one, and is often lonely and isolating.

Having someone to walk that hard road with you, make sure you eat and sleep, and who pushes you out into the sunshine when you find yourself surrounded in self-doubt makes the path a lot smoother. Charlie has been on this adventure with me every step of the way, and kept me going even when things were tough or seemed hopeless. Thank you Charlie for letting me bounce ideas off you and for helping me see through problems when the solution just wouldn’t appear; for listening carefully even when I was prattling on about technical minutiae that would make a lesser person’s eyes glaze over; for trekking up to La Baeza carrying the total station even when you weren’t feeling well; and for always making me laugh no matter what was going on in our lives. Te quiero, mi amor.

Finally, to Gizmo my beloved cat. She moved with me to Tucson and then down to

Oaxaca, and several more times after that, only tolerating each disruption because her mommy was there with her. She was a comfort and a companion for twelve years. I miss her every day.

DEDICATION

To my Grandpa in his loving memory, Maybe one day I’ll find the jeep you had To leave behind in the jungle

TABLE OF CONTENTS

Tabla de contenido LISTS OF FIGURES 17 LIST OF TABLES 30 ABSTRACT 33 CHAPTER I 36

1.1 INTRODUCTION 36 1.2 UNDERSTANDING PREHISPANIC BALLGAMES IN THE 21ST CENTURY: BALLCOURT DISTRIBUTION AND SOCIAL ORGANIZATION IN NEJAPA, OAXACA 40 1.3 PROYECTO ARQUEOLOGICO JUEGOS DE PELOTA NEJAPA-TAVELA 2015: RESEARCH DESIGN AND METHODS 47 1.4 ORGANIZATION OF THE DISSERTATION 49 1.5 CONCLUSION 53 CHAPTER II 55

2.1 INTRODUCTION 55 2.2 THE BALLGAME AS A SOCIAL FIELD: THEORY AND METHOD 57 2.2.1 RELIGION, RITUAL, AND FEASTING 59 2.2.2 SPACE, ARCHITECTURE, AND THE BUILT ENVIRONMENT 62 2.2.3. SPORTS, GAMES, AND SOCIETY 70 2.2.4 FRONTIER ZONES AND BORDERLANDS: FROM PERIPHERIES TO DYNAMIC SPACES 82 2.2.5 SCALING THE COMMUNITY: REGIONAL STUDIES IN ARCHAEOLOGY 86 2.2.6 DISCUSSION 89 2.3 CONSTRUCTING AND NEGOTIATING COMMUNITY AND SOCIAL IDENTITY IN MESOAMERICA 94 2.4 CONCLUSION 99 CHAPTER III 101

3.1 INTRODUCTION 101 3.2 A MESOAMERICAN GAME: AN OVERVIEW OF BALLGAME SCHOLARHIP AND INTERPRETATIONS 102 3.3 INDIGENOUS SPORTS AND GAMES OF THE AMERICAS 105 3.3.1 DISCUSSION AND SUMMARY: SPORTS AS AN IMPORTANT SOCIAL FIELD IN INDIGENOUS AMERICA 113 3.4 ES NUESTRA TRADICIÓN: MODERN BALLGAMES IN MEXICO AND OAXACA 116 3.4.1 TRADITIONAL SPORTS OF MEXICO TODAY 118 3.4.2 PELOTA MIXTECA DE HULE IN OAXACA 122 3.4.3. DISCUSSION 146 3.5 THE ARCHAEOLOGICAL DATA ON THE MESOAMERICAN BALLGAME 151 3.5.1 A NOTE ON BALLCOURT TERMINOLOGY 152

3.5.2 A REGIONAL OVERVIEW OF BALLGAME AND BALLCOURT MATERIAL EVIDENCE 156 3.5.3 THE BALLGAME IN OAXACA 179 3.5.4 DISCUSSION OF THE MATERIAL EVIDENCE 194 3.5.5 SUMMARY 204 3.6 THE COMMUNAL PRACTICES OF MESOAMERICAN BALLGAME TRADITIONS 206 3.6.1 RITUALIZATION VS. SECULARIZATION IN THE BALLGAME 207 3.6.2 SACRIFICE, SACRED COSMOLOGY, AND BALLGAME IDEOLOGY 211 3.6.3 BALLGAME AS RITUAL VS. BALLGAME AS SPORT 214 3.6.4 DISCUSSION 218 3.7 CONCLUSION 224 CHAPTER IV 226

4.1 INTRODUCTION 226 4.1.1 A NOTE ON TERMINOLOGY 228 4.2 A BRIEF INTRODUCTION TO ARCHAEOLOGY IN OAXACA 231 4.3 HOW DID WE GET HERE? A SHORT SOCIOPOLOTICAL HISTORY OF OAXACA 235 4.3.1 TERRA (SOMEWHAT) INCOGNITA: HISTORY FROM THE REGIONS AROUND NEJAPA 244 4.3.2 DISCUSSION 249 4.4 NEJAPA FROM THE FORMATIVE TO THE POSTCLASSIC 252 4.4.1 FROM THE FORMATIVE TO THE CLASSIC PERIOD (2000 BCE – CE 900) 261 4.4.2 THE BIG BANG: POPULATION GROWTH, SETTLEMENT EXPANSION, AND SOCIAL CONFLICT IN THE POSTCLASSIC PERIOD 264 4.4.3 NEJAPA (NEXAPA) IN THE COLONIAL ERA 272 4.4.4 DISCUSSION: CHALLENGING NARRATIVES OF HISTORY AND ETHNIC SOCIAL IDENTITY IN NEJAPA 275 4.5 CONCLUSION 287 CHAPTER V 289

5.1 INTRODUCTION 289 5.2 USING GIS TO INVESTIGATE THE SOCIAL ROLE OF THE BALLGAME 291 5.2.1 GIS THEORY AND CRITIQUES 294 5.2.2 FINDING BALLCOURTS ON THE LANDSCAPE: PREDICTIVE MODELING AND CLUSTER ANALYSIS 299 5.2.3 MOVEMENT ON THE LANDSCAPE: COST-PATH SURFACES AND PATH-DISTANCE ANALYSIS 300 5.2.4 SIGHTING BALLCOURTS ON THE LANDSCAPE: INTERVISIBILITY, VIEWSHEDS, AND AFFORDANCES 305 5.2.5 DISCUSSION 310 5.3 MAPPING METHODS AND ANALYSIS 313 5.3.1 FIELD SURVEY AND MAPPING 313 5.3.2 DATA PROCESSING AND ANALYSIS 320 5.3.3 PANT SURVEY, EXCAVATION, AND ANALYSIS 324 5.4 CONCLUSION 326 CHAPTER VI 327

6.1 INTRODUCTION 327 6.2 CASE STUDIES – MUNICIPALITY OF NEJAPA DE MADERO 330 6.2.1 EL CUCHARITAL (YAU-NEJ-NEJ-005) 332 6.2.2 LOS LIMARES (YAU-NEJ-NEJ-007) 348 6.2.3 CERRO MALUCO (YAU-NEJ-NEJ-012) 358 6.2.4 TEREZONA (YAU-NEJ-NEJ-013) 369

6.2.5 LOS MOGOTES DE PANTEÓN (YAU-NEJ-ANM-002) 377 6.2.6 EJIDO DE LOS CANSECO (YAU-NEJ-CAM-001) 390 6.3 CASE STUDIES – MUNICIPALITY OF SANTA ANA TAVELA 396 6.3.1 LACHIXOBA (YAU-TAV-TAV-003) 399 6.3.2 LA PUERTA (YAU-TAV-TAV-004) 406 6.3.3 LLANO LAS CASAS (YAU-TAV-TAV-015) 416 6.4 CASE STUDY – SAN CARLOS YAUTEPEC 424 6.4.1 THE PREHISPANIC SETTLEMENTS OF SAN CARLOS YAUTEPEC (SO FAR) 426 6.4.2 LA BAEZA (YAU-SCY-BAE-001) 433 6.4.3 SUMMARY: LA BAEZA AND THE PREHISPANIC COMMUNITIES OF SAN CARLOS YAUTEPEC (SO FAR) 470 6.5 CASE STUDIES – BALLCOURT EXCAVATIONS 476 6.5.1 EL SITIAL (YAU-TAV-TAV-010) – MUNICIPALITY OF SANTA ANA TAVELA 476 6.5.2 MUNICIPALITY OF SAN JUAN LAJARCÍA 489 6.6 A BALLCOURT IN THE CLOUDS - LOS PICACHOS (YAU-TAV-TAV-011) 502 6.7 DISSCUSION: A HISTORY OF THE BALLGAME IN NEJAPA 504 6.7 SUMMARY AND CONCLUSION 527 CHAPTER VII 529

7.1 INTRODUCTION 529 7.2. UNDERSTANDING THE NEJAPA BALLCOURTS AS A GROUP 530 7.2.1 BALLCOURT DIMENSIONS: OVERALL AND PLAYING SURFACE LENGTHS AND WIDTHS 532 7.2.2 LATERAL AND TERMINAL MOUND DIMENSIONS 550 7.2.3 BALLCOURT SLOPES 564 7.2.4 NORTH-SOUTH BALLCOURTS VS. EAST-WEST BALLCOURTS IN NEJAPA 569 7.2.5 SUMMARY 580 7.3 A COMPARATIVE CASE STUDY: THE BALLCOURTS OF THE CENTRAL VALLEYS AND NEJAPA 581 7.3.1 THE CENTRAL VALLEY BALLCOURTS 581 7.3.2 BALLGAME TRADITIONS IN OAXACA: COMPARING NEJAPA AND THE CENTRAL VALLEYS 592 7.3.3 SUMMARY 608 7.4 CONCLUSION 609 CHAPTER VIII 611

8.1 INTRODUCTION 611 8.1.1 BUILDING THE GIS SPATIAL DATABASE 612 8.2 THE DISTRIBUTION OF BALLCOURTS IN NEJAPA: SPATIAL ANALYSES AND RESULTS 615 8.2.1 EXPLORING BALLCOURT DIMENSIONS AND ATTRIBUTES ON THE LANDSCAPE 615 8.2.2 BALLCOURTS AND COMMUNITIES: CLUSTER ANALYSES AND RESULTS 634 8.3 TO THE BALLCOURT AND BACK: LEAST COST-PATH ANALYSES, LANDSCAPE MOVEMENTS, AND THE OCCASIONAL FAILURE 646 8.4 SIGHTING BALLCOURTS ON THE LANDSCAPE: VIEWSHED ANALYSES AND RESULTS 660 8.5 SUMMARY AND CONCLUSION 686 CHAPTER IX 688

9.1 INTRODUCTION 688 9.2 GAMES AT THE FRONTIER: EXPLORING THE BALLCOURTS OF NEJAPA 689

9.2.1. EAST AND WEST, NORTH AND SOUTH: BALLCOURT ORIENTATION AND SIGNIFICANCE 690 9.2.2 NON I-SHAPED BALLCOURTS IN NEJAPA 693 9.2.3 COMMUNITIES WITH MULTIPLE BALLCOURTS 695 9.2.4 THE BALLGAME AND BALLCOURTS IN NEJAPA OVER TIME: SHARED TRADITIONS, LOCALIZED EXPRESSIONS 699 9.3 THE BALLGAME IN NEJAPA AND OAXACA: LOCALIZED EXPRESSIONS OF REGIONAL TRADITIONS 723 9.5 FUTURE DIRECTIONS IN BALLCOURT RESEARCH IN NEJAPA 733 9.6 CONCLUSION 734 APPENDIX A 736

A.1 INTRODUCTION 736 A.2 SURFACE CERAMICS AND MATERIALS FROM EL CUCHARITAL (YAU-NEJ-NEJ-005) 736 A.3 SURFACE CERAMICS AND MATERIALS FROM MOGOTES DEL PANTEÓN (YAU-NEJ- ANM-002) 747 A.3 ARCHITECTURE, SURFACE CERAMICS AND MATERIALS FROM LA BAEZA (SCY- SCY-BAE-001) 759 REFERENCES CITED 769

LISTS OF FIGURES

Figure 1.1: The southeastern state of Oaxaca and the Nejapa study region. 37 Figure 1.2: Map showing the ballcourts with two-kilometers buffers placed around them. 38 Figure 1.3: Playing pelota mixteca in the town of Magdelena Jaltepec during the patron feast day. 41 Figure 3.1: Volcanized rubber ball used for playing pelota Mixteca. This ball specifically belongs to the team from the suburb of Buena Vista. 123 Figure 3.2: A pelota Mixteca glove, brightly decorated and studded with over a hundred nails. 124 Figure 3.3: Evolution of the guante or glove in pelota Mixteca, from the earliest leather swatch (upper) to the modern version used today (lower). 126 Figure 3.4: Modern pelota Mixteca gloves. The lighter blue glove in the back features the player's nickname Lento (Slow). 127 Figure 3.5: The patio or court in the polideportivo at the Tecnologíco. The patio is a 100 m field surrounded by high fences. 128 Figure 3.6: The informal patios at San Antonio Arrazola (upper) and Buena Vista (lower). 129 Figure 3.7: The botadora stone is used in pelota Mixteca to launch the play into play. Two have been set up in the patio at Buena Vista for a torneo where multiple teams would be playing. 131 Figure 3.8: The saque (left) and resto (right) teams playing at Buena Vista. 132 Figure 3.9: The chacero or referee observes game play during a match at Guadalupe Etla. 133 Figure 3.10: A chacero marking a raya where a ball has bounced out of bounds. 134 Figure 3.11: Chaceros also keep track of each quinta's score. Score marks from a game at the Tecnologico (upper) and a chacero marks the score during a game at Santa Cruz Amilpas (lower). 136 Figure 3.12: Many players have lasting injuries from playing, especially on the glove hand. 138 Figure 3.13: Score board for the day's torneo at Guadalupe Etla, showing the different towns the teams are coming from. 140 Figure 3.14: Design marked in chalk with the name of the town 146 Figure 3.15: The four general ballcourt forms that have been documented so far in Mesoamerica: a) open; b) T-shaped; the I-shaped court which can be c) open or d) enclosed; and e) palangana or washbasin type. 155

Figure 3.16: Features typically documented at ballcourts, although these features vary and may not be present in some regions and/or during certain time periods. 156 Figure 3.17: Map of Mesoamerica showing the different geocultural macroregions. 157 Figure 3.18: The Central Mexican Zone includes the Federal District as well as the states of Mexico, Tlaxcala, Puebla, Morelos, Hidalgo, Queretaro, and San Luis Potosi. 158 Figure 3.19: The Gulf Coast includes the Huaxteca, Veracruz, and Tabasco. 162 Figure 3.20: The Maya Macroregion can be divided into four separate geological and ecological zones: the Northern Lowlands, the Central Lowlands, the Southern Highlands, and the Pacific Coastal Zone. 166 Figure 3.21: Map showing other macoregions of Mesoamerica as well as areas outside of Mesoamerica (mentioned in the text) where ballcourt structures have been found. 174 Figure 3.22: Ceramic model of a ballcourt and game from West Mexico. These models, which show lively scenes of people enjoying a game, have usually been found in tombs (photograph by Tom Aleto, reproduced under license terms cc-by 2.0) 176 Figure 3.23: Map of the different regions and subregions of Oaxaca discussed in the text and where ballcourts have been documented. 180 Figure 3.24: One of the ballcourt sculptures from the Cueva del Rey Kong-Oy. You can see a small stairway set into the terminal mound (photograph by Stacie M. King, reproduced here with permission). 185 Figure 3.25: Detail of the Rey Kong-Oy figure, located near the ballcourt model in Figure 3.24 (photograph by Stacie M. King, reproduced here with permission). He wears a jaguar-style mask (upper) and holds a rubber ball in his hand (lower). 186 Figure 3.26: Drawing showing how ballcourts are commonly depicted in Mixtec codices, based on an icon from the Codex Zouche-Nuttall. 190 Figure 3.27: Maps showing the frequency of hachas, palmas, and yugos found in the different regions of Mesoamerica. 198 Figure 4.1: A feral cow gazes out onto the landscape, Santa Ana Tavela, Yautepec District, Oaxaca. 227 Figure 4.2: Map of Oaxaca and the different regions and subregions mentioned in this chapter. 231 Figure 4.3: Map of Oaxaca and some of the sites mentioned in the text. 237 Figure 4.4: The Main Plaza at Monte Albán, looking north from the South Platform. 239 Figure 4.5: Map of the Nejapa study region and the modern towns found there. 252 4.6: The Nejapa subregion has a mixed ecology that includes highland scrubs and trees (upper) and more humid zones where you can grow tropical fruits such as papaya (lower). 256 Figure 4.7: Men from Nejapa de Madero performing their tequeo and harvesting watermelons from a communal field. 257 Figure 4.8: Map of the Nejapa study area and some of the sites mentioned in the text. 261

Figure 4.9: The grand wall at Cerro la Muralla (photo by Stacie M. King, reproduced here with permission). Some sections of the wall have a stepped interior (upper), and an overhang can be observed on the exterior (lower). 269 Figure 4.10: 18th century French map of Mexico. Nejapa is one of the places included on the map. 274 Figure 4.11: Side A of the Nejapa Viejo carved stone (drawing by Stacie M. King). 277 Figure 4.12: Some of the carved glyphs found at the Arroyo de Sauso site (drawing by Javier Urcid and the PANT team. 279 Figure 4.13: The pectoral fragment found by Don Adrian near Los Limares (upper) and Urcid's analysis of the glyphs in the pectoral (lower). 282 Figure 5.1: Don Adrian uses an axe to chop down a particularly stubborn tree. 316 Figure 5.2: Setting up the Sokkia total station to map the ballcourt at Lachixoba (photo by Charles Collins). 317 Figure 5.3 Don Taurino standing with a prism rod in front of a controlled fire behind the west lateral mound at El Cucharital. 318 Figure 5.4: Illustration of the order of operations for mapping each ballcourt. 319 Figure 6.1: Map of sites and ballcourts in the Nejapa subregion. 328 Figure 6.2: Map of sites, ballcourts, and modern villages in the municipality of Nejapa de Madero. 331 Figure 6.3: Map of El Cucharital site and ballcourts. 332 Figure 6.4: 3D model of the main ballcourt at El Cucharital, looking from the southwest. 337 Figure 6.5: Composite image showing the east lateral mound and south terminal mound, from the top of the west lateral mound. 337 Figure 6.6: View of the main ballcourt from the northeast corner looking south. 338 Figure 6.7: The cistern on the southern end of the west lateral mound. 339 Figure 6.8: Squared platform on the northwest corner of the north terminal mound looking from the south (upper) and north (lower). 341 Figure 6.9: Interior corner of north and west mounds showing defined edge and possible stair. 342 Figure 6.10: The interior of the south terminal mound showing the possible stairs leading into the court. 342 Figure 6.11 View towards the North Plaza from the north terminal mound (upper); and a view of the South Plaza from the west lateral mound, looking south (lower). 344 Figure 6.12: View towards where a neighborhood ballcourt was located prior to its destruction in 2013. 346 Figure 6.13: Map of Los Limares, which was likely part of the hilltop settlement on Cerro Tanaguixi. 348

Figure 6.14: 3D model of the ballcourt at Los Limares, looking from the southwest. 350 Figure 6.15: The ballcourt at Los Limares (Tanaguixi) looking north (upper) and south (lower). 351 Figure 6.16: A view of the west lateral mound and the surrounding landscape from the east lateral mound at Los Limares. 354 Figure 6.17 Slump that could be a possible staircase on the interior of the west terminal mound. 355 Figure 6.18: The first carved stone with glyphs, El Zapote, showing Side 1 (left) with the abstract figure and Side 2 (right) possibly showing the head of a bird (original photo by Stacie M. King). 357 Figure 6.19: Stone 1 (left) has a symbol very similar to the Aztec gold glyph and Stone 2 (right) has another abstract figure (from King 2010, reproduced here with permission). 357 Figure 6.20: The site of Cerro Maluco, with the largest ballcourt in the Nejapa subregion. 358 Figure 6.21: The north mound on the Main Plaza at Cerro Maluco. 359 Figure 6.22: Composite of the east platform on the Main Plaza, facing east. 360 Figure 6.23: Fire pits on the southern end of the east platform. 361 Figure 6.24: Square room or altar in the center of the Main Plaza, open on the west side. 362 Figure 6.25: 3D models of Cerro Maluco ballcourt, looking from above (upper) and from the southwest (lower). 363 Figure 6.26: View of the ballcourt at Cerro Maluco looking east. 364 Figure 6.27: Bedrock outcropping in the west terminal alley that appears to have a rounded corner. 365 Figure 6.28: The site of Terezona with a T-shaped ballcourt. 369 Figure 6.29: 3D model of the ballcourt at Terezona looking from the southwest corner. 372 Figure 6.30: Composite of the ballcourt looking south from the north terminal alley. 372 Figure 6.31: Composite of the sounth terminal end of the Terezona ballcourt. 373 Figure 6.32: Map of the Mogotes del Panteón site. 377 Figure 6.33: Small platform on the north side of the north lateral mound in the Main Plaza. 379 Figure 6.34: View of Platform A from the south. Platform A forms the northern border of the Main Plaza. 380 Figure 6.35: Platform B on the east side of the Main Plaza. 380 Figure 6.36: 3D model of the ballcourt at MDP from the south. In the background are Platforms A and B; the small platform next to the north lateral mound is also visible. 383 Figure 6.37: Composite of the MDP ballcourt from the west looking east. 383 Figure 6.38: The south lateral mound on the south side with Don Jose Enrico Martinez for scale. 385

Figure 6.39: Stones lining the summit and sides of the west terminal mound. 387 Figure 6.40: Raised feature on the southwest corner of the west terminal mound. 387 Figure 6.41: The east terminal mound at MDP. The superior surface of the mound is level with the east plaza, a common feature of Nejapa ballcourts. 388 Figure 6.42: Fragment of stucco flooring found in MDP ballcourt showing that repairs had been made at least once. 389 Figure 6.43: Map of the Ejido de los Canseco site. 390 Figure 6.44: 3D model of ballcourt at EC. 392 Figure 6.45: View of ballcourt from the east and looking west. 392 Figure 6.46: Line of stones marking the summit and eastern end of the north lateral mound. 394 Figure 6.47: Views of the surrounding landscape from the south lateral mound, looking north (upper left), south (upper right), east (lower left), and west (lower right). 396 Figure 6.47: Figure 6.48: Map of the ballcourts in the municipality of Santa Ana Tavela. Ballcourts here are located much further apart and are not as clustered close together as in Nejapa de Madero. 397 Figure 6.49: Map of the Lachixoba site, which was part of the Greater la Amontonada community. 399 Figure 6.50: Three layers of stucco floor visible from a posthole originally dug for the screen. These floors would have been the interior living surface of the platform. 402 Figure 6.51: 3D model of the ballcourt at Lachixoba, looking from the south. 403 Figure 6.52: Composite view of the Lachixoba ballcourt from the south. 403 Figure 6.53: Map of the La Puerta site. Figure 6.53: Map of the La Puerta site. 406 Figure 6.54: View of the Rio Tehuantepec from the La Puerta site, looking from the southern end (upper) of the site and the western end (lower). 408 Figure 6.55: Piedra la Iglesia, a sacred site still in use today that is visible from La Puerta. 410 Figure 6.56: 3D models of the La Puerta ballcourt from above (left) and from the southwest (right). 411 Figure 6.57: View of the La Puerta ballcourt from the north. 411 Figure 6.58: Southern entrance into the ballcourt at La Puerta. It is possible that a natural rise was modified somewhat to create this entrance. 414 Figure 6.59: Large stone set into the southern end of the La Puerta court. 414 Figure 6.60: Map of the Llano las Casas site. 416 Figure 6.61: 3D models of the Llano las Casas ballcourt from above (upper) and from the southwest (lower). 418 Figure 6.62: View of Llano las Casas ballcourt from the south. 419

Figure 6.63: The double inset entrance on the south terminal mound, which would have allowed access to the court from this side. 421 Figure 6.64: View of the carved glyphs on the stone marking the corner of an attached platform, showing the north (upper) and south (lower) faces. 423 Figure 6.65: Map of the sites documented by PANT so far in the San Carlos Yautepec municipality. 425 Figure 6.66: Map of sites recorded in the territory of San Francisco Guichina 426 Figure 6.67: Map of the sites recorded in the territory of San Isidrio Manteca and San Carlos Yautepec. 430 Figure 6.68: Map of the sites recorded in the vicinity of Santiago Vargas. 431 Figure 6.69: The basketball court at La Baeza, looking east from the agencia building. The whole town is perched on the hillside. 435 Figure 6.70: The site of La Baeza based on current mapping data. We expect the settlement to be much larger once fully mapped. 436 Figure 6.71: Map of Group A structures at La Baeza. 438 Figure 6.72: Structure A1, a possible double temple built on a platform. This structure still has preserved earthen/adobe walls. 439 Figure 6.73: Map of Group B structures at La Baeza. 441 Figure 6.74: Structure B3 (upper) may have been a temple. On its west side there may have been stairs with access to the plaza on this side (lower). 442 Figure 6.75: Group C is located southwest of Group B and may have been a residential group (photo by Júan Jarquin). 443 Figure 6.76: The tlecuiles or hearths found at Group C (photo by Júan Jarquin). 444 Figure 6.77: Examples of La Baeza style bricks that had been looted from a nearby structure previously. 445 Figure 6.78: A room foundation using La Baeza style bricks. Many of the room foundations used these bricks. 446 Figure 6.79 Map of the Group D structures, including the ballcourt. 447 Figure 6.80: Structure D2 is a long rectangular building with much of its western wall preserved. 448 Figure 6.81 We did not map Structure D3, but it appears to be a foundation for a possible altar, and is similar in appearance to the altar structure at Cerro Maluco. 449 Figure 6.82: The foundation for a room on top of Structure D6. 450 Figure 6.83: Map of Group E structures at La Baeza. This may have been an elite residential and/or administrative building. 451 Figure 6.84: View of Structure E1 from the west. 453 Figure 6.85: Map of Group F structures at La Baeza. 455

Figure 6.86: Structure F2 is a large platform with room superstructures on top and a sunken patio. 456 Figure 6.87 The east and west sides of Structure F3, which show very different construction techniques. 457 Figure 6.88: A standing earthen/adobe wall can still be seen on Structure F6. 458 Figure 6.89: Map of the Group G structures at La Baeza. 459 Figure 6.90: Structure G1 is the first platform of the Group G complex. It stands at over a meter tall and features massive stones. 460 Figure 6.91: Structure G2 is the second platform and lies directly on top of Str. G1. An inset staircase gives access to the rooms and other structures on this platform as well as the western extent of the site. 461 Figure 6.92: Map of Group H, which has several buildings that were not mapped. Group H may be superstructures on top of a large temple. 463 Figure 6.93: Standing wall of Structure H1. This wall features shaped stones set into a matrix of some type. 464 Figure 6.94: A large temple and plaza observed in 2013 at the western extent of the ridgeline. Group H may be the superstructures on top of this temple (photos by Stacie M. King). 465 Figure 6.95: 3D model of the ballcourt at La Baeza. 467 Figure 6.96: View of the La Baeza ballcourt from the east and looking west along the ridgeline. 467 Figure 6.97: View of the ballcourt from the west, with the south plaza visible on the right. A possible entrance may have been on the southwest side. 470 Figure 6.98: A large plaza bordered by a terrace wall, looking north. Structure Groups F - H lie on the hilltop to the left. 472 Figure 6.99: Map of the El Sitial site and ballcourt. 477 Figure 6.100: Operation D excavations at the El Sitial ballcourt. 479 Figure 6.101: The exterior end of the drain on the southwest side of the west terminal mound. The drain exterior was constructed in a half-moon shape. 480 Figure 6.102: Operation D - the drain apreture was located in the upper right quadrant. The opening had been covered by wall slump. 482 Figure 6.103: The drain aperture after the wall slump had been removed and the interior of the drain cleaned out. 484 Figure 6.104: Operation E - excavation exposed a terrace wall that connects with the back wall of the lateral structure on the north lateral mound. 486 Figure 6.105: The El Sitial ballcourt has many of the same architectural features as courts in the Central Valleys. 488 Figure 6.106: Map of the San Juan Lajarcía municipality and some of the sites documented there. 490

Figure 6.107: Map of the Cerro del Convento site and ballcourt. 492 Figure 6.108: The limestone cliffs of Cerro del Convento are a prominent feature that can be seen from multiple viewpoints in the landscape. 493 Figure 6.109: The arroyo is a controlled access point to the site, making Cerro del Convento naturally defensible. 495 Figure 6.110: Architectural drawing of the Cerro del Convento ballcourt. 498 Figure 6.111: Excavation drawing showing the different layers as well as the hearth element found in the center of the unit. 500 Figure 6.112: Map of the Los Picachos site and associated settlements, including the Los Chorros site where the ballcourt is located. 502 Figure 6.113: Bar graph showing the overall lengths and widths of the ballcourts in Nejapa. 506 Figure 6.114: Bar graph showing the central playing alley lengths and widths. 506 Figure 6.115: Terminal mounds are very small in Nejapa, and one mound is always nearly or at level with the ground surface. 507 Figure 6.116: Example of the construction methods used to build the court mounds in Nejapa. The west lateral mound at Terezona had been cut during road construction. 509 Figure 6.117: Map of Late Formative and Early Classic sites and ballcourts. 511 Figure 6.118: Map of Middle and Late Classic sites and ballcourts. 514 Figure 6.119: Map of Early Postclassic and Middle Postclassic sites and ballcourts. 520 Figure 6.120: Map of Middle Postclassic and Late Postclassic sites and ballcourts. 523 Figure 7.1: A box plot comparing the distributions of the overall and central alley lengths (upper) and widths (lower). 534 Figure 7.2: A boxplot comparing the central alley and overall court lengths by time period. 536 Figure 7.3: A box plot comparing the central alley and overall court widths by time period. 536 Figure 7.4: The overall court and central alley L/W ratios of each ballcourt. 539 Figure 7.5: Histogram of the central alley L/W ratios. 539 Figure 7.6: A box plot comparing the overall and central alley court L/W ratios. 540 Figure 7.7: A box plot of the central alley L/W ratios for the Classic and Postclassic periods. 542 Figure 7.8: Box plots comparing the distributions of terminal 1 and 2 alley lengths (upper) and widths (lower). 544 Figure 7.9: A box plot comparing the terminal 1 and 2 alley lengths by time period. 546 Figure 7.10: A box plot comparing the terminal 1 and 2 alley widths by time period. 546 Figure 7.11: A box plot comparing the distributions of the terminal 1 and 2 alley L/W ratios. 549 Figure 7.12 A box plot of the terminal 1 and 2 alley L/W ratios by time period. 549

Figure 7.13: Box plots comparing the lengths (upper) and widths (lower) of the lateral and terminal mounds. 556 Figure 7.14: Box plots comparing the interior (upper) and exterior (lower) heights of the lateral and terminal mounds. 557 Figure 7.15: A box plot comparing the volumes of the lateral and terminal mounds. 558 Figure 7.16: Box plots comparing the L/W ratios of the lateral (upper) and terminal (lower) mounds by orientation. 559 Figure 7.17: A box plot comparing the lengths of the different mound groups by time period. 562 Figure 7.18: A box plot comparing the widths of the different mound groups by time period. 562 Figure 7.19: Box plots comparing the interior (upper) and exterior (lower) heights by time period. 563 Figure 7.20: A box plot comparing the volumes of the different mound groups by time period. 564 Figure 7.21: A box plot comparing the distributions of the minimum, maximum, and mean values for the interior lateral surface slopes. 567 Figure 7.22: Bar chart displaying the minimum, maximum, and mean values of the interior lateral surface slope grades by site. 568 Figure 7.23: The estimated degree of slope range for the interior lateral mound surfaces. 568 Figure 7.24: A box plot of the distributions of the minimum, maximum, and mean values by time period. 569 Figure 7.25: The number of north-south and east-west oriented ballcourts by time period. 574 Figure 7.26: Box plot comparing the distribution of the overall and central alley lengths of the EW ballcourts versus the NS ballcourts. 575 Figure 7.27: Box plot comparing the distribution of the overall and central playing alley widths of the EW versus NS ballcourts. 576 Figure 7.28: Box plot comparing the overall and central playing alley L/W ratios of the EW vs NS ballcourts. 576 Figure 7.29: Box plot comparing the L/W ratio of the terminal alleys by orientation. 577 Figure 7.30: Box plot comparing the lengths of the lateral and terminal mounds by orientation. 578 Figure 7.31: Box plot comparing the widths of the lateral and terminal mounds by orientation. 578 Figure 7.32: Box plot comparing the L/W ratios of the lateral and terminal mounds by orientation. 579 Figure 7.33: Box plot comparing the central playing alley lengths of the Central Valley Courts by time period. 584

Figure 7.34: Box plot comparing the central playing alley widths of the Central Valley courts by time period. 584 Figure 7.35: Box plot comparing the central playing alley L/W ratios of the Central Valley courts by time period. 585 Figure 7.36: A histogram of the central playing alley L/W ratios shows a clear bimodal distribution. 585 Figure 7.37: Box plot comparing the lateral and terminal mound lengths of the Central Valley courts by time period. 590 Figure 7.38: Box plot comparing the lateral and terminal mound widths of the Central Valley courts by time period. 590 Figure 7.39: Box plot comparing the lateral and terminal mound L/W ratios of the Central Valley courts by time period. 591 Figure 7.40: Frequency distribution of the Central Valley ballcourt orientations, which show a clear bimodal distribution. 592 Figure 7.41: Frequency distribution (upper) and box plot (lower) comparing the lengths of the central playing alleys of the ballcourts in the Central Valleys and Nejapa. 595 Figure 7.42: Frequency distribution (upper) and box plot (lower) comparing the widths of the central playing alleys of the ballcourts in the Central Valleys and Nejapa. 596 Figure 7.43: Frequency distribution (upper) and box plot (lower) comparing the L/W ratios of the central playing alleys of the ballcourts in the Central Valleys and Nejapa. 597 Figure 7.44: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal mound lengths of the Central Valleys and Nejapa ballcourts. 600 Figure 7.45: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal mound widths of the Central Valleys and Nejapa ballcourts. 601 Figure 7.46: Frequency distribution (upper) and box plot (lower) comparing the L/W ratios of the lateral and terminal mounds of the Central Valleys and Nejapa ballcourts. 602 Figure 7.47: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal interior mound heights of the Central Valleys and Nejapa ballcourts. 604 Figure 7.48: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal exterior mound heights of the Central Valleys and Nejapa ballcourts. 605 Figure 7.49: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal mound volumes of the Central Valley and Nejapa ballcourts. 606 Figure 7.50: The number of north/south versus east/west-oriented ballcourts in the Central Valleys and Nejapa. 607 Figure 8.1: Map of the ballcourts in Nejapa by time period. 617 Figure 8.2 Map of the Nejapa study region and the areas where PANT conducted survey. 618 Figure 8.3 Map of the overall lengths (upper) and widths (lower) of the ballcourts. 620 Figure 8.4: Map of the central lengths (upper) and widths (lower) of the ballcourts. 622

Figure 8.5: Map showing the Mound 1 (north/east laterals) volumes of the ballcourts. 624 Figure 8.6: Map showing the distance of each ballcourt from a major waterway, measured in meters. 626 Figure 8.7: Map showing the distance in meters of the ballcourts from nearby waterways. 627 Figure 8.8 Map showing the elevation of each ballcourt in Nejapa. 628 Figure 8.9: Map showing the degree of slope where each ballcourt was constructed. 629 Figure 8.10: Map showing the percentage or rise in slope where each ballcourt has been constructed. 630 Figure 8.11: Box plots comparing the degree of slope (upper) and percent change in rise (lower) between the random points and the ballcourts. 632 Figure 8.12 Box plots comparing the degree of slope (upper) and percent change in rise (lower) between the non-ballcourt sites and ballcourts. 633 Figure 8.13: Graph comparing the normalized values of the different Ballcourt Indices. 640 Figure 8.14: Map showing Ballcourt Index 4, comparing the random sample points against the ballcourts. 641 Figure 8.15: Map of the Model 1 results for the Classic period. 643 Figure 8.16: Map of the Model 1 results for the Postclassic period. 644 Figure 8.17: Climbing the stairway in a narrow crevice to reach the ridgeline of the cliffs at Cerro la Muralla. 653 Figure 8.18: The road that passes by the ballcourt at Lachixoba, looking towards the southwest. 655 Figure 8.19: Results of the Path Allocation Distance analysis for Classic period sites and ballcourts, using the different impedance surfaces (1 = Mogotes del Panteón; 2 = Ejido de los Canseco; 3 = Llano las Casas; 4 = El Sitial; 5 = Los Chorros; 6 = Los Limares; 7 = Cerro Maluco). 657 Figure 8.20: Results of the Path Allocation Distance analysis for Classic period sites and ballcourts, using the different impedance surfaces (1 = Mogotes del Panteón; 2 = Ejido de los Canseco; 3 = Llano las Casas; 4 = El Sitial; 5 = Los Chorros; 6 = Los Limares; 7 = Cerro Maluco; 8 = El Cucharital; 9 = Lachixoba; 10 = La Puerta; 11 = Terezona; 12 = Cerro del Convento). 659 Figure 8.21: The cumulative viewsheds of the Classic period sites. 663 Figure 8.22: The cumulative viewsheds of the Postclassic period sites. 667 Figure 8.23: The cumulative viewsheds of all habitational sites. 673 Figure 8.24: The cumulative viewsheds of the random points sample. 675 Figure 8.25: Visibility map of where the ballcourts can see and be seen in the Classic period. 678 Figure 8.26: Visibility map of where the ballcourts can see and be seen in the Postclassic period. 681

Figure 9.1: The 706 Figure 9.2: The yoke with water-toad symbolism showing the carved top (upper) and the side/front (lower). It is currently in the collection of the Casa de Cultura, Nejapa de Madero. 708 Figure 9.3: Map of the two ballgame traditions present in Nejapa during the Middle Classic period. 714 Figure 9.4: Stepped aprons can be seen at several ballcourts in the Central Valleys, including the ballcourt on the Main Plaza at Monte Albán. 727 Figure 9.5: Map of showing the locations of the ballcourts and ballcourt icons in the Nejapa and Quiechapa subregions. 731 Figure A.1: Example of the ceramic sherds and rims observed on the surface around the ballcourt at El Cucharital. 737 Figure A.2: Ceramic sherd with incised design, El Cucharital. 738 Figure A.3: Ceramic supports in two different styles, El Cucharital. 739 Figure A.4: Fragment of a possible patojo, a vessel used ethnographically to cook beans. 740 Figure A.5: Broken sahumador with partial handle, bowl, and rim, El Cucharital. 741 Figure A.6: Examples of miniture ceramics that were often used in ceremonies in the Nejapa subregion. In the bottom right corner is a miniature cup. 742 Figure A.7: Fragment of red-painted stucco, likely a piece of wall from a nearby structure. 743 Figure A.8: Figurine piece, possibly part of a headdress with the vertical, curved lines representing feathers. This piece is flat on the back. 744 Figure A.9: Fishing stone weight from top (upper) and side (lower) showing the groove where the net rope would have been tied around the stone. These were found all over El Cucharital. 745 Figure A.10: Obsidian blades found near the ballcourt at El Cucharital. Such blades were often used in rituals and broken afterwards. 746 Figure A.11 Ceramic grayware serving bowl measuring over 10 cm in diameter, MDP. 747 Figure A.12: Serving vessels and bowls from the midden behind the East Mound near the ballcourt. The sherd on the upper right has a applique. 748 Figure A.13: Serving vessels and bowls from the midden behind the East Mound. The sherd in the lower left has a brown wash and red border, a style commonly found in the Mixteca during the Ramos phase or Late Formative. 749 Figure A.14: Bottom section of a possible ceramic spoon, MDP. 750 Figure A.15: A face-neck jar roughly dating the Late Formative and Early Classic. 751 Figure A.16: Rim sherd with braided lip, found near the ballcourt at MDP. 752 Figure A.17: Examples of different incised sherds, many with rims present. The sherd in the bottom middle was found in the East Mound midden, while the rest were found on the ballcourt. 753

Figure A.18: Examples of miniatures and sahumadores found near or on the ballcourt. The sahumador fragment is in the upper left corner; the miniature on the bottom row may have been an expedient incense burner based on the hole and attachment scar on the bottom. 754 Figure A.19: Examples of figurines found on or near the ballcourt. In the top row we have a possible leg; on the bottom, a torso. 755 Figure A.20: Handle from possible utilitarian serving jar, MDP. 756 Figure A.21: Examples of stucco fragments found in the ballcourt at MDP. 757 Figure A.22: Three metates were found in the ballcourt possibly reused as stones to line the mound. One mano was also found, suggesting that food preparations may have taken place near the court as part of game activities and ceremonies. 758 Figure A.23: Fragment of preserved standing wall at Structure A2, one of the first architectural groups encountered at La Baeza. 759 Figure A.24: Handle of courseware jar found near the ballcourt at La Baeza. 760 Figure A.25: Other ceramic artifacts found near the ballcourt. 760 Figure A.26: Grayware sherd, possibly from a narrow necked fine jar, with flared rim. 761 Figure A.27: Another example of a fine grayware jar rim, possibly a cantero. 762 Figure A.28: Course grayware sherd with an impression of woven material or tela. We found several examples similar to this. 763 Figure A.29: Pieces of a nearly whole vessel, found in Structure E. It is made of course orange- gray ware and the side walls featured two ceramic nubbins. 764 Figure A.30: One of the many brown colored chert nodules observed on the ground surface at La Baeza. 764 Figure A.31: Green obisdian microblade, likely reworked from a much larger piece. Obsidian of this type typically comes from Puebla. 765 Figure A.32: Bottom half of an arrowhead manufactured from white chert. 766 Figure A.33: Tenon nail stone found at Structure E. These were typically placed into the walls. 767 Figure A.34: A second tenon stone found at the Structure E.. 768

LIST OF TABLES

Table 3.1: Different indigenous names for the patolli dice game (Kendall 1980:12; Walden and Voorhies 2017). 113 Table 3.2: Location and dates of pelota game matches visited for the ethnographic field project. 118 Table 3.3: Number of ballcourts documented in the regions and subregions of Oaxaca. 181 Table 3.4: Ballgame equipment by macroregion. 197 Table 3.5: Iconography and symbols of the ballgame by macroregion. 203 Table 5.1: Total number of data points recorded at each ballcourt. 321 Table 5.2: Definitions of the spatial dimensions of the ballcourts. 322 Table 5.3: Description of the different structure element codes for some of the spatial dimensions. 323 Table 6.1: Sites with ballcourts in the Nejapa subregion. 328 Table 6.2: Site code definitions. 329 Table 6.3: Ballcourts registered in the municipality of Nejapa de Madero. 331 Table 6.4: Size dimensions of El Cucharital ballcourt and playing surfaces. 339 Table 6.5: Size dimensions of El Cucharital court mounds. 339 Table 6.6: Size dimensions of Los Limares ballcourt and playing surfaces. 352 Table 6.7: Size dimensions of Los Limares court mounds. 352 Table 6.8: Size dimensions of Cerro Malcuo ballcourt and playing surfaces. 365 Table 6.9: Size dimensions of Cerro Maluco court mounds. 366 Table 6.10: Size dimensions for Terezona ballcourt and playing surfaces. 374 Table 6.11: Size dimensions of Terezona court mounds. 375 Table 6.12: Size dimensions of Mogotes del Panteón ballcourt and playing surfaces. 384 Table 6.13: Size dimensions for the Mogotes del Panteón court mounds. 384 Table 6.14: Size dimensions for Ejido de los Canseco ballcourt and playing surfaces. 393 Table 6.15: Size dimensions for Ejido de los Canseco court mounds. 393 Table 6.16: Ballcourts registered in the municipality of Santa Ana Tavela. 398 Table 6.17: Size dimensions for the Lachixoba ballcourt and playing surfaces. 404

Table 6.18: Size dimensions of Lachixoba court mounds. 405 Table 6.19: Size dimensions of La Puerta ballcourt and playing surfaces. 412 Table 6.20: Size dimensions for La Puerta court mounds. 413 Table 6.21: Size dimensions for Llano las Casas ballcourt and playing surfaces. 420 Table 6.22: Size dimensions for Llano las Casas court mounds. 420 Table 6.23: Size dimensions for the La Baeza ballcourt and playing surfaces. 468 Table 6.24: Size dimensions for La Baeza court mounds. 469 Table 6.25: Size dimensions of the El Sitial court mounds. 478 Table 6.26: The overall size and central alley playing surfaces dimensions of the Nejapa ballcourts. 505 Table 7.1: The different spatial dimensions of the ballcourts analyzed in this chapter. 531 Table 7.2: Overall court and central alley size dimensions and statistics in total and by time period. 533 Table 7.3: The overall and central alley length/width (L/W) ratios and descriptive statistics in total and by time period. 538 Table 7.4: Length and width dimensions and statistics of the terminal playing alleys in total and by time period. 543 Table 7.5: The L/W ratio statistics for the terminal 1 and 2 alleys in total and by time period. 548 Table 7.6: Descriptive statistics for the north/east lateral mounds (Mound 1) and south/west lateral mounds (Mound 2). 554 Table 7.7: Descriptive statistics for the north/east (Mound 3) terminal mounds and south/west terminal mounds (Mound 4). 555 Table 7.8: Descriptive statistics of the minimum, maximum, and mean values for the interior lateral surface slope grades in total and by time period. 566 Table 7.9: The ordinal (degree) and cardinal orientations of the Nejapa ballcourts by time period. 571 Table 7.10: Frequencies of the different cardinal orientations of the Nejapa ballcourts. 572 Table 7.11: The percentage of ballcourts oriented north-south (NS) and east-west (EW) by time period. 573 Table 7.12: Central alley size dimensions of the Central Valley ballcourts in total and by time period. 582 Table 7.13: Size dimensions of the north/east (Mound 1) and south/west (Mound 2) lateral mounds of the Central Valley ballcourts. 587 Table 7.14: Size dimensions of the north/east (Mound 3) and south/west (Mound 4) terminal mounds of the Central Valley ballcourts. 588

Table 7.15: Summary statistics of the lateral and terminal mound L/W ratios of the Central Valley ballcourts. 589 Table 7.16: Summary of the central playing alley dimensions of the Central Valley versus Nejapa ballcourts. 594 Table 7.17: Size dimensions of the lateral and terminal mounds of the Central Valleys and Nejapa ballcourts. 599 Table 8.1: Codes and definitions for habitational and non-habitational site types. 614 Table 8.2: The normalized values for the different variables and the Ballcourt Indices of the ballcourt sites. 637 Table 8.3: The efficiency scores for each model comparing habitation sites and ballcourts overall and by time period. 645 Table 8.4: The LCP analyses Series 1 - 5 and the origin and destination points. 647 Table 8.5: The original and adjusted slope values and their reclassified values as well as the original distance from river values and their reclassified values. 647 Table 8.6: Table showing the cells visible to observer points (Classic-period ballcourts) and sets of observer points (highlighted in gray). 679 Table showing the cells visible to observer points (Postclassic-period ballcourts and sets of observers (highlighted in gray). 682

ABSTRACT

The Prehispanic ballgame was an important ritualized sport and shared cultural tradition throughout much of Mesoamerica, and has long been linked to local and regional sociopolitical relationships between different communities and social groups. The widespread popularity of ballgames prior to the Conquest can be noted in many different ways, and the persistence of games through the centuries to the modern day demonstrates their continuing importance for indigenous and mestizo communities. However, the role of the ballgame in local and regional politics would have varied over time as different places and people innovated and reacted changing social and political circumstances. The Nejapa subregion of the eastern Sierra Sur, where multiple ballcourts of various types where recently documented, was a multicultural and multilingual boundary zone. The presence of several ethnolinguistic groups with possibly distinct social identities, who were alternately hostile and cooperative towards each other, would have greatly impacted local and regional political machinations and how people made sense of their changing world. Because of its very nature as a competitive sport requiring at least two teams to play, the ballgame would have been one way to negotiate movement across different social and cultural boundaries. How ballgames would have been used as a social or political strategy, would have changed over time through the decisions and actions of the people who hosted, played, and watched these games.

In the present dissertation, I examine the spatial attributes of the ballcourts in Nejapa, and the distribution of both the structures and their attributes on the physical and social landscape from the Late Formative through the Postclassic (300 BCE – CE 1521) in order to assess the

33 degree to which the ballgame was interwoven, or not, with local and regional community politics. While traditional methods such as excavation can be time intensive and cost- prohibitive, collecting mapping data and analyzing it using Geographic Information System

(GIS) is a non-invasive method that can be used to explore ballcourts and the ballgame at the regional scale. In this dissertation, I explore the ballcourts using a synthesis of in-depth geospatial analyses, archaeological survey and excavation data, and ethnographic information gathered from local communities to understand what distribution patterns of ballcourts and their attributes reveal about the ballgame, social relationships and political organization in the Nejapa region and the southeastern Sierra Sur.

Analysis of the data shows that there is both a remarkable degree of similarity but also some marked differences in the Nejapa ballcourts from the Late Formative through the

Postclassic, suggesting that were shared ways of playing and perhaps even similar rule systems but that communities still constructed their courts according to their specific needs and preferences. In other words, I argue, there were local traditions of shared practices or traditions such that communities could negotiate rules and play techniques between then. Based on both the material and linguistic evidence as well as the different ballcourt types present here, I conclude that people in Nejapa were potentially making connections to various ethnolinguistic groups and regions farther off, resulting if not in mixed, heterogeneous communities, then in relationships that frequently crossed ethnolinguistic boundaries that many have previously treated as being more static or impermeable.

Nejapa’s strategic position along critical trade corridors and its rich alluvial land made it an attractive location for many people, who through their shared experiences of living in this frontier region forged new identities that at times were communal or contrasting while at the

34 same time exposing them to novel ideas and practices that were differentially adapted and innovated. Over the course of history we have periods in Nejapa where multiple courts were in use, and control of the game may have been decentralized, and other periods in which there was only one court and access would have been controlled by the leaders of the communities that built them, signaling different strategies for employing ballgames and their associated ceremonies depending on the particular social and historical circumstances in which they were played. Despite these changes, the game still had a critical role to play in the local and regional politics of the various heterogeneous communities of Nejapa. Overall, the dissertation demonstrates how GIS technology can be used to investigate ballgames and ballcourts as social fields and spaces at different spatial and temporal scales.

CHAPTER I INTRODUCTION

“Although we know that Mixtec kings and queens constructed and used ball courts near their palaces, the persistence of pictographs for ball courts at outlying sites suggest that the game was played on boundaries between groups to settle land disputes, wager or exchange gifts to consolidate alliances…[they were places to celebrate] their common origins and their factional distinctiveness” Pohl et al. 1997:225

“…es parte de la tradición, antes era negocio de los ricos and los reyes, piezas de oro, animales de pluma fina… en el pasado, jugaban en las fiestas religiosas…”

[it] is part of the tradition, before it was [the] business of the rich and the kings, pieces of gold, animals with fine feathers…in the past, they played in the religious festivals…” Don Jose Angel

1.1 INTRODUCTION

Over the course of two field seasons in 2009 and 2011, the team from the Archaeological

Project Nejapa-Tavela (PANT in Spanish) documented eight previously unknown ballcourts in the Nejapa subregion, a large valley located roughly halfway between the highland Central

Valleys and the tropical Isthmus of Tehuantepec in Oaxaca, Mexico (Figure 1.1). During the

2013 field season, the year I joined the team, an additional seven courts were recorded, almost doubling the number of known structures in the area (Figure 1.2). After a long day of analyzing ceramics in the hot July summer, and with this knowledge in mind, I decided to take a quick look at how the ballcourts were distributed on the landscape. A casual visual assessment in ArcGIS indicted that many of the ballcourts appeared to be located close together. Running a simple

36 analysis in ArcGIS using the Buffer tool to place 2 km buffers1 around the sites with ballcourts, I could see that many were located within walking distance from each other (see Figure 1.2). If we estimate the average person’s walking speed at 5.0 km per hour (Aspelin 2005; Pingel 2010), then most of the ballcourts in Nejapa would have been located within a few hours’ walk of each other.

Figure 1.1: The southeastern state of Oaxaca and the Nejapa study region.

1 We first tried 5 km buffers, but the map results proved to be too messy and inconclusive. Decreasing the size of the buffers refined the results. 37

Figure 1.2: Map showing the ballcourts with two-kilometers buffers placed around them.

Previously, researchers working in the Central Valleys region had identified a pattern of distribution where most ballcourts were located in settlements on the regional or subregional boundaries of polities (Kowalewski et al. 1991). Other researchers have also proposed that elsewhere in Oaxaca, ballcourts served as sites for boundary-making or pilgrimage rituals2, neutral spaces where elites could interact socially with each other in peer-polity networks

(Byland and Pohl 1994; Zeitlin 1993). Additionally, some have proposed that courts located in political centers or militarized frontiers were used as practice spaces for warriors in between

2 Though this conclusion is not always based on the spatial distribution of the courts themselves. 38 periods of conflict (Kowalewski et al. 1991; Redmond 1983). Contrastingly, it appears that the ballcourts in Nejapa are not only found in close proximity to each other, but they were also constructed in settlements that were neither located on any clear political boundaries nor clearly acting as regional or subregional capitals. Looking at the PANT data, I wondered if there was indeed a pattern of ballcourt distribution in Nejapa that was distinct from other regions of

Oaxaca, and therefore indicating a different social motive for playing the game, or if this seeming density of courts disappeared when we factored in time and our current understanding of the settlement chronology of the subregion. In other words, what role might the ballgame have played in the sociopolitical relationships among the communities in Nejapa, and how did this role change or fluctuate over time in response to larger historical events?

Nejapa was a multicultural and multilingual boundary zone (sensu Love 2011) where the territories of three distinct ethnic groups converged, which only complicates our understanding of the complex relationship between courts, games and social and/or physical boundaries. The presence of several ethnolinguistic groups with possibly distinct social identities, who were alternately hostile and cooperative towards each other, would have greatly impacted local and regional political machinations and how people made sense of their changing world. Such relationships would have been complicated further still during the Postclassic period (CE 900 –

1521), when Zapotec and Aztec military forces passed through Nejapa on conquest campaigns to the Isthmus. Scholars have long linked ballcourts and games with religion and politics, placing emphasis on the importance of ballcourts and the ballgame during times of sociopolitical change

(Feinman and Nicholas 2011; Scarborough and Wilcox 1991). Ethnographic research on the pelota Mixteca (Chapter 3), a modern ballgame played today in Oaxaca further shows how ballgames can both integrate across social and physical boundaries as well as maintain

39 separateness between groups and individuals, much like how sports operate in our society today

(Stoll 2015; Stoll and Anderson 2017). Did ballgames serve to heighten or lessen tensions between Nejapa’s different communities? Moreover, can we connect changes in ballcourt distribution over time to periods of hostility or interethnic conflict both locally and with outside groups, especially during the Postclassic?

In the present dissertation, I attempt to answer these questions and others posed in this chapter by examining the spatial attributes of the ballcourts in Nejapa, and the distribution of both the structures and their attributes on the physical and social landscape. In doing so, I aim to assess the degree to which the ballgame was interwoven, or not, with local and regional community politics, and how the game and related ritual activities may have been manipulated as

Nejapa’s diverse residents adjusted to shifting sociopolitical landscapes from the Late Formative

(BCE – CE 250) through the Postclassic periods (CE 900-1521).

1.2 UNDERSTANDING PREHISPANIC BALLGAMES IN THE 21ST CENTURY: BALLCOURT DISTRIBUTION AND SOCIAL ORGANIZATION IN NEJAPA, OAXACA

Archaeological, ethnohistorical, and cultural material evidence strongly suggests the

Prehispanic ballgame was an important ritualized sport and shared cultural tradition throughout much of Mesoamerica. It was played in, though not necessarily restricted to, masonry ballcourts

(Miller and Houston 1987; Stark 2012) that across regions and time periods share remarkable similarities in design and architectural elements (Taladoire 1981). Over 2500 ballcourts in some

1200 archaeological sites in Mexico alone have been identified and recorded since the past century (Taladoire 2017). Depictions of both ballcourts and the different ballgames that were played are found in a variety of pictorial media, ranging from Prehispanic and Colonial codices

40 to modeled figurines and carved stone monuments (Day 2001; de la Garza 2000; Eckholm 1991;

Miller 2001; Miller 1989). The widespread popularity of ballgames prior to the Conquest can be noted many different ways (Chapter 4), and the persistence of games despite the frequent and often harsh intervention of Colonial Spanish authorities demonstrates their continuing importance for indigenous and mestizo communities. Even to this day, variants of the ballgame are played in different states in Mexico as well as in the United States wherever Oaxacan migrant communities are found (Stoll 2015; Turok 2000) (Figure 1.3).

Figure 1.3: Playing pelota mixteca in the town of Magdelena Jaltepec during the patron feast day.

According to arguments presented in the current literature, both ballcourts and the ballgame(s) (see Cohodas 1975; Fox 1996; Hendon et al. 2009; Miller 1989; Pasztory 1976;

Santley et al. 1991; Whalen and Minnis 1996; and Wilkerson 1991 for varying interpretations) played multilayered and complex roles in Mesoamerican societies. Beyond the religious and cosmological aspects typically associated with the game, scholars have also suggested that it served some social purpose such as solving or heightening conflicts within and between communities (Fox 1994; Fox 1996; Hill 1999; Kowalewski et al. 1991; de Montmollin 1997;

Stark 2012). Additionally, games also presented opportunities for economic transactions to occur, as markets were often organized as part of the festive activities accompanying game events. Colonial documents indicate that land, resources, and tributary goods often exchanged hands through the practice of gambling by players, sponsors, and audience members, including the elites (Hill et al. 1998; Santley et. al 1991).

Despite the amount of investigation into ballcourts and the ballgame since the 20th century (Taladoire 2012), several gaps remain in our knowledge and understanding of this important sociopolitical institution. As I will discuss further in Chapter 3, research has primarily focused on the Maya, Gulf Coast, and Basin of Mexico areas. This is not surprising since this is where the evidence on ballcourts and ballgames is, on the surface at least, the richest and most extant. Investigation into this phenomena in other areas of Mesoamerica outside of the aforementioned regions has been less intensive in comparison, consisting mainly of region-wide surveys where ballcourts are documented as a part of the study but are not the primary focus; or intensive excavations at one or a few sites only, typically limited to the architectural structure itself. In many cases, not enough attention has been paid to the specific systemic and cultural context in which the ballcourts and the game were active within a given region and time period.

In some respects, the very ubiquity of ballcourts and the ballgame, in both physical and iconographic contexts has led to assumption that they actively played similar roles (political,

42 religious, economic, etc.) or held similar meanings (ideological, social, cultural, etc.) in the sociopolitical relations among and between Mesoamerican communities across space and time.

Yet the sociopolitical processes that made both the games and the courts on which they were played efficacious, or not, to the participants and audience members who were invited have not been adequately accounted for within these explanations. What was it about ballgames that caused people from different communities, social and ethnolinguistic groups, and social statuses to be invested in their planning, performance, and outcomes, and to be so invested over such a long period of time? Similarly, what caused people to lose interest in games, such that game material culture disappears from some places for periods of time, or even permanently? Why do we see areas where there are multiple courts and others where there is only one, and how does that link back to local and regional sociopolitical organization?

Over two decades ago, Fox (1994) suggested that in order for archaeologists to understand how ballcourts were actively used by people, it is important to thoroughly examine their sociopolitical functions within specific regions. He argued that what was most often missing from previous ballgame studies was an “understanding of ballcourts as ‘lived space’

(l’espace vécu)” (Fox 1994:1). In-depth analysis of multiple ballcourts in a region using traditional methods such as excavation, however, can be time intensive and cost-prohibitive.

Collecting mapping data with a total station and analyzing the data using Geographic

Information System (GIS) programs is one non-invasive method that can be used to explore ballcourts at the regional scale. Therefore, the present research has been designed to explore the

Nejapa ballcourts using a synthesis of archaeological survey and excavation, in-depth geospatial analyses, and ethnographic information gathered from local communities. In particular, I focus on the following three analytical questions:

1. Are there observable spatial patterns associated with the ballcourts in Nejapa in

respect to their attributes and distribution?

2. Based on these patterns, what strategies of ballcourt emplacement and ballgame

practice were used by Nejapan communities more broadly?

3. What do these patterns and strategies tell us about the ballgame, social

relationships and political organization in the Nejapa region and the southeastern

Sierra Sur overall?

The construction and organization of social space influences the shaping of social relations and daily practices (Dovey 1999; Foucault 1979, 1984). Through the construction and variable use of social space by a diverse range of people, the built environment becomes a material manifestation of the practices employed by both communities and individual actors to negotiate power and community identity (Casella and Fowler 2005; de Montmollin 1995; Kolb and Snead 1997; Tsukamoto and Inomata 2014; Turnbull 2002; Villamil 2005). One important goal of research on public architecture is to consider why monuments were constructed in the specific locations where they are found on the landscape (Turnball 2002). How buildings are distributed both within sites and on the landscape often has sociopolitical significance relevant to the settlement hierarchy, the social networks among individual communities, and residents’ access to political power.

Architecture and the built environment are social practices that constitute community in several ways (Barber 2005:59; de Montmollin 1995:6). Chief among them, and most critical to the research presented here, is that the construction of large-scale public monuments necessarily involves the participation of large numbers of people. Communal participation in labor projects creates group cohesion and a sense of communitas (sensu Turner 1969) among individuals

(Canuto 2002:71–72; Joyce et al. 2013:136). Subsequently, architectural styles, construction techniques, and spatial layouts often reflect particular sets of practices and cultural preferences that are learned within specific communities however defined, i.e. communities of practice (Lave and Wenger 1991; Wenger 1998, 2000). Such differences may be especially marked in situations where multiple groups with different cultural histories lived alongside one another

(Aldenderfer and Stanish 1993; DeCorse 1989; Stark et al. 1998; Voss 2005). Thus, how a ballcourt was constructed and used as well as where it was emplaced within public space would have expressed particular underlying social, political, and symbolic meanings (Fox 1994; Fox

1996:19; Villamil 2005).

If the ballgame was deeply imbricated with the political and religious practices of many

Mesoamerican societies, then those communities that constructed ballcourts and hosted ballgames would have distinguished themselves socially and politically from other communities that did not (Canuto 2002:728; Hill and Clark 2001; Joyce and Hendon 2000). Moreover, as ballcourts are evidence for both periodic and sporadic (i.e. special or momentous occasions) interactions with neighboring and non-local groups, then the presence of these important facilities in particular sites would have worked to further reinforce intra- and intercommunity social distinctions (Canuto 2002:740; Hendon 2010). In this way, as I explore further in

Chapters 3, ballgames would have helped to establish and maintain relationships both negative and positive between social groups. Public buildings like ballcourts were therefore probably instrumental in marking the community as identifier and a distinct place (Clark 2004).

Spatial analyses have long been used as a means of exploring the built environment and how social space has been consciously subdivided and allocated, taking into consideration elements such as access, visibility, capacity, and use (Fairclough 1992). As spatial creatures,

45 human beings articulate with the surrounding environment through both our physical bodies

(Butler 1990) and culturally mediated practices (Bender 1993; Thomas 1993). Human landscapes both natural and cultural are socially constructed and historically contingent, and they are marked by distinct institutions and systems through space and time (Bender 1993:1999;

Smith 2003; Thomas 1993; Wansleeben and Verhart 1997; Witcher 1999). If we consider the ballcourt as a lived space where human social activities integral to communal relations were performed, then taking into account where ballcourts were emplaced in the landscape and how they might have been used is a fruitful avenue for investigating the ballgame’s role in local and regional politics. My goal therefore in this dissertation is to present a three-dimensional or experiential view of the ballcourts through an integrative approach combining statistical analyses, GIS spatial analyses, and phenomenological descriptions.

To add to Fox’s suggestion of looking at ballcourts as “lived spaces”, I would suggest that we must also pose the central question of how sports are connected to communities and social identities. Sports, games, and other similar activities are global and historical phenomena, and they are present in nearly every known society (Blanchard 1996; Kelly 2007; 2009).

Heavily commercialized as they are, modern sports still have real social, political, and economic consequences for many different types of communities, from fans of teams (both individually and collectively) to whole nations. For example, several observers have remarked on the so- called “Olympic effect”, where cities selected to host the Olympic Games are expected to commit large sums of money, with national leaders hoping to realize economic benefits that actually never materialize. Riots by celebrating fans in Philadelphia after the local NFL team, the Philadelphia Eagles, won the 2018 Superbowl Championship, led to much social commentary about the way African-Americans advocating for social justice are treated by police

46 and presented in the media versus these white sports fans who were causing damages to public property in their celebratory enthusiasm. The importance of sports even in today’s society indicates strongly that sports and games played similarly important roles in the past. By looking at how social theorists have viewed sports (Chapter 2) and by placing Mesoamerican sports in their geographic and ethnographic context (Chapter 3), we can better understand how the ballgame could be linked to community politics.

1.3 PROYECTO ARQUEOLOGICO JUEGOS DE PELOTA NEJAPA-TAVELA 2015: RESEARCH DESIGN AND METHODS

Despite its proximity to the Central Valleys – where archaeologists have been conducting research for decades – and the Isthmus, the Nejapa subregion had seen little investigation apart from a few ethnohistorical studies in the 20th century. In 2009, Dr. Stacie King initiated the

PANT with the goal understanding the Prehispanic and Colonial history of Nejapa via archaeological survey and excavation, and archival research. Over the course of three field seasons, Dr. King and crew have documented a complex archaeological and historical record with occupation beginning as early as the Early Formative (BCE 1500 - 750) period and continuing through to the present day (Chapter 4).

According to archaeological and ethnohistoric research by PANT, sites in Nejapa were occupied by people from distinct sociolinguistic, political, and ethnic groups. Though the composition of these communities with regards to their specific ethnolinguistic identity is not certain, current research indicates such identities were likely varied and mixed across the region, resulting in the distinct material assemblages observed at many sites (King and Zborover 2015).

Similarities in the material assemblages between sites at locations on the valley floor and in the mountains suggests some degree of interaction among these communities that was probably

47 alternatively competitive and cooperative. At the very least, access to the same productions centers or trade networks was actively negotiated between members of different, contemporaneous settlements. Different styles of architecture within and between settlements, on the other hand, may reflect distinct cultural preferences and construction techniques; in other words, localized practices that perhaps link back to fluid and nested social identities.

One way in which Nejapa residents may have actively negotiated internal and external political tensions as well as resolved (or deepened) socioeconomic relationships was through constructing ballcourts and alternatively hosting and attending ballgame events. Competitive, probably friendly but occasionally hostile sport matches and associated communal performances

(via feasting, gambling and other rituals) may have been important arenas for sociopolitical negotiations in situations where multiple languages are spoken, and multiple, heterogeneous social identities are at play3. Whether the outcomes of these communal gatherings resolved or deepened social situations, they still were efficacious and affected social relationships in both predictable and unforeseen ways. For the people of Nejapa, living in an area of interregional movement and transportation and surrounded by ethnolinguistic groups that were often hostile to each other, the ability to negotiate in this way might have been especially important (King 2012).

In the present dissertation, I investigate the role of ballgames in the sociopolitical relationships within and among communities in Nejapa, and between these communities and other Oaxacan regions nearby. I do so by comparing and contrasting 1) the spatial dimensions and design characteristics of the ballcourts; and 2) the distribution of these same courts at the regional and temporal scale, placed within the larger context of macro- and microregional settlement histories. By tracing the history of ballgame in Nejapa, how it may have waxed and

3 Of course, ballgames could also resolve or deepen social tensions among homogenous populations and/or highly centralized settlement systems as well. 48 waned in importance as politics and settlements changed, we can move towards recovering the social meaning of the Nejapa ballcourts as lived “public spaces” (Fox 1994:30).

Mapping took place from the end of May to the end of July, 2015. At each site, detailed information on the ballcourt form, its location within the site, the structures nearest to the court, as well as the access points to and from the court, if present and observable, was collected and recorded in a field notebook4. Each ballcourt was then systematically mapped using a Sokkia

SET610-313 total station in order to collect more accurate measurement of basic structural details, measurements, and elevations. The resulting data points were then entered and processed in an ArcGIS spatial database. Once processed, these data were analyzed using both ArcGIS and the statistical package SPSS.

1.4 ORGANIZATION OF THE DISSERTATION

Over the following chapters, I explore the arguments and ideas about the Mesoamerican ballgame, using the Nejapa subregion as a case study. I discuss the theoretical framework that structured the research design and questions in Chapter 2. Using current perspectives grounded in poststructuralist approaches, I highlight the particular interacting and overlapping social fields that I identify as being salient to the social efficacy of the Mesoamerican ballgame: ritual practices and beliefs, feasting events, and architecture and spatial organization. Poststructuralist approaches emphasize the situational relationships between politics, power, and identity as expressed or manifested through human agency and practice. Thus, by examining how our daily and extraordinary practices are expressed through both material culture (DeMarrais et al. 1996) and spatial organization (de Certeau 1984; Foucault 1979), archaeologists can best study how

4 Given destruction of sites from modern agricultural activities, access points may have existed in the past that are not seen today. 49 communities and social identities were created, negotiated, maintained, and contested in the past

(e.g. Barrett 2000; Bell 1992; Dornan 2004; Fogelin 2007; Hodder 2000; Joyce and Hendon

2000:143; Pauketat 2001; Swenson 2008). In particular, the insights gained from poststructuralist approaches can be especially useful in looking at the role of sports in community and social identities.

One the larger goals of the dissertation is to push ballgame theory forward to more actor- based perspectives. I do so by addressing several issues I see as critical if we are to expand our view of the Mesoamerican ballgame and how they figured into local and regional sociopolitical relationships. In Chapter 3, I begin by reviewing the current literature on the ballgame and the ballcourts, focusing on arguments concerning the game as a sociopolitical institution linked to complex political relationships and social identity in Mesoamerica. I then begin challenging some of these prevailing assumptions by exploring the ethnohistorical evidence on indigenous sports of the Americas, building on the discussion about sports and games from Chapter 2.

Because modern sports and games still have something to tell us about similar practices in the past, I present ethnographic data on the traditional ballgames played today in Mexico, focusing specifically on pelota Mixteca de hule, an indigenous ball sport of Oaxaca. With respect to the archaeological data, part of the issue with ballgame studies is that the material culture from different regions is treated uncritically and diverse items are often lumped together as a singular phenomenon. Therefore, I take a critical look at the material culture of the archaeological ballgame at the regional scale, examining the diversity that existed in game artifacts and iconography across different regions and time periods. Finally, I use the preceding evidence to address the prevailing assumptions about the Mesoamerican ballgame laid out in this chapter. I then present possible models for how ballcourt spatial distribution could be linked back to

50 political organization and social negotiations.

Ballcourts, as I will argue in this dissertation, represent a multitude of embodied meanings and social relationships, more so since many courts continued to be used long after the associated site was abandoned, if not for ballgames, then for other ceremonial activities (Pohl and Byland 1990). Such diversity stems in part from the variable nature of the games themselves. As an event, a game would have been preceded by and concluded with many other rituals and ceremonies; a game could represent in one event the ritual enactment of a cosmological myth while also mediating a conflict between hostile parties. Understanding the spatial and temporal variation in the material and symbolic culture associated with the game and the courts, and how scholars have interpreted and understood this variation, helps to contextualize the Nejapa data within the overall milieu of Mesoamerican ballgame traditions.

Chapter 4 provides the geographical and historical background of the Nejapa subregion, set within the larger context of the trajectory of sociopolitical development and change in

Oaxaca. Although often viewed today as a provincial frontier backwater, the evidence shows that the dynamic changes that occurred throughout Mesoamerica and in Oaxaca directly impacted settlement organization and community relationships in Nejapa. I present the ongoing research from PANT studies and describe what we currently know about the subregion’s history, beginning with the appearance of the earliest villages in the Early and Middle Formative (2000 -

500 BCE) periods through to the political developments of the Classic and Postclassic periods

(CE 250 – 1521) (King 2010, King et al. 2012, King et al. n.d. Vols I and II). Through my presentation of this historical background, I aim to expand our understanding of Oaxacan history beyond just the Central Valleys and Mixteca Alta.

I present the research design and methodology that guided the field project and data

51 analyses in Chapter 5. I specifically discuss the different spatial analytic techniques that were employed in this dissertation to further interrogate the ballcourt mapping data. In doing so, I also address critiques of these same analyses (Gaffney et al. 1996; Kvamme 1995; Lake et al. 1998;

Wansleeben and Verhart 1997; Wheatley and Gillings 2000), and how I try to account for them here. Chapter 6 describes the results of the site reconnaissance conducted at each site, focusing on each individual ballcourt, its measurements, and emplacement within its particular settlement.

Additionally, I provide the background on each site, giving the social context for each court based on site descriptions from PANT.

Chapters 7 and Chapter 8 present the results of the statistical and spatial analyses, respectively. Similarities in measurements, proportions, design and/or style, and construction techniques can point to shared practices among communities at different spatial scales. For ballcourts, certain structural dimensions, such as the size and proportions of the central playing surface and terminal alleys, would have more directly impacted game play (Baron 2006).

Statistical analyses help to identify and determine the significance of patterns in ballcourt size dimensions (Chapter 7). Spatial analyses in GIS computer programs allow for further interrogation of the spatial data through the identification and visual display of significant patterns.

I synthesize the different lines of evidence explored in the dissertation in Chapter 9, drawing on the results from the spatial and statistical analyses as well as the study data from

PANT to answer the research questions posed earlier in this chapter. I look at how the ballcourts, through their emplacement and use, and the ballgames themselves were social strategies for negotiating political power in multiethnic and multilingual Nejapa, and how these strategies may have shifted over time as larger historical events and trends unfolded. I then link

52 the data from Nejapa to a larger interpretation of ballgame traditions in Oaxaca, comparing and contrasting local and regional patterns. I also address the limitations of a research study of this type and the gaps that exist when using spatial data alone. Finally, I conclude the chapter with future directions for research on the ballcourts in Nejapa, and in Oaxaca and Mesoamerica more broadly.

1.5 CONCLUSION

The research described in this dissertation represents the first detailed study of the recently documented ballcourts in the Nejapa region of Oaxaca. I have designed the research to explore how the ballgame was linked to community and regional politics using non-invasive methods. The goal of the dissertation is to elucidate how local and regional community politics may have been mediated through both the social field of the ballgame and the public, lived social space of the ballcourt, using data from a specific region and sociohistorical context. The information on the Nejapa ballcourts in combination with the larger corpus of knowledge on ballgames in Mesoamerica improves our understanding of two critical research issues within ballcourt studies. First, how and why the ballgame endured as an important social institution over the course of several millennia; and second, why diverse groups living in a wide range of geographic and cultural settings adopted and adapted ballgame rituals and practices, not only in

Mesoamerica but also as far north as Arizona in the Southwestern United States, as far east as

Puerto Rico, and as far south as the Orinoco River.

The results of this research also contribute both to the ongoing investigation into the

Prehispanic past of Nejapa and, through a regional focus on ballcourts in a unique place and time period, to a richer and more-encompassing picture of the political salience of the ballgame both in Nejapa and Mesoamerica more broadly. Finally, the research contributes to expanding our

53 knowledge of those regions outside the more heavily investigated areas of Oaxaca through exploration of the archaeological evidence from the understudied Nejapa region.

CHAPTER II

IMAGINING COMMUNITY AND SOCIAL IDENTITY IN ARCHAEOLOGICAL PRACTICE: THEORETICAL AND METHODOLOGICAL IMPLICATIONS

“The local community is a fundamental unit of human society, and all people are members in at least one community” Kolb and Snead 1997:609

On frontiers: “Here it is possible both to escape from the cultural conventions of one’s own society and to make contact with people carrying other conventions, other ways of living, thinking, and organizing social groups” Rodseth and Parker 2005:9

2.1 INTRODUCTION Ballgames have long been associated with the ritual activities of Prehispanic elites and their ability to consolidate political power through these performances. But how did ballgames come to carry such social and political weight? More importantly, why and how do ballgames, and sports in general, play a role in the formation and negotiation of communities and social identities? A fruitful way to explore these questions is to imagine all sports, ballgames included, as social fields, or relational settings with interacting actors, structures, and practices. Bourdieu

(1984, 1985) saw social fields as “enduring domains of habitual practice”, while the Manchester school (Turner 1957) described them as dynamic, changing, and volatile (Postill 2013). As we actually live and experience them, social fields probably lie somewhere in between these two perspectives. That is, social fields contain both conservative elements and the potential for change, they are habitual and emergent. Importantly, social fields link to, overlap with, and are

55 imbricated in other social fields – they can be both horizontally linked and hierarchical, nested within varying relational networks. Leaving aside Bourdieu’s notions of class relations – as explicitly understood within a Western historical tradition – exploring how sports act as a social field can be a workable approach for investigating the role of the Mesoamerican ballgame in community and social identity production and negotiation.

In Mesoamerica, ballgame events were often accompanied by rituals and feasting. They occurred in specially marked places, whether in the ballcourts themselves or on playing grounds staked out using decorated field markers (discussed further in Chapter 3). Many times, a ballgame match was played as a part of larger religious and political ceremonies. Evidence suggests that games were important legitimation strategies in both hierarchical and heterarchical political systems. Some ballgame practices were strongly identified with specific cultural traditions both local and transregional, while other practices cross-cut social and spatial boundaries, tying different regions and communities of Mesoamerica together. The ballgame endured for several millennia, but also waxed and waned in popularity and importance.

Therefore, the Mesoamerican ballgame was arguably a social field through which people negotiated sociopolitical relationships, community and group membership, and social identities.

In this chapter, I present the theoretical considerations that guided the research and analysis stages of the present thesis. With respect to Mesoamerican ballgames, I propose that there are specific practices that, broadly speaking, make it especially efficacious and impactful on community and identity formation and negotiation. These include associated cultural traditions (i.e. social fields) such as feasting, ritual, and architecture. I explore the current theories on these themes throughout this chapter. Finally, since our job as archaeologists and social scientists is to translate the realities of our data to the thickness of social theory, I also look

56 at the different methods and lines of material evidence that are useful in investigating the practices highlighted here.

2.2 THE BALLGAME AS A SOCIAL FIELD: THEORY AND METHOD Human communities and social identities are multi-faceted, dynamic, and situational

(Bray 2003:3; Barber 2005). They are rooted in the shared daily practices and historical experiences of individuals, and are subject to continuity, transformation, and discontinuity

(Blanton 1994; Brumfiel and Fox 1994; Canuto 2002; Canuto and Yaeger 2000; Dobres and

Robb 2000; Jones 1997; Kolb and Snead 997; Knapp 1998; MacEachern et al. 1989; Marshall

2002; Pauketat 2001; Saitta 1994; Schwartz and Falconer 1994; Wills and Leonard 1994). The practices and experiences that constitute and define community and social identity take place in several different contexts within or across regions at different scales, in both intimate and public settings. In the following sections, I consider the specific practices that I argue are linked to the ballgames and that turn them into social fields through which people construct and negotiate their communities and social identities, including: ritual practices and feasting; social space, architecture and the built environment, and sports and games.

Critical questions in archaeology are how, where, when and with what communities and identities were created and negotiated in the past. Individuals use both novel and traditional daily practices as forums for reworking identity, often through employing material culture

(Silliman 2010:195). As I explore throughout this chapter, community and social identities, the act of group membership or belonging, are performed; they are things that people both think about and do. As a result, there often remains the material manifestations of past communities, social identities, and the practices that inform them, which can be identified through various lines of evidence. Doing so, however, is not simply a matter of drawing group boundaries based on

57 the spatial continuities and discontinuities in the distributions of material objects and practices.

This is because community and social identities are differentially manipulated and mobilized in various situations, the symbols of membership change through time, and groups can have varying perceptions of what constitutes social and/or ethnic differences (Jones 1997; Lightfoot and Martinez 1995; Rothschild 2006; Stone 2003). Frontier or boundary zones such as Nejapa, areas where individuals and groups from different homelands interact and form complicated sociopolitical and economic networks, present a different sort of challenge to archaeologists investigating community and social identity. In frontier interactions, social groups in contact with each other often share a variety of material cultural markers, but such markers may not figure into a group’s identity. Following from this, we should always take into account that communities and social identities are not simply monocultural aggregations of people who come together for planned and contingent reasons (Marshall 2002:215).

Poststructuralist perspectives have changed the way archaeologists approach rituals, feasting, landscape and the built environment, and the ways in which social actors interact with these overlapping social fields. While Hawke’s ladder of inference placed ritual and architecture at opposite ends of interpretive ability, postructuralist perspectives see them as intertwined and recursive. Both ritual and architecture carry social meanings that are subject to individual and group interpretation and social contestation. The relation of sport and society, on the other hand, is still somewhat undertheorized in anthropology, although interest in the topic has been growing over the past 30 years. Research from the field of sociology, however, demonstrates that sports play a more significant role in society than is currently acknowledged. Finally, our ability to connect these social fields together in order to understand community and social identity is

58 challenged by frontier and boundary zones, such as Nejapa, where the material evidence may be more equivocal or even expressed in novel way.

2.2.1 Religion, Ritual, and Feasting

Stated simply, ritual and religion are not solely ephemeral phenomena, but have always played a central role in the political dynamics of communities, both internally and in relation to larger sociopolitical systems (Earle 1990:116). The intersection of economic and ritual actions, the ritual economy, links ritual and religious beliefs to the negotiation of status and political authority at all scales of human social organization (Barber and Joyce 2007). Rituals, both those conducted in private and in public performances, communicate specific ideas or meanings through different types of media (Bloch 1989). Because rituals are linked to power and politics and can therefore help to alternatively preserve, alter, or destroy the social order, they are frequently subject to manipulation by different social groups (Bradley 1991, 2005; Kertzer

1988).

Ritual has been defined as an activity that is somehow set apart from daily practice (Bell

1992; Bell 1997; Humphrey and Laidlaw 1994; Humphrey and Laidlaw 2007). Turner (1982) describes ritual as a liminal phenomenon because participants undergoing ritual processes exist as beings-in-transition. The liminal dimension of ritual is important because it is how society reasserts allegiance to old values while simultaneously rejuvenating itself with new values and relational patterns (Rowe 2008:129). At the core of liminality is the playful experimentation that opens up social structures to contestation and negotiation, what Turner (1982a) refers to as the

“ludic” element or dimension. Although not often considered in current discussions of ritual, I agree with Rowe (2008) that the element of play Turner applies to liminality is important when

59 we consider how ritual and politics exist are interlocking social fields, especially in past societies. Novelty is brought onto the “field of possibility” through the spirit of spontaneity and creativity that play engenders and supports, thereby constructing both new expressions of social relationships and new arrangements of the social order (Rowe 2008:129). Social drama, marked by heightened tension in which unknown positive or negative outcomes are possible, is a universal form of cultural experience and an important element of ritual activities. The elements of social drama are organized by ritual into performances that resonate with the “collective experience, values, and cosmological vision of the community,” (Rowe 2008:131). Ritual is a collective effort put on by members of the community, and the ability to participate in planning and executing ritual ceremonies is a critical source of power and authority. The efficaciousness of ritual lies in its presentation and integration of different genres of performance into a cohesive event. In this manner, both change and sameness become important features of ritual performances (Rowe 2008:134).

Yet, meaning in performance or ritual activity can sometimes be ambiguous, even to the performers and spectators themselves (Bell 1992; Bell 1997; Dirks 1994; Humphrey and Laidlaw

1994; Humphrey and Laidlaw 2007; Inomata 2006; Inomata and Coben 2006b). Moreover, the ritual commitment or intent of the actor when performing ritual actions is highly variable

(Humphrey and Laidlaw 1994). This can make measuring the effectiveness of performance using material traces in the archaeological record especially difficult. There are many instances, for example, when a ritual is effective at affecting political change even when it does not happen at all (Dirks 1994), or does not happen in the proscribed way (Humphrey and Laidlaw 2007)

Feasting is a communal ritual activity that was particular significant socially and politically significant in pre-industrial societies (Bray 2003a; Dietler 1996; Hendon 2003; Potter

2000; Mills 2004, 2007). Feasting is distinguished from everyday consumption practices because food and drink are consumed communally and usually involve a large cross-section of the community and sometimes even outsiders (Bray 2003b; Brown 2000:370; Dietler 1996).

Food-related practices such as cooking, serving, and eating are subject to many social rules that are both prescriptive and proscriptive, especially when performed together as a community

(Valeri 1999). In this way, feasts become social fields where identity and status are constructed, communicated, asserted, negotiated, and marked through consumption rituals (Bray 2003b;

Weismantel 1988; Yaeger 2000), linking these activities to power, agency, practice, and the construction of value through symbolism (Hendon 2003).

As an important component of many ceremonial events, feasts were also opportunities for forming and maintaining alliances with other social groups or communities because they helped to create extended social networks (Brown 2000:386; Kan 2015 [1989]; Monaghan 1995).

Ashmore and colleagues (2004:311) view feasting as a way to forge intra-settlement bonds, making local identity more explicit while still defining and reinforcing local and regional hierarchies. Feasting events were also political legitimation strategies, serving to perpetuate and reify social positions and at the same time present opportunities for their renegotiation and contestation (Bray 2003a, 2003b; Blitz 1993; Hendon 2003; Lau 2002Wright 2004; Masson and

Orr 1999; Spielmann 2002). Feasts were not only important social practices for the elite, but were important events for all different kinds of social groups (Brown 2000:368).

Many have argued that feasts gain their efficacy through the commensal politics and practices, which are played out before, during and after the event (Bray 2003a; Bray 2003b;

Dietler 1996; Dietler 2003; Pollock 2003; Potter 2000; Pauketat et al. 2002). Like the dramatic spectacles that often accompanied them, there is a “repetitive and cyclical nature” to feasting

61 events that gives order to the socio-temporal network as perceived by social members (Hendon

2003: 24). The social relations engendered and reinforced at feasting events between hosts and guests are reciprocal in nature but also carry a sense of obligation (Hendon 2003; Monaghan

1996). This is because, like the rituals of gift-giving and receiving (Maus 1990) that sometimes accompanied these events (such as the Potlatch), feasts put into motion cycles of new and old debts. Ethnographic examples of such practices are still documented among Mixtec villages in

Oaxaca today (Monaghan 1996). The reciprocal relationships established between hosts and guests, benefactors and recipients, via cyclical exchanges of credits and debts in the fiesta system, were and still are often compared by the villagers themselves to the relationships between the gods and people (Brown 2003:368; Hendon 2003; Monaghan 1996).

2.2.2 Space, Architecture, and the Built Environment

After nearly five hundred years of scholarly inquiry and consideration, space has moved from an absolute, almost mechanical, Cartesian ontology – space as empty container –to a relational ontology where space is not constant, but constantly emergent and transforming based on the changing relationships between immovable and movable bodies (Conolly and Lake 2006;

Smith 2003:35-70). The division, categorization, and use/or manipulation of agricultural, residential, and public spaces demonstrates that space is equally implicated in both practice and structure (Bourdieu 1973; 1977; Giddens 1979, 1984; Knapp 1992). Without question, society and culture shapes the form, function, and meaning of space (Hall 1966). Extending from and articulating with spatial theory is the concept of landscape, which possesses a certain amount of ambiguity as a result of two competing apprehensions of its meaning. Landscape is alternatively seen as an object of scientific and geographic inquiry, subject to the rational classification and

62 measurement of its related parts; and as an ideological construct reflecting the ways in which human social groups experience and make meaning of their world in specific historical and cultural contexts (Cosgrove 1984:15-16 Anschuetz et al. 2001, 160-161; Ashmore 2002; Bender

2002; Bender et al. 1997; Kantner 2008; Knapp and Ashmore 1999; McCoy and Ladefoged

2009). Yet it is in this very ambiguity that we find the theoretical richness that makes landscape such a powerful analytic subject for archaeological inquiry.

Landscapes become heavily invested with symbolic and cultural meaning through their use and modification by human social groups, and in their role as mediators of cultural practice.

The mapping of “individual, local, national, and mythical histories” onto a landscape creates this articulation of the spatial and social (Richards 1998:98). At the same time, the meanings attached to landscape are neither singular nor static. Multiple, often contesting and contrasting, meanings of landscape exist within many different cultures, and an individual’s experience of their landscape depends largely on the specific set of economic, political, and social relations under which they live (Bender 1993:2, 246). Smith (2003:9) sums up landscape thusly: “[a] broad canvas of space and place constituted within the histories of social and cultural life”, emanating from the historically-contingent conditions that produce and bring together spaces, places and their representations.

The social histories and memories that constitute community and community membership are often tied to specific physical locations, i.e. places, on the landscape (Barber

2013:173; Barber and Joyce 2007). Place, inextricably linked but often confused with both landscape and space, has also experienced a recent revival in academic interest. Casey (1997: ix) argues that as with space and time, humans are “immersed in” place, proclaiming that “to be at all – to exist in any way – is to be somewhere, and to be is to be in some kind of place”. If space

63 is everywhere, if landscape is broad, then place is marked, it is located. It is our house, out of all homes, a sacred mountain where pilgrimages are made, out of all mountains. Places are both mundane and extraordinary, and no matter the scale of their importance, they loom large in the many dimensions of human experience. They are continually being made and remade (Joyce and Hendon 2000) because different social and cultural factors combine together to create senses of place (Kolb and Snead 1997:611). People engage with the specific places marked in their landscape in ways that are culturally and socially conditioned, reflecting emic worldviews. An interesting example of the dialectical relationship and intimacy between landscape, history, and community comes from the Western Apache. In their oral traditions, where events occurred is emphasized over when they happened – the siting of history in place roots the past in the landscape (Basso 1996:21) and facilitates the recalling of historical events in the present.

Following Allan Pred (1984, 1990), Tringham (1991, 1994) argues that the places observable to archaeologists are the result of multiple and overlapping processes in which the life-histories of various material things, including buildings, and the social histories and memories of people intersect in the ongoing work of social reproduction.

One way in which socially meaningful places are constructed, marked, and used is through architecture (Joyce 2009; R Joyce 2004). How societies arrange architecture across the landscape, intentional or otherwise, at some level reflects their social, political, and economic structures (Mehrer 2000:45). Monuments draw people into a community of experience whether they are physically present or not (e.g. Basso 1996; Hegmon 1998; Wright 2012). The spatial organization and structure of monuments, how they are arranged into groups, their intervisibility at various spatial scales, the labor organization requirements, and geographic positioning make them critical features of the landscape. As they themselves are stages for human activities and

64 social experiences, so too are monuments staged against their natural geological and topographic backdrops (Wright 2012:146). Structures can both mark the continuity of a place, by installing a statue or plaque at an important battle site for example, or even attempt to erase it, as when the

Spanish built their churches directly on top of the destroyed temples of conquered indigenous communities (Moore 1986; Tringham 1991, 1994). The built environment both constrains and allows for particular types of interactions and practices (Canuto 2002:65), and architecture plays a particularly unique role by communicating social messages that direct and regulate human actions in a tangible medium (Kolb and Snead 1997:613; Moore 1996: 15). Buildings and other architectural forms become critical in structural transformations because historical events, like the Storming of the Bastille that marked a turning point in the French Revolution, are just as much spatial as they are temporal (Beck et al. 2007). At the same time, structures in the built environment can be sites of resistance or contestation in other ways as well. People may use and move through and about these spaces in ways unintended by their designers, or that change the social and structural meanings of architecture even in the course of carrying out quotidian activities (de Certeau 1980).

Although we often associate monumental architecture with state-level societies, archaeological evidence from different locations all over the world demonstrate that monumental constructions are some of the first large-scale communal projects that humans undertake (Burley

1996; Dietrich et al. 2012; Solis et al. 2001). This is because monumental constructions need massive amounts of labor, and often represent a huge investment involving many members of a community (Joyce et al 2003), and may even involve the labor of others. In other words, the construction of monumental structures brings individuals together for participation in group work projects that benefit the community and/or communal social networks in tangible and

65 intangible ways. The mobilization of voluntary, but still obligatory, labor materializes both corporate actions and political authority (Barber and Joyce 2007:235). Monumental spaces also provide stable points of reference for different cultural practices because they are often the preferred loci for public performances and rituals.

In other words, through the construction of monumental structures, the social collectivities involved were able to achieve material permanence through architecture. The identities of social collectivities were then continuously recreated through the subsequent use of these monumental spaces as well as their continual or even occasional renovation (Barber and

Joyce 2007:233; Joyce et al. 2013:154). The performance of communal labor and shared practices in monumental spaces engenders a sense of common purpose, pride, group cohesion and self-perceived community among individuals. Thus, monumental architecture itself is a practice of place-making (Joyce et al. 2013). Chiefly position in Tonga, for example, was manifested upon landscapes through several types of public constructions. Most important of these were the sia heu lupe, artificially constructed mounds that were essential to the chiefly sport of pigeon snaring (Burley 1996:422). Although the sport was officially restricted to the elite, the construction of the mounds would still have required the mobilization of large corvee labor forces drawn from the entire community.

Clearly, the built environment in which monumental structures are located is a culturally constructed landscape and includes elements that are innovative and conservative, utilitarian and non-adaptive (Moore 1996: 10). Though monumental buildings are created by real human actors, perhaps with specific intentions in mind, they are by no means passive. Instead, they enter into recursive relationships with the community through their construction, use, and abandonment by people. Thus, a single building may embody or reflect a wide range of cultural

66 decisions, experiences, and different levels of meaning, as well as play an important role in shaping the social memories of communities and individuals (Connerton 1989; Moore 1996;

Rowlands 1993). In this way, monuments act as mechanisms of group cohesion and solidarity

(Wright 2012). Similarly, because they are sites of meaning and social memories, monumental spaces often become contested spaces (Low and Lawrence-Zuniga 2003), especially when there is the potential for them to transform from community-orientated integrative structures to tools for defining, promoting, and demonstrating hierarchical differences (Wright 2012:156), and vice versa.

The visible exteriors of domestic architecture, the exterior and interior appearances and design of communal architecture, and the organization of public space are some of the more highly visible items of material culture, and they are unique because they are also fixed on the landscape and are relatively public (Stone 2003:44). Several studies have demonstrated that formal variation in architecture can be a reliable index of cultural affiliation, use-function, and social status (Abrams 1989; Aldenderfer and Stanish 1993; Canuto 2002; de Montmollin 1989;

Healan 1993; McGuire and Schiffer 1983; Rathje 1983; Villamil 2005). Rapaport’s (1969, 1982,

1990) studies have been particularly influential in defining vernacular architecture and connecting it to different settings and activity systems. The most frequent or pervasive types of architectural markers are organized at the basic level of the family or household. The family project is small in scale and lacks the obvious displays of suprafamily status differentiation5, drawing on a labor pool recruited from within ascribed kinship groups for help with construction projects. As a result, the spatial organization of residences is a tangible manifestation of group and individual decisions, making the physical household an outcome of community decisions. In

5 Excluding of course the households of elite families that often involved labor from multiple people outside of the family group. 67 some ways, the construction and use of space are subject to shared and sometimes unconscious norms, akin to those that structure language (Sutro and Downing 1988:33-36).

Larger-scale projects beyond the household are organized by suprafamily relations and social hierarchy, using the labor of the general populace often vis a vis networks of social obligations. These types of projects are highly variable. At lower end of the continuum, they include festive projects organized by communities or low-level elites where labor is exchanged for non-labor commodities. At the higher end of continuum are the corvee projects that require large-scale coordination and (involuntary or obligatory) participation. The functional range and size of monuments found at the corvée level include large-scale agricultural field systems, fortifications, elite residences such as palaces, and ceremonial architecture including temples, platforms, burial structures (Kolb and Snead 1997:613), and even ballcourts.

Construction techniques and materials are themselves a line of evidence into past social structures and organization. A lack of standardization in construction techniques within one structure suggests that multiple work parties were involved, indicating the mobilization of communal labor at a large scale (Levine et al. 2004). There are multiple cases in the Andes of different social groups marking the adobe bricks they contributed, along with their labor, to monumental construction projects as part of their mit’a or tax with special identifying marks

(Dell 2010; Earle 1994). The use of variable construction techniques within a settlement can point to different social identities such as ethnicity and/or social group, while construction materials can inform us about social and economic status – lower-status residences often use simpler materials and construction techniques, such as wattle and daub, while higher-status residences utilized more labor-intensive techniques and materials such as adobe, stone, stucco and plaster.

2.2.2.1 Interpreting and Interpolating Space and Landscape

Interpreting past perceptions and experiences of human-scaled landscapes is an important area of research in archaeology (Rennell 2012). Landscape archaeology is a suite of theories and methods closely aligned with postmodern critiques in anthropology and social theory, in contrast with historical ecology, which concentrates more on the functional-economic relationships between the physical environment and human societies (Gillings 2012:602; Kantner 2008:56-

58). The interactions between cultural memories and places in creating and defining social identities, and how the meanings of places can be contested by different sociocultural groups with conflicting memories of the landscape, are of central concern to landscape archaeologists

(Kantner 2008:58; Colwell-Chanthaphonh 2003; Shackel 2001; Stewart et al 2004).

Over the past decades, two competing methods for investigating the lived experiences of past landscapes have developed, one rooted in the subject-centered field observations and analyses emphasized in the phenomenological perspective; and the other emerging out of GIS modelling (Rennell 2012:510). Subject-centered landscape archaeology and phenomenology has been extensively described (Bender 1993; Bender et al. 1997; Cummings and Whittle 2004;

Thomas 1993b; Tilley 1994, 2004) and counter-argued (Brück 2005; Fleming 1999. 2006). The lack of rigor in phenomenological observations, for instance, often makes them untestable and difficult to corroborate (Brück 2005; Fleming 2006:268; Llobera 2001:1005). However, the emphasis on a dwelling perspective, that is a subject-centered and embodied experience of landscape, is a critical corrective for the sometimes mechanistic and overly deterministic approaches of more traditional landscape approaches.

As I explore further in Chapter 5, digital spatial analysis techniques available in GIS computer-based programs are increasingly used to explore these more phenomenological aspects of landscape. For example, elevation models are commonly used to investigate visibility in past landscapes (Lake and Woodman 2003; Llobera 2003; Wheatley 1995), while combining elevation with terrain data helps researchers build models to understand landscape movements themselves (Bell et al 2002). For example, graphic maps known as least-cost paths are important visual techniques used to explain why sites are located where they are (Carballo and Pluckhahn

2007; Bell and Lock 2000), to identify long-distance trade routes (Sherman et al. 2010), understand regional social and economic networks (e.g. Howey 2007), and to identify possible routes for the colonization of new territories (e.g. Rademaker et al. 2012). Site visibility and location can also be investigated in GIS through the spatial and computational analysis of ecological factors (Bevan et al. 2003). The process of modelling in GIS helps to "explore, clarify, and define [spatial] patterns and relationships" through the extraction of data stored in a spatial database (Conolly and Lake 2006:46).

2.2.3. Sports, Games, and Society

Although not often thought of as so, sports and games are another social field through which communities and social identities are constructed and negotiated. Despite an early interest in sports as a heuristic tool for investigating the origins and spread of cultural traits (Culin

1894:355; Norbeck 1977:2; Norbeck and Farrer 1979), there has been a pervasive and dismissive perception of sports in anthropology, which has been labeled as peripheral to so-called more relevant concerns like economics, politics, and kinship systems. Sociology, on the other hand, has long legitimized sports as a mainstream topic of investigation (McGarry 2010:152).

Sociologists, along with physical educators and sports psychologists, have been making important contributions towards a better understanding of how social life interacts with the structure, values, and organizations of sport. The limitations of these studies, however, is that they primarily emphasize how these interactions are materialized and expressed in Western industrial societies (Gruneau 1975:122). Early theorists on the importance of sports and games include Thorstein Veblen (2009 [1899]), whose Theory of the Leisure Class was the first well known attempt to examine the relationships between sport and social structure, class, and status inequalities. In many ways, hierarchy is a cornerstone of the sporting ethos because of the emphasis on outcomes where winning is valued, a basic statement of distributive relationships

(Gruneau 1975:129).

There are several terms used in the literature on sports that are critical for understanding how researchers frame and interpret the relationship between sports and society: liminality, liminoid phenomena, leisure, work, and play. To start, the concepts of liminality and liminoid phenomena were first developed by Victor Turner, one of the few early anthropologists to seriously consider sports as a critical component of culture and society. Turner (1969, 1977,

1982, 1985) distinguishes ancient from modern sports, arguing that because modern industrial societies are said to be secular and lacking in ritual, modern sports are liminoid phenomenon and not genuine ritual, unlike their ancient predecessors. Liminoid phenomena do feature the playful elements of ritual, but instead of being part of the broader social cohesion in the form of highly formalized activities, they manifest as independent genres of industrial leisure, i.e. as distinct arts or modes of personal expression, including modern sports. According to Turner (1982), liminoid phenomena only emerge in industrial societies because: 1) there is a shift during the Industrial

Revolution from a communitarian paradigm to an individualist one; and 2) the new distinction

71 between leisure (free) time and work time, with liminoid phenomena pursued as leisurely activities according to an individual’s tastes (Rowe 2003:130-132). Modern sports are further distinguished because of their commercialization and separateness from modern religious institutions, thereby reducing them to popular forms of secular entertainment only. But this distinction rests on an ahistorical and erroneous belief in the “purity” of ancient sports, when in fact sports have always been heavily exploited and supported by the economic and political interests of the societies they are embedded in (Rowe 2003 133-134). Interestingly, a similar argument is made about the Mesoamerican ballgame, wherein scholars argue that the game became increasingly secular in the Postclassic period, a problem I explore further in Chapter 3.

Nevertheless, contrary to Turner’s formulation, contemporary sports do contribute the

“ludic” element that is supposedly missing in modern society; moreover, ritual liminality does lie at the core of sports even when they appear to be entirely “secular” (Rowe 2003:133-134).

Aside from the fact that they are structured as games and understood within the language of play, sports both ancient and modern are liminal phenomena because they unfold in open-ended contexts where outcomes are nearly impossible to predetermine. Moreover, sports have the dual capacity for both innovation – through the presentation of ourselves to ourselves in liminal states

– and conservativism in our social self-images. The determination of the final outcome of a sporting event brings a close to the “what-if” sensations that permeate the event during play, thereby ending the liminal performance and connecting us back to concreteness of a now slightly altered social reality (Rowe 2003:129-130). As a social metaphor and a social field, sports therefore both conserve traditional social and cultural values while also mediating structural and political change (Balbus 1975; Hargreaves 1987a, 1987b; Lipsky 1981).

It is not inaccurate to say that the athletes and game symbols of sports gain meaning and significance from the collective (Rowe 2003:136). In the case of modern sports, for example, without the imagination and support of fans who place all their hopes and aspirations on the fortunes of favored teams, they would not be as effective at reflecting and capturing the Gestalt, or collective spirit, of the group. Modern sports are thus just as effective at drawing in large collectives of people at different scales who come to identify with them as were ancient sports.

The Olympic Games are an excellent example of sports as a global ritual (Peacock 1985:81), where appeals are made both to the national identity represented by each country’s athletic team, and to a larger global identity as a human community gathered in one place to participate in sport activities together. Ceremonies occur throughout, marking the beginning and ending of the

Games with parades, dramatic performances, and an elaborate – and quite expensive – show meant to display the history and “character” of the host country. Moreover, the opening ceremonies culminate in the grand lighting of the Olympic Flame. The flame is lit in some grandiose and theatrical way with the Olympic Torch, an object that has made a global journey around the world, carried by different athletes in each country it travels through. Muhammad Ali famously had the honor of lighting the Flame at the 1996 Summer Olympics in Atlanta, Georgia.

The narrative surrounding Ali repeatedly referred to him as a national American hero, an incredibly stark contrast to the general opinion of Ali due to his protest of the Vietnam War and segregation in the United States during the 1960s. In many ways, nationalist narratives are constructed around sports as much as they present, through their liminality, the ability to effect change. As Rowe (2003:136) explains:

“…sports do not merely represent; they also transform. Sports construct a context in which we, as humans, in fact, transform ourselves; creating ourselves anew with each record; extending our potential as we establish new limits to our present being. When records are set, human horizons are literally redefined”.

Sports are also liminal because they are a sensorial experience for fans, competitors and coaches alike, centered around the drama of competition as well as the human body and the physical limits to which it can be pushed. In this way, the world is experienced in corporeal and material ways through sports (Dyck and Archetti 2003). Additionally, sports gain their liminal quality because they are seasonal and somewhat redundant (Rowe 2003:136), especially in our modern context. The coordination of the timing of major sports, such that teams and individuals across large geographic spaces can participate together, results in a cyclical rhythm in game scheduling. Some sports are even embedded in our experience of specific seasons, depending on where we live. In the United States, spring and summer are strongly associated with baseball, such that baseball players are called the “Boys of Summer”, while the arrival of fall signals the start of football season. Internationally, fútbol or soccer is played at ever increasing spatial scales, with games seasons progressing from the nation-state to the interregional and finally the global, culminating every four years in the World Cup. Thus, much like ritual events ordered by a calendar, we live our lives according to the seasonal rhythm of sports. Our daily lives are not simply ordered around commercial sports alone – many a child is enrolled in a team sport through school. Especially in American society, sports are also tightly linked to academic education (Sage 1979). School athletics involve a large commitment not only from students and teachers, but also the greater community. In many small towns and cities across the country, local college or high school sports teams are an important focus for smaller scale collective identities. The social influence of sport is further supported by its economic effects –

74 commercialized sports are highly lucrative businesses worth billions of dollars, and even sports pursued on an individual basis represent a significant amount of money. Sports equipment can be very expensive depending on the investment and time commitment; non-professional bicyclists who can afford to do so, for example, still spend thousands of dollars on what is still often considered to be a “hobby”.

While some have characterized sports as being non-serious because of its ludic elements

(cf. Huizinga, 1955), other scholars have argued that sports serve the very serious social function of maintaining inequality (e.g. Burley 1996; Donnelly 1996; Gruneau 1975). Institutionalized divisions and social inequalities not only impact sports, but are also reflected in how they are structured, i.e. rule systems, and the ways in which the scope and definition of sports can change

(Gruneau 1975:121). Weber (Bendix 1977) observed that sports and games in European and

Japanese feudal societies were important in helping to maintain dominant and legitimizing traditions, echoing similar concerns with class inequalities in the late 19th and early 20th century

(Gruneau 1975:125). Lévi-Strauss (1966:31-32) refers to the “disjunctive effect” of competitive games because of their ability to create inequality, i.e. winners and losers, where none had existed before. The connection between social inequality and sport is not limited to Western or modern societies. By the first millennium CE in Tonga, pigeon hunting had ceased to be a subsistence activity and had transformed into a highly-structured sport reserved exclusively for the chiefs (Burley 1996). The artificially constructed mounds used for the sport became physical manifestations of chiefly authority in Tongan society, marking the social landscape (Burley

1996: 422).

But inequality is not the only hallmark of sports. They are also ideal forums for creating and maintaining communitas (sensu Turner 1969) as well as reinforcing local identities among

75 group members through aggressive competition and team loyalty (Baron 2006:53; Williamson and Cooper 2017: 60). In this way, sports resemble religious rituals in form and effect because of their ability to integrate individuals. (Bell 1997). For many modern societies, sports also importantly function as contemporary “tribal” rites of passage (Beisser 1967), reflected not only in the different age and professional classes in organized sports but also in the way sports are often linked to our concept of masculinity and modern manhood. Moreover, they are also easily politicized and ritualized because of the heightened emotions, both positive and negative (joy, excitement disappointment, anger, antagonism, etc.), that are produced from playing, watching, and the very liminal qualities of sports themselves (Baron 2006; Stoll and Anderson 2017).

Sport myths often reflect cultural traditions and reinforce social attitudes, and sports themselves express the underlying social values or even ideological differences between interest groups (Rowe 2003a, 2003b). In human societies, sports and games are a function of established and emergent power relations; they are therefore closely tied to the material conditions of life

(Gruneau 1975; 129; McClancy 1996). This is evidenced by the very ubiquity of sports-based metaphors in many languages (Balbus 1975; Rowe 2003b; Sage 1979). English is replete with idiomatic expressions based around sports, many of them inked to success and/or sex. In the capitalist, communist, and even fascist societies of the 20th century, sports were instrumental in building and reflecting nationalistic ideologies (Rowe 2003a:139-140). Returning to our earlier example of the Olympics, the world games were often the site of political contests and demonstrations between global adversaries and their allies. The 1936 Olympic Games set in

Berlin were a chance for the Hitler regime to display the power of Nazi ideology and the Aryan

Superman. Instead, African-American athletes from the United States defeated German athletes in key events (Kessler 2011). In the latter half of the 20th century, the United States led a boycott

76 of the 1980 Summer Olympics, which took place that year in the Soviet Union, in protest of the latter’s 1979 invasion of Afghanistan. Eventually, 65 other nations would join the boycott.

Earlier that same year, the US hockey team had defeated the Soviet team at the Winter Olympics in a game that would be known as the “Miracle on Ice”, an event viewed by the American media and public as a symbolic victory in the ongoing Cold War. In response to the US-led 1980 boycott, the Soviet Union itself spearheaded a communist-bloc boycott of the 1984 Summer

Olympics in Los Angeles.

The increasing globalization and corporatization of sports allows us to further analyze how critical social fields are shaped by transnational contexts and the intersections between the production and mediation of national, transnational, and even localized identities (McGarry

2010:153). Modern day examples such as Colin Kaepernick’s 2016 anthem protests, his subsequent blackballing by National Football League (NFL) teams, and the dramatic follow-up in 2017 with more players kneeling in response to attacks on their First Amendment rights by representatives of the Federal Government, assertively demonstrate that sports and sporting events are still “key sites for struggles over power and agency” in society (Cheska 1984).

Indeed, the close relationship between politics and sports is definitively demonstrated by this ongoing struggle within the NFL, touching on issues of racial injustice, police brutality, free speech, government intervention, private corporate issues, and even patriotism.

Arguably, sports act as organizing features of societies because of their usefulness in cross-cultural interactions, identity production, and the reorganization of work and leisure time during periods of sociopolitical change (McGarry 2010; Struna 1996). The near universality of sports makes them ideal as an index of culture change both because of their ubiquitous presence in everyday life and their central role in negotiating social identities and interactions within

77 different cross-cultural contexts (Cheska 1979, 1984; McGarry 2010). This is not a new phenomenon. An interesting example of sports and social identity comes to us from Medieval

Italy, where the elites of Siena promoted a traditional competition between the city’s different contrade or wards in order to encourage hyperlocal identities and discourage class-based identities and solidarity among the lower classes (Baron 2006:52).

Sports also often play a role in the breaking down or the strengthening of boundaries between different social groups. In situations where there is persistent or regular contact between different ethnic groups, some particularly marked sport behaviors will change over time, but still work to reinforce ethnic diversity (Barth 1969; Cheska 1984:253). Organized celebrations involving Americans, Mexicans-Americans, and Diné (Navajo) tribal members in the American Southwest, for example, included separate sporting events for each group, enforcing a social separateness based on ethnicity (Cheska 1979). The relationship between sport and society is therefore both recursive and incredibly complex (Blanchard 1981; Blanchard and Cheska 1985; MacAloon 1984).

2.2.3.1 Work and Play: Sports and Games in Anthropology

Within anthropology, ethnography and ethnohistory have been the leading fields of study on sports. Although important, initial studies by scholars such as Edward Tylor (1879), James

Mooney (1890), and A.C. Haddon (1898) often documented their subjects in a haphazard and cursory way (Blanchard and Cheska 1985; McGarry 2010), and sports were primarily studied as a means to understand other cultural phenomena, a reflection of the general attitude of anthropologists in their day. One important exception during this time period was Stewart Culin

(Blanchard 1996; Cheska 1975; Voorhies 2017). Beginning with his first publication in 1889,

Chinese Games with Dice, Culin embarked on a decades-long mission to document and describe different kinds of games. His most important works, the article “American Indian Games”

(Culin 1898) and the seminal book Games of the North American Indians (Culin 1907), are still referenced today. Though some of his positions have since been refuted, his treatment of sports, games, and the seriousness of play would influence later anthropologists (Blanchard 1996:14).

In the 20th century, one of the first and more systematic attempts in anthropology to look at games and sports was the article “Games in Culture” (1959) by John Roberts, Malcolm Arth, and Robert Bush. The authors set out to find the constant features that define games, and in turn inspired a debate among anthropologists about the “general role of play and the special role of sport in human society,” (Blanchard 1996:17). More detailed ethnographies followed in the

1960s and 70s by Cheska (1979), Geertz (1973), and Norbeck (1971; 1974). By the 1980s the volume of research had increased immensely (e.g. Allison 1980, Azoy 1982 [updated 2011];

Blanchard 1981; MacAloon 1984; Oxendine 1988; Poliakoff 1987), contributing to the ever- growing field of sport studies in anthropology. Much of the ethnographic research on sport has historically positioned itself squarely within explanatory systems approaches, such as structural- functionalism, symbolic approaches that fixate on meanings, or explanatory systems like globalization that seek to determine the final solution for cultural phenomena (McGarry

2010:158). More recent scholarship has shifted to include poststructuralist and agency-based perspectives.

There is general agreement among anthropologists that the activities of play, games, and sports are related, but there is still much debate over their precise meanings (Blanchard 1996:29)

Play is often posed as the polar opposite of sport, with games placed in the intermediate position along a continuum from expressiveness (play) to instrumentation (sport) (Cheska 1979:228,

79 citing Edwards 1973:43-61). In a sense, play is associated with simplicity and unorganized physical activity, while sport is viewed as much more complex, structured, and rule-defined, especially with regards to “participant behavior, object-subject-temporal arrangements, and goal consummation” (Cheska 1979: 228). Another debate in how we define sports centers on whether they are appropriately described as work or as leisure, i.e. play. Early anthropologists and social theorists viewed sport as leisure, an activity that only the wealthy classes could participate in because of the costs in terms of money and time (McGarry 2010). This attitude has changed somewhat over the 20th century, with some anthropologists viewing sports as a type of play

(Huizinga 1955), while others declaring that sports have absolutely nothing in common with play

(Edwards 1973:55-56).

Slowly though, there has been a shift towards seeing sports as work, especially as various sports have become professionalized and even corporatized, or as a varying combination of both.

Rowe (2003:142) suggests that the transformative capacity of sport illustrates a blend of work and play. Indeed, in contrast to the 19th century view of sport as a leisure activity pursued out of boredom or status, being a professional athlete is now considered to be a legitimate (and often highly lucrative) job, with the achievement of athletic success seen as the result of years of concentrated effort and training. The best performances are demanded not only by contractual obligation and fan expectation, but also by the creativity, joy, and even spontaneity displayed by athletes in contributing to multimillion dollar entertainment. At the same time, causal or non- professional athletes still demonstrate many elements of work, and like the professionals they invest a lot of time and energy in their chosen sport. While the choice may be voluntary, it is also utilitarian and enjoyable in purpose (Rowe 2003:143). This expansive view of sport is best

80 expressed by Blanchard (1996:48), where different sports are defined as varying combinations of

“work, leisure, play, and not-play”.

Games, a term sometimes distinguished from and at other times used interchangeably with sports, are predominantly described as pleasurable, recreational activities that involve some level of organized play, competition, two or more sides, agreed upon rules and specific criteria for determining the winner (Blanchard 1996:17; Cheska 1979 227; Huizinga 1955:6-13;

Voorhies 2017:1 citing Roberts et al. 1959:597). They are then further divided according to whether outcomes are determined by physical skill, strategy, or chance (Chick 1998:186).

Games of physical skill can be found nearly everywhere, whereas games of strategy are believed to model social interaction, and many assume that they are only present in complex, but not egalitarian, societies (Voorhies 2017:7-8, 10). Finally, games of chance are supposed to have emerged out of the divinatory practices of religious ceremonies (Csikszentmihalyi and Bennett

1971:47; Culin 1894:355). Perusal of the ethnographic evidence on games, however, shows that these categories are not mutually exclusive. Gambling games, for example, involve a combination of chance and strategy (Voorhies 2017:2, 10). Even in games of physical skill, players and spectators will risk their wealth, political positions, social statuses, and personal well-being via placing bets on game outcomes (Stauffer and Reilly III 2017); that is, the act of gambling is not simply restricted to one type of game. Indeed, as I will demonstrate in Chapter

3, the neat demarcations between games, sports, and gambling drawn by Western scholars are often blurred in the Americas.

2.2.4 Frontier Zones and Borderlands: From Peripheries to Dynamic Spaces

Archaeologists use different lines of material evidence in order to identify, trace, and/or reconstruct communities, social identities and the practices that formed them. However, the ability to link material evidence to past identities become somewhat more complicated when we consider regions or places where the social fields linked to community and social identity are constantly subject to negotiation, tension, contestation, and renewal. Frontier zones and other boundary regions, created by both geography and politics, are places where people, practices, ideas, and objects from different homelands are often exchanged in novel and distinct ways. In this way, frontiers can be productive sites of ethnogenesis, or more broadly the engendering of new identity groups, through the integration or disintegration of social boundaries (Voss 2008).

At the same time, social identities – or certain markers thereof – can also become more entrenched through unequal power relations between different groups. Frontiers can therefore be complex areas to investigate. In frontier regions such as Nejapa, for example, the evidence for multiple ethnolinguistic groups both occupying the area and traveling through challenges traditional interpretations of community formation and social identity in the Prehispanic era

(Chapter 4).

Frontiers have been a topic of interest in anthropology, archaeology and history since at least the last century (Parker and Rodseth 2005). Most studies after the 1950s have primarily focused on cross-cultural comparisons of frontiers around the world (Wyman and Kroeber 1957,

Hartz 1969mad, Miller and Steffan 1977, Lamar and Thompsen 1981; Parker and Rodseth 2005).

The primary goal was to understand processes of cultural diffusion and ethnicity (Rodseth and

Parker 2005:6-7). Frontier studies are most associated with the work of Frederick Jackson

Turner (1893, 1896, 1932), who in essence distilled the popular representation of the American

West as an imaginative frontier. Turner’s Frontier Thesis (1893) advanced the idea that the

American frontier formed American democracy through the impact it had on pioneers, resulting in egalitarianism, violence, and lack of interest in high culture. The openness of the frontier and the constant movement and penetration of the “frontier” into unknown territories released

Americans from their European mindsets, helping to establish liberty in the process. The imagined values of the Frontier, including rugged individualism, become closely identified with our understanding of the American character and our own history.

Recent attempts have been made to extricate frontiers from the Turnerian tradition, especially by scholars who wanted to focus on the experience of minority and subaltern groups in the American frontier while also holding a critical eye to myths about the American West

(Rodseth and Parker 2005). Today, we understand borderlands, frontiers, and boundary zones to be space or places where cultural and geographic distinctions are fluid (Cusick 1998; Adelman and Aron 1999), and where interactions between peoples from many different ethnolinguistic, cultural and social groups occur, are encouraged, or even precluded to varying degrees in spatial domains (Canter 1977; Lightfoot and Martinez 1995). Alternatively, frontiers and border regions may also be considered as abstractions of cultural intentions and differences through the linkage of the social and the spatial (Trouillot 1995). The term boundary is a broader term that encompasses the more specific “border” and “frontier” (cf. Anderson 1996:9). Whereas a border is often defined as a legally recognized line or “crystallized” boundary, a frontier is described as a region rather than a line (Rodseth and Parker 2005:10). Frontiers are therefore fluid zones of complex social relations that include competition and negotiation among different (sometimes radically so) communities with distinct social identities and practices, and polities or societies

83 can have different kinds of frontiers operating simultaneously or at different time periods (Naum

2010; Rice and Rice 2005).

Frontiers are closely associated with three recurring processes: 1) the emergence of the frontier itself in relation to a center or core area; 2) the mutually structuring interactions between a frontier and core area; and 3) the development of social exchange, merger, or conflict between previously separate populations brought together on the frontier (Rodseth and Parker 2005:4). In modern global studies, frontiers as well as borders, diasporas, and contact zones are considered to be transnational fields where people, commodities, and cultural ideas mingle and recombine

(e.g. Appadurai 1996; Hannerz 1996; Clifford 1997; Donnan and Wilson 1999; Rodseth and

Partker 2005). Despite the rapidity with which population movements and cultural exchanges occur, the transnational phenomena occurring today under rapid globalization are argued to be extensions of the same frontier processes that have been developing over thousands of years

(Abu-Lughod 1989; Frank and Gills 1993).

In contrast to core-periphery models or theories where frontiers only emerge in response to actions taking place in the political center (Redmond 1983), Lightfoot and Martinez

(1995:471-472) argue for reconceptualizing frontiers as “socially-charged places” of innovative creation and transformation of cultural constructs in contact situations through the interactions between cross-cutting and sometimes overlapping social units. In pluralistic contacts, people will take advantage of new opportunities and social relationships that they see as beneficial to their own or related interests (Lightfoot et al. 1998). For Rodseth and Parker (2005:4), the frontier is “a shifting zone of innovation and recombination through which cultural materials from many sources have been unpredictably channeled and transformed”. Because frontiers are zones of variously overlapping but not necessarily congruent political, economic, and

84 sociocultural boundaries (Elton 1996:3-9), they give rise to hybrid forms of culture and ambiguous identities composed of selected elements from each previously distinct cultural repertoire (White 1991; van Dommelen 2002). Many societies have been formed and transformed in relation to their frontiers throughout history (Rodseth and Parker 2005). This is because novel social and ethnic identities, and the attendant processes of social production and reproduction that accompany them, are often asserted as part of the strategic responses to the changing circumstances of evolving frontiers (Rice and Rice 2005). Thus, frontiers are therefore important sites of ethnogenesis through the merger or fragmentation of different social groups

(Parker and Rodseth 2005).

How do groups with objectively different cultures interact with each other and organize themselves given equal or unequal power relationships? As I discussed in the previous sections, ethnographic groups that are in contact for a long time often share cultural practices that are viewed as socially useful across different boundaries, while others become distinct markers for specific groups and group membership (Buchignani 1987). Importantly, the adoption of external cultural practices is not necessarily a signal of membership in that group (Casella and Fowler

2005). In fact, some argue that the strength of the boundary maintenance between groups and the degree to which actors adopt the material objects, cultural traditions and/or beliefs of the other is a measure of the power disparity and the direction in which these relationships flow

(McGuire 1982; Rothschild 2006). Understanding these disparities and the objects, traditions, and beliefs involved can help in understanding how identity is constructed and negotiated in frontier situations.

Research on communities and social identities in frontier contexts should therefore be oriented to how people establish and maintain interethnic relationships in politically

“ambiguous” or emergent regions, and how multiple kinds of interactions can and did take place within and between groups (Lightfoot and Martinez 1995:474). Lightfoot and colleagues (1998) advocate for a refined archaeology of pluralism, an alternative approach that attempts to take into consideration the material remains of daily practices and the use of space in “pluralistic social settings”. Critically, their approach emphasizes investigating a suite of habitual practices in a multiscalar study that is both broadly diachronic and comparative in scope. Although it may seem beyond the capacity of most research projects, through focused research questions and targeted sampling of sites, this kind of study can be done. The PANT study is an excellent example of this approach. Over the course of just three seven-month field seasons, PANT has successfully been able to collect enough data to establish the complexity of identity and social dynamics in the Nejapa subregion.

2.2.5 Scaling the Community: Regional Studies in Archaeology

This dissertation emphasizes analysis at the geographical scale of the “region”, placing it firmly within the tradition of regional archeology. While archaeologists have always considered regions as units of analysis, regional archaeology as a distinct subdiscipline was greatly influenced by the New Archaeology of the mid-1960s (Anschuetz et al 2001:170-171; Kantner

2008:39). Flannery’s (1968) human ecology and ecosystem approach was emblematic of this new direction in regional studies through its explicit focus on the recursive relationships between human behavior and the environment. His research in the Central Valleys of Oaxaca helped to establish the geological and archaeological history of the region. Further development in regional studies occurred with the 1976 publication of Carol A. Smith’s Regional Analysis, which introduced well-established geographical methods to archaeology.

More excellent studies soon followed, building on the strengths of classic analytical techniques, while also developing unique approaches for exploring archaeological data at the regional scales (e.g. de Montmollin 1989; Gaffney et al. 1996; Lake et al. 1998; Ruggles and

Medyckyj-Scott 1996; Zubrow 1994). The postprocessualist critiques of the 1980s opened the discipline to using qualitative techniques in regional and landscape studies (Kantner 2008:40).

The postprocessualists themselves were influenced by humanistic geography, a field that has existed as far back as the work of Carl Sauer (1925) in the early 20th century. Finally, the increasing power of computing technology in the late 1990s further impacted regional studies in archaeology because of the ease with which complicated spatial analyses could be performed in order to analyze, visualize, and model multidimensional landscapes using qualitative and quantitative data (Aldenderfer and Maschner 1996; Allen et al. 1990; Conolly and Lake 2006;

Kantner 2008; Kvamme 1989; Maschner 1996).

Obviously, any consideration of regional archaeology requires a good working definition of the “region” as a scale or unit of analysis. This is perhaps the biggest issue with regional studies. The word “region” itself is often conceptualized at many different scales that vary widely in size and scope (e.g. Lekson 1999; Madry and Rakos 1996). Indeed, it appears that how we define a “region” is largely dependent on the particular theoretical perspectives and questions that frame the research (Kantner 2008:41-42). Several criteria used for defining regions include topographic boundaries as well as geographic patterning in material culture (e.g.

Duff 2000; Hegmon et al. 2000), assuming that the contiguous distribution of artifacts and features is representative of spatially-bounded social groups (Kantner 2008:42). Typically, archaeological regions are conceptual spaces where meaningful relationships can be defined between past human behavior, the material signatures people left behind, and/or the varied and

87 dynamic physical and social contexts in which human activities occurred (Kantner 2008:41; e.g.

Parsons 1976:137-144, 2004:9-12; Stafford and Hajic 1992:138-143, Wandsnider 1998). They obtain the appearance of territorial coherence through the “instantiation of clear social and spatial boundaries” such as urban vs. rural, public vs. private, ceremonial vs. domestic (eg

Alonso 1994; Kosiba and Bauer 2013:67).

However, how archaeologists define a region is not necessarily reflective of how people in the past would have – thus any models of human settlement used to identify geopolitical or geocultural regions must not automatically assume that the criteria defined by archaeologists corresponds to criteria in the past (Kantner 2008:42). In reality, the boundaries and relationships between people, and between groups of people and their environments were constantly changing

(Ashmore 2002; Bender 2002; Dewar 1991; Dewar 1992; Kantner 2008; Lightfoot and Martinez

1995; Wandsnider 1992). Ideally, archaeologists should be more interested in the distribution of actual material cultures across a region rather than in arbitrary units that disguise variability or impose interpretation (Dunnell 1992:32-33; Dunnell and Dancey 1983:271-274), although this is always easiest to do once a sufficient amount of regional data has been collected. Techniques used in regional studies have centered on quantitative and graphical methods derived from economic and cultural geography as well as spatial modeling in ecology (Ebdon 1985; Hodder and Orton 1976; Smith 1976) and geography (Kantner 2008:42), most especially with respect to spatial analytical techniques in GIS applications (e.g. Conolly and Lake 2006; Goodchild and

Janelle 2004; Kvamme 1990). I explain my own regional definitions and criteria for

Mesoamerica, Oaxaca, and Nejapa in Chapter 4.

2.2.6 Discussion

Complex social relations, which are historically and spatially contingent, emerge from negotiations that are both cooperative and competitive (Barber 2005:30). As individuals interacted in place, community and social identity were constituted through the shared, non- discursive experiences of daily life, the supra-domestic collective actions that took place in locations both domestic and public, and the extraordinary events that ramped up these experiences (Barber 2005:58-59; Canuto 2002:11). The collective actions that brought individual members and social groups together involved consumption, the exchange of material goods, feasting, and hospitality in these important locations (Barber 2005: 59; Bray 2003b:3;

Canuto 2002: 11; Dietler 1996:91; Dodd 2005). However, collective actions or practices of communities are not passed down blindly or passively from the past. Instead they are active negotiations by people in the present, made and remade each time an individual enacts them, uses them for their own benefit, or questions their efficacy and appropriateness (Douglas 2013;

Keen and Zeitlyn 2007; Silliman 2005:283, citing Pauketat 2001). Even when practices are seen as cultural traditions, they are often both revisited and recast in novel situations. Indeed, because material culture plays an active role in constituting what are considered to be “traditions”, they become important arenas or social fields through which communities express their self-defined senses of cultural distinction (Jones 1997:69; Poyer 1988:480; Silliman 2005:284, 289), while also serving as critical sources of legitimacy in political sphere (Ortiz 2005:145).

In the previous sections, I highlighted how certain linked practices or social fields interact with the ballgame to make it efficacious and affective as an integrative, albeit competitive, social field in and of itself. Rituals, for example, are important loci of social negotiation because members of different social groups, however they are joined or divided, are separated in two

89 important ways: as participants versus non-participants; and between those who do and those who do not possess the esoteric knowledge needed to make ritual meaningful and successful

(Stone 2003:43-44). The physical intimacy of mass consumption in the form of feasting and its inclusion in systems of socially obligatory reciprocity and gift-giving enhance the celebratory, communal nature already present in sports. Furthermore, the way we conceptualize and organize space both reflects and influences our social relations, practices, and interactions, such that architecture becomes an important expression of cultural identity, especially in multi-ethnic contexts (Dodd 2005; Voss 2005:467). Finally, it is critical that when looking at why the

Mesoamerican ballgame was such a prominent practice, we consider the very appeal of sports themselves. Like rituals, feasting, architecture, and sociocultural landscapes, sports provide a social context and the cultural content for the negotiation and/or maintenance of community and social identity across different scales (Cheska 1984).

The spatial layout and organization of communities and households play a key role in sociocultural growth and change because they reflect and shape social structures, and they can be used to evaluate how social systems are organized at different times (Whalen 1988:249, 254).

Building a structure acts as a practice of definition, especially with respect to monumental public architecture that is meant to be viewed by the members of a community and even external groups, through the mobilization of work parties, the communal participation in physical labor, and the materialization of group “tastes” and style preferences as seen in structure design and layout (Barber 2005; DeMarrias et al. 1996). Thus, the distinctions observed in the ceremonial architecture at the sites of San Ildefonso and Catalina, on the North Coast of Peru, suggests that non-elite communities differentially adapted elite ideology in order to formulate their own power strategies (Swenson 2008). Such strategies were themselves linked to culturally specific

90 conceptualizations of proper or accepted ceremonialism (Swenson 2008: 238). Cross-cultural spatial analysis of plazas in Peru demonstrates that large open plazas were key elements of an

Inkan imperial strategy that emphasized large public feasting ceremonies (Moore 2004: 792). In contrast, Chimu ceremonies were used to maintain and rearticulate the social distances between rulers and ruled. Such ceremonies were conducted in the enclosed plazas directly incorporated into the ruler’s residence; audience membership would have been exclusive and linked to social power (Moore 2004: 794). The organization and design of the plaza spaces amongst both groups reflected different political strategies for legitimizing authority and power.

Space is crucial to power, and the ability to control space is a source of authority and legitimation. Political authority in the Urartian polity, for example, was actively and visually produced through specific spatial practices assembled together as geopolitical landscapes, emphasizing particular forms of architecture and spatial layout that were represented in artistic pictorial media via the arrangement of figures in bronze relief pieces (Smith 2001). These practices mediated between the ruling regime and its subjects, and they were themselves historically contingent. The architectural elements in these pieces evoked in the mind of knowledgeable viewers dramatic state spectacles through explicit references to the specific physical spaces where they took place, thus linking performances to legitimizing authority through the display of the aesthetic dimensions of political life (Smith 2001:123-126). Social complexity, hierarchies, and monumental architecture, both public and access-restricted, are instantiated by social groups and emerge out of the instantiation of social relations between individuals and among communities (Joyce 2004). Standing architecture, gaming, and feasting are all fields of social negotiation (Pauketat 2000). In this way, entire ritual grounds of

Mississippian centers could come to symbolize the construction and reconstructions of

91 community as meaning and identity, with other material symbols and meaningful practices working together as parts (though not always coherent or cohesive) of this whole (Bell 1992;

Pauketat 1997; Pauketat 2001).

It is crucial that archaeologists not automatically interpret shared material culture as concrete evidence of shared identity. Most importantly, social identities themselves should never be assumed to be singular and unproblematic, but rather plural and contested (Joyce and Hendon

2000:147). While archaeological communities may be defined in part by a shared identity among members, these identities may just as likely crosscut spatial limits as be bounded by them

(Goldstein 2000:182). Indeed, shared identities can be found even among extended and non- contiguous populations; in other words, spatial proximity does not always equal social proximity

(Knappett 2011). Therefore, in any complex society, we should expect to find communities coexisting as both nested and/or crosscutting entities (Yaeger 2000:126) or modules of social identity (Barber 2005), especially in frontier or boundary zones where social boundaries are often more fluid and the potential for novel group formation, hybrid forms of culture, and/or ambiguous identities is greater (Emberling 1997:317; van Dommelen 2002; Lightfoot and

Martinez 1995:471-472; Parker and Rodseth 2005; Voss 2008).

We can find many cross-cultural examples of the complex relationship between material culture, practice, community, and social identity. Early Postclassic potters in the Central Petén region of Guatemala, for example, did not assert their distinctive identities via ceramics, as they decorated their vessels in similar ways no matter where they lived, or the materials they used.

Instead, social identity was expressed through other mediums and social fields such as incense burners, public architecture, and mortuary ritual (Rice and Rice 2005). Among the Luo pottery- producing peoples of Kenya, the boundaries of style zones fall in areas that are of no cultural or

92 social significance, and even though territorial boundaries and group identities are important to people, they are not reflected in the distributions of ceramic styles (Dietler and Herbich

1998:254). For the Luo, house form and construction, reinforced by rituals and the threat of supernatural sanctions, are much more conditioned by habitus as well as technological and environmental limitations than are ceramic styles.

The production and consumption of ritual items can also constitute corporate or communal identities that crosscut status boundaries by transforming economic action into

“statements of corporate affiliation” (Barber and Joyce 2007). These statements extend also to the labor applied to monumental construction projects, since such structures materialize political authority and power relations through the scaling up of the local community to regional networks of settlement and social group organization. Furthermore, the senses of community identity produced through engagements with material culture and monumental architecture provided the frameworks by which people negotiated the larger sociopolitical conditions in which they lived, especially as these conditions changed and became more complex and even hierarchical (Barber and Joyce 2007). Understanding creation and negotiation of community and social identities in this way is useful for interpreting social networks among geographically widespread communities in Mesoamerica, where we find both localized expressions of regional material culture and distinctive cultural forms unique to specific ethnolinguistic groups, geographic areas, or temporal periods.

Throughout this chapter, I have discussed several different theoretical approaches and methodological techniques, most oriented within a poststructuralist framework and emphasizing human agency and practice. Despite any flaws, the humanist-perspective in the approaches presented here is critical for interpreting a topic as complex as the Mesoamerican ballgame.

Bridging the gap between theory and method, on the other hand, is a much trickier path to navigate. In the analysis and interpretation I present in the following dissertation, I very much follow recent trends in archaeology by following a mixed approach combining qualitative fieldwork and quantitative statistical and GIS-based spatial analyses (McEwan 2012; Rennell

2012).

2.3 CONSTRUCTING AND NEGOTIATING COMMUNITY AND SOCIAL IDENTITY IN MESOAMERICA

There has been a long tradition of community and ethnographic studies in Mesoamerica

(Canuto 2002; for an excellent synopsis of this history, see Chambers and Young 1997).

Despite this long focus, there are still several problems with the general approach to social identity, in particular ethnicity, in Mesoamerican studies (Emberling 1997), especially within archaeology. For the most part, the archaeological site is automatically equated to an existing, bounded community in that past (Yaeger 2000), where group membership is clearly signaled in the material record (Brubaker and Cooper 2000; Jones 1997; Meskell and Preucel 2004; Sackett,

1977; Schuyler 1980; Upton 1996). However, the evidence is much more complicated than that.

Aztec communities, for example, vary so greatly in time and space that emic perceptions of community and urbanism clearly do not necessarily coincide with archaeologically observable settlement systems (Hare 2000:79). Compounding this issue further, modern indigenous group identities are often imposed on local archaeological sites without much consideration for how these identities may have changed after centuries of Spanish colonialism and Mexican nationalism. The direct equation of communities with archaeological settlements therefore overlooks the potential complexity of spatial and non-spatial networks of social, political, and economic relations (Hare 2000:79).

More recent studies have attempted to address this problem by applying poststructuralist and practice-based approaches to questions of community and social identity in Mesoamerica

(Annis 1987; Browner and Perdue 1988; Hawkins 1984; Martin 1990; Rothstein 1988; Schryer

1987; Siverts 1985). Such research is increasingly showing that social identities, whether linked to ethnolinguistic group membership or not, were much more complicated and variable in the

Prehispanic past than has previously been assumed. The evidence directly questions our assumptions about the geographical distribution of social identities and demonstrates clearly that

Mesoamerica did not consist of static social groups living in neatly bounded territories. For example, there is evidence for the cohabitation of ethnic groups in many areas throughout

Mesoamerica both in the Prehispanic era and historically (Messenger 1987). Aside from obvious examples such as Teotihuacan and the Aztec capital of Tenochtitlan, there is increasing evidence for multiple ethnic groups living together in and around many important regional polities in the

Maya area, such as Classic-period city of Copan. This is especially interesting because the Maya area is often treated as being largely ethnically homogenous. Broadly speaking, local communities in Mesoamerica during the Prehispanic era were supradomestic social groups that shared a common history and connection to specific, though not always contiguous, geographic locations, but this does not need to mean that communities were always or necessarily ethnically homogeneous (Barber 2013:67).

In this chapter, I identified several social fields as particularly important for understanding the role of the ballgame in the production and negotiation of community and social identity in Mesoamerica. Despite Spanish efforts at stopping “idolatrous” practices, the sacred 260-day ritual calendar was in use until the 17th century (Forde 2002), and many rituals went underground, hidden from the view of prying priests. The continuing importance of

95 religious organizations in community life is well documented not only in Oaxaca, the geographic focus of this dissertation, but throughout Mexico. Domestic- and community rituals are still critical focal points for the construction of corporate identities in the present day, and today many practices are a synthesis of Prehispanic and Catholic liturgical traditions. The focus of these organizations are the periodic festivities and related activities carried out in veneration of local religious figures and important Catholic saints. Sports competitions, ranging from basketball games to rodeos and races, are still important events during the fiestas. The fiesta system is maintained by committees of people, often a ranked group of individuals who fulfill these social obligations during their period of service or cargo, expending time and money in preparation for and hosting these public festivities (Earle 1990:115).

Earle’s (1990) study of highland Maya communities importantly shows how local religious organizations take an active response to the evolving sociopolitical and economic conditions of the larger world in which they are embedded. For these communities, the periodic and ritualized reiteration of ethnic identity helps to socially unify and define the group, an especially important process when they are under acculturative and exploitative pressures. The group-wide veneration of the spiritual and physical embodiment of the community creates social bonds that help to define it as a sociopolitical entity (Earle 1990:121). Similar forces are very much at play among communities in Oaxaca, particularly in the more rural regions that still follow the traditional system of usos y costumbres (uses and customary law), as I discuss further in Chapter 4. Annual celebrations in honor of a town’s patron saint, which take place over a period of anywhere from two to five days and involve multiple (and expensive) events, reinforce a community’s sense of self as well as individual members’ ties to it.

Public and ceremonial architecture has long been considered instrumental in the construction of the community as an identity marker and place in Mesoamerica (Clark 2004).

How exactly these architectural spaces relate to the emergence of social complexity and the establishment of entrenched social hierarchies has been frequently debated. Basically, do public and ceremonial buildings precede or follow social complexity? For some, monumental architectural spaces, including temples, palaces, plazas, and ballcourts, were the direct outcomes of social complexity (Clark 2004), while others argue that social hierarchies were instead the unintended consequences of these socially marked features in the built environment (Joyce ).

But new evidence is being uncovered every day that complicates these previous interpretations.

Elite spaces, for example often started out earlier in their use-life sequence as communal spaces that were the focus of integrative, community-focused events (Barber 2005; Inomata et al. 2013;

Inomata et al. 2015). Oftentimes, these spaces were co-opted by elites because of their communal, integrative nature and thus potential as a source of power and authority.

As spaces for the public performances integral to community maintenance and function, monumental public buildings were some of the most visible symbols of sacred power in

Mesoamerica (Joyce and Winter 1996: 37). Generally speaking, in more heterarchical organizations, because communal labor constructed public buildings they were considered to be communally owned, and thus were sites of community-focused ceremonies (Brown 2000:382).

In contrast, in hierarchical societies, elites performed dual-purpose rituals that defined the exclusivity of their group membership while also helping to create political identities at a larger scale that included, as much as possible, a wide swath of the community membership itself

(Barber 2005). In this way, not only did monumental spaces become associated with large-scale identities, but the ritual practices that took place there as well (Barber 2005:72), creating a

97 recursive citation of spaces and practices. This is especially evident in Mesoamerica. However, while many structures such as temples, platforms, ballcourts, adoratorios or altars, and plazas are found throughout Mesoamerica in sometimes strikingly similar arrangements, there are enough clear distinctions between regions that should be explored further (Joyce and Winter

1996: 41). For example, sites in the Central Valleys of Oaxaca emphasized pyramidal mounds that tended to be more massive in scale and size; monumental design in the Mixteca Alta region, on the other hand, appears to have focused more on large platforms and monumental terraces

(Balkansky et al. 2000: 377; Kowalewski et al 2009: 303). These distinctions may reflect different ways of performing sacred rituals and distinct modes of communication and organization between various social groups in a community (Stoll 2014).

As a specific type of public monumental architecture in Mesoamerica, ballcourts were common enough that they would have been familiar structures recognizable to or even seen by many people, but not so ubiquitous as to be commonplace. That is, while many places have a church/temple, not everyone has the political power or resources to have a stadium/ballcourt. In many ways, ballgames and ballcourts were liminal phenomena. Both – one an intangible practice and the other the physical embodiment of its performance – occupy a continuum from the “profane” to the “sacred”, the everyday to the extraordinary, while also encompassing commensality and exclusivity, egalitarianism, and elitism. The rituals and ideologies associated with the ballgame and ballcourts emphasize liminality through reference to opposing forces of lightness and darkness, destruction and creation, pleasure and pain. In many ways, the presence of ballcourt is an immediate statement about the organization of power in a region (and at smaller scales) even when there is only one, especially in areas that may lack obvious centralized structures of power such as palaces or are organized heterarchically.

2.4 CONCLUSION

A critical aim of this dissertation is to understand how ballcourts were placed on the landscape in socially significant ways in the Nejapa subregion of Oaxaca. Investigating the complex relationships between human beings and landscapes requires a multidisciplinary approach drawing on ethnographic, environmental, archaeological, environmental, and even experiential sources (Conolly and Lake 2006:43), despite some of the methodological problems associated with the latter. In this chapter, I emphasized the need to understand the social and political processes that are associated with ballgames in Mesoamerica. The concept of processes used here is derived from postructuralist approaches including practice (Bourdieu 1977; de

Certeau 1984; Fogelin 2007, 2008; Pauketat 2001; Lefebvre 1991) and performance theories

(Dornan 2004; Inomata and Coben 2006; Pauketat 2001). Here I treat standing architecture, sports and gaming, ritual activities, and ceremonial feasting as critical social fields for the construction, negotiation, and contestation of community and social identity (Forde 2002;

Pauketat 2001ќ). The practices linked with these social fields were active in the creation and negotiation of cultural traditions, power relations, and the spaces where these practices take place as well as the social structures that these spaces support (Smith 2003: 72).

What establishes particular polities at a specific place and time are certain configurations of local and/or regional political practices and social fields, established through the experience, perception, and imagination of landscapes and material realities by human actors, which themselves establish and reproduce relationships among individuals (Barber and Joyce 2007;

Smith 2003). While complex societies present a greater range of relationships that must be negotiated (Hendon 2003), heterarchical societies in frontier regions present us with novel or

99 ambiguous forms of these same relationships because traditional social boundaries are more fluid. Ambiguity in such cases should not be feared, but rather viewed as deserving of special consideration particularly in culture contact studies (Lightfoot and Martinez 1995). The negotiation of social identities and local communities among multiple ethnolinguistic groups was clearly an important process in creating the political formations and organization we see in

Nejapa during the Classic and Postclassic periods. The ballgame was one way by which this negotiation took place.

In fact, Mesoamerican ballgames and the practices that accompanied them formed an important component of community and social identity at different times and in different places.

Indeed, as the archaeological, historical, and ethnographic evidence will show (Chapter 3), there were many instances where games occurred outside the control and/or knowledge of elites, were sponsored by other status groups, or were the highlight of smaller-scale community events.

Moreover, many games took place outside the ceremonial precincts and/or elite-sanctioned spaces for ballgames; ballcourts were not the sole location for ballgames, and portable court markers are evidence that the game could be and was played in many places. Their widespread material and ideological presence across time and space supports their incorporation into the messiness of human political organization and social negotiation.

100

CHAPTER III

A GAME FOR GODS AND A GAME FOR MEN

“Many people work very hard at play and take such activities very seriously” Blanchard 1981:64

3.1 INTRODUCTION

Mesoamerican scholars have always recognized the prominence and importance of indigenous ballgames. Despite the diversity in the material evidence, researchers have often assumed first, that there was a singular ballgame; and second, that this ballgame primarily served the interests of elites in the sociopolitical relations among and between Mesoamerican communities across space and time. The sociopolitical processes that made both the games and the courts on which they were played efficacious to participants are still relatively not well understood. Similarly, the interpretations of the iconography, and thus the meanings and/or functions that are assigned to these icons and symbols, are often rigid and anachronistic

(Cohodas 1975; Uriarte 2001; Wilkerson 1991), while ballcourt studies have mostly been limited to questions of typology or construction sequences. Thus, change or flexibility in this 3000-year- old tradition is not always adequately accounted for.

Within the past twenty years, however, there has been a small but noticeable shift in how scholars approach the study of ballgames and ballcourts. Researchers are beginning to incorporate a perspective of the ballgame as agentive social praxis in Mesoamerica, and are increasingly recognizing that games may have had a more multifaceted function in Prehispanic

101 society than commonly assumed. Mirroring Vennum’s approach outlined in Chapter 2, in this chapter I bring together the material and the experiential, the tangible and intangible evidence about the ballgame from archaeological, ethnohistorical, and ethnographic sources. First, I summarize the prevailing interpretations about the Mesoamerican ballgame, laying out the some of the issues I see in how the subject has been typically approached. In the following section, I look at the anthropology of sport in general and indigenous games in particular. I then present data from an ethnographic research project I conducted on a modern ballgame played in Oaxaca today. Linking (carefully) the ethnographic present to the archaeological past can help to provide insight into the human dimensions of the ballgame that are missing from the material record. Next, I review the archaeological data and material evidence related to the ballgame and ballcourts at the regional scale, with a special focus on Oaxaca. Finally, I pull the various threads together and discuss the Mesoamerican ballgame in light of new perspectives, theoretical considerations, and the current evidence.

3.2 A MESOAMERICAN GAME: AN OVERVIEW OF BALLGAME SCHOLARHIP AND INTERPRETATIONS

The Mesoamerican ballgame has always been a topic of interest among Western scholars.

Paul Kirckoff (1966 [1943]) even included the ballgame as a key characteristic in his designation of Mesoamerica as a unique geocultural region. How scholars have interpreted the ballgame can be roughly grouped into either religious cosmological-agricultural-fertility themes, or more secular political-economic-conflict themes (Day 2001; Fox 1994; Fox 1996; Gillespie 1991; Hill and Clark 2001; Taladoire 2000; Taladoire 2001; Taladoire and Colsenet 1991; Whalen and

Minnis 1996; Wilkerson 1991). For those who emphasize secular functions, the ballgame is associated with rites of accession and the legitimation of authority between elites. It is also seen

102 as an outlet for social aggression; a social field for inter-elite competition; a substitute for warfare; and a way to avoid or terminate interpolity conflict (Kowalewski et al. 1991; Joyce and

Winter 1996:38; Stern 1948:96–97; Taladoire and Colsenet 1991:174). Some have even proposed that idle warriors played the game when not at war in order to maintain their fighting readiness (Kowalewski et al. 1991; Redmond 1983). The appearance of formal structures for playing right as social hierarchies were emerging suggest the game was linked to developing stratification and a rising elite class (Hill 1999; Hill and Clark 2001; Hill et al. 1998).

Convincing arguments have been made that ballcourt construction increased in times of intense political competition, particularly in situations where there are multiple polities, or a centralized power is breaking down (Feinman and Nicholas 2011). Santley and colleagues (1991:4) contend that the simultaneous use of multiple ballcourts signified regional political decentralization with elites in constant competition (see Stark 2012 for a contrasting opinion). According to this argument, while the sociohistorical meaning of the ballgames would have varied, the essential objective for playing was always the same – economic and personal gain by ruling elites who either participated themselves or sponsored teams (Santley et al. 1991:15).

Scholars favoring a more religious or ritual function argue that the game was primarily a cosmological and/or agricultural fertility ritual that involved acts of sacrifice that was exclusively controlled by elites. As a cosmological ritual, the game represented the fight between contradicting and opposing forces such as lightness and darkness, the forces of the underworld, or between the Sun and the Moon (Berger 2009; Day 2001; Fox 1991; de la Garza 2000;

Gillespie 1991; Uriarte 2000). Uriarte (2000:30-31) points to Venus and solar imagery in the artistic canons as evidence of the game’s celestial character. In the Central Mexican Codex

Borgia, sacrificed ballplayers are painted with the red stripes symbolically associated with the

103 deity Tlahuizcalpantecuhtli-Quetzalcoatl, or Venus in its cycle as the Morning Star. Xólotl, or

Venus as the Evening Star, was another Mexican patron deity of the game (Uriarte 2000:30).

Because the codices often depict gods playing, several authors also suggest that the game could have symbolized the cosmos, with the movement of the ball across the court representing the sun as it passes through the sky and the underworld (Day 2001; de la Garza 2000; Gillespie 1991).

In addition to solar and cosmic images, liminal creatures such as butterflies, frogs, turtles, and crocodiles also appear in ballgame iconography, evidence that the ballgame ritually symbolized

“unification through opposition” (Uriarte 2000:31-33).

Additionally, scholars suggest that the morphology of the ballcourt, its very form and design, represented the entrance to the underworld or was a metaphor for the underworld itself

(de la Garza 2000; Schele and Freidel 1991; Taladoire 2000:27; Taladoire 2001; Uriarte 2000); or it symbolized the channel for the birth of the sun and cosmos (Uriate 2000:33). We can find support for this interpretation in many of the ethnohistorical accounts. In the famous Quiche-

Maya legend of the Popol Vuh, the ballgame mediates the combative interaction between the

Hero Twins and the lords of the underworld that took place in that dark realm (Tedlock 1996).

The indigenous views presented in the codices pair the courts and game with symbolically valued animals such as jaguars, coyotes, eagles and quetzals. In turn, the jaguar and coyote themselves are linked to the earth, night and war, while the eagle and the quetzal are associated with the daytime world, the heavens, dryness, and the warrior. Aside from being both a symbol of the underworld and the birthplace of the sun and light, the codices also depict the tlachco or court as a symbol of fertility and as a place of confrontation between opposing and complementary forces (Taladoire 2015a).

These interpretations are by no means incorrect as the material evidence does generally

104 support them. However, the exclusive focus on elite expressions of ballgame participation, or only viewing the game as an elite political strategy or ritual tradition, means that we may be missing the larger picture (Stoll and Anderson 2017). The increasing incorporation of poststructuralist perspectives over the last twenty years has led to a noticeable shift

Mesoamerican ballgame studies. Beginning with Fox’s (1994; 1996) argument that ballcourts acted as facilities for social integration and political ritual, a small group of researchers have begun looking at the game as an actively lived sociocultural practice by incorporating a poststructuralist perspective as well as theory from the anthropology of sport (Baron 2006; Hill

1999; Ramos 2012; Stoll and Anderson 2017). Hill (1999), for example, was one of the first ballgame scholars to seriously consider sociological and anthropological discussions of sports, drawing a connection between competitive games and the evolution of social and political inequality in Mesoamerican society. In doing so, these researchers have independently arrived at similar conclusions regarding the ballgame and the role it played in how Mesoamerican communities interacted with each other, both internally between different social classes and externally in peer polity networks. Collectively, they conclude that as a competitive sport, ballgames also played a critical role in community and network building for Mesoamerican societies (Stoll and Anderson 2017). Over the course of the following sections, I lay out the theoretical, ethnographic, and material evidence that supports this new perspective on this ancient and enduring tradition.

3.3 INDIGENOUS SPORTS AND GAMES OF THE AMERICAS

The primary impression we have of the Mesoamerican ballgame and its social importance is as a cosmological, fertility, unification/opposition, or conflict-avoidance ritual that was fundamental to elite political strategy. That the ballgame is also a sport is often dismissed or

105 seen as secondary (Altuve 1997; Bernal and Seuffert 1979; Cohodas 1975; Day 2001; Fox 1991;

Koontz 2008; Leyenaar 1978; Miller 2001; Miller and Houston 1987; Schele and Freidel 1991;

Uriarte 2000; Zeitlin 1993). Following from this dismissal is the unnecessary separation of the ballgame as sport from the ballgame as ritual. Yet, as I discussed in Chapter 2, sports are liminal phenomena just as rituals are, and as a social field they are easily ritualized and politicized. This explains why elites and other invested social classes would seek to engage in or align with friendly (and not) organized, competitive social activities. Having laid the theoretical foundation for why sports are affective phenomena that play a role in mediating social relationships, I now explore the historic indigenous games and sports played by native peoples in the Americas.

Much of the literature focuses on sports in Western society, and while there are plenty of cross- cultural similarities, understanding the Mesoamerican ballgame requires we place it in its appropriate context as there are significant emic differences that must be considered between

Western and indigenous attitudes towards sports (for example, indigenous views on gambling).

The various games and sports played by the indigenous peoples of the Americas have been studied by anthropologists since the 19th century. One of the first scholars to look at indigenous sports in-depth was James Mooney (1899), whose article “Cherokee Ball Play” was a lengthy description of a Cherokee “racquetball” match that occurred in September 1899.

Mentioned previously, Culin’s Games of North American Indians (1907) remains an important resource for ethnographic information on indigenous games, many of which no longer exist.

Ethnographic studies of indigenous sports and games continued throughout the 20th century, including Cheska’s (1979; 1984) research on Navajo sports and Blanchard’s (1981) study of

Choctaw stickball and native attitudes towards Western sports. Finally, an edited volume organized by Barbara Voorhies (2017) on the Prehistoric games of North America brought

106 together several archaeologists whose research focuses on games from different geographic regions and time periods. In this section, I draw on these sources to present archaeological and ethnographic information on indigenous American games, mostly North America.

Indigenous games are typically divided into four general categories: kinetic competence, chance, games of representation, and strategy. Emically, the different types of games that fall into these four categories are generally related to diverse social processes such as sex role differentiation, group identity, decision-making models, symbolic identification, the development of important physical skills necessary in adult life, communal integration, and the distribution of resources through wagering (Cheska 1979:229; Cheska 1984:250). In this way, the meanings of traditional games and sports extend beyond physical activity to other important cultural beliefs and traditions (Cheska 1984:250; Stoll and Anderson 2017; Voorhies 2017).

Lacrosse, for example, was frequently associated with warfare by the native tribes that played it

(McGarry 2010), and Choctaw stickball was known as the “little brother of war” (Vennum

1994). This connection to warfare has influenced some interpretations of the Mesoamerican ballgame and its function in Prehispanic societies (Redmond 1983). However, Vennum (1994) has also demonstrated that lacrosse had different and changing social meanings for First Nations people, showing how the game was variably linked to spirituality, medicinal practices, shamanism, warfare, and gambling, both regionally and temporally. On the other hand, ethnohistorical and ethnographic data also clearly demonstrate that indigenous sports and games were viewed by the indigenous as recreational activities as well (Cheska 1984:250).

Gambling was (and still is) a common activity that accompanied nearly all indigenous sports and games (Williamson and Cooper 2017:31; Yanicki 2017; Zych 2017) and can even be considered a universal aspect of pre-contact indigenous society (Cameron and Johansson 2017:1-

107

2). The popularity of gambling practices may be in part because it acts as a social leveler, redistributing material wealth and creating some stability between social groups and nations – there is always, for example, the possibility that losses can be regained (Williamson and Cooper

2017:34). In contrast to Western and Christian views, gambling for native societies was a religious behavior (Walden and Voorhies 2017:7), but there was also pleasure to the activity itself and the drama it added to games (Cameron and Johansson 2017:1-2; Evans 2017).

Gamblers would bet their property, clothes, family members, and even their own lives (Cameron and Johansson 2017; Yanicki 2017). Men often engaged in high stakes bets, particularly in large public settings when playing or betting with men from other clans, tribes, or ethnolinguistic groups (Voorhies 2017:22). In fact, the Blackfoot word for gambling is used interchangeably with the words for gaming and play (Yanicki 2017:13).

As I established earlier, through their performance, games create a shared sense of unity among members and opportunities for strengthening group affiliation. The Oldman River was the playing grounds for the Blackfoot Peoples of Southern Alberta’s Rocky Mountains (Yanicki

2017:1). Ute tribes gathered in the spring in the Provo River region (Great Basin) to engage in various sports such as horse-racing, trading, gambling, foot racing, and wrestling (Janetski

2017:1). At the same time, in some contact situations, different sport activities only served to deepen ethnic divisions. An excellent case study comes to us from the Southwest United States, where culturally-specific sports marked the ethnic boundaries between Navajo, Mexicans, and

Euro-Americans (Culin 1903; Cheska 1984). Culin’s (1903) description of a Fourth of July celebration where all three of these groups participated in various sporting events scheduled either at the same time, or at different times of the day, but spatially separated, demonstrates that instead of being integrated, the prescribed behavior, temporal and spatial structuring, and social

108 relationships of each group helped to create and maintain the boundaries between them.

Indigenous sports and games were often included in the rituals performed for sacred ceremonies and social celebrations and were also tied to symbolic identification and cultural maintenance (Acuna 1978; Cheska 1979:236-237; Cheska 1984:250; Gutierrza 2017; Lopez

Austin 1967; Voorhies 2017:3; Yanicki and Ives 2017:2). Sports such as lacrosse and stickball, with their highly stylized pregame and game procedures, have been particularly highlighted as providing a symbolic mechanism for group identity (Cheska 1979:232). Many game-related myths explain the earliest origin of things; the gods played games amongst themselves and in turn gave them to man, officially giving games divine sanction (Cheska 1979:237). The reason the Blackfoot Peoples of Southern Alberta gathered at the Oldman River was because this was the place where, according to their oral histories, the sacred Napi had taught the different nations how to play the wheel game in order to make and maintain peace (Yanicki 2017:1). Like the

Mesoamerican ballgame, balls used in many North American games also carried religious significance, alternatively symbolizing the earth, the sun and the moon, while also representing fertility, rain, wind, thunder, and lightning (Cheska 1979). Other sports paraphernalia such as chunkey stones, bowls, dice, game sticks, etc., were closely associated with nostalgia, memory and intra- and interpersonal reconstructions of past events (McGarry 2010:159). For example, many chunkey stones were heirloom objects belonging to specific families and communities.

Their use in the present recalled previous games in which that same stone had been played (Zych

2017).

When we look closely at the ethnographic case studies, we easily see interesting parallels between these North American sports and the Mesoamerican ballgame. George Catlin’s descriptions of Choctaw stickball or toli games that he attended in the late 19th century provide

109 some of the richest descriptions of the game as it was played historically, when stickball still retained many of the native traditional practices and attitudes. Feasting, ceremonial activities, and other games such as handball and a version of jacks would take place before, during, and after the main event (Blanchard 1981:23-43; Catlin 1953:290, 293). Matches were announced several months in advance, with arrangements made between the captains or champions of the respective teams. People would travel 10-20 miles on average to attend these events. The playing ground was usually prepared by older men who also often served as judges (Catlin

1953:291). These officials, known as apisaĉi, kept score using visual mnemonic devices, typically sticks that represented points for the competing teams that were inserted into the ground. Music also played an integral role in stickball games. Drummers would play throughout the match and attendant ceremonies, keeping up the excitement of the game by pounding out a consistent percussion beat (Blanchard 1981:35). Other key figures in Choctaw stickball matches, aside from the players and spectators, were Choctaw doctors (ritual specialists), singers or italowa, clowns, and men who as stakeholders supervised the gambling and bet-making (Blanchard 1981:36). Although many of traditional elements are no longer included as part of stickball game matches, the sport is still actively played by the Choctaw in

Mississippi today, along with other Western sports. In many ways modern Choctaw stickball, like its historic predecessor, plays a role in community integration and identity creation.

The Apalachee living in Protohistoric North Florida played a stickball game that also carried ideological significance. The motions performed in the game were likened both to the movement of celestial bodies and the way heroes played in a timeless primordial past (Stauffer and Reilly III 2017:1, citing Aveni 2002, 2010). The game was an important part of the seasonal activities and celebrations linked to the agricultural cycle, and it was also tied to mobilization for

110 warfare as well as shifts in the political order (Stauffer and Reilly III 2017:2). Apalachee stickball shared many elements in common with other indigenous sports: for instance, ceremonies occurred before and after the game, including processions, costuming, ritual medicine consumption, and the placing of wagers. Today, the Muskogean-speaking peoples living in the Southeast still play a ballgame now known as hottí iscósi, where it continues to mitigate conflicts between various groups (Stauffer and Reilly III 2017:5).

Of the many various games played by indigenous peoples in prehistoric North America, perhaps none was more popular or widespread than chunkey. The earliest evidence for chunkey can be found at Cahokia (Zych 2017:1). During the height of the Middle Mississippian Period

(CE 1050-1400), the game was apparently co-opted and professionalized by the elite in the

Greater Cahokian region (Zych 2017:2). Originally, chunkey fields were open and accessible, constructed on the edges of towns. This created a prescribed space for spectators and participants to enjoy the game. However, over time these fields became more rectangular and were moved to the town center (Zych 2017:8-9), possibly representing a shift in the political organization and significance of the game. Hoop and disk games, of which chunkey is a type, were emically related to various native mythological beliefs. Through their active engagement in these sports, individuals connected themselves not only to other people, but also to other cosmic worlds. Again, the fact that chunkey stones were curated and therefore heirloom objects means that the stones themselves invoked past games and (re)created meanings (Zych 2017:.6-7).

Chunkey stones have even been found in high-status burials, likely placed with individuals who were excellent athletes when they were alive. The similarities among the several variations of the chunkey game point to shared ways of playing, while the differences suggest that certain practices were also localized (Zych 2017:16-35).

111

Lacrosse was a popular stickball game played by the Wendat, Iroquoian, Anishinaabeg, and Algonquian tribes (along with other traditional games such as dish and snow snake)

(Willliamson and Cooper 2017:2). Game matches sometimes involved up to hundreds of players, especially when played between different nations, among adjacent communities and even the clans within larger communities. The matches took place on extensive meadows or grass plains with natural boundaries, with the goal lines set hundreds of feet apart. The distance between goal lines and the number of players meant that games could sometimes take up to several days (Williamson and Cooper 2017:9-10). Aggressive physical contact was often encouraged, which caused everything from minor injuries to permanent deformity and death

(Williamson and Cooper 2017:11). Like with many indigenous games, betting was heavily involved and even considered to be as important as the game itself, with everyone from players to spectators wagering (Williamson and Cooper 2017:12). Lacrosse games were extremely popular among native peoples because, as mentioned previously, they touched on multiple aspects of cultural and daily life, including physical prowess, warfare, prestige, gambling, dreaming, mourning, curing, and shamanism (Williamson and Cooper 2017:1; Vennum 1994).

The early chronicler Jesuit Jean de Brebeuf, for example, observed that the Wendat or Huron played both lacrosse and dish games in healing rituals, to influence the weather, and as a way to honor the memory of good players (Williamson and Cooper 2017:6, 9).

Dice games were also very popular among indigenous societies. They are assumed to have originated with divination rituals. Astragali-type dice were regularly used for gambling throughout North America (Dye 2017), linking playing with divination and ritual. The Mi’kmaq women of Canada's Maritime Provinces and the Gaspé Peninsula of Quebec play a version of the bowl and dice game known as waltes. While a favorite pastime of many, it was strangely

112 considered to be exclusively a women’s game (Leonard 2017). The Chiricahua Apache favored a stave dice game that they would play when the multiple local bands gathered together for ceremonies and socially integrative activities (Seymour 2017). Multiple Mesoamerican societies played a dice game as well, commonly known as patolli among the Aztec, patol in Michoacan, petol for Nahuatl speakers from Huiztzilan and Puebla, and by an array of other names as well

(see Table 3.1). There were even professional players who traveled with their gaming implements and lived itinerant lifestyles (Walden and Voorhies 2017:3).

Table 3.1: Different indigenous names for the patolli dice game (Kendall 1980:12; Walden and Voorhies 2017). Group Location Name Totonac Zapotitlan lizla Aztec-Mexica Central Mexico patolli Various Michoacán patol Nahuatl Huiztzilan/Puebla petol Lowland Mopan and K'ekchi' Maya Southern Belize bul or baac Chita-Acaxee Northwest Mexico patole Tarascans Michoacán/Jalisco/Guanajuato kolia atarua K'iche' Quiche Maya Highands bag Tarahumara Chichuahua romavoa or guinze

3.3.1 Discussion and Summary: Sports as an Important Social Field in Indigenous America

Sports and games are captivating to spectators and players because their outcomes depend on a combination of chance as well as feats of athleticism involving mental strategy and physical skill; this combined with their sociability and communality makes them socially and culturally important (Hann 1988:74). All sports share several general features in common: they involve two or more individuals or teams who follow prescribed rule systems and playing techniques; and matches are often played in combination with religious and secular festivals, either forming part of the activities or are the main focus of these celebrations. Among indigenous societies in the Americas (and worldwide overall), people participated in sports and

113 games through a range of roles as hosts and guests, players and spectators, hated rivals and friendly competitors (Taladoire n.d.:15); these games and sports were played for recreational reasons as well as for larger sociopolitical motivations. Furthermore, sporting events were, and still are, accompanied by feasting, ceremonial activities, pregame dances and postgame celebrations, betting, and other adjacent games involving spectators of all ages and genders that together create group solidarity (Catlin 1953; Williamson and Cooper 2017:37). Audiences for sporting events additionally gave games a sense of theatricality and entertainment (Zych

2017:42-43).

Sport paraphernalia are often associated with nostalgia, memory, and personal reconstructions of past events (McGarry 2010:159). There is a huge market in the United States for people who collect sports memorabilia, including signed baseballs, jerseys, and playing cards. Through these shared experiences, via the games and the activities that accompanied them as well as the material culture produced from these experiences, communal bonds and a common identity are forged, once again turning individuals into community members (Inomata 2006:805;

Inomata and Coben 2006:11; Swenson 2008). For First Nations peoples, lacrosse was just as much about competition and athleticism as it was about warfare and spirituality; these meanings changed as one moved through time and space according to the needs and concerns of the players (Vennum 1994), game sponsors, and audience members (Stoll and Anderson 2017). In these ways, sports can have phenomenological meanings that are neither rational nor utilitarian

(Gruneau 1975:129), a fact we can see reflected in the way chunkey matches, game objects, and playing fields were important in creating, recalling, and negotiating social memory (Zych

2017:41).

Importantly, athletic rivalries exist even in the absence of actual or real hostility. Team

114 rivalries across different American sports are well known: in baseball, New York Yankees’ fans despise fans of the Boston Red Sox, while the rivalry between the Chicago Bears and the Green

Bay Packers dates back to the 1920s, and is still the oldest rivalry in American football. Some of the strongest rivalries in modern American athletics exist between fans of teams who play in the same city, so-called “subway” or “cross-town” rivalries. The need for the reaffirmation of regional identity most surely contributes to the intense soccer rivalries within and among

European nations (Bale 1997). According to 16th century Spanish Colonial accounts, ballgames among indigenous communities created team and/or village loyalties, heightening the sense of belonging to a communal identity (Ramos 2012:113). Baron (2006:52) explicitly links community identity to the universal properties of athletic competition in games.

The perception, selection, and expression of behaviors or practices by the group members reinforces cultural differences and similarities as needed. Sports in this sense are just like other sociocultural behaviors that serve to mark group membership and identity (Cheska 1984:252-

253). The Diné adoption of a Euro-American sport such as basketball was not a sign of assimilation because it has been integrated into the repertoire of Diné cultural performance and group expression (Cheska 1984:250). How the sport is played echoes particular group ethos of cooperation among teammates in ways that contrast with American styles of playing basketball.

Finally, under circumstances in which the identity of a community, and therefore the social order, is threatened, sports and games will be co-opted by elites in order to reference and reinforce group solidarity (Baron 2006:54). This especially happens when the interactions, material resources, and aesthetics shared among the elite members set them apart from other members of their community, leading the latter to question the benefits of shared identity and materially supporting the needs of the elite.

115

Indigenous games such as lacrosse, stickball, chunkey, and dice, among others, offered opportunities for social (via prestige from athletic or gambling prowess) and physical mobility

(via travel to other communities and locations for game matches) for both spectators and players

(Zych 2017:37-39). The playing grounds of the Oldman River, the sacred location where they had been given games for peace, was also a gathering place for the different bands and tribes of the Blackfoot Peoples (Yanicki 2017:1). Ute tribes would come together in the springtime in the

Provo River region (Great Basin) to engage in various sports such as horse-racing, trading, gambling, footracing, and wrestling (Janetski 2017:1). Historic Choctaw stickball matches drew tribal members together who otherwise lived in dispersed communities, and today people still travel to attend matches for both traditional stickball and Western sports (Blanchard 1981), as they do among many rural communities in Mexico as well (especially in Oaxaca). Accounts of gambling games across North America show that there was a “[broad] phenomenon of intertribal competition within the socially mediated context of play” (Yanicki 2017:2). In this way, sports established – and continue to do so – sociopolitical networks among otherwise spatially dispersed but culturally linked communities.

3.4 ES NUESTRA TRADICIÓN: MODERN BALLGAMES IN MEXICO AND OAXACA It is an interesting observation that in the bulk of archaeological research on the

Mesoamerican ballgame, few scholars (see Taladoire 2003) have looked at ethnographic data on the modern ballgames played in Mexico today (or the numerous other diverse games for that matter). During my first field season working in Oaxaca in 2008, I conducted a small research project on ballcourts in the Central Valleys. A colleague of mine, inquiring about my project, told me about a traditional ballgame match he had attended the year before in the town of

Yanhuitlan in the Mixteca Alta. He mentioned that it was during the town’s patron saint festival,

116 and that the players had used heavy gloves and a small rubber ball. Intrigued by the possibilities of comparing the ethnographic present of the ballgame to the archaeological past, in 2009 I designed an ethnographic research project to learn more about the modern ballgame played in

Oaxaca known as pelota Mixteca (Mixtec ball). I spoke with several informants on the game’s connection to community functions, whether ritual, political, or economic. What I learned from the ethnographic research helped to shift my perspective on the Mesoamerican ballgame.

In the following sections I present information on the traditional and autochthonous sports of Mexico, focusing specifically on those games that involve balls6 as playing equipment.

I begin with a general overview of these sports, their rule systems, ways of playing, and their cultural significance. I then move to the ethnographic data I gathered on pelota Mixteca over the course of two summers in 2009 and 2010 (Table 3.2). The sampled population included current ball players and audience members who often were former players themselves. Questions focused on technical aspects such as game rules, how often the ball players meet to play, and how games are organized, as well as more social aspects including how they learned to play, what playing pelota Mixteca meant to them, and if they knew how the game originated.

6 An excellent website with information on all the federally recognized autochthonous games of Mexico can be found here: http://www.jcarlosmacias.com/autoctonoytradicional/. There are 66 official games divided into three categories: sports (11); strategy games or mental dexterity (8); and simple game activities (47). 117

Table 3.2: Location and dates of pelota game matches visited for the ethnographic field project. Location Date 2009 San Antonio Arrazola June 14, 2009 Buena Vista June 21, 2009 Oaxaca de Juarez July 12, 2009 July 19, 2009

2010 Oaxaca de Juarez July 11, 2010 Guadalupe Etla July 18, 2010 Santa Cruz de Milpa July 19, 2010 Magedelena Jaltepec July 22, 2010 Santa Cruz de Milpa July 26, 2010

3.4.1 Traditional Sports of Mexico Today

Despite the concerted efforts of Spanish friars and other authorities, indigenous and mestizo campesinos continued to secretly play ballgames throughout the Colonial and National periods. In 1988, the Mexican government officially recognized the cultural importance of these games with the formation of the Federación Mexicana de Juegos y Deportes de Origen, or the

Mexican Federation of Original Games and Sports. Traditional ballgame associations in the

Distrito Federal and the states of Sinaloa, Guerrero, Oaxaca and Michoacán were the first initial participants. Today, the organization is now known as the Federación de Deportes Autóctonos y

Tradicionales, or the Federation of Autochthonous and Traditional Sports (FDAT). Members include at least 13 state associations and a multiplicity of different groups that play a diverse number of mostly traditional indigenous games, divided into physical sports and strategy games

(Turok 2000: 59).

The surviving thirteen ballgames, called modalidades in Spanish, are incredibly interesting in the contrasts and similarities of their rule systems and game play. The modalidad

118 considered to have conserved the most Prehispanic elements is ulama de cadera (hipball), said to be descended from the ullamaliztli (alternatively tlachtli or ōllamalīztli) or hipball game most often witnessed and described by the 16th century scholars. The Jesuit priest Francisco Javier

Calvijero (1945 [1780]) recorded that indigenous Northern Mexicans such as the Nayarita,

Opata, and Tarahuamra tribes played ulama. Although formerly played throughout Northern and

Western Mexico, it is now predominantly found in the states of Nayarit and Sinaloa. Ulama de cadera uses a 3-4 kilogram ball that, like the Prehispanic version, is struck with the hips (Turok

2000: 62). Matches take place on a taste or field that is approximately 60 m long by 4 m wide, and is divided by a center line called the analco that is said to signify the other side of a shore or river. The field is enclosed by end zones called chichis, and the judges (the veedores or juezes) who keep track of the score and make decisions during rule disputes are stationed in these areas

(Ramos 2012; Turok 2000). Two teams of 3-5 players are assigned to different playing positions on the court. Players positioned in front are the best and known as topadores, while those positioned toward the back are the golpeadores and they supposedly receive the hardest impacts

(Ramos 2012).

Once the ball is put into play the aim is to keep it in motion and not let the ball hit the ground. If the ball bounces out of bounds the team serving loses a raya or point. Additionally, if the ball hits the player outside of the hip-thigh zone, the opposing team then receives a point.

The men interviewed by Ramos (2012:210) describe the rules as complicated and taking years to learn, made much more so by the fact that each town has its own rule system. In contrast to other traditional sports, ulama de cadera players wear specific outfits called the fajado that consists of the gamuza (leather loincloth), chimalli (leather belt), and the faja (a long cloth wrapped around the stomach for support) (Ramos 2012:207). Face and body decorations are

119 common among the players, who are primarily men. However, even women who play decorate themselves. Ramos (2012:81) describes how young female ballplayers from Esquinapa, Sinaloa decorate themselves when they perform in special exhibits and parades; they also paint certain design motifs on their faces to create the impression that they are ancient ballplayers.

Related to ulama de cadera are the games ulama de antebrazo (arm ball), ulama de mazo

(mallet ball), and alternatively ulama de palo (stickball). Like ulama de cadera, ulama de antebrazo is played on a field known as a taste, however it is much longer and narrower at 100 m in length by 1.4 m in width. Again, the anlaco divides the court in the center, although the end zones are known in this game as chivos. A veedor is placed in each end zone while a referee stands in the center. The ball is much smaller for antebrazo, weighing only about 500 g. Much as the name implies, the ball is struck with the forearm, which is wrapped in protective cotton.

Sometimes the right thigh and/or the big toe is also covered with deer skin. Teams are smaller, with only 2 to 3 players. Finally, as in ulama de cadera, game points are referred to as rayas

(Turok 2000:61). Ulama de mazo is played on a court similar to antebrazo, with teams of 2 to 3 players using hand-made sticks (usually a tree trunk that has been sanded and trimmed). One stick weighing between 4 and 7 kg is set against a more rounded stick and both are used to hit a rubber ball that weighs 600 g (Turok 2000: 62). Today ulama de antebrazo and ulama de mazo are played in the Sinaloan towns of El Guamuchil and Mocorito, while ulama de cadera is the preferred game in Esquinapa, Los Llanitos, La Mora Escarbada, El Chamizal, and La Savila

(Aguilar 2004; Ramos 2012:6).). Supposedly, ulama de palo is more popular in Michoacán and

Oaxaca (Ramos 2012:206-207), although I have not seen this game played in the latter state. In all three variations, women have been or still are game participants.

The three pelota purépecha games use sticks of various sizes to strike the ball (Turok:

120

62). Pelota purépecha encendida is played in Jalisco and Sinaloa, where in the latter it is known locally as quiche. It is perhaps one of the few games that can be said to still have heavily symbolic overtones. A cultural tradition “rescued” by the local FDAT of Sinaloa, the game is supposed to represent the fight between the old and new suns. Players use a wooden baton, similar to a hockey stick, to strike a ball made of maguey roots that is soaked in petroleum and lit on fire. For this reason, the game is played at night (Turok 2000: 62-63). Using the same point system as ulama de antebrazo, pelota purépecha de trapo also involves a field hockey-like stick and a small round ball that could be made with maguey leaves, wax and cotton, a foam rubber ball, or a wooden ball lined with cotton cloth. There is an offensive and defensive team, and each has a captain. The object of the game is to pass the ball into the other team’s goal zone

(Turok 2000:62). The game is popular in Michoacán, but is also played occasionally in Puebla,

Nuevo Leon, Tlaxcala and Ixtlan. Finally, in the game pelota purépecha de piedra, players use a wooden flat paddle and a ball made of volcanic stone, playing on a court measuring between 600 to 900 m in length and 3 to 5 m in width. Game play involves both teams advancing towards the center from the end zones. The lanzador or shooter of the offensive team attempts to hit the ball as far as possible toward the front player on the defensive side, who then has to hit the ball and return it (Turok 2000:62).

Another ballgame that uses a wooden implement is pelota tarasca or frontón azteca

(interesting given that the Aztecs and Tarascans were traditionally enemies). Players strike either an industrial rubber ball made by the Naila Company or a tennis ball with the puño, a small, round wooden stick. The court they play on measures 120 m by 10 m and is divided into four sections: the zona de saque or strike zone (20 m); the traversal de adelante (5 m) or front advancing zone; the zona de resto or rest zone (5 m), and the contrarresto or counter zone,

121 which is considered to be free territory (Turok 2000:62). The game is popular in Michoacán,

Guerrero, and even Estado de Mexico. Play is between two teams of four members including a captain, a bolillero or striker, and two orilleros or side players, with another man serving as an alternate. Games are divided into three sets; teams compete to win four points in each set, in order to win twelve points in total and ultimately the game (FDAT).

Clearly, these traditional sports present a variable mix of many Prehispanic as well as

European characteristics. Turok (2002:61) argues that the games’ point systems can serve as guides for determining just how many Prehispanic elements have survived in each game. Based on the complexity of the scoring and rules, she claims that ulama de cadera and antebrazo are the closest to their Prehispanic antecedents, with the other games being more syncretic in nature.

3.4.2 Pelota Mixteca de Hule in Oaxaca

There are currently three versions of pelota Mixteca played around the state of Oaxaca: de hule (rubber), de forro (skin or covering), and del Valle (from the Valley). Each uses a different type of ball and playing gear (Don Jóse Angel 2009: personal communication). Pelota

Mixteca de forro uses a ball made with yarn, wool, or worsted yarn lined with suede. Players wrap their hands in electrician’s hide work gloves and cloth to protect their hands while striking the ball. Pelota del Valle involves a small skinny tablet or plank that measures 20 cm by 2 cm and is used to strike a small foam ball. These versions of pelota Mixteca are limited to certain regions of the state. De forro and del Valley, for example, are mostly played among Afro-

Oaxacan and indigenous communities on the Pacific coast.

The third and final version is pelota Mixteca de hule, the most popular and widely played of the three. The Enciclopedia de los Municipios de México (EMM) lists 55 municipalities in

122

Oaxaca that still have a pelota Mixteca field, the majority of which are in the Mixteca Alta and

Central Valleys (Taladoire n.d.:14). Many more fields existed in the 20th century; in the state capital of Oaxaca de Juarez, for example, there had been five such playing fields located in different neighborhoods, where today there is only one. In this game, players use a 1 kg vulcanized rubber ball that is specially produced by one factory in Mexico City. The balls are very colorful and usually painted with designs and either the name of the player or the community it belongs to (Figure 3.1). Community balls are used in the torneos or tournaments where multiple teams from different villages are playing. Someone from the community usually acts as a custodian for the ball.

Figure 3.1: Volcanized rubber ball used for playing pelota Mixteca. This ball specifically belongs to the team from the suburb of Buena Vista.

123

Pelota Mixteca de hule (hereafter pelota Mixteca) is famous primarily for the heavy leather glove that players use to strike the ball. These gloves are made of about 36 layers of stiff cow leather that are pierced by hundreds of metal studs or nails, and usually weigh at least three to five kilograms themselves (Figure 3.2). Players specially request them from the glove maker or guantero, a man who goes by the apodo or nickname of El Caballo (The Horse). El Caballo is the grandson of the original glove inventor, and he learned the techniques of glove making from his father. On a visit to his house one day, he showed me the different gloves made over the years and told me about their invention and how they have changed, evolving from the small swatch of leather used by his grandfather to protect a hand wound to the heavy glove used now.

Figure 3.2: A pelota Mixteca glove, brightly decorated and studded with over a hundred nails.

124

Previously, the way people had played pelota Mixteca for years was with their bare hands, known then as pelota Mixteca de mano fría. Men would dip their hands into boiling water in order to toughen up the skin and to hit the ball farther. The grandfather of El Caballo was a leather worker and pelota Mixteca player who, one day in 1911 before going to play pickup games with his friends, accidentally cut himself deeply on the hand. He grabbed a swatch of leather, cut it into a square and attached it to his hand using leather strings in order to protect the wound. He discovered that day that the leather helped him strike the ball even further than had been possible before. He soon crafted himself something that kind of resembles a golfing glove, covering only the palm and leaving the figures free. Later in the 1930s, metal rivets were added to strengthen the leather and keep the glove from falling apart after so many strikes from the ball (Figure 3.3). Soon the rivets were replaced with metal nails, which necessitated the addition of several more layers of boiled and hardened leather. Eventually, more layers and nails were added over the years, as well as paint and decoration. Early on, other players noticed that the gloves made by El Caballo’s grandfather were very effective in striking the ball and sending it over farther distances and grander heights. They requested that he make gloves for them as well. Today these gloves are brightly painted, featuring Mitla greca-inspired designs and on occasion the ballplayer’s apodo (Figure 3.4). They can cost up to $5000 pesos (in 2010) and take about two months to make.

125

Figure 3.3: Evolution of the guante or glove in pelota Mixteca, from the earliest leather swatch (upper) to the modern version used today (lower).

126

Figure 3.4: Modern pelota Mixteca gloves. The lighter blue glove in the back features the player's nickname Lento (Slow).

In the past, pelota Mixteca was played on designated canchas or courts, sometimes on a street near the church or central plaza, or wherever open space was available. Some patios were located directly in front of the town church (e.g. Magdalena Zhuatlán, Chalcatongo when they still played there, and in Nochixtlán up until the 1950s) (Berger 2009: 9-11). Most courts used to play today have been constructed in so-called polideportivos, usually alongside other fields designated for Western sports such as soccer and baseball (Berger 2009: 9), or are located close to the town center, such as at Magdalena Jaltepec in the Mixteca Alta. The Estadio Tecnológico is a polideportivo on the outskirts of Oaxaca de Juarez that has several different sports fields, including a patio de pelota Mixteca where I attended several games (Figure 3.5), and is the only

127 formal court still left in the capital.

Figure 3.5: The patio or court in the polideportivo at the Tecnologíco. The patio is a 100 m field surrounded by high fences.

However, an official field is not necessary to play. The patio at San Antonio Arrazola, for example, is a cleared space located on terreno escolar, or school land, behind the basketball court and municipal buildings. The patio at Buena Vista is another cleared space that lies behind the house of Don Quique and next to his milpas of corn (Figure 3.6). At Santa Cruz de Milpa, the patio was next to a major four lane road on a patch of land between two buildings. To get there, I was told to get off the bus, turn to my right and look for the big tree, where I would find the field. The patio at Guadalupe Etla was far on the outskirts of town; to get there, I met with a

128 player nicknamed El Negro (the Black One) in the town square. From there he drove us to where the men were playing that day. Neither of these courts are bordered by fences like the official courts are and are in fact quite modest and informal in appearance.

Figure 3.6: The informal patios at San Antonio Arrazola (upper) and Buena Vista (lower).

129

On average the patio measures 100 meters by 10 meters and, like pelota tarasca, is divided into three zones: the zona del saque, which covers around three quarters of the field or about 70 meters; the zona del resto; and the cajón (which can be translated as drawer or bin), an

8-meter section of the zona del resto (Berger 2009; Turok 2000). Where the ball falls in the cajón determines which player can hit the ball on the opposing team. In the zona de saque there is the botadura, a flat, inclined stone against which a player also called the saque strikes the ball to send it into play (Figure 3.7) (Don Jóse Angel 2009: personal communication). Prior to the commencement of play, the field is carefully measured and outlined, with the length of the field negotiated between the team captains. At the games I observed, the men used chalk to mark the lines, although sometimes cords are used or lines are simply scratched into the ground (Berger

2009: 6). Each time it was one of the older players who did this, similar to the tradition in

Choctaw stickball or toli where older men prepared the field.

How much of the field is actually in play during the game depends on the force and strength of the saque (según la capacidad del saque), according to Don Jóse Angel (2009: personal communication). For example, when I attended games at Buena Vista, the patio had lines marking 10, 25 and 30 meters. Players are placed into the categories of las primeras, las segundas, and las teceras (firsts, seconds, and thirds) depending on how far they strike the ball, a decision made by the coime, or the custodian of the court, and the captains of the teams scheduled to play beforehand. Those that can hit the ball the farthest are the best players, and some can even hit the ball up to a distance of 80 meters. While game time is variable and can last for hours, matches between primera teams can finish play as quickly as within a half hour.

130

Figure 3.7: The botadora stone is used in pelota Mixteca to launch the play into play. Two have been set up in the patio at Buena Vista for a torneo where multiple teams would be playing.

Teams are composed of five players called quintas, and game play takes place between two quintas. Many of the men noted that teams were previously composed of only four players – why this change in the number of players occurred was not made clear, although one player claimed that during the Mexican Revolution (1910-1921), teams were smaller because there were not enough men to play. The team that plays in the zona de saque and serves the ball is referred to as the saque or contrarresto (counter), while the opposing team is called the resto 131

(Figure 3.8) (Don Jóse Angel 2009: personal communication). The saque begins play by bouncing the rubber ball against the botadura and striking it across the court to the resto. The players on the responding team return the ball, keeping it in the air and not letting it hit the ground. Play continues this way until one team fails to return the ball, at which point the resto gains a tanto or point. If there are many teams playing, such as during a torneo, then the quintas will rotate in and out of the playing field when their ball goes out of bounds. This means that you can have two to three sets of quintas playing practically at the same time (cuando una sale, hay otra que entra, when one leaves, another enters). At some of the more well-attended matches I went to, there were so many quintas playing that two botaduras had been set up and one saque would launch his ball seconds after the other did (see Figure 3.7).

Figure 3.8: The saque (left) and resto (right) teams playing at Buena Vista.

The point system is similar in many respects to that of tennis. Points awarded to the quintas are counted as quinces (fifteens) using the pattern of 15, 30, and 40. Once one team has achieved 40 points, then a set or juego (game) is declared. The first team to win three sets out of five wins the partido or match (Don Jórge, el que ganó la ventaja, he who wins the advantage).

Many of the men acknowledged this similarity, but they were also quick to point out that this

132 was the only resemblance between the two sports and that pelota Mixteca is muy distinto (very distinct). One of the more complicated ways to win a juego is by winning rayas (a different use of the term from ulama de cadera and other ballgames). As Don Pablo described it, si sale el bote, donde cruza es una raya (if the ball leaves [the court], where it crosses is a line); that is, a raya or line is marked by the chacero or referee at that spot where the ball has gone out of bounds after bouncing once in the playing field, or where it has bounced twice in the field but remains in bounds (Berger 2009: 8) (Figure 3.9 and 3.10). A raya does not give a point to the team that made it; instead, if a second raya occurs, then the two teams switch sides, with the team playing the position of saque changing to resto.

Figure 3.9: The chacero or referee observes game play during a match at Guadalupe Etla.

133

Figure 3.10: A chacero marking a raya where a ball has bounced out of bounds.

According to several players, si no hay quinces y hay dos rayas, se cambian, pero si no hay rayas, no se cambian – if there are no points, and there are two lines, they change, but if there are no lines, they do not change. Once the serving team has won their set, even if there is only one raya marked they must still switch sides – as Don Pablo explained it, lo que hace es hacer rayas para cambiarse...se ven acá, hay una raya y quince, trienta, cuarenta y si hay una raya, se cambia, what they do is they make lines to change, they come here, there is a line and fifteen, thirty, forty, and if there is a line, they change. Otherwise, if there are no rayas marked and neither team has won the set, they continue to play in their positions. Marking the rayas also serves to reduce the playing field, making it more difficult for the resto team to score. Should

134 the resto team hit the ball past the raya, then the tanto is awarded to them. Failure to do so awards the point to the other team (Berger 2009: 7-8), thus the goal of the saque is to block the ball from going past the raya.

Every player that I spoke to stressed the importance of knowing and understanding the rules and how the sport is played. Don Jóse Angel described the reglamiento or rule system as being muy sagrado, very sacred, and for Don Claudio the rules have never changed, son las mismas…siempre esta contabilidad, they are the same…always this counting system. Don

Catarino, who claimed to have rules from 1901 and 1966, also asserted that they have not changed. The rules suggest to some scholars that land disputes may have been resolved through the ballgame and that people often waged land when betting on game outcomes (Pohl et al.

1997:212). The chaceros keep track of the point scoring, and when there are multiple teams playing a chacero is assigned to specific quintas. They keep score in a variety of ways, including making small rips in leaves, scratching the points into the dirt using the bamboo stick, or using what I can only describe as point stones (Figure 3.11). Sometimes the chaceros use a combination of all three.

According to Don Jorge, the coime fulfills other important duties related to the game.

They keep track of the quintas that are playing that day and must be familiar with rules as él define si es buena o es mala la pelota, he decides if the ball is good or bad. Furthermore, if there is a dispute over a point or rule, the coime tiene encargado si hay problema, he is in charge if there is a problem. As both the patron of the San Antonio Arrazola ballplayers and the coime of the field there, Don Jóse Angel pays for its use and maintenance. The coime also receives some of the money made through bets (Turok 2000: 64), in most cases about ten percent of all bets made (Berger 2009: 5). Anyone can be a coime and own a court, but there is a lot of

135 responsibility attached to the position. They are expected to provide food and refreshments for purchase for the players and spectators. The coime must also ensure that the court is regulated and approved by the authorities (cada patio debe estar reglamentado, every field should be regulated). Because drinking beer and mezcal is common at games, fights can and sometimes do start between players. If the police are called and the court is not legally approved, the coime and the players could go to jail.

Figure 3.11: Chaceros also keep track of each quinta's score. Score marks from a game at the Tecnologico (upper) and a chacero marks the score during a game at Santa Cruz Amilpas (lower).

136

Men typically start learning how to play pelota Mixteca between the ages of eight to fourteen. Don Jóse Luis informed me that while they use a normal game ball, children are given smaller-sized gloves made especially for them. On a few occasions I observed young boys practicing by tossing a ball against a wall and striking it as it bounced. Boys will start using heavier gloves and participate in training matches as they grow older, forming teams with friends or joining their family’s team once they are ready for the real game (Berger 2009: 11).

Typically, an older player will teach a boy who is interested in learning. Most boys learn by watching older men play, sometimes even contracting with a good player to learn. When I asked the men who taught them to play, I received a wide variety of answers. Some, like Don Claudio and Don Catarino, learned from their fathers. Others like Don Elfino learned from their older friends (de un amigo…un día me dijo, ¡vamos a jugar a pelota!...from a friend, one day he told me, we’re going to play ball!). Don Jóse Angel indicated that it was actually pretty common for players to learn from their friends or from relatives other than their fathers, as his own father learned from his tíos or uncles.

On the other hand, Don Elfino mentioned how some fathers choose not to teach their sons because they consider the game to be dangerous. The ball, which again weighs around a kilogram, travels very fast when struck. A popular story among players concerns an incident that occurred sometime in the 1970s when a player was struck in the chest by a ball and died as a result of the impact. In between the 2009 and 2010 field seasons, a player I had interviewed was struck in the face by the ball and his jaw had been broken by the impact. Many players present injuries or changes in bone morphology from repeated strikes by the ball, especially on the playing hand (Figures 3.12). Over the course of years working in Oaxaca, I have met quite a few men whose fathers had been ballplayers but who had refused to teach them because they think

137 the game is too dangerous. Sometimes the sons themselves are simply not interested in learning

– Don Elfino lamented that he had five sons, enough for a quinta, but that they did not want to play.

Figure 3.12: Many players have lasting injuries from playing, especially on the glove hand.

Game matches are typically held on Sundays. Those that are not prearranged or agreed upon beforehand are referred to as partidos libres or free games, and teams are made of whoever shows up to play that day. If it is a pledged match or partido de compromiso, arrangements are made beforehand about where and when the game will take place. As Don Jóse Angel explained, there are two captains, hay dos capitanes, and they are responsible for discussing the game arrangements, los que unien, ellos son responsibles y hablan los capitanes (those that unite, they are responsible and the captains talk). A rather specific form of compromiso game is

138 the torneo. These are considered to be the most important matches and are usually held as part of the celebrations for a town’s fiesta patronal or Catholic saint feast day, as well as other special holidays. The tournaments tend to be the best attended, with players and teams coming from different towns and villages throughout the state to participate. The more important the tournament, the higher the number and quality of teams, and the more likely there are to be teams composed of players at the level of primera fuerza (first strength).

Placing your apuesto or bet is a critical component of pelota Mixteca matches today.

Don Claudio mentioned that the game was not just for diversión or fun but also for gambling, and that players and spectators must do so. For smaller matches or partidos libres, smaller amounts of money are wagered, but at larger games like the torneos the men will bet anywhere from $2,000 up to $5,000 MX pesos ($108 up to $250 USD in 2010) on a single game. Making good on your bet is referred to as winning la ventaja, the advantage. However, men also lose a lot of money on these games. Don Catarino explained that sometimes they lose everything through betting, while Don Pablo laughed and added that this is why their families prefer that they not play too often. When a game de compromiso is arranged between team captains and a date set, many times the captains will place bets beforehand and put up substantial sums of money or valuable objects as the prize. This is also to ensure that the teams will show up to play.

In their stories about the game’s origin, many of the players I spoke with stressed its importance in maintaining community relationships not only in the past, se usó para unir los pueblos (it was used to unite the people/villages) but today as well. In fact, Don Jorge said directly that the games were to maintain amistades, friendships, between people from different and often widely dispersed communities. At all of the games I attended, it was clear that players and spectators were coming from all over the state of Oaxaca. Don Jóse Luis for example, who

139 was a frequent attendee and occasional player at the Tecníco on Sundays, traveled down from

San Pablo Huitzo, a town about an hour away from Oaxaca de Juárez. Some were coming from even further. Don Jorge had actually traveled down from the Mixteca Alta to attend the torneo at the Tecníco on July 19, having come at the invitation of friends just to watch. El Negro, who I met at the Tecníco, invited me to games at Guadalupe Etla in the Etla Valley and to Magdalena

Jaltepec, tournaments that he had been told about by players he knew. In the interviews many of the men pointed out to me what pueblos the quintas were coming from. As Don Claudio described it, viene de distantes lugares, they come from distant places (Figure 3.13).

Figure 3.13: Score board for the day's torneo at Guadalupe Etla, showing the different towns the teams are coming from.

There are certain expectations – tiene sus reglas, it has its rules – regarding community participation in playing pelota Mixteca. Don Jorge said that when a team is invited to play “de compromiso”, they are obligated to go. He likened the compromiso games to the custom of guelaguetza, a Prehispanic Zapotec tradition of social obligatory reciprocity that is also linked to

140 traditional corn festivals and harvest celebrations, at least in the Central Valleys. He explained it this way, si me das, te doy, if you give to me, I give to you. When there is pelota tournament in a pueblo, the local team will invite teams and other people, including friends and relatives, from other villages. As an invited guest you are expected to come, and in turn you can expect to be hosted and feed, este es el respeto a vistante, this is the respect paid to the visitor, according to

Don Jóse Angel. A further expectation is that you will return the favor and issue your own invitation to come and play pelota Mixteca in your pueblo. Don Jóse Angel described it as amor con amor se paga, you pay love with love. If you refuse an invitation to compete, then no one will come to your field to play. In his description of the pelota Mixteca as being like guelaguetza, Don Jorge expounded further by saying that para vivir y para compartir, el juego es parte de esta tradición, for living and sharing, the game is a part of this tradition.

Ballgame matches, especially the torneos, are still used to celebrate important religious or political events. The torneo I attended in Arrazola, for example, was to celebrate that town’s patron saint, San Antonio, as was the game match played in Magdalena Jaltepec and Santa Cruz

Amilpas (which took place over two days). Aside from these patron saint feast days, games are often played to celebrate weddings or for political party events (para celebrar bodas…partidos politicos). For these events, Don Jorge said that municipal presidents will offer a prize for the winners. Although Sunday is the day when game matches are usually played, Don Claudio explained that for these special fiesta days, players will come to play on whichever day the celebration falls on. This means that they take time off of work to travel to towns where they have been invited to play. Players can spend up to half their year traveling on pelota Mixteca match circuits (Taladorie n.d.).

Many authors have commented on the similarity between pelota Mixteca and several

141 handball games from Spain, including boce lucea and the game called de largo y rebote, (Turok

2000: 65). Berger makes an interesting argument that the antecedent to pelota Mixteca, pelota a mano fría, may have been introduced by Spanish friars or monks sometime during the 16th through 18th centuries. He proposes that after the Spanish forbid playing ullamaliztli, the indigenous population may have adopted European handball games as a substitute (2009: 55-57).

He rightly cautions, however, that this does not mean that pelota Mixteca is not a traditional

Mixtec and Zapotec game. Obviously, the balls and gloves used in the game are local developments. However, we do have evidence in support of a Prehispanic handball game in the indigenous dictionaries recorded in the Colonial period (Taladoire 2003). In Cordova’s (1942

[1578]) Vocabulario Castellano Zapoteco, there are several indigenous terms for different ballgames. The hipball game (Jugar a la pelota de los yndios con las nalgas [ballgame of the

Indians with the hips]) is listed as tiquija láchi, cotija, qquiquijaya, and i tiquijaya. On the other hand, there is another game that is specifically referred to as a handball game (Jugar à la pelota nuestra con la mano [our ballgame with the hand]) and is known as tigàapayapitipi, pelólo

(perhaps a Zapotec approximation of pelota), and cotàapaya. Moreover, we see the same distinction in Mixtec (yocotenidinama [hipball]; dizen de boleo [handball]), (Alvarado 1962

[1593]); Tarascan (apantzequa chanaqua and apantzeti [handball]; taranduqua chanaqua and taranduni [hipball]) (Gilberti 1901); and even in Nahuatl, where handball was known as nematotopeuiliztli (Molina 1977 [1880]).

Interestingly enough, the men I spoke to explain the origin of the game in a different way.

According to the description of the game provided to me by the Asociación Oaxaqueña de

Pelota Mixteca, it started sometime either before the 14th or 15th centuries, originating in the

Mixteca region most likely during the florescence of the Postclassic sites of Tilantongo,

142

Yanhuitlan and Tututepec (Mixteca Alta and Costa, respectively). Don José Luis claimed that it was a sacerdote or priest who created the rules. He also mentioned that in the cancha chiquita

(small court) at Monte Albán hay un circúlo con perforación este es la pelota Mixteca original, there is a stone circle with a perforation in it (possibly the botadora), and that this is the original

Mixtec ballgame. Don José Angel told me the same story but elaborated further, explaining that in the 14th and 15th centuries there were many kings and mucho oro, mucha riquesa, a lot of gold, a lot of riches. This caused many problems among the people. When they began to die from fighting, they invented pelota Mixteca para evitar guerra, to avoid war. The game was then brought down from the Mixteca Alta to the Baja and Costa, and once the communities had been united there it was given to the Zapotecs. Don Gustavio, who claimed to not know the origin of the game, still stressed that it is a long tradition con miles de años, with thousands of years, passed down from the ancestors to the players today. He even went as far to say that it was invented muy antes del Cristo, long before Christ, and that any changes that have occurred only happened after the Conquest. Don Elfin Trujillo also claimed that pelota Mixteca had been played before the Spanish arrived and that it had originated in the Mixteca. The most interesting origin story was related to me by an old campeón or champion, Don Catarino Peréz, also known as El Oficial (The Official). According to Don Catarino, when the Spanish arrived they asked who had invented this game. They were told by informants that it was a woman named

Anagalla, and that she played with a tiny ball. After Anagalla had invented the game, everyone began to play it.

The players I interviewed insisted strongly that they played pelota Mixteca because according to them “it was a tradition” to do so. For Don Gustavio, playing pelota Mixteca was toda la tradición de la familia, a tradition of the whole family, a sentiment echoed by many

143 players. As one of two young brothers, known by their collective apodo as Las Ranas (The

Frogs), explained to me:

“Yo juego porque a mi el juego me llama la atención más que el fut [futbol] o el béisbol…un día pasé y ví un juego que nunca nunca lo ví…y me llamó la atención porque jugaban con guantes y me lo pareció bien diferente, me dirigí a la cancha para platicar con los señores, me lo explicaron el juego…y cuando aprendí que es tradición de Oaxaca…pues tradición de nuestra gente, me dio ganas de aprender y jugar…”

I play because [pelota Mixteca] is more interesting to me than [soccer] or baseball…one day I was passing by and I saw a game I had never ever seen before…and it interested me because they were playing with gloves and it looked so different, I went over to the playing field to talk with the men, they explained the game to me…and when I found out that it was a tradition of Oaxaca…well [a] tradition of our people, I really wanted to learn and play… Las Ranas, Sta. Cruz Milpa 07/09/09

Don Claudio stressed that Oaxacans have to take care of the game to ensure its survival, because it is nuestra tradición, our tradition. Thus, while it may not formally resemble or relate to a

Prehispanic game, it fulfills an important sociopolitical role in indigenous and mestizo communities and arguably represents a syncretic continuation of a Prehispanic practice (Berger

2009:57).

My ethnographic research focused on the Central Valleys and the Mixteca Alta, primarily in the Nochixtlán Valley. But pelota Mixteca is still also actively played on the Costa Chica

(Taladoire n.d.). As in the Central Valleys and Mixteca Alta, games are played on Sundays, in fact el domingo tenemos compromiso (Sunday we have an engagement) is a common saying.

The playing field is usually located outside the town limits, in some cases close to the terreno escolar such as in the communities of Corralero, El Ciruelo, and Collantes. Although not every community has a ballcourt, they will still have a team, (e.g. Mancuernas, Lagunillas, or Lo de

Candela [Taladoire n.d.:14]), a common occurrence elsewhere in Oaxaca where pelota is played.

144

Wagering also frequently occurs at games on the coast. Prize money at tournaments is gathered by municipal authorities and festival organizers, who also make money from the sales of beverages and food. Again, providing food and drink to guests is required, especially to those who regularly play on Saturdays and Sundays and in the pledged games (Taladoire n.d.:14-16).

Pelota Mixteca on the Costa Chica is exactly the same as elsewhere but has a racial element to it not quite present in other areas of Oaxaca. Here the game is strictly played by Afro-Oaxaqueños

(descendants of escaped slaves who settled on the Pacific Coast) and indigenous peoples

(Chatinos and Mixtecos); mestizos simply do not play. The Afro-Oaxaqueño communities play with each other and neighboring indigenous communities.

Outside of the game, the relationship between the three groups is characterized by mutual discrimination and rejection that has its roots in colonial systems of structural and social racism

(Jackson 1999; Lewis 2003). Both the mestizos and Afro-Oaxaqueños view using an indigenous language as a regressive act (Taladoire n.d.:6), marking an individual as a social inferior.

Despite the negative opinions of indigenous culture and lifestyle, the Afro-Oaxaqueños treat their participation in pelota Mixteca matches with their indigenous neighbors with the utmost seriousness. As Taladoire (n.d.18) observes, the “process of racialization that characterizes

[these] forms of interaction takes on a distinct role” through the medium of the game, where participation carries a “strong sense of honor, prestige, and…compromise and reciprocity”

(translated from Spanish). The integration of otherwise separate ethnolinguistic groups via pelota Mixteca stands in contrast to the Fourth of July celebration described by Culin, where sporting events where an occasion for mutual celebration among Mexicans, Navajos, and Euro-

Americans, but each group mostly participated in separate activities. In this way, we easily see the complex role of sport in identity production.

145

3.4.3. Discussion

While many of the modern indigenous ballgames have lost their overtly religious or ritual elements, there are certain activities that are still performed with regularity and solemnity. This is obvious even to outside observers such as a bewildered graduate student watching a pelota

Mixteca game for the first time. The marking of the court lines with chalk is performed with all seriousness by a senior official or member of the community where the game is held. During particularly important games or tournaments, special markings are added – sometimes the town name or even Bienvenidos (welcome) (Figure 3.14). It is especially during these games that a blessing is given by the coime prior to the commencement of game activities, wishing good luck to all the players (Stoll 2010).

Figure 3.14: Design marked in chalk with the name of the town "Jaltepec", a ball, and the word "Bienbenidos" (bienvenidos or welcome) on the patio. 146

Gambling can also be seen as one of these ritualized (but not necessarily non-religious) behaviors associated with playing. As mentioned previously, players are required to place bets if they want to participate in a pelota Mixteca game, even for the partidos libres. Meanwhile, a quinta that does not show up to play at a previously arranged compromiso game automatically forfeits the winnings to the hosting team. Betting was once part of the ritual activities involved in the traditional indigenous games and sports in the historic period, and we know from the

Spanish chroniclers that even during ceremonial ballgames, gambling occurred frequently (Hill

1999: 30). Players, spectators, and hosts would bet large sums of money, goods, slaves, and even land and family members; those who found themselves in debt would sell themselves or their kin into slavery to the person they owed (Durán 1971:313-318; Evans 2017; Cameron and

Johansson 2017). In the United States of America, indigenous games and sports were often targeted by missionaries because the “evil vice” of gambling was a frequent activity.

Politics are not entirely decoupled from pelota Mixteca and the organization of game matches today. The state government has a general interest in the ballgame through its regulation of the courts and the monetary support it gives for certain tournaments and to the ballgame associations in general. The federal government also sponsors special tournaments through FDAT, including an annual pelota Mixteca event that occurs in San Diego every

September between players from Oaxaca and players from California. However, many of the men still expressed frustration at the lack of apoyo or support in the way of money or attention from the local Oaxacan government. Pelota Mixteca matches take place during the annual

Guelaguetza celebrations in July as part of the local celebrations, but they are never featured in the tourist literature or in advertisements for the festival – a missed opportunity for sure. Don

147

Pablo even wanted to arrange an interview between a local paper and myself in 2009, expressing the hope that if it appears that foreign scholars were interested in studying the pelota Mixteca, that this might spark the government’s attention and interest, and the Asociación Oaxaqueña de

Pelota Mixteca would receive more becas or grants.

At the same time, there are also political divisions among the ballplayers themselves.

Don Jóse Luis often complained that most players only knew the rules or “reglamiento” of pelota

Mixteca and not its true history. During a game on July 19, 2009 at the Técnico, I was told by both Don Jorge and Don Catarino that there was going to be a gran torneo (big tournament) there the following Sunday to celebrate Los Lunes del Cerro (Mondays on the Hill), another name for the Guelaguetza Festival. When I arrived there on July 26, I was surprised to see so few players. In the last interview I conducted that day with Don Jóse Luis, what started out as a conversation about the “true” history and origins of the game quickly turned to an expression of anger and frustration by him about the political issues that were affecting the players and the organization. Apparently, two torneos had been arranged for the same day – a not uncommon occurrence. However, it appears that because prize money was being offered at the torneo in

Santa Cruz Amilpas, many of the players went there instead. By doing so, many broke the compromiso to play at the Tecníco. Don Jóse Luis declared quite angrily that those playing at the Tecníco were there por el gusto, for the love the game, while those at Santa Cruz de Milpa were there only for the promised prize money.

Anyone who has spent some time in Mexico, particularly in the rural and/or majority indigenous areas, will have witnessed firsthand the community-level organization of work- parties for local construction or agricultural projects. These communal labor projects often involve all members without any initiation, interference, or influence from the state or federal

148 governments, and are known as tequeo. The organization of work parties made up of community members are common throughout small villages and rural towns in the Americas (Lau 2002).

Commensal politics also figure largely in the fiesta system in which many indigenous and rural communities participate. In the Mixteca Alta for example, households participate in a cycle of hosting and attending fiestas involving three different kinds of exchange: the reciprocal exchange of wealth by fiesta participants; the pooling of wealth at a designated center, such as with the municipal president or another household; and the distributions made by fiesta sponsors to guests. These cyclical exchanges of goods and wealth link fiestas together in such a way that each event cannot be treated singularly, as households participate at different levels depending on their debts and credits within the system (Monaghan 1990: 58-62). The complex system of cyclical exchanges and community obligations, known by its Mixtec name of saa sa’a, thereby creates a multitude of sociopolitical meanings for each fiesta.

While not technically part of the saa sa’a system, I argue that pelota Mixteca plays a similarly important role in the community-building practices of local indigenous and mestizo campesino communities, in which commensality, game-playing, gambling, and reciprocity all form part of the social system, more so in centuries prior. Pelota Mixteca fits within a larger moral framework operating within many rural indigenous and mestizo communities today, which rests on particular notions of how individuals should interact with each other as members of the same community via reciprocal obligation. That is, an individual is expected to help out with communal tasks and honor their debts, and they can in turn expect help in the future and have the debts owed to them paid in kind.

Because pelota Mixteca players travel frequently in order to fulfill their obligations to play, they establish various social networks at different scales via these game circuits. As stated

149 previously, communities that extend invitations to other teams can expect to receive invitations and be hosted in turn. By refusing invitations to play by other communities, or by failing to show up to a game, an individual or team could lose their calidad moral or moral quality (Stoll

2015). These invitations to play at tournaments and festivities mean that players frequently travel to neighboring communities, and the resulting social networks can be quite extensive.

Afro-Oaxaqueños on the Costa Chica, for example, regularly attend games in Mixteca Alta towns such as Juxtlahuaca, Huajuapan, Tecomaxtlahuaca, and Putla principally; neighboring localities such as in the Pinotepa Nacional, the Distrito de Jamiltepec; and even all the way to

Acapulco in Guerrero (Taladoire n.d.15-16).

In many of the places where I have worked in Oaxaca, I have spoken with men who remember when they or their fathers used to play pelota Mixteca, and how they would travel to other communities to play. Before cars, one remarked to me, people would travel on foot or by donkey, sometimes traveling up to three weeks to get to a game. While mapping the ballcourt at

Ejido de los Canseco in Nejapa, I discussed my research on ballgames and the pelota Mixteca with members of the ejido. They told me about how they used to play in the field near where the ejido offices are located in El Camarón, and that the team would travel to play in nearby towns such as El Gramal (7.5 km by road), Soledad Salinas (14.8 km by road), San Pedro Totolapan

(30 km by road), and even as far as Santa Maria Zoquitlán (71 km by road). Thus, whatever its origins – and the evidence is strong that it is likely a syncretic sport or at least a Prehispanic tradition with European elements grafted on – pelota Mixteca “generates pleasure, emotions, happiness” (Taladoire n.d.20), and above all relationships that are not only regional and state- wide, but also national and even international.

Perhaps one of the most striking similarities between pelota Mixteca and other

150 indigenous sports I have discovered in my research is the use of nicknames. Many of the players

I spoke with had apodos, earned through their participation in game matches. I met men who were called “Lento” (Slow), “El Diablo” (The Devil), “El Caballero” (The Horse), and “No

Gano” (I Don’t Win), among many others. In fact, many of the men only knew one another by their apodos – if I asked about the whereabouts of a player on game day, I would have to use their nickname, “Dónde está el Diablo hoy?” (Where is the Devil today?). Nicknames were also popular among the historical Choctaw when they played toli. The name pałki was given to the fastest players, while slower players were known as saláha wašoha. Fans also gave nicknames such as čanáša (moccasin snake), sįti (snake), and opa niškin (owl or owl eyes) to notable players who stood out as exceptional athletes (Blanchard 1996: 37).

3.5 THE ARCHAEOLOGICAL DATA ON THE MESOAMERICAN BALLGAME

The majority of research into the ballgame has largely focused on three macroregions:

Central Mexico, the Gulf Coast, and Maya area. Often the evidence from these distinct regions and different time periods are conflated together in interpretations of the sociopolitical significance of the ballcourts and the game. Meanwhile, investigation of this phenomenon in other areas of Mesoamerica has been limited. Because of this, local and regional variation through time are unfortunately ignored or simply not considered, leaving unexplored the significance that these differences and similarities might have had for intra- and intercommunity social networks mediated through the ballgame. Reexamination of the iconographic and material culture – including the architecture, objects and symbols associated with game play and related activities – can push our analysis towards a greater understanding of this diverse tradition as encompassing active and lived “communities” of practices.

In the following section, I review the material and historical evidence related to the

151 ballgame and ballcourts in the different macroregions. For this survey I rely on several sources, primarily Stephen de Borhegyi’s (1980) excellent synthesis of regional ballgame data. I then move to a more intensive look at the archaeological record of Oaxaca and how scholars have interpreted the ballgame here specifically. The evidence, I argue, demonstrates that instead of a single, unified “Mesoamerican ballgame”, the different combinations of practices, symbols and ideologies represent multiple complementary, competing, and sometimes even overlapping local and regional ballgame traditions.

3.5.1 A Note on Ballcourt Terminology

While there is great deal of diversity in ballcourt design, there also exists a standardized body plan that makes them immediately recognizable during survey and excavation. Over the past decades, a specific terminology has been developed to describe both the courts themselves and the architectural features that mark them as a specific structure type.

Ballcourt shape: Mesoamerican courts predominantly have four main forms. The individual features (described below) of each court will vary by region and time period (Figure

3.15).

 Open courts –the earliest court form, consisting of two parallel lateral structures with

only a central playing alley and no defined terminal end zones.

 T-shaped courts – courts of this form feature two parallel lateral structures and one

terminal or axial structure that creates a single end zone. The court overall has the

appearance of a capital T.

 I-shaped courts – these courts have two parallel lateral structures and two terminal

structures, a central playing alley, and two defined end zones. This is the most common

152

and popular ballcourt form.

o Enclosed – I-shaped courts where the terminal structures connect to the lateral

structures.

o Open – I-shaped courts where the terminal structures do not connect to the lateral

structures.

 Palangana – distinctly shaped courts that have all four mounds but lack defined end

zones. The four mounds meet at their corners, creating a rectangular structure.

 Finally, sunken courts are structures where the central playing alleys and end zones are

much lower in elevation than the surrounding ground surface.

Ballcourt features: Although a ballcourt does not have to have all of these features, these are the common characteristics of court design (Figure 3.16).

 Lateral structures – two parallel mound structures that are longer than they are wide,

usually rectangular in shape. Sometimes there is only a single rectangular mound, with

another structure, such as a temple or platform, bordering the other side of the court.

o Apron – the interior sloping surface of the lateral structures and sometimes the

terminal structures. The degree of slope can vary greatly between regions and

even time periods. The court aprons of some lateral structures do not slope at all

and instead are vertical (i.e. Chichen Itza).

o Upper wall or cornice – a smaller platform or lateral element on top of the larger

lateral structure.

o Backwall – the exterior wall of the cornice.

o Bench and bench wall – a feature that projects from the interior of the lateral

153

structures, usually very low and close to the playing alley surface.

 Terminal or axial structures – many courts have one or two terminal structures that

mark the end zones of the structure.

o Terminal structures can either be a parallel, rectangular mound or a U-shaped

mound with two lateral arms extending from the mound. Courts with attached U-

shaped terminal structures are called enclosed.

o The interior of the terminal structure can be vertical, referred to as end walls, or

sloped with an apron like the lateral walls.

o Sometimes, a temple or platform borders the terminal end zone, effectively

replacing the terminal or axial structure.

 Playing Alleys – the flat interior of the court is the playing surface. All courts will have

a central playing alley. Depending on the shape of the court, there may also be one or

two end zones or terminal playing alleys where gameplay would have taken place.

154

Figure 3.15: The four general ballcourt forms that have been documented so far in Mesoamerica: a) open; b) T-shaped; the I-shaped court which can be c) open or d) enclosed; and e) palangana or washbasin type.

155

Figure 3.16: Features typically documented at ballcourts, although these features vary and may not be present in some regions and/or during certain time periods.

3.5.2 A Regional Overview of Ballgame and Ballcourt Material Evidence

Archaeological evidence for ballgames in Mesoamerica (Figure 3.17) is found everywhere and includes murals; figurines; stone sculptures such as yugos (yokes) or horseshoe- shaped waist protectors, hachas, the thin, axe-shaped figures attached to the yoke, and palmas or stylized leg protectors, which replaced the hachas in Central Veracruz sometime in the Late

Classic period; and pottery vessels among other items. Ballcourts themselves are also significant sources of data on the ballgame. Other evidence is found in the surviving Prehispanic and

Colonial-era codices that were produced shortly after the Conquest, such as the Codices Borgia,

Dresden, Bodley and Nuttall. In addition to these sources, scholars also have access to Colonial and ethnohistorical documents written by Spanish priests and native informants. Spanish friars

156 such as Fray Bernardino Sahagún (1970), Fray Juan de Torquemada (1975), Fray Francisco

Javier Calvijero (1945 [1780]), and Fray Diego Durán (1994), to name but a few, recorded important ethnographic information on the ballgame, and even wrote first-hand witness accounts of the games they witnessed.

Figure 3.17: Map of Mesoamerica showing the different geocultural macroregions.

157

3.5.2.1 Central Mexico

Figure 3.18: The Central Mexican Zone includes the Federal District as well as the states of Mexico, Tlaxcala, Puebla, Morelos, Hidalgo, Queretaro, and San Luis Potosi.

The ethnohistorical sources on ballcourts and ballgames are extensive for Central Mexico

(Figure 3.18). The Colonial-era authors who wrote about the ballgames observed them in this region, and other authors detail Nahua myths and beliefs about the games. According to the tribute lists recorded in the Codex Mendoza, over 16,000 rubber balls were imported into

Tenochtitlán every year (Santley et al. 1991:5), although it should be noted that not every rubber ball was used for the ballgame (Taladoire 2015b). In these historical accounts, there are descriptions of priests and other ritual specialists performing pre- and post-game rituals, Aztec

158 kings and lords placing very large bets on game matches, and spectators cheering on matches and making bets themselves. Several authors mention that native lords would organize and host games involving other communities – similar to how game matches are organized today – while some even fielded their own professional teams (Berger 2009; Durán 1994; Macazaga Ordoño

1982).

The archaeological evidence, on the other hand, is somewhat limited in Central Mexico due to the extensive urban sprawl and population density. Preclassic clay figurines clearly show that the game was played here at an early date. They also reveal what ballplayers wore, and the different costumes are often regionally specific and may link back to different ballgame traditions (Ramos 2012). Some Central Mexican male figurines wear turbans, rounded hats or helmets, thick protective bands on their arms, wrists or hands, and knees, and finally a wide belt on the hips, while others wear buccal face masks (de Borgeyhi 1980:8; Ramos 2012:61-62).

Figurines also provide the only archaeological evidence for female ballplayers. At the Preclassic site of Tlatilco, excavators found many female ballplayer figurines, the majority of which had been buried with skeletons (Ramos 2012:46). Female ballplayer figures from the site of

Tlapacoya wear only hip garments, while the male outfit is more elaborate and may perhaps point to a complex role in ballgame ceremony and ritual. Meanwhile, vessels from Teotihuacan show Classic-period ballplayers wearing similar gear, with the addition of yokes and zoomorphic hachas (de Borhegyi 1980:6-7).

Courts are difficult to document in the Basin of Mexico due to the extensive modern urban construction from Mexico City (de Borhegyi 1980:8; Santley et al. 1991:10). In northern

Central Mexico, the few I-shaped ballcourts documented here date primarily to the Postclassic period. What we know of the ballcourts in the infamous Aztec-Mexica altepetl of Tenochtitlan

159 comes from the colonial documents, since these structures were destroyed shortly after the

Conquest (although one of these courts was recently discovered during underground excavations in the Centro Histórico district near the Templo Mayor). According to these sources, the Mexica played on masonry courts alternatively called tlachcoi, tlachtli, teotlacho, and ollamaloyan.

These courts were I-shaped with rings and benches for spectators, and measured on average 12 –

15 m long and 6 – 9 wide m between the court sidewalls (Ramos 2012:135). Interestingly, outside of the main ceremonial precinct there were two courts located in the marketplace, and a lesser court in the suburbs (Motolina 1971:380; Stern 1948:51; Sahagún 1970). Going further back in time, there is still much debate over whether there were ballcourts at Teotihuacan, and many scholars have argued that ballcourt construction activity ceased when this altepetl was dominate in the Classic period (de Borhegyi 1980; Santley et al. 1991). Some have proposed that games may have been played on the Avenue of the Dead (Taladoire 2001; Marcus Winter, personal communication), while others have argued otherwise and have even gone as far to say that the ballgame fell out of favor at this time (Santley et. al 1991). However, murals found in the Tepantitla apartment complex show many ballplayers in a wide-range of dress playing different types of ballgames and using various parts of the body as well as pallets and sticks to play (Ramos 2012; Uriarte 2006). Future excavations may yet locate Teotihuacan’s ballcourt.

Taladoire (2009) suggests that a structure found in the Ciudadela compound could possibly be a ballcourt, but confirmation awaits further investigation.

Many more ballcourts dating to the Late Preclassic (Terminal Formative) and Classic periods have been found in southern Central Mexico and the Tehuacán Valley where modern settlement is less dense (Santley et al. 1991:10-12). One such place is the incredible site of

Cantona, located in the Western Basin of Puebla near the border with Veracruz. The site boasts

160 twenty-seven ballcourts (Taladoire 2000), the majority (n = 19) of which are located in the

Acropolis, or civic-ceremonial zone, in the Southern Unit area of the site, while five are found in the Central Unit and three in the Northern Unit (Rivera 2017:46). Except for two palanganas, the are I-shaped courts that consist of two lateral mound structures with benches, sloped aprons, and upper cornices with variable superior elements as well as terminal end walls (Rivera 2017). Half of the courts are found in a specific architectural arrangement known as the Cantona Type

Ballcourt Complex and is composed of three elements: a pyramid or stepped-platform structure;

1 to 2 plazas that are bordered by elongated platforms that sometimes also have other superimposed structures on top of them; and the ballcourt itself, usually located at the other extreme from the pyramid (Cook and Carrión 1998).

Other courts are found alone or associated with a single platform, pyramid, or structure foundations. The isolated courts are located on the periphery of the settlement in association with habitational patios and open spaces, and not near any structures that would suggest civic- ceremonial activities. Likely they served more recreational functions. Courts in association with a structure, on the other hand, were probably linked to administrative, economic, and/or ceremonial activities but not to the same degree as the those constructed in more formal arrangements (Rivera 2017:47). The construction history of the ballcourts at Cantona reveals how changes in both their location and arrangement is connected to changes in the sociopolitical relationships among the social groups living in the settlement. Interestingly, the style and form of the Cantona ballcourts are very similar to courts identified not only at El Tajin but also the

Central Valleys of Oaxaca, and roughly date to the same time periods, suggesting an interregional tradition of game playing (Rivera 2017: 51)

161

3.5.2.2 Gulf Coast

Figure 3.19: The Gulf Coast includes the Huaxteca, Veracruz, and Tabasco.

The archaeological and iconographic record related to the ballgame appears to be most dense in the Gulf Coast macroregion (Figure 3.19). In fact, the oldest rubber ball (dated to 1600

BCE) was found at El Manatí in Veracruz (Baron 2006:4dan). The greatest number and variety of solid and hollow ballplayer figurines come from this area and, like the figurines in Central

Mexico, they wear protective helmets and padding. Most of these figurines are male, although

Huastec figurines from the northern Gulf Coast depict both male and female players (Ramos

2012:9). In addition to the rubber ball, we know from figurines and stone sculptures that regional game equipment also included paddles, clubs, and bat-like objects. The stereotypical

162 body equipment of yokes, hachas, and palmas probably originated here, as the largest number of carved stone representations of these objects comes from this region (de Borhegyi 1980:8).

Formal masonry ballcourts were constructed in this region as early as the Middle

Preclassic (1000 – 300 BCE). A palangana-style or washbasin-shaped court dating to 500 – 400

BCE was documented at the site of San Lorenzo Tenochtitlan. The Epiclassic (CE 600 – 1200) period site of El Tajín is perhaps the best representative of the ritualized ballgame in this macroregion, as the site features over 17 ball courts and multiple ballgame images (Aguero and

Daneels 2009; de Borhegyi 1980:11; Cohodas 1975; Koontz 2008). The archaeological and iconographic evidence points to the critical role games must have played in the local political system. Throughout the Gulf Coast, ballcourts are found in the primary and secondary centers of both centralized and heterarchical systems, demonstrating that their distribution varies widely.

In centralized systems, the courts located at primary centers likely drew the population in from the whole territory and concentrated them there, while the smaller courts of the heterarchical systems probably only mediated more localized relationships (Aguero and Daneels 2009:118).

The importance of the game on the Gulf coast is underscored by the fact that in the upper and lower Cotaxtla Basin, the Mixtequilla region, the Nautla Valley, and the Tlahuanapa Valley around El Tajin, no one lived more than one hour away from a court, and many were less than half an hour away (Aguero and Daneels 2009:121).

The iconography of the Gulf Coast is incredibly rich and diverse. A variety of different subjects are depicted on the yokes, hachas, and palmas, the carved wall panels, stelae, and other stone sculptures. The alligators, frogs, and butterflies described by Uriarte (2000) are present, as are a diverse range of other animals including jaguars, parrots, rabbits, monkeys, deer, dogs, and coyotes (de Borhegyi 1980:9). Severed heads, dismembered limbs, skeletons, and defleshed

163 skulls, commonly cited as evidence of the connection between ballgames and sacrifice, are frequently depicted, but so are dancers, warriors, dwarfs, and Olmecoid heads. Along with the deities linked to Venus, other gods are also included in the iconographic images, such as various fire-gods, jaguar gods, Xochiquetzal (Precious Flower), Xipe Totec (Our Lord the Flayed One), and long-nosed gods (de Borhegyi 1980; Daneels 1997, 2008; Zeitlin 1993).

Based on the associated iconography as well as the distribution of the ballcourts, many scholars view the Gulf Coast ballgame as a political strategy employed by elites to attract followers. The cultural epicenter for this regional tradition was the Central Veracruz region, primarily during the Classic period (Aguero and Daneels 2009). From the Protoclassic (100 BC-

AD 100) period on, the ballgame here involved high-cost paraphernalia, a distinctive scroll style in the art, and decapitation sacrifices of humans. Together, these activities along with the game itself were used by elites to achieve and maintain a popular following (Aguero and Daneels

2009:118). Given a low-risk environment where a dispersed population lived on self-sufficient estates, political elites had to make efforts to attract followers. Zeitlin (1993:123) suggests that the sacrificial ballgame and its ideological theme of supernatural death and rebirth was so widely recognized among contemporary Gulf Coast cultures that it served as central rite that unified diverse religious practices.

The archaeological evidence does seem to support the connection between ballgames, ruling elites, and sacrifice. At the sites of Cerro de las Mesas and Carrizal respectively, archaeologists encountered rulers who had been buried in tombs with stone yokes in the

Protoclassic. One yoke was plain while the other had been carved with a representation of the earth monster, which Daneels (1997) interprets as the earliest evidence of the Classic-era decapitation fertility ritual associated with the Gulf Coast game. The burial at Cerro de las

164

Mesas included a decapitated adult male interred with a yoke (Daneels 2008). Generally, the consensus among scholars is that decapitation rites were broadly related to fecundity as well as the growth of maize in particular via the death and rebirth of the maize god (Koontz 2008:332).

The decapitated human is often depicted artistically with drops of blood coming out from the neck wound, in a metaphorical depiction analogous to the way the silky hairs look as they emerge from the corn husk. This supports the general conclusion that the ballgame, at least in the Central Veracruz region, was linked to agricultural fertility through the decapitation rites held in association with the game.

165

3.5.2.3 The Maya Macroregion

Figure 3.20: The Maya Macroregion can be divided into four separate geological and ecological zones: the Northern Lowlands, the Central Lowlands, the Southern Highlands, and the Pacific Coastal Zone.

For the most part, how we understand the ballgame in Maya society has been greatly influenced by the work of Linda Schele (Baron 2006:5; Freidel et al. 1993; Schele and Freidel

1991). She argues that the game was an elite arena for reenacting sacred creation myths that would later be immortalized in the Popol Vuh. Other scholars have also linked the Classic period ballgame to Maya warfare and conflicts, wherein elite captives taken in battle were forced to play against the king of the altepetl that had captured them (Baron 2006: 6, citing Freidel et al.

1993; Miller 1989; Miller and Houston 1987). But what about the ballgame during the

166

Preclassic and Postclassic? Did it also only serve the interests of the elite in those time periods?

The Maya area is large, variable, and internally complex with regards to geography, ecology, and human cultures. While there is an overall shared culture that ties different Mayan groups together, there are still important local and regional variations. De Borhegyi in fact divides the

Maya macroregion into four separate regions: the Northern Lowlands, the Central Lowlands, the

Southern or Highland Maya zone, and the Pacific Coastal zone (Figure 3.20). I follow this scheme in my discussion of the archaeological and material data of the Maya ballgame(s).

3.5.2.3.1 Northern Lowlands

The Northern Maya Lowlands (the semi-arid and semi-tropical coastal and inland zones of Campeche, Yucatan and northern Quintana Roo) have long played a role in the larger discussion of the Mesoamerican ballgame. Indeed, the ballcourt most referenced in popular culture and imagination is the immensely sized and complexity of the Great Ball Court at

Chichen Itza, which is also associated with the powerful symbols of elite Maya ideology

(Kurjack et al. 1991). The images of death and violence carved into stone in and around the large court connect the game in our minds to human sacrifice. While this court is impressive, it is a singular case and the archaeological evidence from the Northern Lowlands reveals a much more complicated picture of the game.

Previously, the evidence for the ballgame in this area seemed to date primarily to the

Middle to Late Classic (AD 400-900), aside from the Terminal Classic-Postclassic site of

Chichen Itza. De Borhegyi (1980:11-12) described the ballcourts in this area as having vertical walls with horizontally tenoned, paired stone-rings (although the latter may actually be

Postclassic additions). However, Taladoire’s (1981; 2001; 2017) more recent research shows that the ballcourts actually come in several variable types. There are open-ended courts, semi-

167 enclosed courts with one or more simple or U-shaped axial (i.e. terminal) structures; and enclosed courts with large end zones. Benches, cornices, and aprons are present but optional.

Solid and hollow ballplayer figurines, remarkably similar to those from the Gulf Coast, wear elaborate headdresses, jewelry and body padding. Although these figures – and the ballplayers on sculpture – are depicted with yokes, hachas and palmas, stone effigies of these objects are rarely found in this area. It is more than likely that the body padding worn by players was instead made of perishable materials. I explore geographical and temporal issues with the yoke- hacha-palma complex later in the chapter.

A more complete picture of ballgame history in the Northern Lowlands has only recently emerged. Regional survey carried out by the Proyecto Costa Maya (PCM), under the direction of Anthony P. Andrews and Fernando Robles (2004), documented a network of Middle

Formative settlements with multiple examples of ballcourt architecture in Northwest Yucatan. A total of twenty-five sites (out of 116) documented featured a court (Anderson 2003). Edgar Rene

Medina’s (2005) review of the ballcourt corpus here demonstrated that despite some variability, we find all of the traditional attributes of court architecture, including benches, aprons, upper walls and defined end fields (after Taladoire 2001). Excavations at one of the ballcourts revealed a possible central marker in the playing alley that was made up of a series of small cut limestone blocks arranged in a circle (rather than the usual single stone piece). The presence of ballcourt markers had not been previously noted in this area, establishing material continuity with later versions of the ballgame in the larger Maya macroregion (Stoll and Anderson 2017).

In contrast to the popular image of grand ballcourts near soaring Mayan temples and impressive palaces, these Middle Formative courts were primarily found at small villages with a minimal corpus of architecture and elite structures, demonstrating little if any sociopolitical

168 stratification. A notable counterexample is the 67-ha site of Xtobo, one of the largest Formative settlements in the region with many traditional indicators of stratified sociopolitical organization, such as monumental architecture, elite residences, and sacbes or causeways (Anderson 2003,

2011; Stoll and Anderson 2017). The site's ballcourt was located just to the south of the main plaza. While it resembles courts typical to this region, at just under 12 meters long it is the smallest in the sample. Moreover, the small population and modestly-sized pyramidal architecture suggests that the site’s elites likely did not exert control over Northwest Yucatan’s ballcourt network, nor the other 100 additional Middle Formative sites in the subregion (Stoll and Anderson 2017: 223-225). Complicating our understanding of sociopolitical relationships at this time, Nancy Peniche May (2012) excavated what she argues to be a throne room at the site of Xaman Susula that is reminiscent of later Classic Maya presentation throne rooms.

Interestingly, though it is a much smaller site, the elites at Xaman Susula were able to construct a symbol of sociopolitical power that Xtobo’s rulers could not (Stoll and Anderson 2017:225). On the other hand, Xaman Susula did not have a ballcourt. This Middle Formative ballcourt network seems to stop the farther one moves away from the core zone surrounding Xtobo, suggesting that there existed a distinct regional tradition in Northwest Yucatan. What emerges is a complex landscape where sociopolitical power was likely heterarchical and mediated through the ballgame, at least during this time period.

3.5.2.3.2 Central Lowlands

At the time de Borgheyi’s study was published, not many ceramic figurines and effigy vessels had been registered in the Central Lowlands (the tropical rainforest inland and coastal zones of Campeche, Tabasco, Quintana Roo, northern Chiapas, Guatemala [El Petén], Belize, and western Honduras), in comparison to other Maya regions. Thus, our knowledge of

169 ballplayer attire and game equipment primarily comes from carved stone stelae, ballcourt stone panels, and drum-shaped floor makers. Based on these images, Central Lowland ballplayers were elaborately dressed in headdresses, protective eye googles, tasseled gloves, and heavy body padding, among other items. With respect to game attire, there appears to have been a more local tradition in Lubaantun during the Late Classic, as most figurine players wear visored helmets. Some individual figurines even sport a human head, perhaps a trophy, fastened to their belts. Otherwise, representations of severed and trophy heads are not common in the Central

Lowlands (de Borhegyi 1980:12). As is the case for the Northern Lowlands, only a few stone yokes and hachas have been documented here. With respect to the ballcourt themselves, archaeologists have observed both open-ended and enclosed stone slab courts, sloping aprons and benches, and at least three carved stone floor-markers placed in the center and at each end of the playing alley (de Borhegyi 1980:12-13).

Two recent regional studies by Baron (2006) and Lohse and colleagues (2004) offer a clearer picture of ballgame traditions in the subregion of Northwest Belize that adds nuance to our larger understanding of the ballgame in the Central Lowlands. Ballcourts in this subregion are usually formed of two parallel structures open at each end (Baron 2006), and most were constructed in the Late Classic during a period of regional florescence under La Milpa or perhaps

Tikal (Lohse et al. 2004). Lohse and colleagues (2004: 99) tie the appearance of these Classic courts to ongoing regional transformations, suggesting that the rulers of neighboring communities were engaging with each other through ballgame matches and related ceremonies that invoked broadly shared ritual and ideological beliefs. They view the use of the ballgame as part of a “service”-oriented political strategy that specifically linked these beliefs to the elites

(Lohse et al 2004: 99). Baron (2006) analyzed several qualitative and quantitative measurements

170 from ballcourts in the Belize River Valley, the Three Rivers Regions, and the site of Yalbac, which lies in a natural boundary between the two subregions. Her analysis reveals some statistically significant differences between the subregions with respect to ballcourt form, orientation, and probably even game play (Baron 2006:18-31). Excavations and radiocarbon dates show that the Yalbac ballcourt more closely resembled the Belize River Valley when it was initially constructed in the Late Preclassic. After ritual termination in the Early Classic, the court was reconstructed later in the Late Classic and modified to exhibit closer affiliation with the

Three Rivers Region (Baron 2006:35-42).

3.5.2.3.3 Southern or Highland Maya

The Southern or Highland Maya region includes the volcanic highland zones of Chiapas,

Guatemala (including Alta Verapaz), southwestern Honduras, and western El Salvador.

Ballplayer figurines, though again not as numerous, resemble those of the Lowlands. In general players wear plain, channeled, and glyph-decorated open yokes, knee and hip pads, and wrist wrappings, although it appears that in the Alta Verapaz subregion outfits of heavy, quilted padding paired with elaborate headdresses were preferred. Stone hachas and yokes have been identified, while palmas are for the most part completely absent from the record. There are numerous ballcourts in the Highlands, and in fact it has the highest count of the four regions.

Taladoire (2015) notes that in the 1980s, only 158 courts had been documented; by 2000 that number had grown to over 300!

Courts are open, semi-enclosed, and enclosed I-shaped forms with variably sloped apron walls, low benches, and both large and small end zones. We also find the palangana court, a type that otherwise is documented only in parts of Oaxaca, southern Puebla, and the singular case from the Gulf Coast mentioned earlier. Unlike other court types, palanganas were enclosed,

171 unroofed structures that are rectangular instead of the classic I-shape. Importantly, Highland courts do not have the stone-floor markers and carved stone wall panels common elsewhere in the Maya macroregion. Another distinct feature of the courts here are the matched pairs of stone or stucco heads that were set into the court walls. These carved stone heads often depict parrots, serpents (occasionally with a human face emerging from the open jaw), jaguars, and humans (de

Borhegyi 1980:13-16). Ballplayers are also shown with cacao pods and maize cobs, which is interesting in comparison to the maguey and pulque images found in Central Veracruz.

3.5.2.3.4 Pacific Coast

Only one ballplayer figurine had been found in the Pacific Coast zone (the coastal plains and slopes of Chiapas, Guatemala, and El Salvador) when de Borhegyi’s study was published, although thousands of other figurine types were documented. However, we do have numerous stone monuments and panels from the different settlements such as Santa Lucia Cotzumalhuapa,

Tonala, El Baul, El Castillo, and Palo Gordo, among others, that represent ballplayers in different poses. These sculptural figures mostly lack the ornate headdresses seen in other regions, although there are images where a few wear conical hats. Other types of gear include tasseled hip pads, yokes, and effigy gloves. Frequently depicted alongside the ballplayers are sky-gods, jaguars, rabbits, flying sun-vultures, and human skulls and skull racks or tzompantli

(the latter of which is considered a Mexican “trait”), with an emphasis on acts of human sacrifice and dismemberment. Carved stone heads similar to those from the Highlands have been documented in the ballcourts here. De Borhegyi (1980:17) describes the area as having a particularly gruesome style of imagery, suggesting that here and in the Gulf Coast, the Northern

Lowlands, and coastal Oaxaca where we see similar iconography, the ballgame had a “deeply religious and sinister character,” employing “black magic” and bloodshed.

172

At the time of this study, only twenty formal masonry courts had been recorded; and de

Borhegyi (1980:16-17) proposed that some monumental plazas may have been used to play the game (this possibility has been suggested for other regions of Mesoamerica where formal masonry ballcourts are not found but ballgame material culture is present). Since then, various courts have been found and excavated on the Pacific Coast, including at Paso de la Amada (Hill et al. 1998); El Ujuxte (Love et al. 1996) and Takalik Abaj (Schieber de Labarreda 1994), bringing the total known courts to over 50. Paso de la Amada is probably the earliest known monumental ballcourt in Mesoamerica, constructed in 1400 BCE during the Early Formative period. However, the culture that built this court were the Mokayan, and not ethnically Maya.

173

3.5.2.4 Material and Iconographic Evidence from Other Areas of Mesoamerica

Figure 3.21: Map showing other macoregions of Mesoamerica as well as areas outside of Mesoamerica (mentioned in the text) where ballcourt structures have been found.

Material and ethnohistorical evidence from the three macroregions discussed previously represent the core sources for the majority of research on the Mesoamerican ballgame. But sporting activities involving athletic intra- and intercommunal competitions played with rubber, stuffed grass, cotton, wooden, and/or leather balls, were popular all throughout Mesoamerica.

Moreover, the widespread survival of traditional and syncretic ball sports in diverse states such as Michoacán, Chihuahua, Oaxaca, Sinaloa, Colima, Zacatecas, Guerrero, San Luis Potosi, and

174

Jalisco, shows that it is worth our while to review the archaeological evidence available from outside the core macroregions (Figure 3.21).

West Mexico, consisting of the Mexican states of Nayarit, Jalisco, Colima, and

Michoacán, has a strong but understudied ballgame tradition. Formalized ball courts are not common (n=115) except in very specific regions. In Jalisco, ballcourts are associated with the

Teuchitlán tradition, which emerged in the core zone of the highland lake district during the

Classic period. At the site of El Arnel (300 BCE – CE 200) for example, courts are attached to circular platform complexes, or clusters of architectural circles surrounded by platforms that form precincts of different social statuses. More formal ballcourts — described as open I-shaped courts — are located in the central precincts, while less formal versions are found in lower-status precincts. Interestingly, few courts have been found outside the core lake district, suggesting that the ballgame probably served an integrative role between different social segments of the core, but did not figure much into the relationships between the core and the hinterlands

(Weigand 2000:47–55).

On the other hand, ballplayer figurines and portable game art objects are abundant from the Preclassic period onward. Game equipment included rubber balls of various sizes, both long and short sticks, bats, racquets, and paddles – clearly indicating that a variety of games were played here. Unique to the West Mexican tradition are ceramic ballcourt models which show games in progress (Figure 3.22) (Ramos 2012:8). The court models are I-shaped with more or less vertical walls and side benches, sometimes surrounded by two to four stepped structures.

Incredibly, spectators (numbering anywhere from 12 – 30) are shown seated on these structures and include men and women, musicians, embracing lovers, and even umpires engaged in various ballgame-related and other activities (Ramos 2012:70)! Two opposing teams of two to three

175 ballplayers are on the court, some wearing gloves and chin-strapped helmets or “conical bills”, and with anywhere from one to four balls in play (de Borhegyi 1980:6). The generally populous nature of these ballcourt models obviously shows games as celebratory communal traditions.

Ramos (2012:67) purposes that the combination of players and the ritual outfits observed on male players points to the more secular nature of the entertainment-related activities of ballgame ritual in West Mexico.

Figure 3.22: Ceramic model of a ballcourt and game from West Mexico. These models, which show lively scenes of people enjoying a game, have usually been found in tombs (photograph by Tom Aleto, reproduced under license terms cc-by 2.0)

176

Not much is known about Prehispanic ballgames in Northern Mexico (Guanajuato,

Zacatecas, and Southern Durango), although we do find good evidence that games were played in this region. Only 35 ballcourts have been documented at a total of 26 sites, including La

Quemada in the Sierra Madre region of Durango. Open ended courts with low apron walls date to the Early and Late Classic, while enclosed courts with aprons, cornices and end structures, and enclosed courts with benches, aprons, cornices and large end zones date exclusively to the Late

Classic. A fourth form, open-ended with steep aprons and end structures, appears in the Late

Classic and endures through the Postclassic (Taladoire 2000, 2015). At the time of de

Borhegyi’s study, no stone yokes, palmas, or hachas had been found.

Guerrero, once lumped in with West Mexico by many scholars, is slowly emerging as its own unique macroregion with a distinct cultural historical trajectory. Indeed, most strikingly many later sites in Guerrero do not show the signs of the Classic to Postclassic collapse and transition that appear in other regions of Mesoamerica from the 7th to the 10th centuries CE.

Much more investigation needs to be done, but we still find evidence of the ballgame here. As of

CE 2000, 36 ballcourts have been documented at 32 sites, including Xochipala, Piedra Labrada, and Tehuacalco. In fact, Piedra Labrada has five ballcourts (meaning we may have to revise our figures since these courts were only documented after 2000) and ballplayer sculptures were found there. Hundreds of female ballplayer figurines have been found at Xochipala wearing padding on hips, wrists, and knees that resemble the outfit on the more numerous male figurines

(Ramos 2012:55, 68-69).

Numerous ballcourts have been found at the northern and southern margins of

Mesoamerica. In the Cuyumapa River drainage of Honduras during the Late Formative to the

Terminal Classic, the differences in the orientations of the ballcourts were linked by Hendon and

177 colleagues (2009) to seasonal ritual cycles and patterns of intraregional social interaction. Those aligned to the summer solstice were for the use of the local communities close to them, while the courts aligned to the winter solstice were larger and located in major centers where multiple communities would have gathered together to participate in ballgame rituals (Hendon et al.

2009). In their survey in and around Paquimé (CE 1130 – 1450), Whalen and Minnis (2001) documented an extensive ballcourt network. Three I-shaped courts are located at Paquimé itself, while T-shaped courts have been found at smaller sites within the region. Given that Paquimé’s control over the region was minimal at best, Whalen and Minnis (1996: 743-744) relate these distribution patterns to different levels of centralization and interaction between different sites.

Paquimé shares certain cultural and social affiliations with Mesoamerica but most likely was occupied by people from the Mogollon macroregion in the Southwest. Around 206 ballcourts have actually been registered in the Southwest region, which includes both parts of the

American Southwest and Northwest Mexico (Taladoire 2015b). Recent reviews of published data suggest that there may be even more courts than previously thought in Arizona (Taladoire, personal communication), the northernmost extension of ballgames played in dedicated architectural structures. Alongside the architectural evidence, archaeologists find paddles that may have been used to strike the ball during play. Similar such paddles have been documented in Mesoamerica. Evidence for ballgames occurs even in the Caribbean islands, beginning as early as 600 CE in Puerto Rico (Ramos 2012:182-183, citing Algeria 1983), showing just how far the game had spread. Spanish chroniclers observed that ballgames were an important part of

Taino life, and the game they played was similar to those in Mesoamerica and northern South

America (Alegria 1983). Game events were communal affairs that featured great amounts of entertainment. Courts were constructed in the areas between Taino polities where a whole range

178 of interactions such as trade, ceremonies, intermarriage, and other exchanges would take place

(Ramos 2012:184). While we do not find parallel structures further south, a ballgame was played in South America. The Otomoc people of the Orinoco River region in Venezuela used a ball as big as that used by the ancient Maya, and would play on fields located on the outskirts of town (Gumilla 1745:469). The night before a match, the caciques would assign playing positions and other duties. An important elder from the community served as judge, dictating the game’s rules and acting as custodian of the items placed as bets (Ramos 2012:189), in many ways similar to how game matches are organized and regulated in modern pelota Mixteca.

3.5.3 The Ballgame in Oaxaca

Despite the sport’s long, enduring history in Oaxaca (Figure 3.22; Table 3.3), there has been relatively limited investigation with regards to the ballcourts and ballgame in this macroregion, with only two published regional ballcourt studies, one from the Valley of Oaxaca

(Kowalewski et al. 1991), and another from the Pacific Coast (Zeitlin 1993). For both studies, the courts were not the primary focus of the original investigations. Taladoire (personal communication) is currently conducting a formal count of the total number of registered courts; current research up to the year 2000 shows that there are at least 100 court structures located at different sites throughout the state (Taladoire 2015). Again, there is certainly many more that remain unreported. Only a small number of registered courts have actually been excavated.

179

Figure 3.23: Map of the different regions and subregions of Oaxaca discussed in the text and where ballcourts have been documented.

180

Table 3.3: Number of ballcourts documented in the regions and subregions of Oaxaca7 Region Subregion Ballcourt Count Central Valleys 64 Centro District 22 Ejutla 5 Etla 11 Ocotlan 5 Tlacolula 18 Costa Chica 23 Chontalpa 2 Hualtulco 5 Manialtepec 7 Pochutla 2 Rio Verde 1 Cañada 5 Quiotepec 2 Dominguillo 1 Cuitlacan Canada 1 Isthmus 17 Tehuantepec 1 Jalapa de Marquez 11 Mixteca 38 Mixteca Alta 35 Mixteca Baja 2 Mixteca Costa 1 Papaloapam Tutuxtepec 2

Sierra Norte 11 Ixtlan 2 Sierra Mixe 8 Sierra Sur 31 Chontalpa 2 Miahuatlan 4 Nejapa 15 Quiechapa 3 Sola de Vega 7 Note: for some courts only the regional location is available.

7 These numbers are drawn from two sources: 1) my own research using articles and field reports; and 2) Eric Taladoire generously shared his own database on registered courts in Oaxaca. This data was collected and entered into the Microsoft Access database. 181

When exactly the ballgame first appeared in Oaxaca is subject to debate. Blomster

(2012) claims that there is not much evidence for Oaxaca’s early involvement with the ballgame, but he does cite a ballplayer figurine excavated at the site of Etlatongo in the Mixteca Alta that dated to between 1399 and 899 BCE, or the Early Formative Period, that may suggest otherwise.

This find is important because Olmec Early Horizon (1400-1000 BCE) ballplayer figurines are not usually found in primary contexts in the Central Valleys and are very rare, not really appearing there and elsewhere in Oaxaca until the Middle Formative period (900 – 300 BC), with the possible exception of the Etlatongo figurine (Blomster 2012:8022-8023). But there exists the possibility that a ballgame may have existed in Oaxaca beginning as early as the

Archaic, as evidenced by an intriguing structure at the site of Geo-Shih that consists of two 20- meter-long parallel lines of stones placed seven meters apart. While some interpret it as a dance floor (Flannery and Marcus 2004), there is also the possibility that it may have been an early playing field (Blomster 2012:8022; Hill 1999:6). Winter (2009) found a site identical to Geo-

Shih that also features parallel stone alignments as well, perhaps implying a network of open-air

Archaic sites dedicated to ritual gatherings (Lohse et al. 2004). Certainly, the Early Formative date for the Etlatongo figurine shows that Oaxacans were at least aware of the ballgame early on, and it’s costuming and Olmecoid appearance may be evidence of a more formalized game spreading throughout Mesoamerica during this time period (Blomster 2012:8025).

Still, even when they do appear ballplayer figurines are not very numerous in the

Formative period, nor is there much in the way of ballgame art and iconography in general (de

Borhegyi 1980). Those figures that have been identified as ballplayers wear buttock belts, and many hold small balls in their hands. During excavations at the site of El Cucharital in Nejapa,

182 we found the partial leg of a figurine with protective wrapping around its knee. While it could possibly be a warrior, the wrapping suggests that it is also likely the leg of a ballplayer. Another group of figurines has been found in the Central Valleys where the players wear masks and other items associated with the ballgame. Four of the figurines are male and two are female, one of which is standing while the other is seated (Ramos 2012:120). While only a few ceramic vessels have been documented that depict ballplayers (de Borhegyi 1980:17), other graphic representations of ballplayers were painted on tomb walls, such as at the sites of Huitzo (Urcid

2005) and Atzompa (García Robles y Cuatle 2011) in the Central Valleys. Stone effigies of ballgame equipment such as yokes are known but few, while hachas are more common and palmas are completely absent. Graphic depictions of courts, and sometimes even games, are most prominent in the Mixtec codices (Joyce et al. 2004). In fact, of the surviving codices, it is the Mixtec group that preserves the most ballgame symbolism. The Mixtecs even produced exquisitely carved bones, some made of human femurs, which feature ballcourt icons.

Ballcourt sculptural art is also limited in comparison to other regions. Some simple, undecorated center-alley stone markers have been documented at sites throughout the state. The center marker discovered in the ballcourt at El Palmillo in the Tlacolula Valley, for example, was an ovoid-shaped, thin sheet of mica stone; interestingly, the palace next to the court featured a mica-lined entryway. Free-standing stone figures possibly representing dead ballplayers have been also been observed, mostly at sites in the Costa Chica region on the Pacific Coast and one example from Cerro de las Minas in the Mixteca Baja (Zeitlin 1993). These cross-armed sculptures resemble the figure found at Piedra Labrada in Guerrero, which makes sense given the cultural interactions between these areas. There are also possible ballplayer monuments from

Tlacochauaya in the Central Valleys and Tequixtepec in the Mixteca Baja (Taladoire 2003:328).

183

The most famous carved representations of ball players are found at Dainzú (Bernal and Seuffert

1979; Orr 1997; Urcid 2014; see Taube and Zender 2009 for alternative arguments that the

Dainzu figures are gladiatorial boxers). These carved stelae depict warrior or warrior-jaguars dressed as ballplayers, their bodies contorted in different positions as they throw small handballs

(there is some disagreement as to whether these balls were made of rubber or stone). Additional ballplayers were carved along a path leading up the hill directly behind the large stelae.

Recently, three incredible and large clay ballcourt sculptures (Figure 3.24) were recorded at the

Late Formative cave site of Cueva el Rey Kong-Oy in the Mixe region (just north of Nejapa), and only a few kilometers away from the site of Chuxnabán with its three ballcourts (Winter et al. 2014). Alongside one of the ballcourt sculptures is a life-sized mud-sculpted human figure bedecked in ballplayer regalia with a ball in each hand (Figure 3.25).

184

Figure 3.24: One of the ballcourt sculptures from the Cueva del Rey Kong-Oy. You can see a small stairway set into the terminal mound (photograph by Stacie M. King, reproduced here with permission).

185

Figure 3.25: Detail of the Rey Kong-Oy figure, located near the ballcourt model in Figure 3.24 (photograph by Stacie M. King, reproduced here with permission). He wears a jaguar-style mask (upper) and holds a rubber ball in his hand (lower).

186

The appearance of masonry ballcourts in the Late Formative (300 BCE – CE 300, or

Monte Albán Period II [100 BCE – CE 200] in the Central Valleys) during a period of increasing conflict appears to establish an ideological link between religion and warfare (Joyce and Winter

1996:39), with state ideologies possibly being promulgated through the ballgame

(Blomster2012:8024; Feinman and Nicholas 2011:101). In their survey of the Central Valleys,

Kowalewski and colleagues (1991:37) noted that half of the ballcourts they identified were located on regional or subregional boundaries. Most are found in “lower ranking” settlements and not in the major centers, including boundary capitals. In fact, non-ball court sites are as large or larger than those with ballcourts (Kowalewski et al. 1991: 37-38). Boundary-marking may have occurred in other regions of Oaxaca as well, where some courts can be found at isolated locations between sites or ritual centers (Gillespie 1991; Pohl et al. 1997; Winter et al.

2014). Over time this changes and during later periods we begin to see more ballcourts constructed at settlements within regional polities.

Due to a paucity of intensive investigation, we do not know exactly how many ballcourts there are in many of the regions of Oaxaca outside the areas described previously. Yet, what has been documented (see Table 3.3) demonstrates that these areas do deserve more attention from archaeologists. We still only know of two ballcourts in the Mixeca Baja: one located at the site of

Cerro de las Minas, close to the modern town of Huajuapan de León; and the other at Cerro de la

Caja near Tequixtepec, respectively. Additional ballcourts have been found in the Mixteca

Coast, especially in and around the Postclassic Mixtec kingdom of Tututepec (Joyce et al. 2004;

Zeitlin 1993). As we can see in Table 3.3, there are even more courts in the Isthmus, including two at the famous Zapotec “fortress” of Guilengola. Evidence of the ballgame in this region implores us to investigate further. For example, nine ballcourts were documented by INAH

187 crews at the Late Formative site of Tres Picos (Markens and Santiago 2014). Here we find both open and T-shaped ballcourts; in contrast, courts in Oaxaca are predominantly I-shaped. The antiquity of the site and the sheer number of ballcourts may indicate that Tres Picos was another significant center for ballgames apart from Monte Albán in the Formative period. In the

Isthmus, we also find what has been described as a distinct Oaxacan tradition, named Mixe-

Zoque after the ethnolinguistic group that dominated the region throughout most of the

Prehispanic era (Winter and Santiago 2014), which may partially explain some of the differences we see with regards to the ballgame. Certainly, the geographic position of the Isthmus close to the Soconusco region, where one of the earliest formal courts was documented at Paso de la

Amada, suggests that this region had an important role in the development of ballgames in

Oaxaca.

Of the ballcourts in the north of the state, not much is known as there has little intensive investigation in the Mazateca and Chinantla. It is important that we survey these areas in the future, especially since they border Veracruz and likely had closer cultural ties to the Gulf Coast.

Five ballcourts have been documented in the Cuitlácan Cañada, many dating to the Late

Formative and Classic periods, but these studies are from the 1970s and it would be worthwhile to revisit the area. We have a little more evidence of the ballgame in the Mixe region, but we are still missing a lot of information. Two ballcourts were observed, but not registered, on the private lands of a local owner from the town of Santo Domingo Narro, in the municipality of San

Júan Juquila Mixes. As mentioned previously, three other ballcourts were documented at the site of Chuxnabán, which dates to the Late Formative and was occupied during the Classic and

Postclassic periods (Markens and Winter 2014; Winter 2008). Another was documented at the site of Moctum, near the modern town of San Marcus Moctum.

188

Insight into the significance and role of the ballgame among Oaxacan communities can be seen in the way ballcourt icons are employed in the Mixtec codices. In all, there are eight principal Mixtec codices that have survived to the present day, representing the largest corpus of

Prehispanic books in existence (Pohl 1994). Four of these codices – the Codex Vindobonensis or

Vienna, the Codex Zouche-Nuttall, the Codex Selden, and the Codex Bodley – all share significant sections that relate the histories, biographies, genealogies, and ritual affairs of the

Mixtec kingdoms during the Epiclassic and Postclassic periods (AD 900-1521). Occasionally cited but rarely contextualized, the Mixtec codices are particularly useful for thinking about ballcourts as active social places for the negotiation and contestation of social relationships and political power. There are some instances where the court, represented pictographically as an I- shape drawn horizontally (Figure 3.25), is shown in association with deities. On page 15 of the

Codex Zouche-Nuttall, for example, there is a bird deity seen devouring a ballcourt. In another scene on page 3 of the same codex, a water goddess is decapitated in the center of a court

(Macazaga Ordoño 1982:64). Of course, these sections of the codex deal with a time immemorial that is prior to the historical events related in the rest of the document.

189

Figure 3.26: Drawing showing how ballcourts are commonly depicted in Mixtec codices, based on an icon from the Codex Zouche-Nuttall.

For the most part, however, the ballcourt icon in the Mixtec codices is used in two ways: as toponyms or parts of place names; and as places or sites where important historical events occurred. Many of these historical scenes are related to the life of Lord Eight Deer Jaguar Claw, who is often shown playing the ballgame against rivals, and performing rituals in a court (Byland and Pohl 1994; Joyce et al. 2004; Pohl 2004). In one scene, we see him meeting in a ballcourt with Lord Six Jaguar, with the latter performing a Toltec-Chichimec investiture ritual in which he pierces the nose of Eight Deer with a chalchihuite or special jade stone nose plug (Joyce et al.

2004; Macazaga Ordoño 1982). Two of the codices preface the arrival of Lord Eight Deer on the

Pacific Coast with scenes of activities that take place in ballcourts (Joyce et al 2004:284).

Placing these scenes within the narrative of Lord Eight Deer’s political and military campaigns, we see that playing the ballgame and performing ritual actions in the court were critical in legitimizing his rule and establishing Tututepec (Joyce et al. 2004:283). Whether these events actually occurred or not, the centering of games in an important cultural historical narrative highlights the links between ballgames and sociopolitical relationships. 190

The Mixtec depictions stand out in comparison to the greater corpus of ballcourt icons in the Prehispanic codices. As mentioned previously, the largest set of icons are found in these documents, and most of the courts shown with four colors, usually red paired with blue and green, occur here. Ballplayers in the Mixtec codices tend more to be leaders than deities, in contrast to the Central Mexican documents. There are also depictions of women such as 11

Serpent (Xochiquetzal), who do not participate in the game but appear on the side as spectators.

Interestingly, of the 21 occurrences where the ballcourt is used as an anthroponym, all but one are from the Mixtec codices. With regards to the use of the icon as a toponym, the court icons more or less resemble each other, I-shaped icons sometimes divided into sections by lines. One example from the Codex Nuttall does stand out as it was drawn with tenoned stoned rings, which have not been documented in Oaxaca. This specific example may actually be referring to sites on the Central Mexican Plateau where these sculptures occur (Taladoire 2015b

), demonstrating that indigenous groups in the Prehispanic past recognized regional differences in ballcourts.

Based on the available evidence, it appears that Oaxacan communities utilized and manipulated the ballgame in different ways. Many scholars have argued that games played a social mediating role, with ballcourts structures serving as physical boundary markers between different polities (Feinman and Nicholas 2011; Finsten et al. 1996; Gillespie 1991; Joyce and

Winter 1996). As stated previously, similar arguments have been made for the Mixteca Alta, where Classic era ballcourts often marked the boundaries between later Postclassic altepetl

(Byland and Pohl 1994; Pohl and Byland 1990; Pohl et al. 1997). Zeitlin (1993) proposes that in the southern Isthmus and Costa Chica, the ballgame was integral to a widespread peer-polity network. Redmond (1983) suggests that the ballcourts found at the very northern edge of the

191

Cuitlácan Cañada represented (and were proof of) a militarized frontier zone of the expansive

Terminal Formative Zapotec empire based at Monte Albán. Kowalewski and colleagues even

(1991:42) claim that the game served the dual purpose of maintaining the fighting readiness of warriors during idle times while also reinforcing key state and elite ideologies. In a similar vein,

Orr (2001; 2003) views the ballgame as a mock combat ritual, performed in conjunction with pilgrimages to sacred hill centers.

Perhaps one of the most interesting interpretations of the ballgame’s role in local and regional Oaxacan politics comes to us from the Cuitlácan Cañada. Redmond (1983) uses the supposedly “martial” character of the ballgame as evidence that two sites in the Cuitlácan

Cañada with ballcourts were fortified settlements guarding the Quiotepec frontier. Assuming that the game was commonly played by warriors, who would have played it when not on military duties, she suggests that the fact that the courts are only found on the alluvial plain during the

Las Lomas phase is further proof of Zapotec conquest and subjugation of the northern boundary of the Cuitlácan Cañada (Redmond 1983 111-112). However, the reported dimensions of these courts are significant enough to give one pause: the I-shaped court at Cerro de Quiotepec measures 70 m long, while the I-shaped court at Paso de Quiotepec is 65 m long. These measurements mean that the courts fall at the larger end of the size range for Oaxacan ballcourts, indicating that these structures were significant investments for the people who built them.

Ballcourts at any size are sites of social activities, and courts of these sizes would surely not have been used exclusively by the inhabitants of that community alone; their construction alone would have required a large investment of labor. Of the handful of courts recorded in Oaxaca that measure over 60 m, all are located at sites that were residential and ceremonial settlements with residential, not military fortifications. For the ballcourts recorded at sites in other regions of

192

Mesoamerica that measure over 60 m, again this pattern holds. Therefore, if the Quiotepec courts were for the sole purpose of warriors playing when not engaged in military duties, it would make them extremely exceptional.

Redmond bases some of her evidence for linking the ballgame to war on descriptions of the Cherokee ballgame by Gearing (1958, 1962), who studied artistic renditions of four structural poses assumed by warriors of an 18th-century Cherokee village. While I do not doubt the validity of this study, it is admittedly narrow in scope, and as I demonstrate earlier, ballgames among the Cherokee and many other American indigenous groups were played for a variety of reasons, from influencing the weather to healing rituals and yes, even as a substitute for warfare.

Above all, ballgames were social activities, drawing community members and neighboring communities together for both friendly and hostile competition, depending on the needs of the hosts and guests. We should therefore be cautious in assuming that 1) only warriors were playing the ballgame at these particular courts; and 2) that the presence of ballcourts necessarily means that these were fortified settlements guarding a militarized frontier. In essence,

Redmond’s argument for a military purpose for the courts is based on a single comparison study, previous (but not necessarily erroneous) interpretations of the game, and the conclusion that the area had been conquered by the Zapotecs from Monte Albán. We should therefore reserve caution in concluding that the ballcourts of the Quiotepec alluvial plain were militaristic in character, and that their presence is indicative of a fortified frontier.

On the other hand, the ballcourt sculptures discovered in the Cueva del Rey Kong-Oy, a

Mixe cultural hero, challenge our assumptions about this automatic link between conflict and the

Oaxacan ballgame. I have already mentioned the large male figure, El Rey Kong-Oy, dressed in ballplayer attire and holding two small balls in his hands. Alongside these earthen clay court

193 models were other life-sized models of jaguars and human figures with exaggerated genitalia.

Some of the male and female figures are depicted in coitus, while there are other female sculptures in birth positions with prominently displayed vaginas – images not frequently found in

Mesoamerican art (Winter et al. 2014:312-313). The sculptures have been dated to the Late

Formative, and the cave in which they were found is located close to Chuxnabán. While their meaning is still unknown, as well as the relationship between the cave and Chuxnabán

(considering that the site did not reach its apogee until the Classic period), these sculptures suggest other possible, unexplored ideological meanings linked to the ballgame that were unrelated to conflict or boundary marking.

3.5.4 Discussion of the Material Evidence

In the popular imagination, the Mesoamerican ballgame involved two teams of players bouncing a rubber ball between them using only their hips, engaged in what was actually a large- scale ritual event linked to agricultural fertility and sacred cosmology. Points were scored by hitting the ball through hoops attached to the side walls of the court. At the end of the game members of the losing team were sacrificed. This image is not too far off: two teams did engage in play, often using a rubber ball. Sometimes a sacrifice did occur, along with other rites and celebratory events including grand feasts. But can we really argue that the game was played the same way and that it meant the same thing to all people in all places over the course of more than three millennia? Evidence presented in the previous and following sections show that the variability seen in ballgame material culture strongly suggests that not only did different kinds of ballgames and rule systems exist, but that the ideologies, symbols, and ritual practices linked to the games were very diverse as well.

194

3.5.4.1 Ballcourts

In Mesoamerica, the majority of courts documented are I-shaped – that is, there are two parallel lateral structures with a usually narrow playing field between them, with two terminal structures that are perpendicular to the lateral structures forming end zones. Although there is this more or less standardized form, individual ballcourt features such as lateral structures, terminal structures, benches and bench walls, the sloped apron, cornices, playing alleys, and end zones were incredibly variable both regionally and by time period (see Taladoire 1981). For example, the slope of the aprons ranges from nearly flat to perfectly vertical, while benches can be present or absent, short or extended, massive or small, and sloped as well. Baron (2006:17) argues that it is difficult to create a typology of courts because of the diversity of forms, which only increases as we discover more, although numerous attempts have been made. Taladoire

(1981, 2000) has created a typology that is the most complete and comprehensive. He continues to revise the evolutionary scheme and typology as more evidence is discovered, but even this scheme runs into problems because of the incredible diversity of ballcourt forms.

What the evidence does show is that the popular ballcourt of our imagination only existed in specific places at particular times. The tenoned stone rings seen in most modern representations of ballcourts are in fact geographically restricted to Central Mexico and the

Northern Lowlands respectively. They are also a much later development, appearing only in the

Postclassic period (Taladoire 2012). Despite the perception that ballcourts always have center markers, there are in fact only a small number in comparison to the over 2550 courts that have been registered through to now (Eric Taladoire, personal communication). Ballcourts also clearly occur in a range of architectural arrangements and in different positions within site

195 layouts and settlement hierarchies, pointing to a range of variability in their meaning, importance, and centrality to the community where they were located and in within the regional social hierarchy.

3.5.4.2 Ballgame Equipment

We know from the evidence that a range of ballgame equipment was used for play (Table

3.4). There is ample evidence for the use of batons, bats, paddles and other similar stick-like equipment (de Borhegyi 1980). There are also manoplas or hand stones that may have been used in a special type of handball game (but probably mainly for gladiatorial battles). Many of the recorded manoplas come from Oaxaca, the adjoining area of Veracruz, and the highland and

Pacific piedmont of Guatemala – a distribution similar to that of the palangana court (Taladoire

2003:329). Protective clothing is common, and it is clear that helmets and padding were fairly standard body equipment throughout Mesoamerica, although we also see local and regional styles. On the other hand, the stone yokes, hachas and palmas often described as common ballgame equipment are actually found most frequently in the Gulf Coast. However, despite their close association with the ballgame in the Gulf Coast, yokes and hachas are really only common in central and southern Veracruz, while palmas apparently cluster in the north-central area, and then only in the Late Classic period (Daneels 2008; Koontz 2008:325). In other macroregions of Mesoamerica, these objects – or at least their stone icons – occur with much less frequency or are completely absent (Figure 3.26). Although hip padding and belts are frequently molded on ballplayer figurines, yokes are a distinct type of hip garment. They are mostly depicted on figurines and are worn by ballplayers in ballcourt sculpture in the Maya macroregion, but the actual number of stone yokes are few. Moreover, the style, form and

196 designs of the stone game equipment vary by region. For example, in the Maya region we find mostly plain, horse-shoe shaped, open-ended yokes, while the Gulf Coast appears to be the only place where closed yokes were fabricated.

Table 3.4: Ballgame equipment by macroregion. Macroregion Game Equipment Body Playing Central Mexico Helmets, body padding, goggles Rubber balls Racquets, bats clubs, Gulf Coast Helmets, body padding rubber balls Maya Northern Lowlands Helmets, body padding Rubber balls Feathered headdresses, Central Lowlands Large rubber balls helmets, body padding, goggles Decorated headdresses, Southern Highlands Rubber balls body padding Pacific Coastal Lowlands Hip and knee padding, gloves Rubber balls

Helmets, wrist/elbow/knee padding, Oaxaca Rubber balls hip padding Clubs, racquets, Western Mexico Helmets, masks small and large balls, sticks

197

Figure 3.27: Maps showing the frequency of hachas, palmas, and yugos found in the different regions of Mesoamerica.

198

Yuguitos, shaped like smaller versions of the yokes, are often identified as proof of the game when other evidence is not present, however Taladoire (2009:194) rightly points out that morphological similarity does not equal function, and that the yuguitos could have had a different meaning from the yokes. Furthermore, he argues that scholars have never really been able to precisely determine what relationship the stone yokes actually had to the game (Taladoire

2009:194). In fact, a yoke has actually never been found in or around a ballcourt, only in funerary or depositional contexts (Kurosaki Maekawa 2006; Roose 2006). Interestingly, the one exception so far is an example from an archaeological site near Lachiguiri, Oaxaca, found by

Fernando Getino Granados in recent salvage excavations, which are as yet unpublished (Stacie

King, personal communication. Taladoire (2009:194) argues that the relationship between the game and the yoke-hacha-palma complex was more symbolic or ritual than actual practice. Of course, there is the possibility that the stone versions of palmas, hachas, and yokes were not actually worn for game play as many have previously argued. For one thing, the athleticism for which Mesoamericans were famous would have been greatly hampered by wearing body equipment made of stone, and there would have been much more serious injuries as a result of play. Instead, it is more likely that these stone objects were worn only during the ceremonies before and after the game, or they may have been used as molds for creating perishable forms made of stiffened leather stuffed with cotton (Elkholm 1946). Additionally, in many of the sculptures and ceramic vessels where ballplayers are wearing elaborate yokes or hip garments, it is obvious that they are made of woven plant materials that were also probably stuffed (Taladoire

2015b).

199

3.5.4.3 Ballgame Art Objects and Sculpture

Ballplayer figurines represent not only the earliest concrete evidence for ballgames in

Mesoamerica, but they are also proof of just how widespread the sport was even during early periods of political and social development. We find figurines dressed in ballplayer attire in nearly every macroregion beginning in the Middle Formative period, although earlier examples in specific areas do exist. Much of our information on what ballplayers wore is based on the figurines, and the different styles we observe support the existence of local and regional ballgame traditions. We also see figurines with variable sized balls – some hold a large ball likely used for the ulama hipball game, while others hold much smaller balls in their hands.

Moreover, figurines are really the only evidence we have for the participation, at least early on, of women playing ball. Female figurines have been found in West and Central Mexico as well as in Guerrero and the Gulf Coast. Collectively, these figurines show young girls and women in ballgame attire, indicating that they played hip, forearm, and kneeball (Ramos 2012:9).

Center floor markers are considered to be standard sculptural features of ballcourts in

Mesoamerica. However, there have only been 128 cases of floor marker sculptures documented by excavators so far, and of those, 57 (or 44%) were found in the center while the rest (n=71, or

56%) were discovered in other zones of the court (Taladoire 2015a:157). The documented center stone markers are circular disks that are either smooth (n = 54) or decorated with glyphic motifs, icons, or scenes (n = 49) (Taladoire 2015a:165). The purpose of these stone markers is still not very clear. Certainly, those stones with glyphs as well as those placed in the center would not have functioned as botaderas, since these are normally placed in the extreme ends of the court and would have been worn smooth through repeated strikes with the ball (Taladoire

2015a:165). Smooth disks generally come from Arizona, West Mexico, Puebla (Cantona and

200

Metlaltoyuca) or Preclassic courts (Chiapas and Belize), while those with imagery are Classic

Maya in origin.

Though still unknown, the location and iconography of the decorated disks points to their ritual and symbolic value (Taladoire 2015a:165). In cases with a single disk, the marker is placed in the center of the court (Arizona, Oaxaca [Atzompa, El Palmillo, Monte Albán, and

Yagul], Puebla [Cantona] and the Maya macroregion), while in cases where there are multiple disks they are usually placed on a longitudinal axis (Maya macroregion). An extreme case is

Tenam Rosario, with eight disks that were found in the center, the banquets, and the corners

(Taladoire 2015a:168). Taladoire (2015a:165, 173-174) concludes that the iconography on

Maya ballcourt markers were part of the royal iconography (political-ritual) seen on monuments throughout the macroregion, while the images in the codices reflect the Postclassic cosmovision generally shared among communities living in Central Mexico

3.5.4.4 The Codices

In total there are 141 images of ballcourts in the 47 surviving codices from Central

Mexico, the Mixteca region, and the Borgia group from Puebla. Over half (n = 76 or 54%) are from the Mixtec region, while the second largest group hails from Central Mexico (n = 65)

(Taladoire 2015a: 183, 185). Most examples are delimited by a double line, perhaps an artistic device referencing the sunken character of the court (n=19 or 13.6% where it’s only single line)

(Taladoire 2015a:186). Interestingly the ollin glyph, which is commonly linked in the archaeological literature to the ballgame, is relatively absent in the codices except for two examples (Taladoire 2015a:201). Of the ballcourt representations pictured in the codices, only

32 (22.8%) had central elements, 17 (53%) of which are simple circles that are either black or

201 white (Taladoire 2015a:160). More common than central elements are perpendicular lines (n=58 or 41% of the total depictions) that cross the center and divide the court into different colored quadrants (Taladoire 2015a:160). These transverse lines have been documented in modern ethnographic ballgames, including pelota Mixteca, and were likely made with perishable materials in the past.

3.5.4.5 Ballgame Iconography

As with the courts and game equipment, the iconography of the ballgame is incredibly diverse, much more so than we would assume based on scholarly descriptions (Table 3.5).

Animal subjects included not just the stereotypical butterflies, alligators, and turtles, but also jaguars, parrots, rabbits, monkeys, deer, dogs, and coyotes (de Borhegyi 1980:9). Many different types of gods were linked to the ballgame. Although deities tied to Venus are the most often cited in connection to the ballgame, a review of the evidence shows that there were in fact many that were associated with game play. While sacrificial images are overtly emphasized in our interpretations of game symbolism, many other symbols were also common, including ballcourt icons, footprints, dancers, ball players; and plant imagery such as cacao, chilies, maize cobs, agaves, pulque, and flowers. All the aforementioned representations are in fact found associated with ballcourts and ballgames throughout Mesoamerica (de Borhegyi 1980:10).

202

Table 3.5: Iconography and symbols of the ballgame by macroregion. Macroregion Gods and Deities Animals Sacrifice Others Central Mexico Venus MORE Frogs Skulls, heart sacrifice Human Old gods, sky-gods, Xipe Jaguars, monkeys, footprints, Totec, Tlaloc, Skulls, human head coatimundis, deer, players, dancers, Quetzalcoatl, Xolotl, trophies, dismembered Gulf Coast dogs, coyotes, Olmecs, etc,, Xochipilli, Xochiquetzal, arms and legs, tongues rabbits, tapirs, owls, cacao, chilies, Butterfly God, were- and cut eyes fish, serpents, etc. corn, flowers, jaguars, etc. etc. Maya Northern Lowlands Venus unknown Decapitation, skulls Central Lowlands unknown Dismembered heads Jaguars, parrots, Serpents with human Humans cacao, Southern Highlands serpents heads in mouth corn cobs, Pacific Coastal Jaguars, serpents, Flying sun Sky-gods Skulls and tzompantli Lowlands rabbits vultures

Venus, water and Frogs, monkey Wrinkled old Oaxaca fertility gods, sky-gods, heads, vultures, Human heads men, water water-gods serpents, bats symbols Dancers, Western Mexico musicians, players, acrobats

Ballgame artwork and symbolism are most extant in the Gulf Coast; however, even within this region a wide range of subjects were depicted. Severed heads appear alongside dismembered limbs, skeletons, skulls, dancers, warriors, dwarfs, heart extraction rites, stylized death heads and skeletonized manikins, composite monster deities, cross armed human figures, celestial diving gods, sun vultures, human heads in serpent jaws, flowering vines, and anthropomorphic cacao pods (Zeitlin 1993:123)! Clearly, while certain symbols or images are repeated often and frequently co-occur, there is still a remarkable diversity regionally, temporally, and even by subject matter. Even in this respect, the Gulf Coast is not alone. Both

Venus deities and the gods of pleasure were important figures in Central Mexican ballgame iconography, pointing to the cosmological nature of the game, its more experiential and pleasurable aspects, as well as the variability in this aspect of game ideology. We find similar

203 variety in the game material from Oaxaca. While not many yokes have been documented there, those that have been found often feature frogs and symbols related to water. On the other hand, the more common hachas are carved with vastly different iconography – wrinkled old men, wrinkled monkey faces, and bats, the latter an important symbol in Central Valleys iconography

(and still mentioned in modern Oaxacan oral traditions where a malevolent bat spirit-creature turns into an alluring woman to capture male essence).

3.5.5 Summary

As even a cursory review of the evidence shows, there is a remarkable degree of similarity in ballgame material culture but also great deal of inter- and intra-regional variation. It is therefore not difficult to conclude that the data points to the presence of multiple and perhaps even overlapping ballgame traditions. This point is strengthened by the fact that there are multiple sites throughout Mesoamerica that have several ballcourts, many of which – both sites and ballcourts – are contemporaneous. These inter-regional differences are important and informative, though what they reveal about cultural interactions between Prehispanic communities has yet to be teased out by researchers. We have different manifestations of ballgame sports as early as the Early Formative period. Moreover, the earliest appearances of various elements of ballgame culture occur in widely disparate and dispersed geographic areas, with the earliest figurines found in West Mexico, the first monumental ballcourt constructed in southern Chiapas, and the first sculpted ballplayer imagery appearing in the Gulf Coast region during the Early/San Lorenzo Horizon (1400-1000 BCE) (Blomster 2012:8020).

The hipball game or ullamaliztli is usually either seen as the only game played, or at the very least as the dominant elite game or ritual mode of play – other ballgame types being secular

204 in function (Berger 2009). But the idea that there were different types of ball-based sports that were variably popular throughout Mesoamerica, rather than a single dominant ballgame, is not all that radical. In the United States today, people play a range of ball-based sports using equipment specific to that particular sport. Tennis, for example, uses racquets, a small compound rubber ball covered in felt, a net stretched across a rectangular court, and is played between two to four people. Basketball, on the other hand, uses a larger inflatable rubber covered in leather or more rubber, is played on court with two baskets at each end, and is played between two teams of 5 players. Both are popular to varying degrees and are concurrently played. Yet basketball carries different social and political implications with regards to race and class than tennis does. Similarly, non-hipball games were also popular at different times and in different regions, and even co-existed with each other and with hipball (Stoll, in press). The

Tepantitla murals from Teotihuacan clearly support this proposition. While scholars acknowledge the existence of other ball sports apart from hipball, rarely do analyses take this into consideration, instead describing them as simple modalities of tlachtli. Doing so, however, reduces the richness and expansiveness of the larger ballgame and Mesoamerican sporting culture while eliding the significance of their differences and similarities. Even the hipball game was variable, as each mode had its own set of rules, counting system, and specific practices

(Taladoire 2009:193).

Across Mesoamerica we see both conservative and innovative elements in the different regional and local ballgame traditions. A good example is the I-shaped ballcourts, which first appear in the Late Preclassic and continue to be constructed until the Spanish conquest. This form is found as far north as Paquimé in Chihuahua, and as far south as Honduras. Clearly, there is something particular about the I-shape that held special cultural and ideological significance

205 throughout the millennia. Yet each region and local community interpreted this form in their own way. Though drawing on a template shared widely across Mesoamerica, communities also drew on their own specific cultural logics as well as the social and historical contexts in which they lived in order to construct a court. The same was true for ballgames: each game was an active negotiation of cultural sporting traditions that connected players and spectators to past games and to larger regional trends, in this way making and remaking them in novel situations each time they performed them, or even questioned their efficacy and appropriateness or particular elements thereof (Douglas 2013; Keen and Zeitlyn 2007; Silliman 2005:283, citing

Pauketat 2001).

3.6 THE COMMUNAL PRACTICES OF MESOAMERICAN BALLGAME TRADITIONS

In the preceding sections, I presented several different lines of evidence in order to establish the arguments that I make in the following section about the Mesoamerican ballgame

(really, ballgames). Here, I address the problems I have identified over the course of my research about the ballgame. First, I look at the claim that the ballgame was highly ritualized during the Classic period but had become increasingly secularized in the Postclassic. Next, I examine the claims about ballgame ideology, including its connection to sacrifice and cosmology. Finally, I address what I see as the most critical issue in ballgame literature – is it a ritual, or is it a sport? I conclude with my own interpretation of Mesoamerican ballgames, based on the material, ethnographic, and iconographic evidence previously described. This interpretation informs my discussion of the Nejapa ballgame evidence in Chapter 9.

206

3.6.1 Ritualization vs. Secularization in the Ballgame

Aguero and Daneels (2009:123) claim that because early games lacked related rituals, they can be considered as secular sports, defined by the authors as non-religious and very old.

This apparently changes in the Classic era, when the ballgame seemingly becomes more ritualized and achieves the status of a ritual cult (Zeitlin 1993), as seen in the architecture, sculpture, and iconography of the game and courts across Mesoamerica (Aguero and Daneels

2009:118). Then, over the course of the Postclassic period, the game once again becomes secularized and political. The participation of commoners as well as the elites in game matches, the placing of bets on game matches, and the loss of overtly religious iconography in ballgame artwork, are often cited as evidence for this increasing secularization in the latter period

(Cohodas 1975; Day 2001; De Borhegyi 1980; Fox 1996; Gillespie 1991; Hill and Clark 2001;

Koontz 2008; Kowalewski et al. 1991; Macazaga Ordoño 1982; Miller and Houston 1987;

Santley, et al. 1991; Taladoire 2001; Taladoire 2003; Whalen and Minnis 1996 Ramos 2012).

This is an interesting claim to make, as secularization would imply a decrease in the ritual significance and symbolic meaning of the game, with a concurrent shift in control over sponsorship from religious to political institutions (Bell 1997:198-199).

Supposedly, the clearest evidence for “secularization” – the disappearance of ballgame symbolism and portrayals of ballplayers from archaeological contexts in the Classic to

Postclassic transition – may be understood in different ways. It is true that in the Postclassic, stone walls and sculpture are used less as a medium for ballgame art. However, such images do not disappear as some have claimed. Instead, it appears that the Prehispanic codices were the new forum for artistic expression of ballgame iconography, and other media such as ceramics continued to be produced. Thus, ballgame art never completely disappears. Second, it is not

207 clear as to why the practice of betting or the involvement of the lower classes in game play indicates the process of secularization in the ballgame. This may in fact be a reflection of

Western attitudes towards gambling. As previously descirbed, there is plenty of historical evidence on the importance of betting throughout Native American societies, especially with regards to indigenous games (Voorhies 2017). Gambling today is still a central practice in pelota

Mixteca and other modern ballgames played in Mexico – indeed, one cannot play if they do not place their apuesto. It is not entirely unreasonable to infer that gambling may have been practiced by different societies at various times throughout Mesoamerica, and that it wasn’t an entirely new activity in the Postclassic. In some ways, gambling was a form of wealth redistribution, and materially poorer players or spectators could conceivably attain new levels of wealth or prestige through socially profitably wagering on ballgame outcomes. We should not allow modern Western perspectives on gambling as an unserious activity of luxury or vice to lead us into assuming that betting was not present in earlier time periods.

I see a few critical errors in the secularization argument: 1) a misunderstanding of what secularization actually implies; 2) misidentifying certain practices, such as betting, as evidence for increasing secularization of the game when in fact they can be (and often are) ritual elements themselves, i.e. the presence of gambling does not mean the diminishment of the ritual experience; and 3) concluding that the presence of commoners or a change in iconographic media are signs of decreasing ritualization. As defined, secularization is a kind of self-limiting process operating in the ritual systems of societies past and present (Bell 1997:200). It implies a movement of the ritual and religious commitment of practitioners from the public sphere to the private, often as a result of the undermining of religious institutions in traditional societies. In some definitions, religion becomes increasingly separated from the political, economic, and

208 social realms, and is seen as “separate” from governing institutions. It does not, however, imply a “loss” of ritual or religion within a given society. In fact, it may even “demand more personal commitment,” (Bell 1997:201) – particularly in the mind of the practitioner, simply because there is no overarching authority enforcing religious practice or engagement.

Research on daily life in the Postclassic demonstrates that both communal and elite ritual practices were still actively performed in the public sphere, participated in by priests and political lords, and attended by people of different social groups and statuses (Byland and Pohl

1994; Carrasco 1999; Clendinnen 1991; Durán 1994; Matos Moctezuma 1988; de Sahagún 1970;

Smith and Berdan 2003; Spores 1984). One need only review the Colonial and ethnohistorical documents cited by scholars for information on ballgame symbolism for evidence of rituals, ceremonies, and religious traditions in the 11th through 16th centuries. Indeed, as I discuss further in Chapter 4, ritual practices and religious institutions continued in similar forms from the

Formative era to the Postclassic, despite numerous region-wide sociopolitical reorganization of political systems. The 20-month sacred calendar used by the Aztec empire, for example, featured numerous public ceremonies and festivals that were celebrated every month. These events involved a large number of priests, sacrificial victims, noble lords, and, if appropriate, the general population of Tenochtitlan (de Sahagún 1970).

The Colonial authors themselves make the observation that the ball was the sun and that the court was the inframundo or underworld (Fox 1991; Leyenaar 1978; Macazaga Ordoño 1982;

Uriarte 2001). Moreover, the descriptions of ballgame rites also demonstrate that games still involved ritual practices or were part of ceremonial events even in the Postclassic. The Spanish chronicler Fray Toribio de Benavente Motolinia, who was either a direct witness of these events or somehow obtained details of them, describes ceremonies associated with both the construction

209 and dedication of a ballcourt. These sacred activities involved the careful placement of idols, tenoned stone rings, and chanting, among other practices, and occurred on a day considered to be auspicious by priests after consultation with the sacred calendar (Motolinia 1903:337–338, cited in Clune 1963:67 and Fox 1996:40).

Conversely, we simply do not know if the involvement of the lower or “common” classes” was limited or non-existent in the Classic period. The power of the elite classes did increase and access to formerly public ceremonial spaces was restricted in the Classic (Joyce

2004; Joyce and Winter 1996); however, there is also evidence for ballcourt use by villages or social groups lower in the social hierarchy during this same period (Hendon et al. 2009; Joyce et al. 2001; Kowalewski et al. 1991; de Montmollin 1997). Formal masonry ballcourts were not always a requirement for playing the game; in fact, some authors suggest that a number of open spaces, both formal and informal, may have been utilized for play (Taladoire 2003; Zeitlin

1993). In fact, when an exhibition match of pelota Mixteca was hosted at Monte Albán in the

1970s, the game took place on the Main Plaza itself. When speaking of the involvement of non- elites, it is important to consider what time period or geographic region we are referencing.

While in the Maya region the game may have been more directly controlled by elites, especially in the Classic era, lower-status individuals would certainly have been audience members, at least at some games. As stakeholders in the successful performance of agricultural fertility rituals, farmers would certainly have been interested in the outcomes of ballgame matches.

Apart from that, the excitement and thrill of watching physically-demanding athletic performance would have made non-elites aficionados of the game as well (Baron 2006:44).

Ramos (2012:64-65) makes the excellent point that if we based our assumptions about American baseball on the preponderance of artistic representations, one would conclude that game was

210 limited to professionals and men only, and that baseball is controlled by a handful of powerful individuals. Instead, while it has waned in popularity, baseball in its heyday was played wherever appropriate space was available, with children playing pick-up games even in the urban streets – despite the constant danger of breaking a neighbor’s car window. With Mesoamerican ballgames, the preponderance of images in specific contexts has more to do with the political nature of image-making than with the nature of the games themselves and elite control over them. The appeal of the ballgame to people of all social statuses meant that elites were still just as interested in co-opting or associating themselves with the ballgame thousands of years after the construction of the first ballcourt, as they were when games became important social activities (Baron 2006:14-15).

3.6.2 Sacrifice, Sacred Cosmology, and Ballgame Ideology

Sacrifice is always linked in our minds to the ballgame. Such connections do find support from the various sources. According to Sahagún (1970:128), for example, during the festival of Panquetzaliztli, one of the many rites performed included the sacrifice of four slaves in the middle of the tlachco of Tenochtitlán. Torquemada, elaborating further, explained in his accounts that these sacrifices occurred in the morning and afternoon (Macazaga Ordoño

1982:67). As mentioned previously, ballgame iconography is replete with severed heads, dismembered limbs, skeletons, and other images of death (Zeitlin 1993:123). Ritually beheaded players are argued to be the main protagonists in ballgame imagery (Aguero and Daneels 2009), such as in the sculpture of Aparicio and the murals of Las Higueras in Veracruz (Koontz 2008).

Decapitation also appears on palmas, in hachas styled as severed heads, and in the monumental art of several ballcourts in Central Veracruz (Koontz 2008:331), Chichen Itza (Yucatan), and

211

Dainzu (Oaxaca), among others.

Depictions of sacrifice are most prevalent in the Gulf Coast, which was probably the ideological center of a deeply ritualized regional ballgame tradition that overtly emphasized cosmological, sacrificial, and agricultural themes. This so-called ballgame cult reached its apogee during the height of the Classic period (CE 650-1000) (Koontz 2008:11), when game rites were supposedly grafted onto the Plumed Serpent cult after the decline of Teotihuacan

(Daneels 2008:214). Ballgames featured prominently in the regional iconography, most especially in the Central Veracruz region (Pasztory 1972, 1978; Aguero and Daneels 2009).

Koontz (2008, citing Wilkerson 1991) goes as far to say that the right to govern was directly linked to the sacrifice of a victim in the ballcourt after a game match (what evidence there is for this, however, is not certain). How we understand the iconographic themes in Gulf Coast ballgame ideology is largely based on the El Tajín murals (Zeitlin 1993). Interpretations, however, have varied. Where Cohodas (1975) sees a dominant yet general theme of agriculture fertility expressed in two parallel narratives of the death and rebirth of the Sun god (with each mural depicting its own set of seasonally reenacted ceremonies), Wilkerson (1991) argues for a specifically Gulf Coast variant of the mythical underworld journey theme involving pulque and an important ruler of El Tajín known as Thirteen Rabbit. Another interpretation proposes that the pulque gods were propitiated during sacrificial ceremonies coinciding with the 9-day disappearance of Venus (Zeitlin 1993). Meanwhile, Pasztory (1972) and Kowalski (1998) assert that the El Tajín narratives relate human and divine sacrifice to political power, creative generation, and plant growth.

Problematically, there are some rather significant differences in the ballgame material culture that complicate our reading of Gulf Coast ballgame ideology. The stone yokes and

212 hachas, so closely associated with the ballgame in this macroregion (and the ballgame in general), are really, as previously described, only common in Central and Southern Veracruz, while palmas cluster in the North-Central area, and then only in the Late Classic period (Daneels

2008; Koontz 2008:325). Some of the iconographic elements that are depicted at El Tajín are also found in South-Central Veracruz but on different media, while other elements are not grouped together in similar ways (Koontz 2008:333, 341). While the decapitation ritual cited frequently by scholars is depicted at least nine times, these images only occur at three sites

(Aguero and Daneels 2009). Most representations of decapitated heads in fact are on the stone hachas that would have probably been worn as ceremonial attire but not for actual game play.

Meanwhile, ballcourts in nearby zones lack images or representations of this decapitation rite

(Daneels 2008). These differences, and the elaboration of ballcourt imagery at El Tajin, may indicate that this altepetl was the main propagator of this specific ritual ballgame cult. The proposal makes additional sense when we consider that there are 17 (or 18, reports vary) ballcourts at El Tajin, many of which have vastly different orientations and dimensions. Some have suggested that the courts were dedicated to separate religious subcults or perhaps even to distinct competing lineages (Zeitlin 1993:123). Taking all this evidence in to account, we may have to consider that there was never an orthodox version of a Gulf Coast ritualized ballgame cult as popularly imagined. Instead, the material evidence reflects localized expressions of a broader regional tradition that existed during a specific time period (Zeitlin 1993).

Surprisingly, similar iconographic themes and patterns can be found all the way to the

Pacific coastal zones of Oaxaca, Chiapas, and Guatemala during the same time period (Koontz

2008:351; Zeitlin 1993: 121-126). These areas have long been grouped together as the Pacific

Coastal Lowlands, as there appears to have been significant cross-cultural contact and affiliations

213 between these regions and subregions (Winter 2008; Winter and Santiago 2014; Zeitlin 1993).

Although the iconographic themes echo the Gulf Coast tradition, the different artistic conventions, while possibly related to changes through time, more than likely represented separate instances of local leaders differentially embracing and promoting the ballgame themes and rituals related to this tradition. Arguably, this is certainly the case for the Oaxacan coast

(Zeitlin 1993:135). The broad appeal of this particular ritualized ballgame may have been in the symbolic power of the ideology itself, which again emphasized generative sacrifice, supernatural death, and divine rebirth linked to political authority. Because it was so widely recognized, at least during the Classic period, local Oaxacan elites on the Pacific Coast may have differentially adopted this ballgame tradition in attempts to legitimate their authority and enhance interpolity interactions (Zeitlin 1993:136).

3.6.3 Ballgame as Ritual vs. Ballgame as Sport

Throughout the literature, there is a somewhat dismissive attitude toward the idea of the ballgame as a competitive team sport. Though he has shifted his position since, Taladoire

(2000:20-21) once cited the sheer number of masonry ballcourts as evidence that the ballgame was more than just “a thousand-year old sports practice”. Day (2001:66-67) specifically refers to a “ritual ballgame” to distinguish it from more ordinary games and sports. While noting several

Colonial authors’ description of the game as recreational and “joyful” (though prohibited because the conquistadores believed it disguised a ritual practice), Macazaga Ordoño still makes the claim that “despite the enormous bets arranged by the lords of ancient Mexico and the excellent players who were, practically, professional, it was not solely a sport” (1982:50). A firm distinction is made between a juego por los dioses (ballgame for the gods) and a juego de

214 los hombres (ballgame for men). The first is considered to be the earthly recreation of cosmic events and the second a profane act (de la Garza 2000:53). Although this division does find support in the Colonial documents, this distinction may have more to do with Colonial attitudes and (mis)perceptions of indigenous ritual concepts and practices, rather than with there actually being distinct “sacred” and “profane” ballgames.

Here I want to bring in recent scholarship from the anthropology of religion and ritual.

Bell’s concept of ritual density addresses the oft-observed phenomenon that some societies, and even historical periods, seem to have more ritual, or appear to be more ritually dense, than others

(1997). Bell observes that the relationships between various rites or ritual activities “within a ritual system” and, subsequently, how these systems are or may be distinct from each other, are fairly undeveloped themes in ritual studies. Instead the distinctions scholars make between different ritual systems are more often a by-product of larger evolutionary or typological classifications of religion (Bell 1997:173). Yet the interactions between rites and their ritual elements as well as the different levels of ritual life are complex, and ritual density often depends on the context within which these actions occur (Bell 1997:174-209). Another variable in what defines actions as ritual is “ritual commitment”. As defined by Humphrey and Laidlaw (1994), ritual commitment refers to the intent of the actor when performing ritual actions. They posit that what makes ritual action distinctive is the relation between the intention of the actor and the action performed. In “doing” ritual, the actor has a specific idea of the right actions to perform and rules to follow. A person will remove their shoes both outside the home and outside the temple, but it is the setting of the temple that renders this action “ritual” and affects the attitude of the person performing it.

It may be reasonably assumed that games played in the formal masonry courts attached to

215 ceremonial complexes were the most ritually dense and restricted. Other games, either played in these same spaces or elsewhere with the use of portable stone markers, may have had equal ritual importance but less ritual density, perhaps with certain ritual elements left out or modified all together. Ritual elements may have been varyingly incorporated into each event perhaps as part of before and after celebratory rites, or as the whole event itself. Different elements would have been included and excluded depending on the requirements of the participants and their commitment to the ritual in either action, thought, or both, all of which can be independent of the ritual’s (and its constitutive elements) meaning. The degree of ritual density present in a ballgame event can be attributed to the motivations for sponsoring the game in the first place.

Even small, informal games could have been very ritualized (maybe even more so), just in different ways. An important lesson to draw from this is that variability in ritual density can account for the variability in the material culture of the ballgame.

In effect, the ballgame was neither strictly ritual nor sport, but often both at the same time. I would even go as far to say that it is not fruitful to keep the two separate. By restricting the ballgame to only being a mythic-religious event, a false dichotomy is needlessly created. A critical point is that ritual is not always religious, it can be ideological, political, and social as well – all aspects that characterize ballgames. Ritualized ballgames, even those performed in the service of religious cults, would still have provided occasions for individuals to socially interact, participate in economic exchanges through gambling and gift-giving, and express (or at least attempt to) their political authority through the religious and ideological symbolism associated with ballgame matches and attendant ceremonies (Zeitlin 1993:138). But saying so does not discount ballgames as physically challenging, emotionally-charged, and competitive sports that appealed to audiences of all social classes. In this perspective, the lack of clarity or rather

216 fuzziness between the sacred, i.e. ritual, and the profane, i.e. sport, is not a problem, but rather a clue to the game’s efficacy as it was viscerally experienced by players and observers (Stoll in press). Embracing the ambiguity provides an opportunity for studying how games can be both sport and ritual, how sport is ritualized, how ritual can be both practical and fun, and how games come to carry social, political and religious meaningfulness for participants and spectators.

As De Borhegyi (1980:3) succinctly states, because sports participation begins in childhood, we “cannot summarily declare that the ancient ballgames of Mesoamerica were purely esoteric religious events”. Ballgames would have ranged in intensity from informal pickup matches to large spectacles with huge crowds gathered to witness emotional events where valuables, the prestige of sponsors and players, and the outcome of the agricultural season was at stake (Baron 2006:8; Berger 2009:55-57; Stoll and Anderson 2017). Stern (1948: 71) rightly cautions that by overemphasizing the cosmological, agricultural, and warfare connections, we lose sight of the “appeal, direct and uncomplicated, which the game made through its sportive qualities”. After all, the Aztec-Mexica considered the gods of pleasure, Xochipilli and

Xochiquetzal, to also be patrons of the game. Baron (2006) importantly notes that Maya ballplayers were lords, captains, warriors, and other important men – not just ritual specialists.

Scenes on polychrome Mayan vessels are very revealing because they mostly show ballgame preparation, play, and rule or scoring disputes involving referees. The presence of a referee figure strongly indicates that not all games were predetermined events where the losers were always sacrificed. Baron (2006:10) argues that while at times ceremony was key, sport and entertainment were also important, and sometimes even more so.

217

3.6.4 Discussion

|The material evidence and current research attest to the fact that we must incorporate discussions about sports and competitive cooperation back into our analyses of ballgames in the

Prehispanic past. In linking gambling to the ballgame’s role in the development of sociopolitical inequality during the Formative period, Hill and Clark (2001: 338) make the excellent point that as an event the ballgame would have offered numerous opportunities for creating future social obligations among community members. Community organization of ballgames would have involved not the exchange of labor or goods, but social exchanges as well, creating cyclical obligations among households and local civic-religious officials.

Ballgames may have been a critical mechanism for the development of leadership in early Mesoamerican society as well. Fox (1994) hypothesizes that Mesoamerican elites competed for prestige and respect through competitive generosity and providing entertainment in the form of ballgame matches. In a similar fashion, Hill and Clark (2001) argue that the accumulation of symbolic capital and social prestige generated by hosting games evolved to heritable power through the differential control of material goods, court ownerships, and gambling debts that led to wealth and status differentiation. Hill and Clark’s model is insightful, but unfortunately casts only the elite as a primary force behind the promotion and promulgation of the ballgame tradition, thereby reducing the other social classes to passive receivers (Baron

2006). Arguably, the efficacy of the ballgame and its symbolism would not have had much salience or meaning across all social levels if it was only restricted to either elite players or spectators, or to events sponsored by elites. Such meanings would have needed to be actualized and embodied on a more routine basis perhaps through more informal exhibitions and game matches that still would have reinforced certain actualizations of particularly salient meanings

218

(Dornan 2004: 29). I argue that ballgame exhibitions, whether they were large spectacles sponsored by elites where players may have represented cosmic or divine forces, or simple affairs hosted by local communities for diversion, were events where social relationships were negotiated, contested and/or confirmed.

What the players of pelota Mixteca stressed repeatedly in the interviews were the social aspects of the game, the sense of community they derived from it, and the fun they had playing with friends and rivals in what they see as a rich and deeply important tradition integral to their identity as Oaxacans (Stoll 2015). Like the pelota Mixteca matches of today, different social groups in the Prehispanic past would have traveled in local and regional networks to attend ballgames and other celebratory activities, much as some individuals did with their patolli pieces and gameboards. This would have applied not only to players themselves, but perhaps also to spectators and other invested stakeholders. Even in our modern sports, people often follow their favorite teams on the road, traveling when they can afford to do so to attend important games, disseminating both culture and capital while constructing reputation and community through social ties made along geographically extended networks.

As archaeologists studying the ballgame, it is important that we do not lose sight of these critical facets of sports and games. While overtly ritual activities or elements are not as apparent in the modern Mexican ballgames, attitudes towards playing do approach ritualized practice; that is, playing the game is just as much about participating in a competitive sport as it is about indigenous concepts of reciprocal obligations and the moral qualities of players vis-à-vis the respect shown between communities as cyclical hosts and guests (Stoll 2010). If we think about

Mesoamerican ballgames in these terms, linked to systems of reciprocal obligation that also have their roots in the Prehispanic era (Monaghan 1990, 1995), then it is easy to see how ballgames

219 became an important social metaphor, a physically-enacted representation of the relationships of the individual to the community, and the community to the cosmos.

There is the question as to how much real effect ballgame outcomes would have had on everyday life. Baron (2006:44), citing Geertz (1973), argues that games – unlike warfare or conflict – do not change the status quo but instead are simply metaphors. Thus, sports and warfare are supposedly exclusive of each other. Accordingly, legendary ballgames, such as the game played to determine the ruler of the Toltec Empire, cannot be firmly relied upon as a model for historical competition. She argues against ballgames being played between political enemies as a way to decide social conflicts, given that game outcomes are unpredictable (Baron 2006:48-

49). On the other hand, Zych (2017) counters that games do have the ability to affect relationships between people via their direct engagement. Cheska (1979:227) asserts that while materials or resources are not produced or created though playing, the outcomes of games do rearrange them, especially as a result of betting. As the review of the ethnographic and ethnohistoric evidence has shown, indigenous games and sports did affect different dimensions of both every day and extraordinary life. That ballgame outcomes could and did have real material and social effects on Mesoamerican societies should be fairly conclusive.

What the preponderance of evidence does reveals about ballgames is that they played critical but highly variable roles in Mesoamerican communities across space and time. The importance and/or popularity of games waxed and waned as people became more or less invested in them. In some areas, ballgames survived for over 3,000 years, while in other places games were played for a few centuries before disappearing completely. Determining the degree to which games were part of political negotiation can be done through regional analysis of ballcourt distribution coupled with the regional settlement data. Moreover, understanding this distribution

220 and what patterns can be observed can aid in our interpretations of local and regional political organization and historical developments.

For example, where there are multiple ballcourts widely dispersed across the landscape and within the settlement hierarchy – i.e. at major and minor settlements – may indicate that the game was less exclusive and perhaps decentralized with respect to scheduling, game ceremonialism, and the audience composition. Different communities of varying size and composition would have alternatively served as hosts and guests. Ballgames and other ritual activities, including feasting events, would have been opportunities to negotiate social tensions through participation in a shared communal tradition among many different social status groups.

Travel to major settlements by minor elites or individuals from smaller settlements would have carried different sociopolitical meaning than travel by higher-status elites or individuals from major settlements, if they did so. The network of courts in the Cuyumapa drainage, for example, were used at different times of the year and likely by different populations, with those at minor settlements drawing in a wider and more mixed audience (Hendon et al. 2009).

At the other end of the continuum, where ballcourts are found in regional political centers or major settlements only, ballgames were likely tightly controlled by regional elites who had centralized authority over game ceremonies, scheduling, and who could participate. Regional elites would have also exercised authority over who could construct a ballcourt and where. In some instances where access was highly restricted, such as courts attached to palaces, ballgame audiences were likely composed of the elites or other political leaders of that settlement alone.

Ballgame ceremonialism and access was tightly controlled in many polities in the Maya macroregion during the Classic period, where ball playing was closely tied to kingship, and political legitimacy and ceremonialism.

221

Ballgames also served as political mechanisms for networking and competition among neighboring polities. This explains why we see some ballcourts located on the boundaries between regional polities and communities. Such courts were mostly used as boundary markers and as sites for carefully orchestrated events between distinct groups, either in conflict or for communal celebration. Boundary-marking rituals would have been the prerogative of local and regional elites. Communities would have participated in games as a way to affirm boundaries, evoke or evade conflict, and/or have neutral opportunities for social interaction. Similarly, ballgames that took place in courts located at major centers and/or polity capitals were also part of the political networking strategies for local and regional elites participating in peer-polity networks. The circumstances regarding sponsorship and organization of the games as well as who was invited to see them, however, would likely have been different for those taking place at courts in the major centers than at boundary courts, since the audience would have included elites or foreign dignitaries from other polities who had to be invited into the settlement.

Boundary courts were more often neutral spaces, like those found on the boundaries of

Postclassic Mixtec polities that were formerly part of Classic period settlements.

Obviously, the rituals and ceremonies conducted during ballgame exhibitions are not mutually exclusive, and all games would have been opportunities for social interactions and the negotiation of power relationships, whatever form they may have taken. What would have varied is the degree to which people of different social classes could participate in these events and how much control the elites would have had over them. Elites who exercised control over the game would have exerted this power through their role as owners of the courts, hosts of the games, and/or team sponsors. At the same time, by its very form the ballcourts restrict visibility, and their design and location within the spatial layout of settlements affected audience size and

222 composition (Stark and Stoner 2017). Thus, even in decentralized or heterarchical polities, elites probably still held primary control over game scheduling, but games would have been more accessible to many different social status groups.

Of course, not every regional distribution of ballcourts fits neatly into the situations described above. There are cases where there are several ballcourts in a region, but they are concentrated in only a few settlements or even within one settlement, such as at El Tajin with its

17 different ballcourts, or in Jalisco, where the courts are only found in the core settlement around Lake Arnel. Moreover, it has been suggested that the individual courts at El Tajin belonged to specific elite lineages, while in the Lake Arnel example the courts are associated with different social segments living in both upper and lower status precincts (Weigand 2000).

In both cases we have multiple ballcourts in a single settlement. However, in the former the ballgame was a legitimation strategy for competing elites; the games themselves were likely only attended by higher status members of the community. In the latter, the presence of courts in lower status precincts indicates that games were more accessible to a wider audience, but there is still some degree of hierarchy as more formal courts are only found in upper status precincts.

Differences in how the game may have figured into local and regional sociopolitical relationships, then, can be further assessed by looking at how ballcourts are distributed, and not just where. At Cerro Palenque in Honduras, for example, a single ballcourt served a

“community” of around 500 houses and was located among other special-purpose architecture surrounded by small, nucleated mound groups. Meanwhile, in the Cuyumapa drainage there was one court for every 75 residential buildings, with courts occurring among both overlapping and non-congruent small- and large-scale mounds groups (Joyce and Hendon 2000:153-154). Again however, this interpretation focuses mostly on formalized court spaces, and does not account for

223 informal or non-court spaces where games were played. Much like modern sports today, and even pelota Mixteca, ballgames also occurred in spaces outside of the formalized ballcourt structures – as further evidenced by the portable ballgame markers references earlier. While games held in formal court spaces may have been inaccessible to a majority of people, ballgames themselves probably were not.

3.7 CONCLUSION

Linked with concepts such as war, life, death, fertility, and cosmology, but also with community identity, social negotiation, and political negotiation, ballgames were enacted in various forms as well as understood and experienced on many different levels. The specialized courts in which games took place were also heterogeneous; while the general outline was the shape of a capital I, there still existed a great variety of architectural elements and designs.

Courts were also used as spaces for performance of social festivities or political events not related to the game, and West Mexican ceramic ballcourt models provide excellent examples of the some of the ceremonies and celebratory activities that took place within them (Ramos

2012:99-100).

The dominant paradigm assumes that the Mesoamerican ballgame was a ritual and political practice of the elite alone. As a result, we have often lost sight of the communal role this game may have held as a competitive sport. Archaeological, ethnohistoric, and ethnographic evidence suggests that ballgames were important in community-building, and not simply a function of elite political theater (Stoll and Anderson 2017). This brings us to a new understanding of Mesoamerican ballgames as competitive and integrative communal traditions that are not easily reducible to grand rituals or a simple sport. At its core, the Mesoamerican ballgame is a sport that involved incredible physical skill and mental strategy, while also

224 embodying communal discourses of power and identity, much like many of the games in the indigenous cultures of the Americas as well as in sport cultures generally around the world.

In this chapter, I explored the archaeological, ethnographic, and material evidence using a poststructural and historic-contextual approach, critically examining the research on past and contemporary practices of the ballgame itself. Arguably, when looking at ballcourts and ballgames in different regions of Mesoamerica, any assumptions about the meaning, significance, or function of ballgames should be evaluated for their appropriateness to the social reality of those communities. The goal should be to understand how and why ballgames served important social roles, as well as to consider the more pragmatic aspects of sportive play. We should be looking at the processes of political negotiation between and among communities that were made possible through participation in the ballgame as hosts and guests, rivals and supporters, players and audience members, and how those processes lead to the construction of communal senses of identity inherent to the region and people involved.

Having established that the ballgame was an important social field through which communities and individuals both interacted with each other and even negotiated political relationships, in the following chapter I provide the geographic and sociopolitical context for the dissertation. The interesting spatial patterns and different types of ballcourts observed in Nejapa, described further in Chapters 6 through 8, suggests that in this heterogenous frontier zone, the ballgame may have been one way in which people mediated relationships, expressed community identities, and interacted with the foreigners traveling through their territories. Chapter 4 explores the current evidence on the historical development of Nejapa and Oaxaca, helping to place the ballgame data from here within the context of the overall development of this important sport.

225

CHAPTER IV

OAXACA: THE CROSSROADS OF MESOAMERICA

Here in these central valleys of Oaxaca, among the peaks and pinnacles of the Sierra Sur, five thousand feet above sea level, the smallest pocket of soil is cultivated, for literally anything will grow in it… Miles 1961:97

O, Tierra del Sol (land of the sun) Suspiro por verte (I yearn to see you) Canción Mixteca, José López Alavez, 1915

4.1 INTRODUCTION

The state of Oaxaca is a rugged, mountainous region in the southern highlands of modern

Mexico (Figure 4.1). This simple sentence – which opens many an Oaxacan scholar’s background chapter – elides much of the beauty, wonder, and magic that not only inhabits every corner of the state, but also emanates from the landscape, local culture, and even the people themselves. Oaxaca is at times a strange and enigmatic place, and one of few corners of the world where I believe that ghosts were real, and where blessings from shamans reveal

Prehispanic deities hidden in the clouds. Its dramatic geography, shaped by the converging

Sierra Madre Oriental and Occidental mountain chains and cut by numerous narrow valleys and riverine canyons, engenders a wide-range of ecological and climatic zones found at varying elevations. In fact, seven of Mexico’s eleven climate zones are found here. The diversity seen in the flora, fauna, and topology of Oaxaca is reflected in its people. One of the few states in

Mexico with a majority indigenous population (with 53% of the country’s indigenous people in

226 total), Oaxaca is home to 16 officially recognized ethnolinguistic groups, and certainly many more that are not.

Figure 4.1: A feral cow gazes out onto the landscape, Santa Ana Tavela, Yautepec District, Oaxaca.

Though the analytical focus of this dissertation is, more or less, on a singular example of architecture within a specific geographic and cultural area, regional scale studies of ballcourts still require a broad and inclusive perspective on the larger-scale social, political, and economic patterns. The ever-emergent and dynamic social nature of ballcourts as sites of communal activities, through feasting, ceremonies, or the games themselves, necessitates that we locate them squarely in history. Therefore, in this chapter I began by exploring the greater context of the sociopolitical developments in Oaxaca. Of special concern here is the paucity of investigation in lesser-known regions, as the geographical limits of our research conditions our

227 knowledge of Prehispanic history in Oaxaca. This sets the stage for the following section, where

I present the archaeological and historical data on Nejapa from the Early Formative through to the Postclassic period. The archaeological data comes from excavations, surveys, and material analyses conducted by the Nejapa/Tavela Archaeology Project (PANT in Spanish), directed by

Dr. Stacie King from 2009-2013. These data provide the necessary context through which I explore the role of the ballgame and ballcourts in intra- and intercommunity relationships in

Nejapa in later chapters.

4.1.1 A Note on Terminology

As archaeologists, we often wrestle with the different scalar and sociocultural terms we use to discuss our research, specifically with respect to what descriptors are most appropriate for clearly imparting our meaning and intentions. Clarity in the use of our terms – which for the most part are really just heuristic tools that at best incompletely capture the complexity of what we’re making reference to – helps to avoid confusion when discussing the data and interpreting it. This is especially important when we are attempting to do a regional analysis (Chapter 2) – what are the analytical units we are working with? To wit, there are several terms available for discussing various geographic or spatial scales; many are often used interchangeably, such as area, region or zone. Because Mesoamerica is a geocultural zone (defined by Paul Kirchoff 1966

[1943], for all intents and purposes an outsider) consisting of varying nested spatial scales, how we discuss the different ‘areas’ that are distinguished as much by geography and ecology as they are by groups of people and material cultures, can sometimes become overcomplicated due to the usage of similarly meaning words.

228

For ease of discussion, I refer to the major recognized geocultural zones within

Mesoamerica as macroregions; this includes North Mexico, Central Mexico, West Mexico,

Guerrero, Oaxaca, the Gulf Coast, and the Maya homeland. Within these macroregions are smaller, distinguishable geocultural regions. In Oaxaca, for example, we recognize seven8 such regions that I discuss further on in this chapter. Finally, I use the term subregion for smaller, but still distinct, ecological-cultural-ethnolinguistic zones within these regions. In the parlance I use here, Nejapa would be a subregion within the Sierra Sur region of the Oaxaca macroregion. In defining my analytical terms in this way, however, I am not assuming that the boundaries between subregions, regions and/or macroregions were ever neat or clearly demarcated. Frontier zones were numerous, and boundaries overlapped or were even nonexistent. Nor do I see these geographic units as isomorphic with the lived reality of Prehispanic peoples since they are in many respects modern constructions imposed on patterns of material distributions from the past.

Another issue I see in defining terms is how to properly describe the urbanized political organizations that characterized Prehispanic society(ies). Archaeologists have previously borrowed the terms ‘city-state’ and ‘polity’ from Classical archaeology, derived from the Greek polis. While city-state seems to be appropriate, when we examine the social and cultural data more closely we see that this term does not in fact adequately capture large-scale complex political units in Mesoamerica. One obvious problem is that not all Mesoamerican cities qualify as states. Nor does city-state accurately reflect how these units were organized socially, economically, and politically. As friends and I have animatedly discussed after several rounds of

8 The designation of seven cultural regions in Oaxaca is somewhat debatable, and in many ways they do reflect modern divisions and politics. Yet there is also archaeological, historical, and ethnographic evidence to support a rough approximation of seven (or even eight) identifiable regions that can be fuzzily defined by geography and culture. 229 beer over the course of several nights in Oaxaca City’s cantinas, we need a more Mesoamerican term to describe a Mesoamerican phenomenon.

I propose that we use the words that the people themselves used to describe their settlements. In the Mixtec language, large-scale political units were known as yuhuitayu, a compound combining the words for mat reed and (royal) couple (Blomster 2008:22), while the

Zapotecs used quechecoquie, another compound word that signified a community (quiche) with its own ruler (coqui) (Zeitlin1994:275). While similar in concept, the slight semantic differences do express differing concepts of how social relationships are organized. While many Oaxacan researchers use cacicazgo or señorio, these are Spanish colonial terms and are problematic. The first term derives from the Taino word cacique and could more properly be understood as a chiefly position. The application of this term to Mesoamerican leaders by the Spanish was probably meant to diminish or delegitimize their positions as rulers. The second term is reflective of a more Western understanding of sociopolitical communities. For that reason, I will use the appropriate terms – and quechecoquie in general –when referring to polities in Oaxaca, and the Nahuatl word altepetl (water-mountain) when referring to polities outside of Oaxaca. I use these terms in order to more accurately capture the local, sometimes ethnolinguistically- based – although there were also many that were multiethnic and multilingual – political entities that encompassed not only the actual settlement, but the neighborhoods, family lineages, land properties lying outside the main settlement, and the allied smaller dependents or hamlets that make up the larger community, as well as the ruling couple that represented that community in their royal personages. In this way, these quechecoquie or altepetl were not just physical places, but they were also embodied and autonomous political entities (Zeitlin 1994:276).

230

4.2 A BRIEF INTRODUCTION TO ARCHAEOLOGY IN OAXACA

Figure 4.2: Map of Oaxaca and the different regions and subregions mentioned in this chapter.

Oaxaca (Figure 4.2) has paradoxically been the intense focus of archaeological research for decades while also considered to be a lesser or “backwater area” that did not contribute much to the overall historical developments in Mesoamerica (Blomster 2008). While macroregions such as Central Mexico, the Gulf Coast (in particular Central Veracruz), and the Maya homeland are traditionally seen as centers of Mesoamerican culture and politics, the contributions made by

Oaxaca’s Prehispanic societies have received little attention. Yet archaeological evidence has increasingly shown that these societies were significant participants in pan-regional economic and sociopolitical networks (Blomster 2008; Balkansky 1998 Balkansky et al. 2004; Flannery

231 and Marcus 1983; Flannery and Marcus 2000; Joyce 2010; Winter and Santiago 2014). Oaxacan societies urbanized early on, since at least the Middle and Late Formative period (900 BCE – CE

200), and had their own traditions of hieroglyphic writing, monumental architecture, and fine crafts (Joyce 2010; Joyce and Winter 1996).

Although the Central Valleys and Mixteca Alta have been heavily investigated since the early 20th century, research has been more limited outside of these regions. This means that there are large gaps in our knowledge of Oaxacan history, leading to a somewhat skewed vision of the political and social events in this macroregion. Only since the 1970s have archaeologists really began looking into these lesser-known areas. Research in Cuitlácan Cañada by Spencer and Redmond was one of the first major interdisciplinary projects to explore this unknown region, shedding light on a possible early expansionist Zapotec state centered at Monte Albán.

One of the longest and currently ongoing investigations has been the Rio Verde Archaeological

Project, directed by Arthur Joyce, which has enriched our knowledge about the political history and social developments on the Costa Chica.

Perhaps the next best documented area is the southern portion of the Isthmus of

Tehuantepec. While intermittent research has taken place there since the late 1950s, salvage projects carried out from 2002 – 2007 by INAH archaeologists greatly expanded our knowledge of this region. Over 100 new sites were documented, including the impressive Formative-period city of Tres Picos (Winter et al. 2008; Winter et al. 2014). Meanwhile, despite its proximity to the Mixteca Alta, the Mixteca Baja was only briefly surveyed and investigated in the 1980s

(Winter 2007). Although there are some cultural similarities with the Alta, the Ñuiñe of the Baja were a distinct people with their own particular style, traditions, and histories (Blomster 2008:14;

Winter 2007). Intense investigation will likely show that the history of the Baja is just as

232 complex and dynamic as the Alta, but also distinct. We know, for example, that the Ñuiñe experienced a cultural florescence in the Classic period, centered at the major yuhuitayu of Cerro de las Minas (Winter 2007) Moreover, colonial lienzos from the Baja depict a large number of important Postclassic settlements that, through rigorous survey, could be linked to archaeological sites on the ground.

Within the last two decades, more research projects have been conducted or are being conducted in other, even lesser-known regions of Oaxaca. The mountainous Mixe region to the east and northeast of the Central Valleys is better known ethnographically, and not much is known about the Prehispanic period. Recent rescue projects directed by Marcus Winter of

Centro INAH Oaxaca, have documented the presence of several small urban centers that were densely populated and internally complex (Markens and Winter 2014). Incredibly life-sized clay statues of humans, jaguars, and other creatures discovered in the Cueva de Rey Kong-oy near the site of Chuxnabán (dated to the Late Formative), demonstrates the need for further investigation in the Sierra Mixe (Winter et al. 2014). Very little research has been done in the Chinantla, a mountainous area on the northern extreme of Oaxaca close to the border with Veracruz, as well as the Mazateca, another area of highland and river valleys just northwest of the Chinantla

(Delgado 1965; Munn 2014; Winter 2008). In fact, until his death in 2014, Henry Munn had been the only scholar actively working in the Mazateca – and he was not an archaeologist but a writer and ethnopoet who had spent years among the Mazatecans. Finally, Danny Zborover’s

(2014) dissertation research in the highland area of the Chontalpa fleshes out our knowledge of this understudied region that is located in the south and southeastern area of the Sierra Sur, adding to the previous efforts of Peter Kroefges (2001, 2004) who had worked on the Pacific coastal area of the Chontalpa.

233

Meanwhile, recent and ongoing research in eastern Guerrero, close to the border with

Oaxaca, will inform us about the social interactions between the multiple cultural groups that lived in this frontier region; indeed, peoples living in western Oaxaca and the Mixteca Baja may have had more in common culturally with the societies of eastern Guerrero – who likely spoke local variants of Mixtec and Amuzgo, both found in Oaxaca, in addition to Tlapaneca and Nahua

– than with the Central Valleys or even the Mixteca Alta. Investigations in multiethnic and multilingual frontier or boundary zones in Mesoamerica will shed further light on intra- and interregional social relationships more broadly. One of the most recent research projects to do so is the Nejapa/Tavela Archaeology Project (Proyecto Arqueológico Nejapa/Tavela or PANT in

Spanish). In 2009, Dr. Stacie King initiated the first investigation of the Prehispanic and

Colonial history in the Nejapa subregion. Nejapa was (and still is to a degree) a multiethnic and multilingual frontier region through which multiple groups traversed as they followed interregional trade routes. Previously, our only sources of information on Nejapa were colonial documents from the 16th, 17th, and 18th centuries. Over three field seasons carried out in 2009,

2011, and 2013, the PANT team has documented and registered 162 sites in a 192 km2 area. I explore this research and place it within the overall context of Oaxaca further on in this chapter.

Expanding research into the understudied regions, as well as returning to those areas that have not seen investigation in some time (i.e. the Mixteca Baja, the Cuitlácan Cañada, and

Chinantla regions), will help enrich our understanding of Prehispanic Oaxacan history as well as answer many long-standing debates among Oaxacan archaeologists. For example, the ethnohistorical and colonial histories describe the conquest of the Isthmus and several other areas in Oaxaca by Zapotec armies in the Postclassic period. According to these same documents, the royal lineage from Zaachila in the Central Valleys left with a large military force and embarked

234 on a series of conquest campaigns towards the Isthmus of Tehuantepec. During their march, the

Zaachileño armies established settlements and mountaintop fortresses along their route (King and Zborover 2015; Oudijk 2000, 2008). Additionally, these same sources claim that Nejapa, the geographic focus of this dissertation, was conquered by the Zaachileño Zapotecs during their march to the Isthmus. Several mountaintop sites in Nejapa have even been described as Zapotec fortresses, including Cerro de la Muralla and Cerro del Convento, among others (King et al.

2014). As I will show later in the chapter, however, the evidence for conquest in Nejapa is not as definitive or obvious as the histories suggest. Indeed, signs of conquest or subjugation in other regions assumed to have been under Zapotec control are not evident or are lacking entirely, while in other areas Zapotec penetration from the Central Valleys is much clearer (Balkansky 2002;

Barber et al. 2013; Levin 2013; Spencer and Redmond 2001a, 2001b, 2004; Workinger 2013).

Given how much our interpretation of Oaxacan history is shaped by the idea of an aggressive

Zapotec state, this is problematic. Undoubtedly, much work remains to be done in order to understand fully the sociopolitical developments and historical trajectory of Oaxaca’s various regions.

4.3 HOW DID WE GET HERE? A SHORT SOCIOPOLOTICAL HISTORY OF OAXACA

Though largely considered to be uninvolved in the transformations and larger political developments because of its supposed extreme balkanization (especially in the Postclassic) as we will see in the following sections, Oaxaca did not lie “outside” of history (Blomster 2008:3).

Instead, the available ethnohistorical and archaeological evidence point to both early developments in social complexity concurrent with other, better known centers of Mesoamerican culture, and the long and deep involvement of Oaxacan societies in international alliances and

235 long-distance trade networks stretching from Central and West Mexico to the Guatemalan highlands (Figure 4.3). Indeed, extensive archaeological projects during the 1960s and 70s demonstrate that human beings have been living in Oaxaca since the Paleoindian period, at least

8,000 – 10,000 years ago. The transition to settled village life began in the Early Formative period, around 2000 to 1500 BCE, if not earlier. Recent research by Guy Hepp (2015) at the site of La Consentida in the Lower Rio Verde Valley documents one of the earliest confirmed sedentary villages (calibrated 1947 – 1530 BCE) ever discovered in Oaxaca. The evidences also reveals nascent interregional trade relationships as well as some of the earliest public buildings constructed in Oaxaca, and indeed probably so for the Pacific coast of Mexico in general.

236

Figure 4.3: Map of Oaxaca and some of the sites mentioned in the text.

Throughout the Formative period, the predominant settlement pattern consisted of small groupings of houses clustered around a centrally-located head town, suggesting a high level of interaction and integration between different distinct settlement clusters (Barber 2005;

Kowalewski et al. 2009). While these head-towns would typically have more mounds than other smaller sites, indicative of some hierarchical organization tied to settlement size, public architecture was still generally open and accessible. More than likely, this type of arrangement suggests that incipient power was coalescing around specific local lineages, though sociopolitical organization on a whole was still communally-based. As population increased and settlements became more complex, these smaller sites would have developed into neighborhoods or barrios.

237

Culturally, the suite of shared symbols and new artifact types that began to appear throughout the

Formative period at different archaeological sites throughout Mesoamerica also appear here in

Oaxaca (Blomster 1998:309; Joyce 2010:88), while the evidence for feasting and ceremonial gatherings involving large groups of people increases.

The founding of Monte Albán in 500 BCE. (Figure 4.4), one of the earliest urban sites in

Mesoamerica, marked a profound social transformation in Oaxacan settlement patterns (Barber and Joyce 2006: 223; Blanton 1978: 39; Joyce 2010: 118-119; Joyce and Winter 1996: 35).

Only a century or two after the founding of Monte Albán, urban centers such as Monte Negro,

Huamelulpan, and Cerro Jazmín appeared in the Mixteca Alta (Balkansky 1998; Joyce and

Winter 1996: 39 Perez-Rodriguez et al. 2018). By 200 BCE, this shift to urbanism was well established through the Oaxaca macroregion, with competing centers established everywhere from the Mixteca Alta down to the Isthmus of Tehuantepec (Balkansky 1998a; Balkansky

1998b; Balkansky et al. 2000). The sociopolitical developments that had emerged in the latter half of the Formative period continued through the Classic period as Oaxacan and Mesoamerican societies became increasingly more urbanized. However, we observe a clear difference between the Formative and Classic periods in the decline of corporate, communal power, and the concurrent rise of an aggrandizing nobility who negotiated power and position amongst themselves in a heavily competitive political landscape (Joyce 2010). The crystallization of social hierarchies is reflected in the increasing spatial distancing between households of different status groups and the decreasing access to what were formerly communal sacred spaces.

238

Figure 4.4: The Main Plaza at Monte Albán, looking north from the South Platform.

Such developments meant that commoners were more and more excluded from leadership positions at the supra-community level. Despite this exclusion, non-elites still had access to various social roles through which they could negotiate status and power within their communities and/or status groups (Joyce 2010:224). For example, evidence shows that specialization in craft production was organized at the household- and community-level. These production and trade relationships were not imposed on them by the ruling nobility or other political authorities, but instead were pursued independently for the most part. In many aspects of their lives, not much was different for people of lower or non-noble statuses than how it was in the Formative period. Where they appeared to be most affected by the machinations of nobles and rulers was in the increasingly restricted access to particular rituals and sacred realms (Joyce

2010:225).

The Classic to Postclassic transition was also marked by major changes in demography, economies, and sociopolitical relationships, both in Oaxaca and Mesoamerica overall. In comparison to the relatively rapid end (around 200 years) of powerful Formative period urban sites, the collapse of Classic-era settlements and political systems took place over several centuries, from about CE 650 – 1100 (Byland 2008:331-332), while in other areas this transition did not happen at all, such as eastern Guerrero (Blomster 1998). The destruction of the civic-

239 ceremonial center of the powerful and widely influential altepetl of Teotihuacan around CE 650, arguably marks the beginning of this dramatic social reorganization. Even though the city itself continued to be occupied into the 9th century, albeit at much reduced population levels, the collapse of its political and ideological civic-ceremonial core had a reverberating effect throughout Mesoamerica (Blomster 2008:6).

While the fall of Teotihuacan was impactful, it was not the sole cause of the Classic to

Postclassic collapse and reorganization – varying combinations of different political, economic, and even environmental factors were at play. While environmental factors are not so obvious in

Oaxaca, communities here were certainly affected by the political changes slowly sweeping across Mesoamerica. Even the powerful Monte Albán, which had survived the Terminal

Formative reorganization and had continued to exert considerable political influence through most of the Classic period, was not immune to these dramatic transformations, experiencing political instability and turmoil for almost a century before the main core of the city was finally abandoned around CE 850. Towards the end of the Classic period, the increasing power of other quechecoquie such as Lambityeco, Zaachila, Mitla, and Jalieza, continually challenged the political dominance of Monte Albán. The rise of these new competing centers, coupled with

Monte Albán’s eventual and final collapse, clearly indicates that political alignments were already shifting, and new ones were being created as newly ascendant elites negotiated a transforming political landscape (Blomster 2008:17).

Only a few centuries after the abandonment of Monte Albán, a dramatic conflict between several urban centers in the Central Mixteca Alta led to the complete collapse of some of the yuhuitayu there. The saga, known as the War of Heaven (also the War that Came from Heaven or War of the Stone Men), is famously depicted in the Mixtec codices and likely involved three

240 different Mixtec factions (Balkansky et al. 2000; Pohl 1994; Pohl and Byland 1990; Pohl and

Byland 1994). How much of this conflict was localized only to the Nochixtlán Valley (Williams

2009:44, 99), or had spread throughout the Mixteca Alta, remains to be seen9. While today we do not quite understand the significance of the War of Heaven in Mixtec history in its entirety, it is clear that the outcomes of this conflict, i.e. which ruling lineages survived and held on to power and which did not, were critically important to the political legitimacy of later Postclassic

Mixtec kingdoms (Byland and Pohl 1004: 14).

Understanding the Classic to Postclassic transition in Oaxaca is somewhat complicated by chronological issues that unfortunately also affect how we interpret data from the Postclassic as well. One of these issues is the calibration curve for radiocarbon samples, leading to wide date ranges when analyzing samples from possible Postclassic contexts. This can make it difficult to create a settlement chronology and understand the sequence of occupation and abandonment in the area. The other issue has to do with the ceramic chronologies developed for many areas in Oaxaca; oftentimes, the entire Postclassic period is a single ceramic phase, such as the Natividad phase in the Mixteca Alta and Monte Albán V in the Central Valleys. As with the calibrated radiocarbon dates, large ceramic phases make reconstructing past sociopolitical landscapes difficult. Moreover, changes in ceramic styles and forms between the Classic and the

Postclassic and how such changes occurred is not well understood. Currently, I am unaware of any active projects working on refining Postclassic ceramics, but it would be worth the time and effort. Above all, the range of responses to the sociopolitical transformations that were occurring across Oaxaca after 700 CE is clear evidence of the complexity and variety of integration and contact between the different regions and ethnolinguistic groups (Blomster 2008:19).

9 Jansen (1988, 1990) even argues that the War of Heaven is actually about the fall and eventual collapse of Monte Alban. 241

The political upheavals of the 8th through the 11th centuries lead to some dramatic changes in settlement patterns throughout the Oaxacan macroregion. Some areas were completely depopulated, while others were abandoned entirely. By the 13th century, population levels had recovered and there was a huge boom in settlement, with people expanding into even the most remote areas (Blomster 2008:20; Winter 2008:394). Social relationships in Oaxaca during the Postclassic were supposedly characterized by deep factionalism and the constant negotiation of competition and cooperation (Pohl et al. 1997: 224), consisting of multiple ruling families in a complex network of interdependent relationships and nested hierarchies (Joyce

2010; Pohl 1994: 69-70). There was apparently constant flux in the regional Postclassic political structures as these smaller, “balkanized” communities interacted through varying means of kinship networks, martial alliances, trade relationships, and warfare (Balkansky et al. 2000: 379-

380; Blomster 2008; Kowalewski et al. 2009: 315). One area of life where we perhaps see the most change from previous periods was in the economies of Postclassic societies, affecting members of all status groups. While markets had always been an important facet of

Mesoamerican life, they became considerably more formalized in the Postclassic, with an increasing role for merchants as well as new forms of currency. The mass production of many goods meant that prestige objects that formerly had been restricted to the elites were available for sale as “commercial luxury goods” (Blomster 2008:27-28).

However, there were also strong continuities from earlier time periods, including the neighborhood-based organization of settlements; and the rituals and religious ceremonies that were significantly important arenas for accessing and legitimizing political power for both elites and other social groups (Clark 2004; Elson 2006; Finsten et al. 1996; Flannery and Marcus 1976;

Gonlin and Lohse 2007; Hamann 2002; Hutson 2002; Joyce 2004; Joyce and Weller 2007; Joyce

242 and Winter 1996; Marcus and Flannery 1994; Masson and Orr 1999). Conflict and warfare were also still common in Oaxaca, with both local skirmishes between neighboring communities and long-range conflicts between communities located far apart: Tilantongo against Teposcolula

(both yuhuitayu in the Mixteca Alta) and the Central Valleys; the community of Tejupan versus various communities of Chochoteco-speakers in the northwest Coixtlahuaca Valley; Mitlatongo against the allied forces of Tlaxiaco and Tututepec; and finally, Tututepec warring against

Tamazola (Spores 1967:14). However, except for a few instances, most of these conflicts were not over territorial acquisition but rather boundary disputes (if close to each other) and/or control over trade or alliance networks.

Though often described as balkanized, the extensive alliances and trade relationships between Oaxacan communities meant that events occurring in one area could dramatically affect events elsewhere. Such is the case for Nejapa in the Postclassic. In the Central Valleys, several newly established urban centers emerged into the vacuum of power left by Monte Albán during the Terminal Classic (CE 850 – 1000) and Early Postclassic (CE 1000 – 1200). The ruling dynasty at one of these centers, Zaachila, exerted widespread influence in both regional and international politics through globalized trade networks and social alliances that extended from the Oaxacan coast to Central Mexico. Coqui Cosijoeza 6 Water and his son Quixicayo, along with their retainers and followers, left Zaachila at some point to begin a series of conquests towards the Isthmus (CE 1280-1440). Along the way, they supposedly set up villages and established fortresses along the way in primarily Huave, Mixe and Zoque-controlled territories.

These villages and fortresses were then used to control the primary commercial route from the

Soconusco and Coatzacoalco regions to the Central Valleys, creating a Zapotec-dominated corridor (Oudijk 2008:106).

243

According to these native histories, the fortresses were built at Quiavicuzas, Quiegolani,

Quiechapa, and, most importantly, Nejapa. In fact, the histories also claim that the entire Nejapa subregion had been conquered by the Zaachileños (Oudijk 2008:106). Eventually, a dynastic struggle in the 15th century after the death of Coqui 1 Grass lead to a split in the noble lineages at

Zaachila, an event which would have profound effects on the regional politics in Oaxaca. Half- uncle to the previous ruler Coqui Cosijoeza 6 Water, Coqui 1 Grass’ death caused a succession crisis for Zaachila’s royal rulers. Coqui Cosijopii then led his followers and many allies to establish a quechecoquie at Tehuantepec before or after CE 1440 (Oudijk 2008:104), once again traveling through the Nejapa subregion and reasserting their dominance there.

4.3.1 Terra (Somewhat) Incognita: History from the Regions Around Nejapa

Much of the history presented in the previous section is based on ethnohistorical, colonial, and archaeological evidence from the Central Valleys and Mixteca Alta, since these have traditionally been the areas of focus for research in Oaxaca. Research from other regions, however, complicates this picture of Oaxacan history. Research from the Lower Rio Verde

Valley, for example, challenges some of the historical narrative with respect to the transition to sedentary village life, the development of social complexity and urbanism, and even the question of Zapotec imperialism during the Formative and Postclassic periods. In this section, I explore what is currently known about the regions immediately surrounding Nejapa: the Sierra Mixe

(Sierra Norte), the Southern Isthmus, and the Chontalpa (Sierra Sur). This gives the broader historical context to the evidence from Nejapa that will be described later in this chapter.

To the north of Nejapa lies the Sierra Mixe subregion, which includes a high mountainous zone and a lower piedmont and river valley zone that here slopes down east to the

244

Isthmus of Tehuantepec. Archaeological work started here since sometime in the 1950s and continued sporadically through to today, and we have good documentation of early and continuous occupation here. Ceramics from the site of Juquila Mixe, for example, date to the

Late and Terminal Formative periods, and resemble forms from both the Isthmus (the Kuak phase 200 BCE – 1 CE) and the Central Valleys (Danibaan to Nisa phases 500 BCE – 200 CE)

(Hutson 2014; Winter 2008). Other materials conclusively demonstrate that several Late

Formative sites, such as Chuxnabán, Juiquila Mixes, and San Marcos Móctum, continued to be large, important urban centers through the Classic and Postclassic periods (Markens and Winter

2014:290; Winter 2008:407-410). Additional sites that were established in the Postclassic are now covered by modern Mixe settlements, including Totontepec and Tlahuitoltepec, among others (Winter 2008:410). Still, much remains unknown about this area because of the paucity of research here.

The Isthmus of Tehuantepec is the next best studied region outside of the Central Valleys and Mixteca Alta. People would have traveled through the Nejapa subregion on their way here.

The Isthmus is currently split between two modern Mexican states – Oaxaca and Tabasco. The evidence discussed here relates to the Southern Isthmus, that portion which pertains to Oaxaca and geographically includes an area of low mountains bordered by the Gulf Coastal Plain to the north, and the flat, dry Pacific coastal plain to the south (Winter 2008:410-411). This geographic variability is matched by the ethnic and linguistic diversity of the region, with Zapotec communities dominating the coastal plains, the Chontales living in the mountains to the west,

Zoques villages to the east and northeast, Mixe villages in the north and west, and finally small groups of Huaves living along the Pacific Coast. Urbanized sites first began to appear in the

Late Formative. Material assemblages are from the Formative period are mixed. While we do

245 see a closer form and stylistic affinity with Chiapas and the Gulf Coast than say the Central

Valleys, there are also multiple lines of evidence for a distinctive and local Isthmian culture

(Winter 2008:411-412). One of the most important sites during this time period was the immense urbanized quechecoquie of Tres Picos, which had around seven or nine ballcourts, the most of any known settlement in Oaxaca (Marken and Winter 2014).

After a brief period of abandonment at the end of the Late Formative, new communities were established in the Classic period along the Ríos Los Perros and Tehuantepec. Along with these new communities were novel local ceramic styles, including Maya-style figurines that resemble those from Jaina. Interestingly, communities along the Río Tehuantepec, lying about

23 km to the southwest of the Río Los Perros, seem to have participated in a different ceramic complex altogether (Winter 2008:412). People mostly abandoned the Southern Isthmus towards the end of the Classic period, and the apparent paucity of settlement on the coastal plain during the Early Postclassic (Aguadas phase, CE 900 – 1300) lead many to assume that population levels did not recover here until the Late Postclassic (Ulam phase, CE 1300 – 1521). Recent investigations by Winter and other archaeologists in the 2000s, however, have documented a substantial Early Postclassic occupation at the site of Paso Aguascalientes, located at the juncture of the Ríos Tequisistlán and Tehuantepec (Winter 2008:413).

By the Middle to Late Postclassic, the arrival of the Zaachileños – again via the Nejapa route – dramatically impacted the sociopolitical landscape of the Southern Isthmus. They first attempted to take control of the Isthmus and its resources in the 1300s, supposedly establishing fortresses such as the remarkable Guiengola (Oudijk 2008) and the recently discovered Cerro

Negro, near the modern town of Jalapa del Marquez at the very western end of the region

(Winter et al. 2008). According to Burgoa’s (1934 [1674]) account, the invading Zaachileños –

246 this time the rebellious split-off lineage led by Coqui Cosijopi and his followers – later successfully established their royal quechecoquie at Cerro Padre Lopez in the mid-15th century near modern-day Tehuantepec. During their military campaigns, they pushed some groups such as the Huaves out of their traditional territories, turning the Río Los Perros into a loose boundary between the Zoque- and Zapotec-speaking communities (Zeitlin 1994:277). These same sources also claim that Tehuantepec, the royal seat of Coqui Cosijopi and later his son Cosijoeza, was the center of an expansive tributary state that controlled most of the coastal plain stretching to the east and back west to the Sierra Sur. From there, the Zaachila Zapotecs were able to claim tribute rights from distant places such as Ixtepeji and even Nejapa (Zeitlin 1994:277), while also controlling the all-important coastal trade routes to the Soconusco region and the Maya in highland Guatemala.

The Chontalpa subregion lies to the southwest of Nejapa and extends across the Sierra

Sur down to the lowlands and coastal plains of the Pacific in a southeast direction. Research by

Kroefegs (2004) in the latter zone and Zborover’s (2014) study in and around the modern town of Santa María Zapotitlán in the piedmont area provide much of the information we have on the

Prehispanic-era in the Chontalpa, apart from its rich ethnohistorical tradition. While there maybe earlier settlements that have yet to be documented, current evidence shows that occupation in the

Chontalpa began sometime in the Late Formative, however who these people were is still unknown (Zborover 2014:326). The Chontal arrive sometime in the Late Classic (Winter

1986:136) and there is a corresponding increase in the population during this time period

(Zborover 2014:331). The subregion is more intensely occupied in the Early and Late

Postclassic periods, especially in the upper piedmont region where lowland Chontales were often pushed to during the various conflicts between the Mixtec yuhuitayu of Tututepec and the

247

Zapotec quechecoquie of Tehuantepec. Several local lienzos indicate that many communities in the Chontalpa were tribute subjects to Tututepec. During the Postclassic, the Mixtec capital exerted indirect influence over four Chontal communities in the mountains through its Zapotec- subject town of Suchixtepec, and three others through its Pochutec-subject town of Huatulco on the coast (Zborover 2014:338-339). Archaeological evidence supports the ethnohistorical sources, showing that the Chontal highlands had closer ties to the communities of the west and central coast than to the Isthmus, even though the latter is closer geographically (Zborover

2014:347). In contrast, Chontal communities on the coast showed greater quantities of Isthmian and Zapotec-style ceramics from Tehuantepec, highlighting the complexity and nuance of local social relationships (Zborover 2014:342).

West of the Nejapa region by 34 km is the subregion of Quiechapa, on the northern slopes of the Sierra Sur (Badillo 2017). Surprisingly, as in the Chontalpa there is a rich corpus of ethnohistorical documents and oral traditions that record the different events that took place here going far back in time. One such story describes a period of conflict between the Chatinos and the Chontales; today people in Quiechapa speak only Zapotec, indicating that at some point in the past three different languages were spoken here. It is hypothesized that the area was predominantly Chontal- and Chatino-speaking during the Formative and/or Classic periods until the arrival of Zapotecs from Mihuatlan, Coatlan, Amatlan, and Ocelotpec sometime in the 9th and

10th centuries CE (Badillo 2017:39-40). In the Late Postclassic, according to these same sources, the Zaachilenos established a fortress or fortified settlement in Quiechapa during their military campaigns to the Isthmus. During his survey of Quiechapa, Badillo (2017) documented two modest ceremonial centers with the only examples of monumental architecture in the subregion, including ballcourts. The natural topography and hydrology effectively divides Quiechapa in

248 two, and each zone is dominated by its ceremonial center (Badillo 2017:41). Ceramic evidence indicates that both sites were first occupied in the Late Formative period, although it is likely that the monumental architecture is much later. Both sites continued to be occupied into the

Postclassic. Overall though, there are clear differences in settlement patterns over time;

Postclassic sites tend to be higher in elevation and more dispersed than in earlier periods, perhaps indicating a shift in occupation in response to heightened conflict during this time (Badillo

2017:43).

4.3.2 Discussion

The different altepetls and social groups of Mesoamerica were tied together in complex ways that cut across different ethnolinguistic, social, and physical boundaries. In Oaxaca, the sociopolitical relationships between the different communities were also characterized by shared but still distinct cultural-historical trajectories that were marked by intensive episodes of sociopolitical and economic interaction while also bound by sociopolitical relationships built on marriage alliances, shared ideologies, and economic networks (Balkansky 1998a; Flannery and

Marcus 2004; Pohl 2003a; Pohl 2003b; Pohl 2003c; Zborover 2014). Evidence for very early interactions between communities in Oaxaca, Central Mexico, and the Veracruz Gulf Coast point to the former’s participation in a developing pan-regional ideological system that emerged as early as the Formative era (Flannery and Marcus 2000). In the Central Valleys, Mixteca Alta,

Costa Chica and Isthmus regions, we see the early and rapid development of both settled village life and urban centers during this same period.

In their interpretation of political history and social development in the Postclassic, Smith and Berdan (2003) take an explicitly World Systems theory approach. Different regions of

249

Oaxaca are cast as peripheral zones to the more economically centralized Central Valleys, and by extension Oaxaca is viewed as a periphery to Central Mexico. However, as Blomster (2008:28) and Winter (2008) make clear, some of these areas are only considered to be peripheral because of the lack of research and data. Blomster (2008:31) makes the excellent point that rather than being an outsider to these newer patterns of regional interactions, Oaxaca was instead a strategic crossroads for the movement of peoples, practices, and goods due to its geographical position, and whose communities also participated in interregional trends and sociocultural practices.

The complex settlement patterns in Oaxaca reflected combinations of practices and interregional practices. Communities varied considerably in how they were governed, what place they occupied in regional social hierarchies, and the power and influence they were able to exert through political relationships and alliance networks (Blomster 2008:22-23). The mountainous and rugged geography, combined with incredibly diverse ecological zones that change as one moves up or down in elevation, meant that factional social groups had to constantly interact with one another, negotiating conflict and cooperation, power and influence among them. At times, some communities had greater success at exerting power and influence over others via military and political forces; but we must understand that this was accomplished in a context where ethnolinguistic and perhaps other social identities were likely fluid.

Factionalism may have characterized Oaxacan political organization, but it was never a hindrance to developing and negotiating local and regional intercommunity relationships.

While Monte Albán probably exerted more control than most, there were still several other competing urban settlements that held influence over surrounding communities and competed with each other for available economic, social, and political resources. Furthermore, verticality, or the exploitation of different ecological zones from the valley floors to the

250 mountain zones, would have linked physically dispersed communities together in ways that cut across different sociopolitical and ethnolinguistic boundaries (Winter 2008:394). Group or community territories often included, and still do, many different and not always contiguous ecological zones. Many Oaxacan subregions – including Nejapa – include a high mountain zone, a middle piedmont zone, and lowland alluvial bottoms or coastal plains. As Winter (2008) points out, the practice of verticality, and how it may have shaped or impacted interregional relationships among different groups, has been undertheorized by archaeologists in Oaxaca.

The intensity of cross-cultural contact and episodes of conflict, aversion, collaboration and cooperation between Oaxacan communities varied considerable across time and space, but their interactions were critical to the development of the sociopolitical organization of all. More than a backwater or provincial area that lay outside of history, Oaxaca was instead an important contributor to the overall development of political society in Mesoamerica. Broadly shared religious practices, art styles, and architectural traditions found in public and elite domestic contexts throughout Mesoamerica are also evident in Oaxaca (Joyce 2010:220; Urcid 2005), the result of intense interactions and expansive intercommunity relationships rather than political integration or unification (Joyce 2010:220). Oaxaca’s position as the geographical crossroads of

Mesoamerica, connecting Western, Northern, and Central Mexico with the Maya region and

Central America, meant that Oaxacan communities had plenty of opportunities to participate in these global networks and interact with distinct ethnolinguistic groups. At the same time, the ruggedness of the terrain probably meant that these communities had more choice in shaping these interactions, determining for themselves whether to participate or not – although there are plenty of examples where outside groups were successful in dictating terms to their advantage.

As we shall see in the following section, Nejapa was also a crossroads of sorts connecting

251 different regions of Oaxaca, funneling movement in Mesoamerica from the northwest to southeast, just on a much smaller scale.

4.4 NEJAPA FROM THE FORMATIVE TO THE POSTCLASSIC

Figure 4.5: Map of the Nejapa study region and the modern towns found there.

The subregion of Nejapa lies in the Yautepec District of Oaxaca in the eastern part of the

Sierra Sur range (Figure 4.5). It is bordered by the Isthmus of Tehuantepec on the east, coastal districts to the south, the Central Valleys to the northwest, and the Sierra Mixe region to the north. The name Yautepec means “Hill of the Black Corn” in Nahuatl. Encompassing around

4772 km2of territory, it is a very mountainous area that includes a wide range of ecological environments. Lower elevations are hot and humid, while higher in the mountains and piedmont

252 zones it can be quite cool at night, especially in the winter months. Overall, the predominant ecological regime is a temperate climate zone with mahogany, cedar, Guanacaste trees, as well as pine and oak forests. Reflective of this mixed character, parrots are commonly found nesting in the tops of pine trees (primarily of the ocote species) at the highest elevations. There even exists a flock of green army macaws that live close to the area of Cerro de la Muralla, near the modern town of San Bartolo Yautepec. Local fauna includes wild boar, ocelots, snakes, mountain lion, coyotes, and foxes. Closer to the towns of Nejapa de Madero and Santa Ana

Tavela, there is still a small population of Mexican white-tailed deer.

Farmers in Yautepec grow cash crops such as papaya, coffee, maize, peanuts, and beans, while also raising goats, cattle, chickens, sheep, and pigs. A variety of fruit tree crops also grow here, including mango, peach, tamarind, plum, sapodilla, lemon, and avocado. Many villages also cultivate maguey for sale to Santiago Matatlán in the Central Valleys and the distant state of

Jalisco, although a few magueyeros do make their own mezcal. Census data as of 2005 counted a total of 31,688 people, a substantial number of whom are indigenous Zapotec, Mixe, and

Chontal-speaking peoples. There are 12 municipalities in the district in total. The smallest is

Santa Catarina Quioquitani at 46 km2 and 439 individuals, the majority of whom are indigenous

(84%). The largest is San Carlos Yautepec, which encompasses an astounding 2,492 km2 and a total population of 9,857, 38% of whom are indigenous.

The journey to Nejapa officially begins once you pass the Portillo de San Dionisio, a narrow passage that marks the transition from the broad alluvial plains of the Tlacolula Valley (a subarm of the Central Valleys) to the dramatically shaped mountains of the Sierra Sur. The

Highway 190 twists and turns as it winds its way through the mountains, occasionally dipping close to the river before climbing several hundred meters above it. Along the way, you get stuck

253 behind loud, belching diesel trucks that crawl along at 30 km per hour, dodge large boulders that have slid off the mountainside after a hard rain, and sometimes even donkeys and cows that wander along the roadside seeking patches of grass. After an hour and a half of these dips, dives, and hairpin turns, you go around an extreme curve and suddenly! The valley opens before you, the white limestone cliffs of the Cerro del Convento mesa looming large off in the distance. The road continues on, hugging the piedmont that skirts the southwestern edge of the alluvial valley floor before continuing on through the mountains towards the Isthmus of Tehuantepec. Turning off the main highway at the town of El Gramal takes you onto a roughly paved road that continues north and east into the Nejapa Valley.

The Nejapa region lies about halfway between the Central Valleys of Oaxaca and the

Isthmus of Tehuantepec, and is the largest open space along this route through the Sierra Sur mountain range that divides the two regions. The subregion includes the surrounding piedmont zones, mountains, and the alluvial bottomlands that are cut through by several perennial rivers.

The Río Grande enters from the west and eventually joins with the Rio la Virgen from the south to form the Río Tehuantepec that exits the valley moving north. In total, the subregion measures approximately 953 km2. Today, most of the Prehispanic and Colonial territory of Nejapa falls within three different municipalities: Nejapa de Madero, Santa Ana Tavela, and San Pedro

Quiatoni, but also includes sections of San Carlos Yautepec, San Juan Lajarcía, and San Bartolo

Yautepec. PANT survey and excavations focused on these municipalities, with the exception of

San Pedro Quiatoni, as we did not receive permissions to conduct fieldwork there. Given the density of settlement in the Nejapa, it is entirely possible that additional sites (and ballcourts) have yet to be documented in that area. Spatial data from INEGI (Digital Elevation Model 2009,

50 x 50 m, data ITRF92) indicates that the valley floor in this specific area of the Yautepec

254

District varies in elevation from 620 to 800 m above sea level, while the surrounding mountains reach as high as 2700 m above sea level at the eastern edge of Santa Ana Tavela.

The local vegetation is a mix between the different ecologies of the dry scrub brush and pines of the highlands, and the lush tropics of the mid-range lowland and coastal zones (Figure

4.6). The rivers that flow through Nejapa create the rich alluvial lands of the valley bottom that in the past were used to grow corn, beans, and chilies. Later in the Colonial period agricultural production was shifted to sugar cane crops and cochineal cultivation. These same fields now grow papaya, bananas, corn, melons, mangos, and maguey. Land is owned both privately and communally, with the communal lands belonging to each town, agencía, ejido, and municipal head. Local governments follow the usos y costumbres (uses and customary laws) tradition that governs many indigenous villages in Mexico. While most people closer to the valley lands speak

Spanish, many smaller communities in the mountains still speak indigenous languages (in some towns up to 80% of the population). Communal work parties (similar to the Prehispanic Zapotec tobicollaba and tobicozaana, kin or neighborhood-based work parties [Zeitlin 1994]), performing their tequeo or communal service, are still organized today to work on community construction projects, fight forest fires, and cultivate commonly-held agricultural fields (Figure

4.7). Many of these products are then sold at markets in the Oaxacan capital.

255

Figure 4.6: The Nejapa subregion has a mixed ecology that includes highland scrubs and trees (upper) and more humid zones where you can grow tropical fruits such as papaya (lower).

256

Figure 4.7: Men from Nejapa de Madero performing their tequeo and harvesting watermelons from a communal field.

In the Prehispanic past, the Nejapa subregion was an important stop on the international exchange route between the Central Valleys and the Isthmus of Tehuantepec. By extension,

Nejapa also connected the Basin of Mexico to the Soconusco region of Chiapas. Merchants would pause here to refill their travel supplies, trade for local goods, and use the subregion as a spring board to also travel to more mountainous areas to the north and south (Brockington, 1989;

Machuca Gallegos, 2007; Paso y Troncoso, 1905). These same trade routes were later used by the royal lineages from Zaachila in the Central Valleys, first in the 14th century and once again in the 15th century, in order to access and conquer the Isthmus. The Aztec ruler Azhuizotl also sent his military forces through Nejapa on their way to the Soconusco region of coastal Chiapas only

257 a few decades later, after having conquered several sites and established fortresses in the Mixteca

Alta and the Central Valleys (King 2012; King and Zborover 2015; Oudijk 2000; Oudijk and

Restall 2007; Paso y Troncoso 1905) . Shortly thereafter, Nejapa was annexed by the Aztecs, becoming an important of the Aztec Empire (Gasco 2005).

Today, Nejapa stands out as a point of convergence between the modern territories of the

Mixe peoples to the north and northwest, the Chontales to the south, and the Zapotec-speaking peoples to the west, southwest, and east (King 2012). Historical evidence indicates that these ethnolinguistic groups were already living in Nejapa when the Spanish armies and early chroniclers first arrived (Burgoa 1934 [1674]; Paso y Troncoso 1905; King and Zborover 2015).

These documents describe the Zapotecs as living primarily on the valley floor, while the Mixe occupied the mountains and the Chontales lived like “wild savages” in mountain caves (King

2012). A petition to the local colonial authority by the Spanish living in the first settlement established in Nejapa (then spelled Nexapa) gives insight into the state of interethnic relationships at the time. The vulnerable settlers wanted to move their town physically closer to the Zapotecs on the valley floor, they claimed, in order to be safe from attacks by allied Mixe and Chontal warriors.

Ethnographic information on interethnic relationships in Nejapa stretches back even further to before the arrival of the conquistadores. Again, according to those documents related to Zaachila, Coqui and his son Quixicayo established several fortresses in Nejapa, effectively taking over the subregion. The colonial documents detail battles between different Zapotec communities against various Chontal and Mixe communities, with some Mixe and Chontal towns occasionally forming alliances against their common enemy in the Postclassic period

(King and Zborover 2015; Zborover 2014). To this day, the Mixe regard their Zapotec neighbors

258 with suspicion and distrust. Despite long-standing tensions between these groups, interethnic alliances were still possible. In the 1600s, the people of Nejapa and neighboring Villa Alta, located to the north in the mountainous Sierra Mixe, formed a regional alliance that cut across ethnic and linguistic boundaries to rebel against burdensome Spanish policies (Yannakakis

2008). Despite the Spanish effort to emphasize ethnic differences and create division among indigenous peoples, communities in Nejapa likely activated ancient social networks to fight against this oppressive ‘foreign’ system, much as they might have done when previous invasive forces had passed through (King and Zborover 2015).

The complex reality of ethnic identity and social relationships in Nejapa is further supported by the 1580 Relación de Nejapa, which makes mention of Nahuatl translators. The use of the word “translator” rather than “speaker” suggests that those individuals who knew

Nahuatl were native speakers of a different language. This raises two interesting points: 1) that individuals in Nejapa learned to communicate with the Nahuatl-speakers passing through their territories, indicating that a few communities may have allied or cooperated in some way with the Mexica armies, or felt the imperative to be able to communicate with foreigners moving through their territories; and 2) that some people in Nejapa were probably accustomed to speaking, or having to learn to speak, multiple languages.

The picture of Nejapa given to us by the colonial documents is of a multilingual frontier zone where clearly defined ethnic groups lived in discrete territories, a boundary area that only truly came into history when it was brought under the control of foreign armies. But do we actually see signs of a Zapotec or Aztec conquest in the Postclassic, such as extreme site destruction (such as through burning), a complete replacement of local ceramics and/or other artifact classes, the presence of newly established military fortresses, and/or settlement

259 reorganization? Additionally, most of the historical evidence only describes events of the

Postclassic period, but what was happening in Nejapa during earlier periods? Finally, are sites clearly identifiable as Zapotec, Mixe or Chontal? As documented by PANT, the archaeological evidence presents us with a much more complicated story than initially may be assumed. In the following sections, I use the archaeological data collected by PANT study to describe the settlement history of the Nejapa subregion so far. The sequence of settlement occupation and abandonment related here is based on several lines of evidence: 1) radiocarbon dates from excavated contexts; 2) local ceramic chronologies using diagnostic sherds from surface collections and excavated contexts that were also referenced to other, established ceramic chronologies in Oaxaca; and 3) ethnohistorical and Colonial-era documents.

260

4.4.1 From the Formative to the Classic Period (2000 BCE – CE 900)

Figure 4.8: Map of the Nejapa study area and some of the sites mentioned in the text.

While it is entirely possible that small groups may have traveled through and even stayed briefly in Nejapa during the Paleoindian and Archaic periods, we do not find much evidence indicating the presence of people that early. Instead, the oldest documented occupation in the region dates to the Early Formative period (2000 – 500 BCE), at the archaeological site of

Colonia San Martín (YAU-NEJ-NEJ-004). Colonia San Martín is located at an advantageous and fertile spot next to the confluence of the Río Grande and Río de la Virgen on the valley floor

(Figure 4.8). People continued to live there through to the Middle Formative (750 – 350 BCE)

261 and later periods, occasionally abandoning the site only to reoccupy the area at other times (and people still live there today on top of these earlier occupations). Only one other Middle

Formative site has been identified so far, Llano las Casillas (YAU-NEJ-NEJ-002). Although the site center has mostly been destroyed by modern agricultural activities, it is still possible to see a high density of ceramic and lithic materials on the surface extending along for quite a distance, as well as wall foundations and small platforms scattered throughout. From the ground surface, it is evident that occupation at Llano las Casillas was organized into neighborhoods or barrios, demonstrating the antiquity of this type of social organization. Given that known settlement in the Early to Middle Formative was primarily on the valley floor, the low levels of occupation may have more to do with site loss due to centuries of alluvial flooding and modern intensive agricultural practices than a lack of settlement during this time (King et al. n.d. Vols I and II).

As a location for early agricultural villages, Nejapa would have been an ideal place to settle down and farm due to the expansive valley floor, fertile soils, and perennial water flow, where the rivers would bring rich silts from the highlands during the yearly rainy seasons. It is reasonable to assume that during most of the Formative period, there were more low-density settlements consisting of a few households in the subregion than is currently documented.

Settlement expanded in the Late Formative Period (350 BCE – CE 250), with occupation in strategic locations throughout the valley floor. People were still living at Colonia San Martín at this time. Another prominent Late Formative site was Mogotes del Panteón (YAU-NEJ-

ANM-002), which occupied a strategic, fingerlike extension of piedmont overlooking the Rio

Grande near the western entrance to the valley till at least until the Early Classic. Such a location would have allowed the people living there to control access in and out of the valley and exert influence over important trade networks (King 2015). Mogotes del Panteón is also

262 remarkable for its small but complex ceremonial center with several large public structures, including a massive ballcourt (described in Chapter 6). To the east of Mogotes del Panteón lies

Nejapa Viejo, a large urbanized site that grew and thrived from the Late Formative through to the Classic period (CE 250 – 900). The PANT survey crew documented pyramids, elite residences, a grand plaza, and various groupings of smaller households surrounding these other structures. Much of the ceramic material at the site bears similarity to popular styles dating to the Late Formative and Classic periods in the Central Valleys, including fine grayware outleaned bowls with incised horizontal lines on the border. On the other hand, there are also other ceramic materials that stylistically resemble Isthmian ware, such as vases with textile impressions and kaolin/cream pastes. The material evidence suggests trade relations or at least social contact with other regions of Oaxaca. At the same, there also is clear evidence for autochthonous ceramic styles, such as a common and very popular style of cylindrical support.

The other large and impressive Classic site is the Cerro Maluco, which has been roughly dated to the Late Classic or Xoo phase – although there is some evidence for earlier occupation perhaps dating as early as the Late Formative. Cerro Maluco includes residential houses on some of the terraces as well as an impressive ceremonial center with a large plaza, several monumental structures, and one of the largest ballcourts in the state of Oaxaca (described in

Chapter 6). Based on surface ceramic materials, three additional sites generally dating to the

Classic period were also documented during the 2013 field season, based on the surface ceramic materials: El Callejón de Tejabana (YAU-NEJ-NEJ-010), Arroyo del Sauso (YAU-NEJ-NEJ-

023), and Arroyo de Limón (YAU-NEJ-NEJ-009). Of the three, Arroyo Limón was most likely a substantial settlement, given the density of ceramics, a line of shaped foundation stones observed in a cut on one side of an agricultural field, and reports from the locals of finding what

263 were likely Prehispanic burials (King et al. n.d. Vols I). El Callejón de Tejabana is known only through ceramics emerging out of a canal cut that ran through cultivated fields. This would suggest that additional occupation is likely buried under the soil or has been destroyed through modern agricultural activities in this location, but what evidence remains probably belonged to a small settlement, likely a neighborhood or town. There are other settlements that likely probably date to the Classic period as well, but additional excavations are needed. Radiocarbon evidence, for example, from the sites of El Sitial (YAU-TAV-TAV-010) and Cerro del Convento (YAU-

SJL-SJL-002) gave both Middle Classic and Postclassic dates. Interestingly, both sites have ballcourts (Chapter 6).

4.4.2 The Big Bang: Population Growth, Settlement Expansion, and Social Conflict in the Postclassic Period

The Nejapa subregion reached its greatest population and settlement extent during the

Postclassic period (CE 900/1000-1521) (see Figure 4.8). Large and small settlements would grow to occupy all the extant ecological zones, intensifying on the valley lowlands while also reaching the highest elevations in the surrounding piedmont and mountain zones. Some of the large mountain sites feature large stone and adobe structures, temples, palaces, and platforms with white and sometimes red-painted stucco floors, while others even have ballcourts.

Settlement diversity, observed in the architectural styles and spatial organization, increased, while access to trade networks also greatly expanded. We also find a diverse array of ceramic styles identified from surface collections and excavations; some wares are local in style and origin, while others clearly come from elsewhere, including the polychrome Mixteca-Puebla style vessels that were exchanged throughout Mesoamerica in the Postclassic.

264

The extent of Nejapa’s links to other regions and groups in Oaxaca and beyond can be seen in other areas as well. Gold mined here (Paso y Troncoso 1905:159) and the Chontalpa

Alta (Zborover 2014:338) was traded by the Pochutec of Huatulco to the Mixtecs of Tututepec

(Acuña 1984:191), attracting the attention of many outside groups. Glyphs similar to the Aztec symbol for gold have even been found in the Chontal subregion (Zborover 2014) and perhaps even in Nejapa (King 2010) (see Figure 6.19). Today, gold is still being mined just northwest of

Nejapa near the town of San Jose de Gracia, and foreign mining companies are still trying to open operations in the mountains near Santa Ana Tavela. The Chontal highlands, just south of

Nejapa, were likely used by the Mixtecs living on the coast to travel from the Huatulco area on the Pacific Coast to Jalapa del Marques in the southwestern Isthmus, avoiding the hostile

Zapotecs that controlled the latter (King and Zborover 2015:141). Other trade routes traveled through the Chontalpa Alta northwards through Nejapa to the Sierra Mixe, allowing people to cross the Continental Divide and descend down onto the coastal plain of the Gulf of Mexico

(King and Zborover 2015: 141; Zborover 2014).

Several notable urbanized sites were established at the beginning of the Postclassic.

Greater La Amontonada (GLA) is an Early to Middle Postclassic settlement (calibrated date of

CE 1042-1400) that encompassed a large ceremonial center and about 12-14 residential neighborhoods. A combined total of 94.75 m2 were excavated at these sites by PANT archaeologists over the course of the 2009-2013 field seasons, including large portions of at least three structures in three different neighborhoods. The site complex gets its name from the civic- ceremonial center also known as La Amontonada, located on an extension of piedmont to the south of the Río Grande. The high density of miniatures, incense burners, and serving vessels indicate that the ceremonial center was used by the elites and non-elites living in the different

265 neighborhoods for formal and informal politico-religious events. Six other sites make up the various neighborhoods forming the larger Amontonada community; three of these neighborhoods feature ballcourts (described in Chapter 6). Based on the excavated data, it is likely that the residents of Greater La Amontonada’s neighborhoods were fairly autonomous in their daily lives, able to make their own decisions with regards as to how to construct and arrange their houses and neighborhood public spaces. The monumental architecture of the center, on the other hand, would have required the mobilization and coordination of work parties drawn from the different neighborhoods (Konwest 2017), likely organized along actual and/or fictive kin relationships. Likewise, the uniformity of the ceramic and obsidian evidence also suggests some form of top-down organization or coordination of access to products.

La Puerta (CE 1050 – 1270) was a contemporary of Greater La Amontonada and is located in the upper piedmont region about 4 km or so to the northeast of the latter. Settlement, including public architecture and habitational terraces, stretches across the sides and ridgelines across several hills. Because of the danger presented by the location, the PANT team could only document those structures that were most accessible. Architecture was still observable even beyond these areas, however, indicating that the settlement is much larger than officially recorded. La Puerta is notable for the small ballcourt located on a saddle between two hilltops, below the area of primary occupation. It may have been even one of the first structures encountered by visitors to the community if approaching from the south/southeastern side.

Interestingly, the material culture at La Puerta is similar to Greater La Amontonada, reflecting equal access to foreign trade networks and local production sources at both communities. Given that the sites are contemporary, this would suggest that some people in Nejapa were choosing to reside in the mountains, while others preferred living on the valley floor.

266

After an apparent period of abandonment sometime in the Classic period, Colonia San

Martín (calibrated dates from CE 1268-1638) is once again occupied in the Middle Postclassic, and again continues to be so to today. Though not much remains of the Postclassic settlement because of ongoing construction, a remaining elite house mound was excavated in 2011 and

2013. PANT excavators uncovered rooms covered in several layers of red-painted stucco on the walls and floors. The architectural style and material evidence of Colonial San Martín stands out from other settlements in Nejapa. The majority of the Mixteca-Puebla style ceramics documented in Nejapa come from this site in particular. Additionally, a large number of imported ceramics from other areas of Oaxaca have been recorded at Colonial San Martín, in percentages that are much higher than other sites that PANT has excavated. The evidence may indicate that Colonial San Martín was a merchant community that exerted a high degree of influence over the trade of exotic prestige goods in Nejapa.

In their 2011 survey of the piedmont and mountain zones in Nejapa, PANT crews documented at least five large fortified sites: Cerro del Convento, Los Picachos (YAU-TAV-

TAV-008); Cerro de la Muralla (YAU-SBY-SBY-001), El Mantecón (YAU-SCY-GUI-007), and

Cerro San Antonio (YAU-SCY-GUI-014). Described as military forts in colonial sources

(Martínez Gracida 1910b), these sites instead exhibit a great deal of variability that suggests they did not all have military purposes. Of the five, Cerro del Convento (calibrated CE 1027-1492), is the most likely candidate for being a fortified site, given that there is relatively less evidence for domestic use, a less formal temple, and three large defensive trenches that had been constructed on the most accessible side. On the other hand, Cerro del Convento also features a ballcourt, so it may not have been a military fort in the traditional European understanding of the

267 term. As I describe later in Chapter 6, other evidence from Cerro del Convento also shows its long-term use as a site of refuge, storage, ritual ceremonies, and other important purposes.

Of the three remaining “forts”, Los Picachos (calibrated CE 945-1047) and Cerro de la

Muralla (calibrated 1261-1641 CE) were less fortified settlements and more densely urbanized but still defensible sites with complex public architecture and extensive occupation. Cerro de la

Muralla shows internal hierarchical divisions, with lower-status residences found toward the edges of the site separated from higher-status neighborhoods by grand walls (Figure 4.10) – a pattern similar to many mountain top sites throughout Oaxaca. These large walls would have also served defensive purposes if the settlement came under attack. The high-status neighborhood features a seven-room palace and a temple-patio-altar complex. The site of Los

Picachos stretches out along a narrow ridgeline and appears to be heavily fortified; at the same time, its western side was directly accessible to the trade corridors that connected the Nejapa subregion to the Isthmus (King and Zborover 2015:138; I describe Los Picachos in more detail in

Chapter 6). Ceramic and architectural evidence at both sites shows that though they lived in defensive locations high up in the mountains, people in these communities still had access to interregional trade networks. They constructed their structures following regional styles while still using local building techniques and producing goods according to their own customs. El

Mantecón, while less densely occupied, was also a habitational site that may have had some religious importance.

268

269

El Sitial (calibrated CE 1412 – 1619), mentioned earlier, is another complex, mountaintop site, with large temple structures, plazas, a palace with white stucco floors, and a ballcourt with an altar and other, smaller structures nearby. The slopes of the mountainside are covered with habitational terraces, indicating a substantial population in the past. Despite its defensive characteristics, like many mountain sites in Oaxaca, El Sitial is primarily a residential and ceremonial settlement, possibly with several neighborhoods. In 2013, the PANT crew camped further down the mountain slope from the main center of El Sitial. Surrounding the campsite were several previously unknown structures, including a small ballcourt that appears to be a palangana-style court (described in Chapter 6). Given the proximity of this structure group to the much larger site, it was the ceremonial center of a neighborhood attached to El Sitial, likely composed of lower-status elites or mixed-status groups. Alternatively, it could also represent a much earlier phase of occupation.

Other sites in the Postclassic show the range of settlement diversity in Nejapa, although we still do not understand very well their sequences of occupation nor how they relate to other, contemporary sites. Terezona for example, located on a hilltop near Nejapa de Madero, appears to have been a large, possibly urbanized settlement with several areas of mounded architecture, extensive residential terraces, and an unusually shaped ballcourt (described in Chapter 6).

However, we have only been able to date this site to the Postclassic in general. La Baeza is another urbanized center that has been roughly dated to the Late Classic to Early Postclassic periods based on surface ceramics. Located on the very eastern edge of the Nejapa survey area,

La Baeza is somewhat of a mystery, with architectural styles and ceramic artifacts that connect it to communities both in Nejapa and Quiechapa. The monumental architecture at La Baeza

270 clearly points to its local and regional prominence and, as I discuss further in Chapters 6 and 9, the site may in fact represent a different subregion or unique area separate from Nejapa.

Additional large and dense communities, documented in 2013 on the southern and eastern edges of the Nejapa subregion and perhaps representing yet another distinct zone of occupation, might also be Postclassic in age. Although PANT was unable to determine dates based on the surface ceramics, further investigation should help determine to what time periods these sites may date.

Despite this, the settlement density does demonstrate that Nejapa had been intensely occupied in the Postclassic when the Spanish arrived in the area in the 1520s.

The Postclassic period was marked by increased tensions and political strife between the different ethnolinguistic groups whose territories surround Nejapa. The Mixes, for example, were engaged in a series of devastating battles with Zapotecs from the Central Valleys as they attempted to stop the latter’s incursions into their mountain home (Barabas and Bartolomé

1984:11). Hostilities increased even further beginning sometime in the Middle Postclassic, when the Zaachileños arrived in Nejapa during their first campaign of conquest towards the Isthmus in the 1300s. According to the calibrated radiocarbon dates obtained from sites on the valley floor, this incursion may have occurred around the time (or just after) Greater La Amontonada was abandoned and just as or before Colonia San Martín is reoccupied. As mentioned previously, the

Zaachileños supposedly established a series of fortresses here as they annexed the Nejapa subregion. Under the Coqui Zaachila II, the Zapotecs also initiated a military campaign in the

Chontalpa area in the 1400s just to the southeast (Martínez Gracida 1910a). If the Zapotecs did control the territory as the ethnohistorical documents indicate, then they would have used Nejapa as a springboard for these military actions towards the Chontales.

271

A second Zapotec incursion occurred later in the 15th century, when the armies of the renegade Zaachileño lineage took control of the Isthmus around CE 1460. Nejapa was claimed as part of the Zapotec-Isthmanian sphere of influence (Zeitlin 2005). However, these events were followed shortly afterward by the arrival of the Aztec imperial armies during the reign of

Huey Tlatoani Ahuizotl in the years CE 1469 – 1486 (Gasco 2005). Although ultimately their purpose for passing through was to conquer the rich lands of the Isthmus and Soconusco, the

Aztecs were also likely attracted by the gold already being mined in the area. How communities in Nejapa reacted to the Nahuatl presence there is not entirely clear from either the histories nor the archaeological evidence, apart from the mention of Nahuatl translators in the former. For contrast, in the Coixtlahuaca Valley where the Aztecs established a garrison we find Aztec ceramics and diagnostic sherds all over the ground surface. Colonia San Martín remains occupied throughout these events in a location that was not very defensible; at the same time, we see fewer settlements on the valley lands while mountain sites continue to be occupied.

Different communities in Nejapa probably reacted and adapted to the presence of the Central

Mexican forces, with some forming alliances with the Aztecs while others either fled or resisted.

4.4.3 Nejapa (Nexapa) in the Colonial Era

Even after the Spanish conquest of Mexico, the Nejapa subregion continued to be an important stop on the camino real or royal road that connected the different regions of the new empire together. Pedro de Alvarado was the first Spaniard to visit Nejapa when he passed through the area on his way to Guatemala in 1523 (Matthew 2007:106; Oudijk and Restall

2007:49). The first official Spanish settlement was founded in 1533, probably as a result of the arrival of the conquistador Francisco Maldonado, but it was abandoned quickly soon after

272

(Burgoa 1934 [1674]; Gerhard 1972:195). Despite initial setbacks, the richly productive alluvial floodplains and piedmont were very attractive to the Spanish, making the subregion an important contributor to the colonial Oaxacan economy. Cochineal, a highly valued red dye, was a lucrative cash crop produced both in Nejapa and the nearby Chontalpa Alta (Bartolome and

Barabas 2006; Baskes 2000, 2005; Hamnett 1971; King and Zborover 2015). The subregion was a central hub for the distribution of Isthmian salt as well as an important meeting place for traveling merchants, much as it was in the Prehispanic era (Kuroda 1984:11-12; Laviada

1978:169; Machuca Gallegos 2007). Nejapa also continued to be an important gateway into the

Sierra Mixe to the north during the Colonial period (King and Zborover 2015:141).

To this day, there are still several prominent haciendas that have long since fallen into ruins in the municipality of Nejapa de Madero, including the Hacienda San José (YAU-NEJ-SJF-

001), the Hacienda las Casillas (YAU-NEJ-NEJ-001), the Hacienda Chihuiro near San Isidro

Chihuiro, and finally the Hacienda San Isidro Manteca, which can be found near the town of the same name in the San Carlos Yautepec muncipalit1y. The Hacienda San José was probably constructed by the Dominican order sometime in the 1560s, as they had a ranch for small livestock during this period (AGN Mercedes 5, 175r, in Chance 1989:156). These haciendas primarily produced sugar, another important cash crop grown in Nejapa during the Colonial period. Though seen as peripheral today, Nejapa’s importance in the Colonial era is further demonstrated by an 18th century French map of Mexico. Of the few places depicted in Oaxaca

(Guaxaca) on the map, one of these is Nejapa (here spelled Nixapa) (Figure 4.10).

273

Figure 4.10: 18th century French map of Mexico. Nejapa is one of the places included on the map.

Spanish control of Nejapa was constantly challenged in the Early Colonial period. The indios, as they are referred to in the Colonial documents, would often flee into the mountains when the colonial authorities arrived to collect tribute (Paso y Troncoso 1939:64). The

Dominican friars constantly complained that the mountain people would come down to eat, but never stayed for the Catholic mass afterward (Gay 1982:315-316). In 1528, both Nejapa de

Madero and the nearby pueblo of La Asunción de Majaltepec (spelled at this time as

Maxaltepeque) were named as part of the Mayorazgo or marquisate given to Hernán Cortés

(Romero Frizzi 1996). Majaltepec10 was a Transitional-Colonial period settlement founded by

10 The settlement shows an interesting mix of Prehispanic and Colonial characteristics; during excavation of a test pit for example, the PANT crew found the remains of several individuals. These individuals had been buried according to Prehispanic customs (under or near a structure) but with clearly Colonial objects such as glass beads and a steel winch. 274 local Nejapeños (el pueblo viejo de los mixes or old town of the Mixes) sometime prior to or even during the early decades of the Spanish penetration into the area. From the 1530s to the

1550s, Majaltepec and Nejapa were both required to pay tribute in money and food crops as well as render service to the encomenderos and the Spanish authorities (Paso y Troncoso 1905b).

However, the Majaltepecaños actively resisted Spanish control, to the great frustration of the clergy. After years of conflict between the inhabitants of Majaltepec and the Spanish laity and clerical authorities, the people were forced in the mid-1600s to resettle in the location where

Santa Ana Tavela is now. Some people resisted the Spanish policy of congregación

(congregation) and attempted to return to Majaltepec after a few years, resulting in a bloody conflict that left at least 100 dead and 80 wounded (King et al. 2012). Finally, during the millennial rebellion of 1660 in the Isthmus, Nejapa acted as a principal route for the rebels

(Barabas and Bartolomé 1984; Gay 1982:491-492; Martínez Gracida 1910a; Yannakakis 2007).

4.4.4 Discussion: Challenging Narratives of History and Ethnic Social Identity in Nejapa

Our picture of local and regional relationships in Nejapa, particularly with regards to the

Zapotec presence in the subregion, is further complicated by the pictographic evidence that

PANT has documented there. During a flooding episode in the 1980s, a carved stone with

Zapotec glyphs was discovered at Nejapa Viejo (it was later moved to the zócalo of modern

Nejapa de Madero). Rectangular in shape, the stone measures 1.70 m in height, 1.34 in width, and 14 cm in thickness, and is carved on both sides (Figure 4.11). Based on the condition of the glyphs, it has been suggested that the side with the thickly and deeply incised lines was done much earlier, while the lighter, thinner lines on the obverse were made much later. As determined by Javier Urcid, the earlier glyphs are similar to the Central Valley Zapotec-style that

275 dates to 200 BCE – 200 CE or the Terminal Formative period. However, unlike the Central

Valley style where glyphs were arranged in specific orders or linked with special personages, the

Nejapa signs are individually placed. The later glyphs, on the other hand, are much more clearly identifiable with common signs in Zapotec writing used from the Terminal Formative to the Late

Classic period (200 BCE – CE 850). While it is possible that the later carvings could represent a local version of 9th-century Zapotec-style writing conventions, Urcid also suggests that these glyphs may have been drawn in the 10th or 11th centuries, which may explain the use of new graphic conventions. A later date for the glyphs would suggest the survival of Zapotec writing in

Nejapa even after the collapse of Monte Albán (King et al. 2010: Appendix C, 152-159).

276

Figure 4.11: Side A of the Nejapa Viejo carved stone (drawing by Stacie M. King).

Arroyo del Sauso, another Classic period site, is remarkable because of the concentration of glyphs that have been engraved onto a bedrock spur on the west side of the arroyo (Figure

4.12). Carved over several different episodes, the glyphs extend west from a cascading waterfall to the upper part of the hillside; more glyphs likely existed in the past but have since been

277 destroyed over the centuries due to natural and artificial damage. Most of the signs are not recognizable as canonical glyphs of any known writing, and probably date to between the 6th and

9th centuries. It is possible that the makers of these carvings were not specialists in writing, explaining their unconventional style. The performance of carving the glyphs on a rock face near a waterfall, and hence an important source of water, may have been more important to the individuals who made them than creating readable content that would be recognizable to an audience. Finally, the unconventional glyphs may have been made by scribes or divinatory specialists who engraved them while under the influence of an altered or hallucinatory mental state experienced during ritual ceremonies (see King et al. n.d. Vol. I: Appendix B, 155-159).

278

Figure 4.12: Some of the carved glyphs found at the Arroyo de Sauso site (drawing by Javier Urcid and the PANT team.

At the same time, several of the signs are recognizable and are reminiscent of glyphs from both the Zapotec and Ñuiñe writing traditions. For example, there is a compound glyph composed of a cutaway view of a shell ornament on top and a stripe of knotted fabric on the bottom (see King et al. n.d. Vol. I: Appendix B, 158-159). Although the logophonic values are unknown, this compound shell-rope glyph is often accompanied elsewhere by Zapotec

279 inscriptions and the symbolic image of the sun. In some narrative scenes, it is portrayed as an object that is presented, obtained, or transferred in the context of genealogical archives. It also appears in emblematic texts and with frequency in agglutinated Ñuiñe texts. While some logographic values can be guessed at through their ritual and mantic connotations, others have more obvious denotative meanings related to the land and rain. Perhaps the motive for carving at the Arroyo del Sauso was as much about the negative as well as the positive potential of a cyclical and powerful river. This mountain river provided the water that not only supported nearby agricultural production and human reproduction, but also occasionally brought much destructive power, especially in the case of flash floods. It may be that the people who were most affected by the cyclicality of the river and its possibilities for creation and destruction made seasonal processions to this area, enacting rituals of petition and divination that included engraving the rock surfaces in order to bring balance to the unknown and unpredictable aspects of life (King et al. n.d. Vol. I: Appendix B, 158-159).

While mapping the ballcourt at Cerro Maluco (YAU-NEJ-NEJ-012), one of the men working for me brought a small ceramic piece for me to look at. Don Adrian had found it on his terreno located near another Classic period ballcourt site, Los Limares (YAU-NEJ-NEJ-007).

The fragment was a pectoral applique that personified the rain deity and is commonly seen on

Zapotec effigy vessels (Figure 4.13). The emblem conflates together three primary signs11. In the center is an oval with three vertically placed circles, symbolizing the three seeds placed into the cleft made by the coa or digging stick in an idealized corn field (A). On the sides of the cartouche are four loops with two circles set in each, suggesting the icon of nose or lip plugs (B).

When they appear in other contexts, these signs substitute as drops of rain, and their semantic

11 The emblem was kindly analyzed by Javier Urcid. 280 and logophonic value appears to be “precious”. Finally, below the central cartouche is an upside-down corn husk, or Glyph J in a non-canonical position (C). Glyph J, when in canonical position, represents an actual cob – normally, the trefoil element on top references the budding or the ripening of the corncob, while the lateral elements represent the husks’ open leaves (Stoll

2018: Appendix D: 93-94). In this non-canonical position, we cannot exactly be sure what

Element C represents, but given that glyphs in Nejapa are often placed in atypical positions, it is likely that this piece originated here and had not been imported from the Central Valleys. While it is difficult to determine the exact age for this piece, we can infer that it was made sometime between the 3rd century BCE and the 9th or 10th centuries CE, based on other glyphic evidence from Nejapa as well as the date for the site of Los Limares.

281

Figure 4.13: The pectoral fragment found by Don Adrian near Los Limares (upper) and Urcid's analysis of the glyphs in the pectoral (lower).

282

Additional evidence for the use of Zapotec-style glyphs are found at several other sites in

Nejapa. Three large stones located not far from the ballcourt at Los Limares (Chapter 6) feature glyphs, and there is also a boulder with glyphs that’s set at the northwest corner of a platform at another Classic period site, Llano las Casas, which also features an unusual style of ballcourt. I describe both in more detail in Chapter 6. While we have been unable to definitively identify an ethnic Zapotec presence or link Classic period sites in Nejapa to a “Zapotec”-identified ethnic group, these glyphs demonstrate that some people in the region at least understood Central

Valley Zapotec language and imagery (King 2010: Appendix C). That some of the glyphs closely resemble Ñuiñe script may hint at other trade and social relationships further away.

Overall, the Formative and Classic-period evidence points to social complexity and interregional trade networks developing early in Nejapa even though population levels were likely low.

As for the arrival of the Zaachileño Zapotecs in the 14th and 15th centuries, no direct evidence of conquest in the Postclassic has been found so far. We did not see evidence of burning, the complete supplementation of the local ceramic tradition by a foreign tradition, nor sharp disjunctions in settlement patterning or site layout – evidence commonly associated with episodes of conquest. One structure excavated at the site of El Órgano, a neighborhood of GLA, showed some signs of burning in the past – on the southwest side of the house, a nearly complete burnt adobe brick and fragments of other burnt bricks were found. Additionally, our excavations suggest that there had been an earlier structure under the surface structure, and that the orientation of the building had shifted or changed completely in between the two construction episodes. The burnt adobe brick and fragments were associated with the earlier structure.

However, the evidence indicates that the burning is more likely related to an accidental house

283 fire rather than conquest. The nearby temple, which would have been an excellent target for invading forces, was neither burnt nor destroyed. However, the abandonment of Greater La

Amontonada may have occurred just before or around the time of the Zaachileño migration.

Where the people went, we are not sure at this point, but nearby Colonial San Martín does overlap in time with GLA and is occupied long thereafter. Material evidence at both communities is distinct, which may point both to different peoples and changes due to time period, or simply material changes due to time.

On the other hand, defensive structures such as large walls at many of the Postclassic mountain sites suggests that some in Nejapa may have regarded the different foreign armies with suspicion or even outright hostility. Many of the mountain communities also have sight lines down into the valley lands. Such views would have allowed the inhabitants to monitor movement below and choose when to interact with any foreign travelers passing through (King and Zborover 2015:138-139). Sound travels very well in the mountains, and communication was certainly possible over long distances, allowing these mountain communities even more control over their interactions with neighbors and passing travelers (King and Zborover 2015:139). To this day, people moving through these spaces will make special calls to locate each other or announce their presence when approaching. Thus, people suspicious of the foreigners moving through the Nejapa area may have sought refuge at mountain settlements.

Meanwhile, other communities such as Colonia San Martín may have entered into cooperative alliances with the foreign Zaachila-Zapotec, Aztec-Mexica, and later Spanish armies

(King 2012). Considering Colonia San Martín’s location on the valley flood plains during what would have been the height of Zapotec and Aztec movement to the Isthmus, it appears that the people in this settlement, at the very least, were not overly concerned with defense nor did they

284 fear (or flee) the Zaachila-Zapotec and/or Nahuatl presence. That there were Nahuatl translators in Nejapa certainly points to the complicated relationships between communities here and external ethnolinguistic groups. Finally, the Zapotec –style glyphs in the earlier Classic period complicate our timeline for Zapotec migration into the Nejapa region. While we have better evidence for Zapotec penetration into the formerly Chontal-dominated Quiechapa, the evidence is much more ambiguous in Nejapa. That some of these glyphs resemble Ñuiñe iconography perhaps suggests that people in Nejapa were participating in pan-regional writing traditions, rather than a definable Zapotec-ethnic identity or indirect/direct imperial control emanating from

Monte Albán in the Central Valleys.

The archaeological evidence also does not point to any clearly defined ethnic identities linked to any specific group. Some communities shared material culture even though they are located several kilometers away and at drastically different elevations; while other communities present radically different material culture even though they are right alongside each other.

Interestingly, what seems to affect ceramic assemblages the most, at least at the mountain sites, is proximity to the trade corridors; settlements closer to these routes obviously had more access to and interactions with foreign traders and goods. In particular, there is a large concentration of a type of plain grayware resembling similar wares commonly argued to be exclusively

“Zapotec”, and often used as evidence for Zapotec control or presence (King 2010; King et al.

2012). However, other scholars have argued (King 2011; Levin 2013; Workinger 2013;

Zborover 2014) that given the widespread popularity of this ceramic type in the Postclassic (and earlier), graywares should not be seen as indicative of economic or political hegemony by

Zapotec-speaking peoples. Instead, communities appear to have drawn on locally shared ideas as well as techniques and styles that linked them to other groups. Studies of grayware ceramics

285 in other regions of Oaxaca have demonstrated that these ceramics were produced using local techniques and design preferences (Levin 2013; Workinger 2013)

Support for a de-emphasis on distinctly marked ethnic identities can also be seen in the historical documents from colonial Nejapa. As mentioned previously, there is the alliance in the

1600s between the people of Nejapa and Villa Alta in order to fight against Spanish authorities.

Although in some documents the Zapotecs, Mixes, and Chontales are described as living in specific geographic zones (valley vs. mountains vs. caves, respectively) within the Nejapa subregion, these same Spanish Colonial sources also use these exonyms interchangeably. A single community is described as Mixe in one Dominican source, while in another document they record the story of a priest having to learn Chontal in order to communicate the liturgy to the people living there (King and Zborover 2015:143). Badillo (2017:40) reports similar interchangeable use of ethnic exonyms by Spanish in Quiechapa.

As I discussed in Chapter 2, identity can be highly flexible in borderlands, boundary areas, and frontiers zones. In these pluralistic contexts, frontiers are sites of novel ethnic group formation through the merger, fragmentation, disintegration, and renegotiation of ethnic boundaries. Frontiers also give rise to cultural hybrids and ambiguous identities through interactions with other cross-cutting and overlapping social units. Given these lines of evidence, it is likely that in Nejapa, ethnolinguistic-based identities were seen as fluid or as less important than other group ties (following Horn 2014). King and Zborover (2015) propose that the shared histories and settlement traditions of the people who inhabited the northern corridor of the eastern Sierra Sur worked to foster commonly-held social identities. These social identities may have been much more meaningful to the people living in Nejapa and surrounding areas than the ethnoyms used in the current literature and Mexican ethnic classification systems (King and

286

Zborover 2015:132). They further argue that this diversity meant that the people living here would “have had to be comfortable and skilled in moving between different cultural worlds, languages, and environments,” (King and Zborover 2015:132). That contradictory interpretations of the political and cultural composition of the eastern Sierra Sur existed in the past as well are evidenced by ethnohistoric documents and lienzos from the Early Colonial period. Apart from the interchangeable use of ethnic exonyms, we also have an example where the Mixe and Chontal made land claims in the colonial period (via these lienzos) that would have meant that their original territories overlapped each other in Nejapa (Acuña 1984; Barros van

Hövell tot Westerflier 2007; King and Zborover 2015; Zborover 2008).

4.5 CONCLUSION

In this chapter, I have described the known evidence for the sociopolitical development and settlement history of Oaxaca from the Formative period to the Postclassic. I have attempted to show that our understanding of this development and history is conditioned by an intense research focus on the two most accessible regions of Oaxaca, giving us only part of the story of this critical macroregion. Nejapa is one such subregion in Oaxaca that has been traditionally seen as peripheral to the larger events and historical trends occurring in the Central Valleys and

Mixteca Alta, but that actually played a much bigger role in interregional politics and trade networks in the Prehispanic era than previously thought. Evidence from Nejapa also complicates our interpretation of Oaxacan settlement history and social identity in the past.

Given its close ties to other regions and community identities that do not seem to be strictly based on defined ethnolinguistic affiliations, it is likely that in Nejapa, ethnolinguistic identities were not overtly emphasized or marked, and more importance was placed on other types of group ties. This was likely a widespread attitude in Oaxaca, as it is increasingly

287 becoming clear that there was never a unified “Zapotec” or “Mixtec” state or even identity.

Instead, factional divisions cut across, and within, sociocultural and ethnolinguistic boundaries

(Blomster 2008:12). The northern Mixteca Baja, one example among many, was also a multilingual zone where people spoke various Mixtecan dialects as well as Chochopopoloca

(Joyce 2010:237), while in Coixtlahuaca during the Postclassic Mixtec and Chochoteca-speakers lived alongside each other, albeit in separate neighborhoods (Kowalewski et al. 2017; Maunter

2015). Evidence from Quiechapa suggests that Chatino and Chontal speakers lived there before the arrival of the Zapotecs, with the former two eventually disappearing overtime (Badillo 2017).

The Mixtecs of the Costa Chica exerted influence over Chontal communities via Zapotec or

Pochutec allies (Zborover 2014). These and other examples demonstrate that multiethnicity was not an unusual phenomenon in Oaxaca. With respect to Nejapa, rather than being bounded solely by ethnic ties, people were potentially making connections to various groups and regions, resulting in possibly heterogeneous communities.

288

CHAPTER V

BALLCOURTS DO NOT PHOTOGRAPH WELL: THE METHODOLOGY OF MAPPING BACLLOURTS

“If materiality is the means through which social actors transform fleeting identities into historical facts, then the different forms of permanent marking of the landscape documented in these examples must be understood as the result of conscious actors using architecture to write different forms of community onto the landscape” Joyce and Hendon 2000:154

“Living in a landscape requires moving through it via one locale to another” Robinson 2010:801

5.1 INTRODUCTION

Spatial analysis has long been used as a means of exploring the built environment and how social space is consciously subdivided and allocated. As I discussed in Chapter 2, human beings are spatial creatures that articulate with their surrounding environment through our physical bodies as well as our socially and culturally mediated practices (Bender 1993; Thomas

1993). Natural and cultural landscapes are socially constructed and historically contingent, marked by distinct institutions and systems, and possessing multiple meanings for individuals and communities throughout space and time (Bender 1993; Smith 2003; Thomas 1993;

Wansleeben and Verhart 1997; Witcher 1999). How the built environment is organized reflects, constrains, and influences the power relations inherent in social structures (Dovey 1999;

Foucault 1979). Moreover, architectural design, construction, and layout often represent

"communities of practice" (Wegner 2000) that can be linked back to different groups, locations, and social identities. The arrangement of architecture and settlements and the spatial

289 relationships between sites can be analyzed to study the human activities that produce them

(Kolb and Snead 1997:614). By taking into consideration elements such as access, visibility, capacity, and use (Fairclough 1992), we can learn much about how societies in the past organized themselves culturally, materially, and ideologically.

Because ballcourts were public structures used for staging community-oriented events, understanding their sociopolitical functions and how they may have actually been used by people is most effectively accomplished at the regional scale (Fox 1994). Again, monuments are more than just individual symbols of status and power. In a collective sense, they create cultural and historical landscapes through which local, chiefly, sociopolitical and religious ideologies were enculturated, modified, and normalized (after Burley 1996:421). Monuments gain(ed) power and sociocultural capital or influence through communal participation in their construction, the physicality of labor and bodily presence, and from the activities that took place there over time, which often involved citations of socially efficacious practices and appeals to dominant or communal religious beliefs. By focusing on the spatial properties of public monuments such as ballcourts at the community (local) and supracommunity (regional) scales, we can move towards recovering some of the social meaning of these specially marked structures. The social and political implications surrounding the sponsorship and performance of the ballgame depended in part on its emplacement within both the built environment and the larger landscape overall.

This dissertation specifically considers two things: first, what is similar or different about the Nejapa ballcourts as a group; and second, why they are located where they are in the landscape (Turnball 2002), and what that distribution can tell us about sociopolitical organization in Nejapa during the Classic and Postclassic periods. There are many tools available to archaeologists for studying the spatial properties of both landscapes and structures. While

290 traditional12 spatial analysis techniques have been used for decades, Geographical Information

Systems-based (or GIS) analyses available through many computer programs are especially useful for integrating and interrogating spatial data, and have become increasingly popular in archaeology over the last 20 years. In this chapter, I look at the GIS spatial analysis techniques used in this dissertation, exploring the theory and methodology underlying its use as well as the critiques against it, especially with regards to the specific analytical techniques I employ here

(Chapter 7). Finally, at the end of the chapter I summarize the field and laboratory methods of the PAJP-NT research project.

5.2 USING GIS TO INVESTIGATE THE SOCIAL ROLE OF THE BALLGAME As visible, public structures that were often integral structures to their communities, however defined, ballcourts are excellent subjects for site and regional scale spatial analysis. By measuring their centrality and exclusivity with respect to placement within and between communities, where they are located on the landscape with respect both to the physical environment and the social hierarchy, and how each structure could have been perceived visually from different locations in that landscape can tell us much about the importance of the ballgame in intra- and intercommunity relationships. While traditional spatial techniques can be used to do this, the analytical power offered by computer-based spatial analysis programs based on

Geographic Information Systems really offers archaeologists the ability to organize and investigate the spatial properties of archaeological phenomena of varying types in a single database (Conolly and Lake 2006; Kantner 2008; Kvamme 1990; Lock 2003; Wheatley and

Gillings 2002). GIS has become popular due in large part to its capacity to integrate different forms of social and environmental data with archaeological data, and for creating models that

12 I use traditional in the sense that these techniques are largely non-computer based. Although computers may have been used to perform some of the calculations, the programs themselves are not specifically spatially-oriented. 291 can help provide further insight into past behaviors, facilitating our ability to analyze human landscapes and human-environment (both cultural and natural) interactions (Couclelis 1999

(Gaffney et al. 1996; Kvamme 1995; Lake, Woodman, and Mithen 1998; Wansleeben and

Verhart 1997; Wheatley and Gillings 2000; Kantner 2008; Stancic and Kvamme 1999; Lock

2000; McCoy and Ladefoged; Wheatley 2003; Wheatley and Gillings 2002; Berry 2003; Dyson-

Bruce 2003; Ford 2007; Gregory and Ell 2006; Limp 2001; Ryavec 2001; Stichelbaut 2006).

GIS is a powerful tool for studying regional migration and settlement patterns (Elliot

2005; Galaty 2005; Jennings and Craig 2003; Jones 2006; Peterson and Drennan 2005), ancient agriculture (Bevan and Connolly 2002; Friedman et al 2003; Ladefoged and Graves 2000, 2008;

McCoy 2005, 2006); warfare (Field 2005) and even intrasite spatial patterning (Jones and

Munson 2005; Holdaway et al. 2005). Most research has focused on how past social and economic organization was influenced by how the environment was both used and perceived

(Kosiba and Bauer 2013), and archaeologists today rely heavily on GIS as the principal analytical tool to investigate human-environmental relationships (e.g. Aldenderfer and Maschner

1996; Bauer et al. 2004; Casana and Cothren 2009; Chapman 2006; Connolly and Lake 2006;

Howey 2007; Kosiba 2011; Lake and Woodman 2013; Llobera 2003, 2007; Lock 2000; Spikens

2000; Wernke 2007; Williams and Nash 2006). But research also shows the significance of agency and historical contingency in explaining archaeological phenomena (see chapters in

Meskell and Preucel 2004), and new developments in visualization offer tools that can incorporate and analyze contextual and qualitative data (McCoy and Ladefoged 2009:281).

Recent research efforts have especially focused on incorporating culture and belief systems into spatial models via the inclusion of more subjective measurements (Fitzjohn 2007; Gaffney et al.

1996; Llobera 1996; Wansleeben and Verhart 1997; Witcher 1999), as well as by taking into

292 account the historicity of sensory perceptions such as vision and hearing (Giles 2007).

Paradoxically, GIS can reveal elements of the past that might “otherwise fall outside the Western gaze” (McCoy and Ladefoged 2009:266; see also Bender 1999), while at the same time maps and visual representations can hide complexities and reinforce expectations (Hodder 2000).

In GIS, datasets can be manipulated mathematically to produce new rasters displaying network, visibility, movement, cost-surfaces, and erosion, among others. Elevation data in particular can be used to help understand how human beings use space and landscape by building visibility, movement, and ecological models. Surface modelling is an important analytical tool that uses digital elevation models derived from the interpolation of sets of discrete observations in order to create a continuous surface (Connolly and Lake 2006:90). In line with the traditional visual-structural analyses used by Higuichi (1987) and Moore (1996a, 1996b), there is much emphasis on measuring visibility and perception within GIS models (Clark 2007; Fitzjohn 2007;

Frieman and Gillings 2007; Giles 2007; Witcher 1999 Wheatley 1993, Witcher 1999, Wheatley and Gillings 2000; Connolly and Lake 2006), and some have even attempted to extend analysis to other possible human sensory perceptions (Frieman and Gillings 2007). In fact, most landscape studies in GIS use visibility measurements and analysis as a way to identify why certain places were chosen as sites of activity (Frieman and Gillings 2007; Lake and Woodman

2003; Llobera 1996; Wheatley and Gillings 2000: 1).

Another approach using spatial data is dynamic modelling, which studies how phenomena change over time. This is often done through agent-based simulation models, a way for archaeologists to both predict and explain past human behaviors and decision-making using actual landscape factors (Connolly and Lake 2006:47). Visibility graphs and agent-based computer models generated in GIS programs such as ESRI's ArcGIS (the program used in this

293 dissertation) can be used to more effectively model the 3D experience of space and the role of human agents (Clark 2007; Giles 2007). Using these models in combination with archaeological data can be useful for looking at the unusual shapes and sizes of buildings or architectural elements as well as to reconstruct the sensory ecologies of past environments (Dawson and Levy

2005; Dawson et al. 2007) that would have affected their perception.

5.2.1 GIS Theory and Critiques

GIS studies in archaeology have employed sharply contrasting theoretical approaches to landscapes and environments (see Anschuetz et al 2001; Ashmore and Knapp 1999, David and

Thomas 2010; Kosiba and Bauer 2013; Smith 2003). Some have used GIS to examine systemic, cultural adaptations to natural climatic and geographic conditions, thereby framing landscapes or environments as terrains of social or economic resources (e.g. Anderson and Gillam 2000; Jones

2006; Wescott and Brandon 2000). Others have applied GIS to explore how societies assign cultural significance to their environment, treating landscape more as a topography of meaning and memory (e.g. Chapman 2003; Llobera 1996, 2001) than as a network of resource patches.

Kosiba and Bauer (2013:62) refer to the former as the econometric approach, which concentrates on the physical attributes of a landscape or environment and the human relationships that center around those attributes. In this respect, GIS is used more in a processual, empirical or quantitative approach (McEwan 2012:527). Economic theories associate archaeological activities with environmental patterns and propose that human behavior can therefore be predictably modeled since it is continually affected by local and regional environmental conditions (Ebert 2000; Kvamme 1990; Parker 1985; Woodman and Woodward 2002). The latter is the interpretive approach, which focuses more on the places, practices, and concepts that

294 shape people’s perceptions and experiences of that same environment (Kosiba and Bauer

2013:62). The interpretive approach draws on phenomenological theory, itself grounded in postmodern geographical theories as well as postproccesual critiques in archaeology. I have previously discussed the phenomenological definition of landscape in Chapter 2. Suffice to say, people engage with their landscape through both bodily experience and perception, reproducing cultural meanings through those interactions (Kosiba and Bauer 2013:64; e.g. Bender 1998; Feld and Basso 1996; Gosden 2001; Tilley 1994, 2004; Tuan 1977, 1989). Interpretive studies therefore emphasize a subject-oriented perspective within a “hermeneutic and inductive framework” (Kosiba and Bauer 2013:64).

There has been a long history of debate between these two different approaches to landscape in GIS (see Bruck 2005; Fleming 2006; Gillings 2012; Johnson 2006; Llobera 2011).

Econometric approaches, the phenomenologists argue, are problematic because model data is often incomplete, biased, and not appropriate for quantifying or qualifying past human behavior

(Wheatley 2004). Because of their seemingly narrow focus on the economic function of environment, they ignore or leave out the sociopolitical context in which societies and settlement systems are produced and reproduced (Smith 2003; Kosiba and Bauer 2013:64). This has led to charges of environmental determinism (see Gaffney and van Leusen 1995). Proponents of the interpretive approach argue that GIS-based mapping methods and predicative techniques dehumanize and distort the past because it is based on the Western-oriented Cartesian gaze

(Bender 1999; Cosgrove 1984; Thomas 1993a,b, 1999, 2004:198-201, 2008; Tilley 1994, 2004).

However, there have also been many critiques of the interpretive (i.e. phenomenological) approach (e.g. Bruck 2005). In particular, critics have argued that phenomenological research methods are highly subjective and/or not sufficiently rigorous (Fleming 1999, 2005, 2006;

295

Johnson 2011; Shennan 2002, 2011:325). The subject-oriented perspective advocated by interpretive approaches is based on the researcher’s own bodily engagement with the landscape, but we cannot assume that such observations are isomorphic to peoples’ engagement with or perception of the landscape in the past, nor the sociocultural values and meanings embedded within it. The highly-subjective nature of the interpretative approach ultimately means that the results are not replicable. As Kosiba and Bauer (2013:66) rightly point out, both approaches ultimately fail because they tend to generalize human behavior and assume that all people valued the environment in similar ways.

Though GIS has been increasingly incorporated into more subject-centered research, there is still much resistance towards its use (e.g. Gaffney et al. 1996 Tilley 2004: 218; Thomas

2004) because the techniques are seen as anathema to the emerging progressive theoretical agenda in recent studies of space and landscape (e.g. Gaffney and van Leusen 1995). Those that did attempt to use GIS often ran into the toolbox problem, wherein archaeological data was sometimes compromised in order to fit the requirements of the GIS analytical tool (Gillings

2012:604), most frequently seen in visibility or viewshed analyses (Wansleeben and Verhart

1997: 60-61). There are certainly empirical and methodological limitations to applying GIS techniques to interpretative studies that have to be acknowledged (see Fontijn 2007, Lake and

Woodman 2003; Llobera 2007; Tschan et al. 2000; Wheatley and Gillings 2000). For example, viewshed analyses can be too coarse to replicate human perception (Kosiba and Bauer 2013:65).

Moreover, the digital visibility of raster cells is often equated with their visibility in the physical environment (Llobera 2007; Ogburn 2006). Another common assumption made in GIS analyses is that past actors used, perceived, and experienced the environment in similar ways, although a few studies have attempted to address how experiences and perceptions of the environment may

296 have been different for people of different social status and groups (cf. Fitzjohn 2007; Kwan

2002; Koisba and Bauer 2013).

5.2.1.1 Summary

One of the primary concerns regarding the use of GIS is environmental determinism and whether the program, because of the assumptions built into the analytical tools, biases our interpretations towards environmental explanations for historical change (Aldenderfer and

Maschner 1996; Allen et al 1990; Lock 2000, 2003; Lock and Stancic 1995; McCoy and

Ladefoged 2009). This issue is further compounded by that fact that GIS spatial technology is often applied without first considering a workable theoretical framework, leading to “excessive gratuitous application without regard to purpose, appropriateness, or theory” (McCoy and

Ladefoged 2009:282, citing Drennan 2001:668). While some see GIS as another way of

“exploring place and landscape as experiential phenomena” (Rennell 2012:512); others argue that it is theoretically sterile and regressive (Sui 1994).

Even given these critiques, when carefully applied and using the appropriate statistics,

GIS-based computer programs can be used to measure the more subjective properties of the landscape, such as those linked to visibility and even other human sensory perceptions. In their analysis of recumbent stone circles constructed between the Middle Neolithic and the late

Bronze Age (c. 3500-1000 BCE), Lake and Woodman (2003) found that there was a statistically significant preference for placing these structures in specific locations where the far horizon enhanced the circularity of the stone circles. While the social and symbolic significance of this circularity is not clear, the GIS results do provide a fruitful starting point by confirming that

297 there indeed existed specific practices or beliefs that influenced the choice of location for constructing the circles.

Overcoming the reductive constraints of GIS approaches requires that we as researchers do not simply limit ourselves to environmental properties, but also explore the social (and sometimes more intangible) properties – differences and similarities, boundaries and accessible areas –that constituted ancient landscapes (e.g. Bauer 2011; Kosiba 2011; Lindsay 2011; Rizvi

2011; Wernke 2007). The increasing focus on generating socially meaningful results from GIS analysis in archaeology has come about in part due to the growth in landscape archaeology, which has improved our use of viewshed and cost surface functions and lead to more efforts to properly model human perception and movement in time and space (McCoy and Ladefoged

2009:282). When we ourselves “engage” with computer-based spatial programs in order to measure the tangible and intangible attributes of social landscapes and the human-environment dynamic, we should remember that GIS is neither a neutral or passive process (Kwan 2002, see also Wheatley 2000). Instead, like many analytical methods in archaeology, GIS results should be treated as an “interim stage” in interpreting the data, rather than the final explanation (Lock

2000:61).

In the following sections, I take a critical look at the specific methods I use in my spatial analysis, evaluating their appropriateness for analyzing ballcourts in meaningful ways. I also address the criticisms of each method and how I attempt to accommodate for these critiques.

These specific methods are predictive modeling and cluster analysis, least-cost path analysis, and visibility or viewshed analysis.

298

5.2.2 Finding Ballcourts on the Landscape: Predictive Modeling and Cluster Analysis

Attribute-based analysis is a critical technique for evaluating archaeological phenomena in GIS. This primarily takes the form of predictive modelling which, as the name suggests, is a method for predicting the value of a dependent variable in unsampled locations using one or more independent variables (Aldenderfer and Maschner 1996; Conolly and Lake 2006; Gillings et al. 1999; Kvamme 1989, 1990, 1995, 2006; Lock and Stancic 1995; Mehrer and Wescott

2006; Wescott and Brandon 2000). Predictive modelling in GIS is popular because archaeologists can quickly collect and analyze environmental data from locations of interest for statistically significant patterns (Kvamme 1990; Wescott and Brandon 2000). Because regional environmental information is organized into data layers in GIS applications, simple mathematical procedures can be performed that relate variables, i.e. “attributes” together that allow archaeologists to use known site locations to identify those variables that influenced past human settlement patterns (e.g. Duncan and Beckman 2000; Perkins 2000; Warren and Asch 2000), as well as predict where archaeological sites should be found (e.g. Ebert and Singer 2004; Elliott

2005). These variables can include known factors such as elevation, slope, aspect, visibility, and distance to water, among others (Connolly and Lake 2006:46). Essentially, we can use these factors to test models for why sites are located where they are, find sites in unknown locations, and explore alternative variables that influenced settlement patterns.

Once the appropriate attributes have been selected and confounding has been accounted for (Woodman 2000b:452), the next step in the process is to build the predictive model

(Connolly and Lake 2006:183). The preferred technique for doing this is via logistic regression analysis (Carrer 2013; Menard 2001), which can use a combination of variables of different scales (i.e. nominal, ordinal, interval, and/or ratio data) (Connolly and Lake 2006:183), allowing

299 for the inclusion of environmental and social attributes. Understandably, this type of analysis is often critiqued for being environmentally determinative because of the overreliance on a rather limited range of environmental variables (Connolly and Lake 2006:179-180; Gaffney and van

Leusen 1995), although the inclusion of social variables should be a useful corrective to this problem. Predictive modeling can be further enhanced with linear regression analysis, which is used to ascertain the relationship of two continuous variables, including spatial factors (Connolly and Lake 2006:45). This can help to evaluate to what degree these same variables are influencing the distribution of archaeological phenomenon, as some variables may have more of an affect than others.

5.2.3 Movement on the Landscape: Cost-Path Surfaces and Path-Distance Analysis

Another important issue to consider is where ballcourts are positioned with respect to the movements of people through the landscape. This touches on the accessibility of courts to people in a given settlement area, and what courts may have been used by which communities.

How people move through a landscape is greatly determined by both environmental and social factors (Llobera et al. 2011; White and Barber 2012). Why a person may bypass certain locations in favor of others can reveal much about the economic and sociopolitical organization of the landscape at several scales. For example, an individual may travel from their location at

Place A to Place D, even though Place C may be closer physically, because Place D is easier to travel to in terms of energy costs. At the same time, an individual may bypass Place C, even though it is more accessible, because it lies in hostile or forbidden territory. How sites or archaeological phenomena are related to each other based on the costs of travel or energy expended between them is explored in GIS using path-distance modeling (e.g. Anderson and

300

Gillam 2000; Gaffney and Stancic 1991; Harris 2000; White and Surface-Evans 2012; Whitely and Hicks 2003), which is also known as least-cost analysis.

Cost or friction surfaces, generated in GIS using DEMs or digital elevation models and slope rasters, calculate the costs of movement across 3-dimensional space and can therefore theoretically identify the likely pathways used by people in the past (Anderson and Gilliam 2000;

Harris 2000; 2002; Gaffney et al. 1996: 135; Kantner 2008; Llobera 2000; Whitley and Hicks

2003; Wheatley and Gillings 2002; White and Barber 2012; White and Surface-Evans 2012), as well as define past territorial boundaries (e.g. Hare 2004) and the relationships between settlement patterns, site locations, and economic resources (Lock and Harris 2006). Like predictive modeling, cost-path analysis can also be used to predict site locations based on variables such as hydrology, soil types, vegetation, slope, and/or potential agriculture productivity (Mehrer and Wescott 2006; Wescott and Brandon 2000). But it also can be help archaeologists understand regional social and economic networks (e.g. Howey 2007); identify possible routes for the colonization of new territories (e.g. Rademaker et. al 2012); and to define relationships between regional types and environmental variables through catchment-area modeling, optimal-foraging behavior, and modeling prehistoric pathways (e.g. Limp 1991;

Madry and Rakos 1996).

Studies of movement in archaeology primarily use GIS to create mathematical prediction models that map where movement most likely occurred, or was channeled, on a specific landscape (White and Barber 2012:2684). In modeling transport routes, the accumulated cost of travelling, i.e. the cost surface, is calculated outward from an origin point and then the route of steepest reduction in accumulated cost is calculated from the destination point back to the origin

(Connolly and Lake 2006:252). In GIS terms, a numeric expression is generated that expresses

301 how difficult movement would have been between any given individual cells in the raster

(Collischonn and Pilar 2000:397; Llobera et al. 2011; White and Barber 2012:2684). Next, an accumulated cost-surface is created to represent the movements costs both away from and towards a targeted cell on the raster maps (Howey 2007:831; Llobera et al. 2011:844; White and

Barber 2012:2684). A final step is to represent graphically the least costly routes of movement going to and from the origin cell (Harris 2000; Tobler 1993; Surface-Evans and White 2012;

White and Barber 2012:2684). While cost-path or least-cost analysis is rather popular in archaeology, Wansleeben and Verhart (1997: 59) argue that cost surfaces are too close to catchments analysis, which problematically posits a straightforward relationship between settlement, landscape, and economy. Therefore, human land-use pattern must be understood in terms of both the potential movements of people through a landscape and the relative accessibility of the different locations under study (Connolly and Lake 2006:214).

The Nejapa landscape is not flat – in fact, it ranges in elevation from 600 m above sea level at the valley floor to 2800 m at the highest peak on the eastern side of the valley floors.

Travel through this landscape would have been affected by the changes in slope as one moved from one location to another. Llobera (2000) suggests that we can combine the cultural influence of monuments with the energetic costs of traversing the terrain. Thus, path-distance modeling can help to identify which ballcourts may have been accessed by which communities based on the ease of travel between these sites. Such analysis can be further refined using chronological controls. Unfortunately, path-distance modeling does not include vegetation, which can facilitate or delay travel, nor the social factors that may affect the places a person can travel through or to safely. Conflicts between communities, a common occurrence even today in

302 modern Oaxaca13, can make movement into another community’s territory dangerous. However, path-distance modeling is still a useful heuristic tool for an initial assessment of how people may have moved through ancient landscapes, and can help to identify if factors other than energy costs linked sites together.

5.2.3.1 A Cost-Path Case Study: Predicting Transportation Networks in Oaxaca

The cost-path or path-distance techniques in GIS can only really compute one-to-many and many-to-one relationships (Fabrega-Alvarez 2006; Llobera et al. 2011), an approach that is most applicable for research focused on a small number of known locations (White and Barber

2012). Identifying potential transport movements across regional landscapes is much more difficult because it requires modelling many-to-many relationships (White and Barber

2012:2685). Using Oaxaca as their case study, White and Barber (2012) develop a From

Everywhere to Everywhere (FETE) approach that allows researchers to generate probabilistic landscape movements when the origins and destinations are not assumed. Their results are especially useful for archaeologists studying intercommunity networks in Oaxaca, since the mountainous terrain tends to (forcibly) channel movement through a “limited range of topographic breaks,” (White and Barber 2012:2685).

The movement corridors generated by their FETE approach are independently corroborated by Oaxaca’s ethnohistoric record, suggesting that the results closely match the approximate routes utilized in the Prehispanic era (White and Barber 2012: 2692). Important transportation routes identified in the FETE results that are historically documented include: a

13 While I was writing this dissertation in July and August 2017, there were issues with garbage pickup due to internal problems in the community where the garbage dump is located. The community had split into two, there were no government officials in charge, and several people were murdered. 303 high-traffic route through the Central Valleys north to the Tehuacan Valley and southeast to the southern Isthmus (Redmond 1983; Zeitlin 1978b); and another major route into the Central

Valleys through the mountain town of Sosola, which was the location of an Aztec garrison (Ball and Brockington 1978:111), a route the FETE results also linked to other important trade centers such as Tamazulapan in the Miahuatlán subregion of the Sierra Sur (White and Barber

2012:2692). Another high-traffic route identified by the FETE output included travel along the

Rio Santo Domingo, a tributary of the large Rio Papaloapan connecting the Cuicatlan Cañada with the Gulf Coast (White and Barber 2012:2692). This route is also documented by archaeological and ethnohistoric evidence (Ball and Brockington 1978; Rees 1975). At the same time, there is a high-traffic route that bypasses the Rio Santo Domingo and connects the Central

Valleys to Tuxtepec on the Gulf Coast. Another series of high-traffic routes extend northwest- southeast between Ejútla and the coast near Huatulco as well as the west and central coasts

(White and Barber 2012:2692). Importantly, marine-shell production has been extensively documented at the Ejútla site, showing that this connection to the Pacific coast figured prominently in their economic networks (Feinman and Nicholas 1990).

The FETE output also highlighted routes that currently receive little attention by archaeologists. The results predict a high-traffic east-west corridor passing into the Sola Valley from the southern Central Valleys. Less heavily trafficked corridors branch from this route south to the Pacific coast. Other predicted high-traffic routes travel between the coast and the Mixtec highlands that completely bypass the Central Valleys, and hence Monte Albán. In fact, the

FETE results reveal a number of routes that traverse the Mixtec highlands, supporting the archaeological evidence for diverse and long-distance connections in the area generally (White and Barber 2012:2692-2693). The output also predicts a series of routes that run east-west along

304 the coast, suggestive that there was extensive trade in coastal goods occurring outside of the control of Monte Albán. Importantly for my purposes here, the FETE maps show that the

Nejapa subregion not only falls along the Tehuantepec route, a major route of transportation between Central and Southern Mesoamerica, but also connects in the north-northwest with the

Mitla-Isthmus route that runs partly through the Sierra Mixe region (White and Barber

2012:2693-2694). The ethnohistoric record in Oaxaca clearly shows that movement in these landscapes “embodied specific histories, social values, and spatial perceptions,” (e.g. Zeitlin

2005:14; White and Barber 2012:2694). When evaluating the results of path-distance analyses, we should take these facts into consideration, especially in Oaxaca.

5.2.4 Sighting Ballcourts on the Landscape: Intervisibility, Viewsheds, and Affordances

As mentioned previously, most attempts to replicate sociocultural perceptions of the environment and archaeological phenomena in the past in GIS are done through visibility or viewshed analyses, which is the modeling of the visibility of places, land attributes, and other archaeological phenomena a class of observers (Gaffney at al 1996; Llobera 1996, 2000;

Maschner 1996; Pollard and Gillings 1998; Ruggles and Medyckyj-Scott 1996; Wheatley 1995,

1996; Rennell 2012). Most research has utilized GIS to examine the intervisibility between sites and features and how this may have influenced people’s perception of their relationships to places, other people, and even to their own past (Kosiba and Bauer 2013:65), as well as to assess how the visibility of environmental features might have been helpful for monitoring game, supervising agricultural fields, and/or overseeing important resource patches and access routes to them (Krist and Brown 1994; Lock and Harris 1996; Maschner 1996); and how archaeological phenomena were incorporated into ancient peoples’ decision making with regards to location and

305 feature forms (Ayala and Fitzjohn 2002; Jones 2006; Lake 2007; Ogburn 2006). However, a potential issue with visibility analysis is that explanations of site locations must assume that visual qualities are linked to perceptual and symbolic associations (Gaffney et al 1996; Wheatley

1996).

Intervisibility or visibility analysis in GIS is based on the “automatic determination of whether any given pair of points are intervisible”, an operation typically performed on DEM rasters that works by projecting straight-of-line sights from a viewpoint to the selected target

(Connolly and Lake 2006:226). Technically, a viewshed is the set of target cells that are visible from a defined viewpoint or observer. Single viewsheds are binary maps where cells are simply marked as visible or not visible from the viewpoint, while multiple or cumulative viewsheds are created using a logical union to join two or more viewshed maps (Wheatley 1995). In these maps, each cell records whether it is visible from at least one viewpoint (Connolly and Lake

2006:226). Combined cumulative viewsheds generated for every possible viewpoint, where each map cell records the number of other map cells to which it is visible, are known as a total viewshed map (Llobera 2003). While earlier applications of visibility analysis relied on simple viewsheds, recent studies have begun to incorporate more realistic estimations of the limitations of human vision using Higuichi and fuzzy viewsheds (see Fisher 1991, 1992, 1994, 1995, 1996,

1998 for earlier discussions about developing cumulative viewshed models). First developed by

Wheatley and Gillings (2000), Higuchi or fuzzy viewshed more closely model the environmental conditions a viewer might encounter and how distinct an object might be, or how visible or distinguishable a location is, by including factors that affect visual perception (Fisher 1994: 168).

Additionally, an increasing decay value is assigned to cells located further from the targeted object, much as our ability to perceive objects decreases the further they are away.

306

Much of the research regarding viewsheds has been on the placement of monuments and other social features on the landscape (Connolly and Lake 2006:225; Wheatley 1995, Woodman

2000a, Fisher et al. 1996, Llobera 2001). Swanson (2003) uses digital elevation data to quantify the intervisibility of hilltop features that had been hypothesized to be fire-signaling stations around the site of Paquimé. The results reveal that while these features can be linked together into line of sight network, the topography of the region naturally promotes intervisibility. Kosiba and Bauer (2013) use viewshed analysis to test whether specific architectural categories correspond to differences in the visibility of surrounding spaces. The results show that the more elaborate residences were built in locations where the visibility of particular spaces and sites could be maximized, while commoner households were built in places where civic or ceremonial spaces would be most visible to them. At the same time, the more elaborate residences were also positioned for “maximum surveillance of the commoners’ residence” (Kosiba and Bauer

2013:82).

Other studies have used even more subjective measurements to understand human perception and engagement with the landscape. Williams and Nash (2006), for example, analyze site viewsheds in the Andean mountains to see how they might be linked to ancient relations beliefs. Briault’s (2007) viewshed analysis of peak sanctuary cults on the island of Crete, on the other hand, suggests that visibility and landscape location were less important than had been previously assumed. To test the idea that Mesolithic sites were ideally positioned to provide more extensive landscape views than in other areas on the landscape, Lake and Woodman (2002) ran a series of viewshed analyses to see how much of the landscape was visible from known site locations in comparison to non-site locations. Yet still in other studies, viewshed analyses of architectural structures reveal that both religious and residential buildings were constructed in

307 such a way that they both restricted and channeled intervisibility within the site as well as people’s experience of these sites (Clark 2007; Dawson et al 2007).

Of course, there is always the danger that GIS-viewsheds become uncritical orthodoxy in archaeology (McCoy and Ladefoged 2009:237; Wheatley and Gillings 2000:1), and several researchers have expressed skepticism about the total applicability of viewshed functions to every case (eg Fitzjohn 2007; Freiman and Gillings 2007; Wheatley 2004). Thomas (2004,

2008) strongly critiques the god’s eye perspective predominant in GIS visualizations that only work to encourage an objectified view of the world in the past. Wheatley and Gillings (2000) rightly point out several potential problems with visibility analysis in GIS, including computational, experimental, substantive, and theoretical concerns. For example, little consideration is made about how the appearance of past landscapes might compare to modern- day, as well as the effects of the curvature of the earth, intervening vegetation, object- background clarity, seasonal cycles, temporal and cyclical dynamics, time of day, climate, weather conditions, observer knowledge of the target, and even the social and physical conditions placed on human mobility itself (if you were not permitted into an enclosed ceremonial precinct, for example, you would never know what it looked like) on visibility and human perception (Fisher 1994; Wheatley and Gillings 2000). Additional problems include experimental issues, or the results of the way researchers conduct visibility analysis; substantive issues, such as the choices made in determining parameter values and data for the analysis; and theoretical issues, or how the frame of reference and the purpose of the GIS-based visibility analysis are defined by the researcher, also affect visibility (Connolly and Lake 2006:229-232).

Thus, while viewshed models provide information on the visual structure of a landscape, giving some basis for how we interpret visual experiences, they are not synonymous with

308 perceptions or experiences of the landscape or monuments in the past (Rennell 2012; Wheatley

1996). With regards to monuments, we must take care not to automatically assume a straightforward link between a location and the views it affords towards or from a particular monument. Many line-of-sight models do not take into account the important factors that would have affected visibility mentioned previously, such as atmospheric conditions, the angle of the sun, intervening vegetation, etc. We should not assume that our calculated lines-of-site are isomorphic with actual lines-of-sight in prehistory (Wheatley and Gillings 2000: 6). As well, we must exercise caution when interpreting viewshed analyses because it is easy to falsely assume that visibility was always necessarily important in prehistoric societies (Wansleeben and Verhart

1997: 60-61); sometimes monuments were purposefully hidden from sight.

Thus, when performing viewshed analyses it is important that we carefully consider how people in the past engaged with places as they moved in and around these locations and took part in social activities and practices (Rennell 2008). The roles that landscape topography, the direction of the view, and the proximity of culturally important features may have played in visibility and perception needs to be accounted for as much as possible (Lagerås 2002). As with least-cost models, we must treat the viewsheds results as entry points into understanding visibility in the past. One solution is to use Higuchi or fuzzy viewsheds when possible, especially if the environmental data is available. Viewshed analyses can be further refined by using distance-based zones of visibility established through subject-centered field study (Rennell

2012:517). Indeed, bringing in field observations to our interpretations can help to flesh out viewshed analyses and correct for the mechanistic assumptions and environmental determinism that can come from GIS results.

309

5.2.5 Discussion

In balancing between the econometric and interpretive approaches present in current GIS analysis, McEwan (2012) and Rennell (2012) advocate a mixed approach combining qualitative fieldwork (i.e. subject-centered field survey techniques) and quantitative (i.e. GIS-based spatial analysis) analysis in such a way that they complement each other, accompanied by archaeological predictive modeling. Rennell (2012:514) calls this “reflexive methodology”, and is the approach that frames the analyses used here. The application of such approaches has been rare (Hamilton et al 2006; Jerpasen 2009; Sims 2009; Sturt 2006), with the best examples coming from visualization studies (Gillings 2009; Lake and Woodman 2003; Llobera 1996,

2000, 2011; Llobera et al. 2004). Much effort has been made towards creating models that are based on measurable social, cultural, and emotional variables (Sebastian and Judge 1988;

Kamermans 2007; Verhagen et al 2007; Verhagen and Whitley 2012), rather than just simply presenting model outputs as significant “statements about site distributions and estimates of predictive power” (McEwan 2012:528; see also Gibson 2005; Graves 2009, 2011; Hudek et al

2002; Kvamme 1988, 1990; Verhagen 2007; Wescott and Brandon 2000).

Qualitative analysis can help to improve the models on theoretical rather than just statistical grounds (McEwan 2012:542; Rennell 2012:510). For example, viewsheds can be generated in GIS and then checked through formalized qualitative survey and then fed back into the modelling process (McEwan 2012:543). Rennell (2012) combines GIS-generated continuous and cumulative viewshed models with subject-centered field survey trips carried out between targeted sites and points in the study region. The research goal was twofold: first, to explore how a landscape’s visual structure may have changed as a person moved about; and second, investigate how the experience of these places affected the monumentality of Iron Age sites. The

310 results interestingly enough show that the unique experiential characteristics of the islet sites under study were not reflected in the GIS-generated viewshed maps (Rennell 2012:516), underscoring the importance of actually visiting the locations under analysis.

With respect to GIS analysis, several authors have demonstrated the fruitfulness of combining the different techniques outlined in this chapter. Robinson (2010) examines where rock art is positioned with respect to the immediate environs as well as the larger landscape in order to understand the role of this art in the social relationships among groups living in the San

Emigdio Hills of South-Central California. Using a series of least-cost path and viewshed analyses, he determined that rock-art sites were positioned at locations where families would gather socially to process acorns seasonally. Kantner and Hobgood (2003) also employ both cost-surface and viewshed analyses to evaluate the cultural landscape of the Chaco region in the northern Southwest of the United States, comparing the Chaco roads to a simulated network of

GIS-generated cost paths that connect Puebloan villages together. The results suggest that the roads were not part of a regional network linking disparate villages into a unified economic system, but rather that they served local functions related to small-scale socio-ideological dynamics. Similar studies include Mack’s (2004) use of both viewshed analysis and cost-path modeling to reconstruct the dramatic landscape experiences of pilgrims visiting 16th century

Vijayanagara, India; and Byerly and colleagues’ (2005) application of the two techniques to assess the function of a purported bison jump in Texas through understanding its place on the

Paleoindian landscape. Several other studies have researched cognition and landscape using viewsheds and cost surfaces created in GIS to reconstruct perception in the past (Gaffney et al.

1995, 1996; Llobera 1996, Lock and Harris 1996; ; Mack 2004:68; Madry and Rakos 1996;

311

Maschner 1996; Ruggles et al. 1993; Ruggles and Medyckyj-Scott 1996; Stead 1995; Wheatley

1993, 1995, 1996).

Cluster analysis has also been fruitful for understanding sociopolitical dynamics at human-scaled experiences and perceptions of landscape. The results of Borgstede and Mathieu’s

(2007) cluster analysis reveal a 3-tier settlement hierarchy in the Husta-Acatec region in the

Guatemalan Highlands. Here, ballcourts are found only at 1st-tier sites that also feature other examples of monumental architecture such as pyramids and range structures, and extensive residential zones located around the monumental core (Borgstede and Mathieu 2007:194).

Problematically, random distributions of attributes are often treated as the statistical null hypothesis despite the fact that these types of distributions are also conditioned by biological, social, and environmental factors (Maschner and Stein 1995; Woodman 2000), even though they may be less obvious. It is therefore important to remember that distinct types of patterning can exist at various spatiotemporal scales, thus the size of the study area itself can effect on the detection and characteristics of patterning when performing cluster analysis, as it may be hard to detect subtle patterns at larger scales (Connolly and Lake 2006:163).

I propose that we can investigate the spatial properties of both ballcourts as architectural spaces and the ballgame as a sociopolitical ritual by combining in-field observations and GIS analytical techniques within a cultural landscape studies perspective. Geospatial analysis can also be used to create logistic regression models predicting where ballcourts should be located given certain factors such as population, site complexity (i.e. number and monumentality of public structures; degree of diversity in residential zones, etc.), and geographical position, among others (Vaughn and Crawford 2009). Comparing the spatial models against the archaeological data can help to identify what other factors may have influenced ballcourt placement and

312 distribution. Combining spatial techniques with statistical analyses can reveal other patterns as well. For example, some ballcourt features are assumed to have influenced how the game was played at each site. Thus, in regions where the lateral mound aprons, the height of the court structures, and the central alley lengths and widths are standardized, there may have been a high degree of interaction between communities via the ballgame (Baron 2006), as these features would have affected the ways in which the ball would have bounced as it struck these surfaces, and how much room players would have had to move in. A case study from the Yalbac Hills, that form a natural boundary between the Three Rivers Region to the north and the Belize River

Valley to the south, shows how statistical analysis of the similarities and differences in court measurements can reveal Yalbac’s political orientation over time.

5.3 MAPPING METHODS AND ANALYSIS

Aside from a few examples, the majority of ballcourts are visible and accessible on the ground surface, making systematic mapping a viable option for identifying the range and social significance of variation in architectural dimensions and patterns (Aldenderfer and Stanish

1993:5). Mapping the ballcourts documented by PANT in the Nejapa subregion was the primary focus of the present dissertation. In the following sections, I describe the field collection methods and data analysis procedures used to conduct the research.

5.3.1 Field Survey and Mapping

Field research took place between May and August of 2015 in the municipal towns of

Santa Ana Tavela and Nejapa de Madero, the agencías of Las Animas and San Carlos La Baeza, and finally the territory of Ejido de los Canseco. GPS points and maps from the PANT survey data were used to relocate the ballcourts at each site where they had been documented. After

313 locating each court, our first task was to remove the trees and overgrowth in order to facilitate mapping. Given that fieldwork took place (not by choice) in the rainy season, the vegetation was particularly thick and verdant. Depending on the size of the court and the conditions of the vegetation, cleaning would sometimes take up to several days. As is typical in rural Oaxaca, field crews used machetes and the occasionally axe to cut down the trees and clear other debris.

I often joined in, taking time in between photographing and recording notes to swing a machete for the common cause. Photographs of the courts before, during, and after cleaning were taken for documentation purposes. The only ballcourts where we did not have to clear away vegetation were those at the sites of Ejido de los Canseco and Los Limares, since they were currently being used for cultivating both cash and subsistence crops. At the other courts, we easily cleared a few tons of overgrowth, reflecting the rich agricultural productivity of the soil in the Nejapa region.

All the cleared overgrowth was allowed to dry, and then was carefully burned to keep the area neat and free of debris under the supervision of local officials.

The Sokkia SET610-313 total station would be set up once enough plant overgrowth had been removed to allow for mapping. Oftentimes, two field workers would continue cleaning other sections of the court while another field worker assisted me with the total station. At each court, we followed a specific order of operations for setting up the total station. When we first arrived at a site, I would walk around the court and identify locations suitable for setting up the total station while the field crew began cleaning (Figure 5.1). Obviously, a flat surface was always critical, but I also looked for places where it would have been possible to map as much of the court structures without having to move the total station too much. Once the first location for the total station (recorded as TS #1) and a suitable backsight point had been selected (recorded as

BS #1), I marked both points (or benchmarks) with nails and field tape. I used a Trimble GTX

314

2005 GPS unit to establish these mapping points using the Projection Coordinate System

WGS1984, Zones 15N and in the case of La Baeza, 14N. The GPS data from the Trimble included the northing (N), easting (E) and elevation (Z) coordinate information of each point.

When we were ready to map, this information would be entered into the total station, allowing the machine to “know” where it is located in geographical space relative to the earth’s surface

(Figure 5.2). A total station records critical spatial measurements such as the horizontal and vertical angles as well as the linear distances from itself to the target. These measurements are then automatically converted using a trigonometric formula into eastings, northings, and elevation values (Connolly and Lake 2006:62).

To ensure as much accuracy as possible given the amount of time we had at each court, the GPS measurements for the first total station and backsight points were taken at least twice prior to set up. If the second set of measurements were not within 0.01 cm of the first set, the measurements were taken for a third time. Depending on the results of the third set, either the first or second set of measurements were input into the total station. During mapping, we would reshoot the first total station point and the backsight to reestablish the machine in geographic space and only recorded when the measurements were within 0.01 cm of the original measurements.

315

Figure 5.1: Don Adrian uses an axe to chop down a particularly stubborn tree.

316

Figure 5.2: Setting up the Sokkia total station to map the ballcourt at Lachixoba (photo by Charles Collins).

Once that information had been entered, we would start mapping with the total station and the prism rod. The person holding the prism rod would walk back and forth across each structure of the court, taking points at regular 1-meter intervals (Figure 5.3). Sometimes this interval was shortened or lengthened depending on the topography and inclination of the structures, especially where the terrain was flat. When considering the important measurements to target for mapping the ballcourts in Nejapa, I focused on the playing surfaces, lateral mounds, terminal mounds (if present), and, if time permitted, any associated cultural and natural features that were located within the immediate vicinity of the court (stairs, altars, other structures, etc.).

I also, when possible, mapped some of the ground surface around the ballcourts to show in the maps the differences in heights between the mounds, the playing surfaces, and the surrounding

317 terrain. Data collection with the total station followed a specific order of operations for accurate and precise capture of architectural and topographical features, and usually we started with the north terminal mound.

Figure 5.3 Don Taurino standing with a prism rod in front of a controlled fire behind the west lateral mound at El Cucharital.

Each series of points pertaining to a particular structure was recorded on a sheet designed for this specific purpose (Figure 5.5), according to the structure we were mapping – A for playing alleys, B for lateral structures, C for terminal structures, and D for non-court structures or features. At times, we had to move the total station to better capture certain areas of the court, depending on its size and the nature of the surrounding terrain. In these instances, after selecting a new location we would shoot in that point using the total station and record in a field notebook the northing, easting and elevation coordinates (TS #2). Upon setting up the total station at the

318 new station point, I would enter the information for the point, while the measurements for TS #1 would be used as the new backsight point (BS #2). We did this as often as needed in order to capture as many points as possible while mapping the court. Accuracy was checked by only saving the new backsight point when the measurements were within 0.01 cm of the original measurements.

Figure 5.4: Illustration of the order of operations for mapping each ballcourt.

At each site, I also recorded details about the location of the courts within their respective settlements as well as any structures that were nearby in a field notebook. This was done in

319 order to understand the ballcourt’s spatial relationships to important natural or architectural features, and what this might mean with regards to internal and external sociopolitical organization. I also recorded observations about the court as a person in that space, writing down what I could see both looking outward from the court and looking towards the court from different vantage points around it. At many of the courts there were ceramic sherds and other artifacts scattered on the surface. Because our permit did not include collecting surface artifacts, we took photographs of the ceramics in the field using a photo scale. Although we did not collect them, the ceramics were still very informative about what activities may have taken place in or around the court (discussed further in Chapters 6 and 9).

By taking numerous mapping points of the ballcourts, I was able to precisely measure each court’s respective length, width, depths, elevations, and cardinal and spatial alignments within a sub-meter accuracy of the actual ballcourt size. These data were used in the statistical analyses in order to quantitatively assess the Nejapa courts as a group, as well as for some traditional spatial analyses such as determining audience size and composition. Spatial data on additional site features – such as plazas, temples, terraces, and other nearby structural elements as well as natural features that may carry symbolic meaning – located near and around the ballcourt were taken from the site reconnaissance and PANT survey data.

5.3.2 Data Processing and Analysis

Over the course of the field project, we recorded a total number of 6,848 points for 10 ballcourts (Table 5.1). The goal was to record as many points as possible in order to capture accurately the 3-dimesional space of the courts. Due to time limitations, we only collected 75 data points at the La Baeza ballcourt. However, the resulting map produced from these points

320 was much smoother, and in some respects, presents an improved visualization of the court once displayed in ArcGIS. The methods for mapping ballcourts in the future can therefore be further refined to comprise between too little and too much coverage.

Table 5.1: Total number of data points recorded at each ballcourt. Site Code Name Total Points YAU-NEJ-ANM-002 Mogotes del Panteón 924 YAU-NEJ-CAM-001 Ejido de los Canseco 340 YAU-NEJ-NEJ-005 El Cucharital 1121 YAU-NEJ-NEJ-007 Los Limares 670 YAU-NEJ-NEJ-012 Cerro Maluco 1481 YAU-NEJ-NEJ-013 Terezona 459 YAU-SCY-BAE-001 La Baeza 75 YAU-TAV-TAV-003 Lachixoba 1020 YAU-TAV-TAV-004 La Puerta 432 YAU-TAV-TAV-015 Llano las Casas 259

The mapping data was downloaded from the Sokkia SET610-313 using the proLink transfer software. The data files were saved as *.txt documents and then imported into an Excel spreadsheet, where I was able to clean up the data by removing erroneous points and adding in the structure codes. Each series of points for a ballcourt was saved first as a *.xlsx file (for record-keeping purposes) and then as a separate *.csv file for importing into ArcGIS. The individual *.csv files were imported in as XY tables and then projected into the correct geographic projection system (WGS1984 UTM 15N except for La Baeza, which is WGS1984

UTM 14N).

From the XY table I created a point feature class, and then from those points generated

Triangulated Irregular Networks (TIN) surfaces of each ballcourt. TIN surfaces are generated by building a tessellation of triangular polygons derived from sets of mass points (Mack 2004:68;

Marozas and Zack 1990:167). Often these points are derived from spot heights, contour vertices, or a combination of the two (Connolly and Lake 2006:107). TIN models are useful because they

321 are versatile, allowing researchers to derive attributes related to elevations, slope, aspect, shaded relief, and viewsheds. I also used the TIN raster to draw in the playing surfaces, mounds, and the overall footprint of the ballcourt as polygon shapes in individual feature classes. I then ran the

Minimum Bounding Geometry tool in ArcGIS, which calculated the dimensions and orientations of the court structures and features based on those polygon shapes. These measurements were then entered into a Microsoft Access database.

Table 5.2: Definitions of the spatial dimensions of the ballcourts. Measurement Description Ballcourt overall court length length of the entire court from one terminal end to the other overall court width The width of the court from the exterior of one lateral structure to the other

Playing Surfaces central alley length length of the central playing surface, not including the terminal alleys central alley width width of the central playing surface, not including the terminal alleys terminal alley length length of terminal playing surface, perpendicular to the central playing surface terminal alley width length of terminal playing surface, perpendicular to the central playing surface

Mounds mound length length of the lateral/terminal mound from one extreme to the other mound width width of the lateral/terminal mound from one extreme to the other mound height 1 (H1) interior height of the lateral/terminal mounds from the superior surface of the mound to the playing surface mound height 2 (H2) exterior height of the lateral/terminal mounds from the superior surface of the mound to the surrounding ground surface mound volume estimated volume or internal fill of the lateral/terminal mounds

Other orientation cardinal and compass degree orientation of the ballcourt along its central axis slope interior angle of the lateral mounds length/width ratio ratio of the length of the structure element to its width

322

Table 5.3: Description of the different structure element codes for some of the spatial dimensions. Element Code Description Terminal 1 north/east terminal alleys Terminal 2 south/west terminal alleys

Mound 1 north/east lateral mounds Mound 2 south/west lateral mounds Mound 3 north/east terminal mounds Mound 4 south/west terminal mounds

The orientation value for the overall ballcourt polygon shape was recorded as the general orientation of the court; for statistical purposes these measurements were recorded in the database in a range from 0° - 180 °. GIS gives these values in directional degrees. To convert the directional degree value to the compass designation value I used the website Degree to

Compass Designation14. The interior and exterior heights of the mounds were measured by comparing the recorded elevation height of the mound minus the elevation height of the central playing alley and the exterior surface terrain, using the following formula:

H1 = Zm – Za

H2 =Zm – Zb,

whereas Zm is maximum mound elevation, Za is the central playing surface elevation,

and Zb is the elevation of the exterior ground surface

Mound volumes were estimated using the ballcourt TIN raster surfaces and the Polygon

Volume tool in ArcGIS. This tool allows you to calculate the volume for areas of special interest, in this case the lateral and terminal mounds. Values calculated this way tend to

14 http://www.csgnetwork.com/degrees2direct.html 323 underestimate volumes, but they are generally more accurate than traditional methods using shape formulas (ESRI ArcGIS). The interior slopes of the lateral mounds were measured using the DEM rasters, then reclassified into zones in order to determine the maximum, minimum, and mean of each slope zone. A point feature class was created for use in the cluster, least-cost, and viewshed analyses. An average elevation for the court was also calculated by sampling the elevation of the ballcourt points. The individual feature classes were then collated into single feature point and polygon classes and projected into WGS1984 15 N to facilitate analysis and for data storage.

The qualitative data recorded in the field notebook and the quantitative data generated from GIS were entered into a Microsoft Office Access database I created specifically for collecting and storing information specific on the Nejapa ballcourts as well as those found in

Oaxaca in general. The Access database allowed for the quick generation of data tables that were then used in the statistical package SPSS for the statistical analyses. Analysis of the spatial properties of each ballcourt, including the lengths, widths, volumes, structure slopes, orientations, among other spatial information, provides information on the range of variation in the design and form of the ballcourts in total, which can be used to assess the degree of integration (or not) of ball playing in a study region (Barron 2006; Taladoire 1981).

5.3.3 PANT Survey, Excavation, and Analysis

The initial surveying and mapping of the Nejapa region was conducted by the

Nejapa/Tavela Archaeological Project (PANT) during field seasons in 2009, 2011, and 2013, under the direction of Dr. Stacie King. Survey teams, where permitted, walked several transects each field season that targeted specific areas of the Nejapa subregion, covering in total 192 km2.

324

These targeted areas were carefully selected prior to each field season in order to fill in gaps in coverage. A Trimble ProXH GPS unit with sub-meter accuracy (< 20cm after postprocessing) to map all visible architecture encountered while on survey and determine site boundaries (some of which was done in post-analysis). Artifact collections of surface scatters were made at each location and marked with a GPS point. Each site identified in the field was documented and registered with the Catálogo Nacional de Sitios Arqueológicos and the Sistema Único de

Monumentos y Zonas Arqueológicos e Históricos15.

Several one-by-one meters test pits were excavated in the 2009 and 2011 field seasons, with more extensive horizontal excavations and two-by-two meter test pit excavations at 11 different sites taking place in 2013. Excavation methods followed long-established protocols in

Oaxaca used by most archaeologists. We excavated according to the natural and cultural layers we encountered; or by removing the soil at 10 cm layers when no obvious changes in the soil or cultural features were observed. All artifacts found in excavation units and in the screened soil were bagged and tagged with the appropriate information, including ceramic sherds, lithic materials, carbon samples, bone material, and any other objects of interest. The artifact collections were recorded in an Excel database and assigned a bag number. The artifacts were later analyzed at the laboratory in Santa Ana Tavela. Some artifacts and materials were exported to the United States for further processing and analysis, including XRF analysis (obsidian),

INAA analysis (ceramics), and radiocarbon C14 dating, among others. The survey and mapping data from PANT contribute the supplementary information on the settlement patterns in Nejapa and the site contexts for the ballcourts. Additionally, the excavation data and analysis results provide the chronological and historical context for the ballcourt mapping data.

15 National Catalogue of Archaeological Sites and the Unique System of Archaeological and Historical Monuments and Zones. 325

5.4 CONCLUSION

Research projects that focus on ballcourts within a specific historical and cultural context at the regional scale have been limited. The evidence suggests that such regional studies are not only feasible but extremely informative, and are a good starting point for understanding the sociopolitical processes that made ballcourts and ballgames so prominent (and not) in

Mesoamerican communities. Regional studies are important because localized chronological and architectural patterns can be identified. Spatial analysis techniques available through

ArcGIS allow for a more in-depth investigation of regional interactions because of the ability to integrate together environmental, archaeological, and social data into a single database. Adding a visual-spatial analysis focusing on sight-view perspectives, the thresholds of human perception, and the different viewsheds afforded by the landscape can help us to further interrogate the role of ballcourts and the ballgame in the sociopolitical relationships among competing and cooperating ethnolinguistic communities over time. In this way, assumptions about the meaning, significance, or function of ballgames in a given region can be evaluated for their appropriateness to the social reality of those communities.

Description of the research results will be given in subsequent chapters. Chapter 6 discusses the ballcourt sites and surrounding settlements in detail, providing a qualitative analysis of the structures and their contexts. Chapter 7 and Chapter 8 present the results of the statistical and spatial analyses for a quantitative look at the courts, respectively. Finally, in

Chapter 9, I offer my interpretation of the ballcourts and how they were situated sociopolitically and historically with respect to the Nejapa subregion.

326

CHAPTER VI

THE BALLCOURTS OF NEJAPA: SITE CASE STUDIES AND DESCRIPTIONS

“La época en que fueron construidos los edificios que en él [La Baeza] se encuentran se pierde en la oscuridad de los tiempos, pero por el aspecto que presentan se cree que existieron antes de la conquista de México…”

The epoch in which the buildings that you find there [La Baeza] were constructed is lost to the obscurity of time, but by the aspect they present it is believed that they existed before the conquest of Mexico… ([Juan] Carlos—Nov. 23, 1892 [originally transcribed by Stacie M. King, translated by author])

6.1 INTRODUCTION

I have argued in previous chapters that regional studies of ballcourts can provide insight into community social relationships because architecture can reflect, constrain, and influence human perceptions of social space and power relations, and that the meanings of monuments are fluid and situational depending on their spatial and historical context. Moreover, the location of these structures in site layouts and on the sociopolitical landscape can both highlight, emphasize, ignore, or downplay particular meanings linked to ballgame events and their effects on local communities. In this chapter, I describe in detail the qualitative characteristics of the ballcourts documented and mapped in the Nejapa subregion (Figure 6.1; Tables 6.1 and 6.2). Because my aim here is to place these structures within their social and historical context as much as possible,

I include current data on the sites where the ballcourts are located and the settlements closest to them, both physically and temporally. I begin with the sites we were able to map during the

2015 field season, in the municipalities of Nejapa de Madero, Santa Ana Tavela, and San Carlos

327

Yautepec. I especially focus on La Baeza, a monumental site in the mountains on the western border of Nejapa that was previously unknown to archaeology. I also discuss the four ballcourts that I was unable to map due to safety concerns, in the municipalities of Santa Ana Tavela and

San Juan Lajarcía. We excavated two of these courts in 2013, and I include the excavation data and results here. Contextual data on the archaeological sites in the Nejapa subregion is drawn from PANT field reports submitted to INAH (King 2010; King et al. 2012; King et al. n.d. Vols I and II) as well as dissertation fieldwork by my colleagues, Elizabeth Konwest (2017) and Alex

Badillo (2017). (Table 6.2)

Figure 6.1: Map of sites and ballcourts in the Nejapa subregion.

328

Table 6.1: Sites with ballcourts in the Nejapa subregion. Site Municipality Site Location Time Period Size (m2) Type Nejapa de Madero El Cucharitalab Alluvial Early/Middle 49586.69 CCB plain Postclassic Los Limares Piedmont Classic 939.33 RIT Cerro Maluco Piedmont Late Classic 197318.17 CCRS Terezona Piedmont Postclassic 493099.55 CCRL Los Mogotes del Piedmont Late Formative/ 182379.34 CCRS Panteón Early Classic Ejido de los Canseco Alluvial Classic CCRS 3909.95 plain Santa Ana Tavela Lachixobaa Alluvial Early/Middle 170908.98 CCB plain Postclassic La Puerta Piedmont Early/Middle 21138.55 CCRS Postclassic El Sitialb Mountain Classic/Middle- 364926.43 CCRL Late Postclassic Los Picachos (Los Mountain Classic/Postclassic 224027.16 CCRL Chorros) Llano las Casas Mountain Classic 20386.74 RIT San Juan Lajarcía Cerro del Convento Mountain Middle-Late 288714.93 CCRL Postclassic San Carlos La Baeza Mountain Early Postclassic CCRS 196960.29 Yautepec aSites form part of the Greater La Amontonada community bSites with multiple ballcourts.

Table 6.2: Site code definitions. SITE CODE DEFINITION RES residential, general CCRS civic-ceremonial residential small, possibly part of a community but not confirmed; < 20 structures CCB civic-ceremonial barrio or neighborhood that is clearly part of a community

CCRL civic-ceremonial residential large, sometimes a head town or main neighborhood in a community; ≥ 20 structures RIT ritual site, general

329

6.2 CASE STUDIES – MUNICIPALITY OF NEJAPA DE MADERO

Today, there are about 8000 people living in the municipality of Nejapa de Madero

(Nexapa: from the Nahuatl nextli, atl ‘ash, water’, ‘Ashy Water’ or ‘River of Ash’ [Río de

Ceniza]), whose total territory encompasses 370 km2 of alluvial plain, piedmont and mountain topography (Figure 6.2). The municipal head (cabecera del municipio) is the eponymous Nejapa de Madero, and the municipality itself includes several agencías (agencies, or smaller towns that are not the municipal head) and ejidos (communal lands belonging to smaller, separate communities or corporate groups that may also form neighborhoods in these towns), such as El

Camarón (pop. 2170), Las Animas (pop. 674), San Martín de Porres (pop. 281), El Gramal (pop.

319), and San Jose las Flores (pop. 295), among others. So far, 55 sites have been registered in the municipality, or 33% of the total sites documented (n=168), with the majority in the territory of Nejapa de Madero. Of the total ballcourts registered in the Nejapa subregion, the majority (n

= 8) are found in this municipality (Table 6.3). Given that Nejapa de Madero occupies the majority of the alluvial land in the subregion, and thus would have been the focus of much of the

Prehispanic settlement, this distribution makes sense.

330

Figure 6.2: Map of sites, ballcourts, and modern villages in the municipality of Nejapa de Madero.

Table 6.3: Ballcourts registered in the municipality of Nejapa de Madero. Site Name Site Code Court Orientation Elevation (masl) Type I-shape N-S 1° 658 El Cucharital YAU-NEJ-NEJ-005 I-shape E-NE 60° 629 Los Limares YAU-NEJ-NEJ-007 I-shape S-SE 150° 645 Cerro Maluco YAU-NEJ-NEJ-013 I-shape E-W 84° 806 Terezona YAU-NEJ-NEJ-015 T-shape N-S 0° 741 Los Mogotes del YAU-NEJ-ANM- I-shape E-W 98° 660 Panteón 002 Ejido de los Canseco YAU-NEJ-CAM- I-shape E-SE 121° 693 001

331

6.2.1 El Cucharital (YAU-NEJ-NEJ-005)

Figure 6.3: Map of El Cucharital site and ballcourts.

The site of El Cucharital (Figure 6.3) was first surveyed in 2009. It lies about 500 m south of the principal plaza and ceremonial center of Greater La Amontonada (GLA) on the other side of a large arroyo. The site technically has two separation registrations, since the

Prehispanic occupation today straddles the boundary between the Nejapa de Madero and Santa

Ana Tavela municipalities (in Tavela it is registered as YAU-TAV-TAV-002) (King 2010).

Survey in 2011 and 2013 expanded the boundaries of the site and documented additional structures. In total, the site measures 112.3 hectares and includes 213 mapped architectural structures, including public architecture and residences belonging to different social statuses. El

332

Cucharital is the largest site in the GLA community (described in Chapter 3), and likely consisted of several different neighborhoods or zones, each with their own small complex of public architecture (similar to the local town or neighborhood center in modern Western cities).

One of these neighborhoods was clearly occupied by higher-status elites (Konwest 2017). Both the arroyo separating El Cucharital from the ceremonial precinct and the alluvial terraces that borders the northwest side of the principal hill bluff (where both Cucharital and Amontonada are located) were probably utilized for cultivation, with some perishable straw residences constructed among the milpas (a common practice among rural subsistence-based communities in Oaxaca today).

Test pit excavations were first carried out on the Tavela side of El Cucharital in 2009 and again in 2011. More extensive excavations took place in 2013: a 1 x 1m test pit at a residential group on the Tavela side; and horizontal excavations at a residence on the Tavela side of

Cucharital and a public ceremonial structure on the Nejapa side. Shovel test pits were also conducted in some of the alluvial zones separating La Amontonada, El Cucharital, and

Lachixoba. I directed the test pit excavation (Operation B or Op B.) at the Tavela residential group, which lies in the northwest zone of the site – and just on the other side of the road (that divides the site in two) from the main ballcourt. In this part of the site, there are three platforms surrounded by terrace walls that were likely elite residences, given their size and proximity to ceremonial architecture. We placed the test pit next to a line of stones on the west side of the westernmost platform (measuring 12.6 meters and 8.5 meters wide) that we had originally assumed to be a structure wall. Excavations instead showed that the line of stones were in fact a wall foundation for a perishable superstructure. To construct the platform, the residents first placed a layer of fill, likely taken from a nearby midden, in order to level the original natural

333 surface. Over this fill they added a layer of large stones on which they then placed a careful mix of smaller stones and dirt to create a leveled living surface (Konwest 2017).

The first horizontal excavation, Operation A (Op. A), focused on a group of structures about 716 m southeast of Op. B. These structures were in a field that had been recently cleared by the owner, Don Nahum Carreño Nolasco. Survey in 2009 documented the presence of multiple structures here, including a temple, sunken patio, and a tall L-shaped building (which evidence indicates was covered in red stucco in the past) that were surrounded by smaller platforms, patios, and structure foundations. The presence of these public structures indicates that this was a ceremonial precinct of a neighborhood within Cucharital, and was occupied by residents of mixed social statuses. However, any smaller residential structures that had existed in the past did not preserve, as they would have been made of more perishable materials, and thus more susceptible to destruction due to agricultural activities and time. The excavation targeted a structure lying 28 meters to the south of the temple, focusing on the eastern half. I assisted

Konwest with the excavations, which took place from April 2 to 12, 2013.

As in Tavela Op. B, we observed a similar sequence of construction events: earthen fill placed on the natural surface, followed by a layer of larger stones, and then a prepared stone floor. The evidence also showed that the land had been occupied prior to the construction of this structure, as it lies on a high density of earlier cultural material (Konwest 2017). The structure itself is rectangular in shape and was probably built in one episode, consisting of a simple stone foundation on which a wattle and daub superstructure was likely mounted. We did not observe evidence of modification or reuse during the occupation of the structure or after its abandonment.

The prepared stone floor extended outside of the structure on its eastern side, indicating that a patio had once been there. A piece of carbon was found discovered inside a broken ceramic

334 vessel excavated from a structure midden. This carbon sample dates the occupation of the structure to between A.D. 1207 and 1288 (768 ± 35 BP).

The second horizontal excavation, NEJ-005 Operation A (Op. A) took place in the higher-status neighborhood on the western side of the site. The principal ballcourt at Cucharital may be part of this same neighborhood. South of both the court and a complex of four large platforms surrounding a plaza (likely ceremonial as well), is a low square platform with a small 2 m by 1 m rectangular room in the southeast corner, and two C-shaped block structures located on either side of the west entrance to the room. A large, flat plaza and possible altar were constructed on the west side of this platform, while to the south there is a temple, additional residences, and multiple terraces. Konwest (2017) excavated on and around the southern C structure. She discovered a stucco floor that surrounds the structure on both the exterior and interior. Several layers of stucco floor indicate a much longer occupation for these buildings than the structure in TAV-002 Op. A. Like the other structures though, these stucco floors were laid on a prepared stone layer. A lens for a post mold was also found, indicating that these structures were roofed in the past. The presence of red painted stucco, which was not found in situ in the excavations, points to collapse from other superstructures.

The C structure documented by Konwest is certainly unique for GLA and in the Nejapa subregion overall. The walls are 90 centimeters thick with faced stones on both sides, and rise some 55 cm up from the stucco floors (although it is likely that they were taller in the past). The interior of the walls had been filled with soil and irregular rocks, while their exteriors were covered with multiple layers of white stucco plaster. The multiple layers of stucco on the walls and floors indicate that people actively maintained the building for quite some time. The eastern wall of the C structure corners with another large, low platform, while the retaining wall to the

335 west that separates the platform from the plaza is different from construction techniques used to build the C walls. Evidence shows that the front of the retaining wall was once covered in a thin layer of stucco, and that the bottom course of stones extended some 6 centimeters out past the other stones, creating a distinct wall base. Konwest (2017) suggests that both the large square platform and the C structures were constructed in a single episode, but that the former was built first. The architectural plan of the buildings as well as the choices in construction material indicate that this was likely a complex of public but elite-purpose buildings.

6.2.1.1 The Ballcourt

The central neighborhood of El Cucharital was most likely the residential barrio for the higher-status elites. The ballcourt lies at the very northern end of this barrio. Currently it is bordered to the north by a large plaza; however, there may have been additional structures nearby that have since been destroyed by agricultural activity (today papaya is grown in these the fields bordering the north, east, and south sides of the court). From this plaza and also by standing on any of the ballcourt mounds, you can see towards the ceremonial center at La

Amontonada. A majority of the monumental buildings, including the ballcourt, and the residential elite structures of El Cucharital sits on top of a prominent bluff, the same extension of piedmont where La Amontonada is found at the extreme northern end.

336

Figure 6.4: 3D model of the main ballcourt at El Cucharital, looking from the southwest.

Figure 6.5: Composite image showing the east lateral mound and south terminal mound, from the top of the west lateral mound.

337

Figure 6.6: View of the main ballcourt from the northeast corner looking south.

The ballcourt itself is medium-sized and I-shaped, measuring 49.7 m long by 41 m across overall (Figure 6.4 – 6.6). It is an enclosed and sunken I-shaped court and is oriented almost perfectly N-S (1° E of N). The central playing alley is 26.5 m long by 6.5 m wide, while the north end zone measures 4.4 m wide by 16 m long, and the south end zone 4.9 m wide by 16.8 m long (Table 6.4). The south end of the west lateral mound and the east corner of the north terminal mound have been affected by modern agricultural activity, including the building of a water pump and cistern on the west mound (Figure 6.7)

338

Table 6.4: Size dimensions of El Cucharital ballcourt and playing surfaces. Overall Central Playing Surfaces Terminal Alleys North South Length 49.7 26.5 16 16.8 Width 41.0 6.5 4.4 4.9 Note: all dimensions in m

Figure 6.7: The cistern on the southern end of the west lateral mound.

Table 6.5: Size dimensions of El Cucharital court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 east lateral 27.8 18.5 4.2 4.2 1099.5 Mound 2 west lateral 28.8 15.8 3.8 2.9 965.6 Mound 3 north terminal 33.4 6.6 1.0 0.34 267.9 Mound 4 south terminal 33.6 8.3 1.7 1.2 382.9 Note: all dimensions in m.

339

The north terminal mound measures 33.4 m long by 6.6 m wide, 1 m high on the interior and only 34 cm on the exterior, with an estimated volume of 267.9 m3 (Table 6.5). There is a single stair on the exterior side. The west corner of the mound is distinct, as there is an observable squared shape measuring 4.2 m long by 6.4 m wide that is clearly marked by a line of roughly faced stones (Figure 6.8). This small platform may have been utilized in the performance of rituals, or as viewing platform for important dignitaries watching the game, such as a referee. On the western edge of the north end zone, between the north mound and the west mound, there appears to be lines of stones that are suggestive of a small set of stairs leading into the ballcourt from this side (Figure 6.9). There are no similar architectural features on the east side of the north mound. This could be due to modern construction activities; however, based on the mound’s morphology I do not believe that there was a corner platform here or a set of stairs into the ballcourt as was observed on the west side. The south terminal mound is much larger than the north mound, measuring 33.6 m long and 8.25 m wide, 1.7 m on the interior and 1.2 m on the exterior, with an estimated volume of 382.9 m3. Based on the lines of stones observed on both sides of the mound, there is a 3-step staircase on the southern exterior, and a staircase of at least 3 – 4 steps on the interior side of the mound leading into the ball court (Figure 6.10). This may indicate that the south terminal mound was the main entrance into the court.

340

Figure 6.8: Squared platform on the northwest corner of the north terminal mound looking from the south (upper) and north (lower).

341

Figure 6.9: Interior corner of north and west mounds showing defined edge and possible stair.

Figure 6.10: The interior of the south terminal mound showing the possible stairs leading into the court.

342

The east and west lateral mounds are among some of the largest in Nejapa. The east lateral mound measures 4.2 m in height, 27.8 m in length by 18.5 m in width, with an estimated volume of 1099.5 m3. The west lateral mound is somewhat smaller at 3.8 m high on the interior and 2.9 m on the exterior, 28.8 m long by 15.8 m wide, with an estimated volume of 965.2 m3.

The west mound is also higher at the south end than at the north, despite the construction of the water tank, with a 50-cm difference between the two ends. The west mound also lies close to the western edge of the bluff. On this side, there is a small area about 2 m wide that separates the mound from the terraced edge of the bluff and allows north-south passage on this side of the court. The average slope of the lateral mounds is 24° (21.5 – 28.5). The superior surfaces of both mounds are very narrow at the top, measuring only 0.5 m across. Viewers of the game would thus have likely stood at the north and south sides of the court, standing both on the terminal mounds and in the North and South plazas (Figures 6.11), rather than the lateral structures as has been proposed for other ballcourts in Oaxaca.

343

Figure 6.11 View towards the North Plaza from the north terminal mound (upper); and a view of the South Plaza from the west lateral mound, looking south (lower).

344

Right behind the ballcourt on the western side, a pipeline had been dug into the hillside to run water from the river below up to the water pump that has been installed on the west lateral mound. About 15 m downslope of the terraced edge of the bluff, we observed a high density of broken and discarded sherds, including multiple fragments of serving ware, sahumadores or incense burners, miniatures, a piece of red stucco wall, and figurines (see Appendix A; Figures

A.1 – A.10). We even documented a fishing net weight-stone (see Appendix A; Figure A.9), examples of which have been found in the other excavations at Cucharital. The density and diversity of the sherds strongly suggests that this area was a midden, or at least where the trash from ballgame events was thrown. The sherds appear to be mostly gray ware bowls, but we did observe some thicker utilitarian wares, such as the coarse orange-gray pastes obtained from local sources and used primarily for heavy jars. The ceramic materials are suggestive of ritual and feasting activities, but this awaits further confirmation from excavation and analysis.

As late as 2013, it was possible to see from the top of the western mound (as well as from the nearby terraced edge), a second ballcourt court that fell within the site limits of El Cucharital

(Figure 6.12), but likely was in a separate neighborhood (Konwest 2017). Unfortunately, this court was destroyed only a day after it was recorded. Thus, the only available information we have is a quick outline recorded using the Trimble GPS unit. Based on the lineal evidence, I drew in the polygon shapes in ArcGIS in order to measure the central playing alley, the north and south mounds, and estimate the overall size of the court. The results reveal that this second ballcourt measures about 56 m long by 51 m wide overall, with a central alley that is 40 m long by 8 m wide. The north mound is about 36 m long by 23 m wide and the south mound is 40 m long by 25 m wide. The court itself is oriented about 60° E of N (E-NE), and lies 479 m to the southeast of the principal Cucharital court on the main bluff. Interestingly, these measurements

345 would make this court much larger than the latter, more prominently located court. However, given that these are estimates based on field survey data and post-hoc assumptions made in GIS, we cannot be certain how accurate the measurements are. For this reason, this ballcourt was not included in statistical analyses, but was counted in the study sample for the spatial analyses.

Figure 6.12: View towards where a neighborhood ballcourt was located prior to its destruction in 2013.

The positioning of the principal court at El Cucharital close to the western edge of the bluff afforded incredible views of the valley lands to the west and north. You can also see towards several other sites with ballcourts, such as Terezona (YAU-NEJ-NEJ-013) and the Cerro

Maluco (YAU-NEJ-NEJ-012), which has the largest ballcourt in the subregion (to be described shortly), although only the former was likely occupied at the same time as Greater La

Amontonada (what would have been actually visible I discuss further in Chapter 8). From the east and south sides of the ballcourt, one would have seen the other, mixed-status neighborhood

346 of El Cucharital that is now located on Tavela municipal land. With regards to its position within the GLA community as well as the sightlines to other ballcourt sites, the main El

Cucharital court is located in a pivotal position. It occupies an important place between the main ceremonial zone that was the focus of community-wide events, and the residential barrio of the greater elite.

347

6.2.2 Los Limares (YAU-NEJ-NEJ-007)

Figure 6.13: Map of Los Limares, which was likely part of the hilltop settlement on Cerro Tanaguixi.

Los Limares (YAU-NEJ-NEJ-007) is a large terrace located on the southwestern side of a hill that lies about 2 km to the southwest from modern Nejapa (Figure 6.13). The terrace rises a few meters above the valley floor, and today lies close to where they currently cultivate lime trees using water from a feeder river that flows alongside the site. It is likely that Los Limares is associated with the dense settlement that lies further up the hill and on the summit, known locally as Tanaguixi (YAU-NEJ-NEJ-011). Survey by PANT in 2013 documented multiple

348 rectangular foundations and stone-and-earth retention terraces, as well as a large, irregularly shaped platform on the west side of the summit (King et al. n.d. Vol. I). Lines of stones on the platform also indicate that there were once superstructures built on top. Interestingly, some of the foundations had faced stones, while others did not. The faced-stone foundations were not regulated to one zone, but were instead distributed throughout the site, leaving the significance of their distribution unclear. Further analysis may help to reveal other spatial or social patterns that are not readily apparent at this time. In the southern zone of the site, at the highest point on the flattened ridgeline of the hill, there are two additional groupings of rectangular foundations made of faced stone, separated by 220 m and each measuring 12 by 12 m. The purpose of these structures is currently unclear. Based on surface ceramics, the site dates to the Classic period.

Nearby to Tanaguixi is El Callejón de Tejabana (YAU-NEJ-NEJ-010), lying about 1.3 km to the northeast from the ballcourt and only 300 meters or so from the boundaries of

Tanaguixi. This small 2.4 ha site, also located next to a channel of water that then passes by cultivated fields, appears to have been a religious site, as the evidence consists mostly of ceramic material scatters that appear to be date to the Classic period. About 800 meters away from the eastern boundary of Tanaguixi is Loma de la Cruz, located on a large hill southeast of Nejapa de

Madero. Lower on the foothills of this same site, the PANT survey crew located three structure groups, while two additional structure mound groups were found at the northeast and northwest limits of the site. Closer to the center of the site, the crew documented a fourth structure group, that consisted of four rectangular foundation lines – rather than mounds – surrounding a patio, with terraces bordering the north and south foundations. Two additional large structures lie on a terrace at the extreme west end of the site, one a low foundation and another a tall platform.

Multiple foundations of Prehispanic houses were observed on the west and north sides of the

349 hilltop. Importantly, the site not only has a magnificent view of the alluvial bottomlands below but has a direct line of sight to the Portillo de Nejapa, a geographical feature that marks the southeastern terminus of the Nejapa subregion. At 70 ha, Loma de la Cruz is one of the larger sites in the region. Today, the hilltop is used to celebrate the fiesta of Santa Cruz on May 3rd of every year.

6.2.2.1 The Ballcourt

As noted previously, the ballcourt at Los Limares (Figure 6.14 and 6.15) is found on the southwest side of the same hill as Tanaguixi. Other than the terraces that cover the hill slope to the east of the ballcourt, there are no other visible structures nearby. No structures were observed on the surfaces of the terraces themselves, suggesting three possibilities: 1) these terraces were residential, but any stone foundations that may existed have since been buried or destroyed; 2) the houses were made entirely of perishable materials that would leave no trace; or

3) the terraces on this side of the hill were not residential, but instead were used for cultivation as well as movement from the ballcourt to the site above. Like Tanaguixi, the ceramic materials on the surface suggest a Classic period date for the site.

Figure 6.14: 3D model of the ballcourt at Los Limares, looking from the southwest.

350

Figure 6.15: The ballcourt at Los Limares (Tanaguixi) looking north (upper) and south (lower).

351

Table 6.6: Size dimensions of Los Limares ballcourt and playing surfaces. . Overall Central Playing Surfaces Terminal Alleys North South Length 46.7 24.0 13.0 14.5 Width 30.5 9.5 4.4 5.8 Note: all dimensions in m

The ballcourt itself appears simple in design, not being as elaborate as the court at El

Cucharital. Oriented 150° E-SE, it is an enclosed and sunken I-shaped court. Currently, the local landowner uses both the court and the terrace it sits on for watermelon and maize cultivation. From one exterior end to another, the court measures 46.7 m long and 30.5 m across

(Table 6.6). On the interior, the central playing alley measures 9.5 m wide and 24 m long. The south end zone is 14.5 m wide and 5.5 m long, while the north end zone is 13 m long and 5.8 m wide.

Table 6.7: Size dimensions of Los Limares court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 east lateral 30.8 12.3 1.9 1.9 470.2 Mound 2 west lateral 27.5 11.0 2.4 2.4 360.2 Mound 3 north terminal 25.5 7.6 0.5 0.6 93.4 Mound 4 south terminal 33.6 8.3 1.7 1.2 38.3 Note: all dimensions in m.

The terminal mounds are very small (Table 6.7). The north terminal mound measures

25.5. m long by 7.6 m wide, 50 cm in height on the interior and 60 cm on the exterior, with an estimated volume of 93.4 m3. The south terminal mound is somewhat smaller and shaped differently, measuring about 21.4 m long by 5.4 m wide, 70 cm in height on the interior and only

20 cm on the exterior, with an estimated volume of 38.3 m3. Interestingly, while it wasn’t quite clear during the mapping, once the mapping points had been digitized it was apparent that half of the south terminal mound is not present. There is about a 50-cm difference between the top of

352 the south mound and the playing floor. In the southwest corner there also appears to be a sudden drop that may be due in part to erosion. However, it is probable that building only half of the mound was intentional.

The east and west lateral mounds are much smaller than those at El Cucharital. The west lateral mound measures 27.5 m long, 2.4 m high, 11 m wide, with an estimated volume of 470.2 m3 (Figure 6.16). Its interior slope has probably suffered from more erosion than the east mound. There’s an interesting slump in the middle of the mound that could be due to post- abandonment activities, or that could have been a stair or access point into the court (Figure

6.17). Future excavation should be able to confirm this. The east mound is somewhat larger in its dimensions but still smaller at 30.8 m long, 12.3 m wide, and 1.9 m high with an estimated volume of 360.2 m3. The average slope of the lateral mounds on the interior is 21° (18.5-27.5°).

Due to modern agricultural activity, any stones that would have been on the surface are missing or out of place.

353

Figure 6.16: A view of the west lateral mound and the surrounding landscape from the east lateral mound at Los Limares.

354

Figure 6.17 Slump that could be a possible staircase on the interior of the west terminal mound.

The terrace surrounding the ballcourt to the west is a large, flattened space that may at some point have also served as a plaza. The west lateral mound would have effectively blocked views of the game from this side, but anyone standing here would have had a great view of the alluvial floor below and the surrounding piedmont. There is also a possible tomb located in about the center of the plaza/terrace, indicated by a buried, flat lintel stone and what appears to be two vertical stones directly underneath it on either side. A small opening under the lintel stone appears to lead to a chamber of some depth – further exploration is needed to see what this feature could be. Views of the game would have been most accessible from the north side of the ballcourt, as there is level space there and the mound is low enough to see into the court. On the south side of the court, the south mound levels out and then immediately slopes downward towards a natural drainage gully, where during the rainy season excess water flows down the

355 hillsides and drains into the river that runs below. The land then slopes back upward before leveling out again, where the terrain is covered in forest. On the north and east sides, the land slopes downward slightly before it slopes upwards to the hillside. A modern footpath lies in this small depression. The ballcourt at Los Limares also has views towards the Classic-era site of

Cerro Maluco, which lies only 1.2 km to the southeast of the court.

Apart from the plaza/terrace, there are no structures within a 100 m of the court. To the north of the court by about 127 m are three natural stones with glyphs carved on them. One of the stones is known locally as “El Zapote”, and it has three figures carved on it – an abstract figure, a simple face with possible feather plumes and ear spools, and a serpent that is climbing up the stone (Figure 6.18). On the other side of the stone there appears to be carved the head of a bird. The second stone features a stylized cross carved in a style similar to the Aztec “gold” glyph. The third stone is much more eroded, but there is a figure that could be an anthropomorphic figure or smoke coming out of a vessel (Figure 6.19). The relationship between the ballcourt and the three carved stones is not known at this time.

356

Figure 6.18: The first carved stone with glyphs, El Zapote, showing Side 1 (left) with the abstract figure and Side 2 (right) possibly showing the head of a bird (original photo by Stacie M. King).

Figure 6.19: Stone 1 (left) has a symbol very similar to the Aztec gold glyph and Stone 2 (right) has another abstract figure (from King 2010, reproduced here with permission).

357

6.2.3 Cerro Maluco (YAU-NEJ-NEJ-012)

Figure 6.20: The site of Cerro Maluco, with the largest ballcourt in the Nejapa subregion.

The site of Cerro Maluco (Figure 6.20) is located 2 km south of Nejapa de Madero and can easily be seen from the valley floor. Structure foundations were observed on the hillside terraces of the southeast side, likely the habitational zone for the common people. The hill summit features several large mound groups organized around patios. There are other large mounds as one ascends the hill towards the ballcourt, all bordering a large plaza that is located adjacent to the north lateral mound of the court. The grandest mound is a possible Prehispanic temple on very north end of the Main Plaza, measuring at best estimate 51 m by 33 m (Figure

358

6.21). A barbed-wire fence running along the center of the temple made it difficult to explore the rest of it. Without further exploration, we cannot be certain, but it does appear that the residents took advantage of the natural terrain to construct the temple. Another, much smaller mound can be found on the west side of the north plaza, measuring 10.8 m by 7 m.

Figure 6.21: The north mound on the Main Plaza at Cerro Maluco.

Finally, the East Platform on the east side of the plaza is also located only 10 m north from the east terminal mound of the ballcourt (Figure 6.22). It measures about 26 m long by 13 m wide and is oriented north-south, with views towards the hills to the east and south. An additional nine mound groups were documented in the northern part of the site. Based on their observations, the PANT survey crew determined that Cerro Maluco was a residential, civic, and ritual site occupied by people of different socials statuses. Due to time limitations, it was not possible to map all of the visible architecture present, indicating that the site is either much larger or more densely occupied than the current digital map suggests.

359

Figure 6.22: Composite of the east platform on the Main Plaza, facing east.

Interestingly, the East Platform structure on the Main Plaza features two large fire pits at its southern end (Figure 6.23). The fire pits were constructed into the mound slightly east of the structure’s center line. Fogón 1, or the west fire pit, looks slightly collapsed and does not have a clear square shape, or may have been slightly rounded to begin with. It measures about 84 cm at the north-south axis and 80 cm at the east-west axis, although this is at best an estimate. Fogón

2, or the east fire pit, is better preserved and has a much clearer squared-off shape, with faced stones on the east side. It measures 89 cm on the north-south axis and 82 cm at the east-west axis. The purpose of these fire pits is unclear, but they may have been used for feasting or ritual purposes, and perhaps even as bonfires for games or other activities taking place in the ballcourt at night.

360

Figure 6.23: Fire pits on the southern end of the east platform.

Several smaller structures can be found going west to east across the Main Plaza. These structures consist of stones foundations, typically a single course of stones only, in different arrangements. One structure is a single course of stones in the shape of a square with one side open and may possibly have been the foundation for a perishable altar structure (Figure 6.24).

Another line foundation seems to be a double room, consisting of a single course of stones in the shape of a rectangle (can be seen in bottom corner of Figure 6.22). It is also entirely possible that these stones are derrumbe or rockfall from collapsed structures; at this time, without excavation, it is difficult to determine what these lines of stones were or what their purpose may have been.

361

Figure 6.24: Square room or altar in the center of the Main Plaza, open on the west side.

Based on the ceramic materials collected during surface survey, Cerro Maluco roughly dates to the Classic period. Two sherds collected at the summit of the hill resemble Central

Valley sherds that have been securely dated to the Xoo phase, or Late Classic. This may mean that the site is a contemporary of Nejapa Viejo, the large site that dominated the Nejapa subregion during the Classic period, as well as Tanaguixi (and Los Limares) and Mogotes del

Panteón (another ballcourt site to be discussed shortly).

362

6.2.3.1 The Ballcourt

Figure 6.25: 3D models of Cerro Maluco ballcourt, looking from above (upper) and from the southwest (lower).

363

Figure 6.26: View of the ballcourt at Cerro Maluco looking east.

East from the other mound groups that form the core of the site, on the western side of the Cerro Maluco, lies the ballcourt and the Main Plaza complex. At 87.3 m in length from the exterior of the west terminal mound to the exterior of the east terminal mound, and 36.6 meters in width, the ballcourt at Cerro Maluco (Figure 6.25 – 6.26) is the largest in the Nejapa subregion and easily one of the largest in Oaxaca, given current information (Taladoire, personal communication). The court is an enclosed and sunken I-shaped court, and one of the few in the sample that is oriented east-west (84° or E-NE). This ballcourt is the most complex observed so far with regards to its architectural features and design and is rather unique in that sense. The central playing alley measures 40.4 m long and 8.6 m wide (Table 6.8). The west end zone is about 23 m long, 11.7 m wide at the north end and 24 m wide at the south end. The east end zone measures 24 m long and 6.8 m wide. In the west end zone, there is a prominent

364 outcropping of bedrock that looks like it has been modified, as part of it does appear to have been shaped with a flat edge and curved corner (Figure 6.27). Although not entirely level, it is possible that the bedrock was utilized for ritual purposes or may have even figured into game play in a manner similar to the botadora. Given the extensive length of the playing alley, this feature would likely not have interrupted or affected game play in any significant way.

Table 6.8: Size dimensions of Cerro Malcuo ballcourt and playing surfaces. Overall Central Playing Surfaces Terminal Alleys East West Length 87.3 40.4 23.9 23.1 Width 36.6 8.6 6.8 ? Note: all dimensions in m

Figure 6.27: Bedrock outcropping in the west terminal alley that appears to have a rounded corner.

365

Table 6.9: Size dimensions of Cerro Maluco court mounds.

Orientation Type Length Width Height 1 Height 2 Volume Mound 1 north lateral 30.8 12.3 1.9 1.9 1721.4 Mound 2 south lateral 27.5 11.0 2.4 2.4 1056.7 Mound 3 east terminal 33.7 12.5 2.7 2.7 666.3 Mound 4 west terminal 20.6 14.2 2.5 2.5 604.7 Note: all dimensions in m.

The west terminal mound is unusual in that only the north half is present, while the southern end is non-existent; that is, only half of the west mound was constructed, with the entire south and west sections of the end zone having been left open and accessible (Table 6.9). In this corner, the playing surface extends out and meets with the edge of the terrace, from there the terrain slopes down towards the next terraced surface. Based on the morphology, it is clear that this was intentional. To construct the west mound, they modified and built on the bedrock where it protruded out of the ground surface on this side of the hilltop. The motive for leaving this side open may have to do with the wide and encompassing views of the alluvial flood plains below and towards the surrounding mountains as well as other sites. There is also no protruding bedrock on this end of the west mound. Asymmetrical end zones do not occur with any regularity in Mesoamerica, but neither are they entirely unusual, as evidenced by the courts at

Chinkultic and Toniná, both Late Classic sites in Chiapas (Taladoire 2001). However, the

Chinkultic end zones are simply different sizes, and are not shaped unusually like the west terminal zone at Maluco. On the superior surface of the west mound there is a small, double platform, consisting of a 10 by 5.2 m one-course stone platform on top of another, slightly larger one-course stone platform – not quite the same form or height as a typical cornice or upper wall.

This feature may have served as a platform for observing the game and performing other related activities, similar to but much more complex than the squared platform on the northwest corner

366 of the El Cucharital court. From this upper feature, the west mound slopes down on the exterior to the next terrace below. From its summit, the interior of the west mound slopes down 2.5 m to the central playing alley. Overall, the west mound measures 20.6 m long and 14.2 m wide, with an estimated volume of 604.7 m3.

The north lateral mound is the largest of the four court structures, measuring 44.2 m long and about 14 m wide, 4.7 m in height from the playing surface and only 10 cm on the exterior, making the top surface of the mound nearly level with the surface of the Main Plaza. The estimated volume of the north mound is 1721.4 m3. The south lateral mound is smaller than the north at about 3.4 m in height on the interior, and 45.2 m long and an estimated 12.6 m wide, with an estimated volume of 1056.7 m3. The mound is much higher at the east end than at the west end, with about a 1.3 m difference between the two sides (3.4 m versus 2.1 m, respectively).

The average slope of the mound interiors is 26.1° (22.5 – 36.5°). The exterior (south side) of the south lateral mound descends (at an average decline of 24%) directly down for several meters to the next terrace below from where the ceremonial core is located. We were not able to map the mound’s exterior because it was too risky and time-consuming to cut down the thick vegetation growth on this side, and would have required extensive cleaning to even set up the total station.

The east terminal mound is the most complex of the four court mounds. In total, it measures 33.7 m long by 12.5 m wide, 2.7 m in height from the playing surface, and with an estimated volume of 666.3 m3. Based on the surface slope (16-19°) and the lines of stones (at least five) in the northeast corner, where the north mound connects to the east mound, there was probably a large staircase here that allowed access from the plaza down into the ballcourt to the playing surface. This needs to be confirmed through excavation. There is an extreme height difference of at least 3.2 m from the north end of the east mound, at 4.3 high, to the south end, at

367

1.1 m high. In fact, where the superior surface levels out (at about the center of the mound) and then connects to the south lateral mound, the east mound consists of 2 courses of medium-sized rectangular blocks interspersed with large boulders that have been roughly faced. Unlike many of the terminal mounds in Nejapa, the east terminal mound of Cerro Maluco’s ballcourt does not slope, but instead resembles the vertical back walls more common to Central Valley courts.

Moving north to south, the east mound descends as a series of small platforms from the Main

Plaza, terminating in an L-shaped platform bordered on its east side by a small patio. The first and superior-most platform begins as small patio on the south end of the East Platform, linking these two structures – the East Platform and the ballcourt – together. Between these platforms and the interior edge of the east terminal mound, there is a staircase allowing passage along the mound as well as access to the platforms. Two sets of small steps in the center of the mound permit access into the ballcourt from this side, where the mound levels out.

The design of the ballcourt, including the multiple, descending platforms of the east mound, as well as the size of the Main Plaza, would have allowed a substantial audience to view the game. Certainly, any games or activities taking place in the court would have been readily visible from the east, west, and north mounds. The west end zone, with its broad, open, and irregular space in the southwest corner, could have also been used for watching the game even during play. Aside from these areas, it may also have been possible to see the game while standing on the southern end of the East Platform where the fire pits are located, as one can see into the eastern end zone of the court from this angle. Activities taking place in the interior of the court were likely not visible from the North Temple, but people would have been able to observe anything happening on the court mounds. Certainly, the sounds of game activity, or

368 even ceremonies taking place in the court, would have at least been heard from anywhere on the plaza or the North Temple.

6.2.4 Terezona (YAU-NEJ-NEJ-013)

Figure 6.28: The site of Terezona with a T-shaped ballcourt.

The Postclassic site of Terezona (YAU-NEJ-NEJ-013) was first surveyed and registered in 2013 (Figure 6.28). It lies about a kilometer to the east of Nejapa de Madero and, based on surface artifact scatters, measures 49.3 ha in total. From both the ballcourt and the upper plaza, one has commanding views of the valley lands below and to the north, including several sites with ballcourts. There is a remarkable density of structures such as house foundations, mounds

369 of different sizes, residential and agricultural terraces, and architectural complexes. In the southeast zone of the site, the survey team documented simple stone foundations that likely represent lower-status households on the terraces. A large architectural group of four mounds surrounding a plaza is found on a modified plain in the northwest sector of the site. To the east of this first group there is another plaza that is demarcated by several descending terraces.

Cultivation terraces cover the eastern flank of the hill. One of these terraces is on a small knoll, whose summit has been defined by a line of 30 by 30 cm-sized stones. The exact purpose of this line of stones is not clear, but I would propose that the summit may have been used for observing the terrace fields, or even performing localized agricultural ceremonies. To the west of the knoll is a mound measuring 9 m long by 5 m wide. A grand patio and a possible set of stairs lie in front of this mound, and all three are oriented with a view towards the valley floor. Another series of architectural groupings are found in the northeast sector of the site, consisting of sets of four structures surrounding patios. Altogether, they form a grand complex with 11 structures and four patios in total. This complex is located on the highest part of the hill from which you can see the valley floor as well as the large mounds to the southwest.

Thus, there are several locations in Terezona – including where the ballcourt is located – with impressive views of the valley floor as well as other archaeological sites. Given the site’s orientation towards and proximity to other sites, Terezona was likely part of a network of several communities. On the eastern border of Terezona only 73 m away is the 65.1 ha site of El

Maneadero (YAU-NEJ-NEJ-019). Most of the architecture here is concentrated on a ridgeline, which rises from the lower eastern end west to the summit of the hill. Construction consists primarily of a series of 1 m tall terraces connected by a system of paths and stairs that permitted both passage and access. On the surface of these terraces, the PANT survey crew recorded 2 m

370 wide by 3 m long stone foundations. The west side of the summit features a complex of foundations separated by a 5 m wide track that travels upwards towards the summit, a possible

Prehispanic road. Moving towards the southeast there are two mounds with a patio between them, one of which measures 2 by 4 m. On the north side of the site there is another patio that possibly had a mound on one side that is no longer there, while on the principal peak an architectural complex was constructed on a flattened plain. This peak is surrounded on all sides by terraces, again many with stone foundations on them. It is possible that Terezona and El

Maneadero may have formed a single community, given their close proximity.

Southwest of Terezona is Terreno de Julio Leon (YAU-NEJ-NEJ-020), located only 150 m away. A small 6.3 ha site, the survey crew registered a stone foundation measuring 9 m long by 3.6 m wide, and a terrace measuring 5.4 m in length and 30 cm in height. Both structures suggest a small household that may have been affiliated in some way with Terezona. To the north about 466 m away is El Campo de Aviación (YAU-NEJ-NEJ-037), a 21-ha site with a stone foundation and a mound in the eastern part of the site. It is possible that the El Campo de

Aviación may pertain to the Terezona community, but further clarification is still needed.

Today, the site is a communal field for cultivating watermelon, but it may have been a ceramic production center or a pottery workshop in the past, as many of the surface ceramics show evidence of fabrication mistakes. If it was a ceramic production center, then it could explain where local communities both on the valley floor and in the mountain zones were obtaining their vessels, given the strong similarities in type and paste ware that have been observed in many of the ceramic assemblages at Nejapa sites.

371

6.2.4.1 The Ballcourt

Figure 6.29: 3D model of the ballcourt at Terezona looking from the southwest corner.

Figure 6.30: Composite of the ballcourt looking south from the north terminal alley.

The ballcourt at Terezona lies just south of the site center, respectively (Figures 6.29 –

6.30). This court is especially unique for Nejapa and Oaxaca overall. There is no south terminal mound and defined south end zone; instead, the playing alley on the southern end remains flat and slopes downward a meter to a lower terrace, rather than sloping upwards as is the general 372 pattern for Oaxaca. We at first assumed that the west side of the southern end of the court had been left open as we had observed at Cerro Maluco, with the perhaps a small plaza or patio at the south end of the west lateral mound. With the east side covered in heavy vegetation, it was difficult to see if the eastern half of the south mound was present. After clearing the plant overgrowth on the east side, we soon realized that there was neither an enclosure nor any structure connected to this end of the east lateral mound. In fact, further cleaning revealed a set of meter-wide steps that lead from the edge of the south end of the playing alley to the lower terrace below (Figure 6.31). I could only see two steps, but it is possible that more existed and are currently buried, as the stairs only begin halfway down the slope of the terrace. The shape of the terrace edge gives the terminus of the playing alley in the south a slightly curved appearance.

Figure 6.31: Composite of the sounth terminal end of the Terezona ballcourt.

Given the missing south terminal mound and the lack of a clearly defined south end zone,

I tentatively identify this structure as a T-shaped ballcourt. According to Taladoire’s (2015) most recent schema of ballcourt types, T-shaped courts (or at least those observed so far) typically consist of two parallel structures with one terminal (axial) structure (Type I variant 1) or one U-shaped axial structure (Type 1 variant 2). These court types are open; that is, the

373 terminal mound does not connect to the lateral mounds. Courts of this form have primarily been documented in the Maya Lowlands (Taladoire 2015), but Whalen and Minnis have also identified T-shaped courts at sites around the major center of Paquimé (all TI.v1). A T-shaped court was also found at Tres Picos in the Isthmus; JP2 is a T1.v1, or open court with a terminal mound. A second court, JP4, may also be T-shaped, with a temple mound defining the west end zone, based on site map drawings (Winter and Santiago 2014). The Terezona court is somewhat different from these other examples because the north terminal mound connects with the east lateral mound, creating an enclosed north end zone. Further excavation is needed to confirm this identification, and to test if the north end zone is actually enclosed or if it is a result of post- occupational infill or hill slump (there is a terrace above the court). Even if the court is not T- shaped, it is still distinguished by its unusual design that is atypical of both Nejapa and Oaxaca.

Table 6.10: Size dimensions for Terezona ballcourt and playing surfaces. Overall Central Playing Surfaces Terminal Alleys North South Length 48.8 21.5 22.2 26.8 Width 32.6 4.4 6.7 9.0

The structure itself is oriented 0° N. The playing alley, measuring from the north terminal mound to about the termini of the east and west lateral mounds, is about 21.5 meters in length and 4.4 m in width; overall, from one extreme to another the court measures about 48.8 m long and 32.6 m wide (Table 6.10). Both the north and south end zones are somewhat difficult to calculate, given the effects of the road construction as well as the morphology of the court in the south. Roughly, the north end zone is 22.2 m long by 6.7 m wide, while the south end zone is 26.8 m long by 9 m at its widest. The dimensions of the north terminal mound are also a little difficult to determine, but I estimate it to be half a meter in height on the exterior and 2 m in

374 height on the interior, 7.7 m wide and only 15.8 m long, with an estimated volume of 210.3 m3, due to the eastern half missing as a result of the road that was constructed to give access to the water tank. I asked the locals if they could recall whether the mound was complete in the past and if it had connected to the west lateral mound. They were not entirely sure, but one confirmed that the north mound was much larger before the water tower and road construction, though he could not recall if it connected to the west mound. Given two other similar examples, it is also possible that the west end of the north mound did not exist either. This would make the

Terezona court even more unusual in form and design, and if a T-shaped court, then it appears to be a hyperlocal example.

Table 6.11: Size dimensions of Terezona court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 east lateral 23.4 8.8 2.7 2.7 440.6 Mound 2 west lateral 23.5 16.7 2.7 3.0 669.9 Mound 3 north terminal 15.8 7.7 2 0.5 210.3 Note: all dimensions in m.

Both the east and west lateral mounds of the Terezona court fall in the middle range of mound sizes in Nejapa (Table 6.11). However, the west mound is much higher and larger than the east mound, reaching about 2.7 m in height on the interior and 3 m on the exterior, 23.4 m long and 16.7 m wide, with an estimated volume of 669.9 m3. Since the northern edge of the west mound was sheared off by the road construction, it was probably larger in the past

(although perhaps not by too much more). Although the destruction is unfortunate, it does allow us to observe how the mounds were constructed using a fill of mixed earth and small stones, with then much larger stones were placed on the surface. The east lateral mound measures 2.7 m high by 23.4 m long and 8.8 m wide, with an estimated volume of 440.6 m3. There is a slight

375 difference in height from the north to the south end of the mound of about 60 cm. This could be due to hill slump, or it could be a feature of the structure. Of the three court structures, it is the only one that is completely intact. It connects with the north mound, creating a defined and enclosed north end zone. The summit of the east mound is level with the ground surface of an upper-level terrace on which there is a sloped surface that leads to an upper plaza measuring about 20 m by 20 m. This sloped surface was probably a wide set of stairs that gave access to and from the plaza to the ballcourt. On the east side of this upper plaza, there is a long mound or platform. Northeast of the plaza, but on a slightly lower terrace, is an architectural group of four small mounds surrounding a sunken patio. The average interior slope of the two lateral mounds is 22° (19.5 – 28.5°).

376

6.2.5 Los Mogotes de Panteón (YAU-NEJ-ANM-002)

Figure 6.32: Map of the Mogotes del Panteón site.

Los Mogotes de Panteón (MDP) (Figure 6.32) lies within the territory of the agencía of

Las Animas on an extension of piedmont with a view of the Rio Grande as it enters the valley from the northwest. Las Animas is the first community you encounter taking the highway into the Nejapa region once you’ve passed the district boundary line between Tlacolula and Yautepec at the Boquero de Yautepec (roughly translated, mouth of Yautepec). MDP is an interesting site that should definitely be investigated further. It is well-positioned close to the entrance to the valley and is also one of the earlier occupations in the Nejapa region, likely dating to the Late

Formative to Early Classic period, based on the ceramic evidence. It was first surveyed by King

377 in 2007 and later registered in 2009. Unfortunately, PANT was initially denied permission by the agencía committee to continue working in Las Animas, which meant that we could neither survey in the area nor excavate at MDP in 2011 and 2013. However, by 2015 there was fortunately a change in the members of the agencía committee and they were much more open to archaeologists working in the area. While having lunch with the local land owner where the court is located and a nearby neighbor, they told me about Prehispanic restos or remains they’ve encountered in the Las Animas territory, including rock shelters with red-painted hands. Clearly, there is additional evidence of occupation in this area of the Nejapa subregion. With the recent change in committee members, and with more friendly relations between PANT and the agencía finally established, we look forward to exploring more of Las Animas territory in the future.

The architecture at MDP spreads out over three large terraces flanking the sides of the foothill. MDP is dominated by an impressive ceremonial center that includes a ballcourt and several additional structures and plazas. Attached to the north lateral mound on the exterior side, at about its center, is a small square platform that measures about 3.9 m by 5.2 m, and about 40 cm in height (Figure 6.33). Located immediately north of this same mound, and at the north end of the ceremonial center, is the Main Plaza. It measures about 53 m by 15 m and is bordered on all three sides by mounds. On the north side of the plaza is Structure A, a possibly split-level platform with a lower, westerly extension that might have had stairs in the past allowing access to the plaza (Figure 6.34). Meanwhile, Structure B on the plaza’s east side is the largest of the three plaza structures, and may have been a temple (Figure 6.35). A significantly smaller mound borders the west side of the Main Plaza (not pictured). Structures A and B were originally described in 2007 as an L-shaped platform, but when I walked around the area it was clear that they were in fact separate structures. Excavation should be able to determine whether this

378 separation was intentional or is from a more recent event. While there was not enough time to map both structures in their entirety, the measurements we did take reveal that Structure A measures 4 m in height and Structure B is about 4.2 m high.

Figure 6.33: Small platform on the north side of the north lateral mound in the Main Plaza.

379

Figure 6.34: View of Platform A from the south. Platform A forms the northern border of the Main Plaza.

Figure 6.35: Platform B on the east side of the Main Plaza.

380

A second plaza lies directly east of the court itself (not pictured). An additional mound, possibly another temple, lies on the eastern edge of this same plaza. This mound is located only

2 meters south of Structure B. The exterior side of the mound directly overlooks the Rio Grande as it flows past below. Here I found a looter’s pit, around which there were several scatters of broken sherds, and even several pieces that were nearly whole. From this pit, we were able to document a substantial number of different diagnostic ceramic sherds, among them a face-neck jar and other finely decorated wares (Appendix A; Figures A.11 – A.16). These sherds have been identified as Classic-period ceramics, and some may even be earlier as they are similar to

Monte Albán I and Monte Albán II Late Formative types (500 – 100 BCE and 100 BCE – CE

200, respectively) from the Central Valleys. Additionally, there were also multiple fragments of ritual items such as figurines, sahumadores/incensarios, and expedient miniatures (Appendix A;

Figures A.18 and A.19) found all over the plazas and the court.

Residential occupation at MDP is found on a second terrace located below the main terrace with the ceremonial center. This second terrace is also flat and substantially-sized, with a great quantity and diversity of domestic artifacts on the surface. Unfortunately, any small house foundations that would have been present have eroded from the surface due to modern agricultural activities. The third terrace at MDP was dedicated to agriculture in the past, and it continues to be used for growing magueys today (we were also once roadblocked by a donkey on our way to the site). The highly productive lands surrounding the hill on which MDP sits form part of the alluvial plain located north of the Rio Grande.

Given the relatively early date for the site, its exact relationship to other settlements in the area is still unknown. Closest to MDP is Llano las Casillas (YAU-NEJ-NEJ-002), a 57-ha site that lies about 1 km to the south. Although the site has suffered severe damage from agricultural

381 activities, there still remains some architecture such as stone walls and small, low platforms as well as artifact scatters throughout the site. Based on the observations of the survey crew, it appears that the architecture is grouped into distinct neighborhoods that are themselves dispersed across the terraces. However, Llano las Casillas has been dated to the Middle Formative period

(but was reoccupied later in the Postclassic), leaving doubt as to it being a contemporary of

MDP. Located further away at 3 km to the west is Nejapa Viejo, the large, complex site that was first occupied in the Late Formative and later dominated the Classic period in Nejapa. MDP was probably a contemporary of Nejapa Viejo, although again its exact relationship to the larger site is unknown at the moment. Further excavation should hopefully clarify this. Nevertheless, it is interesting that we find the ballcourt at MDP, and not at the more prominent Nejapa Viejo. At the same time, we cannot discount that one may have existed at the latter site in the past, but was been destroyed by agricultural activity, erosion, or urban construction over subsequent centuries.

Indeed, much of the site has washed away in modern floods and continues to do so today.

382

6.2.5.1. The Ballcourt

Figure 6.36: 3D model of the ballcourt at MDP from the south. In the background are Platforms A and B; the small platform next to the north lateral mound is also visible.

Figure 6.37: Composite of the MDP ballcourt from the west looking east.

The ceremonial center of MDP was constructed in the western portion of the site but is technically located on the southern end of the hill (the Rio Grande flows west of this same hill).

The ballcourt (Figures 6.36 and 37) lies at the southern extreme of the ceremonial center and has a very imposing presence. Although not the largest in Nejapa by any means, the ballcourt certainly feels massive, with some of the biggest lateral and terminal mounds apart from Cerro

383

Maluco and El Cucharital. It is an enclosed, sunken I-shaped ballcourt that is oriented more or less east-west at 98° (Table 6.12). The central playing alley measures 24 m long by 5.9 m wide.

Overall, the court measures 52.7 m long by 36 m wide. The west terminal end zone measures 16 m long by 5 m wide, while the east terminal end zone is 15.8 m long by 5.3 m wide.

Table 6.12: Size dimensions of Mogotes del Panteón ballcourt and playing surfaces. Overall Central Playing Surfaces Terminal Alleys North South Length 52.7 23.9 15.8 16.0 Width 36.0 5.9 5.3 5.1 Note: all dimensions in m

Table 6.13: Size dimensions for the Mogotes del Panteón court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 North lateral 24.6 14.3 3.5 1.9 808.3 Mound 2 South lateral 26.5 16.0 3.2 4.6 724.4 Mound 3 East terminal 30.6 8.6 0.0 3.0 489.3 Mound 4 West terminal 33.6 11.2 1.7 1.2 436.9 Note: all dimensions in m.

Unlike many of the other courts in the sample, the interiors and exterior heights of all of the court mound structures vary incredibly, likely due to the topography of the hill (Table 6.13).

The north lateral mound measures 3.5 m as it rises from the playing alley surface, but only 1.9 m on the exterior. It is also 24.6 m long, 14.3 m wide, and has an estimated volume of 808.3 m3.

The summit of the north mound is rather narrow, only about 1.6 m across, indicating that it probably was not really used as a standing or sitting area for viewing the game. As mentioned previously, there is a small square platform attached directly to the exterior the north mound.

The purpose of this square platform is not clear at the moment. The south lateral mound is 3.2 m high on the interior and measures 4.6 m above the ground surface of the middle terrace on the exterior (Figure 6.38). The mound measures 25.4 m long and 15.3 m wide, with an estimated volume of 724.4 m3. There may have been a plaza on this side of the south mound as well given

384 the extensive flatness of the area. The top of the south mound is also very narrow, only 1.2 m at its flattest. Given the narrowness of the summit and the steeper slope down to the next surface

(about 28° – 32°), the south mound would not have been a great place for viewing the game

(similar to the later Cerro Maluco). The average slope of the interior surfaces of the lateral mounds is 21° (17.5 – 25.5°).

Figure 6.38: The south lateral mound on the south side with Don Jose Enrico Martinez for scale.

385

The west terminal mound measures 33.6 m long and about 11.2 m wide, with an estimated volume of 436.9 m3. It slopes 1.4 m upwards from the playing surface, and then downward 1.5 m to the middle terrace below. There is a 70-cm difference in height from the north end to the south. Although some of the stones are out of place, we did observe a defined line of stones marking the western and eastern edges of the mound. In fact, the entire summit of the west mound is lined in heavy boulders. In some photos, it is possible to see that this continues down both the interior and exterior sides of the mound (Figure 6.39). On the western exterior of the mound at the southern end, we could see that there are at least 2 courses of stones before the mound slopes out and downward, creating a raised, square corner (Figure 6.40). This is an interesting feature and recalls the squared corner platform of the north terminal mound at El

Cucharital court (although if it is the same or similar, the MDP feature is much older). There is no such similar feature in the northwest corner. We also discovered a small hole on the west mound, likely belonging to a burrowing animal. The hole revealed that the interior of the mound was constructed using a mixed fill of earth and small to medium-sized stones, much like the west mound at Terezona.

386

Figure 6.39: Stones lining the summit and sides of the west terminal mound.

Figure 6.40: Raised feature on the southwest corner of the west terminal mound.

387

The east terminal mound rises 3 m above the playing surface, while its superior surface is level with the ground surface of the East Plaza. In total, the mound measures 30.6 m long by 8.6 m wide, with an estimated volume of 489.3 m3. Here, there is a 90-cm difference between the north and south ends, indicating that the natural slope of the terrain inclines more at this end of the court than at the west end. Unlike the west mound, the east mound is defined only by a line of stones that mark the very interior superior edge and the interior surface (Figure 6.41). In other words, the superior surface of the east mound leads directly into the East Plaza. These same lines of stones outline the “arms” of the terminal mound that meet with each lateral mound to create the enclosed end zone.

Figure 6.41: The east terminal mound at MDP. The superior surface of the mound is level with the east plaza, a common feature of Nejapa ballcourts.

Scatters of ceramic artifacts were found all over the ballcourt, including fragments of figurines, miniatures, sahumadores, and decorated sherds, many with incised designs (see

Appendix A, Figure A.17). Most of the sherds are fine gray wares, but there were also examples

388 of utilitarian vessels, such as handles for large jars. We observed several broken metates that were probably used in feasting events and then perhaps later utilized as building stones, and even a broken mano. Perhaps some of the most remarkable finds were fragments of stucco in the court, including one piece that clearly shows two layers (Figure 6.42) – evidence the court once had stucco floors that had also been repaired at some point in the past. This indicates that the ballcourt was in use for some time. So far, this is the only court where we have found evidence of stucco. In the two courts where we excavated, El Sitial and Cerro el Convento, we did not find stucco fragments nor evidence of stucco floors in the units.

Figure 6.42: Fragment of stucco flooring found in MDP ballcourt showing that repairs had been made at least once.

389

6.2.6 Ejido de los Canseco (YAU-NEJ-CAM-001)

Figure 6.43: Map of the Ejido de los Canseco site.

Ejido de los Canseco (EC) (Figure 6.43) is a site located on the communal lands administered by an ejido that is based in the agencía of El Camarón. These agricultural fields, used today for growing maize and agaves, lie along the road leading from Soledad Salinas (in the

Quiatoni municipality, Tlacolula District) to San José las Flores (the Nejapa municipality, the

Yautepec District). There is only one large structure anywhere close by, a small mound located about 500 meters away from the court. More structures probably existed at some point in the past but they have since been leveled or destroyed. Currently, the site itself encompasses 3.9 ha,

390 based on the ceramic scatters on the ground surface. When survey crew members worked here in 2009, they observed small structure foundations on the eastern side of the ballcourt, possibly households, but these have since been heavily disturbed. They also documented a fire pit or some other, similar element that had ash and ceramic sherds within it. This element may have once been utilized in rituals or for cooking food for feasting events.

The site closest to EC is located about a kilometer away to the southeast. Ejido de

Animas (YAU-NEJ-ANM-001) (which belongs to the agencía of Las Animas) measures 26.6 ha site and features a series of domestic units of small mounds separated by patios. Much further away at 1.8 km towards the southeast, and only about 483 meters from Ejido de Animas, is

Nejapa Viejo. King (2010) suggests that both sites may be related in some way with Ejido de los

Canseco, and even includes MDP in this community network. This would mean that there were at least four ballcourts in use in the Nejapa municipality during the Classic period – Ejido de los

Canseco, Los Limares, Mogotes de Panteón, and Cerro Maluco. MDP is only 3.2 km southwest from EC.

391

6.2.6.1 The Ballcourt

Figure 6.44: 3D model of ballcourt at EC.

Figure 6.45: View of ballcourt from the east and looking west.

The ballcourt at Ejido de los Canseco (Figures 6.44 and 45) is similar in appearance to the ballcourts at Los Limares and La Puerta (and even in some respects to Terezona). The

392 ballcourt is oriented 159° S-SE), and is an enclosed, sunken I-shaped court with U-shaped terminal structures (Table 6.14). From one extreme to another, the court measures 52.8 m by 37 m overall, while the playing alley itself measures 23.5 m long and 5.8 m wide. The east terminal zone measures about 17.8 m by 5.8 m while the west terminal zone measures 11.8 by 5.9 m; the difference in length may be due to erosional in-fill.

Table 6.14: Size dimensions for Ejido de los Canseco ballcourt and playing surfaces. Overall Central Playing Surfaces Terminal Alleys East West Length 52.8 23.5 17.8 11.8 Width 37.0 5.8 5.8 5.9 Note: all dimensions in m

Table 6.15: Size dimensions for Ejido de los Canseco court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 north lateral 29.6 14.8 2.8 1.8 677.1 Mound 2 south lateral 34.4 15.9 2.4 1.9 735.5 Mound 3 east terminal 25.5 7.6 0.5 0.6 143.3 Mound 4 west terminal 33.6 8.3 1.7 1.2 239.9 Note: all dimensions in m.

The south lateral mound measures 2.4 m in height on the interior and 1.9 on the exterior,

34.38 m long and 15.9 m wide, with an estimated volume of 735.5 m3 (Table 6.15). On the southern exterior, the mound slopes down to the ground surface, where they are currently cultivating maguey. The north lateral mound measures 29.6 m long by 14.8 m wide, 2.8 m in height on the interior and 1.8 on the interior, with an estimated volume of 677.1 m3. There is a noticeable slope of the superior surface of the north mound, the west end being 50 cm higher than the east. On the northern exterior, the mound surface gradually slopes down 150 m to a tributary river below (about a 1 m difference in actual elevation from the bottom of the mound to the river). At the east end of the mound, there appears to be a strange feature that became more

393 visible once the mapping points were digitized. A line of stones was observed at here, close to the exterior side. At this same eastern end of the mound the line of stones turns south, creating a corner (Figure 6.46). The feature, which measures 6 m by 6 m, is reminiscent of the squared platform at El Cucharital, and the line of stones may have demarcated this feature as well as the exterior edge mound from the natural slope down to the river. The average interior slope of the east lateral mound is 20° (17.5 – 26°), while for the west lateral mound it is 15° (13.5 – 17.5°).

Figure 6.46: Line of stones marking the summit and eastern end of the north lateral mound.

Like the terminal mounds observed at the other ballcourt sites, the east and west terminal mounds are much smaller than the lateral mounds. The east mound measures some 1.5 m above the playing surface, but only rising 10 cm above the ground surface on the exterior. In total, the mound is 17.7 m long by 7.6 m wide, with an estimated volume of 143.3 m3. The southern half

394 of the east mound is missing, and it is possible that it was destroyed since the court is currently used for growing maize, squash, and watermelon. However, I did observe what appeared to be a stone edge and corners defining the southern terminus of the east mound, similar to the west mound at Cerro Maluco and possibly Terezona and Los Limares (although not clear in field photos). The west terminal mound is larger, measuring 21.4 m long by 11.3 m wide. With respect to height, the mound measures 2 m high on the interior and only 60 cm on the exterior.

The estimated volume is 239.9 m3. Aside from the corner and line of stones on the north lateral mound and the east terminal mound, there is not much evidence of preserved architecture on or around the ballcourt, probably due to modern agricultural activities. While there was a good density of ceramics on the ground surface, they were mostly non-diagnostic fine gray wares. It is possible that to construct the ballcourt they excavated down from the surface and used the excavated dirt as fill for the lateral mounds.

Although located right on the valley floor, the ballcourt has been constructed in an area with some of the best views of the surrounding lands and mountains (Figures 6.47). Thus, despite the plainness and simplicity of the court itself, its geographic position makes it rather remarkable. Because the elevation of the terrain where the court sits is slightly higher than the alluvial plain, the valley floor spreads out before you on the south and eastern sides, and one can appreciate much of the surrounding mountains. It would have been possible to view movement through much of this area.

395

Figure 6.47: Views of the surrounding landscape from the south lateral mound, looking north (upper left), south (upper right), east (lower left), and west (lower right).

6.3 CASE STUDIES – MUNICIPALITY OF SANTA ANA TAVELA

The town of Santa Ana Tavela today has 1200 inhabitants and encompasses around 82 km2 of territory, including the alluvial bottomlands and the mountains on the eastern and southeastern edges of the subregion. The tallest peak in the entire subregion is found here, in the eastern extreme of the municipality, the Cerro Capulín. Of the total number of sites registered,

64 or 38% have been found here. Although a new settlement, the history of Santa Ana Tavela actually stretches back deeper into the Colonial era. Even today Majaltepec, known as the pueblo viejo or old town of the Mixe people16, continues to figure prominently in the oral histories as well as the cultural identity of the Taveleños, with half the town still claiming

16 Although again the evidence from Majaltepec does not point to a clearly defined ethnolinguistic identity. 396 descent from the people who formerly lived there. In fact, people from Tavela were making pilgrimages to Majaltepec until the 1950s, and to this day they celebrate the patron feast day of the Virgen de La Asunción Majaltepec in mid-August.

In the following sections, I describe the courts we documented and mapped in the Santa

Ana Tavela municipality territory (Figure 6.48; Table 16). Due to concerns over the rainy season and later several forest fires that occurred near the mountain sites where some of the courts are located, it was decided that for safety reasons we would not map there. Therefore, only 3 of the 6 ballcourts in the Tavela municipal territory were mapped. The non-mapped courts are discussed later on in the chapter.

Figure 6.48: Map of the ballcourts in the municipality of Santa Ana Tavela. Ballcourts here are located much further apart and are not as clustered close together as in Nejapa de Madero.

397

Table 6.16: Ballcourts registered in the municipality of Santa Ana Tavela. Site Name Site Code Court Type Orientation Elevation (masl) Lachixoba YAU-TAV-TAV-003 I-shape N-NE 32° 626 La Puerta YAU-TAV-TAV-004 I-shape S-SE 162° 794 I-shape E-SE 111° 1211 El Sitial YAU-TAV-TAV-010 palangana n/a n/a n/a Los Picachos YAU-TAV-TAV-011 palangana n/a n/a 2080 (Los Chorros) Llano las Casas YAU-TAV-TAV-015 palangana S-SE 159° 949

398

6.3.1 Lachixoba (YAU-TAV-TAV-003)

Figure 6.49: Map of the Lachixoba site, which was part of the Greater la Amontonada community.

Konwest (2017) describes Lachixoba (Figure 6.49) as a neighborhood or barrio in the

Early to Middle Postclassic site of Greater La Amontonada. It is located only 100 meters east of

El Cucharital. Like the other neighborhoods in this community, Lachixoba has its own ceremonial precinct. Unlike the others though, this one includes a rather sizable ballcourt, indicating that the residents of Lachixoba probably had a high enough status to mobilize the necessary labor to build a court. The site was first registered and surveyed in 2009. It is located south of the Rio Grande and extends across three east-west oriented hills as well as the flat

399 terrains further south of the road connecting Santa Ana Tavela to Nejapa de Madero. This same road bisects the site into the north and south zones. In total, Lachixoba covers 17.1 hectares and incorporates 79 architectural features, but more than likely there existed other structures in the past. According to the señora who runs the local papelería or paper supply store, there used to be a rather large mound in the fields where they today grow watermelon and cantaloupe. This mound existed when she first came to Tavela from Guerrero several decades ago.

Archaeological evidence supports her statement – high densities of artifacts including ceramics, obsidian, and chipped stone, are found all over the agricultural fields even though any architecture that was there has since been destroyed, except for at the far south and west edges of the fields. Further south in this zone, the PANT team did record terraces and stone house foundations that probably belonged to households of a higher-status, since the stones are faced.

The zone north of the road is the largest part of the site, covering the flat terrains where the ballcourt and a platform are located as well as the flanks and summits of the three interconnecting hills. Structure preservation is much better on these hills. The platform, the nearest surviving structure to the court, is located about 73 meters to the west-northwest, and measures an estimated 36 m long by 27.2 m wide. Additional platforms, temples, and terraces that are most likely residential, were documented in 2011 and 2013 on the hills, primarily on the eastern-most. During the 2013 season, I directed a test pit excavation on the summit of the westernmost hill at a possible residential structure on a 16.9 m long by 8 m wide platform. In the west and northeast of the platform we could see other surviving sections of the structure foundation. The north and west portions of both the structure and the platform have been lost due to erosion as well as their position close to the western cliff edge. Under Konwest’s (2017) direction, we placed the unit next to an east-facing wall line that we assumed were the remnants

400 of a rectangular superstructure constructed on the larger platform. Excavating just a few cm below the ground surface, we found a layer of poorly preserved white stucco floor that directly overlies a second and better preserved white stucco floor covering the western side of the unit and running along and under the line of stones. The stones used for the foundation are small, consisting of only a single course that does not extend very much below the ground surface.

Based on the evidence from the excavations, we concluded that the line of stones was actually a small step from the stucco floor observed in the unit up to the structure’s interior. This stucco floor was most likely the surface of the platform.

When installing the screen for the excavated soil (1.5 m northwest of the test-pit), several layers of white stucco flooring were found just below the ground surface in the interior of the structure (Figure 6.50). In total, there were 3 layers of stucco, with each surface separated by about 2-5 cm of sediment. Clearly, the interior structure floors had been repaired more frequently than the exterior floors, indicating long-term occupation of this hilltop. Separately, scattered about the platform were several fragments of red painted stucco on the ground surface.

Given that we only observed white stucco when we excavated, it is likely that the red stucco covered the walls of the unpreserved superstructure that had existed on the platform.

401

Figure 6.50: Three layers of stucco floor visible from a posthole originally dug for the screen. These floors would have been the interior living surface of the platform.

402

6.3.1.1 The Ballcourt

Figure 6.51: 3D model of the ballcourt at Lachixoba, looking from the south.

Figure 6.52: Composite view of the Lachixoba ballcourt from the south.

Given the presence of a nearby platform, and the temple mound that once existed in the fields to the south, the ballcourt was likely located in the main ceremonial zone of the Lachixoba

403 neighborhood (Figures 6.51 – 52). It is an enclosed, sunken I-shape ballcourt, and is oriented

32° E of N (N-NE). Overall, the court measures 50.3 m long by 39.4 m across (Table 6.17). The central playing alley is 29.7 m long and is 4.4 m wide, while the north terminal zone is 19.6 m long by 4.4 m wide, and the south terminal zone measures 15.4 m long by 4.2 m wide. The east lateral mound measures 3.5 m in height on the interior and 3.2 m on the exterior, and 33.8 m long and 18 m wide with an estimated volume of 1378.9 m3 (Table 6.18). The west lateral mound is 3.2 m high on the interior and 2.5 m on the exterior, and 35.5 m long by 15.8 m wide with an estimated volume of 1164.7m3. A line of stones was observed on the south corner of the west mound, marking that corner of the mound. Both lateral mounds have clearly suffered damage from modern activities. The average slope of the lateral mounds is 21° (17.5-26.5°).

Table 6.17: Size dimensions for the Lachixoba ballcourt and playing surfaces. Overall Central Playing Surfaces Terminal Alleys North South Length 50.3 26.7 19.6 15.4 Width 39.4 4.4 4.4 4.2 Note: all dimensions in m

Again, the terminal mounds of the Lachixoba court are much smaller than the lateral mounds. The north terminal mound measures about 1 m in height on the interior (and less than

10 cm on the exterior), 30.4 m long by 8.2 m wide, with an estimated volume of 303.9 m3. The south terminal mound measures 20.5 m long by 5 m wide, with an estimated volume of 137.6 m3. Similarly, the mound slopes 1.2 m down to the playing alley, but it is not very extant on the exterior side. The PANT survey crew proposed that the south mound may have been destroyed by construction in 2009, when the town bulldozed a road to the new refuse dump right next to the ballcourt. However, it is also possible that like several of the courts in Nejapa, the superior

404 surface of this mound was only 10-30 cm in height on the exterior. Indeed, there is a 30-cm height difference between the east and west sides, with the former probably representing the mound’s actual height prior to the road construction.

Table 6.18: Size dimensions of Lachixoba court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 east Lateral 33.8 18.0 3.5 3.2 1378.9 Mound 2 west Lateral 35.5 15.3 3.2 2.5 1164.7 Mound 3 north Terminal 25.5 7.6 1.0 0.0 303.9 Mound 4 south Terminal 33.6 8.3 1.2 0.0 137.6 Note: all dimensions in m.

Many of the mound surface stones have been moved or are out of place but given the density of cobble stones it is likely that, when initially constructed, the entire surface of the mounds was covered with stones much like the court at MDP. These stones would have been obtained from the nearby Rio Guitihuini, which lies on the western boundary of the Lachixoba site limits. As such, we were able to observe that the mounds had been constructed using an earthen and stone-mixed fill. We documented several ceramic sherds on the surfaces in and around the court. Most of the sherds were plain fine gray and orange-fired gray ware, although we also observed other ceramic types such as course orange-gray jars, two solid supports, a possible tecomate or cylindrical vessel with a very narrow neck, and a fragment from the neck of a canter or small liquid vessel. One sherd, probably a hemispherical bowl, measures over 10 cm in width, indicating that large serving vessels were used and discarded near the court. The surface material strongly supports the hypothesis that feasts and other ceremonies tool place in and around the ballcourt.

405

6.3.2 La Puerta (YAU-TAV-TAV-004)

Figure 6.53: Map of the La Puerta site.

The site of La Puerta (YAU- TAV-TAV-004) (Figure 6.53) occupies the piedmont zone lying on the edge of the mountains that continue north to the Sierra Mixe. Occupation spreads across several hilltops that flank the Río Tehuantepec as it heads northeast out of the Nejapa subregion (Figure 6.54). Residents were able to not only see the river as it enters and exits the valley, but also the surrounding mountains and other sites nearby. Officially encompassing 2.1 hectares with 45 documented structures, the site is probably much larger since the PANT crew could observe additional structures and terraces further north along the ridgeline that were

406 unfortunately inaccessible due to the dangers of the topography. The survey crew documented a terrace and a poorly-preserved platform on the southernmost hill. A small saddle connects the southern hill and a lower area with the east and west peaks. Most of the occupation is found on the west peak, with well-preserved terraces – some over a meter high – and houses with small rooms. Many of the terraces here are more formal with stucco floors and structure remains, some of which were likely domestic units. Between the east and west peaks there are additional terraces and smaller constructions. Based on radiocarbon samples obtained from excavations, La

Puerta dates to the Early Postclassic (calibrated 1050-1270 AD). Other indications of full-time occupation at this site include a flexed burial under a stucco floor, extensive pottery scatters on the surface, and several manos and metates.

407

Figure 6.54: View of the Rio Tehuantepec from the La Puerta site, looking from the southern end (upper) of the site and the western end (lower).

408

In 2011, PANT crews excavated three test pits on residential terraces located in different sectors of the site. Operation 1 was placed along a small foundation wall on the interior of a possible house that had been constructed on a terrace on the western hill. Interestingly, excavated artifacts from the lower strata indicate an earlier occupation possibly dating to the Late

Classic period in whichever, currently unknown location they had taken the construction fill from. The surface occupation was covered by a thick cap of large stones and ash fill, and included ceramic sherds and stucco fragments that were also likely taken from elsewhere.

Another excavation was placed on a different terrace on the same hill, in the middle of a series of terraces descending down this side. This terrace has a view of the river as it leaves the valley. A white stucco floor was found about 21-24 cm below the surface, lying on a layer of fill that was used to level out the original surface. There is no evidence that this floor was repaired at any time. The burial mentioned previously was found in this excavation. Found almost a meter below the surface, the flexed and wrapped burial was placed directly on the bedrock on its left side with the arms around the legs. A fine gray-paste bowl had been put upside down on the right ribs.

Several sites contemporary with La Puerta can be found nearby. Visible from La Puerta is Piedra la Iglesia (YAU-TAV-TAV-005), a ritual site that is said to be inhabited by spirits, and in fact is still in use to this day (Figure 6.55). The 9.4-hectare site of Los Mogotes del Burro

(YAU-TAV-TAV-007) is located on a hill just opposite of La Puerta. The architectural features at Los Mogotes del Burro are particularly impressive given its location, with over 178 structures including plazas, mounds, platforms, terraces, residential structures, and even defensive walls.

There are many similarities between La Puerta and Los Mogotes del Burro. Both sites were well positioned to monitor anyone entering the region from the north as well as into the valley from

409 the south. Access to each site is restricted and difficult, suggesting a defensive strategy for building in these locations. Ceramic evidence collected at both sites is simple and without decoration, primarily gray and orange paste wares. To the south by 3.1 kilometers is the contemporary community of Greater La Amontonada, which probably would have been visible from the highest points in La Puerta and likely Los Mogotes del Burro as well.

Figure 6.55: Piedra la Iglesia, a sacred site still in use today that is visible from La Puerta.

410

6.3.1.1 The Ballcourt

Figure 6.56: 3D models of the La Puerta ballcourt from above (left) and from the southwest (right).

Figure 6.57: View of the La Puerta ballcourt from the north.

411

The La Puerta ballcourt was constructed on the saddle linking the southern hill peak to the two other peaks where the residential occupation is located (Figures 6.56 – 6.57). The orientation of the court is 162° N-NW, following the contours of the saddle. Overall, the court is

34.8 m long and 19.8 m wide, while the central playing alley itself measures about 22.4 m long by 4.8 m wide (Table 6.19). It is fairly simple in appearance and is probably the least complex of all the courts in the study sample. The north terminal zone is 17.4 m long by 5.1 m wide, and the south terminal zone is 17.7 m long by 4.8 m wide.

Table 6.19: Size dimensions of La Puerta ballcourt and playing surfaces. Overall Central Playing Surfaces Terminal Alleys North South Length 34.8 22.4 17.4 17.7 Width 19.9 4.9 5.1 4.8 Note: all dimensions in m

The east lateral mound measures 1.8 m high, 23.6 m long and about 7.7 m wide, with an estimated volume of 169 m3 (Table 6.20). On the eastern exterior side, which looks towards the mountains marking the eastern boundary of the Nejapa subregion, the mound surface continues directly into the slope of the hillside, such that there is no clear distinction between the structure and the natural surface. The west lateral mound only measures 1.4 m in height, and is 22.3 m long by 7.5 m wide, with an estimated volume of 147.3 m3. On the western exterior of the west mound the ground surface levels out for several meters before meeting the western edge of the saddle. The significant height different between the two mounds could be due in part to modification of the west mound by a local farmer in the 1950s, who used some of the west mound’s surface stones to create a corral on the playing alley for his animals. The east and west lateral mounds of La Puerta court are some of the smallest in the Nejapa subregion. Both

412 mounds are higher at the south end than the north; there is a 45-cm difference on the west mound, and a 50-cm difference on the east mound. The sloping surface of the mounds may be due to erosion or may be intentional.

Table 6.20: Size dimensions for La Puerta court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 east lateral 23.6 7.7 1.8 1.8 169.0 Mound 2 west lateral 22.3 7.5 1.4 1.4 147.3 Note: all dimensions in m except volume (m3).

Rather than terminal mounds of mixed earth and stone as seen at other sites, the existing bedrock at the north and south ends was shaped to form edges demarcating the terminal zones.

A seemingly natural, ascending entrance can be found at the south end of the court (Figure 6.58), and it even appears that the bedrock may have been shaped here to form a stair or entryway. In fact, when you take the modern vereda or path up to the site, you walk directly up and into the ballcourt. A small outcrop of bedrock, or perhaps simply a well-set stone, lies in southern end of the ballcourt alley (Figure 6.59). While the purpose of this stone is not clear, it did remind me of a botadora, the playing stone used in pelota Mixteca. From the northern end of the court, the hill slopes upwards again, leading to the residential occupation with some small public buildings.

413

Figure 6.58: Southern entrance into the ballcourt at La Puerta. It is possible that a natural rise was modified somewhat to create this entrance.

Figure 6.59: Large stone set into the southern end of the La Puerta court.

414

There is a limited amount of space in and around the ballcourt for viewing the game.

While it would have been possible to stand on the west lateral mound and see into the court, the same would not have been likely for east lateral mound given the narrowness of its superior surface and the extreme slope of the hillside on the exterior. The west mound was not so high as to entirely block the view of those standing in the small plaza on the west side of the ballcourt, but not much would have been visible from this perspective. An audience would most likely have stood at the north and south ends of the court, especially at the north where the end zone was much larger. Games and other activities taking place around the court were probably most visible from terraces on the hillslopes of the western peak. Certainly, the sounds of games would have been heard from many different points in the site.

415

6.3.3 Llano las Casas (YAU-TAV-TAV-015)

Figure 6.60: Map of the Llano las Casas site.

The site of Llano las Casas (YAU-TAV-TAV-015) stands out from the other communities with ballcourts due to its location and isolation (Figure 6.60). Llano las Casas

(LLC) is located in a small intermountain valley lying close to the boundary between Santa Ana

Tavela and the nearby San Juan Lajarcía municipality. To get to the site, we took a dirt road that veers off from the main road between Santa Ana Tavela and Nejapa de Madero, just after you pass the watermelon fields of Lachixoba. After ascending into the mountains for 45 minutes, we arrived at the corral where the herd of cows communally owned by Tavela will gather for salt

416 lick. This location is itself a site known as Bajo de los Hornos (YAU-TAV-TAV-038). We parked the car there, and then proceeded to hike for an hour and a half on a tiny footpath that heads south and west, crisscrossing a feeder river as we went. Halfway along the journey to the

LLC site we passed another path that takes you directly to El Camarón, a modern town an hour south from Tavela by road and only a 4-hour journey walking. After following this narrow vereda we came upon an intermountain valley or llano. The same river that we had crossed several times runs along the western edge of the site, about 11 meters from the western side of the ballcourt (and about 2 m lower in elevation). Aside from the court and a small platform, there is no other evidence of occupation at the site. The terrain surrounding the ballcourt is more or less level.

Although isolated, there are several sites located about a kilometer or more away from

LLC. Only 1.22 km to the northeast is El Sitial, a large and complex site with multiple formal public structures, a palace, and two unusual ballcourts (to be described later in the chapter). A kilometer to the northwest is La Loma del Orejón (YAU-TAV-TAV-032), which may have been a neighborhood of El Sitial but was registered as a separate site. There are also a few sites to the south that lie about 1.5 km away. Cerro La Virgen (YAU-SJL-SJL-005) lies on the southeast skirt of the hill with the same name, and primarily consists of terraces for cultivation and ceramic sherds scatters. Close by to Cerro La Virgen is Peña del Rodeo (YAU-TAV-TAV-045), where about 16 structures were documented. Most likely people lived at Peña del Rodeo and farmed at

Cerro La Virgen. The relationships between the sites to the north and south of LLC is unclear at the moment, as we only have secure dates for El Sitial. The ceramics we observed at LLC were rather generic and therefore inconclusive with regards to a date.

417

6.3.3.1 The Ballcourt

Figure 6.61: 3D models of the Llano las Casas ballcourt from above (upper) and from the southwest (lower).

418

Figure 6.62: View of Llano las Casas ballcourt from the south.

The ballcourt at Llano las Casas (Figures 6.61 and 6.62) is unique because it is a palangana or washbasin-style court, unlike the others in the sample. Courts of this type are typically found in Guatemala, although a number have been documented in central (n = 5) and northern Oaxaca (n = 6) close to the border with Puebla. Palangana courts are generally rectangular in shape with low walls that connect at the corners, forming right angles. In many cases, the lateral mounds are larger than the terminal mounds. Many of the recorded palangana courts have smaller buildings constructed on the enclosing structures (Taladoire 2003); however, only four of the courts have a set of stairs, usually located in the center of an end wall.

Palangana courts are also highly variable in their widths and lengths as well in the profiles of the structure walls. Indeed, there appears to be no regional standard or pattern when it comes to

419 palangana court construction (Taladoire 2003:332-334). Most have been dated to the Classic period, and thus may indicate a similar date for the LLC site.

The court here is indeed rectangular, with lateral mounds larger than the terminal mounds, and is oriented 159° N-NW. It is one of the smaller courts for this area, measuring 38 m long and 14.5 m wide overall, while the playing alley is 27.6 m long and 5.2 m wide (Table

6.21). The east lateral mound is slightly mounded above the level of the ground surface on its eastern exterior, and slopes 1.4 meters down to the playing alley (Table 6.22). It measures 28.8 m long by 5.1 m wide, with an estimated volume of 128.4 m3. The west mound, on the other hand, is more clearly defined, rising 1 m above the ground surface on the exterior side and 90 cm on the interior. It measures 28.6 m long by 6.3 m wide, with an estimated volume of 67.3 m3.

The mound slopes upwards 70 cm from the north to the south end. The average slope of the lateral mounds is 11.6° (9.5-13.5°).

Table 6.21: Size dimensions for Llano las Casas ballcourt and playing surfaces. Overall Central Playing Surface

Length 37.9 14.5 Width 27.6 5.2 Note: all dimensions in m

Table 6.22: Size dimensions for Llano las Casas court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 east lateral 28.8 5.1 1.4 0.0 128.4 Mound 2 west lateral 28.6 6.3 0.9 1.0 67.3 Mound 3 north terminal 12.6 3.4 0.5 0.2 16.2 Mound 4 south terminal 14.5 6.5 1.3 0.7 84.0 Note: all dimensions in m.

420

The north terminal mound rises only 20 cm above the ground surface on its exterior, sloping 50 cm down to the playing surface. It measures 12.6 m long by 3.4 m wide, with an estimated volume of 16.2 m3. The south terminal mound – like the west lateral mound – has been clearly constructed up from the ground, with stones lining the surface. It measures 1.3 m in height from the playing surface, but only 70 cm on its exterior side. It measures 14.6 m long and

6.5 m wide, with an estimated volume of 84 m3. At about the center of the south mound, there is a double-inset entrance that would have allowed access into the court from this side (Figure

6.63). Like several of the courts in Nejapa, the people who built it took advantage of the natural topography.

Figure 6.63: The double inset entrance on the south terminal mound, which would have allowed access to the court from this side. Apart from the ballcourt and the probable plaza or flattened space surrounding it, the only other architecture present is a small, 13 m by 7.9 m platform that rises 30 cm above the

421 surrounding terrain and is attached to the southwest end of the court. This platform feature fits the general description of palangana courts – small structures constructed on the enclosing wall.

Here the only difference is the platform is attached to the court, rather than constructed on it (see contrasting examples in Taladoire 2003). Interestingly, in the northwest corner of this same platform is a large boulder with glyphs carved on its western face, consisting of a spiral, a double cross, and a scroll (Figure 6.64). On the southern face, there appears to be another double cross

(or a butterfly glyph) and perhaps a double circle.

422

Figure 6.64: View of the carved glyphs on the stone marking the corner of an attached platform, showing the north (upper) and south (lower) faces.

423

Ballgames and game activities taking place in and around the court would have been highly visible to any audience that was present. The low height of the mounds meant that they would not have blocked the view of anyone standing to the north, east, or west of the court (there would have been no room to stand on the south, as only a meter or two from the exterior the terrain gradually slopes down to the river). The bigger question is, who would have used the court? Who would have had control over the games and activities at this court, and who would they have invited? At the moment, it is difficult to answer these questions with the data we currently have. Further investigation at this site and the communities nearby should help to resolve them in the future.

6.4 CASE STUDY – SAN CARLOS YAUTEPEC

The municipality of San Carlos Yautepec is the largest in the Yautepec District, with territory covering nearly the entire district north from the mountains of the Mixe subregion, across the valley floors of Nejapa subregion towards the south, nearly touching the Pacific coast

(Figure 6.64). Indeed, at its southernmost point the municipal boundary is only about 11 km from that southern coast. In total, the municipal territory encompasses 2491 km2 of mountains and valleys crossing a range of environments. The municipal head is the eponymous San Carlos

Yautepec, in the north-central part of the municipality, and is located just southwest of Highway

190. Survey in this area was briefly conducted in 2011 and 2013, and so far, 29 sites have been documented (17% of the total sites registered). PANT survey crews covered the territories around San Carlos Yautepec and towns to the south, including San Francisco Guichina (3.42 km to the south-southeast), San Isidro Manteca (about 5 km to the east-southeast), and Santiago

424

Vargas (about 12 km to the southeast). With the registration of La Baeza, the total known number of sites in this area is now 30.

Figure 6.65: Map of the sites documented by PANT so far in the San Carlos Yautepec municipality.

Since this area is relatively unknown to archaeology, in the following section I describe the sites registered so far in the San Carlos Yautepec municipality. In doing so, I aim to not only give context to La Baeza, but also to tease out settlement patterns and identify other communities that may have existed in this area and that perhaps were even contemporaries of La Baeza.

Much still remains to be explored, so these interpretations are preliminary at best. Indeed,

425 photographs taken over 100 years ago of archaeological sites that have not yet been registered with INAH in the San Carlos municipality and around La Baeza (recorded in C. George

Rickard’s Ruins of Mexico [1910]) demonstrate that this area was likely densely populated in the past.

6.4.1 The Prehispanic Settlements of San Carlos Yautepec (So Far)

Figure 6.66: Map of sites recorded in the territory of San Francisco Guichina.

The following sites are found about 3 kilometers to the southeast of the modern pueblo of

San Francisco Guichina (Figure 6.66). The site of La Loma del Estoraque (YAU-SCY-GUI-

426

001) consists of about 12 or so terraces covered with stone foundations. There also appears to be a small architectural grouping of four rectangular structures surrounding a square plaza. On the southeast side, we find a 2 m2 mound measuring over a meter in height. This site was likely part of a larger community that also included Cerro de Algarrobales (YAU-SCY-GUI-003), which lies only a kilometer towards the southeast from Estoraque. A much larger settlement, Cerro de

Algarrobales measures over a kilometer in length from the western end of the ridgeline towards the east-southeast. The site is full of stone line foundations, structure walls and terraces as well as six possible architectural groupings. In the northwest portion of the site there is a square patio that is surrounded by large, rectangular structure mounds, along with other smaller structure groupings. Towards the southeast there is a temple-patio-alter group (TPA)17, a common ceremonial complex found throughout Oaxaca. This temple is an exceptional building and even features two columns. The Cueva del Huizache (YAU-SCY-GUI-002) is a rock shelter found only 50 m to the west of the western boundary of Cerro Algarrobales. The wall of the shelter is covered in red hands painted in negative form. A few ceramic sherds were also documented, suggesting that the wall shelter had served as a place for ritual activity in the past.

Northeast of Cerro Algarrobales by about 900 m is Loma de Tío Ramiro Sur (YAU-SCY-

GUI-016), a site that extends over the ridgeline of a nearby hill. Here the PANT survey team documented several structures, platforms and terraces. El Manguito (YAU-SCY-GUI-004), located 1.1 km north of Cerro Algarrobales, is a large stone with a painted hand-print that, based on its size, may have belonged to a child. The stone was found in an arroyo that runs alongside the access path going to Cerro Algarrobales. About 70 m towards the southwest from the large stone where the arroyo terminates in a rock shelter, the survey crew found ceramic sherds,

17 Temple-patio-altar groups usually consist of a temple mound next to a patio or small plaza, with an altar structure in the middle, and are usually Late Classic in date. 427 including fine gray ware and orange paste ware decorated with designs. Given the proximity of these sites each other, if they are contemporaneous then they likely represent a single community centered around Cerro Algarrobales. At the same time, Loma del Tío Ramiro Sur is located across a large barranca separating the Cerro Algarrobales from the Cerro Grande, which may mean that this settlement may pertained to the community on that hill, Cerro Antonio (described further on).

El Mantecón (YAU-SCY-GUI-007) is site located on the nearby Cerro Grande. Two large defensive walls measuring over 2 m in height and lying about 58 m apart begin in the southernmost part of the site, and continue for over 40 m. Stone terraces cover the north, east, and western flanks of the hill and were probably used for both habitation and cultivation. In the southern part of the site there is a large structure measuring 5 by 6 m, most likely some type of public building. El Mantecón has been interpreted by PANT as a residential and religious settlement, based on the site’s location, the views afforded by the location, and the diversity of the architecture present. Close to El Mantecón are the Cueva Grande del Mantecón (YAU-SCY-

GUI-005), with red hand-prints in negative form (n=5), and Las Manos Pintadas (YAU-SCY-

GUI-006), where some of the 22 handprints also appear to be child-sized. Continuing north from

El Mantecón by about 1 km lies La Mesa (YAU-SCY-GUI-008), appropriately named as the site is located on a small mesa. Here they found multiple stone lines and foundations. In the southeast zone of the site there is a large 17 m by 10 m mound on top of a platform that itself measures 32 m by 50 m. Almost a kilometer to the northeast of El Mantecón is Loma del

Peñasco (YAU-SCY-GUI-017), a series of terraces with structures built on top that spread across the hillside and extend further to the southeast. Only 550 m from Loma del Peñasco is El Abrigo del Encanto (YAU-SCY-GUI-012), another rock shelter with nine painted hands in negative

428 form. Again, if these sites are contemporaneous it is possible that they constituted a community centered at El Mantecón.

Loma de Tío Ramiro (YAU-SCY-GUI-011) is a residential hilltop site with multiple terraces on which rooms and structure walls had been constructed. Only 80 m to the north from there is the ritual cave site of Cueva del Nopal (YAU-SCY-GUI-010). Loma de Tío Ramiro itself lies only 620 m to the southeast of Cerro San Antonio (YAU-SCY-GUI-014), a site found within the communal territory of San Francisco Guichina that is also readily visible from there.

Located on a rounded hill that links to the larger Cerro Grande, Cerro San Antonio consists of huge terraces on the north, west, and south sides of the hill, with grand walls on the south that limit access to the settlement on this side. These terraces are covered with metates, ceramic sherds, and the remains of what were likely habitational structures. Grander and more formal structures are found on the summit of the hill, including a stucco plaza, platforms, mounds, and even subterranean tombs! One tomb had been constructed under the plaza, and the arrangement of the two is intriguing. The entrance to the tomb is located on the open, northern margin of the plaza; the tomb’s chamber extends into the nearest terrace on that side. The other three sides of the plaza are defined by rectangular structures. Three other terraces further down also feature structures, many of which surround patios. One of the most unusual architectural groups is defined on one end by a structure in the form of a U that is also bordered on the east by a large open plaza. The town of San Francisco Guichina itself sits upon a Prehispanic site, and many of the modern constructions have actually been built on Prehispanic mounds. Close to the house of

Doña Benita López (YAU-SCY-GUI-015), the PANT crew found a small hill that had been cut by machinery, revealing stucco floors, ceramics, and lithics. If contemporaneous, then it is

429 possible that Cerro San Antonio was a major center in this area of Nejapa, with occupation extending from the Loma de Tío Ramiro site to where modern Guichina is today.

Figure 6.67: Map of the sites recorded in the territory of San Isidrio Manteca and San Carlos Yautepec.

A much smaller Prehispanic community might have also existed close to the modern pueblo of San Carlos Yautepec and the smaller community of San Isidro Manteca (Figure 6.67).

Only 1.5 km to the northwest is the site of Cerro Quebrantadero Sur (YAU-SCY-SCY-001). Not much was found here except for ceramic sherd scatters, worked lithics, and a metate. The site appears to have had multiple occupations, as we identified ceramic sherds from the Postclassic,

Colonial, and the Historic periods. A kilometer further northwest from Cerro Quebrantadero Sur is Cerro Quebrantadero Norte (YAU-SCY-SCY-003). There the PANT survey crew 430 documented two structures. The building in the north section of the site was open on its south side. A complete incensario was found left as an offering at the side of the structure foundation.

The nearby Cueva de Don Luis (YAU-SCY-SCY-002) is another ritual cave site with rock paintings. Here the images consist of geometric forms and a painted hand, while there were ceramic sherds scattered at the opening of the cave.

Figure 6.68: Map of the sites recorded in the vicinity of Santiago Vargas.

The survey crew also documented several sites in and around the modern village of

Santiago Vargas (Figure 6.68). The large site of Cerro Macahuitillo (YAU-SCY-STV-002) lies only 130 m north from the town cemetery. On the top of the hill and extending north by 850 m

431 there are several stone lines, terraces and foundations. In the highest part of the hill there appears to be a collapsed stone adoratorio or altar measuring 1 m in length and 70 cm in width.

Moving further north on the east side of the hill there is a large structure that had been constructed right next to the crag, measuring about 10 m in length and 5 m wide, making it much larger than the other buildings found scattered throughout the site. Bajo del Cerro de

Macahuitillo Sur (YAU-SCY-STV-005) is another site that lies close to Cerro de Macahuitillo.

Climbing the hill going north, the survey crew registered foundations and terraces, and even four complete, well-preserved structures. Evidence suggests that there may have been even more structures in this section of the site as there is a lot of rubble and stone collapse.

Closer to Santiago Vargas, only 125 m to the north, is Bajo del Cerro de Macahuitillo Oeste

(YAU-SCY-STV-006). There is a 10 m long by 5 m wide isolated stone structure on a small knoll in the eastern portion of the site. A set of stairs on the southeast side of the knoll heads downwards and continues for about 40 meters. Likely all three sites were part of the same community, given their proximity to each other and their location on and around the Cerro

Macahuitillo.

Located just next to Santiago Vargas is a large 40 m by 25 m stone-and-earth platform that measured over 2 m in height at a place known locally as Bajo de Palacio (YAU-SCY-STV-

001). To the south of the village about 330 m away is Bajo de Palacio Sur (YAU-SCY-STV-

003), where the survey crew documented an isolated 5 m by 2 m room and, 45 m northwest of this small structure, a mountain of collapsed stones that perhaps belonged to a much larger structure. Bajo del Palacio Este (YAU-SCY-STV-004) is another site just south of Vargas by about 370 m, with terraces and rooms on the hillside. Most likely this was a habitational site or

432 perhaps even barrio for another community nearby, with additional structures that existed in the past since lost due to severe erosion.

The settlement evidence documented by PANT suggests that there were several communities with possible multiple neighborhoods dispersed over the hillsides and ridgelines in this part of the San Carlos Yautepec municipality. I suggest that these communities were centered around the larger settlements of Cerro de Algarrobales, El Mantecón, Cerro San

Antonio, Santiago Vargas, and Cerro Macahuitillo. It is interesting that there are several ritual sites with red-painted hands near to these communities – rock art sites are incredibly common throughout Nejapa, and obviously had a significant social and ritual meaning. Of course, further investigation is needed to determine if these sites are contemporaneous or not. But if similar material patterns bear out, it is likely that the communities here were organized much in the same way as Greater La Amontonada: a central neighborhood or ceremonial center with other, dispersed neighborhoods, some of which had their own, smaller ceremonial precincts or public buildings. Some sites are only households made up of related family groups. Dispersed across the landscape was a network of ritual sites centered on rock shelters and caves that likely were part of communal religious ceremonies. If these communities are indeed contemporaneous, then it would suggest that this area was densely settled in the past. Exactly when these settlements were occupied remains to be determined.

6.4.2 La Baeza (YAU-SCY-BAE-001)

The village of San Carlos La Baeza is a small community of only 73 inhabitants, perched on the eastern hillside of the Cerro San Antonio (formerly known as Las Tijeras) at about 1580 m above sea level (Figure 6.69). It is located about 7.6 km directly north of San Carlos Yautepec

433 and is relatively isolated – both San Carlos and the town of El Camarón are about an hour and a half away by car, while the village of Aguas Frías is five hours away walking (there is no road to there), although still within the agencía territory of La Baeza. From the main road that goes through the village to the town of La Hacienda in the municipality of San Pedro Totolapan

(whose territory lies north and northwest of La Baeza in the Tlacolula District), there is another dirt road that goes west and then north on its way through the mountains. Where it veers north there is another dirt road, constructed as recently as 2014, that winds up to and across the summit, passing by the archaeological site of La Baeza (Figure 6.70) an astounding 1900 meters above sea level. Much of what we have observed of the site core runs along the ridgeline of the

Cerro San Antonio, and from multiple vantage points there are great views of the surrounding mountains and valley floors below, including places where one can directly see the towns of El

Camarón, El Gramal, and San Carlos Yautepec. There are also vistas where it is possible to see towards the municipalities of San Pedro Totolapan, Nejapa de Madero, Santa María Zoquitlán,

San Bartolo Yautepec, San Francisco Guichina and San Pedro Mártir Quiechapa. On our first visit in 2013 we asked our guides how long it would take to travel to some of these places from

La Baeza. They reported that it takes the entire day walking to reach Zoquitlán, about 7-8 hours walking to Totolapan, and about 4-5 hours to reach San Carlos Yautepec.

434

Figure 6.69: The basketball court at La Baeza, looking east from the agencia building. The whole town is perched on the hillside.

435

Figure 6.70: The site of La Baeza based on current mapping data. We expect the settlement to be much larger once fully mapped.

Manual Martínez Gracida was the first person to describe La Baeza, including the mysterious site in his list of Zapotec fortresses running from the Central Valleys to the Isthmus.

Likely he received his information from local sources or visitors who had made trips to the area.

Agustín Delgado (1965) also wrote about La Baeza as a Zapotec fortress. However, based on our brief survey, we determined that the defensive characteristics of the site were more a result of the location than any intentional construction by the Prehispanic inhabitants. There were no defensive walls or features that typically characterize fortified sites, and the extent of the occupation as well as its density and monumentality instead points to a continually inhabited settlement. There are also several ways to access the site, although they would still have been

436 well-controlled and monitored. During our informal survey, we observed several, well-preserved platforms, multiple plazas, often with structures on all sides, and many smaller residences. Many of the structures were clearly defined with stone walls that are still standing. On our return to the village, we took a more northerly route and observed additional temple-patio-altar complexes, or

TPAs. King counted about 15 TPAs in total in the time we were there. A second visit in 2015 to map the ballcourt revealed additional structures that had not been seen in the previous visit nor again when we returned in 2016.

On our March 14, 2016 visit, we began the day by leaving the village around 5:15 in the morning, traversing the dirt road that heads out of town and ascending higher into the mountains before the sun had even risen. After walking 2.8 km from the village, the first structure we encountered was a small, blown-out mound foundation measuring 3 by 3 m. The terrain surrounding the structure had obviously been flattened and was probably a small plaza in the past. Apart from this first structure, no further architecture is visible from the road for another kilometer, where we observed a possible TPA or other special type of public architectural group.

Group A consists of two structures (Figure 6.71). Structure A1 is 28.4 m long by 26.518 m wide double-platform, with an 8 m long by 8.4 m wide superstructure that was itself likely a double temple (Figure 6.72). Some of the adobe walls on the east side of the building superstructure, close to where the entrance would have been, are still standing. A square outline can be seen on the surface of the patio, perhaps the stone foundation for an altar that no longer exists. A second platform (A2) lies on the east side of the plaza, measuring 22.4 m long by 24.8 m wide. Two room superstructures are still preserved; the north room is 5.5 m by 3.3 m and the south room 7 m by 3.7 m. Here too there were some wall remnants that were likely constructed of adobe

18 All measurements are listed north-south and then east-west. 437 bricks (Appendix A, Figure A.23). A line of stones, likely the top of a terrace retaining wall, borders the north side of the patio and the hill slopes immediately downward from this line.

Figure 6.71: Map of Group A structures at La Baeza.

438

Figure 6.72: Structure A1, a possible double temple built on a platform. This structure still has preserved earthen/adobe walls.

Southeast by 130 m of this first group is another architectural complex, Group B (Figure

6.73). First, there is a massive platform (Structure B1) with two superstructure patio groupings.

Patio Group B1-1 is a grouping of three room foundations: the north (Room 1) room is 4.7 m by

5 m, the south room (Room 2) measures 5 m by 3 m, and the east (Room 3) 5.8 m by 4.5 m. The three rooms surround a small 7 m by 11 m patio. The patio group is open on the south side.

Twelve meters to the east is Patio Group B1-2, a second grouping of three room foundations.

Room 1 on the west side measures 6.5 m by 2.6 m, while on the south side Room 2 is smaller at only 3.2 m by 4.1 m. The third room on the east side is not as well preserved and only a remnant of the stone line foundation remains, measuring 3.5 m by 2 m. The rooms surround a patio that roughly measures 6 m by 9 m, and that is open on the north side. Only 15 m to the south of these

439 two structures is a line of faced stones, another wall line, and a metate stone that had broken in half. The faced stones are in the shape of a rectangle measuring 2 m by 1.7 m by 1.4 m, and they may have been the foundation for another room or patio group.

About 30 meters to the southeast is Structure B2, a large earth-and-stone platform roughly 29 m by 29 m with a small 6.3 m by 3.3 m superstructure, possibly a room, on its southern end. Only about 20 m or so from this large platform is a 13 m by 3.4 m structure with standing preserved walls over a meter in height, Structure B3 (Figure 6.74). This may have been constructed on a small platform that is now poorly preserved. A line of thin, rectangular faced blocks lies on the first course of stones, with two more courses of larger stones on top. It is certainly possible that the walls were much higher in the past. On the western side of B3 there was a large staircase that leads out to a substantial, flattened space, most likely a plaza (see

Figure 6.74). On the east side of this structure is a stone and earth foundation (Structure B4), also not well-preserved, that measures 7.7 m by 7 m. The relationship between B3 and B4 is not clear and will need to be explored further with excavation. We observed a third group of structures, Group C, to the south of Group B but did not have sufficient time to map it (Figure

6.75). In this group, there were several structures, including room foundations and even two sunken patios. We even observed two large tlecuiles or stone-lined hearths that appear to be located in the patio itself (Figure 6.76). We are not certain of the relationship between this group of structures and Group B. Both Group B and C are located on mostly flat ground surface where the ridgeline widens out considerably.

440

Figure 6.73: Map of Group B structures at La Baeza.

441

Figure 6.74: Structure B3 (upper) may have been a temple. On its west side there may have been stairs with access to the plaza on this side (lower).

442

Figure 6.75: Group C is located southwest of Group B and may have been a residential group (photo by Júan Jarquin).

443

Figure 6.76: The tlecuiles or hearths found at Group C (photo by Júan Jarquin).

Groups B and C are also where we see an architectural technique that appears to be unique to La Baeza so far (although a similar constructions style has been seen at Zaachila).

Room stone-line foundations feature thin, planed rectangular blocks that are about 15-25 cm in length and bed depth, and 5-7 cm in height (Figure 6.77). Oftentimes, these foundations are a mix of faced square stones and the stylized blocks (Figure 6.78). The blocks also appear to have been manufactured from different types of stone, as they vary in color and grain from dark and rough, to smooth and white and a sandpapery brown of various shades. Of course, this is based on observation alone, and petrographic studies are needed to evaluate this claim further.

Nonetheless, these blocks represent an interesting and distinct characteristic of the La Baeza site

444 that has not been observed elsewhere in the Nejapa subregion. In the rest of this section I refer to them as La Baeza-style blocks.

Figure 6.77: Examples of La Baeza style bricks that had been looted from a nearby structure previously.

445

Figure 6.78: A room foundation using La Baeza style bricks. Many of the room foundations used these bricks.

Walking another 210 meters, we arrived at a small saddle between two peaks of the Cerro

San Antonio. Here we find Group D (Figure 6.79), which encompasses several monumental structures including the ballcourt (described in the following section). What always strikes me, both in 2013 and in subsequent visits, is the large rectangular building that is more or less attached to the southeast corner of the ballcourt, measuring 24 m by 6 m. Aside from the proximity of Structure D2 to the ballcourt (Structure D1), the building is impressive because a large portion of the west wall is well preserved, standing at over 2.4 m (Figure 6.80). The wall has been constructed with a lower foundation of large, roughly triangular shaped stones alternating with vertically stacked smaller stones filling the gaps in-between. This foundation is followed by a course of horizontally placed thin rectangular stones. The third course somewhat

446 repeats the pattern from the foundation, except here the large stones are more ovoid in shape.

Another layer of horizontally placed stones follows this third course, and the pattern continues for several more layers, the only difference being that closer to the top the larger stones are directly together, with no smaller stones placed in between.

Figure 6.79 Map of the Group D structures, including the ballcourt.

447

Figure 6.80: Structure D2 is a long rectangular building with much of its western wall preserved.

Structure D2 lies on the east side of a 30 m by 43 m plaza that itself is directly south of the ballcourt. A terrace retention wall borders the south end of the plaza, running along the edge of the saddle. In the rough center of the plaza there is a small square mound of stones (Structure

D3) that we did not map. It appears to be similar in size and appearance to the square stone line foundation – also roughly in the center of the plaza – at Cerro Maluco (Figure 6.81). This may be the altar structure of the TPA complex. An additional structure lies in southwest corner of the plaza, Structure D4, a 7.2 m by 7.3 m rectangular foundation made of earth and stone. The purpose of Structure D4 is unclear, but it parallels the relatively smaller mounds that border the west sides of the plazas at Cerro Maluco and Mogotes del Panteón.

448

Figure 6.81 We did not map Structure D3, but it appears to be a foundation for a possible altar, and is similar in appearance to the altar structure at Cerro Maluco.

On the western end of the plaza there is a massive terrace or platform (Structure D5), 29 m by 4.5 m, which appears to have been partially constructed of modified bedrock. There is a grouping of rooms on top, Patio Group D5-1. The west room is the largest at 8.5 m by 2 m. The other two rooms are much smaller, with the south room measuring 1.4 m by 3.4 m and the north room only about 2.6 m by 2.8 meters. There likely was a 10 m by 7 m interior patio in between these rooms. Behind and up from the west room about 11 m to the southwest lies another stone and earth foundation measuring about 6.3 m by 2.5 m (Structure D6, not mapped), which may have been constructed on a possible terrace as we did observe several stone lines that were probably retaining walls. There may have been additional room structures here that simply did not preserve well, as some of the lines of stones we could see – particularly on the north side of

449

Structure D6 – were faced. We also observed a tlecuil only about 3 m west of D6’s northwest corner. Structure D6 also stands out because there is a clear entrance marked by two large upright stones on the east side (Figure 6.82).

Figure 6.82: The foundation for a room on top of Structure D6.

Continuing west for another 40 m from Structure D6 we encountered Group E, Structure

E1 – an impressive building complex consisting of several patio and room groupings (Figure

6.82). What you first notice as you approach Group E is a massive outcropping of bedrock that has been squared off to form the northeastern edge of the platform on which the superstructures had been built. The first grouping (E1-1) in Structure E1 consists of what were likely four room structures around a patio. The north room is a 5.4 m by 6.3 m mound, the east room 8 m by 3.2 and the south room is 2.9 m by 7.8 m, while the room mound on the west side of the patio is the

450 largest at 13.3 m by 3.7 m. This room I refer to as the Main Room due to its size and location, as it also marks the east side of the second patio grouping on the west side of Structure E1. The second patio grouping E1-2 is smaller than E1-1; the north room mound measures 2.7 m by 7.7 m, while the south room mound is much squarer shaped at 4.9 m by 4.9 m. A 3.3 m by 1.8 m stone foundation mound is located in the southwest corner of E1-2, and its function is not quite clear at the moment. There is an interesting architectural difference between the two patio groupings. The room mounds of E1-2 features La Baeza-style block foundations, while the rooms of E1-1 were constructed of mixed faced and roughly-shaped stones.

Figure 6.83: Map of Group E structures at La Baeza. This may have been an elite residential and/or administrative building.

451

To the south of E1-1 and E1-2 are two lines of stones that appear to mark a hallway or passageway between the two patio groupings. This passageway also steps down to a series of possible retention terraces or a third patio grouping, as on the west side of this terrace there is 6 m by 3 m room mound. In the south room of E1-1 we observed another tlecuil lined with La

Baeza-style blocks. While the inhabitants took advantage of the bedrock outcropping to create the eastern portion of the platform for Structure E1, the western portion was clearly constructed of earth and stone. Based on the complexity of the architecture, we interpret Structure E1 as an important elite residence or administrative building, or at a combination of both (Figure 6.83).

Only 24 m to the northwest of Structure E1 is Structure E2, a 3.5 m by 3.3 m line foundation of

La Baeza-style bricks. This structure could be part of a larger superstructure that is below the surface or has eroded away.

452

Figure 6.84: View of Structure E1 from the west.

Further to the west along the ridgeline, about 265 m from Structure E2, is Group F – consisting of two ascending terraces and a large platform with several mounds, possibly rooms, and a sunken patio (Figure 6.85). On the southern end of the terrace itself we observe a 3.7 m by

3.2 m mound (Structure F1). The platform (Structure F2) itself measures 34 m by 21 m and lies on a northwest-southeast axis. It was constructed using faced stones blocks and earthen fill

(Figure 6.86). The mound (Structure F3) on the southeast side of the interior sunken patio measures 15.6 m by 3.3 m. On its eastern exterior, the mound foundation was constructed using a course of large stone blocks as the foundation, followed by a second course of medium-sized stone blocks. Interestingly, the third course is the La Baeza-style blocks. A line of larger stone boulders, many faced on one side, lies on top of the thinner blocks. These stones are jumbled and appear to be out of position; they may have been a fourth course of stones or fall from a

453 collapsed wall. The western interior of the mound, facing the sunken patio, is markedly different as there are only three courses of roughly shaped stone boulders of various sizes (Figure 6.87).

On the southwest side of the patio, we found the remnants of an earth and stone structure

(Structure F4) of which only the north and west sides remain, measuring 8.2 m by 13 m. There appears to be a small 3 m line of stones extending from the southeast corner that may connect to

Structure F3. A small 4.1 m by 3 m earth and stone mound (Structure F5) can be found on the west side of the patio. Finally, on the northern end of the platform, outside of the sunken patio, is Structure F6, another earth and stone mound measuring 3.6 m by 10.4 m. The wall marking the boundary between the patio and Structure F6 consists of a two-course stone block foundation with a standing earthen wall, possibly made of adobe bricks (Figures 6.88).

454

Figure 6.85: Map of Group F structures at La Baeza.

455

Figure 6.86: Structure F2 is a large platform with room superstructures on top and a sunken patio.

456

Figure 6.87 The east and west sides of Structure F3, which show very different construction techniques.

457

Figure 6.88: A standing earthen/adobe wall can still be seen on Structure F6.

To the west of Group F by 44 m is Group G, a massive double platform with mound superstructures on the second platform (Figure 6.89). The first platform (Structure G1) measures

25.6 m on the north-south axis, 12.6 m on the south end and 8.3 m on the north end. At over a

1.5 m in height, the platform is a significant construction. From the northwest corner, the platform wall continues northeast by 9.4 m, then turns to the northwest and extends 15.6 m in this direction. The east wall of G1 was constructed using a mix of faced stone blocks and roughly shaped boulders (Figure 6.90). The first course consists of vertical boulders with other stones of various sizes, including some placed horizontally, wedged in between. The second course is a line of horizontally oriented and shaped blocks. The third course repeats the pattern of the first. Access to the top of GI is by an incline or ramp on the southeast corner. On top of

G1 is a second massive platform, Structure G2, which features a large five-step staircase in the

458 center allowing access to the superstructures on top (Figure 6.91). This second platform measures 24.2 m by 21 meters. The wall of G2 is markedly different from G1 in that it was constructed using only large stone boulders of various sizes, roughly faced, with some reaching over a meter in height and half a meter in width. Three such boulders define the edges of the staircase, and it is possible to see horizontally oriented boulders extending back from the wall face in the sides of the stairwell. The platform itself is over 2 meters in height. Group G lies about 1970 m above sea level.

Figure 6.89: Map of the Group G structures at La Baeza.

459

Figure 6.90: Structure G1 is the first platform of the Group G complex. It stands at over a meter tall and features massive stones.

460

Figure 6.91: Structure G2 is the second platform and lies directly on top of Str. G1. An inset staircase gives access to the rooms and other structures on this platform as well as the western extent of the site.

On the surface of G2 we find two mounds. Structure G3 is an earth and stone mound located on the north side of the platform measuring 4 m by 11.7 m. In the northwest corner of the same platform is Structure G4, an 8.8 m by 11 m earth and stone foundation with a small, preserved wall. This structure is oriented northwest-southeast. From here the ridgeline takes a turn to the northwest, and the following structures are oriented accordingly. Additionally, the ridgeline at this point is at its narrowest, measuring only 10-20 m across and surrounded by steep drops on each side. To the northwest of Structure G4 by 10 m is Structure G5, a 12 m by 5 m stone and earth foundation, this time using La Baeza-style blocks. On the west side of G5 is a

17.7 m by 15 m plaza, bordered on the northeast and southwest sides by retaining walls.

Continuing northwest by 25 m we find two additional structures, G6 and G7, that are actually

461 located lower on the slope descending from the ridgeline on the northeast side. Structure G6 is a stone line foundation of La Baeza-style blocks and measures 13 m by 3 m. Directly across from

G6, about 14 m to the northwest, lies Structure G7. An unusual earth and stone building measuring 8 m by 7.7 m, G7 is open on the north side, perhaps indicating that this was a U- shaped structure similar to what was found at the other Cerro San Antonio, near Guichina.

After another 110 m, we observed several more mounds group around patios, Group H

(Figure 6.92). Due to time limitations, we could only map one of these mounds. Structure H1 measures 4 m by 16.8 m and lies on the north side of a patio. There is a clear entrance on the patio-facing side of the building, with two pillars on either side. The wall remnants still standing present a rather distinct construction style not observed elsewhere at the site so far: a mix of small to medium-sized stones set into an earth/adobe matrix (Figure 6.93). In what would be the middle of the standing wall there is an apparent line of long rectangular stones oriented sideways, creating a horizontal division between the upper and lower stones. Some of the larger stones seem to have been shaped somewhat, and at least three seem to have been given a rough triangular shape, although this has yet to be confirmed. While my colleagues mapped Structure

H1, I continued exploring west about 10 – 20 m and encountered yet another patio surrounded by mounds. During our brief survey in 2013 we had observed a major temple that had been constructed on a massive bedrock outcropping that probably had been partially modified. We estimated that this temple measures about 15-20 m in height and was large enough to host two front and two back rooms on top. King described in her notes that one of these rooms had the remnants of columns. A large plaza was located to the west of this temple, and there was a sizable altar in its center (Figure 6.94). Likely, Structure H1 is this same room with the columns described by King, and that we had reached as far as the summit of the same temple we had seen

462 in 2013. Of the many possible TPA groups we observed at the site, this would be the largest and easily the most impressive.

Figure 6.92: Map of Group H, which has several buildings that were not mapped. Group H may be superstructures on top of a large temple.

463

Figure 6.93: Standing wall of Structure H1. This wall features shaped stones set into a matrix of some type.

464

Figure 6.94: A large temple and plaza observed in 2013 at the western extent of the ridgeline. Group H may be the superstructures on top of this temple (photos by Stacie M. King).

465

Towards the end of our survey we found a looter’s pit that appears to be old, and in our conversations with the local men they confirmed that it was probably dug before their grandparent’s time (hace muchos muchos años, many many years ago). The pit was located in the northeast corner of a patio or room. Upon further inspection, we discovered what appeared to be a large horizontal stone, on top of which we could see construction fill of mixed stone and earth. The pit measures some 50 cm in height from the inferior edge of the horizontal stone to the bottom of the pit, but there is a lot of leaf litter and dirt, so it may have been larger in the past. How far back the pit extends beneath the structure is unknown at this point; while it was possible to crawl in, we choose not to do so since there may be animals living inside. We suspect that this may have been a tomb based on the following evidence: the location of the pit in what would have been a structure with restricted access; the structure itself is on top of a massive temple that is located in a large ceremonial complex with restricted access; and what appears to be a lintel stone roof, a common feature of underground tombs in Oaxaca. Obviously, future investigation is needed to determine what this pit represents and what, if anything, may have been looted from here. We marked the pit with a GPS point but did not map it officially in order to protect it from any further looting.

466

6.4.2.1 The Ballcourt

Figure 6.95: 3D model of the ballcourt at La Baeza.

Figure 6.96: View of the La Baeza ballcourt from the east and looking west along the ridgeline.

467

The ballcourt at La Baeza (Figures 6.95 and 6.96) is located 1900 meters above sea level on a saddle between two small hill peaks. It is part of the Group D structure complex that includes several monumental structures and platform-terraces. After leaving the larger saddle or llano (plain) where we mapped Group B, we skirted south around a small peak and descended down to the next saddle. The first structure one immediately sees is the ballcourt, Structure D1, as it lays directly at the end of the descent. The space around the ballcourt has been clearly flattened on the north, south, and west sides. Of course, the flattened area on the south is the large plaza (described in the previous section). The flattened area north of the ballcourt is much narrower, and is boarded on its northern side by a terrace retention wall that also marks the edge of the saddle as the terrain slopes down. Likely this space served as a passageway around the court as well as access to the east and west sides of the saddle. The rest of the complex,

Structures D2 – D5, are all located south and southwest of the ballcourt. As far as we can tell, there are no structures to the east, west, or north. However, we did not explore these areas as intensely, and it is possible that there is some architecture on the small peak lying just to the east and north. Further investigation in the future will help us to assess the presence of architecture in these areas.

Table 6.23: Size dimensions for the La Baeza ballcourt and playing surfaces. Overall Central Playing Surface Terminal Alleys East West Length 44.9 25.0 19.3 15.8 Width 26.1 7.1 6.3 5.4 Note: all dimensions in m

The ballcourt at La Baeza is an enclosed, sunken I-shaped court that measures a total 45 meters in length and 26.1 m wide (Table 6.23). According to the locals, the court is called the

“Laguna” because it fills with water during the rainy season. It is oriented more or less east-west

468

(84° E of W), like the courts at Cerro Maluco and Mogotes de Panteón. The playing alley itself is 25 m long by 4.4 m wide. The east terminal zone measures about 19.3 m long by 6.3 m wide; the west terminal zone is 16.8 m long and 5.4 m wide. The north lateral mound measures 25.2 m long, 10.5 m wide, and 2.5 m high with an estimated volume of 293 m3 (Table 6.24) The north mound is larger and taller than the south lateral mound (see Figure 6.95), which measures 1.7 m high, 10.2 m wide and 26.4 m long, with an estimated volume of 240.1 m3. The average slope of the lateral mounds is 15° (9° - 19°). Both were constructed using a mixed earth-and-stone matrix that was then covered in larger stone blocks, some roughly-shaped while others were faced.

Today, many of these stones are out of place. Preservation of the south mound is better than the north, the latter having suffered damage particularly on its northern exterior.

Table 6.24: Size dimensions for La Baeza court mounds. Orientation Type Length Width Height 1 Height 2 Volume Mound 1 north lateral 25.2 10.5 2.5 2.5 293.0 Mound 2 south lateral 27.5 10.2 2.4 2.4 240.1 Mound 3 east terminal 24.1 3.8 0.5 0.6 48.7 Mound 4 West terminal 21.0 5.9 1.3 1 85.0 Note: all dimensions in m.

The east terminal mound rises 1.3 m above the playing surface. Similar to the east mound at Mogotes del Panteón, the interior of the mound here is lined with stones, and on its exterior only rises slightly above the ground surface. The east mound measures 24.1 m long and

3.8 m wide, with an estimated volume of 48.7 m3. The west terminal mound is much larger in both mass and height, measuring 1.3 m on the interior and 1 m on the exterior, 21 m long by 5.9 m wide, with an estimated volume of 85 m3. The west mound is an earthen construction with stones, some of which are shaped, placed on top. Today, many of the blocks have fallen out of place and at this time it is difficult to ascertain how it might have appeared in the past. There

469 appears to be a small set of stairs on the southwest side of the west mound, and in fact the height of the mound appears to be smaller on the south end than on the north – indicating that access to the court may have been from this side (Figure 6.97).

Figure 6.97: View of the ballcourt from the west, with the south plaza visible on the right. A possible entrance may have been on the southwest side.

6.4.3 Summary: La Baeza and the Prehispanic Communities of San Carlos Yautepec (So Far)

During our brief survey in 2016, we were able to map about 86 structures ranging from terraces and platforms to room structures, foundations, and standing buildings. In our earlier

2013 visit, we covered more of the northern side of the ridgeline on our return to San Carlos La

Baeza, and noted additional structures, walls, room foundations, and TPAs we did not come

470 across again in the 2016 survey and mapping. During the 2015 field season, I observed other walls and room foundations that I did not see again in 2016 because we took a different route from Group D to Group E, approaching from the south the year prior rather than the north. We also know that there is another architectural group south/southeast of Group D, a large structure with mounds that surround a sizable patio first observed in 2013, though not seen again in the

2016 survey. Even during the last survey we saw more structures than it was possible to map in the time allotted, including other mound foundations, patios, and a large flattened space bordered by a one course terrace retention wall (Figure 6.98). This feature, possibly an immense plaza, can actually be seen using satellite imagery software such as Google Earth. It directs movement to the hill above where we find Groups G and H, as well as the large doubled-room temple described earlier. Without a doubt, there are many more structures on the hilltops and sides of the Cerro San Antonio. The men from the village informed me as much, commenting that additional occupation zones are located even further west of where we had ended our survey, possibly extending onto another saddle and peak. The areas to the north and south of the main ridgeline have yet to be explored.

I would argue that what we observed in 2013 and 2015, and what we were able to later map in 2016 probably represents the urban core of the site, given the multiple ceremonial complexes and the medium- to high-status residencies. Future surveys are planned in order to record those structures we have previously observed as well as determine if there are additional neighborhoods, outlying occupational areas, and/or small affiliated sites around the perimeter or nearby the main occupation. Satellite images certainly suggest the presence of additional areas of occupation on the hillslopes and piedmont extensions of the Cerro San Antonio. Indeed, one

471 of the most interesting features is a possible modified path that leads to a cliff-face where there might be caves.

Figure 6.98: A large plaza bordered by a terrace wall, looking north. Structure Groups F - H lie on the hilltop to the left.

La Baeza clearly stands out as an archaeological site for several reasons. The main core of the site – at least what we were able to map – is distributed over about a kilometer and a half of narrow ridgeline that runs east-west. At times, the amount of available flat terrain is only 10-

20 m across. Yet, even in these narrow confines we find architecture, including patios and room foundations. The distribution of the architecture across the ridgeline suggests that movement through the site’s core from east to west was tightly controlled and directed, with restriction in

472 access seeming to increase as one moves higher in elevation. There are several large public spaces that were likely used for communal gatherings, with the audience probably composed mostly of local elites and invited members of the community, and on some occasions even other high-status guests from surrounding settlements. These public spaces include Group B, Group

D, and the immense flattened area (see Figure 6.91). Group B, consisting of a few platforms with superstructures and a possible temple (Structure B3) and plaza, was likely the most accessible, and would have included a sizable and mixed audience.

Group D, which includes the ballcourt, has a much smaller gathering space and is located after a small hill knob. People would have been observed descending down to this structure group from both the east and west sides. Arguably, the western end of the ridgeline was occupied and/or used by higher status elites and the eastern end by lower status elites or other important community members. It is interesting that the ballcourt lies on a saddle that is both between two hill knobs and these two distinct areas of the site. In fact, Structure D5, the platform with room structures on top, may have been an elite residence. The ballcourt would have been “attached” to this residence and therefore the prerogative of the people who lived there, a common pattern observed in the Central Valleys during the Late Classic period (Dante

Ríos García, personal communication). Finally, the immense flattened area was probably the most restricted in access as a gathering space, despite its size. We find this flat area after Groups

D and E, the latter of which I have argued is likely a high-status residence given its size and complexity. As previously mentioned, the rise seen behind the plaza in Figure 6.91 leads to

Groups F, G, and H where the ridgeline narrows precipitously, and movement is only permitted by ascending platforms. Group H, of course, are the rooms possibly on top of the great temple that forms the largest TPA group in the whole site. Given the restricted access, flow of

473 movement, and the monumentality of this part of the site, the western section of the ridgeline was probably exclusive used by the higher-status elites, likely priests or other similar social functionaries.

The architecture is distinct at La Baeza and varies throughout the site. Very few of the monumental buildings repeat the same exact construction techniques, although the technique of using larger and roughly shaped stones with smaller stones set in between as the foundations of buildings is repeated throughout the site. This technique has been observed in the Central

Valleys (Late Classic period), Mixeca Alta, and at some sites in Nejapa (Early to Middle

Postclassic). Other construction techniques observed at the site are similar to Quiechapa to the southwest. Somewhat unique to the site are the room foundations made with the thin La Baeza- style planed blocks and/or medium-sized roughly squared stones, both of various grades and colors. The stones, rocks, and boulders used in the construction of La Baeza’s structures must have been obtained nearby, and in future surveys I hope to locate a quarry (or quarries, as we have examples of chert as well). Some of the structure walls were constructed using an earthen matrix set in between the courses of stones. Perhaps most surprising, at a few of the structures we observed short, standing earthen brick walls that may also possibly be adobe. Their presence speaks to the remarkable preservation present at the site. One of the best examples is a small remnant of the Structure A1 temple wall, which interestingly has some faced stones set into it.

There is a substantial density of ceramics on the surface throughout, again indicating intensive occupation here. Much of what we found was non-diagnostic plain gray ware, likely from serving bowls and other similar type vessels. However, the diagnostic ware we did observe, including a few G.3 bowl rims, suggest that La Baeza was occupied from the Late

Classic to Early Postclassic (estimated date between CE 700-1200). We also found utilitarian

474 ware such as coarse orange/brown paste jars, including a massive jar handle. One of the most interesting sherds we found was made of thick gray ware paste and had a fabric impression. We also found several supports and a possible handle. The lithic pieces observed on the surface were especially interesting. Sizable chunks of chert nodules were found throughout the site, indicating that lithic production may have been a major source of economic activity at La Baeza.

The most surprising finds were a green obsidian microblade and the bottom half of an arrow point crafted from white chert. At Structure E1, we found two different stone tenons that would have been placed into the corners or walls of the superstructures. Similar stones were found at El

Órgano, a neighborhood in Greater La Amontonada, and Nejapa Viejo, although these latter examples had been decorated with faces (see Appendix A; Figures 23-33).

Without excavation and confirmed dates for occupation, it is difficult to assess the relationship between La Baeza and the communities tentatively identified so far in the municipality of San Carlos Yautepec as well as those in Nejapa, if such relationships even existed. The settlement’s location far from the major concentration of occupation in Nejapa makes it an outlier in the study area, and strongly suggests that it belonged to a different political system altogether. Moreover, the mountain on which the settlement was built stands out on the landscape. La Baeza therefore appears to be a unique site, given its strategic geographic position at the possible boundaries between different Late Classic and Postclassic quechecoquies, as well as the numerous examples of distinctive and monumental architecture, some examples of which have been not observed at the other sites nearby so far. One such structure is the ballcourt, which appears to be located between two distinct residential occupations of the site. Currently, this is the only one recorded so far in the San Carlos Yautepec area, but there could likely be more at other archaeological sites that have yet to be documented and registered (as evidenced

475 by Rickard’s photographs from the early 1900s). At the same time, La Baeza does share some architectural features and/or construction techniques with other sites in Nejapa, Quiechapa to the southwest, and possibly even with communities in the Central Valleys and further north. It is possible that La Baeza was either a leading center during the Late Classic into the Early

Postclassic in this area of the Sierra Sur, or it may have served some other sort of influential role

– perhaps as a religious site – given the number and variety of monumental structures.

6.5 CASE STUDIES – BALLCOURT EXCAVATIONS During the 2013 field season, we conducted excavations in two of the ballcourts, at the sites of El Sitial and Cerro El Convento respectively. Both are mountaintop sites on the southern and eastern borders of the Nejapa subregion. Although I was unable to map these courts due to concerns over the rainy season (which makes many of the mountain sites inaccessible) and several subsequent forest fires that occurred, the data we collected in 2013 provides useful information about the construction and use of each court. In the following section, I describe the excavation data and offer an interpretation of the results.

6.5.1 El Sitial (YAU-TAV-TAV-010) – Municipality of Santa Ana Tavela

El Sitial (YAU-TAV-TAV-010) is a large, complex site located in the piedmont region south of the modern town of Santa Ana Tavela (Figure 6.99). The site was first surveyed and registered in 2009, and PANT conducted excavations there in 2013. The ballcourt is located on a saddle between two large peaks and is easily one of the best conserved courts in the Nejapa subregion. Like most of the structures in the sample, the Main Ballcourt is an enclosed, sunken

I-shaped court oriented 111° (E-SE), similar to only five other courts in the study. Stairs permitting access into the court and onto the mounds are located on the north and south lateral

476 mounds, as well as the southeast corner of the west terminal mound and on the north side of the east terminal mound. To the west and north of the court there is a possible adoratorio (explored in Operation C), while on the exterior of the southeast corner there is another, larger structure whose purpose is unknown at the moment, but interestingly is similar to the rectangular structure attached to the southeast corner of the La Baeza ballcourt.

Figure 6.99: Map of the El Sitial site and ballcourt.

Overall, the court measures roughly 45 m long by 25 m wide, and the central playing alley is 21 m long by 5 m wide. The east terminal zone measures about 19 m long by 5 m wide, while the west terminal zone is 17 m long by 5 m wide (Table 6.25). It is difficult to calculate the east and west terminal mounds based on the spatial data available, but the north lateral

477 mound measures about 21 long by 10 m wide and the south lateral mound about 22 m long by 10 m wide. We do not have not have any measurements for the heights of the court mounds with the exception of the west lateral mound, which measures roughly over 1.5 m on its interior. With the exception of the exterior height of the same mound, all the other mounds do not have exterior mounds higher than about half a meter.

Table 6.25: Size dimensions of the El Sitial court mounds. Overall Central Playing Surface Terminal Alleys North South Length 45 21 19 17 Width 25 5 5 5 Note: all dimensions in m

I directed the excavations at Operation D, (Op. D) located on the interior of the west terminal mound of the ballcourt from March 20 – 26, 2013 (Figure 6.100). In the 2009 survey, the PANT crew had observed a sizable drainage gutter located just slightly outside and down

(south) from the exterior of the wester terminal mound (Figure 6.101). While the interior of the west mound conforms to the terminal contour of the I-shape, the exterior is shaped like a half- moon on its southern end, with the drain outlet in the center of the curve. Here the ground surface levels out for a few meters before meeting the down slope of the hillside. The goals for

Op. D were first, to locate the drain opening and second, observe the methods of construction and use history of the ballcourt. After exploring the drain outlet to see in which direction the gutter traveled, we set up the 2 x 2 m test pit slightly to the right of center on the interior slope of the west terminal mound, as the drain appeared straight but deviates at a slight angle to the north as it moves west to east. Measuring the drain as best as possible, we estimated that its total length is about 5.5 m from west to east and placed the test pit accordingly.

478

Figure 6.100: Operation D excavations at the El Sitial ballcourt.

479

Figure 6.101: The exterior end of the drain on the southwest side of the west terminal mound. The drain exterior was constructed in a half-moon shape.

The ground surface, or Layer 1, was a dark gray, loose fine soil (10 YR 5/1) with lots of roots and small stones mixed in. Layer 2 was a much sandier soil, so fine grained that it was almost powder-like, with only a few rock inclusions. The soil in this layer was mostly loose, although compact in parts, and was a whitish-pink in color (7.5 YR 8/2). In this layer, we found rock fall from the interior wall of the mound, and it was this same type of soil that was filling the drain once we had located it. Based on our observations, we concluded that Layer 2 was a result of soil collapse and accumulation after the ballcourt had been abandoned or ceased to be used.

Layer 3 was only observed in the northern, eastern, and southern profiles of the northeast and southwest quadrants. While the texture of the soil was the same, the soil color was a light

480 brown (10 YR 8/2). There was a lot of disturbance due to tree roots in this layer. I would argue that this layer represents a period immediately after the court ceased to be actively used, as this layer appears only in the interior of the court and it is where there was soil mixture likely due to water action after past rain storms. The fourth layer is similar to Layer 2, a whitish-pink (7.5 YR

8/2) sandy soil with some stone inclusions ranging between 5 mm to 5 cm in size. However,

Layer 4 is much more compact and leveled, and it is only seen in the northeast quadrant and eastern profile of the unit. It is possible that this was the original floor of the court, which would indicate that the playing surfaces were compacted earthen floors, rather than stucco. Towards the end of the operation we observed Layer 5, a very compact and hard brown soil (10 YR 7/4).

This soil was encountered in different areas of the unit, including behind and below the stones in the southwest quadrant (at the top of the mound wall); above and behind the stone lintel that roofed the drain opening; in front of the drain; and the floor of the drain itself. Layer 5 was probably the original ground surface before the construction of the court and would have been used in creating the interior fill of the west mound.

The west wall of the ballcourt was not well preserved, and it was clear based on their positioning that some of the stone rows had fallen or slumped over at some point in the past

(Figure 6.102). Based on the size and position of the larger stones that had formed the base of the wall, we suspect that these stones fell first, and by that action the middle and top wall stones had fallen behind the base stones, while others had been thrown forward. After the events that caused the initial destruction to the wall, it continued to decompose over time. Although we cannot be completely certain about how the wall originally appeared at this time, based on the morphology of the east wall, it is likely that the interior walls of the terminal mounds were not as sloped as those of other courts in the Nejapa subregion.

481

Figure 6.102: Operation D - the drain apreture was located in the upper right quadrant. The opening had been covered by wall slump.

Towards the end of the excavation, we finally located the drain opening in the northeast corner of the unit after removing some of the fallen or slumped-over wall stones (Figures 6.103).

The lintel stone that formed the roof of the drain was similar in appearance to the lintel stone observed at the drain outlet – planed on the inferior edge and curved on the superior surface, but more or less rectangular in form. The lintel stone measures 65 cm in length and 32 cm at its widest point. The roof of the drain rests on two small walls, only one stone across and 3 stones deep, with both walls measuring approximately 45 cm in height. The wall stones of the drain vary in size, but all are more or less rectangular. Behind the lintel stone the roof continued as a

482 series of similarly sized and shaped stones, based on best estimates. It is possible to infer that the entire roof of the drain consists of these large flat stones that are roughly 60 cm in length and 30-

40 cm in height, and that the drain walls were constructed using similar faced stones of various sizes as we observed at the opening. In total, the opening of the drain measures 60 cm in height, from the inferior edge of the lintel roof to the drain floor, and 38 cm in width, from the north to the south wall. The floor of the drain is 1.76 cm below the top of the west terminal mound. The soil around the drain is compact and orange in color, while inside the drain there was a lot of infill with soil from Layer 2. Cleaning out the drain, we did not find many artifacts except for a few pieces of carbon, which gave a calibrated date of CE 1412 – 1619 (with an 88% probability of dating between CE 1412 – 1518). This would indicate that the ballcourt was probably last occupied sometime in the Late Postclassic, although the court itself is maybe older.

483

Figure 6.103: The drain aperture after the wall slump had been removed and the interior of the drain cleaned out.

484

Based on the available evidence, we propose that the residents of El Sitial took advantage of a natural depression or drainage on the hill in order to construct the ballcourt, using the dirt excavated during the construction of the I-shape to build the west terminal mound around the drain. In this way, they used the natural slope of the terrain to drain the court. After constructing the west mound, the interior was faced with irregularly shaped stones, the largest placed at the bottom of the wall with much smaller stones towards the top. The wall was constructed without using a cement or earthen matrix. While it is possible that the exterior of the mound was also faced with stones, we did not observe any stones on the surface, in contrast to the interior. We also did not find any evidence of stucco, neither in the excavation unit nor on the surface – even though stucco has been observed elsewhere at the site. Therefore, we suspect that the floor was indeed a compacted, earthen-clay and pebble mix, and that the walls of the court had been faced with irregular stones (but not decorated with stucco). Aside from the lack of stucco, there is also no evidence for modification or reconstruction episodes within the interior of the court, although with a compact earthen floor such modifications may not have been necessary or observable.

After abandonment, the ballcourt began to refill naturally, and at some other point the interior wall of the west terminal mound slumped forward and collapsed.

A second excavation unit was placed at the northwest corner of the north lateral mound on its exterior side, Operation E (Op. E). On this side of the ballcourt a second platform had been constructed with two sets of stairs located equidistant from the east and west ends and permitting access to the top of the north mound. Because the ground surface outside of the ballcourt inclines downward from west to east, the eastern stairs are much larger than those at the opposite end. A terrace retention wall corners with the north lateral mound at its western end, extending northeast from the ballcourt. The terrace does not run perpendicular to the ballcourt

485 but was constructed at an angle to the backwall of the north lateral structure, perhaps oriented to the contours of the hillside (Figure 6.104).

Figure 6.104: Operation E - excavation exposed a terrace wall that connects with the back wall of the lateral structure on the north lateral mound.

Op. E was placed directly in front of the western stairs where the architecture was best preserved. Given the position of the terrace with respect to the exterior of the north lateral mound, we suspected that there might be a midden in this corner. The excavations revealed that there had been three construction episodes on this side of the court. Similar in many respects to architectural techniques observed elsewhere at public structures in Nejapa (such as La Baeza and

El Órgano), the upper wall or cornice as well as the exterior of the north lateral mound was

486 constructed using large, vertically-placed stones. Smaller stones were then used to fill the spaces in between. The exterior of the mound originally featured a 125-cm wide three-step staircase of flat stones, with each step measuring 20 cm in height. Several stones were later added on top of the staircase, shortening the stairs to 65 cm. However, the added stones are much smaller than the basal foundation of the original platform, leaving the line dividing the two staircases visible from the surface. Excavations also showed that the contoured terrace was constructed before the ballcourt, with the terrace stones set deeper into the soil and extending into and under the north lateral mound. The terrace wall was constructed with two or three courses of irregularly-sized stones. A substantial number of ceramic sherds and fragments from large vessels and incensarios was collected in the test pit, confirming the presence of a midden. Some of the identified ceramics include miniature sahumadores, large jars, apaxtles (big round jar with small, narrow opening used for water), a comal, and a patojo (a type of vessel used ethnographically for cooking beans). The evidence clearly demonstrates that feasting and other ritual activities did take place near and perhaps even inside the court.

What is most striking about the ballcourt at El Sitial is that of all the courts in the sample, this one most closely resembles the Zapotec-style court design in the Central Valleys (Figure

6.105). There are several features seen here that are not present or observed elsewhere in

Nejapa, such as upper walls or cornices, lateral structures, back walls, benches, and bench walls.

These features are commonly found on Central Valley courts. The significance of this distinct court design is not clear at the moment but may point to social connections to Central Valley

Zapotecs. In contrast, where we do find structures on court mounds, these are located on the terminal structures, not the laterals. Another striking difference is that the exterior heights of the lateral mounds are much smaller than other lateral mounds in our sample, likely an artifact of the

487 mountain’s topology and how the court was constructed. Finally, the terminal mounds are not much smaller in height than the lateral mounds with respect to their overall height, unlike all the other I-shaped courts in the Nejapa subregion.

Figure 6.105: The El Sitial ballcourt has many of the same architectural features as courts in the Central Valleys.

During the 2013 field season, while scouting out a place to set up camp for our 10-day excavation at El Sitial, we located a second, much less formal ballcourt19 constructed on a terrace that marks the western access point to the main settlement further up on the hilltop. The court has the distinctive rectangular shape of a palangana court. The mounds of the court were built

19 Unfortunately, there are no clear photos of this ballcourt. 488 using a single file of large, rough stones that had been put together in a very informal manner. It is possible that the whole court was constructed by simply excavating out the space for the central alley, then lining the edges with the large stones. Unlike the palangana court at Llano las

Casas, no defined mounds are present. We did observe other structures near the court, but they are much less formal and monumental than those found on the summit. The court was likely part of the ceremonial precinct of a smaller neighborhood associated with the El Sitial settlement.

Unfortunately, apart from these observations, not much else is known about this court.

6.5.2 Municipality of San Juan Lajarcía

The municipality of San Juan Lajarcía lies in the eastern portion of the Yautepec distract, with the Highway 190 bisecting the municipal territory in half (Figure 6.106). In total, Lajarcía includes 160 km2 of land and around 700 inhabitants. Survey was conducted here in 2011 and

2013. The PANT survey crew recorded 18 sites here, or about 11% of the total sites registered.

I have previously described Cerro La Virgen, one of the sites lying only a kilometer and a half from Llano las Casas. Perhaps what is most interesting about Cerro la Virgen is that it is located on a ridgeline running between Lajarcía and Tavela. Again, it is likely that this site was used for cultivation by the inhabitants of Peña del Rodeo (in Tavela territory). Aside from the impressive

Cerro del Covento, several other habitational sites were recorded in the Lajarcía municipality, including a settlement on the western flank of the Cerro Gavilán (YAU-SJL-SJL-014) where the

PANT crew recoded 31 terraces and 25 structure foundations along with various concentrations of ceramic, lithic, and groundstone artifacts. The most interesting features at this site are three small, rounded rooms measuring 1.5 m in diameter and 1 m in height, constructed below a large stone that juts out from a rock shelter on the southeast side of the hill. The rooms on the eastern

489 side of this stone appear to have been used for storage, based on their size and similar modern ethnographic examples. Several other features in addition to the storage rooms suggest that

Cerro del Gavilán was a Postclassic farming community. At the summit of the site there is a broad plain where one can see the entire valley area of Nejapa as well as south towards the Cerro

Manteca, a prominent mountain near Guichina in the San Carlos Yautepec municipality.

Figure 6.106: Map of the San Juan Lajarcía municipality and some of the sites documented there.

Another habitational site is Cerro Verde (YAU-SJL-SJL-003), which features terraces, possible structure foundations, and two stone mounds that possibly could be graves or tombs on the western hill peak, while the eastern hill peak is bare. Paraje el Algarrobal (YAU-SJL-

SJL017) is a settlement that lies only 850 m to the north of Cerro del Convento and may be

490 related to the latter. The PANT survey crew observed at least four architectural groupings, several stone terraces, and various concentrations of ceramic materials. North of one of these structure groups is another large, squared structure defined by stone walls, measuring 6.4 m in length by 4 m in width and 30 cm high, and featuring stone and earth rubble mounds that suggest divisions of the interior space as well as two small stone rubble mounds at the entrance that were likely columns. These features suggest that this structure was a temple.

There are also several ritual sites located near many of these same settlements. Mano de

Gente (YAU-SJL-SJL-001), for example, is probably a ritual site located only 1.3 km west of

Cerro Verde (and 2.7 km northeast of Cerro del Convento), given the presence of red-hands painted in negative form on a wall. Southwest of Mano de Gente by 600 m is Portillo de Mano de Gente (YAU-SJL-SJL-016), which consists of two adjacent stone foundations measuring approximately 2 m by 2 m. Given the large stone located on the south side of the north stone foundation that appears to be rubble from the structure that would have been built on top, and that the north side of the foundation is open, this may have been an adoratorio.

Finally, about 3 km north of SJL, very close to where the highway is now, is a possible pilgrimage site on the Cerro del Sauso (YAU-SJL-SJL-012). At the ritual site of Monte Grande

(YAU-SJL-SJL-009), the survey crew found six stone foundations on a flattened peak of a local mountain as well as offerings and incensarios.

491

6.5.2.1 Cerro del Convento (YAU-SJL-SJL-002)

Figure 6.107: Map of the Cerro del Convento site and ballcourt.

The archaeological ruins of Cerro del Convento (YAU-SJL-SJL-002) lie on the top of one of the most visible mountains separating the Nejapa subregion from the Isthmus (Figures

6.107). More of a tall mesa than a mountain, the surrounding 100 m high cliff faces consist of a white-yellow limestone that make it one of the most striking and visible geographic features in the area, and it is one of the first sights you notice from the moment you enter the valley from the north on the highway (Figure 6.108). The site is located right along the Colonial-era camino real or royal road that traveled between the Central Valleys and the Isthmus of Tehuantepec. Today the modern Highway 190 passes by on the left just before arriving at the Portillo de Nejapa.

492

Cerro del Convento figures prominently in the oral histories of many of the local people in

Nejapa, with many tales abounding about the ruins, the caves that dot the cliff faces of the mesa, and the supposed supernatural powers that are still attributed to the site. It was first described by

Fray Francisco de Burgoa in 1674 as a Zapotec fortress in the mountains around Nejapa, and was assumed to have been constructed during the Zaachileño conquest campaigns towards the

Isthmus (Burgoa 1934; Wallrath 1976).

Figure 6.108: The limestone cliffs of Cerro del Convento are a prominent feature that can be seen from multiple viewpoints in the landscape.

Burgoa reported that located on the mesa top was the tomb of a famous Zapotec warrior, supposedly the nephew of the Coqui Cosijoeza. This tomb was also a pilgrimage site for local

493 peoples where they would bring their “idols” (Burgoa 1934:241-248). During the Spanish

Colonial era, the caves were utilized as secret sites for rituals and sacrifices conducted by locals hoping to escape Catholic oppression. Unfortunately, a Dominican friar by the name of Juan

Ruiz, who was in charge of disseminating Church doctrine to the indigenous communities, took it upon himself to end these idolatrous activities and destroyed all the religious items as well as the tomb itself in a two-day period (Burgoa 1934:241-248). Despite the efforts of the overzealous priest, Cerro del Convento continued to be visited as a sacred space.

Despite the abundance of stories, both oral and recorded, about Cerro del Convento, there was no initially no official information on exactly where the site was located. The mystery was solved when King (2010) found a report buried within the inedited papers of Manuel Martínez

Gracida (stored in the archives of the Biblioteca Central del Estado de Oaxaca), written by

Rafael Isaac Ramírez in 1892. In his report, Ramírez (1892) identified the site described by

Burgoa as “Peña del Convento”, helpfully including several drawings and a map with measurements of the architecture, along with detailed descriptions of the site. In a later report

Martínez Gracida (1910c) himself wrote a description of the Convento caves in a document about the various famous caves and grottos of Oaxaca. With this information in hand, King and the PANT crew were able to arrange a visit with local officials to Cerro del Convento in 2009.

During this visit, they identified several Prehispanic mounds and a ballcourt on the top of the mesa, as well as incensario fragments, rock-wall paintings, and lithic scatters. The site was subsequently surveyed and mapped in 2011. The survey crew observed that the mesa is naturally fortified on three sides by the tall cliff faces. The only access to the top are through small paths

(including one that can only be used during the dry season, as it follows a seasonal arroyo) on the

494 northeast and south sides that would have only been known to inhabitants or frequent visitors to

Convento (Figure 6.109).

Figure 6.109: The arroyo is a controlled access point to the site, making Cerro del Convento naturally defensible.

495

Most of the surviving architecture is concentrated on the east side of the mesa, on top and in front of the tallest point, and where the soil is best conserved (the north, west, and south zones having suffered from erosional and agricultural activities in the past). The structures they observed included residences and ceremonial buildings, with both square and rectangle-shaped stone foundations dispersed throughout this zone. Ceremonial buildings other than the ballcourt include a collapsed mound that is more or less circular, measuring 18 m in diameter and 2 m in height; and a 40-cm high platform with a 5-m tall superstructure on top. There are also smaller structures that appear to be organized in groups and arranged around small patios, likely households.

South of the structures lies the highest point of the site, a small hill where you can see terrace walls as well as another architectural compound with rectangular structures, and an irregularly shaped structure with evidence of recent heavy looting. King argues that the tomb of the Zapotec warrior, the nephew of Cosijoeza, may have been actually been located here. To the southeast of the compound there is a defined entrance with a small stair measuring 20 cm high and 2.2 m long. Not far from here is the arroyo that descends down this side, effectively dividing the southern section from the rest of the site. If you continue to follow this path, you will arrive at the summit of the cliff faces and enter a small rock shelter that has been reinforced with walls and terraces, forming a path that takes you to the caves. The path itself is intersected with small dividing walls, stopping at a zone where the pathway thins out to the extent that mountaineering equipment is needed to pass. From this point, it is possible to see a 4 or 5 m tall wall that had been constructed between the exposed bedrock of the cliff face and would have served to form and reinforce both the path and access to the caves.

496

Access to the mesa is on the south side, and the path passes three large, extended walls that Ramírez described as trenches in 1892. Some of the walls have stairs on the interior sides, with heights on the exterior measuring over 2 m tall. The tallest wall measured 45 m long, but it could have been longer as the crew was unable to measure the northern extreme; the second wall is 90 m long, and the third is 82 m in length. The walls controlled access on this side of the mesa where it was most vulnerable, providing protection to the site on top. The terraces behind the walls lack evidence of ceramic fragments and house foundations, and so may have been constructed for cultivation and/or controlling movement to and from the site.

King proposes that the Cerro del Convento mesa was a place where the local people of

Nejapa could directly meet with the various foreigners that passed through the subregion, including the Zapotecs, Aztecs, and eventually even the Spanish, given its high visibility and prominent location in the valley. The ethnohistorical sources describe the various ethnic groups as frequently traveling along the camino real in the 15th and 16th centuries. Again, according to

Burgoa, Cerro del Convento was established by the Zaachileños as a fortress in order to have a secure, defensive point where they could observe and quickly respond to Aztec military advances. However, the evidence we collected suggests a much longer use for the site, especially for the caves dotting the limestone cliffs where some of the material remains date to as early as the Archaic period.

497

6.5.2.2 The Ballcourt

Figure 6.110: Architectural drawing of the Cerro del Convento ballcourt.

In 2013, the PANT crew returned to Cerro del Convento to conduct excavations at some of the structures in the eastern zone of the site. Operation C (Op. C) was located in the estimated center of the ballcourt, and excavations of the 2 x 2 m test pit took place from February 8 to 18,

498

2013 (Figure 6.110). The goal of the excavations was to understand the construction and occupational history of the court as well as to see if there was a central ballcourt marker. The ballcourt is I-shaped and enclosed, and although a sunken court, the difference in height between the ground surface and the playing alley floor is not as great as observed at other sites, likely due to the toughness and shallowness of the surface soil. The court is oriented 7° W of N (353° N-

S), and in total measures 40 in length from north to south and 19 m wide from the eastern exterior to the west. The playing alley itself is 29 m long and 4 m wide. The north terminal mound is smaller than the south terminal mound, and it is only really observable by the sloped incline to the playing alley. Given the amount of overgrowth at the time, it was difficult to ascertain the complete dimensions of the south terminal mound.

Layer 1 represents the most recent ground surface, a very soft, dark gray (7.5 YR 4/1) and fine-grained soil with small stone inclusions. Ceramic sherds, including a fine-orange ware piece, lithics, and pieces of bajareque or daub were found in this layer. The debris suggests that at some point in the past, the court was used as a trash pit, likely after the structure was abandoned. The following layer was similar to the first, except less fine and with larger rock inclusions. The majority of the ceramic sherds found in this unit came from this layer. Although a darker gray color (7.5 YR 3/1) than the previous layers, Layer 3 was a very distinct from the first two as it was more pebbly and compact. Given that it is also consistently level throughout the entire unit, we identify this layer as the playing floor. The final layer was much sandier, possibly mixed with eroded material from the bed rock. Layer 4 is distinct in color, much more pinkish white (2.5 YR 8/2) than the other layers. No ceramic sherds were found in this layer, and we assume that this is the sterile level prior to the occupation of the mesa. At this point we terminated the excavation at 75 cm below the ground surface. Complicating our understanding

499 of the use-history of the court was an element that we found directly in the center of the test pit, most likely a hearth. The superior edge of the hearth was first encountered in Layer 2, and it continued through Layers 3 and 4. Stones ranging in size from 10 to 30 cm in length were arranged in a circle. In total, the hearth measures 45 cm at its highest point and 110 cm at its widest. Pieces of carbon and burned ceramics were found all around the hearth, and a substantial-sized lens of carbon was observed at its base. This suggests that hearth was used in the past, although we are not sure if it was a single episode or multiple events (Figure 6.111).

Figure 6.111: Excavation drawing showing the different layers as well as the hearth element found in the center of the unit.

Based on the evidence from the excavation, some preliminary conclusions can be made about the construction and form of the ballcourt. The inhabitants of Cerro del Convento likely selected this area of the mesa to build the court because the natural topography was mostly level here. Because we were not able to excavate the mounds themselves, we do not know precisely

500 how they were constructed and with what materials. Not many stones remained on the mound surfaces, although it is still entirely possible that they were constructed like many of the Nejapa courts with a mix of earth and stone that were then covered with much larger surface stones. It appears that the inhabitants of Convento used a compact earth, pebble, and sand matrix for the central playing surface. There was no evidence of multiple construction episodes or later modifications to the court, at least with respect to the central alley. However, as at El Sitial, there is the possibility that modifications may have been made to the mounds, but without further exploration we cannot be sure.

With respect to the hearth, we still do not understand the relationship between this feature and the ballcourt. It is interesting that the hearth is located directly in the center of the court. In several regions in Mesoamerica it was common to place a stone or other type of marker in this area of the court. A radiocarbon sample dates the use of the hearth to the Middle Classic period

(calibrated CE 430 – 643). But according to radiocarbon dates taken from samples found in other excavations, structures on the mesa, including the ballcourt, were more likely constructed during the Early Postclassic (calibrated dates CE 1027 – 1264) or Middle to Late Postclassic (CE

1324 – 1612). Without a doubt, if the hearth and ballcourt were contemporaneous, the position and height of the former would likely have affected gameplay, perhaps even rendering the court inoperable. Feinman and Nicholas (DATE) reported an oven feature in the corner of the recently discovered ballcourt at Lambityeco, so there may be some precedent for this type of feature.

Perhaps the original hearth was used for rituals or fiestas during the Middle Classic, and through these activities had acquired a sacred significance, or the historical memory thereof. Then, perhaps much later when the inhabitants of Convento were deciding on a location for the ballcourt, they choose this location specifically because it was a sacred place, positioning the

501 center of the court over the hearth. It is also possible that there is no relationship between the hearth and the court, and their positioning with respect to each other is only a coincidence.

Without a doubt, further exploration in the future will allow us to reconstruct the sequence of events with more security.

6.6 A BALLCOURT IN THE CLOUDS - LOS PICACHOS (YAU-TAV-TAV-011)

Figure 6.112: Map of the Los Picachos site and associated settlements, including the Los Chorros site where the ballcourt is located.

Los Picachos, as described in Chapter 3, is located on the southeastern edge of the Santa

Ana Tavela municipal territory (Figure 6.112). The site itself sits at an astounding 2100 m above sea level, one of the highest elevated sites in the subregion. While the site is highly defensible

502 and access would have been tightly controlled, evidence also clearly shows that residents actively participated in both local and interregional grade networks (King and Zborover

2015:138). Due to the extremely high elevation at which the site is located, and the danger from nearby forest fires that field season, I was also unable to map the ballcourt at Los Picachos. The information presented here is therefore drawn from the 2009 and 2011 survey and excavation data. The site itself lies directly north and only 200 m above Colonial-era Majaltepec, and it is possible that residents from the former moved to the latter after the arrival of the Spanish.

Occupation extends for almost a kilometer and a half along the ridgeline, despite the fact that the mountain summit is no more than 50 m in width with almost vertical drops on all sides.

Architectural structures have been built on large terraces that cover the lateral sides of the summit, some with stone walls reaching over 3.5 m in height forming all three sides of the terrace. Traversing from one terrace to another is possible but somewhat difficult, as there are no stairways permitting easy movement, further pointing to the defensive nature of the site.

A complex of structures can be found at the highest part of the site, around 2150 m above sea level, and was probably the civic-ceremonial center. There the PANT survey crew documented two plazas surrounded by high walls, an adoratorio, and a patio with views towards the north end of the site. At the northern end of this restricted zone there are several high walls – one of which is over 6 m in height. Coupled with the extremely difficult topography of the summit, the walls suggest that this area was used specifically for defense. There also may be subterranean tombs in the possible civic-ceremonial center, pointing to its exclusiveness. The ceramics here are also much finer than those that were collected on the terraces below, which tend to be more utilitarian wares made from locally-sourced clays, and they demonstrate clear

503 trade relations with other regions of Oaxaca. Clearly, people of higher status lived higher on the summit while those of lower status lived on the terraces.

Technically the ballcourt of Los Picachos is actually located at Los Chorros (YAU-TAV-

TAV-029), which is just downslope of the former. However, given the proximity of the two sites to each other and the tendency for dispersed settlement at Oaxaca mountain sites, more than likely Los Chorros was part of the larger Los Picachos community. There are a series of terraces in the southwest sector of the site with several structures built on them. To the east there is another ridgeline with even more terraces. An isolated structure lies even further east of this.

Although simple in appearance, there is still a panoramic vista from this structure. Moving down and west, between the simple structure and some former miners’ cabins is the ballcourt.

According to the survey data, the court measures 30 meters in length, 10 m in width, and 2 m in height. It is not clear if the court is I-shaped, but based on the description and the measurements, it is possible that it is a palangana structure. Based on radiocarbon dates, Los Picachos dates to the Early Postclassic (calibrated 83.6% likely between 945-1047 AD), although if the court is indeed a palangana, there may also have been a Classic-period occupation.

6.7 DISSCUSION: A HISTORY OF THE BALLGAME IN NEJAPA

Overall, the majority of the ballcourts in the Nejapa subregion (n = 11 or 73% of the total; or n = 8 or 80% of the study sample) are enclosed, sunken I-shaped courts. In total, the ballcourts in Nejapa range in overall length from 30 m (Los Chorros) to 87.3 m (Cerro Maluco) and in overall width from 10 m (Los Charros) to 40.8 m (Cerro Maluco) (Figure 6.113). The central alley playing surfaces are obviously much smaller and narrower: lengths range from 21.5 m (El Sitial) to maximum of 40.4 m (Cerro Maluco), while in width the alleys range in size from

4.4 m (Lachixoba) to 9.5 m (Los Limares) (Figure 6.114). Each court has parallel lateral mounds

504 that are on an order of magnitude much larger than the attached U-shaped terminal mounds.

Many of the terminal mounds, primarily the north or east terminal mound, have exterior heights that were nearly level or are level with the surrounding terrain (Figure 6.115). Often this level space is a plaza or patio next to the court. While this stylistic choice is found in other regions of

Oaxaca, such as the ballcourt at Cerro la Campana in the Central Valleys and Guiengola in the eastern Isthmus, in Nejapa it is the standard – even among the non I-shaped courts. Both the lateral and terminal mounds feature medium-sloped20 aprons on their interiors, but only the exteriors of the lateral mounds are sloped, with the exception of the north lateral mound at Cerro

Maluco.

Table 6.26: The overall size and central alley playing surfaces dimensions of the Nejapa ballcourts. Range Minimum Maximum Overall Length 57.27 30.00 87.27 Overall Width 30.82 10.00 40.82 Central Length 18.88 21.50 40.38 Central Width 5.12 4.38 9.50 Note: all dimensions in m.

20 Neither vertical (90°) nor nearly horizontal (≤ 15°). 505

Figure 6.113: Bar graph showing the overall lengths and widths of the ballcourts in Nejapa.

Figure 6.114: Bar graph showing the central playing alley lengths and widths.

506

Figure 6.115: Terminal mounds are very small in Nejapa, and one mound is always nearly or at level with the ground surface.

The Nejapa courts lack the features often considered to be typical of the Oaxacan court types (Taladoire 1981), including low benches, stepped lateral interior slopes with lateral structures on top, and vertical back walls. Of course, these elements mostly occur in the Central

Valleys21, and even then they are not all present at every court. Where we do find additional

21 Even then, stepped lateral slopes and vertical back walls appear to occur more in courts the main Zimatlan Valley (where Monte Alban is located) and the Tlacolula Valley. 507 structural elements in Nejapa are on the terminal mounds, such as the platforms and descending patios of Cerro Maluco, the northwest corner square platform at El Cucharital, and similar looking features at Mogotes del Panteón. The one exception is El Sitial, which features more of the typical elements associated with Oaxacan court types and also has terminal mounds nearly as tall as the lateral mounds.

Based on in-field observations and excavations in two of the courts, it is clear that similar techniques were used to construct them (Figure 6.116). Lateral mounds were constructed using a mix of local earth, the same as the surrounding soil, and small to medium-sized stones. Some of the earthen matrix used for the interior fill of the lateral mounds likely came from material excavated to create the sunken playing surfaces, which are on average a meter below the surrounding ground surface. Larger-sized, roughly shaped (if at all) stones were then placed on the surface of the mounds. The terminal mounds, depending on their height, were constructed in a few different ways. Some were built just like the lateral mounds, with an earthen-stone matrix covered in roughly shaped surface stones. Others are simply a single course of stones rising only centimeters above the surrounding ground surface. At La Puerta, the only court where this occurs, the terminal ends were made by modifying the bedrock at either end. The stones used to cover the mound surfaces were locally sourced, obtained from the immediate area of the site.

Thus, ballcourts closer to the rivers on the valley floor were covered in more rounded, river cobble stones, while courts higher up in elevation feature much rougher and sharply angled stones. Shaped stones are used in some aspects of court construction, primarily to define corners and edges, or other specially marked features. Again, the exception is El Sitial, the only court where faced stones had been used to cover the mound surfaces.

508

Figure 6.116: Example of the construction methods used to build the court mounds in Nejapa. The west lateral mound at Terezona had been cut during road construction.

Without excavation in more of the courts, we cannot not say for certain if the construction techniques were the same for building the actual playing surfaces as they were for the court mounds. Excavations at Cerro del Convento and El Sitial revealed that both central playing surfaces were compacted earthen floors made from locally obtained soil, mixed with pebbles. This type of matrix would have created a hardened floor that would have also allowed a ball to bounce easily. It is more than likely that the floors are similar in the other ballcourts.

Interestingly, the court at Mogotes del Panteón is the only one where we found fragments of stucco in the interior, one of which clearly shows one layer laid on top another. This suggests that this floor at least had stucco and then repaired in the past. Playing surfaces at other courts may yet prove to have had stucco as well. Overall, the similarities in architectural form, construction techniques, and court sizes suggest if not a shared ballgame tradition, then perhaps a

509 common suite of localized practices that inhabitants drew on when they constructed their ballcourts. I evaluate this hypothesis in Chapters 7 and 8.

While several dates remain to be confirmed, based on the current evidence we can roughly outline a history of the ballgame in Nejapa, tracing its presence and importance through the construction of the courts and the settlement history of the subregion. The earliest court to be constructed was the structure at Los Mogotes de Panteón, sometime in the Late/Terminal

Formative or Early Classic (Figure 6.117). At over 52 m in overall length, this was a massive structure, especially for the time period. With its large lateral mounds, smaller to almost non- existent terminal mounds, and construction techniques, the ballcourt at MDP becomes the standard design for future I-shaped courts in the Nejapa subregion. The arrangement of the architecture around the court suggest that these structures were likely built as part of the same construction program. Given the likely low population settlement and the massiveness of the court and its surrounding structures, building the ceremonial complex may have required labor beyond the immediate community itself, involving work parties drawn from neighboring settlements. This would have made the ballcourt a supracollective monumental project.

510

Figure 6.117: Map of Late Formative and Early Classic sites and ballcourts.

Two other confirmed Late Formative/Early Classic sites are Colonia San Martin and

Nejapa Viejo. All three sites are located near the major rivers, with Mogotes de Panteón on a small extension of piedmont and the latter two on the valley floor near their confluence. Both of these are also located only 3 km away from Mogotes del Panteón over terrain that is relatively easy to traverse. While we are not certain of the extent of the Late Formative occupation at these sites, it can be reasonably assumed that Colonia San Martin was the largest of the three given that it is much older, beginning sometime in the Early Formative. It is interesting that the court was constructed at Mogotes del Panteón, rather than at the older settlement of Colonia San

Martin, or at Nejapa Viejo, a community that would come to dominate the region in the Classic.

511

Although we have evidence for domestic structures at Panteón, the occupation was probably not as intense or extensive as at the other two settlements. There still exists the possibility, however, that there was a court at the other two sites, especially Nejapa Viejo, as both have been affected by flooding events, erosion, and modern construction.

On the other hand, if the court at MDP was the only one constructed and used during this period of time, then a possible reason for the court’s existence here is the strategic position of the settlement itself. People traveling into the Nejapa subregion west-northwest from the Tlacolula

Valley would have first passed by Mogotes del Panteón. Views from both the site itself and the ballcourt would have allowed the inhabitants to monitor movement coming in from that direction as well as along the river towards the other two settlements. Visitors may have been impressed by such a grand court on the hill at the western entrance to the alluvial plain. Would the inhabitants of Colonia San Martin and Nejapa Viejo have traveled to MDP for ballgame events?

It is difficult to ascertain at the moment without excavation to determine the relationship among these communities, but given their relative proximity to each other, the strategic location of MDP at the entrance to the subregion, and the massive size of the court itself as well as the surrounding public space, more than likely people, at least specially invited members from other communities, traveled to MDP to attend games accompanied by ceremonies and feasting.

By the Early Classic, Colonial San Martin has been abandoned while occupation continues at Nejapa Viejo and MDP (Figure 6.118). Population increased as settlement expanded along the rivers and alluvial plains. We also see the construction of additional courts.

Again however, without confirmed dates we cannot be sure when exactly the courts at Los

Limares, Ejido de los Canseco, Llano las Casas, and the palangana court at El Sitial were built.

Thus, the relationship among Classic-period communities and Classic-period ballcourts remains

512 a little murky at the moment. The court at Ejido de los Canseco, a settlement possibly related to

Nejapa Viejo, was probably constructed sometime in the Early to Middle Classic. It may have overlapped in use with the court at MDP. However, by the end of the Early Classic the latter community had been abandoned, perhaps leaving Ejido de los Canseco as the only court in use for some time. If Ejido de los Canseco is associated with Nejapa Viejo, this could mean that the leaders at the latter were consciously moving control of the game closer to them. At the same time, the court at Canseco is much more informal than the court at MDP, while the settlement itself lies nearly 2 km north of Nejapa Viejo. Additionally, not much remains in the way of occupation or evidence of other structures at Canseco, although survey crews did observe some structure alignments in 2009 and possible fire pits around the court, the traces of which had disappeared by 2015 when I worked there. Why a court was constructed here at this particular settlement is unclear, given its location far from other settlements and the apparent lack of complex architecture apart from the court itself and the one small mound lying some 500 m away that remains. Despite its informal appearance, the court at Canseco is as large overall as the earlier court at MDP, and is similar with respect to the overall design (i.e. enclosed I-shaped court with low or non-extant U-shaped terminal mounds).

513

Figure 6.118: Map of Middle and Late Classic sites and ballcourts.

Los Limares, a valley floor ballcourt that resembles the one at Canseco, may also date sometime to the Middle Classic, although we cannot be sure. As I discussed earlier, this site most likely pertains to Tanaguixi, the larger settlement on the top and flanks of the same hill where we find the Limares court. This area of the settlement appears to have had ceremonial purposes, given the presence of the court, a possible tomb, and two stones carved with glyphs lying only 100 m north of the former as well as the lack of habitational structures nearby.

Moreover, the panorama views of this section of the Nejapa subregion from the terrace where the court was constructed are expansive, likely influencing the selection of this location for building the court. The physical separation of the court from the ceremonial architecture on the hilltop is

514 interesting and stands in contrast to the court at MDP, which was clearly an integrated part of the ceremonial zone of that settlement. This striking difference in the location and spatial context of the Limares court may be linked back to the motives for playing games there in the first place.

Perhaps, heightened tensions between Tanaguixi and its neighbors led the former’s inhabitants to construct the court downslope of their residences and other sacred structures, keeping outsiders away from these areas but still hosting a space in which to interact, compete, and negotiate political tensions and relationships with others. At the same time, the terrace lacks defensive structures and is relatively accessible. The western mound of the court would have been visible from the alluvial plain, and the sounds of game playing would have traveled far. Certainly people would have been able to see and the crowds gathered on the terrace.

The Classic period is also when we see the first ballcourts constructed in the mountains, the palangana courts at El Sitial and Llano las Casas. It has been suggested that palangana courts date to the Early to Middle Classic, and the Middle Classic date from an excavated context at El

Sitial strongly suggests that these courts probably date to this same time period. As I explained earlier, palangana courts are a distinct court type that appears only in a few areas – the

Guatemalan Highlands and Chiapas, where most palangana courts have been documented, and the border area between Puebla and Oaxaca. The two palangana courts occur in very different social contexts: the Llano las Casas court was constructed in an isolated and hidden plain deep in the mountains; and the second surrounded by informal public architecture in a neighborhood likely associated with the Classic settlement on the El Sitial hilltop.

This would mean that around the Middle Classic, there is a dramatic increase in the number of courts, representing two different ballgame traditions. First, we have two I-shaped courts on the valley floor, one to the north of the confluence of the major rivers, and the other on

515 the eastern side of the Rio de la Virgen. Interestingly, both of these courts are not located within a ceremonial center or in association with other public monuments. The stone line foundations observed in 2009 around the Canseco court likely belonged to simple room structures, while the

Los Limares court sits at the bottom of the hill downslope from the main settlement and the public monuments on its summit. Interestingly, any one approaching Tanaguixi from the southeastern side of the hill would have encountered the court first. Second, the two (and possibly three, if we consider the court at Los Chorros) palangana courts located in the mountains. The mountain courts may represent a second network of courts where a different type of game was played (Taladoire 2003). The difference in the court forms found on the valley and in the mountains is striking. We know that there was interaction between the peoples living at these different elevations, but did they play at each other’s courts given how different they are? It is not entirely without possibility that ball players in Mesoamerica may have been versed in different game styles, especially in a frontier region and trade corridor such as Nejapa where they were already accustomed to speaking different languages or interacting with many individuals from different groups.

Two additional courts were constructed in the Late Classic court: at Cerro Maluco

(although it could have been constructed earlier) and possibly La Baeza, respectively. The

Maluco court is the largest in the Nejapa region, and is bigger than many of the courts in the

Central Valleys during this time period. It is also the most complex with its multiple platforms and staircases. Despite its unusual size, like the other courts on or near the valley floor, it is an enclosed I-shaped court with terminal mounds much smaller than the lateral mounds. However, unlike these others, the Cerro Maluco court features a similar architectural arrangement as the much earlier and by now long-abandoned court at MDP, just on a more massive scale. The size

516 of the court and the associated public spaces indicates that a large population could have gathered there. The settlement occupying this hill itself is not very extensive, and in fact nearby

Tanaguixi is much larger. In many ways, it is surprising that such an impressive court would be constructed here on the Maluco hill south from other major Classic period settlements, including the more dominant Nejapa Viejo, in a way nestled into the piedmont. Yet the size of the court strongly suggests again that labor from neighboring communities may have been needed to build it. Despite the similarities in the structural arrangement, the public gathering space on the north side of the ballcourt is much larger than at Maluco than at MDP. As well, because the superior surface of the north lateral mound is practically level with the plaza, activities within the court would have been much more visible to a larger audience at Maluco.

While La Baeza was probably part of a different political system, the court itself resembles the Nejapa structures in its size and overall design, including lateral mounds much larger than the U-shaped attached terminal mounds, with one terminal mound nearly or at level with the ground surface. Moreover, it is interesting that the architectural arrangement of the court is similar to that at Maluco (and the earlier court at MDP), except with the plaza structures to the south rather than the north. Apart from these striking similarities, there are also some notable differences. The La Baeza court is much smaller in comparison to the Maluco court, which is easily twice the size of the former, and the MDP court. Although this is likely due mostly to the amount of space available on the ridgeline where the Baeza court was constructed, topography does not explain the extreme size of the Maluco court. Moreover, this court was located in a public space probably meant for large gatherings and was distinct from the other zone of monumental architecture further up the hill. As I suggested earlier, the Baeza court may have been attached to an elite residence that probably also served some administrative purpose –

517 while certainly people would have gathered there for games and ceremonies, the audience was probably much more restricted. Thus, hosting and attending games may have carried different meanings altogether for the people who attended them. At the moment, we do not know if the elites from Maluco would have traveled to attend and participate in games at La Baeza and vice versa, although the possibility exists that it may have happened on rare occasions. Still, it is more than likely that the elites at La Baeza were aware of the oversized and elaborately designed court on a hill in the valley where the rivers meet.

Without more dates from excavated contexts, resolving the chronology of court construction in the Classic period in the Nejapa subregion remains an issue. There may have been two courts in use in the Early Classic (MDP and Canseco), and perhaps as many as four (or five) in the Middle Classic representing two different game types. Depending on when the

Limares and Canseco courts were abandoned, there could have been as many as three courts in use during the Late Classic, or as few as one. If the former situation, then we would have a network of courts on the valley floor resembling each other in design and shape, only one of which is located in association with other public buildings. On the other hand, if these other ballcourt settlements were indeed abandoned by the Late Classic, then Cerro Maluco would have been the only court in use at this time (again, although La Baeza was likely a contemporary it was probably part of or its own distinct political system). This would have been a significant change, from communities going to games at different courts at several locations in the subregion, perhaps as part of seasonal ceremonies, to people traveling to a single court for perhaps different types of ceremonies, albeit the largest in the region. Such a shift implies changes in the meanings of game ceremonies and the priorities of hosts, players, and the audience members who attended.

518

Of course, there remains the question of the court at Los Chorros, which has tentatively been identified as belonging to the greater community of Los Picachos (King 2010).

Descriptions of the courts indicate that it is a palangana type, dating the structure to the Early and/or Middle Classic. However, a radiocarbon sample from the latter site dates Los Picachos to probably a Classic to Early Postclassic transitional period. Given the narrow temporal range observed at other palangana courts, more than likely the Los Chorros court was in use no later than the Middle Classic, making it a part of that period’s unique network of mountain ballcourts.

It is possible that Los Chorros was an earlier settlement and that occupation had shifted further up in elevation to the ridgeline of Los Picachos later on. Another possibility is that there may be a Classic-era occupation at Los Picachos, and that Los Chorros was an associated barrio whose ballcourt was abandoned when the ballgame played in that type of structure fell out of popularity. Excavation at the Los Chorros will help to resolve this chronological question.

The Early to Middle Postclassic (Figure 6.119) saw not only an explosion in population and settlement expansion, but perhaps also the greatest florescence of the ballgame in the Nejapa subregion. During this time period we see a number of courts with contemporary or overlapping occupations, including at the Greater La Amontonada community with its three (and possibly four) neighborhood ballcourts. Along with the valley courts, there are three more located in the mountains: La Puerta, Cerro del Convento, and the court at La Baeza, which may have continued to be used into the Early Postclassic even as the settlement itself waned in influence. In contrast to the Middle Classic period, the courts both in the mountains and on the valley floor are all enclosed and I-shaped. All of the ballcourts identified in the Nejapa subregion so far (with the exception of La Baeza) dating to this time period are found on the eastern side of the Rios la

519

Virgen and Tehuantepec; more of course may exist in those areas that have not yet been surveyed.

Figure 6.119: Map of Early Postclassic and Middle Postclassic sites and ballcourts.

As a major center with the greatest concentration of population at this time, Greater La

Amontonada likely attracted people from other communities to attend games at the different neighborhood courts. Given their proximity, location along the Rio Tehuantepec, and shared material culture, teams from Greater La Amontonada may have traveled to La Puerta to attend games and vice versa. However, the composition of the audience would have been much different at GLA than it would have been at La Puerta. The courts at the former are much larger,

520 with ample public space for a sizable audience. The court at La Puerta is much smaller, and its location on a narrow saddle would have greatly limited the size of the audience, although we cannot discount people standing on the terraces to watch the game below. Access to the mountain courts was probably much more restricted in general. The courts at Cerro del

Convento is a naturally defensible site with restricted access while visitors to La Baeza had to pass through several residential and monumental zones before arriving at the court. Nonetheless, the public space around both courts could have hosted a sizable but selective audience for game events and ceremonies. Games played at the mountain courts would have carried much different meanings than those hosted closer to the valley floor because of the limited accessibility to both the settlements and the courts, and the attendant social rules attached to accessibility. By the end of the Early Postclassic, La Baeza and its court were abandoned. If it was the only court in this part of the study area, then its abandonment may have marked significant changes in the popularity of the game and/or its role in local politics as well as in the organization of local sociopolitical relationships and hierarchies itself. On the other hand, if we find evidence for other ballcourts in use after the Early Postclassic, then it would indicate only the latter scenario with a continuing although perhaps altered role for the ballgame. Even if the La Baeza court turns out to be the only structure of this type in the immediate area, it would not necessarily have meant that people ceased to play the game here once that court had been abandoned, as people may have played in other spaces such as open fields.

We do not have a confirmed date for the unusual, T-shaped ballcourt at Terezona. It may have been constructed sometime in the Middle or Late Postclassic, although at the moment it is difficult to say more. The hill on which it sits is relatively accessible from the valley floor, and nearby communities likely traveled there for games as there is amble public space around the

521 court and on the upper plaza for viewing game events and related ceremonies. Ample views of the valley floor below and the surrounding mountains would have made the court an ideal place to host important ceremonies. Interestingly, the court was not constructed in what can be considered the main monumental zone of the Terezona settlement, but in a secondary group to the west. Unfortunately, without a more definitive date for Terezona, we cannot be sure how the court may have figured into the ballcourt network nor the political relationships of the

Postclassic period.

The Middle to Late Postclassic marks a period of transition in the Nejapa subregion

(Figure 6.120). Towards the end of the Middle Postclassic, the GLA community was abandoned and Colonia San Martin was once again reoccupied. Radiocarbon dates indicate that these occupations may have overlapped, yet Colonia San Martin represents a distinct settlement in

Nejapa, with its higher proportions of thick adobe architecture, red stucco floors, and imported ceramics such as polychromes in the Mixteca-Puebla style. The abandonment of GLA may have coincided with the arrival of the Zaachileños from the Central Valleys, although we do not find obvious signs of conquest. People may have simply abandoned their homes and left for settlements in the mountains. While occupation continued at some of the communities established during the Early and Middle Postclassic period, the number of ballcourts decreases dramatically. Middle to Late Postclassic courts include Cerro del Convento, El Sitial, and possibly Terezona. For the first time, we see more courts in the mountains than on or near the valley, in places with more restricted access than before. In fact, depending on when the

Terezona court was constructed and then abandoned, it may have been that only the mountains courts were in use during the Late Postclassic at all. The complete abandonment of courts on the valley floor and the shift to high-elevation mountain courts where access would have been more

522 restrictive suggests another dramatic shift in both the ballgame and the significances and purposes of the ceremonies for the people of the Nejapa subregion.

Figure 6.120: Map of Middle Postclassic and Late Postclassic sites and ballcourts.

Based on the evidence we have so far, there appear to be two periods of florescence where the number of courts increases: sometime in the Middle Classic, and once again in the

Early to Middle Postclassic. There are interesting differences between these two periods. In the

Middle Classic, we have possibly two distinct ballgames and thus game traditions, one that was played in I-shaped courts on or near the valley in the settlements around the major rivers; and the other localized to the mountains and played in palangana courts. Material culture evidence does indicate that people living in the mountains and the valley floor interacted often, but the question 523 of whether they played in each other’s court is difficult to ascertain at the moment. The Middle

Classic florescence of courts may be contemporary in age with the appearance of Zapotec-style glyphs, showing an increase in social complexity and ceremonialism in Nejapa. Nejapa’s strategic position along the Mexico-Isthmus-Guatemala trade corridor must have brought new ideas and people, such that two communities high in the mountains constructed exotic palangana courts, while the palangana at Llano las Casas may have been a communal effort by people from different settlements nearby.

The second period of game florescence in the Early and Middle Postclassic was much different. Here all the ballcourts are I-shaped and are only found in association with other public, monumental architecture within the settlement itself. Thus, of the courts documented and dated to this period, none are found in isolated contexts as in the Classic period. The construction of several new courts beginning in the Early Postclassic likely co-occurs with the reestablishment of interregional trade relations after the Classic to Postclassic transition in

Mesoamerica. From this time period through to the Middle Postclassic, people traveling in and around Nejapa would have encountered a small network of ballcourts concentrated along the eastern banks of the Rio Tehuantepec. Interestingly, the majority of these courts are found within the neighborhood communities of Greater La Amontonada. In other words, the increase we see in formal courts is for the most part hyperlocalized, suggesting that the elites of GLA exerted some control over the scheduling and hosting of ballgames and game ceremonies within their sphere of influence in the Nejapa area. However, the higher-status elites of Cucharital probably did not exercise exclusive control over games, given that the other courts were located in other neighborhoods and would have been under the control of the elites or corporate groups that lived there. The other courts in use at this time are found higher in the mountains. Did

524 teams from the mountains and valleys play in each other’s courts? I would suggest that teams from GLA probably played against a team from La Puerta, but not much can be said beyond that.

Still, the number of courts at GLA indicates that ballgames were very important in the internal politics of the neighborhoods that made up the greater community. Most of these courts, with the exception of Cerro del Convento, will be abandoned by the end of the Middle Postclassic.

Outside of these periods of florescence, the evidence so far shows that for most of

Nejapa’s history there are only ever one or a few courts in use at a time. Still, this would not necessarily mean that the game did not play some role in local politics. The first formal ballcourt at MDP was constructed at a strategic location where inhabitants could observe and even control movement into the valley. The MDP community was likely the first settlement encountered by travelers passing through; the monumental architecture at the site, including the court, would have been impressive. The MDP ballcourt was used for some time before the more informal structure at Canseco was constructed, marking a period where the number of courts begins to increase. At the end of the Classic there was probably only one court again, at Cerro Maluco.

The construction of this was clearly a grand declarative statement of power by those who built it, demonstrating the influence of Maluco’s elites and perhaps even the importance of Nejapa in regional exchange networks even as the power of Monte Albán in the Central Valleys waned.

It is interesting that while we do not see signs of conquest or drastic changes in material culture during the Middle Postclassic when the Zaachileños supposedly marched through Nejapa

(with the exception of the singular Colonia San Martin, which is once again reoccupied), we do see a contraction in the number of courts, with only Terezona and El Sitial likely built during this time. In some respects, access to the game becomes much more restricted than it was in previous periods. With the exception of Terezona, the only other courts in use were the mountain courts

525 at Convento and El Sitial. This pattern remains even into the Late Postclassic when we see a further contraction in the number of courts with the likely abandonment of Terezona and a shift in much of the settlement from the alluvial plain and major rivers to the piedmont. The expansion and contraction of the courts in Nejapa may be linked to similar changes in population and settlement patterns as inhabitants adapted to changes in the sociopolitical landscapes and relationships.

There are striking differences between the Classic and Postclassic periods with respect to the distribution of ballcourts and their relative site contexts. Classic-period courts, are found in a variety of contexts – neighborhood precincts, TPA architectural groups, at the foot of a hill far from the main settlement, and an isolated hidden valley. None of the settlements where we find ballcourts were major communities or dominant centers, and even the largest court is found tucked away on a hill far from the main areas of occupation. The diversity of site contexts and ballcourt types suggests that the courts may have been used for radically different purposes, depending on the needs and motivations of the hosts and guests. On the other hand, Postclassic courts are only found in neighborhood precincts or with other, public monumental buildings.

That is, we do not see courts occurring in other contexts outside of the public ceremonial zones at both major and minor settlements as we do in the Classic period. Similar contexts and ballcourt types in the Postclassic suggests a more “unified” ballgame ceremonialism, at least in contrast to the Classic period. This would especially be true for the neighborhood courts at

GLA.

From the construction of the first court to the last, we see several shifts in the role ballgames played in the relationships among the communities in the Nejapa subregion. At particular points in time we have multiple ballcourts that may have formed a network of places

526 for hosting and participating in games and ceremonies. In other periods, we find courts at only one or a few sites. These courts were either used exclusively by that community and any allied or associated settlements, or were places that did draw people in from neighboring communities, but under the tightly controlled conditions of the hosting elites. The cycling between expansion and contraction suggest that the role of ballgames and their attendant ceremonies in local politics changed in response to shifts in the sociopolitical landscape. In Nejapa, people were potentially making connections to various ethnolinguistic groups and regions farther off, resulting if not in mixed, heterogeneous communities, then in relationships that frequently crossed ethnolinguistic boundaries that have been previously treated as being more static or impermeable. Nejapa’s strategic position along critical trade corridors and its rich alluvial land made it an attractive location for many people, who through their shared experiences of living in this frontier region that over the centuries forged new identities that at times were communal or contrasting.

Because of its very nature as a competitive sport requiring at least two teams to play, the ballgame may have occasionally been one way to negotiate movement across these boundaries.

6.7 SUMMARY AND CONCLUSION

Data collected on the documented ballcourts in the Nejapa subregion through survey, mapping, and excavation demonstrate the long history of the ballgame in this area. Nejapa is ideally positioned geographically near the areas where we see both the early emergence of formalized game structures, and where ballgames achieved the status of a religious cult. This positioning perhaps explains why we not only see a massive, formalized court structure as early as the Early Classic period (and perhaps even earlier in the Late Formative), but also what may be the largest ballcourt in Oaxaca (at Cerro Maluco), a high density of courts at a variety of sites, and court types that are not all too common to Oaxaca, in an area previously considered to be a

527 provincial, backwater frontier. The evidence also shows that ballcourts in Nejapa were used for feasting and ceremonial activities, much like elsewhere in Oaxaca. In Chapters 7 and 8, I discuss the results of different statistical and spatial analyses conducted on the mapping data in order to evaluate the different patterns and relationships with regards to the ballcourts, Nejapan communities, and the ballgame. I interpret these analyses within the larger geopolitical and social-historical context of Nejapa and Oaxaca during the Classic and Postclassic periods in

Chapter 9.

528

CHAPTER VII

A QUANTITATIVE ASSESSMENT OF THE BALLCOURTS IN NEJAPA: STATISTICAL ANALYSES AND RESULTS

“When the court had been completed and plastered, at midnight on an auspicious day they put the heart in the court with certain witcheries…and others went to the temple to inform the priests that they had made a ballcourt and fulfilled therein all the solemnities and ceremonies… Motolinía 1903:337-338

7.1 INTRODUCTION

In Chapter 2, I discussed how formal variations in architecture, including size, design, and exterior/interior appearances, can be used to evaluate degrees of cultural affiliation within a region (Aldendurfer and Stanish 1993; Abrams 1989; Canuto 2002; Healan 1993; Kent 1984;

McGuire and Schiffer 1983). The spatial organization of public and/or communal structures is a tangible manifestation of both group and individual decisions and preferences, and a common vocabulary of monumental components suggests widespread knowledge about their use and construction (Ingold 2000; Sutro and Downing 1988). Observable patterns in monument building and design can index the particular meanings and function to the communities that built and used them (Wright 2012). In Mesoamerica, we have long recognized that many structures share similar architectural features while also reflecting local decisions and preferences.

Moreover, Taladoire (1981, 2000, 2003) and other scholars have observed varying degrees of patterning in ballcourt form and design by geographic region and time period.

As I proposed in Chapters 3 and 5, statistically significant similarities could indicate the presence of a ballgame tradition in Nejapa, i.e. both commonly-held practices or ways of playing

529 as well as building these structures. A common ballgame tradition in Nejapa would further imply interaction via gaming events and celebrations between communities likely composed of different ethnolinguistic groups but also sharing a local social identity as people of this region.

Contrastingly, strong statistical differences would suggest more marked social boundaries between these communities, with each community developing its own game practices and architectural court vernacular, perhaps linked to other social or ethnolinguistic group traditions.

In this chapter, I present the statistical analyses and results of the court spatial dimensions, the measurements for which were derived from the mapping data collected during the dissertation fieldwork (Chapter 5). I first explore the spatial dimensions of the Nejapa ballcourts overall and by time period. Then, I compare these data to the evidence from the Central Valleys, the region with the most complete dataset on ballcourts next to Nejapa. I interpret these results and the results of the spatial analyses in the context of the archaeological, ethnohistorical, and ethnographic evidence from the PANT study in Chapter 9.

7.2. UNDERSTANDING THE NEJAPA BALLCOURTS AS A GROUP

Using the mapping data collected during the 2015 field season, I was able to generate multiple measurements of the different ballcourt dimensions22 in ArcGIS. These included the overall length and width of the ballcourt; the lengths and widths of the playing surfaces (central and terminal alleys); the lengths, widths, heights and estimated volumes of the lateral and terminal mounds; the internal slopes of the lateral mounds; and the orientation of the overall court structure (Table 7.1). The data were then entered into a Microsoft Access database, from which I was able to generate data tables to be imported into the statistical analysis package

22 The methods for determining these measurements were described in more detail in Chapter 5. 530

SPSS. In the following sections, I evaluate the Nejapa ballcourts23 overall and by time period in order to assess the degree of similarity or difference between them. Because La Baeza is geographically an anomaly within the Nejapa study area, I did not include the court’s measurements in the analyses. Additionally, I also removed Cerro Maluco from the analyses of the overall dimensions and the central playing alleys because in these measurements it is an extreme outlier.

Table 7.1: The different spatial dimensions of the ballcourts analyzed in this chapter. Measurement Description Ballcourt overall court length length of the entire court from one terminal end to the other overall court width width of the court from the exterior of one lateral structure to the other

Other orientation cardinal and compass degree orientation of the ballcourt along the central axis lengthwise slope interior angle of the lateral mounds length/width ratio ratio of the length of the structure element to its width

23 The dimensions of each court were first reported in Chapter 6. 531

7.2.1 Ballcourt Dimensions: Overall and Playing Surface Lengths and Widths

Table 7.2 presents the descriptive statistics for the overall size and central playing alleys of the Nejapa ballcourts, as a group and by time period. In total, the overall court lengths (σ =

9.3) and widths (σ = 10) show a greater degree of variation than the central alleys (length σ =

5.6; width σ = 1.7). Boxplots (Figure 7.1) comparing the distributions of the overall and central alley lengths and widths demonstrate the relatively small size range of these two dimensions, but also clearly show that this range is much narrower for the latter. The preference for a specific size range for the central alley lengths and widths is suggestive of similar ways of playing the game or rule systems, even though there is no clear standardization. Although not included here, even the court width for the immense Cerro Maluco falls within this size range (w = 8.6 m).

Meanwhile, the relatively greater variation in overall length and width is perhaps more reflective of local needs, tastes, and/or the spatial constraints of where they were constructed.

A series of statistical analyses tested the strength of the relationships between the length and width measurements of the overall and central court dimensions. There is a strong and statistically significant (r = .851; p = .001) correlation between the overall length and width dimensions of the ballcourts, suggesting that we can expect a ballcourt of a certain length to also be of a certain width, and that we would not expect to see extremely long and skinny courts. On the other hand, although the central alley dimensions fall within a very specific and narrow range of sizes, there is no significant relationship between length and width; that is, longer alleys are not necessarily wider and shorter alleys are not necessarily narrower. (r = -.188; p = .629). There is also no significant correlation (r = .204; p = .598) between overall and central alley length as well as between the overall width of the court and the width of the central playing alley (r = .174;

532 p = .655); wider ballcourts do not necessarily have wider playing alleys, and vice versa. The result points to the greater influence of the lateral and terminal mound widths on these size dimensions of the central alleys than the overall dimensions of the court.

Table 7.2: Overall court and central alley size dimensions and statistics in total and by time period. Overall Overall Central Central Time Period Length Width Length Width Classica Mean 44.03 25.61 24.75 6.60 Std. Deviation 9.93 12.53 1.91 1.96 Minimum 30.00 10.00 23.54 5.15 Maximum 52.79 36.94 27.60 9.50 Range 22.79 26.94 4.06 4.35 Variance 98.59 157.06 3.65 3.86 Postclassic Mean 43.06 29.42 26.27 4.98 Std. Deviation 9.74 9.60 7.63 .91 Minimum 24.82 19.00 21.50 4.38 Maximum 50.31 40.82 39.43 6.57 Range 25.49 21.82 17.93 2.19 Variance 94.77 92.22 58.25 .83 Totala Mean 43.50 27.69 25.59 5.70 Std. Deviation 9.33 10.63 5.58 1.61 Minimum 24.82 10.00 21.50 4.38 Maximum 52.79 40.82 39.43 9.50 Range 27.97 30.82 17.93 5.12 Variance 87.08 112.89 31.14 2.59 Note: all dimensions in m. a. excluding Cerro Maluco.

533

Figure 7.1: A box plot comparing the distributions of the overall and central alley lengths (upper) and widths (lower).

534

When we look at these spatial dimensions by time period, we see that Classic and

Postclassic courts are remarkably similar with respect to their measurements. There are some small differences. Courts in the Classic period were narrower on average (Classic mean = 25.6;

Postclassic mean = 29.4), while Postclassic courts were much more consistent in size with respect to their overall widths (Classic width σ2 = 157.1; Postclassic width σ2 = 92.2). Central alley lengths are slightly longer and much more variable in the Postclassic than in the Classic, but their widths are much more consistent. The box plot diagrams for the overall and central alley dimension visibly support these interpretations, showing again similar size ranges in both time periods – especially with respect to the overall court length (Figure 7.2 and 7.3). Where we see the greatest difference is in the central alley width – the size range in the Classic is much greater and more variable than in the Postclassic period (Classic σ = 1.9; Postclassic σ = 0.9). T- tests results, on the other hand, show no statistically significant differences between the Classic and Postclassic periods with respect to the overall24 and central alley25 dimensions.

As we saw when we looked at the overall and central alley court dimensions in total, the Classic and Postclassic courts fall within specific size ranges that more or less overlap from one time period to another. This could suggest that not only did there exist similar ways of playing the game or rule systems, but that they persisted over time. Alternatively, rule systems or ways of playing may have changed over time, but when it came to constructing their courts Postclassic

Nejapa communities referenced the older Classic structures.

24 Length: t(8) = .699, p = .523; width: t(8) = .014, p = .989. Equal variances assumed. 25 Length: t(7) = .210, p = .840; width: t(7) = 1.649, p = .143. Equal variances assumed 535

Figure 7.2: A boxplot comparing the central alley and overall court lengths by time period.

Figure 7.3: A box plot comparing the central alley and overall court widths by time period.

Another important metric for evaluating degrees of similarity or difference among

536 regional architecture is the length/width ratio (L/W) of the court dimensions. Even when there is great variation in structural dimensions, there may still be preferred or standardized proportions that tie seemingly disparate buildings together into a regional tradition. The L/W ratio also informs us about the general shape of the court dimensions. Ratios measuring between 0.0 – 0.5 are more rectangular, while ratios between 0.5 – 1.0 are squarer in shape. For these analyses, I included the Cerro Maluco court dimensions because despite its extreme measurements, its L/W ratio surprisingly fell within the size ranges for Nejapa. Table 7.3 shows that the overall L/W ratio ranges from 0.33 to 0.82, with an average of 0.58, while the central alley L/W ratio ranges from 0.15 to 0.40, with an average of 0.24. Figure 7.4 graphically compares the L/W ratios of each ballcourt. The frequency distribution for the central alley suggests a trending preference for a specific L/W ratio range (Figure 7.5), with the greatest frequency ranging between 0.20 – 0.25

(σ = 0.06). When we look at a box plot of the overall and central alley L/W ratio distributions

(Figure 7.6), we again see this preference for a narrow range for the latter dimension. The results also show that with regards to the overall shape of the Nejapa courts, they tend to be squarish; the central alleys, on the other hand, are definitely rectangular, which makes sense given the morphology of ballcourts in general. Finally, there is no correlation between the overall and central alley L/W ratios (r = .147, p = .685), revealing that more rectangular-shaped ballcourts do not necessarily have more rectangular central alleys.

537

Table 7.3: The overall and central alley length/width (L/W) ratios and descriptive statistics in total and by time period. Overall Central Time Period L/W ratio L/W ratio Classic Mean .53 .26 Std. .17 .08 Deviation Minimum .33 .19 Maximum .70 .40 Range .37 .21 Variance .03 .007 Postclassic Mean .65 .21 Std. .14 .04 Deviation Minimum .48 .15 Maximum .82 .25 Range .35 .10 Variance .02 .001 Total Mean .60 .24 Std. .16 .07 Deviation Minimum .33 .15 Maximum .82 .40 Range .49 .25 Variance .03 .004

538

Figure 7.4: The overall court and central alley L/W ratios of each ballcourt.

Figure 7.5: Histogram of the central alley L/W ratios.

539

Figure 7.6: A box plot comparing the overall and central alley court L/W ratios.

What trends do we see in the L/W ratios over time? Interestingly, while Classic courts tends towards squarishness and Postclassic courts are more rectangular in their overall dimension, there is significant overlap between the two ranges (Classic: 0.33 – 0.70; Postclassic:

0.48 – 0.82). With respect to the central alley L/W ratio (Figure 7.7), again there is significant overlap in the ranges from one time period to another, showing a fairly persistent preference for a specific ratio range in this critical dimension of game play26. Statistical tests showed no significant difference between the overall and central alley L/W ratios of the Classic and

Postclassic periods27. Interestingly, we see that the L/W ratio range narrows considerably in the

26 Classic: σ =.08, σ2 = .007; Postclassic: σ = .04, σ2 = .001. 27 Overall: t(9) = -.882, p = .401; central alley: t(8) = .288, p = .234. 540

Postclassic, and in fact, 80% of the courts fall in the 0.20 – 0.25 range (vs. 60% of Classic courts).

Thus, with respect to the overall size of the courts, Nejapa communities constructed these structures according to their local tastes and needs across time, but still following the general preferences for how big or small courts should be in the subregion. That is, with the singular exception of Cerro Maluco, ballcourts in Nejapa are of similar size more or less; they are variable but not distinctive. On the other hand, the preference for specific proportions in the size dimensions of central alley suggests that there were shared ways of playing within each time period and that these practices may have continued to be shared across time. If rules or game practices did change, they did so in the negotiation of practices between communities or social groups when playing. The small change in the proportion we observe in the central alleys from the Classic to the Postclassic is most likely due to the abandonment of the palangana courts – and whatever game tradition may have been associated with it – and the shift to I-shaped ballcourts only (Terezona’s court is T-shaped but its central alley size and proportions fall in the same ranges).

541

Figure 7.7: A box plot of the central alley L/W ratios for the Classic and Postclassic periods.

With respect to the terminal alley playing surfaces, we can see from Table 7.3 that the lengths of the north and east terminal alleys (terminal 1 or T1) are fairly similar (mean = 18.31 m; σ = 3.36) to the lengths (mean = 17.68 m; σ = 4.59) of the south and west terminal alleys

(terminal 2 or T2). The T1 (mean = 5.49 m; σ = 0.89) and T2 (mean = 4.59; σ = 1.72) widths are also somewhat similar and overlap in their ranges. Figure 7.8 compares the distributions of the terminal alley lengths and widths. This visual assessment shows that there is more variation in the T1 lengths and the T2 widths, but that the measurements for both cluster around similar values. While there is no significant correlation between T1 length and width (r = .512, p =

.159), there is a very strong relationship between T2 length and width (r = .808, p = .008), meaning that while T1 alley dimensions are highly variable, T2 alley dimensions were more predictable – i.e. longer south and west terminal alleys are also wider.

542

Table 7.4: Length and width dimensions and statistics of the terminal playing alleys in total and by time period. Terminal 1a Terminal 1 Terminal 2b Terminal 2 Time Period Length Width Length Width Classic Mean 17.60 5.94 16.35 6.31 Std. Deviation 4.66 .66 4.86 1.60 Minimum 12.96 5.25 11.75 5.14 Maximum 23.94 6.84 23.14 8.67 Range 10.98 1.59 11.39 3.53 Variance 21.72 .44 23.61 2.57 Postclassic Mean 18.87 5.13 18.75 5.57 Std. Deviation 2.31 .96 4.60 1.92 Minimum 16.11 4.36 15.36 4.17 Maximum 22.22 6.73 26.84 8.96 Range 6.11 2.37 11.48 4.79 Variance 5.35 .93 21.19 3.69 Total Mean 18.31 5.49 17.68 5.89 Std. Deviation 3.36 .89 4.59 1.72 Minimum 12.96 4.36 11.75 4.17 Maximum 23.94 6.84 26.84 8.96 Range 10.98 2.48 15.09 4.79 Variance 11.27 .81 21.05 2.96 Note: all dimensions in m. a. North and east terminal mounds. b. South and west terminal mounds.

543

Figure 7.8: Box plots comparing the distributions of terminal 1 and 2 alley lengths (upper) and widths (lower).

544

We see some interesting differences between the terminal alley lengths and widths when we break them down by time period (Table 7.4). Most obviously, Classic courts had much more variable lengths and widths, while the size ranges of these same dimensions for Postclassic courts are much narrower (Figures 7.9 and 7.10). Statistical tests show that there is no statistically significant correlation between the lengths and widths of both terminal alleys in the

Classic period28. On the other hand, although there is also no correlation between the lengths and widths of the Postclassic T1 alleys29, we do see a very strong and significant relationship between the lengths and widths of the T2 alleys (r = .995, p = .000). That is, the south or west terminal mounds (depending on orientation) in the Postclassic were more predictable in their dimensions, i.e. longer alleys are wider. Perhaps, south and west terminal alleys were standardized to some degree during this period. Despite these apparent differences, t-tests reveal that the terminal alley lengths and widths are actually not statistically different from the Classic to the Postclassic30. Thus, while these measurements varied within and between each time period, these variations were not so significant as to distinguish Classic from Postclassic courts by size.

28 Classic T1: r = .818, p = .182; Classic T2: r = .846, p = .154. 29 Postclassic T1: r = .795, p = .108. 30 T1: length t(7) = -.536, p = .608; width t(7) = 1.424, p = .197. T2: length t(7) = -.759, p = .473; width t(7) = .612, p = .560. 545

Figure 7.9: A box plot comparing the terminal 1 and 2 alley lengths by time period.

Figure 7.10: A box plot comparing the terminal 1 and 2 alley widths by time period.

546

What about the L/W ratios of the terminal alleys? Generally, for both the T1 (mean = 0.31; σ =

0.06) and T2 (mean = 0.40; σ = 0.21) these ratios are fairly similar (Table 7.5). The boxplot comparing the T1 and T2 ratios shows a similar distribution of values for both, suggesting that despite variation in size measurements, the builders for these courts were aiming for similar proportions (Figure 7.11). Moreover, there is no statistical difference between the T1 and T2 alleys overall. By time period, it seems that Postclassic terminal alleys are much more similar to each other with respect to their proportions than Classic courts were. Indeed, the range of proportions is much narrower in the later period, reflecting a similar observation made when comparing the length and width value distributions of each. Statistically, there is some degree of difference in the Classic and Postclassic L/W ratios, but only with T1 alleys does it approach significance31. This makes sense given that their values only overlap slightly at the lower end of those ranges, and that Postclassic terminal alleys are more consistent in their proportions.

Overall, the results strongly suggest that despite some degree of variation in the lengths and widths of the terminal alleys, there is a strong preference for a specific and narrow size range for these dimensions. Moreover, the preference for a particular size proportion for the terminal alleys supports the idea that there were shared ways of playing ballgames. While different activities may have taken place in the terminal alleys32, even during games, it is generally assumed that depending on the type of game played, some players would stand in these end zones to prevent the ball from entering this area. Ethnographically we still have players assigned to defend the end zones of modern ballgame courts. More so than their lengths, the widths of the terminal alleys would have affected gameplay because the available space would determine how many players could be in that part of the end zone immediate to the central alley, as well as how

31 T1: t(7) = 2.285, p = .056, T2: t(7) = 1.973, p = .089. 32 Many ballcourts in the Central Valleys have niches in the interior corners of the walls of the terminal alleys. 547 they played and interacted with the ball while remaining active in the game. Over time, the preferred size range and proportions for the terminal alleys narrowed, perhaps indicating a shift in game playing practices from the Classic to the Postclassic related to standardization or an increase in the “sharedness” of practices.

Table 7.5: The L/W ratio statistics for the terminal 1 and 2 alleys in total and by time period. Time Period T1 L/W ratio T2 L/W ratio Classic Mean .35 .54 Std. Deviation .07 .28 Minimum .29 .32 Maximum .45 .94 Range .16 .62 Variance .005 .078 Postclassic Mean .27 .29 Std. Deviation .03 .03 Minimum .22 .27 Maximum .30 .33 Range .08 .06 Variance .001 .001 Total Mean .31 .40 Std. Deviation .06 .21 Minimum .22 .27 Maximum .45 .94 Range .22 .67 Variance .004 .046

548

Figure 7.11: A box plot comparing the distributions of the terminal 1 and 2 alley L/W ratios.

Figure 7.12: A box plot of the terminal 1 and 2 alley L/W ratios by time period.

549

7.2.2 Lateral and Terminal Mound Dimensions

In this section, I discuss the different dimensions of the lateral and terminal mounds of the ballcourts. In analyzing these measurements, I first grouped the mounds according to their orientation and position in the layout of the ballcourt. Thus, because they occupy comparable positions in the court layout, north and east lateral mounds were grouped together as Mound 1, south and west laterals as Mound 2, north and east terminals as Mound 3, and south and west terminals as Mound 4. With a few exceptions, t-tests determined no significant in-group differences in the lateral and terminal mounds according to orientation and position (i.e. lateral vs. lateral). I discuss these exceptions later in the section. The mound dimensions discussed here include the length (L), width (W), interior (H1) and exterior (H2) heights, estimated volumes (V); and the L/W ratios, as described in Table 7.1.

The descriptive summaries of the Mound 1 and Mound 2 dimensions (Table 7.7) and

Mound 3 and Mound 4 dimensions (Table 7.8) show overall that within group they are very similar in size ranges, including even within the mean statistic of each measurement. Box plots comparing the distribution of the values of each mound dimension (Figures 7.13-15) confirm these observations and demonstrate 1) there is a variability in the size values but they occur within specific ranges; and 2) the size ranges Mounds 1 and 2 significantly overlap, as they do for Mounds 3 and 4. The volumes of the lateral and terminal mounds especially show a striking similarity between the different groups. Where we do see some differences between is with H2, or the exterior height. The observed difference in H2 for Mounds 1 and 2 is in part due to the north lateral mound at Cerro Maluco, which is nearly level with the ground surface of the Main

Plaza. After removing this outlier, there was still some difference in the Mound 1 and Mound 2

550

H2 means that suggests on average that the latter mounds have higher exterior heights.

However, this slight difference33 is not statistically significant (t(14) = -.959, p = .354). With respect to Mounds 3 and 4, the small difference in the ranges and means of H2 reflects the in- field observation that more north and east terminal mounds tend to be at or nearly level with the surrounding ground surface, in comparison to the south and west terminal mounds. As with the lateral mounds, this difference is not statistically significant. The box plots also show that lateral mounds are much larger than terminal mounds. Finally, a series of t-tests evaluating the differences between the mounds shows that north/east lateral and terminal mounds do not differ significantly from south/west lateral and terminal mounds with respect to their different dimensions.

Despite the strong preference for mounds of specific size ranges, there is no significant correlation between their respective lengths and widths, i.e. longer mounds are not necessarily wider mounds, and shorter mounds lengthwise are not necessarily narrower. Contrastingly,

Mound 1 widths show a strongly significant relation to their interior (R = .775, p = .014, 60% of variation) and exterior (R = .745, p = .021, 55.5% of variation) heights. The same relationship is seen between the widths and the interior (R = .819, p = .007, 67.1% of variation) and, to a lesser degree, exterior (R = .693, p = .039, 48% of variation) heights of Mound 2. In other words, wider lateral mounds tend to be taller in height while narrower lateral mounds are shorter, which makes sense given the structural form of court mounds. On the other hand, we only see a significant relationship between the widths and interior heights of Mound 3 (R = .705, p = .051,

49.7% of variation) and Mound 4 (R = .868, p = .011, 75.3% of variation). Thus, wider terminal mounds tend to have deeper central alleys and narrower terminal mounds shallower alleys. The

33 Mound 1: σ = 1.85; Mound 2: σ = 2.46. 551 lack of correlation or relationship between the terminal widths and exterior heights again reflects the general design of Nejapa ballcourts where these mounds are nearly or at level with the surrounding ground surface.

As mentioned previously, I grouped the mounds by orientation and position in the layout of the court for the analysis, i.e. north and east laterals were compared against south and west laterals. A series of t-tests evaluating these measurements showed few statistically significant differences (no differences were observed between mound pairs, i.e. north and south laterals and east and west laterals, for example). There is a significant difference between the north and east terminal mounds in their interior heights (t(6) = 2.042, p = 0.03), and the widths of the south and west terminal mounds (t(5) = -4.4953, p = .004). These are the only measurements where we see difference in the mounds, highlighting not only the general similarity of these structures but also the relatively greater variability of the terminal mounds.

There is no statistically significant difference in the L/W ratios of the lateral and terminal mounds between the mound groups; in other words, despite their variability in their respective length and width measurements, the proportion of the mounds are similar. A box plot of the north, east, south, and west lateral L/W ratios shows just how similar the proportions of these particular structures are (Figure 7.16). The L/W ratios of the north, south, east, and west terminal mounds are much more variable, but there is overlap between paired structures (north and south, east and west). Again, though present this variation is not statistically significant, and points to the overall greater variability observed in the respective measurements of the terminal mounds when compared in groups and as individuals.

Based on the design of the courts, the ballgame in Nejapa appears to have been primarily watched from the terminal mounds and the viewing areas near these structures, such as plazas,

552 patios, and nearby terraces. The low height of the terminal mounds probably meant that they were not used in game play as the lateral structures were, instead they were used mainly as viewing platforms and entrances into the court itself. Thus, the measurements of the terminal mounds were likely linked to the needs and tastes of the community more so than the lateral mounds, which would have been more subject to degrees of standardization because they were involved in game play. The one exception is the north lateral mound at Cerro Maluco, whose exterior height is nearly level with a large plaza and thus it too would have been used for viewing events taking place in the court.

553

Table 7.6: Descriptive statistics for the north/east lateral mounds (Mound 1) and south/west lateral mounds (Mound 2). Mound 1 Mound 1 Mound 1 Mound 1 Mound 1 Mound 2 Mound 2 Mound 2 Mound 2 Mound 2 Time Period Length Width Height 1 Height 2 Volume Length Width Height 1 Height 2 Volume Classic Mean 31.60 12.13 2.86 1.1400 761.05 32.42 12.35 2.4600 2.66 588.81 Std. Deviation 7.41 4.04 1.31 .99 595.06 7.79 4.03 .98 1.39 381.80 Minimum 24.60 5.11 1.40 .00 128.41 26.48 6.26 .90 1.00 67.28 Maximum 44.18 14.83 4.70 1.90 1721.38 45.24 16.04 3.40 4.60 1056.69 Range 19.58 9.72 3.30 1.90 1592.97 18.76 9.78 2.50 3.60 989.41 Variance 54.97 16.30 1.71 .99 354090.10 60.69 16.22 .97 1.93 145769.30 Postclassic Mean 27.16 13.27 3.05 2.30 772.01 27.50 13.79 2.78 2.45 736.89 Std. Deviation 4.84 5.81 1.03 1.82 562.43 6.05 4.26 1.02 .73 442.51 Minimum 23.44 7.74 1.80 .00 169.01 22.25 7.46 1.40 1.40 147.33 Maximum 33.76 18.53 4.20 4.20 1378.88 35.51 16.66 3.80 3.00 1164.74 Range 10.32 10.79 2.40 4.20 1209.87 13.26 9.20 2.40 1.60 1017.41 Variance 23.391 33.780 1.070 3.320 316325.23 36.596 18.14 1.043 .54 195816.13 Total Mean 29.6233 12.6333 2.9444 1.66 765.92 30.24 12.99 2.60 2.57 654.62 Std. Deviation 6.46054 4.60194 1.12596 1.45 543.78 7.13 3.94 .95 1.09 390.39 Minimum 23.44 5.11 1.40 .00 128.41 22.25 6.26 .90 1.00 67.28 Maximum 44.18 18.53 4.70 4.20 1721.38 45.24 16.66 3.80 4.60 1164.74 Range 20.74 13.42 3.30 4.20 1592.97 22.99 10.40 2.90 3.60 1097.46 Variance 41.739 21.178 1.268 2.12 295700.35 50.81 15.49 .90 1.18 152405.98

554

Table 7.7: Descriptive statistics for the north/east (Mound 3) terminal mounds and south/west terminal mounds (Mound 4). Mound 3 Mound 3 Mound 3 Mound 3 Mound 3 Mound 4 Mound 4 Mound 4 Mound 4 Mound 4 Time Period Length Width Height 1 Height 2 Volume Length Width Height 1 Height 2 Volume Classic Mean 24.03 7.92 1.64 .18 282.28 22.3020 9.7060 1.5800 .6000 280.7780 Std. Deviation 8.80 3.24 1.182 .25 280.60 6.94644 3.65967 .69065 .57879 238.97821 Minimum 12.57 3.39 .50 .00 16.22 14.54 5.40 .70 .00 38.33 Maximum 33.69 12.47 3.00 .60 666.26 33.62 14.20 2.50 1.50 604.70 Range 21.12 9.08 2.50 .60 650.04 19.08 8.80 1.80 1.50 566.37 Variance 77.52 10.474 1.40 .06 78738.07 48.253 13.393 .477 .335 57110.584 Postclassic Mean 26.50 7.48 1.33 .28 260.84 27.08 6.64 1.45 .60 260.27 Std. Deviation 9.39 .79 .58 .26 47.03 9.26 2.28 .35 .85 173.49 Minimum 15.80 6.60 1.00 .00 210.69 20.53 5.02 1.20 .00 137.59 Maximum 33.35 8.15 2.00 .50 303.95 33.63 8.25 1.70 1.20 382.95 Range 17.55 1.55 1.00 .50 93.26 13.10 3.23 .50 1.20 245.36 Variance 88.12 .64 .33 .07 2211.46 85.81 5.22 .13 .72 30100.77 Total Mean 24.96 7.76 1.53 .28 274.24 23.67 8.83 1.54 .60 274.92 Std. Deviation 8.43 2.49 .96 .24 213.89 7.20 3.47 .59 .56 207.82 Minimum 12.57 3.39 .50 .00 16.22 14.54 5.02 .70 .00 38.33 Maximum 33.69 12.47 3.00 .60 666.26 33.63 14.20 2.50 1.50 604.70 Range 21.12 9.08 2.50 .60 650.04 19.09 9.18 1.80 1.50 566.37 Variance 71.11 6.22 .92 .06 45748.22 51.91 12.04 .34 .34 43190.65

555

Figure 7.13: Box plots comparing the lengths (upper) and widths (lower) of the lateral and terminal mounds.

556

Figure 7.14: Box plots comparing the interior (upper) and exterior (lower) heights of the lateral and terminal mounds.

557

Figure 7.15: A box plot comparing the volumes of the lateral and terminal mounds.ù

558

Figure 7.16: Box plots comparing the L/W ratios of the lateral (upper) and terminal (lower) mounds by orientation.

559

When we examine the measurements of the mounds by time period, we see several interesting similarities and differences that may be linked back to game practices. Lateral mounds in the Classic period were much longer than the terminal mounds, whereas in the

Postclassic the lateral and terminal mound lengths are much more similar (Figure 7.17). The mound lengths influence the overall shape of the ballcourt, and this difference we see in the lengths by time period may explain why Classic courts tend towards a rectangular shape and

Postclassic courts are more squarish. For the mound widths, the situation is somewhat reversed

(Figure 7.18). In the Classic period there is significant overlap in mound width, although north/east terminal mounds are somewhat narrower, while in the Postclassic lateral mounds are generally wider than the terminal structures.

With respect to the mound heights by time period, the box plots (Figures 7.19 and 7.20) show that in general lateral mounds are much taller than terminal mounds, and that the exterior heights of the latter structures are shorter than the interior heights. Classic period terminal mound structures tend to be shorter in their interior heights than the lateral mounds, but there is overlap in the value ranges. Exterior heights are more variable, but still show that lateral mounds are generally taller than terminal mounds. The box plot also reflects the in-field observation that the exteriors of north and east terminal mounds are nearly or are at level with the surrounding ground surface. In the Postclassic, there is some overlap, but the value clusters demonstrate that lateral mounds are much taller than terminal mounds. This observation is supported by t-tests that show that the lateral interior and exterior heights are indeed significantly different from the terminal heights34. Volumetrically speaking, the box plot (Figure 7.21) reveals two things: first,

34 H1: t(11) = 3.301, p = .007; H2: t(11) = 3.225, p = .008). 560 that lateral mounds are larger than terminal mounds; and two, the volumes of the lateral and terminal mounds of the Classic are remarkably similar to those of the Postclassic period.

Indeed, statistical analyses show no significant differences in the lateral and the terminal mounds between the Classic and Postclassic periods with respect to their measurements. The similarities from one time period to another point to an overall persistence in shared preferences for courts of certain sizes and design as well as some aspects of game playing. At the same time, the differences we do observe also point to some changes in game playing. In the Classic period, we have I-shaped and palangana courts as well as courts that fall at either end of the preferred size range for courts and their mound structures. The measurements of the Postclassic courts, on the other hand, are much more regularized and the preferred size ranges much narrower. Many of the courts in the Postclassic are contemporary, explaining this greater degree of regularity in the respective size ranges of the lateral and terminal mounds, even the T-shaped court at

Terezona. In contrast, current evidence indicates that ballcourts in the Classic period were probably built centuries apart – the general similarity in the sizes and designs of Classic courts may suggest that communities building newer courts looked to older structures as references.

Although not mapped, the palangana court at El Sitial is similar in appearance and size to the court at Llano las Casas. I explore meaning of these similarities and differences with respect to the ballgame in Nejapa and the larger historical context in Chapter 9.

561

Figure 7.17: A box plot comparing the lengths of the different mound groups by time period.

Figure 7.18: A box plot comparing the widths of the different mound groups by time period.

562

Figure 7.19: Box plots comparing the interior (upper) and exterior (lower) heights by time period.

563

Figure 7.20: A box plot comparing the volumes of the different mound groups by time period.

7.2.3 Ballcourt Slopes

The interior surfaces of the lateral mounds are thought to have played a significant role in how ballgames were played (Baron 2006). Interior lateral mound surfaces range from completely vertical – such as the ballcourt Chichen Itza – to nearly horizontal – like many of the courts in the Maya macroregion – but the majority incline outward at various degrees. During game play, the ball likely bounced against the sloped lateral mound surfaces, thus the degree of incline would have affected the speed of the ball and its trajectory. For example, balls hitting against gently sloped surfaces would have bounced and changed direction at obtuse angles, decreasing slightly in velocity; balls striking vertical or more extremely sloped walls would have ricocheted sharply like in racquet ball, maintaining or even increasing in speed. The degree to which the interior surfaces were inclined probably depended on the type of game played and the

564 rule systems in place. Therefore, a preference for a specific range of incline or slope for the interior lateral surfaces in a given region may reflect a shared tradition of at least court construction, if not game play itself. Indeed, Taladoire (1981) includes variations in the slope grade of the interior lateral surfaces in his evolutionary schema of different regional court types.

GIS makes it possible to measure the degree of slope incline from the mapping data, however these measurements are at best estimations only. For one thing, it is difficult to evaluate the effects of erosion and time on the interior surface of the lateral mounds.

Additionally, based on in-field observations, we know that –with the exception of El Sitial and the north lateral and west terminal mounds at Cerro Maluco – the lateral mounds were constructed using a mixed earth-and-stone matrix that was then covered in a layer of large stones. Many of these stones are missing from the mound surface today. Nonetheless, enough surface stones are present that we can assume the present degree of incline is close to the original slopes of the interior lateral surfaces. To account for the unevenness of the surface, I include here the minimum, maximum, and the mean values of the inclines in slope degrees (hereafter slope). In two cases, Ejido de los Canseco and La Puerta, the two lateral mounds had different slope degree values, in both cases the west lateral mounds. However, this is most likely due to differential effects of erosion and time.

The value distributions of the minimum, maximum, and mean degrees of slope of the interior lateral mound surfaces show similar and overlapping ranges (Table 7. Figure 7.21).

When we visual these values by site we again see a general overlap in the degree values (Figure

7.22). The differences in the spreads between the minimum and maximum are expected, however this difference is not statistically significant. More than likely, the maximum degree of slope value is closest to the original surface grades when the courts where in use, and the

565 minimum values are the result of erosion and the effects of time. Still, the visual and statistical results suggest that the interior lateral mound surfaces were inclined at a slope grade of between approximately 15-40°, taking into account the effects of erosion and assuming that the lowest

(minimum) values represent those effects (Figure 7.23). Courts with shorter lateral mounds probably featured slopes at the lower end of this proposed range, and courts with higher lateral mounds at the upper end of this range. At the very least, the slope grade would not have been greater than 40°, indicating a relatively gentle incline for these interior surfaces.

Table 7.8: Descriptive statistics of the minimum, maximum, and mean values for the interior lateral surface slope grades in total and by time period. Time Period Min Slope Max Slope Mean Slope Classic Mean 16.50 24.42 19.10 Std. Deviation 4.47 8.08 5.07 Minimum 9.5 13.5 11.6 Maximum 22.5 36.5 26.0 Range 13.0 23.0 14.4 Variance 20.00 65.24 25.74 Postclassic Mean 15.50 23.50 18.80 Std. Deviation 6.04 6.48 5.54 Minimum 6.5 13.5 10.0 Maximum 21.5 28.5 24.0 Range 15.0 15.0 14.0 Variance 36.50 42.00 30.70 Total Mean 16.05 24.00 18.96 Std. Deviation 4.99 7.05 5.02 Minimum 6.5 13.5 10.0 Maximum 22.5 36.5 26.0 Range 16.0 23.0 16.0 Variance 24.87 49.65 25.18

566

Figure 7.21: A box plot comparing the distributions of the minimum, maximum, and mean values for the interior lateral surface slopes.

567

Figure 7.22: Bar chart displaying the minimum, maximum, and mean values of the interior lateral surface slope grades by site.

Figure 7.23: The estimated degree of slope range for the interior lateral mound surfaces.

By comparison, there is not much difference in the incline of the interior lateral surfaces between the Classic and Postclassic periods. The ranges for the minimum, maximum, and mean values overlap (Figure 7.24), and independent t-tests reveal no statistically significant distances from one time period to another. Where we do see some difference is in the spread of the ranges;

Postclassic interior lateral surfaces vary more in their inclines, while the different slope values

568 for the Classic surfaces cluster closer together (save for the outliers of Cerro Maluco and Ejido de los Canseco). Again, these differences are not significant and suggest that there were no dramatic changes in slope over time.

Figure 7.24: A box plot of the distributions of the minimum, maximum, and mean values by time period.

7.2.4 North-South Ballcourts vs. East-West Ballcourts in Nejapa

It has been frequently suggested that the orientation of ballcourt structures was socially and culturally meaningful (Hendon et al. 2009; de Montmollin 2007), and that regional preferences for specific orientation are evidence for shared game playing or ideological traditions

(Baron 2006). Some scholars have even proposed that during the Classic period, east-west oriented I-shaped ballcourts were tied to the Mexican version of the hipball game, while open- ended courts with north-south alignments were linked to the Maya hipball game (de Montmollin

1997:24;Taladoire and Colsenet 1991:162-163). Globally, multiple studies have documented

569 cross-cultural examples where structure orientations can be linked to specific social and ideological practices (García and Belmonte 2014; Moore 1996; Pearson and Richards 2003;

Smith 2007). Within Mesoamerica, on the other hand, regional studies show that this formulation is somewhat simplified, and many of the traits associated with a distinct “Mexican” or “Mayan" tradition often overlap spatially and temporally, including orientation (see de

Montmollin 1997 for an excellent discussion of this debate). Nonetheless, we cannot discount that in some instances the orientation of the ballcourt was linked to specific cultural or social meanings, or associated with certain status groups.

At the same time, while orientation may not have carried a particular or defined social or ideological meaning, it may have still signified something related to the season or time of day at which games were played in a court. For example, how the ballcourt was oriented would have changed how the direction of the sun would have affected game play at different time of the day.

Moreover, the degree of effect would have varied by season as the sun moved along the horizon.

In east-west oriented ballcourts, games played in the morning or the evening would have been greatly affected by the angle and direction of the sun as it rose and set. Games at north-south oriented ballcourts could have been played at different times of the day because the sun would not have affected the players as much. Thus, the orientation of the ballcourt may have more to do with the angle and direction of the sun as it would have affected games at the particular location where the courts were built. Controlling for the sun may have been important so as to give the community or sponsor group the home-field advantage.

Table 7.9 shows the degree and cardinal orientation of each ballcourt in Nejapa by time period. Table 7.10 displays the frequencies of the compass designation values or cardinal orientations for the Nejapa ballcourts; for ease of analysis and comparison all orientations were

570 recorded between 0 – 180 degrees. As we can see, the greatest percentages are courts oriented

NS (0° – 11°, 169° – 180°; 27.3%) and S-SE (147° – 168°; 27.3%), with EW (79° – 100°;

18.2%) and E-SE (101° – 123°; 18.2%) following close behind. Obviously, these numbers are too small to determine a standardized orientation in Nejapa. When we simplify the categories into two, NS and EW, we see that 58.3% are oriented north-south and 41.7% are oriented east- west, respectively. Again, this confirms that there is no standardized orientation nor a clear preference for a specific alignment.

Table 7.9: The ordinal (degree) and cardinal orientations of the Nejapa ballcourts by time period. Time Cardina Period Site Name Ordinal l Classic Cerro Maluco 84 EW Ejido de los 121 E-SE Canseco Llano las Casas 159 S-SE Los Limares 150 S-SE Mogotes del 98 EW Panteón Postclassic Cerro del 173 NS Convento El Cucharital 1 NS El Sitial 111 E-SE La Puerta 162 S-SE Lachixoba 32 N-NE Terezona 0 NS

571

Table 7.10: Frequencies of the different cardinal orientations of the Nejapa ballcourts. Cumulative Orientation Frequency Percent Valid Percent Percent E-SE 2 18.2 18.2 18.2 EW 2 18.2 18.2 36.4 N-NE 1 9.1 9.1 45.5 NS 3 27.3 27.3 72.7 S-SE 3 27.3 27.3 100.0 Total 11 100.0 100.0

As stated previously, ballcourt orientations are known to vary by time period and geographic area. Do we see any kind of patterning with respect to orientation and time period in

Nejapa, given that over half of the courts are oriented north-south? Running a simple cross- tabulation analysis, we can see from Table 7.10 that a majority of the NS courts date to the

Postclassic period (71.4%), while the majority of EW courts date to the Classic (75%).

Conversely, 60% of the courts in the Classic period are east-west oriented, while 83.3% of the courts are north-south oriented in the Postclassic. The difference between the two time periods, however, is not significant (p = 0.137). Thus, even though there does appear to be a shift in ballcourt alignment from majority EW-oriented in the Classic to majority NS-oriented in the

Postclassic, the small sample may be too small to detect significant differences. When we look at the bar chart results (Figure 7.25), we see that ballcourts in the Classic period are split between north-south oriented courts and east-west oriented courts, while in the Postclassic only one court is oriented east-west while the rest are oriented north-south. Thus, the shift is not from one predominant orientation over time period, but from variety to one standard or preferred (more or less) orientation.

572

Table 7.11: The percentage of ballcourts oriented north-south (NS) and east-west (EW) by time period. Time Period Classic Postclassic Total Cardinal Simple EW Count 3 1 4 % within Cardinal 75.0% 25.0% 100.0% Simple % within Time Period 60.0% 16.7% 36.4% % of Total 27.3% 9.1% 36.4% NS Count 2 5 7 % within Cardinal 28.6% 71.4% 100.0% Simple % within Time Period 40.0% 83.3% 63.6% % of Total 18.2% 45.5% 63.6% Total Count 5 6 11 % within Cardinal 45.5% 54.5% 100.0% Simple % within Time Period 100.0% 100.0% 100.0% % of Total 45.5% 54.5% 100.0%

573

Figure 7.25: The number of north-south and east-west oriented ballcourts by time period.

The next series of questions to consider is if there is a statistically significant difference between the sizes and dimensions of the north-south and east-west oriented ballcourts. Box plots comparing the overall and central playing alley lengths, widths, and L/W ratios (Figures 7.26 –

7.28) show that the distributions of these values overlap considerably (with Cerro Maluco standing out as an exceptional case). NS oriented courts vary much more in their overall dimensions than ES courts, although this is likely a result of the small sample size of the latter.

What is most striking is just how similar NS and EW oriented courts are in the same respective dimensions of the central playing alleys, which would have been more important for ballgame play. Box plots of these same measurements for the terminal alleys also demonstrate generally overlapping distributions (Figure 7.29). Indeed, the close similarity in the distribution of the

L/W ratios for the north, south, and east terminal alleys points to a preferred and narrow size range for this particular feature of the court, suggesting again a shared tradition or similar ways

574 of playing and rule systems in Nejapa. On the other hand, west terminal alleys show a much greater range of variability in the L/W ratio. In this respect, NS courts were much more standardized while EW courts had west terminal alleys that varied much more from one court to another. This may mean that these alleys may have been more subject to local needs and preferences and thus may have served other purposes apart from game playing, such as spaces for ceremonies or for audiences to watch. On the other hand, the sample size of EW courts is small and hence the results likely do not reflect any meaningful difference, cultural or otherwise.

The similarity in the length, width, and L/W ratios of the courts and their playing alley surfaces is supported by the results of several independent t-tests.

Figure 7.26: Box plot comparing the distribution of the overall and central alley lengths of the EW ballcourts versus the NS ballcourts.

575

Figure 7.27: Box plot comparing the distribution of the overall and central playing alley widths of the EW versus NS ballcourts.

Figure 7.28: Box plot comparing the overall and central playing alley L/W ratios of the EW vs NS ballcourts.

576

Figure 7.29: Box plot comparing the L/W ratio of the terminal alleys by orientation.

When we plot out the distributions of the different dimensions of the mounds, we see again that there is significant overlap between the north/south lateral mounds and the east/west lateral mounds, as well as the north/south terminal mounds and east/west terminal mounds

(Figures 7.30 – 7.32). Independent t-tests confirm no significant difference in the various dimensions of the NS and EW-oriented lateral mounds (all p-values > 0.05, although the differences between their lengths of north and east lateral mounds (t(6.139) = 13.657, p = 0.057) does approach significance. Where we do see significant differences are in the widths (t(13) = -

4.174, p = 0.001), interior heights (t(13) = -3.587, p = .005), and volumes of the NS and EW terminal mounds (t(13) = -3.051, p = 0.009). Why we see differences in these particular dimensions and only in the terminal mounds of NS and EW ballcourts is not clear at this time.

Finally, overall differences in these measurements between individual mounds (i.e. north vs. south vs. east vs. west laterals, etc.) are also not statistically significant (all p values > 0.05).

577

Figure 7.30: Box plot comparing the lengths of the lateral and terminal mounds by orientation.

Figure 7.31: Box plot comparing the widths of the lateral and terminal mounds by orientation.

578

Figure 7.32: Box plot comparing the L/W ratios of the lateral and terminal mounds by orientation.

In general, the results further demonstrate that although there was no strict standardization in court measurements, there was still a preference for specific size ranges and, most importantly, proportions in Nejapa even when the ballcourts are differentially oriented.

The shift from no clear preference for court orientation in the Classic period to predominantly north-south courts in the Postclassic is intriguing but a little more difficult to explain. Court orientation in Oaxaca is highly variable and there appears to be no preference for a particular orientation by region or time period. Thus, there appears to be no specific cultural or social meaning attached to north-south or east-west orientations that could help explain the shift in orientation over time in Nejapa. Rather, the shift may be explained by the apparent increase in standardization, or at least the increase in preference for specific and narrower size ranges, in the

Postclassic. As the data results show, Postclassic courts are more similar to each other in their size dimensions and overall appearance, perhaps these more of these courts were

579 contemporaneous. Thus, there may have been a higher degree of coordination in building the courts, at least during the Early and Middle Postclassic, so that teams from the different communities could be played at each other’s courts. This would have meant having courts oriented in the same general direction.

We must also consider though that the orientation of the courts may have more to do with the local topography and the space where the courts were constructed than social or cultural meanings and/or ways of playing. For communities who constructed courts, the first consideration was probably choosing a location with favorable topography for building this type of structure. The orientation of the court was likely a result of where people chose to build more than anything else, at least for some of the courts such as those in the mountains.

7.2.5 Summary

When we look at the ballcourts in Nejapa as individual structures, they are all highly variable and appear to be distinct. Certainly, many feature unique elements that reflect local decision-making and tastes, such as the corner platform at the El Cucharital ballcourt and the series of stepped platforms that form the east terminal mound of Cerro Maluco court. However, when we evaluate the ballcourts as a group, several patterns emerge. First, while it is clear that there is no standardization in the measurements, there is a preference for specific size ranges in both court dimensions and proportions. The exception is Cerro Maluco, which stands out as an unusually-sized court in comparison to the others in the sample with respect to overall size and central alley playing length and lateral mound lengths. Still, in other important measurements and proportionally the ballcourt at Cerro Maluco falls within the preferred range for Nejapa. The quantitative evidence is highly suggestive of a regional ballgame tradition or at least shared ways

580 of playing that endured over time. Even if ways of playing or rule systems changed over time, such changes emerged from the interactions and negotiations between communities. However, while there are similarities between the Classic and Postclassic periods, the differences we do see may point to changes in the role of the game in local and regional politics. I explore these differences and similarities further in Chapter 9.

7.3 A COMPARATIVE CASE STUDY: THE BALLCOURTS OF THE CENTRAL VALLEYS AND NEJAPA

The data analyses so far suggest that there was indeed shared ball playing practices among the communities in Nejapa that persist through time, even as these may have changed. In order to understand these results in context, it is useful to compare the Nejapa data against another region or subregion in Oaxaca with ballcourts. By comparing the case studies, we can evaluate whether the patterns we are observing in Nejapa are unique in comparison, or part of larger regional trends, and whether there existed shared or distinct regional ballgame traditions and/or the difference/similarity in the sociopolitical role of the game. In the following sections, I briefly explore and analyze the ballcourt data from the Central Valleys, then compare these data to the information from Nejapa. I conclude with a preliminary interpretation of what the comparative results may mean for the ballgame in these two regions.

7.3.1 The Central Valley Ballcourts

The information on the ballcourts documented and registered in the Central Valleys region represents the most complete database of court dimensions in Oaxaca after Nejapa. Most of the data comes from the regional survey projects conducted in the 1970s and 80s, including

Blanton’s Valley of Oaxaca Settlement Pattern Project, among others (Blanton 1978; Blanton et

581 al. 1982). The ballcourt data was then collated and presented in the edited volume The

Mesoamerican Ballgame (Scaraborough and Wilcox 1991). Because the ballcourt dimension are based on in-field survey methods, there are some issues with their accuracy; certain dimensions were also not recorded or were unavailable based on what data had been recorded. Moreover, the data is frankly decades old at this point, and does not include any new ballcourts that have since been found and recorded. Despite these issues, the Central Valleys ballcourt database is still an important source for comparative information. In this section, I briefly review the statistical results on the Central Valley courts.

Table 7.12: Central alley size dimensions of the Central Valley ballcourts in total and by time period. Time Period Length Width L/W ratio Late Formative Mean 28.03 7.28 .27 Std. Deviation 7.20 3.90 .14 Minimum 15.00 1.00 .06 Maximum 44.00 15.00 .50 Variance 51.89 15.20 .02 Classic Mean 28.14 7.92 .31 Std. Deviation 14.63 4.86 .14 Minimum 9.00 3.00 .11 Maximum 70.00 24.00 .57 Variance 213.97 23.59 .02 Postclassic Mean 26.55 7.40 .30 Std. Deviation 9.78 2.02 .11 Minimum 16.60 5.70 .15 Maximum 38.60 10.00 .38 Variance 95.64 4.09 .01 Total Mean 27.95 7.60 .29 Std. Deviation 11.33 4.22 .14 Minimum 9.00 1.00 .06 Maximum 70.00 24.00 .57 Variance 128.33 17.80 .02

582

Because the overall widths are not reported, I only analyzed the measurements of the central playing alleys. As Table 7.12 shows, there is a much wider range of size variation in

Central Valley courts, especially in the Classic period. Box plots (Figures 7.33 – 7.35) show the distributions of the lengths, widths, and L/W ratios of the central alleys. The distribution of the

L/W ratios shows is practically bimodal (Figure 7.36), with the first peak occurring at 0.15 –

0.20 (the global or first local maximum), and the second peak occurring at 0.40 – 0.45 (the second local maximum). This suggests that there may have been two distinct populations with respect to the L/W ratios or proportions of the central playing alleys. Additionally, there is a moderate but strongly significant relationship between the length and width dimensions (r =

0.519; p = 0.000), indicating that on average longer central alleys are wider, while shorter alleys are usually narrower, in contrast to the Nejapa central alleys which show no correlation (these differences will be explored further in the following section and Chapter 9). When we compare these same measurements by time period, we do not see any statistically significant differences between the Late Formative, Classic, and Postclassic periods (all p-values > 0.05).

583

Figure 7.33: Box plot comparing the central playing alley lengths of the Central Valley Courts by time period.

Figure 7.34: Box plot comparing the central playing alley widths of the Central Valley courts by time period.

584

Figure 7.35: Box plot comparing the central playing alley L/W ratios of the Central Valley courts by time period.

Figure 7.36: A histogram of the central playing alley L/W ratios shows a clear bimodal distribution. 585

Much like Nejapa, in the Central Valleys we see similar size ranges, means, and standard deviations in the different dimensions between the north/east lateral mounds and the south/west lateral mound, as well as between the north/east terminal mounds and the south/west terminal mounds (Tables 13 – 15). However, we also see greater variations in these same measurements, revealing that there are both many very small mounds and many larger-sized mounds, and other sizes in between. When we compare some of these measurements by time period, we see overall that there is much overlap from the Late Formative to the Classic period, the only difference being that Classic period south/west laterals mounds are significantly larger volumetrically than

Late Formative south/west lateral mounds (t(6.647) = -2.340, p = 0.054).

586

Table 7.13: Size dimensions of the north/east (Mound 1) and south/west (Mound 2) lateral mounds of the Central Valley ballcourts. Mound 1 Mound 1 Mound 1 Mound 1 Mound 1 Mound 2 Mound 2 Mound 2 Mound 2 Mound 2 Time Period Length Width Height 1 Height 2 Volume Length Width Height 1 Height 2 Volume Late Formative 28.64 12.38 2.93 2.50 932.69 29.07 12.38 2.15 2.15 788.54 7.64 3.50 1.98 1.09 673.20 7.97 4.35 .80 .80 450.99 16.00 7.00 1.00 1.00 290.00 15.00 5.00 1.00 1.00 210.00 44.00 17.00 9.00 4.00 2640.00 44.00 18.00 3.00 3.00 1836.00 28.00 10.00 8.00 3.00 2350.00 29.00 13.00 2.00 2.00 1626.00 58.40 12.26 3.92 1.19 453196.23 63.46 18.92 .64 .64 203391.77 Classic 27.92 11.09 3.42 3.42 1061.08 29.64 11.73 3.67 3.82 1376.60 4.58 5.15 1.51 1.51 762.09 5.77 5.88 2.23 2.27 913.61 19.00 5.00 1.00 1.00 25.00 23.00 2.00 1.00 1.00 140.00 36.00 20.00 6.00 6.00 2340.00 43.00 20.00 8.00 8.00 2835.00 17.00 15.00 5.00 5.00 2315.00 20.00 18.00 7.00 7.00 2695.00 20.99 26.49 2.27 2.27 580785.72 33.25 34.62 4.97 5.16 834677.60 Total 28.74 12.12 3.18 2.96 1079.15 29.58 12.60 2.85 2.88 1075.71 6.58 4.50 1.74 1.34 815.05 6.93 5.54 1.76 1.79 735.88 16.00 5.00 1.00 1.00 25.00 15.00 2.00 1.00 1.00 140.00 44.00 20.00 9.00 6.00 3200.00 44.00 25.00 8.00 8.00 2835.00 28.00 15.00 8.00 5.00 3175.00 29.00 23.00 7.00 7.00 2695.00 43.28 20.28 3.04 1.81 664310.70 48.09 30.67 3.10 3.19 541513.95 Note: all measurements in m. a. Because measurements were only available for one Postclassic ballcourt, it was not included in the summary statistics.

587

Table 7.14: Size dimensions of the north/east (Mound 3) and south/west (Mound 4) terminal mounds of the Central Valley ballcourts. Mound 3 Mound 3 Mound 3 Mound 3 Mound 3 Mound 4 Mound 4 Mound 4 Mound 4 Mound 4 Time Perioda Length Width Height 1 Height 2 Volume Length Width Height 1 Height 2 Volume Late Formative 28.00 8.25 1.50 1.80 423.50 33.00 6.67 2.00 2.00 491.00 9.59 2.87 1.22 1.10 482.13 11.53 2.52 1.15 1.15 564.00 18.00 4.00 .00 1.00 80.00 20.00 4.00 1.00 1.00 80.00 42.00 10.00 3.00 3.00 1134.00 42.00 9.00 3.00 3.00 1134.00 24.00 6.00 3.00 2.00 1054.00 22.00 5.00 2.00 2.00 1054.00 92.00 8.25 1.50 1.20 232449.00 133.00 6.33 1.33 1.33 318097.00 Classic 26.00 9.17 3.17 3.17 702.17 26.00 11.20 2.67 2.67 866.80 8.00 3.76 1.17 1.17 376.42 8.15 5.07 .82 .82 494.84 17.00 4.00 2.00 2.00 136.00 17.00 4.00 2.00 2.00 136.00 35.00 15.00 5.00 5.00 1080.00 35.00 17.00 4.00 4.00 1428.00 18.00 11.00 3.00 3.00 944.00 18.00 13.00 2.00 2.00 1292.00 64.00 14.17 1.37 1.37 141695.37 66.50 25.70 .67 .67 244867.20 Total 26.91 8.80 2.33 2.55 590.70 27.67 9.56 2.36 2.36 689.67 8.36 3.29 1.44 1.29 420.61 9.29 4.39 .92 .92 496.85 17.00 4.00 .00 1.00 80.00 17.00 4.00 1.00 1.00 80.00 42.00 15.00 5.00 5.00 1134.00 42.00 17.00 4.00 4.00 1428.00 25.00 11.00 5.00 4.00 1054.00 25.00 13.00 3.00 3.00 1348.00 69.89 10.84 2.06 1.67 176910.68 86.25 19.28 .85 .85 246857.00 Note: all measurements in m. a. Because measurements were only available for one Postclassic ballcourt, it was not included in the summary statistics.

588

Table 7.15: Summary statistics of the lateral and terminal mound L/W ratios of the Central Valley ballcourts. Mound 1a Mound 2b Mound 3c Mound 4d Time Period L/W Ratio L/W Ratio L/W Ratio L/W Ratio Late Formative .40 .40 .26 .20 .14 .21 .20 .01 .00 .00 .00 .19 .56 .81 .56 .21 .56 .81 .56 .03 .02 .04 .04 .00 Classic .36 .40 .38 .45 .19 .23 .23 .26 .00 .00 .24 .24 .60 .65 .83 .83 .60 .65 .60 .60 .04 .05 .05 .07 Total .38 .41 .32 .38 .16 .21 .21 .22 .00 .00 .00 .19 .60 .81 .83 .83 .60 .81 .83 .64 .03 .05 .05 .05 a. North and east lateral mounds. b. South and west lateral mounds. c. North and east terminal mounds. d. South and west terminal mounds.

589

Figure 7.37: Box plot comparing the lateral and terminal mound lengths of the Central Valley courts by time period.

Figure 7.38: Box plot comparing the lateral and terminal mound widths of the Central Valley courts by time period.

590

Figure 7.39: Box plot comparing the lateral and terminal mound L/W ratios of the Central Valley courts by time period.

Kowalewski and colleagues (1991) noted no observable patterns in the orientations of the

Central Valleys ballcourts in their original analysis. Indeed, at first glance there is no obvious standardization in orientation or even preference for a specific degree range. However, a visual display of the data shows a clear bimodal distribution, suggesting that there were two distinct populations: one group of ballcourts oriented between 0° – 50° (N/NE) and another group oriented between 80° – 150° (E/S-SE) (Figure 8.40). Given that most Central Valleys courts are assumed to date to the Classic period, it seemed safe to assume that there was no relationship between time period and orientation. When I evaluated the data based on where the ballcourts are located, i.e. the north or Etla Valley, the east or Tlacolula Valley, and the south or Ocotlan

Valley, some further patterning emerged. Due to the small sample sizes, I could only perform a pair-wise comparison between the ‘north’ and ‘east’ valleys. The results show that there is a statistical difference in the orientations of ballcourts in the ‘north’ valley as compared to the ‘east

591 valley (z = -2.395, p-value = 0.017). In other words, ballcourts in the Tlacolula are predominantly oriented E/S-SE, whereas in the Etla Valley the courts are nearly evenly split between N/NE and E/S-SE-oriented courts.

Figure 7.40: Frequency distribution of the Central Valley ballcourt orientations, which show a clear bimodal distribution.

7.3.2 Ballgame Traditions in Oaxaca: Comparing Nejapa and the Central Valleys

Both the ballcourts from the Central Valleys and Nejapa show internal patterning with respect to the different dimensions of the court structures and playing surfaces. Based on the summary, statistical results, and visual representations of the data presented so far, there does appear to be some interesting differences between the two regions. Are these differences significant enough to suggest distinct ways of playing, i.e. ballgame traditions? Alternatively, if there are similarities, what do they mean in the larger context about ball playing not only in

Nejapa and the Central Valleys, but in Oaxaca in general? Similarly, what do these differences

592 tell us about ballplaying and community interactions as well? In this section, I explore three specific metrics for which there was sufficient data from both regional datasets for comparison: the central playing alley dimensions; the lateral and terminal mound dimensions; and structure orientations (ordinal and cardinal directions).

Table 7.16 compares the dimensions of the central playing alleys of ِthe Central Valleys ballcourts to those of the central playing alleys of the Nejapa ballcourts. Clearly, there is a much greater range of variation in the measurements of the Central Valleys courts than in the Nejapa courts (length: CV = 61.0, N = 18.9; width: CV = 23.0, N = 5.5; L/W ratio: CV = 0.52, N =

0.26). This observation is supported by both the frequency distributions and box plots comparing the spreads of the values for the central playing alleys in each region (Figures 7.41 –

7.43). The visual displays also show that the Central Valleys values overlap the values of the

Nejapa courts and, at the same time, confirms that there is a preference for a narrower size range in Nejapa compared to the greater variation in the Central Valleys. Statistically, while there is no difference between the two regions with respect to their lengths (t(56) = 0.381, p = 0.705), we do see significant differences in their widths (t(45.353) = 2.331, p = 0.024) and L/W ratios

(t(45.353) = 2.531, p = 0.016), respectively. In both cases, equal variances between the two populations cannot be assumed, further underscoring how different they actually are with respect to these measurements.

593

Table 7.16: Summary of the central playing alley dimensions of the Central Valley versus Nejapa ballcourts. Region Length Width LW ratio Central Valleys Mean 27.95 7.60 .29 Std. Deviation 11.33 4.22 .14 Minimum 9.00 1.00 .06 Maximum 70.00 24.00 .57 Range 61.00 23.00 .52 Variance 128.33 17.80 .02 Nejapa Mean 26.67 5.74 .22 Std. Deviation 5.31 1.72 .07 Minimum 21.50 4.00 .14 Maximum 40.38 9.50 .40 Range 18.88 5.50 .26 Variance 28.21 2.96 .00 Note: all measurements in m.

594

Figure 7.41: Frequency distribution (upper) and box plot (lower) comparing the lengths of the central playing alleys of the ballcourts in the Central Valleys and Nejapa.

595

Figure 7.42: Frequency distribution (upper) and box plot (lower) comparing the widths of the central playing alleys of the ballcourts in the Central Valleys and Nejapa.

596

Figure 7.43: Frequency distribution (upper) and box plot (lower) comparing the L/W ratios of the central playing alleys of the ballcourts in the Central Valleys and Nejapa.

597

Turning now to the lateral and terminal mounds, we can see from the frequency distributions that the values for the different mound dimensions overlap for the most part (Table

7.16, Figures 7.44 – 46), indicating that there are similar size ranges for both regions. While many of the distributions for the Central Valley mound dimensions are nearly normal or are even bimodal, distributions for the same in Nejapa primarily skew heavily to the right, showing a preference for values at the lower end of the range. Still, there is clearly a greater range of variation in the mounds of the Central Valleys than in the Nejapa subregion. Perhaps one of the most interesting differences between the two regions is the L/W ratios. The frequency distributions for the lateral and terminal mounds in the Central Valleys are bimodal, more clearly so for the latter, suggesting that there are two distinct “populations” proportion-wise and perhaps even different modes of playing. Contrastingly, in Nejapa we see a more normal distribution for the lateral mounds and a slightly left-skewed distribution for the terminal mounds, indicating a preference for specific proportions.

598

Table 7.17: Size dimensions of the lateral and terminal mounds of the Central Valleys and Nejapa ballcourts. L/W Region Mound Type Length Width Height 1 Height 2 Volume Ratio Central Valleys lateral Mean 29.33 12.12 3.12 2.91 1072.35 .41 Std. Deviation 6.82 4.91 1.97 1.58 775.65 .16 Minimum 15.00 3.00 1.00 1.00 140.00 .09 Maximum 44.00 25.00 10.00 8.00 3200.00 .81 Range 29.00 22.00 9.00 7.00 3060.00 .72 Variance 46.55 24.07 3.90 2.50 601635.39 .03 terminal Mean 26.28 8.58 2.68 2.70 767.84 .34 Std. Deviation 10.59 3.91 1.68 1.68 670.42 .19 Minimum .00 .00 .00 1.00 20.00 .05 Maximum 45.00 17.00 7.00 7.00 2430.00 .83 Range 45.00 17.00 7.00 6.00 2410.00 .78 Variance 112.06 15.29 2.83 2.83 449468.06 .04 Nejapa lateral Mean 29.93 12.81 2.77 2.26 1154.25 .44 Std. Deviation 6.61 4.16 1.03 1.23 776.30 .15 Minimum 22.25 5.11 .90 .00 169.79 .18 Maximum 45.24 18.53 4.70 4.60 2927.47 .71 Range 22.99 13.42 3.80 4.60 2757.68 .53 Variance 43.65 17.29 1.05 1.50 602647.79 .02 terminal Mean 24.35 8.26 1.53 .74 331.19 .36 Std. Deviation 7.63 2.93 .78 .87 328.19 .14 Minimum 12.57 3.39 .50 .00 25.56 .20 Maximum 33.69 14.20 3.00 2.70 1134.30 .69 Range 21.12 10.81 2.50 2.70 1108.74 .49 Variance 58.24 8.58 .61 .76 107708.38 .02 Note: all measurements in m.

599

Figure 7.44: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal mound lengths of the Central Valleys and Nejapa ballcourts.

600

Figure 7.45: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal mound widths of the Central Valleys and Nejapa ballcourts.

601

Figure 7.46: Frequency distribution (upper) and box plot (lower) comparing the L/W ratios of the lateral and terminal mounds of the Central Valleys and Nejapa ballcourts.

602

Statistical analyses of the lateral and terminal mound measurements showed no significant differences between their lengths, widths, and the L/W ratios when we compare the

Central Valley and Nejapa ballcourts. On the other hand, we do see significant difference in the interior (t(42.849) = 2.985, p = 0.005) and exteriors heights (t(38.658) = 4.761, p = 0.000) as well as the volumes (t(35.258) = 2.601, p = 0.014)of the terminal mounds35. These differences can be seen visually in Figures 7.47 – 7.49. Overall, terminal mounds in Nejapa are much shorter in height and smaller in form than those in the Central Valley, especially with respect to the exteriors which are often very low to nearly level with the surrounding surface. While ballcourts in the Central Valleys have been documented with low or non-extant terminal mounds

(Feinman and Nicholas 2011), it is not the predominant pattern there as it is in Nejapa.

35 Equal variances not assumed for all three measurements. 603

Figure 7.47: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal interior mound heights of the Central Valleys and Nejapa ballcourts.

604

Figure 7.48: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal exterior mound heights of the Central Valleys and Nejapa ballcourts.

605

Figure 7.49: Frequency distribution (upper) and box plot (lower) comparing the lateral and terminal mound volumes of the Central Valley and Nejapa ballcourts.

606

With respect to the orientations of the ballcourt themselves, we observe some interesting but relatively small differences between the Central Valleys and Nejapa (Figure 7.50). North- south-oriented courts predominate in Nejapa (NS = 63.6% versus EW = 36.4%), whereas there are slightly more east/west oriented courts in the Central Valleys (NS = 42.6% versus EW =

57.4%). However, this difference is not statistically significant (p = 0.202), and could be a result of sample bias. I explore the implications of these results, especially with respect to the social meaning of orientation, further in Chapter 9.

Figure 7.50: The number of north/south versus east/west-oriented ballcourts in the Central Valleys and Nejapa.

607

7.3.3 Summary

While we could only compare three metrics from each database, the results reveal some interesting patterns of similarity and difference between ballcourts in the Central Valleys and those in the Nejapa subregion. In all dimensions, the courts in the Central Valleys show a much greater range of variation than the Nejapa courts, whose measurements occur within a much narrower range of values. For example, although the central playing alleys are similar in length, we see significant differences in the widths and, most importantly, their proportions in the

Central Valleys. At the same time, we see differences in the distributions of the values for several dimensions; where distributions for the Central Valleys are highly non-normal and even bimodal, value distributions in Nejapa tend to be nearly normal or skewed towards one direction, again showing preferences for specific size ranges and proportions.

The statistical analyses and results both within each region and in comparison, also confirm observations made in the field about certain style differences in the architectural design of the Nejapa and Central Valleys ballcourts. As mentioned earlier, one of the dimensions where we see statistical significance is in the exterior heights of the terminal mounds, reflecting the in- field observation that all terminal mounds in Nejapa are very low or practically level with the surrounding ground surface. We find courts like this in the Central Valleys, but this is not the predominant pattern by far. The main ballcourts at Monte Albán and Atzompa, for example, have terminal mounds with vertical back walls and exterior heights over 2 meters. The ballcourt at Cerro la Campana, near the modern towns of Suchiquiltango and Huizto, has an eastern terminal mound whose exterior height is level with the adjacent plaza. However, the western terminal mound is actually a temple mound that serves the dual purpose of forming the eastern

608 border of the west plaza, and has a double staircase on that side. Interestingly, we have several examples of ballcourts whose lateral or terminal mounds were temples or other similar structures in other group configurations in the Central Valleys. In Nejapa, on the other hand, ballcourts are separate structures and we have no examples where a mound was another structure. Even the east lateral mound of Terezona, which does provide passage to the upper plaza, is clearly set out from the upper staircase leading directly to that same plaza.

Finally, perhaps one of the most marked distinctions between the two regions is the density of ballcourts per square kilometer. For the Central Valleys, this density is 1.7%. For

Nejapa, the density of ballcourts per square kilometer for the total study region is 6.7%. When we consider only the amount of land surveyed by the PANT team, this number increases to

18.5%, an increase over the Central Valleys by over 900%! Should we find more ballcourts in the areas that have not been surveyed, then this density may increase. What is the significance of this density? If we factor in time, then the density of ballcourts in Nejapa is not very significant because there would have been several periods of time in which only one court would have been in use. However, if we consider instead the location of the courts themselves, then the density is significant. Barring any courts that may be found in the future, where we find ballcourts in

Nejapa appears be primarily in areas that would have experienced the most traffic as people moved through the valley. With the exception of the unique palangana court at Llano las Casas, the courts in the mountains are found at settlements that had access to trade networks and/or were sites of ritual significance. I explore this theme further in Chapter 9.

7.4 CONCLUSION

In this chapter, I presented the results from statistical analyses of the Nejapa ballcourt mapping data. The purpose of these analyses was to evaluate the degrees of difference and

609 similarity between the courts as a group. The results indicate that despite their uniqueness as individual structures, there is also strong patterning with respect to specific size ranges and proportions, in addition to certain architectural style elements and construction methods that tie the subregion together in a shared ballgame tradition. I then compared these results to data from the Central Valleys in order to determine what the significance of the differences and similarities in the Nejapa ballcourts is in a cross-regional perspective. The results reveal interesting patterns that help us interpret how ballgames may have figured into the relationships between communities in similar and distinct ways in these regions. I explore more of these differences and similarities, and their significance, between the Nejapa subregion and the Central Valleys, as well between Nejapa and other regions in Oaxaca, further in Chapter 9.

610

CHAPTER VIII

VISUALIZING THE BALLCOURTS OF NEJAPA ON THE LANDSCAPE: SPATIAL ANALYSES AND RESULTS

“Sí, tuvimos un equipo de pelota mixteca. Jugamos en frente de la oficina…donde nos encontraste, ¿recuerdas?…jugamos con Camarón…y también allá con El Gramal, Soledad Salinas…hasta [Santa María] Zoquitlán y [San Pedro] Totolapan, viajamos hasta allá para jugar.”

Yes, we had a pelota Mixteca team. We played in front of the office…where you found us, remember?…we played with Camarón…and also with El Gramal, Soledad Salinas….up to [Santa María] Zoquitlan and [San Pedro] Totolapan, we traveled all the way there to play. Don Jesús, Ejido de Canseco

8.1 INTRODUCTION

In the previous chapter, I evaluated the mapping data from the Nejapa ballcourts using

statistical methods. But to understand the courts in a regional contextً, it is useful to consider how they are distributed across the Nejapan landscape with respect to: 1) regional settlement patterns; 2) cultural and/or natural features; and 3) the ballcourt dimensions themselves, their attributes, and types of courts present. For example, are the largest courts found only on the valley floor or at they located on higher elevations? Are ballcourts found in certain types of environments, such as near waterways, and do we find only find ballcourts at complex, large residential sites with monumental public architecture? In Chapter 5, I described how GIS spatial analysis can assist archaeologists in two critical ways. First, by helping us to visualize and interrogate multiple lines of archaeological data in three-dimensional space through the

611 combination of vector objects (i.e. archaeological features) and raster layers (DEM, slopes, vegetation, hydrology). Symbolizing the vector objects and raster layers based on different attributes can further elucidate other distributions patterns. Second, the analytical tools available in GIS programs such as ESRI ArcGIS, the program used here, can help archaeologists evaluate the strength or character of these visual associations as well as create predictive models useful for understanding the distribution of archaeological phenomena.

The following sections visually and spatially explore the ballcourt mapping data. First, I visually assess the distribution of the ballcourt dimensions and attributes on the landscape. Then,

I used a combination of different spatial techniques, specifically the least-cost path, cluster distribution, and intervisibility and viewshed analyses to further interrogate the spatial data.

Along with the visual display maps, the results of these analyses should help us to understand the ballcourts as permanent and public social structures constructed in a shifting sociopolitical landscape. Of course, these results are more heuristic tools that help us to visualize what may have occurred in the past, rather than reflections of a past reality. They are therefore useful for exploring possibilities or avenues for further inquiry into the archaeological data. Comparing the spatial analyses results against the in-field observations and the ethnohistorical and archaeological evidence further helps us to think about observable spatial patterns in ballcourt distribution and how these patterns might be linked to local and regionally sociopolitical relationships.

8.1.1 Building the GIS Spatial Database

Spatial data on the sites recorded in the Nejapa region were taken from the larger PANT

GIS database. I combined site boundary feature classes from each field season and then cleaned

612 up the data where necessary. For example, Colonial-period sites, such as the ex-haciendas and the community of Majaltepec, were removed from the database entirely. Additional social attributes such as time period ('Classic', 'Postclassic', and 'unknown'), associated community if known, site type ('habitational', nonhabitational'), and site code were added to the site feature table, based on information from PANT field reports and databases (see Table 8.1). A site point feature class was generated from the site boundary feature class using the Feature to Point tool for use in the spatial analyses. A separate feature class was created specifically for the ballcourts, with each point recording the exact location of each ballcourt along with its spatial data. Both the polygon (site boundaries) and point (site points) feature classes were saved along with the ballcourt point feature class in a separate MASTER file geodatabase as backup safety files. For ease of data storage, I created a different file geodatabase for each separate spatial analysis I ran. Although La Baeza was included in the spatial database, I did not include it in the spatial analyses presented here. Again, La Baeza likely belonged to a different political system altogether.

613

Table 8.1: Codes and definitions for habitational and non-habitational site types. Habitational Code Definition RES residential, general RAN rancheria, isolated household civic-ceremonial residential small, possibly part of a community CCRS but not confirmed; < 20 structures civic-ceremonial barrio or neighborhood that is clearly part of a CCB community civic-ceremonial residential large, sometimes a head town or CCRL main neighborhood in a community; ≥ 20 structures Non-habitational Code TC terreno de cultivar, cultivated terrain RIT ritual site, general CRIT cave ritual site CRP cave ritual site with painting SV sitio de vigilar, lookout point TMU temporary multiuse site PG piedra grabada, site with stones with glyphs civic-ceremonial nonresidential, usually ceremonial precinct of a CCNR community

All analyses used a 30 x 30 m DEM raster surface created by PANT by clipping the original INEGI data to the Nejapa study region. DEM surfaces are critical in GIS analysis for creating first-order derivatives of slope and aspect (e.g. Barton et al. 2010) as well as second- order derivatives of terrain curvature (e.g. Bevan and Conolly 2004); visibility (Wheatley and

Gillings 2002; Lake and Woodman 2003; Llobera 2003); movement (e.g. Llobera et al. 2011) and cost-surfaces (Bell and Lock 2000; Bell et al. 2002). Finally, additional geological and geographic information on Nejapa and Oaxaca overall comes from INEGI data purchased by and generously provided to me by PANT.

614

8.2 THE DISTRIBUTION OF BALLCOURTS IN NEJAPA: SPATIAL ANALYSES AND RESULTS

One of the central questions framing this dissertation is if there are observable spatial patterns associated with the ballcourts in Nejapa communities. As I discussed in Chapter 2, how buildings are distributed both within sites and on the landscape often has sociopolitical significance relevant to the settlement hierarchy, the social networks among individual communities, and residents’ access to political power. Several regional studies have already demonstrated the link between ballgames, ballcourt distribution, and regional political organization (De Montmollin’s 1989, 1995, 1997; Hendon et al. 2009; Santley et al. 1991; Stark n.d.; Whalen and Minnis 1996). Therefore, it is arguably possible to assess the sociopolitical role of the ballgame in Nejapa at the regional scale by examining the distribution of ballcourts, their dimensions, and their attributes across the sociopolitical landscape (de Montmollin

1995:158; Hendon 2010). In the following section, I take a critical look at these distribution patterns via visual maps and cluster distribution analysis.

8.2.1 Exploring Ballcourt Dimensions and Attributes on the Landscape

Digital maps allow for a quick visual assessment of how archaeological data and their attributes are distributed spatially on the landscape. I generated maps symbolizing the ballcourt measurement data and attributes such as elevation, slope degree and percent change in rise, and distance from waterways in order to see, for example, where the largest courts were located, how far courts were located from major rivers and other sources of water, and at what elevations or slope grades were they preferentially constructed. For the maps of court measurements, I only included courts in the study sample; for the maps displaying geographic and topographic data

615 with respect to the courts, I included all structures documented by PANT. Visualizing these results revealed some interesting patterns.

At first glance (Figure 8.1), it appears that a majority of the ballcourts recorded in the

Nejapa subregion are located on or close to the alluvial valley floor, with a large concentration (n

= 6) of these on the eastside of the Río de La Virgen (before it joins the Rio Grande) and

Tehuantepec. Although it lies at a much higher elevation than the valley floor courts, the La

Puerta court follows this same pattern. Two other courts are also on the valley floor: the structure at Mogotes del Panteón, directly north of the Río Grande, and Ejido de los Canseco, which lies about 3 km to the northwest of the confluence of the Ríos Grande and La Virgen.

Interestingly, when we consider the chronological age of the ballcourts (see Figure 8.1), we see that Classic-era courts were constructed on either side of the major rivers, while Postclassic-era courts near or on the valley floor are concentrated on the east side of the rivers. Five more courts are located at higher elevations in the mountains. If we include the palangana court at El Sitial, then there are six mountain courts in total. Mountain ballcourts are more mixed in age – two are

Postclassic in age, while two are likely Classic-period and another possibly Classic or Classic-

Postclassic Transition36. These mountain courts are found in the ranges in the southeast of the study region37.

36 The most approximate date we have for the Los Chorros court is from an excavated context at the nearby site of Los Picachos, with which the former is likely associated, dated to the Classic-Postclassic Transition (calibrated CE 945 – 1047). At the same time, while a few palangana courts date to the Late Classic, most are assumed to date to periods centuries earlier. This issue should be resolved with excavations in the future. 37 Excluding La Baeza, which may have belonged to a different political system altogether. 616

Figure 8.1: Map of the ballcourts in Nejapa by time period.

At the same time, we must consider if this distribution of courts we observe in Nejapa is an approximate reflection of reality or an artifact of survey boundaries (Figure 8.2). When we examine where the ballcourts have been recorded and where PANT has conducted survey, it does appear as if the ballcourts concentrate in specific places, in particular along the major rivers or in the eastern mountain ranges. These areas likely fell along important Prehispanic travel corridors, especially as people were passing through the Sierra Sur on their way to the Isthmus, and even south into the Chontalpa subregion. However, there are also clearly several large gaps in the survey coverage, especially in the northern and western reaches of the study region. We know that more sites exist in the general area around La Baeza, and several people reported

617 finding archaeological remains including ritual sites with red-hand paintings in the territory.

More than likely there are ballcourts to be found in these areas too. The two ballcourts to the north of the Rio Grande and the rivers’ confluence – MDP and Canseco – strongly suggest that the additional ballcourts may be located here. Given the observed pattern of courts along the major rivers, future survey on either side of the Rio de La Virgen south of the confluence will likely reveal other ballcourts. Still, the survey map does seem to show a general tendency for courts to be built along the major waterways on or near the alluvial valley floor.

Figure 8.2 Map of the Nejapa study region and the areas where PANT conducted survey.

618

Is there a difference between ballcourts located on the valley floor and those found in the mountains? Looking at the maps of the court measurements, some interesting patterns emerge.

With respect to the overall widths and lengths (Figures 8.3), we see clearly that the largest courts are indeed located on or near the valley floor along the major rivers, while those courts at the smaller end of the size range are found at much higher elevations. Those courts on the valley floor fall between 45 – 55 m long. The exception is the court at Cerro Maluco, which at 87 m long 36.5 m wide is the largest in the Nejapa subregion, and indeed probably in this part of the

Sierra Sur (the court at El Cucharital is wider by four meters but certainly not as long).

However, if we were to include the two other mountain courts at Cerro del Convento and El

Sitial, then the pattern changes somewhat. The estimated length of the Sitial court would place it at the lower end of the preferred size range for the valley courts, while the Convento court comes in just under at 40 m. Meanwhile, the palangana courts at Los Chorros and El Sitial measure around 30 m long. Thus, mountain ballcourts vary more in overall length. Width-wise, we have a much clearer distinction between valley and mountain courts, even when we consider the mountain courts not depicted here. Valley courts fall into two ranges: 30 – 35 m wide (n = 2) and 35 – 40 m wide (n = 5); mountain courts, on the other hand, are around 20 m or less. The limitations of available terrain suitable for building a court likely affected the sizes of the courts in the mountains more so than those on the valley.

619

Figure 8.3 Map of the overall lengths (upper) and widths (lower) of the ballcourts.

620

We see a slightly different pattern when we examine the central alley lengths and widths

(Figure 8.4). There is no obvious spatial pattern with respect to these dimensions as there is to the overall sizes. For the central alley lengths, both mountain and valley courts tend to fall in the 21 – 25 m range. The exceptions are the main ballcourt at El Cucharital and the court at

Lachixoba, which are somewhat bigger, and the extremely long Cerro Maluco. Despite its unusual alley length, the central alley width of Maluco’s court still falls within the preferred size range for Nejapa courts. Instead of there being a difference between mountain courts and valley courts, the narrow size range for central alley widths is clear. The visual results reflect the trends observed in the statistical results of a preference for a specific size range, and that such sizes occur at all elevations and locations.

621

Figure 8.4: Map of the central lengths (upper) and widths (lower) of the ballcourts.

622

What about the court mounds? For the sake of brevity, here I look specifically at the

Mound 1 volumes (north and east lateral mounds), using volume as an indicator of size. The results show that the largest mounds do occur on the valley floor, but that there are also valley courts with mounds similar in size to the mountain courts, which fall towards the lower end of the mound size range (Figure 8.5). This may change once the courts not included in the study sample have been mapped. For example, based on in-field observations, the lateral mounds of the main ballcourt at El Sitial likely fall towards the larger end of the size range, while Cerro del

Convento falls somewhere toward the middle and the other mountain palangana courts, like the court at Llano las Casas, are at the lower end of this same range. Again, the amount of space on suitable terrain that would have been available for building a court would have affected some of the structure’s size dimensions much more at mountain sites than those closer to the valley floor.

La Puerta’s court is on a narrow and small saddle, thus it is the smallest court in the study sample. At the same time, other dimensions were either not affected at all or were not affected greatly. Central alley dimensions are related to ways of playing and were apparently independent of site size or location, while the size of the mounds were subject both to space limitations and local tastes and needs.

Overall, the general conclusion we can draw is that on average, the courts constructed on or near the valley floor are larger in their overall size dimensions than the courts in the mountains, likely because space limitations of the latter. Mound sizes were more variable but larger mounds still tend to be found at lower elevations. The critical exception is the central alley lengths and widths, which show no similar spatial patterning. Because these dimensions are probably related to particular ballgames, playing styles, and/or rule systems, the narrow size

623 ranges and the lack of difference between the courts overall and by their location (mountain vs. valley) strongly suggests shared or similar ways of playing in Nejapa.

Figure 8.5: Map showing the Mound 1 (north/east laterals) volumes of the ballcourts.

During fieldwork, there was a general impression that the ballcourts seemed to be located close to water sources, specifically rivers. Indeed, when we look at how far the ballcourts are located from the major rivers (Figure 8.6), i.e. the Ríos Grande, de La Virgen and Tehuantepec, we see that many are found within 2 km distance from a major river (the court at Los Chorros is located about 4 km from a major river on the other side of that mountain range). Six of the courts are even closer than that, falling within 1500 m of a major river. Apart from the Los

624

Chorros court, all the courts located far from major rivers (n > 4 km) are those in the mountains, obviously. When we consider all waterways, including the intermittent rivers and streams that flow primarily during the rainy season, we see that nearly all the courts can be found within 350 m of a waterway, with a majority (n = 8) found even closer than that at within 150 m of a nearby river38 (Figure 8.7). The two courts located furthest from water sources are Cerro del Convento and La Puerta, the former a mesa and the other at the southern end of a narrow ridgeline. Even then, nearby water sources are only about 500 m away for both, respectively. Of course, it is not surprising that we find courts near waterways as we would expect people to prefer settling where water is accessible. Yet, the visual map supports the in-field observations that many of the courts were close to or within sight of waterways.

38 This number includes the second palangana court at El Sitial, which was only about 100 – 150 m up from a nearby river. 625

Figure 8.6: Map showing the distance of each ballcourt from a major waterway, measured in meters.

626

Figure 8.7: Map showing the distance in meters of the ballcourts from nearby waterways.

Moreover, the observation that most of the courts are located on or near the valley is supported by elevation map (Figure 8.8). We see again that a majority (n = 7) are found at 625 –

750 m asl. Three of the courts are found at extremely high elevations – Cerro del Convento

(1505 m asl), La Baeza (1925 m asl), and Los Chorros (2080 m asl). When we consider the degree of slope and percentage change in rise (Figure 8.9 and 8.10), we see that the courts were constructed where the terrain was relatively flat with gentle changes in topography. Most are found on slope grades of less than 7°, especially those closest to the valley floor but even at sites located high in the piedmont and mountain regions such as La Puerta, El Sitial, and Los Chorros.

Where we see courts constructed at grades higher than 8° (but still less than 13°) are Cerro

627

Maluco, Cerro del Convento, and Llano Las Casas. With respect to the percentage of slope rise, again a majority are found where there are relatively gradual changes in topography (≤ 8.56%).

Most of the courts constructed at slightly higher slope percentages between 10 – 13.1% are all found at mountain and piedmont sites, reflecting the natural topography of the local settlement.

Because El Cucharital is located on the summit of a narrow extension of piedmont, it falls within this higher range of percent change in rise.

Figure 8.8 Map showing the elevation of each ballcourt in Nejapa.

628

Figure 8.9: Map showing the degree of slope where each ballcourt was constructed.

629

Figure 8.10: Map showing the percentage or rise in slope where each ballcourt has been constructed.

Is there any significance to the spatial patterns I have described here? For example, are ballcourts preferentially located at sites that are closest to major rivers or waterways? To evaluate this, I generated 100 random points within the area surveyed by PANT (see Figure 8.2) and sampled their elevations, slopes (both degree and percent change in rise), and distance from sources of water. I then compared the spatial attributes of the random points to those of the ballcourts. The t-test results were rather surprising; while there was no statistically significant differences between the random points and the ballcourts with respect to elevation and distance from waterways, the degree of slope (t(30.015) = 5.049, p = .000) and percent change in rise

(t(30.015) = 5.173, p = .000) was actually significantly different. The results were the same

630 when I compared these same spatial attributes between the ballcourts and those sites without ballcourts. Again, the degree of slope (t(19.615) = 4.719, p = .000) and the percent change in rise (t(20.998) = 4.862, p = .000) were significantly different. These differences can be observed visually in Figures 8.11 and 8.12. The results show that ballcourts are found at different elevations and within relatively close distances to waterways because sites in Nejapa share these attributes in general. At the same time, they demonstrate at the level of the individual site, courts were selectively constructed on flat or flattened terrain where there was sufficient space for a structure with a sizable and rectangular-shaped footprint.

631

Figure 8.11: Box plots comparing the degree of slope (upper) and percent change in rise (lower) between the random points and the ballcourts.

632

Figure 8.12 Box plots comparing the degree of slope (upper) and percent change in rise (lower) between the non-ballcourt sites and ballcourts.

633

8.2.2 Ballcourts and Communities: Cluster Analyses and Results

Having examined the ballcourts on the landscape, we now critically examine the sites themselves where they were built. When we look at these sites, we see that most are large and small civic-ceremonial residential settlements and neighborhoods (Figure 8.13). But this does not tell us much. Are all the sites where ballcourts were built the same? Attribute-based or cluster analysis is a technique for evaluating archaeological phenomena in GIS based on predictive modelling (Aldenderfer and Maschner 1996; Conolly and Lake 2006; Gillings et al.

1999; Kvamme 1989, 1990, 1995, 2006; Lock and Stancic 1995; Mehrer and Wescott 2006;

Parker 1985; Wescott and Brandon 2000), using factors or attributes associated with the spatial features that the researcher is interested in testing. These same variables can then be used to build a predictive model via logistic regression analysis (Stopher and Meyburg 1979; Hosmer and Lemeshow 1989; Hosmer et al. 1989; Menard 2001). Where there is a lack of correspondence between the model and the actual data is an indication that there are alternative variables influencing the location of sites or other archaeological phenomena.

For the analysis presented here, I selected several variables that might explain why we find ballcourts at particular settlements, including the size, accessibility, and complexity of the site, the degree of slope and percent change in rise, and finally its distance from sources of water.

I then tested a series of assumptions involving different combinations of these variables. First, that ballcourts are found only at large, complex settlements that were accessible based on the surrounding geography. Second, that these large, complex, and accessible settlements were located on more or less flat terrain that shows little change in topography. Finally, that these same sites also tended to be located close to sources of water. The goal was to not only tease out different details about the site themselves but also to see which variables and models best predict

634 where ballcourts may be found.

Site size was simply the value of the shape feature, in this case the size of the polygons drawn for each site based on in-field observations, survey, and post-hoc processing in ArcGIS. I used elevation as a proxy for accessibility, making the assumption that sites at lower elevation are more accessible than those found at higher elevations. Site complexity was calculated based on the number of structures recorded at each site, such as plazas, temples, platforms, structure foundations, and montículos, or mounds in Mexican Spanish. Mounds often represent domestic structures, and thus are an indication of the number of families that would have been living there.

For those sites where there were no structures recorded in the GIS database but were described in field reports as having structures present, I did one of two things: 1) counted the number of structures mentioned in the field reports and used that value; and 2) used a minimum structure count value based on the site size and type. For those sites without a structure count, I estimated the minimum number of structures based on the average number of structures recorded at sites and the site size (≥ 5000 m2 = 1; ≥ 10000 m2 = 10; ≥ 10000 m2 = 20). The degree of slope and percent change in rise were sampled from rasters containing these values, assuming that ballcourts are more likely to be found at sites that are relatively flat with little change

Cluster analysis tools in GIS tend to prefer higher values. Since the models assume that sites with ballcourts are located at lower elevations on relatively flat terrain near to water sources, I inverted the values for these attributes using the following equation:

((Original value – Zmax) * -1) + Zmin

For the analysis, I excluded most of the nonhabitational sites, since these tend to be very small and lack any architectural features or structures. After combining all of the variables into a single attribute table linked to the point feature class, I then normalized the variable values using

635 the following equation:

Normalized value = (original value – minimum value) / (maximum value – minimum value)

Normalizing rescales the attribute values to between zero and one, and was calculated for all six variables. This helps to simplify comparison as it converts the variables to equivalent values.

The normalized values of each variable were then summed together to create a ballcourt index for each series of assumptions. These values were then also normalized for comparison (Table

8.2).

636

Table 8.2: The normalized values for the different variables and the Ballcourt Indices of the ballcourt sites.

Time Norm. Norm. Norm. Norm. Norm. Norm. Norm. Norm. Norm. Norm. Period Site Name Access Area Strc. Count Slope Deg. Slope Perc. Major Rivers Waterways BCI 1 BCI 2 BCI 3 Classic Cerro Maluco .90 .18 .27 .75 .21 .85 1.00 .45 .46 .57 Ejido de los .96 .00 .00 .96 .04 .74 .86 .32 .34 .42 Canseco Llano las Casas .83 .02 .01 .97 .03 .02 1.00 .29 .30 .25 Los Chorros .28 .20 .16 .43 .51 .62 .95 .21 .21 .32 Los Limares .98 .00 .01 .92 .07 1.00 .80 .33 .35 .48 Los Mogotes .98 .16 .12 .88 .11 .99 .93 .42 .43 .57 del Panteon Postclassic Cerro del .55 .26 .38 .24 .70 .30 .74 .40 .40 .33 Convento El Cucharital .98 1.00 1.00 .74 .22 .97 1.00 1.00 1.00 1.00 El Sitial .71 .32 .95 .79 .18 .15 1.00 .66 .67 .55 La Puerta .90 .02 .09 .50 .44 .97 .41 .34 .34 .37 Lachixoba .99 .15 .13 .95 .04 .99 1.00 .43 .44 .59 Terezona .93 .44 .30 .95 .04 .89 1.00 .56 .57 .67

637

Based on these values, we can make the following conclusions about the sites where we found ballcourts. For the most part, sites with ballcourts are relatively accessible as they are located at lower elevations. The exceptions are Los Chorros, which has an exceptionally low value at 0.28, and El Sitial (0.71) and Cerro del Convento (0.55). El

Cucharital is the largest ballcourt site with a value of 1.00, however the other settlements show much smaller values, suggesting that ballcourts were built at sites that fall on the lower end of the site size range. We see similar values for the site structure count – while two sites show high values, El Sitial (0.95) and El Cucharital (1.00), respectively, the results show that most sites had a low density of structures. Unsurprisingly, except for those sites located at much higher elevations, sites with ballcourts are located on relatively flat terrain (range 0.50 – 0.95). At the same time, most sites are also found on terrain that varies somewhat in topography with respect to the percent change in rise (range 0.04 –

0.44). All of the sites are located within some distance of a water source; even those sites found far from a major river (those with lower values) are still located close to an annual or intermittently-flowing river.

When we look at the ballcourt indexes themselves, we see that in each set of assumptions, most of the sites with ballcourts fall in the lower to middle end of the value ranges. For Ballcourt Index 1, that range is 0.21 – 0.66 and for Ballcourt Index 2, it is 0.21

– 0.67; these ranges are very close and show that the variables of slope degree and percent change in rise do not greatly affect the overall results. The greatest shift in the value of each ballcourt site is observed in Ballcourt Index 3 (range 0.25 – 0.67). Although overall the range is similar to the previous two sets of assumptions, the individual values shifted on average ± 0.05 points (see Table 8.2; Figure 8.13). The exception is El Cucharital, a settlement that encompassed different neighborhoods, which consistently falls at the higher 638 end of index value range (1.00 for each index). One of these neighborhoods was a high- status elite barrio with complex, monumental public architecture and multiple residences.

When we compare these values across time, we see that there is not much difference overall statistically (all variable p-values > 0.5). While sites with ballcourts in the Classic period have relative fewer structures (p = 0.69), in all other respects we do not see dramatic differences in the types of sites where ballcourts are found over time. Are there significant differences between non-ballcourt sites and ballcourt sites with respect to these attributes?

T-test results show that with respect to individual attributes (using the normalized values), sites with and without ballcourts are remarkably similar (p values ≥ 0.05). However, when we sum these variables together as indexes, we see that BC Index 1 (p = 0.76) and BC

Index 2 (p = 0.70) approach significance, while the BC Index 3 values for non-ballcourt and ballcourt sites are statistically different (p = 0.20) – perhaps reflecting the observed trend for ballcourts to be primarily located at small-to-medium sized settlements with some degree of complexity on relatively flat or gentleһ sloping terrain, somewhat close to rivers.

639

Figure 8.13: Graph comparing the normalized values of the different Ballcourt Indices.

To further evaluate if sites with ballcourts share certain characteristics in common, in comparison to other settlements or places on the landscape, I generated a random sample of feature points and measured their elevations, degree of slope and percent change in rise, and distance from water sources. Because these are random points and not actual sites recorded during survey, the size and structure count variables could not be included. As was done previously, I inverted and normalized the values to create an index. Again, the ballcourt sites fall in the middle of the index range (Figure 8.14), supporting the general conclusion that the sites where we find ballcourts are relatively accessible, on flat terrain with gentle topography, and near sources of water. When we compare the variables of the ballcourt sites against those of the random sample, the t-tests results show significant differences between the two groups with respect to accessibility (t(110) = -1.982, p =

0.050), degree of slope (t(110) = -2.139, p = 0.035), percent change in rise (t(110) = 2.059,

640 p = 0.042), and distance from annual or intermittent water sources (t(18.478) = -3.892, p =

0.001), with the latter variable being so different in each sample that equal variances cannot be assumed. However, when summed together as an index, there was no statistical difference between the random sample and the ballcourt sites, suggesting that certain individual variables are better predictors of ballcourt sites than others.

Figure 8.14: Map showing Ballcourt Index 4, comparing the random sample points against the ballcourts.

The results of the cluster analyses support the in-field observations that ballcourts were constructed at sites that varied in size but tended to be small residential settlements, were mostly complex and relatively accessible, and were located on more or less flat terrain

641 with gentle changes in topography. While most were constructed at habitational sites with other public architecture, some courts are found in isolated contexts and others were constructed in places that were less accessible and feature steep topography. When we compare these same scores by time period, we see there is little statistically significant difference between Classic ballcourt sites and Postclassic ballcourt sites. In the next section, I use logistic regression models to explore just how powerful the variables I identified are in explaining the distribution of ballcourts on the Nejapa social landscape.

8.2.1.1 The Logistic Regression Model: Predicting Where Ballcourts Could Be Found

Regression models in GIS are valuable tools as they predict attribute values that are tied to locations; in other words, it predicts the location of a class of phenomena. They are also useful in examining the relationships between independent and dependent variables to assess which factors exert more or less influence on site or feature location by comparing a control group against a study group. First, I evaluated the three assumptions modeled in the previous section: 1) site with ballcourts are large, complex and accessible; 2) moreover, they are located on relatively flat terrain; and 3) finally, they are also located close to sources of water. Here the control group contains the non-ballcourt sites, while the study group is comprised of the ballcourt sites. I then ran a series of binary logistic regression analyses using SPSS, as ArcGIS cannot perform these statistical tests, one for each set of assumptions comparing all sites to all ballcourt sites, Classic period sites to Classic period ballcourts, and Postclassic sites to Postclassic ballcourts. For every regression set, each variable was weighted for every set of assumptions by their relevant regression coefficients.

The results were summed with the constant (Y-intercept) to create the model using the following equation: 642

model = constant + weighted variable1 + weighted variable2 +…..weighted variableN

I then created the probability model using the following equation:

1 / (1 + exp(-model))

Figures 8.15 and 16 show some of the results of the logistic analysis for each series based on the three sets of assumptions. While many of the sites predicted by the models to have a ballcourt in fact do not, most of the sites that do have ballcourts were identified as potential court sites. To test the efficiency of each model, I used the following equation:

Efficiency = 1 – (percent of total sites within most likely category / percent of ballcourt sites within most likely category)

Figure 8.15: Map of the Model 1 results for the Classic period.

643

Figure 8.16: Map of the Model 1 results for the Postclassic period.

The efficiency score for each set of assumptions (i.e. the logistic regression models) are described in Table 8.3. The models prove to be more or less efficient at identifying potential ballcourt sites, at an efficiency rate of a little over 50%. Overall, there was little difference between the results of the models, i.e. which sites were identified as potential ballcourt sites and which ballcourt sites were correctly identified by the model. Model 2, which tested the assumption that ballcourts are found at large complex sites located on relatively flat terrain, proved to be the most efficient, although Model 3 is slightly more efficient for Classic period sites and ballcourts. Interestingly, the models work best for

Postclassic sites and ballcourts. This is probably because Classic sites with ballcourts were

644 somewhat more variable in their size and number of structures, such as the isolated site of

Llano las Casas, whereas ballcourts in the Postclassic are all found within the civic- ceremonial zones of settlements or neighborhoods. Thus the model is more likely to identify Postclassic ballcourt sites based on the assumptions built into the models.

However, these differences again are not statistically significant.

Table 8.3: The efficiency scores for each model comparing habitation sites and ballcourts overall and by time period. Series Efficiency Model 1 Model 2 Model 3 Average Habitation Sites / Ballcourt Sites 47.2% 53.8% 48.1% 49.7% Classic Sites / Classic Ballcourts 49.1% 56.2% 57.1% 54.1% Postclassic Sites / Postclassic Ballcourts 66.9% 68.3% 66.9% 67.4% Average 54.4% 59.4% 57.4% 57.1%

Next, I ran a logistic regression to evaluate the model created for comparing the ballcourts sites against a random sample. These variables were site accessibility, degree of slope and percent change in rise, and distance from waterways. Surprisingly, this model has about the same rate of efficiency as the previous models, at a rate of 57%. Therefore, we can reasonably expect to find ballcourts at relatively accessible sites showing some degree of complexity; and further, that sites with ballcourts will feature flat terrain with gently sloped topography that can accommodate these large structures with unusual architectural elements. Finally, while sites with ballcourts are located near to major rivers and/or annual or intermittent water sources, this a general feature of habitational sites in

Nejapa. On the other hand, these models have only a moderate efficiency rate. There are clearly other, more social variables that explain why the ballcourts were constructed at these particular settlements.

645

8.3 TO THE BALLCOURT AND BACK: LEAST COST-PATH ANALYSES, LANDSCAPE MOVEMENTS, AND THE OCCASIONAL FAILURE

Even though the ballcourts in Nejapa are located not far from each other, travel between them would have been affected by factors both social and physical. Path-distance modeling uses the costs or ease of travel based on slope to determine which sites would have been more accessible when moving to and from other sites39. The goal of the least cost-path (LCP) analyses was to evaluate which ballcourts are considered to be (at least according to the model) most accessible, and therefore preferred, to each non-ballcourt community based on the energy costs of travel to that ballcourt. Of course, not all of the sites recorded are contemporaneous; moreover, the PANT database includes habitational and non-habitational sites. In modeling which ballcourts people from other communities would have traveled to, these are important factors to take into account. Based on descriptions from PANT field reports, sites were coded according to type (habitational and non-habitational) and time-period (Classic, Postclassic, and unknown when no date was indicated) where the information was available. Separate feature classes were then created based on these attributes: habitational, non-habitational, Classic-period habitational sites, and Postclassic-period habitational sites. For modeling movement during different time periods, I also divided the ballcourts into two feature classes based on this attribute

(Classic, Postclassic). I then ran a series of LCP analyses modeling movement between the ballcourts and different origin points (Table 8.4), using different impendence surfaces and different combinations of these same surfaces (Table 8.5).

39 Critiques of LCP analyses and their uses in archaeology were discussed in Chapter 5. 646

Table 8.4: The LCP analyses Series 1 - 5 and the origin and destination points. LCP Analysis Origin Destination Series 1 All ballcourts All sites Series 2 All ballcourts Habitational sites Series 3 Classic ballcourts Classic habitational sites Series 4 Postclassic ballcourts Postclassic habitational sites Series 5 All ballcourts Postclassic habitational sites

Table 8.5: The original and adjusted slope values and their reclassified values as well as the original distance from river values and their reclassified values. Original Value Slope Reclassified Value Slope 0 – 5 1 5 – 10 4 10 – 20 16 20 – 30 64 30 – 75 1000

Original Value Slope Adjusted Reclassified Value Slope Adjusted 0 – 10 1 10 – 20 4 20 – 30 16 30 – 45 64 45 – 75 1000

Original Value Distance from River (m) Reclassified Value Distance from River 0 – 100 500 100 – 250 400 250 – 500 300 500 – 1000 200 < 1000 100

In the first step in LCP analyses is to create a basic impedance surface by reclassifying a slope surface generated from a 30 x 30 m DEM of the Nejapa region, using the values listed in Table 8.5. In general, any slope that is higher than 30 degrees is considered to cost too much energy to travel over, and will generally be avoided when possible (Minetti 1995). In order to account for the fact that travel in Nejapa would have required movement across more rugged terrain, I ran a second LCP set using adjusted

647 values for the slope impedance surface so that higher slope values would be less costly (see

Table 8.5). These values, of course, are arbitrary and work best as heuristic tools for exploring how people may have moved through mountainous landscapes and not definitive statements on what slopes are less costly to travel over.

Because LCP models search for and prefer cells with the lowest slope values, they tend to channel movement towards the valley floor and along the major waterways of

Nejapa. While rivers were certainly important modes of transportation during the

Prehispanic era, they also would have been more subject to restrictions upon movement, given their dual importance as sources of water for human consumption and agriculture.

Moreover, mountainous regions such as Oaxaca would have required people to travel over rough terrain on a regular basis. In order to account for the way people who lived in mountainous regions actually moved through them, we must model pathways that give preference to travel along the midlines of hill- and mountain slopes. One way to do this is by adding additional frictions surfaces to the original cost-impedance surface raster that make travel along or near riverways more “costly” and travel on the piedmont midline less so. Following methods used by Badillo (personal communication), for the next of set of

LCP analyses I created two additional friction surfaces. First, I created a waterways friction surface that allocates differential costs to those pixels containing waterways (n =

100) and those that do not (n = 0). Second, I then created a distance-from-waterways friction surface that assigned inverse cost values to different Euclidean distances from major rivers, with higher costs allocated to cells closer to the waterways (see Table 8.5).

Again, these values are arbitrary and heuristic tools only, but one thing that I did have to consider carefully was not to set the values so that travel between sites close to or along the river was excluded completely. I then used the Raster Calculator tool to create two new 648 friction surfaces: a third LCP set that combined the slope and waterways surfaces; and a fourth LCP set aggregating the three friction surfaces of slope, waterways, and distance- from-waterways. I then reran the LCP series analyses using these friction surfaces.

However, there were several issues immediately apparent when reviewing these results. The LCP models often failed to identify paths between sites with ballcourts and other settlements nearby; these problems increased with each series I ran using the three different sets of LCP impedance surfaces. For example, when we only consider Classic- period habitational sites and the ballcourts dating to this same period, the models did not identify any paths leading to Cerro Maluco from other contemporary habitational sites.

This is a result that more than likely does not reflect reality – the ballcourt at Cerro Maluco is simply too large and too complex to have only been constructed and/or used by the local community alone. At the same time, Cerro Maluco is a low hill (788 m asl) located on the southern side of the valley less than 2 km from the Río de La Virgen and about 500 m southwest of Tanaguixi, another Classic-period site with its own ballcourt (located at the site of Los Limares). People would not have avoided traveling there simply because the hill slopes were steep, especially considering that these settlements would have paved roads and stairs that would have allowed (and controlled) access to the sites, in addition to the paths that directed or facilitated movement between them.

When we compare the results of the three LCP sets, we immediately see some important differences. The first LCP forces most movement along the major waterways

(and likely some intermediate river drainages as well) and channels the movement from multiple sites through one site in order to reach the ballcourt, while the second LCP set shows several paths moving over higher elevations and rugged terrain, and even shows multiple alternative paths between several sites and how people may have accessed 649 ballcourts while avoiding travel through other sites. The results from third LCP model, unfortunately, do not show much improvement over the first two, and identify even fewer paths linking sites to ballcourts. In this set, the modeled paths travel through higher elevations where movement would have been still less costly while avoiding rivers. In a sense, these paths may likely be more accurate than those modeled in LCP set 1 since modern foot travel in the Nejapa subregion occurs along higher elevation footpaths. At the same time, many of these mountain trails will crisscross rivers, even major waterways, when necessary.

To this day, Oaxacans in rural areas travel to and over steep slopes in the course of their daily lives. During the 2015 field season, one of my field workers sent his brother as a replacement for a day or two while he traveled to his terreno or land. For most people, this would seem extreme – why would this amount of time be needed to visit land they own? I asked Don Taurino about this, and he explained that his terreno was located about 5 hours away from Nejapa de Madero walking, that there were no roads out there, and that to visit would require rough travel deep into the mountains and an overnight stay. This was a matter of course for him, and while he did not do it every day, those occasions when he did have to travel there were not seen as extraordinary instances, but rather routine occurrences in a series of cyclical actions. In another instance, my partner and I were hiking along some footpaths to the summit of a ridgeline near the town of San Jose del Pacífico, nearly

3000 m above sea level. On our way back down, legs sore and out of breath, we heard the bell ringing out the end of the day for the local high school. Another 1000 m further down, we passed a young schoolgirl, aged 14, on her way towards the same ridgeline we had just left on her journey home to one of the small hilltop towns we had seen on the other side.

She traversed the narrow and precarious footpaths, created through the sheer determination 650 of thousands of previous steps, in that automated manner one does when they have

(literally) taken the same path thousands of times before. In my discussions with pelota

Mixteca players, they often talked about how in the days before many of the roads had been constructed, their fathers and grandfathers would travel distances up to three weeks away by foot to attend games. No one viewed these travel times as extraordinary or excessive, but simply as a necessary part of making trips to these distant places.

While these examples are anecdotal, they highlight a large gap not only in Oaxacan research but even in other LCP analyses as well. While several authors have considered travel in mountainous terrain, they mainly discuss it in terms of the energetic or caloric costs of traversing over different slope grades (Balstrøm 2002; Minetti 1995; Pingel 2010;

Rees 2004; Whitley and Hicks 2003). Obviously, steeper grades require more energy, and people will tend to avoid or minimize travel over these areas when they can. But travel through mountainous areas is not simply accomplished via vertical movement. People also move laterally along the contours of hill flanks and mountainsides, and many hills in

Oaxaca are crisscrossed by switch-back dirt paths that allow people to gradually ascend and descend. Thus, while slopes above 30° may cost more with respect to energy, this does not mean that people will not regularly travel over these steeper slopes, especially when they have to, or simply have no other choice, in order to access food, water, and other needed resources. Even today, one can see agricultural fields carved onto the mountain sides that are easily 75° (and perhaps over) in slope40, with a small household or field house usually located on a nearby ridge. The intense occupation of the hilltops, ridgelines, and slopes in

Oaxaca in the past (and present) tells us that people did indeed traverse high-cost terrain in

40 The maximum slope value over which people will travel in traditional LCP analyses, and also considered to be the costliest in terms of energy (see Table 8.5). 651 their daily lives, not just via vertical passages but also in lateral movements that would have decreased the energetics or caloric expenses of travel when possible without sacrificing elevation. The Los Picachos settlement is located at a high elevation on a ridgeline that can be very difficult to transverse, yet it was still intensely occupied.

Closer to the settlements themselves, people would have modified natural pathways and constructed stairs to facilitate and channel movement, as I mention earlier. For example, access to the ridgeline of the cliffs nearby to Cerro la Muralla, where we find substantial evidence of occupation, is via a set of stone stairs set in a narrow crevice (Figure

8.17), whereas passage along the same follows tightly along dramatically steep drops. We do not have to look far for similar examples in other places – the cliff-dwelling communities of the American Southwest were accessible by hidden footpaths known only to the inhabitants (Matson et al. 2013; Whitehead 1990). While the context and motivations for occupying these spaces were not the same, the Southwest evidence shows that people will often move about difficult terrain when necessary or when that is the predominant geography. Though the exact travel routes have probably changed somewhat over time with shifting political alliances and settlement patterns, the inhabitants of Nejapa today move through the landscape much as their ancestors would have. People still visit ritual sites such as Piedra la Iglesia and Piedra la Boluda (YAU-TAV-TAV-009) following probably the same access routes used in the past. Thus, the likely paths used by the people of Nejapa in Prehispanic era lie somewhere between the models of LCP set 2 and LCP set

3. That is, people preferentially sought out paths higher in elevation that allowed them to move laterally through rugged terrain, but they would have also traveled along or across rivers when necessary.

652

Figure 8.17: Climbing the stairway in a narrow crevice to reach the ridgeline of the cliffs at Cerro la Muralla.

At the same time, we also must consider the possibility that sites with ballcourts were not meant to be accessible, at least some of them. Indeed, the failure of the LCP model sets to identify paths between ballcourts sites and non-ballcourts sites may be a result of people intentionally constructing courts in places that were not easily accessed by 653 people simply moving over the landscape. This can certainly explain why we do not see paths drawn for many of the sites located in the mountains, since they are located in areas that are naturally inaccessible and defensive. Certainly there is plenty of evidence globally and in Mesoamerica that sacred structures or sites were often constructed in places that are hard to get to and require a lot of labor to reach. Even within settlements, the design of buildings and the site’s spatial layout can restrict or make more difficult access to sacred places or features; many important temples in Maya altepetls have exaggerated stairs that made ascent to the top harder. The difficulty of getting to these places as well as their inaccessibility is part of the ritual process itself, thus heightening their sacred character and ceremonial power. This probably best applies to sites like Cerro del Convento and Llano las Casas.

At the same time, the results of the earlier spatial analyses discussed in previous sections demonstrate that with a few exceptions the sites with ballcourts are relatively accessible, using elevation as a proxy, especially those sites closer to the valley floor and located along the major rivers. The ballcourt in the Lachixoba neighborhood of Greater La

Amontonada, for example, would have been very accessible if we only considered the energetic costs of travel (today, it’s even easier to get to as a dirt road leads you directly past the court itself, Figure 8.18). Even those sites located on bluffs or hills but still on the valley floor would again have had paths, paved roads, or stairs to direct movement to and through the site for lesser cost than just simply climbing up vertically. Thus, while intentional inaccessibility may explain why the models failed to identify paths between the ballcourts and other settlements, it is clear that other factors are affecting movements across the landscapes and between sites.

654

Figure 8.18: The road that passes by the ballcourt at Lachixoba, looking towards the southwest.

One solution may lie in the GIS tool Path Distance Allocation (PDA), which produces rasters that identify zones where travel from one cell to another produces the least energy costs. Thus, it can be assumed that people would have preferentially traveled from their community to any sites, locations, or places that fall within these zones and back.

Because it does not draw a singular path between a predetermined location and destination, this tool may be more appropriate for mountainous areas with more variable topography such as Nejapa. I ran the PDA analysis for the Classic and Postclassic ballcourts and settlements confirmed to date to these time periods, using the four different impedance surfaces described earlier. Visually, the differences both between time periods and the different impedance surfaces is striking (Figure 8.19 and 8.20). For the Classic period, the results group together the ballcourts at Ejido de los Canseco and Mogotes del Panteón,

655

Cerro Maluco and Los Limares, and the palangana courts respectively – suggesting the people living at settlements falling within these same zones would have preferentially to those specific courts. The results support other lines of evidence which tie these courts together based on age, geographical proximity, and in the case of the mountain sites, the specific court type.

On the other hand, it appears that people living at those sites not falling in the same zones as the ballcourt would have experienced increased costs traveling to the ballcourts and may not have therefore traveled to those courts. The one exception is El Callejón de

Tejabana, whose inhabitants would have traveled to the palangana courts under the conditions of Impedance Surface 4. However, it is still difficult to definitively conclude that these allocation zones reflect reality or that people would not have traveled to courts simply because they are trying minimize the energy costs of travel. For example, Nejapa

Viejo appears in its own allocation zone in each result using a different impedance surface; but considering that it was the dominant settlement during the Classic period, people would have traveled outside this zone to access other sites, settlements, and places in the Nejapa subregion. Again, the PDA tool shows zones where movement would have incurred the least energy costs; people would still have traveled to other zones, just at an increased cost.

656

Figure 8.19: Results of the Path Allocation Distance analysis for Classic period sites and ballcourts, using the different impedance surfaces (1 = Mogotes del Panteón; 2 = Ejido de los Canseco; 3 = Llano las Casas; 4 = El Sitial; 5 = Los Chorros; 6 = Los Limares; 7 = Cerro Maluco).

The results for the Postclassic settlements and ballcourts are more difficult to

interpret, and show how least cost path analyses may noِt be entirely appropriate for understanding movement in mountainous regions. For example, assuming travel only occurred to and from the places where the accumulation energy costs were minimal, then the people living in the different neighborhoods of Greater La Amontonada would not have traveled to each other’s courts except under the conditions of Impedance Surface 4, largely due to the assumptions in both the GIS tool itself and each impedance surface, and the particular topography of the Greater La Amontonada settlements, which are found on or

657 near rivers spread across low-lying hills and extensions of piedmonts. In fact, the results show a more “fractured” landscape, with some sites falling into their own specific zones.

However, again given proximity and contemporaneity, people probably did play in each other’s neighborhood courts. Rather than limiting travel to only those places where the energy expended traveling was minimal, we should instead assume that people did travel between zones, just at increasing accumulations of energy costs. Obviously, the next step would be to measure the maximum amount of energy people would have been willing to expend to reach places in order to determine how far or to which ballcourts people actually traveled to in the past. Still, given the amount of time spent and distance traveled by pelota

Mixteca players to attend games both now and in the past – sometimes up to three weeks by foot before many modern roads were built – measuring maximum distances or where people traveled from and to based on energy costs alone may not be an appropriate metric for understanding which ballcourts people would have visited in Nejapa.

658

Figure 8.20: Results of the Path Allocation Distance analysis for Classic period sites and ballcourts, using the different impedance surfaces (1 = Mogotes del Panteón; 2 = Ejido de los Canseco; 3 = Llano las Casas; 4 = El Sitial; 5 = Los Chorros; 6 = Los Limares; 7 = Cerro Maluco; 8 = El Cucharital; 9 = Lachixoba; 10 = La Puerta; 11 = Terezona; 12 = Cerro del Convento).

Additional radiocarbon dates from future excavations at more sites in Nejapa should do much to help refine the chronological issues in the path-distance modeling, although it remains to be seen if more sites with dates will produce better results. Clearly, factors other than slope, energy costs, or proximity to waterways affected travel and influenced the paths people took to go from one site to another. Slopes and rugged terrain were and are not clearly not barriers to movement in Nejapa, and would not have hampered travel as much as traditional LCP models assume. Again, inaccessibility may have been an intentional

659 choice, especially for those courts in the mountains. Modern pilgrimages in Mexico demonstrate that the journey itself is a ritual process. Thus, the difficulty of the journey and the restricted access of the settlement and/or ballcourt would have underscored the sacredness of the court and heightened the power of the rituals or ceremonies performed there. Clearly other, more intangible social factors were probably at play as well.

Moreover, the limitation of LCP models in only allowing for one-to-many or many-to-one relationships severely limits our ability to really model paths of movements between communities with ballcourts and those without, as people were likely traveling to multiple different settlements to attend games, perhaps depending on the season or particular festival. Although PDA analysis can correct for some of these issues, there are still limitations because assumptions based on energy costs are built into the tool. Social and political factors, such as intentional inaccessibility via topography or social proscriptions, therefore had a much greater influence on how and where people traveled in Nejapa and which ballcourts they would have accessed.

8.4 SIGHTING BALLCOURTS ON THE LANDSCAPE: VIEWSHED ANALYSES AND RESULTS

Whenever I looked out at the surrounding landscape of each ballcourt I visited, what stood out to me the most was the seemingly expansive views afforded by the structure’s location. Indeed, it seemed as if the ballcourts had been specifically constructed in places where one would be able to monitor the valley floors and/or nearby mountains. But does that mean that the ballcourts themselves were visible to other, contemporaneous settlements, or even other ballcourts? The social importance of visibility, after all, is often based on what can and cannot be seen. If ballcourts were socially significant for Nejapan

660 communities, then we can assume that they should be visible from the places where people lived, or even places they considered important such as sacred sites. As I described in

Chapter 5, viewshed analyses attempt to replicate the sociocultural perceptions of environmental and archaeological phenomena in the past (Gaffney at al 1996; Llobera

1996, 2000; Maschner 1996; Pollard and Gillings 1998; Ruggles and Medyckyj-Scott 1996;

Wheatley 1993, 1995, 1996; Rennell 2012), and are therefore another useful way to investigate the placement of monuments and other social features on the landscape

(Connolly and Lake 2006:225; Wheatley 1995, Woodman 2000a, Fisher et al. 1996,

Llobera 2001). Several studies have shown that structures are often constructed in locations that either highlight or hide their visibility, depending on cultural attitudes towards buildings with specific social functions (Wansleeben and Verhart 1997).

For the analyses presented in this section, I calculated a series of cumulative viewsheds using the Visibility tool in ArcGIS, first starting with all habitational sites and then examining the site viewsheds by time period in order to see which ballcourts would have been visible to people at these locations. For each set of analyses, I used several parameters in order to model as best as possible what people would have been able to see, including such factors as observer height and the distance limits human visual perception.

By default, viewshed analyses in GIS model what theoretically can be seen, but do not take into account any factors (described in Chapter 5) that would have limited or affected visual perception. This can lead to the false perception that people could see landscape features hundreds or even thousands of kilometers away. Of course, we cannot account for all the different factors that affected or limited vision in Nejapa’s Prehispanic past, such as atmospheric conditions. Vegetation is often cited as an important factor affecting visibility for both observers and the objects they are viewing. However, we don’t usually have 661 access to this information, and can only interpret how landscapes might have looked in the past. For Nejapa, we estimate that populations were higher in the past, especially during the Postclassic, and that the land was more intensely occupied either for habitation or agricultural use. Therefore, the assumption can be made that there was less tree cover then than in the modern day.

In each set of analyses, the height of the observer at each site was set at 1.5 m, roughly the average size of Prehispanic peoples, while the outer radius of the viewshed was set at 5 km, the limit of human vision given the curvature of the earth. Additionally, the processing extent for the analyses was limited to the Nejapa study area. Although from certain locations, especially those at higher elevations, it is possible to see other regions or places outside of this area, for our purposes here we are only interested in what people can see within the Nejapa subregion itself. In order to further elevate the visibility of the ballcourts from the habitational sites, I generated 100 random points within the PANT survey area and ran a viewshed analysis using these points – are ballcourts just as likely to be seen from random points on the landscape as they are from the settlements where people lived? Finally, I ran a viewshed analysis using the locations recorded as sacred sites to test whether ballcourts would have been visible from other places that were culturally important. Because we have evidence for the long-term use and reuse of many ritual or sacred sites even into the modern day, and because we do not have dates for most of these sites, I did not include time period in this analysis. The same parameters were set with respect to observer height, human vision, and processing extent as in the previous analyses.

662

Figure 8.21: The cumulative viewsheds of the Classic period sites.

Beginning with the Classic period settlements and ballcourts, when we examine the viewshed results we see some interesting differences between which ballcourts would have been visible and which would not (Figure 8.19). Cerro Malcuo (1) does not fall within the viewshed of other settlements, and even within its own site it is fairly well hidden. This makes sense when we consider the physical topography of both the court’s location and the court itself. Views of the court from the north would have been blocked by the large temple on the north side of the plaza, while from the south people would have only seen the back of the south lateral mound, and then only once they were on the lower terrace. Access from the court would have been possible from the west, but like the south mound people

663 would have seen only the back of the west terminal mound as they were approaching it.

Moreover, the superior surface of the north lateral mound is nearly at level with the plaza surface, the central playing surface is sunken, and effectively the mounds are all of similar height (even though the east mound is much shorter in comparison) – making the interior

“invisible” until an individual was standing close by or on the mounds themselves.

However, we must consider that while only the backs of the mounds would have been visible to people, and perhaps only as they were actually moving closer to it, the knowledge of what those mounds represented would have still marked the structure as a ballcourt. The survey boundaries end less than 500 m away, so there might yet be Classic period sites to the southwest of this limit. If so then, the back of south mound may have been visible to them.

Similarly, the court at Los Limares (5) would have been hidden from the people living on the upper slopes and summit of the Tanaguixi settlement to the northeast.

However, there are terraces on the southwestern slope, and although there is no evidence for structures or other clear signs of habitation, people would probably have used them to view games and other activities on the court as it can be seen very well from here, especially at higher elevations. If the terraces were used for agriculture, then the people working them would have seen the court and perhaps even walked past it on a routine basis.

The court itself falls just outside the viewshed results to the southwest. Visibility from this direction was likely affected by the fact that the court was constructed on a modified terrace opposite from its outer edge, closer to the hill side. The terrace itself rises some 2 m above a drainage gully, after which the alluvial plain rises again in elevation. Thus, from other settlements the exterior of western lateral mound would not have been very distinguishable, and probably did not come into view until one was much closer or had ascended the terrace 664

(although the mounds are somewhat visible when a person stands in the drainage gully).

As with Cerro Maluco, the survey boundary lies not very far the site itself; if there are sites to the west and south of this limit it is possible that at least the western lateral mound was visible from this area.

In Chapter 6, I described Llano las Casas (3) as a ballcourt hidden away in the mountains and isolated from other sites. Indeed, finding it almost feels like stumbling upon a secret. The viewshed results support this conclusion. The site does not fall within the viewshed of any of the contemporary habitational settlements nearby, and as we will see later this trend continues even when we consider all habitational sites regardless of time period. The isolation of LLC is therefore purposeful, its hidden nature adding to the sacredness of the ceremonies and ballgames that took place there. The lack of other structures or signs of habitation meant that the site was used primarily for games and related activities. More than likely, people from nearby settlements traveled here as a part of pilgrimage rituals or because it was a neutral space for social engagement under hostile or friendly conditions. The ballcourt at LLC is therefore very unique due to its isolation, hiddenness, and distinct form.

In contrast, the courts at Ejido de Canseco (2), Los Chorros (4), and Mogotes del

Panteón (6) all fall within the viewsheds of Classic period habitation sites. Given the relative flatness and gentle sloping of the terrain surrounding this court, the ballcourt at

Canseco was highly visible as people moved about their daily lives. Of course, it is difficult to account for the structures that may have existed in the past but did not survive to the present day, and it is difficult to estimate what their height would have been and if they would have blocked views of the court. The foundation lines observed by PANT that once could be found around the court were rather simple, and the one surviving mound that may 665 pertain to the site is small as well. At almost 2 m tall, the lateral mounds of the ballcourt may have towered over the smaller surrounding structures. It is difficult to say how visible the ballcourt at Los Chorros would have been, since most of the present information on this court is drawn from survey descriptions. Although the court falls within the viewshed of the site where it’s located (given the high elevation of the court and the remoteness of the site, it would not have been visible to other settlements), without knowing what structures were nearby or the height of the court mounds we cannot say for certain what views people may have had, if at all.

The ballcourt at Mogotes del Panteón, on the other hand, appears to have been the most visible of all the courts in Nejapa during the Classic period, well within the viewshed of contemporary habitational sites south of the court. As I described in Chapter 6, the ballcourt sits on a prominent extension of piedmont near the Rio Grande, and is most visible from the vantage point of the river. Of course, what people would have been able to see of the court is the exterior of the south lateral mound, which rises some 4 m above the terrace below it. It would have appeared as a large mound of stones, but not much else of the court itself – such as the playing alleys or other mounds – would have been seen as the south lateral mound would have blocked people’s views. Part of the western mound may have been visible, but probably only within a certain distance range. There are also several vantage points from within the site itself where the court would have been visible from other directions, although only to those on the uppermost terrace in and around the ceremonial zone.

666

Figure 8.22: The cumulative viewsheds of the Postclassic period sites.

In the Postclassic period (Figure 8.20), we again see interesting differences between which ballcourts would have been visible to people as they moved about the landscape.

While the white limestone cliff face of Cerro del Convento (1) is one of the most visible features of the Nejapa landscape, it’s ballcourt would not have been. Obviously, the high elevation of the settlement on the summit of a mesa and the court’s location just west of the highest promontory point meant that effectively views of the court would have been blocked. One perhaps could have had views towards the court’s general location from other mountain sites, but it’s doubtful that the structure itself would have been visible or distinguishable. Given the topography of the mesa’s top and the other buildings

667 documented there, the court was probably only partially visible to the site’s inhabitants from certain vantage points. The ballcourts at El Cucharital (4) and Lachixoba (6), on the other hand, were much more visible not only to the people who lived in those settlements but from other vantage points and settlements nearby as well. The main court at El

Cucharital was probably especially visible given its location on the highest part of a prominent bluff, and to this day stands out as a defined feature on the landscape, particularly from the east. Again though, what people would mostly likely have seen and what would have been most impressive are the lateral mounds. For the people who lived in the other neighborhoods to the east, some of which were mixed-status and/or lower status, they would have seen the massive, earth and rock mounds daily. From the west, people would have been able to see the west lateral mound or at least the upper portion, but only from a certain distance. Closer to the bluffs, the ability to see the west lateral mound would have decreased until one had ascended to the top; in other words, at a specific point in the landscape as one approached the bluff, the upper portion of the west mound would have come in and out of view. Interestingly, the residents of the high-status barrio would not have seen the court as easily given intervening structures that would have blocked their view. The court would have become more visible as they moved closer to it from the south, and from the four mound platform group (about 54 m south) onward people could probably see into the court.

Like Ejido de Canseco, the ballcourt at Lachixoba lies directly on the alluvial plain.

Visibility of the court, therefore, would have been most affected by the height of any surrounding structures. However, we can make certain guesses about what people would have been able to see based on the structural evidence that does remain. The one surviving structure in any proximity is a low but long platform, likely indicating that the ballcourt had 668 been located in the residential public ceremonial zone of that neighborhood. To the south of that court, across the modern road where they now grow melons, were smaller households and milpas. Depending on the height and density of the structures that may have existed in the past, the very top of the lateral mounds may have been visible, or completely blocked. Still, given the proximity of these households to the court, people would have been aware of this structure and the activities that took there. In contrast, on the three hills to the north we find higher status buildings stretching across the upper hillsides and summits, including platforms, temples, terraces, and high-status residences.

On the surface of the platform I excavated on the western most hill, we found a fragment of red-painted stucco that likely came from a wall. From certain vantage points, the court mounds and perhaps even the playing alleys would have been visible, although the further east and north one moved along the hills the more difficult it would have been to see the court.

The second ballcourt documented at the El Cucharital site likely belonged to a different neighborhood, perhaps of mixed-status like Lachixoba and those areas of

Cucharital below the bluffs to the east. No surviving structures remain, so again it is difficult to determine what view people would have had of the court from there. Perhaps, like the court at Lachixoba, people either saw the upper part of the lateral mounds as they moved about the settlement, and would have seen more as they drew closer, or their view was blocked by intervening buildings but they were aware of the structure and the activities that took place there because of they lived in proximity to it. Interestingly, people living on the bluffs in the higher status neighborhood would have had excellent views of this barrio ballcourt they walked along the western edge; from some angles, given the orientation of the court and the lower terminal mounds, it probably would have even been possible to see 669 into the court itself.

The ballcourt at Terezona (12) is located on a lower western part of the hill summit, and its west lateral mound – or at least the upper portion – may have been visible from the valley floor below from the northwest, west, and southwest. Again though, we must consider the existence in the past of intervening structures that may have obscured views of the west lateral mound from below, that simply did not survive to the modern day

(especially after construction of the water tower and road). On the other hand, the downward slope west of the court, which drops 3 m only 20 out from the lateral mound

(with a percent change in rise increase from 7.5% to 10%) suggests that any buildings located on these slopes may not have blocked people’s ability to see the court mound. The court mounds would have been visible to the inhabitants of Terezona from several vantage points in the settlement, including from other monumental structures, while from other locations people would have been able to see directly into the court, especially from the south. Where the court would have been impossible to see is from the east and southeast – the upper platform to the east of the court effectively blocks all views from that direction, and the hillside slopes steeply from there

Although located in the upper piedmont region, the analysis shows that the ballcourt at La Puerta (7) would have been visible from nearby settlements, at least the exteriors of the lateral mounds. The court is the only structure on the small and narrow saddle connecting the southernmost peak to the other hills that make up the site, thus there would have been no other buildings to block views of the court. From the west, people may have been able to see the west lateral mound, but only at higher elevations – it is doubtful that the court could be seen from lower elevations or the river below. From the east, the exterior of the east lateral mound was probably visible at different elevations as one 670 ascended the hill. However, given that the lateral mound blends into the hillside, it may have been visible but not necessarily distinguishable. Within the site itself, views of the court’s playing alleys would have only been available from the terraces on the slopes of the east and west peaks facing the court.

As I have mentioned before, the dating for the ballcourt at Los Chorros is somewhat problematic – palangana courts have been dated to the Early to Middle Classic but radiocarbon evidence from Los Picachos indicates a Classic to Postclassic transitional period occupation. The results suggest that if Los Chorros and its ballcourt were in use during this time period, the court would have fallen within the viewsheds of nearby Los

Picachos and Loma de los Sitios. Residents on the summits and eastern slopes may have been able to see the court, but at a distance of around 800 m it may have been hard to distinguish; and if there were intervening buildings their views would have been blocked entirely. Even if the ballcourt was not distinguishable, people would have been able to see towards where it was and could easily point out its location. While most of the ballcourts in the Postclassic were visible to both residents and people from other settlements nearby, the main court at El Sitial (5) only falls within the viewshed of the site itself. Simply put, the local topography and high elevation blocked views of the court for residents from other nearby sites. Within El Sitial itself, the people who lived and moved about on the hill peak opposite from the court would have seen the exterior of the north lateral mound. On the peak where the court is located, as the largest structure there (apart from the probably- modified temple) it would have been highly visible.

There are several examples in Oaxaca of Classic-period ballcourts that were reused in the Postclassic periods, especially in the Mixteca Alta. Do we have evidence for the same in Nejapa? We do not know if Classic-period ballcourts fall within the viewsheds of 671

Postclassic Mixtec yuhuitayu, as these types of analyses have not be done there. If, however, visibility was an important factor, then it is interesting that none of the Classic- period ballcourts really fall within the viewsheds of Postclassic settlement, although

Canseco and MDP do fall right at the edge of the visibility threshold. However, I would argue that this is a result more of the parameters of the analyses, the local topography, and the location of these courts in places with high visibility. On the other hand, Mixtec

Classic period ballcourts may not fall within the viewsheds of the Postclassic sites, but there is still evidence for the reuse of these courts based on material assemblages and reconstruction episodes. Given a lack of evidence for post-abandonment reconstruction of the court or material remains dating to the later period, we can reasonably assume that the

Classic-period Nejapa courts were not reused in the Postsclassic. Instead, the evidence demonstrates that once these courts (and the settlements) were abandoned they ceased to be part of the social network, visually and materially.

672

Figure 8.23: The cumulative viewsheds of all habitational sites.

When we consider the viewsheds of all habitational sites, we see similar results to the viewsheds by time period with general increase in the visibility of the ballcourts (Figure

8.21). The court at Cerro Maluco (2), for example, becomes visible to residents living in settlements to the south, as does Terezona (12). The Los Limares (10) court, on the other hand, still remains just outside the viewsheds of nearby settlements. Visibility of the courts increases for Ejido de Canseco (3), La Puerta (7), and the neighborhoods of Greater La

Amontonada (4 and 6). The court at Cerro del Convento now falls within the viewsheds of two settlements to north but at the very edge of visibility, suggesting that it would still have been very difficult to see it. While the results show that the ballcourt at El Sitial (5)

673 becomes visible to the residents at Loma de la Oreja, views of the court are still limited only to the people living on and around the mountain where these sites are located. The greatest increase appears to be for the ballcourt at Terezona (12), which now is shown as visible to settlements to the southeast, El Maneadero,, and southwest, Terreno de Julio

Leon. Given that the former lies at a higher elevation (about 900 masl) than Terezona’s court (740 masl), residents at El Maneadero probably had a better view of the court from certain vantage points than the people at Terreno de Julio Leon. Finally, even when we add in additional habitational sites, the ballcourt at Llano las Casas (8) remains hidden from view, and would have been impossible to see from any nearby settlement. This underlines the isolated and hidden nature of the LLC court.

674

Figure 8.24: The cumulative viewsheds of the random points sample.

To further understand the relationship between visibility, the ballcourts, and settlements, I compared the viewsheds of habitational sites against the viewsheds from a random sample of points (Figure 8.22; Table 8.6). When we look at the results, we see some interesting differences between the two. The courts of the Greater La Amontonada neighborhoods (4 and 6) are still highly visible, suggesting that these structures were defined features of the Early and Middle Postclassic sociopolitical landscape. On the other hand, some of the sites that in previous analyses fell within the viewsheds of one or more nearby settlements are no longer visible, including Ejido de Canseco (3), La Puerta (7), and

Los Chorros (9). The visibility of El Sitial (5), Mogotes del Panteón (11), and Terezona

675

(12) is somewhat more restricted, especially in the case of Terezona, but still more or less the same. Finally, the most striking difference between the previous analyses and the random sample is that the court at Llano las Casas (8) is visible in the latter, although the viewshed is still somewhat restricted. Los Limares (10), which always fell outside of the viewsheds of nearby settlements, is also now highly visible as is Cerro Maluco (2).

Previously, Cerro Maluco was only visible when we included all habitational sites. The results strongly suggest that some of the ballcourts were intentionally built in visible locations while others were purposefully hidden from view.

In addition to identifying what would have been visible to a set of observers (known as Frequences), the Visibility tool in ArcGIS can also be used to determine what can be seen from an observer point and from where that observer point is visible to others

(Observers). Additionally, the tool also identifies what areas or points in the landscape different combinations of observer points could see and be seen from. This allows us to see not only what could be seen from the ballcourts, but also which ballcourts were visible to other courts, which courts have intersecting (overlap) or converging (meeting at common point or area) viewsheds, and from where one could see one or more ballcourts.

Intervisibility among contemporary structures, for example, may indicate social links between their respective settlements; while intervisibility among Classic and Postclassic courts may explain the locations for the latter, as people may have been intentionally making reference to ancestral settlements and past events. Intersecting or converging viewsheds may indicate that those places visible to one or more observers were socially and/or culturally important.

The resulting raster surface displays the viewshed from both each observer point, in this case the ballcourts, and from different combinations of observer points, i.e. areas of the 676 landscape that can be seen by one or more observers. The attribute table records the cells visible to each observer and the cells that are visible to combinations of different observer points. Thus, when we do not see recorded instances of overlap or cells visible in common to two or more observer points, we can assume that they were neither intervisible to each other nor had views in common. I ran two separate analyses evaluating the ballcourt viewsheds by time period. As in the previous analyses, I set parameters in order to best model what would have been visible to people; I used the same processing extent and visibility threshold as before, but this time for the observer height I added the external height of Mound 1 (north/east lateral mounds) to the average person height of 1.5 m. Thus, the analysis models the viewshed from the courts as if someone was standing on one of the lateral mounds. For those courts where we do not have measurements for the mound, I used the average external mound height for each time period (Classic = 1.6 m; Postclassic =

2.7 m).

The results show that in general there is not much intervisibility between the ballcourts in the Classic period, nor do their viewsheds overlap, intersect, or even converge upon important natural or cultural features on the landscape, i.e. sections of the landscape where the viewsheds intersect, such as sacred sites. The one exception is Cerro Maluco (1) and Los Limares (5) – the viewshed of the former is oriented south-southeast, while the latter’s viewshed is south-southwest, and both converge in the middle. Perhaps there is an important site or cultural feature located there that simply has not been recorded; on the other hand, this could simply be a coincidence. From this same location, both courts would have been visible to an observer standing there. Ejido de Canseco (2) and Mogotes del

Panteón (6) have the most expansive viewsheds of all the courts, but they were not intervisible to each other and there is only one small vantage point from which one could 677 see both courts. From the court at MDP, individuals would have been able to directly monitor movement along the Rio Grande as well as observe the piedmont zones and mountains in several directions. At the same time, people traveling along the river would have been able to see the impressive south lateral mound of MDP’s ballcourt. The results also suggest that people moving about the piedmont zone on either side of the river could also see the court.

Figure 8.25: Visibility map of where the ballcourts can see and be seen in the Classic period.

678

Table 8.6: Table showing the cells visible to observer points (Classic-period ballcourts) and sets of observer points (highlighted in gray). Mogotes Cerro del Los Ejido de Los Los Count Maluco Panteón Limares Canseco Limares Chorros Site 5 0 0 1 1 0 0 LL-EC 29 1 1 0 0 0 0 CM-MDP 41 0 0 0 0 1 1 LLC-LC 109 0 1 0 1 0 0 MDP-EC 160 1 0 0 1 0 0 CM-EC 194 0 1 1 0 0 0 MDP-LL 2341 0 0 0 0 1 0 LLC 3691 1 0 1 0 0 0 CM-LL 4289 0 0 1 0 0 0 Los Limares (LL) 7085 0 0 0 0 0 1 Los Chorros (LC) 9786 0 0 0 1 0 0 Ejido de Canseco (EC) 12220 1 0 0 0 0 0 Cerro Maluco (CM) 15346 0 1 0 0 0 0 Mogotes del Panteon (MDP) 1540448 0 0 0 0 0 0 Not visible

The views from the court at Ejido de Canseco are focused primarily towards the north-northeast and the southeast; these are also the areas from which the court could have been visible. However, the intervening topography blocks intervisibility between them for the most part. Still, it is interesting how the viewshed of MDP falls to the west, while that of Canseco falls to the west. As was observed from the previous viewshed analyses, Llano de las Casas (3) is hidden and isolated from the other courts and has a relatively limited viewshed, reflecting its location in a small, narrow plain next to an intermittent waterway.

Similarly, Los Chorros (4) is also isolated visually from the other courts, largely due to its location high up in the mountain range bordering the eastern edge of subregion, but its viewshed is more expansive than LLC and it would have been visible from different vantage points in the landscape – although perhaps not distinguishable as an individual structures. Views are predominantly towards and from the mountains to the south and east. 679

Finally, when we examine where sacred sites fall within the viewsheds of the courts, we see that they were not oriented towards these locations; thus their viewsheds do not converge upon specific, important places in the landscape.

In the Postclassic period, in contrast, we see much more overlap among the viewsheds of the ballcourts, although we do not see a concurrent increase in intervisibility between the courts (Figure 8.24, Table 8.7). Of the Postclassic ballcourts, Terezona (12) has the largest viewshed, with the greatest number of cells, i.e. the landscape, visible to individuals standing on the court mounds and from where the ballcourt would be visible.

This makes sense given the elevation of the court as well as the position of the hill itself.

The court’s viewshed encompasses a large swathe of the alluvial floodplain as well as the piedmont and mountain areas looking towards the northwest to the southwest. Individuals standing on or around the court on this side of the hill would have been able to monitor what was going on throughout much of the alluvial flood plain, especially the area around the confluence of the three rivers as well as the northwest and southwest entrances into the subregion. The results suggest that the court was visible from these areas; however, given the court’s position on the summit of a hill, it was more likely that people could see towards the court’s location but the structure itself was not distinguishable. The next largest viewsheds are El Sitial (5) and La Puerta (7), likely because of their high elevations.

The viewshed from La Puerta’s court is somewhat limited because of its position on a narrow saddle, but the results show that an individual would have been able to monitor movement along the piedmont on either side of the Rio Tehuantepec, while from the east lateral mound people would have had ample view of the mountains to the east. Today, this area is the border between Tavela and neighboring communities just to the north in the

Sierra Mixe. Would people have been able to see the ballcourt as the descended down from 680 the Sierra Mixe? It may have been possible to see the exterior of the east lateral mound from certain vantage points, but intervening vegetation may have often blocked visibility.

Figure 8.26: Visibility map of where the ballcourts can see and be seen in the Postclassic period.

681

Table 8.7: Table showing the cells visible to observer points (Postclassic-period ballcourts and sets of observers (highlighted in gray). La El Los Count CC Terezona CM Lachixoba Puerta Sitial Chorros CB Site 1 0 1 1 0 1 0 0 0 T-CM-LP 2 0 1 0 0 1 0 0 0 T-LP 2 0 1 1 1 1 0 0 1 T-CM-L-LP-CB 5 0 1 1 1 1 0 0 0 T-CM-L-LP 7 0 1 0 1 1 0 0 0 T-L-LP 13 0 0 0 0 1 0 0 1 LP-CB 30 0 0 0 1 0 0 0 1 L-CB 35 0 0 1 0 1 0 0 1 CM-L-CB 106 0 0 1 1 1 0 0 1 CM-L-LP-CB 118 0 0 1 1 1 0 0 0 CM-L-LP 129 0 1 0 1 0 0 0 0 T-L 193 0 0 1 0 1 0 0 0 CM-LP 193 0 1 0 0 0 0 0 1 T-CB 386 0 1 1 1 0 0 0 1 T-CM-L-CB Cucharital Barrio 677 0 0 0 0 0 0 0 1 (CB) 751 0 1 1 1 0 0 0 0 T-CM-L 826 0 0 0 1 1 0 0 0 L-LP 1010 0 0 1 1 0 0 0 1 CM-L-CB 1130 0 0 0 0 0 1 1 0 ES-LC 1454 0 1 1 0 0 0 0 1 T-CM-CB 2005 0 0 1 1 0 0 0 0 CM-L Cerro del 2872 1 0 0 0 0 0 0 0 Convento 3619 0 0 1 0 0 0 0 1 CM-CB 5279 0 0 0 1 0 0 0 0 Lachixoba (L) 6084 0 0 0 0 0 0 1 0 Los Chorros (LC) Cucharital Main 7268 0 0 1 0 0 0 0 0 (CM) 8361 0 1 1 0 0 0 0 0 T-CM 8748 0 0 0 0 1 0 0 0 La Puerta 11185 0 0 0 0 0 1 0 0 El Sitial 15920 0 1 0 0 0 0 0 0 Terezona 1517335 0 0 0 0 0 0 0 0 Not visible

While the viewshed of La Puerta overlaps with those of several other courts on the

valley floor, including those of the Greater La Amontonada community, the viewshed of El

682

Sitial’s court demonstrates that visually it was much more isolated. There is some overlap between what was visible to people at Sitial’s court and what would have been visible to people standing around the court at Los Chorros (9). PANT survey did not reach this area, so there could be sacred sites and/or settlements there from which both ballcourts, or at least their locations, could be seen. However, we must also consider that these courts are more than likely not contemporary, so the overlap in their viewsheds is probably not significant. Interestingly, although people standing on Sitial’s court mounds would have been able to see towards Los Picachos; the site falls just at the edge of the court’s viewshed.

Thus, while people at Sitial probably could not distinguish structures or individuals, if present, the location of the settlement would have been within view, and vice versa. El

Sitial, Los Picachos, and Majaltepec are often linked together in the minds of local

Taveleños. Perhaps, these settlements – along with Los Chorros – represent different occupations of the same community through time. The inhabitants of Postclassic Sitial may have built their court purposefully so that when playing the game and performing ceremonies they could also gaze upon their previous home.

The main ballcourt at El Cucharital (4) has the next most expansive viewshed, with views primarily towards the north from west to east as well as the southwest; views towards the east, on the other hand, were much more restricted. The viewshed of this court overlaps significantly with that of Terezona’s, particularly around the confluence of the three rivers.

People standing at this confluence would thus have been able to see both courts from this vantage point. Although there is some intersection between the viewsheds of the main ballcourt at Cucharital and the ballcourt at Lachixoba (6), people standing on the lateral court mounds of the latter would not have been able to see the courts in the other neighborhoods – indeed the bluff effectively blocks intervisibility among these courts. 683

Instead, people at Lachixoba would have been able to see somewhat to the south and north with an extensive view towards the east. From this same court, people would have been able to see La Puerta; though the results suggest that the ballcourt and much of the site itself would have been visible, but more than likely structures would have not been distinguishable. Moreover, the viewsheds for Lachixoba and La Puerta overlap on the former’s court, indicating that people at the latter could see the former – although again we cannot assume that individual structures were actually distinguishable. Still, it is interesting these are the only courts in the Postclassic courts that are intervisible.

The viewshed of the second ballcourt at Cucharital (4) overlaps with those of several other courts, although most of these are rather small and not visually meaningful as they converge upon small portions of the landscape. On the other hand, the results show that there was some of convergence among the viewsheds of the three courts of the Greater

La Amontonada community, areas in the landscape that were visible to people at the courts and where the courts would have been visible. There is even one vantage point where it would have been possible to see all three courts at one time. The courts themselves were

noِt intervisible to each other, with the exception that people could see the second

Cucharital ballcourt from the main court on the bluff. From the courts, the views to the east would have been somewhat limited by the bluff on this side, but people would have been able to see the alluvial lands from north to south looking west, especially around the confluence of the rivers, and vice versa. Finally, the viewshed from the ballcourt at Cerro del Convento (1) is extremely limited, and does not intersect or overlap with the viewsheds of any other courts. People standing on the mounds of this court would have had a narrow sliver of view towards the north and south; views towards the west were blocked by the naturally topography of the mesa. Likewise, people may have been able to see towards the 684 court, but not the structure itself, along this is narrow line of site. Much like Llano las

Casas in the Classic period, this court is hidden and isolated. Once again, the “invisibility” of the court and the difficulty in accessing the summit of Cerro del Convento likely highlighted the sacredness of this important place of ritual and pilgrimage.

As I explained earlier, the Observers analysis records those cells that the observer and one or more other observers can see in common. Table 8.8 shows the number of times that each court’s viewshed overlaps with another. Looking at the new cell counts, we see that the ballcourts at Terezona and the elite barrio of El Cucharital have the largest viewsheds, followed again by El Sitial and then the courts at Lachixoba and La Puerta.

Interestingly, the greatest number of overlaps are between the ballcourts of Greater La

Amontonada, La Puerta, and Terezona, especially between the main ballcourt at El

Cucharitaandl and Terezona. The results show that while these ballcourts were not necessarily intervisible to each other, there were many vistas in common from where people standing on or around these courts could see and the courts could be seen (from these same points in the landscapes). Much of the overlap, or the points of visibility in common, falls around the confluence of the rivers and the alluvial lands surrounding this area; in other words the viewsheds are directed towards and from the west running north to south, while views to and from the eastern side of the study region are more limited. Of course, this is partly a result of the natural topography, which slopes upward moving east towards the eastern edge of the subregion. At the same time, it could also be a result of where we have done survey, as there may yet be more ballcourts in those areas where we have not surveyed that may have viewsheds directed towards the east. Still, it is interesting that the river confluence and alluvial plains to the west are both where people would have been able to see the ballcourts as well as the area that would have been most visible to the 685 courts.

Table 1. Table showing the combined cells visible to ballcourt sites and the number of viewshed with other ballcourts.

Site No. of Overlaps Overlap Sites Total Cell Count Cerro del Convento 0 None 2872 El Cucharital Main Barrio ballcourt, 25314 Ballcourt 14 Lachixoba, La Puerta, Terezona El Cucharital Barrio Main ballcourt, 7525 Ballcourt 10 Lachixoba, La Puerta, Terezona El Sitial Los Chorros 12315 1

Lachixoba El Cucharital ballcourts, 10654 12 La Puerta, Terezona La Puerta El Cucharital ballcourts, 10056 11 Lachixoba, Terezona Los Chorros El Sitial 7214 1

Terezona El Cucharital ballcourts, 27211 11 Lachixoba, La Puerta

8.5 SUMMARY AND CONCLUSION

The viewshed analyses described here reveal some interesting patterns not only regards to the spatial dimensions of the courts, but also their locations within the Nejapa subregion, the types of sites in which they were constructed, and their visibility from other settlements and/or different places in the landscape. Ballcourts on or near the valley tend to be larger and have larger mounds than those in the mountains, but these dimensions were still subject to local needs and tastes. Central alley lengths and widths, on the other hand, show no spatial patterning and all courts fall within a narrow size range. Ballcourts tend to be located at lower elevations and near sources of water, although this describes many sites in Nejapa. What does stand out about where ballcourts are located is that they were

686 constructed in areas of the settlement that were relatively flat with gently sloping topography.

With a few exceptions, most courts were constructed at sites that on average were small, relatively accessible and complex residential settlements or neighborhoods where the terrain was gently sloping and more or less flat. Most courts would have been located in the public ceremonial centers. Other courts are found in isolated contexts or in places that were less accessible and feature steep topography. When we think about accessibility in terms of the costs of travel, then many of the courts would have been inaccessible to people from other settlements, especially those in the mountains. However, the limitations of least-cost path analyses and the assumptions that are built into these models preclude us from making definition conclusions about travel between ballcourts for games. Finally, while some of the ballcourts were visible from other settlements, most were not. Ballcourts were only visible from certain points in the landscape, and there is not much evidence for intervisibility among them. Interestingly, the viewsheds noticeably change from the

Classic to Postclassic period, reflecting perhaps changes in both settlement distributions and what people wanted to be visible from and to the court. I explore the results, patterns, and their significance in the following chapter.

687

CHAPTER IX

DISCUSSION AND CONCLUSIONS

“They’re all getting screwed you know…you know what THEY care about? They care about the ballgame…” Mark Baum (Steven Carell), The Big Short 2015

9.1 INTRODUCTION

When I originally began research on the ballcourts in the Nejapa subregion, the density of the courts seemed distinct – a cluster of structures located primarily in and around the rivers on the alluvial plains, with some found in the mountains. Given the proximity of the courts, it appeared that the ballgame may have played an unusually important role in local politics, and Nejapa looked like an excellent opportunity to test my interpretations that the ballgame played a communal role in community relationships.

However, the qualitative descriptions of the ballcourts and their site and settlement contexts

(Chapter 6), and the statistical (Chapter 7) and spatial (Chapter 8) analyses of the mapping data present a more complex narrative of ballgames, the ballcourts, and their changing roles in the local and regional sociopolitical relationships of Nejapan communities and their neighbors. In this chapter, I bring together the qualitative descriptions and quantitative results and set this information within the broader historical context of the Nejapa subregion, the eastern Sierra Sur, and Oaxaca in general,

688

I began by discussing the results from Chapters 7 and 8, answering the question: are there observable patterns associated with the ballcourts in Nejapa in respect to their attributes and distribution. Even the lack of observable patterns, or instances where we see mostly differences, can give us insight in the importance and meaning of the ballgame and game ceremonialism, or the absence thereof, to Nejapan communities. Based on these patterns, whatever form they may take, I then interpret the data within a historical context to consider what strategies of ballcourt emplacement and ballgame practice were used by

Nejapan communities more broadly, expanding on the ballgame history described in

Chapter 6. Sometimes these strategies may have been more internally directed, with the purpose of negotiating relationships among different social groups in a community, rather than interacting with neighboring settlements. Finally, I expand the discussion to consider what local patterns and strategies can tell us about the ballgame, social relationships, and political organization in the Nejapa region and the southeastern Sierra Sur overall

9.2 GAMES AT THE FRONTIER: EXPLORING THE BALLCOURTS OF NEJAPA

The qualitative descriptions of the ballcourts in Chapter 6 documented the individual variability of each structure and the many unique architectural elements present at each. As individual structures then, the Nejapa ballcourts appear to be highly varied and a reflection of local decision-making, tastes, and needs. On the other hand, the statistical results presented in Chapter 7 show several patterns with respect to their size dimensions that tie the courts together into what could possibly be a shared ballgame tradition, similar ways of playing, and/or that communities were drawing on similar references for constructing their courts. Chapter 8 revealed yet further patterns of similarities and

689 differences in the spatial distribution of the courts on the sociopolitical landscape of

Nejapa.

I have previously discussed the qualitative and quantitative results in Chapters 6 through 8. In this section, I summarize and interpret the results of those previous chapters, exploring the significance of these observed patterns. I begin by examining specific themes that stood out when I was exploring the data, namely a particular structure arrangement associated with east-west oriented ballcourts; the examples of non-I shaped present in

Nejapa; and the two case studies of settlements with multiple courts. Then, picking back up the preliminary history of the ballgame sketched out in Chapter 6, I discuss the results of the analyses against the social and political backdrop of the Classic and Postclassic periods in Nejapa.

9.2.1. East and West, North and South: Ballcourt Orientation and Significance

While the statistical results (Chapter 7) showed no significant patterning with respect to court orientation nor differences between north-south and east-west courts, we do find other evidence that marks four of the east-west courts apart from the others in the sample41. First, the ballcourts at Cerro Maluco, Mogotes del Panteón, and La Baeza appear in a particular arrangement of group architecture. The arrangement of structures is as follows: a plaza bordered on all four sides by four structures with a small square structure more or less in its center. Three of the structures are temple or platform mounds, with two always larger than the third; the fourth structure is a ballcourt. Apart from these strong

41 Although also oriented roughly east-west, we find no similar arrangement of structures in association with the ballcourt at Ejido de Canseco. It is possible that such structures did exist at one time, but have since been destroyed by agricultural activities. 690 similarities, there are also significant differences. At Mogotes del Panteón and Cerro

Maluco, the plaza is located north of the north lateral terminal mound. The two larger structures are found on the north and east sides of the plaza, with the smaller third structure on the west. The eastern structure at Mogotes del Panteón was most likely a temple, while the north structure is an elongated platform with a set of stairs on its west side permitting access to the plaza. The square structure is a one-course stone platform attached directly to the north lateral mound of the ballcourt.

This arrangement is somewhat different at Cerro Maluco. Here the east structure is an elongated platform with two fire pits at its southern end, and the north structure is an immense temple likely modified from the bedrock. The square structure is just west of center in the plaza, and is a stone-lined foundation open at the western end (the west structure is similar to the one at MDP). In contrast to the first two, the court is on the north side of the plaza at La Baeza. The eastern structure is a rectangular building that had stone walls at least 2.5 m in height. Here the smaller structure is on the south side of the plaza, while the western structure is a large, terraced platform with a room and possible patio group in its surface. The square structure in the center is a stone foundation, taller than the platform at MDP, and of the three most clearly resembles a standing altar. No similar such arrangements of structures occur in association with north-south oriented ballcourts in

Nejapa, although we cannot preclude that such structures may have existed in the past.

MDP’s court roughly dates to the Early Classic, while La Baeza and Cerro Maluco are estimated to be Late Classic settlements.

The fourth east-west oriented ballcourt at El Sitial also has structures arranged around; the pattern is somewhat similar to the other three but is also very distinct. There is a possible plaza just north of the north lateral mound. The adoratorio is at the west side of 691 the plaza rather than in or near the center. Opposite of the north lateral mound is another formation that could be a structure or modified hilltop utilized as a large temple, although it is not clear at the moment and may in fact just be a natural feature. In contrast to the first three examples, there is no structure on the east side of the plaza. Instead, there is a terrace edge that meets the court at an angle and has a scalloped appearance. At the southeast corner of the court there is a rectangular building similar to Str. D2 at La Baeza. The court at El Sitial is also distinguished by its architectural design elements that, as I mentioned previously, include more of the traditional ballcourt features observed in Central Valley structures and elsewhere.

The differences in the arrangement of structures between the main El Sitial ballcourt and the three other east-west courts could be due to the location of the former on the steep mountain side. Space limitations may have forced the inhabitants to rearrange the architectural structures as needed even as they attempted to keep close to this spatial configuration. Another explanation could simply be time; the first examples date to the

Classic period (although Baeza does continue into the Early Postclassic), while El Sital’s court is Late Postclassic in age. Though Baeza’s court is on the north of the plaza instead of the south, the overall similarities among the three courts at Mogotes, Maluco, and Baeza would seem to tie them closer together. At the same time.

Given the small sample size, we do not have enough evidence to definitively call this architectural arrangement among the east-west oriented Nejapa courts an identifiable ballcourt group type, as they have been able to do at Cantona. Moreover, the three courts are at the very least separated in construction and likely use by hundreds of years. Still, the similarities are striking, and may even suggest that the builders of the latter courts may have been referencing the earlier structure at Mogotes. It is not quite known at this juncture 692 what the significance is of the similarities or subtle differences between the structural arrangements of the four east-west ballcourts in the Nejapa region.

9.2.2 Non I-shaped Ballcourts in Nejapa

As a reflection of Nejapa’s position as a place of international movement and exchange, we find ballcourt forms that are not all together common in Oaxaca. Terezona has a “T” -shaped court, but in contrast to those previously documented elsewhere in the

Maya Lowlands, Northwest Mexico, and the Isthmian site of Tres Picos, this example is a sunken court with a “U”-shaped terminal structure that connects to the lateral mounds, at least on the east side. Why there is a T-shaped court in Nejapa is not clear at the moment.

The community closest to the Nejapa subregion with documented T-shaped courts is Tres

Picos in the eastern Isthmus, but this settlement dates to the Late/Terminal Formative and we estimate that Terezona is Postclassic in age, based on surface finds. Of course, we still do not know if there are T-shaped courts in in the north of Oaxaca. The few courts documented in the Cuitlácan Cañada and the Sierra Mixe subregions are I-shaped, but that does not preclude the possibility that they may be there and in the Papaloapan region (the

Sierra Mazateca and Chinantla subregions). Thus, it is unclear at the moment why the inhabitants of Terezona constructed a ballcourt of this form, particularly when the region closest to Nejapa where we frequently see T-shaped courts is the Maya Lowlands. We must also consider the possibility that this was simply a design choice of the people who built the court, and that its form has no larger meaning or connection beyond local tastes and needs. Leaving the south end open may have just been a way to open up viewing space

693 for watching the game. Despite being of a different form, Terezona’s court still shares similarities to other courts in Nejapa, most especially in its size dimensions.

Another non I–shaped, and definitively confirmed, ballcourt form we see in Nejapa is the palangana. Two courts have been securely identified as palangana types, while the court at Los Picachos is more than likely a third, based on current descriptions. All three are located in the mountainous region bordering the southeastern edge of the study region:

Llano las Casas (948 masl); El Sitial (1121 masl) and Los Chorros (2 046 masl). As I previously stated, El Sitial is only 1.22 km northeast from Llano las Casas, while Los

Chorros is located 5.17 km to the northeast of El Sitial. The palangana court was a type specific to a certain time period, the Middle Classic, and a particular geographic region; it may even have hosted a special ballgame that was popular for a while before falling out of favor. The location of the palangana courts here in a specific area of the Nejapa subregion must surely be significant. The settlements where we find these courts, would have been accessible to each other via mountain trails and passes – indeed, the path one takes to Llano las Casas also leads to another that takes you to the western access of the El Sitial settlement, where one finds that site’s palangana court, although we cannot at the moment say with any certainty that they were contemporaneous. These settlements were not necessarily near trade routes or travel corridors, although based on material assemblages they would have had access to trade goods. In fact, they were rather isolated from other settlements both geographically and with respect to their visibility.

Palangana courts have been reported previously in north and central Oaxaca, with most occurring in the Mixteca Alta close to the border with the Tehuacan Valley. Thus, it is not entirely unusual to see them here in Nejapa, although their presence is still interesting when we consider the fact that outside of these areas, palangana courts have not been 694 observed or documented elsewhere in Oaxaca. Outside of Oaxaca, palangana courts have only been found in the Maya highlands and Chiapas. It is tempting to speculate that the appearance of palangana courts in Oaxaca where they are has to do with the international trade routes linking this macroregion to the Maya macroregions. We know historically that trade routes ran from the Basin of Mexico through modern Puebla and the Tehuancan

Valley, and on into Oaxaca via either the Mixteca Alta or the Cuicatlan Canada to the

Central Valleys, then through the Nejapa subregion and the Isthmus before arriving in the

Soconusco region with access to the rest of the Maya macroregion. It is interesting that the few palangana courts found in Oaxaca fall close to or along these trade routes, and may point to the spread of this court type and its attendant game via these same exchange networks. Nejapa therefore may be the subregion linking the palangana courts in northern

Oaxaca to those in the Maya region, and explain how this court type may have spread between these two macroregions. A small portion of the obsidian evidence documented in

Nejapa does come from sources in the Guatemalan highlands, so it is not entirely out of the realm of possibility. However, at this stage it is difficult to make a definitive conclusion as to why we have these unusual court types – and both types more common in the Maya macroregion than elsewhere – here in Nejapa. Future survey and excavations can help to resolve these questions.

9.2.3 Communities with Multiple Ballcourts

There are two communities in Nejapa with multiple ballcourts: Greater La

Amontonada, with three confirmed and a possible fourth court that existed still at some point within recent memory; and the two ballcourts at El Sitial. As I described in Chapter

695

4, Greater La Amontonada was a dispersed, urban settlement with multiple residential neighborhoods dating to the Early to Middle Postclassic periods (CE 1042-1400). It could certainly be considered a regional center in Nejapa during this time as its ceremonial core features the largest pyramid structure in the area, and the higher-elite status neighborhood had public monumental architecture not seen elsewhere. Each neighborhood had their own self-contained administrative areas, with larger public architecture surrounded by residences, and were small enough that daily face-to-face interactions among residents were possible (Konwest 2017). While there is no ballcourt in the civic-ceremonial center – with its giant pyramid, multiple plazas, and large platforms – there is the main court on the north side of that higher-status elite neighborhood (El Cucharital), with views towards the main ceremonial center. Another, more informal court existed on the western slopes of the same extension of piedmont as the main court, and would have been part of the ceremonial precinct of that neighborhood. The third court at Lachixoba was also a neighborhood structure, and was probably part of a larger ceremonial precinct that is no longer present.

Locals also report that there may have been another possible court at the site of Guitihuini that has since been destroyed. What each neighborhood has in common is their location along the smaller, seasonal tributary Rio Guitihuini (which eventually joins with the Rio

Tehuantepec). Konwest (2017) argues that residents of Greater La Amontonada likely had a fair degree of autonomy and decision making when it came to constructing and arranging not only their houses, but also their inner neighborhood public spaces, including the ballcourts.

As mentioned earlier, El Sitial (CE 1412 – 1619) is another site with two ballcourts

– the palangana court in a nearby neighborhood, and the main court, an I-shaped structure on the saddle between two hilltops. The civic-ceremonial center of the settlement is 696 dispersed across these hilltops and the saddle, with plazas, temples, platforms, and even a palace on the hilltop opposite the ballcourt. Both Greater La Amontonada and El Sitial have Middle Postclassic occupations; although the radiocarbon dates do not overlap, they could still have been occupied simultaneously for some time, or people may have been the process of abandoning the former as others were reoccupying the latter. While we cannot determine for sure what types of ballgames were played at each court at El Sitial, it is interesting nonetheless that there are two different courts reflecting what might be distinct ballgame traditions found within the same settlement. Depending on when they were used, these ball courts may have served some integrative purpose between the different neighborhoods, and possibly even between these neighborhoods and other communities.

However, it is also possible that the area of the site where the palangana is located may have been abandoned by the time the main ballcourt was constructed; without secure dates from the former court or structures nearby, we cannot say for certain that they were in use at the same time. More than likely, the palangana court is linked to the earlier, Classic- period occupation; certainly the Middle Classic date we have for El Sitial fits in with the generally accepted dates for the construction of palangana courts. At the same time, while the palangana court may be an earlier structure, it could have still been used well into the

Postclassic when the larger, more formal I-shaped court was built. Future excavations should help to resolve this chronological issue.

It is interesting to consider the differences between these two communities with multiple ballcourts. Greater La Amontonada was a regional center located on the alluvial plains, piedmont extensions and hilltops near the Rio Tehuantepec, with ballcourts that were contemporaneously used and closely resemble each other in general appearance and size, i.e. enclosed, I-shaped with sunken playing surfaces . El Sitial, on the other hand, is a 697 high-elevation, mountaintop site that has two dramatically different courts that more than likely represent two separate occupations and were used hundreds of years apart, although again we cannot discount the possibility that the earlier palangana court was reused. Why these communities in particular have multiple courts remains to be seen, but are surely related to distinct motivations and uses of the game. The multiple courts at Greater La

Amontonada are likely linked to the community’s prominent geographic and political position during the first half of the Postclassic. Moreover, Greater La Amontonada’s neighborhoods were populated by people of mixed-status who constructed their residences and public monuments in their own styles while also having access to the same local and interregional trade networks. The courts likely worked to integrate a heterogeneous population together, while also providing several public spaces for hosting larger events to which people from other communities may have been invited. El Sitial, on the other hand, is much more isolated, relatively speaking, from other contemporary communities and would have been much more difficult to access. Ballgames hosted at the courts here would have carried different connotations than those that took place at Greater La Amontonada simply due to its elevation and location. Moreover, this court was most likely used in the

Late Postclassic, when we see a contraction of courts to defensible and restrictive mountain top sites, at a time of heightened conflict or increased tensions as imperial armies moved through the Sierra Sur. Even if people from other communities had been invited to attend games in either of El Sitial’s courts, the journey to these places would have changed the meaning of these games and what role they would have played in local sociopolitical relationships.

698

9.2.4 The Ballgame and Ballcourts in Nejapa Over Time: Shared Traditions, Localized Expressions

The statistical analyses show that across many of the different court dimensions there is a consistent preference for specific size-ranges. Although not all of the ballcourts’ individual measurements fall within these ranges, they are not so far off as to stand out from the rest as unusual. In other words, we do not have two distinct populations of large ballcourts and small ballcourts, but courts with measurements that cluster around similar values. Of course, there is the exceptionally-sized court at Cerro Maluco, which again is one of the largest in Oaxaca. But while it is unusual in its overall length, central alley length, and the length of the lateral mounds, in many other dimensions it falls within the range of the other courts. Importantly, there appears to be a preference specific size range for the central alley lengths and widths, which along with the terminal alley lengths and widths and the interior slopes of the lateral mounds have been identified as the specific dimensions identified that most influenced game play. Even the court width for the immense Cerro Maluco falls within this size range (w = 8.6 m). At the same time, while we see a strong correlation between overall length and width, there is no correlation between the central alley lengths and widths. Proportions of structure lengths and widths can also indicate degrees of similarity among architecture even when individual measurements may vary. The L/W ratios of the overall and central alley lengths and widths of the Nejapa courts also cluster close together. People who constructed the courts were probably referencing earlier or contemporaneous examples, thus building their courts with a particular set of proportions, knowingly or unknowingly, in mind. Thus, there appears to be both a preference for specific size ranges and evidence that the size dimensions reflected the local needs, tastes, and/or the spatial constraints of where they were constructed.

699

These patterns continue when we consider the spatial distribution of the ballcourts.

The visual maps from Chapter 8 revealed that so far the ballcourts are found in the center and eastern areas of the Nejapa subregion, primarily along the Rios de la Virgen and

Tehuantepec, or in the east-southeastern mountain range. However, there may yet be ballcourts in areas that have not been surveyed that may change this pattern. Five of the courts (six if we include La Baeza) are found high elevations, most are close to or on the valley floor. These latter courts also tend to be larger, with greater overall lengths and widths and bigger lateral mounds on average than the mountain courts. On the other hand, when we look at the central alley lengths and widths, a measurement considered important to game playing practices, we see much more similarity across all the courts no matter where they are located. This supports the results of the statistical analyses that there were shared ways of playing or at least shared references on which people drew to build their court. When we consider the types of sites where ballcourts are found, we observe that they primarily were constructed in small to medium, civic-ceremonial sites, these sites vary in their size, complexity, and accessibility – ranging from large neighborhoods to small towns. There are a few exceptions. At one end of the scale is the site of El Cucharital, the largest and most complex of the sites and a highly visible ballcourt; at the other is Llano las

Casas, which stands out because of its lack of occupation and other structures, complete isolation from other settlements, and hiddenness from view. The court at Los Limares is also an interesting case because while there is no occupation near the court itself, the structure was likely part of the hilltop community of Tanaguixi.

Which ballcourt people may have traveled to from other sites is not very clear if we consider only the natural topography of the area as a cost to movement. Ethnographic evidence demonstrates that steep slopes are not a detriment to people’s ability to move 700 about their landscape. Older men from the villages in Nejapa still get up early in the morning to travel kilometers to their milpas that are often located higher up in the piedmont. Importantly, as I discussed in the previous chapter, much of this travel occurs laterally and at higher elevations where people walk the ridgelines following topographic contours. That we find settlements in the mountains and lower on the alluvial plain with similar material culture and architecture styles shows that these communities interacted with each other to a high degree. Thus, movement did occur among these sites in ways that the LCP analyses cannot capture. Other, more intangible barriers must have acted on where and when people traveled. At the same time, the lack of paths may have been purposeful; certainly we can argue that some of the ballcourts, especially those located in the mountains, were meant to be inaccessible. This can certainly explain the location of courts at Cerro del Convento and Llano las Casas, both ritual sites that would have been difficult to access and have courts “hidden”, so to speak, from view.

While it is not improbable that people from one community only used one ballcourt, or only had access to one, it is also likely that people traveled to different ballcourts depending on who was hosting games. This was most the likely the scenario for the Early to Middle Postclassic, when we have the greatest number of courts at sites either occupied contemporaneously or overlapping in chronology. At other times in Nejapa, there may have only been one or two courts in use at a time, a situation that probably occurred in the

Classic where there may be centuries separating each ballcourt, and in the Middle to Late

Postclassic when only the ballcourts at inaccessible mountain settlements were in use.

The results show that in general there is not much intervisibility between the ballcourts in the Classic period, nor do their viewsheds overlap, intersect, or even converge upon important natural or cultural features on the landscape, i.e. sections of the landscape 701 where the viewsheds intersect, such as sacred sites. There is much more overlap among the viewsheds of the ballcourts in the Postclassic period, but no concurrent increase in intervisibility. However, while these ballcourts were not necessarily intervisible to each other, there were several areas of the landscape where the courts could “see”, really the people standing on or around these courts, and could be seen. Where we see the most convergence or intersection in visibility is between the ballcourts of Greater La

Amontonada, La Puerta, and Terezona. This intersection falls around the confluence of the rivers and the alluvial lands surrounding this area. Of course, this is partly a result of the natural topography, which slopes upward moving east towards the eastern edge of the subregion. At the same time, it could also be a result of where we have done survey, as there may yet be more ballcourts in those areas where we have not surveyed that may have viewsheds directed towards the east. Still, it is interesting that the river confluence and alluvial plains to the west are both where people would have been able to see the ballcourts as well as the area that would have been most visible to the courts.

As previously noted, ceramic assemblages from both surface survey and excavated contexts primarily consisted of serving bowls, with some fragments of ritual objects such as sahumadores, figurines, and miniatures. Though at a much smaller rate, we did find fragments of large jars made of rough orange-gray, or brown paste. These vessels were likely used for carrying liquids or other foodstuffs. In the midden behind the north lateral mound, we found a fragment of a patojo, a vessel shaped like a boot with spikes on it.

While we are not entirely certain of its use, modern ethnographic examples (sans spikes) are used for cooking beans. Many of the ceramic vessels observed both on the surface and, most especially, in the middens near the ballcourt at Mogotes del Panteón, were decorated wares. Many pieces were of an impressive size, indicating that serving vessels used in 702 activities taking place around the court involved a sizable number of individuals, and that these events were socially marked. Several studies (Mills 2004, 2007, 2008) have demonstrated that the size of serving vessels is linked to the number of feasting participants and status groups. This would suggest that feasting and other ceremonial activities were carried in and around the ballcourts in Nejapa.

Of all the ballcourts in the study sample, Llano las Casas is perhaps the most distinct. It’s extremely limited viewshed reflects the narrowness of the small valley where it was constructed; visually, it’s also hidden from other settlements. You only see it once you arrive at the court. Along with a small platform, the other prominent feature is the large boulder carved with Zapotec-style glyphs, described in Chapter 6. Llano las Casas itself lies between the two residential settlements, El Sitial to the north and Peña del Rodeo to the south. However, without more radiocarbon dates from excavated contexts, it is difficult to say much more about the relationship that existed among these communities, or who might have used the palangana court at Llano las Casas, considering that there is no evidence for settlement in direct association with the court itself. If we assume contemporaneity, then the Llano las Casas court may have been located at the possible boundaries between the communities of Peña del Rodeo and El Sitial. A radiocarbon date does indicate an earlier, Classic-period (CE 406 – 548) occupation for the latter. A second palangana on the same hill as El Sitial but perhaps an associated settlement or neighborhood strongly suggests that Llano las Casas and El Sitial were contemporaneous, or at least overlapped in occupation, and that people living on El Sitial occasionally descended down from the mountain to play games at the court of Llano las Casas.

The multiple lines of evidence strongly suggest that people may have shared a common suite of references and practices related to the ballgame in Nejapa. There are 703 similarities in the architectural forms (sunken courts with low terminal mounds, most enclosed and I-shaped), construction techniques, and the specific size ranges for those court dimensions that influenced game play. Even the non I-shaped ballcourts are similar in their construction and size ranges. One drawback is time, as construction of the courts likely occurred centuries apart. This is especially true for the Classic period, where the first

(Mogotes de Panteón) and last (Cerro Maluco) courts to be constructed were built at most

500 to 800 years apart. However, these similarities persist more or less over time from the

Classic to the Postclassic periods, with some small differences. Classic period ballcourts narrower and more variable on average in their overall widths, while their central alley widths are wider and also variable. Postclassic period ballcourts, on the other hand, are much more consistent in size. Central alley lengths are also much longer, while the size range for their widths is narrower. However, the differences we observe are not statistically significant. Ballcourts in both periods were constructed using more or less the same methods, and all courts have very low terminal mounds. Together, the evidence could suggest that there existed similar ways of playing the game or shared rule systems that persisted over time. Alternatively, rule systems or ways of playing may have changed over time, but when it came to constructing their courts Postclassic Nejapa communities referenced the older Classic structures.

Of course, this argument hinges on the assumption that ballplayers would have been unlikely to play in another community’s court if it differed substantially from their own or what they were used to, as this would give one team clear home field advantage.

Standardization in playing fields certainly would have made it possible for individuals and teams to play at the ballcourts of other communities (Baron 2006:42). This same factor plays into the construction of modern sports fields, whose measurements are heavily 704 regulated by sports organizations. A football field is a standard 120 yards (110 m), with specific dimensions determining the lengths and widths of different zones. On the other hand, baseball fields show a much greater range of variation. The fences surrounding the fields can be placed anywhere between 300 to 420 feet (91 to 128 m), while the backstop can be 50 to 100 feet (15 to 30 m) behind home plate. While the distances between the bases is more heavily regulated, the placement of the outer fence and backstop do affect gameplay, and both batters and pitchers have to adjust their playing style for different stadiums. The infamous Green Monster at Fenway Park is an excellent example of this effect (Figure 9.1). Most pelota Mixteca fields are standardized, but the length of the field involved in game play is dependent upon the ability of the players; in other words, the actual playing field is shortened for lower ranked players, but lengthened for players of first rank. A degree of shared fluency in gaming traditions is important in order for there to be intergroup relations on both local and regional scales (Yanicki and Ives 2017:3). Thus, even when the field is only somewhat standardized, game play could be easily adapted when necessary because the proportions of the field were similar enough that technique does not have to be dramatically changed.

705

Figure 9.1: The "Green Monster" at Fenway Park, an 11.3 m (37.2 ft) wall that is the highest among Major League Baseball stadiums (original photograph by Jerry Reusse, reproduced under license terms of cc-by-sa-2.0).

It may be more accurate to say, then, that in Nejapa there existed localized references on which the inhabitants drew on in order to construct their ballcourts. That is, when building their courts, Nejapa’s residents may have looked to other, older courts as reference guides. This may have been truer in the Classic, when we have many less courts, than in the Postclassic where we have examples of several courts that were contemporaneous and/or overlapping in date. However, what we observe in Nejapa was not a regional ballgame tradition, or localized expression thereof, in the manner of the ritual ballgame cult of the Peripheral Coastal Lowlands. Although we have an example of a yoke

(Figure 9.2) with water toad symbolism, currently housed in the collections of the Casa de

Cultura in Nejapa de Madero, there is not much other evidence for a shared ideological symbolism of suite of symbols used by Nejapan communities. Instead, they built their ballcourts using similar designs, size ranges, proportions and construction techniques.

706

More than likely, people played a ballgame in Nejapa whose rules and playing styles were close enough that different communities could play at each other’s courts on those occasions if and when they did. Any differences in playing styles or rule systems could be negotiated beforehand between community elites, game sponsors, and/or team captains.

707

Figure 9.2: The yoke with water-toad symbolism showing the carved top (upper) and the side/front (lower). It is currently in the collection of the Casa de Cultura, Nejapa de Madero. 708

What was the role of the ballgame and the ballcourts in the sociopolitical relationships of Nejapan communities, and how did this role change over time? As I discussed in Chapter 3, the ballgame could play various and overlapping roles in communal sociopolitical relationships at local and regional scales. Ballgames could unite different social strata while also highlighting social differences both within and between communities. Sometimes, the hosting and attending of ballgames was tightly controlled by regional elites who had centralized authority over game ceremonies and ritual events.

Oftentimes they were political mechanisms for networking and competition among neighboring polities. Ballcourts themselves are found in variable distributions within regions across Mesoamerica, there is no consistent pattern. For example, in some places courts are more widely dispersed across the landscape and within the settlement hierarchy.

Different communities of varying size and composition would have alternatively served as hosts and guests, indicating perhaps that the game was less exclusive and perhaps decentralized. In others, ballcourts are only found at the regional political centers or major settlements only; while in such situations ballgames may have been more strictly controlled, they could still unite different social strata together. Even when ballcourts are restricted to the frontiers zones or boundaries between regional polities and communities, ballgames would have still fulfilled these roles. Where we find ballcourts on the landscape and in the settlement hierarchy can tell us about how ballgames might have figured into local politics in Nejapa.

From the Late Formative period to the Postclassic, the historical contexts, the nature of social relations, and interregional relations changed significantly. Similarly, we also see a shift in the role and meaning ballgames played in the relationships among the

709 communities in the Nejapa subregion, from the construction of the first court to the last.

The expansion and contraction of the courts in Nejapa first described in Chapter 6, for example, may be linked to similar changes in population and settlement as well as the larger historical trends. I explore these themes in the following sections.

9.2.4.1 Ballgames and Ballcourts in the Formative and Classic Periods

Overall, Classic period ballcourts are found in a variety of contexts such as architectural groups in community ceremonial zones, at the foot of a hill far from the main settlement, and an isolated hidden valley. It is also in the Classic period where we see two distinct court types, enclosed and sunken I-shaped courts with U-shaped terminal mounds and palangana courts, as well as the largest court in the subregion and perhaps in this area of the Sierra Sur. There is not much intervisibility among the ballcourts at this time; people would have seen and be able see to and from the courts at Ejido de los Canseco and

Mogotes del Panteón from opposite sides of the valley floor and piedmont areas. Perhaps, the lack of intervisibility, convergence, or intersection between ballcourt viewsheds is an indication that courts were not contemporary. Thus, the court at MDP may have been abandoned by the time the structure at EC was constructed. Following from this, the overlap in the viewsheds of the ballcourts at Los Limares and Cerro Maluco may suggest that these courts were in fact contemporary. In general, the diversity of site contexts and ballcourt types suggests that the courts may have been used for different purposes depending on the needs and motivations of the hosts and guests.

Masonry ballcourts were first constructed in Oaxaca sometime in the Late

Formative period, appearing in many places in the Central Valleys, the Isthmus, Mixteca

Alta and Baja, and the Sierra Mixe. In Nejapa, the first ballcourt was constructed at this 710 time or not long after at Mogotes del Panteón, in either the Late Formative or Early Classic.

At this time, the settlement of Colonia San Martin – the oldest so far in Nejapa – is waning, and will be abandoned by the end of the Formative or shortly thereafter. Nejapa de Viejo is a new settlement that will soon ascend in influence and come to dominant the subregion in the Classic period. Both sites are located near or around the confluence of the three rivers.

Perhaps, while people were settling in Nejapa Viejo, they were moving to the extension of piedmont strategically located near the entrance to the valley. Trade relations between far- flung regions intensified in some places while waning in others; at the same time, Oaxacan society was rapidly urbanizing, engendering great political changes among communities and individuals as social status were also increasingly differentiating and solidifying. Some even argue that elites at Monte Albán were fielding imperial armies to conquer other regions, which surely would have impacted regional politics. However, there are doubts about this “Zapotec Empire”, as evidence instead shows that trade relations between the

Central Valleys and the Rio Verde region on the coast ceased at this time and communities in the latter region began producing grayware bowls in local styles. Because their preferred exchange route had been cut off, Zapotecs from Monte Albán may have sought alternative routes to access coastal products. One route would have been through Nejapa, either to the

Isthmus or to the Pacific Coast via the Chontalpa.

The strategic position of the MDP settlement near the entrance to the valley and from a location that would have allowed people to monitor movement along the Rio Grande may have been the reason why people moved there, especially during this time of political and economic intensification. While people would have used this subregion as a conduit for movement since Paleoindian times, the changing political climate of the Late Formative would have motivated people to control access in and out of the valley, especially with 711 regards to trade. The ballcourt at MDP is a massive structure, and the exterior of the south lateral mound would have been visible to people traveling along the Rio Grande, as well people standing on or near that mound could see the river and alluvial plains below.

Though we cannot discount the possibility that ballcourts were also constructed at Colonia

San Martin and Nejapa Viejo, MDP’s strategic location may explain why we find the first court here instead of these much larger settlements.

The MDP ballcourt was part of a building program that also included other monumental buildings such as plazas, temples and platforms. The layout is very formal and clearly planned. While the Main and East Plazas could have probably fit at least 100 people comfortably, only those standing on the latter and the west terminal mound would have been able to see the game. People standing on the plazas or Structures A and B would not have been able to see the game or any activities taking place in the court, but they would have been able to hear the sounds. Given the limited capacity of the space and the restricted views of the game, the audience was likely composed of local elites and other persons of high social standing. Perhaps, important visitors were invited, either from nearby communities or foreign dignitaries who came for trading purposes. The quality of pottery discarded in middens found nearby certainly points to the amount of investment put into game ceremonies by the residents of MDP.

It is difficult to place the ballcourt at Ejido de los Canseco in a proper historical and social context. Likely the second ballcourt in our sequence, it is vastly different from the

MDP court. We are not entirely sure when it was constructed, it may have been in use at the same time as the MDP court, overlapped with, or was constructed after that settlement had been abandoned. Although nearly as large, it is much simpler and the structures that may have existed around it were likely household or smaller-scaled public buildings – in 712 contrast to the monumental architectural program at MDP. The evidence suggests that this was a neighborhood court or a court belonging to a smaller settlement. The hosting and scheduling of games and ceremonial activities was probably decided by important members of the community, who may or may not have had defined social statuses. Perhaps the game was opening up to people other than the elites; alternatively, there may have been more ballcourts in the past located at other neighborhoods or small towns that have since been destroyed, and the game was played by people of many statuses. There also exists the possibility that ballgames at this time were only played at Canseco only, for whatever reason. Still, except from the north and south terminal mounds, people would not have been able to see the games though they would have been able to hear them.

By the Middle Classic, we have possibly two distinct ballgame traditions, one focused on or near the valley in the settlements around the major rivers and played in I- shaped courts, Ejido de los Canseco and Los Limares; and the other localized to the mountains and played in palangana courts, El Sitial, Llano las Casas, and possible Los

Chorros (Figure 9.3). Though smaller in size, the ballcourt at Los Limares is similar in appearance to the Canseco court. Except from the terraces and the north and south terminal mounds, game activities would have been hidden from view, although again the sounds from the court would have been audible. As with Canseco, the Limares court is not located in a ceremonial precinct; instead, it was constructed down and away. While viewership of the game was probably restricted, there appears to be a de-emphasis in the courts' exclusivity as a public monument. Moreover, viewsheds from the courts, as well as perception of them, is no longer directed towards the river confluence. Nejapa Viejo is the predominant center at this time, and probably controlled movement along the trade routes and the rivers. The dramatic shift in the site context and location of the courts may mean 713 that while the game was still played, it wasn’t as important in negotiating social relations as before.

Figure 9.3: Map of the two ballgame traditions present in Nejapa during the Middle Classic period.

The second ballgame tradition are the palangana courts in the mountains bordering the southeastern part of the study area (see Figure 9.3). Here we also see a variety of contexts. The palangana located near El Sitial was constructed in the ceremonial precinct of the occupation there, near rather simple temple mounds. It was built using rough construction techniques. It is more difficult to comment on the palangana at Los Chorros, although a nearby structure does indicate that the court was probably part of a ceremonial

714 precinct or was associated with other public buildings. The palangana court at Llano las

Casas, on the other hand, was isolated, hidden and probably a sacred site perhaps used by people from different communities as part of pilgrimage or boundary-making rituals. As I have mentioned previously, palangana courts are not common in Oaxaca, but their appearance where we do find them makes sense since these locations are along the trade route traveling from the Altiplano to the Guatemalan highlands. For whatever reason, various communities along this route, from Chiapas through Nejapa and the Mixteca Alta into the Tehuacan Valley, adopted this particular court type and the game that may have gone with it. Taladoire (2003) proposes that it was a handball game, and perhaps explains as this game type first appeared in Oaxaca. Yet, the palangana courts in Nejapa are located high in elevation, in inaccessible places where people would have been able to control movement to their settlements, at a distance from this trade route. Perhaps, people traveling from Guatemala followed coastal routes before turning north through the

Chontalpa and arriving in Nejapa, bringing a game with them that appealed to some of the mountain communities who later adopted it.

Material culture evidence suggests interactions between the people living in the mountains and the valley floor, but the question of whether they played in each other’s court is difficult to answer at the moment, especially if we assume that different games were played in I-shaped and palangana courts respectively. The Middle Classic florescence of courts may be contemporary in age with the appearance of Zapotec-style glyphs, showing an increase in social complexity and ceremonialism in Nejapa. At two of the ballcourts we find stones carved with glyphs in a localized style distinct from the Central

Valleys. Nejapa’s strategic position along the Mexico-Isthmus-Guatemala trade corridor must have brought residents into contact with new ideas and people. By this time, Oaxacan 715 society was highly urbanized and elites were slowly consolidating their power by co-opting formerly communal rituals and public spaces. Exotic and luxury trade items and consumables were critical parts of elite legitimation strategies, and the exchange and consumption of these goods would have increased traffic traveling through Nejapa.

Communities here would have employed various strategies to negotiate with outsiders and amongst each other. In the mountains, people adopted the palangana court and game, while on the valley floor they continued to build I-shaped courts similar in design to the earlier

MDP court, but much simpler and not as part of formal architectural programs.

The Cerro Maluco ballcourt was constructed sometime in the Late Classic, if not earlier. As the largest structure of its kind in the Nejapa subregion and likely the Sierra

Sur, it was a grand declarative statement of power by those who built it, perhaps demonstrating the importance of Nejapa in exchange networks. Depending on when the courts at Canseco and Los Limares were abandoned, the Maluco court was probably the only one in use during this time period. This would signify a severe contraction in the number of courts in Nejapa in the Late Classic, effectively moving game ceremonies and events from the areas around the confluence of the rivers and the older center of power,

Nejapa Viejo, further south and away from the alluvial plains. The Cerro Maluco settlement is a much smaller than its neighbor communities to the north like Tanaguixi, so the massiveness of its court is very interesting. The structures and plaza associated with the court are also some of the largest structures in Nejapa during the Classic period, and would have probably accommodated a significantly-sized audience. Perhaps the court was used only by the people of the Maluco community, however given the size of the court and the monumental structures around it, individuals of social status from other settlements may have been invited. Aside from its size, the Malcuo court also stands out because it is part of 716 an architectural group that echoes the arrangement of structures at the MDP from centuries earlier. People may have referenced the older court when building this one.

In the Late Classic, the power of Monte Albán in the Central Valleys was beginning to wane and other queichopeque were filling the vacuum of power and competing for influence. At a much larger scale, the collapse of Teotihuacan was having a major reverberating impact throughout Mesoamerica. Between the two, and amid the collapse of other urbanized settlements in Oaxaca, trade relationships were probably disrupted, and people may have shifted to using other routes in order to access needed trade items. One of the most dramatic changes in Nejapa is the shift in court viewsheds from around the confluence of the rivers and the alluvial plains and the piedmont zones north and west in the early half of the Classic period to the Rio la Virgen drainage and the piedmont zones around the river and to the southeast. Evidence from Quiechapa indicates a movement of

Zapotecs into the subregion, putting pressure on the Chontal and Chatino communities that already lived there. This may have increased hostilities throughout the Chontalpa and lead to tensions with neighboring regions. Maluco would have been hidden from view to communities north of it, but would have been visible from areas south and people traveling northwest along the Rio la Virgen. Perhaps, people at Maluco were turning their focus to their southern neighbors, either because of the increasing hostilities in Quieichapa and the

Chontalpa or because disrupted trade relations shifted travel routes, driving traffic north from the Chontalpa rather than running east-west (and vice-versa) from the Central Valleys to the Isthmus.

The ballcourt at La Baeza is probably Late Classic as well, but it is more difficult to place in the sociohistorical context of the Sierra Sur. On maps it appears isolated as the only court registered in the western part of the study area. However, we know that more 717 settlements existed here. Survey and excavation will fill in these gaps and help us build a picture of the settlement patterns and history of the western Nejapa subregion. This area may just be another multilingual frontier zone, as there would have been Zapotecs living to the north and northwest, and Chatinos and Chontales to the south from east to west. Unlike the other Nejapa courts, the structure at La Baeza is possibly associated with an elite residence, a pattern observed in the Central Valleys. Should this court prove to be the only such structure in the western part of the study area, then we would have a different political configuration here, with the scheduling and hosting of games tightly controlled by Baeza’s elite. Although likely part of another frontier zone, polity, or regional settlement system,

Baeza’s court falls within the preferred size ranges for Nejapan courts, is constructed using similar techniques, and is in a similar architectural arrangement as at Cerro Maluco.

9.2.4.2 Ballgames and Ballcourts in the Postclassic Periods

Postclassic ballcourts in Nejapa are primarily enclosed, sunken I-shaped ballcourts; the exception is Terezona, whose court has a T-shaped form but still is partially enclosed and sunken. Overall, the size dimensions for the different court features are somewhat more standardized, falling within a much narrower size range than in the Classic period.

On the other hand, although there are some differences in these size dimensions Postclassic courts were constructed using the same methods as in the Classic and their design was the same, that is terminal mounds that are much smaller than the lateral mounds with exterior heights that are at or nearly level with the surrounding ground surface. Whereas courts in the previous period occur in different contexts, Postclassic courts are mostly all located in the ceremonial precincts of neighborhoods or settlements in association with public buildings. La Puerta is unusual because the court is on a narrow saddle between what 718 would have been the main occupation and perhaps a second zone with terraces and some structures that may have been public buildings. Still, similar contexts and ballcourt types in the Postclassic suggests a more “unified” ballgame ceremonialism, at least in contrast to the

Classic period.

After the collapse of Classic-period societies, new settlements emerged and population expanded in the Early to Middle Postclassic as sociopolitical relationships and interregional alliances were reconfigured and trade networks reestablished. In Nejapa, similar trends occur and we see an expansion of settlements and population. During this time there is also a second period of game florescence that was much different from Middle

Classic, i.e. primarily I-shaped courts found in ceremonial precincts or with other, public monumental buildings. People traveling through Nejapa would have encountered a network of ballcourts along the eastern banks of the Rio Tehuantepec, including Greater La

Amontonada with its multiple neighborhood courts (although again there is the possibility that more courts may be found south and west of the river confluence). GLA was probably one of the more influential communities in Nejapa at this time. However, unlike in the

Classic period or later in the Late Postclassic, I would suggest that the scheduling and hosting of games and the construction of a court was not centrally directed by that community’s higher status elite. In other words, ballgames and rituals were somewhat decentralized in the Early to Middle Postclassic.

The different architectural styles observed across GLA’s neighborhoods may indicate that this community was composed of people with variable social – perhaps ethnolinguistic – identities. Games were probably one way people living in these neighborhoods negotiated social conflict and integration between them. There are some important differences between the main ballcourt on the bluff and the neighborhood courts. 719

The former is much more massive, better constructed, and more formal. Although this court was more visible, access to the court was probably more restricted. Thus, even though there is a general degree of decentralization with respect to control over game scheduling, there is also some status differentiation that marked the main ballcourt as more exclusive or important. Whereas audiences for games at the neighborhood courts were likely composed of members of different social statuses or groups, at the main court attendance would have been limited to elites from the main barrio, lower-status elites from the neighborhoods, and other community members of special status. In contrast, the small size of the ballcourt as well as the limited space available around it would mean that audiences were much smaller at La Puerta. Games were probably more for the people living here. Even though the audience was smaller, given the level of complexity and material status differentiation observed at La Puerta, there would not have been a defined hereditary or high-status elite class that exclusively controlled games. In this sense, the La

Puerta court is analogous to the neighborhood courts at GLA.

The reemergence of the ballgame and construction of new courts beginning in the

Early Postclassic likely co-occurs with the reestablishment of trade relations after the

Classic to Postclassic transition in Mesoamerica. Results from the viewshed analyses for this period seem to support this as views are once again directed towards and from the west running north to south, centered on the confluence of the three rivers and the alluvial plains that surround them as well as the piedmont zones nearby. Courts at this time are much more visible, and there are several points on the landscape (around the confluence of the rivers and their watersheds) where it would have been possible to see one or more of the courts. Along with the narrower preferred size ranges, this suggests a greater degree of integration in ballgame practices and game events. The results seem to indicate that 720 visually the court at Terezona was part of this Early to Middle Postclassic ballgame network. Although I have tentatively placed Terezona in the Middle Postclassic, there exists the strong possibility that the court was constructed earlier and was contemporaneous with or overlapped the courts of GLA and La Puerta. This also makes sense if we consider that – although on a hill – Terezona is located much closer to the alluvial plain and major rivers in comparison to the later period courts discussed shortly.

Sometime in the late Middle Postclassic and the Late Postclassic, there is a dramatic contraction in the number and spatial distribution of the courts from the alluvial plain and major rivers to the piedmont and mountain summits. Sites with ballcourts at this time are highly defensible and would have been difficult to access. Viewshed results show that, in contrast to the high visibility of the courts in the Early and Middle Postclassic, the courts at

Cerro del Convento and El Sitial are generally hidden from view, highlighting the exclusivity of the court and probably the restricted access to the game as well. This contraction in courts and shift to the mountains may coincide with the first arrival of the

Zaachileño Zapotec armies in the 14th century. Overall, we do not see signs of conquest or drastic changes in material culture related to the supposed Zaachileño invasion and conquest. However, this change in the number, location, and exclusivity of the courts may be evidence of people leaving their communities closer to the valley floor and to settlements in the mountains in response to the arrival of these foreign armies.

Overall in Nejapa, people were potentially making connections to various ethnolinguistic groups and regions farther off, resulting if not in mixed, heterogeneous communities, then in relationships that frequently crossed ethnolinguistic boundaries that many have previously treated as being more static or impermeable. Nejapa’s strategic position along critical trade corridors and its rich alluvial land made it an attractive location 721 for many people, who through their shared experiences of living in this frontier region forged new identities that at times were communal or contrasting. Because of its very nature as a competitive sport requiring at least two teams to play, the ballgame was one way to negotiate movement across these boundaries. However, the role the ballgame played and how communities strategically used the game to negotiate sociopolitical relationships internally, with their neighbors, and with outsiders from other regions changed over time, and likely multiple different strategies were used.

We have periods in Nejapa where multiple courts were in use, and control of the game may have been decentralized, and other periods in which there was only one court and access would have been controlled by the leaders of the communities that built them.

The game in Nejapa clearly reached its apogee in the Early to Middle Postclassic, but I would argue that even as its sociopolitical meaning and strategic use may have changed over time, its importance never waned. Even in the Classic when there was probably only one court in use, MDP at the beginning of this period and Cerro Maluco at the end, we can still conclude that the game was important because both of these structures were massive

(especially Maluco’s court), oriented east-west, part of formal architectural groups, and located at sites on hills overlooking rivers that would have been major transportation routes.

In the second period of contraction during the late Middle to Late Postclassic, the game still retains a degree of importance. The main ballcourt at El Sitial lies opposite the palace and other monumental structures, while Cerro del Convento has been an important ritual site for millennia.

722

9.3 THE BALLGAME IN NEJAPA AND OAXACA: LOCALIZED EXPRESSIONS OF REGIONAL TRADITIONS

In its social and historical context, Nejapa was a diverse multiethnic and multilingual frontier of communities accepting of various practices and ways to live, yet also sharing similar expressions of material culture and expression. This observation is reflected in the subregion’s ballcourts, which show an interesting mix of differences and similarities demonstrating both common references and local preferences in ballgame practices. What the ties the ballcourts of Nejapa together is that they were sites of social interaction through games, rituals, and feasting. Such events still occur today in the fiestas patronales, which gather people in the valley together for food, drink, and a little friendly competition.

In Chapter 7, I compared ballcourt data from the Central Valleys to the data from

Nejapa in order to see if there were differences or similarities significant enough to suggest distinct or shared ways of playing, i.e. ballgame traditions, and what would this mean in the larger context of ball playing these area and in Oaxaca as well. The results show that the central alley lengths, widths, and L/W ratios of the Central Valleys ballcourts overlap with the values of the Nejapa courts but show a greater range of variation. Court alley lengths in the Central Valleys, for example, range from 9 m to 70 m! In contrast, Nejapa courts have central alley lengths that range from 21.5 m to 40 m, confirming that there is a preference for a narrower size range in Nejapa in comparison to the Central Valleys. Statistical tests revealed significant differences in the central alley widths and L/W ratios. Again, there is much greater variation with respect to these dimensions in Central Valley courts, while in

Nejapa measurements fall within a narrower size range. The bimodal distribution of the

L/R ratios may point to two distinct populations in the Central Valleys, very narrow

723 rectangular central alleys vs. broader, less rectangular alleys. While we generally see the same in Nejapa, the sample is perhaps too small to tease modality in the data if present.

However, it is interesting that in both cases the central alleys L/W ratios shift in value from the Classic (Central Valleys = .11 – .57; Nejapa = .19 – .40) to the Postclassic (Central

Valleys = .15 – .38; Nejapa = .15 – .25)

Similarly, the values of the lateral and terminal mound dimensions also overlap.

Again though, there is much greater variation in the Central Valleys than in Nejapa.

Distributions of these values in the former are nearly normal or are even bimodal, while in the latter they tend to skew to the right or left, indicating a preference for specific mound sizes. Statistically, there is only significant difference in the interior and exteriors heights and volumes. Terminal mounds in Nejapa are consistently much shorter in height and smaller in form, especially with respect to their exterior heights. The evidence again shows that low or non-extant terminal mounds were standard in Nejapa. By comparison, this style appears in the Central Valleys but only at certain courts (Feinman and Nicholas 2011).

Where we also see a difference between the Central Valleys and Nejapa is in the density of ballcourts per square kilometer. When we consider only the amount of land surveyed by the PANT team, the density is over 900% in the latter over that of the former at 18.5% vs. 1.7%. As I discussed in Chapter 7, over the course of time this density in

Nejapa is not significant, as there would have been several periods of time in which only one court would have been in use. In comparison, in each time period there would have been several contemporaneous courts in use in the Central Valleys. On the other hand, geographically we do see some significance in Nejapa’s court density. Central Valley courts are more dispersed throughout the region with some clustering, especially around

Monte Albán. Ballcourts in Nejapa appear primarily in and around the rivers or other areas 724 that would have experienced the most foot traffic. While it is likely that more ballcourts will be found Nejapa, I suspect that we will still find these courts in the alluvial plains and piedmont areas along the major rivers. Over the course of time, Nejapans continued to build courts in these areas even as settlement patterns shifted. With respect to the Central

Valleys, with more data it would be possible to see how the ballcourts may have shifted in distribution over time (unfortunately, GPS locations were not available). However, we can safely assume that ballcourts would still have been fairly dispersed in the Formative,

Classic, and Postclassic period. Shifts in distribution would have accompanied changes in the political boundaries of polities as well as the status of influential quechecoquie in settlement hierarchies.

Do these differences add up to distinct ballgame traditions, or are there enough similarities to suggest shared ways of playing and rule systems? It is difficult to say which types of games would have been played in the I-shaped ballcourts or how rule systems and playing techniques may have changed over time. Certainly, the various terms recorded for different ballgames in the colonial Zapotec dictionary indicate that people were familiar with various games. Perhaps the best indication of different games are the different court forms. In the Central Valleys, open courts occur only in the Late Formative, while I-shaped courts were constructed from the Late Formative onward. In Nejapa, we primarily have I- shaped courts in every time period, with some palangana courts that appeared in the Middle

Classic. If we assume that a specific ballgame was played in the I-shaped form, then the court dimensions would suggest that people in Nejapa and the Central Valleys were playing a similar game – maybe hipball – but with local rule systems and slight differences in playing techniques (more so in the Central Valleys where there is more bimodality in the data). These differences, however, would not necessarily mean that a team from Nejapa 725 could not have played in a Central Valleys court, should such an event had occurred in the past. Nejapa court dimensions fall within the size ranges of the Central Valleys, so the size of some of the courts in the latter would not have been unfamiliar to people in Nejapa.

There may have been enough degree of fluency that rule systems could be negotiated between teams, similar to stickball games between different indigenous groups in North

America.

Other differences in the courts also suggest local trends and preferences. Design- wise, the ballcourts are very different in the Central Valleys from those in Nejapa. In addition to the differences in the terminal mounds, Nejapa courts lack many of the features found on Central Valley courts, such as vertical back walls on the terminal mounds, benches, and lateral structures on the lateral mounds. Only the main ballcourt at El Sitial has these features. In contrast, when we do see additional court structures in Nejapa, they tend to be on the terminal mounds. Construction methods also vary somewhat. People in the Central Valleys more often used shaped stones to line the façades of the court structures and playing alleys were covered in stucco surfaces. Nejapa courts were lined with unshaped stones, and so far only one court has evidence of a stucco floor. While excavations in the future may uncover stucco floors in other Nejapa courts, the evidence so far indicates that floors were made of compacted earthen or clay floors often mixed with small pebbles. Perhaps the most striking difference are the interior slopes or aprons of the lateral mounds. Although not present at all courts, there are several examples in the Central

Valleys where the aprons are stepped (Figure 9.4), a characteristic not observed in Nejapa so far. Clearly, the appearance of the courts was much more subject to local tastes and preferences, with a greater degree of similarity in Nejapa as compared to the Central

Valleys. 726

Figure 9.4: Stepped aprons can be seen at several ballcourts in the Central Valleys, including the ballcourt on the Main Plaza at Monte Albán.

We also see some differences between the Central Valleys and Nejapa in the site contexts where ballcourts are found, at least in the Classic period. In both regions, ballcourts can be found in the ceremonial precincts of settlements in association with other public monuments. However, in Nejapa we have examples of courts occurring in contexts not observed in the Central Valleys. The courts at Los Limares and Llano las Casas are not found in association with other structures. The former is set apart and down the hill from the main settlement, while the latter lies hidden in an isolated valley. During the Late

Classic, a trend emerged among Central Valley elites to build small ballcourts next to their house compounds or palaces. We do not see this pattern in Nejapa (not even in the later

727

Postclassic) except for perhaps at La Baeza where the ballcourt may be associated with an administrative/residential building, supporting the hypothesis that this settlement may have belonged to a separate political system.

Kowalewski and colleagues (1991) observed that ballcourts in the Central Valleys are found either in lower ranking settlements on regional or subregional boundaries or at higher status regional capitals. In contrast, it is more difficult to identify regional or subregional boundaries in Nejapa, if they even existed. While some communities may have exerted more influence at certain points in time, such as Nejapa Viejo in the Classic period and Greater La Amontonada in the Early to Middle Postclassic, we do not find evidence for the same level of power and influence, or even consolidation, as exerted by Central Valley quechecoquie. Thus, in Nejapa the ballcourts were probably not marking boundaries or major capitals but instead were the prerogatives of individual communities who at times shared a fluid identity as a multilingual people living in a trade corridor and frontier zone.

Interestingly, courts in both regions appear to fall along major routes of foot traffic. In the

Central Valleys, most courts are found in the northern Etla, central Valle Grande, and southwestern Tlacolula Valleys. This would have generally been the route followed by people traveling from the Altiplano through Oaxaca and Nejapa on their way to the

Isthmus.

How do the Nejapa ballcourts compare to those regions and subregions closer to the study area? The Isthmus region to the southeast was clearly an important center of development for ballgames in Oaxaca since at least the Late Formative. For example, the incredible site of Tres Picos, which dates to this period, has both open and T-shaped ballcourts. The latter court form may link this region to the gaming traditions of the Maya area where we find this type. Winter and Santiago (2014) consider the Isthmus to be part of 728 the Mixe-Zoque tradition. Thus, games played here may have had considerable different rule systems and playing techniques. I-shaped courts appear later; those at Postclassic

Guiengola, for example, resemble Central Valley courts – perhaps suggesting that at some point in time communities began playing a game different from what they were playing previously. Along the coast during the Classic period, some communities participated in the pan-regional ritual ballgame cult of the Peripheral Coastal Lowlands zone. Isthmus- style wares do appear in ceramic assemblages in Nejapa, so people were trading with each other. However, there is not sufficient data to draw links between the ballgame in Nejapa to the Isthmus traditions.

We know little of the ballgame in the Sierra Mixe region to the north. Two ballcourts were observed, but not registered, on the private lands of a local owner from the town of Santo Domingo Narro, in the municipality of San Juan Juquila Mixes. Recent salvage work along a super highway being constructed near Narro has documented additional court structures (Stacie M. King, personal communication). As mentioned previously, three other ballcourts were documented at the site of Chuxnabán, which dates to the Late Formative and was occupied during the Classic and Postclassic periods (Markens and Winter 2014; Winter 2008). Another was documented at the site of Moctum, near the modern town of San Marcus Moctum (Taladoire, personal communication). Not much information is available on the size or form of these courts, although likely they were I- shaped, as evidenced by the I-shaped court models found in the Cueva del Rey Kong-Oy.

Many more ballcourts probably remain to be documented. Even less is known about ballgame material culture or ideology here. The exemplary sculptures of the Cueva site appear to link the game to human sexuality and fertility, in contrast to the more common symbolism of agriculture fertility through regenerative violence. Moreover, the sculptures 729 are evidence for a Late Formative handball game. As with the Isthmus, there is not sufficient data to link the ballgame in Nejapa to the traditions of the Sierra Mixe.

In contrast to the density that we observe in Nejapa, there are far fewer ballcourts in the Sierra Sur subregions to the south. Zborover (2014) documented two courts in the

Chontalpa Alta. Only one of these, in the Late Classic Cerro Estibo site, is securely identified as a ballcourt. Surprisingly, its form and architectural features are very distinct – here we have an elongated depression with two parallel mounds on either side, similar in a fashion to the open-court form. Here Zborover also found ballcourt material culture that connects the ballgame tradition in the Chontalpa to the Sola de Vega Valley and the Central

Valleys as well as the Peripheral Coastal Lowlands (PCL) zone. The other, probable ballcourt is also an elongated depression, except with only one narrow mound instead of two. In the Quiechapa region to the southwest, Badillo (personal communication) observed three ballcourts (Figure 9.5). The two courts that were mapped are both I-shaped courts.

One is oriented north-south and dates to the Postclassic, while the other is oriented east- west but the date is unknown. Although there are few courts in Quiechapa, Badillo did document ballcourt-shaped icons that had been carved into the bedrock in places with ritual characteristics. So far, no such icons have been found in Nejapa, and their presence in

Quiechapa is intriguing, given the vast difference in the number of ballcourts between the two regions.

730

Figure 9.5: Map of showing the locations of the ballcourts and ballcourt icons in the Nejapa and Quiechapa subregions.

Within the eastern Sierra Sur region, the number, density, and age of Nejapa’s ballcourts make this subregion stand out from its neighbors to the immediate south and southwest. Likely, these differences are related to Nejapa’s position as a frontier zone, a place of interregional movement and trade, and the home of multiple ethnolinguistic groups. We have better evidence for earlier occupation in Nejapa, with the first permanent settlement appearing in the Early Formative. As a crossroads between regions to the north, west, east, and south, people in Nejapa would have had greater opportunities to participate in interregional relationships, and would have experienced or been exposed to ballgames earlier on and from different places. Over the course of time, these communities developed

731 shared preferences for ballcourts design and sizes, while still expressing their own local tastes and needs. In contrast, ballcourts appear much later in the Chontalpa and Quiechapa.

While the latter’s courts are I-shaped, those in the Chontalpa are open and sunken, suggesting a different style of playing altogether. It is difficult to say if communities in

Nejapa hosted visiting teams from Quiechapa or even the Chontalpa. However, we can certainly imagine that the number and age of Nejapa’s courts as well as those periods when there were multiple courts in use probably had some effect on intra and interregional relationships.

As I mentioned earlier, we do not have much ideological or iconographic symbols specifically associated with the ballgame in Nejapa. The glyphs found on carved boulders near courts occur in other ritual contexts and are not particularly identified with ballgames.

On the other hand, the figurine leg with wrapping around the knee found in Op A. at El

Cucharital on the Tavela side resembles other ballplayer figurines found in Oaxaca. The other example of ballgame material culture in Nejapa is the yoke described previously, which is decorated with water-toad symbolism. Though rare, yokes in Oaxaca feature this type of symbolism, while the more common hachas have wrinkle-faced old men and monkeys. Because it is a singular example, the yoke may be evidence more of Nejapa’s trade networks than anything else. Though symbols related to the PCL ballgame tradition were found in association with a court in Chontalpa (and other areas on the Pacific Coast), we see no such similar material culture in Nejapa. Ceramic evidence from middens found around the courts do show that rituals took place there, so we do know that Nejapan communities participated in game ceremonialism. What form this ceremonialism may have taken, and whether it was linked more to Central Valleys, Sierra Mixe, or the PCL traditions, is unknown at the moment. Likely, as in many regions of Oaxaca, people 732 participated in broader or more regional game traditions but had their own localized expressions of them.

9.5 FUTURE DIRECTIONS IN BALLCOURT RESEARCH IN NEJAPA

Much of the interpretations presented here, while based on multiple lines of evidence, are still highly speculative. Without a doubt, much more investigation is needed, especially if we are to resolve many of the chronological issues that make it difficult to interpret the data. It would be interesting, for example, to see if the patterns observed so far in the Nejapa subregion hold out at the sites I have not mapped, and in those areas where we have not surveyed, but where ballcourts may exist. Therefore, additional survey and excavations should help to resolve these issues and either confirm or challenge the history of the ballgame in Nejapa sketched out so far. Survey will target current the gaps in the regional coverage, especially around La Baeza, the agencía of Las Animas, and the area west and south of the Rios Grande and la Virgen respectively, with the goal of documenting additional sites and, hopefully, identifying more ballcourts. Excavations in the mapped ballcourts in order to understand construction sequence and chronology. In the meantime, the available mapping data can be used for additional spatial analyses that could not be included here in order to ascertain several factors including: 1) the size and composition of the audience given the available space; and 2) where the audience would have stood in order to view activities taking place in the court. Finally, further refinement of the LCP models and the viewshed analyses should produce more useful results, helping us to link communities to the ballcourts they would have used as well as understand more about their location on the landscape. Such information will provide further insight into the ballgame in Nejapa.

733

9.6 CONCLUSION

The ballgame was a salient social field across all levels and types of social groups and boundaries in Mesoamerica. The meaning of games and the spaces in which they took place would have been actualized and embodied during extraordinary events and even on a more routine basis through more informal exhibitions and game matches. Whether large spectacles or simple affairs, ballgames were always events where social relationships were negotiated, contested and/or confirmed. In this dissertation, I have advocated for a regional approach to studying ballcourts because as special-purpose and marked public structures, their patterning and spatial distribution can provide insight into local and regional political organization. The Nejapa subregion, with its unique density of courts and variable court types, proved an excellent case study to test these ideas. Similar such approaches should help direct future investigations to flesh out more in-depth ballgame traditions in other regions of Mesoamerica, especially in areas where more research is needed. By doing so, we can hopefully understand in better ways the endurance and appeal of this ancient sport.

We see these variable patterns repeated throughout Mesoamerica: places with only one ballcourt or a few ballcourts for large populations, cities with multiple ballcourts, and areas of low and high densities that fluctuated across time as games may have waxed and waned in popularity, or became more or less controlled by specific social groups. Above all, the courts and the events that took place there were imbricated in community and social identity, not only through shared material culture and a common architectural vernacular, but also through shared ways of playing and, simply put, the sheer emotional experience of

734 watching physical competition, and the sense of bonding and communitas that such experiences produce among invested viewers and participants.

735

APPENDIX A

BALLCOURT SURFACE CERAMICS AND MATERIALS

A.1 INTRODUCTION

Surface ceramic and other materials were frequently observed in and around many of the ballcourts mapped over the course of the 2015 field season. Two sites had abundant surface artifacts because nearby middens had been exposed: in the case of the ballcourt at El Cucharital, a midden west of the west lateral mound was exposed during excavation of a pipeline to draw water from the

Rio Guitihuini below; and with respect to Mogotes del Panteón, along with the surface scatter there were also two middens that had been looted (all described in Chapter 6). While there were not very many materials observed on the ground surface around the ballcourt at La Baeza, there was a marked density of surface artifacts throughout the site, and are included here because these materials have not been described or presented previously. The surface materials from all three sites reveal important information about the kinds of activities that took place at the ballcourts (and for La Baeza in general), the relative chronology of the courts and the sites themselves, and the local economy and trade relationships in which residents of these settlements participated.

A.2 SURFACE CERAMICS AND MATERIALS FROM EL CUCHARITAL (YAU-NEJ-NEJ-005)

The surface materials found at the El Cucharital ballcourt were found in and around the court itself as well as from a possible midden located west of and below the west lateral mound. The ritual

736 items, such as the sahumadores and miniatures, all came from this midden and are likely refuse from ceremonial activities.

Figure A.1: Example of the ceramic sherds and rims observed on the surface around the ballcourt at El Cucharital.

737

Figure A.2: Ceramic sherd with incised design, El Cucharital.

738

Figure A.3: Ceramic supports in two different styles, El Cucharital.

739

Figure A.4: Fragment of a possible patojo, a vessel used ethnographically to cook beans.

740

Figure A.5: Broken sahumador with partial handle, bowl, and rim, El Cucharital.

741

Figure A.6: Examples of miniture ceramics that were often used in ceremonies in the Nejapa subregion. In the bottom right corner is a miniature cup.

742

Figure A.7: Fragment of red-painted stucco, likely a piece of wall from a nearby structure.

743

Figure A.8: Figurine piece, possibly part of a headdress with the vertical, curved lines representing feathers. This piece is flat on the back.

744

Figure A.9: Fishing stone weight from top (upper) and side (lower) showing the groove where the net rope would have been tied around the stone. These were found all over El Cucharital.

745

Figure A.10: Obsidian blades found near the ballcourt at El Cucharital. Such blades were often used in rituals and broken afterwards.

746

A.3 SURFACE CERAMICS AND MATERIALS FROM MOGOTES DEL PANTEÓN (YAU-NEJ-ANM-002)

The surface materials observed in and around the ballcourt at Mogotes del Panteón and from the middens nearby are some of the most impressive seen in the Nejapa subregion. Excavations in the future should uncover additional middens and ceramics materials.

Figure A.11 Ceramic grayware serving bowl measuring over 10 cm in diameter, MDP.

747

Figure A.12: Serving vessels and bowls from the midden behind the East Mound near the ballcourt. The sherd on the upper right has a applique.

748

Figure A.13: Serving vessels and bowls from the midden behind the East Mound. The sherd in the lower left has a brown wash and red border, a style commonly found in the Mixteca during the Ramos phase or Late Formative.

749

Figure A.14: Bottom section of a possible ceramic spoon, MDP.

750

Figure A.15: A face-neck jar roughly dating the Late Formative and Early Classic.

751

Figure A.16: Rim sherd with braided lip, found near the ballcourt at MDP.

752

Figure A.17: Examples of different incised sherds, many with rims present. The sherd in the bottom middle was found in the East Mound midden, while the rest were found on the ballcourt.

753

754

Figure A.19: Examples of figurines found on or near the ballcourt. In the top row we have a possible leg; on the bottom, a torso.

755

Figure A.20: Handle from possible utilitarian serving jar, MDP.

756

Figure A.21: Examples of stucco fragments found in the ballcourt at MDP.

757

Figure A.22: Three metates were found in the ballcourt possibly reused as stones to line the mound. One mano was also found, suggesting that food preparations may have taken place near the court as part of game activities and ceremonies.

758

A.3 ARCHITECTURE, SURFACE CERAMICS AND MATERIALS FROM LA BAEZA (SCY-SCY-BAE-001)

Over the course of survey at La Baeza, we observed several preserved standing, earthen/adobe walls. Their state of preservation is excellent considering the possible age of the site, which we currently estimate to have been occupied between the Late Classic to the

Early Postclassic (about CE 700 – 1200). In addition to these walls, there were also ceramic, lithic, and ground stone materials found on the surface throughout the site. The numerous chert nodules could indicated that lithic production was the primary economy of the La Baeza settlement, and that there may be a quarry nearby.

Figure A.23: Fragment of preserved standing wall at Structure A2, one of the first architectural groups encountered at La Baeza.

759

Figure A.24: Handle of courseware jar found near the ballcourt at La Baeza.

Figure A.25: Other ceramic artifacts found near the ballcourt.

760

Figure A.26: Grayware sherd, possibly from a narrow necked fine jar, with flared rim.

761

Figure A.27: Another example of a fine grayware jar rim, possibly a cantero.

762

Figure A.28: Course grayware sherd with an impression of woven material or tela. We found several examples similar to this.

763

Figure A.29: Pieces of a nearly whole vessel, found in Structure E. It is made of course orange- gray ware and the side walls featured two ceramic nubbins.

Figure A.30: One of the many brown colored chert nodules observed on the ground surface at La Baeza.

764

Figure A.31: Green obisdian microblade, likely reworked from a much larger piece. Obsidian of this type typically comes from Puebla.

765

Figure A.32: Bottom half of an arrowhead manufactured from white chert.

766

Figure A.33: Tenon nail stone found at Structure E. These were typically placed into the walls.

767

Figure A.34: A second tenon stone found at the Structure E.

768

REFERENCES CITED

Abrams, E. M. 1989 Architecture and Energy: An Evolutionary Perspective. Archaeological Method and Theory 1: 47–87.

Abu-Lughod, Lila 1989 Zones of Theory in the Anthropology of the Arab World: 42.

Acuña, Rene 1978 Farsas y Representaciones Escénicas Del Os Mayas Antiguos. Centro de Estudios Mayas, 15. Mexico: Universidad Nacional Autónoma de México.

Adelmon, Jeremy, and Stephen Aron 2018 From Borderlands to Borders: Empires, Nation-States, and the Peoples in Between in North American History. American Historical Review: 28.

Aguero, Adriana, and Annick Daneels 2009 Playing Ball - Competition as a Political Tool. In Blood and Beauty: Organized Violence in the Art and Archaeology of Mesoamerica and Central America. Heather Orr and Rex Koontz, eds. Pp. 117–138. Los Angeles, CA: Cotsen Institute of Archaeology Press.

Aguilar-Moreno, Manuel 2004 Filosofía y Simbolismo Del Juego de Pelota Mesoamericano. Estudios Jaliscienses 56: 10–28.

Aldenderfer, Mark, and Herbert D. G. Maschner 1996 Anthropology, Space, and Geographic Information Systems. Oxford University Press.

Aldenderfer, Mark S., and Charles Stanish 1993 Domestic Architecture, Housel Archaeology, and the Past in the South-Central Andes. In Domestic Architecture, Ethnicity, and Complementarity in the South-Central Andes. Mark S. Aldenderfer, ed. Pp. 1–12. Ames: University of Iowa Press.

Alegria, Ricardo 1983 Ballcourts and Ceremonial Plazas in the West Indies. Publications in Anthropology, 79. New Haven, C.T.: Yale University Press.

Allen, Kathleen M.S>, Stanton W. Green, and Ezra B. W. Zubrow, eds.

769

1990 Interpreting Space: GIS and Archaeology. London: Taylor & Francis.

Allison, Maria 1980 A Structural Analysis of Navajo Basketball. ProQuest Dissertations Publishing. http://search.proquest.com/docview/302970894/?pq-origsite=primo, accessed August 14, 2018.

Alonso, Ana Maria 1994 The Politics of Space, Time and Substance: State Formation, Nationalism, and Ethnicity. Annual Review of Anthropology 23: 379–405.

Altuve, Eloy 1997 Juego, Historia, Deporte Y Sociedad En América Latina. Maracaibo, Venezuela: Universidad del Zulia, Vicerrectorado Académico, Centro Experimental de Estudios Latinoamericanos.

Alvarado, Francisco de 1962 Vocabulario En Lengua Mixteca. Edición Facsimilar. Edi, Mexico: Instituto Nacional de Antropología e Historia.

Anderson, David G., and J. Christopher Gillam 2000 Paleoindian Colonization of the Americas: Implications from an Examination of Physiography, Demography, and Artifact Distribution. American Antiquity 65(1): 43–66.

Anderson, David S. 2003 El Asentamiento Preclásico En La Región Noroeste de Yucatán. In In Proyecto Costa Maya: Reporte Interino, Temporada 2002: Reconocimiento Arqueológico de La Esquina Noroeste de La Peninsula de Yucatan. A.P. Andrews and F. Robles Castellanos, eds. Fernando Robles, tran. Pp. 46–61. Mérida: Instituto Nacional de Antropología e Historia, Centro INAH Yucatán. 2011 Xtobo, Yucatán, México, and the Emergent Preclassic of the Northern Maya Lowlands. Ancient Mesoamerica 22(2).

Andrews, Anthony P., and Fernando Robles Castellanos 2004 An Archaeological Survey of Northwest Yucatan, Mexico. Mexicon 26(1): 7–14.

Annis, Sheldon 1989 Poverty, Natural Resources, and Public Policy in Central America. Transaction Publishers.

Anschuetz, Kurt F., Richard H. Wilshusen, and Cherie L. Scheick 2001 An Archaeology of Landscapes: Perspectives and Directions. Journal of Archaeological Research 9(2): 157–211.

Appadurai, Arjun 1996 Modernity at Large: Cultural Dimensions of Globalization. St. Paul: University of Minnesota Press.

770

Ashmore, Wendy 2002 “Decisions and Dispositions”: Socializing Spatial Archaeology: Archeology Division Distinguished Lecture 99th AAA Annual Meeting, San Francisco, CA, November 2000. American Anthropologist 104(4). New Series: 1172–1183. 2004 Classic Maya Landscape and Settlement. In Mesoamerican Archaeology: Theory & Practice. Julia A. Hendon and Rosemary A. Joyce, eds. Pp. 169–191. Malden, MA: Blackwell Publishing.

Ashmore, Wendy, and AB Knapp 1999 Archaeologies of Landscape: Contemporary Perspectives. New York: Wiley- Blackwell.

Aspelin, Karen 2005 Establishing Pedestrian Walking Speeds. Portland State University.

Aveni, Anthony 2002 Behind the Crystal Ball. Boulder, CO.: University Press of Colorado. 2010 Indian Origins of Lacrosse. Colonial Williamsburg Journal Winter 2010.

Ayala, G., and M. Fitzjohn 2002 Seeing Is Believing: Questions of Archaeological Visibility in the Mediterranean. Antiquity 76(292): 337–338.

Azoy, G. Whitney 2011 Buzkashi: Game and Power in Afghanistan, Third Edition. Waveland Press.

Badillo, Alex Elvis 2017 Narraciones de Quiechapa: Considerando La Evidencia Material Del Recorrido Sistematico. Cuadernos Del Sur 42. January-June: 35–54.

Balbus, Ike 1975 Politics as Sport: An Interpretation of the Political Ascendancy of the Sports Metaphor in America. In Stress and Contradiction in Modern Capitalism. Leon N. Lindberg, ed. Pp. 26–39. Lexington: Lexington Books.

Bale, John 1997 Identity, Identification and Image: Football and Place in the New Europe. In Football and Regional Identity in Europe. Siegfried Gehrmann, ed. Pp. 279–294. New Brunswick, N.J.: Transaction Publishers.

Balkansky, Andrew K. 1998a Origin and Collapse of Complex Societies in Oaxaca (Mexico): Evaluating the Era from 1965 to the Present. Journal of World Prehistory 12(4): 451–493. 1998b Urbanism and Early State Formation in the Huamelulpan Valley of Southern Mexico. Latin American Antiquity 9(1): 37–67. 2002 The Sola Valley and the Monte Albán State: A Study of Zapotec Imperial

771

Expansion. Memoirs of the University of Michigan Museum of Anthropology No. 36, 36. Ann Arbor: University of Michigan.

Balkansky, Andrew K., Stephen A. Kowalewski, Verónica Pérez Rodríguez, et al. 2000 Archaeological Survey in the Mixteca Alta of Oaxaca, Mexico. Journal of Field Archaeology 27(4): 365–389.

Ball, Hugh G., and David L. Brockington 1978 Trade and Travel in Prehispanic Oaxaca. In Mesoamerican Communication Routes and Cultural Contacts. Thomas Lee and Navarrete, eds. Pp. 107–114.

Balstrøm, Thomas 2002 On Identifying the Most Time-Saving Walking Route in a Trackless Mountainous Terrain. Geografisk Tidsskrift-Danish Journal of Geography 102(1): 51–58.

Barabas, Alicia, and Miguel Alberto Bartolomé, eds. 1999 Configuraciones Étnicas En Oaxaca: Perspectivas Etnográficas Para Las Autonomías, vol.1. 1a ed. Coleccion Obra Diversa. México, D.F: CONACULTA, Instituto Nacional de Antropología e Historia : Instituto Nacional Indigenista.

Barber, Sarah B. 2013 Defining Community and Status at Outlying Sites During the Terminal Formative. In Polity and Ecology in Formative Period Coastal Qaxaca. Arthur A. Joyce, ed. Pp. 165– 192. Boulder, CO: University Press of Colorado.

Barber, Sarah B., and Arthur A. Joyce 2007 Polity Produced and Community Consumed: Negotiating Political Centralization in the Lower Río Verde Valley, Oaxaca. In Mesoamerican Ritual Economy. E. Christian Wells and Karla L. Davis-Salazar, eds. Boulder, CO: University Press of Colorado.

Barber, Sarah B., Arthur A. Joyce, Arion T. Mayes, José Aguilar, and Michelle Butler 2013 Formative Period Burial Practices and Cemetaries. In Polity and Ecology in Formative Period Coastal Qaxaca. Arthur A. Joyce, ed. Pp. 97–134. Boulder, CO.: University Press of Colorado.

Barber, Sarah Barr 2005 Heterogeneity, Identity, and Complexity: Negotiating Status and Authority in Terminal Formative Coastal Oaxaca. Ph.D., University of Colorado at Boulder. http://search.proquest.com.ezproxy2.library.arizona.edu/pqdtft/docview/305009876/abstrac t/EBDFF2E125F54BBAPQ/1?accountid=8360, accessed May 22, 2014.

Baron, Joanne P 2006 Sport, Community, and the Ballgame at Yalbac, Belize. University of Pennsylvania.

Barrett, John C. 2000 A Thesis on Agency. In Agency in Archaeology. John E. Robb and Marica-Anne Dobres, eds. Pp. 61–68. London: Routledge.

772

Barton, C. Michael, Isaac I. Ullah, and Sean Bergin 2010 Land Use, Water and Mediterranean Landscapes: Modelling Long-Term Dynamics of Complex Socio-Ecological Systems. Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences 368(1931): 5275–5297.

Baskes, Jeremy 2000 Indians, Merchants, and Markets: A Reinterpertation of the Repartimiento and Spanish-Indian Economics. Stanford: Stanford University Press. 2005 Colonial Institutions and Cross-Cultural Trade: Repartimiento Credit and Indigenous Production of Cochineal in Eighteenth-Century Oaxaca, Mexico. The Journal of Economic History 65(1): 186–210.

Basso, Keith H. 1996 Wisdom Sits in Places: Landscape and Language Among the Western Apache. Albuquerque, N.M., Albuquerque: University of New Mexico Press. http://ezproxy.library.arizona.edu/login?url=http://www.netLibrary.com/urlapi.asp?action= summary&v=1&bookid=22747, accessed August 13, 2018.

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.

Beisser, Arnold R. 1967 The Madness in Sports: Psycholosocial Observations on Sports. Appleton-Century- Crofts.

Bell, Catherine M. 1992 Ritual Theory, Ritual Practice. New York: Oxford University Press. 1997 Ritual Perspectives and Dimensions. New York: Oxford University Press.

Bell, Tyler, and Gary Lock 2000 Topographic and Cultural Influences on Walking the Ridgeway in Later Prehistoric Times. In Beyond the Map: Archaeology and Spatial Technologies. Gary Lock, ed. Pp. 85– 100. Omaha: IOS Press.

Bender, Barbara 1993 Landscape : Politics and Perspectives. Oxford: Berg Publishers. http://www.bcin.ca/Interface/openbcin.cgi?submit=submit&Chinkey=204407, accessed November 27, 2014. 1999 Subverting the Western Gaze: Mapping Alternative Worlds. In The Archaeology and Anthropology of Landscape. Peter J. Ucko and Robert Layton, eds. Pp. 632–634. London: Routledge. 2002 Time and Landscape. Current Anthropology 43(S4): S103–S112.

Bender, Barbara, Sue Hamilton, and Christopher Tilley 1997 Leskernick: Stone Worlds; Alternative Narratives; Nested Landscapes. Proceedings

773 of the Prehistoric Society 63: 147–178.

Bendix, Reinhard 1977 Max Weber: An Intellectual Portrait. Berkeley: University of California Press.

Berger, Martin 2009 Pelota Mixteca. University of Leiden.

Bernal, Igancio, and Andy Seuffert 1979 The Ballplayers of Dainzú. Verlagsanst: Graz : Akadem. Druk-u.

Berry, Joseph W. 2003 Historic Sanborn Maps in the Digital Age. Journal of GIS in Archaeology 1: 73–78.

Bevan, A., C. Frederick, and A. Krahtopoulou 2003 A digital Mediterranean countryside: GIS approaches to the spatial structure of the post-medieval landscape on Kythera (Greece). Archeologia e Calcolatori 14: 217–236.

Bevan, Andrew, and James Conolly 2004 GIS, Archaeological Survey, and Landscape Archaeology on the Island of Kythera, Greece. Journal of Field Archaeology 29(1–2): 123–138.

Blanchard, Kendall 1981 The Mississippi Choctaws at Play: The Serious Side of Play. Urbana: University of Illinois Press. 1996 The Anthropology of Sport: An Introduction. ABC-CLIO.

Blanchard, Kendall, and Alyce Tayce Cheska 1985 The Anthropology of Sport: An Introduction. South Hadley, MA: Bergin & Harvey.

Blanton, Richard E. 1978 Monte Albán: Settlement Patterns at the Ancient Zapotec Capital. Studies in Archeology; New York: Academic Press. 1994 Houses and Households: A Comparative Study. Springer Science & Business Media.

Blanton, Richard E., Stephen A. Kowalewski, Gary M. Feinman, and J. Appel 1982 Monte Albán’s Hinterland, Part 1: Prehispanic Settlement Patterns of the Central and Southern Parts of the Valley of Oaxaca, Mexico, vol.Prehistory and Human Ecology of the Valley of Oaxaca. Memoirs of the University of Michigan Museum of Anthropology, 15. Ann Arbor: University of Michigan.

Blitz, John H 1993 Big Pots for Big Shots: Feasting and Storage in a Mississippian Community. AMERICAN ANTIQUITY 58(1): 18.

Bloch, Maurice

774

1989 Ritual, History, and Power: Selected Papers in Anthropology. Monographs on Social Anthropology, 58. London: London School of Economics.

Blomster, Jeffrey P., ed. 2008 After Monte Albán: Transformation and Negotiation in Oaxaca, Mexico. Mesoamerican Worlds; Variation: Mesoamerican Worlds. Boulder, CO.: University Press of Colorado. 2012 Early Evidence of the Ballgame in Oaxaca, Mexico. PNAS 109(21): 8020–8025.

Borgstede, Greg, and James R. Mathieu 2007 Defensibility and Settlement Patterns in the Guatemalan Maya Highlands. Latin American Antiquity 18(2): 191–211. de Borhegyi, Stephan Francis 1966 A Miniature Ceremonial Ballgame Yoke from Mexico. American Antiquity 31(5): 742–744.

Bourdieu, Pierre 1973 The Berber House or the World Reversed. In Rules and Meanings. Pierre Bourdieu and Mary Douglas, eds. Pp. 98–110. New York: Penguin. 1977 Outline of a Theory of Practice. Cambridge: Cambridge University Press. 1984 Distinction: A Social Critique of the Judgement of Taste. Cambridge, Mass.: Harvard University Press. 1985 The Social Space and the Genesis of Groups. Social Science Information 24(2): 195–220.

Bradley, Richard 1991 Ritual Time and History. New York: Oxford University Press. 2005 Ritual and Domestic Life in Prehistoric Europe. London: Routledge.

Bray, Tamara L. 2003a The Commensal Politics of Early States and Empires. In The Archaeology and Politics of Food and Feasting in Early States and Empires. Tamara L. Bray, ed. Pp. 1–16. New York: Kluwer Academic Publishers. 2003b The Archaeology and Politics of Food and Feasting in Early States and Empires. Boston, MA: Springer US. http://ezproxy.library.arizona.edu/login?url=http://dx.doi.org/10.1007/b100538, accessed August 12, 2018.

Briault, Camilla 2007 Making Mountains Out of Molehills in the Bronze Age Aegean: Visibility, Ritual Kits, and the Idea of a Peak Sanctuary. World Archaeology 39(1): 122–141.

Brockington, Lolita Gutiérrez 1989 The Leverage of Labor: Managing the Cortés Haciendas in Tehuantepec, 1588- 1688. Duke University Press.

775

Brown, Linda A. 2000 From Discard to Divination: Demarcating the Sacred Through the Collection and Curation of Discarded Objects. Latin American Antiquity 11(4): 319–333.

Browner, C. H., and Sondra T. Perdue 1988 Women’s Secrets: Bases for Reproductive and Social Autonomy in a Mexican Community. American Ethnologist 15(1): 84–97.

Brubaker, Roger, and Cooper Frederick 2000 Beyond Identity. Theory and Society 29: 1–47.

Brück, Joanna 2005 Experiencing the Past? The Development of a Phenomenological Archaeology in British Prehistory. Archaeological Dialogues 12(1): 45–72.

Brumfiel, Elizabeth M., and John W. Fox 1994 Factional Competition and Political Development in the New World. New Directions in Archaeology. Cambridge [England] ; New York, NY: Cambridge University Press.

Buchignani, Norman 1987 Ethnic Phenomena and Contemporary Social Theory: Their Implications for Archaeology. Ethnicity and Culture: 15–24.

Burgoa, Francisco de 1934 Geográfica Descripción. Publicaciones Del Archivo General de La Nación XXVI. Mexico: Talleres Gráficos e la Nación.

Burley, David V. 1996 Sport, Status, and Field Monuments in the Polynesian Chiefdom of Tonga: The Pigeon Snaring Mounds of Northern Ha’apai. Journal of Field Archaeology 23(4): 421– 435.

Butler, Judith 1990 Gender Trouble and the Subervision of Identity. New York and London: Routledge.

Byerly, Ryan M., Judith R. Cooper, David J. Meltzer, Matthew E. Hill, and Jason M. LaBelle 2005 On Bonfire Shelter (Texas) as a Paleoindian Bison Jump: An Assessment Using GIS and Zooarchaeology. American Antiquity 70(4): 595–629.

Byland, Bruce E., and John M.D. Pohl 1994 In the Realm of 8 Deer: The Archaeology of the Mixtec Codices. Tulsa: University of Oklahoma Press.

Calvijero, Francesco 1914 Historia Antigua de México : Sacada de Los Mejores Historiadores Españoles, y de

776

Los Manuscritos y de Las Pinturas Antiguas de Los Indios Dividida En Diez Libros: Adornada Con Mapas y Estampas. Mexico City: Departamento Editorial de la Dirección General de las Bellas Artes.

Cameron, Catherine M., and Lindsay D. Johansson 2017 The Biggest Losers: Gambling and Enslavement in Native North America. In Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 273–285. Salt Lake City: University of Utah Press.

Canter, David 1977 The Psychology of Place. St. Martin Press.

Canuto, Marcello A., and Jason Yaeger 2000 The Archaeology of Communities: A New World Perspective. London: Routledge.

Canuto, Marcello Andrea 2002 A Tale of Two Communities: Social and Political Transformation in the Hinterlands of the Maya Polity of Copan. Ph.D., University of Pennsylvania. http://search.proquest.com.ezproxy2.library.arizona.edu/pqdtft/docview/305556673/abstrac t/AA003C73845D4AFEPQ/1?accountid=8360, accessed June 22, 2014.

Carballo, David M., and Thomas Pluckhahn 2007 Transportation Corridors and Political Evolution in Highland Mesoamerica: Settlement Analyses Incorporating GIS for Northern Tlaxcala, Mexico. Journal of Anthropological Archaeology 26(4): 607–629.

Carrasco, David 1999 City of Sacrifice: The Aztec Empire and the Role of Violence. Boston, MA: Beacon Press.

Carrer, Francesco 2013 An Ethnoarchaeological Inductive Model for Predicting Archaeological Site Location: A Case-Study of Pastoral Settlement Patterns in the Val Di Fiemme and Val Di Sole (Trentino, Italian Alps). Journal of Anthropological Archaeology 32(1): 54–62.

Casana, Jesse, and Jackson Cothren 2008 Stereo Analysis, DEM Extraction and Orthorectification of CORONA Satellite Imagery: Archaeological Applications from the Near East. Antiquity 82(317): 732–749.

Casella, Eleanor Conlin, and Chris Fowler 2005 Beyond Identification: An Introduction. In The Archaeology of Plural and Changing Identities. Eleanor Conlin Casella and Chris Fowler, eds. Pp. 1–8. New York: Kluwer/Plenum.

Casey, Edward 1997 The Fate of Place: A Philosophical Inquiry. Berkeley: University of California Press.

777

Catlin, George 1953 A Choctaw Ball Game. In Primitive Heritage: An Anthropological Anthology. Margaret Mead and Nicholas Calas, eds. Pp. 289–295. New York: Random House. de Certeau, Michel 1984 The Practice of Everyday Life. Berkeley: University of California Press.

Chambers, Erve J., and Philip D. Young 1979 Mesoamerican Community Studies: The Past Decade. Annual Review of Anthropology 8: 45–69.

Chance, John K. 1989 Conquest of the Sierra: Spaniards and Indians in Colonial Oaxaca. Norman: University of Oklahoma Press.

Chapman, Henry P. 2006 Landscape Archaeology. Gloucestershire: Tempus.

Cheska, Alyce Tayce 1975 Stewart Culin: An Early Ethnologist of Games. 1979 Native American Games as Strategies of Social Maintenance. In Forms of Play of Native North Americans. Edward Norbeck and Claire R. Farrer, eds. Pp. 227–247. Proceedings - American Ethnological Society ; 1977.

Cheska, Alyce Taylor 1984 Sport as Ethnic Boundary Maintenance: A Case of the American Indian. International Review for the Sociology of Sport 19(3–4): 241–257.

Clark, John E. 2004 Mesoamerica Goes Public: Early Ceremonial Centers, Leaders and Communities. In Mesoamerican Archaeology: Theory & Practice. Julia A. Hendon and Rosemary A. Joyce, eds. Pp. 43–72. Malden, MA: Blackwell Publishing.

Clendinnen, Inga 1991 Aztecs: An Interpretation. Cambridge: Cambridge University Press.

Clifford, James 1997 Routes: Travel and Translation in the Late Twentieth Century. Cambridge, MA.: Harvard University Press.

Clune, Francis J. 1963 A Functional and Historical Analysis of the Ball Game in Mesoamerica. PhD, University of California.

Cohodas, Marvin 1975 The Symbolism and Ritual Function of the Middle Classic Ball Game in

778

Mesoamerica. American Indian Quarterly 2(2): 99–130.

Collischonn, Walter, and Jorge Victor Pilar 2000 A Direction Dependent Least-Cost-Path Algorithm for Roads and Canals. International Journal of Geographical Information Science 14(4): 397–406.

Colwell-Chanthaphonh, Chip 2003 Native American Perspectives of the Past in the San Pedro Valley of Southeastern Arizona. KIVA 69(1): 5–29.

Connerton, Paul 1989 How Societies Remember. Cambridge University Press.

Conolly, James, and Mark Lake 2006 Geographical Information Systems in Archaeology. Cambridge Manuals in Archaeology. Cambridge, UK ; New York: Cambridge University Press.

Cook, Angel García, and Beatriz Leonor Merino Carrión 1998 Cantona: Urbe Prehispanica En El Altiplano Central de Mexico. Latin American Antiquity 9(3): 191–216.

Cosgrove, Denis E. 1984 Social Formation and Symbolic Landscape. Croom Helm Historical Geography Series. London: Croom Helm.

Couclelis, Helen 1999 Space, Time, Geography. Geographical Information Systems 1: 29–38.

Csikszentmihalyi, Mihaly, and Stith Bennett 1971 An Exploratory Model of Play. American Anthropologist 73(1): 45–58.

Culin, Stewart 1889 Chinese Games with Dice. Franklin printing Company. 1894 The Value of Games in Ethnology. Abstract presented at the Proceedings of the American Association for the Advancement of Science, 43: 355-358. 1898 American Indian Games. Journal of American Folk-Lore 11: 245–252. 1903 Report, Museum Expedition on a Collecting Expedition. Unpublished papers. New York: Brooklyn Museum. 1907 Games of the North American Indians. Bureau of American Ethnology Annual Report, 1902-1903, 24. Washington, D.C.: Smithsonian Institution Press.

Cummings, Vicki, and A. W. R. Whittle 2004 Places of Special Virtue: Megaliths in the Neolithic Landscapes of Wales. Cardiff Studies in Archaeology. Oxford : Oakville, CT: Oxbow ; David Brown Book Co.

Cusick, James 1998 Historiography of Acculturation: An Evaluation of Concepts and Their Application

779 in Archaeology. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology. James Cusick, ed. Pp. 126–145. Carbondale: Southern Illinois University Carbondale.

Daneels, Annick 1997 Settlement History in the Lower Cotaxtla Basin. In Olmec to Aztec: Settlement Patterns in the Ancient Gulf Lowlands. Barbara Stark and Philip J. Arnold III, eds. Pp. 206–252. Tucson, AZ.: University of Arizona Press. 2008 Ballcourts and Politics in the Lower Cotaxtla Valley: A Model to Understand Classic Central Veracruz? In . Philip J. Arnold III and C.A. Pool, eds. Pp. 197–223. Washington, D.C.: Dumbarton Oaks.

David, Bruno, and Julian Thomas 2016 Handbook of Landscape Archaeology. Routledge.

Dawson, Peter C., and Richard M. Levy 2005 A Three-Dimensional Model of a Thule Inuit Whale Bone House. Journal of Field Archaeology 30(4): 443–455.

Dawson, Peter, Richard Levy, Don Gardner, and Matthew Walls 2007 Simulating the Behaviour of Light inside Arctic Dwellings: Implications for Assessing the Role of Vision in Task Performance. World Archaeology 39(1): 17–35.

Day, Jane S. 2001 Performing on the Court. In The Sport of Life and Death: The Mesoamerican Ballgame. E. Michael Whittington, ed. Pp. 64–77. Charlotte, NC: Thames & Hudson.

DeCorse, Christopher R. 1989 Material Aspects of Limba, Yalunka and Kuranko Ethnicity: Archaeological Research in Northeastern Sierra Leone. In Archaeological Approaches to Cultural Identity. Stephen J. Shennan, ed. Pp. 125–140. London: Unwine Hyman.

Delgado, Agustin 1965 Archeological Reconnaissance in the Region of Tehuantepec, Oaxaca, Mexico. Papers of the New World Archaeological Foundation, 18. Provo, UT: New World Archaeological Foundation, Brigham Young University.

Dell, Melissa 2010 The Persistent Effects of Peru’s Mining Mita. Econometrica 78(6): 1863–1903.

DeMarrais, Elizabeth, Luis Jaime Castillo, and Timothy Earle 1996 Ideology, Materialization, and Power Strategies. Current Anthropology 37(1): 15– 31.

Dewar, Robert E. 1991 Incorporating Variation in Occupation Span into Settlement-Pattern Analysis. American Antiquity 56(4): 604–620.

780

1992 Remnant Settlement Patterns. In Space, Time, and Archaeological Landscapes. Jaqueline Rossignol and LuAnn Wandsnider, eds. Pp. 227–256. New York: Plenum.

Dietler, Michael 1996 Feasts and Commensal Politics in the Political Economy. In Food and the Status Quest. P. Wiessner and W. Schieffenhovel, eds. Pp. 87–125. Providence, RI: Berghahn Books. 2003 Clearing the Table: Some Concluding Reflections on Commensal Politics and Imperial States. In The Archaeology and Politics of Food and Feasting in Early States and Empires. Tamara L. Bray, ed. Boston, MA: Springer.

Dietler, Michael, and Ingrid Herbich 1998 Habitus, Techniques, Style: An Integrated Approach to the Social Understanding of Material Culture and Boundaries. In The Archaeology of Social Boundaries. Miriam T. Stark, ed. Pp. 232–263. Washington, D.C.: Smithsonian Institution Press.

Dietrich, Oliver, Manfred Heun, Jens Notroff, Klaus Schmidt, and Martin Zarnkow 2012 The Role of Cult and Feasting in the Emergence of Neolithic Communities. New Evidence from Göbekli Tepe, South-Eastern Turkey. Antiquity 86(333): 674–695.

Dirks, Nicholas B. 1994 Ritual and Resistance: Subversion As a Social Fact. In Culture, Power, History: A Reader in Contemporary Social Theory. Nicholas B. Dirks, Geoff Eley, and Sherry B. Ortner, eds. Pp. 483–503. Princeton, NJ: Princeton University Press.

Dobres, Marica-Anne, and John E. Robb 2000 Agency in Archaeology: Paradigm or Platitude? In Agency in Archaeology. John E. Robb and Marica-Anne Dobres, eds. Pp. 3–17. London: Routledge.

Dodd, Lynn S. 2005 Territory, Legacy, and Wealth in Iron Age Anatolia. In Untaming the Frontier in Anthropology, Archaeology and History. Bradley J. Parker and Lars Rodseth, eds. Pp. 238– 260. Tucson, AZ.: University of Arizona Press. van Dommelen, Peter 2002 Ambigous Matters: Colonialism and Local Identities in Punic Sardinia. In The Archaeology of Colonialism. C. Lyons and J. Papadopoulos, eds. Pp. 121–147. Los Angeles: Getty Research Institute.

Donnan, Hastings, and Thomas M. Wilson 1999 Borders: Frontiers of Identity, Nation and State. Oxford, Oxford ; New York: Berg. http://ezproxy.library.arizona.edu/login?url=http://site.ebrary.com/lib/arizona/Doc?id=1000 6746, accessed August 14, 2018.

Donnelly, Peter 1996 Approaches to Social Inequality in the Sociology of Sport. Quest 48(2): 221–242.

781

Dornan, Jennifer L. 2004 Beyond Belief: Religious Experience, Ritual, and Cultural Neuro-Phenomenology in the Interpretation of Past Religious Systems. Cambridge Archaeological Journal 14: 25– 36.

Douglas, Gavin 2013 Performing Ethnicity in Southern Shan State, Burma/Myanmar: The Ozi and Gong Traditions of the Myelat. Ethnomusicology 57(2): 185–206.

Dovey, Kim 1999 Framing Places: Mediating Power in Built Form. London: Routledge. http://www.tandfebooks.com/doi/abs/10.4324/9780203017579, accessed November 1, 2013.

Duff, Andrew I. 2000 Scale, Interaction, and Regional Analysis in Late Pueblo Prehistory. In The Archaeology of Regional Interaction: Religion, Warfare, and Exchange Across the American Southwest. Michelle Hegmon, ed. Pp. 71–98. Boulder, CO: University Press of Colorado.

Duncan, B., and Kristen A. Beckman 2000 Application of GIS Predictive Site Location Models within Pennsylvania and West Virginia. In Practical Applications of GIS for Archaeologists: A Predictive Modelling Toolkit. Konnie L. Wescott and R. Joe Brandon, eds. Pp. 23–33. London: Taylor & Francis.

Dunnell, Robert C. 1992 The Notion of Site. In Space, Time, and Archaeological Landscapes. Jaqueline Rossignol and LuAnn Wandsnider, eds. Pp. 21–42. New York: Plenum.

Dunnell, Robert C., and William S. Dancey 1983 The Siteless Survey: A Regional Scale Data Collection Strategy. In Advances in Archaeological Method and Theory. Michael B. Schiffer, ed. Pp. 267–287. New York: Academic Press.

Durán, Fray Diego 1971 Book of the Gods and Rites of the Ancient Calendar. Fernando Horcasitas and Doris Heydon, trans. Norman: University of Oklahoma Press. 1994 The History of the Indies of New Spain. Norman: University Oklahoma Press.

Dyck, Noel, and Eduardo Archetti 2003 Sport, Dance, and Embodied Identities.

Dyson-Bruce, Lynn 2003 Historic Landscape Assessment: The East of England Experience Paper Product to GIS Delivery. Journal of GIS in Archaeology 1: 63–72.

782

Earle, Duncan 1990 Apprppriating the Enemy: Highland Maya Religious Organization and Community Survival. Class, Politics, and Popular Religion in Mexico and Central America 10: 115– 139.

Earle, Timothy 1994 Wealth Finance in the Inka Empire: Evidence from the Calchaqui Valley, Argentina. American Antiquity 59(3): 443–460.

Ebdon, David 1985 Statistics in Geography. Oxford: Basil-Blackwell.

Ebert, James I., Eileen L. Camilli, and Michael J. Berman 1996 GIS in the Analysis of Distributional Archaeological Data. In New Methods, Old Problems: Geographic Information Systems in Modern Archeological Research. Herbert D. G. Maschner, ed. Pp. 25–37. Occasional Paper, 23. Carbondale: Center for Archaeological Investigations, Southern Illinois University Carbondale.

Eckholm, Susanna M. 1991 Ceramic Figurines and the Mesoamerican Ballgame. In The Mesoamerican Ballgame. Vernon L. Scarborough and David R. Wilcox, eds. Pp. 241–251. Tucson, AZ.: University of Arizona Press.

Edwards, Harry 1973 Definitions and Clarification in Sociology of Sport. Homewood, IL: Dorsey.

Ekholm, Gordon F. 1946 The Probable Use of Mexican Stone Yokes. American Anthropologist 48(4): 593– 606.

Elliott, Michelle 2005 Evaluating Evidence for Warfare and Environmental Stress in Settlement Pattern Data from the Malpaso Valley, Zacatecas, Mexico. Journal of Anthropological Archaeology 24(4): 297–315.

Elson, Christina M. 2006 Intermediate Elites and the Political Landscape of the Early Zapotec State. In Intermediate Elites in Pre-Columbian States and Empires. Christina M. Elson and R. Allen Covey, eds. Pp. 44–67. Tucson, AZ.: University of Arizona Press.

Elton, Hugh 1996 Frontiers of the Roman Empire. Bloomington: Indiana University Press.

Emberling, Geoff 1997 Ethnicity in Complex Societies: Archaeological Perspectives. Journal of Archaeological Research 5(4): 295–344.

783

Evans, Susan T. 2017 Aztec Gambling and Magical Thinking. In Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 259–272. Salt Lake City: University of Utah Press.

Fábrega-Álvarez, Pastor 2006 Moving without Destination. A Theoretical, GIS-Based Determination of Routes (Optimal Accumulation Model of Movement from a Given Origin). Archaeological Computing Newsletter: 7–11.

Fairclough, Graham 1992 Meaningful Constructions: Spatial and Functional Analysis of Medieval Buildings. Antiquity 66: 348–366.

Feinman, Gary M., and Linda M. Nicholas 1990 At the Margins of the Monte Albán State: Settlement Patterns in the Ejutla Valley, Oaxaca, Mexico. Latin American Antiquity 1(3): 216–246. 2011 The Ballcourt at El Palmillo: Implications for Late Classic Oaxaca, Mexico. Nicholas, Linda M. Mexico Berlin: Verlag A. Saurwein 33(4): 98–104.

Feld, Steven, and Keith H. Basso, eds. 1996 Senses of Place. Sante Fe: School of American Research Press.

Field, Julie S. 2005 Land Tenure, Competition and Ecology in Fijian Prehistory. Antiquity 79(305): 586–600.

Finsten, Laura, Stephen A. Kowalewski, Charlotte A. Smith, Mark D. Borland, and Richard D. Garvin 1996 Circular Architecture and Symbolic Boundaries in the Mixtec Sierra, Oaxaca. Ancient Mesoamerica 7(01): 19–35.

Fisher, Kevin D. 2009 Placing Social Interaction: An Integrative Approach to Analyzing Past Built Environments. Journal of Anthropological Archaeology 28(4): 439–457.

Fisher, Peter 1991 First Experiments in Viewshed Uncertainty: The Accuracy of the Viewshed Area. Photogrammetric Engineering & Remote Sensing 57: 1321–1327.

Fisher, Peter F 1992 First Experiments in Viewshed Uncertainty: Simulating Fuzzy Viewsheds. Photogrammetric Engineering and Remote Sensing 58: 345–352.

Fisher, Peter F. 1994 Probable and Fuzzy Models of the Viewshed Operation. In Innovations in GIS. Michael F. Warboys, ed. Pp. 161–175. London: Taylor & Francis.

784

1995 An Exploration of Probable Viewsheds in Landscape Planning. Environment and Planning B: Planning and Design 22(5): 527–546. 1996 Reconsideration of the Viewshed Function in Terrain Modeling. Geographical Systems 3: 33–58. 1998 Improved Modeling of Elevation Error with Geostatistics. GeoInformatica 2: 215– 233. 1999 Probable and Fuzzy Models of the Viewshed Operation. In Innovations in GIS 1. Michael F. Warboys, ed. Pp. 161–175. London: Taylor & Francis.

Fitzjohn, Matthew 2007 Viewing Places: GIS Applications for Examining the Perception of Space in the Mountains of Sicily. World Archaeology 39(1): 36–50.

Flannery, Kent V. 1968 Archaeological Systems Theory and Early Mesoamerica. Anthropological Archaeology in the Americas: 67–87.

Flannery, Kent V., and Marcus, eds. 1983 The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. A School of American Research Book; Variation: School of American Research Book. New York: Academic Press.

Flannery, Kent V., and Joyce Marcus 1976 Evolution of the Public Building in Formative Oaxaca. In Cultural Change and Continuity: Essays in Honor of James Bennett Griffin. Charles E. Cleland and James B. Griffin, eds. Pp. 205–221. New York: Academic Press. 2000 Formative Mexican Chiefdoms and the Myth of the “Mother Culture.” Journal of Anthropological Archaeology 19(1): 1–37.

Fleming, Andrew 1999 Phenomenology and the Megaliths of Wales: A Dreaming Too Far? Oxford Journal of Archaeology 18(2): 119–125. 2006 Post-Processual Landscape Archaeology: A Critique. Cambridge Archaeological Journal 16(03): 267.

Fogelin, Lars 2007 The Archaeology of Religious Ritual. Annual Reviews in Anthropology 36: 55–71.

Fontijn, David 2007 The Significance of ‘Invisible’ Places. World Archaeology 39(1): 70–83.

Forde, Jaime 2002 Ideology, Identity, and Icons: A Study of Mixtec Polychrome Pottery from Late Postclassic Yucu Dzaa (Tututepec), Oaxaca, Mexico. M.A., Unversity of Colorado at Boulder.

Foucault, Michel

785

1979 Discipline and Punishment: The Birth of the Prison. New York: Vintage Books. 1984 Structuralism and Post-Structuralism. In Michel Foucault Aesthetics Pp. 433–463. Essential Works of Foucault.

Fox, John G. 1994 Putting the Heart Back in the Court: Ballcourts and Ritual Action in Mesoamerica. PhD, Harvard University. 1996 Playing with Power: Ballcourts and Political Ritual in Southern Mesoamerica. Current Anthropology 37(3): 489–509.

Fox, John W. 1991 The Lords of Light Versus the Lords of Dark: The Postclassic Highland Maya Ballgame. In The Mesoamerican Ballgame. Vernon L. Scarborough and David R. Wilcox, eds. Tucson, AZ.: University of Arizona Press.

Frank, Andre Gunder, and Barry K. Gills 1996 The World System: Five Hundred Years Or Five Thousand? London: Routledge.

Freidel, David M., Linda Schele, and Joy Parker 1993 Maya Cosmos: Three Thousand Years on the Shaman’s Path. New York: William Morrow and Company, Inc.

Friedman, Richard A., John R. Stein, and Taft Blackhorse Jr. 2003 A Study of a Pre-Columbian Irrigation System at Newcomb, New Mexico. Journal of GIS in Archaeology 1: 4–10.

Frieman, Catherine, and Mark Gillings 2007 Seeing Is Perceiving? World Archaeology 39(1): 4–16.

Gaffney, V., and M. van Leusen 1995 Postscript: GIS, Environmental Determiism, and Archaeology: A Parallel Text. In Archaeology and Geographical Information Systems: A European Perspective. Gary R. Lock and Z. Stancic, eds. Pp. 367–376. Oxford: Taylor & Francis.

Gaffney, V., Z. Stancic, and H. Watson 1996 Moving from Catchments to Cognition: Tentative Steps towards a Larger Archaeological Context for GIS. In Anthropology, Space, and Geographic Information Systems. Mark S. Aldenderfer and Herbert Maschner, eds. Pp. 132–154. New York: Oxford University Press.

Galaty, Michael L. 2005 European Regional Studies: A Coming of Age? Journal of Archaeological Research 13(4): 291–336.

García, A César González, and Juan Antonio Belmonte 2014 Sacred Architecture Orientation Across the Mediterranean: A Comparative Statistical Analysis. Mediterranean Archaeology Archaeometry 14(2): 95–113.

786

García, Nelly Robles, and Augustín E. Andrade Cuatle 2011 El Proyecto Arqueológico Del Conjunto Monumental de Atzompa. In Monte Albán En La Encrucijada Regional y Disciplinaria. Memoria de La Quinta Mesa Redonda de Monte Albán. Oaxaca, Mexico: CENTRO INAH Oaxaca. de la Garza, Mercedes 2000 El Juego de Pelota Según Las Fuentes Escritas. Arqueología Mexicana: 50–53.

Gasco, Janine L. 2005 The Consequences of Spanish Colonial Rule for the Indigenous Peoples of Chiapas, Mexico. In The Postclassic to Spanish-Era Transition in Mesoamerica. In The Postclassic to Spanish-Era Transition in Mesoamerica: Archaeological Perspectives. Susan Kepecs and Rani T. Alexander, eds. Pp. 77–96. Albuquerque: University of New Mexico Press.

Gay, Jose Antonio 1982 Historia de Oaxaca. Porrúa, México.

Gearing, Frederick O. 1958 The Structural Poses of I8th Century Cherokee Villages1. American Anthropologist 60(6): 1148–1157. 1962 Priests and Warriors: Structures for Cherokee Politics in the Eighteenth Century. American Anthropological Association Memoir, 93. Washington, D.C.: American Anthropological Association.

Geertz, Clifford 1973 Deep Play: Notes on the Balinese Cockfight. In The Interpretation of Cultures. New York: Basic Books.

Gibson, T. H. 2005 Off the Shelf: Modeling and Management of Historical Resources. In Predictive Modelling for Archaeological Heritage Management: A Research Agenda. Nederlandse Archeologische Rapporten 29 Pp. 205–223. Amersfoot: Rijksdienst voor het Oudheidkundig Bodermonderzoek.

Giddens, Anthony 1979 Central Problems in Social Theory. Berkeley: University of California Press. 1984 The Constitution of Society: Outline of a Theory of Structuration. Berkeley: University of California Press.

Gilberti, Fray Maturino 1901 Diccionario de La Lengua Tarasca o de Michoacán. Mexico: Ed. A Peñafiel.

Giles, Kate 2007 Seeing and Believing: Visuality and Space in Pre-Modern England. World Archaeology 39(1): 105–121.

787

Gillespie, Susan D. 1991 Ballgames and Boundaries. In The Mesoamerican Ballgame. Vernon L. Scarborough and David R. Wilcox, eds. Pp. 317–346. Tucson, AZ.: University of Arizona Press.

Gillings, Mark 2012 Landscape Phenomenology, GIS and the Role of Affordance. Journal of Archaeological Method and Theory 19(4): 601–611.

Gillings, Mark, D. Mattingly, and J. Van Dalen, eds. 1999 Geographical Information Systems and Landscape Archaeology. Oxford: Oxbow Books.

Goldstein, Paul S. 2000 Communities Without Borders - The Vertical Archipelago and Diaspora Communities in the Southern Andes. In The Archaeology of Communities: A New World Perspective. Marcello A. Canuto and Jason Yaeger, eds. Pp. 182–209. New York: Routledge.

Gonlin, Nancy, and Jon C. Lohse 2007 Commoner Ritual and Ideology in Ancient Mesoamerica. Boulder, CO: University Press of Colorado.

Goodchild, M. F., and D. G. Janelle 2004 Spatially Integrated Social Science: Examples in Best Practice. Oxford: Oxford University Press.

Graves, Dorothy 2011 The Use of Predictive Modelling to Target Neolithic Settlement and Occupation Activity in Mainland Scotland. Journal of Archaeological Science 38(3): 633–656.

Gregory, Ian N., and Paul S. Ell 2006 Error‐sensitive Historical GIS: Identifying Areal Interpolation Errors in Time‐series Data. International Journal of Geographical Information Science 20(2): 135–152.

Gruneau, Richard 1975 Sport, Social Differentiation, and Social Inequality. In Sport and Social Order: Contributions to the Sociology of Sport. Donald W. Ball and John W. Loy, eds. Pp. 121– 134. Boston, MA: Addison Wesley Publishing Company.

Gumilla, José 1745 El Orinoco Ilustrado y Defendido, vol.3. P. I. Obregon, ed. España: España Misionera. MDCCLXXXXI. Historia Natural Civil y Geográfica de Las Naciones Situadas En Las Riveras Del Rio Orinoco. Barcelona: Imprenta Carlos Gibert y Tuto.

Gutierrez, Gerardo 2017 Acrobatic Dances and Games of Mesoamerican as Ritual-Entertainment. In

788

Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 235–258. Salt Lake City: University of Utah Press.

Haddon, Alfred Cort 1898 The Study of Man. G.P. Putnam’s Sons.

Hall, Edward Twitchell 1966 The Hidden Dimension. Doubleday.

Hamann, Byron 2002 The Social Life of Pre‐Sunrise Things: Indigenous Mesoamerican Archaeology. Current Anthropology 43(3): 351–382.

Hamilton, Sue, Ruth Whitehouse, Keri Brown, et al. 2006 Phenomenology in Practice: Towards a Methodology for a ‘Subjective’ Approach. European Journal of Archaeology 9(1): 31–71.

Hamnett, Brian R. 1971 Dye Production, Food Supply, and the Laboring Population of Oaxaca, 1750-1820. The Hispanic American Historical Review 51(1): 51–78.

Hann, John H. 1988 Apalachee: The Land between the Rivers. Gainesville, FL.: University Press of Florida.

Hannerz, Ulf 1996 Transnational Connections: Cultures, People, Places. London: Routledge.

Hare, Timothy S. 2000 Between the Household and the Empire: Structural Relationships Within and Among Aztec Communities and Polities. In The Archaeology of Communities: A New World Perspective. Marcello A. Canuto and Jason Yaeger, eds. Pp. 78–101. London: Routledge. 2004 Using Measures of Cost Distance in the Estimation of Polity Boundaries in the Postclassic Yautepec Valley, Mexico. Journal of Archaeological Science 31(6): 799–814.

Hargreaves, J. A. 1987 Victorian Familism and the Formative Years of Female Sport. Victorian Familism and the Formative Years of Female Sport.: 130–144.

Hargreaves, John 1987 Sport, Power and Culture. The Times Literary Supplement 4388: 491.

Harris, Trevor 2000 Moving GIS: Exploring Movment Within Prehistoric Cultural Landscapes Using GIS. In Beyond the Map: Archaeology and Spatial Technologies. Gary R. Lock, ed. Pp. 116–123. Amsterdam: IOS Press.

789

Hartz, Louis 1969 The Founding of New Societies: Studies in the History of the United States, Latin America, South Africa, Canada, and Australia. Houghton Mifflin Harcourt.

Healen, D. M. 1993 Urbanism at Tula From the Perspective o f Residential Archaeology. In Prehispanic Domestic Units in Western Mesoamerica: Studies of the Household, Compound, and Residence. Robert S. Santley and Kenneth G. Hirth, eds. Pp. 105–119. Boca Raton: CRC Press.

Hegmon, Michelle 1998 Technology, Style, and Social Practice: Archaeological Approaches. In The Archaeology of Social Boundaries. Miriam T. Stark, ed. Pp. 264–280. Washington, D.C.: Smithsonian Institution Press.

Hegmon, Michelle, Kelley Hays-Gilpin, Randall H. McGuire, A. E. Rautman, and S. H. Schlanger 2000 Changing Perceptions of Regional Interaction in the Prehistoric. In Archaeology of Regional Interaction: Religion, Warfare, and Exchange Across the American Southwest and Beyond. Michelle Hegmon, ed. Pp. 1–21. Boulder, CO.: University Press of Colorado.

Hendon, Julia A. 2003 Feasting at Home: Community and House Solidarity among the Maya of Southeastern Mesoamerica. In The Archaeology and Politics of Food and Feasting in Early States and Empires. Tamara L. Bray, ed. Pp. 203–234. New York: Kluwer Academic Publishers. 2010 Houses in a Landscape: Memory and Everyday Life in Mesoamerica. Duke University Press.

Hendon, Julia, Rosemary Joyce, and Russell Sheptak 2009 Heterarchy as Complexity: Archaeology in Yoro, Honduras. Anthropology Faculty Publications: 1–15.

Hepp, Guy 2015 La Consentida: Initial Early Formative Settlement, Subsistence, and Social Organization on the Pacific Coast of Oaxaca, Mexico. Doctoral, University of Colorado at Boulder.

Higuchi, Tadahiko 1983 The Visual and Spatial Structure of Landscapes. Cambridge, MA.: MIT Press.

Hill, Warren D. 1999 Ballcourts, Competitive Games, and the Emergence of Complex Society. PhD, University of British Columbia.

Hill, Warren D., Michael Blake, and John E. Clark

790

1998 Ball Court Design Dates Back 3,400 Years. Nature 392(6679): 878–879.

Hill, Warren D., and John E. Clark 2001 Sports, Gambling, and Government: America’s First Social Compact? Americ 103: 331–345.

Hodder, Ian 2000 Agency and Individuals in Long-Term Processes. In Agency in Archaeology. John E. Robb and Marica-Anne Dobres, eds. Pp. 21–33. London: Routledge.

Hodder, Ian, and Clive Orton 1976 Spatial Analysis in Archaeology. Cambridge: Cambridge University Press.

Holdaway, Simon, Rod Wallace, Russell Gibb, et al. 2005 Archaeology Without Squares: A Complicated System for Recording and Visiualizing the Excavation of a 19th Century Maori Village. Journal of Polynesian Society 114(3): 267–283.

Hosmer, David W., Borko Jovanovic, and Stanley Lemeshow 1989 Best Subsets Logistic Regression. Biometrics 45(4): 1265–1270.

Hosmer, David W., and Stanley Lemeshow 1989 Applied Logistic Regression. New York: Wiley.

Howey, Meghan C.L. 2007 Using Multi-Criteria Cost Surface Analysis to Explore Past Regional Landscapes: A Case Study of Ritual Activity and Social Interaction in Michigan, AD 1200–1600. Journal of Archaeological Science 34(11): 1830–1846.

Huizinga, Johan 1955 Homo Ludens; A Study of the Play-Element in Culture. Beacon Paperbacks ; 15. Boston: Beacon Press.

Humphrey, Caroline, and James Laidlaw 1994 The Archetypal Actions of Ritual : A Theory of Ritual Illustrated by the Jain Rite of Worship. New York: Clarendon Press. 2007 Sacrifice and Ritualization. In The Archaeology of Ritual. Evangelos Kyriakidis, ed. Los Angeles: University of California Press.

Hutson, Scott R. 2014 Artefactos Prehispánicos de La Sierra Mixe. In Panorama Arqueológico: Dos Oaxacas. Marcus Winter and Gonzalo Sánchez Santiago, eds. Pp. 267–278. Oaxaca, Mexico: CENTRO INAH Oaxaca.

Inomata, Takeshi 2006 Plazas, Performers, and Spectators: Political Theaters of the Classic Maya. Current Anthropology 47(5): 805–842.

791

Inomata, Takeshi, and Lawrence S. Coben 2006a Overture: An Invitation to the Archaeological Theater. In Archaeology of Performance: Theaters of Power, Community, and Politics. Takeshi Inomata and Lawrence S. Coben, eds. Pp. 11–46. Oxford: AltaMira Press. , eds. 2006b Archaeology of Performance: Theaters of Power, Community, and Politics. Oxford: AltaMira Press.

Inomata, Takeshi, Jessica MacLellan, Daniela Triadan, et al. 2015 Development of Sedentary Communities in the Maya Lowlands: Coexisting Mobile Groups and Public Ceremonies at Ceibal, Guatemala. Proceedings of the National Academy of Sciences: 201501212.

Inomata, Takeshi, Daniela Triadan, Kazuo Aoyama, Victor Castillo, and Hitoshi Yonenobu 2013 Early Ceremonial Constructions at Ceibal, Guatemala, and the Origins of Lowland Maya Civilization. Science 340(6131): 467–471.

Jennings, Justin, and Nathan Craig 2001 Politywide Analysis and Imperial Political Economy: The Relationship between Valley Political Complexity and Administrative Centers in the Wari Empire of the Central Andes. Journal of Anthropological Archaeology 20(4): 479–502.

Johnson, M. H. 2006 On the Nature of Theoretical Archaeology and Archaecological Theory. Archaeological Dialogues 13(2): 117–132.

Jones, Eric E. 2006 Using Viewshed Analysis to Explore Settlement Choice: A Case Study of the Onondaga Iroquois. American Antiquity 71(3): 523–538.

Jones, Geoffrey, and Gene Munson 2005 Geophysical Survey as an Approach to the Ephemeral Campsite Problem: Case Studies from the Northern Plains. Plains Anthropologist 50(193): 31–43.

Jones, Sian 1997 Discourses of Identity in the Interpretation of the Past. In Cultural Identity and Archaeology: The Construction of European Communities. Paul Graves-Brown, Sian Jones, and Clive Gamble, eds. Pp. 62–80. London: Routledge.

Joyce, Arthur A. 2004 Sacred Space and Social Relations in the Valley of Oaxaca. In Mesoamerican Archaeology: Theory & Practice. Julia A. Hendon and Rosemary A. Joyce, eds. Pp. 192– 216. Malden, MA: Blackwell Publishing. 2009 The Main Plaza of Monte Albán: A Life History of Place. In The Archaeology of Meaningful Places. Brenda Bowser and María Nieves Zedeño, eds. Pp. 32–52. Salt Lake

792

City: University of Utah Press. 2010 Mixtecs, Zapotecs, and Chatinos: Ancient Peoples of Southern Mexico. The Peoples of America; Variation: Peoples of America. Malden, Mass.: Wiley-Blackwell.

Joyce, Arthur A., Laura Arnaud Bustamante, and Marc N. Levine 2001 Commoner Power: A Case Study from the Classic Period Collapse on the Oaxaca Coast. Journal of Archaeological Method and Theory 8(4): 343–385.

Joyce, Arthur A., Marc N. Levine, and Sarah B. Barber 2013 Place-Making and Power in the Terminal Formative: Excavations on Rio Viejo’s Acropolis. In Polity and Ecology in Formative Period Coastal Qaxaca. Arthur A. Joyce, ed. Pp. 135–163. Boulder, CO: University Press of Colorado.

Joyce, Arthur A., and Erinn T. Weller 2007 Commoner Rituals, Resistance, and the Classic-to-Postclassic Transition. In Commoner Ritual and Ideology in Ancient Mesoamerica. Nancy Gonlin and Lohse, eds. Pp. 141–182. Boulder, CO: University Press of Colorado.

Joyce, Arthur A., and Marcus Winter 1996 Ideology, Power, and Urban Society in Pre-Hispanic Oaxaca. Current Anthropology 37(1): 33–47.

Joyce, Arthur A., Andrew G. Workinger, Byron Hamann, et al. 2004 Lord 8 Deer “Jaguar Claw” and the Land of the Sky: The Archaeology and History of Tututepec. Latin American Antiquity 15(3): 273–297.

Joyce, Rosemary A. 2004 Unintended Consequences? Monumentality as a Novel Experience in Formative Mesoamerica. Journal of Archaeological Method and Theory 11(1): 5–29.

Joyce, Rosemary A., and Julia A. Hendon 2000 Heterarchy, History, and Material Reality: Communities in Late Classic Honduras. In The Archaeology of Communities: A New World Perspective. Marcello A. Canuto and Jason Yaeger, eds. Pp. 143–160. London: Routledge.

Kamermans, Hans 2007 Smashing the Crystal Ball. A Critical Evaluation of the Dutch National Archaeological Predictive Model (IKAW). International Journal of Humanities and Arts Computing 1(1): 71–84.

Kan, Sergei 2015 Symbolic Immortality: The Tlingit Potlatch of the Nineteenth Century, Second Edition. University of Washington Press.

Kantner, John 2008 The Archaeology of Regions: From Discrete Analytical Toolkit to Ubiquitous Spatial Perspective. Journal of Archaeological Research 16(1): 37–81.

793

Kantner, John, and Ronald Hobgood 2003 Digital Technologies and Prehistoric Landscapes in the American Southwest. In The Reconstruction of Archaeological Landscapes through Digital Technologies. Muarizio Forte, Patrick Ryan Williams, and J. Wiseman, eds. Pp. 117–123. Oxford: Archaeopress.

Keen, Anne Elise, and David Zeitlyn 2007 Language, Diet, and Ethnicity in Mayo-Darlé, Adamaoua, Cameroon. Anthropos 102(1): 213–219.

Kelly, William W. 2007 Is Baseball a Global Sport? America’s ‘National Pastime’ as Global Field and International Sport. Global Networks 7(2): 187–201. 2009 Samurai Baseball: The Vicissitudes of a National Sporting Style. The International Journal of the History of Sport 26(3): 429–441.

Kent, Susan 1984 Domestic Architecture and the Use o f Space: An Interdisiplinary Cross-Cultural Study. Cambridge: Cambridge University Press.

Kertzer, David I. 1988 Ritual, Politics, and Power. New Haven, CT: Yale University Press.

Kessler, Mario 2011 Only Nazi Games? Berlin 1936: The Olympic Games between Sports and Politics. Socialism and Democracy 25(2): 125–143.

King, Stacie M. 2010 Informe Final: Proyecto Arqueologica Nejapa y Tavela 2009. Informe. Bloomington: University of Indiana-Bloomington. 2012 Hidden Transcripts, Contested Landscapes, and Long-Term Indigenous History in Oaxaca, Mexico. In Decolonizing Indigenous Histories: Exploring Prehistoric/Colonial Transitions in Archaeology. Maxine Oland, Siobhan M. Hart, and Liam Frink, eds. Pp. 230–262. Tucson, AZ.: University of Arizona Press.

King, Stacie M., Elizabeth R. Konwest, Alex Elvis Badillo, et al. n.d. Vol II Informe Final: Proyecto Arqueológico Nejapa/Tavela, Temporada III, 2013 Vol II: Excavaciones Arqueologicas. Final informe. Bloomington: University of Indiana- Bloomington. 2012 Informe Final: Proyecto Arqueológico Nejapa/Tavela, Temporada II, 2011. Informe. Bloomington: University of Indiana-Bloomington.

King, Stacie M., Elizabeth R. Konwest, Alex Elvis Badillo, and Juan Jarquín Enríquez 2014 Más Que Conquista: Un Cuento de Dos Fortalezas En La Región de Nejapa. Cuadernos Del Sur 19(36): 21–41.

King, Stacie M., Elizabeth R. Konwest, Alex Elvis Badillo, and Juan Juaqin

794

n.d. Vol. I Informe Final: Proyecto Arqueológico Nejapa/Tavela, Temporada III, 2013 Vol I: Recorrido Sistematico. Informe. Bloomington: University of Indiana-Bloomington.

King, Stacie M., and Danny A. Zborover 2015 Beyond Ethnonyms: Interdisciplinary Research on Mountain Identity in the Sierra Sur of Oaxaca, Mexico. In Engineering Mountain Landscapes: An Anthropology of Social Investment. Laura L. Schieber and María Nieves Zedeño, eds. Pp. 131–146. Salt Lake City: University of Utah Press.

Kirchoff, Paul 1966 Mesoamerica: Its Geographic Limits, Ethnic Composition and Cultural Characteristics. In Ancient Mesoamerica: Selected Readings. J.A. Gahram, ed. Norman A. McQuown, tran. Pp. 1–14. Palo Alto, CA: Peek Publications.

Knapp, A. B. 1992 Archaeology and Annales: Time, Space, and Change. In Archaeology, Annales, and Ethnohistory. A. B. Knapp, ed. Pp. 1–21. Cambridge, England: Cambridge University Press. 1998 Social Approaches to the Archaeology and Anthropology of Mining. In Social Approaches to an Industrial Past. A. B. Knapp, Vincent Piggot, and E. Herbert, eds. Pp. 1– 23. Los Angeles: Routledge.

Knapp, A. B., and Wendy Ashmore 1999 Archaeological Landscapes: Constructed, Conceptualized, Ideation. In Archaeologies of Landscape: Contemporary Perspectives. Wendy Ashmore and A. B. Knapp, eds. Pp. 1–31. Oxford: Blackwell Publishing.

Knappett, Carl 2011 An Archaeology of Interaction : Network Perspectives on Material Culture and Society. New York: Oxford University Press.

Kolb, Michael J., and James E. Snead 1997 It’s a Small World after All: Comparative Analyses of Community Organization in Archaeology. American Antiquity 62(4): 609–628.

Konwest, Elizabeth R. 2017 Material Markers of Community Identity in Postclassic Nejapa, Oaxaca, Mexico. Doctoral, Indiana University-Bloomington.

Koontz, Rex 2008 Ballcourt Rites, Paradise, and the Origins of Power in Classic Veracruz. In Pre- Columbian Landscapes of Creation and Origin Pp. 11–29. Springer.

Kosiba, Steve 2011 The Politics of Locality: Pre-Inka Social Landscapes of the Cuzco Region. In The Archaeology Politics: The Materiality of Political Practice and Action in the Past. Peter G. Johansen and Alexander A. Bauer, eds. Pp. 114–150. Newcastle upon Tyne: Cambridge

795

Scholars Publishing.

Kosiba, Steve, and Andrew M. Bauer 2013 Mapping the Political Landscape: Toward a GIS Analysis of Environmental and Social Difference. Journal of Archaeological Method and Theory 20(1): 61–101.

Kowalewski, Stephen A., Andrew K. Balkansky, Laura R. Stiver Walsh, and Verónica Pérez Rodríguez 2009 Origins of the Ñuu: Archaeology in the Mixteca Alta. Boulder, CO.: University Press of Colorado.

Kowalewski, Stephen A., Gary M. Feinman, Laura Finsten, and Richard E. Blanton 1991 Pre-Hispanic Ballcourts from the Valley of Oaxaca, Mexico. In The Mesoamerican Ballgame. Vernon L. Scarborough and David R. Wilcox, eds. Pp. 25–44. Tucson, AZ.: University of Arizona Press.

Kowalski, Jeff K. 1999 Natural Order, Social Order, Political Legitimacy, and the Sacred City: The Architecture of Teotihuacan. In Mesoamerican Architecture as a Cultural Symbol. Jeff K. Kowalski, ed. Pp. 76–109. Oxford: Oxford University Press.

Krist, Frank J., and Daniel G. Brown 1994 GIS Modeling of Paleo-Indian Period Caribou Migrations and Viewsheds in Northeastern Lower Michigan. Photogrammetric Engineering and Remote Sensing 60(9): 1129–1138.

Kroefges, Peter 2001 Archaeological Survey in the Coastal Chontalpa de Oaxaca, México. Report. FAMSI. http://www.famsi.org/reports/00045/index.html. 2004 Sociopolitical Organization in the Prehispanic Chontalpa de Oaxaca, México: Ethnohistorical and Archaeological Perspectives. Doctoral, University at Albany-SUNY.

Kurjack, Edward B., Ruben Maldanado C., and Merle Green Robertson 1991 Ballourts of the Northern Maya Lowlands. In The Mesoamerican Ballgame. Vernon L. Scarborough and David R. Wilcox, eds. Pp. 145–160. Tucson, AZ.: University of Arizona Press.

Kurosaki Maekawa, Mitsura 2006 Estudio Sobre Los Yugos. Licenciatura thesis, ENAH-INAH.

Kvamme, K.L. 1989 Geographic Information Systems in Reginal Archaeological. In Archaeological Theory and Method. Michael B. Schiffer, ed. Pp. 139–202. Tucson, AZ.: University of Arizona Press. 1990 The Fundamental Principles and Practice of Predictive Modeling. In Studies in Modern Archaeology, Vol. 3. A. Vorrips, ed. Pp. 257–295. Bonn: Holos-Verlag. 1995 A View across from the Water: The North American Experience in Archaeological

796

GIS. In Archaeology and Geographical Information Systems. G. Lock and Z. Stancic, eds. Pp. 1–14. London: Taylor & Francis. 2006 There and Back Again: Revisiting Archaeological Locational Modeling. In GIS and Archaeological Site Location Modeling. Mark W. Mehrer and Konnie L. Wescott, eds. Pp. 3–38. Boca Raton: Taylor & Francis.

Kwan, Mei-Po 2002 Feminist Visualization: Re-Envisioning GIS as a Method in Feminist Geographic Research. Annals of the Association of American Geographers 92(4): 645–661.

Ladefoged, Thegn N., and Michael W. Graves 2000 Evolutionary Theory and the Historical Development of Dry-Land Agriculture in North Kohala, Hawai’i. American Antiquity 65(3): 423–448.

Ladefoged, Thegn N., Charlotte T. Lee, and Michael W. Graves 2008 Modeling Life Expectancy and Surplus Production of Dynamic Pre-Contact Territories in Leeward Kohala, Hawai‘I. Journal of Anthropological Archaeology 27(1): 93–110.

Lagerås, K.E. 2002 Visibile Intentions? Viewshed Analysis of Bronze Age Burial in Western Scania, Sweden. In Monuments and Landscape in Atlantic Europe: Perception and Society During the Neolithic and Early Bronze Age. C. Scarre, ed. Pp. 177–191. London: Routledge.

Lake, M.W., and P.E. Woodman 2000 Viewshed Analysis of Site Location on Islay. In Hunter-Gatherer Landscape Archaeology: The Southern Hebrides Mesolithic Project, 1988–98, Vol. 2, Archaeological Fieldwork on Colonsay, Computer Modelling, Experimental Archaeology, and Final Interpretations. Steven J. Mither, ed. Pp. 497–503. Cambridge: The McDonald Institute for Archaeological Research. 2003 Visibility Studies in Archaeology: A Review and Case Study. Environmental and Planning B. Planning and Design 30: 689–707.

Lake, M.W., P.E. Woodman, and S.J. Mithen 1998 Tailoring GIS Software for Archaeological Applications: An Example Concerning Viewshed Analysis. Journal of Archaeological Science 25(1): 27–38.

Lamar, Howard Roberts, and Leonard Monteath Thompson 1981 The Frontier in History: North America and Southern Africa Compared. New Haven: Yale University Press.

Lau, George F. 2002 Feasting and Ancestor Veneration at Chinchawas, North Highlands of Ancash, Peru. Latin American Antiquity 13(3): 279–304.

Lave, Jean, and Etienne Wenger 1991 Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge

797

University Press.

Laviada, Inigo 1978 Los Caciques de La Sierra. Mexico: Editorial Jus.

Lefebvre, Henri 2001 The Production of Space. London: Routledge.

Lekson, Stephen H. 1999 The Chaco Meridian: Centers of Political Power in the Ancient Southwest. Rowman Altamira.

Leonard, Kevin 2017 Why Waltes Was a Woman’s Game. In Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 19–33. Salt Lake City: University of Utah Press.

Levine, Marc N. 2013 Examinging Ceramic Evidence for the Zapotec Imperialism Hypothesis in the Lower Rio Verde Region of Oaxaca, Mexico. In Polity and Ecology in Formative Period Coastal Qaxaca. Arthur A. Joyce, ed. Pp. 227–264. Boulder, CO.: University Press of Colorado.

Levine, Marc N., Arthur A. Joyce, and P. Goldberg 2004 Earthen Mound Construction at Rio Viejo on the Pacific Coast of Oaxaca, Mexico. Paper presented at the 69th Annual Meeting of the Society for American Archaeology, Montreal, Quebec, Canada.

Lewis, Laura A. 2003 Hall of Mirrors: Power, Witchcraft, and Caste in Colonial Mexico. Duke University Press.

Leyenaar, Ted J. J. 1978 Ulama: Perpetuación En México Del Juego de Pelota Prehispánico: Ullamaliztli. Culiacán, Mexico: Gobierno del Estado de Sinaloa, Dirección y Fomeno de Cultura Regional.

Lightfoot, Kent G., and Antoinette Martinez 1995 Frontiers and Boundaries in Archaeological Perspective. Annual Review of Anthropology 24: 471–492.

Lightfoot, Kent G., Antoinette Martinez, and Ann M. Schiff 1998 Daily Practice and Material Culture in Pluralistic Social Settings: An Archaeological Study of Culture Change and Persistence from Fort Ross, California. American Antiquity 63(2): 199–222.

Limp, W. Fredrick

798

1991 Continuous Cost Movement Models. In Proceedings of the Applications of Space Age Technology in Anthropology, ASA, John C. Clifford A. Behrens and Thomas L. Sever, eds. Mississippi: Stennis Space Center. 2000 Geographic Information Systems in Historic Preservation. In Science and Technology in Historic Preservation Pp. 231–247. Advances in Archaeological and Museum Science. Boston, MA: Springer. http://link.springer.com/chapter/10.1007/978-1- 4615-4145-5_10, accessed August 17, 2018.

Lindsay, I. 2011 Holding Down the Fort: Landscape Production and the Sociopolitical Dynamics of Late Bronze Age Fortress Regimes in the Southern Caucuses. In The Archaeology of Politics: The Materiality of Political Practice and Action in the Past. Peter G. Johansen and Andrew M. Bauer, eds. Pp. 151–15. Newcastle upon Tyne: Cambridge Scholars Publishing.

Lipsky, Richard 1981 How We Play the Game: Why Sports Dominate Amerian Life. Boston: Beacon Press.

Llobera, M. 2003 Extending GIS-Based Visual Analysis: The Concept of Visualscapes. International Journal of Geographical Information Science 17(1): 25–48.

Llobera, M., P. Fábrega-Álvarez, and C. Parcero-Oubiña 2011 Order in Movement: A GIS Approach to Accessibility. Journal of Archaeological Science 38(4): 843–851.

Llobera, Marcos 1996 Exploring The Topography of Mind: GIS, Social Space and Archaeology. Antiquity-Oxford 70: 612–622. 2001 Building Past Landscape Perception With GIS: Understanding Topographic Prominence. Journal of Archaeological Science 28(9): 1005–1014. 2007 Reconstructing Visual Landscapes. World Archaeology 39(1): 51–69. 2011 Archaeological Visualization: Towards an Archaeological Information Science (AISc). Journal of Archaeological Method and Theory 18(3): 193–223.

Lock, Gary R. 2000 Beyond the Map: Archaeology and Spatial Technologies. IOS Press. 2003 Using Computers in Archaeology Towards Virtual Pasts. London ; New York: Routledge.

Lock, Gary R., and Trevor Harris 2006 Enhancing Predictive Archaeological Modeling: Integrating Location, Landscape, and Culture. In GIS Archaeolgoical Site Location Modeling. Mark W. Mehrer and Konnie L. Wescott, eds. Pp. 41–62. Boca Raton: Taylor & Francis.

Lohse, Jon C., Kerry L. Sagebiel, and Joanne Baron 2013 The Ballgame, Community Ceremony, and Political Development in Northwestern

799

Belize. In The Ballgame, Community Ceremony, and Political Development in Northwestern Belize. In Classic Maya Political Ecology: Resource Management, Class Histories, and Political Change in Northwestern Belize. Jon C. Lohse, ed. Pp. 99–126. Los Angeles, CA: Cotsen Institute of Archaeology Press.

López Alavez, José 1915 Canción Mixteca.

Lopez Austin, Alfredo 1967 Cuarenta Clases de Magos En El Mundo Nahuatl. Estudios de Cultura Náhuatl 7: 87–117.

Love, Michael 1996 Investigaciones Arqueológicas En El Ujuxte,MRetalhuleu, 1995-1996. Informe Preliminar. Report. Guatemala: Instituto de Antropología e Historia. 2011 Cities, States, and City-State Culture in the Late Preclassic Southern Maya Region. In The Southern Maya: The Rise and Fall of an Early Mesomaerican Civilization in the Late Preclassic. Michael Love and Jonathan Kaplan, eds. Pp. 47–75. Boulder, CO: University Press of Colorado.

MacAloon, John J. 1984 Olympic Games and the Theory of Spectacle in Modern Societies. In Rite, Drama, Festival, Spectacle: Rehearsals Toward a Theory of Cultural Performance. John J. MacAloon, ed. Pp. 241–280. Philadelphia: Institute for the Study of Human Issues.

Macazaga Ordoño, César 1982 El Juego de Pelota. Mexico: Editorial Innovación.

MacEachern, Scott, David J. W. Archer, and Richard D. Garvin, eds. 1989 Households and Communities: Proceedings of the Twenty-First Annual Conference of the Archaeological Association of the University of Calgary. Calgary, Alberta: Archaeological Association of the University of Calgary.

Machuca Gallegos, L. 2007 Comercio de Sal y Redes de Poder En Tehuantepec En La Época Colonial. Mexico City: Centro de Investigaciones y Estudios Superiores en Antropología Social.

Mack, Alexandra 2004 One Landscape, Many Experiences: Differing Perspectives of the Temple Districts of Vijayanagara. Journal of Archaeological Method and Theory 11(1): 59–81.

Madry, Scott L.H., and Lynn Rakos 1996 Line-of-Sight and Cost Surface Techniques for Regional Research in the Arroux River Valley. In New Methods, Old Problems: Geographic Information Systems in Modern Archaeological Research. Herbert D. G. Maschner, ed. Pp. 104–126. Occasional Paper, 123. Carbondale: Center for Archaeological Investigations, Southern Illinois University Carbondale.

800

Marcus, Joyce, and Kent V. Flannery 1994 Ancient Zapotec Ritual and Religion: An Application of the Direct Historical Approach. In The Ancient Mind: Elements of Cognitive Archaeology. Colin Renfrew and Ezra B. W. Zubriw, eds. Pp. 55–74. Cambridge: Cambridge University Press. 2004 The Coevolution of Ritual and Society: New 14C Dates from Ancient Mexico. Proceedings of the National Academy of Sciences 101(52): 18257–18261.

Markens, Robert, and Marcus Winter 2014 La Tumba 1D de Chuxnabán, Quetzaltepec, Mixes. In Panorama Arqueológico: Dos Oaxacas. Marcus Winter and Gonzalo Sánchez Santiago, eds. Pp. 279–291. Arqueología Oaxaqueña, 4. Oaxaca, Mexico: INAH and Consejo Naciónal para la Cultura y las Artes.

Marozas, Bryan A., and James A. Zack 1990 GIS and Archaeological Site Location. In Interpreting Space: GIS and Archaeology. Kathleen M.S> Allen, Stanton W. Green, and Ezra B. W. Zubrow, eds. Pp. 165–172. London: Taylor & Francis.

Marshall, Yvonne 2002 What Is Community Archaeology? World Archaeology 34(2): 211–219.

Martin, Joann 1990 Motherhood and Power: The Production of a Women’s Culture of Politics in a Mexican Community. American Ethnologist 17(3): 470–490.

Martínez Gracida, Manuel 1910a Los Indios Oaxaqueños y Sus Monumentos Arqueológicos. Unpublished report in five volumes. Civilización Mixteco-Zapoteca. Oaxaca, Mexico: Biblioteca Pública Central del Estado de Oaxaca. 1910b Civilización Chontal, Historia Antigua de La Chontalpa Oaxaqueña. Mexico: Imprenta del Gobierno Federal. 2010 Los Indios Oaxaqueños y Sus Monumentos Arqueológicos, Tomo IV, Arquitectura. Unpublished report in five volumes. Civilización Mixteco-Zapoteca. Oaxaca, Mexico: Biblioteca Pública Central del Estado de Oaxaca.

Maschner, Herbert D. G. 1996 The Politics of Settlement Choice on the Northwest Coast: Cognition, GIS, and Coastal Landscapes. In Anthropology, Space, and Geographic Information Systems. Mark S. Aldenderfer and Herbert D. G. Maschner, eds. Pp. 175–189. Oxford: Oxford University Press.

Maschner, Herbert D. G., and Jeffrey W. Stein 1995 Multivariate Approaches to Site Location on the Northwest Coast of North America. Antiquity 69(262): 61–73.

Masson, Marilyn A., and Heather Orr 1999 The Writing on the Wall: Political Representation and Sacred Geography at Monte

801

Albán. In The Sowing and the Dawing: Termination, Dedication, and Transformation in the Archaeological and Ethnographic Record of Mesoamerica. Shirley B. Mock, ed. Pp. 165– 175. Albuquerque: University of New Mexico Press.

Matos Moctezume, Eduardo 1988 The Great Temple of the Aztecs : Treasures of Tenochtitlan. New York: Thames & Hudson.

Matson, R. G., William D. Lipe, and William R. Haase 1988 Adaptational Continuities and Occupational Discontinuities: The Cedar Mesa Anasazi. Journal of Field Archaeology 15(3): 245–263.

Matthew, Laura E. 2007 Whose Conquest? Nahua, Zapotec, and Mixteca Allies in the Conquest of Central America. In Indian Conquistadors: Indigenous Allies in the Conquest of Mesoamerica. Laura E. Matthew and Michel R. Oudijk, eds. Pp. 102–126. Norman: University of Oklahoma Press.

Mauss, Marcel 1990 The Gift: The Form and Reason for Exchange in Archaic Societies. Routledge Classics. London: Routledge.

McClancy, Jeremy 1996 Sport, Identity and Ethnicity. Oxford: Berg Press.

McCoy, Mark D. 2005 The Development of the Kalauapapa Field System, Moloka’i Island, Hawai’i. The Journal of the Polynesian Society 114(4): 339–358. 2006 Landscape, Social Memory, and Society: An Ethnohistoric Archaeology Study of Three Hawaiian Communities. Doctoral, University of California.

McCoy, Mark D., and Thegn N. Ladefoged 2009 New Developments in the Use of Spatial Technology in Archaeology. Journal of Archaeological Research 17(3): 263–295.

McEwan, Dorothy Graves 2012 Qualitative Landscape Theories and Archaeological Predictive Modelling—A Journey Through No Man’s Land? Journal of Archaeological Method and Theory 19(4): 526–547.

McGarry, Karen 2010 Sport in Transition: Emerging Trends on Culture Change in the Anthropology of Sport. Reviews in Anthropology 39(3): 151–172.

McGuire, Randall H. 1982 The Study of Ethnicity in Historical Archaeology. Journal of Anthropological Archaeology 1: 159–178.

802

McGuire, Randall H., and Michael B. Schiffer 1983 A Theory of Architectural Design. Journal of Anthropological Archaeology 2(3): 277–303.

Medina Castillo, Edgar 2005 El Juego de Pelota Del Preclásico Medio En El Noroeste de Yucatán, México. Licensiatura thesis, Universidad Autónoma de Yucatán.

Mehrer, Mark W. 2000 Heterarchy and Hierarchy: The Community Plan as Institution in Cahokia’s Polity. In The Archaeology of Communities: A New World Perspective. Marcello A. Canuto and Jason Yaeger, eds. Pp. 44–57. London: Routledge.

Mehrer, Mark W., and Konnie L. Wescott, eds. 2006 GIS and Archaeological Site Location Modeling. Boca Raton: Taylor & Francis.

Menard, Scott 2001 Applied Logistic Regression Analysis. Thousand Oaks, CA.: Sage Publications.

Meskell, Lynn, and Robert Preucel 2004 Identities. In A Companion to Social Archaeology. Lynn Meskell and Robert Preucel, eds. Pp. 121–141. Malden, MA: Blackwell Publishing.

Messenger, Lewis C. 1987 Community Organization of the Late Classic Southern Periphery of Mesoamerica: “Expressions of Affinity?” In Interaction on the Southeast Mesoamerican Frontier: Prehistoric and Historic Honduras and El Salvador. Eugenia Robinson, ed. BAR International Series, 327. Oxford: British Archaeological Research.

Miles, Beryl 1961 Spirit of Mexico. London: Murray.

Miller, David Harry, and Jerome O. Steffan 1977 The Frontier: Comparative Studies I. 1st ed.. Norman: University of Oklahoma Press.

Miller, Mary Ann 1989 The Ballgame. Record of Art Museum, Princeton University 48(2): 22–31. 2001 The Maya Ballgame: Rebirth in the Court of Life and Death. In The Sport of Life and Death: The Mesoamerican Ballgame. E. Michael Whittington, ed. Pp. 78–87. Charlotte, NC: Thames & Hudson.

Miller, Mary Ellen, and Stephen D. Houston 1987 The Classic Maya Ballgame and Its Architectural Setting: A Study of Relations between Text and Image. RES: Anthropology and Aesthetics 14(Autumn): 46–65.

803

Mills, Barbara J. 2004 The Establishment and Defeat of Hierarchy: Inalienable Possessions and the History of Collective Prestige Structures in the Pueblo Southwest. American Anthropologist 106(2). New Series: 238–251. 2007 Performing the Feast: Visual Display and Suprahousehold Commensalism in the Puebloan Southwest. American Antiquity 72(2): 210–239. 2008 Colonialism and Cuisine: Cultural Transmission, Agency, and History at Zuni Pueblo. In Cultural Transmission and Material Culture: Breaking Down Barriers. Miriam T. Stark, Brenda Bowser, and Lee Horne, eds. Pp. 245–262. Tucson, AZ.: University of Arizona Press.

Minetti, A. E. 1995 Optimum Gradient of Mountain Paths. Journal of Applied Physiology 79(5): 1698– 1703.

Molina, Fray Alonso de 1977 Vocabulario En Lengua Castellana y Mexicana y Mexicana y Castellana. Porrúa, México.

Monaghan, John 1990 Reciprocity, Redistribution, and the Transaction of Value in the Mesoamerican Fiesta. American Ethnologist 17(4): 758–774. 1995 The Covenants with Earth and Rain: Exchange, Sacrifice, and Revelation in Mixtec Sociality. Norman: University of Oklahoma Press.

Montmollin, Olivier de 1988 Scales of Settlement Study for Complex Societies: Analytical Issues from the Classic Maya Area. Journal of Field Archaeology 15(2): 151–168. de Montmollin, Olivier 1989 The Archaeology of Political Structure: Settlement Analysis in a Classic Maya Polity. New Studies in Archaeology. Cambridge ; New York: Cambridge University Press. 1995 Settlement and Politics in Three Classic Maya Polities. Monographs in World Archaeology, no. 24. Madison, Wis: Prehistory Press. 1997 A Regional Study of Classic Maya Ballcourts From the Upper Grijalva Basin, Chiapas, Mexico. Ancient Mesoamerica 8(01): 23–41.

Mooney, James 1890 The Cherokee Ball Play. American Anthropologist 3(2): 105–132.

Moore, Henrietta L. 1986 Space, Text, and Gender: An Anthropological Study of the Marakwet of Kenya. Cambridge [Cambridgeshire] ; New York, NY, USA: Cambridge University Press.

Moore, Jerry D. 1996 Architecture and Power in the Ancient Andes: The Archaeology of Public Buildings. New Studies in Archaeology; Variation: New Studies in Archaeology.

804

Cambridge ; New York, NY, USA: Cambridge University Press. 2004 The Social Basis of Sacred Spaces in the Prehispanic Andes: Ritual Landscapes of the Dead in Chimú and Inka Societies. Journal of Archaeological Method and Theory 11(1): 83–124.

Motolinia, Fray Toribio de Benavente 1903 Historía de Los Indios de La Nueva España. Mexico City: Casa del Editor.

Munn, Henry 2014 La Arqueologıa de La Sierra Mazateca. In Panorama Arqueológico: Dos Oaxacas. Marcus Winter and Gonzalo Sánchez Santiago, eds. Pp. 63–99. Oaxaca, Mexico: CENTRO INAH Oaxaca.

Norbeck, Edward 1971 Men at Play. Natural History 10: 48–53. 1974 Anthropological Views of Play. American Zoologist 14(1): 267–273. 1977 The Study of Play, Johan Huizinga and Modern Anthropology. In The Anthropological Study of Play: Problems and Prospects. David F. Lancy and Allan B. Tindall, eds. Pp. 13–22. Cornwall, NJ: Leisure Press.

Norbeck, Edward, and Claire R. Farrer, eds. 1979 Forms of Play of Native North Americans. Proceedings - American Ethnological Society ; 1977. St. Paul: West Pub. Co.

Ogburn, Dennis E. 2006 Assessing the Level of Visibility of Cultural Objects in Past Landscapes. Journal of Archaeological Science 33(3): 405–413.

Orr, Heather 1997 Power Games in the Late Formative Valley of Oaxaca: The Ballplayer Sculptures at Dainzú. PhD, University of Texas.

Ortiz, Laura Velasco 2005 Mixtec Transnational Identity. Tucson, AZ.: University of Arizona Press.

Oudijk, Michael R. 2008 The Postclassic Period in the Valley of Oaxaca: The Archaeological and Ethnohistorical Records. In After Monte Albán: Transformation and Negotiation in Oaxaca, Mexico. Jeffrey P. Blomster, ed. Pp. 95–118. Boulder, CO: University of Colorado Press.

Oudijk, Michel R. 2000 Historiography of the Bènizàa: The Postclassic and Early Colonial Periods (1000- 1600 A.D.). Research School of Asian, African, and Amerindian Studies (CNWS). Leiden, Netherlands: Universiteit Leiden.

Oudijk, Michel R., and Matthew Restall 2007 Mesoamerican Conquistadors in the Sixteenth Century. In Indian Conquistadors:

805

Indigenous Allies in the Conquest of Mesoamerica. Laura R. Matthew and Michel R. Oudijk, eds. Pp. 28–63. Norman: University of Oklahoma Press.

Oxendine, Joseph B. 1988 American Indian Sports Heritage. Champaign, IL: Human Kinetics.

Parker, Bradley J., and Lars Rodseth, eds. 2005 Untaming the Frontier in Anthropology, Archaeology, and History. Tucson: University of Arizona Press.

Parsons, Jeffrey R. 1972 Archaeological Settlement Patterns. Annual Review of Anthropology 1: 127–150. 2004 Critical Reflections on Forty Years of Systematic Regional Survey6. Paper presented at the 69th Annual Meeting of the Society for American Archaeology, Montreal, Quebec, Canada.

Paso y Troncoso, Francisco del, ed. 1905 Papeles de Nueva España, Publicados de Orden y Con Fondos Del Gobierno Mexicano. Segunda serie. Madrid: Establecimiento tip. “Sucesores de Rivadeneyra,.” 1939 Epistolario de Nueva España 1505-1818, Tomo III, 1533-1539. Biblioteca Historia Mexicana de Obras Ineditas, segunda serie 3. Mexico: Antigua Libreria Robredo de José Porrua e Hijos.

Pasztory, Esther 1972 The Historical and Religious Significance of the Middle Classic Ball Game. In Religion En Mesoamerica: XII Mesa Redonda de La Sociedad Mexicana de Antropologia Pp. 441–455. Mexico. 1976 The Murals of Tepantitla, Teotihuacan. New York: Garland Publishing. 1978 Artistic Traditions of the Middle Classic Period. In Middle Classic Mesoamerica: A.D. 400-700 Pp. 108–142. New York: Columbia University Press.

Pauketat, Timothy R. 1997 Specialization, Political Symbols and the Crafty Elite of Cahokia. Southeastern Archaeology 16(1): 1–15. 2001 Practice and History in Archaeology: An Emerging Paradigm. Anthropological Theory 1: 73–98.

Pauketat, Timothy R., Lucretia S. Kelly, Gayle J. Fritz, et al. 2002 The Residues of Feasting and Public Ritual at Early Cahokia. American Antiquity 67(02): 257–279.

Peacock, James 1985 An Anthropologist Goes to the Olympics. Social Science Newsletter 70(2): 77–81.

Pearson, Michael Parker, and Colin Richards 2003 Architecture and Order: Approaches to Social Space. London: Routledge.

806

Peniche May, Nancy 2012 The Architecture of Power and Sociopolitical Complexity in Northwestern Yucatan during the Preclassic Period. In The Maya of Ancient Mexico: Reinterpreting the Northern Maya Lowlands. G.E. Braswell, ed. Pp. 65–87. London: Equinox Press.

Perkins, Philip 2000 A GIS Investigation of Site Locations and Landscape Relationships in the Albegna Valley, Tuscany. In Computer Application and Quantitative Methods in Archaeology. K. Lockyear, J. T. Sly, and Mihailescu-Birliba, eds. Pp. 133–140. Oxford: Archaeopress.

Peterson, Christian E., and Robert D. Drennan 2005 Communities, Settlements, Sites, and Surveys: Regional-Scale Analysis of Prehistoric Human Interaction. American Antiquity 70(1): 5–30.

Pingel, Thomas J. 2010 Modeling Slope as a Contributor to Route Selection in Mountainous Areas. Cartography and Geographic Information Science 37(2): 137–148.

Pohl, John M.D. 2003a Ritual Ideology and Commerce in the Southern Mexican Highlands. In The Postclassic Mesoamerican World. Michael E. Smith and Francis F. Berdan, eds. Pp. 172– 180. Salt Lake City: University of Utah Press. 2003b Creation Stories, Hero Cults, and Alliance Building: Confederacies of Central and Southern Mexico. In The Postclassic Mesoamerican World. Michael E. Smith and Francis F. Berdan, eds. Pp. 61–66. Salt Lake City: University of Utah Press. 2003c Royal Marriage and Confederacy Building among the Eastern Nahuas, Mixtecs, and Zapotecs. In The Postclassic Mesoamerican World. Michael E. Smith and Francis F. Berdan, eds. Pp. 243–248. Salt Lake City: University of Utah Press. 2004 The Archaeology of History in Postclassic Oaxaca. In Mesoamerican Archaeology: Theory & Practice. Julia A. Hendon and Rosemary A. Joyce, eds. Pp. 217–238. Malden, MA: Blackwell Publishing.

Pohl, John M.D., and Bruce E. Byland 1990 Mixtec Landscape Perception and Archaeological Settlement Patterns. Ancient Mesoamerica 1(01): 113–131.

Pohl, John M.D., John Monaghan, and Laura R. Stiver 1997 Religion, Economy, and Factionalism in Mixtec Boundary Zones. In Códices y Documentos Sobre México. Segundo Simposio. Salvador Rueda Smithers, C. Vega Sosa, and R. Martínez Bracas, eds. Pp. 202–235. Mexico City: INAH and Consejo Naciónal para la Cultura y las Artes.

Poliakoff, Michael B. 1987 Combat Sports in the Ancient World: Competition, Violence, and Culture. Yale University Press.

Pollard, Joshua, and Mark Gillings

807

1998 Romancing the Stones1: Towards a Virtual and Elemental Avebury. Archaeological Dialogues 5(2): 143–164.

Pollock, Susan 2003 Feasts, Funerals and Fast Food in Early Mesopotamian States. In The Archaeology and Politics of Food and Feasting in Early States and Empires. Tamara L. Bray, ed. Boston, MA: Springer.

Postill, John 2013 Fields as Dynamic Clusters of Practices, Games, and Socialities. In Sociality: An Anthropologial Interrogation P. 2014. Oxford ; New York: Berghahn.

Potter, James M. 2000 Pots, Parties, and Politics: Communal Feasting in the American Southwest. American Antiquity 65(03): 471–492.

Poyer, Lin 1988 Maintaining “Otherness”: Sapwuahfik Cultural Identity. American Ethnologist 15(3): 472–485.

Pred, Allan 1984 Place as Historically Contingent Process: Structuration and the Time- Geography of Becoming Places. Annals of the Association of American Geographers 74(2): 279–297. 1990 Making Histories and Constructing Human Geographies: The Local Transformation of Practice, Power Relations, and Consciousness. Boulder: Westview Press.

Rademaker, Kurt, David A. Reid, and Gordon R. M. Bromley 2012 Connecting the Dots: Least Cost Analysis, Paleogeography, and the Search for Paleoindian Sites in Southern Highland Peru. In Least Cost Analysis of Social Landscapes: Archaeological Case Studies. Devin A. White and Sarah L. Surface-Evans, eds. Pp. 32–45. Salt Lake City: University of Utah Press.

Ramírez, Rafael Issac 1892 Descripción Del Cerro de La Muralla, Peña Del Convento, Cerro Del Jabalí, y Horno de Los Zapotecos. Biblioteca Central del Estado de Oaxaca, Documentos inéditos de Manuel Martinez Gracida.

Ramos, Maria Isabel 2012 Women Playing a Man’s Game: Reconstructing Ceremonial and Ritual History of the Mesoamerican Ballgame. University of California.

Rapaport, Amos 1969 House Form and Culture. Foundations of Cultural Geography Series. Englewood Cliffs, N.J.: Prentice-Hall. 1982 The Meaning of the Built Environment: A Nonverbal Communication Approach. Tucson, AZ.: University of Arizona Press. 1990 System of Activities and System of Settings. In Domestic Architecture and the Use

808 of Space: An Interdisciplinary Cross-Cultural Study. Susan Kent, ed. Pp. 9–20. Cambridge: Cambridge University Press.

Rathje, William L. 1983 To the Salt of the Earth: Some Comments on Household Archaeology Among the Maya. In Prehistoric Settlement Patterns: Essays in Honor of Gordon R. Willey. Evon Z. Vogt and Richard M. Leventhal, eds. Pp. 23–34. Albuquerque, N.M.: University of New Mexico Press.

Redmond, Elsa M. 1983 A Fuego y Sangre: Early Zapotec Imperialism in the Cuicatlán Cañada, Oaxaca. Studies in Latin American Ethnohistory & Archaeology, v. 1. Ann Arbor: University of Michigan, Museum of Anthropology.

Rees, Peter W. 1975 Origins of Colonial Transportation in Mexico. Geographical Review: 323–334.

Rennell, Rebecca 2008 Exploring “Everyday” Places in the Iron Age Landscapes of the Outer Hebrides. In Changing Perspectives on the First Millennium BC: Proceedings of the Iron Age Research Student Seminar 2006. Oliver Davis, Niall MacPherson Sharples, and Kate Waddington, eds. Pp. 43–60. Oxford: Oxbow Books. 2012 Landscape, Experience and GIS: Exploring the Potential for Methodological Dialogue. Journal of Archaeological Method and Theory 19(4): 510–525.

Rice, Prudence M., and Don S. Rice 2005 The Final Frontier of the Maya: Central Peten, Guatemala, 1450-1700 CE. In Untaming the Frontier in Anthropology, Archaeology and History. Bradley J. Parker and Lars Rodseth, eds. Pp. 147–173. Tucson, AZ.: University of Arizona Press.

Richards, Julian D. 1998 Recent Trends in Computer Applications in Archaeology. Journal of Archaeological Research 6(4): 331–382.

Rickards, Constantine George 1910 The Ruins of Mexico. London: HEShrimpton,. http://hdl.handle.net/2027/njp.32101078172200, accessed August 18, 2018.

Rivera, Maria Zamora 2017 Los Juegos de Pelota En Cantona, Puebla. Arqueología Mexicana 146(July- August): 46–51.

Rizvi, Uzma Z. 2011 Subjectivities and Spatiality in Indus Urban Forms: Mohenjo Daro, the Body and the Domestication of Waste. In The Archaeology of Politics: The Materiality of Political Practice and Action in the Past. Peter G. Johansen and Andrew M. Bauer, eds. Pp. 221– 244. Newcastle upon Tyne: Cambridge Scholars Publishing.

809

Roberts, John M., Malcolm J. Arth, and Robert R. Bush 1959 Games in Culture. American Anthropologist 61(4): 597–605.

Robinson, David W. 2010 LAND USE, LAND IDEOLOGY: AN INTEGRATED GEOGRAPHIC INFORMATION SYSTEMS ANALYSIS OF ROCK ART WITHIN SOUTH-CENTRAL CALIFORNIA. American Antiquity 75(4): 792–818.

Rodseth, Lars, and Bradley J. Parker 2005 Introduction: Theoretical Considerations in the Study of Frontiers. In Untaming the Frontier in Anthropology, Archaeology and History Pp. 3–21. Tucson, AZ.: University of Arizona Press.

Romero Frizzi, Maria de los Angeles 1996 El Sol y La Cruz: Historia de Los Pueblos Indígenas de México. Mexico: CIESAS.

Roose, Ninon 2006 Le Complexe Des Hachas, Jougs et Palmes En Méso-Amérique. Doctoral, Université de Paris 1, Panthéon Sorbonne.

Rothschild, Nan 2006 Colonialism, Material Culture, and Identity in the Rio Grande and Hudson River Valleys. International Journal of Historical Archaeology 10(1): 72–107.

Rothstein, Frances 1988 Global Views And Human Action In A Rural Community In Mexico. Urban Anthropology and Studies of Cultural Systems and World Economic Development 17(4): 365–381.

Rowe, David 2003a Sport, Culture & Media: The Unruly Trinity. McGraw-Hill Education (UK). 2003b Sport and the Repudiation of the Global. International Review for the Sociology of Sport 38(3): 281–294.

Rowe, Sharon 2008 Modern Sports: Liminal Ritual or Liminoid Leisure. In Victor Turner and Contemporary Cultural Performance. Graham St John, ed. Pp. 127–148. New York: Berghahn Books.

Rowlands, Michael 1993 The Role of Memory in the Transmission of Culture. World Archaeology 25(2): 141–151.

Ruggles, Clive L. N., and David J. Medyckyj-Scott 1996 Site Location, Landscape Visibility, and Symbolic Astronomy: A Scottish Case Study. In New Methods, Old Problems: Geographic Information Systems in Modern

810

Archaeological Research, Occasional Paper No. 23. Herbert Maschner, ed. Pp. 127–146. Carbondale: Southern Illinois University Carbondale.

Ruggles, Clive L. N., David J. Medyckyj-Scott, and A. Gruffydd 1993 Multiple Viewshed Analysis Using GIS and Its Archaeological Application: A Case Study in Northern Mell. In Computing the Past: Computer Applications and Quantitative Methods in Archaeology. Jens Anderson, Torsten Madsen, and Irwin Scollar, eds. Pp. 125– 131. Aarhus, Denmark: Aarhus University Press.

Ryavec, Karl 2002 Land Use/Cover Change in Central Tibet, c. 1830–1990: Devising a GIS Methodology to Study a Historical Tibetan Land Decree. The Geographical Journal 167(4): 342–357.

Sackett, James A. 1977 The Meaning of Style in Archaeology: A General Model. American Antiquity 42(3): 369–380.

Sage, George H. 1979 Sport and the Social Sciences. The ANNALS of the American Academy of Political and Social Science 445(1): 1–14. de Sahagún, Bernadino 1970 Historia General de Las Cosas de Nueva España: English and Nauhuatl. O. Anderson and Charles E. Dibble, trans. Salt Lake City: SAR Press.

Saitta, Dean J. 1994 Class and Community in the Prehistoric Southwest. In The Ancient Southwestern Community. W. H. Wills and Robert D. Leonard, eds. Pp. 25–43. Albuquerque: University of New Mexico Press.

Santley, Robert S., Michael J. Berman, and Rani T. Alexander 1991 The Politicization of the Mesoamerican Ballgame and Its Implications for the Interpretation of the Distribution of Ballcourts in Central Mexico. In The Mesoamerican Ballgame. Vernon L. Scarborough and David R. Wilcox, eds. Pp. 3–24. Tucson, AZ.: University of Arizona Press.

Sauer, Carl O. 1925 The Morphology of Landscape. In University of California Publications in Geography Pp. 19–53.

Scarborough, Vernon L., and David R. Wilcox, eds. 1991 The Mesoamerican Ballgame. Tucson, AZ.: University of Arizona Press.

Schele, Linda, and David M. Freidel 1991 The Courts of Creation: Ballcourts, Ballgames, and Portals to the Maya Otherworld. In The Mesoamerican Ballgame. Vernon L. Scarborough and David R. Wilcox, eds. Pp.

811

289–316. Tucson, AZ.: University of Arizona Press.

Schieber de Lavarreda, Christa 1994 A Middle Preclassic Clay Ballcourt at Abaj Takalik, Guatemala. Mexicon 16(4): 77–84.

Schryer, Frans Jozef 1987 Class Conflict and the Corporate Peasant Community: Disputes over Land in Nahuatl Villages. Journal of Anthropological Research 43(2): 99–120.

Schuyler, Robert, ed. 1980 Archaeological Perspectives in Ethnicity in America: Afro-American and Asian American Culture History. Farmington, NY.: Baywood.

Schwartz, Glenn M., and Steven E. Falconer 1994 Archaeological Views from the Countryside: Village Communities in Early Complex Societies. Washington: Smithsonian Institution Press.

Sebastian, Lynee, and W. James Judge 1988 Predicting the Past: Correlation, Explanation and the Use of Archaeological Models. In Quantifying the Present and Predicting the Past: Theory, Method and Application of Archaeological Predictive Modeling. W. James Judge and Lynee Sebastian, eds. Pp. 1–18. Denver: US Department of the Interior, Bureau of Land Management.

Seymour, Deni J. 2017 Social Aspects of an Apachean Stave-Dice Gaming Feature at Three Sisters. In Prehistoric Games of North American Indians: Subartic to Mesoamerica Pp. 163–177. Salt Lake City: University of Utah Press.

Shackel, Paul A. 2001 Myth, Memory, and the Making of the American Landscape. Gainesville, Gainesville, FL: University Press of Florida.

Shennan, Stephen 2002 Genes, Memes, and Human History: Darwinian Archaeology and Cultural Evolution. London: Thames & Hudson. https://repository.library.georgetown.edu/handle/10822/547237, accessed August 18, 2018.

Sherman, R. Jason, Andrew K. Balkansky, Charles S. Spencer, and Brian D. Nicholls 2010 Expansionary Dynamics of the Nascent Monte Albán State. Journal of Anthropological Archaeology 29(3): 278–301.

Silliman, Stephen 2010 Indigenous Traces in Colonial Spaces: Archaeologies of Ambiguity, Origin, and Practice. Journal of Social Archaeology 10(1): 28–58.

Silliman, Stephen W.

812

2005 Social and Physical Landscapes of Contact. In North American Archaeology. Timothy R. Pauketat and Diana DiPaolo, eds. Pp. 273–296. Malden, MA: Blackwell Publishing.

Sims, Lionel 2009 Entering, and Returning From, the Underworld: Reconstituting Silbury Hill by Combining a Quantified Landscape Phenomenology with Archaeoastronomy. Journal of the Royal Anthropological Institute 15(2): 386–408.

Siverts, Karl 1985 Sowing and Shaping - Men and Women: A Discussion o f Meaning and Identity in a Tzeltal Community o f Highland Chiapas. Folk 2\: 17–35.

Smith, Adam T. 2001 The Limitations of Doxa: Agency and Subjectivity from an Archaeological Point of View. Journal of Social Archaeology 1(2): 155–171. 2003 The Political Landscape: Constellations of Authority in Early Complex Socieites. Berkeley: University of California Press.

Smith, Carol A. 1976 Regional Analysis. Studies in Anthropology. New York: Academic Press. http://ezproxy.library.arizona.edu/login?url=http://www.sciencedirect.com/science/book/97 80126521016, accessed August 15, 2018.

Smith, Michael E. 2007 Form and Meaning in the Earliest Cities: A New Approach to Ancient Urban Planning. Journal of Planning History 6(1): 3–47.

Smith, Michael E., and Francis F. Berdan 2003 The Postclassic Mesoamerican World. Salt Lake City: University of Utah Press.

Solis, R. S. 2001 Dating Caral, a Preceramic Site in the Supe Valley on the Central Coast of Peru. Science 292(5517): 723–726.

Spencer, Charles S., and Elsa M. Redmond 2001a Multilevel Selection and Political Evolution in the Valley of Oaxaca, 500–100 B.C. Journal of Anthropological Archaeology 20(2): 195–229. 2001b Multilevel Selection and Political Evolution in the Valley of Oaxaca, 500–100 B.C. Journal of Anthropological Archaeology 20(2): 195–229. 2004 A Late Monte Albán I Phase (300-100 B.C.) Palace in the Valley of Oaxaca. Latin American Antiquity 15(4): 441–455.

Spielmann, Katherine A. 2002 Feasting, Craft Specialization, and the Ritual Mode of Production in Small-Scale Societies. American Anthropologist 104(1): 195–207.

813

Spores, Ronald 1967 The Mixtec Kings and Their People. The Civilization of the American Indian Series,; 85; Variation: Civilization of the American Indian Series ;; v. 85. Norman: University of Oklahoma Press. 1984 The Mixtecs in Ancient and Colonial Times. Norman: University of Oklahoma Press.

Stafford, C. Russell, and Edwin R. Hajic 1992 Landscape Scale: Geoenvironmental Approaches to Prehistoric Settlement Strategies. In . Jaqueline Rossignol and LuAnn Wandsnider, eds. Pp. 137–166. New York: Plenum.

Stark, Barbara n.d. Canchas de La Mixtequilla. In El Juego de Pelota. Annick Daneels, Natalia R. Donner, and Johnathan Hernández Arana, eds. Primer Coloquio Temas Selectos Del Centro-Sur de Veracruz. Mexico City: Universidad Nacional Autónoma de México.

Stark, Miriam T., Mark D. Elson, and Jeffrey J. Clark 1998 Social Boundaries and Technical Choices in Tonto Basin Prehistory. In The Archaeology of Social Boundaries. Washington, D.C.: Smithsonian Institution Press.

Stauffer, J. Grant, and F. Kent Reilly III 2017 Playing the Apalachee Ballgame in the Fields of the Thunder God: Archaeological and Ideological Evidence for Its Antiquity. In Prehistoric Games of North America: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 34–47. Salt Lake City: University of Utah Press.

Stead, Stephen 1995 Humans and PETS in Space. In Archaeology and Geographical Information Systems: A European Perspective. Gary R. Lock and Z. Stanic, eds. Pp. 313–317. London: Taylor & Francis.

Stern, Theodore 1948 The Rubber-Ball Games of the Americas. New York: J. J. Augustin.

Stewart, Andrew M., Darren Keith, and Joan Scottie 2004 Caribou Crossings and Cultural Meanings: Placing Traditional Knowledge and Archaeology in Context in an Inuit Landscape. Journal of Archaeological Method and Theory 11(2): 183–211.

Stichelbaut, Birger 2006 The Application of First World War Aerial Photography to Archaeology: The Belgian Images. Antiquity 80(307): 161–172.

Stoll, Marijke 2010 Es Nuestra Tradición: The Archaeological Implications of an Ethnography of the Modern Ballgame in Oaxaca, Mexico. Paper presented at the 75th Annual Meeting of the

814

Society for American Archaeology, St. Louis, MO. 2014 The Mesoamerican Ballgame in Oaxaca from the Prehispanic Past to Modern Times. Arkeopatías 6(Winter): 49–60. 2015 The Practice of Play in the Sport of Life and Death: Exploring Regional Variation in Ballgame Material Culture and Ideology. Paper presented at the Society for American Archaeology 80th Annual Meetings, San Francisco, CA, April.

Stoll, Marijke, and David S. Anderson 2017 Sport and Ritual as Social Bonding: The Communal Nature of Mesoamerican Ballgames. In Prehistoric Games of the North Americans Indians: Subartic to Mesoamerica. Barbara Voorhies, ed.

Stoll, Marijke M. 2018 Proyecto Arqueológico Juegos de Pelota Nejapa / Tavela 2015. Informe Final. Tucson, AZ.: University of Arizona.

Stone, Tammy 2003 Social Identity and Ethnic Interaction in the Western Pueblos of the American Southwest. Journal of Archaeological Method and Theory 10(1): 31–67.

Stopher, Peter R., and Armin H. Meyburg 1979 Survey Sampling and Multivariate Analysis for Social Scientists and Engineers. Toronto: Lexington Books.

Struna, Nancy L. 1996 People of Prowess: Sport, Leisure, and Labor in Early Anglo-America. Sport and Society. Urbana: University of Illinois Press.

Sturt, Fraser 2006 Local Knowledge Is Required: A Rhythmanalytical Approach to the Late Mesolithic and Early Neolithic of the East Anglian Fenland, UK. Journal of Maritime Archaeology 1(2): 119–139.

Sutro, Livingston D., and Theodore E. Downing 1988 A Step Toward a Grammar of Space: Domestic Space Use in Zapotec Villages. In Household and Community in the Mesoamerican Past. Richard R. Wilk and Wendy Ashmore, eds. Pp. 29–50. Albuquerque, N.M.: University of New Mexico Press.

Swanson, Steve 2003 Documenting Prehistoric Communication Networks: A Case Study in the Paquimé Polity. American Antiquity 68(4): 753–767.

Swenson, Edward R. 2008 The “Disjunctive Approach” to the Archaeological Analysis of Ritual Politics. In Religion, Archaeology, and the Material World. Lars Fogelin, ed. Pp. 237–260. Center for Archaeological Investigations. Carbondale: Southern Illinois University Carbondale.

815

Taladoire, E. 1981 Les terrains de jeu de balle (Mésoamérique et Sud-Ouest des Etats-Unis). Etudes Mésoaméricaines/Estudios Mesoamericanos. Série II Mexico(4). http://cat.inist.fr/?aModele=afficheN&cpsidt=12434080, accessed November 27, 2014. 2017 Los Juegos de Pelota En Mesoamérica: Investigaciones Recientes. Arqueología Mexicana 146(July-August): 26–34.

Taladoire, Eric n.d. El Juego de Pelota Mixteca Entre Los Afrodescendientes de La Costa Chica: Relaciones Interétnicas a Través Del Juego. 2000 El Juego de Pelota Mesoamericano. Origen y Desarrollo. Arqueología Mexicana: 20–27. 2001 The Architectural Background of the Pre-Hispanic Ballgame: An Evolutionary Perspective. In The Sport of Life and Death: The Mesoamerican Ballgame. E. Michael Whittington, ed. Pp. 97–115. Charlotte, NC: Thames & Hudson. 2003 Could We Speak of the Super Bowl at Flushing Meadows?: La Pelota Mixteca, a Third Pre-Hispanic Ballgame, and Its Possible Architectural Context. Ancient Mesoamerica 14(02): 319–342. 2012 Ballgames and Ballcourts in Prehispanic Mesoamerica: A Bibliography. BAR International Series, 2338. Oxford: Archaeopress. 2015a El Centro Del Tlacho, El Centro Del Mundo? Anales de Antropología 49(1): 157– 181. 2015b Las Aportaciones de Los Manuscritos Pictográficos Al Estudio Del Juego de Pelota. Anales Del Instituto de Investigaciones Estéticas 37(106): 181–221.

Taladoire, Eric, and Benoit Colsenet 1991 Bois Ton Sang, Beaumanoir: The Political and Conflictual Aspects of the Ballgame in the Northern Chiapas Area. In The Mesoamerican Ballgame. Vernon Scarborough and David R. Wilcox, eds. Pp. 161–174. Tucson, AZ.: University of Arizona Press.

Taube, Karl, and Marc Zender 2009 American Gladiators: Ritual Boxing in Ancient Mesoamerica. In Blood and Beauty: Organized Violence in the Art and Archaeology of Mesoamerica and Central America. Heather Orr and Rex Koontz, eds. Pp. 161–220. Los Angeles: Cotsen Institute of Archaeology.

Tedlock, Dennis 1996 Popol Vuh: The Definitive Edition of The Mayan Book of The Dawn of Life and The Glories of Gods and Kings. New York: Touchstone.

Thomas, Julian 1993 The Politics of Vision and the Archaeologies of Landscape. In Landscape: Politics and Perspectives. Barbara Bender, ed. Pp. 19–48. Oxford: Berg Press. 2004 Archaeology and Modernity. London: Routledge.

Tilley, Christopher Y. 1994 A Phenomenology of Landscape: Places, Paths, and Monuments. Explorations in

816

Anthropology. Oxford, UK ; Providence, R.I.: Berg. 2004 The Materiality of Stone Explorations in Landscape Phenomenology: 1. Explorations in Landscape Phenomenology ; 1. Oxford ; New York: Berg. http://ezproxy.library.arizona.edu/login?url=http://site.ebrary.com/lib/arizona/Doc?id=1014 6606, accessed August 18, 2018.

Tobler, Waldo 1993 Three Presentations on Geographical Analysis and Modeling. Technical Report, TR-93-1. Santa Barbara: National Center for Geographic Information and Analysis, University of California.

Torquemada, Juan de 1975 Monarquía Indiana. 5th edition. Mexico: Editorial Porrúa.

Tringham, Ruth 1991 Households With Faces: The Challenge of Gender in Prehistorical Architectural Remains. In Engendering Archaeology: Women in Prehistory. J. M. Gero and Margaret Conkey, eds. Pp. 93–131. Oxford: Blackwell Publishing. 1994 Engendered Places in Prehistory. Gender, Place & Culture 1(2): 169–203.

Trouillot, Michel-Rolph 1995 Silencing the Past: Power and the Production of History. Boston, MA: Beacon Press.

Tschan, Andre P., Wlodzimierz Raczkowski, and Malgorzata Latalowa 2000 Perception and Viewsheds: Are They Mutually Inclusive? In Beyond the Map: Archaeology and Spatial Technologies. Gary R. Lock, ed. Pp. 28–48. Amsterdam: IOS Press.

Tsukamoto, Kenichiro, and Takeshi Inomata, eds. 2014 Mesoamerican Plazas: Arenas of Community and Power. Tucson, AZ.: University of Arizona Press.

Tuan, Yi-Fu 1977 Space and Place: The Perspective of Experience. E. Arnold.

Tuan, Yi-fu 1989 Topophilia: A Study of Environmental Perception, Attitudes, and Values. New York: Columbia University Press.

Turnbull, David 2002 Performance and Narrative, Bodies and Movement in the Construction of Places and Objects, Spaces and Knowledges: The Case of the Maltese Megaliths. Theory Culture Society 19(5/6): 125–143.

Turner, Frederick Jackson 1893 The Significance of the Frontier in American History: 19.

817

1896 The Problem of the West. 1932 The Significance of Sections in American History. H. Holt and Company.

Turner, Victor W. 1957 Schism and Continuity in an African Society; a Study of Ndembu Village Life. Manchester, Eng.: Published on behalf of the Rhodes-Livingstone Institute, Northern Rhodesia, by Manchester University Press. 1969 The Ritual Process: Structure and AntiStructure. Chicago: Aldine Publishing. 1982 From Ritual to Theatre: The Human Seriousness of Play. Performance Studies Series ; 1st V. New York City: Performing Arts Journal Publications. 1985 Experience and Performance: Towards a New Processual Anthropology. In On the Edge of the Bush: Anthropology as Experience. Victor W. Turner, ed. Pp. 205–226. Tucson, AZ.: University of Arizona Press.

Turok, Marta 2000 Entre El Sincretismo y La Superviviencia: El Juego de Pelota En La Actualidad. Arqueología Mexicana: 58–65.

Tylor, Edward B. 1879 The History of Games. The Fornightly Review.

Upton, Dell 1996 Ethnicity, Authenticity, and Invented Traditions. Historical Archaeology 30(2): 1–7.

Urcid, Javier 2005 The Zapotec Social Tradition: Knowledge, Memory, and Society in Ancient Oaxaca. Coral Gables, FL: Foundation for the Advancement of Mesoamerican Studies Inc. 2014 Otra Narrativa de Jugadores de Pelota En Dainzú. In Panorama Arqueológico: Dos Oaxacas. Marcus Winter and Gonzalo Sánchez Santiago, eds. Pp. 43–56. Oaxaca, Mexico: INAH and Consejo Naciónal para la Cultura y las Artes.

Uriarte, María Teresa 2000 Práctica y Símbolos Del Juego de Pelota. Mariposas, Sapos, Jaguares y Estrellas. Arqueología Mexicana: 28–35. 2001 Unity in Duality: The Practice and Symbols of the Mesoamerican Ballgame. In The Sport of Life and Death: The Mesoamerican Ballgame. E. Michael Whittington, ed. Pp. 40– 49. Charlotte, NC: Thames & Hudson.

Valeri, Valerio 1999 The Forest of Taboos: Morality, Hunting, and Identity Among the Huaulu of the Moluccas. Madison: University of Wisconsin Press.

Vaughn, Sallie, and Tom Crawford 2009 A Predictive Model of Archaeological Potential: An Example from Northwestern Belize. Applied Geography 29(4): 542–555.

818

Veblen, Thorstein 2009 The Theory of the Leisure Class. S.l.: The Floating Press.

Vennum, Thomas 1994 American Indian Lacrosse: Little Brother of War. Washington, D.C.: Smithsonian Institution Press.

Verhagen, Philip, ed. 2007 Case Studies in Archaeological Predictive Modelling. Amsterdam University Press.

Verhagen, Philip, Hans Kamermans, Martijn van Leusen, et al. 2007 First Thoughts on the Incorporation of Cultural Variables into Predictive Modelling. In Case Studies in Predictive Modeling. Philip Verhagen, ed. Pp. 203–208. Leiden, Netherlands: Universiteit Leiden.

Verhagen, Philip, and Thomas G. Whitley 2012 Integrating Archaeological Theory and Predictive Modeling: A Live Report from the Scene. Journal of Archaeological Method and Theory 19(1): 49–100.

Villamil, Laura Paola 2005 Divergent Cityscapes: Urban Patterns at Two Ancient Maya Centers in Central Quintana Roo, Mexico. Ph.D., University of Michigan. http://search.proquest.com.ezproxy2.library.arizona.edu/pqdtft/docview/305451056/abstrac t/A5605008D24C4E33PQ/1?accountid=8360, accessed September 3, 2014.

Voorhies, Barbara 2017a Introduction. In Prehistoric Games of North American Indians. Barbara Voorhies, ed. Pp. 1–18. Salt Lake City: University of Utah Press. , ed. 2017b Prehistoric Games of North American Indians: Subartic to Mesoamerica. Salt Lake City: University of Utah Press.

Voss, Barbara L. 2005 From Casta to Californio: Social Identity and the Archaeology of Culture Contact. American Anthropologist 107(3): 461–474.

Walden, John, and Barbara Voorhies 2017 Ancient Maya Patolli. In Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 197–218. Salt Lake City: University of Utah Press.

Wandsnider, LuAnn 1998 Regional Scale Processes and Archaeological Landscape. In Unit Issues in Archaeology: Measuring Time, Space, and Material. Ann F. Ramenofsky and Anastasia Steffen, eds. Pp. 87–102. Salt Lake City: University of Utah Press.

819

Wansleeben, M., and L. Verhart 1997 Geographical Information Systems: Methodological Progress and Theoretical Decline. Archaeological Dialogues 4(1): 53–70.

Warren, Robert E., and David L. Asch 2000 A Predictive Model of Archaeological Site Location in the Eastern Prairie Peninsula. In Practical Applications of GIS for Archaeologists: A Predictive Modelling Toolkit. Konnie L. Wescott and R. Joe Brandon, eds. Pp. 5–32. Philadelphia: Taylor & Francis.

Weigand, Phil C. 2000 The Evolution and Decline of a Core of Civilization, The Teuchitlán Tradition and the Archaeology of Jalisco. In Greater Mesoamerica: The Archaeology of West and Northwest Mexico. Michael S. Foster and Shirley Gorenstein, eds. Pp. 43–58. Salt Lake City: University of Utah Press.

Weismantel, Mary 2004 Moche Sex Pots: Reproduction and Temporality in Ancient South America. American Anthropologist 106(3). New Series: 495–505.

Wenger, Etienne 1998 Communities of Practice: Learning, Meaning, and Identity. Cambridge University Press. 2000 Communities of Practice and Social Learning Systems. Organization 7(2): 225–246.

Wernke, Steven A. 2007 Negotiating Community and Landscape in the Peruvian Andes: A Transconquest View. American Anthropologist 109(1): 130–152.

Wescott, Konnie L., and R. Joe Brandon 2000 Practical Applications of GIS for Archaeologists: A Predictive Modelling Toolkit. Philadelphia: Taylor & Francis.

Whalen, Michael E. 1988 Small Community Organization During the Late Formative Period in Oaxaca, Mexico. Journal of Field Archaeology 15(3): 291–306.

Whalen, Michael E., and Paul E. Minnis 1996 Ball Courts and Political Centralization in the Casas Grandes Region. American Antiquity 61(4): 732–746. 2001 Casas Grandes and Its Hinterland: Prehistoric Regional Organization in Northwest Mexico. University of Arizona Press.

Wheatley, David 1995 Cumulative Viewshed Analysis: A GIS-Based Method for Investigating Intervisibility, and Its Archaeological Application. In Archaeology and Geographic Information Systems. G. Lock and Z. Stancic, eds. Pp. 171–185. London: Taylor & Francis.

820

2000 Spatial Technology and Archaeologial Theory Revisited. In CAA ’96—Computer Applications and Quantitative Methods in Archaeology. K. Lockyear, J. T. Sly, and Mihailescu-Birliba, eds. Pp. 123–132. London: Archaeopress.

Wheatley, David, and Mark Gillings 2000 Vision, Perception, and GIS: Devleoping Enriched Approaches to the Study of Archaeological Visibility. In Beyond the Map: Archaeology and Spatial Technologies. G. Lock, ed. Pp. 1–27. Amsterdam: IOS Press. 2002 Spatial Technology and Archaeology: The Archaeological Applications of GIS. New York: Taylor & Francis.

White, Devin A., and Sarah B. Barber 2012 Geospatial Modeling of Pedestrian Transportation Networks: A Case Study from Precolumbian Oaxaca, Mexico. Journal of Archaeological Science 39(8): 2684–2696.

White, Devin A., and Sarah L. Surface-Evans 2012 Least Cost Analysis of Social Landscapes: Archaeological Case Studies. University of Utah Press.

White, Richard 1991 The Middle Ground: Indians, Empires, and Republics in the Great Lakes Region, 1650-1815. Cambridge Studies in North American Indian History. Cambridge ; New York: Cambridge University Press.

Whitehead, Gibson 1990 Warfare and the Evolution of Tribal Polities in the Prehistoric Southwest. In The Anthropology of War. Johnathon Haas, ed. Pp. 171-.

Whitley, Thomas G., and Lacey M. Hicks 2003 A Geographic Information Systems Approach to Understanding Potential Prehistoric and Historic Travel Corridors. Southeastern Archaeology 22(1): 77–91.

Wilkerson, S. Jeffrey K. 1991 And Then They Were Sacrificed: The Ritual Ballgame of Northeastern Mesoamerica Through Time and Space. In The Mesoamerican Ballgame. Vernon L. Scarborough and David R. Wilcox, eds. Pp. 45–72. Tucson, AZ.: University of Arizona Press.

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.

Williamson, Ronald F., and Martin S. Cooper 2017 “He Must Die Unless the Whole Country Shall Play Crosse”: The Role of Gaming in Great Lakes Indigenous Societies. In Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 48–62. Salt Lake City: University of Utah Press.

821

Wills, W. H., and Robert D. Leonard, eds. 1994 The Ancient Southwestern Community: Models and Methods for the Study of Prehistoric Social Organization. 1st ed.. Albuquerque: University of New Mexico Press.

Winter, Marcus 2007 Cerro de Las Minas: Arqueología de La Mixteca Baja. Arqueología Oaxaqueña, Serie Popular, 1. Oaxaca, Mexico: Centro Nacional de Antropología e Historia. 2008 Classic to Postclassic in Four Regions: The Mazateca, the Chinantla, the Mixe Region, and the Southern Isthmus. In After Monte Albán. Jeffrey P. Blomster, ed. Pp. 393– 426. Boulder, CO.: University Press of Colorado.

Winter, Marcus, Tamara Ballensky, Jason Ballensky, and Javier Pérez Guerrero 2014 La Cueva Del Rey Kong-Oy. In Panorama Arqueológico: Dos Oaxacas. Marcus Winter and Gonzalo Sánchez Santiago, eds. Pp. 293–320. Arqueología Oaxaqueña, 4. Oaxaca, Mexico: INAH and Consejo Naciónal para la Cultura y las Artes.

Winter, Marcus, Alma Z. Montiel Ángeles, Victor M. Zapien López, et al. 2008 Proyecto Salvamento Arqueológico Carretera Oaxaca-Istmo 2006-2007: Tramo Jalapa Del Marqués KM 177-190. Final informe. Oaxaca, Mexico: Centro INAH Oaxaca.

Winter, Marcus, and Gonzalo Sánchez Santiago, eds. 2014 Panorama Arqueológico: Dos Oaxacas. Arqueología Oaxaqueña, 4. Oaxaca, Mexico: INAH and Consejo Naciónal para la Cultura y las Artes.

Witcher, R.E. 1999 GIS and Landscapes of Perception. In Geographical Information Systems and Landscape Archaeology. Mark Gillings, D. Mattingly, and J. Van Dalen, eds. Pp. 13–22. Oxford: Oxbow Books.

Woodman, P.E. 2000 A Predictive Model for Mesolithic Site Location on Islay Using Logistic Regression and GIS. In Hunter–Gatherer Landscape Archaeology: The Southern Hebrides Mesolithic Project 1988–1998, Archaeological Fieldwork on Colonsay, Computer Modelling, Experimental Archaeology, and Final Interpretations. Steven J. Mither, ed. Pp. 444–464. Cambridge: McDonald Institute Monographs.

Workinger, Andrew G. 2013 Coastal/Highland Interaction in Oaxaca, Mexico: The Perspective from San Francisco de Arriba. In Polity and Ecology in Formative Period Coastal Qaxaca. Arthur A. Joyce, ed. Pp. 193–226. Boulder, CO.: University Press of Colorado.

Wright, James C. 2004 The Mycenaean Feast. ASCSA.

Wright, Joshua 2012 Landscapes of Inequality? A Critique of Monumental Hierarchy in the Mongolian

822

Bronze Age. Asian Perspectives 51(2): 139–163.

Wyman, Walker D., and Clifton B. Kroeber 1957 The Frontier in Perspective. Madison: University of Wisconsin Press.

Yaeger, Jason 2000 The Social Construction of Communities in the Classic Maya Countryside: Strategies of Affiliation in Western Belize. In The Archaeology of Communities: A New World Perspective. Marcello A. Canuto and Jason Yaeger, eds. Pp. 123–142. London: Routledge.

Yanicki, Gabriel M. 2017 Reinventing the Wheel Game: Prestige Gambling on the Plains/Plateau Frontier. In Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 104–118. Salt Lake City: University of Utah Press.

Yanicki, Gabriel M., and John W. Ives 2017 Mobility, Exchange, and the Fluency of Games: Promontory in a Broader Sociodemographic Setting. In Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 139–162. Salt Lake City: University of Utah Press.

Yannakakis, Yanna P. 2007 The Indios Conquistadores of Oaxaca’s Sierra Norte. In Indian Conquistadors: Indigenous Allies in the Conquest of Mesoamerica. Laura E. Matthew and Michel R. Oudijk, eds. Pp. 227–253. Norman: University of Oklahoma Press.

Zborover, Dan A. 2014 Decolonizing Historical Archaeology in Southern Oaxaca, Mexico: Late Formative to Republican Periods. University of Calgary.

Zborover, Danny A. 2007 El Proyecto de Arqueología Histórica de La Chontalpa, Oaxaca. 05038. Foundation for the Advancement of Mesoamerican Studies, Inc. 2008 Identidades Faccionales En “Narrativas Territoriales” de La Oaxaca Colonial: Un Enfoque Desde Las Montañas Chontales. In Pictografia y Escritura Alfabetica En Oaxaca. Bas van Doesburg, ed. Pp. 233–270. Oaxaca, Mexico: Instituto Estatal de Educacion Publica de Oaxaca.

Zeitlin, Judith Francis 1993 The Politics of Classic-Period Ritual Interaction. Ancient Mesoamerica 4(01): 121– 140. 1994 Precolumbian Barrio Organization in Tehuantepec, Mexico. In Caciques and Their People: A Volume in Honor of Ronald Spore. Joyce Marcus and Judith Francis Zeitlin, eds. Pp. 275–300. Ann Arbor: University of Michigan. 2005 Cultural Politics in Colonial Tehuantepec: Community and State Among the Isthmus Zapotecs, 1500-1750. Palo Alto, CA: Stanford University Press.

823

Zeitlin, Robert N. 1978 Long-Distance Exchange and the Growth of a Regional Center: An Example from the Southern Isthmus of Tehuantepec, Mexico. In Prehistoric Coastal Adaptations: The Economy and Ecology of Maritime Middle America. Barbara L. Stark, ed. Pp. 183–210. New York: Academic Press.

Zubrow, Ezra B. W. 1994 Cognitive Archaeology Reconsidered. In The Ancient Mind: Elements of Cognitive Archaeology. Ezra B. W. Zubrow and Colin Renfrew, eds. Pp. 187–190. Cambridge: Cambridge University Press.

Zych, Thomas J. 2017 Chunkey and the Historic Experience in the Mississippian World. In Prehistoric Games of North American Indians: Subartic to Mesoamerica. Barbara Voorhies, ed. Pp. 63–86. Salt Lake City: University of Utah Press.

824