Garbage, society, and environment in a Mexican municipio: The case of Coxcatlán, Puebla, Mexico By Andrew M. Hilburn ©2014 Submitted to the graduate degree program in Geography and the Graduate Faculty of the University of Kansas in partial fulfillment of the requirements for the degree of Doctor of Philosophy. __________________________________ Chairperson, Peter H. Herlihy __________________________________ Jerome E. Dobson __________________________________ Brent E. Metz __________________________________ Garth A. Myers __________________________________ Barney Warf __________________________________ William I. Woods Date defended: November 22, 2013 The Dissertation Committee for Andrew M. Hilburn certifies that this is the approved version of the following dissertation Garbage, society, and environment in a Mexican municipio: The case of Coxcatlán, Puebla, Mexico __________________________________ Chairperson, Peter H. Herlihy Date approved: January 25, 2014 iii Abstract This dissertation examines the practice of garbage management and its construction as a set of environmental issues across the municipio of Coxcatlán, Puebla, Mexico. The generation and sustainable management of garbage are pressing environmental issues around the world. This is especially true in a rural Mexican municipios, like Coxcatlán, where municipal govern- ments attempt to deliver garbage service within larger policy directives with limited infrastruc- ture and accessibility challenges. The different geographies of garbage management created by differential accessibility to formal management, local governance, the influence of traditional practices, and public education among other factors contribute to how garbage is perceived of as an environmental issue. To understand the practical and discursive management of garbage across Coxcatlán municipio, this study employs 1) an analysis of the current state of garbage management in Mexico as a whole and Coxcatlán in particular 2) a household garbage census to illustrate the management of garbage in the home from a practice-based perspective, 3) a Q-method study to understand the subjectivities regarding what garbage is and how it should be managed in the home, 4) a risk mapping survey to describe the heterogeneity of perception of garbage-related issues across the municipio, and 5) a litter quantification and classification study to compare perceptions of litteredness to the material realities of garbage that fall outside of man- agement. The results of the broader analysis of the garbage issue in Mexico and Coxcatlán frames the material magnitude of garbage and the trajectory of management through which this material passes. The extensive municipio-wide garbage management census shows how differential ac- cessibility to formal garbage management essentially creates two spatialities or models of gar- bage management with different effects on residents and the environment in each. The Q-method study identifies six different perspectives on what garbage is and how it should be managed and the participatory risk mapping study describes precisely what are the most pressing garbage is- sues to local residents and how this is contingent on spatial and demographic factors. Lastly, the litter study results highlight the importance of estimating the perception of an issue in addition to more quantitative framings. All of the results in the context of a regional geography of Coxcatlán provide a fuller picture of garbage, society and the environment beyond simple garbage genera- tion narratives. Acknowledgements This is a dissertation about garbage and its management in small-town Mexico. You may read that and ask yourself, “What?”. Presuming that few others besides me would immediately find this a worthwhile topic to spend the past five years working on, some explanation is required to show how that came to be. The following preface section goes over that. Here I’d like to men- tion those along the route who without them this would not have been possible or as fun. These are friends, advisers, family, functionaries, and mix-and-matches of all of those who opened doors for me and gave me the right advice at a number of points in my life. I never would have gotten into geography in the first place had it not been for the great faculty at the Department of Earth Sciences at the University of South Alabama. Before see- ing the light, I was a struggling, semi-disinterested biology major and lumberyard lackey with a general interest in natural science, anthropology, history, political science. I had no idea that geography was all that and more. Dr. Roy Ryder, my adviser there, was instrumental in awaken- ing me to what geography was as well as nurturing my interest in Latin America. Besides being a caring and considerate adviser, his example of being an ex-pat Scotsman who had lived and taught in Costa Rica, Ecuador, and then Mobile, and did all that as a career, opened my eyes to what geography was. My thanks also go to Drs. Mimi Fearn, Glenn Sebastian, Lary Dilsaver, and Vicky Rivizzigno who also floated geography as not only a topic of interest but a potential career and made the Life Sciences Building seem like a second home. I am indebted to the folks at the Department of Geography and Geology at the University of Southern Mississippi for offering me an opportunity to come to Hattiesburg and study and be a teaching assistant. The two years I spent getting my Masters and living in Hattiesburg were some of the happiest of my life. It’s really hard to quantify my thanks to the people at USM in the past tense as I still remain friends and colleagues with them. At the time, the department faculty was really young and dynamic and they opened not only their offices to me but their lives as well. I am especially grateful to the big Don of the department, Dr. Clifton “Skeeter” Dixon, for offering me the teaching assistantship and keeping me on track. My thesis adviser, Dr. Joby Bass quickly grasped the fact that I was interested in Mexico, and had me jump headfirst into a topic quickly. Joby was and still is today a dedicated and timely adviser, editor, and counselor. Dr. David Co- chran, while not officially my adviser, was as influential as Joby in my career development and iv vadvisement. He has also given me great advice over the years in addition to graciously picking up the tabs at Hattiesburg bars and coffee shops when I pass through town. While I never got into palynology or climatology at USM, I count Dr. Andy Reese as an adviser and friend. In addition to being fun to work with as his TA, Andy and his wife Olivia were also kind enough to have get togethers for the grad students. Thanks also go to Drs. Jason Janke, George Raber, Jerry Griffith, and George Roedl. Beyond the faculty, I had a surrogate family of fellow grad students at USM with Tom Strange and Chris Storm who remain two of my best friends today, but whom I don’t see often enough. Danny Redo, Eli Peneva, Caleb Smith, and David Barber were also great fun. Lastly, thanks go to the Hardy Street McDonald’s, La Fiesta Brava, Nick’s Ice House, The Thirsty Hippo, and Peter’s Poboys for the greater part of my subsistence. At KU, I would never be what and where I am today without Dr. Peter Herlihy. I met Dr. Herlihy at the Chicago AAG meeting when I was unsure of what I was going to do after USM following a complete lack of offers from any Ph.D. program. After a good talking up by Joby and David and seeing I knew some GIS and had worked in Mexico, Dr. Herlihy found support and took me on as a graduate student at KU. Since then, Dr. Herlihy has been an “all in” adviser, meaning that he doesn’t take on many graduate students, but for those that he does, he ensures his involvement in every important step in their careers. I have had the pleasure of working with him on the Bowman Expeditions over the years and have spent considerable time with him in the field. Dr. Herlihy has been a devoted editor and academic counselor in the Carl Sauer “Berkleley School” tradition and has graciously included me and his other students into his publications. Dr. Herlihy takes the approach that advisors have an obligation toward the future wellbeing of their students in their professions, and to a degree, their wellbeing in their personal lives. This holistic approach to advising is perhaps the thing for which I am most thankful. He has created a familial group of friends and scholars who have worked and still work in Room 202 of Lindley Hall. In addition to Dr. Herlihy, Dr. Bill Woods is the other reason I am at KU. Dr. Woods offered me the first teaching assistantship position I had here when he was director of the Envi- ronmental Studies program and has served as a personal mentor and benefactor since I have been here. His support for my research and professional wellbeing and countless letters of support has been invaluable to my career here at KU. Thanks go to Dr. Garth Myers for his support of my research on garbage, service on my committee, and for keeping the goal empty and the refs hon- est on the soccer pitch. I appreciate Dr. Barney Warf’s precise editing, committee service, and re- freshingly dry sense of humor. The career advice, tireless writing of letters of support, and friend- ship from Dr. Brent Metz has been above and beyond the call of duty for the extra-departmental committee member. Dr. Jerry Dobson has been a key figure in my professional development and has kept me employed as a Bowmanista for a number of years. As a native Georgian, he is also the only other Southerner in the department and is a fun person with whom to share this conceit. Besides my committee, thanks go to Drs. Terry Slocum and Johannes Feddema who have served as chairs for the department during my stay here. Additional thanks go to Darin Grauberger, from whom I learned a lot about production cartography. During these years at KU, I have had the pleasure and distinctive pleasure of sharing time in the office and in Latin America, as well as share in the personal lives of John Kelly, Aida Ra- mos, Andy Norris, and Taylor Tappan. None of us came from Lawrence or anywhere nearby but we consider the pack that has offered mutual support and unconditional friendship over the years a family of sorts. Besides the obvious academic aspect of my stay here in Lawrence, I consider building these relationships one of the most rewarding experiences of the past seven years. In ad- dition to the self-dubbed “202 crew”, I am glad to have met other friends and colleagues such as Luke Struckman, Aaron Gilbreath, Andy Allen, Heather Putnam, Lisa Rausch, Mike Bergervoet, Gerardo Hernández, Stephs Kozak and Willis, Jenny (Brackhan) Young, Norberto Baldi, Ismael Hinojosa, Geoff Leonard, Rebecca Croswaith, Pat Green, Andrew Gottsfeld, Brendon Asher, Mark Hummer, Yoshi Nakazawa, Stephan Fuchs, Jared Beeton, Josh Long as well as many oth- ers that comprise the pan-KU grad student cohort. At other universities, I have had tremendous opportunities as a graduate student to meet and even work with some amazing Latin Americanist geographers. I am glad to know Dr. Derek Smith and Dr. Scott Brady with whom I have worked in Mexico and shared many not-so-cold beers and stale botanas. Thanks are also due to Dr. Bill Doolittle who has kept a supportive inter- est in my developing scholarship. Over the years, Dr. Doolittle has been an outside adviser of sorts and I feel honored to know him. I am also thankful for the great advice and friendship from other Latin Americanist geographers such as Drs. Bill Davidson, Craig Revels, Matt Fry, Alex- andra Ponette-González, David Salisbury, Mike Steinberg, Julie Velásquez, Chris Brown, and Christian Brannstrom. Institutionally, thanks go out to Dr. Charles Bankhart who was the grant coordinator for the Fulbright-Hays application and awarding process. Thanks also to Dr. Amy Bishop at vi the US Department of Education. Also, I thank Dr. Miguel Aguilar-Robledo at the Universidad Autónoma de San Luis Potosí who graciously offered himself and the Coordinación de Ciencias Sociales y Humanidades as a host and host institution in Mexico. Esther Mendoza was an im- mense help at the COMEXUS office in Mexico City for all things immigration and during my medical emergency. Thanks go to the offices of the Puebla Secretariat of Environmental Sustain- ability and Land Use and the Tehuacán-Cuicatlán Biosphere Reserve. In Coxcatlán, where I lived during my field research, I give immense gratitude to Coxcatlán’s presidente municipal Efigenio Atilano who approved my research, answered my questions, and offered his support during my stay. Additional thanks go to the previous presidente Miguel Zerón and his office for drafting an earlier letter of support.Thanks are due to the cabildo de regidores (regents board) in the munici- pal government in Coxcatlán, the individual presidentes auxiliares (auxiliary presidents) in each of the towns where I worked, and the Coxcatlán municipal sanitation crew. Thanks go to Dr. Jóse Luis Hernández and the rest of the Sagrado Corazón Hospital staff in Tehuacán for the expert ap- pendectomy and aftercare. I can never thank the Alva and Armas families enough for letting me invade their lives during my ten months in Coxcatlán. Miguel Alva, his mother Doña Teresa, and daughter Tere opened their home, dinner table, daily lives, and hearts to me while I was in Coxcatlán. I have a similar debt of gratitude to Jóse Luis “Pepe” Armas with whom I lived for the latter half of my stay. Acknowledging my Coxcateco family would not be complete without noting Salvio Mier, Emiliano Alva, Berna Armas, and Paty and the entire Aguilar clan. My surveys would not have been as comprehensive had it not been for the efforts of Ricardo Olaya who besides being an excellent survey taker is a super chingón saxofonista. I hope to keep them all as friends for life. Most importantly, I thank my own family for their support of my research and career choice. I can honestly say I would never have entered graduate school had I never met my wife Alison Hadley during my second-to-last semester at South Alabama. Her passion for archaeology as a career inspired me to pursue mine for geography. I also owe her years of life and together- ness for the days, weeks, months, and probably aggregate years we’ve lived spatially apart doing our own research. Our mutual experience of doing our own dissertations has been pleasurably frustrating. I can’t think of anyone better with whom to do any of this. Thanks are also due to my step-mother, Gayle Hilburn, for convincing me to return to my undergraduate studies after a brief existential crisis. I should also thank my brother J.D. for always being available to talk and vii coming to Mexico when I had my medical emergency. The most thanks are due to my father and mother, Jere and the late Sue Hilburn, who seeded all of my interest in geography by keeping a house full of books, atlases, and field guides, and always unconditionally encouraged any intel- lectual pursuit of mine. This dissertation is dedicated to my grandmother Iva Nell Hilburn who passed away dur- ing the course of my fieldwork. viii Table of Contents ABSTRACT iii ACKNOWLEDGEMENTS iv TABLE OF CONTENTS ix LIST OF FIGURES xv LIST OF TABLES xxi GLOSSARY OF SELECT TERMS xxiv CHAPTER 1: Garbage, environment, and society in Coxcatlán, Puebla, Mexico 1 INTRODUCTION 1 A note on definitions 3 OBJECTIVES OF THE DISSERTATION 3 METHODS 6 Introduction 6 Household garbage census 7 Q-method survey 8 The participatory risk mapping of garbage 9 Studying roadside litter 10 Participant observation, hanging out, and gathering information 11 A CULTURAL AND POLITICAL ECOLOGY OF GARBAGE 13 A brief history of the human-environmental approach 13 CURRENT GEOGRAPHICAL AND SOCIAL SCIENCE RESEARCH ON GARBAGE 19 Environmental governance and garbage 20 Informal waste sector research 22 Environmental justice and waste 24 ix x Previous research on garbage in Mexico 25 Littering 26 FORWARD TO THE NEXT CHAPTERS 28 CHAPTER 2: Land and life in Coxcatlán, Puebla, Mexico 29 LOCATING COXCATLÁN 29 The municipio 29 Coxcatlán in space and time 30 COXCATLÁN’S “NATURAL” SETTING 33 Landforms, relief, and hydrology 34 Climate in the valley and sierra 39 Vegetation 44 PROPERTY, SETTLEMENTS, AND POPULATION 49 Land Tenure and Ejidos 49 Toponyms 56 The cabecera municipal: Coxcatlán 58 Calipan 65 Tilapa, Pueblo Nuevo, and San Rafael 67 The Sierra Communities 69 Between the Valley and Sierra in numbers 74 DAILY LIFE AND ECONOMY 78 Agriculture in the valley 78 Agriculture, sierra style 85 Transportation and communication 90 Consumption in Coxcatlán 94 Municipio Governance in the Neoliberal Era 97 The Narco Wars 99 SUMMARY 101 CHAPTER 3: Garbage Management in Coxcatlán 103 GARBAGE MANAGEMENT IN MESOAMERICA, NEW SPAIN, AND MEXICO 103 Middens, heaps, and scatters 103 Dooryard gardens: A window into the past 104 A brief note on urban garbage management in the historical period 108 CONTEMPORARY GARBAGE MANAGEMENT IN MEXICO 109 Framing the garbage issue in Mexico 109 Household municipal solid waste generation in Coxcatlán 115 Legal foundations of garbage management in Mexico 118 MUNICIPAL MANAGEMENT APPROACHES 124 Local organization and practice of MSW collection 124 Service provision and collection estimates 130 THE PRACTICE OF HOUSEHOLD GARBAGE MANAGEMENT IN COXCATLÁN 133 Collecting and storing 133 Reuse and recycling 135 Dumping and burying 137 Burning 140 Composting, “abono”, and keeping animals 143 THE GARBAGE MANGAGEMENT TRAJECTORY IN COXCATLÁN 143 CHAPTER 4: Household garbage management in Coxcatlán 148 OBJECTIVES AND METHODS 149 Project intent 149 The household and population sample 150 RESULTS 154 Participant and Household Demographics 154 Garbage Management Practices 166 DISCUSSION AND CONCLUSION 174 Participation in municipal service 174 Traditional practices 177 Spatialities of Garbage Management 181 Conclusion 188 xi CHAPTER 5: Q-Methodology, subjectivity, and household garbage in Coxcatlán 190 INTRODUCTION 190 METHODS AND STUDY POPULATION 191 What is Q-Methodology? 191 Doing Q-method research in Coxcatlán 192 ANALYSIS OF GARBAGE COMPOSITION SORT 203 Representing what we [think we] throw away 203 Factor 1: Mostly Foodies 206 Factor 2: Packaged 207 Factor 3: Homebodies 209 ANALYSIS OF GARBAGE MANAGEMENT SORT 210 The management of particular garbage classes 210 Factor 1: Sanitary Citizens 212 Factor 2: Home Recyclers 213 Factor 3: Convenience Independents 214 DISCUSSION AND CONCLUSION 216 The garbage generation sort 216 The garbage management sort 218 Concluding remarks 219 CHAPTER 6: The construction of garbage as a set of environmental issues in Coxcatlán 221 INTRODUCTION 221 METHODS 222 “Participatory” “Risk” Issue “Mapping” 222 Data collection and analysis 224 Fieldwork and study population 226 RESULTS AND DISCUSSION 232 Mapping issues across the municipio 232 Burning 233 Dumping 242 xii Public cleanliness 244 Other issues 245 Chi-square and logistic regression results 247 Conceptualizations of garbage 249 CONCLUSION 251 CHAPTER 7: Roadside littering in Coxcatlán 253 INTRODUCTION 253 BACKGROUND 254 Litter and Littered Landscapes 254 Research Approaches 259 SITE SELECTION AND METHODS 262 LITTER COLLECTIONS 267 Initial Clean 267 First Collection 269 Second Collection 272 Total collection results 274 INFERENTIAL STATISTICS TESTS AND RESULTS 277 Methods 277 Results 278 DISCUSSION AND CONCLUSION 279 Generation of litter and spatial modeling 279 Composition 284 Inferential tests results 287 Management Implications 288 Conclusion 289 Conclusion: Geographies of garbage in Coxcatlán, Puebla, Mexico 292 GENERAL RESEARCH FINDINGS 292 Garbage generation, governance, and management in Coxcatlán 292 The state of garbage management in Coxcatlán’s households 294 xiii The subjectivities of garbage generation and management 297 The heterogeneity of opinion regarding garbage-related issues 299 Material realities and perception of litter in Coxcatlán 301 CONCLUSION 304 Final commentary 304 Coming back and the future of garbage in Coxcatlán? 306 REFERENCES 308 Appendix 1: Researching Garbage in Small Town Mexico 338 INTRODUCTION 338 STUDYING GARBAGE IN MEXICO 338 Before the dissertation 338 January to June 2011 344 Late June: Work Interruption 353 Getting back to it: August to December 2011 357 Appendix 2: A Prehistoric and historical regional geography of Coxcatlán 363 Palaeoindian to Archaic 364 Archaic Coxcatlán 366 Preclassic beginnings 367 Classic Coxcatlán and Sansuanchi 369 Conquest, Corregimiento, Haciendas, and Persistence 372 Revolution and Institutions 377 Neoliberal Era to Today in Mexico and the Tehuacán Valley 379 xiv List of Figures CHAPTER 1 Figure 1.1 Municipal waste generation by OECD and select countries, 2007 2 Figure 1.2 Locations of long-term, peer-reviewed research on generation and classification in Mexico 19 CHAPTER 2 Figure 2.1 Map of municipios in Mexico in 2010 29 Figure 2.2 Municipio of Coxcatlán, Puebla, Mexico 32 Figure 2.3 Route from Coxcatlán - Tehuacán - Puebla - Mexico City 33 Figure 2.4 Tehuacán-Cuicatlán Biosphere Reserve and Coxcatlán’s location within it 35 Figure 2.5 Topographical relief and surface hydrology, municipio of Coxcatlán, Puebla, Mexico 36 Figure 2.6 Barranca Atempango streambed in December 37 Figure 2.7 Engineered stream used for the transmission of water to cabecera municipal 37 Figure 2.8 Elevation gradient, municipio of Coxcatlán, Puebla, Mexico 38 Figure 2.9 Climagraph for Coxcatlán and Zoquitlán, 1971-2001 40 Figure 2.10 Dry-season selva baja caducifolia 41 Figure 2.11 Dry-season temporal field outside Coxcatlán cabecera 41 Figure 2.12 Wet-season matorral vegetation outside Coxcatlán cabecera 42 Figure 2.13 Wet-season sugarcane field near Venta Salada 42 Figure 2.14 Vegetation communities in Coxcatlán municipio 45 Figure 2.15 Madrone and oak woods above Xacalco 46 Figure 2.16 Induced pasture vegetation class in foreground with irrigated sugarcane in the background 46 Figure 2.17 Matorral crausicaule vegetation south of the cabecera municipal 47 Figure 2.18 Selva baja caducifolia below Tequexpalco 47 Figure 2.19 Hypothetical spatial-land use divisions within a nucleo agrario 51 Figure 2.20 Social property (ejidos) in Coxcatlan, Puebla 54 Figure 2.21 Urban area and colonias of Coxcatlan, Coxcatlan, Puebla 59 x v x vi Figure 2.22 Looking south to Coxcatlán’s centro from Colonia Calvario 61 Figure 2.23 Looking north to Coxcatlán’s centro from Colonia Benito Juárez 61 Figure 2.24 Busy morning in the parque (plaza) in the center of town 62 Figure 2.25 Ayuntamiento building 62 Figure 2.26 Looking west at Colonia San Isidro and Colonia Donato Bravo Izquierdo 63 Figure 2.27 Colonia Benito Juárez streetscape 63 Figure 2.28 Calipan from the sierra above town 66 Figure 2.29 Pueblo Nuevo from Cerro Colorado 68 Figure 2.30 Newly paved street in San Rafael 68 Figure 2.31 View from above Ocotlamanic on the road from Tepeyoloc 71 Figure 2.32 Looking toward the town of Xacalco from a house lot 71 Figure 2.33 Tequexpalco homes with near-lot fields 72 Figure 2.34 Ranchería Pala with Coxcatlán in the background below 72 Figure 2.35 Sugarcane lands near the Río Salado, Coxcatlán, Puebla 79 Figure 2.36 Main source irrigation canal, Coxcatlán, Puebla 81 Figure 2.37 Terraces and bordas with freshly planted sugarcane, Coxcatlán, Puebla 81 Figure 2.38 Ingenio Calipan trucks waiting to be unloaded, Calipan, Cox., Puebla 83 Figure 2.39 Burning the remnant husks after the harvest, Coxcatlán, Cox., Puebla 83 Figure 2.40 Infield milpa agriculture above Xacalco, Cox., Puebla 87 Figure 2.41 Smallholder fields planted below Tecoltepec, Coxcatlán, Cox., Puebla 87 Figure 2.42 Maguey (Agave spp .) amid a milpa on the dirt road to Ocotlamanic, Coxcatlán, Cox., Puebla 88 Figure 2.43 Small bull standing between milpa of mixed maize above and a planting of oats for fodder below, Xacalco, Cox., Puebla 88 Figure 2.44 Roadways in Coxcatlán municipio 91 Figure 2.45 Bridge across the Barranca Soyolapa in February and April 2011 100 CHAPTER 3 Figure 3.1 Mesoamerican house lot model showing discard and use areas 106 Figure 3.2 Municipal solid waste generation by state (entidad federativa) 2010 111 Figure 3.3 Per capita municipal solid waste generation by class, 2010 112 Figure 3.4 Garbage management practice by household in Mexico and the State of Puebla, 2010 113 Figure 3.5 Final disposition of collected garbage in Mexico (1,000s metric tons), 2000 -2010 114 Figure 3.6 Per capita municipal solid waste generation by class in Coxcatlán, Cox., Pue., 2010 116 Figure 3.7 Prioritization of management in LGPGIR 121 Figure 3.8 Municipal government structure and garbage management 124 Figure 3.9 Garbage collection truck in action 126 Figure 3.10 Recycle collection truck parked outside the ayuntamiento 126 Figure 3.11 Garbage dump south of the cabecera municipal 127 Figure 3.12 Shuttered and closed transferral station 127 Figure 3.13 Putting garbage out for collection 128 Figure 3.14 Service provision in the study area 130 Figure 3.15 Old Pepsi bottle being reused as a container for pulque 134 Figure 3.16 Buckets being reused as planters 134 Figure 3.17 Local recyclables collector pile 136 Figure 3.18 Fierro viejo collector truck 136 Figure 3.19 Small dumpsite in the Barranca Soyolapa 138 Figure 3.20 Larger tiradero clandestino along the highway 139 Figure 3.21 Large burn pit 141 Figure 3.22 Burn pile outside a home 141 Figure 3.23 A bucket of compost before its emptying in a pit 142 Figure 3.24 Organic refuse pit 142 Figure 3.25 The garbage trajectory in the study area 144 Figure 3.26 Pepenadores picking at the municipal dump 146 CHAPTER 4 Figure 4.1 Garbage census participant households by community 151 Figure 4.2 Native status of participants 159 Figure 4.3 Burn pile in a barranca 178 x vii Figure 4.4 Medium-sized tiradero clandestino 181 Figure 4.5 Two garbage management models 182 Figure 4.6 Smoldering garbage piles at the Coxcatlán municipal dump 184 CHAPTER 5 Figure 5.1 Half of the 16-class Q-sample 194 Figure 5.2 Half of the 16-class Q-sample 195 Figure 5.3 Half of the 28-class Q-sample 196 Figure 5.4 Half of the 28-class Q-sample 197 Figure 5.5 Q-Study participants by colonia 198 Figure 5.6 Example of a performed household generation Q-sort 200 Figure 5.7 Example of a performed garbage management sort 200 Figure 5.8 Forced-normal distribution for the household generation q-sort 201 Figure 5.9 Forced-normal distribution for the garbage management q-sort 201 Figure 5.10 Ideal sort for the “Mostly Foodies” factor 206 Figure 5.11 Ideal sort for the “Packaged” factor 207 Figure 5.12 Ideal sort for the “Homebodies” factor 208 Figure 5.13 Ideal sort for the “Sanitary Citizen” factor 212 Figure 5.14 Ideal sort for the “Home Recyclers” factor 213 Figure 5.15 Ideal sort for the “Convenience Independents” factor 215 CHAPTER 6 Figure 6.2 Participant households by settlement and spatial divisions used in statistical analyses 231 Figure 6.3 PRM “map” of issues by incidence (x-axis) and importance (y-axis) for the entire 433 participants 233 Figure 6.4 PRM “map” of issues by incidence (x-axis) and importance (y-axis) divided by valley and sierra households 234 Figure 6.5 PRM “map” of issues by incidence (x-axis) and importance (y-axis) divided by cabecera households and households from other valley settlements 239 Figure 6.6 PRM “map” of issues by incidence (x-axis) and importance (y-axis) divided by x viii centro households and households from other colonias in the cabecera municipal 240 Figure 6.7 PRM “map” of issues by incidence (x-axis) and importance (y-axis) divided by women and men 241 CHAPTER 7 Figure 7.1 Litter alongside the highway in Coxcatlán 255 Figure 7.2 Litter in an irrigation canal in Calipan 255 Figure 7.3 Litter cleanup by an Oportunidades-organized crew in the cabecera municipal of Zoquitlán, Puebla 258 Figure 7.4 Posting notifying the illegality of littering along the Barranca Soyolapa in the cabecera municipal 258 Figure 7.5 Diagram of a typical 250 square meter site 263 Figure 7.6 Looking south on the INEGI Marker site 263 Figure 7.7 Map of study sites along the highways 264 Figure 7.8 Initial clean total weights by study site 268 Figure 7.9 First collection abundance and weights across study sites 270 Figure 7.10 First collection normalized abundance by study site 271 Figure 7.11 Second collection abundance (inner ring) and weights (outer ring) across study sites 272 Figure 7.12 Second collection normalized abundance by study site 273 Figure 7.13 Both collection abundance and weights across study sites 275 Figure 7.14 Both collections normalized abundance by study site 276 Figure 7.15 Roadside typology and the sites associated with each type 283 APPENDIX 1 Figure A1.1 Garbage dump on the outskirts of Ayutla Mixes, Oaxaca 340 Figure A1.2 Pico Orizaba or Citlaltépetl from the northern Tehuacán Valley near San José Ixtapa 342 Figure A1.3 Another look at Pico Orizaba or Citlaltépetl from the northern Tehuacán Valley near San José Ixtapa during the rainy season 342 xix Figure A1.4 Intermunicipal sanitary landfill outside Ajalpan, Puebla 343 Figure A1.5 Looking east from Chilac across the southern Tehuacán Valley toward Calipan, Coxcatlán, and Sierra Negra 345 Figure A1.6 The south side of the park in Coxcatlán’s centro 345 Figure A1.7 Arriving in the early morning to San Luis Potosí 346 Figure A1.8 A political sign for the winning PAN candidate for state governor 348 Figure A1.9 View from the author’s residence in Coxcatlán 350 Figure A1.10 Running errands in Colonia Benito Juárez 352 Figure A1.11 The author’s appendix in a jar of formaldehyde 356 Figure A1.12 Going back to the Mexico City airport by bus for recovery 356 Figure A1.13 The view from my room and mosquito pabellón 360 Figure A1.14 The rodeo for the 20 de noviembre celebration 361 APPENDIX 2 Figure A2.1 Tehuacán Valley Project excavation crew at the Cueva del Maíz 363 Figure A2.2 Cueva de Maíz 365 Figure A2.3 The ancient Presa Purrón or Mequitongo 367 Figure A2.4 Coxcatlán Viejo or Sansuanchi, the site of the postclassic cacicazgo situated at the eastern end of the cabecera municipal 369 Figure A2.5 The ruins of Hacienda La Soledad 373 Figure A2.6 The Serdán family house. Today it is the Museo de la Revolución Casa de los Hermanos Serdán in Puebla City, Puebla 376 Figure A2.7 Ruined Revolutionary-era building in Coxcatlán town 376 xx List of Tables CHAPTER 2 Table 2.1 Ejidos in Coxcatl án, Puebla 55 Table 2.2 Toponyms in Coxcatlán, Puebla 57 Table 2.3 Population by settlement, 1990-2010, Coxcatlan, Puebla 64 Table 2.4 Indigenous language speakers, religious affliliation, and educational attainment by settlement 75 Table 2.5 Households size, floor type, electricity, piped water, television, and car ownership by settlement 76 Table 2.6 Transportation rates by Mexican pesos and US dollars. Prices and exchange rates as of Dec. 2011 92 Table 2.7 Prices for utilities and construction materials in Coxcatlán. Prices and exchange rates as of Dec. 2011 94 Table 2.8 Prices for basic foodstuffs in Coxcatlán. Prices and exchange rates as of Dec. 2011 95 Table 2.9 Prices for household goods and non-essentials in Coxcatlán. Prices and exchange rates as of Dec. 2011 96 CHAPTER 3 Table 3.1 Legal responsibilities of the state and municipio under the LGPGIR 122 Table 3.2 Estimates of garbage generated and potentially collected and uncollected in the study area. Only the valley settlements have collection. 132 CHAPTER 4 Table 4.1 Garbage census participant households as apercentage of total households in a community community (n = 520) 152 Table 4.2 Sex, head of household status, age, and educational attainment by participant 155 Table 4.3 Occupation of census participant 158 Table 4.4 Indigenous language speakers by language 161 xxi Table 4.5 Household size by total, adult, and minor residents. An adult is anyone over 18 years of age 162 Table 4.6 House building type by material and form 164 Table 4.7 Farming and gardening frequency by household 165 Table 4.8 Frequency of collection service by town 167 Table 4.9 Use frequency of collection service by town 167 Table 4.10 Recycling, sale of recyclables, and reuse by town 170 Table 4.11 Traditional garbage management practices by town 172 CHAPTER 5 Table 5.1 Factor correlation and characteristics for garbage generation sort 204 Table 5.2 Z-scores and ranks by factor for the garbage generation sort 205 Table 5.3 Factor correlation and characteristics for garbage generation sort 210 Table 5.4 Z-scores and ranks by factor for the garbage management sort 211 CHAPTER 6 Table 6.1 Surveyed households as a percentage of the total by settlement 226 Table 6.2 Incidence and importance by issue from all participants, and the valley and sierra subsamples. 235 Table 6.3 Incidence and importance by issue from all participants, and the cabecera and rest of valley subsamples. 236 Table 6.4 Incidence and importance by issue from all participants, and the centro and colo- nia subsamples from the cabecera municipal. 237 Table 6.5 Incidence and importance by issue from all participants, and the men and women subsamples. 238 Table 6.6 Results from the binary Chi-Square tests 248 CHAPTER 7 Table 7.1 Litter classes and subclasses used in this study 266 Table 7.2 Initial clean results by study sites 267 xxii Table 7.3 Mann-Whitney U test results by ecological indeces between urban-rural and high way stratifications 280 Table 7.4 Mean abundance by sites and litter classes between collections 280 Table 7.5 Study sites’ litter abundance and weight and the municipio-wide estimates for abundance and weight 281 Table 7.6 Roadside litter composition by percent weight in Coxcatlán compared to other lit ter studies and MSW estimates in Coxcatlán and Mexico 285 Table 7.7 Roadside litter composition by abundance in Coxcatlán compared to Muñoz et al. 2009 286 xxiii xxiv Glossary of select terms Abarrote : A small store, usually in a private residence, that sells convenience items, soft drinks, and snacks. Agencia de policia : A branch of the municipal government in settlements outside of the cabecera municipal. The presidente municipal typically appoints the officials who hold these offices. Aguador : The manager of water resources in a community or social property. In Coxcatlán there are aguadores for residential water usage by colonia and there are also aguadores who control irrigation districts within the ejido ’s irrigated sugarcane lands. Ama de casa : Housewife. This is the most commonly mentioned occupation by survey partici- pants in this study. The occupational bounds of this vocation can often go beyond typical domes- tic activities and can involve running an abarrote or some other home-based business. Apantle : A small irrigation canal that typically connects irrigated fields to a larger source canal. Article 27 of the 1917 Mexican Constitution : Article 27 defines the territoriality of Mexico and laws concerning land tenure. This article was the foundation law for the establishment of social properties known as núcleos agrarios or ejidos. Changes to this Article in 1992 removed the inalienability of these communal landholdings. Article 115 of the 1917 Mexican Constitution : Article 115 establishes political hierarchies and roles of federal, state, and municipal governments. Relevant to garbage management, this article lists garbage management under the responsibilities of the municipio. Changes to Article 115 in 1982 gave greater de jure autonomy as well as responsibility to municipios. Ayuntamiento : The joint municipal government of a Mexican municipio. The ayuntamiento is comprised of the presidente municipal with a treasurer, syndicate, advisors, secretary, and a cabi- net of regents. xx v Barranca : A heavily incised stream. Many of the waterways in Coxcatlán’s valley portion are referred to as barrancas. Cabecera municipal : The seat of government in a Mexican municipio. This is often the larg- est settlement in the municipio and its name is often the same as the municipio as in Coxcatlán, Coxcatlán, Puebla, Mexico. Cacicazgo (archaeological) : A small sub-state polity headed by a cacique or chief that was char- acteristic of the southern Tehuacán Valley during the Postclassic Period (AD 900 – AD 1520). The abandoned settlement of Sansuanchi (Coxcatlán Viejo) was likely one such cacicazgo . Calipeño : A resident or native of Calipan, Coxcatlán, Puebla. Campesino : A farmer, field hand, or rural laborer. This is most often mentioned occupation by male survey participants in this study. Like the ama de casa occupation, off-farm and extra- household work often supplements strictly agricultural work. Additionally, many men whose income comes from outside of agriculture identify as campesinos if they farm a small piece of land or hail from a rural locale. Carretera : A highway above the size of a street that connects towns. These can be paved or un- paved. CDI : Comisión Nacional para el Desarollo de los Pueblos Indígenas or National Commission for the Development of Indigenous Peoples. Established in 2003 from earlier less-centralized offices, this federal commission coordinates development programs and projects for indigenous commu- nities across Mexico. Centro : The downtown area of an urban area. In the cabecera of Coxcatlán the centro is the his- torical center, has the highest rents, and has the most urbanized landscape. Chalan : Laborer. With regard to garbage collection, the chalan rides on the back of the truck, picks up and dumps the garbage in the truck, and operates the compactor. Chofer : Driver. With regard to garbage collection, the chofer drives the truck and is the de facto leader of the sanitation crew. Choferes have prior experience in heavy truck operation and are thusly paid more than chalanes for their work. Colonia : A sub-settlement division similar to a neighborhood or barrio. Colonias represent the expansion of a town and are often less formal in terms of land tenure, infrastructure, and gover- nance. Combi : A small collective bus used for public transportation. Comisariado : The executive governing body of an ejido. The comisariado can refer to the entire executive committee consisting of a president, secretary, and treasurer or it can often just mean the president who is often referred to as the comisariado or comi. Corregimiento : A co-regency or form of early Spanish colonial governance. In contrast to the more common and more abusive encomienda, corregimientos were administered directly by a co-regent under the viceroy in Mexico City. CONAGUA : Comisión Nacional del Agua or National Water Commission. In Mexico, all water resources are property of the federal government and the CONAGUA is the oversight commis- sion on water rights, water usage, and data collection. Coxcateco : A resident or person from Coxcatlán’s cabecera municipal or someone from the muni- cipio of Coxcatlán. The former definition is a more common usage. Cristiano : A non-Catholic, Protestant Christian. While Catholicism is the predominant religious identification in Coxcatlán, different Cristiano denominations are seeing increasing member- ships. xx vi Cumbre : Ridgeline. The cumbre atop Coxcatlán’s sierra defines the municipio’s eastern bound- ary. Ejido : A communally-managed social property in Mexico. The ejido is the dominant form of land tenure in Mexico and in Coxcatlán. In Coxcatlán’s valley portion, the most productive ejido lands are irrigated and produce sugarcane. Ejido members are known as ejidatarios. Encomienda : The encomienda system was an early Spanish colonial practice of controlling indigenous labor and territory in the New World. Under encomienda, indigenous communities’ labor and lands were “entrusted” to a colonist who could extract labor and tribute from the popu- lations under their jurisdiction. Estado or Entidad Federativa : The level of government beneath the federal government as defined in Articles 115-122 of the Mexican Constitution. In Mexico, there are 31 estados (states) and 1 Distrito Federal (federal district) which are generally classified as entidades federativas (federative entities). Faena: The faena is communal labor among members of a community or social property. The faena is either conducted by the members themselves or is substituted by paid labor. Fierro Viejo : A collective term for small-scale private scrap collectors. The fierro viejo collectors drive through communities in old pickup trucks offering their purchasing prices from loudspeak- ers. These collectors buy metal from households which they sell to larger recyclers in nearby urban areas. Hacienda : A Spanish colonial landed estate. The hacienda system remained in place after the end of encomienda and was the dominant form of land ownership in Mexico prior to the Revolu- tion in the early 20th Century. The southern Tehuacán Valley’s haciendas were primarily dedi- cated to the production of sugarcane. xx vii House lot : The lot on which a household resides. House lots, also known as solares, dooryards, and other names are the sites of numerous domestic activities of a household such as gardens, small-manufacture, recreation, and relevant to this study, garbage management. INEGI : The Instituto Nacional de Estadística, Geografía, e Informática or National Institute of Statistics, Geography, and Informatics. INEGI is a federal bureau that is charged with conducting censuses as well as collecting and distributing geographic information. Ingenio : A sugar refinery. There is an ingenio in Calipan that buys as well as finances much of the sugarcane produced in the area. Ley Federal sobre Metrología y Normalización : This law outlines Mexico’s regulatory frame- work in matters of industry and commerce. Relevant to garbage management, the Federal Law about Metrology and Normalization sets standards for how garbage is disposed and how its quantities are measured through standards known as Normas Oficiales Mexicanas (NOM) or Of- ficial Mexican Norms. Ley General de Equilibrio Ecológico y Protección al Ambiente (LGEEPA) : Passed in 1988, the General Law of Ecological Equilibrium and Environmental Protection is the general founda- tion of environmental law in Mexico. Before the LGPGIR , the LGEEPA was the primary law that governed garbage management policy at the federal level. Ley General para la Prevención y Gestión Integral de Residuos (LGPGIR) : This is the cur - rent federal law, passed in 2003, outlining the management of all wastes, including garbage, and the responsibilities of all actors, from all levels of government to the private sector, involved in managing waste. Matorral crasicaule : An open woodland vegetation assemblage characterized by Acacia and Mimosa species, mesquite (Prosopis luisana), shrubbery such as Lantana camara and Mexican oregano (Lippia graveolens ) among many others, all punctuated by cacti species like Opuntia species and many smaller cacti. The large cardón and xoconostli cacti comprise some of the xx viii larger flora in this association. Maquila(dora) : An export-oriented factory in Mexico whose inputs and outputs are generally tariff-free. In the southern Tehuacán Valley, garment and especially blue jeans sewing plants are common, low-wage jobs. Milpa : The definition of milpa varies according to context. Milpa means “field” in Hispanicized Nahuatl and generally refers to a set of field crops or the field on which they are cultivated. In the Sierra Negra in Coxcatlán municipio, milpa usually refers to maize, beans, fava beans, some oats, and other field crops. MSW : Municipal Solid Waste. In Mexico, this designation applies to any non-hazardous, non- industrial waste produced by homes and businesses. Municipio : The political entity below the state or entidad federative in Mexico whose powers and responsibilities are described in Article 115 of the Mexican Constitution. The municipio of Coxcatlán, Puebla is the study area for this study. Related to garbage management, the municipio is charged with the practical, day-to-day governance of garbage. NAFTA : North American Free Trade Agreement. Known as the Tratado de Comercio Libre in Mexico, NAFTA is an international free trade agreement between Canada, the United States, and Mexico. Nahuatl : An Uto-Aztecan language that is the most widely-spoken indigenous language in the state of Puebla with nearly a half million speakers in 2010. In Coxcatlán municipio’s sierra com- munities, Nahuatl is the predominant language there. Neoliberalism : A political economic ideology that favors free markets, deregulation, and priva - tization. Neoliberalism intends a diminishment of the public sector and a decoupling of govern- ment from social services. In Mexico and across the world, neoliberalism began to replace the heavy state-client system in Mexico following economic crises in the 1980s and 1990s. xxix Normas Oficiales Mexicanas (NOM): Official Mexican Norms or standards. The NOMs are standardized measures or procedures for industrial, commercial, or governmental practices. Related to garbage, a set of NOMs regulate collection, disposal, and measurement of garbage generation and classification. OECD : The Organisation for Economic Co-operation and Development is an international or- ganization whose goal is to promote free-market economic development and liberal democracy across the world. There are 34 member states, including the United States and Mexico. OOSELITE : Organismo Operador de Servicio de Limpia de Tehuacán is the parastatal organism that the Municipio of Tehuacán charges with managing garbage collection and the operation of the sanitary landfill. Oportunidades : A Mexican federal social welfare, poverty alleviation program administered by SEDESOL that ties cash transfers with education, nutrition, and domestic workshops. Relevant to garbage management, Oportunidades (Opportunities) agents host mandatory home economics classes on household sanitation as well as mobilize recipients into occasional litter clean-ups. PAN : Partido Acción Nacional or National Action Party. The PAN ’s is a right-of-center political party across Mexico. Prior to the recent presidential victory by the PRI in 2012, the PAN had held the two previous national presidencies that broke the 76-year domination political domination of the PRI . In Coxcatlán, the current Presidente Municipal and his administration is affiliated with the PAN and the current state governor is also Panista (PAN affiliated). Pastizal inducido : A grassland vegetation assemblage that results from human activity. Pastiza - les inducidos occurs in areas so often disturbed that a grassy and scrub vegetation of Poacae, Lantana spp. and Lippia graveolens with short small, understory cacti persists. Pepenador : A waste picker who makes a living off of scavenging recyclables out of garbage at the dump. Pepenadores are integral to garbage management but less-than formally employed in many Latin American and Global South cities. In Coxcatlán, there is occasional and opportu- xxx nistic scavenging at the municipal dump, typically by Fierro Viejo collectors as they can carry recyclables to a regional recycler. PET : Polyethylene terephthalate. Also known as #1 plastic, PET is used in the manufacture of cheap plastic bottles and food and beverage receptacles. In Coxcatlán, PET , along with card- board is collected by the municipal government for recycling. Porfiriato: The period in Mexican history (1876-1910) where the presidency was dominated or occupied by Porfirio Díaz. The Porfiriato was a time characterized by a Western, European- ized model of development from immense foreign investment and increasing marginalization of Mexico’s lower and working classes. Díaz maintained his grip on Mexico through strict, institu- tionalized political hierarchies and a willingness to crush any challenge to his rule. His continued involvement in the Mexican presidency was one of the factors that precipitated the Mexican Revolution. PRD : Partido de la Revolución Democrática or Party of the Democratic Revolution. The PRD is a social democratic, left-of-left-of-center party that has seen growing popularity in the past 20 years. The PRD is strong in the Federal District (Mexico City) and southern states. PRI : Partido Revolucionario Institucional or Institutional Revolutionary Party. The PRI is slight- ly left-of-center and was the dominant political party of Mexican national politics from after the Mexican Revolution (1910-1918) to the end of the 20th Century. The PRI and its predecessors held the presidency from 1924 until 2000 when PAN candidate Vicente Fox won. After 12 years, the PRI recently won back the presidency in 2012. The PRI also dominated Coxcateco politics until the 2010 election when the PAN candidate won the municipal presidency. PROCEDE : Programa de Certificación de Derechos Ejidales y Titulación de Solares or Program for Certification of Ejidal Rights and Titling of House Lots. The goal of PROCEDE was to cer- tify and title social properties across Mexico. xxxi PROFEPA : Procuraduría Federal de Protección del Ambiente or Federal Attorney General for Environmental Protection. This office is the legal enforcement for environmental law in Mexico. Programa Municipal para la Prevención y Gestión Integral de Residuos Sólidos Urbanos (PMP-GIRSU) : The PMP-GIRSU or Municipal Plan for the Prevention and Integral Manage- ment of Urban Solid Wastes is the LGEEPA -required municipal garbage management plan. The PMP-GIRSU is both a document and a practical management plan that diagnoses the magnitude of garbage and prescribes a best practices plan for managing it. The PMP-GIRSUs are typically created with the assistance of the state government. Pulque : A lightly-alcoholic, slightly-effervescent beverage made from fermented aguamiel or liquid extracted from a decormed agave plant. Pulque is a traditional beverage for Coxcatlán’s sierra communities and is often consumed by laborers working in their fields. RAN : Registro Agrario Nacional or National Agrarian Registry. The RAN is in charge of curat- ing and administering documents related to land tenure for the agrarian sector. Regidor(a) : Municipal regent in charge of some function. The regidores are appointed by the presidente municipal and are collectively known as the cabildo de regidores. In Coxcatlán the office of the regidor de ecología is in charge of garbage collection service provision. SEDESOL : Secretaría de Desarollo Social or Secretariat (Ministry) of Social Development. The SEDESOL is involved in poverty alleviation programs, most notably Oportunidades , as well as the urban land tenure formalization program CORETT . Selva baja caducifolia : A type of cactus, thorn forest vegetation whose canopy averages 7 -15 meters and loses 75 percent of its leaves during the dry season. The lower slopes of the Sierra Negra and hills of the valley portion of the municipio are typically covered by selva baja caduci - folia or SBC. xxxii SEMARNAT : Secretaría del Medio Ambiente y Recursos Naturales or Secretariat (Ministry) of the Environment and Natural Resources. SEMARNAT is the federal environmental protection office. Relevant to waste, SEMARNAT is charged with oversight of hazardous waste manage- ment and public education on garbage management. Sequía : The dry season that lasts from mid-October to mid-May. Almost no precipitation falls during the middle portion of the sequía . This dry season is the result of the seasonal southern mi- gration of the subtropical high pressure system combined with the municipio’s placement within a rain shadow from Sierra Negra. Sexenio : The standard six-year term of the Mexican federal presidency and state governors. In Mexico, there is no possibility for reelection. SMRN : Secretaría del Medio Ambiente y Recursos Naturales or Secretariat (Ministry) of the En- vironment and Natural Resources. This was the State of Puebla’s ministry in charge of garbage management until it changed into the SSAyOT in 2011. Solar: The lot on which a family in rural Mexico lives. Traditionally, the solar includes the house, yard, garden area, and relevant to garbage management, a small dumping zone. This term is largely synonymous with house lot . SSAyOT : Secretaría del Sustentabilidad Ambiental y Ordenamiento Territorial or Secretariat (Ministry) of Environmental Sustainability and Land Management. The SSAyOT is the current state office in Puebla in charge of performing its role stated in the LGPGIR. Tiradero clandestino : An informal, illegal dumpsite. In the study area, tiraderos clandestinos can vary from a meter in diameter to about 200 square meters and are found on roadsides, barrancas, and in accessible parts of wooded areas. Toft area : The area just outside the house and in the main activity area of a house lot. The toft area is where refuse accumulated in Mesaomerican village house lots. xxxiii Trecenio : The standard three year term of a municipal presidency. At the end of a trecenio, the majority if not all of the municipio’s governmental staff are replaced. UN : United Nations. Relevant to garbage, the UN’s Environment Programme collects global data on garbage generation and characterization. Usos y costumbres : Indigenous and campesino customary laws. Usos y costumbres are not offi- cially recognized in Puebla as they are in Oaxaca but they govern small civic matters, the hosting of public events, sanitation, and public works. Zafra : The sugarcane harvest. In Coxcatlán, the zafra begins in early December and lasts until early February. xxxiv CHAPTER 1: Garbage, environment, and society in Coxcatlán, Puebla, Mexico INTRODUCTION The production and disposal or destruction of waste is one of the most basic and intrinsic human ecologies (Peet et al. 2011: 30) and today garbage generation and disposing of it remain two critical environmental issues across the globe. Increasing garbage generation linked to the inescapable coupling of consumption to economic growth makes them a particular preoccupation for development agencies such as the United Nations and the Organization for Economic Co- operation and Development (OECD) as well as governments across the globe (Yepes 1990; UN 2009; OECD 2012) (Figure 1.1). This is especially true in Mexico, where garbage generation has increased 36 percent since 2000 with 47.8 million metric tons produced in 2011 (SEMARNAT 2012). Yet framing how this topic is a major environmental issue is largely left imprecise and general, pointing vaguely to garbage’s health, environmental, and aesthetic hazards along with Malthusian fears of increasing consumption and garbage generation. Thus, the issues of genera- tion and disposal take precedence around most environmental debates about garbage. Addition- ally, most of the research on the perceived garbage crisis has a significant urban bias, under- standably as cities are sites of increasing consumption amid more intensely commodified space. Without presuming too greatly an urban-rural dichotomy, there is notable lack of discussion of garbage management in the less connected, more economically marginal places that straddle the urban - rural divide. To address this lack, this dissertation describes the realities and perceptions of garbage and its management in the rural municipio of Coxcatlán, Puebla, in south-central Mexico. Con- ceptually, this dissertation contends that these issues are contingent on local and regional con- texts. Garbage, understood here as anything non-hazardous, non-industrial managed as waste, is matter that everyone and every household generates (UN 2009; Moore 2011: 134). How this matter is managed differs according to unequal access to infrastructure, traditional practices, and various forms of governance. Also, what constitutes garbage and how it is best managed is contingent upon individual perception, opinion, and various environmental discourses, creating contested spaces and conflicting issues of environmental governance (Barr 2002; Drackner 2005; Moore 2012). In the municipio of Coxcatlán, Puebla, about 20,000 people in 4,500 households (INEGI 2010) and 18 settlements covering nearly 250 square kilometers compost, burn, dump, 1 Fi gu re 1 .1 . M un ic ip al w as te g en er at io n by O EC D a nd s el ec t c ou nt rie s, 2 00 7. M ap b y A . H ilb ur n. D at a de riv ed b y ea ch c ou nt ry a nd c om pi le d by O EC D . 2 put out for municipal collection, recycle, litter, and feed to animals, among other practices, their garbage every week. Fittingly, in Coxcatlán, there are different perceptions of what garbage is and almost as many opinions on how it should be dealt with. These opinions, perspectives and discourses help to construct a number of environmental issues related to garbage management in Coxcatlán, in Mexico, and beyond. The overall goal of this dissertation is to provide an environ- mental-geographical perspective on garbage beyond the urban bias by describing how people in less-than-urban and rural places manage their garbage and how they perceive these practices as well as more broadly-defined garbage management as environmental issues. A note on definitions For the sake of clarity throughout this dissertation, I must precisely distinguish what I am calling “garbage” here. Garbage is any non-hazardous, non-industrial waste produced by house - holds and small commercial operations. It is the stuff we, as residents, individual citizens, and families, collectively throw out as the result of daily activities. Garbage in a more technical sense is conceptually synonymous with what is typically called “municipal solid waste” or “MSW” and “urban solid waste” in the managerial and academic literature. In Spanish, the terms used are basura for “garbage” and residuos sólidos urbanos or residuos sólidos municipales for “urban solid waste” and “municipal solid waste”. The word tlazolli , which can signify many other more specific filth and waste-related concepts, is the general term for garbage in Nahuatl, the primary indigenous language in Coxcatlán. In the following nine chapters, the word “garbage” will be used generally to refer to all of the aforementioned words and concepts. In some instances, “municipal solid waste,” “MSW,” “basura,” and “tlazolli” will be used in reference to specific contexts and citations but unless specified otherwise, will basically touch upon the meaning of “garbage” noted above. OBJECTIVES OF THE DISSERTATION The following dissertation achieves the following four research objectives: 1) This dissertation documents how household garbage is managed in a rural-but-urbaniz - ing Mexican municipio at a number of scales. Here, garbage is generated and managed by indi - viduals and households who use a variety of practices to collect and dispose of garbage. After it leaves the home, a number of actors, from municipal to state to federal governments and beyond, 3 govern what happens to that garbage, as well as sanction and prohibit certain practices to dis- pose of it. Using the theoretical lenses of cultural and political ecology to focus on data taken from extensive surveying and archival fieldwork, this project will describe in detail the cultural- political-garbage-environmental nexus in Coxcatlán, Puebla, Mexico. To address this objective, I conducted a household census from 520 households across the municipio to observe the different practices of household garbage management. This cultural-political ecological approach pro- vides a window into understanding the daily management of garbage in study area households as a set of practices that combine deep-rooted tradition with differing access to and use of more contemporary garbage infrastructure. This approach also better contextualizes the circumstances that produce more recent forms of consumption and garbage, as well as the official policies and practices to dispose of it. 2) This dissertation examines the individual and consensus group perceptions of gar- bage generation, garbage management practice, and garbage related issues in Coxcatlán, Puebla, Mexico. The definition of what constitutes garbage and the practices by which it is managed are socially-produced and are contingent on locally-constructed environmental knowledges. This approach assumes that garbage and its management are issues that are not homogeneously and unilaterally understood by all, but are conditional upon perception, opinion, and environmental discourses (Barr 2002; Drackner 2005; Moore 2012). These perceptions and opinions are con- tingent upon place, uneven access to infrastructure, and global environmental discourses as well as many other influences. Garbage, thus, is not an easily straightforward management concern of generation, disposal, and environmental impact, but is a political issue of environmental man- agement about control over the end point of consumption and the practices to deal with it. To address this issue, this research employs “risk mapping” to identify the heterogeneity of opinion regarding garbage across the municipio. This research also applies Q methodology to identify subjectivities regarding garbage generation and appropriate management practices of garbage. 3) This following dissertation project engages what is one of the most noticeable and often mentioned garbage-related issues in Coxcatlán and perhaps all over rural Mexico, litter- ing. Littering is at times synonymous with the garbage problem in the study area. Its capacity to contaminate landscapes, be it chemically, biologically, or aesthetically, makes it a pressing issue in the minds of residents there. Litter’s importance is also owed to its status as garbage-out- of-place and the potential political capacity to indicate poor governance, lack of education and 4 manners, and a lack of civic responsibility. Approaching litter empirically is remarkably difficult as the topic is very broad, suffers from a negative stigma, and is often cleaned up or re-deposited elsewhere. Finding adequate public space to conduct a study is also difficult. To address this ob- jective, I conducted a litter quantification and characterization study alongside Coxcatlán’s paved highways from June to November 2011. The primary goal of this objective is to derive an esti - mate of what type of and how much litter exists. This will allow the comparison of its perception as a major environmental issue to how much and what type of it exists materially. 4) Lastly, this dissertation provides a descriptive geography of people and place in a rural municipio in southern Mexico. In the study area, garbage is not generated and managed outside of a definition of place by a general homogeneous society divorced from history. Garbage man- agement practices and the construction of garbage as a set of environmental issues are place- contingent. Understanding Coxcatlán municipio and its surrounding area as an interconnected agglomeration of places produced by its diverse society through engagement with the nation and world abroad, situate these issues geographically. Descriptive regional geographies in the vein of the cultural-historical or “Berkeley School” tradition that are updated to include newer social theoretical perspectives can provide a useful model to bring together culture, history, the environment, political economy, and present information on settlements and society in non-urban Mexico. This research comes out of the participant observation field work I conducted during my stay in Coxcatlán from January-December 2011. In terms of its organization, this research com - ponent is inspired by works such as Oskar Schmieder’s (1930) The settlements of the Tzapotec and Mije Indians, state of Oaxaca, Mexico , Robert C. West’s (1948) Cultural Geography of the Modern Tarascan Area , Donald Brand’s (1951) Quiroga: A Mexican Municipio , and David Hill’s (1964) The Changing Landscape of a Mexican Municipio Villa Las Rosas, Chiapas . While the aforementioned are much more comprehensive studies than this single research objective, my general description of land and life in Coxcatlán will serve to better contextualize the more fo - cused research on garbage and its management but also provide an account of life in a municipio straddling rural-urban and local-global divisions. Synthesized together, these four objectives form a geographical narrative about how Cox - catlán municipio “takes out” its garbage while constructing and contending with environmental issues related to it. The rest of the following chapter will situate this dissertation project within its theoretical and disciplinary research themes. The second chapter offers a descriptive geogra- 5 phy of the municipio of Coxcatlán, Puebla. Following a section outline, Chapter 2 provides the general location, natural setting, contemporary settlements, society, and lifeways in Coxcatlán. The rest of the dissertation focuses on garbage management and litter in Coxcatlán. Chapter 3 outlines the garbage issue in Mexico and Coxcatlán from the past to today. The chap- ter begins by setting the historical precedent of garbage management from prehistory through the colonial era to the near present. Next, a description of the legal frameworks for governing garbage management help locate the current management situation in the municipio of Cox- catlán. Chapter 4 presents the results of the municipio-wide garbage census. Chapters 5 and 6 investigate how garbage and its management are framed as environmental issues by Coxcatlán municipio’s residents. Chapter 5 is a two-part Q-methodology study that identifies the subjectiv- ity of cabecera residents regarding their own garbage generation and how it should be managed,. Chapter 6 is a risk mapping study taken across a large, extensive sampling of residents across the municipio that identifies the heterogeneity of opinion about garbage-related issues. Chapter 7 outlines the issue of litter in Coxcatlán and includes a systematic study to quantify and classify it on Coxcatlán’s roadsides. The final chapter, a conclusion, offers a summary of the entire disserta- tion. METHODS Introduction Garbage is more than simply an issue concerning rising generation rates and increas - ingly inorganic contents with vaguely defined environmental and social impacts. Rather, garbage is a multi-faceted human-environmental issue. Studying these multiple facets in one place, like Coxcatlán, requires varied methodological approaches in order to provide an ample and precise understanding of these impacts. Garbage as matter is one thing. It is the material result of the process of consumption by people and is continually produced within some definition of the environment. Much of the literature on garbage generation and litter studies engage it from this material perspective. This matter also must be managed or dealt with in some manner by house- holds or individuals, making garbage a practical matter. The practices by which garbage is man- aged and dealt with are contingent upon differential access to formal infrastructure, tradition, and other sociodemographic variables, creating different geographies of garbage management. By utilizing participant observation and an extensive household garbage census, my research identi - 6 fies the material and practical aspects of garbage in Coxcatlán. Garbage and its management are also contested, perceptionally-dependent issues as well. The Q-method surveys and the participa- tory risk mapping data identify local constructions of garbage-related issues there. Litter, as one of these garbage-related issues, bridges the material and perceptional aspects of garbage. Results from the litter study provide an empirical basis to gauge potential incongruencies between litter’s material and perceived realities. These multiple methodologies interrogate the geographic place of garbage and its management as material, practical, and perceptional issues resulting in a re- gionally contextualized geography of garbage management in Coxcatlán. The following section offers descriptions of these methodological approaches. Household garbage census Gauging the material magnitude of garbage by households is a major preoccupation of garbage-oriented research. This focus is represented by a sizable and growing literature on gar - bage generation in Mexico and across the world (Buenrostro, Bocco, and Bernache 2001; Buen- rostro, Bocco, and Vence 2001; Buenrostro, Ojeda, and Marquez 2007; Buenrostro, Marquez, and Pinette 2008 among many others). Yet, how this garbage is managed after its generation is typically overlooked. Is the garbage dumped, burned, set out for collection, recycled, stored, or buried? This is an important question as the practice of garbage management represents the key engagement with the waste matter itself, the households and society that generate it, and the en- vironment. All of the practices mentioned beforehand transfer as well as transform garbage into the air, soil, water, and living spaces. The practices used by households varies according to place, accessibility, age, tradition, educational attainment, economic marginality, and sex among other factors. Thus, garbage management practice can be placed at the nexus of society and the envi- ronment, making it of critical importance to the broader scope of the dissertation. Currently there are numerous garbage generation studies across Mexico and even one conducted in Coxcatlán’s cabecera municipal (Sánchez-Méndez and Olmos-Torres 2009) but as of yet there are no studies on household garbage management in the region. To address this lack, I conducted a nearly- mu- nicipio-wide garbage management census to provide empirical evidence how garbage is managed there. The fieldwork for this research began in the last week of August and lasted past the first week of November 2011. The primary survey instrument was a 28-item questionnaire that ad - 7 dressed household demographics and garbage management practices. The research team consist- ed of myself and a paid local investigator with questionnaire surveying experience. In total, we conducted 520 distinct interviews with participants who each represented a different household. All of the interviews were conducted in Spanish door-to-door and in individual homes with the exception of 34 being conducted individually but during a health clinic seminar and outside of an elementary school. The data from the 520 questionnaires were summarized and entered into Excel spreadsheets where descriptive statistics were calculated. Further detail on the fieldwork is provided in Chapter 4. Q-method survey From March through May of 2011, I conducted a Q-method survey to describe and analyze the subjectivities of a sample of Coxcatlán’s population regarding two garbage-related themes, garbage generation and the practices of managing it. Q-method is a quantitative ap- proach to systematically measure human subjectivity. This unique method, whose use is grow- ing in human-environmental geography, can identify common perspectives, viewpoints, or discourses and then statistically measure how individuals relate to those views (Robbins 2000; Robbins and Krueger 2000; Eden et al. 2005; Danielson 2007; Hawthorne, Krygier, and Kwan 2008; Brannstrom 2011; Brannstrom, Jepson, and Persons 2011; Jepson, Brannstrom, and Per- sons 2012; Ward 2013). In this case, I try to unpack the subjectivities of a sample of Coxcatlán’s residents on the garbage generation issue and the practices of dealing with it. In Q-method surveys, a respondent is instructed to sort and rank-order a sample of state - ments regarding an issue along a forced, quasi-normalized distribution that represents their view - point, permitting the comparison of opinion within the sampled group (McKeown and Thomas 1988). Photographs have been used successfully in previous studies employing Q-method, as well (Hawthorne et. al. 2008; Green 2005; Fairweather and Swaffield 2000, 1995; Swaffield and Fairweather 1996). In this example, I defined the sample of statements, also called the Q- sample, as all garbage classes. I took photographs of different garbage classes defined by the state environmental ministry to construct the Q-sample. The representative photos were printed out on 3-inch by 5-inch cards for a 16-class q-sample and a 27-class q-sample. During the in- terviews, respondents arranged photo cards on one of two 42 by 36-inch professionally printed canvas templates, each with its own instructions and distributional scale. Following the sorting 8 of photos, referred to as a Q-sort, the respondents gave an explanation of how and why they constructed this response, in addition to responding to a short demographic survey. The Q-sorts were recorded by photographing the arranged cards on the canvas template. In all, I conducted 70 interviews with two Q-sorts per respondent. These respondents were sampled from the centro ( n = 37) and the peripheral colonias ( n = 28) of the cabecera municipal with five being residents of nearby towns. I interviewed 36 women and 34 men in a variety of contexts, ranging from an unused computer lab to certain participants’ homes. Following data collection, the results were entered into Excel spreadsheets for posterity. The data were then entered into PQMethod, a freeware software that is typically used in applied Q-method studies whereby the data were factored, flagged, and rotated to construct the factors used in later analysis. The results of this analysis along with a fuller description of Q-method, the research process, the results, and a discussion regarding the meaning of the factored subjec- tivities forms the majority of Chapter 5. The participatory risk mapping of garbage The Q-method study unpacks the subjectivities of a relatively small group of residents in the cabecera municipal of Coxcatlán municipio surrounding two key issues about garbage: generation and its ideal management. Yet garbage is much more expansive than its framing as a generation-management issue. The diversity with which garbage is constructed as a set of socio- environmental issues varies according to place and population characteristics. Thus, one compo- nent of my dissertation seeks to highlight how Coxcatlán’s residents from across the municipio perceive garbage as a set of socio-environmental issues. To do this I used and modified participa- tory risk mapping, a method used in development research to identify sources of risk (Smith et al. 2000; 2001; Baird 2009). Participatory risk mapping is not a mapping method in the traditional geographic sense. Rather, it “maps” the incidence and severity of perceived risks along a two-dimensional graph. In practice, it involves respondents listing perceived risks and then rank-ordering them by their assumed severity (Baird et al. 2009: 467). The data are then normalized and analyzed to produce indices of incidence (frequency of a risk being noted) and severity (the normalized rank-order of an issue). The main intent of this method is to highlight the heterogeneity and relative severity of perceived risks across a sample population. I modified the data collection process to expand 9 beyond the concept of risk and to apply it toward garbage. In place of risks and their relative severity, I had respondents list garbage related issues and then rank-order them according to their importance. These data then allow the calculation of incidence and “importance” values for particu - lar issues which can then be subsequently “mapped” or plotted on a two-dimensional coordinate grid. The incidence index indicates the proportion of respondents that listed a particular issue thus quantifying the breadth of a particular issue across the sample or subsample of the study popula- tion. Incidence is measured from 0 (no mention) to 1 (always mentioned). The“importance” or “severity” index, as it is listed in other uses of participatory risk mapping, involves more complex calculations than the incidence index because of the variable number of issues listed by respon- dent, which require the creation of uniform intervals. Ultimately each issue receives a mean “im - portance” value from 0 (no importance) to 1 (always ranked highest). The mean of an identified issue’s “importance” index and its incidence value can then be “mapped” with incidence values on a two-dimensional grid plot. These graphs or “maps” can then be used in discussions regarding the issues. Pearson’s Chi-Square statistic was used to determine relationships among spatial and de - mographic variables on the incidence of particular issues being mentioned. I also applied logistic regression to relate any effect of age and educational attainment on the odds of mentioning the ten most frequently listed issues. The data were analyzed in Excel and the statistical testing was conducted in SPSS. The fieldwork for this research was conducted in conjunction with the garbage census mentioned above from August to November 2011. The participatory risk mapping exercise require additional time beyond the garbage census and some respondents did not have the time or did not wish to participate. Nevertheless, I conducted 433 interviews in twelve communities across the municipio. Further description of the method along with results and discussion form the majority of Chapter 6. Studying roadside litter The majority of this dissertation examines garbage and its management. It presumes that garbage is both a material and perceptionally-contingent human environmental issue that is constructed and governed by diverse actors living in different circumstances with differential 10 access to resources and power. Yet, what about garbage that goes unmanaged? Litter is unman - aged garbage and it is an important and present garbage-related issue due to its abject status, its potential to contaminate landscapes, and its previously noted political capacity that can indicate poor governance, incomplete education and manners, and lacking of civic pride. By comparing the material reality of litter with its perceptions, an empirical litter study that gauges the quantity and character of litter represents the intersection of garbage’s perceptional and material construc - tions. However, gauging litter generation empirically can be difficult as litter imparts a negative stigma to where it remains. Also, public places where it can accumulate unaffected by external factors are rare. Thus, I selected roadsides along Coxcatlán’s 69 kilometers of highway as my study site as these are largely anonymous public spaces where littering is an observable problem. To bring about my research goals, I modified methods used by U.S. state departments of transportation and engineering firms to quantify and classify highway litter. The methods involve the selection of random sites in an area of interest, removing all litter within them, and then returning at regular intervals to classify and measure the deposited litter. In June 2011, using GIS and field observations, I selected ten 5 by 50 meter sites and cleaned them. Later, in August and November, I made two collections at 10-week intervals. All collected litter was sorted in accor - dance to predetermined litter categories, where it was counted and weighed. The abundance (count) and weight data were then area normalized by 10 square meters in order to compare the relative littered-ness of Coxcatlán to other case studies in the United States. These results highlight the power of casual perception relative to empirical data regarding an en- vironmental issue like garbage or litter. The per area estimates also allow for absolute estimates across the municipio. Lastly, I conducted differences of means tests on litter abundance between sites classed as “urban” and “rural” as well as between sites on the two paved highways in order to infer if any locational attributes related to litter accumulation exist. I also extended this analy - sis to examine potential differences between collections to identify any seasonal effect on litter- ing. Participant observation, hanging out, and gathering information Throughout my stay in Coxcatlán and surrounding areas in 2011, I conducted other research that greatly informed this dissertation but does not fit into one of the aforementioned 11 methods. One could call some of this participant observation. By that, I observed and partici- pated in household garbage management in the homes where I lived and worked. Just by living in Coxcatlán as well as taking door-to-door formal surveys, I watched and was impressed by how households use various practices besides placing their garbage out for collection or just burning it. I rode along on a few occasions with the municipio’s sanitation crew, seeing how they kept to a schedule and a route, as well as watching how people interact with them. I spent time at the dump, watching pepenadores (waste pickers) pick through garbage, and I took walks out into the monte (woods) around town to look at and count the tiraderos clandestinos or informal, illegal dumps. I also spent time with people, talked with them about my research, and what I was trying to do there. This usually had the effect of forcing me to consider what I was actually studying but it also allowed me to find out the believability of information I received or how officially stated policies concerning garbage are brought about practically to everyday residents. More formally, I spoke with officials at government offices in the state capital, in Tehu- acán, in the cabecera municipal of Coxcatlán, as well as its outlying settlements. I submitted questions and information requests from the State of Puebla’s Secretariat of Environmental Sus - tainability and Land Use (SSAyOT) over the course of my stay to better understand the state gov - ernment’s role in garbage management. I consulted with and purchased data from the state office of INEGI (Instituto Nacional de Estadística, Geografía e Informática), the federal census and geographic institute, to get the best population, geospatial, and environmental data. I met with officials from the federal Tehuacán-Cuicatlán Biosphere Reserve, both in their Tehuacán-based office and in the field. In Coxcatlán, the municipal president and his staff were typically available for information regarding the daily practice of garbage management. The local officials in the towns and small settlements across the municipio provided me with insight on how customary rules ( usos y costumbres ) govern sanitation and consequently, garbage. These experiences allow me to validate and verify my own conclusions taken from the more formal, empirical methodolo- gies. Lastly, the ample time I was able to spend in Coxcatlán gave me the opportunity to un - derstand how people live and make a living in a rural but urbanizing Mexican municipio. I lived with families. I made friends. I helped out with household chores. I went to work with people. I bought things and rode or walked nearly every length of road in the municipio. I learned about the local politics. I came to appreciate the varied cultural landscapes, from the towns, sugar - 12 cane fields, and arid scrub forests of the valley to indigenous communities and their milpas set out about the pine-and-oak covered slopes of the adjacent sierra. In short, I spent a year learn - ing about how people, places, and the environment of Coxcatlán come together to create what esteemed Latin Americanist geographer Robert West (1993) calls, the “personality” of a place. Chapter 2 draws heavily on observations and field notes from this aspect of my inquiry. A CULTURAL AND POLITICAL ECOLOGY OF GARBAGE Introduction The theoretical connection of this dissertation to geography as a discipline centers on and contributes to cultural and political ecology. It also adds to general geographical and social science research on waste and garbage. First, this dissertation engages the field of geography from the human-environmental approach through cultural and political ecology. Consumption, for necessity or leisure, inevitably leads to undesired matter or waste. Producing and dealing with waste is a fundamental human-environmental practice (Peet et al. 2011: 30) and this dissertation is at its core a narrative of a society and the waste it generates and how this matter is managed within some definition of the environment. Garbage is the waste of individuals, families, small- scale commerce, and municipalities. This research frames the daily waste management activities of these actors, both practically and discursively, in the environment. Second, this dissertation’s examination of accessibility to garbage management from the state and the effects, both material and as perceived, contributes to the larger body of geographical and social science research on waste around the themes of environmental governance and environmental justice. By investigat- ing how individual households as well as other non-governmental actors manage garbage, this dissertation draws upon and adds to the informal waste sector and waste as a commodity litera- ture. Lastly, this dissertation expands its analysis beyond notions of managed garbage to inves- tigate unmanaged garbage or litter, which is a well-worn subject in academic circles, but rarely investigated outside large urban areas and tourist destinations (Silva-Iñiguez and Foster 2003; Rodrigues-Santos et al. 2005; Muñoz-Cadena et al. 2009). A brief history of the human-environmental approach The following brief outline on the development on human-environmental research as a subfield in geography situates this work into the broader corpus of cultural and political ecol- 13 ogy1. The human-environmental approach has deep roots in geography and remains today one of the largest subfields within the discipline. In the Annals of the Association of American Geogra - phers , arguably the most widely read and premier journal in Anglophone geography, articles are organized around four key themes, one of which is called “Nature and Society”. Today, the in - tertwined approaches of political ecology and cultural ecology largely define this nature-society dialectic. Cultural and political ecology in the United States can be traced back to the paradig - matic shift away from environmental determinism that defined human-environmental research during the first two decades of the 20th century. By the 1920s, the idea that humans and society adapt to environmental conditions rather than being subject to them became a more predominant paradigm with the advent of human ecology and the cultural-historical approach (Martin and James 1993: 344). Human ecology, as summarized by Harlan Barrows (1923), was an approach that conceived of the Earth as the home of humans and implied the mutualism of humans and their environment. Human ecology broke the unilateral deterministic link between the environ - ment and humans by proposing that people and societies2 adjusted to environmental conditions. Later, Gilbert White, a student of Barrows, and his students Robert Kates and Ian Burton, began to examine the strategies and responses of individuals to natural hazards, essentially starting a subfield of human-environmental research on environmental hazards that persists today. Critics, though, argued that this “natural hazards” school of research was far too narrowly focused on human behavior and failed to get at the socioeconomic and historical circumstances behind risk aversion (Zimmerer 1996; Liverman 1990; Watts 1983) The approach that human-environmental engagement is determined in part by cultural expression came about just after Barrows’s address by Carl O. Sauer at the University of Califor- nia-Berkeley. In his famous 1925 position, “The Morphology of Landscape”, Sauer (1963) out- lined a definition of a human agent-centric model of geography as the study of the environment as modified by cultural expression. Using a synthesis of historical evidence combined with meth- ods from the natural sciences and physical geography, Sauer and the “Berkeley School” students and colleagues sought to interpret landscapes as cultural artifacts or palimpsests of human activ- ity rooted in culture-specific traits, technologies, and ideologies. This approach demanded inten- sive and long-term field work typically in rural environments as well as focused archival research (Leighly 1987). Sauer’s students like James Parsons and Robert West in addition to their stu - 14 dents set a tradition of regional geographic research in Latin America that remains strong today (Brown and Mathewson 1999). Many of these topics involved landscape changes that resulted from the interaction of Spanish colonialism and the indigenous peoples of the New World (Sauer 1950; Parsons 1955; Denevan 1961, 1992; Johannesen 1963). Similar to the critiques of the “hazards approach,” critics of the cultural-historical or “Berkeley School” approach claim that its loose “superorganic” definition of culture fails to attribute agency to the complex social relations within cultures (Duncan 1980) as well as its tendency to gloss over place-specific details, not to mention class and gender, to create a broad generalist argument regarding landscape change that may not hold up to deeper scrutiny (Brookfield 1964; Zimmerer 1996). Cultural ecology, an approach largely founded by anthropologist Julian Steward (1955), sought to identify culture as reflective of adaptation to the environment, centering on subsistence activities. In this nascent form of cultural ecology, culture occupied a mostly functionalist role, linking nature and society. Technology, settlement landscapes, and the structural underpinnings of society, among other cultural traits, were all relative adaptations to environmental condi- tions (Butzer 1989). Rural, subsistence-oriented societies, from both pre-modern and then con- temporary contexts, were typically the subjects of this research as they provided a presumably closer relationship between conceptualizations of “nature” and society (Butzer 1976; Parsons and Denevan 1967; Carneiro 1960; Aschmann 1959; Netting 1968). The systems ecology focus arose within cultural ecology later in the 1960s and 1970s and conceptually situated humans and societies as constituents of a holistic ecosystem (Grossman 1977). In this focus of cultural ecology, culture and human society also played a functional role but one that was conceived to achieve some form of ecosystemic homeostasis (Ackerman 1963; Fosberg 1963; Geertz 1963; Rappaport 1968; Vayda and Rappoport 1968; Neitschmann 1973; Bergman 1980). Energy flows, equilibrium, and trophic levels, metaphors borrowed from a natural science-based definition of ecology, were used to describe human-environmental engagement in this approach (Zimmerer 1996: 173). Critics of this approach claim that such studies too narrowly framed the culture and environments that were their study subjects. These studies could illustrate how societies adapt their subsistence to environmental conditions and change, but not why they came to be like that (Vayda and McKay 1975; Watts 1983; Zimmerer 1996). By that, this approach was largely ahistorical and ignored larger socioeconomic processes such as global and regional economies as well as involvement by the state (Robbins 2004). 15 While the general systems approach persisted into the 1980s, another cultural ecology approach arose in the early 1970s, known as the adaptive dynamics approach. Like its predeces - sors cultural ecology and general systems ecology, adaptive dynamics examined adaptations to environmental conditions, but did so at an individual level. These individual adaptations could be scaled up and aggregated to describe more widespread regional environmental change (Zim - merer 1996: 174). Adaptive dynamics research maintained a topical focus on subsistence, pri - marily peasant agroecology and agricultural landscape evolution (Waddell 1972; Boserup 1975; Denevan and Schwerin 1978; Denevan 1983; Turner 1983; Doolittle 1984, 1990; Wilken 1987; Knapp 1991; Zimmerer 1994). Many of the studies to come out of adaptive dynamics examined relic agricultural landscapes such as raised fields (Dunning and Beach 1994; Doolittle 1992; Knapp 1991; Mathewson 1990; Turner 1983). This research approach remains vibrant in archae - ologically-oriented geography today (Dull 2007; Woods et al. 2009). However, when applied to contemporary societies, the criticism that adaptation cannot sufficiently account for environmen- tal change, landscape modification, and the human activity behind them remains apparent (Wolf 1982: 78; Watts 1983a; Zimmerer 1996: 176). For almost 30 years or so, political ecology has been an expansive and contested ap - proach that geographers, among other social and environmental scientists, use to explain the interrelation of society, political economy, and the environment or its social construction (Zim - merer and Bassett 2003; Robbins 2004). The term “political ecology” was likely first coined by the anthropologist Eric Wolf in a brief entry in Current Anthropology in 1972 (Wolf 1972) but it was not until the 1980s that a more defined approached came together. Political ecology very broadly seeks to elicit “chains of explanation” or “networks” (Blaikie 1985; Blaikie and Brookfield 1987) for environmental change from sources such as global development initiatives, markets, and international funding, as well as the national, state, and local policies that bear their mark. Political ecology’s constitution has not remained static and this approach has been rede - fined and appended upon since its emergence in the mid-1980s. In the 1980s and 1990s, as cul- tural ecology became more inclusive of policy initiatives, political economy, global conservation initiatives, development, and new paradigms borrowed from humanities as actors and conceptual frameworks in human-environmental interactions, it began to coalesce with political ecology as a subfield (Zimmerer 1994, 2007; Bassett and Zimmerer 2003). For the first decades of its existence, cultural and political ecological research largely 16 engaged male subjects, maintained a male-normative perspective, or assumed a gender-neutral approach. It also largely represented the scholastic preoccupations of Western academics without noting the inherent positionalities within them (Escobar 1996, 1999). The advent of feminist po - litical ecology acknowledged the influence of gender on the struggle over environmental resourc- es and power (Carney 1993, 1996; Rocheleau, Thomas-Slayter, and Wangari 1996; Rocheleau 2009). Additonally, political ecologists sought to expand beyond the approach’s preoccupation with rural areas and into urban locations (Swyngedouw 1997; Keil 2005; Heynen, Kaika, and Swyngedouw 2006; Robbins 2007). Still, a large spatial focus of cultural and political ecology research remains on rural areas in the developing world, with a considerable regional emphasis in Latin America (Zimmerer and Bassett 2003) and Mexico in particular. The major themes of this research in Mexico primarily involve forest management (Klooster 2000, 2003; B.L. Turner et al. 2001; Bray et al. 2006;), property regime/land tenure change (Toledo 1996; Herlihy et al. 2008; Perramond 2008; Smith et al. 2009, 2012; Kelly et al. 2010; Herrera-Rodriguez 2012; Hernández-Cendejas 2012), water rights (Barkin 2006; Wilder and Romero-Lankao 2006; Kelly 2013), agricultural and ranching systems (Perramond 2010; see Doolittle et al. 2002 for an overview), and conservation areas (Si - monian 1995; Young 1999a, 1999b; Keys 2005; Ericson 2006). Only Moore (2008, 2009, 2011) specifically examined garbage from an obvious political ecological approach in Mexico. Robbins (2004: 215) lists “the political ecology of trash, waste management, and gar- bage subsistence” under a list of “new substantive research mandates,” pointing to productive new spaces for its investigation within political ecology. Garbage has seen a growing interest in explicitly political ecological studies (Nchito and Myers 2004; Myers 2005; Davies 2008; Moore 2008, 2009, 2011; Hartmann 2012) and there is a growing human-environmental focus on gar - bage in geography and allied disciplines, yet the existing literature remains small. Also, nearly all of these take place in large cities, even though garbage is oftentimes an overt presence in small town and rural landscapes. Garbage is produced by and in households, communities, towns, cities, municipalities, states, and countries. Each of these entities and scales has its own ways, policies, and practices of managing this garbage. This dissertation is theoretically grounded in cultural and political ecol- ogy through its focus on how Coxcatecos as residents of their households, communities, the state of Puebla, as Mexicans, and as world citizens manage their garbage, both practically and discur - 17 sively. This dissertation is methodologically situated in cultural and political ecology through a grounded focus on a detailed accounting of the practice of garbage management by households. Households have remained an important scalar focus for cultural and political ecology for de - cades to show a more precise resolution beyond more spatially conflated categories of a region, and in this case, community and municipio (Brookfield 1964; Agrawal 1994; Netting 1993; Watts 1983; Grossman 1977; Bassett and Zueli 2000; Zimmerer 2004). The intent of the garbage cen- sus outlines a practice-centered household economy of garbage with the goal of highlighting the diversity of garbage management practices in Coxcatlán beyond a simplistic dichotomy of those with or without collection service (Bulkeley and Gregson 2009). Here households and commu- nities employ a varied mix of management practices that serve tradition, perform some func- tionality, and provide measures of ease and convenience. Not content to rely on a matter-of-fact representation of household garbage management, the Q-method study attempts to describe the interpersonal subjectivities, opinions, and discourses behind these fundamental environmental practices at the household level. The risk mapping study (Chapter 6) seeks to show how house - holds and communities from diverse socioeconomic backgrounds given differential access to garbage management infrastructure and different regimes of garbage governance frame garbage as an environmental issue. The literature of environmental awareness and public perception of environmental risk is well-developed, yet much of the work has exhibited a developed, industrialized world bias (Beck 1992; White and Hunter 2009). Environmental concern is a global phenomenon and people in more economically marginalized settings maintain similar if not greater environmentalist per - spectives than their counterparts in the developed world (Hunter, Hatch, and Johnson 2004; Dunlap, Gallup, and Gallup 1993). Through a political ecology lens, garbage becomes politi - cal; evincing a struggle for control over waste, its impact, and the best practices to deal with it. Within geography, this research is located within a growing focus of political ecology research investigating the social construction of environmental issues (Harrison and Burgess 1994; Watts and Peet 1996; Escobar 1996; Braun and Castree 1998; Eden 1998; Bridge and McManus 2000; Castree and Braun 2001; Demeritt 2001) that seeks to point out the underlying discourses behind their construction and practical employment. In Latin America and Mexico, similar research has focused of water depletion (Von Bertrab 2003), soil erosion (Zimmerer 1993), and post-colonial environmental history (Sluyter 1999). This dissertation joins this literature in its constructivist 18 Figure 1.2. Locations of long-term, peer-reviewed research on generation and classification of MSW in Mexico. Map by author. view of the environment through use of Q-methodology (Robbins 2006) to describe the subjec - tivities regarding garbage generation, its impact, and best management and risk mapping (Smith et al. 2000, 2001; Baird 2009) to highlight the heterogeneity of opinion and the construction of environmental issues related to garbage in Coxcatlán. CURRENT GEOGRAPHICAL AND SOCIAL SCIENCE RESEARCH ON GARBAGE This dissertation draws from and contributes to social science-oriented studies on waste and garbage, most of which begins with garbology. Garbology describes the research agenda of William Rathje’s “Garbage Project” at the University of Arizona that began in 1973 and contin - ued strong through the mid-2000s (Rathje 1974; Philips, Restrepo, and Rathje 1984; Rathje and Murphy 2001). Rathje and his team of researchers applied an anthropologically-informed archae - ological approach to garbage in this ground-breaking, multi-decade, domestic and international research project. They examined garbage from a wide-ranging, holistic perspective that sought to understand contemporary culture and society from the things it threw away. Important among the multitude of Rathje et al.’s findings is the understanding that what people say and think they 19 throw away may differ from what they actually do. Currently, there is a massive body of research on garbage in peer-reviewed literature with journals specifically dedicated to garbage, waste management, and recycling. Waste Manage - ment, Waste Management and Research , and Resources, Conservation and Recycling are top- tier, multi-issue per year, English language journals that cover mostly technical issues of formal garbage management. Geographical and allied disciplinary research into garbage, recycling, and waste in general is an emerging yet strong body of scholarship that is seeing an increase in interest due to the growing recognition of garbage as one of the most critical human-environ- mental issues facing the world today (Robbins 2004; Moore 2011). The geographical and social science-based research is notably less concerned with the technical aspects of waste management and more focused on waste management as a socio-spatial, economic, and environmental issue. Current geographical research on garbage is organized around four interrelated themes: 1) envi - ronmental governance of garbage, 2) the informal waste sector, 3) environmental justice of waste and its global trade, and 4) littering. Environmental governance and garbage Environmental governance is an expanding avenue of political ecology and the gover - nance of garbage is a notable theme in this larger body of research. Much of this literature uses garbage as a focal issue for understanding the interaction and too often functional breakdown of different scales of environmental governance regarding garbage management. Davies and O’Callaghan-Platt (2008) examined the multiple and fluid scales of garbage governance in their native Ireland, while Davies (2009; 2008) expanded this line of investigation in a cross compari - son of Ireland and New Zealand. Davoudi (2009) and Davoudi and Evans (2005), using garbage policy as a central environmental issue, explored how the inclusion of broader regional gover- nance from the EU has worked its way into garbage management policy in the United Kingdom in the past decade. Fitting with the difficulties in applying regional-based governance to local practices, the inflexibility of upper-level supranational and state-driven policy is laid bare in Bulkeley and Askins’ (2009) research on informal household management of organic garbage in the UK. Murray (2009) unpacks the power relations fundamental to Irish waste policy in an attempt to point out the inadequacies of Beck’s (1992; 1996) reflexive modernity thesis by show- ing the overruling influence of economic development. Oosterveer (2009) also takes a general 20 critique on rigid state-based environmental governance in East Africa, offering an alternative in what he calls the “networked state” where garbage is managed by hybrid formal and informal actors. Myers (2005) also uses inadequate garbage management as an example to unpack the neoliberal strategies embedded in so-called “good governance” initiatives of global develop - ment agencies, notably the World Bank, in three cities in sub-Saharan Africa. A common, unit- ing theme points out the incongruities between garbage policy goals across scales with the daily practice of garbage service provision. The empirical realities described in these studies bring to light how policy needs might not intersect at all with the populations who must abide them. Not all research on environmental governance provides evidence for the gaps between different scales of garbage governance. Forsyth (2005), through case studies in Indonesia and India, indicates how a highly communicative and participatory approach of public-private gover - nance over garbage dump regulation can be successful for local communities and waste service providers. Bhuiyan (2010), in case studies from Dhaka and Chittagong, Bangladesh, essentially proves the same point, showing how community-based garbage management can supplement the often inadequate state-run garbage collection service. Joseph (2006) also points to how increased social participation of informal actors in more formal garbage management can result in increased access to sanitation that ultimate is in-line with policy goals. In Coxcatlán, a similar more-than-what-policy-intends model of environmental governance with a mix of formal and informal actors exists. While Mexico’s General Law for the Prevention and Integral Management of Wastes ( Ley General para la Prevención y Gestión Integral de los Residuos ) is a compre- hensive waste law that is more inclusive of all sectors of society in its conceptual design, it still intends by practice a more hierarchical state-run garbage management policy. This dissertation, while not entirely focusing on environmental governance, contributes to the aforementioned en- vironmental governance of garbage literatures by giving a household-based perspective (Bulke- ley and Askins 2009) in contrast with the more policy-oriented research on environmental gov - ernance. Drawing attention to agency within the household regarding garbage management, Barr (2002) analyzed the behavioral determinants of reuse, recycling, and waste minimization patterns in a detailed household study in Britain. The dissertation research follows both Barr’s household- scale focus as well as its concern with attitudes and subjectivities of households regarding gar- bage issues. 21 Informal waste sector research Garbage is waste that comprises unwanted and rejected matter. Yet given the subjective nature on what constitutes waste and as well as the growing commoditization of waste amid its reconceptualization as a resource, across the developing world the informal labor sector of waste pickers that makes a living from garbage collection, recycling, and reuse. Geographers and other allied social scientists have studied the informal waste labor force and their struggles for control over garbage resources and a livelihood. This is a massive body of literature that spans much of the developing world (Katusimeh, Burger, and Mol 2013). There are even recent feature docu- mentaries on the informal waste sector, such as 2010’s Waste Land , a story about making art out of Brazil’s large dumps, a 2009 documentary on Egypt’s Zabaleen waste pickers titled Garbage Dreams , and in a Global North example, there is 2009’s Dive! an hour-long view of hipsters and homeless alike subsisting off of “dumpster diving” in Los Angeles. Much of the formal academic research on the informal waste sector comes from case studies in Latin America (Medina 2000). Fergutz, Dias, and Mitlin (2011) give empirical evi - dence on waste generation and waste picking in Brazil to highlight the economic benefits of informal recycling while making a claim to reward this sector with more formal recognition and economic security. Moreno-Sánchez and Maldonado (2006) also prove this point, statistically showing the economic benefits and noting the social benefits of including informal waste work- ers into garbage management regimes in a case study in Colombia. Hartmann (2012) looked at the struggle of waste pickers for control of garbage resources at the massive La Chureca dump in Managua, Nicaragua following development agencies’ plans to formalize management there. Medina (1997) reported on how the privatization of waste management and recyclables recovery in Cali, Colombia marginalizes the already existing informal scavenger sector. Medina (2000) also surveyed case studies in Latin America and Asia to assess the common ways waste scaven- gers are treated and offer alternatives for more equitable treatment and increased environmental sustainability. On a more positive note, Hernández, Rawlins, Schwartz (1999) described a com - munity enterprise-based garbage collection system and conducted a survey on recycling partici- pation in Quito, Ecuador that pointed out needs for greater participation in the program’s design. In Mexico, Castillo Berthier’s (1983) in-depth research on the garbage cycle in the Mexico City metropolitan area was a groundbreaking study that portrayed garbage management as a social and economic issue beyond previous material-based studies on generation and clas- 22 23 sification. In La Sociedad de la Basura: Caciquismo en la Ciudad de México , Castillo-Berthier outlines in detail the cycle of garbage collection, recycling, and disposal while showing how the control over garbage management created a boss-client system that integrated populist politics of waste workers with the dominant political regime (the PRI) in Mexico City. Medina (1998) com - pared informal recycling regimes on both sides of one point at Laredo-Nuevo Laredo, along the US/Mexico border. Medina also (2005) expanded his comparative study, exploring the scalar dif- ferences between informal waste workers in a small city (Nuevo Laredo), a large city (Monter - rey), and a megalopolis (Mexico City) in Mexico paying attention in all of these cases to the vital yet informal and socially unsupported role these people play in the functioning of everyday life. Guzmán and Macías (2012) analyzed the struggles of pepenadores or waste pickers in San Luis Potosí to defend the nature of their work amid top-down structural changes in management at the large Peñasco dump there. Ojeda-Benítez, Armijo de Vega, and Ramírez-Barreto (2002) exam- ined the interaction of municipal governments, private concessions, and informal waste workers in the recovery of recyclables in Mexicali, Baja California. Moore (2008, 2009) described how garbage was used as a political tool by marginalized Oaxaca City residents who blocked access to the city dump. Moore highlights how control over the dump challenged the civic order of the municipal government because of the government’s inability to remove the most obvious but often-forgotten ordering of the city, garbage collection. In this sense, garbage by its abject nature, challenged contemporary perceptions of orderliness and was thus used by a marginalized popula - tion to voice their concerns over their own sanitation and environmental risk. The most obvious project in the body of research on informal waste management sector is its important role in avoiding garbage crises in the developing world. In spite of the informal waste sector’s importance in collection service, minimizing waste in landfills, in its primary role in a global recyclables commodities market, its work, like that of the rest of the informal labor sector, largely goes under-recognized and under-compensated by governments and the general public. While there is no dedicated informal waste labor sector in Coxcatlán, the communities, households, and individuals who are without collection service in the municipio do, in a sense, fall outside official garbage management policy ideals and represent a form of informality when it comes to dealing with their garbage. This dissertation research contributes to this body of work by opening the definition of informal waste management to households in underserved communi- ties across the developing world. Environmental justice and waste Garbage is pollution (Eriksen et al. 2013; Njeru 2006) and implies risk (Douglas 1966; Hawkins 2003; Riley 2008). Moving garbage away from lived space and depositing it elsewhere through collection is the predominant management strategy in industrialized countries (Moore 2011: 140). Yet garbage and its by-products such as smoke and leachate have to go somewhere. Typically these “somewheres” are economically or socially (racially) marginalized places (Pulido 2000). The spatial fix that garbage collection performs is often paid in terms of environmental quality by the most powerless and stigmatized populations (Moore 2011: 137). This makes gar - bage and waste a focus of environmental justice research. In fact, Williams (1999) and Bullard (1994) cite a 1982 protest over a North Carolina hazardous waste landfill siting as the marquee event that started the current environmental justice movement. Much of this body of work focus- es on more dynamic, hazardous, and toxic waste than garbage, making its material focus some - what outside the scope of this dissertation (Maantay 2006; Kurtz 2005, 2003; Ishiyama 2003; Girdner and Smith 2002; Pulido 2000; Williams 1999; Szasz 1994; Bullard 1994; 1992) and very little directly addresses the matter of garbage (Pellow 2002; Moore 2008, 2009, 2011). Nonethe - less, this literature makes useful parallels to garbage in the relation of waste and marginalized people and spaces. This dissertation contributes to the environmental justice and garbage debate in an oblique sense. Coxcatlán’s 2.5 hectare dump does not have a resident population of pepenadores nor does anyone really live near it for the time. Rather, the nearly thirty percent of Coxcatlán’s population that is currently without garbage service has to contend with their own garbage on their own property and in their communities could be considered an environmental justice is- sue. Furthermore, this population is almost entirely indigenous, which is the most economically, socially, and ethnically marginalized stratum of Mexican and pan-Latin American society today (Radcliffe 2007). However, accusing direct environmental racism in this case as a local effect is an incorrect and imprecise assumption that I do not intend to imply. Coxcatlán’s meager garbage collection infrastructure is insufficient to service the geographically isolated and spatially exten- sive communities of its eastern sierra half. It is not from a lack of effort that no service exists. Thus, the causal indirectness of environmental justice and racism issues arise more from inade- quate access to financial and technical resources than a deep nefarious plot to further marginalize indigenous peoples. When considering the entire municipio and its garbage collection infrastruc- 24 ture, the claim could be made that the lack of a sanitary landfill and the continued existence of the open air dump there is a possible environmental justice issue, just at a much broader scale. The global trade of waste synthesizes the waste as a global commodity and the environ - mental justice literature. Waste, through its construction as a resource by its ability to be recycled or turned into energy as well as its ubiquity and abundance, has made it a global commodity. Yet its fundamental character as something unwanted and polluting makes it the subject of spatial othering that lands it in the most marginalized places. Moore (2011) outlines the global waste trade, conceptualizing it both from a material and policy perspective. Lepawsky and McNabb (2010) draw attention to the more regional flows of electronic waste or “e-waste”, complicating the simplistic conceptual “pollution haven” model of dumping from global North to global South by showing a less than linear model where e-waste is revalued, converted, and resold, thus stay- ing in a state of flux. Previous research on garbage in Mexico Since the early 2000s, there has been a growing body of academic literature on garbage management in Mexico. Most of it concerns municipal solid waste (MSW) management in major metropolitan areas or tourist areas. Figure 1.2 shows where peer-reviewed research on MSW in Mexico has taken place since 2001. Very little of this research, however, concerns garbage management in rural areas or small cities and towns. This focused effort on large metropolises is appropriate given that approximately 52 percent of Mexicans live in such urban areas (INEGI 2005). Yet for small towns and rural municipios, the only available information comes from difficult-to-access grey literature archived in state environmental offices’ archives to assess the garbage situation there. Estimating how much garbage and of what classes or substances it comprises is a promi - nent theme in this body of research. Generation and classification studies have been conducted in Guadalajara (Bernache et al. 2001; Bernache 2003), Morelia (Buenrostro, Bocco, and Bernache 2001; Buenrostro, Bocco, and Vence 2001; Buenrostro, Ojeda, and Marquez 2007; Buenrostro, Marquez, and Pinette 2008), Chihuahua City (Gómez et al. 2008), Mexicali (Ojeda-Benítez, Armijo de Vega, and Ramírez 2000, 2002, 2003; Buenrostro, Ojeda, and Marquez 2007; Ojeda- Benítez, Armijo de Vega, and Marquez 2008; Ojeda-Benítez et al. 2008) and the tourist desti- nation Lago Cuitzeo in the western state of Jalisco (Buenrostro et al 2008). Other studies have 25 focused on the impact of leachates (Reyes-López et al. 2008; Ticante et al. 2004), decomposition rates (Hernández et al. 2008), and typologies (Ojeda-Benítez and Beraud-Lozano 2003) of land- fills in Mexico. Researchers have also conducted studies examining the management of garbage and recycling behavior on Mexican university campuses (Armijo de Vega, Ojeda, and Ramírez 2003; Ayala-Rodríguez 2004; Maldonado 2006; Espinosa et al. 2008). While most of the academic literature on garbage in Mexico appropriately outlines the considerable issue of garbage management there by investigating generation rates, final disposi- tion, and landfill metrics, only a few studies have approached the issue from a more geographical and social science perspective. Beside the aforementioned work of Sarah Moore (2008; 2009), Buenrostro and Bocco (2003) indirectly touch upon a political ecology perspective in their sum- mary of the potential environmental impact of garbage across Mexico. Noting the consumption pattern changes due to Mexico’s rapidly urbanizing population to a greater and more mixed and inorganic solid waste stream, the authors situate this issue within Mexico’s garbage manage - ment infrastructure, drawing attention to the potential environmental issues from such a garbage managment system. Funes Aguirre (2004) examined the management of garbage in Mexico from an environmental law perspective by looking at the competing scales of environmental law at international, federal, state, and municipal scales. Galván-Meraz (2009) conducted an exhaustive and thorough analysis of contextualized Mexico’s current solid waste laws in a larger environ - mental legal perspective. Guevara, Quintanar, and Rodríguez (2004) conducted a survey measur - ing environmental attitudes regarding garbage collection, recycling, and separation in a neigh- borhood of Puebla City, Puebla, finding out that the mostly pro-environmental attitudes correlate well with the moderate success of a recycling and separation project there. Littering This dissertation examines garbage from both a management and non-management per- spective. Litter is garbage that falls outside of formal management (Geller et al. 1980). If garbage is abjected matter (Kristeva 1982; Moore 2008) or matter-out-of-place (Douglas 1966) then litter is garbage doubly out-of-place. It has the potential to contaminate landscapes chemically (Lemieux et al. 2003; Nduka 2006; Solorzano-Ochoa et al. 2012), biologically (Wokekoro and Inyang 2007), and visually (Pandey 1990). Litter challenges civic order by representing the ab - sence of management. As such, it is often synonymous with environmental and social problems 26 related to garbage (Kelling and Coles 1996; Ligget et al. 2001). By looking at the litter problem in Coxcatlán, the penultimate chapter (Chapter 7) counterbalances the previous chapters’ exami - nation of garbage’s management there. Through its comparison of how much people anecdot - ally believe Coxcatlán is littered against field-sampled data on its accumulation by class, it also synthesizes the previous chapters’ conceptualization of garbage as both a material and discursive object. Littering is a universal environmental issue and as such, the literature on it is vast across disciplines and temporally deep. This body of scholarship coalesces around the behavior behind littering, the environmental and social effects of litter, and quantitative studies to measure it. Littering is generally believed to be a behavior that is outside the norms of social propriety and environmental stewardship and identifying who litters has been a preoccupation for many behav- ior and psychology researchers (Heywood 2002) but no identifiable trope of a litterer is really identifiable. Partly this is because most people end up littering, whether actively or passively, at some point in time but also because the conditions for observing littering are nearly impossible to create. Thus, in order to gauge the magnitude of littering at an individual level, many studies have tried to identify the frequency with which people litter given certain environmental or social characteristics of the setting (Finnie 1973; Geller et al. 1977; Reiter and Samuel 1980; Horsley 1988; Cialdini, Kallgren, and Reno 1991; Reno, Cialdini, and Kallgren 1993; Hansmann and Scholz 2003; Sibley and Liu 2003; DeKort et al. 2008; Schultz et al 2013) with the larger proj - ect of reducing it. Litter is matter, so another preoccupation for researchers is its quantification. Litter cleanup is an immensely expensive burden on municipal governments and as a result there are numerous grey literature studies on gauging litter in public space. Beaches, because of their economic as well as ecosystemic value, have been the subject of numerous litter quantification studies (Silva-Iñiguez and Foster 2003; Rodrigues-Santos et al. 2005; Ivar do Sul et al. 2009). Litter could be considered a predominantly urban issue, however because of the difficulties in scientifically quantifying litter in such a dynamic environment where cleaning takes place, there are few studies on the subject (Moriwaki et al. 2009; Muñoz-Cadena et al. 2009; Seco Pon and Becheruchi 2012). The Keep America Beautiful survey is a periodic national sampling of lit- tering in urban environments as well as roadsides across the country (MSW Consultants 2009). Australia’s Keep Australia Beautiful campaign updates its annual National Litter Index in what is probably the largest and most comprehensive data source on littering (MacGregor Tan Research 27 2012). Roadside litter studies funded by various states’ departments of transportation or environ - mental agencies have been around since the 1970s (IAR 2006). These studies have taken place in Florida (Butterfield and Myers 1999), Texas (NuStats 2009), Ohio (Davey Resource Group 2004), Washington (Cascadia Consulting Group 2005), and New Jersey (Stein and Syrek 2005). While there are slight methodological differences between these studies, most involve selecting sites, cleaning them of refuse and periodically measuring the litter that accumulates. In Mexico, Muñoz-Cadena et al. (2009) examined littering in two barrios of different socioeconomic char - acteristics in Mexico City and Silva-Iñiguez and Foster (2003) measured litter on a beach in Ensenada, Baja California but thus far no such studies have been published or conducted along any of Mexico’s highways. Chapter 7 seeks to fill that gap. FORWARD TO THE NEXT CHAPTERS The following chapters first describe tland and life in Coxcatlán, Puebla, Mexico, where this study on garbage and its management occurred. The subsequent chapters will frame the garbage issue at the federal, state, municipal scales in Mexico, finishing with an empirical study on household generation and management. This latter component provides insight into garbage management from the often neglected household scale. Juxtaposed against the material realities of garbage and its management in Coxcatlán, are how garbage is framed as a set of environmen- tal issues. Synthesizing the material and perceptional aspects of garbage is the issue of littering. The penultimate chapter compares the perceived extent of littered-ness to its material reality in Coxcatlán. The final chapter provides a brief summary. Endnotes 1. The following outline of the development of the human-environmental approach is inspired heavily and draws upon more in-depth disciplinary histories by Zimmerer (1996) and Robbins (2004) with supplements from Martin and James (2003). 2. Barrows, Sauer, and most other preeminent geographers at the time persisted in the use of the word “man” to collectively identify more gender-neutral terms such as “humanity”, “humans”, “people”, or “society”. For the sake of some form of erudition, I use “humans” or “society”, although I recognize these latter concepts are perhaps loaded terms that will eventually be dispensed in future scholarship. 28 CHAPTER 2: Land and life in Coxcatlán, Puebla, Mexico LOCATING COXCATLÁN The municipio What is a municipio? The municipio of Coxcatlán, Puebla is the primary spatial frame for this dissertation. Since “Coxcatlán municipio” or the “ municipio of Coxcatlán” will be referenced for its entirety, it is important to elaborate what a municipio is and does. In Mexico, as well as in other Latin American countries, the municipio is the political-governmental entity below the state ( estado or entidad federativa ). While there is no direct analog for a municipio in the United States, most would place a municipio in the same spatial frame as a U.S. county. Just like coun - ties in the U.S., there is a great deal of variation in size for municipios across Mexico (Figure 2.1). In Puebla, whose territorial expanse is 34,290 square kilometers 1, there are 217 municipios. Puebla’s 217 municipios is second in number only to its southern neighbor, Oaxaca, which has Figure 2.1. Map of municipios in Mexico in 2010. Map by A. Hilburn. Data, INEGI 2010. 29 570 municipios but with almost three times its territory. On the other end, Baja California Sur 2, at over twice the size of Puebla, is divided into only five municipios (INEGI 2012). The number and size of municipios in a state is usually determined by recognition of preexisting territorial divisions. Both Puebla and Oaxaca still have large indigenous populations and their municipal maps reflect these groups’ long, entrenched territoriality and the colonial boundaries upon which they are predicated (Dennis 1987). The municipio is defined as a political unit and its powers are described in Article 115 of the 1917 Constitution. While each is called a municipio libre, most derive the vast majority of their budgets from state and federal sources (Rodríguez 1993). Each municipio’s government is seated in the cabecera municipal. For the purposes of clarity, the eponymously titled cabecera of Coxcatlán (pop. 6,009) will be referred to as “Coxcatlán town” in cases where both the mu- nicipio and cabecera are being discussed. The cabecera is typically, but not always, the largest settlement within the municipio. In states like Puebla or Oaxaca with numerous municipios, the municipio usually bears the same name as its cabecera. Additionally, many muncipio names are prefaced by a patron saint, such as San Juan Bautista Coxcatlán, or are followed by a figure from Mexican history like nearby Teotitlán de [Enrique] Flores Magón. This can reflect earlier eccle- siastical territories, as is the case with Coxcatlán or representing the home region of a famous person, like nearby Teotitlán. These saint or national names also help distinguish municipios in states like Puebla or Oaxaca with such large counts of municipios. Each municipio’s government is headed by the presidente municipal (municipal presi - dent) who is elected every three years on July 4 th. The presidente municipal is the chief execu- tive in all settlements within the municipio. In more rural municipios like Coxcatlán, smaller settlements outside of the cabecera have municipal representation and authority through local inspectorias de policia (police inspectors) or presidentes auxiliaries (auxiliary presidents) who are appointed by the municipio’s president. The municipio is a strong functional region, one that can levy taxes, pass ordinances, run the police, administer most public services, like garbage collection, and conduct all other local government business. Spatially, it operates as a network of settlements within a defined area. Coxcatlán in space and time Where is Coxcatlán, Puebla? Without delving into the more nuanced humanistic notions 30 of place (Tuan 1990; Cresswell 2004; Lefebvre 1991; Soja 1996; Sack 1997), geography, and its sister discipline geodesy, provide numerous spatial methods and metaphors for determining where something is. From a basic Cartesian or geodetic understanding of location, the muni- cipio of Coxcatlán, Puebla, México covers about 249 square kilometers with its centroid located at 18°14’46”N, 97°7’5”W (Figure 2.2). Thus, Coxcatlán sits well below the tropics. Coxcatlán is within the southern border of the State of Puebla and is adjacent to the neighboring State of Oaxaca. Bordering muncipios in the State of Puebla include Ajalpan, San Sebastian Zinacatepec, Coyomeapan, Zoquitlán, and San José Miahuatlán. Teotilán de Flores Magón and San Antonio Nanahuatípam form Coxcatlán’s southern border in the state of Oaxaca. Physiographically, the municipio is bifurcated east-west with the western half in the Tehuacán Valley while the upland sierra half to the east forms part of the Sierra Negra Range of the larger Sierra Madre de Sur chain of southern Mexico. Portions of Coxcatlán are within the Tehuacán-Cuicatlán Biosphere Reserve designated and managed by the federal government. Coxcatlán is about 35 kilometers south-southeast by state highway and 30 kilometers as-the-crow-flies from the outskirts of the city of Tehuacán (pop. 274, 900, INEGI 2010). The state capital, Puebla, is about 160 kilometers northwest by highway from Coxcatlán and the federal capital, Mexico City, is another 115 kilo - meters west by federal highway from Puebla (Figure 2.3). What time is it in Coxcatlán? Coxcatlán, as well as most of Mexico, is located in the Cen - tral Time Zone (-6:00 GMT) and observes a slightly shorter Daylight Savings Time (DST) cal - endar than the one used in the US. However, the indigenous communities in the sierra half of the municipio and across indigenous southern Mexico typically reject DST as an absurd and foreign idea imposed upon them by the government and globalizing forces. These communities refer to their own timekeeping as la hora de Dios or “God’s time”, which never wavers from Central Standard Time. Yet, in Coxcatlán there is not really any observable confusion with scheduling between the two time systems as most understand this discrepancy and are familiar with events associated with the official time and those with la hora de Dios . Whether it be an absolute location, in the case of a bounded area in Euclidean space, or one of the relative locations described above, Coxcatlán is much more than a location or destina- tion, it is a place, a region, a study site, and a home as well as many other embodiments of space. The following sections of this chapter as well as the rest of this dissertation will provide a much fuller regional geography of Coxcatlán and surrounding area. 31 Figure 2.2. Municipio of Coxcatlán, Puebla, Mexico. Map by A. Hilburn. Data, INEGI 2010. 32 Figure 2.3. Route from Coxcatlán - Tehuacán - Puebla - Mexico City. Map by A. Hilburn. Data, INEGI 2010. COXCATLÁN’S “NATURAL” SETTING The physical setting or natural context that Coxcatlán municipio bounds are by no means purely physical or anywhere close to natural. Rather, a nature-society hybrid (Zimmerer 2000) or a cultural landscape (Sauer 1963) is implied for what constitutes “nature” there. Simply put, I am certain that no square meter of the municipio has been left untouched or unchanged through history by some form of plow, axe, flame, bulldozer, dynamite, or sheer human might. Coxcatlán as much as any place in the inhabited world remains to this day the site of dynamic human-envi- ronmental processes of change. This is evident as the larger urban areas such as Coxcatlán town, Calipan, and Tilapa cover significant portions of the municipio. Irrigated sugarcane fields domi- nate the riparian areas of the valley floor and the hillsides of the sierra communities are almost entirely made up of fields in different forms rotation and fallow. Even the areas set as reserve by the social properties ( ejidos ) and the portions within the Tehuacán-Cuicatlán Biosphere Reserve are managed landscapes by the presumable suspension of development and extractive activities 33 within their boundaries. Nevertheless, these changes do occur within some definition of an “en- vironment”, “ecological setting”, or “cultural landscape” that does have flora, fauna, a geologic base, soils, and climate. The following sections outline this setting, however tenuous its claim to nature may be. Landforms, relief, and hydrology At a continental scale, Coxcatlán municipio is found near the intersection of three differ- ent physiographic provinces in Mexico, the Sierra Madre Oriental, the Neovolcanic Axis, and the Sierra Madre del Sur. The formation processes and origins of all three physiographic provinces are relevant to the broader region of the Tehuacán Valley and the sierra to its east. Yet it is the lat - ter province, the Sierra Madre del Sur, in which one places Coxcatlán. The Neovolcanic Axis of Mexico is to the north of the study area and does not reach Coxcatlán, yet on a clear day from the tops of the sierra, one can get a clear view of this region’s most impressive feature, the stratovol - cano Pico de Orizaba, or Citlaltépetl, Mexico’s highest mountain at 5,636 meters above sea level. The surface geology of Coxcatlán’s sierra is comprised mostly of schists from the Up - per Cretaceous (INEGI 1999). The orogeny that produced the Sierra Madre Oriental and Sierra Madre del Sur and subsequently the Tehuacán Valley began later in the Palaeocene of the Ter - tiary and is contemporaneous with the Laramide Orogeny that produced the Rocky Mountains (Maldonado-Koerdell 1964: 15). In Coxcatlán, there is a narrow intermediate area between the sierra and valley floor of Lower Tertiary origin composed of a sandy, highly friable conglomer- ate. The upland portions of Coxcatlán town and the Postclassic site of Sansuanchi or Coxcatlán Viejo are sited in this area. The intermontane Tehuacán Valley, of which Coxcatlán comprises a small part at the southern end of its approximately 5,000 square kilometer expanse, extends and gently drops in elevation from the municipio of Tepanco de Lopez (1,820 m) in the north, south - east through the metropolitan area of Tehuacán (1,650 m), to Coxcatlán (1,190 m), terminating in the Cuicatlán Cañada (650 m) in northern Oaxaca state (Figure 2.4). This elevation change is gradual and is almost imperceptible as one traverses the valley. The majority of the valley’s sur - face is composed of alluvial deposition from the Quaternary. Before a rapid drainage caused by the opening of the Río Santo Domingo watershed in the Lower Quaternary, the Tehuacán Valley and surrounding area was a shallow lake environment from its formation in the Lower Tertiary. Many of the surface sediments and young sedimentary rocks are indicative or this sort of envi- 34 ronment and are impregnated with lacustrine fossils. This drainage process, which is believed to have been brief and dramatic, connected the formerly-hydrologically isolated Tehuacán Valley to what today is the Río Papaloapan watershed that eventually flows into the Gulf of Mexico at Coatzacoalcos, Veracruz (Brunet 1967: 73 – 75). In the study area, all streams drain westward into the Río Salado, which forms the major - ity of the western border of Coxcatlán (Figure 2.5). Although it is the highest order stream in the Tehuacán Valley, the Río Salado is barely perennial. Other water courses that approach having Figure 2.4. Tehuacán-Cuicatlán Biosphere Reserve and Coxcatlán’s location within it. Map by A. Hilburn. Data, INEGI 2010., SEMARNAT 2009. 35 year-round water are the Río Comulco that forms Coxcatlán’s northern border, the Río de los Mangos or Tepazalco that drains the northern sierra portion of the municipio and empties into the Salado just below Calipan, and the Río Tilapa, providing the hydrological outlet for the southern portion of the municipio. Most of the water courses in the municipio are highly incised intermit- tent streams. These features are not typically given the designation arroyo , or stream, but are typically referred to as barrancas, or gullies, owing to their usual lack of surface water and steep banks of alluvium or sedimentary rocks (Figure 2.6). The majority of water used by Coxcatlán’s Figure 2.5. Topographical relief and surface hydrology, municipio of Coxcatlán, Puebla, Mexico. Map by A. Hil- burn. Data, INEGI 2010. 36 Figure 2.6. Barranca Atempango streambed in December. Photo by A. Hilburn, December 2011 Figure 2.7. Engineered stream used for the transmission of water to cabecera municipal. Photo by A. Hilburn, November 2011 37 settlements comes from springs dotting the landscape and channeled water courses with origins in the sierra (Figure 2.7). Coxcatlán municipio’s topographical relief is visually striking. It covers only approxi - mately 249 square kilometers, but along a transect from its western border with the municipio west of Coxcatlán town at the Río Salado (960m above sea level) to one of its highest elevations on the ridgeline above Tecoltepec (2660 m asl), the elevation rises 1,700 meters in less than 17 Figure 2.8. Elevation gradient, municipio of Coxcatlán, Puebla, Mexico. Map by A. Hilburn. Data, INEGI 2010. 38 kilometers of horizontal distance (Figure 2.8). The municipio’s eastern border is the last, west - ernmost ridgeline of the Sierra Negra range before arriving to the Tehuacán Valley, giving an impressive elevation contrast in very little horizontal distance. Climate in the valley and sierra Coxcatlán municipio’s microclimates vary according to the dramatic elevation differ - ences across its extent with higher elevations of the sierra being wetter and cooler than the valley below. One can see these differences clearly using climatological data from 1971 to 2000. These data come from two stations, one from Coxcatlán town (elev. 1,145m asl), which is sited on an alluvial fan just above the riverine terraces of the valley floor and represents the valley climates, and the other from Zoquitlán 3 (2,260m asl), to represent the sierra (Figure 2.9). The common climatic attribute between the valley and sierra is wet-dry seasonality. In the valley, the dry season or sequía is more pronounced than it is in the sierra and generally lasts from October to mid-May. This seasonal aridity is caused by the southern migration of a subtropical high pres- sure system, causing general atmospheric stability in the northern tropics during this time (Byers 1967: 49-50). Across Middle America, the effect is a dry season of varying lengths depending on relative distance from the equator. Topography provides an additional factor for explaining the valley’s aridity and truncated wet season as well as the sierra portion of Coxcatlan municipio’s relative dryness to similar elevations eastward. A rain shadow effect falls over the Tehuacán Valley, produced by the ranges of the Sierra Madre Oriental to its northeast and Sierra Madre del Sur to its southeast. While the southeastern Tehuacán Valley is only about 150 kilometers in straight line distance from the Gulf of Mexico, these mountains act as barriers against some of the moisture brought by the northeast trade winds during the wet season. Even the more humid sierra portions of Coxcatlán municipio are noticeably drier than similar elevations to their east, as they comprise the final, westernmost slopes of these mountain masses. During the dry season there is virtually no rainfall in the valley, producing a notable vegetative dormancy. Grasses and annual vegetation dry to desiccated husks while the perennial shrubs and trees shed leaves and scale back their water consumption and transpiration (Figure 2.10). All surface water in the smaller barrancas dry up completely and the larger streams like the Río Salado, Río de los Mangos, and Río Tilapa have only small trickles or stagnant pools along their courses. The sequía also greatly affects the human populations of the valley. All dry 39 Figure 2.9 Climagraph for Coxcatlán and Zoquitlán, 1971-2001. The lines represent temperature and colums represent precipitation. Light grey represents Zoquitlán, dark grey respresents Coxcatlán. Graphic by A. Hilburn. Data, CONAGUA 2001 land (non-irrigated) or temporal agriculture is suspended for the dry season (Figure 2.11). Local water resources are taxed and the weekly allotments of household water from municipal springs slow or stop completely in many of the municipio’s valley settlements. Wells run dry. During the sequía, especially in the latter, hotter months of April and May, dry soil kicked up by strong winds accumulates on most household surfaces. The large volume of airborne dust during this time produces a thick, yellowish haze that serves to obscure normally expansive vistas of the surrounding areas. The sierra’s dry season follows roughly the same schedule, but has some (ap - prox. 25-50 mm/month) precipitation from drizzle from October to December and fog through - out the dry season when the valley receives virtually none (2-5 mm/month). The transition from the seemingly endless sequía to the brief wet season begins towards the middle of May in the valley and lasts until early August, where a less intense wet season per- 40 Figure 2.10. Dry-season selva baja caducifolia. Photo by A. Hilburn, October 2011 Figure 2.11. Dry-season temporal field outside Coxcatlán cabecera. Photo by A. Hilburn, October 2011 41 Figure 2.12. Wet-season matorral vegetation outside Coxcatlán cabecera. Photo by A. Hilburn, July 2009 Figure 2.13. Wet-season sugarcane field near Venta Salada. Photo by A. Hilburn, August 2011 42 sists about three to four weeks, followed by a month-long precipitation increase known locally as the veranillo or “little summer” that typically ends by mid-October. The brief wet season is marked by a dramatic and beautiful greening of the valley as well as of the sierra (Figures 2.12 and 2.13). By late June, water courses through the valley streams and springs run with increased volume. Temporal fields are planted and the dooryard gardens around homes produce significant shade and wet-season produce. The clouds of June bring needed relief from the oppressive heat of April and May. But not all aspects of the wet season are welcome. High intensity and long duration rains, notably those from late wet-season tropical cyclones can produce natural hazards. These rainfall events can cause frequent and disruptive road wash-outs and massive landslides. In 1975, the sierra community of Pala4 was buried in one such landslide, causing much of the town to resettle in the cabecera municipal of Coxcatlán, where today its former residents live in a colonia of the same name. The erosion scar from that event is still visible on the landscape. Flash flooding of the barrancas can be a threat to goat herders, their livestock, and playing children (Foody et al. 2004). Excessive rains can also damage irrigation systems and massive erosion events can destroy planted cropland (Vinet 2008). The rainy season also brings about the threat of mosquito- borne illnesses, mainly dengue fever. Public health crews from the state government scour the region to control mosquito populations through fumigation and public education campaigns. While the public awareness of dengue loomed large in the form of posters, wall murals, discus - sions, public health surveys, and the presence of public health workers, I was unaware of any real dengue epidemic in the study area during my stay. Average daily temperature in both the valley and the sierra does not vary too greatly through the year with a peak in May and a low in December or January. In the valley, where I spent most of my stay, the warmest season falls at the end of the dry season, beginning in March, leading up to considerably hotter days and nights in early May, and ending with the first clouds of the wet season later that month or in June. The highest average daily temperature for Coxcat - lán town is 27.8°C in May. Calipan, the second largest settlement and only about three kilome - ters from the cabecera, has very similar temperatures as Coxcatlán town. The valley settlements of Pueblo Nuevo (also known as Guadalupe Victoria), San Rafael, and Tilapa experience slightly but noticeably warmer temperatures due to their lower elevations near the Río Salado and the lower southern portions of the municipio. While not extremely hot, most valley residents com - 43 plain about the late dry season heat, especially those who labor in agricultural fields. In Zoquit- lán’s cabecera, which represents the sierra settlements’ climate, the peak average daily tempera - tures are in May as well, but the mercury only rises to 17.9°C. The sierra in April and May offers a far more comfortable climate for its residents than the Coxcatecos in valley below. The coolest month in Coxcatlán town is January with an average daily temperature of 19.6°C. In Zoquitlán, December is slightly cooler than January with an average daily tempera - ture of 13.7°C. During the coldest months of November, December, and January, strong polar cold fronts, called nortes, can bring winter precipitation or frosts in the sierra (Byers 1967). Near zero or sub-zero temperatures are uncommon and very rare in the valley. Considering human comfort, the winter months offer mild daytime highs around 27°C and daytime lows around 12°C in the valley. This means wearing short sleeves during the day and a sweater at night. Daytime in the sierra during the winter brings comfortable temperatures to around 20°C, but the nights can become quite chilly, around 5°C, especially when no indoor climate control exists besides a hearth and blankets in all of the residences there. Vegetation The Tehuacán Valley and adjacent hillsides are recognized worldwide as an important conservation area due to the high endemism of plant life there. This high endemism was one of the motives for creating the Tehuacán-Cuicatlán Biosphere Reserve in 1998 (Dávila et al. 2002). Fitting with the climatic differences across its breadth, the municipio of Coxcatlán also features considerably different vegetation types as plants have adapted to the different precipitation, temperature, and disturbance regimes (Figure 2.14). A national plant community census (INEGI 2009) recognized six different vegetation types within municipal boundaries but most could eas - ily condense these down to three or four. Some examples of vegetation types are found in Figures 2.15-2.18. Beginning at the highest elevations above 1,800 meters, one encounters forests domi - nated by oaks ( Quercus brachystachya and Q. obtusata ), pine ( Pinus pseudostrobus and P. hart - wegi ), and mixes of each. Typically, oak dominates where more moisture is present. A species of madrone ( Arbutus xalapensis ) is also present at the higher elevations above 2,000 meters (Figure 2.15). Bromeliads such as Spanish moss or heno ( Tillandsia usneoides ) and tecolote or xoloch ( Tillandsia imperialis ), as well as ferns ( Alsophila spp. ) with an array of other canopy-dwelling flora are also present in the canopies where oak dominates (Smith 1969; Acosta et al. 2009). In 44 Figure 2.14. Vegetation communities in Coxcatlán municipio. Map by A. Hilburn. Data. INEGI 2006. the understory and in disturbance zones an assemblage of Ericaceae, Asteraceae and many other assorted annuals, forbs, and grasses (Poaceae) takes hold (Smith 1964). Some cycads ( Dioon spp. ) are present in and just below the forest ecotone (Dávila et al. 2002). In the drier elevations below 1,800 meters one encounters more meso-to-xerophytic plant associations with shorter heights (5-7 meters) and wider umbrella-shaped canopies. The most extensive, covering approximately 80 square kilometers or about one-third of the study area is the Selva Baja Caducifolia or SBC. SBC, by definition, is any warm climate, dense vegetation 45 Figure 2.15. Madrone and oak woods above Xacalco. Photo by A. Hilburn. Figure 2.16. Induced pasture vegetation class in foreground with irrigated sugarcane in the background; Ejido lands, San Juan Bautista Coxcatlán, Cox., Pue. Photo by A. Hilburn. 46 Figure 2.17. Matorral crausicaule vegetation south of the cabecera municipal, Coxcatlán municipio, Puebla. Photo by A. Hilburn, November 2011 Figure 2.18. Selva baja caducifolia below Tequexpalco, Cox. Pue. Photo by A. Hilburn, September 2011 47 assemblage between 7 and 15 meters in height where more than 75 percent of the foliated veg - etation loses its leaves during the dry season (SEMARNAT 2005). This dense vegetation associa - tion in the southern Tehuacán Valley could also be dubbed thorn scrub-cactus forest and is found in the intermediate zone below the pine-oak forests of the sierra and above the sugarcane fields and pasturelands of the valley. SBC tends to exist where there is little or low-intensity distur - bance. Some of the marquee species of this assemblage are cacti. The giant candelabra cactus, Pachycereus weberi 5 , known locally as cardón or candelabro is the most massive cactus in this association. Stenocereus stellatus , or xoconostli, is another large cactus present in large numbers in this assemblage. The chende ( Polaskia chende ) with helically radiating branches produces a popular fruit in the latter part of the dry season and Cereus species, with their tall columnar struc- ture, punctuate the landscape. Various Opuntia species such as the tuna de viboras are present with other species of Mamillaria , Ferrocactus , and Echinocactus as understory. The main tree assemblages are from the Leguminosae, Burseraceae, and Anacardaceae families (Smith 1969). One of the largest and easily recognizable tree species of the SBC is the pochote or Ceiba aesculifolia . The caterpillar larvae of a moth species ( Arsenura polyodonta ), called pochocuiles , live exclusively on pochote trees during the wet season and are considered a local delicacy when boiled and pan-fried6 (Avendaño et al. 2006).The red, smooth-trunked cuajiote are also common as stand-alone individuals. Acacia species like the uña de gato , pirul ( Schinus molle ), and mesquite ( Prosopis luisana ) are also important trees. The aptly-named man- tecoso ( Fouiquieria formosa ) tree with its buttery latex sap and its bright yellow flowers is fairly ubiquitous throughout this assemblage. The cholulo ( Zizyphus pedunculata ) and the chupandía ( Cryptocarpa procera ) are typical trees whose fruits are used to make a natural shampoo and relieve thirst, respectively. The spiny mala mujer or “bad woman” ( Cnidoscolus spp. ) grows near areas with increased soil moisture and has the ability to inflict acute pain from its myriad thin needles on anything or anyone not paying attention (Smith 1969). In terms of structure, the SBC is very dense and in areas of little recent disturbance, the canopy appears closed. Another common vegetation association in the lower elevations of Coxcatlán is what is known scientifically as matorral crasicaule. This vegetation assemblage is fairly similar in species composition to the SBC noted above in Coxcatlán, but has a more open structure of tall shrubs and low trees interspersed with cacti. The “crasicaule” distinction denotes the presence of cacti (INEGI 2007a). The matorral floristic landscapes are characterized by Acacia and Mimosa 48 species, mesquite ( Prosopis luisana ), shrubbery such as Lantana camara and Mexican oregano ( Lippia graveolens ) among many others, all punctuated by cacti species like the ever-present Opuntia species and many of the smaller cacti mentioned earlier. The large cardón and xoconost- li cacti comprise some of the larger flora in this association. Grasses (Poaceae) and small forbs fill in the gaps. Consistent clearing, burning, and erosion have produced in some parts of the valley a vegetation assemblage known as pastizal inducido or induced pasture. This association primarily features grasses (Poaceae) with shrubs such as Lantana camara and fragrant Mexican oregano with scattered Opuntia and Mammalaria cacti. Pastizal typically occurs around human settlement areas that have been cleared, plowed, and grazed for decades (Figure 2.16). 49 PROPERTY, SETTLEMENTS, AND POPULATION Land Tenure and Ejidos Social property, in the form of núcleos agrarios , is the primary form of land tenure in the study area. Eight ejidos comprise almost the entire land property base of the study area. They remain to this day important social-environmental institutions that control the majority of land and surface natural resources in the study area. However, considering the shift toward a more free-market, neoliberal Mexican state in the past 20 years, the degree to which these institutions and the land they control is changing. Much of the impetus behind the Mexican Revolution was the landlessness and the re - sulting lack of a secure economic base for the campesino and worker class. In 1910, fewer than two percent of the population owned 80 percent of available land in Mexico. The Revolutionary Constitution of 1917 and the post-Revolutionary state sought to reform land tenure, the property regime, and possession of natural resources by redistributing land to its agrarian constituents. This was to be effected through the expropriation and redistribution of the large haciendas and foreign-owned land grants to peasant communal groups through the ejido structure (González 2002). In general, this reform tried to return land tenure to a more native structure, albeit a somewhat idealized form. The redistribution of inalienable communal lands to corporate entities composed of peasants also had the effect of garnering the support of the peasant class by the PRI party, which controlled Mexican politics until the 2000 elections (Raat 2004). Following the rati - fication of Article 27 of the 1917 Constitution and the 1934 Código Agrario, land redistribution occurred across the Mexico and ejidos subsequently came into being, with many reflecting extant 50 community boundaries. Throughout the rest of the 20 th century, agrarian reform continued with nearly 104 mil - lion hectares granted to 30,000 núcleos agrarios , a generic term that encompasses all ejidos and comunidades agrarias 6 . These social properties were managed around the portioning of rights to land and access to community resources for each ejidatario . The ejido system became a fully in- tegrated Mexican institution that seemed to reflect the agrarian ideals of the Revolution but also the PRI 7 party that was in charge. The ejido is governed by three entities from within. The asamblea or assembly consists of the all ejido members known as ejidatarios . This body and their vote govern major ejido busi- ness. Decisions are determined by majority vote and the assembly typically meets bimonthly, monthly, or every other month, depending on the ejido . The comisariado is the executive branch of ejido governance and consists of a presidente, secretaria, and tesorero that oversee the practi- cal administration of ejido business and officially represents the ejido in external matters. A new comisariado is elected every three years. An oversight committee, called the consejo de vigi - lancia, is an elected board that ensures compliance with ejido regulations as well as checks the power of the comisariado (Barnes 2009: 395). The land base of the ejido is traditionally divided into basic use zones; however, there is a large degree of diversity as to specifically how land use is allotted within an ejido . Typically, an ejidatario is allotted perennial use to one or a number of parcels in a parceled area and is granted certain use rights to a common use area. However, many ejidos , especially those with extensive holdings that practice shifting cultivation for subsistence, maintain most or all of their lands in common use. Parceled areas usually occur where soil quality is highest and water is more abundant. The common use area is typically larger than the parceled area and is under passive management and covered in forest, scrub, or pasture. A planned or pre-existing urban settlement area of the ejido is set aside as its own distinct allotment within the entire property. In the human settlement area, ejidatarios maintain use rights of their own urban house lot, known as a solar, in cases where there was no pre-existing settlement. At times, some ejidos maintain collective farming, ranching, logging, and even fishing operations and land is set aside for these activities. A schematic of a hypothetical ejido is illustrated in Figure 2.19. How an ejido allots land use is largely left to the discretion and tradition of its members and their chosen livelihoods and no single model can account for the immense variation in land use zoning from ejido to ejido across Mexico. For nearly 60 years, this model was the primary land tenure unit for rural Mexico. Yet, the ejido system was not without major structural problems. First, ejido membership is rarely economically lucrative and many subsistence-based ejidos are in general poor (Gravel 2007; Gledhill 1995). While small-scale, minifundista 8 agriculture and ranching was the dominant eco- nomic activity within ejidos , some based their production in forestry, fishing, or mining. Primary sector activities like these, though, are rarely lucrative and more often are oriented primarily towards subsistence. Depending on the land granted to them, ejidos could be resource and land rich but were nearly always cash poor. The thorough democratic process and collective decision- making enforced by the ejido structure placed barriers for capital and its investment. Addition- ally, the fixed and small economies of scale inherent in ejido production did not make them ideal sites for capital accumulation. Another major problem of the ejido system was the bureaucratic inefficiency of federal land redistribution, which combined with growing populations resulted in Figure 2.19. Hypothetical spatial-land use divisions within a núcleo agrário. Illustration by A. Hilburn. 51 52 a burgeoning landless demographic in rural Mexico (Barnes 2009). After marked economic downturns in the 1970s and 1980s, Mexico’s former patron-client state ideology was giving way to more neoliberal approaches. This shift can be generally attrib- uted to a combination of the rapid fall of global oil prices upon which Mexico had previously hedged development loans, the subsequent economic structural adjustment demanded by credi- tors (Raat 2004), a contemporaneous global retraction from Keynesian economic policy (Harvey 2005), and the emergence of a technocratic political elite in Mexico schooled in neoclassical eco - nomics (Teichman 1995, 2004). This new, market-based policy course set its sights on the agrar - ian sector. Privatization was to be the panacea for what was seen to be faltering rural production. The tool for titling, which was hoped to lead to economic growth through privatization, was PROCEDE ( Programa de Certificación de Derechos Ejidales y Titulación de Solares). The promoters of this project saw communal smallholder agriculture and rural industry as inefficient, non-competitive, and retarding of progress in a globalized economic context (Gates 1996). All of these characteristics were embodied in the ejido structure. The privatization of the núcleos agrári - os through survey, registration, and titling was promoted as a way to give property security and collateral to rural producers. It was also believed that the removal of ejidal governance would make engagement with larger capital interests, such as agribusiness or lumber companies, much more efficient, thereby maximizing the economic potential of Mexico’s agrarian sector (Bartra 2004; Appendi 1998; Del Castillo 1996). Also packed in with land titling and privatization was an end to further redistribution of land and the formation of new ejidos (de Ita 2003). Addition - ally, reduced farm subsidies9 and a scaling back of technical assistance and marketing programs were grafted onto this a larger agrarian counter-reform project of which PROCEDE covered land tenure. Early critics pointed to many problems with the reformation of state-led, post-Revolu - tionary agrarian policy into a general neoliberal, market-led model. Long-term fears were that this would soon signal a re-appropriation of land back into the hands of large landholders or would be swallowed up by urban and peri-urban development interests (Vázquez-Castillo 2004; Appendini 1998; Goldring 1996). Yet, this disposession has not yet occurred to a notable de - gree with ejidos and external private landowners (Barnes 2009; Perramond 2008; Haenn 2006; Nuitjen 2003). The rise of bosses and elites within pre-PROCEDE ejidos has been noted (de Ita 2003). Also, many of the problems hypothesized to occur because of the Article 27 reforms were already extant, such as poverty, increasing out-migration, and overpopulation (Lee 2007). Some communities have maintained their systems of communal governance amid PROCEDE and have used it to clarify boundary conflicts or strengthened and even expanded their land claims and rights (Haenn 2006; Nuitjen 2003). Yet others have seen the commodification of their communal lands and resources through PROCEDE participation and privatization erode their communal structures, diminishing the future social and economic sustainability of the community (Bait - enmann 1998; Baños-Ramírez 1998). While the economic effects of privatization vary greatly by community across Mexico, a common theme regarding its potential negative impact on the environment is apparent (Toledo 1996, Haenn 2006; Liverman and Vilas 2006). Much of this speculation centers on the changing human-environment relationships linked to the capitalization through privatization of communal lands whereby land and natural resources become marketable10 (Lewis 2002; Goldring 1996). This monetary valorization of the environment and its reconceptualization of being a source of marketable natural resources and not a site of cultural reproduction inherent in neoliberal policy poses a threat to invaluable but agro-ecosystems protected by ejido structure (Toledo 1996). The weakening of ejido governance by the 1992 reforms allows individuals or small groups within ejidos to potentially overexploit previously communal resources such as forests, water, gathering zones, or pastures (Lewis 2002; Baitenmann 1998; Zabin 1997). In Coxcatlán municipio, there are eight núcleos agrarios , all of which are ejidos (Figure 2.20). These eight ejidos cover 28,783 hectares which is larger than the entire municipio’s areal extent of 24,866 hectares as the boundaries of some ejidos cross municipal boundaries causing the entire area of its social properties to be larger than the area of the municipio. The largest in size and political influence in the study area is that of San Juan Bautista Coxcatlán, which spans 16,785 hectares and is headed in the cabecera municipal (Table 2.1). This ejido covers the entire central portion of the municipio, from the sierra to the sugarcane fields in the valley. Technically speaking, the San Juan Bautista Coxcatlán ejido includes the lands used by the indigenous sierra communities in the northeastern portion of the municipio, but these communities are allowed to form their own assemblies and conduct agrarian business as annexes largely without much influ- ence from the ejido ’s governing body. This ejido is also the only one in the municipio that did not participate in PROCEDE. The majority of the study area’s ejidos were created in the mas- sive land redistribution during the sexenio (six-year term) of populist president Lázaro Cárdenas 53 54 (1934-1940), when approximately 18 million hectares of land were redistributed to various núcleos agrários across Mexico (Klooster 2003). As a social institution, ejidos are much more than a disappearing corporate group of farm- ers. Here, ejido membership can offer substantial supplemental income due to the availability of inexpensive agricultural labor and the relatively low inputs involved in its production locally. The ejido is also a considerable political institution in the municipio and ejido membership al- lows greater access to local and regional political power channels. It is the ejido , and not the Figure 2.20 Social property (ejidos) in Coxcatlan, Puebla. Map by A. Hilburn. Data, INEGI 2012. Table 2. 1. E jid os in C ox ca tla n, P ue bl a. D at a, P H IN A 2 01 2. 55 56 municipio, which sponsors and pays for the November 20 th celebration in honor of the Mexican Revolution. Lastly, being an ejidatario offers someone a visible higher social status and sense of belonging to the community with the ejido’s ties to the agrarian struggles of the Revolution. Lastly, the ejido system provides the land property base for most of the municipio. Outside of individual home lots, ejido common use lands are where garbage often ends up in Coxcatlán municipio. On the formal side, the municipal dump, located about two kilome - ters south of the cabecera sits on ejido land unofficially set aside for the dump. Most of the land around this growing two–hectare site was formerly temporal (non-irrigated agriculture) in previ - ous, more agriculturally-dependent times but remains important to goat herders. The thorny scrub forest ( matorral crausicaule) on the alluvial fan between the Tehuacán-Huautla highway and the irrigated sugarcane fields bordering the Río Salado are home to a number of the larger tiraderos clandestinos or informal clandestine dumps where residents from indeterminate places take their garbage to dump it en masse. If plans to site a sanitary landfill in Coxcatlán are fully realized, then it is likely the ejido on whose land it will be constructed will likely become involved in its location or outright rejection and resistance. Toponyms Toponyms or place names in central and southern Mexico often have a Nahuatl origin. These toponyms are the result of Nahuatl being the predominant language spoken in central Mexico before the Spanish, the extensiveness of the Triple Alliance (Aztec) tribute state that spoke it, and the collaboration of Nahuatl speakers with the Spanish in the early colonial period. The ethnocentric Aztec state that dominated central and southern Mexico in the 15 th and early 16 th centuries gave its own place names following conquest instead of using the native extant ones. Some were translations of the conquered place into Nahuatl, others were completely dif - ferent and usually referential to some tribute commodity or obvious land feature (Mundy 1996: 139-141). After the fall of Tenochtitlán, the Spanish maintained the Nahuatl toponyms or slightly altered their form due to a lack of Nahuatl orthography or to reflect Spanish phonetic preferences during the colonial era. Today, Nahuatl or Nahuatl-based place names exist in places where it has never been nor is spoken. The majority of the settlement names in Coxcatlán municipio are Nahuatl in origin. It is unknown whether or not Nahuatl was spoken in Coxcatlán at the time of conquest by the Span - ish, but the señorio of Cuzcatlán was known to be under tribute to Tenochtitlán, where Nahuatl was spoken. The name of the municipio and cabecera means “place of necklaces”, perhaps referring to a regionally renowned prehispanic lapidary industry there. In the valley, Calipan and Tilapa also bear hispanicized Nahuatl etymologies, whose meanings indicate “place of the high house” and “Black River” respectively. Pueblo Nuevo and San Rafael, the other two settlements in the valley, were founded in the 1930s and have Spanish Catholic and Mexican national place names. The settlements of the sierra have almost exclusively Nahuatl toponyms. However, it is unclear whether these names date to prehispanic or even colonial times as none are found on any precolonial or colonial document. They may, in fact, be more recent place names, from per- haps the colonial period to today, given by the local populations who happen to speak Nahuatl. It was recounted to me by area residents that many of the smaller sierra hamlets did not really have names until the early 20 th century. A few hybrid Nahuatl-Spanish toponyms exist in the municipio with one example, Tres Ocotes or “Three Pines”, being a Hispanicized version of the Nahuatl “ocotl” which signifies “pine”. Besides settlements, many of the physical features in the study area and surrounding region such as mountains, hills, and streams bear Nahuatl names. A list of select toponyms with select etymologies is found in Table 2.2. Table 2.2. Nahuatl toponyms in Coxcatlán municipio. Translation from Franci 1976 and Alva Alva 2011. 57 58 The cabecera municipal: Coxcatlán Coxcatlán or Coxca11, as it is called locally, is the cabecera municipal of the municipio and the largest settlement in the study area. The breadth of Coxcatlán’s urban area spans over two kilometers east-west and about one kilometer north-south on an alluvial fan and adjacent rolling uplands between the valley floor below and the sierra above town. The town sits on a considerable but even slope that goes upward about five percent from the west with a difference of over 100 meters from the lowest (~1,000m asl) to highest (1,125m asl) point. The barranca Soyolapa flows through the town into the Río Salado below and to the west. A car bridge and an elevated metal footbridge span its mostly dry course. Local directions are rarely given in cardinal terms, rather “up” or “down” or on this or that side of a barranca. For the most part, Coxcatlán is laid out in a rectilinear grid pattern, centering on a small, approximately 2,000m 2 plaza, referred to locally as the parque surrounded by its church and gov- ernment building, a pattern found in most Latin American urban areas (Stanislawski 1947) (Fig - ure 2.21). The palacio municipal also called simply ayuntamiento houses the municipal govern- ment and occupies the east end of the parque . The recently frescoed and painted Catholic church, which dates from the late 16 th century, sits on the park’s south side. Parish buildings and a small courtyard lie within the walls of the church compound. Residences and businesses surround the rest of the park and a kiosko or bandstand occupies its center. Mangoes and mimosa trees provide shade over the park’s palimpsest of beautification projects from past municipal presidencies. Figures 2.22 to 2.27 offer a flavor of urban landscapes in the cabecera. The parque is the commons of the town and is the social center of Coxcatlán where people of all ages come to meet, talk, hang out, buy from street vendors, and attend public and religious functions. Town parades begin and end at the park. During the patron saint holiday (De - cember 24-27), Holy Week (varies), Independence Day (September 15-16), and the commemora - tion of the Revolution (November 20), the park is filled with games, carnival rides, music events, and vendors. Public announcements and the monthly social welfare payments attract large numbers as well. A covered market building, about half the size of a football field, shares a wall with the church compound and is the site of Coxcatlán’s Sunday market in addition to wedding receptions, quinceañeras , and bailes or town dances. Coxcatlán is subdivided into 16 colonias or neighborhoods. The largest colonia, Colonia Centro, occupies the urban core and extends to the historical limits of the town. Its paved street Fi gu re 2 .2 1. U rb an a re a an d co lo ni as o f C ox ca tla n, C ox ca tlá n, P ue bl a. M ap b y A . H ilb ur n. D at a, IN EG I 2 01 0, 2 01 2; F ie ld d at a 20 11 . 59 60 grid follows the standard grid pattern found in most Latin American towns and cities. Colonia Centro, unlike many other urban areas in the Tehuacán Valley and adjacent sierra, has preserved to a large extent its colonial architecture. These buildings are usually one-story with high ceil- ings, rectangular, and constructed around an open patio area. The walls are of painted, frescoed adobe and can be up to a meter in thickness. This architectural style is ideally suited to providing cool spaces during Coxcatlán’s hot days and adequate protection from its cool nights. Colonia Centro also has greater shade than adjacent colonias from plantings of mango, tempesquixtle, Indian laurel, and various zapote varieties. Colonia Centro is arguably the most affluent section of town and its house lots are typi- cally larger than those found in the peripheral colonias. Going only on anecdotal evidence, rents and property values increase with proximity to the town park. Most of the property owners in this part of town are natives of Coxcatlán and from long-entrenched families. Compared to the more peripheral colonias, Colonia Centro is almost completely non-indigenous. Colonia Centro has the closest proximity to common amenities such as best-stocked abarrotes or general stores, the town market, tortillerías , and recreational spaces. It is also the most well-connected by bus and taxi service. Municipal water is typically more available there than most other colonias. In response to natural growth and population influx from migration from the nearby sierra communities, Coxcatlán’s population expanded considerably in the 20 years from 4,863 in 1990 to 6,577 in 2010 residents (Table 2.3). As the demand for housing grew, land adjacent to the town was allotted by the ejido for transformation into lots to be sold to newcomers. These allotments are today’s peripheral colonias. The relative age of these colonias increases with distance from the center of town as most of they were settled gradually in response to petitions for housing. One exception was that of Colonia Pala, also known as Palita (Figure 2.19), which was settled more rapidly due to the mass resettlement of the sierra community of Pala following a 1975 land - slide. The residents of Palita have maintained a strong community identity that has allowed them to be comparably more successful in obtaining paved streets, improved urban infrastructure, police protection, and a primary school than older, more established colonias. Currently, there are six colonias, 20 de Noviembre, Cruz de Palo, La Joya, Ojo de Agua, El Pedregal, and San Isidro, which are experiencing active growth and settlement intensification. Very few, if any, lots are open for development in the remaining nine. In the centro-adjacent co- lonias, the infrastructural development is almost indistinguishable from the urban core. Outside Figure 2.22 Looking south to Coxcatlán’s centro from Colonia Calvario. Photo by A. Hilburn, June 2011 Figure 2.23. Looking north to Coxcatlán’s centro from Colonia Benito Juárez. Photo by A. Hilburn, February 2011 61 62 Figure 2.24. Busy morning in the parque (plaza) in the center of town. Photo by A. Hilburn, Month unknown 2011 Figure 2.25. Ayuntamiento building during park refurbishment. Photo by A. Hilburn, August 2013 Figure 2.26. Looking west at Colonia San Isidro and Colonia Donato Bravo Izquierdo. Photo by A. Hilburn, Sep- tember 2011 Figure 2.27. Colonia Benito Juárez streetscape. Photo by A. Hilburn, March 2011 63 64 Table 2.3. Population by settlement, 1990-2010, Coxcatlan, Puebla. Data, INEGI 1990-2010. of the center, in the more peripheral colonias, infrastructure is typically of an ad-hoc nature with unpaved streets, unburied tubed water or open wells, and few street lamps. Some lots have yet to be cleared of monte and may be still used as pasture for goats. Electricity is found nearly in every household ( 99 percent ) due to the considerable efforts of the federally-controlled electric com- pany, La Comisión Federal de la Electricidad (CFE), in regularizing and subsidizing electricity accessibility and affordability. Water provision in the dry season remains a problem in some of the newer colonias. Homes made of cement blocks, referred to locally as block or bloq , and thin-walled, un- frescoed adobe brick are the typical architectural forms in the peripheral colonias. The lots are less built-up, often with no external walls and a simple small fence. Some are even planted in milpa (corn field) or some other garden or tree crop. Mesquite and Acacia trees and wild shrubs are the common urban flora of the colonias as opposed to the centro’s older growth fruit trees. Some larger, compound-style homes in varying stages of construction also can be found in the colonias, the result of remittances sent home by family members in the United States or larger Mexican cities. In the farthest reaches of the colonias and even beyond their boundaries are camp houses built by ejidatarios ( ejido members) for relaxation, family events, or rest during agri- cultural work. Most colonias have a patron saint with a dedicated shrine and a small chapel. A colonia-wide sense of community also seems to be shared among residents. While affluence varies across and within colonia households, they are noticeably less af- fluent than the centro and some are impoverished by even Mexican standards. Many of the newer colonias are home to residents from the adjacent Sierra Negra communities who have migrated in search of work in the sugarcane fields of Coxcatlán or to provide a more comprehensive edu- cation for their children. Some of the female residents of these colonias work as domestic labor- ers in many of the homes in Colonia Centro or wealthier neighbors. The peripheral colonias as a whole are more indigenous than those surrounding the urban core of Coxcatlán. Property and land titling in the colonias can be a complicated issue. The degree to which residents have established title security to the land on which their houses and businesses sit var- ies depending on where one resides. Much, but not all, of the land occupied by today’s colonias technically remains a part of the land held in communal trust by the ejido San Juan Bautista Cox - catlán. While no future conflicts with regard to land ownership are foreseen, the lack of certainty in urban land ownership in Coxcatlán could be problematic with expanding urban growth and the changing dynamics of the social property regime from a government-backed entity to a more land commodity ideal in Mexico. Calipan Calipan, two-and-a-half kilometers due north of Coxcatlán (Figure 2.2), is the second largest urban area of the municipio with 4,211 residents as of 2010 (INEGI 2010). Situated at about the same elevation as Coxcatlán, the most recognizable aspect of Calipan is sugar produc - tion. The smokestack from the large ingenio or sugar refinery is visible from miles away and the refinery itself occupies the most central acreage of the town (Figure 2.28). While Calipan was 65 66 the site of a pre-Hispanic village, colonial and contemporary Calipan was built around the sugar industry owing to its relative abundance of water. Irrigation canals and raised aqueducts that provide water to the adjacent and interwoven sugarcane fields run openly throughout the town. A large open reservoir owned and used by the ingenio is a notable feature of the town’s urban landscape. Another important industry resulting from Calipan’s water abundance is water puri - fication. A number of small, household-sized potable water producers and distributors provide more economical, but less-purified drinking water options for area residents. Calipan is home to a branch campus of CONALEP (Colegio Nacional de Educación Profesional Técnica), a federal - ly-run vocational-technical junior college. While Calipan is under the administrative jurisdiction of Coxcatlán, the town in general maintains an air of independence and cultural uniqueness from its neighbors. Calipeños, as they are known, have at times protested the imposition of auxiliary presidents selected by the mu- nicipal president, who almost always is from Coxcatlán town. While I was there, I witnessed multiple small protests by Calipan residents against the appointed auxiliary president as well as the municipal government based in the cabecera. In terms of the agrarian sector, Calipan’s ejido Figure 2.28. Calipan from the sierra above town. Sugar ingenio is in the foreground. Photo by A. Hilburn, October 2011 was given a much smaller allotment of land than nearby Coxcatlán, generating some resentment and competition between the two towns. Although only separated by a few kilometers, Calipeños typically speak with a distinct accent, similar to one found along the Veracruz coast that is anec - dotally attributed to the influx of Veracruz sugar workers and Afro-Mexicans in the 19th century. Calipan is divided in four colonias and a Colonia Centro. Many of the characteristics between Coxcatlán’s centro and its peripheral colonias are similar in Calipan. Calipan lacks the historical architecture of Coxcatlán, however, and its urban plan is largely built around the sugar refinery and irrigation canals. Calipan is surrounded by the ejido ’s sugarcane fields to the north, east, and west. Current urban growth is occurring southward towards Coxcatlán. One colonia in Calipan is partially sited on land donated by the San Juan Bautista Coxcatlán ejido . In spite of having paved streets and organized physical infrastructure, another colonia, Colonia Obrera, was built to house sugar workers in the 1970s, but sits largely abandoned due to its isolated location from the center of town. Tilapa, Pueblo Nuevo, and San Rafael Highway 980 leaves the barrancas and gently sloping alluvial fan covered in selva baja and enters the valley sugarcane lands of the towns of Guadalupe Victoria approximately five kilometers south of Coxcatlán, known everywhere besides the agrarian registry as Pueblo Nuevo, and San Rafael (Figure 2.2). Here, the climate is noticeably warmer and these towns were found - ed and exist today primarily for the planting, cultivation, and harvest of sugarcane instead of an agreeable climate or some aesthetic value. Pueblo Nuevo (pop. 416) is a small, dusty linear grid of about three main streets intersected by six smaller ones aligned along the highway between the sugar fields west of town and Cerro Colorado, a small but steep ridge about 100 meters above town named for the reddish color of the its rock (Figure 2.29). The settlement is an asentamiento humano zone for the ejido of San José Tilapa. Most people work in town work in the sugarcane fields, either on their property or as chalanes or field hands. The ejido of Tilapa on whose land Pueblo Nuevo sits and subsequently the town was created in the late-1930s from the expropri - ated lands of a nearby hacienda. The town features a primary school and is accessible by bus and urban taxi from Coxcatlán. San Rafael, three kilometers south of Pueblo Nuevo, is similarly attached to the highway between the sugarcane fields below and to the west and the selva baja above and east of town. 67 68 Figure 2.29. Pueblo Nuevo from Cerro Colorado.Photo by A. Hilburn, September 2011 Figure 2.30. Newly paved street in San Rafael. Notice the lack of garbage. Photo by A. Hilburn, September 2011 San Rafael is a very small town with only 258 residents in 2010. The town layout consists of four blocks west and two parallel streets east of the highway with a primary school, a town hall, a wa- ter tower, and a small Catholic church (Figure 2.30). Like Pueblo Nuevo, San Rafael sits on the old hacienda lands that now belong to the ejido in Tilapa. Also like Pueblo Nuevo, the town is deeply involved in the cultivation of sugarcane and is surrounded by some of the best sugarcane lands in the ejido . The most notable thing about San Rafael besides sugarcane is its orderliness. The streets are clean, almost no litter or stray dogs are visible, and each resident’s lot is ordered and maintained. This cleanliness is produced by a strict enforcement of a set of customary town rules known as usos y costumbres whereby every household has to participate in weekly cleaning and tidying regimens or face fines. Tilapa, the third largest town in Coxcatlán with 2,081 residents, is another six kilometers south of San Rafael. Situated in a strip of the valley amid dissected canyonlands, Tilapa forms the southern terminus of the municipio and is only four kilometers north of Teotitlán de Flores Magón, Oaxaca. Listed on the Relación Geográphica of Coxcatlán as a tributary town in 1580, a village in the vicinity of Tilapa has existed since pre-colonial times (Schwaller 1987; Mundy 1996). Just to the north of town is the archaeological site of Presa Purrón, the famous ancient irrigation reservoir. Like the aforementioned two smaller towns within its ejido , Tilapa’s local economy is also dependent on the cultivation of sugarcane. A closed but still-maintained ingenio lies just below the town hall in the center of town. Before it closed in the late 1960s, the mill used water stored from the Río Tilapa, whose course runs along the north side of town. Tilapa is arranged along streets that follow a crooked grid pattern that radiates outward from a park, a town hall, and a salon ejidal . The Sierra Communities This dissertation presupposes a division within the municipio between the valley and the sierra regarding their population and settlements. Divided not only by the impressive physi - cal boundary that the steep westernmost ridgeline of the Sierra Negra range provides, the sierra communities in Coxcatlán municipio are notably different from the aforementioned valley com- munities in terms of their populations, language, economies, culture and social organization, and urban morphologies, to name a few attributes germane to geography. Notions of environmental determinism and facile spatial relations between landscape and 69 70 ethnicity not implied, one of the most obvious things one notices in Coxcatlán municipio is what appears to be an almost totally indigenous-speaking sierra and largely non-indigenous valley. Using people over five years of age who speak an indigenous language, in this case Nahuatl, as a proxy for some notion of “indigeneity”, one can see a fairly deep divide between what is almost total Nahuatl fluency in the sierra and very small percentages in the valley communities. Yet, calling Nahuatl speakers in Coxcatlán’s sierra “indigenous” is not entirely uncomplicated. By the definition of the federal indigenous affairs agency, Centro del Desarollo de Pueb - los Indígenas , an indigenous home is defined as any home where any family member speaks an indigenous language (CDI 2006). Yet what it means to be indigenous today goes beyond language and transcends any simplistic attribution of previously observed cultural traits with a particular indigenous “group”. In my experience there, most people who speak Nahuatl as a first language would not likely first identify them as indigenous. “Serrano ”, “ de la sierra”, both of which mean “from the sierra”, are common responses and more often they would express an identity with their home town than say, “ Soy indígena ” or “I am indigenous”. Most serranos do recognize their distinctness in terms of language, culture, and customs, yet a direct identity as “indigenous” is just not as apparent in the study area as opposed to other, arguably more politi - cally active areas of Mexico where autochthonous languages are spoken, like in parts of Oaxaca and Chiapas states. The non-indigenous people in the region as well as the government would classify residents of the sierra communities as “indígenas” by their use of Nahuatl and Nahuatl- inflected Spanish, slightly different physical stature, involvement in smallholder agriculture, rela- tive poverty, and more traditional and pragmatic approach to dress. Most of the sierra communities, with the exceptions of Pala and Vigastepec, are situated a few hundred meters below the ridgeline with maintained dirt roads leading to the paved highway above. A few, such as Barranca Vigas, Tequexpalco, and Cobatepec have no paved road access and are connected by dirt road to a valley settlement like Tilapa for the former and Coxcatlán for the latter two. The population size varies considerably among the settlements in the sierra with towns of over 1,000 inhabitants to small hamlets of about 100 people. Tecoltepec (pop. 1,292), Ocotlamanic (pop. 1,112), and Xacalco (pop. 960) are the largest towns in the Coxcatlán sierra (INEGI 2010). In response to their growing population, they have extensive settlement mor - phologies covering vast areas of up to approximately 3.5 square kilometers stretching over 700 meters in vertical relief in the case of Xacalco. The urban subdivisions here are called barrios as Figure 2.31. View from above Ocotlamanic on the road from Tepeyoloc. Photo by A. Hilburn, October 2011 Figure 2.32. Looking toward the town of Xacalco from a house lot. Photo by A. Hilburn, October 2011 71 72 Figure 2.33. Tequexpalco homes with near-lot fields. Photo by A. Hilburn, October 2011 Figure 2.34. Rancheria Pala with Coxcatlan in the background below. Photo by A. Hilburn, October 2011 opposed to colonias in the valley, although their meanings as sub-town area units are not visibly different. A multi-story town hall, which serves as the meeting place for the auxiliary municipal officers, is usually located at the center of town. A Catholic church, a covered auditorium, and some of the towns’ schools are typically the center as well. Some streets, mainly those near and around the town hall/church/auditorium complex are paved in cement while the rest of the town proper is accessible by dirt road and footpaths called veredas in Spanish. A few storefronts are present and some homes maintain small assortments of household goods for sale, but there is very little retail or service economy there. Smaller sierra towns in Coxcatlán municipio range in size from about 100 to 600 resi - dents. The larger towns in this class, such as Chichiltepec, Tepeyoloc, and Tequexpalco have similar, but less extensive layouts to the largest towns in the sierra with public buildings, an audi- torium, and a church. The smallest hamlets in the sierra, those like Tres Ocotes, Cobatepec, and Calpuhuacán usually only have a primary school for a public building and are too small to have a permanent Catholic church. One or two dirt roads with extensive footpaths weave their way through town. A few houses sell goods but no storefronts are present. Figures 2.31 to 2.34 offer a view of the sierra townscapes. House lots throughout the sierra communities exhibit a fairly uniform pattern. The typical size is from a quarter to half of a hectare. Most are planted in an array of fruit trees to optimize seasonality, ensuring a fairly year-round supply of fruit. Some have small gardens of vegetables, ornamental trees and plants, and some houses keep chickens, turkeys, or small livestock. Bur - ros are common. Small milpa plantings and a few rows of flowers for cash sales or donation to the church often attach to home lots. A few homes, far less than in the valley settlements, have automobiles or trucks. Most houses have electricity and tubed water. Some of the more affluent homes have indoor plumbing but outhouses and wash houses are more common. Efficient modes of transportation are considerably more limited for the sierra communi- ties of Coxcatlán municipio. The impressive slopes between the sierra and valley limit transpor- tation between the sierra communities and the valley to two paved roads, one of which requires leaving the state of Puebla briefly before turning back across the state boundary and into the sierra. One-lane dirt roads and footpaths comprise the remainder of the transportation network between the valley and sierra. These narrow, at-times precarious, and winding trails remain important to many households to reach their agricultural plots, herd goats, and travel between 73 74 communities. Between the Valley and Sierra in numbers The following chapters note the differences between households from the sierra and valley portions of Coxcatlán municipio in terms of how they manage their garbage as well as how they perceive garbage as an environmental issue. The distinction between these two por- tions of the municipio is apparent from observation as well as statistical data. Sierra and valley households have unequal access to municipal infrastructure and resources and thus manage their garbage differently. Given these practical and experiential circumstances as well as place-based factors that influence perception of garbage-related environmental issues and risks, residents from these two population subdivisions within the municipio are likely to give differing opin- ions on what they perceive is most important regarding garbage and waste in their household and community. Differences in consumption and material possessions, of course, do exist within these two subsamples. The following is a brief analysis using household demographic statistics from the 2010 population census to highlight material differences between households in valley and sierra settlements in the study area (INEGI 2010). Indigenous language use, religious affiliation, and education attainment all provide insight into aspects that inform opinions that are surveyed in later chapters. Household-level demographics illustrate the material conditions of households in both subsamples. The number of rooms per house and the number of occupants per room indi- cate the size of the home and the value of personal space there, shedding light on the relative affluence of a typical household. Electricity and piped water are perhaps the most basic utilities and their presence serves as a proxy for state-defined development. Additionally, the presence of particular key household goods such as televisions and automobiles serve similar inferences about a household. Conversely, the complete lack of a more comprehensive list of household goods used by the Mexican census bureau, INEGI, intimates material poverty and a rejection of certain modern household wares. A simple difference of means test statistic, Mann-Whitney’s U, indicates where, if anywhere, these differences exist. The most obvious statistically significant difference regarding individuals in sierra and valley settlements is indigenous language use (Table 2.4). Almost the entire sierra subsample, 96 percent, speak an indigenous language, mostly Nahuatl, and only about seven percent speak an Table 2. 4. In di ge no us la ng ua ge s pe ak er s, re lig io us a ffl ili at io n, a nd e du ca tio na l a tt ai nm en t b y se tt le m en t, IN EG I 2 01 0. 75 76 Table 2. 5. H ou se ho ld s si ze , fl oo r t yp e, e le ct ric ity , p ip ed w at er , t el ev is io n, a nd c ar o w ne rs hi p by s et tle m en t, IN EG I 2 01 0. indigenous language in valley settlements. Differences in educational attainment are also notice - able and statistically significant with higher attainment (7.8 years) in the valley subsample than that of the sierra (4.4 years). However, the generational differences across the entire population are perhaps greater and more significant than spatial ones with the increase in access to educa- tion over the past generations (Santíbañez, Vernez, and Rasquin 2005). There are no significant or noticeable differences regarding religion between the sierra and valley subsamples. Roman Catholicism is the predominant religion in the study area, claiming just fewer than 90 percent of the population. Slightly fewer than a tenth of the study area follows a non-Catholic Christian de - nomination, referred to generally as Cristiano, which can mean an array of evangelical churches and Jehovah’s Witnesses. Data at the settlement level highlight significant differences among dif- ferent communities that would have different exposure to and acceptance of non-Catholic Chris- tianity. There are significant differences between valley and sierra households and their mate- rial possessions (Table 2.5). The houses are one such differing attribute, where the percentage of households with two or more rooms is significantly greater in the valley. Conversely, the number of occupants per household and room is significantly greater in the sierra, where there is a higher fertility rate and greater material poverty among the largely indigenous population in the sierra. The valley settlements have higher percentages of houses with covered floors with 87 percent compared to just fewer than 66 percent in the sierra subsample. With regard to household appli - ances, the valley settlements all have significantly greater percentages of households with tele- phones, cell phones, cars, refrigerators, and kitchen sinks. Cars are significantly more present in valley households as well. Predictably, the percent of households without any tabulated goods are greater in the sierra subsample. Piped water and electricity, the most basic public utilities in the study area show no significant difference between the sierra and valley. While access is greater in the valley than the sierra, the efforts of the nationalized electrical company Comisión Federal de Electricidad to electrify almost the entire country and promote its usage have been largely successful. There is almost no difference between subsamples regarding tubed water arriving from springs and tanks. Across the municipio, just above three-quarters of the population has access to this utility which is typically either free or very cheap (approx. 150 pesos/year or $10.80 US/year). Peripheral co- lonias in the valley settlements and particular sierra communities have decreased access to piped 77 78 water because of both an infrastructural lack and scarcity. Vertical wells serve the homes without tubed water. Television and radio are more present in valley homes than sierra ones but there is no significant difference between them. Government census statistics only really report on things people have or do. These data are useful in showing standardized differences between and among populations. Yet the need to generalize the complexity of things such as population, culture, and economy to numerical or ordinal data discounts more nuanced illustrations of daily life in particular place. The following brief section outlines the practices and activities of people in the study area. DAILY LIFE AND ECONOMY Agriculture in the valley While the amount of land being used for agriculture and number of people involved in it is declining in Coxcatlán municipio, as well as across Mexico, it still remains the predominant economic activity there. What defines agriculture and other land-based economic activities is a very diverse set of practices producing both food and agroindustrial commodities. Out of the 24,866 hectares of the entire municipio of Coxcatlán, just over 40 percent (10,034 hectares) are regularly active in primary sector work, from farming to small-scale ranching and plant resource- gathering. Over half of this area, or 5,471 hectares are actively planted or improved farmland (INEGI 2007b). Coxcatlán is best recognized as a sugar-producing municipio. Hacienda sugarcane pro - duction arrived in the early 18 th century (see previous chapter) and continued after the Mexican Revolution. Even after the expropriation of hacienda lands in the early 1930s to form the three ejidos of Calipan, San Juan Evangelista Coxcatlán, and San José Tilapa, the valley’s irrigated lands remained almost entirely devoted to sugarcane growing. The present sugarcane lands cor- respond roughly to the immediate few river terraces of the Río Salado and the most proximate sections of its tributaries (Figure 2.35). While the land of the ejido is owned corporately by all of its members, each ejidatario maintains perennial rights to their particular parcel(s), which tend to be anywhere from one to three hectares in size on average. Evapotranspiration in the southern Tehuacán valley is high, precipitation is seasonal, and sugarcane requires high water inputs and year-long maturation. A consistent water supply and irrigation infrastructure are prerequisites. All of the lands under cultivation for sugar are irrigated by earthen canals fed by both springs and access to the Río Salado (Figure 2.36). This extensive network of canals is fed by perennial springs, both within and outside of the ejido boundaries. The source canals are owned collectively by the ejido . These are typically one-to-two meters wide and a meter deep. These canals travel primarily above ground but some can cover distances of up to a few hundred meters below ground through engineered tunnels. In the municipio’s sugarcane lands, the large source canals are named after a patron saint, a Nahuatl toponym, or a famous person. Access to a particular canal’s water is granted by the ejido authorities through an aguador or waterer. Each parcel of land is allotted a particular time of irrigation per month measured in hours. The amount of hours a parcel receives depends upon its relative location to a source canal or a natural watercourse. These irrigation events are coordinated according to a predetermined schedule to maintain optimal flow through the source canals, varying from 12 (driest parcels) to 48 (wettest parcels) hours per month. Incidental irrigation can be petitioned if upstream watercourses are plentiful or if a rare rain event occurs in the dry season. A sugarcane parcel requires considerable engineering before it is ready to be sown and irrigated. Most parcels are terraced to conform to the varying slopes surrounding the Río Salado and its tributaries and to allow for irrigation water to flow through the entire parcel (Figure 2.37). Figure 2.35. Sugarcane lands near the Rio Salado, Coxcatlán, Puebla. Photo by Alison Hadley, June 2011 79 80 The slope of a parcel determines the width of a terrace with steeper parcels having multiple, shorter terraces. A moderately sloped parcel will have terraces 12 – 15 meters wide. Each ter - race is separated by a borda or heavy earthen wall made by compacting soil into a linear berm. Bordas are typically 0.3 meters tall on the uphill side and 0.5 – 0.75 meters on the downhill side in areas of moderate slope. More sloped parcels obviously have greater downslope heights or are more frequently terraced. Each parcel accesses irrigation water from the source canals through concretized sluice gates called compuertas adjacent to the canal. The compuertas are usually 0.5 meters by 0.5 meters, and are opened and closed using a replaceable wooden gate that slides out from above. Irrigation water, once the compuerta is open, flows along smaller canals called apantles that direct water to the different terraces on the parcel. Terraces are rowed and graded so that water is evenly distributed across its area. The irrigation infrastructure on a particular parcel requires constant maintenance and is the responsibility of the parcel holder. The source canals are maintained by the ejido , however all water resources in Mexico are controlled by the federal government according to the Ley de Aguas Nacionales or National Water Law of 1992. Coop - eration between ejidos regarding water usage is enforced by the federal Comisión Nacional del Agua (CONAGUA) when and if conflict arises. Sugarcane ( Saccharum officinarum) is a perennial crop. In the study area, it is harvested once per year and allowed to regrow until the next harvest, usually 10 – 12 months later. Con - secutive harvests decline in yield annually and after 8 - 10 years, a parcel should be replanted to maintain sufficient productivity. New plantings are made by burying pre-cut budded stalks called semilla or seed parallel to the rows. Two varieties, caña morada and caña enicot are used in the study area ejidos . Caña morada is believed to give a typically higher yield and is preferred for its taste as a raw product, but caña enicot is thought to be more drought-tolerant and durable against soil compaction. Weeding, periodic composting or fertilizing, and pest control improve yield. Parcels are irrigated during 12-hour sessions and usually at night. After maturation and ideally before the plant flowers, the stalks are harvested. The harvest, called the zafra , occurs every year from mid-November to the end of Janu - ary. Workers contracted by the ingenio arrive to a parcel with the parcel holder supervising. The parcel is then burned to remove the leaves and dried vegetative matter around the stalks, making it easier for the cane harvesters to manually cut the stalks with machetes. Harvested stalks about 1.5 meters long are then stacked in bundles parallel to the field rows by the harvesters where a Figure 2.36. Main source irrigation canal, Coxcatlán, Puebla. Photo by A. Hilburn, December 2011 Figure 2.37. Terraces and bordas with freshly planted sugarcane, Coxcatlán, Puebla. Photo by A. Hilburn, October 2011 81 82 tractor with a mechanical arm can pick up and transport the bundles to where dilapidated heavy trucks await. Once the truck is full, it leaves for the ingenio. Besides the local ejiditarios growing a raw agricultural input, the ingenio or sugar refin- ery in Calipan produces refined sugar, industrial alcohol, and a few other sugarcane-based indus- trial products (Figure 2.38). The ingenio is the largest buyer as well as primary financier of local sugarcane production. It also sells the stalks and contracts the labor for planting. The ingenio pays each grower a predetermined price per metric ton that is agreed upon by the ejido authori- ties the year before the harvest. In this sense, a sugarcane grower sells his or her crop prior to the zafra . The individual harvest is weighed by the ingenio and payments are made after subtracting fees for the harvesters, trucking, and any other inputs provided by the ingenio. Following the harvest, sugarcane growers prepare their fields for the next harvest. The dried tops and leaf litter that did not burn remain in the field after the sugarcane is removed. This litter is organized in rows that follow the bordas and is then burned to diminish its volume as well as return carbon to the soil (Figure 2.39). Weeks later, some growers lightly plow between the rows to aerate the soil following compaction by the trucks and tractors used in the harvest. Periodic weeding while the new stalks are growing can help optimize yields. Producing a sug - arcane crop is labor intensive during particular stages of its growth. Weeding is typically done manually with hoes and shovels. Extracting the grass and weeds from every row within every terrace is exhausting work. Manuring and chemical fertilizer application also requires consid - erable labor as tractors are not typically used beyond tilling and soil preparation. The bordas that separate the terraces need to be maintained to hold in the irrigation water and considerable quantities of soil must be shoveled and compacted to keep these walls strong. Apantles can erode their courses and must be maintained and the concrete compuertas must also be rebuilt when the cement decomposes. Incidental tasks like pest control and stone removal add to the la - bor invested into a parcel. For more affluent parcel holders, day labor is used exclusively for the maintenance of a parcel. Many of these day laborers come from the sierra communities to work temporarily for cash or from the poorer colonias of the valley settlements. For the parcel holders who do work their fields, day labor is often hired for post-harvest maintenance or weeding. Pay is low, usually around 80 – 100 pesos or 5.75 – 7.25 US dollars per workday, which lasts from 7:30 a.m. to around 3 p.m.. Besides sugarcane, other market-oriented crops are produced in the irrigated zone. Some Figure 2.38 Ingenio Calipan trucks waiting to be unloaded, Calipan, Cox., Puebla. Photo by A. Hilburn, November 2011 Figure 2.39. Burning the remnant husks after the harvest, Coxcatlán, Cox., Puebla. Photo by A. Hilburn, Novem- ber 2011 83 84 parcels are planted in milpa, mainly for the production of dry-season elote or sweet corn. A few ejidatarios grow watermelons and cantaloupes. Growers of these crops generally maintain greater vigilance in their fields to thwart theft. Yet the risk involved in planting these highly per- ishable, more-difficult-to-market crops keeps the majority of parcel holders in the valley growing sugarcane. A few people grow tomatoes in invernaderos or greenhouses, but the capital invest- ment in the greenhouses as well as the competition from factory farmed tomatoes keeps these enterprises small and few. Outside of the irrigation zone in the valley, dry land or temporal agriculture is practiced during the wet season. These fields are planted almost exclusively in milpa on ejido lands previ- ously covered in monte. Some enterprising ejidatarios plant temporal fields but those without eji - datario status can petition the ejido for the rights to plant in exchange for small rents paid in cash or produce. Depending on the parcel, these fields are planted for a few years before their produc- tivity diminishes greatly. Anecdotally, residents noted that in previous generations much of the land now covered in native vegetation was used in shifting temporal agriculture. The shrinking of small-scale agriculture across Mexico due to a lack of profitability and a growing dependence on industrial food sources combined with the greater restrictions on land use imposed by the federal biosphere reserve there are likely causes for this decline in temporal agriculture. A few orchards of mixed zapote negro ( Diospyros digyna ), zapote amarillo ( Pouteria campechiana ), chicozapote ( Manilkara zapota ), and mango ( Mangifera indica ) are scattered across the valley in areas with nearby springs or irrigation canals. Goats, locally called chivas , are the primary small livestock animal of the valley portion of the study area. Goat ranching is almost exclusively small-scale with herds ranging from 10 – 30 individuals. These animals are ideally suited to the arid climate of the Tehuacán Valley. Goats can tolerate long periods of activity with minimal water (Pandeya and Solanki 1991). They also can forage on most any plant mater as well as garbage. During my field work, I observed goat herds feeding on illegal dump piles in the monte surrounding the cabecera municipal. Their abil- ity to eat just about any plant matter makes goat grazing an affordable, low-investment livestock. This same ability to consume most plants, even cacti, makes their herds a considerable threat to the protected vegetation communities in the biosphere reserve as well (Parkswatch 2002: 9). Pas - tores or goat-herders with the assistance of dogs pasture their animals in the selva baja caducifo - lia and matorral vegetation or on abandoned temporal fields. Permission to graze their animals on ejido land is granted by the comisariado or other ejido authority. There are restrictions according to the rules of the biosphere reserve regarding where goats can be herded, but some goat herd- ers graze their goats wherever they can, including protected areas. Their meat is sought after for barbecuing and a regional specialty called mole de caderas. During the time of my fieldwork, a mature goat for retail consumption cost from 1,000 to 1,500 Mexican pesos or $72.61 - $108.92 US dollars, depending on the weight and quality of the animal. Cattle are not very common livestock in the valley. The lack of suitable pasture in the valley limits the production of cattle beyond one or a few individual animals per livestock owner. There are some artisanal, household-scale dairy operations that produce cheese and leche bronca or unpasteurized milk for local markets. While there is currently no industrial-scale animal rais - ing operation in the municipio of Coxcatlán, the Tehuacán Valley is one of the largest chicken, egg, and hog producing regions in Mexico. The neighboring municipios of San Sebastian Zina - catepec, San José Miahuatlán, and Ajalpan are home to large, vertically integrated factory farms. Agriculture, sierra style The most marked difference between the agriculture of the valley and sierra portions of the study area is the end product of what they produce. Food is the goal for most all agricultural activity in the sierra, whether it is for the consumption by the families that grow and raise it or whether it ends up for sale in nearby markets. This is not to simplistically claim that sierra farm- ers do not need money or do not dabble in capitalized farming, rather it points to that readily edible foodstuffs are produced instead of a nutrition-poor, industrial commodity such as sugar- cane. The following is by no means an exhaustive description of southern Mexican campesino agriculture, which is comprised of complex agroecological systems, rather it is a quick glance into the primary economic activity for the sierra communities in the study area. For a more com - prehensive analysis of indigenous and campesino agriculture of the region see Wilken (1987), Smith (1965), and González (2001). Figure 2.40 to 2.42 represent common sierra agricultural landscapes. Most households in the sierra farm milpa and a variety of other crops throughout the year. While the garbage census conducted was not a complete sample of all sierra households, 164 of 167 were involved in some form of agriculture (Chapter 4). Milpa, which literally means “field” in Hispanicized Nahuatl, encompasses a wide definition of meaning in the region. It mostly 85 86 refers to what one could call field crops. The most common form of milpa are plantings of maize ( Zea mays ) with some intercropping of beans, called etl ( Phaseolus vulgaris ). Fields planted exclusively in corn or beans or any other staple crop are referred to as milpa. Fava beans ( Vicia faba ), called habas, are common as well. In fields where there is ample water supply, usually those near the home, an array of calabasas or squashes ( Cucurbita spp. ) are planted, but these are less common than beans or fava. Chícharo or snow peas ( Pisum sativum ) are found in some milpas but can also be found in household gardens as well. These are sometimes eaten green, but most are left to ripen and dry after which they are boiled, seasoned, and served as beans typi- cally are prepared in the region. In areas of extreme slope, avena or oats ( Avena sativa ) are sewn to prevent erosion by providing groundcover as well as for a high-quality forage for livestock. Some oats are allowed to ripen for atole, a hot thick porridge typically made with corn or rice consumed for breakfast or late night supper. Many homes in the sierra plant their house lots in milpa if there is adequate sun and the lot is not already too intensely planted with fruit trees. These milpa plantings are not as suscep- tible to pests and drought as other fields as a household can monitor and hand-water these plant- ings. Many homes also are surrounded by a field, typically planted in milpa that range from 0.4 - 1 hectares (1-2.5 acres) on average. The size depends on the relative density of the house lots in a barrio or community. Portions of these fields are left in acahual or secondary vegetation as plots must be rotated to avoid soil exhaustion. The fallow time for these plots is typically short, usually only a few years, if any, as many homes’ subsistence depend intensely on the produce that these fields provide. More fortunate households maintain agricultural fields outside of their immediate home fields. In the study area, many of these fields can be considerable distances of multiple kilome- ters from the home. These fields tend to be larger, from one to two hectares, and are typically sewn in maize or intercropped maize with beans where they are not left fallow. Smaller plantings of fava or a variety of chili are also common. Maguey ( Agave spp. ) are present along the borders of plots or fields frequently. These plants are harvested by removing the central spike of the plant and allowing the liquid to accumulate in the vacant space, where it is removed and consumed unfermented as aguamiel or fermented to produce the lightly effervescent and slightly alcoholic ( approx. 3 percent alcohol by volume) pulque . Some fields have small 20 – 30 square meter plant- ings of ocote ( Pinus pseudostrobus ). Figure 2.40. Infield milpa agriculture above Xacalco, Cox., Puebla. Photo by A. Hilburn, October 2011 Figure 2.41. Smallholder fields planted below Tecoltepec, Cox., Puebla. Photo by A. Hilburn, November 2011 87 88 Figure 2.42. Maguey (Agave spp.) amid a milpa on the dirt road to Ocotlamanic, Cox., Puebla. Photo by A. Hil- burn, October 2011 Figure 2.43. Small bull standing between milpa of mixed maize above and a planting of oats for fodder below, Xacalco, Cox., Puebla. Photo by A. Hilburn, October 2011 The immediate garden area around the houses themselves is also the site of more di- verse plantings beside field crops. Tomatoes (Solanum lycopersicum ), called xitomatl, are more frequently found around homes than fields due to their susceptibility to drought, pests, or theft. Chayote ( Sechium edule ) vines are found along fence lines or grow near trees that provide structure. Nopal ( Opuntia ficus-indica) are kept in small rows in the garden area and are har- vested when needed. Different varieties of aguacates ( Persea americana ) provide fruit and shade as well. Temperate fruit trees such as apple ( Malus spp. ), peach ( Prunus persica ), pear ( Pyrus communis), and quince ( Cydonia oblonga ) provide fruit during June, July, and August. Anona or chirimoya ( Annona cherimola ), pomegranate ( Punica granatum ), plum ( Prunus domestica ), and tejocote ( Crataegus mexicana ) are also popular trees for sierra households among a wide array of other fruit trees. Certain plants, what many would call weeds, are a useful part of sierra house- hold foodways. Pipicha ( Porophyllum linaria ) and papaloquelite ( Porophyllum macrocephalum ) are common plants that are both removed as weeds and used as condiments in seasoning food. An array of plants generally called quelites , most from Amaranthus and Chenopodium , are boiled and eaten as greens, but are passively cultivated. All livestock raising in the sierra communities is small-scale and non-industrial. Turkeys, called guajalotes , with perhaps the exception of chickens, are the most common animal raised in the sierra communities. In cases where a family has a few individuals, the turkeys gener - ally range across the house lot during the day and are placed in a protected area such as a shed or porch at night. Some homes, those who actively raise turkeys for market, keep their larger numbers of birds in coops. Pigs are raised in small pens and goats are grazed in the monte at lower elevations by some families. Sheep and some cattle are kept by a few families. Most of the livestock in the sierra is raised for sale and can provide a non-labor, farm-based source of cash for many homes in the sierra. While the sierra does not experience the severe aridity in the dry season that the valley does, water availability is a major agricultural issue in the sierra. The sierra communities in Coxcatlán are situated on the last, westernmost slopes of the Sierra Negra and are more arid than ranges to the east where the humid air from the Gulf of Mexico has not yet been removed of its moisture by orographic precipitation. This issue varies in degree of its sever- ity according to particular settlements which may have more or less water resources from springs and the ability to distribute it through tube irrigation. Microclimatic factors where a particular settlement or portion of a settlement may be situated more or less ideally to receive orographic 89 90 precipitation are also a factor regarding precipation budgets of particular communities. Transportation and communication Cars, buses, and trucks move people and goods across the municipio by a network of unpaved and paved roads as well as city streets. Coxcatlán municipio is serviced by two paved highways. State highway 980 provides paved motor vehicle transit in a north-south direction, connecting the municipio to the Tehuacán metropolitan area and the larger cities of Ajalpan and San Sebastian to the north. Highway 980 runs south connecting the municipio to nearby Teo- titlán de Flores Magón, Oaxaca where connections to the Sierra Mazateca and Oaxaca Valley beyond can be made. Within the municipio, this highway connects the valley towns from Calipan through the cabecera to Tilapa at the southern end of the municipio. Prior to 2001, there was no paved access to the sierra portion of the municipio. Today, a paved highway referred to as “ car- retera Coxcatlán-Zoquitlán ” pulls off of Highway 980 at the southern end of the cabecera and leads up the sierra to the cumbre at 2,600 meters above sea level, where it eventually leads to other communities in the Sierra Negra as well as the obvious town of Zoquitlán, two municipios to the east. The sierra towns in the northeast portion of the municipio have unpaved roads that lead uphill to the cumbre along which this paved highway runs (Figure 2.44). Car ownership is fairly low in the study area. Only 18 percent of households in the mu- nicipio have cars and depend on a number of ways to get around and transport things. Private buses and combis provide the majority of mass-transit along the paved highways. Depending on the line, bus stops can be anywhere from formal concrete shelters to parks to any safe roadside. Third-class buses run by two operators, Inter-Oceánicos and Tehcoxteo connect terminals in Tehuacán’s downtown south through Ajalpan and San Sebastian to all valley settlements in the municipio. The former operator is the largest in the area and is a regional subsidiary of national bus company ADO ( Autobuses del Oriente ). Its fleet consists of large, modern 80-100 seat buses and 15-20 seat Mercedes or Volkswagen vans driven by licensed and salaried men. The latter operator is a cooperative of a number of van and bus owners based in Tehuacán. Both lines are priced to within a peso of each other. The travel costs in both time and money are listed in Table 2.6. Additionally, second-class buses on the northbound AU line that connect Huautla, Oaxaca to Tehuacán, Puebla can be boarded for non-stop service that bypasses the rest of the Tehuacán Val - ley towns north of the municipio or its reverse can be taken direct from Tehuacán. 91 A few bus lines from other municipios in the Sierra Negra offer bus services to and from Tehuacán that pass through Coxcatlán on the Coxcatlán-Zoquitlán highway. These older, but well-maintained vehicles are refurbished school and city buses. The subida or trip into the sierra takes considerably longer than the descent with speeds averaging 15-20km/h due to the steep incline and frequent switchbacks on the road. The cost for the trip up into the sierra also costs about 30-50 percent more than the bajada or descent. The routes of these buses vary depending on the community out of which they are based, and do not directly pass through most sierra com- Figure 2.44. Roadways in Coxcatlán municipio, 2010. Map by A. Hilburn. Data, INEGI, Google Earth, and 2011 Fieldwork. 92 Table 2.6. Transportation rates by Mexican pesos and US dollars. Prices and exchange rates as of Dec. 2011. 93 munities within the municipio. These towns must be accessed by leaving the bus at the dirt road entrances that descend from the highway at the cumbre. Coxcatlán and Calipan have three taxi sitios or municipally licensed taxi services be- tween them. These taxis, usually a Mexican-market Nissan Sentra called a Tsuru, generally take people to places around the town out of which they are based but also offer service to other towns in the valley and sierra for a considerably higher cost. A ride in a taxi in town cost five pe- sos (36 cents) and a trip to nearby Calipan or Pueblo Nuevo was 35 pesos ($2.54 US) when I was there in 2011. Trips to sierra communities without access to the paved highway are available by taxi and prices to these places are fluid and debatable but average 80 - 120 pesos ($5.81 - $7.26 US) depending on the taxista or driver. Prices to Tehuacán vary but most are around 200 pesos ($14.52 US) but the trip only takes around 45 minutes compared to the hour-and-a-half by bus. Unpaved but maintained roads link the sierra communities to the valley settlements of Calipan, Coxcatlán, and Tilapa. Besides taxi service, no mass transit to these towns exists. Quick transport on these roads depends on having one of the few cars or trucks in these towns or hitching “ un ride” which are generally freely offered with the obligation of the hitcher to offer a small recompense and for the driver to politely decline it. These winding dirt roads also serve the communities through which they run by providing wider, safer and improved foot transpor- tation. Donkeys and horses are still used by some households and these roads greatly decrease the transit time between towns. In the valley, the ejidos maintain dirt roads to provide access to fields for workers, trucks, and machinery. While roads have greatly improved foot travel, many people still use the old system of veredas or footpaths to connect to communities and access their agricultural fields. These paths are not marked or mapped but every settlement in the study area is linked by an extensive network of veredas that provided the external connection for particular communities in previous generations. 94 Table 2.7. Prices for utilities and construction materials in Coxcatlán. Prices and exchange rates as of Dec. 2011. Data, 2011 Field research Consumption in Coxcatlán Garbage can be considered the end result of the production-consumption chain. There - fore it is important to note the staples of consumption by residents in Coxcatlán and their relative cost. The table is not an exhaustive listing of the items of daily life consumed by most coxcate- cos but it provides an empirical window into the things many people buy every day to reproduce their homes and livelihoods (Tables 2.7 - 2.9). Most if not all of these items eventually become garbage and the services highlighted in the table produce it. Table 2.8. Prices for basic foodstuffs in Coxcatlán. Prices and exchange rates as of Dec. 2011. Data, 2011 Field research 95 96 Table 2.9. Prices for household goods and non-essentials in Coxcatlán. Prices and exchange rates as of Dec. 2011. Data, 2011 Field research Municipio Governance in the Neoliberal Era The supreme law of Mexico is founded in its 1917 Constitution. This document estab - lishes and defines three hierarchical levels of government at the federal, state (estado or entidad federative ), and municipal ( municipio) levels. Article 115 of the Constitution describes the pow - ers granted to the municipio as well as outlines the roles of the 32 states or federative entities, as they are also called. Like the United States, Mexico is a federation of states whose official but not often mentioned name is Estados Unidos Mexicanos (United Mexican States). Yet, the PRI- controlled Mexican government, which controlled Mexico for 72 years following the Revolution, was far more centralized at the federal level than its northern neighbor. This centralized patron- client federal system functioned well for the PRI to create unilateral political consensus across the country and across Mexican society. After decades of economic growth and development, the debt crises and economic decline of the early 1980s forced Mexico to reconsider running a coun - try and provisioning services in such a hierarchical, top-down manner. In 1983, the Miguel de la Madrid administration reformed the constitution to grant more autonomy in addition to greater responsibility to state and municipal governments. The changes to Article 115 of the Mexican largely granted municipios more de jure au- tonomy in its fiscal management and governance. Municipios were given the right to all property tax and service provision revenues and were allowed to manage their finances independently. They could also legally establish their own municipal ordinances where state or federal laws did not specify. Yet the practical autonomy that municipios were given by the 1983 Reforms was still conditional upon the entrenched dependence of many less well-off municipios on federal and state funds to operate and provide services. Municipios following the 1983 Reforms could raise revenues one of three ways. First, they could levy property taxes or service provision fees such as water, garbage, or local retail markets. Second, municipios could obtain federal or state fund- ing through grants for specific projects. These convenios , as they are often called, are typically for one-time only projects and for high-profile infrastructure development. Third, municipios received revenue sharing from the federal government which would be administered by state governments. This last source of funding typically makes up a substantial proportion of a muni- cipio’s budget, making them subject to greater influence by state governments. Additionally, fol- lowing constitutional reforms, the municipios were now responsible for greater service provision including a comprehensive listing of local infrastructure of which garbage management is a part. 97 98 In short, the 1983 Reforms served to shift greater responsibility from the federal government to the state and municipal level while keeping the previous strong centralist political grip on states and municipios. The following chapter will discuss the role of municipal politics as related to garbage in more detail, but it is fitting here to briefly describe the political situation in the municipio that existed during the course of fieldwork. In Coxcatlán and in municipios across Mexico, a new municipal government is selected every three years, called a trecenio. When I arrived in January 2011, the trecenio of the previous administration was ending at the end of the month. The PRI ( Partido Revolucionario Institucional ) municipal presidency of Roberto Zerón, that had granted me permission to work and had written a support letter for funding applications, was winding down its operations while literally cleaning out their offices. In February, the new PAN (Partido Acción Nacional ) presidency led by Efigenio Atilano took power. They also graciously granted me permission and support for my research. The new panista (PAN-controlled) government had come to power following a narrow election in July over the PRI candidate, that when told to me in anecdotal accounts, was decided by less than 100 votes. This was the first panista government in the municipio’s PRI-dominated history. At the same time, a new state governor, Rafael Moreno Valle, the first panista in that position was taking office heralding a major shift away from the center-left PRI to the more socially conservative PAN in the State of Puebla. The presidencies of the comisariados ejidales , while possessing no civil power but considerable political influence in the municipio, however remained predictably priista (PRI-controlled). While there is support for other parties such as the more-left-than-center PRD ( Partido de la Revolución Democratica ) and similarly oriented PT ( Partido de Trabajo ), these party affiliations apply more to offices at state and federal levels. At the muncipio level in Coxcatlán, the PRI and PAN dominate. Across Mexico, when a new municipal president takes over, the entire governmental staff changes. While new appointees are made to the various municipal regidores, or regents, posi- tions, the duties and functions of their offices remain fairly similar. The specific titles and office duties vary from municipio to municipio, but most cabildos de regidores (boards of regents) cover issues such as the treasury, economy, health, ecology, justice, and records. These offices remain constant across municipal presidencies. By customary rule, one of these regents must come from an opposing party. As it is across rural parts of the world, most government offices are held by non-professional politicians. In Coxcatlán this is the case and past presidents have come from a number of professions, from shop owners to heavy machinery operators. Regardless of profession, all come with experience in local and regional political party participation. A new chief of police is typically appointed and the entire police force turns over, some of which are professionals from other municipalities or the military but some have no prior law enforcement experience. The same applies to the staff of the protección civil who provide emergency rescue and paramedic services in the muncipio. The auxiliary presidents ( presidentes auxiliaries or agentes municpales) in the municipio’s towns and villages that fill in for the municipal president are also appointed by the new municipal government. Other auxiliary positions in maintenance and utilities, including sanitation workers, are less appointed and may carry some carry-over from trecenio to trecenio. In addition to changes in the staffing of government offices that accompany a new mu- nicipal government, the municipal infrastructure and personnel changes almost completely (Ward 1998; Assetto et al. 2003). Much of the funds for these changes and improvements come from petitions to the state government. Some new vehicles and equipment are purchased, and small public works projects are begun across the municipio. In Coxcatlán in 2011, the most obvious changes were repainting projects that replaced the green and red, symbolic not just of Mexico but of the PRI party, with the baby blue and white of the PAN party (Figure 2.45). The repair and resurfacing of the major north-south highway through the municipio had also just begun a few months after the switch and still continues to present. The Narco Wars In 2006, the newly inaugurated president Felipe Calderón began a massive, large-scale military and law enforcement effort to diminish or destroy the growing power of the various drug cartels and organized crime organizations in Mexico. The result of what is generally seen as a clumsy and corrupt effort by the federal government has been a continued internecine war between drug cartels, criminal organizations, and the military with immense collateral damage to the law-abiding public. Thus far, it is conservatively estimated that 50,000 Mexicans have been killed in violence related to this war. In the chaos caused by what began as a conflict over the control of drug trafficking, criminal organizations predicated on extortion, kidnapping, and para- militarism have taken hold and have prospered (Bonner 2012). Much of the violence related to 99 100 the drug war has been focused along the US-Mexico border, large urban areas, and in particular states such as Sinaloa, Michoacán, and Veracruz. Despite of being only two municipios away from the state of Veracruz where considerable drug war-related violence has occurred, Coxcatlán and the surrounding region had virtually none of it during my stay. That is not to say that there was not violent crime around, with Tehuacán and many of the small cities around having considerable local gang activity. Even small Coxcat - lán had a group of young men that were suspected of a few minor muggings and burglaries. Yet when I was there in 2011, only the menace of drug violence existed in the minds of area resi - Figure 2.45. Bridge across the Barranca Soyolapa in February and April 2011. The bridge was painted in PRI green to reflect the ruling political party but was repainted PAN baby blue by the new administration. Photos by A. Hilburn, February (top) and April (bottom) 2011. dents. On occasion, reportedly unsavory characters in expensive SUVs and pickups sporting dif - ferent accents were spotted in town, but most were just passing through or were just incorrectly profiled as narcos. SUMMARY Coxcatlán municipio is but one of 217 municipios in the Mexican state of Puebla and one of 2,456 in Mexico. It, as well as other muncipios, is an autonomous functional region, whose powers are defined by Article 115 of the 1917 Mexican Constitution. Coxcatlán spans about 250 square kilometers and is home to approximately 20,000 residents in 18 settlements set between the pine-oak forested slopes of Sierra Negra and the semi-arid Tehuacán Valley. In terms of its population and settlement characteristics, the municipio is divided between an urbanizing val - ley and a largely agrarian indigenous sierra. Over half of the municipio’s approximately 20,000 residents live in and around the cabecera or municipal seat called Coxcatlán and the sugar-refin- ing town of Calipan about five kilometers away. Here, as well as the rest of the valley’s smaller settlements, the cultivation of sugarcane by members of the social properties is the primary eco- nomic activity, although these small towns have the service and retail befitting such places. The Nahuatl-speaking sierra communities are largely dedicated to smallholder, food-oriented agricul - ture. These villages are not entirely rural, however, in the sense that designation implies a lack of cosmopolitanism and stagnant growth. The sierra is economically less affluent, less physi- cally connected by infrastructure, and settlements do tend to be smaller, but a few villages boast populations over 1,000 residents and most are sender communities for national and transnational migration. The following chapters will investigate in more depth the management of garbage and the way this is framed as an environmental issue by residents in Coxcatlán municipio. However, this research took place in this setting and later chapters will draw upon and reference the mate- rial from this biography of a place. Endnotes 1. This is slightly larger than the combined areas of the U.S. states of Maryland (27,092 sq. km) and Delaware (5,295 sq. km). 2. Baja California Sur is considerably larger than Puebla with 73,922km2 of state territory. 3. This weather station is located in the cabecera municipal of Zoquitlán, an adjacent municipio to Coxcatlán that 101 102 is situated at roughly the same elevation of many of the sierra communities in Coxcatlán municipio. 4. This small Nahuatl-speaking settlement is also listed as “Apala” in older censuses and in C. Earle Smith’s 1964 Flora of the Tehuacán Valley and today is largely referred to as “Ranchería Pala” by area residents. The current cen- sus simply calls it “Pala” and for consistency, it will be referred to by this name throughout the text. 5. This cactus is also known as Cereus candelabrum, Lemaireocereus weberi, Stenocereus weberi, and Cereus weberi. 6. The “community” that I am referring to is a comunidad agraria which is a social property similar to an ejido but one granting usufruct lands to an indigenous community. 7. The Partido Revolucionario Institucional or Institutional Revolutionary Party ruled Mexico from the late 1920s until Vicente Fox’s successful electoral victory in 2000. 8. This is a general land tenure/land use system of smallholder agriculture with the minifundio as the individual production unit. These usually consist of 1-10 hectare (2.5-25 acre) holdings (West and Augelli 1989). 9.. Farm subsidies were replaced with the PROCAMPO program that essentially made up the difference between pre-NAFTA and post-NAFTA commodity crop prices. This program expired this year in 2009. 10. However, it should be noted that all land-based economic activity in an ejido prior to the 1992 reforms was not entirely communal. Although technically not legal, many ejidatarios would rent land or at times sell its use rights to outsiders or others within their own community with the asamblea’s approval. Thus, reglamiento interno has in practice trumped federal law in small-scale land dealings (Haenn 2006; Nuitjen 2003). 11. This is pronounced COH-skah. Given the multiple syllables that often comprise place names in the study area, there is the common practice of contracting these names. The nearby towns of San Sebastian Zinacatepec and Teotitlán de Flores Magón are called “Sanse” and “Teoti” respectively. CHAPTER 3: Garbage Management in Coxcatlán This chapter outlines the practices of managing garbage and waste matter across the municipio of Coxcatlán. It begins with a description on how the municipio government structures and provides garbage collection service. This is followed by a discussion of the results of a mu- nicipal solid waste generation and classification study conducted by municipio of Coxcatlán and the state government. Using these results, an estimation of service provision and garbage genera - tion is provided to determine the scope and size of garbage management in the municipio. Not all households in the municipio receive collection service and must rely on a number of practices to dispose, destroy, or convert garbage. Finally, the trajectory that garbage follows, from consump - tion through management to final disposal or reconfiguration, is outlined. GARBAGE MANAGEMENT IN MESOAMERICA, NEW SPAIN, AND MEXICO Middens, heaps, and scatters The majority of this research project focuses on recent and current garbage management in Coxcatlán and surrounding areas. However the past cannot be disposed of easily as ancient but extant traditions regarding garbage disposal are present there. Understanding how waste and garbage were conceived of and dealt with in past societies allows one to reflect upon why the current situation is the way it is. For the first inhabitants and their descendants who lived in impermanent settlements, the archaeological record from middens, cave deposits, and artifact scatters provide some insight into the nature of what people from that time consumed and then threw away. When village life became the norm in the Tehuacán Valley and Mesoamerica abroad, around 1,500 BC, the resulting permanent households developed new ways of dealing with waste that they could not escape or leave behind. The dooryard garden concept elaborated in Latin Americanist cultural geography, behavioral archaeology, and ethnoarchaeology provides a model for understanding refuse disposal and reuse during this time. The record of human occupation in Coxcatlán and the greater Tehuacán Valley begins with garbage. The main proxy for what is known about subsistence, seasonal migration, and pop- ulation of the far past societies from the Paleoindian period to contact with the Spanish colonial front (approx. 10,000 BC – 1521 AD) is the stuff they threw out, could not consume, or simply left behind as summarized in the beginning of Appendix 2. Much of the waste that MacNeish and 103 later researchers recovered came from cave floor deposits and middens or dump heaps at sites across the Tehuacán Valley. These deposits reflect changes in material culture over time, but due to their antiquity and the ephemeral nature of residences in the region, little is known about the practices of dealing with waste besides in situ waste scatters or deposition in middens or dump heaps. The wastes left by the earliest inhabitants during the Ajuereado Phase (10,000-7,000BC), were mostly faunal remains of extinct late Pleistocene mammals and some small contemporary mammals. The chipped stone tools to kill and process them were also part of these assemblages. Some plant-based refuse was deposited at the time but their high decomposability has removed such material from the archaeological record. Baskets and textiles arrive in the El Riego (7,000 – 5,000BC) waste stream showing early evidence for use of plant-based materials. Beyond the Ajuereado and El Riego phases, population growth, climate change, and increasing social com - plexity led to a shift in the food waste record from faunal to plant remains, reflecting changing subsistence strategies from hunting and gathering to the incorporation of horticulture. In the middle Archaic Coxcatlán and Abejas phases (5,000 - 2,300 BC) and early Formative Purrón phase (2,300 – 1,500 BC), early forms of corn ( Zea mays ), cucurbits ( Cucurbita spp. ), beans ( Phaseolus spp. ), chilies ( Capsicum spp. ), and other well-known domesticated plants such as cot- ton ( Gossypium spp. ) become present. Over time, these botanical remains and the ground stone tools used to process them such as manos (pestles) and metates (open mortars) become more predominant than faunal remains and chipped stone cutting tools. Ceramics turn up in the late Purrón phase with the increasing permanence of residences and associated reliance on agricul - ture. (MacNeish 1964; Flannery 1967) Dooryard gardens: A window into the past Village life came to fruition in the Santa María phase (850 – 150 BC) and the dooryard garden model of waste disposal can be used as an analog to understand how individual house- holds managed their garbage in Coxcatlán through to the end of prehistory, the colonial era, and for most of the 20 th century for rural households. The dooryard or house lot, also known as lote or solar in Spanish, is the individual familial residential compound common in Middle American villages. Assuming connections between ancient and contemporary agrarian village households exist in terms of residential spaces, the spatial analysis of how relatively contemporary house lots 104 are spaces of waste management can perhaps provide insight into how waste was managed in Mesoamerican villages. These sites of household reproduction have long been a focus of the cultural-historical approach in Latin Americanist geography (Herlihy and Wiseman 2008; Anderson 1952, 1954; Kimber 1966, 1973, 2004; Denevan 1966; Gade 1976; Smith and Cameron 1977; Works 1990a, 1990b; McKillop 1992; Pulsipher 1994; Steinberg 1998, 2002; Keys 1999; Denevan 2000; Whit - more and Turner 2001; Doolittle 2004). This focus, perhaps due to Carl Sauer and his acolytes’ interests in plants, centered primarily on the gardens and orchards themselves. However, while noting where dump heaps and pits are in these residential landscapes, most geographic research on dooryard gardens, with the exception of Doolittle (2004), largely ignore the spatial ordering of refuse management areas in house lots. Outside of geography, ethnoarchaeological research in the 1980s and 1990s sought to understand waste discard in contemporary peasant households in Mexico and Central America with the goal to better contextualize archaeological assemblages of ancient households and more completely elaborate the processes of archaeological site formation (Hayden and Cannon 1983; Wilk 1983; Deal 1985; Arnold 1990; Killion 1990; Sutro 1991; Hirth 1993). Archaeologists, with their reliance on garbage as a material culture proxy for larger questions about cultural and social relations of ancient peoples, place considerable importance on reconstructing spatial pattern- ing of discarded objects. Their efforts to systematically represent the past with current examples provide more precise models of ancient garbage management than much explicitly geographic research. These studies also, perhaps more importantly, provide contemporary ethnographies of garbage management in Middle American peasant households. The typical Middle American rural village house lot usually consists of a residential structure, kitchen, patio, and outhouse, bordered by some form of garden or orchard (Deal 1985). There is no precise universal configuration to how these elements are arranged in every case, but there are general patterns with regard to the management of refuse within these house lots. Hirth (1993) sums up the maintenance of residential space and garbage management here as a “series of concentric rings extending outward from the residence…each ring farther out from the cen - ter contains more trash until a threshold is reached in the garden or garden-orchard.” Killion’s (1990) model of domestic space use by farming households in the Sierra de las Tuxtlas provides a more precise picture of refuse disposal within the home but highlights the same general concept 105 (Figure 3.1). The typical Tuxtla house lot features a clean area around the house and patio that is kept clean of waste and vegetation while the outer “intermediate” portions of the lot receive both less cleaning and more active dumping of garbage. The “garden-refuse area”, located on the far periphery of the house lot, is where no cleaning occurs and systematic purposive dumping of refuse takes place in piles, pits, and scatters. These peripheries of a house lot are what Lewis (1976) calls the toft area, comprised of the “immediate site of a dwelling and its outbuildings as well as the location of most activities associated with the dwelling”. The simple logic behind the distancing of waste from the inhab - ited zone is that it would impede the necessary activities of the household (Killion 1990). The gardens and orchards, typically the focus of geographers, are useful spaces that have a diverse assemblage of fruit trees, medicinal plants, flowers, and at times a vegetable plot that border the house buildings and patio area (Anderson 1954; Steinberg 1998; Keys 1999). Besides receiving Figure 3.1 . M esoamerican house lot model showing discard and use areas. Figure by A. Hilbur n and adapted from Killion 1990. 106 food waste and other organic refuse that ostensibly enriches the soil, these areas are also the re- positories of more persistent materials that could be classified as “junk”. Here, farther away from the more actively lived space, large pieces of waste like broken boards, busted cement structures, and scrap metal along with useless hazardous materials such as broken glass or jagged, rusty metal are deposited in piles or are accumulated in pits and are later buried. Some of the more us - able pieces may be left aside in a state of semi-storage for potential later use (Hayden and Can - non 1983). Besides dumping on the peripheries of house lots, ancient Coxcateco households likely employed other methods of waste disposal. Burying garbage in pits was one such practice. In ancient Maya house lots, out-of-use pits were filled with encumbering and hazardous waste like broken ceramics (Hayden and Cannon 1983). Sisson (1973), in his excavations of Coxcatlán Viejo, encountered what he thought to be old, previously used storage pits that were filled with what looked to be garbage. Dumping garbage outside of the house lot in water bodies, down hill - slopes, in barrancas, and on side streets was also a likely practice given its use in a few notable ethnoarchaeological studies. Sutro (1991) noticed the common practice of depositing persistent, inorganic garbage in the stream bed that ran through a small, agrarian Zapotec village in the Oaxaca-Mitla Valley. Proximity to this watercourse seemed to govern dumping behavior with households living closest to it dumping there more frequently. However, residents of this village also took broken glass and other hazardous wastes to dumps far outside the village center and away from agricultural fields. Hayden and Cannon (1983) also observed the proximity effect on dumping by Highland Guatemala Maya households that dumped on vacant lots and side streets that were typically less than a two-minute walk from their house. For nearly all the ethnoarchaeological examples, burning refuse was a common practice for households. Burning converts unwanted matter to ash or at least reduces its volume. Some of the burning occurred in pits, but mostly garbage that was not intended for the toft area, pro- visional discard pile, or outside dumping ground was burned in small piles outside of the main lived space of the house lot. Even refuse dumped outside of the house lot was burned to perhaps minimize its blighting effect (Deal 1985). While it would be erroneous to assume that no changes have occurred in household-level strategies to remove or dispose of waste have occurred since the Ajalpan Phase (1,500 – 850 BC), toft area disposal, outside dumping, and especially burning remain common practice in more traditional households of the study area today. 107 A brief note on urban garbage management in the historical period Garbage management and general public sanitation in past urban societies have attracted the attention of a number of historians, however most of the case examples from these histories come from major metropolises and typically have a bias on the United States and Great Britain (Kelly 1973; Pellow 2004; Melosi 2005, 2008; Rogers 2005; Humes 2012). In Mexico, there is a considerable literature on the historical management of wastes and sanitation from the colonial era to the past century in major urban areas such as Mexico City or Puebla (Castillo-Berthier 1983; Dávalos 1997; Flaks 2010; Glasco 2010). During the colonial era and early Mexican state - hood, residents in these large urbanities followed the European model of dumping garbage in the streets (Dávalos 1997; Glasco 2010). The resulting situation was an unsanitary milieu of garbage and excrement-filled streets that impeded crosstown travel and was purported to be an agent of disease (Flaks 2010; Glasco 2010). Considering the greater sense of public space over private space and the thousands of residents without formal living quarters, street dumping was at least a socially equitable and convenient practice. In Mexico City, the sanitation reforms of Viceroy Revillagigedo in the late 18 th century sought to systematize garbage collection and dumping to limited success. This reform made street dumping illegal by imposing fines and the solution to the garbage management problem was collection by municipal street cleaners who would then dump in authorized sites, typically in slum areas. Sanitary conditions improved in more affluent areas of the city where collection was regular, but the positive results of sanitation reform did not extend to poorer areas. By outlawing dumping and littering but not providing comprehensive service, Revillagigedo’s reforms had the effect of persecuting the poor and indigenous citizens while using their residential spaces as dumps (Glasco 2010). This state of unequal collection service with some public cleaning persisted and improved incrementally in large cities through independence, the Reforma Period, the Porfiriato, and the Revolution by a system of sanitation workers on foot and mule cart. It was not until the 1930s when mechanized collection infra - structure as well as city ordinances for garbage collection and landfilling approached what exists today in major cities. In the 1950s, fully amid the post-revolutionary economic boom, collection by trucks had completely replaced carts and mules in Mexico City and other large cities (Castil - lo-Berthier 1983: 30 - 31). Outside of major cities, there are few, if any, historical accounts of public sanitation in Mexico. The historical evolution of state-run garbage management in large cities was vastly 108 different than that of the small towns and villages in Coxcatlán municipio. Yet through anec - dotes, residents in Coxcatlán’s valley settlements recounted that households across the municipio burned, composted, or dumped their garbage in ways similar to those described in the ethnoar- chaeological accounts. It was not until 1996 that the municipio government first began collecting garbage in Coxcatlán town and Calipan, signaling a change in how garbage was managed by these households. This service then expanded incrementally to the rest of the valley settlements by 2005. Today, the sierra settlements have yet to receive this service and typically manage their garbage by burning, burying, or dumping. One town operates a communal dump. The following two sections outline in greater detail garbage generation, characterization, and management in contemporary Mexico in general and Coxcatlán specifically. CONTEMPORARY GARBAGE MANAGEMENT IN MEXICO Framing the garbage issue in Mexico The discussion of the management of any matter or phenomenon, whether it is land, water, wildlife, or in this case, garbage, should try to frame the sheer magnitude of what is being examined. With garbage, the most common metrics to gauge garbage from an empirical perspec - tive are generation, usually in kilograms per some time or collective unit, and classification, where garbage is classed into recognizable classes such as food, paper, glass, plastic, or a binary organic/inorganic division. The absolute amount of garbage generated, known technically as residuos sólidos ur - banos or urban solid waste, is increasing year by year in Mexico and across the world. In 2000, Mexico generated nearly 31 million metric tons of garbage. A decade later, a figure just north of 40 million metric tons was estimated to have been generated, representing a 30 percent increase in that time. Annually, the average rate of increase in generation for the past ten years sits just shy of three percent. Greater increases in generation have been seen in the State of Puebla with nearly 1.4 million metric tons generated in 2000 and over 1.8 million metric tons in 2010, repre - senting a 35 percent increase, five percent greater than the national estimate. While some of this increase is natural, due to population growth, the rate of increase in garbage generation exceeds that of population growth in both Mexico and Puebla. The average Mexican generated 0.63 kilograms of garbage per day in 2000. As of 2010, the same Fulano, or Mexican John Doe, generated 0.71 kilograms per day, showing an 11 percent 109 increase. This amount of garbage is a national average, however. Generation rates are spatially variable, at various scales, within towns and across the country, reflecting demographic and economic differences. This spatial variation is contingent upon the unequal distribution of wealth in addition to cultural practices regarding sanitation, and rejection/acceptance of contemporary consumerism, among other factors. For one, garbage generation rates generally increase with greater economic affluence. A glance at Figure 1.1 points to this trend where countries classified as “global north” are some of the highest producers. Compared to the average Norwegian, Irish - man, or American, Mexicans produce a relatively less garbage per person. However, the absolute quantity of garbage generated in Mexico places it sixth among OECD countries where data ex - ists. Within Mexico, particular regions and states generate more and less garbage per person (Figure 3.2). For the sake of comparison and to assist in understanding the spatial variations of garbage generation, the Mexican government defines four garbage generation regions in Mexico: the north, comprising the border states and adjacent states; the center, made up of a latitudinal band that transverses the central part of the country from the Pacific to the Gulf of Mexico; the capital or Distrito Federal ; and the south, stretching from Guerrero on the Pacific east to the Yucatan peninsula states. The north region and the federal capital district are the largest genera - tors of garbage per capita in the country. These regions are the most economically developed and have largely urban and globalized populations. The south region, known for its more rural, in - digenous, and economically marginalized populations, produces the least, with the central region in between, both spatially and in terms of generation rates. These regional estimations illustrate garbage generation beyond the nation-level, but Figure 3.2 points to considerable variability at the state level within SEMARNAT’s regional boundaries. At the level of municipios, individual localities, and households there are also similar patterns of increased economic affluence with garbage generation. The 1973 study by Philips et al. pointed to “conspicuous consumerism” behind upper-income households in Mexico City generating twice the amount of garbage than that of the average Chilango (Mexico City resident) household (Philips et al. 1973). A more recent study of household garbage generation in Morelia, the capital of the State of Michoacán, showed that garbage generation rates increase with socio - economic affluence. The researchers in this study found that garbage contents of more wealthy households had more paper, processed foods, and yard waste than those of poorer households, 110 Figure 3.2 . M unicipal solid waste generation by state (entidad federativa) 2010 with SEMARNAT generation zones in orange. M ap by A. Hilbur n, Data, INEGI 2010, SEMARNAT 2010. whose garbage was comprised of more basic foodstuffs such as tortillas and diapers, reflecting the trend of larger family sizes within low-income households (Buenrostro, Bocco, and Bernache 2001). In metropolitan Chihuahua City, a generation study observed a notable, if statistically insignificant, increase in garbage generation in more affluent parts of the city. Chihuahua as a whole is more economically affluent than other cities where generation and classification studies have taken place and the Chihuahuense generation rate was higher and considerably less organic than other cities in western and central Mexico (Gómez et al. 2008). A study conducted in Mexi - cali, Baja California Norte, on the U.S. border, added the dimension of family structure to gauge differences in generation and composition between socioeconomic classes. The results of this research found that higher-income single-parent and nuclear families produced large amounts of garbage as did middle-class extended families. As expected, poor families produced the least (Ojeda-Benítez et al. 2008a). Contrary to other household-level generation studies, another publication from the previously noted generation study in Mexicali noted that higher educational achievement, typically analogous to economic affluence, was related to a decrease in garbage generation (Ojeda-Benítez et al. 2008b). Socioeconomic status and geographic region were not 111 believed to have an effect on the generation of household hazardous waste such as cleaning prod - ucts, batteries, and automotive products (Buenrostro et al. 2006). In Mexico according to 2010 INEGI statistics, the dominant class of garbage was food, garden, and other organic waste, representing 52 percent of the total waste stream (Figure 3.3). This class is followed by paper and cardboard (14 percent), an “other” class representing unclas - sifiable bits and materials besides those listed (12 percent), and plastics (11 percent). Glass, met- als, and textiles round out the classification at six, three, and just above one percent, respectively. The composition ratio of garbage, somewhat surprisingly, has not noticeably changed since 2000. Some materials have been reclassified, adding to sharp changes in plastics (increase) and house- Figure 3.3. Per capita municipal solid waste generation by class, 2010. Graph by A. Hilbur n. Data, INEGI and SEMARNAT 2010 hold waste (decrease) categories, but the waste stream in Mexico is mostly just greater in size and not much different regarding its contents. How all this garbage is managed varies across space as well (Figure 3.4). At the national level, a considerable majority of Mexican households (79 percent) use some form of garbage 112 Figure 3.4. G ar bage management prac tice by household in M exico and the State of Puebla, 2010. Graphic by A. Hilbur n. Data, INEGI and SEMARNAT 2010 collection, whether by the municipal government or by a semi-private or private contractor, as their primary method of garbage disposal. Comparatively, in the state of Puebla, only 69 percent of state residents use some form of collection. In many underserved and economically marginal urban areas, informal waste collectors, known as carretoneros or burreros, collect household garbage for small fees, recycle what they can, and find a place to dump their garbage (Medina 2005). The lack of any collection service, regardless of provider, is largely attributable to the difficulties of extending collection service to the more economically marginalized communities in remote, rugged parts of the state. Burning garbage is the second most common method across the country (15 percent) and also in the State of Puebla (21 percent). An alternative or addition to garbage collection is the provision of large dumpsters by municipios or contractors into which residents can deposit their garbage. Dumpster use is more common in the State of Puebla (8 percent) than the national average (4 percent). Some settlements, those typically without collec - tion service, have community dumps apart from those managed by the municipio where residents take their garbage to dump. Community dumps are not as common in recent decades and only 113 fewer than one percent of Poblanos (resident of the state and city of Puebla) use them and just over one percent of Mexicans dump their trash in them. Dumping, burying, and other forms of garbage disposal are slightly less utilized across Mexico (over one percent) than in Puebla (under two percent). However, these methods of disposal are largely against legal and moral codes of sanctioned garbage management and are likely underreported in the census from which this data is derived. For the garbage that is collected across Mexico, it eventually is deposited in sanitary landfills, open-air dumps, and sites managed by methods somewhere in between (Figure 3.5). An increasing amount of this waste is being recycled from year to year. The largest portion of garbage in Mexico ends up in sanitary landfills. Most Mexicans live in cities and most, if not all, larger urban areas deposit their garbage in sanitary landfills. In 2011, there were only 196 sanitary landfills in the country compared to the estimated thousands of open-air dumpsites. Of- Figure 3.5. Final disposition of collec ted gar bage in M exico (1,000s metric tons), 2000 -2010. Graph by A. Hilbur n. Data, INEGI and SEMARNAT 2010. 114 ficial open air dumps, though, are more common in less-populated, rural municipios where less garbage is generated by less people. The rate of garbage being deposited in sanitary landfills and other controlled sites is increasing while that of deposition in open-air dumps is declining due to state and federal programs to increase sanitary landfills. This increase in controlled landfill- ing could also be attributable to greater population increase rates in larger cities where sanitary landfills are. Outside of officially sanctioned dumpsites, there are innumerable tiraderos clan- destinos or clandestine dumps where individuals or groups of individuals dump their garbage illegally. These sites vary in size from a hundred square meters to areas the size of half a football field. Unfortunately there is little to no data on these sites as they are periodically cleaned up or, conversely, are neglected by authorities. The amount of waste being recycled is also increasing, although at lesser rates than the amount being landfilled. Household municipal solid waste generation in Coxcatlán In January and February 2009, the Coxcatlán regiduría (regent) of public health in con - junction with the State of Puebla Ministry of the Environment and Natural Resources ( Secre - taría del Medioambiente y Recursos Naturales or SMRN ) conducted an unpublished municipal solid waste generation and classification study in the cabecera municipal (Sánchez-Mendez and Olmos-Torres 2009) . This study was part of the mandatory management plan used to obtain state funding for solid waste management projects. In addition to the generation and classification study, the project included the informal mapping of garbage collection routes, a general study of garbage management across the municipio, the provision of a recycling truck, as well as the con- struction of a solid waste transferal station that would compact, weigh, and prepare for shipping of all inorganic municipal solid waste to the nearby sanitary landfill in Ajalpan. The generation and classification study sampled 100 households’ garbage for seven consecutive days. The research team, composed of SMRN employees and local residents, first conducted a basic questionnaire survey concerning demographics and household garbage man- agement. The research team then left standard-sized polyethylene bags for each participating household. Each day, the research team collected, grouped, weighed, and classified the household garbage according to normas mexicanas (federal standards) NMX-AA-022-1985 and NMX- AA-015-1985 for conducting such a generation and classification study respectively. Genera- tion rates were determined by combining all garbage collected for the day and weighing it. The 115 research team estimated the character or relative proportions by class of the sample using a modi- fied version of the quarter method. This method, where an approximate one-fourth subsample is extracted from a homogenized pile of the total sample, is a much more efficient than classifying the entire sample. After seven iterations of collection and analysis, the resulting generation rate for the sample of households in the cabecera was almost exactly 1 kilogram (999 grams) of garbage per household per day or seven kilograms (6.99 kg) per week (Figure 3.6). This estimate is consid - erably below the 0.95 and 0.76 kilograms per day estimated for individuals in the central and southern parts of Mexico in other studies (INEGI 2010). This low estimate could be some ef - fect of study awareness by the participating households in that they “produced” less garbage on purpose. If not, the one kilogram per household per day appears reasonable considering personal Figure 3.6 . Per capita municipal solid waste generation by class, 2010. Graph by A. Hilbur n. Data, INEGI and SEMARNAT 2010 observations of how much garbage was thrown away for collection in the homes in which I lived during my stay. Also, Coxcatlán’s low level of affluence and low garbage generation corresponds to similar observations made regarding wealth, material possessions, and garbage generation in 116 Mexico (Buenrostro, Bocco, and Bernache 2001; Gómez et al 2008; Ojeda-Benítez et al. 2008a; Ojeda-Benítez et al. 2008b). Glass dominated the mass of the assemblage, comprising 27 percent of the averaged subsamples. Individual counts were not taken, but glass was likely not very abundant and its high representation in Coxcatlán’s household garbage is more likely the result of the relatively greater weight of glass compared to those of plastics, organic material, and paper. The next largest MSW class was PET, short for polyethylene terephthalate and also known as #1 plastic, the plastic used in soft drink bottles and much plastic packaging. PET is currently a recyclable material in Coxcatlán along with the fourth largest MSW class, Tetra pack, or coated cardboard used in food packaging which was 7.9 percent of the averaged subsample. Organic waste formed 11.6 percent of the averaged mass of the subsample. Organic waste is possibly underrepresented in this analy - sis due to the common practice of using vegetable scraps and egg shells to fertilize garden plants or to feed poultry. However, from a management perspective the relatively small presence of organic waste in the total assemblage is accurate given that it represents what will be collected. Steel and tin comprised a significant 7.1 percent of the mass, but like glass, is probably overes- timated due to its high weight per item ratio. Also, steel and tin are typically sold by individual householders to independent recyclers. Aluminum was not present in the sample at all as the consumption of aluminum-canned drinks is not very common and aluminum’s easy recyclabil - ity and token monetary value prevent it from entering the local waste stream. Cardboard from packaging and storage, which accounted for 5.3 percent of subsample weight, is a fairly common waste class but its high reusability may underrepresent its presence in the assemblage. Paper (3.3 percent) is very common in disposed household waste but its light weight and the fact that used toilet paper, which is not flushed in the study area, was omitted from inclusion in the study contribute to its small contribution to the sampled mass. Polystyrene, also known as Styrofoam or unicel, only accounts for 2.2 percent of the weight but is likely underrepresented in terms of abundance due to its almost negligible weight. A mix of plastic types, HDPE (6.1 percent), poly - propylene (2.2 percent), and LDPE (0.9 percent). Cloth (5.3 percent) in the form of old rags and destroyed clothing is a surprisingly significant sample constituent but ceramics, which was prob- ably a major waste class decades and centuries ago in Coxcatlán, only contributed 5.5 percent of the mass even when considering its relatively greater heft. In spite of there being some sense of how much of what is managed how noted above, 117 all of these claims regarding garbage generation, classification, and management come from a Babylonian tower of data unintelligibility. The various datasets from which these studies derive their conclusions are site specific and very idiosyncratic. The above figures on generation, clas- sification, and management must be taken as simple estimates and provide only a coarse general empirical frame to gauge the garbage issue. The lack of congruity with data sets makes under- standing garbage at a global level extremely difficult, but even in Mexico, where census and environmental data is of high quality, there are major issues with this data related to time depth and inter-compatibility. The existence of comparable data of similar quality at similar scales is extremely rare. Garbage generation is a much easier phenomenon to study as it only really requires measuring weight. As such, there is greater availability for data on its generation. Classification studies, on the other hand, require the separation of garbage into distinct categories, then weigh- ing or measuring their volume. Fittingly, these studies are rarer and have smaller study samples. When garbage estimation studies are conducted, the results are highly dependent on the sample size. Ideally, the entire mass of garbage for an area would be measured and from that one could determine how much of what classes are generated. However, sampling everyone’s garbage is impossible and garbage generation studies must be conducted across sample populations with varied demographic traits and different garbage generating behaviors. The problems with gar- bage classification fill a similarly varied grab bag of data. Like any taxonomic attempt, waste classification is arbitrary, flexible, and sometimes inapplicable. The result of this lack of data congruity is that garbage statistics are notoriously difficult to compare across different academic studies, government agencies, and across international boundaries (Buenrostro, Bocco, and Cram 2001; Buenrostro and Bocco 2003; Moore 2011). Nevertheless, there are policy mandates for data standardization within the most recent federal laws regarding waste management in Mexico. These efforts have greatly improved the quality of garbage-related data in the past ten years at the federal and state-level scales, however, the need for local idiosyncratic studies still remains and provides the impetus for a growing academic literature on garbage in Mexico. Legal foundations of garbage management in Mexico Inside Mexico, the official governance of garbage and its management falls to all three levels of government; federal, state, and municipal. Article 4 of the 1917 Mexican Constitution 118 is the highest statute requiring that “every person has a right to the protection of their own health and [that] every person has the right to a healthy environment to pursue their own development and wellbeing”. It also states that the government will guarantee this right and notes right of the government to enforce actions against environmental damage and degradation (Constitución Política de los Estados Unidos Mexicanos 1917: Articulo 4). While these laws originate directly from the foundational document of Mexico’s federal government, there are more specific federal laws regarding environmental protection and waste management. The Ley General del Equilibrio Ecológico y Protección al Ambiente (LGEEPA) or General Law of Ecological Equilibrium and Environmental Protection is the overarching fed - eral legal foundation for environmental protection in Mexico. Generally, this law, with six main themes, covers biodiversity, natural resource use, environmental protection, and environmental education and it defines the standards and modes of enforcement, oversight, and measurement (Diario Oficial de la Federacion 1988). The LGEEPA was first signed into law in 1988 at the end of the Miguel de la Madrid administration after nearly five years of constitutional provisioning for it (Assetto et al. 2003: 254). This motion was part of the 1983 reforms to Article 115 of the 1917 Mexican Constitution that sought to decentralize federal power to states and municipalities (Chapter 2). Whether or not these decentralizing reforms were underwritten by larger motions towards increasing democracy or rather in light of neoliberal structural adjustments following economic instability in the early 1980s is undeterminable, but similar movements occurred in the United States at the time President Reagan took power (Saint-Germain 1995). Initially, this law had a distinct section on all forms of waste, including municipal solid waste or garbage. Reforms to this law in 1996 further outlined these responsibilities with regard to the extent to which they could be satisfactorily met. Article 7 of the LGEEPA outlines the role that state governments have within this law. Section 6 of this article explicitly notes that state governments are charged with the “ regulation of the systems of collection, transportation, storage, management, treatment, and final disposal of municipal and industrial waste that are not considered hazardous…” [italics mine] (Diario Oficial de la Federación 1988). This specifically defines that “regulation” is the purview of the state gov- ernment in garbage management and allows for state government oversight of municipios. This statute gives state governments some governance over local garbage management. This regula- tory role assumes the understanding that state governments are more equipped, both technically 119 and financially, and are more politically disinterested in assisting municipios carry out their responsibilities in accordance to environmental norms and standards. It is also believed to struc - ture a dependency of municipios on state goverments, maintaining a centralist political hierarchy (Assetto et al. 2003). The following article, Article 8, defines the municipio’s responsibility in environmental governance. Section 4 states that “the application of the legal provisions relating to the pre- vention and control of the effects on the environment caused by the generation, transportation, storage, management, treatment, and final disposal of municipal and industrial waste that are not considered hazardous [italics mine] (Diario Oficial de la Federación 1988). Thus, at the federal level, the municipio is charged with the practical application of garbage management including collection and disposal. The municipio also maintains its right to levy fees for providing garbage collection as well as other public services. In October of 2003, the Ley General para la Prevención y Gestión Integral de los Residu - os or General Law for the Prevention and Integral Management of Wastes (LGPGIR) was passed by the Mexican federal government and became law in January 2004. This law more specifically addressed the governance of wastes, from garbage to liquid runoff to hazardous wastes, than did the previously noted LGEEPA. The LGPGIR comprehensively outlines the concept of gestión integral or integral management that underlies its existence apart from the LGEEPA. This newer law, as is defined in its title, takes into consideration the prevention of waste generation and a more holistic approach to waste management including valorization, reduction, reuse, recycling, and conversion of wastes (Galván-Meraz et al. 2008: 27). The LGPGIR’s holism is greatly informed by soft international policy on the environment in general and garbage specifically. Mexico is a signing country for the 1972 Stockholm and 1992 Rio de Janeiro UN Declarations. It is also a signatory member for the UN’s Agenda 21. Regarding the latter two UN documents, the conceptual framework of “sustainable development” of the Rio Declaration and Agenda 21 guide much of the latent discourse within Mexican federal environmental law. The concepts of minimization, garbage prevention, greater social participation, technology transfer, and universal accountability can be traced back to the 1992 UN conventions (Funez-Izaguirre 2004). The LGPGIR establishes a prioritization for waste management practices, in an effort to minimize their perceived environmental impacts (Figure 3.7). In a similar all-inclusive way, the law also carried with it the paradigm that implies that waste management is the responsibility of 120 all entities that generate waste, including private citizens and corporations. Within the law there are seven títulos or themes that cover, 1) the object and scope of the law, 2) the levels of govern- ment and their roles involved in waste’s governance, 3) the classification of waste materials, 4) the policy instruments of the law, 5) hazardous wastes and their management, 6) solid wastes and their management, and 7) security and control measures. At a practical level in the LGPGIR, the federal government is charged with broad policy development regarding wastes. Policy development is periodically defined through a Programa Nacional para la Prevención y Gestión Integral de los Residuos or National Program for the Pre - vention and Integral Management of Wastes (PNPGIR). This general policy guide reiterates the intent of the LGPGIR within the larger goals of SEMARNAT and the acting presidential admin - istration. The PNPGIR is the central federal program that provides a policy template for state and municipal governments. The federal government, through SEMARNAT, is also charged with the oversight of hazardous waste management. The Ley Federal sobre Metrología y Normalización (LFMN) or Federal Law on Metrology and Normalization (Weights and Standards) establishes standards, requirements, and procedures for waste management as well as thousands of other reg- ulatory functions across the federal government. These standards are known as normas oficales mexicanas (NOM) and refer to the approved final destinations of particular wastes among other topics. The Procuraduría Federal de Protección al Ambiente (PROFEPA) or Federal Attorney’s Office for Environmental Protection is charged with enforcement of the LGEEPA and LGPGIR. Figure 3.7. Prioritization of management in LGPGIR from most desired (lef t) to least desired (right) and the per - ceived environmental impac t from least (bottom) to most (top). I llustration by A. Hilbur n 121 Table 3.1. Legal responsibilities of the state and municipio under the LGPGIR. Data summar y from G alván-M eraz et al. 2008. 122 Article 9 of the LGPGIR specifies the role of the state government in the management of wastes. Like the LGEEPA outlines, the LGPGIR charges the state with a regulatory and oversight role over the municipios within its borders. The state environmental ministry, in whatever office it resides, is also required to draft a state-level waste management plan following policy outlined in the federal PNPGIR. In Puebla, this office is currently called the Secretaría de Sustentabilidad Ambiental y Ordenamiento Territorial (SSAyOT) or Secretariat of Environmental Sustainability and Land Management. This office was formerly called the Secretaría del Medio Ambiente y Re - cursos Naturales (SMRN) or Secretariat of the Environment and Natural Resources. The SSAy - OT uses the state program to assist municipios in drafting their own LGPGIR-required manage - ment plans. It also serves as a repository and registry for these municipio management plans. More practically, the state environmental ministry is also authorized to assist municipios in developing waste management infrastructure. For most rural municipios, like Coxcatlán, this typically involves the procurement of service vehicles, fuel purchase support, and when deemed worth the investment, the construction of a sanitary landfill. Sanitary landfills, in the case of small rural municipios with low populations, constitute significantly large investments of capital, land, planning, and administration and are usually only built after inter-municipal agreements are made between multiple municipios. This is the case for the five nearby municipios of Ajalpan, Zinacatepec, Miahuatlán, Chilac, and Altepexi that deposit their garbage in a jointly-managed landfill on the outskirts of Ajalpan. The state government also establishes and registers manage- ment plans for landfills with the federal government. In addition to infrastructural development, public education programs occupy a considerable part of a state environmental ministry’s efforts. Public signage, elementary school campaigns, and the promotion of location-specific projects, such as garden composting or recycling, are part of these efforts. Table 3.1 provides a translation of all the listed duties of the state government noted in Article 9. The following article of the LGPGIR, Article 10, outlines the governance of wastes, specifically municipal solid waste or garbage, by municipios (Table 3.1). The municipio’s role, defined in this article, is largely an operational one that collects and disposes of garbage or pays for a private operator to do it, as well as enforces the LGPGIR at a local level. A municipio’s government conducts these practices under the guidance of its LGPGIR-required Programa Municipal para la Prevención y Gestión Integral de Residuos Sólidos Urbanos (PMP-GIRSU) or Municipal Program for the Prevention and Integral Management of Municipal Solid Waste 123 that the state environmental ministry assists the municipal government to develop. The following section outlines practical day-to-day management of garbage by the municipio government of Coxcatlán, Puebla. MUNICIPAL MANAGEMENT APPROACHES Local organization and practice of MSW collection The presidente municipal is the chief executive in a Mexican municipio’s government. As such, he or she is ultimately in charge of and responsible for the operation of municipal solid waste management services and other requirements defined by Article 10 in the LGPGIR (Figure 3.8). Given the typically busy schedule of a presidente, solid waste service is delegated to one of his or her regidores or municipal regents. While the fields of administration of each regidor vary Figure 3.8. M unicipal gover nment struc ture and gar bage management. Char t by A. Hilbur n. from municipio to municipio, the regidor of ecology, public health, or sometimes public works is usually placed as the person to manage and organize the daily provision of services. In Coxcat - lán, the previous regidor in charge of MSW collection was of public health, currently the regi - dora of ecology manages pickup. Any major decisions regarding solid waste service provision are made by the municipal president. The regidor mostly acts as the manager for the typically small staff of local sanita- tion workers as well as maintains records of expenditures and activities. Regidores are also the primary contact with state environmental ministries and other governmental organizations. They see to any problems with infrastructure repair and manage the activities at the local dump. They 124 are also in charge of the placement and provision of public waste receptacles across the muni- cipio. The regidor serves as the primary public contact with issues regarding garbage. When I had questions about the collection service while conducting field work, I almost always went to the regidora first. The regidor’s role is solely managerial and most typically do not involve themselves in the daily operation of collection service. That job is left up to the sanitation crews. Like in the U.S., a sanitation crew in Coxcatlán consists of a chofer or driver of the garbage truck and two chalanes or workers who collect the garbage. The chofer is the de facto supervisor as their training in operating a commercial truck confers a leadership position. In Coxcatlán, there is one full-time sanitation crew operating out of the cabecera and a part-time crew in Calipan that works on collection days. Additionally, a crew consisting of a driver and a collector operate the recycling truck that follows the garbage collector. The vehicle infrastructure for garbage collection in Coxcatlán consists of one large late model MacNeilus garbage collection truck, a smaller older garbage truck for collection in Cali - pan, and a small Chinese cab-over-engine truck with a tall caged bed that serves as the recycling collection vehicle (Figures 3.9 and 3.10). All of these vehicles with the exception of the truck op - erating in Calipan are on-loan from the state government through what was the Secretariat of the Environment and Natural Resources and is now the Secretariat of Environmental Sustainability and Land Registry. The ownership of the vehicles by the state government prevents the common practice of selling municipal property at the end of the three-year term of a municipal president. Besides the vehicles used to collect garbage, the municipio manages the open-air dump- site that is just under two kilometers south of Coxcatlán town. The dump sits 100 meters west of Highway 980 on over two hectares of land donated by the ejido San Juan Bautista Coxcatlán when garbage collection began in 1996. It sits on an anticline framed in by two barrancas to its north and south and gently slopes to the west toward the Río Salado. It is currently an open air dump that is periodically topped with soil to prevent air contamination. Sections are periodically burned to decrease the volume of garbage at the site. A large pack of feral dogs subsists on the pickings at the site as well. A transferal station, consisting of a tall metal building housing a trash compactor and industrial truck scale, was installed at the dump in 2009 by the state government for the compaction and shipping of inorganic garbage to the nearby sanitary landfill in Ajalpan. Its use, however, was infrequent and the entire project was subsequently abandoned by the end of the terms of the previous municipal president and state governor. The reasons for the short 125 Figure 3.9. G ar bage collec tion truck in ac tion. Photo by A. Hilbur n, Date unk nown Figure 3.10. R ec ycle collec tion truck par ked outside the ayuntamiento. Photo by A Hilbur n, Date, Januar y 2011. 126 Figure 3.11. G ar bage dump south of the cabecera municipal. Photo by A Hilbur n, Date unk nown Figure 3.12. Shuttered and closed transfer ral station. Photo by A Hilbur n, Date, Januar y 2011. 127 operational life of the transferal station are often cited as a lack of training in its use, the lack of a significantly large truck to carry the compacted garbage to the sanitary landfill, and the end of the government’s term that provided the equipment. Today the station is fenced off and locked up and parts of the scale were burglarized recently. It is uncertain that the station will ever operate again. The sanitation crews operate according to prescribed routes and schedule. A typical day for a sanitation crew begins around 7:30 a.m. in the park at the center of Coxcatlán. The crew typically wears work clothes and boots with the chalanes (garbage collector on truck) using face masks and work gloves to protect themselves. The work begins with two chalanes collecting and emptying the contents of the plastic barrels that act as public garbage receptacles into the gar- bage truck, which is parked adjacent to the palacio municipal. The crew then loads the truck, the chalanes jump on the back, and the truck starts on their planned route for the day, announcing its presence to the public with short, discreet blasts of the horn. The recycling truck starts minutes after the garbage truck, running along the same route. Residents place their garbage beside their house or at the corner. The sanitation crew does not pass down every street in a colonia and some Figure 3.13. Putting gar bage out for oollec tion. View from collec tor truck . Photo by A Hilbur n, Date, July 2009. 128 streets become impassible from rain and use damage, so predetermined spots are chosen for the deposition of household garbage in some places. Some residents separate the plastic and card - board apart from the rest of the garbage and others simply mix their household garbage together. A costal or nylon burlap sack that once held sugar, corn, fertilizer, or animal feed is the primary container used to deposit garbage for pick up (Figure 3.13). Sometimes a 5-gallon cubeta or bucket is used. After dumping the contents of the bags or buckets, the sanitation workers place them where they found them and move on. Frequently residents, and especially shop owners, tip the sanitation crews with soft drinks and snacks. The truck moves along, stopping, backing up, and turning around to optimally follow the route without stopping until its entire length has been traversed, usually around 1 p.m. Following the last stop, the crew travels to the municipal dump where an appropriate spot is selected. The chofer then rolls up all the windows to keep out the seemingly millions of flies and horrible stench of burning garbage. A chalan operates the push rod that expels the garbage from the back and the chofer hits the accelerator briefly to dislodge the truck from the compacted mass of garbage. A very quick visual inspection follows and all board the cabin for the five-minute ride back to Coxcatlán’s park. The sanitation crew is paid weekly with chalanes earning 100 pesos (approx. $7.75) per day and the chofer making 250 pesos per day. Besides the collection routine, the municipal sanitation crew also collects the garbage from the public markets in Coxcatlán, Calipan, and Tilapa. In Coxcatlán, municipal garbage collection is offered free of charge. Municipal revenue sources such as fees for other public works and services in addition to general funds received through petitioning the state government pay for the workers’ salaries, fuel, and vehicle main - tenance. In other parts of the Tehuacán Valley, municipal garbage service is paid for by each household. In nearby Ajalpan, a 5-8 peso per receptacle charge depending on container size is levied by the municipal government, although revenue collection is difficult. The city of Tehu- acán’s sanitation department OOSELITE charges a fee, but revenue collection can be problem - atic as well and currently is significantly indebted to the private garbage collection concession there. Given the current low overhead for delivering service, support from the state government, and potential political and practical issues with payment collection, it is unlikely that Coxcatlán’s municipal government will begin to charge a fee for its municipal service any time soon. 129 Service provision and collection estimates Current municipal garbage and recyclables collection service extends only to the commu- nities along Highway 980 in Coxcatlán municipio (Figure 314). This includes Coxcatlán, most of Calipan, Pueblo Nuevo, San Rafael, and Tilapa. The collection crews operate six days a week with Sunday being a day of rest. In Colonia Centro, the colonias on its northern edge, and par- ticular streets ideally located along the route, collection service comes two times a week. Outside of the centro, in the peripheral colonias, once weekly service is the norm. Only the far extremes Figure 3.14. S er vice provision in the study area. M ap by A. Hilbur n. Data, 2011 Field R esearch and INEGI 130 of town do not receive any service. Calipan maintains its own municipal garbage collection truck, but this vehicle is not operated with the regularity of the Coxcatlán-based crew. Collection is typically once-per-week in most of Calipan, but service can be more infrequent along less- accessible streets and far-off colonias. A small private garbage service that operates in Calipan fills in the service void in these parts of town when residents are willing to pay the five peso per costal rate. In Pueblo Nuevo, San Rafael, and Tilapa service is provided once weekly on planned days at or around usually prescribed time schedules along a normal route through town. The sierra communities were, as of 2011, not included in the service provision area. Ac - cessibility and limited collection vehicles are the largest infrastructural obstacles to providing service in these communities. The highway to Zoquitlán, paved in 2001, is very steeply-graded and winding. A trip from the crossroads with Highway 980 in Coxcatlán to the ridgeline above Pala takes anywhere from 30 – 45 minutes and the various desviaciones to the communities themselves requires even more travel on the crumbling pavement at the top. Furthermore, the majority of sierra communities in the northern half of the municipio are situated off and below the paved highway on steeply graded dirt roads that can be subject to washouts and severe rut- ting in the wet season. The communities of Cobatepec, Calpulhuacán, and Tequexpalco are only accessible by a dirt road that is usually in poor condition. Sierra communities in the southern half are even more isolated from convenient road infrastructure as the only paved road access to the general proximity of these towns runs through the adjacent municipio of Teotitlán del Camino, Oaxaca before reentering Coxcatlán. An hour to an hour-and-a-half can be required to reach communities like Vigastepec, Tres Ocotes, and Potrero. Given the fact that currently only one consistently used garbage collection truck and crew operates in the municipio, the travel times and spent fuel resources would either require more trucks and crews or the scaling-back of collection service frequency in the valley communities. Compounding the problem for deliver- ing household garbage collection service, many of the sierra community households are spread apart across fairly large distances and elevations within a given community. Many households are linked to each other and the town center only by footpath. However, nodal collection could offer a solution that seems to work in similar urban layouts in nearby Zoquitlán and Coyomea - pan. Aside from the communities in the sierra, there are a few dispersed hamlets of two to three households on the hillsides raising goats and cultivating wet-season crops that do not have col- lection service. 131 Within the collection service provision area in Coxcatlán municipio, there are 3,204 households (INEGI 2010). While relative access to service varies by location, most households can expect at least once-per-week collection with the exceptions above noted. Assuming the very conservative weekly MSW generation rate of 6.99 kilograms of municipal solid waste produced by the average Coxcateco household (Sánchez Mendez and Olmos Torres 2009), this sums up to 23,396 kilograms of municipal solid waste generated and presumably ready for collection each week. Using this estimate, Coxcatlán potentially produces almost half of the MSW collected with 10,555 kilograms per week. Tiny San Rafael generates the least, with an estimated 398 kilograms per week. Outside of the service area, 1,184 households have to contend on a number of strate - Table 3.2. Estimates of gar bage generated and potentially collec ted and uncollec ted in the study area. Only the valley settlements have collec tion. Data from S anchez-M endez and Olmos-Tor res 2009 and INEGI 2010. 132 gies to dispose of an estimated 8,276 kilograms of municipal solid waste per week. However, this weight estimate assumes that sierra households produce as much MSW as a Coxcateco household, which is doubtful as most sierra households would have significantly less access to material consumables that are destined to be collectable garbage. Yet even three-quarters (6,207 kg), half (4,138 kg), or a quarter (2,069 kg) of this estimate still amount to considerable weekly amounts of solid waste for these households and communities to manage. Considering the sierra communities as individual places that generate MSW rather than a total of summed households, the towns of Tecoltepec with its 240 households and Ocotlamanic, home to 235 households, are potentially generating substantial MSW within their boundaries even with the most conservative estimates of weekly household MSW generation. THE PRACTICE OF HOUSEHOLD GARBAGE MANAGEMENT IN COXCATLÁN Collecting and storing The production and management of household garbage begins with an initial separation of the embodied self and lived space from the unwanted, wasted matter that becomes garbage. This process involves an initial staging of sorts for garbage. Regardless of whether garbage col - lection service exists or not, most households in the study area collect garbage in a number of ways around the house lot until its quantity exceeds a known threshold and must be disposed. The solar or garden area of a house lot offers ideal places to collect, hide, and store garbage and waste outside of sight. If a household recycles, it is common to see recyclables left to accumulate in a hidden spot behind a shed or less-accessible corner until their size merits collection or sale to municipal recyclers or scrap buyers. Large items such as busted concrete, scrap metal, old appli- ances, and even dead animals are either deposited along the perimeters of house lots or put in an inconspicuous but typically organized montón or pile somewhere in the lot, referred to earlier as the toft area. Some designation of later reuse may be implied to some of the items in the pile. The house space is used as a storage area for garbage. In the kitchen, many households keep a bucket or small satchel to throw scraps of uneaten food, food preparation waste, and pack- aging refuse for later disposal. In the study area and most parts of Mexico, the sewer systems, septic pits, and latrines cannot receive toilet paper without stoppage and damage, so used toilet paper is disposed of in plastic bags and waste baskets until it is thrown out with the garbage or burned in a pit. Houses are typically modular and constant construction is a common feature in 133 Figure 3.16. Buckets being reused as planters. This is a common reuse of an item in the study area. Photo by A Hilbur n, November 2011. Figure 3.15. Old Pepsi bottle being reused as a container for pulque. Photo by A. Hilbur n, D ecember 2011 134 the study area. Rooms that are under construction and isolated form the lived space of the house are sometimes used as a depot for garbage and unused items. Reuse and recycling Reuse of ostensible waste products is a common practice in the study area. The forms of reuse are seemingly unlimited but some are so common as to render the use of specifically designed, task -purposive items such as ready-made plastic buckets unnecessary. Packaging is typically the most reused form of items that would typically be considered disposable or garbage. Plastic bags from stores are the most common form of reuse. They are cheap and almost always offered to hold merchandise and prepared food items, even from the humblest of shops. After their immediate use, they are employed in storage of household items, tote bags, marking flags in agricultural fields, and for the collection and deposition of garbage. In contrast to their high reuse value, littered plastic bags are often cited as one of the larger garbage-related concerns for study area residents. The same mixed assessment applies to plastic soft drink bottles. While they are common litter and often discarded and recycled, plastic bottles also serve as drinking canteens for outdoor laborers. Besides their use in the storage of numerous wet and dry goods, they also are used in the sale of locally produced alcoholic beverages such as pulque , a lightly alcoholic beverage made from fermented liquid collected from certain Agave species, and caña , a cheap artisanally- produced liquor distilled from sugarcane (Figure 3.15). In many sierra households and avid gardening households in the valley, the tops of plastic bottles of an array of sizes are removed and used for small planters for the raising of chili and tomato seedlings before transplantation in the garden or fields. The most common maceta or flowerpot in the solares in the study area is likely an old lard, detergent, or motor oil bucket (Figure 3.16). This usually occurs when a bucket’s reuse as a water pail, garbage container, or some other function is no longer desirable. Cardboard boxes are kept by many households not just for storage of items but for use in travel and the long-distance transportation of many or large items. Most public buses in the region have an array of corrugat- ed cardboard boxes reinforced with string that also serves as a handle in their cargo holds. They serve as a type of luggage for less-affluent and frugal residents. Recycling is a growing trend in the study area. For plastics and cardboard waste, the 135 Figure 3.17. Local rec yclables collec tor pile. Photo by A. Hilbur n, O c tober 2011. Figure 3.18. Fierro viejo collec tor truck . Photo by A. Hilbur n, O c tober 2011. 136 increased availability of recycling collection by the municipal government is likely the most con- siderable factor for this growth. Some private collectors store plastics (Figure 3.17). The growing perception that recycling of these materials is more environmental-friendly than throwing them into the waste stream or dumping and burning also likely contributes to increased recycling. Af- ter collection, the PET and cardboard are stored by the municipal refuse workers in a room in the new fiesta hall until it is trucked to Tehuacán and sold to recyclers. Metal is by far the most valu- able waste material in the study area. Besides the recycling service provided by the municipal government, a few independent collectors buy scrap metal and aluminum from households to sell to metal recyclers in Tehuacán. These buyers, called locally fierro viejo or old iron, drive pickup trucks through area streets announcing their presence with an announcement and repetitive jingle on a loudspeaker atop their vehicle (Figure 3.18). The price they pay for the scrap is below market value (1-2 pesos per kilogram) but the convenience service they provide provides them with ample business to continue to operate. On the occasion that PET (#1 plastic or polyethylene terephthalate) bears a high enough value to make it a worthwhile recyclable, the independent col- lectors will buy it at an almost negligible price of 1 peso per kilogram. Dumping and burying Dumping collected refuse in an area outside of one’s own lived space is perhaps the most ancient form of garbage management. It remains the preferred method of most contemporary so - cieties, albeit in a controlled fashion in government-sanctioned space. Many of the most produc- tive archaeological sites excavated by MacNeish and his collaborators in the 1960s were in fact dump sites. Today, the final destination of garbage collected by municipalities is the dump and this accepted form of collective dumping should be distinguished from dumping by individual households. Yet, throwing garbage outside the prescribed acceptable areas is a common practice for households, if contrary to most municipal laws in Mexico. Dumping in this case, is defined as the consistent disposal of refuse in bulk outside of the household. It should be distinguished here from littering, which is less a sustained management practice and more of a one-off instance as well as the difference in magnitude of disposed waste. Regardless of definition, dumping is generally seen as a less-desirable if not morally wrong practice today. However, the lack of per - sistence of solid waste in the past made dumping a less objectionable practice (Deal 1983) and it likely continues today because of old habits. Some residents see dumping as an option for man - 137 aging their trash on their own time and schedule. Those who dump in the monte or wooded areas around the cabecera explained that dumping gives them the freedom to dispose of garbage when they want to and that the dumps are long established and do not pose much of an environmental risk. It was seen as commonplace for a particular family to dump because it had always dumped and was not willing to change. Wooded and heavily vegetated areas, vacant lots, and roadsides, all areas of imprecisely understood ownership, are ideal sites for dumping. Barrancas and stream beds are also highly selected areas in that it is assumed that the next rain event will simply rid the area of the garbage (Figure 3.19). These informal dumpsites are known locally as tiraderos clandestinos or clandes- tine dumps and vary greatly in size from a small pile of about five square meters to almost half a hectare (5,000 m 2). Some of these sites are the result of one to a few social events and some are fairly active dump sites for select members of a community. Others are known dumping areas for people traveling along the highway, almost as if the undefined area of the tiradero clandestino were a trash receptacle of sorts. An almost total lack of enforcement against dumping combined with large unsupervised undeveloped spaces and the general ease of unloading garbage makes Figure 3.19. Small dumpsite in the Bar ranca S oyolapa. Photo by A. Hilbur n, S eptember 2011. 138 the practice fairly widespread across the municipio and surrounding region. Dumping typically is not the only method a household may use to manage its garbage. It is often a highly circumstantial practice. If a family misses out on or is not served by local service for some reason, then taking the trash to a vacant lot or tiradero clandestino becomes an option. Also, the barrancas and roadsides are often sites where the castoff remnants of par- ties or events are dumped due to a lack of trash receptacles or the unwillingness to transport the garbage home for municipal pickup (Figure 3.20). While it is uncommon in Coxcatlán, dumping by groups from outside the municipio, such as animal processing plants and light manufacturing plants known as maquilas, exists across the Tehuacán Valley. Formal agreements and payment regarding dumping rights between ejidos and industries can exist but are typically outside of legitimacy and are not followed to precise terms. In the study area, some communities still maintain small dumps in and around their town limits. Many of these dumps have been in operation for years and served as a place to dispose of garbage that a household did not want to burn or bury at home. Before municipal garbage col - lection these dump sites were the only option for sanctioned dumping. Today, some of these sites have been cleaned up through faena or community organized work projects. Others have been Figure 3.20. Small tiradero clandestino along the highway. Photo by A. Hilbur n, April 2011. 139 repurposed for receiving garden waste and large unusable organic matter such as rotten logs, spiny limbs, and dead animals. In some sierra communities, these dumpsites still function as they always have and remain the best option for large waste disposal as there is currently no collec- tion service there and available land for dumping is increasingly rare. Burying garbage today is not as common as it was in ancient times. Many of the middens and privies found by archaeologists across Mesoamerica are often old unused storage pits. These converted storage pits were excavated at a few household site at Sansuanchi or Coxcatlán Viejo (Sisson 1973). Today, the time and sweat investment in excavating a pit out of the often hard soil just to fill it with garbage significantly limits the practice of burying garbage. Furthermore, most households produce more garbage than they could possibly dig deep enough holes to receive the garbage. Typically, the only things buried are dead animals such as deceased pets, opossums, skunks, and inedible birds. If garbage is buried, old, out-of-use pits are filled instead of fresh excavations. Burning Along with dumping, burning is one of the oldest methods of dealing with unwanted matter. Before garbage collection was provided in the mid-1990s in the cabecera and Calipan and the southern valley communities in 2005, nearly every household in the municipio burned its more persistent garbage in heaps or pits around the house lot. Currently, it is the preferred meth- od of garbage disposal in all of the sierra communities. Fire is easy to make, costs almost noth - ing, and almost entirely destroys the unwanted matter. It also is believed to sanitize matter that is considered unclean, turning infectious rot and decay into presumably neutral ashes. Most of the burning in households that use burning as their primary disposal method takes place in square meter pits lined with stones or bricks to prevent cave-in (Figure 3.21). Gas - oline, kerosene, denatured alcohol, or sometimes dried vegetative matter is used to start the fire when less-than-flammable materials are involved. Households that have collection service and do not typically burn garbage often tend to burn in piles, not wanting to establish a more permanent area for burning. Sometimes households will burn small garbage piles in alleyways and the street so as not to contaminate their own private house lot (Figure 3.22). This also occurs in the small dumpsites located in the barrancas that pass through urban areas. In the sierra communities, most households tend to burn primarily the inorganic garbage 140 Figure 3.21. Large bur n pit. Photo by A. Hilbur n, S eptember 2011. Figure 3.22. Bur n pile outside a home. Photo by A. Hilbur n, July 2009. 141 Figure 3.23. A bucket of compost before its empt ying in a pit. Photo by A. Hilbur n, November 2011. Figure 3.24. Organic refuse pit. Photo by A. Hilbur n, November 2011. 142 they produce. Toilet paper, disposable diapers, plastic bottles and bags, and any unwanted paper are thrown in a small pit for burning. However, most of these households do not produce all that much waste. Also, residents there only really see inorganic garbage as wasted, useless matter. Scraps of food and most organic waste matter do not have the same uselessness implied in the more urbanized parts of the study area, Mexico, and the United States. Rather, those items are typically thrown into the garden, a pit for organic waste, or are fed to animals. Composting, “abono”, and keeping animals The solar or house lot garden area in many homes operates as a garden area. What is planted there, if anything, varies from house to house. Many households have fruit trees, a small garden area, a planting of nopal, and decorative shrubs and flowers or some medicinal herbs. Also, some households in the more rural, agrarian parts of the muncipio live among some of their agricultural fields. As is the case with many gardening households across the world, organic household waste is used to enrich the soils of household gardens. While the majority of organic matter used to enrich the often poor temporal soils of the study area is humus from forested parts of the municipio and surrounding area, the food scraps, egg shells, coffee grounds, fruit peels, and other kitchen detritus are kept in a bucket until they are either dumped directly into the field or garden (Figure 3.23). Some households excavate a pit for a type of passive composting where - by the decomposed contents of the pit are infrequently taken out to mix in with tilled soil or planting media (Figure 3.24). Many of the more rural households keep some form of domesticat - ed animals, as well as pets. Chickens and turkeys are common in the yards of sierra households and these animals typically eat old, inedible tortillas, rotted fruits and vegetables, as well as any non-toxic unwanted matter. If a household keeps hogs or goats, these animals often consume any uneaten food, although the practice is discouraged by health officials as well as folk knowledge to prevent contamination of the animals’ meat. Dogs and cats in the less-affluent households also act as garbage disposals, always scavenging for anything edible. THE GARBAGE MANGAGEMENT TRAJECTORY IN COXCATLÁN From the moment it is produced to its decomposition or reconfiguration into something not seen as waste, garbage enters and passes through a trajectory of management. In Coxcatlán, the trajectory of garbage begins with consumption. Consumption, whether for necessity, for de- 143 sire, for enjoyment, or for conspicuous ostentation (Veblen 1994: 43) invariably produces waste (Kennedy 2007). Garbage is waste and waste is anything that has lost its use value or becomes undesirable. For individual citizens and households, garbage generation is the result of preparing food, eating, drinking, working, breaking things, fixing things, making things dirty, cleaning, and simply allowing things to decay or lose value. Garbage is then removed, turned into something useful, put out of sight, tossed, or destroyed by these individuals and households. At any rate, following its establishment as waste, this matter enters a generally defined management trajec- tory (Figure 3.25). Garbage produced in public and outside of the home can be thrown on the ground or placed in a public receptacle. If it is littered, the litter can be collected by public sanitation work - ers or in some cases cleaned up by private citizens, becoming their garbage. If it is not managed or dealt with, then the garbage just becomes litter and is apart from the larger system described in this section. In the valley settlements, public receptacles are periodically emptied and this pub - lic garbage is collected by the municipal sanitation crew. In the sierra, some towns like Xacalco Figure 3.25. The gar bage trajec tor y in the study area. Char t by A. Hilbur n. 144 have public receptacles that are emptied of their contents which are then recycled, burned, or dumped elsewhere by residents placed in charge of this work. Waste generated by households can first avoid becoming garbage by being repurposed in some way, if only temporarily. A soft drink bottle can be reused as a canteen or an empty lard bucket can become a planter pot for flowers. If not, then that garbage can be managed by tradi- tional methods outlined above to varying degrees and frequencies. Some houses may compost, some burn, others may feed some organic waste to pets and livestock, a few may keep pits or heaps, and a handful dump garbage outside the house. In the sierra communities, households manage their garbage by a combination of the home-based practices listed above. Some of the larger communities, like Tecoltepec, keep a community dump where residents can dump some of their garbage. These community dumps are increasingly becoming disused or converted into other uses. The valley community of San Rafael kept a community dump before collection ser - vice arrived in 2005. Today, this site is only used to deposit what in the US would be considered “trash”; things like unusable wood, tree limbs, old dead cacti, busted rocks, and old concrete. In communities where there is no service, most households understandably tend to take care of their garbage where and when it involves the least effort, which is in the house lot whenever the chance arrives. Most households in the valley put their garbage out for collection when the municipio’s carro or truck passes by to collect it, but most may also use some other more traditional garbage management practices such as composting. The valley households may separate the cartón or carboard and recyclable plastic from their garbage and leave it out for collection by the mun- cipio’s recycling truck. Other households separate only their metal waste, everything from unus - able rebar and large metal junk to food and drink cans. This is typically sold to the scavenging metal recyclers collectively referred to as fierro viejo. In the sierra communities, households manage their garbage by a combination of the home-based practices listed above. Some of the larger communities, like Tecoltepec, keep a community dump where residents can dump some of their garbage. However, these community dumps are increasingly becoming disused or converted into other uses. The valley community of San Rafael maintained a community dump before collection service arrived in 2005. Today, this site is only used to deposit what in the US would be considered “trash”; things like the aforemen - tioned unusable wood, tree limbs, old dead cacti, busted rocks, and old concrete. In communi - 145 ties where there is no service, most households understandably tend to take care of their garbage where and when it involves the least effort, which is in the house lot whenever the chance ar- rives. All of the garbage that is collected by the municipio’s sanitation crew eventually ends up at the municipal dump, south of the cabecera. This garbage is piled up and occasionally set afire to minimize volume. Yet for some of the garbage left here, this does not signify the final stop on the garbage trajectory. Although dumpsite garbage collecting is a highly structured and formalized industry in Mexico’s large cities (Castillo-Berthier 2003; Medina 2000; Medina 2005; Ojeda-Benitez et al. 2002), there are occasional garbage pickers, called pepenadores, who visit the dump to opportunistically find something of reuse value or something to sell to recyclers (Figure 3.26). Some pepenadores may even be the fierro viejo recyclers looking for metal that was discarded in the garbage stream. Besides human agents, there are packs of stray dogs that live in and near the dump that subsist off of scavenging the smoldering piles of garbage for ed- ible items. Garbage that is intended for recycling follows another divergent trajectory. The cardboard and recyclable plastic collected by the municipal recycling crew is taken back to the cabecera Figure 3.26. Pepenadores (waste pickers) at the municipal dump. Photo by A. Hilbur n, June 2009. 146 where it is stored until a sufficient amount or price is reached to merit its shipment to Tehuacán to a recycling plant or scrap collector. In 2011, the recyclables were stored in the new salon de fiestas or town dance hall whose intended use has not yet been fully realized due to the remain - ing popularity to host big events in the more centrally located market building. Once the recy - clables are sent to Tehuacán, the money earned from the sale pays for the transferal shipment and the rest is ostensibly reinvested into the maintenance of the sanitation service. The private scrap collectors will also store the metal and plastic they collect until an assumed least-cost optimal- ity is achieved or the need for liquidity pushes them to sell their materials down the recyclables commodity chain. The garbage trajectory outlined above notes the management path through which gar- bage passes on its way to final disposal, recycling, or reuse. However, the quantity of material garbage that passes through this system is not outlined nor does the garbage trajectory estimate the frequency with which the management practices are used. Estimating the quantity of garbage that passes through the system is an immense task beyond the scope of this project. However, by recording the frequency of practices that households employ to manage their garbage can add further detail to the broader garbage management system outlined above. The following chapter, citing a municipio-wide garbage census, provides this more detailed perspective on how garbage is managed in Coxcatlán. 147 CHAPTER 4: Household garbage management in Coxcatlán While in the planning stages for this dissertation, I felt it would be necessary to conduct a generation/classification study in a smaller settlement to compare to the existing one conducted in April 2009 (Sánchez-Mendez and Olmos Torres 2009) to better understand the garbage issue in Coxcatlán. This research question came from the gap in a vibrant and growing academic lit- erature on garbage generation in Mexico. At the time and still today there were no such studies in contexts outside major metropolitan areas in Mexico. My aims to conduct a generation study in an agrarian village could both fill that gap and provide my larger project with an accurate base- line to frame the materiality of the garbage issue in the municipio. Yet after some months in Cox- catlán I abandoned this idea for a few reasons. In hindsight, staying with this research goal would have been a major challenge with mixed results at best. First, the practicalities of simply doing it were hardly achievable at the time. Getting a sizable sample of households to participate in such a study outside something sponsored by the state or municipal government would have been highly prohibitive. Attempting to use the same methods as the previous study in Coxcatlán would require households to fill a bag given by the researcher (me) that would then need to be collected by the researcher (me) for a period of an entire week. The garbage would then need to be sorted and weighed by class in a clear space and finally be taken for disposal. In reality, to realize this study would have required a considerably sizable team of paid researchers who would need to be trained, given the absence of this line item in my funding, was practically impossible. In terms of community relations, this aspect of my research would have been purely academic; the muni- cipio’s government had its mandatory study for its management plan and any assistance offered by them would really just be support on their dime. More importantly, such a study would have overwhelmed my time and focus and would have taken me away from my overall focus on the management of garbage, both materially and as an environmental issue. In its place, I planned to conduct a more useful and practical set of projects. To date, there were no systematic surveys in a place like Coxcatlán on how households managed garbage. One could assume that those with service put their garbage out for collection, some burn, others compost, and that recycling is going on. Yet the frequency with which households use any and all of these practices and how those relate to particular household demographics remained unclear. Thus, the research goal to conduct a garbage generation study morphed into a garbage manage- 148 149 ment census at the household level from a large and widely drawn sample. While the generation study would have been useful and is a goal for future research, a more refined understanding on how that material is managed is more useful. So is its management. Dealing with garbage, whether it is by dumping, littering, burning, or recycling, is the final and perhaps key engage- ment between society, consumption, and the environment. Waste management caps the trajectory or closes the loop between extraction, production, and consumption by society in the environ- ment. A detailed empirical look into how garbage is managed by households in a number of different locational contexts in Coxcatlán municipio was needed to accurately frame this engage- ment. This chapter presents the results of the municipio-wide garbage census taken between the months of August and November 2011. OBJECTIVES AND METHODS Project intent Simply dividing the Coxcatlán municipio into households with collection service and those without does not give a very useful or interesting description of how garbage and wasted matter is managed there. Across Coxcatlán municipio there are mixed methods each household utilizes to deal with its garbage given unequal access to services, different household ameni- ties, and idiosyncratic preferences. In the previous chapter, one can see these different practices and the circumstances that frame them. In order to get a more full and empirical record of how households “take out the trash” across the municipio of Coxcatlán, I administered a household garbage management census in conjunction with the risk mapping exercise (Chapter 6). The de- scriptive intent of this data is to give an empirical narrative of the daily reality of garbage man- agement in a number of different settlements and across different household types. Additionally, the ostensibly less-related variables of the households and participants provide a fuller picture of daily life and demographics than that provided by national census data and anecdotal observa- tion. Government policy and official collection practice likely does not account for all of the practices used by households to manage their garbage. Lastly, this study is an attempt to what Bulkeley and Gregson (2009) call “opening-up the black box of the household” regarding real and actual garbage management. Often policy stops at the front door, ignoring the diverse prac- tices that households use to manage their garbage. The intent of this study was to knock on those doors and ask about what was going on inside. These are the results. The household and population sample The basic unit of this census was the household. I selected the household as individu- als typically do not manage their own garbage in isolation from other members of a household. While not all members of a household participate in burning, dumping, burying, or placing garbage by the entryway for collection, garbage becomes aggregated within and becomes the assumed property or responsibility of a particular household. The majority of garbage, in a sense, is the result of the production and consumption or “metabolism” of a household. People tend to think of garbage this way, as something a household or family produces, something from the “we” rather than the “I” or self. The head of household or person most involved in dealing with the garbage was sought out for the survey, yet if another member of the household had something to add or discuss between the individual being interviewed, then their comments were included as well. Nevertheless, the individual surveyed does represent the household to a degree as well as indicates individual-specific data regarding management practices. Thus, more individual-based variables such as gender and education in addition to others were solicited. The survey sample population for this research includes 520 households across thirteen communities in the municipio of Coxcatlán. The survey included households from the valley communities as well as those in the sierra communities to the east of the cabecera, representing a diverse sample of situational and community contexts. In the valley communities, 353 house- holds were surveyed. Out of these 353, 145 are from the cabecera municipal with 106 from nearby Calipan and the other 102 spread across Tilapa, Pueblo Nuevo, and San Rafael. The re- maining 167 come from the seven sierra communities of Calpulhuacán, Cobatepec, Ocotlamanic, Pala, Tepeyoloc, Tequexpalco, and Xacalco (Figure 4.1). The relative percentage of households sampled per settlement are posted in Table 4.1. An opportunistic sampling procedure, based on availability in the home, drove our participant selection with recognition of an estimate of the relative size of each settlement and their proportion of the total sample population. The research team for this census was comprised of me and a local investigator from Coxcatlán town. The local investigator I selected was a young man who besides being a mason and a part-time musician had previous experience in questionnaire administration and was well- known in various communities across the municipio and surrounding region. Following a brief 150 negotiation over payment, work hours, and expenses, I conducted four, 4-5 hour training sessions that outlined questionnaire procedures, covered human subjects and consent issues, and included practice through mock scenarios. Besides receiving training for the project, the local investiga- tor provided a useful assessment and critical response to the applicability and feasibility of my prepared questionnaire, which greatly streamlined the survey process. I provided additional Figure 4.1. Garbage census participant households by community (n = 520). One rectangle equals one house- hold. Chart by A. Hilburn. 151 training in spreadsheet data entry as well as a basic introduction to GPS data collection. Follow- ing this initial training, the local investigator and I tested the questionnaire while practicing our survey procedure for two full days in Coxcatlán town. After realizing that particular aspects of the questionnaire were imprecise and inapplicable or unfeasible given the goals and limitations of the project, the questionnaire was then revised and mulled over by the research teamover the next four days before starting data collection. Prior to conducting surveys in communities, the local investigator and I met with the au- thorities of each community in direct meetings or through less formal introductions through the municipal president and his staff. This was usually a meeting where we presented our permission from the municipal president followed by a description of the intent and content of the project. After receiving authorization, we conducted the census in a number of ways that seemed appro- priate or feasible given the town in which we were working. In the majority of towns, we conducted the survey door-to-door, selected participants who Table 4.1. Garbage census participant households as apercentage of total households in a community commu- nity (n = 520). Total household census population from INEGI 2010. Table by A. Hilburn. 152 were willing and were over 18 years of age. Given the generally assumed importance of garbage management in the lives of Coxcatlán municipio’s residents, most were eager to participate. Some were understandably perplexed why an American, first, wanted to live in Coxcatlán of all places, and second, about my decision to study garbage. In spite of that and based on very little but sheer observation, I argue that Coxcatlán residents in general were far more open to partici- pating in a questionnaire or interview than a fellow American. Subtracting my own positionality and the fact that talking to a gringo was a novel event for some, I believe the practice of public surveying, especially door-to-door, is more commonplace in the study area and across Mexico by government and non-government agencies alike. Residents there are more accustomed to donat- ing their time to participate. Most towns in the valley were compact and had fairly simple gridded town plans. It was not difficult traversing the breadth of each town we selected. The sierra communities typically had more dispersed settlement morphology and we would walk considerable distances as well as climb and descend hundreds of meters just to survey one part of a community. In every home that we observed to be occupied and not obviously engaged in eating or some other private activ- ity, we called out or knocked to ask for participation. We conducted our survey during daytime hours, usually following breakfast and before the late afternoon comida or dinner around 3:30 or 4:00 p.m. This time block was chosen because these hours did not interfere with time most families reserve for eating, relaxing, or household chores. With each participant, we gave a brief description of the intent and purpose of the research as well as obtaining consent to participate before administering the census. In many instances, the participant households already knew us, knew of us, or had seen me around before. This prior familiarity with us helped in establishing rapport with the participants. In a few instances in the sierra, groups were addressed in groups and following a descrip - tion of the project, participants were interviewed individually. In Tequexpalco, we interviewed 17 mothers in the town hall building who were coming to drop off and pick up their children at the elementary school next door. The health officer at the Tepeyoloc weekly health clinic was kind enough to allow us to interview 17 visitors coming to occasional public health discussions or platicas. This was helpful for us as many of these communities have difficult-to-access areas and interviewing in a central location greatly diminished the time required to traverse the entire community or portion of the community. Taking the survey outside the home in a public place 153 also perhaps eased trust concerns in these communities, which receive few if any outside visitors. The basic survey instrument of this census was a questionnaire followed by a brief contextual open-ended interview to clarify certain points and to see if the participant had any comments to add. In conjunction with but separate from this survey, we conducted a risk map- ping exercise. Not all participants had the extra time to participate or could not provide any information for the latter exercise so the survey population for this study is significantly larger by 87 participants. All but one interview was conducted in Spanish with the exception being one participant who was eager to practice his fluent English that he had learned working and living in the United States. A few of the participants in the valley and nearly all in the sierra communi- ties spoke Nahuatl but all were fluently bilingual with the exception of one participant, who had her grandson translate. The questionnaire was administered and responses filled in by me or my investigator. To accommodate the sensitivities of the participants, we informed each participant that if there was a question that they did not feel comfortable answering or were unsure of, that it was their right not to answer or answer vaguely. This is reflected in an “unsure or no response” category in the results. While this research was not participatory, we purposefully opened the clipboard and questionnaire form to the participant to allow them to peruse the document in the research process. We also allowed them to view and check their results and responses during and after the survey and interview. RESULTS Participant and Household Demographics While this census focuses primarily on the household as its basic individual unit, the demographic data on the participants who provided the household-level data can provide insight into the character of the household. For this reason each participant’s sex, age, head of house- hold status, education, occupation, residency status (native/not native), and civil status (married) were recorded. These statistics allow us to indirectly gauge the economic and social status of a particular household as well as the community to which it belongs. Moreover, these demographic statistics provide empirical insight across the municipio into the households that manage their garbage. For the entire study population (n = 520), we interviewed 388 women (74 percent) and 154 Table 4.2 . S ex , h ea d of h ou se ho ld s ta tu s, a ge , a nd e du ca tio na l a tt ai nm en t b y pa rt ic ip an t. D at a, D is se rt at io n fie ld w or k 20 11 Ta bl e by A . H ilb ur n. 155 132 men (Table 4.2). This gender bias is a reflection of the fact that the practical heads of the household for most of the daytime hours are women. Housework in Coxcatlán as well as across Mexico is considerably gendered (Christie 2008). Domestic labor and housekeeping, which includes taking care of the garbage, typically fall within the expectations of women. Women, in general, are thus more informed participants regarding household garbage management than are men. Considering the gender bias as a practical matter for conducting this survey, women were typically more available and more present in most households. When men were encountered as the first contact of a household, we were often referred by them to interview the doña or woman- in-charge of the household to interview as it was implied by men that taking care of garbage was something women did in the house, comparable to cooking and laundry. This gender bias was even more apparent in the sierra communities, where 141 of 167 participants or 84 percent were women. A comprehensive study of the gender division of labor in Mexico also noted the gendered nature of domestic labor in Mexican homes (Pedrero, Rendón, and Barrón 1997). Our research considered this and allowed it to guide the participant sampling. I wanted to record the most accurate and comprehensive household information as possi- ble. The likely best choice for a potential participant household would be one of its heads as they would be the most intimate with the daily activities of home maintenance. A head of household would also likely be the most responsible members in that they would give the most honest and considerate responses to the survey questions. Out of the 520 individuals, we interviewed, 381 or 73 percent claimed to be the jefa or jefe of the home. If the head of household was not selected or available, we typically interviewed someone who appeared competent in daily household mainte- nance such as an adult child of the head of household or a cohabitating cuñada(o) or in-law. Age of the participant is perhaps a more useful variable for the risk mapping exercise because of its ability to attach a generational dimension to individual perceptions of garbage and the environment (Table 4.2). Nevertheless age can be utilized in this survey to potentially situate certain garbage management practices with the same generational differences. The relative age of a head of household, along with other variables, can also help infer the relative affluence of a household as it typically takes decades, if ever, to develop household wealth and equity. Fitting with the goal to interview the head of household, the mean age of the participants was notably high at 41.9 (s = 16.0) years of age. The age range of all the participants spanned from a 90 year- old don to a few individuals at the minimum required age of 18 years of age, recognized by the 156 Mexican state as the official age of adulthood. The mode or most frequent age of those inter- viewed were 30-year-olds, with 23 individuals at that age. In Mexico, formal education presumably affects one’s awareness of environmental is- sues (Fernández-Crispín 2009; Guevara-Martínez and Fernández-Crispín 2009). The practice or avoidance of a particular method of garbage disposal such as participation in recycling or litter- ing are conditioned in school for particular generations. Education attainment can also be used as a proxy statistic for affluence as well in rural Mexico (Parker and Pederzini 2000: 37-38). The mean years of school attendance for the entire study population was 6.7 (s = 4.1) years. The sierra communities have a much lower mean educational attainment of 4.3 years (s = 3.0) than the valley communities’ 7.817 years (s = 3.99) for a number of reasons, including the need for school-age children to work in the home, little precedent for and subsequent low appreciation for formal education, and for the older generations, there was little to no access to education beyond primary school. Since 1992, it has been a federal mandate that children attend school from first to ninth grade, but enforcement of this statue was typically low, especially in previous decades (Parker and Pederzini 2000). From the 167 individuals surveyed in the sierra communities, only one had attended formal schooling beyond 12th grade and 23 percent had never attended school. This variable is likely an underestimate of educational levels for the entire sierra population of Coxcatlán municipio because the surveyed population was older, reflecting lower educational attainment of previous generations. Today, primary and secondary education, analogous to elementary and middle school in the U.S., respectively, is more valued and current attendance is around 98 percent (INEGI 2010). Primary and secondary school attendance are also conditions for federal, World Bank-supported social welfare payments called Oportunidades . The gender bias of the survey may also underestimate the educational attainment as women were and still are in some cases less expected to attend school beyond primary school (Parker and Pederzini 2000). Out of the 39 sierra participants who did not attend school at all, 33 were women. Estimating household income to a numerical value by the type of occupation(s) of the heads of household is possible but is perhaps far too indefinite given the diversified nature of employment and money earning in the study area. The occupations of the surveyed participants, however, attempt to give one aspect of a particular household’s economy and the study popula- tion’s households. Unpaid domestic labor was the primary occupation for the survey participants. A full 308 individuals, all female, responded “ama de casa” or housewife to the occupation ques- 157 tion of the survey. This amounts to roughly 79 percent of all females and 59 percent of the entire study population. While some of these amas de casa worked temporary jobs or sold domestical- ly-produced goods and agricultural products, most saw themselves as homemakers first. Tradi- tional gender-role expectations that women be primarily domestic caretakers may overshadow the fact that many rural Mexican women conduct numerous paid and unpaid economic activities to supplement household income. Many homes in the study areas also conduct small retail businesses or general stores known as abarrotes, miscelaneas, or tienditas. Sixty participants or nearly 12 percent of the sur- vey population ran a small store (shop worker) as their primary occupation. The next most com- mon occupation was “campesino” or farmer/agricultural laborer, with 33 individuals or about six percent of all participants. It is worthwhile to note that this occupation also carries a wide range of meanings, from a type of self-employed yeoman who works his own fields to a landless laborer in the agricultural sector, with most individuals falling somewhere in between. Besides living off the produce and cash-derived from their own fields, most campesinos work a variety of supplemental jobs to make ends meet. “Campesino ” and “ ama de casa” are the basic response Table 4.3. Occupation of census participant. Data, Dissertation fieldwork 2011. Table by A. Hilburn. 158 for anyone who is not a professional, tradesperson, office holder, or owner of a single sustaining business. These positions constitute honest, if not lucrative work in the study area. These three occupations, campesino, ama de casa, and shop worker, account for 77 percent of the survey population. The other 40 occupations and their frequency and relative proportions are listed in Table 4.3. The status regarding a participant’s current residency could be related to their opinion on garbage-related issues in their respective community. Someone from outside a community may view daily life and practice in their new community differently than a native. This variable applies more directly to the risk mapping exercise we conducted in conjunction with the garbage census. However, the descriptive results do show some residential mobility of the survey par- ticipants, which indirectly illustrates population patterns in the study area communities. For the entire survey population, 438 or about 84 percent resided in the town of their birth. From the 77 transplants, 22 were from other communities within the municipio, 20 hailed from the immediate Figure 4.2. Native status of participants. Data, Dissertation fieldwork 2011. Chart by A. Hilburn. 159 region of the Sierra Negra and Tehuacán Valley, two others were from farther reaches of the state of Puebla, 15 came from other states in Mexico, and 18 participants claimed too many places to define a “hometown” or chose not give a specific response. Five participants gave no response to confirm or deny their native status. The sierra community households selected had only slightly more native-born participants with 87 percent compared to 83 percent in the valley communities. Civil or marital status can influence what a particular household disposes. A single man who does not cook much or cooks basic essentials would likely dispose of more processed food, less vegetables, more soda and beer receptacles, and in general less garbage than a nuclear fam- ily of five. This census did not seek to quantify and characterize the waste stream but civil status does help contextualize the household size and economic data as well as possibly identify a rela- tion to a particular garbage disposal practice. The majority of all participants were married with 61 percent or 318 out of 520 listing themselves as married. This may seem a low estimate, espe- cially if one considers only those who claimed to be head of household, this percentage becomes 254 of 381 participants, jumping to just 66 percent of total households. Common law marriage, called union libre, helps explain the lower than presumed married frequency and accounts for 35 individuals’ civil status or just less than seven percent of the entire study population. Unmarried and divorced were grouped into one class that accounts for 146 individuals or 28 percent of the total. Six individuals with deceased spouses were counted and eleven individuals preferred not to admit their civil status. The sierra participants we sampled were more likely to be married (74 percent) or in a common law union (9 percent) than participants from Coxcatlán, Calipan, Tilapa, San Rafael, or Pueblo Nuevo that had an aggregated average of 55 percent married and 6 percent in a common law union. For the entire surveyed population, 191, or about 37 percent, spoke an indigenous lan- guage (Table 4.4). By far, the most common indigenous language surveyed was Nahuatl, an Uto-Aztecan language that is the most widely spoken indigenous language in Mexico, with 186 speakers making up 97 percent of all indigenous language speakers surveyed. Besides Nahuatl, two participants spoke Mazateco, a language from the Otomanguean linguistic family that is spoken with great frequency in the sierra south of the study area in Oaxaca. Another two partici- pants spoke Popoloca, a language from the same family that is spoken in the northern Tehuacán Valley and nearby Mixteca Baja, and one spoke Mazahua, another Otomanguean language from west-central Mexico. All of these people had married into the predominately Nahuatl-speaking 160 Table 4.4 Indigenous language speakers by language. Data, Dissertation fieldwork 2011. Table by A. Hilburn. or Spanish-speaking communities. Fittingly, the sierra community participants interviewed had a significantly higher frequency of indigenous language speakers with 146 or 87 percent of total sierra participants compared to the 40 valley settlement indigenous language speakers, represent- ing 11 percent of their subsample. The high frequency of indigenous speakers in the sierra com- munities is a lower estimate than the most recent INEGI 2010 population census that reports and average of 98 percent indigenous speakers for those same communities. In the valley settlements, the cabecera municipal had the highest percentage of indigenous speakers surveyed, with 18.5 percent speaking Nahuatl. Many of the residents from Palita, a colonia in the cabecera municipal of resettled residents from the sierra community of Pala (Chapter 2), speak Nahuatl and of the 13 households surveyed from that colonia, eight households (62 percent) spoke it. The household, the basic enumerative unit for this study, is represented by variables that describe its members as well as its physical attributes (Table 4.5). The number of people 161 who constitute a household as well as the age groups to which they belong illustrate the size and structure of a household as a group of people. The average size of a household in the entire study population consists of 4.9 (s = 2.3) members. The mode household size was just above the mean at 5 members and the largest household surveyed had 18 members living there regularly. We interviewed 31 participants who lived alone. The households in the sierra communities were typically larger with a mean of 5.7 (s = 2.5) residents than the valley community households that had a mean of 4.5 (s = 2.0) residents. This difference is perhaps attributable to the higher fertil- ity rates of the less economically developed and traditional sierra communities. Adults, defined here as anyone 18 years of age or older, comprised the majority of the members of households in the study population with a mean of 3.1 (s = 1.5) per household. The mean frequency of minors, those less than 18 years of age, was 1.8 (s = 1.6) per household. Valley community households surveyed reflect the disparity in numbers of adults and minors, have a mean of 3.2 (s = 1.5) adults and 1.4 (s = 1.2) minors. In the sierra communities, one finds more parity between adults and minors in the household with means of 3.0 (s = 1.6) and 2.7 (s = 2.0) respectively. The Table 4.5 Household size by total, adult, and minor residents. An adult is anyone over 18 years of age. Data, Dis- sertation fieldwork 2011. Table by A. Hilburn. 162 household size and structure are listed by community and age groups in Table 4.5. The building materials used in a particular house’s construction allow for an easy assess- ment of the relative affluence of a household (Table 4.6). Other amenities and features such as modern kitchen appliances, indoor plumbing, number of rooms, flooring and roof type, and do- mestic energy sources could better estimate the affluence of a household but surveying for these variables is a much more invasive process and would potentially be distracting to the survey’s larger goals to study home garbage management. In short, we wanted to understand how people manage their household waste without them being self-referential to their own relative wealth or poverty more than we desired a correlative variable estimating affluence. House construction type can be quickly and accurately assessed as construction styles are fairly easy to identify and homogeneous across the study area. The primary residential building of each participant’s home was categorized as being tall concrete, low concrete, low adobe, wood plank or cabin, river cane ( carrizo ) and thatch, or mixed concrete/adobe or adobe/carrizo/wood. Across the survey population, low concrete houses were the most common with 309 houses representing 54 percent of the sample. In the valley settlements, low concrete buildings are much more common, making up 73 percent of their subsample. Low concrete buildings were also fairly common in the sierra communities, representing nearly a third of all homes surveyed there. While variation among these low concrete structures undoubtedly exists, these houses are commonly one-story rectilinear buildings of concrete post frames fleshed out by concrete blocks made of Portland cement and locally sourced, lightweight aggregates. Most are around three meters in height. Roof type was not recorded but higher quality roofs are made of concrete slabs. More economical options are corrugated metal, corrugated fiberglass canvas, and ceramic tiles. Multi-story concrete houses, denoting greater household affluence or at least greater mate- rial equity, were also fairly prevalent with 48 surveyed representing 9 percent of the total sample (Table). Adobe houses were the second most common house type in the study population. There were 117 of these houses representing 23 percent of the total. These buildings made up the larg- est class of building in the sierra communities with 74 units or 44 percent of the subsample. In the valley settlements, adobe was present in 12 percent of the subsample. Here, adobe houses are more common in the peripheral colonias or in the agricultural settlements of Pueblo Nuevo and San Rafael. Adobe is an effective, fairly sturdy, and extremely economical building mate- 163 Table 4.6 House building type by material and form. Data, Dissertation fieldwork 2011. Table by A. Hilburn. rial. These structures are built by stacking adobe bricks, made from mining clayey soil, wetting it with water to make a mud, then mixing in straw, hay, and ground up detritus, which can include bits of rock, glass, organic refuse like corn cobs, and finally pressing it into mold for sun dry- ing. A frame is then built of timbers or adobe or concrete posts whereby bricks are then stacked within the frame’s dimensions to create walls. After stacking the bricks, a type of daub is applied to the walls in order to coalesce the bricks into a more structurally unified wall that can be left bare in more humble circumstances or frescoed over to allow for a painted, more finished appear- ance. Like concrete abodes, there is variation in the quality and value of adobe buildings. The 42 households surveyed from Colonia Centro in Coxcatlán town are large, colonial-style buildings with over half-meter-thick and 4-5 meter high walls and are some of the highest valued build- ings in the study area. Yet the majority of what constitutes an adobe structure in the study area is a rectangular building of 1-3 rooms with walls of just over two meters high and a brick or two thick. These homes typically have roofs of ceramic tiles called tejas placed over a thin wooden 164 frame. A slightly more expensive but no more effective option is corrugated metal called lamina placed over metal slats embedded in the walls. Some homes were made of other or mixed materials. These houses represent about six percent of the total survey. Mixed houses ranged from houses built with concrete walls of up to a meter and finished with adobe or baked brick. Other mixed houses were adobe walls finished with rough-cut pine planks. In cases of deep poverty or the need of a provisional house while another more permanent version is being constructed, mats of carrizo cane are propped and sewn into a frame of thin straight wooden sticks to form small one-room homes. Mixed homes and other building materials were not nearly as common as the concrete and adobe homes. Access to land is useful as an identifier for households that have a significant relationship to land-based economic activities (Table 4.7). It shows a household has an identifiable economic stake in their land and its sustainability and thus could be useful to correlate with opinions and practices regarding garbage. It also can be a complimentary measure of a household’s relative affluence. In the course of the conducting the survey, we interviewed sugarcane growers, home Table 4.7 Farming and gardening frequency by household. Data, Dissertation fieldwork 2011. Table by A. Hilburn. 165 gardeners, smallholder milpa farmers, small-scale ranchers, and people involved in some or all of the aforementioned. Thus, it was difficult and clumsy to systematically categorize and rate households by their specific agricultural activities. Also, household income estimation was not as great a priority of this research as was identifying an economic or moral connection to conserva- tion and their immediate environment. A more useful, simpler attribute is a measure that gauges how frequently a household engages land-based activities. For this, we identified households that always plant or raise animals, sometimes plant or raise animals, or never plant nor raise animals. Households that never plant or raise animals held a small majority with just less than 52 percent of the entire survey population not engaged in any directly land-based activities. This is an ex- pected outcome given the somewhat urban bias of the survey. The valley settlement stratum was markedly less engaged in land-based activities with 75 percent not engaged in planting or animal husbandry. Only a paltry three households or just below two percent of the sierra subsample did not plant or raise animals. Households that always plant made up 44 percent of the entire sample with nearly 20 percent in the valley and 95 percent in the sierra engaged in land-based activities. Land-based activities appear to be something that a household always does or not at all. Only nine households, five from the valley and four from the sierra, or less than two percent for the entire survey population, claimed to sometimes plant or raise animals. Garbage Management Practices Placing garbage outside of the home for collection by the municipal government is the primary method for household garbage management in the Tehuacán Valley portion of the mu- nicipio. For the five settlements in the valley, 335 of 353 or about 95 percent of the homes we surveyed expressed their use of the collection service as a method for disposing of garbage. Only five households, located at the edge of the peripheral colonias of the cabecera claimed to not have municipal collection service and thirteen were unsure or did not respond. Besides the no response households and the five who claimed no service in Coxcatlán town, every household surveyed in the valley utilized municipal collection (Table 4.8.). Of the households with municipal collection, 58 households purported to have service twice per week. These households were exclusive to the central portions of Coxcatlán and Cali- pan. One household each in Tilapa and San Rafael claimed to have bi-weekly service as well, but this is most likely an error or misjudgment. The overwhelming majority of valley residents, 250 166 Table 4.8. F re qu en cy o f c ol le ct io n se rv ic e by to w n. D at a, D is se rt at io n fie ld w or k 20 11 . T ab le b y A . H ilb ur n. Table 4.9. U se fr eq ue nc y of c ol le ct io n se rv ic e by to w n. D at a, D is se rt at io n fie ld w or k 20 11 . T ab le b y A . H ilb ur n. 167 or 71 percent, have garbage collection once per week. The garbage collection service dedicates one full day to collecting in Pueblo Nuevo, San Rafael, and Tilapa every week. A small number of households, 14 percent of the total, noted that the truck only comes by their house twice per month. This could be true but is more likely due to the fact that the collection routes in many peripheral colonias of the cabecera and Calipan only follow the main streets and people who live off of the route may fail to place their garbage at the designated collection points in time or at all thus bringing the regularity of service into question. By chance there were a few households located on streets that were situated on multiple days’ collection routes. It is important to note that the data only portray access to collection service. They do not indicate the frequency or use of the services offered by the municipio. For that, we asked the par- ticipant households how often they use the garbage service (Table 4.9). Most households we sur- veyed, 69 percent, place their garbage out for collection once per week. Given the considerable regularity with which the collection team operates, most people recognize the window of time on such a day that the carro passes by their residence. For all besides a few colonias in the cabecera and Calipan, garbage disposal is a once-per-week activity. However, increased service frequency does not indicate greater service usage because just 28 households, less than eight percent of the entire valley survey population, claimed to regularly place their garbage for pickup two or more times per week. A small but significant portion of households, ten percent, indicated that they only use the garbage service when they feel it is necessary or when they have “enough” garbage to dispose. These households sometimes explained that they do not typically generate much garbage or tend to compost, feed their animals with whatever organic material they have, or use some other method for ridding their homes of garbage as they saw fit. A small number of partici- pants, 17 individuals or nearly five percent of the surveyed population dispose of their household garbage twice monthly. There were 22 participants who could not identify any regularity to their disposal by collection or chose not to answer. Lastly, there were only four households, all of which were from Coxcatlán town, who left their garbage out daily for collection. To this point it appears that service once per week is an adequate provision of service where service exists in the municipio. The significant discrepancy between the number of households with biweekly service (58) and those actually using the service twice per week (28) supports this claim. Recycling waste material is embodied by two related practices in the study area. Recy- cling by one very general definition means placing recyclable materials for collection and recy- 168 cling by another party, in this case, the municipal government. This service, as mentioned earlier, collects PET plastic and cardboard which it sells to recyclers in nearby Tehuacán. Separating the plastic and cardboard out from the rest of the garbage is promoted through public announcements but separation by waste class is not yet an enforced requirement in the study area, as it is in larger Mexican metropolitan areas such as Mexico City. Most people seemed to participate out of a felt obligation to minimize the amount of inorganic garbage that ends up at the municipal dump. They also saw fit to participate in recycling out of its relative ease given the weekly doorstep col- lection. The previous definition of recycling, however, excludes efforts made by households, namely those in sierra communities, without access to municipal service but do practice some form of waste material recycling through the collection and sale of scrap metal. The sale of scrap metal is typically included in indices of recycling activity in Mexico and it warrants inclusion in this study as well. Scrap metal recycling is a less comprehensive form of recycling than what is offered by the municipio. It is also a monetized form of recycling, stretching thin its definition as a practice of garbage management. However, the almost negligible payment for metal by the fierro viejo collectors is so small as to render it worthless. This economic negligibility makes scrap recycling a believable form of recycling and management of a waste product rather than production of a commodity at the household level. These two forms of recycling, one of donat- ing plastics and cardboard and the other being the sale of scrap metal, were recorded and aggre- gated as separate variables yet both were combined to give a more inclusive view of recycling in the study population. Also, in spite of an absence of regular collection of recyclable materials in the sierra, there are occasional collections by the municipal government of recyclable materi- als in select sierra communities. For that reason, a few sierra households mentioned that they do recycle their plastics and cardboard however irregularly they may do it. Recycling is a common practice for households with regular service (Figure 4.10). In the valley, 262 or 74 percent of all households surveyed, utilize the municipal recycling service regularly. The two settlements that arguably have the greatest access to recycling, Coxcatlán and Calipan, had the lowest participation in recycling with only 72 and 59 percent recycling their plastic and cardboard, respectively. Pueblo Nuevo, San Rafael, and Tilapa all had high instances of recycling with 90, 87, and 91 percent participation respectively. The sierra community sub- sample had 21 households or just shy of 13 percent recycling plastics or cardboard. These house- 169 Table 4.1 0. R ec yc lin g, s al e of re cy cl ab le s, a nd re us e by to w n. D at a, D is se rt at io n fie ld w or k 20 11 . T ab le b y A . H ilb ur n. 170 holds recycled through the occasional collections by the municipality or through particular fierro viejo collectors who collect plastic when the material price makes it worthwhile. Across the entire survey population, close to 40 percent of households sold their scrap metal, usually steel food cans or aluminum drink cans to scrap buyers (Table 4.10). The sierra households had greater participation in scrap metal selling with close to 58 percent selling their metal compared to 31 percent in the valley subsample. Recycling becomes a far more common and ubiquitous practice if one considers a more inclusive definition of recycling that combines households that recycle their plastics or sell their scrap. In the valley, the count rises to house- holds or percent of the survey population. However, this more inclusive definition only raises the participation in the valley households from 74 percent to 80 percent. In the sierra, the only form of “recycling” is selling scrap metal and occasionally plastic, so the inclusive definition of recy- cling does not apply there. Besides the 58 percent of households selling metal, eleven percent of households sell their plastic The reuse of waste materials is part of daily life and just about every household tends to use things beyond its implied lifespan. Thus, this study measures the reporting of reuse or the re- purposing of materials rather than immeasurable instances of reuse of certain objects. The willed and systematic practice of repurposing items was the target statistic for this study during inter- views. Ostensible waste materials, like old buckets, are reused and repurposed with consider- able frequency across the study area. For the entire survey population, 71 percent of households practice some form of repurposing of items whose initially intended use had expired. The sierra households appear to be collectively more frugal than their valley counterparts with 91 percent of all households within their subsample reusing waste items regularly. Reuse was noted by 71 per- cent of all valley households with the agricultural communities of San Rafael and Pueblo Nuevo reporting the highest rates of 93 and 73 percent respectively. For the households in the sierra communities, burning, dumping, bury pits, composting, and reuse as animal feed, as well as a host of other methods were employed to remove perceived waste products from the home (Table 4.11). Even considering the long existence of municipal waste collection in the valley portion of the municipio, many households there still utilize these methods in conjunction with collection to manage their garbage. However, the regularity with which these other methods, such as burning, are used by valley households is likely consider- ably less than households that have no other option. These long-entrenched practices remain 171 Table 4.1 1. T ra di tio na l g ar ba ge m an ag em en t p ra ct ic es b y to w n. D at a, D is se rt at io n fie ld w or k 20 11 . T ab le b y A . H ilb ur n. 172 important and are the only methods for thousands of residents across the study area. They remain examples of household-centered, low-input, and largely independent forms of garbage disposal. Burning garbage in a pit or pile is the most common form of garbage management in the sierra subsample with 97 percent, or all but five households, burning some portion of their garbage. The garbage burned by households here is typically the inorganic component, such as plastic bottles, unusable plastic bags, and soiled toilet paper, leaving the organic component for dumping in the garden or pit. In spite of the almost universal availability of at least weekly gar- bage collection in the valley settlements of the municipio, 28 percent of households we surveyed burn some of their garbage. The type of garbage burned in these households is usually leaf and limb litter, called hojarasca . The almost universal practice of burning in the sierra combined with over one-fourth of valley households burning brings the total to just over 50 percent for the entire survey population. Keeping a small pit, typically about a cubic meter in size, in which to discard mostly organic waste and hojarasca is a widespread practice across the study area. For the entire study population, 21 percent of households maintained an active dump pit, called a pozo or hoyo . Si- erra households were more likely to use a pit or bury their garbage with nearly 41 percent fre- quency compared to only eleven percent of the valley households. Instead of discarding organic garbage from the kitchen into a pit, many households dump this refuse directly into their gardens, potted plants, or nearby fields to enrich the soil. Sierra households had a greater frequency of us- ing organic waste as abono or fertilizer, with nearly a third of sierra households participating as opposed to twelve percent for households in the valley below. Household animals such as chickens, turkeys, hogs, and goats can be sinks of a sort for organic waste and food garbage. Pets such as dogs and cats subsist on table scraps in less affluent households. All of these animals, pets and livestock alike, are natural scavengers and the edible offal of the households to which they belong can be a useful supplement or alternative to animal feed. Old tortillas, fruit husks, moldy bread, vegetable peels, and meat gristle are typically not wasted and make their way to these animals’ stomachs if a household keeps animals. Household animals were not recorded by this survey, and it cannot be easily assumed that the frequency of feeding animals food scraps is congruent to keeping animals as chickens and turkeys, which can subsist on insects and some feed. Also, the practice of feeding animals scraps is stigmatized due to health standards and concerns over perceived backwardness. Only four percent of the entire 173 survey population fed garbage to household animals. The frequency was higher in the sierra subsample with slightly more than seven percent, representing only a dozen households. Possible underreporting of this practice could be attributable to this stigma. Dumping is most likely an extremely underreported phenomenon in the study area. Dumping runs contrary to municipal ordinances and is looked down upon from a general moral perspective. There is no quantifiable empirical data to rigorously support a claim that dumping is a far more intensive practice, but the steady activity observed at numerous dump sites across the municipio in addition to information taken from informal interviews suggest that dumping occurs with far greater frequency than reported in this survey. Only nine households in the entire study area and only one from the sierra subgroup claimed to dump garbage in the monte outside of town, in barrancas, streets, or vacant lots. Besides the aforementioned garbage management practices, only five households, all in the sierra subsample, utilized any other methods. Four households placed their garbage in a large unburned pile on the edge of their home property to decompose and deteriorate and one kept their inorganic garbage in plastic bags to deteriorate away from the home. DISCUSSION AND CONCLUSION Participation in municipal service From the results of the survey, it is clear that most households in the valley utilize the municipal garbage collection service to some extent. The universality of service in the valley settlements is a likely the main cause of such high participation. Of households that knew of an accurate figure or chose to respond, 330 of 335 or just fewer than 99 percent reported some use of the municipal collection service. The majority of households (71 percent) have collection ser- vice at least once per week with a smaller proportion (16 percent) having collection twice week- ly. The other twelve percent claimed just more (>3 times a week) or less service (1x/2 weeks) frequency. Not assuming that everyone uses the garbage collection every time it passes by, we also asked how often participating households use the service. In terms of service use, most val- ley households (70 percent) place their garbage out for collection once per week. This is a seem- ingly adequate frequency due to the fact that the collection truck only arrives once per week for most households but of those with biweekly service, just under half (48 percent) reported to put their garbage out twice per week. About eleven percent of households put out their garbage when 174 some determined threshold had been reached. In one of my residences in Coxcatlán town, I did not put out my garbage with any regularity and mostly did it when it was convenient or when I thought the garbage in the house had become too great. Yet this was a household that did not have much pretense of being a typical home, consisting of myself and my friend of my age who ran a business in its front half and who was helped out in the general maintenance of the house by his close girlfriend. Our household lacked any observable routine and is one such model for inconsistent use of the garbage service. Only nine households or just less than three percent had no service or chose not to use it. Twelve participants or just over three percent could not or chose not to specify how frequently they used the garbage collection service. Participation in the recycling program is not as extensive as that of garbage pickup even though the recycling truck typically follows just behind the garbage truck. This participation gap is perhaps owed to the additional effort required to separate the recyclable plastic and cardboard. Nevertheless, the 74 percent of valley households that reported recycling appears high when considering other case studies across the globe. Convenience is frequently cited as a major cause in promoting recycling. Given that recycling before 2009 was almost impossible and now only requires the at-home separation of recyclables can be attributable to the relatively high participa- tion in the study area. Underneath all of recycling activity is that recycling service, like garbage collection, is free of charge. Being forced to pay for recycling, even amid strong environmental morals, would likely strike a serious blow to participation in such an economically marginalized region. The psychological reasons behind recycling behavior and the demographic factors related to it have been the subject of numerous studies across the globe for decades (White and Hyde 2012; Tang, Chen, and Luo 2011; Kurz, Linden, and Sheehy 2007; Hannsman et al. 2006; Scott 1999; Gamba and Oskamp 1994; Oskamp et al, 1991; Vining and Ebreo 1990, 1992). In Mexico, Corral-Verdugo and colleagues investigated the psychological and demographic drivers of recy- cling by households in Hermosillo, Sonora (Corral-Verdugo 1996, 1997, 2003; Corral-Verdugo and Figueredo 1999; Corral-Verdugo and Armendariz 2000), finding that both economic austerity and a general ethic of conservation were related to recycling and reuse behaviors. Households from the two largest towns in the municipio, Coxcatlán and Calipan, reported the lowest participation in recycling rates with 72 and 59 percent of households recycling, re- spectively. In Tilapa, Pueblo Nuevo, and San Rafael, recycling participation is at 91, 90, and 87 percent respectively. This lower participation is perhaps attributable to the newness and novelty 175 of garbage service let alone recycling in these towns, where before 2006 there was no municipal collection of any sort. In the cases of Coxcatlán and Calipan, the entrenched practice of leaving garbage for collection, regardless of its potential recyclability, could be a factor in lesser partici- pation. More importantly, in Pueblo Nuevo and especially San Rafael, the high sanitation and public cleanliness standards enforced through customary rules (usos y costumbres ) there may engender a sort of normative environmentalism practiced through recycling. Moral obligations have been seen as drivers in increased participation in environmentalist causes (Joiremann et al. 1997; Bamberg and Möser 2007). In the valley settlements, 75 percent of households separate their garbage. A closely related 74 percent of valley households reported that they participate in the recycling program to some extent, indicating the congruence of separation to recycling. While our statistics on garbage separation do not specifically equate recycling with separation, as separation can indicate any garbage partitioning (organic from the rest), the congruity of results from the valley settlements indicate that separation does tend to signify recycling. The selling of recyclable materials, primarily metal, to private recyclers was also a fairly notable practice for valley households. Just under a third, 31 percent precisely, of households claimed to sell metal to the fierro viejo collectors. This phenomenon may even exhibit under- reporting due to the perception of selling scrap for change as fitting for the lower social classes (Castillo-Berthier 2003; Medina 2005). Also, residents noted on a number of occasions that the almost negligible two pesos per kilogram (14-18 cents US) prices paid by fierro viejo collectors in 2011 was not incentive enough to recycle the little metal they had. From a household per- spective, intent of selling garbage for recycling differs from donating recyclables in exchange for their collection, but in the end both are recycling practices. If one includes the selling of recyclable materials such as metal to the definition of recycling, then the percent of households recycling jumps to 80 percent in the valley. In the sierra settlements, the sale of recyclables is much more common, with 57 percent of households selling metal and some plastic for the low prices offered by the fierro viejo collectors. The higher economic marginality in the sierra likely explains this greater percentage but the lack of access to outside forms of garbage collection may also be a draw for households simply wanting to get rid of their metal and plastic without having to burn it, store it, or deal with it otherwise. The reuse or repurposing of materials, while not attached to any service provided by the 176 municipality or private enterprise, is conceptually related to recycling through the conserva- tion rubric “reduce, reuse, recycle” and bears mentioning here. Reuse by the valley households was reported 71 percent of the time, whereas the sierra community subgroup showed 91 percent of households systematically repurposing items. The general lack of availability of goods and greater material poverty of the sierra communities and the smaller valley settlements is perhaps manifest in the data on reuse. The households from Coxcatlán and Calipan, the two largest and most economically developed towns in the municipio, have the lowest percentages of reuse with 62 and 52 percent respectively. A quick comparison between the relative material wealth listed in the 2010 INEGI census further suggests a link, however presumable, between material wealth, access to it, and reutilization of materials (Table 2.5). Traditional practices Almost exactly half of the households surveyed in this study (50.2 percent) burn some portion or all of their garbage. Yet given that almost all (97 percent) households in the sierra subsample burn their garbage, further contextualization of this figure is required. In spite of the near universality of garbage collection in the valley, 99 households or 28 percent of the valley subgroup, still burn some portion of their garbage. How and where this burning occurs also var- ies. Some of this reporting entails the burning of hojarasca or plant and leaf litter from house lot gardens. The garbage collection service does pick up limited amounts of limbs, dead cacti, and unwanted brush but this matter is generally viewed by area residents as matter that can and should be managed on site1. Additionally, the general lack of lawn bags and the sheer size of the yard waste make burning a convenient option for maintaining the order of the house lot garden. Besides yard waste, garbage is burned by valley residents out of convenience or when they wish to deal with it quickly. The usually ample size of house lots makes burning garbage or yard waste an easy, immediate, and unobtrusive management solution. Sutro (1990) observed in a peasant community in the Oaxaca Valley some relation with increasing house lot size and outside dump- ing, alluding to the causality of needing enough space to dump so as to not offend neighbors with the bad smell of decomposing trash. The proximity to neighbors and lot size effects may be applied towards burning as well. Calipan, the settlement with the second lowest reported burning frequency (17 percent), has noticeably smaller house lots than those from Coxcatlán, Tilapa, and Pueblo Nuevo which report higher household burning rates of 30, 39, and 43 percent respective- 177 ly. Additionally, within the cabecera, more households burned (36 percent) in the peripheral co- lonias where lots tend to be larger than those from the smaller, more developed urban lots in the centro or “downtown” of Coxcatlán town. There also may be some relation to how developed or built-up a lot is to the practice of burning. While only relying on anecdotal evidence, I observed or rather smelled almost no burning in the centro of Coxcatlán. Complaints from neighbors, the abundance of collection service, and the lack of a toft zone in the more concretized and built up lots in the town’s center are the likely reasons for the absence of burning there. In the less for- mally developed colonias, where there is more house lot than house, more dirt than concrete, less governance, and a more DIY approach to household maintenance, burning is a still a viable solu- tion to garbage management. Outside of the house lot, in streets and in adjacent creek beds, I observed on numerous occasions the burning of household garbage in piles. In some of the farthest reaches of Coxcatlán town there are large one square meter wide pits along the dirt roads where some households burn their garbage so as to avoid contaminating their house lots. Some of these piles were what could be called sweepings from the house or yard but some were basically the contents of the costales Figure 4.3 . Burn pile in a barranca. Photo by A. Hilburn, September 2011. 178 that people used to store garbage prior to its burning. In a few instances I witnessed large piles of garbage being burned in unoccupied areas such as abandoned lots and the stretch of Barranca Soyolapa that runs through Coxcatlán town. However this practice is more akin to dumping as the burning appears to have been intended to mitigate the blighting effect of the dumped refuse. In the sierra communities, burning garbage is the most frequently cited method of gar- bage disposal in the absence of municipal collection. However, where burning is largely seen as a negative and backwards, not to mention informally illegal, practice in the valley, feelings are more mixed with regard to it in the sierra. The fact that burning plastic and other inorganic waste produces a bad stench, contaminates the air and soil, is a fire hazard, and is seen as culturally backwards is not at all lost on the sierra community households. Yet burning offers sierra house- holds a way to establish order in their private space and maintain a home free of garbage. Many participants informed me that formal home economics instruction given by agents with the social welfare program Oportunidades promotes burning in controlled, sand or brick-lined pits away from water sources and high use areas. This controlled form of burning is a willed systematic ap- proach to garbage management in the sierra. Not every household in the sierra kept burn pits and a more incidental approach was taken where many households placed mostly inorganic garbage in piles until they were deemed too large and were burned. Plastic bottles, toilet paper, diapers and anything perceived as useless and flammable comprise the typical materials burned in sierra households. Yard waste, brush, and any vegetative matter that could not serve as fuel or just left to rot were also burned. In short, things here that get burned have lost all of their use or potential reuse value. Over the course of conducting the surveys, this aspect of garbage as truly useless matter became to be defined much more narrowly in the sierra communities as well as less urban households in the valley. In many cases my research assistant and I figured out that many house- holds did not consider organic waste such as food scraps, old tortillas, and vegetable peels to be garbage or tlazolli . These things had a use, whether it was for food for animals, garden fertilizer, or were so biodegradable so as to not be worth the trouble, unlike real garbage like diapers or plastic bottles. Pits remain another garbage management option for a significant percentage of area households. Across the entire study population, 21 percent of households kept a small pit in which they deposited garbage. Pits were a more common feature in sierra households with 41 percent of household compared to the valley subsample of 11 percent. This was typically organic 179 waste from the kitchen but some threw inorganic garbage in the pits to be covered with soil. Af- ter a pit becomes too weathered, too full, or malodorous, it is filled in and new one is dug. Some pits were used to burn things but most were just holes where the garbage was intended to disap- pear from sight and possibly enrich the soil in and around it. In one of my residences in Coxcat- lán town, the organic waste from the kitchen, things like fruit peels, inedible parts of vegetables, old tortillas, egg shells, and chicken bones were collected in a pail in the kitchen and then were dumped in a small hole in the garden. Sometimes the decomposed contents of the pit were used as abono or compost for the potted plants or other yard plantings. Related to but not exclusive to pits is the practice of using organic household garbage, usually things from the kitchen, to fertilize house lot gardens and nearby agricultural plots. This practice was recorded separately to ascertain the household use of organic garbage in total and not apart from keeping a pit. There exists some statistical overlap with pit usage but many participants explicitly noted the practice of throwing kitchen garbage directly onto their house lot gardens or nearby agricultural fields. Sutro (1990) noted the common practice of discarding kitchen waste in unoccupied areas of house lots in his intensive study on household economics in the Oaxaca Valley. Like pit use, the sierra households (33 percent) were far more likely to employ organic waste as fertilizer in their home gardens than were households in the valley settlements (12 percent). Feeding kitchen waste to domestic livestock and pets was far less common than previous- ly mentioned practices, with only four percent of households noting it during the survey. Again, there was a greater percentage of households feeding old tortillas, vegetable matter, and bones to their pets and animals in the sierra communities (7 percent) than in the valley towns (2 percent). All of these figures are likely underreported due to the negative stigma attached to feeding gar- bage to animals, both from the shame of poverty and the common knowledge of the health risks of garbage-fed livestock. Dumping is definitely underreported in this census. Dumping is not as common as in previous generations before formal garbage management and governance existed, but I assume its virtual absence in the data denotes a mostly inaccurate variable. Informally, some residents noted that they or relatives would store garbage in an unoccupied area of the house lot compound and then carry it to known tiraderos (dumps) in the monte west of town to dump or along the roadside into the sierra. One man I met would take some of his household garbage his family’s recreational parcela on the outskirts of town for provisional dumping. From an outside perspec- 180 tive, this dumping appeared inexplicable given the once to twice per week collection service but these same residents cited the convenience of taking care of garbage when they wanted to and some general dissatisfaction with the garbage collection service. One participant noted that he personally carries his garbage to the municipal dump south of town. Spatialities of Garbage Management Tables 4.8 and 4.11 point to two markedly different garbage trajectories in Coxcatlán. In the valley, garbage is managed largely by the state and in the sierra it is managed mostly by individual households. These two systems produce different spatialities or more simply, spatial models with regard to the movement and final disposition of garbage. The two spatial models of garbage management each engage space and land resources differently in terms of the distribu- tion and flows of garbage as well as land use intensity and spatial extensiveness. The overwhelming dominance of collection as a primary garbage management practice in the valley produces a land-intensive system that funnels garbage from a network of households through the municipio’s sanitation infrastructure and into a concentrated dump area. This system Figure 4.4 . Medium-sized tiradero clandestino. Photo by A. Hilburn, April 2011 181 is how I managed my garbage there and continue to use every week I put my garbage out by the curb in the US. The garbage collection model entails a directionality of waste away from its sites of generation outward to a focused end. Dumps become garbage sinks for a defined “gar- Figure 4.5 . Two garbage management models. Chart by A. Hilburn. 182 bageshed” of sorts. From an intentional perspective, this system is a form of collective dumping. The intent of dumping is to rid one’s own space of unwanted matter and garbage collection even goes a step further by eliminating the individual labor required to go dump it. Garbage collection concentrates the labor inputs involved in the final disposition of garbage and relies heavily on fuel and infrastructure to exist. The garbage collection system is a spatial solution for households and municipios alike1. By that, this management system allows households to avoid the crisis of having to live with, burn, actively use, actively contaminate, or otherwise deal with unwanted matter by removing it to a far-off (relatively speaking) site. Garbage collection provides a solution for contemporary consumerism by offering a spatial solution to the crisis created by the inextricable link between consumption and inevitable waste production. In a practical sense, households in the valley no longer have to deal with the extra effort to manage garbage. They no longer have the stench of burning plastic or their kids do not run the risk of getting burned by smoldering piles in the yard. Parts of their yards do not have to be used, however passively, for garbage management. Stored garbage does not visually contaminate their yards and these households can reflect on having the trappings of an amenity enjoyed by residents of more economically developed urban areas of Mexico. Garbage collection is a nice amenity and is something most households in Coxcatlán want. However, garbage collection is not without its potential drawbacks. First there is the issue of where the garbage goes. The municipal dumpsite is the material spatial answer of garbage collection. The municipal dump is sacrificed space that has been trans- ferred from the individual house lot to a place ostensibly far enough away from those served. The dump becomes a form of spatial garbage in that has no value beyond a depository for valueless matter and is ecologically ruined in the process. Dumps are not new phenomena to urban popu- lations in the study area. Most towns prior to the collection service had dumpsites on the edge of town where people could take dangerous or large debris to dump. However, the magnitude to which these sites were used and the reliance of the population on these sites was not likely as great. Following Sutro (1991) and Hayden and Cannon (1983), who acknowledge the existence of dumping in Middle American peasant villages prior to the implementation of collection ser- vice, dumping then was much more sporadic, spatially diffuse, and small scale. Today, even with the small collection infrastructure and comparably low garbage generation rates of the service 183 area, most of this garbage is being focused into a 2.5 hectare site. The municipal dumpsite is also a largely uncontrolled site, meaning there is no geosynthetic liner to prevent leachate seepage into the groundwater and the piles of deposited garbage are not compacted and frequently capped with earth. Currently, the availability of undeveloped land creates a situation where no residences are very close (less than a kilometer) to the dumpsite. Coxcatlán’s low population density and settlement locations do not have the environmental justice issues of economically and socially marginalized populations living near dumps as is the case in larger cities in Mexico (Castillo- Berthier 1983, 2003; Medina 2005; Moore 2008; Moore 2009) or the United States (Bullard 2000; Cole and Foster 2000; Pellow 2004) but as Coxcatlán’s colonias expand southward toward the dump, proximity effects from living with garbage could become an issue. In terms of control over an environmental practice, submitting to garbage collection forfeits a sense of autonomy and control over garbage management from the household and community level to the municipio. With garbage collection, burning and other more traditional practices of garbage management can become morally taboo in the softest cases and illegal in the strictest form. The garbage truck and sanitation crew become the normative form of garbage Figure 4.6 . Smoldering garbage piles at the Coxcatlán municipal dump. Photo by A. Hilburn. Date unknown. 184 management and using the garbage service creates a type of dependency on it. Table 4.9 high- lights the ubiquity with which garbage collection service is used and Table 4.11 points to the reliance on collection and the related diminishment of other forms of garbage management. Con- versely, for the municipio, the inclusion of more communities, neighborhoods, and households is an added burden on the limited infrastructure to provide service. Currently, the one to two collec- tor trucks, the recycling truck, and their crews appear to be near or at maximum capacity. Garbage collection is also a capital-intensive system and there is the issue of paying for it. The garbage collection service in Coxcatlán in 2011 was funded by a general municipal fund comprised mostly of federal and state money solicited by the municipio as well as revenue from municipal fees from other services such as water service provision, documents and licenses, and market taxes. Through Article 115 of the Mexican Constitution and the Article 10 of the LGP- GIR, the municipio is directed to provide garbage service to the best of its abilities. While there is currently no fee for the garbage collection service, Article 115, following the 1983 neoliberal reforms, also allows the municipio to levy fees for a number of municipal services, includ- ing sanitation. Section 10 of Article 10 of the LGPGIR, specifically reaffirms the ability of the municipio to receive payment for garbage services. Sections 4 and 5 of Article 10 also empower municipios to grant concessions to private and semi-private garbage collectors, recyclers, or dumps, which would obviously require payment by the municipio or from residents directly. If a fee were placed on the service, more economically precarious households would have yet anoth- er fee to account for as well as have less management options, both morally and legally, to deal with their garbage. For the municipio, the duties to collect and account for the fees paid would add an additional administrative burden. The fee administration could also invite corruption or at least bad politics from a mistrusting public. The small size of the service shed allows the two truck crew to conduct their duties where they are expected. Coxcatlán, Calipan, Tilapa, and its two satellite communities are not large settlements and appear to be within the operational means and scheduled expectations of their residents. It is a small operation for now. Population growth and increasing generation as well as the intent to expand the service area would likely force the municipio to exceed the capacity of the budget to provide the service free of charge or give the idea of revenue generation a political backing. More urbanized municipios like San Sebastian Zinacatepec and Ajalpan as well as the city of Tehuacán farther to the north all have per bag fees. Serving the sierra communities would 185 require more vehicles, a substantially greater fuel budget, and more ample sanitation crews as well as perhaps the improvement of road infrastructure to allow them to more easily and safely access the communities. The sierra communities do not have as an effective spatial solution to their garbage as do communities in the valley. There are community dumps as well as a number of tiraderos clan- destinos and littered hill slopes in some of these communities, but the spatial segregation and isolation between the community and these dumps is not as great as that found in the collection system. Here, much of the garbage management occurs within and around the house lot. As such, households and the areas around them become sinks or traps of garbage rather than the mere sites of garbage generation as is found in the collection model. Consumables flow in, are used up, then are managed either by repurposing, decomposition, burning, storage, and burial. Regardless of the management practice, the waste or whatever into which it was converted stays. Even the materials that are dumped by residents do not leave the community or neighborhood. Instead, the house lots absorb garbage and the renderings of the ways to deal with it. A substantial portion of the waste in households that utilize this waste management model is organic and much of this matter is reinvested into the garden soil by broadcasting it into gardens, where it decomposes. Some is eaten by domestic and wild animals. Intact glass containers are reused but broken or sur- plus glass poses a greater management problem. These are typically stored in piles or in contain- ers in some out of the way place. Paper and plastic are burned in pits and piles. In a large-scale sense, the house lot model is a land extensive system of garbage man- agement. As house lots are individual sinks of garbage, their distribution pattern implies a more diffuse and less intensive form of garbage management from a collective whole. Materials flow inward to these sites, they are consumed, and are converted into other forms by burning, rotting, physical breakdown, or burying. Compared to the concentration of garbage at the dump in the previous waste management model, garbage management is diffused over the entire area of a community. Even dumping outside the home does not remove the waste entirely out of a resi- dent’s lived space to the same degree as a municipal dump does. However in another sense, one more grounded in the individual house lot, this model implies a land intensive garbage manage- ment system. This is especially true when considering lived space. The lack of collection forces residents to live with their garbage and dedicate portions of their own home properties to storing, burying, or burning it. The spatial distancing of garbage and the home becomes much more con- 186 tracted. Burning garbage is a material conversion and not a spatial distancing in that unwanted materials like plastic bottles and diapers become something less conspicuous. The home site in the house lot model is a vertically-integrated site of consumption, waste production, and waste management. When considering the organic garbage a household produces, these house lots become importers of nutrients which can be reemployed into house lot garden area production. While the magnitude of a household’s reinvestment of soil nutrients, mainly phosphorus, through the applied and passive use of organic waste, from kitchen garbage to hojarasca , is difficult to determine, recent archaeological and geographical research on an- thropogenic soils point to the significance of nutrient trapping on home and public space in New World antiquity. In regions of perceived nutrient scarcity, such as the Amazon Basin as well as many other regions with poor soil fertility across the New World, the trapping of phosphorus and carbon underwrote subsistence. Today these carbon and phosphorus-rich soils are useful proxies for indicating ancient settlement as well as point to more sustainable agricultural uses in highly trophic or eroded environments (Lehmann et al. 2010; Woods et al. 2008; Glaser and Woods 2004). In many house lot gardens and infields in Coxcatlán, the influx of soil nutrients from household garbage, in addition to other inputs, may contribute to their productive maintenance. On the other hand, the house-lot model implies the trapping of inorganic and less-than- organic waste (paper, cardboard) on the home site. From antiquity to the past few decades, inorganic garbage was mostly ceramic, lithic debitage, rubble, and later glass and some metal. These materials provided some injury risk from puncture, cuts, or scrapes. They still do today. The advent of plastic and plasticized packaging makes households in the house lot model sinks of these items and their burned and decomposed chemical components. In many homes where no collection exists and a few that do have service, plastic bottles, diapers, and toilet paper are burned in piles and in pits. The residues from this burning can contaminate soil with toxic sub- stances such as PCBs (polychlorinated biphenyls) and PCDD/Fs (polychlorinated dibenzo-p- dioxins and polychlorinated dibenzofurans) (Solorzano-Ochoa et al. 2012; Lemieux et al. 2012). Besides providing a contact risk with toxic substances, the soil contaminants can be ingested through produce grown in on-site gardens (Jackson 2011). Burning garbage also releases a slew of toxic substances into the atmosphere, potentially affecting the respiratory health of anyone liv- ing in and around where the burning occurs (Sidhu et al. 2005; Christian et al. 2010). Addition- ally, lower birth weights have been correlated with the practice of burning garbage (Amega et al. 187 2012) The common practice of burning garbage is not without human burn and property damage risks (Forjuoh 2005). Conversely, not burning or ridding the household of garbage poses poten- tial biological health risks from decay and vermin that feed or reside around garbage (Wokekoro and Inyang 2007). The contamination risk from garbage self-management turns homes into small dumpsites, bringing environmental justice concerns that had previously been more focused on people living on or near formal dumpsites (Pellow 2004; Moore 2008; Higgs and Langford 2009; Hartmann 2012) to the rural house lot. Conclusion This chapter describes the daily practical management of garbage by households in Cox- catlán municipio, giving mention to the other actors and practices involved in garbage manage- ment in there. It noted that the five main settlements in the valley portion of the municipio and 69 percent of its population have garbage collection. The remaining 31 percent of its population spread out among fifteen communities in the sierra do not have formal municipal collection and must contend with other practices to manage their garbage. The result of this bifurcated service provision are two waste management models. These two models, the “collection” and the “house lot” models, represent of the two different management regimes in the municipio. For the households outside of the collection service area, this survey provides empirical evidence of the practices used by households to manage a diverse array of waste on their house lots and vicinity. Results from this study have shown that there is a more hybrid and diverse ap- proach to garbage management in the municipio than what is dictated by official policy. Burning garbage is the predominant practice in sierra households, followed by pit disposal, direct disposal of organics to gardens, and use as animal feed. Dumping is likely underreported by this survey. Even when service is present, almost a third of valley households burn some component of their garbage. A small but still significant portion of valley households still separate the organic com- ponents of their garbage for the benefit their gardens and house lot soil. The previous chapter (Chapter 3) outlined the legal structure of garbage management, showing how garbage is managed in a situation fitting with the goals of garbage manage- ment policy in Mexico at the federal, state, and municipal levels. However, garbage manage- ment policy often says very little about how garbage is managed by households (Bulkeley and Askins 2009; Bulkeley and Gregson 2009). Most garbage is generated by households and since 188 this material must be first handled and dealt with by them, the household is perhaps the most significant actor in garbage management, especially where universal collection service does not exist. Yet the household is one of the least visible scales of garbage management in typi- cal garbage management policy. Households are often subsumed within the official levels of governance to which they are subject. The differential access to official garbage management infrastructure such as collection service and the functional role of traditional practices like organics disposal in gardens and burning complicate the simplistic assumptions that garbage is just collected, disposed of, or recycled at best or “poorly managed” at worst. By focusing on management practice at the household-level, presumably unmanaged garbage in underserved areas or those with no access to officially sanctioned management is in fact managed by a number of practices by households. The case of unserved communities and informal recyclers ( fierro viejo) show that additional actors outside of policy prescribed ones govern the manage- ment of garbage in Coxcatlán. Additionally, a household-centered practice-based approach brings to light the inadequacy of assuming that all garbage that is generated within a policy- congruent management regime, is collected, repurposed, or recycled when some portion is also being burned, dumped, composted, or buried. The diversity of garbage management practices that is less-than-congruent with prescribed garbage management policy suggests that real, lived garbage management is also contingent on individual perceptions, opinions, and traditional structures. The following chapter addresses public perception on garbage and its management. Endnotes 1. Through a concept developed by David Harvey (Harvey 1982, 1985, 2002) to describe the dependence or “fix” of global capitalism on the reconfiguring or “fixing” of space to avoid crises in production and consumption, Sar- ah Moore (2011: 137-138) applied the spatial fix to flows of hazardous waste from global north to global south. A slightly different take on the same idea could go further to explain the intent behind the garbage collection system in Coxcatlán but in such a local case study, this theoretical concept does not fully apply. Thus, the more general terms “spatial solution” and “spatial distancing” are used en lieu of “spatial fix” to describe the practices of garbage management used to distance the self and home from unwanted matter. 189 CHAPTER 5: Q-Methodology, subjectivity, and household garbage in Coxcatlán INTRODUCTION The increasing generation of garbage and its changing character are major environmental issues in Mexico and across the globe. Additionally, how this garbage is disposed of, converted, reused, or recycled, in essence, how it is managed is a similarly large issue for governments and everyday citizens (Moore 2011). The previous two chapters have outlined the issues of garbage generation, garbage governance, and garbage management from a material and practice-based perspective. These chapters identified the trajectory of garbage management from consumption to generation to final disposal as well as shown that Coxcatlán municipio is the site of two dif- ferent but not mutually exclusive garbage management models, one based in the household and house lot and the other constitutes a spatial distancing provided by the government away from the home. By that, they have outlined the more matter-of-fact side of garbage and how it is dealt with in Coxcatlán, outlining how a certain amount of garbage of x different types is managed by y practices that are underwritten or not by z sets of obligations and laws enacted through different scales of governance. Garbage, in this sense, is defined by a common, third-party consensus. Yet, how do individuals and groups perceive what this material conflation called garbage is and how do they believe it should be managed? Moore (2012: 780) quotes Myers (2005: x), “After all, what is more material than gar- bage?” to illustrate how garbage is not just a straightforward material and practical issue but is contingent on numerous conditions and factors. Garbage is perceptional, discursive, and subjec - tive. The cliché that “one person’s trash is another’s treasure” references these subjective and conditional properties. This argument about the subjectivity of waste and non-waste can also be extended to how one should deal with that waste. The different subjectivities on garbage and its management can coexist as well as conflict and compete, making them political (Moore 2008, 2009; Hartmann 2012). Unpacking the subjectivities on garbage as a human-environmental issue add to how individual and consensus group subjectivities construct garbage socially and environ- mentally and also how they inform the daily practices of garbage management. Q-method, a research methodology spanning the qualitative-quantitative divide, is in- creasingly being used in human-environmental research to gauge subjectivities on a number of environmental issues ranging from wind energy (Brannstrom, Jepson, and Persons 2011; Jepson, 190 Brannstrom, and Persons 2012) to parks and NIMBYism1 (Hawthorne, Krygier, and Kwan 2008) among others (Danielson 2007; Ward 2013). The following chapter seeks to describe the sub- jectivities of two major garbage-related issues in Coxcatlán: household garbage generation and household garbage management practices, by employing Q-method. With regard to the entirety of the dissertation, this chapter provides a glimpse on how people perceive the material (genera - tion of various garbage types) and practical aspects (how people should manage these types of garbage) of garbage. METHODS AND STUDY POPULATION What is Q-Method? This chapter seeks to investigate the subjectivity of a sample of Coxcatlán’s population regarding two garbage-related themes. It employs Q-method, a quantitative approach to system- atically measure human subjectivity. Q-method was first developed by William Stephenson, a British physicist turned psychologist, in the mid-1930s (Stephenson 1935; Addams and Proops 2000). Stephenson sought to measure performed or practiced subjectivity, which he referred to as “operant”. Q, as it is often referred to as, seeks to identify common perspectives, viewpoints, or discourses and then statistically measure how individuals relate to those views. Q illustrates how individuals load on those factored discourses or viewpoints, using a term from principal components analysis to which Q is statistically if not conceptually related. In terms of how data are analyzed, Furthermore, Q has often been portrayed as an inverted form of R factor analysis in that instead of correlating different variables’ loadings on factors from n individuals, Q corre- lates different individuals’ loadings from n tests. In short, the individual participants’ represented subjectivities become the variables that define the correlated factors. Functionally, these factors represent the different subjectivities regarding a particular issue. The fields of psychology (Brown 1980) and political science (McKeown and Thomas 1988) have long used Q-method; however, it was not until the mid-to-late 1990s and early 2000s that human geographers began to employ Q (Robbins 1996; Robbins and Krueger 2000; Eden et al. 2005). Its use has steadily increased since then, with Brannstrom (2011:535) claiming that “at least two dozen papers published [o]n geographic or environmental topics” have been published since 2007. While this claim may appear excessive due to an overly inclusive assessment of what constitutes geographic and environmental applications of Q-method, this approach’s popularity 191 is growing among human geographers who seek quantifiable evidence of subjectivity that can be qualified by an interview (Brannstrom 2011:533). In light of research regarding the social and discursive construction of the environment and environmental issues, Q-method can be a useful tool to describe and situate environmental perspectives and discourses among and within groups of people (Eden et al. 2005) and has been applied towards forestry policy (Dasgupta 2005; Swaffield and Fairweather 2000; Steelman and Maguire 2000), global climate change (Van Eeten 2000), sustainability (Hooker-Clarke 2002; Barry and Proops 1999), tourism (Green 2005; Hawthorne et. al. 2008) and specifically for political ecology-oriented research (Robbins 1996; Robbins and Krueger 2000; Brannstrom 2011; Brannstrom, Jepson, and Persons 2011; Jepson, Brannstrom, and Persons 2012; Ward 2013). Q-method also allows for what Robbins and Krueger (2000: 636) call an “anti-essentialist” approach to gauging subjectivity and perception, in that the subjectivities are constructed by the participants themselves. While the researcher’s own goals are inexorable in the design and analysis of a Q study, the method permits the respon- dent to construct responses in congruence with their own viewpoints (Eden et al. 2005) rather than merely responding to a researcher’s prescribed construction of the researched phenomenon. Doing Q-method research in Coxcatlán This research project investigates two themes regarding the subjectivity of garbage: a) the perceived household generation of garbage and b) the management of that garbage. Understand - ing how people perceive what the contents of their garbage are allows for the identification of different groups of perceived garbage generators. Additionally, comparing the differing subjec- tivities on what garbage respondents think they generate against existing data on garbage genera- tion of Coxcatlán’s cabecera (Sánchez-Mendez and Olmos-Torres 2009) allows for an interesting framing of the perceived garbage issues within an observed material reality. The second theme regarding garbage management practices looks to identifies different stakeholder groups who show consensus over how particular garbage classes should be managed. The results from this theme can better contextualize some of the garbage census data on why certain practices persist amid presumably adequate collection service. The fieldwork for the photo-based Q-method study was conducted from March – May 2011 in the cabecera municipal of Coxcatlán municipio. Doing Q-method as a researcher is a highly interactive process. Practically, Q-method is based on three main steps. First, the re - 192 searcher selects the representative statements or photos, called a “Q-sample” from a larger field of issues, called the “communication concourse” that encompasses what the researcher believes include all subjective aspects of a particular issue. The often-painstaking process by which the concourse is defined varies according to the issue, with samples usually taken from interviews supplemented by published sources in what McKeown and Thomas (1988: 26) call a “quasi- naturalistic” approach. Similarly, the drawing of representative statements or photos from the concourse to form the Q-sample is also both important and difficult. For this research, I took a different approach and defined the concourse as all garbage classes. I took staged photographs of different garbage classes defined by the state government’s environmental ministry (SMRN then SSAyOT) to construct the Q-sample. Garbage classes are not as apparent as statements in terms of their discourse but still maintain their own subjective properties. Certain items and materials are recycled, reused, isolated, burned, left to rot, and pushed aside based on how people see their use, lack of use, impedance potential, or disgust. I constructed this study’s Q-samples from a concourse of slightly adjusted sets of predefined gar- bage classes to fit to local generalizations of what constitutes garbage. Garbage is material and does not draw from recognizable discourses or perspectives (environmentalist, capitalist, Repub - lican, Democrat, etc.) as a Q-sample using statements, however garbage classes can be organized by their properties such as “organic/inorganic”, “natural/industrial”, “harmful/beneficial”, or “waste/recyclable”. The fundamental properties of the assemblages of garbage classes that repre - sent the constructed factors allude to different subjectivities of participants on material garbage. Photographs were used to represent the phenomena (garbage) that I proposed to the study participants. A photograph of a class of garbage, such as metal, more effectively represents the subject than its written signifier. Also, given what I assumed to be lower literacy rates of the study area as well as my own fluent, but gringo-inflected Spanish, photographs appeared more appropriate instead of basic text. However, to minimize ambiguity in understanding what the photographs represented, I superimposed in a non-intrusive font the written class name in the center of the image. Photographs have been employed in Q-method research successfully and in some cases, like mine, more appropriately represent the sought after communication concourse (Hawthorne et. al. 2008; Green 2005; Fairweather and Swaffield 2000, 1995; Swaffield and Fair- weather 1996). The selected photos were printed out on 3-inch by 5-inch cards with a 16-class q-sample 193 Figure 5.1. One half of the 16-class Q-sample. Graphic by A. Hilburn. 194 Figure 5.2. The other half of the 16-class Q-sample. Graphic by A. Hilburn. 195 Figure 5.3. 15 garbage classes of the 27-class Q-sample. Graphic by A. Hilburn. 196 Figure 5.4. 12 garbage classes of the 27-class Q-sample. Graphic by A. Hilburn. (Figures 5.1 and 5.2) and a 27-class q-sample (Figures 5.3 and Figure 5.4). After the Q-sample was selected, the respondents were chosen for sorting and interviews. Compared to more tradi- tional R statistical surveying, Q-method does not require a significantly large or random popula- tion sample. Purposive sampling that pursues an expansive variety of respondents is preferred. Practically, this amounts to anywhere from 20 to 50 but can be as high as 100 or as low as three 197 Figure 5.5. Q-Study participants by colonia. Data and graphic by A. Hilburn. or four, depending on the level of intensiveness that the researcher seeks (McKeown and Thomas 1988: 36-37). For this study, 70 respondents were selected in a quasi-purposive yet opportunistic sam - pling procedure by targeting equal numbers of women and men from different economic and cultural backgrounds across the town’s fifteen colonias (Figure 5.5). I surveyed 36 women and 34 men with 37 being from Colonia Centro, 28 from other colonias, and five from outside the community. Administering the two2 Q-sorts generally required about 25-30 minutes of a respon - dent’s time. The use of pictures likely sped up this process. Due to the significant amount of time needed for each Q-sorting session and considering that my personal contacts were strongest in the cabecera where I lived, I decided to limit the study area to just there. Also related to time constraints, I made the decision to not conduct a laborious post-sort interview for each Q-state- ment. Instead I would ask a participant why they constructed their Q-sorts the way they did and then jot down notes on their main points. I also had a short questionnaire to obtain basic demo- graphic data on age, sex, colonia of residence, educational attainment, and household garbage 198 management practices. Intensive interviews can be helpful in explaining the meanings of factors and to highlight statistically hidden significance. Yet given the matter-of-factness of the Q- sample (photographs of garbage classes) and the effective time constraints, the questionnaire and short interview proved sufficient to elaborate the meanings of the factors. For the second step, a chosen respondent is instructed to sort and rank-order the Q-sam- ple items regarding an issue along a forced, quasi-normalized distribution of that best represents their opinion regarding a proposed statement or photo, thus allowing the comparison of opinion patterns within the sampled group (McKeown and Thomas 1988). The photo cards were arranged by participants on one of two 42” x 36” professionally printed lona or canvas guides with the instructions and distributional scale (Figures 5.6 and 5.7). Following their ranking of photos, the subjects also provide a detailed explanation of how they constructed this response, called a “Q- sort.” The two different Q-sorts had the following conditions of instruction: 1) This is a representation of the garbage produced in my home ; and 2) This represents how each of these waste classes should be managed The first sort, where participants were asked to represent the garbage produced in their home, used a reduced Q-sample of 16 waste classes and the distribution ranged from -3 (less rep - resentative) to +3 (most representative) (Figure 5.8). This smaller Q-sample accommodates the lack of precise consideration by most people, myself included, of what they actually discard. The latter two used a more comprehensive Q-sample of 27 garbage classes. The second Q-sort, which asked how particular garbage classes should be managed, had a distributional scale that ranged from -5 to +5 and required respondents to rank garbage classes according to their appropriateness of managing them in the home (-5; dumping, burning) to leaving them for the government to manage (+5, landfilling, recycling) (Figure 5.9). The third step in Q-analysis, involves the data analysis. Q first depends on correlation, factor analysis, and calculating the factor scores (McKeown and Thomas 1988: 46). See Brown 1980 or McKeown and Thomas 1988 for a more detailed description of the statistical procedures and their theoretical underpinnings. After each Q-sort was constructed by the participant, photos were taken of their Q-sorts. These were coded and entered in a database along with basic de- mographic data. The Q-sort data were entered into PQMethod, a freeware DOS-based program 199 Figure 5.6. Example of a performed household generation Q-sort. Photo by A. Hilburn, April 2011 Figure 5.7. Example of a performed garbage management sort. Photo by A. Hilburn., April 2011. 200 Figure 5.8. Forced-normal distribution for the household generation q-sort. Graphic by A. Hilburn. Figure 5.9. Forced-normal distribution for the garbage management q-sort. Graphic by A. Hilburn. 201 whose use is fairly universal in published studies. PQMethod allows for a variety of different factor rotations and provides a summary of representative sorts, eigenvalue indexes, and fac- tor loadings by respondents. The loaders are the individuals who correlate strongly with a par- ticular factor. Understanding who the loaders are can help define a factor. During the analysis there is considerable researcher influence on how best to interpret the data. The type of rotation (varimax, manual, centroid) is left to consideration in addition to the number of factors that are included. For this analysis, principal components analysis was used to determine the factors and the varimax rotation appeared to be the most appropriate and is typically the standard in Q-meth- od literature. Q, in short, involves qualitative assessment with statistical analysis. It can be helpful to briefly cover the meaning of particular metrics and values before attempting to explain the signif- icance of the data analyses. Since the assessment of particular garbage classes along the measure- ment scale was made by different participants, or in the case of Q, variables, the z-score offers an intelligible standardized representation of the difference between the absolute score and the mean for a particular garbage class (Watts and Stenner 2012: 9). The Q-score value (Q-SV ) is a normal- ized z-score that approximates the value in whole integers along the measurement scale (-5 to +5 or -3 to +3), facilitating the interpretation of the z-score results. The Q-SV is a useful figure for determining the rank-order of an ideal sort. The p-value represents the statistically significant difference of a particular garbage class’ z-score for a factor, highlighting its potential usefulness as a distinguishing class of garbage for that factor. This analysis assumes significance at both the 95 and 99 percent ( p < 0.05 and p < 0.01) confidence levels. The following sections present the results of the two q-sorts on a) garbage generation and b) garbage management practice. In terms of the factors themselves, the eigenvalue represents the sum of the squared loadings on a fac- tor and any eigenvalue greater than one is typically considered to be significant. The correlation matrix identifies the correlations between factors, pointing out their difference or similarity from 0 (no relation) to 1 (representing the same thing). However, complete reliance on the eigenvalue in determining a factor should be avoided as certain factors may be significant but defy any at- tempts at defining them (McKeown and Thomas 1988: 51). The significant loaders metric is a count of how many participants loaded significantly on a particular factor. Even after the data have been statistically analyzed, the researcher is required to qualita - tively describe and name the different factor groupings by their representative sorts. Additionally, 202 the demographic data, interviews, and observations are used to highlight the subjectivities behind the consensus represented in the factors. There is no easy and systematic process to doing this, either, especially with photographs. The loading data allows the researcher to interpret the factor meaning by allowing them to see the types of participants that more ideally represent a particular factor. The interpretation of what each factor means can rely on statistically significant distin- guishing statements or in this case photographs of more or less discretely defined garbage class- es. These distinguishing garbage classes for a factor have z -scores that are statistically different from the same class in other factors. For a Q-sample constructed of verbal statements, one can describe a factor by assessing the latent discourse from the distinguishing statements as well as by sourcing their provenance by a participant’s loading on it. The cliché that pictures are worth a thousand words does not apply to this Q-methodology survey very well. In this more rigidly- defined, matter-of-fact Q-sample, describing a factor is not as easy as in a text-based Q-sample. For this reason, a conservative approach to the inclusion of factors was taken as there were often too few distinguishing classes or too few loaders on factors to sufficiently elaborate its meaning. ANALYSIS OF GARBAGE COMPOSITION SORT Representing what we [think we] throw away In Mexico, much of the research on garbage comes from an academic and managerial perspective and centers on generation and classification in order to adequately frame the mag- nitude of garbage in a material sense (Ojeda-Benítez, Armijo de Vega, and Ramírez-Barreto 2000, 2002, 2003; Bernache et al. 2001; Buenrostro, Bocco, and Bernache 2001; Bernache 2003; Buenrostro et al 2008; Gómez et al. 2008; Ojeda-Benítez et al. 2008). The increasing generation of garbage, especially that of particular classes like plastic, is one of the most prominent garbage- related issues today in Coxcatlán, Mexico, and the world at large. Chapter 3 presented the 2009 study of garbage generation and classification of 100 households in Coxcatlán’s cabecera mu- nicipal (Sánchez-Mendez and Olmos-Torres 2009). The results show that the average Coxcateco household produces around one kilogram of garbage every day. By weight, over half of this garbage consists of an overrepresented proportion of glass (27 percent) and an underrepresented proportion of plastics such as PET (~14 percent), plastic films (7 percent) and combined other plastics (6 percent), due to their high and low weights per item, respectively. The rest is a mix of combined cardboard (13 percent), organic (12 percent), metal (7 percent), ceramics (6 percent), 203 cloth (5 percent), and paper (3 percent). These empirically-derived figures may provide an objective, quantified, and material view on garbage generation but given their likely confinement to state and municipal manage- ment circles, what frames of reference do most non-expert citizens have to gauge their own garbage generation by class? Also what we think we throw away may in fact differ greatly from what we actually throw away, making the generation of household garbage a discursive process. Identifying the subjectivities of garbage generation can provide a public perception perspective to the more matter-of-fact managerial data on garbage generation. Too often the opinions and beliefs of local, non-authoritative actors are left unconsidered. The following analysis provides that additional perspective on the garbage generation issue. For this analysis, I used a Q-sample consisting of 16 different classes of garbage. This is a smaller, more contracted Q-sample than those used in the following two Q-studies. This was done to facilitate participants’ potential difficulty in recalling the relative abundance by garbage class of the things that they throw away. The conditions for instruction were “this is a representa - tion of the garbage generated in my home”. Participants were instructed to rank-order the cards representing the 16 classes of garbage along a quasi-normal distribution on a scale from -3 (less Table 5.1. Factor correlation and characteristics for garbage generation sort. Data, 2011 Fieldwork. 204 Table 5.2. Z-scores and ranks by factor for the garbage generation sort. Bold indicates 0.05 sig. and * indicates 0.01 significance. Data, 2011 Fieldwork. 205 representative) to +3 (most representative). I ran principal components analysis (PCA) and a varimax rotation to identify three (3) factors that I named “Mostly Foodies”, “Packaged”, and “Homebodies”. The following subsections describe these factors. Tables 5.1 and 5.2 display the various metrics of the garbage generation sort. Factor 1: Mostly Foodies The “Mostly Foodies” factor is largely defined by the distinguishing “food waste” class (Figure 5.10). Participants recounted that “ no ocupamos mucha basura ” or “we don’t produce a lot of garbage” but what they said they produced was mostly food scraps such as stale tortillas, fruit peels, egg shells, and gristle and bones from meat they ate. This is not to say that these par- ticipants did not believe they produced some packaging waste that a perceivably normal family would in the course of a week’s consumption but the comparably low Q-value score and z-score for plastic bottles and receptacles help distinguish the emphasis on food waste’s representation of this factor. The distinguishing class of plastic bags and films was also positioned more toward the middle of the scale as opposed to the following more packaging-representative factors. The middle to negative scoring of other inorganic packaging-related distinguishing classes such as styrofoam, glass, and various plastics also signal the defining importance of food waste for this Figure 5.10. Ideal sort for the “Mostly Foodies” factor in the household generation q-sort. Bold outline and gray background indicates significance at 0.05. Graphic by A. Hilburn. 206 factor. This is not just because food waste has a higher rank-order but also that food-related but not food-exclusive packaging like #1 PET plastic, cardboard, and aluminum define the more representative end of this factor’s ideal sort. Infrequently generated classes such as ceramics and construction materials distinguish the negative end of this factor. None of the factors for the q-sorts on household waste generation can be easily explained by the information of the participants that loaded on it. The Mostly Foodies factor had the high - est amount of loaders with 27 participants. Age and generation may help contextualize the description of the factor. The mean age for loaders on this factor was 42.3 years, over three years higher than the study population mean and approximately six and two years older than the aver- age ages of loaders on the other two factors. Other variables such as sex, educational attainment, and garbage disposal practices offer little inference toward associating subjectivities with demo- graphic groupings for this analysis. Factor 2: Packaged The “Packaged” factor is a far departure from the previous factor in its representation by inorganic garbage that appears mostly to be packaging (Figure 5.11). This considerable differ- ence is evident in the small factor correlation score of 0.1879 between the two factors. The three discretely organic classes that are organic, food waste, garden waste, and rubble have the most Figure 5.11. Ideal sort for the “Packaged” factor in the household generation q-sort. Bold outline and gray back - ground indicates significance at 0.05. Graphic by A. Hilburn. 207 negative (less representative) z-scores. Additionally, there is an extreme difference of 3.171 between the z-scores for food waste between “Packaged” (z = -1.103) and “Mostly Foodies” ( z = 2.068). “Packaged” represents a subjectivity that follows the recognition that household garbage is becoming increasingly inorganic due to changing patterns in consumption and associated lifestyles. This may be linked to increasing obesity rates in Mexico (Olaiz et al. 2007). The entire ideal sort as well as the distinguishing statements help define this factor (Figure 5.11). Plastic bags, various plastics, and styrofoam, all of which mostly represent packaging, are the distinguishing classes that define the positive (most representative) end of the spectrum for the “Packaged” factor. Other inorganic classes such as toilet paper, glass, and metals, some of which are packaging, distinguish the middle of the spectrum. The comparably low representation by food waste and garden waste anchor the definition of this factor on the things left over after consumption, like packaging. The “Packaged” factor had 16 participants load on it whose mean age was around 36 years, a full six years younger than that of the previous factor and three years younger than the entire sample population’s mean age. While the ascription of more processed and packaged foods to the consumptive preferences of younger generations may contextualize this factor more broadly (Guarnaccia et al. 2011), this study cannot definitively establish such a link. Figure 5.12. Ideal sort for the “Homebodies” factor in the household generation q-sort. Bold outline and gray background indicates significance at 0.05. Graphic by A. Hilburn. 208 Factor 3: Homebodies The subjectivity represented by the third factor is a little less apparent than the previous two. This lack of simple distinction arises more from its apparent similarity to the first factor rather than an unintelligible assemblage of garbage class relative abundances (Figure 5.12). This factor, titled “Homebodies”, references a garbage assemblage that many people would find famil- iar in their own homes and draws on a perspective of an array of garbage that a nuclear family generates. A mix of food waste and inorganic packaging largely define this sort. The “Homebod- ies” correlates strongly with the first “Mostly Foodies” factor with a correlation score of 0.4992. Plastic bags and plastic films, often maligned because of their ubiquity as garbage and as litter, are perhaps the most distinguishing class for this factor because of the notable -1.924 z-score dif - ference between factors 1 (Mostly Foodies) and this factor. Frequently, participants would frus - tratingly refer to the frequent presence of plastic bags and plastic in general in their household garbage. It is not likely this subjective representation is reflected by some quantifiable dominant mass or even abundance (count) of plastic bags in their actual garbage. The possible overestima - tion of plastic bags, especially in a begrudging sense, points to the positioning of plastic bags as a proxy for an array of problems attached to garbage and its management (Njeru 2004). Food waste is a distinguishing factor for the “Homebodies” but it maintains a lower yet still positive z-score (lower representation) than that of the “Mostly Foodies” factor. Various plastics and styrofoam, representing packaging and food consumption-related garbage, define the middle of this factor’s Q-sort. These two materials are common but perhaps not everyday items. Styrofoam waste is typically associated with food and drinks eaten “to go” and items made of non-PET (#2-6) plastic comprise a number of bottles, cases, and packs used to contain food and drinks. The highly negative (least representative) distinguishing classes of aluminum and glass are typically less abundant in many homes as plastics are much more ubiquitous, durable, and cheaper materials. Soil and rubble and construction-maintenance materials are easily understood as less-common, less-frequently used waste and help define the less generated side of this factor. 209 ANALYSIS OF GARBAGE MANAGEMENT SORT The management of particular garbage classes The results outlined in Chapter 4 indicate that even with the presence of collection ser- vice in the cabecera municipal there is considerable diversity regarding the practices used by households to manage their garbage. The garbage census showed how households from different situational and demographic contexts deal with their garbage. Yet by focusing on practice, the garbage census survey left the questions on how garbage should be managed and by whom it should be managed unanswered. The subjectivities that inform this diverse array of garbage man- agement practices and the actors who perform them are left open and undefined. This particular Q-sort on garbage management seeks to highlight factors of consensus on how particular garbage classes should be managed and by whom. When considering the aforementioned practices of gar - bage management (collection, burning, dumping, composting, etc.) together, these practices can be divided between those controlled by the household and those provided by the municipio and larger governmental bodies. Where a participant believes particular classes of garbage fit along this spectrum and how their own subjectivity correlates with that of other participants is what this Q-sort attempts to locate. Table 5.3. Factor correlation and characteristics for garbage management sort. Data, 2011 Fieldwork. 210 Table 5.4. Z-scores and ranks by factor for the garbage management sort. Bold indicates 0.05 sig. and * indicates 0.01 significance. Data, 2011 Fieldwork. 211 The Q-sort was based on 27 different classes of garbage ordered along a quasi-normal distribution representing practices that are household-based on the negative3 side (-5 to -1) to garbage management by the municipio on the positive side (+1 to +5) with a “neutral” in the middle. I ran principal components analysis (PCA) and used a varimax rotation to identify three distinct factors and the participant loadings on them. After contextualizing the factors by their defining waste classes, participant attributes, and ancillary data from interviews, I named them “Sanitary Citizens”, “Home Recyclers”, and “Convenience Independents” (Tables 5.3 and 5.4). Factor 1: Sanitary Citizens The Sanitary Citizen factor represents a perspective on a garbage management regime where the municipio is charged to take care of “dangerous” or “hazardous” garbage that has the potential to infect and pollute (Figure 5.13). This factor is defined largely by the high, statisti- cally significant Q-score and z -score values of waste classes known to be harmful to the environ- ment and public health. These high values and scores relate to the positive end of the -5 to +5 scale, indicating agreement that formal, governmental-controlled management of these classes is preferred. The Sanitary Citizens factor alludes to the destruction or sanitary displacement of garbage away from populations more so than a reduction-reuse-recycle strategy. Priority for col- lection was given to managing seemingly the most dangerous, if least abundant, garbage classes Figure 5.13. Ideal sort for the “Sanitary Citizen” factor in the garbage management q-sort. Bold outline and gray background indicates significance at 0.05. Graphic by A. Hilburn. 212 over more everyday garbage classes. Batteries were believed to be best managed by the munici - pal government away from the home site and had a Q-sort value of +5. Medical waste had a similarly high Q-score value of +4. Household chemicals also helped define this sort with a high z -score and positive Q-sort value of +3. Related to medical waste by its potential biohazard, dia - pers had a defining positive Q-value of +2. Seemingly absent from the Sanitary Citizens factor were statistically significant negative z-scores and Q-values. Thus, the overall defining preoccu- pation of the Sanitary Citizens is the management of garbage by the municipio through collection and final disposition. The Sanitary Citizen factor had the highest eigenvalue score (λ = 27.772), meaning that is was the most practically significant or well-defined factor. It also had the highest number of significant loaders with 37 participants or just over half of the study population. Factor 2: Home Recyclers Similar to the Sanitary Citizens factor, the Home Recyclers factor is associated with a garbage management regime based on collection by the municipio. However, the latter factor alludes to a different role of the government in garbage management (Figure 5.14). The Home Recycle factor places recyclable materials in the most forward position in contrast to the Sanitary Figure 5.14. Ideal sort for the “Home Recyclers” factor in the garbage management q-sort. Bold outline and gray background indicates significance at 0.05. Graphic by A. Hilburn. 213 Citizens’ choice of the dangerous or infectious waste classes. In fact, the rank-ordering of waste materials on the extreme positive end (waste classes that should be managed by the government) read off like a recycling brochure. Furthermore, they appear to be ordered by general understand- ings of their perceived recyclability. Recyclable materials like aluminum, plastic bottles, various plastics, cardboard, glass, other metals, paper, and plastic bags define the positive end of this sort. Additionally, this factor exhibits a more defined and recognizable array of waste classes that should be managed within the home. Garden waste or hojarasca , a general definition of rubble or escombro, and vegetable fibers define the factor regarding garbage classes that should be managed at home. Unexpectedly, batteries had the most negative statistically significant z-score value, meaning the item most associated with household management. While it was not a statis - tically significant defining statement, medical waste had the most negative z-score. These gar- bage classes, medical waste and batteries, were rank-ordered at polar opposites by the Sanitary Citizens. A loading participant on this factor noted that the municipal garbage collectors do not accept batteries and discourage mixing them in with their household garbage, leaving households to store them in their houses in order to manage them properly. Others noted that they almost have no battery waste in their houses and it was barely worth mentioning. This latter explana- tion helps to explain the negative z-score of medical waste. There were 21 significant loaders or defining variables (participants) on the Home Recyclers factor. Participants more often noted the economic benefits of recycling instead of the potential ecological benefits. The importance of both formal and public environmental education were noted as agents in constructing this subjec- tivity by participants. Factor 3: Convenience Independents Defining the Convenience Independents factor is more difficult and is less apparent. The diificulty in interpreting this factor is mainly because it references a subjectivity that is not easily recognizable by someone, like myself, who has known no other garbage management regime besides collection service. This factor only had five significant loaders and all were middle-aged to elderly with comparably lower formal educational attainment. All but one were from the colonias outside of the centro. The loaders on this factor also had notably less frequent garbage service due to their presence in the colonias. Also, the household managed garbage classes and state managed garbage classes in this factor do not follow any sort of organic/inorganic divide 214 that is more apparent in the previous two factors. Rather, the classes deemed to be household management-oriented (negative and 0) are mixed (Figure 5.15). In opposition to the other fac- tors, inorganic waste can be managed in the home for this factor. Styrofoam, various plastics, and artificial fibers define the household management-oriented garbage classes for this factor. Distinguishing neutral garbage classes include plastic bags, meat/animal food waste, rubber, and glass. Again, these are mixed and lack the influence of any officially prescribed garbage manage- ment discourse. The Convenience Independents factor lacks the organic/inorganic or recyclable/ non-recyclable or hazardous/non-hazardous distinction that are present in the other factors. This factor has “convenience” in its title both because there is no recognizable garbage class pattern in what is managed in the household and by the municipio but also because what is supposed to be managed by the government or external parties are heavy, cumbersome items such as cement and concrete trash, and general rubble. These items create hindrances for these loaders in their house- holds and the municipio has the infrastructure to perceivably deal with it. A mix of recyclable or dangerous garbage classes on the positive end of the scale such as aluminum, batteries, card- board, and animal waste may point to some of subjective similarity with the previous two fac- tors. The Convenience Independents factor appears to represent traditional views toward garbage management, where garbage is garbage and not an assemblage of different classes along organic/ inorganic lines. It likely has more to do with a perspective based on garbage as a hindrance in Figure 5.15. Ideal sort for the “Convenience Independents” factor in the garbage management q-sort. Bold out - line and gray background indicates significance at 0.05. Graphic by A. Hilburn. 215 that larger and cumbersome objects should be managed by third-parties if possible. This factor represents a perspective of the underserved where the opportunity to perform, if only discursive- ly, the mainstream separation and recycling practices referenced by the previous two factors and thus illustrates a traditional viewpoint. DISCUSSION AND CONCLUSION The garbage generation sort This Q-sort showed some inter-factorial correlation but demonstrates what appear to be three different viewpoints on garbage generation (Table 5.1). There is a relatively high correla - tion ( r2 = 0.4992) between factor 1 (Mostly Foodies) and factor 3 (Homebodies). Both factors have food waste as highly representative of their garbage assemblage but the increased represen- tation of packaging by the latter factor sufficiently distinguishes it from the former in conceptual terms. Nevertheless, the primary intent behind conducting this sort allows some comparison of the subjective estimates of garbage generation to the material estimates of garbage generation at the national (INEGI) and local scales (Sánchez-Mendez and Olmos-Torres 2009) in order to test the hypothesis that these estimates do not correlate. Direct comparisons are not possible because of the difference in measurement scales (Q-SV and percent of a garbage class) as well as the dif- ference in garbage classifications. Nevertheless, a constrained relative comparison will be used in this case. The “Mostly Foodies” factor follows with some of the local material estimations of garbage generation as well as national estimates. Organic waste, which subsumes the food waste class in this study, represents only about 12 percent by weight of the Coxcatlán estimation but represents just over half of the national estimates. Conversely, the neutral-to-moderately positive placement of plastics for this factor may overestimate the national estimate’s approximately ten percent but follows the local estimate’s representation of PET/plastic bottles and plastic bags and films representation. The moderate placement of glass fits the national estimate but underesti- mates its locally estimated proportion. The “Packaged” factor likely over-represents the plastic fraction in both the national and local estimated waste stream. Plastic bags and films are less prominent in this factor than the lo- cal generation estimates. The 10.9 percent of combined plastics from the national estimate also does not believably fit with the high Q-score values of other plastic categories of plastic bottles4, 216 various plastics, and styrofoam. However, it is worth noting that this representation is not eas - ily assessed by the local and national estimates as they are both weight estimates which would underestimate abundance. This factor’s moderate placement for metals fits with local estimates than national ones (3.4 percent). The low placement of food waste as well as garden waste sits in moderate contrast to organic waste’s representation in the local and is in extreme contrast with national estimates of organic waste. The “Homebodies” factor situates plastic bags and films as the most representative class of garbage. Many participants often equated the garbage problem with the ubiquity of plastic bags with sayings like “pura bolsa de plástico ” or “nothing but plastic bags” when looking for a place near the positive end of the scale to put the card (Njeru 2006). Yet, like the previous fac- tor, assessing plastic’s representation in this factor with the national and local estimates is left open to interpretation. On the opposite end, the lowest placement of aluminum is also difficult to assess due to the conflation of aluminum in a general metals classification for both the local and national estimates. Food waste and garden waste’s moderate placements fit the local estimate for organic waste but fall far beneath that of the national estimate. The moderate placement of sty- rofoam and various plastics fit the local percentages of their respective categories of 3.7 percent and 6.1 percent. Beyond measuring the congruity of what people think they generate against what they may really generate, these factors also point to a tension between tradition and globalization. The “Mostly Foodies” factor indirectly implies a traditional, mostly inorganic perspective regard - ing garbage generation. The “Homebodies” factor also touched upon a less-packaging-abundant waste stream. Food waste in this study came to indicate locally available, easily perishable goods such as tortillas, eggs, fruits, and vegetables. Certain loaders on these factor as well as other participants lamented what they perceived as a growing inorganic fraction of waste. The “Pack- aged” factor represented a waste stream that is indicative of what Leatherman and Goodman (2005) term the “cocacolonization” of traditional Mexican foodways by transnational corpora - tions and their packaged foodstuffs. This factor parallels the trend across Mexico where the inorganic fraction of garbage is increasing at a greater rate than inorganic garbage. The “Mostly Foodies” factor and “Packaged” factor bookend this traditional-globalized tension of garbage composition with the second factor (Homebodies) somewhere in between. Whether or not these factors are truly reflective of local households’ garbage or are a type of performed representation 217 of perceived garbage composition, the factors inevitably reflect larger debates on the globaliza- tion of foodways and more relevantly, the solid waste stream. The garbage management sort The Q-study on garbage management practices brings to light the issue on the role of the municipio’s collection service in the study area. By allowing participants to reflect their views on how and by whom particular garbage classes should be managed, the utility of garbage collection in the lives of Coxcatecos can be seen. Conversely, the value of traditional practices can also be assessed. As noted in the previous chapter, residents in the cabecera rely on more than just the collection and recycling service to manage their garbage. Some burn, others compost or bury, and a few store garbage in addition to placing their garbage at the nearest collection spots. This Q-study also contributes to a more complete, perception-based understanding about the environ- mental governance of garbage within the household (Barr 2002; Bulkeley and Gregson 2009). In the previous chapter, the stated practices were just that, frequencies of the incidence of certain practices. While useful in determining their expanse of such practices, the garbage census left off in its inquiry about how classes of garbage should be managed. Applying Q-method to how dif- ferent classes of garbage should be managed places the subjectivity and ultimate decision making of the participant within the entire field of governance of that garbage. The first factor, “Sanitary Citizens”, prioritized the sanitary removal of seemingly dan- gerous waste, be it toxic or bio-hazardous, by the municipio away from the home. This was a fairly widely agreed upon factor with 38 participants significantly loading on it. This view per- ceives certain garbage as a threat or hazard (Moore 2012: 783-784). There was only one defining class on the negative or household management side of the scale. Thus, the management focus of the “Sanitary Citizens” was on other-than-household actors and reliant on the spatial solution of garbage removal of hazardous materials. This factor represents a viewpoint that has bought in to the collection service because they view it as an improvement to sanitation and household safety. The “Home Recycler” is also representative of a viewpoint that defines a role for the municipal collection. This role differs from the sanitation role defined by the “Sanitary Citizens” because the classes that are intended most for management by the municipio are recyclable. From this perspective, waste is less a risk but is an opportunity to engage a previously less-than-available conservation practice. The “Home Recycler” earns the other part of its name by retaining some 218 distinguishing classes that should be managed at home. These classes, like garden waste, veg- etable fibers, and rubble, are garbage classes that have been in the past typical of home manage- ment. The “Home Recycler” factor is a good example of the power of Q-method to highlight the relation of unexpected subjectivities. Here we have a factor that is simultaneously traditional in its reliance on home garbage management and also environmentalist in its deep recognition in the value of formal recycling. Lastly, this fact exhibits the effect of environmental education on attitudes toward recycling, as many participants noted that education and public information were some of the most important issues related to garbage (Fernández-Crispín 2009). Contrasting the previous two factors with regard to the role that the state plays in gar- bage management is the “Convenience Independents” factor. This factor largely represents an underserved or non-reliant perspective concerning garbage collection, representing a subjectivity that has not bought in fully to the garbage collection model of garbage management. The lack of an apparent organic/inorganic dichotomy for home and state managed garbage points to a prag - matic view of garbage as matter that is truly waste and needs to be destroyed or removed from the site rather than representing a commodity, an opportunity for environmentalism, or a threat to home sanitation. However, two loaders noted the importance of separation but another loader mentioned that there is not enough information on how to separate and what classes of garbage should go where and more importantly even less motivation to do so. All but one loader are from the colonias outside the centro and two of the five loaders reported that they burn their garbage. Concluding remarks Q-method is a useful methodological tool to unpack the subjectivities of a population regarding a human-environmental issue like garbage. However, it should be noted that Q-method is not an end in of itself. It adds an enlightening angle on an issue in order to more comprehen- sively understand it, but the results should complement additional research methods rather than be the center of it in the same way excerpts from a semi-structured interview would complement a more concrete understanding of something. As employed in this dissertation, Q permits the subjective voice of a sample of Coxcatlán’s residents towards two garbage-related issues to be statistically factored. These factored subjectivities can then add to more material and practical understandings of garbage generation, garbage management, and its environmental impact. The results of the first q-sort define the viewpoints of what Coxcatecos believe constitutes 219 everyday waste. The generation of garbage and its components is perhaps the most discussed sin- gle garbage-related issue in Coxcatlán and across the world. As Coxcatecos all contribute to the construction of this issue locally, both materially and discursively, then a description of how they view it would also complement its more material estimation by x kilograms per day of y classes of garbage. Also, the second Q-study highlighting viewpoints on how garbage should be man- aged similarly complements the previous two chapters’ focus on how garbage is really managed. The final Q-study on the relative environmental impact of garbage gives a grounded nuanced descriptive account on local perceptions of the issue of environmental degradation from garbage, adding a perceptional dimension to its more quantifiable risks listed in more technical literatures. The previous Q-sorts provide a ground-up, participant-centered dimension on the how two garbage-related issues are constructed. Q-method allowed the participants to construct their own practiced subjectivities as they saw it, yet I a priori framed that garbage generation, management, and environmental impact are in fact garbage-related environmental issues. This chapter falls short of representing a wide expanse of garbage-related issues in Coxcatlán as rep- resented by its residents. Also, this chapter’s study area was confined to just the cabecera. Using participatory risk mapping, the following chapter will demonstrate the heterogeneity of opinion regarding a multitude of issues across the entire municipio. Endnotes 1. NIMBY stands for “Not In My Back Yard” and refers to a spatial-political stance where an individual or group of people can be for a particular thing, land use, or spatial practice but be against its proximity to them. The cited example with Hawthorne, Krygier, and Kwan (2008) examined NIMBYism with regard to the construction of a rails-to-trails park in Ohio. 2. I initially administered three q-sorts. The results of the q-sort focused on the impact of garbage proved to be too dissonant to clearly identify any distinguishable factors. 3. By “negative”, I strictly imply negative numbers and not any semantic negativity toward household management of gar- bage. This was made explicit to participants who inquired about the attribution of the Likert scale used in this sort. 4. Plastic bottles and containers is not significant at the 95% C.I. and is thus not a distinguishing class for this factor but its inclusion helps add to the comparison between real and perceived garbage generation. 220 CHAPTER 6: The construction of garbage as a set of environmental issues in Coxcatlán, Puebla, Mexico. INTRODUCTION The previous chapter intensively examined the subjectivities of a relatively small group of residents in the cabecera municipal of Coxcatlán municipio. It specifically looked at the subjectivities surrounding two issues about garbage: generation and its ideal management. These two issues are important on a global level. As such, their understanding as issues comes largely from academic literature, government agencies, and popular discourse. In essence, they are pre - determined and were so in the Q-method research. Yet there is more to garbage’s framing as an environmental issue beyond generation, management, and environmental impact. Unfortunately, there is little discussion of what precisely these issues are as many are specific to communi- ties and places. Additionally, framing of the garbage issue from just a generation-management perspective glosses over the social aspects of garbage, having it come across as a straightforward managerial issue. Perception of garbage and its construction as a set of socio-environmental is- sues, like other environmental issues, varies according to place and by population characteristics. A better understanding of the plurality of perception and opinion from across the municipio’s di - verse population more accurately frames garbage as a socio-environmental issue there. In short, this chapter allows the residents of Coxcatlán to systematically outline their concerns about garbage. This chapter’s use of participatory risk mapping, a method used in development research to identify sources of risk, seeks to highlight how Coxcatlán’s residents from across the muni- cipio perceive garbage as a set socio-environmental issues from their perspective. This chapter will attempt to answer the following two research questions: 1) What are the most frequently noted garbage-related issues and what are their relative impor - tance or severity? 2) What is the heterogeneity of opinion within the entire study population with regard to spatial and demographic qualifiers? 221 The first question forms the core material of this chapter. The resulting garbage issue “maps” will address the first question by highlighting heterogeneity of perception according to place and a number of relevant demographic variables. Additionally, chi-square statistics will be applied to test whether place and sex have any effect on garbage issue perception. Addressing the second question, statistical testing can identify any within group variation on the perception of garbage-related issues (Baird 2009). This question in particular addresses spatial and demograph - ic differences within the total sample population toward the more frequently mentioned issues with the goal of identifying place and group-specific issues. METHODS “Participatory” “Risk” Issue “Mapping” Participatory risk mapping or PRM1 is a relatively new method to identify and rank per- ceived risks or sources of risk in a population. The approach was developed by Smith, Barrett, and Box (2000; 2001) to study the nature of perceived risks in Ethiopian and Kenyan pastoral- ist communities. Since then it has been used in a few other studies to gauge perceived risks in Tanzania (Quinn et al 2003; Doss et al. 2008; Baird 2009), but as of yet has not been employed outside of East Africa or beyond identifying risk. Most simply, participatory risk mapping is “a two-stage system of ordinal rankings wherein respondents first identify risks and then rank them” (Baird et al. 2009: 467). A primary use of participatory risk mapping is to demonstrate the heterogeneity of per- ceived risk among seemingly homogenous populations. The data from this method can then be taken to examine differences in risk perception within and among study populations. Participa- tory risk mapping not only notes the incidence of a particular risk being mentioned across a study population, it also provides an index of severity or importance of that particular risk compared to other risks. Risk, in the context of this method, is defined as “exposure to uncertain and po- tentially unfavourable consequences” and is used synonymously with “hazard” (Smith et al. 2001: 3). Risk, environmental risk, and perceptions of environmental hazards have long been and still are frequent topics of human-environmental research (Kates 1962, 1971; Lowenthal 1967; Saarinen 1984). The concept of risk has application to a study of garbage because of the hazards, perceived or real, to public health and environmental sustainability that poorly managed gar- bage poses. However, risk also implies some negative or “unfavorable” connotation that may be 222 unwarranted in popular opinions regarding garbage or waste. One cannot assume that all percep - tions of garbage are related to some risk. A lack of service or need for more recycling does not intimate notions of risk in the same sense. Therefore, this study approaches this two-dimensional method (incidence and severity) of gauging perception in a more openly-defined way by not focusing entirely on a narrow definition of risk and instead expanding the method’s scope to all issues regarding garbage and its management. This inclusive approach is more applicable to the previously stated research objectives and it is does not impose a priori assessments of garbage or garbage-related issues as risks that may be notable but less important to a study participant. The “participatory” aspect of participatory risk mapping must be also qualified and cri- tiqued. Participatory research in the social sciences and especially field-based human geographic investigation is a popular methodological paradigm that seeks to de-center the researcher and represent the study population as it chooses to represent itself and then utilize the research results (Yeich and Levine 1992). The popularity of a more horizontal, participatory approach in social science research is a notable acknowledgement of the situated, positional nature of all research (Park 1993). The inclusion of a participatory aspect to a research project is an active step to pro- vide alternative methodologies that counteract researcher-researched, expert-amateur power dy- namics inherent in earlier and more traditional social science research (Herlihy and Knapp 2003: 304). In current human geography there are numerous popular participatory approaches such as public participation GIS or PPGIS (see Sieber 2006 for a comprehensive epistemological ac- count), participatory research mapping, the first method dubbed PRM, (Herlihy and Leake 1997; Herlihy 2003; Herlihy and Knapp 2003), and participatory action research or PAR (see Whyte et al. 1989 for a summary) among others. In contrast, participatory risk mapping’s pedigree is more attributable to the handful of methodologies based out of rapid rural appraisal or RRA (Chambers 1980; Carruthers and Chambers 1980; Beebe 1987; Grandstaff and Grandstaff 1987) and later, participatory rural ap - praisal or PRA (Chambers 1994a, 1994b, 1994c; Campbell 2001), the latter with deeper roots in non-state development than in strictly academic inquiry. RRA was an alternative to the overly- preconceived, time-consuming, and often inaccurate, too hastily done, biased, and rapid rural surveys of rural development research of the 1960s and 1970s. RRA, which rose to popularity in the 1980s is less a singular method and more a development-oriented research approach that favors multiple methodologies that include secondary sources of information with punctuated 223 empirical research. Yet, preconceived research design, researcher-oriented end uses, and the top- heavy dynamic opened the way for more locally-applicable, locally-influenced, and participatory research design and methods. The purpose of adding a participatory approach to development research had the primary goal to identify potential development issues as the target populations perceive them, thereby empowering them (Chambers 1994a). In Smith et al.’s participatory risk mapping research (2000; 2001), groups of people were interviewed and their collective responses and rankings were summarized to represent their vil- lage’s perceived risks, such as human disease and livestock predation, and the hierarchy of those risks. The responses were then tabulated and analyzed. The critique that participatory risk map- ping and its predecessors, RRA and PRA, are only marginally participatory as the design of the method and research parameters are heavily guided and largely predetermined by the researcher appears sound in this light. The primary engagements with participation in participatory risk mapping are a more open dialogue and the use of the subjects’ responses as data. The data come from the participants and utilize their perceptions of risk or particular issues as the framework and focus for development work. However, this participation appears little different than most non-participatory qualitative approaches such as ethnography and focus groups. Ultimately, the researcher asks the questions, collects the responses, and analyzes the data outside of the partici- pation of the “participants”. For this reason, I conceptually omit the “participatory” qualifier for the approach employed in this study but leave its title untouched for epistemological congruence with other studies using the method. Data collection and analysis The two-step procedure for collecting participatory risk mapping data is fairly simple and noted to be “easy to field” or easy to administer (Smith et al. 2000, 2001; Quinn 2003). The interview from which the participatory risk mapping data is obtained is open-ended, allowing the respondent to give their own responses in the way they see fit to describe the most relevant issues or risks. First, the researcher asks the participant to list what they consider the most im- portant risks, or in this study, issues related to garbage. We applied the more open approach to gauging not only risk but any issue related to garbage in our field work by maintaining this open- endedness with our interviewing with the study respondents. Instead of “risk” we used words and phrases such as asunto or “issue”, or lo que nos afecta or “that which affects us”, in our ques - 224 tions and instructions during the interviews. If participants inferred the words problema (“prob - lem”, “trouble”) or riesgo (“risk”) from what we were saying in our instructions, it is intended that these understandings came from the participants themselves. Our broadening of the concept of risk utilized in this methodology is only conceptual and does not change any methodological considerations as devised by Smith et al. 2001 and other practitioners of participatory risk map- ping. The rank-ordering of garbage-related issues by their importance instead of risks and their severity does not change the way the data are calculated or are graphed. In another sense, one could hypothetically have participants list their favorite ice cream flavors and rank them in terms of their relative deliciousness. In our case, the respondent discussed and elaborated on garbage- related issues while the researcher recorded their responses on the questionnaire, all while al- lowing the respondent to observe their recording. The researcher then allows the respondent to review the issues that they mentioned and revise or add to the list. Lastly, the respondent is asked to rank-order by number the issues they mentioned from most important (1) to less important (subsequent numbers). Ties are permitted and numbers that follow account for these ties (1, 2, 2, 4…). These data then allow the calculation of incidence and “importance” values for particular issues which can then be subsequently “mapped” or plotted on a two-dimensional coordinate grid. The incidence score, representing the proportion of respondents that listed that issue, illus- trates the breadth of a particular issue across the sample or subsample of the study population. The index is measured from 0 (no mention) to 1 (always mentioned). A particular issue, such as dumping, may have an incidence value of 0.35, indicating that 35 percent of the sample popula - tion listed it. This incidence index can also be used as a dependent variable in correlation-regres- sion and chi-square hypothesis testing. Calculating the “importance” or “severity” index, as it is listed in other uses of participa - tory risk mapping, is less straightforward than the incidence index. The reason behind a more complex calculation is that the number of issues listed varies by respondent, presuming the need to derive uniform intervals. One respondent may list four issues and another may list two issues and both may list the same variable, say dumping, with different rankings. Setting uniform in- tervals for each respondent mitigates the effect of the difference of listed issues between respon- dents. This use of uniform intervals is preferable to that used in the original publication describ- ing participatory risk mapping (Smith et al. 2000). The updated uniform intervals are calculated 225 by 1/ n with n representing the number of issues mentioned by that participant. Utilizing these intervals, the “importance” factor for a given issue by a respondent is termed R ij = 1 – (r ij -1/n i ) , meaning the importance factor, R, is the rank or r of the issue j in the context of the group of n issues noted by i respondent (Smith et al. 2001; Baird 2009). The mean of an identified issue’s “importance” index values across the sample or subsample allow it to be mapped with incidence values on a two-dimensional grid. Fieldwork and study population The fieldwork for this research was conducted in conjunction with the garbage census survey from late August to late November, 2011. The details of the field work and an elaboration of the variables’ significance for this study are described in more detail in Chapter 4. The survey instrument and the interviews were administered by me and my local research assistant. Most of the surveys were given house-to-house but some respondents were interviewed opportunisti- cally at a health clinic and outside of a school at the request of a few communities’ leaders. Prior to conducting the research, we practiced the interviewing method outlined in previous studies Table 6.1. Surveyed households as a percentage of the total by settlement. Data from 2011 fieldwork.and INEGI 2010. 226 (Smith et al. 2000, 2001; Baird 2009) through numerous mock interviews between each other as well as with friends. We then performed a trial run of interviews while testing the survey instru - ment’s applicability and efficiency. After editing and streamlining the survey and interview pro- cess, we began work in the five communities in the valley portion of Coxcatlán municipio. After a sufficient sample was interviewed in the valley settlements, we began sampling households in seven communities in the sierra portion of the study area. The participatory risk mapping component of the survey required more time and some respondents, who also participated in the garbage census, did not have ample time or desire to continue beyond the garbage census survey. Thus, the entire sample size for the participatory risk mapping study is 433 participants or 83 percent of the 520 surveyed for the garbage cen- sus described in the previous chapter. Described otherwise, every risk mapping participant also participated in the garbage census but not the inverse. The number of sample respondents and households by community and their proportion of total households in their respective communi- ties are listed in Table 6.1. We targeted the heads of households for our study because they publically represent the home and are responsible for taking care of garbage disposal. The head of household was not always available but slightly under three-fourths or 74 percent surveyed claimed to be one of the heads of household. While we targeted heads of households, anyone over the age of 18 was considered a potential participant. In the instances where a head of household could not be in - terviewed, we asked if someone who was involved in household garbage management who was available. In our sample population, there is a considerable gender bias with 320 women of 433 total respondents that represent 74 percent of the sample. Taking care of the household garbage is an observably gendered practice along with cooking, cleaning, and laundry as well as any other form of domestic labor. Perception of environmental risks may be contingent on these gender differences (Flynn et al. 2006) and feminist political ecology has highlighted the differences in resource use and environmental perception by gender (Rocheleau 2005, Rocheleau et al. 2006). Sex or gender is likely an important factor due to the highly feminized nature of the work of gar- bage management within the home. The resulting “maps” should point to differing perceptions of garbage-related issues between what are perhaps two different experiential circumstances (male and female roles) in the study population. The gender bias is discussed in further detail in Chap- ter 4. 227 Besides gender and heads of household, we recorded 26 other variables on participant demographics, household size and characteristics, and household garbage management. We also collected spatial data specific to the colonia or barrio. While not all of these variables are di - rectly relatable to the participatory risk mapping data, some are useful for stratifying the sample to compare the issue “maps” as well as providing independent test variables for statistical hy - pothesis testing. Age of the participant is a useful auxiliary variable for participatory risk map- ping because it can impart a generational dimension to individual perceptions of garbage and the environment (Rohrschneider 1988; Dunlap and Scarce 1991; Jacobs 2002; Anderson et al. 2007). The relative age of a head of household, along with other variables, can also help infer the rela- tive affluence of a household as it typically takes decades, if ever, to develop household wealth and equity. The mean age of the sample population was 41.3 years, perhaps reflecting the high frequency for sampling the head of household. The valley subsample was slightly older than that of the sierra, with means of 42.9 years and 38.7 years, respectively. Environmental education is a considerably ample subfield of education research and there are substantial literatures on the effect of schooling on attitudes regarding the environment (Lynch and Hutchinson 1992). Educational attainment is thus an important variable as formal schooling often entails some form of instruction of the natural sciences that may affect attitudes on garbage and its relation to the environment (Volk and Cheak 2003; Sahil and Erkal 2010). Environmental education in Mexico and across the world typically comes from a Westernized, nature-as-natural-resources or positivist ecosystems approach that obscures indigenous and folk natural science (Gómez-Pompa and Kaus 1992). The average years of school attendance for the entire sample population was 6.8 years, with higher attainment in the valley settlements (8.1 years) than in the sierra (4.4 years). The native status regarding a participant’s current residency could potentially be related to his or her opinion on garbage-related issues in their respective community. While the link to insider/outsider status to attitudes toward garbage and garbage-related issues could be tenuous, someone from outside a community may view daily life and practice in their new community differently and therefore raise a contrasting perspective or new issue yet unmentioned. In the sample population, 83 percent are residents of what they consider their “hometown”, five percent are from other communities within Coxcatlán municipio, and four percent are from the nearby re- gion of the Sierra Negra and Tehuacán Valley. Just fewer than five percent were from other states 228 in Mexico, only two respondents were from other parts of the State of Puebla, and three percent could not claim any hometown at all. Indigenous people can harbor differing perspectives on nature and issues regarding environmental conservation than the mainstream society (Toledo 1994; González 2001; Toledo et al. 2010; Wildcat 2010). Speaking an indigenous language, which would most likely be Na- huatl in the study area, provides an empirically-derived identifier for defining indigenousness. A considerable majority of the respondents we surveyed in sierra communities speak Nahuatl (88 percent) with a few speaking other indigenous languages (3 percent), and few not speaking any indigenous language (5 percent) or not responding (4 percent). Only 33 respondents or less than twelve percent of the valley community subsample spoke an indigenous language and of these, twelve of the 33 individuals were originally from communities outside their current residence. Assuming that indigenous participants harbor different ecological knowledges than those of the non-indigenous sample, one could also assume a different plot of data regarding perception of garbage-related issues. While there are small populations of indigenous-identified residents in the valley, the correlation of indigenousness and residing in a sierra settlement is too great. Thus, the sierra population and indigenousess will be conflated in this study. The number of responses given by the participants varied considerably. In our fieldwork, we avoided instructing them to list a certain expected amount of issues. We wanted responses about the issues that immediately came to mind regarding garbage or were relatable or important to a participant’s daily life rather than a general listing of all the garbage related issues they could conceive. Participants who listed two responses were the largest class (40 percent), followed by those that listed just one (31 percent), who were followed by those who listed three (25 percent). Participants that listed four responses made up three percent and one individual out of the en- tire 433 participant sample listed five responses. The uniform intervals based on the number of responses by an individual make the ranked “importance” factor comparable across these classes (Smith 2001: 5). In participatory risk mapping, the generalization of the risks or issues that people actually mention during the course of interviews to the risks or issues that are analyzable in a database is an important but not often-mentioned step in the process. Smith et al. (2000, 2001) and other studies utilizing the method (Quinn 2003; Baird 2009) do not provide any in-depth explanation on how they reduced what was likely a diverse list of risks to the dozen or so that they graphed. 229 This process reduces what people say to what is essentially a code, which is a form of quantifica- tion used in ethnographies (Hay 2005). The degree of inclusion for a given issue has consider - able effects on the final calculation of importance and incidence indices. One important purpose of this study was to identify the diversity of perceptions regarding garbage-related issues across the study area from a bottom-up perspective. During fieldwork, we came to discover that dump- ing was a major issue. However, some people added specifics to what was being dumped, such as dead animals and where it was being dumped, such as a barranca. From both a managerial and research perspective, both of these factors are important and worthy of inclusion into their own discrete categories rather than grouping dumping into one inclusive category. Therefore, I abstained from heavy-handed generalization and minimized the reduction of what people said to issues. The result was a fairly large set of issues that numbered 80 distinct issues. From an analytical perspective more issues means dilution of the incidence and importance index values, which can problematize analysis by under-representing issues that are essentially the same and over-estimating the representation of less-often mentioned issues. However, the preservation of unique issues is a more important goal of this research and the resulting data indices performed sufficiently in later analysis. This study presupposes that the incidence, importance, and diversity of perceived issues regarding garbage are contingent upon numerous factors beyond simple individuality of opinion. To display the effects of these factors, the study population is stratified and the resulting issue graphs, or “maps” as implied by the method’s inventors, allow the incidence and importance indices of each subsample to be compared beside each other2. I stratified the sample population demographically by sex, age, and educational attainment in years and spatially by valley/ sierra , cabecera /other valley settlements , and centro/colonia within Coxcatlán town . While most these stratifications are binary, the data, with the exception of sex, do not imply natural di- chotomies but rather are paired for ease of graphic display, narrative interpretation, and statistical testing. The previously noted differences between settlements in the valley and the sierra in terms of participant demographics, the unequal access of municipal garbage management infrastruc- ture, as well as different approaches to local governance will also produce incidence-importance graphs worthy of further discussion and analysis. Similar spatially-implied differences can be made apparent between the centro and colonias of the cabecera as well as those between the 230 cabecera and the remaining valley settlements. The implied differences are both in terms of development and demographics as well as differential access to formal garbage collection and disposal. A map illustrating the spatial stratifiers that will be used in this study is Figure 6.1. The “incidence” factor allows for binary logistic regression-based hypothesis testing (Baird 2009). Pearson’s Chi-Square statistic can also highlight relationships between spatial and Figure 6.1. Participant households by settlement and spatial divisions used in statistical analyses. 231 demographic variables and the incidence of particular issues being mentioned when both are bi- nary nominal variables. While the main intent of this chapter is to simply but effectively describe the heterogeneity of opinion regarding garbage, the statistical tests quantitatively infer this dif- ference within the total sample population. Using logistic regression, I focus on the effects of age and educational attainment on the odds of mentioning the ten most frequently listed issues. The binary spatial and demographic variables were cross-tabulated in SPSS to test for statistically significant relationships between them and the ten issues with the highest frequency. Relating the incidence of issues with spatial and demographic groups can identify group or place-specific issues apart from the entire sample. RESULTS AND DISCUSSION Mapping issues across the municipio Reading and interpreting the “issue maps” or incidence-importance graphs is fairly straightforward and simple. The x-axis represents the incidence, which is the proportion of participants that listed a particular issue. The y-axis displays the “importance” factor that rep - resents a normalized rank of the importance if that issue across participants that mentioned it. If one divides the map into quadrants, the top-right quadrant encapsulates those issues that are both noted across a sizable portion of the participant sample as well as maintain a notable rank in terms of importance as a garbage-related issue. Conversely, the bottom left quadrant represents issues that are neither widespread nor are of high relative importance. The relative location of these issues plotted out in space allows one to effectively visualize the differences or similarities in their perception by the total sample or a subsample of participants. Ultimately, this provides a demographically and spatially informed “map” or graph of how garbage is perceived from-the- ground-up in the study area, allowing for the discovery of perhaps yet-unrecorded issues as well as in-depth contextualization of issues that are more formally recognized. The “issue map” of all 433 participants from across the municipio of Coxcatlán is listed in Figure 6.2. The graphs between valley and sierra (Figure 6.3), cabecera and other valley settlements (Figure 6.4), centro and peripheral colonias (Figure 6.5), and men and women (Fig - ure 6.6) are on the following pages. The three most frequently cited issues in order from greatest 232 to least was the problem of burning garbage, followed by general dumping, and the maintenance of public cleanliness or aseo público . These three issues also had fairly high importance values with dumping being listed as the most important out of these three issues followed by public cleanliness and burning. Other issues with lesser, but notably high frequency and relatively simi - lar importance are street littering, lack of or insufficient municipal collection service, a lack or insufficiency of public garbage receptacles, and contamination of water by garbage in order from highest to least of these issues. Issues with frequencies greater than 0.02 are listed by spatial stratifying variables and sex in Tables 6.2 to 6.5. Burning Problematic burning was the most frequently cited issue related to garbage in the entire study population, with 27 percent mentioning it in the participatory risk mapping study. In light of the existence of municipal collection, burning was perceived by those who mentioned it as Figure 6.2. Graph or “map” of issues by incidence (x-axis) and importance (y-axis) for the entire 433 participants. 233 culturally backwards and environmentally damaging. This judgment was moderated to a degree by those who still must contend with their garbage outside of the collection service area, but a noteworthy number of sierra participants perceive it as a less-than-ideal, if necessary practice. The importance index value was also moderately high. Men most frequently listed burning as an issue and to it also attributed a moderately high importance value. Yet the effect of sex is likely not great on this issue as female participants listed problematic burning second in frequency with a similar importance value than that of male subsample. The results from Chapter 4 mention that burning remains a common practice where at least weekly collection is available, with 28 percent of all valley households burning some of their garbage. Of these participants that burned some of their garbage, almost one-third of them listed that garbage burning was a problem as an issue during the participatory risk map- ping exercise, indicating its necessity or practicality in spite of regulations and social consensus against the practice where municipal service exists. The valley subsample listed burning as its Figure 6.3. Graph or “map” of issues by incidence (x-axis) and importance (y-axis) divided by valley and sierra households. 234 235 Table 6.2. Incidence and importance by issue from all participants, and the valley and sierra subsamples. Only issues with 0.02 (2%) incidence are listed. Data from 2011 fieldwork. Table 6.3. Incidence and importance by issue from all participants, and the cabecera and rest of valley subsam- ples. Only issues with 0.02 (2%) incidence are listed. Data from 2011 fieldwork. 236 Table 6.4. Incidence and importance by issue from all participants, and the centro and colonia subsamples from the cabecera municipal. Only issues with 0.02 (2%) incidence are listed. Data from 2011 fieldwork. 237 Table 6.5. Incidence and importance by issue from all participants, and the men and women subsamples. Only issues with 0.02 (2%) incidence are listed. Data from 2011 fieldwork. 238 most frequently noted issue but not comparably important concern. To many valley residents, most of whom have regular weekly garbage pickup, burning is perceived as an unnecessary and dirty method of garbage disposal that also poses the risk of a bush fire that can damage or mar private property and public space. Both the cabecera and the rest of the valley settlements men- tioned burning frequently. The cabecera has a higher incidence in frequency but lower impor- tance index than the rest of the valley settlements. The observably stricter small town governance through usos and costumbres as well as a sense of greater neighbor surveillance in the smaller valley settlements may be attributable to the higher importance value there. In towns such as San Rafael, mentioned in Chapter 2, the importance placed on appropriate disposal methods and general orderliness, is made manifest in these data. Yet the high frequency of problem burning as an issue here and across the valley is likely an indicator of the normative reliance on the garbage collection service in lieu of other methods. In other words, the arrival of the garbage collection Figure 6.4. Graph or “map” of issues by incidence (x-axis) and importance (y-axis) divided by cabecera households and households from other valley settlements. 239 model of garbage management, mentioned in Chapter 3, can turn a previously ubiquitous prac- tice such as burning into a maligned but still extant method of garbage disposal. For households who lack municipal collection, controlled burning of garbage (juntar y quemar ) is considered a positive management practice, indicating a well-run household with notable incidence and importance value. This issue was exclusive to the sierra subsample of par- ticipants. In contrast to the negative perspective on burning garbage, this viewpoint sees garbage burning as a practical and inexpensive solution to ridding one’s household of garbage. It is the antithesis to the perceived negative practices of holding on to garbage or dumping it outside the home. State public health workers in village clinics at times promote controlled, low-intensity burning of garbage in lined pits in lieu of keeping garbage around the house. This practice is also encouraged by SEDESOL agents in the pláticas seminars that are a requisite for receiving federal social welfare payments. Before municipal collection service was present in the valley settlements, the preferred method of removing inorganic, non-decomposing garbage from the Figure 6.5. Graph or “map” of issues by incidence (x-axis) and importance (y-axis) divided by centro households and households from other colonias in the cabecera municipal. 240 household was burning it in small cubic meter pits. Some of these pits were lined with stone or brick. In a positive way, fire sanitizes potentially infectious material such as used toilet paper by turning into biologically neutral ash. It also reduces waste material such as plastic bottles that is high in volume to a mere fraction of its former size. Supporting the importance of a pro-burning discourse is the notable incidence and importance of the issue of keeping a garbage-free home. Relatedly, the practice of collecting and organizing garbage for disposal had a similar incidence and importance values. Admittedly, participants who burn garbage note the often noxious smells and acknowl- edge the potentially harmful chemical by-products of burned plastics, but given the almost non-existence of plastic recycling service and no municipal collection, most participants view a garbage filled home as a bigger risk. While exact figures do not exist, most sierra homes likely produce a smaller fraction of the 1 kg produced per household per day in the cabecera. Through interviews and participant observation, we observed that most people typically only burn toilet Figure 6.6. Graph or “map” of issues by incidence (x-axis) and importance (y-axis) divided by women and men. 241 paper and plastic bottles that cannot be repurposed. In spite of this, burning as a problem fea - tured an almost similar incidence value but a lower importance value than that of positive burn- ing ( juntar y quemar ). The high incidence and remarkable importance of negatively perceived burning highlights the presence of a more mainstream discourse of contamination of soil and air by garbage burning. It also importantly notes how garbage collection and sanctioned dumping or landfilling is perceived as the preferred management practice of garbage. Lastly, there is the notion of people saying that burning is an important garbage-related issue and whether or not they really believe that. While I contend that the data accurately reflect the perceptions of the participants, there can be honest incongruity with believing something is important but ultimately ignoring it en lieu of circumstantial constraints. From my own point of view, I believe climate change resulting from global society’s excessive reliance on fossil fuel- based energy is a key environmental issue, however, I drive my car most places and enjoy bask- ing in the energy-intensive artificial cool and warmth provided by my HVAC unit. In a similar sense, someone in the study area who burns their garbage knowing its effects can still list it as an important environmental issue. Dumping Just behind the negative perception of burning in the incidence index was dumping, which when considering that other forms of dumping (dumping in barrancas, dumping in the woods…) are counted as distinct issues unto themselves, generally makes it the most frequently noted issue. Dumping also carries a higher importance index value than that of burning, illustrat - ing that those mentioning it collectively hold it as a more pressing issue. The valley settlements outside of the cabecera had a higher incidence and higher importance values than that of the cabecera. The more land-based economy-dependent settlements outside of the cabecera may perceive dumping as a greater threat to their livelihoods than the more service-oriented cabecera. They are also perhaps in greater proximity to the larger tiraderos clandestinos in the valley. Lastly, dumping is an outright defiance to the stricter environmental governance associated with usos y costumbres and small town life in general. In the cabecera, dumping is mentioned with almost the same incidence but with higher importance in the centro than the colonias. This is likely due to the greater threat to civic order in the more urban, built up centro than the arguably more dumping-prone but more open urban landscapes of the colonias. Like burning, dumping 242 was perhaps a more common practice prior to municipal collection service (Hayden and Can - non 1983; Deal 1985), but today has been rendered illegal and a much maligned practice. Yet, burning destroys, or rather converts it to ash, the garbage itself removing it from future manage- ment, whereas dumping simply places the burden of final disposal on someone else or simply just persists, visually and physically contaminating the surrounding area. This observation could help explain dumping’s higher importance value and higher frequency when considered a single, composite issue. Improper dumping of garbage was listed as the most frequent concern for participants from the eight sierra communities we sampled. It also registered high in terms of its importance value. This importance score is especially telling due to this issue’s high incidence. Land-based economic activity such as smallholder farming and small-scale forestry and ranching is the pri- mary economic activity in the sierra communities. Dumped garbage can temporarily minimize the productivity of the relatively finite landholdings of sierra communities where livelihoods are more dependent on the intensive use of land for food staple and cash crop or livestock produc- tion. Like in the sierra, dumping is a major concern in the valley subsample as well, where the not-too-uncommon practice to collect one’s garbage until a sufficient quantity merits a trip to a tiradero clandestino or other hidden, neglected space like a roadside or barranca3 to dump household garbage. Dumping was the most frequently listed issue among women participants, with a substantially high importance value among them. It was substantially less frequently men - tioned but ranked slightly more important by male participants. Related to dumping is the issue of street littering. The distinction between the two was made explicit by participants in addition to my own assessment (Chapters 3 and 4). Dumping is a form of mass littering and littering is individual piece dumping. Additionally, dumping is an active form of garbage management, however contrary to legal and moral norms, where litter is more passive, immediate, and of the moment. Across the study area, street littering had the fourth highest incidence and notable importance. This issue was reported similar between the valley and sierra, with the valley having marginally greater relative incidence in the valley than the sierra but slightly lower importance. More important spatial differences in the incidence and importance of street littering are apparent. The more built-up urban landscape of the cabecera has higher incidence and importance than those of the rest of the valley’s settlements. While there is no empirical evidence of greater litter presence in the cabecera, anecdotally one can eas- 243 ily assess greater litter abundance in the more dynamic town landscape of Coxcatlán compared to other valley settlements. Furthermore, the centro of the cabecera has greater incidence and importance than that of the aggregated peripheral colonias, suggesting a similar relationship with greater public activity and litter abundance. The larger town of Calipan even had higher relative incidence than other, smaller valley settlements like Tilapa, San Rafael, and Pueblo Nuevo. One of the most obvious differences of reporting street littering is between women and men, with higher incidence among women than men. Some studies have shown a link between women and reported responsible environmental behavior (McStay and Dunlap 1983; Ojedokun and Balodun 2011; Ojedokun 2011) a few have correlated gender among other demographics with anti-litter- ing perspectives (Arafat et al 2007; Al-Khatib et al. 2009). Public cleanliness Public cleanliness or aseo público as it is also known was an often noted issue, with 21 percent of participants mentioning it. There is very little difference regarding its frequency based on the the previously noted spatial distinctions This issue, as described by participants, encap- sulates general notions of public order regarding dirt, unclean things, and garbage. Aseo público is seen as a responsibility that both citizens and the town or municipal government must uphold. Individual responsibilities involve keeping areas around private property clean, ordered, and free of litter as well as not littering public spaces. Local and municipal authorities are thought to be responsible for some public cleanup, enforcing littering and property maintenance ordinances, and also providing public refuse receptacles. Public cleanliness has the highest importance value and the second highest incidence of the issues noted by the sierra subsample. This issue had a similar incidence value in the valley subsample but a much lower importance value. While numerous indeterminable factors contribute to public cleanliness’ high incidence and importance as an issue in the sierra, local, sub-municipio forms of governance help to contextualize it. The municipal government is the formal legal authority in the municipio, yet communities that are relatively small, socially well-organized, and agriculturally-oriented maintain and enforce local ordinances typically referred to as usos y costumbres (customary laws). These ordinances, while not formally recognized in the State of Puebla, usually govern practical matters such as sanita- tion, infrastructure maintenance, the hosting of public events, and informal policing of civic order (Eisenstadt 2004; 2007). Under these rules, many towns levy small fines on those who 244 are caught dumping garbage, littering, or fail to maintain standards of order and cleanliness on their property. Given the small size and close-knit nature of many of these communities, self- policing is a strong factor in maintaining what is observably a much less-littered and ordered sierra townscape. In the smaller valley settlements of Pueblo Nuevo and especially in San Rafael, usos y costumbres regarding public cleanliness are strictly enforced. In San Rafael, mandatory weekly sweeping is enforced by small fines and garbage cannot be stored. Appointed town sani- tation committee members inspect houses and also patrol for litter. Besides cleanliness in public spaces, participants from a number of sierra communities noted the importance of aseo del hogar or home cleanliness. The less-served, more independent-by-default serrano households from Cobatepec, Xacalco, Ocotlamanic, and Tepeyoloc understand the need to take a more willed approach to garbage management to keep their private spaces clean. The campaigns of public health officials and the social welfare program Oportunidades also bear their mark regarding the importance of a clean home in the sierra. Other issues Issues such as street littering and a lack of public service also maintained relatively high incidence frequencies and notable importance index values. Street littering’s potential to visually contaminate and degrade the perceived order of a place (Pandey 1990) was usually given as the reason for its importance. Considering that there is no collection service in the sierra communi- ties of Coxcatlán municipio, lack of service only had an incidence value there but had a fairly high importance value given by those who noted it. In the valley, where nearly universal collec - tion service exists, lack of service was a prominent issue and some participants mentioned that the schedule and routes were inconsistent or that their street or block did not receive service with any notable frequency. Interestingly, there are no apparent differences in incidence and impor - tance between the cabecera and the rest of the valley settlements. However, within the cabecera, the less-served colonias note the lack of service far more frequently than the centro. Recycling was also notable for its incidence but not its importance. The remarkable but not high incidence indicates the recognition of recycling as a pro-environmental practice but one that is less important or trivial in terms of overall goals of minimizing waste. Some participants noted that recycling was not worth the effort and others mentioned that they believed officials with the municipio pocketed the receipts. There is some spatial or demographic variation for the 245 incidence and importance values on recycling across the study population. Colonia Centro of the cabecera, where the recycling truck passes by frequently and has relatively high participation5 , had remarkably high incidence but only moderate-to-low importance. The precursor to recycling and other more pro-environmental garbage practices, the act of separating garbage was of similar incidence to recycling but had greater assumed importance. Surprisingly, in the sierra, where separation is less rewarded by the state by the provision of recycling service, separation has slightly higher incidence but less importance than those of the valley settlements. The need to separate the compostable and reusable from the truly desechable or disposable is the responsibil- ity of each serrano household as garbage collection is not available there. Campaigns encourag- ing participation in the relatively new (2009) recycling program perhaps add to separation’s high importance in the valley. Across the study area, the desire for more public garbage receptacles or barriles (barrels) was a commonly listed issue due to the perception of their presence being the only real behavior- al solution to litter. Making not-littering more convenient was the intended effect of the barrels, which has an empirically-derived positive effect of litter reduction (Finnie 1973; Huffman 1995; DeKort et al. 2008). The perception of greater governmental rather than communal or individual responsibility for litter and garbage management also likely influences its higher frequency and importance among valley residents than serranos who contend with quasi-governmental commu- nity statutes like usos y costumbres . Water is a precious resource in the semi-arid Tehuacán Valley as well as its adjacent si - erra. Water contamination by garbage in arroyos, wells, irrigation canals, and other small bodies of water was a more or less frequent concern of relatively high importance. In Calipan, a sugar - cane-dependent and water-abundant community, the issue of garbage blocking irrigation canals and contaminating irrigated lands had a notable incidence and importance. Outside of the survey, a resident of Pueblo Nuevo noted how the extensive irrigation canal system of the entire val - ley can act as a conduit for waterborne garbage, with more populated and perceivably “dirtier” upstream communities essentially dumping on smaller, “more ordered” downstream communi - ties. Lastly, the association of vermin and garbage and the public health risk they imply were also notable, with mosquito incubation in garbage and stray dog subsistence having comparably low but remarkable incidences and importance values. 246 Chi-square and logistic regression results The previous results give a narrative comparison of the incidence and importance of the more frequently noted issues across the entire study population and by preconceived spatial and demographic strata. To show statistical relationships between these subgroups and particular is- sues, I applied a simple but effective non-parametric Pearson’s Chi-Square statistic to test wheth- er or not the incidence of particular issues was contingent and statistically significant upon place or sex. Pearson’s Chi-Square statistic can be used to detect a significant relationship between two variables. In this case, I am testing to see if there is a relationship between being from the valley or sierra, the cabecera or another valley settlement, the centro or the colonia of the cabecera, or sex for listing a particular issue. In this case, I selected nine issues for their frequency as well as relevance. These issues along with their relationships by place or sex variables are listed in Table 6.4. For the distinction between participants from the valley and the sierra, there were some notably significant relationships for particular issues. There was a significant relationship (p < 0.05) between the issue of inadequate service provision and the valley-sierra spatial bifurcation, with remarkably greater incidence in the valley than in the sierra. Rather than showing a marked desire for service by sierra households, which while not apparent in this study still exists, the results show that reliance on garbage collection brings up notions of adequacy by frequency of collection. Similar to collection service, there was a significant relationship (p < 0.05) of the valley-sierra division and the adequate provision of public garbage receptacles with greater in- cidence in the valley than the sierra. Again, this result infers a greater responsibility attributed to extra-household actors by valley residents than their neighbors in the sierra. The issue of burning garbage in the home as a pro-sanitation practice also showed a significant (p < 0.05) relation - ship with this spatial division with it mentioned only by sierra households. Dumping ( p = 0.06) and water contamination from garbage ( p = 0.053) just missed the alpha-statistic (0.05) but have higher incidences in the sierra and valley respectively. Other spatial divisions showed fewer significant relationships. Only the provision of public receptacles ( p = 0.053) approached any significant relationship for the cabecera-other settlements distinction, with the visibly more litter-prone cabecera having higher incidence than the combined sample of Calipan, Tilapa, Pueblo Nuevo, and San Rafael. This same issue was farther from the 0.05 alpha-statistic with the division of the centro and peripheral colonias in 247 the cabecera municipal ( p < 0.10). At the same spatial scale, there was a significant (p < 0.05) relationship between recycling incidence with the centro of the cabecera having higher incidence than the peripheral colonias. The gender distinction showed significant relationships, with street littering (p < 0.05) and dumping ( p < 0.05) with greater incidence of these issues being mentioned by women. Research has shown that women are less likely to litter or at least admit that they litter (Cialdini, Reno, and Kallgren 1990; Meeker 1997). In the study area and across Mexico, anti-litter cam- paigns generally involve women. Additionally, garbage management and the maintenance of sanitation are generally perceived to be the domain of women. Thus, a practice that challenges the control of garbage and general order is expectedly more often mentioned by women. There were only two statistically significant logistic regression models for educational attainment with regard to the ten issues with the highest incidence. No statistically significant models were present between any of the ten issues and age. It appears generational differences Table 6.6. Results from the binary Chi-Square tests. Data, 2011 Fieldwork. 248 cannot so easily be drawn out of the data. Results do perhaps show the effect of education attain- ment on pro-environmental behavior and the garbage collection regime. Only the incidence of the issues of public barrels and service provision with educational attainment showed a statisti- cally significant result representing a positive prediction with increasing educational attainment. Research has shown the effect of environmental education campaigns with increasing effective- ness of public waste receptacles (Geller et al. 1980; De Kort et al. 2008; Schultz et al. 2013). Also, increasing formal education as a predictor of the incidence of the service provision issue may indicate that formal education has an effect where state-controlled, non-traditional, “mod - ern” garbage management such as collection is preferable. Conversely, this relationship may indicate a perception of “traditional” methods as unsanitary or environmentally unsustainable. In Mexico and especially Puebla, environmental education at a number of levels (primary, second- ary, and beyond) has been effective in promoting pro-environmental behavior (Guevara Martínez and Fernández Crispín 2009). Conceptualizations of garbage Besides simply indicating the heterogeneity of perception of garbage-related issues, the results also point to different conceptualizations of waste in society. Dumping, burning, public cleanliness, and littering are big issues all over the municipio. So are inadequate service, insuf- ficient public garbage receptacles, and the maintenance of a clean home that is free of garbage. All of these issues in addition to less-frequently noted issues like recycling, separation, and the lack of rule enforcement issues are united by one common theme. They all concern garbage management, or its absence in the cases of dumping and littering. From the data, one can see that Coxcatecos typically perceive garbage issues around the idea of management or as what Moore (2012) calls “waste as manageable object”. This conceptualization of garbage centers it as the subject of varying systems of governance, both formal and informal. Garbage in this sense is everyday, expected out-of-place matter (Douglas 1966); it is something apart from society and needs to be managed by it to maintain order (Moore 2012: 786). How this is optimally done and by whom is indicated by the data, which suggest a desire for a growing role for more formal waste service. The management regime evident by this management-centric conceptualization of gar - bage appears more formal than informal. On one end, the management options available in the 249 home like burning or even dumping are seen as bad practices by the study population. While the practice persists, both of necessity and choice, its assessment is mostly negative. The relatively high incidence and importance given to the adequate provision of waste receptacles also suggests a larger role given to the state in managing garbage. The notable incidence of recycling and sepa- ration also indicate a shift to formal, outside-the-home garbage management. Dumping and burning have a negative connotation. They are mostly seen as poor if ef - fective management solutions for garbage. This suggestion may appear complicated slightly by the notable incidence and importance of collecting and burning as a good practice by some sierra participants. Yet this more controlled burning is promoted by the state and agreed upon by resi - dents as a lesser of two evils alternative to keeping garbage around the house or dumping it else- where. Additionally, the residents who burn out of necessity well understand the health as well as proprietary risks of an unkempt house over burned garbage. Without any ostensible alternative, burning, while still seen as a risky practice, indicates a positive action in sierra homes. Burning and dumping are also common as evidenced in the garbage census, both in burn - ing’s frequent mention and dumping’s highly conspicuous absence. Yet when considering the ethnographic and ethnohistorical record, dumping and burning were also tried-and-true, ancient methods of getting rid of garbage (Hayden and Cannon 1983, Deal 1985; Arnold 1990; Killion 1990; Sutro 1991; Hirth 1993). While it is impossible to assess how pre-modern Coxcatecos perceived burning and dumping, one can assume that it did not have the same toxic connotation, indicating a shift in how garbage is perceived locally. Thus there is an apparent separation be- tween the practice of burning and its perceived potential effects. The discursive, if not practical, maligning of dumping and burning stems largely from its conceptualization as a health and environmental risk, which is yet another way garbage is per- ceived. The negative view of burning and dumping can be framed by what Beck (1992, 1999) and Giddens (1990) call the “risk society” theory. Risk society theory, as applied to the environ- ment, can be summarized as the perceptual replacement of agency toward human and technologi - cal causes of risk as opposed to natural causes in modern societies (Wakefield and Elliott 2000: 1140; Moore 2012: 789). I must note that this is not a recent shift, as most Coxcatecos have known for a long time that burning and dumping inorganic waste releases harmful contaminants and toxic substances. What is important is that the juxtaposition of the ubiquity and persistence of burning and dumping practices with their perception as problems, serves to indicate a lack of 250 control or vacuum of governance regarding garbage management in Coxcatlán. They indicate a lack of governance from formal government bodies as well as individual citizens to protect themselves from environmental degradation and health risks. Moore (2012: 789) claims that the waste as risk conceptualization challenges the existing notions of formal governance. In this sense, burning as an issue both serves as a proxy indicator for the dissatisfaction with the lack of universal service as well as individual responsibility to not do it where collection service is avail- able. It also is a call out to the inability of a number of actors to prevent the perceived environ - mental damage and negative health risks from burning. Imprecise general issues such as water contamination and general environmental degradation also fit these definitions of waste as hazard and risk. CONCLUSION The results of this study indicate a number of garbage-related issues that are perceived to affect residents across the entire municipio. Burning garbage has the potential to contaminate both air and soil and poses a damage risk to property and personal safety. Despite its long tenure and previous acceptance as a practice for eliminating waste, burning’s mostly negative connota - tion is relatively recent and is likely due to the increase in the inorganic and chemically toxic components of garbage. The change in burning’s perception is also due to the recent availability of garbage collection service in the study area. The lower incidence in the sierra where no service exists supports this claim. Dumping’s previous ubiquity and relative acceptance and current ma - ligning fits a similar change. Public cleanliness, seen as both the responsibility of local authori- ties and private citizens, is an important issue for households across the municipio. By identifying these issues as well as others mentioned above, the participatory risk map- ping methodology is useful in identifying how garbage or any other issue affects a target popula- tion as they see it. This method gives a more nuanced view of how people perceive garbage as an issue in a place. That is not to say that these issues are constructed detached from environmental governance, policy, access to infrastructure, or extant traditions. Rather, participatory risk map- ping’s descriptive power combined with statistical inference can contextualize the perception of an issue with particular populations and identities. For human geographers and allied research - ers this method allows one to see the heterogeneity and breadth of issues according to place and demographics. 251 Heterogeneity by place and by demographic variable is most evident between the partici - pants from the sierra communities and valley settlements where unequal access to infrastructure, diverse garbage management practices, varying approaches to governance, and different ethnic and traditional backgrounds contribute to differing perceptions on garbage-related issues. Dif- ferent management regimes between the valley, where there is collection, and the sierra, where there is not, also likely contribute to different perceptions of garbage and garbage-related issues. Lastly, since the issues are given by the study participants themselves, participatory risk map- ping points to how residents perceive of wastes themselves, as managed everyday matter, as a hazard, as risk, and as a challenge to control. These conceptualizations of garbage all arise out of the issue of littering. Littering and the related practice of dumping also have high incidence and importance across the municipio. Litter represents the nexus of the material and discursive properties of garbage in Coxcatlán. In the following chapter that synthesizes this dissertation’s focus on the material, practical, and perceptual aspects of garbage, litter and littering is examined in greater detail. Endnotes 1. The use of the acronym “PRM” in reference to Participatory Risk Mapping is not to be confused with Participa - tory Research Mapping developed by Herlihy (2003), Herlihy and Leake (1997), and Herlihy and Knapp (2003). The former is the method used in this chapter but I insist that its “mapping” qualifier is a misnomer as it largely produces graphs rather than formal maps, which are the result of the latter. 2. The incidence and importance factors used in the text of the results are presented parenthetically with “inc.” representing incidence and “imp.” representing importance. 3. “Dumping in the barrancas” was a notable issue with 0.03 incidence and 0.83 incidence. 4. By comparison the other colonias of the cabecera municipal only have about 67 percent recycling participa- tion. 252 CHAPTER 7: Roadside littering in Coxcatlán INTRODUCTION This chapter examines the issue of roadside litter in Coxcatlán municipio. This study is a departure of sorts from the larger theme of this dissertation, which focuses on the manage- ment of garbage and waste. Litter is waste that is unmanaged. It is garbage outside the purview of control and its analysis here expands the scope of the dissertation beyond managed garbage to unmanaged garbage as well. After the first half of my stay in Coxcatlán, I had observed that many of the problems about garbage mentioned by residents were about litter. It was not difficult to understand how many people had arrived to that conclusion. While Coxcatlán is a visually beautiful place and not some ruined environmental dystopia, litter is visible most everywhere. It appears especially abundant along the roadsides and water courses as well as in vacant lots. But besides concluding that there was the imprecise quantity of “a lot” of litter, I began to ask myself if Coxcatlán were any more littered than anywhere else I knew, notably the United States. This dissertation contends that garbage is both a material and perceptual environmental issue, so it maintains that unmanaged garbage, or litter, embodies a similar framing as well. In order to test the assumption that there may not be as much litter in the municipio as residents and I believed, an estimate of the material quantity of litter needed to be established. In June 2011, I began a study to quantify and characterize the litter alongside the highways there. The approach taken here to study litter diverges from the previous look at garbage man- agement practices (Chapters 3, 4, and 5) or its framing as a wide set of environmental issues (Chapters 5 and 6) by measuring the literal matter that constitutes litter in the study area by its weight, abundance, and its class. In other words, instead of gauging how people see litter as an environmental issue or as a waste removal practice, this study seeks to characterize and quan - tify litter as physical matter. Currently, there are no available datasets and only one systematic study on littering in Mexico (Muñoz-Cadena et al. 2008), much less one in a rural-but-urbanizing municipio like Coxcatlán. This study can provide an empirical backbone to vague assumptions that there is a great deal of litter that is often thought to be mostly plastic. The four main research questions of this study seek to provide an empirical base of how much and of what type of litter exists in order to highlight common assumptions and perceptions regarding litter as a problem. They are: 253 1) What is the quantity (abundance) and weight (mass) of litter by class (plastic, glass, organic, etc.) in the study area? 2) How do these compare to data from similar studies done elsewhere? 3) Is there any spatial pattern to littering in the study area? 4) Is there any temporal or seasonal pattern to littering or litter presence? These questions point to the size and character of litter as well as illustrate some of the underlying contextual factors that are associated with its presence. They also complement the larger scope of the dissertation by engaging the results in previous chapters. Littering or dumping was a widely mentioned issue in the municipio during the PRM study. Litter is obvious, it makes an impression on residents, and people talk about it. Litter is an issue that is generally outside of official management by the government and is beyond the influence of individual households. Litter is outside of the household and not in the dump. It occupies public space. It is both the most obvious and neglected garbage issue in the study area. The following chapter first concep- tually maps out what litter is, then identifies the conundrums in measuring it. What follows is a description of the systematic approach and methods of the study followed by results, discussion, with a summary and conclusion. BACKGROUND Litter and Littered Landscapes Litter is a big deal in Coxcatlán. It is nearly synonymous with the garbage problem there. In the previous chapter, the results from the participatory risk mapping exercise showed that lit - tering is one of the most if not the most important garbage-related issue in the municipio. Yet the previous risk mapping data did not provide the only mention of why littering is such a notable issue in Coxcatlán. In casual discussions regarding my research, the general topic of garbage usually led to some mention of the ever presence of littering or to a self-deprecating nod to how Coxcatecos are generally dirty people with a poor environmental ethic1, to which I do not give any validity, at least not any more than any other group of people. Many of these conversations typically began with more opinion-neutral themes regarding household or municipal manage- ment but when garbage as a universal problem was discussed, the focus often shifted to littering. In other words, most people tend to think that there could be more garbage pickup, better recy - cling, or a sanitary landfill, but litter and dumping was nearly always mentioned by Coxcatecos 254 Figure 7.1. Litter dumped and accumulated alongside the highway in Coxcatlán. Photo by A. Hilburn. September 2011. Figure 7.2. Litter in an irrigation canal in Calipan. Photo by A. Hilburn, October 2011 255 as a concrete problem. Litter appeared as something almost distinct from garbage. Garbage as waste is unavoidable and is the result of daily consumption. Litter appeared to be something dif- ferent. If garbage is matter-out-of-place (Douglas 1966), then litter is garbage-out-of-place or matter twice removed from order. If litter follows this definition as garbage-out-of-place, it is largely an ideal, a concept related to one’s own idea of how things should be ordered rather than how they are ordered because litter is nearly everywhere and unavoidable (Schultz and Stein 2009: 1). The act of littering, while not as common as dealing with one’s garbage in a socially prescribed manner, is also a part of daily life. It can be passive, the product of leaving a wrapper on a bench or a napkin being blown away by the wind (Sibley and Liu 2003). Littering is also a willed act or a behavior. People place garbage where it is believed not to belong for numerous reasons. There is a considerable literature, mainly from psychology and sociology, which seeks to understand the underpinnings of littering behavior (Powers et al. 1973; Geller et al. 1977; Cialdini and Baumann 1981; Sibley and Liu 2003; de Kort 2008). Yet, in both public and academic discourse, litter and littering are perceived to be a problem. Why? Extant litter is an obvious indicator of environmental contamination (Finnie 1973). It is also nearly everywhere (Figure 7.1). Outside of extreme cases, air, water, and soil pollution are less visible and depend on specialized monitors to gauge their intensity whereas litter is there to be seen by all. The problem of litter and littering is global and especially so in economically marginal areas like rural Mexico, where the limited budgets of municipios do not allow for its extensive clean-up, adequate public waste receptacles, or expansive anti-litter campaigns. Gar - bage collection and the educational campaigns regarding appropriate disposal have only existed in the study area for fewer than twenty years, which is a relatively short period of time regarding the changing of entrenched behaviors and beliefs. Litter is an assemblage of objects. While litter is largely the result of human production and consumption, it has its own agency (Latour 1994). Litter’s physical and chemical properties allow it to contaminate soil, air, and water. Intensive littering in and around human populations can pose a chemical hazard (Nduka et al. 2006), as well as a biological hazard as disease vectors such as rodents, flies, and mosquitos scavenge off decomposing garbage (Sridhar and Oedjiran 1983). Litter supports the subsistence of stray dogs, which in a sense are a living form of litter, that pose a risk to citizens by attacking them, as well as being a disease vector (Fielding 2008; 256 Klimpel et al. 2010). Broken glass and other sharp littered objects can be a considerable injury risk for playing children and pedestrians (Al-Khatib et al. 2007; Arafat et al. 2006). Car accidents have even been attributed to litter in the roadway (Forbes 2003). Excessive litter can contami - nate waterways and clog drainage systems. In agricultural zones, litter can sully fields, pastures, and irrigation systems, forcing producers to divert time and labor to its clean up (Figure 7.2). Fire is a not-insignificant hazard in the long dry season. Litter provides both additional fuel load as well as the ignition source when cigarette butts are tossed from vehicles. Aside from effects in both the human and environmental spheres of influence through its numerous physical proper- ties, litter and garbage-out-of-place can be powerful agents in political and social discourse and can challenge notions of governance and control (Moore 2008). From an aesthetic and social standpoint, littered landscapes show a lack of environmental education, stewardship, and general order. Litter is garbage, which to most people is aesthetically repulsive (Pandey 1990). Litter’s unsightliness transfers its properties to the places it inhabits, suggesting disorder and wasted space. The aesthetic blight and visual degradation that litter can create, the so-called “broken window” effect, has been correlated to increased crime rates in urban areas (Ligget et al. 2001; Kelling and Coles 1996). While this last factor likely does not apply to increased crime in a rural municipio like Coxcatlán, the presence of excessive litter may encourage further littering or other forms of environmental degradation due its appearance as wasted space. Litter reflects poorly upon the local population, who presumably but not particularly, are those littering. This also applies to the government that does nothing to prevent or mitigate the litter. The enforcement of laws against littering is almost impossible given the meager resources of most municipal governments. Signs are posted to prevent littering (Figure 7.3) and social welfare community service (Figure 7.4) picks some of it up on clean up patrols, but litter persists and people’s perception of it as a problem persists as well. Littering may not be the most impor - tant garbage-related issue from a management perspective compared to increased service provi- sion or the lack of a sanitary landfill, but it is one of the most obvious issues related to garbage to Coxcatlán’s residents. Garbage can be managed; litter cannot be taken care of so easily. Part of this difficulty places litter to something beyond control. The enormity of the issue can be difficult to gauge because of its wide distribution as well as its sheer amount. In municipios like Coxcatlán, only anecdotal estimates like “ tanta basura” 257 Figure 7.3. Litter cleanup by an Oportunidades -organized crew in the cabecera municipal of Zoquitlán, Puebla. Photo by A. Hilburn, June 2011. Figure 7.4. Posting notifying the illegality of littering along the Barranca Soyolapa in the cabecera municipal. Photo by A. Hilburn, August 2011 258 exist with regard to the litter problem. No systematic approach has yet been taken to estimate the size and class of the litter in a place like Coxcatlán. There is very little to no data on littering in Mexico outside of a few beach surveys (Silva-Iñiguez and Foster 2003) and neighborhood litter sampling in the capital (Muñoz-Cadena et al. 2009). For this reason, I conducted the following study to understand a few important aspects of litter in the municipio. Research Approaches There are a number of ways to research the litter problem. First, it is necessary to distin - guish the act of littering or littering as a behavior from the real, physical matter of litter itself and its agency in the environment. Both are related as the latter is the result of the former and litter gives some indication of littering behavior through the spatial heterogeneity of its deposition. In other words, where people throw what constitutes behavior. Yet the way researchers approach the two pjenomenaare different in that social experiments are the preferred method for under- standing littering behavior and quantification and classification methods taken from the natural sciences are used to gauge litter from a management perspective. The former populate psychol- ogy journals with the latter published in waste management journals and in gray literature for government-sponsored studies. Littering behavior can be difficult to study empirically. Litter as matter is notably less complicated. Littering is typically done on the sly. Besides being against the law, the act of lit - tering is attached to notions of shame, a lack of respect for the environment and other people’s property, laziness, as well as general carelessness (Heywood 2002). It carries a negative stigma. This makes littering behavior arguably more difficult but not impossible to approach from an empirical research perspective. Interviewees or questionnaire respondents are not likely to admit to littering (Rodrigues Santos 2005) and even if they did the researcher could be burdened by ethical concerns over use of the data and the privacy and protection of their subjects. Also com- pounding methodological approaches is that recording acts of littering in real time is highly im- practical if not impossible. Littering is a brief act; it only takes a second to throw a piece of gar - bage and even less to drop it. Dumping requires a little planning but it still can be done quickly. Monitoring littering over a larger area, such as a municipio, as it happens requires constant and omnipresent surveillance as well as an eye for detail, which combined together in the real world do not exist (for good reason as well). 259 Littering laws’ unenforceability likely points to one reason why littering persists. Previ - ous studies on littering-as-behavior have been social experiments to observe and gauge why people littered and what type of environments were conducive to it (Finney 1973; Cialdini et al. 1990, 1991; de Kort 2008). Many of these studies, though, took place in dynamic areas such as university quadrangles and polling places, not on small town street corners in Mexico. In the fol - lowing study, this was not possible for many reasons. The observation of littering, when possible, only infers the behavior behind why litter exists but tells nothing of the matter itself that directly causes the environmental contamination. It provides little insight into the size and qualities of the litter problem. Litter-as-matter, on the other hand, is comparably unambiguous and obvious, and can be collected in a manner similar to a biologist taking field samples. Litter can be measured after it has occurred, but littering cannot. Yet, litter exists in active, dynamic, and managed space. People clean public space, water carries litter downhill and deposits it as it goes, winds blow litter long distances, and even grazing animals eat it. Site selection thus becomes very important in a litter study. Existing studies have measured littering in urban environments (Seco Pon and Becheruchi 2012; Moriwaki et al. 2009; Muñoz-Cadena et al. 2009) and managed recreation areas (Ivar do Sul 2009; Rodrigues-Santos et al. 2005) but these studies sampled highly regular environments where street cleaning was regular and consistent or where very little if any cleaning took place in a highly controlled environment such as a beach. In Coxcatlán, daily litter cleanup only ex - ists around parks and in front of certain residences at irregular times. Stray dogs carry off select pieces of trash and redeposit others elsewhere. Small, intermittent street markets and occasional block parties also prevent any sort of litter depositional regularity and consistency there. Coxcat- lán town’s fairly significant slope allows for high rainfall events to “clean” the streets and strong winds can negate a civic-minded resident’s best tidying effort. From a fieldwork perspective, the problem exists of identifying spaces of littering in a small, highly connected town and the effect of how such information may reflect upon specific blocks or neighborhoods as being dirty or un- kempt. Litter and cleanliness are political. In short, it can show bad manners to conduct a study in such an environment. Highways, though, are a form of common property. Furthermore, they are common property that occupies local space. Highways are owned by everyone as well as no one. They are a linear commons that span large distances giving access to a wide spectrum of people. Besides 260 the road space itself, a highway’s domain spills over into buffers or easements, usually about five meters on both sides of the roadway, giving spaces for litter to accumulate and researchers to potentially collect it. In Coxcatlán, very few people live directly on the highway and very little to none of the areas adjacent to them are actively cleaned nor are the site of any activity besides littering. Highways provide ideal sites on which to accurately record and measure littering across the municipio. The systematic study of littering on roadsides goes back to the late 1970s United States with a few examples from Virginia and Kentucky (IAR 2006). By the 1990s consistent methods had been developed and a wider study site distribution existed across the United States. Litter studies are growing in number today due to their usefulness for planning and cost-estimation purposes but these roadside litter studies appear to be a global north phenomenon (Forbes 2009). In the United States, the non-profit organization Keep America Beautiful conducts its National Visible Litter Survey and Litter Cost Survey that surveys roadside litter as well as litter in other situational contexts as loading docks, storm drains, and retail areas (MSW Consultants 2009). Keep Australia Beautiful, a similarly-titled anti-litter non-profit organization down under, annu- ally updates what is one of the most comprehensive and spatially extensive studies on littering in the world in its National Litter Index whose funding comes from governmental and packag- ing industry sources (MacGregor Tan Research 2012). Most of this research was conducted by environmental engineering firms working with various states’ departments of transportation and outside of peer-review. To date, there are no studies outside of Muñoz-Cadena et al. (2009) for street or roadside litter estimation in Mexico. This chapter has two primary objectives. The first is to estimate what constitutes litter and how much of it exists along Coxcatlán’s roadsides. The rates at which litter accumulates and potentially contaminates roadsides and adjacent areas can also be estimated from this objective. The type, quantity, and generation rates will provide a glimpse into the material problem of litter as it exists in the study area. The roadsides in Coxcatlán are not all the same with regard to the amount and type of litter. One stretch may be heavily littered while another nearby might not have nearly as much. The second objective for this study is to understand the effect of location and time, be it in a settlement, near one, in a rural context, or along a particular highway, or dur- ing a particular time of the year on the class and quantity of litter. This will allow for the identi- fication of potentially at-risk areas and seasons for littering as well as infer about the spatial and 261 locational attributes of littering. SITE SELECTION AND METHODS The study area for this project is limited to the municipal boundaries of Coxcatlán muni- cipio. The site selection and collection methods used for this study are modeled after those used in the Ohio Department of Natural Resource’s 2004 Statewide Litter Survey 2 (Davey Resource Group 2004). Sites for this study were chosen quasi-randomly. Selection was done in ArcGIS 10 by converting a highway line shapefile for the entire study area into points spaced at one meter intervals. Each point was assigned an individual integer and using a random number generator, the points were selected as the nearest end point for a potential site. The coordinates for these sites were recorded and located in the field by GPS. Sites needed to fit particular physical criteria if they were to be included in the final study. Excessive slope or curviness, overgrown vegeta- tion, dispersing or accumulating drainage, insufficient right-of-way, and visible management by people would all disqualify a site3. The sampling was quasi-random in that if a preselected site was determined unsuitable and any site within the field of vision of the research team appeared suitable, then that site would be assessed for possible inclusion in the study. After considerable prior assessment of the highways in the study area, it was determined that a 50 meter by 5 me - ter study site size was ideal. This determination was based the difficulty of finding a relatively straight 50m length of highway, as well as the feasibility of precisely collecting and analyzing each site in a timely manner given the small size of the research team 4. The five meter width of each site was based upon the generally accepted size of highway right-of-ways in the study area and across Mexico. All sites except two have the standard 50 by 5 meter (250m 2) sizes (Figures 7.5 and 7.6). Selected sites were measured with standard metric measuring tapes and marked by using semi-permanent stakes or large flagged boulders. Sites were left so as to look inconspicu- ous to passersby but relatively easy to relocate when collecting litter. The selected sites vary according to their locational contexts. Some were in town, others just outside them, while others were amid agricultural fields and natural reserve areas. For the sake of simplicity, these different contexts will be referred to as “urban” and “rural” in later parts of this study. A total of 10 sites were selected for the entire duration of the study (Figure 7.7). Five sites (sites 2, 3, 5, 6, 8, 10) are considered urban and the remaining five, rural. Four sites were located on the busy Tehuacán to Huautla highway (Puebla 980) and six sites were located 262 Figure 7.5. Diagram of a typical 250 square meter site. This particular site is the Busted Tire site between Coxcat- lán and Calipan. Photo by A. Hilburn, June 2011. Figure 7.6. Looking south on the INEGI Marker site. Photo by A. Hilburn, June 2011. 263 along the smaller Zoquitlán road that begns in the cabecera and heads into the Sierra Negra Other site attributes such as date, time of day, noticeable modification, slope, and ground cover were recorded. The research team took photographs and produced a sketch drawing of each site in addition to a descriptive narrative of each site. Sites were given names instead of numbers to better establish mental associations with anecdotal evidence and field notes. Once sites were selected, they were completely cleaned of all litter down to small bits less than one centimeter in size 5 . Litter is defined here by any anthropogenic matter that appeared Figure 7.7. Map of study sites along the highways. Map by author. Data, INEGI 2010; 2011 fieldwork. 264 “out of place”. Leaves, grasses, and even purposively cut vegetation were not included. In addi - tion to the sites themselves, the immediate areas around each site were also cleaned to prevent future contamination of the site by wind, water, or animals. The litter was collected in 200 liter black polyethylene bags whose weight was recorded and subtracted from weighing. The litter collected was then weighed using a hanging field scale accurate to ten grams and recorded. Used toilet paper, diapers, and the far-too-common plastic bottles of urine were left in situ with their location and counts recorded. The litter from the initial cleanup was not classified due to a num- ber of differences between sites as well as the sheer enormity of litter collected. Also, the length of time that had elapsed since a particular site’s formation or modification could not be deter- mined and some may have experienced cleaning in relatively recent time. Some sites had been burned by low intensity roadside fires caused by cigarette butts6 thrown from cars or fires started for vegetation control. Burning can destroy or affect particular classes of litter more than others. Once all sites were cleaned of all garbage, a ten week interval was allowed for the accumulation of litter across the study area. After ten weeks, the research team returned to the study sites and collected all litter that had accumulated within each site’s boundaries. Immediately following collection, the contents were weighed for a net total after which they were classified according to predetermined classes. The classes are listed below in Table 7.1. One can expect particular classes of garbage in house - hold garbage that one would not find alongside the road. The initial cleaning experience allowed me to better understand the character and contents of roadside litter. The classes used were based on those provided by Ohio DNR study (Davey Research Group 2004) but some modification was necessary to fit particular local expressions of litter7 . Following collection, the litter was classed, counted, and weighed in the field on stable, level areas near the site to best replicate the conditions under which it would be collected by a clean-up crew. The classed litter was counted by individual piece and weighed on an electronic scale with one gram of precision. Weighing litter is useful from a management perspective as the weight of litter must be removed and transported to a final disposal site. Weight, however, should not be the sole descriptor of roadside litter as particular classes such as organic matter may have a higher average weight per individual item, but not be as ubiquitous or persistent as another class, like plastics, in the sample. Each individual piece of litter, described as garbage- out-of-place, has the potential to contaminate, whether chemically or aesthetically, its surround- 265 ings. Thus, individual counts by class were taken of the collected litter. Individual counts or abundance also give the potential to understand the assemblages of litter found at particular sites, giving an indication into the effect of location and time on the behavior of littering. One objective of this study is to identify the quantity and character of roadside litter in the study area, whose results could be extrapolated to other similar parts of Mexico. In short, it seeks to understand what and how much of what people litter on roadsides in a rural Mexican municipio. To date, there have been no such studies of roadside litter in Mexico. Muñoz-Cadena et al. (2009) examined littering in two neighborhoods in Mexico City, but the differences be - tween a densely populated neighborhood in one of the largest metropolitan areas in the world and the easement along state highways in Coxcatlán municipio (pop. 19,639) are so large as to render them distinct enough to proceed with one’s own study. This information will contribute to fram - Table 7.1. Litter classes and subclasses used in this study. 266 ing the issue of littering from an environmental management perspective. Descriptive statistics of weight and individual counts by site, collection period, and locational context will illustrate how much and what is being discarded along the roadside in Coxcatlán. Lastly, one goal of this study is to identify the effect of place and season on the quantity and type of litter present. Using the individual counts data on litter classes, comparisons were made across collection periods, and locational contexts to infer the effect of time and space on litter and littering using simple non-parametric statistical testing. LITTER COLLECTIONS Initial Clean After a week spent conducting the initial cleaning at the eleven sites selected for the study, we disposed at the municipal dump of 28 bags that we had filled to varying degrees of fullness with plastic bottles, burned bits of plastic, CDs, pieces of tire, cigarette butts, bags, corn husks, and a vast array of detritus, as well as a strangely not-insignificant amount of shoes. In total, we collected 105.6 kilograms of litter over the eleven study sites (Table 7.2). The total col - lected amounts to 88.3 kilograms of litter. Divided over the combined area of all ten sites, there 267 Table 7.2. Initial clean results by study sites. Data, A. Hilburn field work, June 2011 was 0.4 kg of litter per 10 m 2 before removing it. In everyday terms, a quantity of litter just shy of a pound (approx. 14 oz.) was present on an area about the size of a small-to-midsize home’s front porch or a typical toolshed across the aggregated study sites. This kilograms or pieces per 10m 2 was the normalized measure for litter throughout this study. Obviously, not all sites were similar with regard to the extent to which they were littered. The most intensely littered site in the study area in terms of absolute weight was the site just south of the cabecera that we jokingly titled the Biodome 8 Site due to its location within the Figure 7.8. Initial clean total weights by study site. Data, INEGI 2010; A. Hilburn field work, June 2011 268 biosphere reserve9 . In the 250m 2 of the Biodome site, we collected 17.1 kilograms of litter. This amount is especially impressive considering that it appeared to have experienced burning on one or more occasions, thus reducing the litter weight as well as eliminating some litter altogether. However, the smaller site (approx. 145m 2) located at the crucero (crossroads) of two highways at the southern end of Coxcatlán town had the highest normalized weight with 0.8kg/10m 2. The least littered site was the Diaper site 10 , at the cumbre (ridgeline) of the sierra above the town of Pala with only 3.7 kilograms of garbage and 0.1 kg/10m 2 (Figure 7.8). Between the 10 sites there was high variability of weight per site with a mean of 8.8 kilograms per site or the normalized average value of 0.4 kg/10m 2 (Table 7.2). First Collection The first collection took place 10 weeks following the initial cleanup in the third week of August, 2011. From all 10 sites, the team collected 313 pieces of litter weighing a total of 6.112 kilograms (Figure 7.9). Distributed across the summed area of all of the sites, there were 1.3 pieces and 27 grams of litter per 10m 2. Plastic dominated the assemblage, comprising the highest percentage by abundance (70.3 percent) as well as weight (38 percent) of the total ag - gregated litter collected. Plastic soda bottles, plastic bags and films, plastic cups, and disposable eating utensils were the majority of this class, inferring a fairly obvious link between eating and drinking on the go and littering along the roadside. Cardboard drink boxes, packs of cigarettes, gum packs, and pieces of paper among other cellulose-based products contributed to paper and cardboard litter being the next highest class with 10.5 percent of the abundance and 14.9 percent of the weight. Related to mobile consumption’s effects on littering, the next highest class of litter in abundance was styrofoam or unicel (7.7 percent), of which plates and cups were the predomi - nant forms. Styrofoam’s low weight per item attributed to its low relative contribution to the total weight in the sample (1.4 percent). On the other hand, glass, usually a nonreturnable Coke or beer bottle, did not figure prominently in the sample in terms of abundance, but its high weight per item ratio contributed to 16.6 percent of the sample weight from only six pieces. Scrap metal is a valuable commodity in and around Coxcatlán and is thus not typically discarded haphazardly or is scavenged quickly. Most of the twelve pieces of metal found on Coxcatlán’s roadsides dur - ing this period were from aluminum soft drink and beer cans and metal bottle caps. The major- ity of the 753 grams of organic litter was contributed by the contents of a large bag of vegetable 269 Figure 7.9. First collection abundance (inner ring) and weights (outer ring) across study sites. Chart by A. Hilburn Data, Field work, August 2011 scraps from a single roadside meal at Diaper Site 11. Organic materials were not very ubiquitous or significantly great. Bottles of urine, paper smeared with feces, diapers, batteries, and automo- tive chemicals comprised the hazardous waste encountered during the first collection which was present at half the sites. Across sites, the site with the most litter abundance and weight during the first collection was the Diaper site, with 61 pieces of litter and 1.241 kilograms of litter across its 250m 2 area due to what appeared to be a small mass dumping from a few individuals (Figure 7.10). This site also has the highest normalized values with 4.9 items and 99 grams of litter/10m 2 of area. The site was highly littered but its total weight was augmented considerably by the dumping of a few large items dumped in one instance. Nevertheless, that the Diaper Site was the most littered site was surprising due to its location in a presumably low traffic area in a wooded area along the cumbre of the sierra above the cabecera. It was also the least littered site just ten weeks prior during the initial cleanup. The Busted Tire site had 4.6 items per 10m 2 (58 items) and the smaller Crucero Site was just under that count at 4.5 items per 10m 2 but had a larger normalized weight at 152 grams/10m 2 than Busted Tire’s 72 grams/10m 2 due to some glass and oiled cardboard 270 items. The Moo Cow site, just about a kilometer north of the Crucero site had the next highest litter abundance (3.3 items/10m 2) but the normalized weight of the Shoebox site (86 grams/10m 2) was greater than its 50 grams/10m 2 due to heavy items such as a pair of shoes, a broom handle, and some thick-walled plastic bottles. The lowest abundance and weight from all ten sites was found at the isolated No Name site located along the sinuous stretch of highway below the sierra community of Pala. This site produced only eight items at a nearly negligible 26 grams. Normal - ized by area, this amounts to a mere 0.6 items and 2 grams per 10m 2. The spatial distribution of Figure 7.10 First collection normalized abundance by study site. Map by A. Hilburn. Data, INEGI 2010; A. Hilburn field work, August 2011 271 litter intensity is shown in Figure 7.9. Second Collection The second collection occurred in mid-November, 2011, ten weeks following the first collection. From the 2,281m 2 of all the sites, we collected 452 individual pieces of litter weigh - ing 5.366 kilograms giving an estimate of 2 pieces and 24 grams of litter per 10m 2 across all sites (Figure 7.11). We collected visibly more litter in terms of abundance (139 pieces) from all ten sites during this period than the previous one. However, the total weight of this collection was slightly lower (746 grams) than the previous collection. I contend that abundance is a more significant measure of overall “litteredness” than weight. Most sites looked measurably more lit- tered or affected by litter in November than they did in August. Weight is highly variable within classes and making inferential analyses on litter contamination is notably more difficult. This intra-class variability is particularly evident when examining the four sites with litter abundance counts from 50 to 61. Yet weight should not be discounted altogether as it is an important mea - sure for estimating litter for management and cleanup purposes. Figure 7.11. Second collection abundance (inner ring) and weights (outer ring) across study sites. Chart by A. Hilburn. Data, Field work, November 2011 272 As with the first collection, plastic was the dominant class, comprising 58.2 percent of the abundance and 61.4 percent of the weight in the second collection’s litter assemblage (Figure 7.11). Plastic drink bottles, plastic bags and packaging, as well as plastic cups and cutlery were the typical items found in this class. The small reuse value of most plastic packaging combined with its low commodity value added to its employment as transportable containers with ephem- eral contents appear to predestine plastic to become litter. Paper and cardboard items, usually some form of packaging, make up the next largest category with 13.9 percent of the abundance. Figure 7.12. Second collection normalized abundance by study site. Map by A. Hilburn. Data, INEGI 2010; Field work, November 2011 273 Following paper and cardboard is styrofoam with 9.3 percent of abundance, suggesting further the link with consumption and packaging’s link to litter. Hazardous waste such as diapers, urine bottles, and used toilet paper made 6.6 percent of the assemblage. Metal, “other”, glass, organic, and rubber represent the lower abundance proportions of the second collection. The most littered site for the second collection was the Busted Tire site with 103 pieces and 1.186kg over its total area (Figure 7.12). Normalized to account for site size, this amounts to 4.1 pieces and 47 grams per 10m 2. The next four sites, the Biodome site, the INEGI Marker site, the Diaper site, and the Crucero sites had 61, 60, 50, and 50 pieces per site respectively, yet noting the Crucero site’s smaller size (146m 2) points to its higher normalized abundance at 3.4 pieces/10m 2, a full piece more per areal unit than the INEGI Marker and Diaper sites. The two least littered sites we encountered during the second collection were the No Name and Dog Skull sites, with 1.2 and 0.2 pieces/10m 2 respectively. These two sites recorded low abundances in the previous collection and initial clean as well. Total collection results In the combined two collections, during August and November, we collected 765 pieces of litter on the 2,281 square meters of the ten study sites, with a normalized estimate of 3.35 pieces of litter for every ten square meters. The combined total weight of the litter collected was a significant 11.48 kilograms or just over five grams for every ten square meters (Figure 7.13). The most intensely littered site was the Busted Tire site with 161 pieces of litter or 6.44 pieces/10m 2 (Figure 7.14) . This site was consistently well-littered and had 2.3 pieces/10m 2 at the first collection and 4.1 pieces/10m2 at the second collection. This site, located on the outskirts of the cabecera, is on one of the busiest stretches of highway between the cabecera and Calipan making it a high traffic area near the municipio’s largest settlement. The Busted Tire site also had a considerable amount of garbage in the initial clean with 11.2 kilograms of litter collected in June. Fitting with its name, the Crucero site’s location at the crossroads of the two major paved highways in the study area made it the second most littered site for the combined collections with 5.68 pieces of litter per 10m 2. The Diaper site, located at another intersection between the Zoquitlán highway and an unpaved road to a sierra community was also highly littered with 4.44 pieces of litter per 10m 2. Crossroads are areas where traffic slows down and people wait on pub- 274 Figure 7.13. Both collection abundance (inner ring) and weights (outer ring) across study sites. Chart by A. Hil- burn. Data, Field work, August and November 2011 lic transportation, offering numerous opportunities to deposit litter. They are also places where passengers who get off from public transportation such as buses may be carrying empty food and drink packaging, that for a number of reasons, are discarded nearby. The INEGI Marker and Biodome sites had nearly identical amounts of litter with 91 and 90 pieces or 3.6 pieces per 10m 2 respectively. Their assemblages with regard to contents were almost identical. These two sites are covered in monte and are located just outside urban areas. The proximity to purchase and consumption points as well as their appearance of osten- sibly not belonging to anyone helps point to why these sites were littered as much as they were. The Shoebox (2.52 pcs./10m 2), Erosion (2.32 pcs./10m 2), and Moo Cow (2.12 pcs./10m 2) sites had lesser litter abundance than the largely undeveloped INEGI Marker and Biodome sites and were all located in areas that showed some development or human edifice. The Erosion site was used as a gravel depot, the Shoebox site was adjacent to sugar cane fields and a roadside gar- den, and the Moo Cow site was adjacent to an artisanal dairy. Perhaps the implied presence of people, property, or simply occupied space acts as a hindrance if not deterrent to littering activ- ity. The Dog Skull (0.76 pcs./10m 2) site was near a small homestead and while I am confident it 275 was not actively cleaned, the presence of surveillance by the owners may have explain the low litter abundance. The No Name site had low litter abundance (1.52 pcs./10m 2) but was located far away from any development on the steep descent from the sierra just below Pala. The high speeds, hard breaking, and hairpin turns that characterize driving the highway here offer very little opportunity to comfortably litter, which help to explain the low litter abundance. The composition of the litter for the combined total is dominated by plastic with 483 pieces making up 63.1 percent of the assemblage. Plastic bags and plastic films were the most Figure 7.14. Both collections normalized abundance by study site. Map by A. Hilburn. Data, INEGI 2010; field work, November 2011 276 abundant subclass followed by plastic bottles. Most of the plastic was likely food or drink pack- aging, suggesting a link with portable consumption and the litter problem. Further implying the packaging and littering connection was cardboard and paper making up the next most abundant class of litter, representing 12.5 percent of the assemblage. Styrofoam was the next most abun - dant class with 8.6 percent of the total, adding further evidence of packaging being synonymous with litter. In the combined collections sample, we counted 29 diapers and toilet paper pieces and three bottles of urine along with four other items ranging from used condoms and bits of soiled toilet paper to unidentifiable chemicals. These items, classified as hazardous litter, make up 4.7 percent of the litter abundance. Metal litter, mostly aluminum drink cans, made up 3.5 percent of the combined collections’ abundance. Litter that did not fit a discrete category (2.6 pcs./10m2), glass (2.4 pcs./10m 2), organic matter (1.8 pcs./10m 2), and rubber (0.7 pcs./10m 2) fill in the less abundant classes. INFERENTIAL STATISTICS TESTS AND RESULTS Methods The previously noted descriptive statistics of how much and of what type of litter oc- curs on the study area’s roadsides frame the litter issue’s magnitude there. These statistics pro - vide a matter-of-fact, quantified representation of littering and its rates of its accumulation. Yet they do not infer much into the distribution of this litter according to space and time nor do they elaborate anything regarding littering behavior. Understanding where littering is occurring can identify potential spaces of litter contamination or the situational conditions that engender it. Littering may also follow seasonal patterns and ancillary temporal data can point to times of the year where litter presence is greater. A Mann-Whitney’s U test was applied to the individual count data to test the relationship of situational context and seasonal difference on littering. This test was used because of the relatively few sites ( n = 10) available for sampling in the study. The Mann-Whitney U or its analogous inverse, the Wilcoxon W, is a rank-order statistic that identi - fies a moderately robust difference of means between two sample populations (McGrew and Monroe 2000: 163). The confidence interval used was 95 percent or p = 0.05. Normalized litter abundance ( A pieces/10 sq. meters) and indices derived from ecology studies were used to compare between population samples. Ecological indices such as evenness ( J ), Shannon diversity ( H’ ), and species richness ( S ) have been used before in similar attempts 277 to gauge the effect of place and season on littering (Seco Pon and Becheruchi 2012). For this study, the subclasses of garbage (Table 7.1) are considered the individual “species” from which the aforementioned indices are derived. Species richness ( S ) is simply the number of different species in a community or litter subclasses found within a site. Shannon’s diversity index ( H ) indicates the distribution of species or litter subclasses by representing the proportion of a par- ticular subclass relative to the total number of subclasses present which is then multiplied by the natural logarithm of the proportion. The evenness index ( J ) takes the Shannon index value ( H ) and divides it by H max giving a value between 0 (no evenness) and 1 (complete evenness). These indices provide added dimensions on litter composition with higher diversity, evenness, and rich- ness indicating greater littering intensity and more purposive littering behavior. The sites were stratified according to their locations in urban or rural environments to test the effect of place or environmental attributes on litter presence. Sites along the larger, higher traffic Tehuacán-Huautla highway and those along the smaller Zoquitlán road to the Sierra Negra were also stratified to test the effect of traffic on littering. In order to understand the effect of sea- sonality, I compared the means of each litter class between the two collection periods in August and November. It was expected that sites around towns, approaching some definition of what it means to be urban, would have significantly greater litter in all of the waste classes than sites in more rural settings. The busier Tehuacán-Huautla highway would presumably have higher litter abundance and greater ecological index values than the less-traveled Zoquitlán road. The second collection, representing litter that accumulated between August and November, was expected to be greater in all measures than that collected in August. Results After analyzing the normalized count data (abundance) with Mann-Whitney’s U, there were no significant differences between the sites classified as urban and those classified as rural in any of the litter classes and their sum by site in the first, second, and combined collections (Ta- ble 7.3). In all but two litter classes, the “urban” sample had a greater normalized mean than the “rural” sample, but these differences were not significant at p = 0.05 or any p-value approaching it. Similarly, there was no significant difference of means between the urban and rural subgroups for richness, Shannon diversity, or evenness in any of the collections or their combined total. The 278 “urban” sites had greater mean values for the three indices for the first collection and combined total but were surprisingly lower for the second collection. In short, there were no statistically significant differences in litter between sites classified as urban and those deemed rural. Comparing mean normalized abundance between sites along the Tehuacán-Huautla high - way and those along the road from the cabecera to Zoquitlán in the sierra also provided statisti - cally insignificant results. The lightly populated “rubber”, “other”, and “hazardous” classes of lit- ter had higher abundance means in the Zoquitlán road subgroup mostly due to a few anomalous litter scatters. The first collection class abundance means were much closer in value than those of the second collection, when the Tehuacán-Huautla highway subgroups were visibly greater except for two classes. For the combined counts, seven of ten litter class means for abundance were greater for the subgroup of sites on the busier Tehuacán-Huautla highway than the subgroup on the slow, winding road to the Zoquitlán. The ecology indices of richness, Shannon diversity, and evenness were also statistically insignificant for the first, second, and combined total collec- tions by highway stratification. The Tehuacán-Huautla subgroup means for these three indices were greater across the board than the Zoquitlán road subgroup but these were not significant at p = 0.05. The effects of different collection periods were tested to see if there was a seasonal effect on litter deposition (Table 7.4). Yet like, the previously noted stratified comparisons, there were no statistically significant differences between the normalized mean abundance between the col- lection instances for any of the ten sites. The second collection from November had higher total normalized abundance and seven of ten sites in November had greater normalized abundance but none of these differences were significant at the 95 percent confidence interval. In short, there is no statistically significant seasonal signal for littering deposition. DISCUSSION AND CONCLUSION Generation of litter and spatial modeling For the entire 2,281 square meters of Coxcatlán municipio’s roadsides surveyed for the twenty-week period from June to November 2011, we collected 765 pieces of litter that together weighed just shy of 11.5 kilograms (Table 7.5). This amounts to a generation rate of 8.69 pieces and 0.130 kilograms of litter per ten square meters per year. Assuming that this rate applies to the entire municipio’s roadsides, one can extrapolate the magnitude of litter generated there by mea - 279 Table 7.4. Mean abundance by sites and litter classes between collections. No differences of means at p = 0.05 was found for Mann-Whitney U tests. Data, Fieldwork 2011. Calculations done in SPSS 14. Table 7.3. Mann-Whitney U test results by ecological indeces between urban-rural and highway stratifications. No differences of means at p = 0.05 was found for Mann-Whitney U tests. Data, Fieldwork 2011. Calculations done in SPSS 14. 280 suring the length of all of the paved highways in the municipio and multiplying that length by 10 to account for the five meter easements on both sides of the road. After establishing the total roadside area for the municipio, one can then extrapolate the combined study sites accumulation rates to this area. Across the municipio, there are 69 kilometers of paved highway with five meter easements on each side of the road, totaling 689,169 square meters of what is considered here as roadside. After extrapolating the weekly litter generation rate for the combined sampled area to all of Coxcatlán’s roadsides for an entire year, the estimated annual accumulation is a surpris - ingly large 600,946 pieces weighing 9,017 kilograms. This mass of litter includes 379,421 pieces of plastic weighing a collective 4,449 kilograms, a hefty 1,021 kilograms of paper and cardboard from 75,413 individual pieces, and a whopping 1,536 kilograms of glass from 14,140 bottles, containers, and broken shards. A complete breakdown by class of the estimated annual litter ac- cumulation is in Table 7.4. While the litter accumulation rates in Coxcatlán appear by sheer magnitude excessively large, they are comparable to littering rates in the United States when normalized into an un - derstandable ten square meters. The Ohio DNR study showed a 0.132 kilograms per 10 square meter accumulation rate, slightly greater than Coxcatlán’s 0.130 kilograms per 10 square meters. Both of these rates are noticeably greater than the 0.054 kilograms per 10 meters square found along the State of Washington’s highways in a 2005 study, but a different methodology and Table 7.5. Study sites’ litter abundance and weight and the municipio-wide estimates for abundance and weight. Data, Field work 2011 281 analytical framework was used, limiting its comparability (Cascadia Consulting Group 2005). In terms of litter abundance, a similar pattern emerges where Coxcatlán’s roadside litter is less than that found on U.S. roads. The large-scale study conducted by the non-profit organization Keep America Beautiful surveyed 240 sites across the United States and estimated 9.15 pieces of litter per 10 square meters surveyed (MSW Consultants 2009). Coxcatlán’s roadsides have only 8.69 pieces per same areal unit. This similarity in litteredness defies the common assumption that many Coxcatlán residents have that the United States and other more developed nations are cleaner countries. These annual generation estimates are useful to quantify the magnitude of the litter issue in Coxcatlán. They are also highly provisional and do not take into consideration the spatial het- erogeneity of litter deposition implied by previous sections of this chapter. A more refined model of litter generation rates interpolated from the existing site data contextualized with anecdotal reference can perhaps provide more accurate insight into the litter issue. Litter studies in Ohio and New Jersey categorized their data by road types and road features such as overpasses and interchanges (Davey Resource Group 2004; Stein and Syrek 2005). Unfortunately a similar clas - sification method would not work in Coxcatlán because there are no overpasses, interchanges, or any really systematically applied road classes. A similar approach, assuming spatial autocorrelation or the similitude of nearer rather than farther phenomena, would apply the litter accumulation rates of particular sites to a stretch of nearby roadway or roadway with similar attributes. A systematized multi-criteria GIS analysis or spatial interpolation could be used to classify stretches of road and group sites, yet given the previous unsuccessful attempt to infer statistically significant differences between simple binary locational contexts, a more experience-based anecdotal classification more accurately character- izes particular stretches of road. After having spent considerable time on Coxcatlán’s highways, there are five recognizable qualitative classes of paved road in Coxcatlán. The relatively short distances of Coxcatlán’s paved roads and the ten months I had to familiarize myself with them supports the validity of these classifications. The stretches of road between the ridgeline or cumbre comprise the classification with the longest distance at 28.2 kilometers. This classification is defined by the sparsely littered No Name site, the Erosion site, and Dog Skull site, all of which are located on the steep, winding Zoquitlán road connecting Coxcatlán to the Sierra Negra. The urban classification is the next lon- 282 gest at 19.4 kilometers and its litter rates are calculated from the Busted Tire, Crucero, Moo Cow, and Shoebox sites. “Urban” roadsides are those in or near towns. The Diaper Site defines the 9.5 kilometers of rolling highway on the cumbre above the valley. The “sugarcane lands” classifica- tion is a 7.0 kilometer long classification represented by the INEGI Marker site. This designation refers to the highway between Pueblo Nuevo and Tilapa that is bounded on both sides by sugar - Figure 7.15. Roadside typology and the sites associated with each type. Map by A. Hilburn. Data, INEGI 2010; field work 2011 283 cane fields. The “monte” classification stretches 4.49 kilometers through the matorral and selva baja between the cabecera and Pueblo Nuevo that sits in reserve by the Coxcatlán ejido. The 316 meter stretch of highway that runs through San Rafael can largely assumed to be free of litter and contributes no litter abundance or weight to this reclassification. Figure 7.15 illustrates their distribution and lists their lengths. When this qualitative contextual model is applied to extrapolate the generation rates across the municipio, the results show a much less littered roadside in Coxcatlán. Using the clas - sified accumulation rates, the municipio-wide abundance drops to 563,362 pieces, a full 37,584 pieces of litter less than the previous optimal extrapolation. Fittingly, the classified model also reports a smaller total weight of 7,062 kilograms, 1,955 kilograms fewer than the optimal model. On an areal unit basis, the classified model estimates 8.17 pieces and 0.102 kilograms per 10 square meters compared to 8.69 pieces and 0.130 kilograms per same areal unit with the opti - mal model. These estimates are even smaller than the previously noted examples from Ohio and across the United States, making the contrast in material terms with Coxcatlán’s litter much more pronounced. Composition Litter, like garbage, is waste but differs in how it is produced. Both are the products of consumption. Yet garbage or municipal solid waste (MSW) is managed waste. It enters a socially prescribed trajectory towards reuse, recycling, or final disposal. Litter conceptually sits apart from garbage and its mutually understood active practices of management (Geller 1980). Litter - ing is an opportunistic, at times active, at times passive, and more public-than-private form of dealing with waste. Litter can become garbage if it is cleaned up, but left alone, it remains litter. Under these different conditions the composition of litter is different than managed garbage. The litter on Coxcatlán municipio’s roadsides is generally less organic and more plastic-based than the MSW produced by its households. Using percent weight by litter class, plastics comprise a far greater portion of the litter stream in Coxcatlán (49.3 percent) than that of its MSW stream (23.7 percent) or all MSW in Mexico (10.9 percent) (Table 7.6). These results are congruent with those of the two neighborhoods in Mexico City where Muñoz et al. (2009) conducted their litter survey. Both the affluent and poorer neighborhoods reported high plastic (74 and 72 percent) compositions12. Conversely, the organic component of Coxcatlán’s litter is proportionally smaller 284 than that of its MSW assemblage and especially lesser than that of the estimated national MSW composition. However, a claim that litter in Coxcatlán is across the board more inorganic is inac - curate. Glass, paper/cardboard, metal, and styrofoam are all inorganic and the MSW of Coxcat - lán has greater proportions in these waste classes than the litter estimated to be on its roadsides. Besides glass, Coxcatlán’s litter proportions are also less frequent in the aforementioned inor - ganic waste classes than streetside litter found in Mexico City’s neighborhoods. Regardless of what materials comprise it, litter in Coxcatlán is mostly packaging. Most of the plastic items are bags, films, cups, and bottles that once held some item of value that had been consumed prior to its disposal (Table 7.7). Out of all the 483 plastic items collected during both collections, 462 or 96 percent were some type of packaging. Plastic bags, the subject of nu - merous citizen complaints (Chapter 6) as well as systematic research into litter problems (Njeru 2006; Warner 2010), were the most abundant plastic litter with 238 pieces. Furthermore much of the litter was related to food and drink consumption. Eating and drinking on the road, whether it is on a bus, in a car, or by a pedestrian is a very common practice in Coxcatlán and across Mexico. Being on the road is a fairly private and anonymous experience, offering an opportunity to litter without the scorn of others. The lack of surveillance by others has been proven to be a condition for littering in a number of studies on littering behavior (Williams et al. 1997; Sibley and Liu 2004). The often maligned #1 PET soft drink bottle was also a common piece of litter at 97 pieces. “Plastic bottles” were noted repeatedly by residents during Q-method interviews as being a major issue, however imprecisely outlined by the interviewee. This same indefinite Table 7.6. Roadside litter composition by percent weight in Coxcatlán compared to other litter studies and MSW estimates in Coxcatlán and Mexico. Data, Fieldwork 2011, Muñoz-Cadena et al. 2009, Davey Research Group 2005 , Sanchez-Mendez and Olmos-Torres 2009, INEGI 2010. 285 noting of “plastic bottles” during the PRM exercise was noted by nearly three percent of partici - pants and had a high final importance index of 0.8. There were also 29 bottles made from other plastics, most notably the drinkable yogurt bottles that children tend to drink on buses while riding to school. We collected 21 plastic cups as well. Of the second most abundant class, paper/ cardboard, only 36 percent was packaging due to the high relative abundance of flyers and sheet paper in this class. Yet all but a few pieces of the Styrofoam, 46 cups and 17 food plates in total, and all but one piece of the metal litter were former containers of consumed items. Outside of Coxcatlán and Mexico, litter from packaging appears to be an almost uni - versal problem. The Keep America Beautiful nationwide litter survey conducted across the US found that 45.8 percent of roadside litter was packaging. Of the packaging found in this study, two-thirds was plastic (MSW Consultants 2009: 3). In a roadside litter study conducted in New Jersey, 61 percent of roadside litter was from food and convenience product packaging (Stein and Syrek 2005: 20). There are certain materials, however, that reflect differences between the US and Mexico. A considerable proportion of the litter in the Ohio DNR study was glass non-re - fundable beer (Percent) and soft drink bottles. Assuming Mexicans drink a similar amount of soft drinks and beer while on the road, one is still not likely to find as many glass bottles on Mexican highways. This assumption is probably not attributable to any higher environmental ethic of Mexicans, either. In Mexico, the overwhelming majority of beer and the most inexpensive soft drinks are sold in refundable bottles. While the contents are sold, these bottles are typically not sold but are tacitly loaned to known parties or consumed in a dine-in only context. In the case of Table 7.7. Roadside litter composition by abundance in Coxcatlán compared to Muñoz et. al 2009. Data, Field- work 2011, Muñoz-Cadena et al. 2009. 286 a permanent transfer, each bottle can cost up to 75 US cents and are likely too valuable to throw away. The recycle value of aluminum and the extra cost of canned beer and canned soft drinks also could potentially limit the amount of litter on the highways. Scavengers may also pick up “valuable” litter like metal, thereby skewing the collection results. The greater commodity valu - ation of things that are ostensibly waste in Mexico may partly explain the lower litter accumula- tion rates there than in the United States. Inferential tests results In contrast to this study, the Ohio DNR study mentioned above noted statistically signifi- cant urban-rural differences on county roads in Ohio, which approximate the highway system in Coxcatlán somewhat accurately. Curiously, in Ohio there was no apparent statistical significance for litter on interstate highways between urban sites and rural sites, perhaps due to the magnitude of an interstate creating a sort of urban space regardless of where it is. This latter fact, that sim- plistic urban-rural dichotomies cannot be used to differentiate the spatiality of litter accumula- tion, points to the lack of comparability of such qualitative assessments of what constitutes urban and rural between Ohio and southern Puebla. This incomparability of urban-rural classifications serves to mitigate the statistical insignificance of my testing. In fact, all of the Coxcatlán’s high- ways would likely be considered “rural” by the Ohio DNR study’s parameters. The lack of sig - nificance between any of the stratifications could also be attributed to the relatively small number of sites sampled ( n = 10). Ten sites were likely a too small of sample population to establish sig - nificant statistical differences between locational contexts. Also, Coxcatlán’s 249 square kilome- ters and approximately 69 linear kilometers of paved highway are probably too limiting and few to allow for any statistically significant spatial difference across sites and contexts. Finding 30 or even 20 suitable sites along such few kilometers of highway would have been extremely difficult, requiring the lowering of selection criteria which could have made the data from each site less intelligible with the others. A much broader survey beyond Coxcatlán’s borders, perhaps across the Tehuacán Valley, would facilitate the inclusion of more sites and provide sufficiently differ- ent locational contexts (more urban, more traffic). The main intent of this study, though, was to frame the magnitude of litter accumulation in Coxcatlán specifically. Additionally, although litter occurs most everywhere, the people and government of Coxcatlán generally only live with and 287 manage the litter that is part of their municipio, their communities, and their properties. Extend - ing the sample intervals could have also allowed for time to develop statistically significant mean differences between collection periods. However, allowing more time between collections would have ignored the intent of the research goal to understand if a seasonal effect exists regarding lit- tering accumulation. Management Implications The creation of anti-litter ordinances is not a difficult task for most rural Mexican muni- cipios because of the general consensus view of littering as a socially frowned-upon practice. En - forcing these ordinances is a much more taxing prospect for the limited law enforcement there. In rural municipios like Coxcatlán, nucleos agrarios, whether they are ejidos or comunidades agrarias, own much of the land on or around which littering occurs. Some of these corporate groups have appointed wardens to monitor land use but they are usually too few and are ultimate- ly powerless in terms of actual law enforcement. Besides anti-litter rule and law enforcement, there are the management options of cleanup and litter prevention. Litter cleanup, unfortunately can be an expensive prospect. The aforementioned study from New Jersey estimated that it cost $1.29US in 2007 to pay to clean up one piece of litter (Stein and Syrek 2005: 27). The Keep America Beautiful study estimated that litter management costs counties with less than 30,000 residents, a similar enumeration unit to Coxcatlán, $1.66US per capita every year. In the same re - port city governments reported $3.59US per capita spent on clean-up in their annual expenditures (MSW Consultants 2009: 93-94). Obviously, these are cost estimates from the United States, and those in rural Mexico would be lower due to cheaper labor costs. Yet when the costs are adjusted for currency differences and considering the meager budgets for many economically marginal municipios like Coxcatlán, paid clean-up can be a far-too-large financial burden. Currently in Coxcatlán and across rural Mexico, the social welfare cash payment program to poor families, Oportunidades , makes participation in public service projects a contingency for payment. Local coordinators from SEDESOL ( Secretaría of Desarollo Social ) or Secretariat of Social Develop - ment organize public cleaning days and set quotas for garbage picked up around town. However beneficial to town and village landscapes, the highway roadsides are often ignored, mostly due to the fundamental danger of working there. This latter fact is one of the reasons most volunteer efforts toward cleaning up roadways remain in the planning stages. The potential for anti-litter 288 volunteerism is great in places like Coxcatlán, where social properties control much of the land base. Communal labor in social properties, called tekios or faenas, could be directed toward litter cleanup. Given the substantial infrastructural resources of trucks and equipment of many ejidos and their members, anti-litter faenas could produce significant results towards removing litter from the roadsides near their lands. All of the above says nothing about litter prevention. Preempting littering is another management option for litter control. From a monetary perspective, anti-littering education is not an inexpensive control, but it can be effective in certain applications (Stein and Syrek 2005: 32). Signage can be used to cut down on littering but the effect only applies to the immediate area around a sign. Positive messages that ask people not to litter appear to be more effective than negative messages that state the fine or penalty for littering (Geller et al. 1977; Reiter and Samuel 1980; Durdan et al. 1985; Texas DOT 2008). Citing empirical evidence that the presence of litter encourages littering, litter control and clean-up could be a first step in preventing littering (Finnie 1973; Cialdini et al. 1990). The addition of trash receptacles along roadsides and other litter-prone environments has also led to decreased litter (Finnie 1973; Geller et al. 1980; Stein and Syrek 2005; DeKort 2008; Schultz et al. 2013). While this would involve the provision and maintenance of more litter receptacles by the municipio, they could be maintained by the exist- ing sanitation crews as they travel along the highways along their routes. With only anecdotal evidence to support the trash receptacle solution, the sierra community of Xacalco appears to keep its unpaved roadsides litter-free through the placing and maintenance of reinforced burlap sacks on posts throughout the town. While there is no panacea to the litter problem in Coxcatlán, efforts can be made to strategically allocate signage, clean-up projects, and receptacles to reduce roadside litter there. Hopefully, the results of this study can assist in that goal. Conclusion Litter is garbage outside the socially prescribed management trajectory of waste (Finney 1973). As a result, it is fundamentally perceived as a problem. Furthermore, litter is seen as an environmental problem by its abilities to chemically, biologically, physically, or aesthetically contaminate. Litter, like garbage, can be an agent of disease. The decomposing components of litter, the possible pathogens left by those who consumed what became the litter, and bodily fluid contaminated pieces all contribute toward litter’s attribution as a biological contaminant. Disease 289 vectors like stray dogs, rats, and insects subsist on litter, attributing an infectious quality to it. Ex - posure to garbage and litter can cause illness from a number of pathogens. If litter is inorganic, especially if it is plastic, it poses the potential to chemically contaminate the soil, air, or water through its photodecomposition, burning, or physical breakdown into particles. It also clogs wa - terways, spread onto agricultural fields, and can spill out onto roadways. The litter problem is an aesthetic issue as litter challenges notions of order and purity (Douglas 1965; Pandey 1990). The presence of litter visually serves to visually contaminate an environment and can lead to greater littering (Kelling and Coles 1996). Litter is also political. Its presence can signify a lack of gov - ernmental and societal capacity to control. It challenges order and propriety. Litter is a big issue in Coxcatlán. It is an obvious feature on streets, on vacant lots, in waterways, in the wooded areas, and along the roadways. Residents talk and complain about it. Arguably, litter that exists around residences and more occupied spaces like parks and markets looms larger in their collective consciousness. Yet it is immensely difficult to systematically identify the magnitude of the litter problem there because of maintenance and other forms of disturbance, not to mention the practical and ethical concerns of surveillance conducted by a researcher. Roadways are public property, are traversed by most people, go largely unmaintained, offer anonymity, and are most importantly, usually littered. The highways that lead into, through, and out of the municipio can be seen to reflect the order and environmental norms of its govern- ment and residents, whether this exists in any form of observable reality or not. The comparably lower rates of litter accumulation along the roadsides in Coxcatlán to those in many parts of the United States is surprising given that many Coxcatecos believe they live in a generally more trash-strewn environment than their northern neighbors. They may be correct, if only partially, as one could assume there may be more litter in Coxcatlán’s more lived spaces such as streets, parks, and peri-residential areas due to its observable presence in what appears to be more than some US towns and cities. Yet as we have seen in the roadside example, such assumptions may prove untrue and should be systematically tested. Then, where does the sense that Coxcatlán is dirty and trashy come from? As one can see from the results of Chapter 6, the litter problem is not so greatly posed in the sierra communities as it is in the more urbanized valley, but it does merit some attention by their residents, making it a universal concern. Perhaps the assumed greater abundance of litter comes from the fact that litter in Coxcatlán is believed to have no pretense of management by its government, social organizations, and residents. More 290 likely, general discourses that Coxcatlán is less economically developed, less educated, less ca- pable and subsequently less managed may be the source of such partially erroneous assumptions. Lastly, the litter issue posed by Coxcatecos contrasted by the estimates of its relatively lesser empirical magnitude testifies to the power of such discourses. Endnotes 1. Nosotros somos cochinos, no? or Pinches cochinos coxcatecos! (We [Coxcatecos] are pigs, huh? or Damn filthy Coxcatecan pigs!) were typical responses after I had satisfied casual inquiries into how much litter I had collected on the side of the road or any sort of reference to roadside litter. 2. This study collected, classified, and measured roadside litter at 56 sites ranging from country roads, state highways, federal highways and interstates, and interchanges across the state of Ohio. Each site was 1,500 feet long from the road to the edge of the right-of-way. 3. The gasolinera site was selected during the initial cleanup of June 2011. However, when the research team returned in August during the wet season, it was obvious that is was where litter carried by a small intermittent drainage was deposited. 4. The research team consisted of the author and his archaeologist wife in June and project research assistant Ricardo Olaya assisted in November. The August collection was conducted solely by the author. 5. Full protective safety equipment was used by all members of the research team. This included mask, leather gloves, fluores- cent safety vests, road markers, and sturdy footwear. 6. Upon returning from the initial clean of one site, I observed a car’s occupant discard of a lit cigarette which caused a brief, low intensity fire along the roadside. See Figure for the exact fire described here. 7. These differences can be notable and this chapter’s discussion section notes the differences between Mexican and US roadside litter composition. 8. Biodome (1996) was a popular low-brow comedy starring Pauly Shore and Stephen Baldwin. 9. The most intensely littered site we encountered was the site we eventually rejected near the gasolinera (gas station). There we cleaned 17.3 kilograms of litter from an area just under 250m2 on a dry early June afternoon. Ten weeks later, following a rain event well into the wet season, we discovered that our eleventh site was a charco or small drainage catchment and the reason for its impressive litter cache was not the location itself but a small ephemeral rivulet that deposited its trash load there. 10. The reason behind the naming of this site points to one reason it had the least amount of litter. Due to safety procedures, we did not collect diapers for weighing, rather we left them on site, counted them, and noted their location to avoid their inclusion in later collections. 11. This bag was the result of one incidence of litter and its contents were aggregated rather than counted separately. 12. It should be noted that the Muñoz et al. (2009) study only considered inorganic litter, thus inflating the proportions of plastic and other inorganic material relative to this survey that considered organic litter in its classification system. 291 Conclusion: Geographies of garbage in Coxcatlán, Puebla, Mexico The generation of garbage and the need to manage it are inevitabilities. Nearly all mate - rial consumption results in some kind of waste, or useless, abjected matter. The production of waste and the practice of dealing with it require both social and environment resources, making garbage management one of the most fundamental and basic human ecologies. Today, managing the estimated 131,000 metric tons of garbage that is generated daily in Mexico (INEGI 2010) is a major human-environmental issue there, as well as in other countries across the world. Gar - bage has the potential to contaminate, whether biologically, chemically, or aesthetically, the lived environment and impact human lives. Therefore, gauging garbage generation’s magnitude while devising ways to safely deal with it are important to governments at federal, state, and municipal levels in Mexico, as well as researchers investigating the issue. In response, much of the existing research on garbage in Mexico has an empirical, managerial perspective, as much of it addresses and defines generation rates and classification of the garbage stream. While this is important re- search, the garbage issue from a geographical perspective goes far beyond generation and classi- fication. This dissertation research, while acknowledging the importance of understanding gen- eration and classification, attempted to describe the governance of its management, the formal and informal practices of dealing with it, while looking at how garbage as an environmental issue is constructed locally and by whom. To answer all of these inquiries required a multi-pronged methodological approach. The intent of this research was to provide a more holistic and expan- sive definition of the issues of garbage and its management in Mexico by highlighting not just their material, but practical and discursive aspects as well. GENERAL RESEARCH FINDINGS Garbage generation, governance, and management in Coxcatlán Chapter 3 began by outlining the current state of garbage generation and management in Mexico in the context of its historical precedents. We see that generation rates are increas- ing each year with economic development being positively correlated with increased generation rates. Similarly, the inorganic fraction of garbage is increasing due to changes in consumption patterns, again generally due to economic development and urbanization. Nearly 79 percent of all the garbage generated in Mexico is collected by municipal governments or by private conces- 292 293 sions. The majority of the garbage collected is deposited in sanitary landfills, while the amount entering recycling plants is steadily increasing. Despite the increased importance of sanitary landfills in formal management, open-air dumps still persist, but they are on the decline. The re- maining 21 percent of garbage generated in Mexico is either burned, disposed of in a small com- munity dump, disposed of in public containers, or is illegally dumped or buried (INEGI 2010). In Coxcatlán, the practices and the actors involved in garbage management define a garbage trajectory that begins in consumption and ends in house lots, wooded areas, community dumps, the municipal dump, roadsides, and recyclers. The data used to construct this narrative comes from the months of living in Coxcatlán municipio and observing how people there man- age their garbage. I interviewed officials at the state environmental ministry and the regidora (regent) of ecology in Coxcatlán several times to understand the formal organization of garbage service provision and the responsibilities that correspond of each level of government. I followed the garbage collector crews in ride-alongs on a few occasions while asking questions and map- ping the collection route as we passed through the colonias and towns. I also spent time at the dump, mapping its extent and curiously looking into what kinds of stuff Coxcatecos throw away. In Mexico, the municipio is charged with the practical management of garbage. Fol - lowing neoliberal constitutional reform in 1983, the municipio’s role in providing services and collecting revenues has expanded, creating some precariousness in governance and service provi- sion for economically marginal municipios, such as Coxcatlán. In these places, state governments still provide major funding support to municipios in order for them to maintain infrastructure and provide services, including garbage collection. In the case of Coxcatlán, the state environmental ministry also plays an important supervisory and technical role as well, providing the municipio with its two collector trucks and assisting in the creation of municipal garbage management plans. At the municipio level in Coxcatlán, the regidor of ecology is placed in charge of formal garbage management by the presidente municipal. This job entails administrating town sanitation through garbage and recycling pickup. Currently, the two sanitation crews collect garbage and re- cyclables in the five valley settlements of Coxcatlán, Calipan, Tilapa, Pueblo Nuevo and San Ra- fael, where approximately 70 percent of the municipio’s population lives. The estimated 22,396 kilograms of garbage per week that the valley settlements generate are deposited in the 2.5-hect - are open-air dump about a kilometer south of the cabecera municipal. The recyclable cardboard and #1 plastic are stored until they are sold to various Tehuacán recyclers, with the funds being reinvested into the municipal budget. Although there is nearly universal service in the valley, households still practice other methods to manage their garbage. In addition to utilizing the collection service, households there burn, dump, compost, feed to animals, bury, sell to private recyclers, and reuse garbage to different degrees on their house lot and immediate areas. For the households in the sierra communities without access to collection, these are the only available garbage management practices. In most cases, the house lot is the site of disposal. Geographic research on peasant dooryard gardens and ethnoarchaeological investigation on household debris deposition serve to tie traditional garbage disposal with its ancient predecessors. The unevenness of collection service availability and the persistence of traditional gar- bage management practices create a bifurcated garbage management trajectory. Here, garbage is generated by consumers and is split into two flows. The first flow, the one largely sanctioned by official government policies, passes first through households, whereby it travels through munici- pal garbage collectors, and ends up in the dump or in a Tehuacán recycling facility. In the second flow, garbage is generated, and then is burned, dumped on-site or elsewhere, composted, fed to animals, sold to scrap recyclers, or repurposed. Here, outside of the garbage collection zone, house lots and tiraderos clandestinos become the final destinations for garbage. The state of garbage management in Coxcatlán’s households Through participant-observation, I gathered an understanding on how garbage manage - ment is governed and practiced in Coxcatlán. The results of this research comprise Chapter 3. One of the more important conclusions was the elaboration of a “garbage trajectory” or flow chart of garbage from consumption through its various managerial actors to final disposal, reuse, or recycling. To get a more accurate understanding of the degree to which households use certain garbage management practices, I conducted a nearly municipio-wide garbage management cen- sus from August to November 2011. Just by conducting the door-to-door surveys and interviewing the participants, I came to the conclusion that household garbage management in Coxcatlán, whether in the valley where there is collection service or in the sierra where there is not, is a highly gendered practice. Gar - bage management in the home is dominated by women. Three-fourths, or 388 of the 520 partici - pants, were women. When greeted by the male head of household during our surveying, we were 294 often referred to speak with the doña , or female head of the house as it was perceived by many that she would know more about how and when they managed their garbage. Where there is garbage collection service, 99 percent of household reported that they use it. While the frequency of garbage collection varies depending on location along the routes, 70 percent of all valley households replied that they put out their garbage at least once per week. This is strong evidence that the garbage collection service is the primary method of waste dis- posal for valley households. Although the recycling truck follows the same schedule and route as the garbage collector truck, recycling participation is much lower than garbage collection use frequency. Only 74 percent of households in the valley report recycling their cardboard, plastic, or metal through the municipal program. Recycling participation was reportedly higher in the smaller valley settlements of Tilapa (91 percent), Pueblo Nuevo (90 percent), and San Rafael (87 percent). The relativeness newness of garbage collection, let alone recycling combined with the high public sanitation requirements from customary law in these towns, are believed to be factors in this higher participation. The selling of scrap metal to private recyclers was also fairly common (31 percent) in valley households. In the more economically marginalized sierra com - munities, the practice of selling to scrap recyclers was much more prevalent, with 57 percent of households reporting selling their metal and some plastic. One of the important conclusions of this research is that Coxcatlán municipio’s house - holds utilize multiple methods of garbage management in spite of 70 percent of residents having comprehensive garbage and recycling service. My research highlights the persistence of tradi- tional methods of garbage management in Coxcatlán municipio. Despite the nearly universally available collection service in the valley, 28 percent of households there replied that they peri - odically burn part of their garbage. Larger lot size and locations away from more urban land - scapes have a positive correlation with burning as a management practice. In fact, many of the households with garbage collection service that continued to burn their garbage were from the peripheral colonias of the cabecera and from the smaller valley settlements. The convenience of burning and the belief that it can be done safely and innocuously if the contents are mostly organic are motivations for burning. For the sierra communities of Coxcatlán municipio, where collection service does not reach, burning is the primary method of garbage disposal with nearly all (97 percent) households reporting that they burn portions of their garbage. The portion that is burned here is typically the inorganic fraction of their household waste. This usually consists of 295 plastic bottles, toilet paper, Styrofoam containers, and diapers. Conversely, the households that burn garbage in the valley appear to mostly burn their yard waste. In lieu of other opportunities, sierra households are encouraged by health and social service officials to control burn in lined pits in designated portions of their lots. Besides burning, other traditional methods of garbage disposal are used by Coxcatlán municipio’s households. Square-meter-sized dump pits are commonplace and 41 percent of sierra households and 11 percent of valley household reporting some use of one in the homes. Tak- ing organic, typically food waste, to the garden or fields for fertilizer is also a common practice with a third of sierra households (33 percent) and 12 percent of valley households mentioning their use of the practice. This is a likely underreported value for the sierra households, as are the reporting of the practices of feeding food scraps to domestic animals and livestock and dumping outside of the house lot. The former practice is generally perceived to be culturally backwards, indicative of poverty, and potentially unsanitary. The latter practice, dumping, is formally illegal and is seen as morally improper, and was only reported by fewer than two percent of households, although the persistence and perhaps growth of tiraderos clandestinos on the landscape testify to the contrary. Beyond the immediate descriptive aspect of the garbage census data, their summary sug - gests two spatial models for garbage management with regard to land use, social relations, and environmental impact. The first one, the garbage collection model, is a land-intensive system whereby households cede control over garbage management to external actors, in this case, the municipio, in order to realize a solution to garbage disposal. This spatial solution removes the need for households to actively and practically contend with their waste; rather it is moved off site and away from the household. Here, households become mere sites of consumption whereby the waste from that consumption and the necessity of doing something with it is exported to a site of abjection and waste, in this case the 2.5 hectare dump south of the cabecera municipal. This garbage management model is land-intensive in that garbage collected from a wide spatial network of households is focused on one area, the municipal dump. It also depends greatly on fuel, collection infrastructure, formal governance, and more importantly, capital, in order for it to function. Lastly, the garbage collection model is a fix in the sense that it relocates the environ- mental impact of garbage, be it chemical, biological, or aesthetic, outside of the household and puts it somewhere else. 296 The other model, the house lot model, largely represents garbage management by house- holds in the sierra communities. This model does not achieve the spatial distancing of the pre- vious model that spatially and practically removes garbage from the home. Rather, house lots become sinks of garbage and its remnants following burning, decomposition, or dumping. Ma- terials flow in, are consumed, and largely remain. While there are customary rules in communi- ties where there is no collection service, the house lot model represents largely self-determined governance over how garbage is managed. There are also little-to-no capital inputs and there are no formal ties between a household and some greater waste management regime, such as a municipal government or private contractor. Compared to the previous model, the house lot model is land extensive when considering at a settlement or broader scale. Here garbage is not focused into one site, rather it is distributed across the breadth of entire settlements, on house lots or in small, but numerous tiraderos clandestinos. In terms of its environmental impact, house lots in this model are the recipients and endpoints of garbage flows. The recycling of organic matter from food scraps and garden waste back into house lot soils can be beneficial for soil fertility and household garden production. However, house lots also become sites of biological and chemical hazard and environmental degradation from the burning, burying, and storage of what is increas - ingly becoming a larger inorganic fraction of garbage. While valley households, with access to garbage collection and traditional garbage management practices, can transcend the rigidity of the two-model system proposed here, this distinction remains binary and fixed for sierra house- holds, creating issues of environmental justice. The subjectivities of garbage generation and management Garbage and its management go far beyond material constructions in their definitions as human-environmental issues. Garbage is material and it is disposed of by x number of ways in y number of places by z actors. Yet, the definition of what constitutes garbage and how by whom should dispose of it are contested issues. From this perspective, garbage is subjective, discursive, and political. The garbage census and the participant observation of garbage management in this dissertation project framed the material and practical aspect of garbage management in Coxcat- lán and in Mexico at large. From this research one can see that there are conflicting subjective qualities of what garbage is and how it should be managed. This is especially apparent when one sees that burning and dumping, as well as other traditional practices persist amid garbage col- 297 lection availability in the valley settlements. Q-methodology, an approach that attempts to gauge practiced human subjectivity can be useful in highlighting the diversity of perspectives regard- ing a particular issue. Q-methodology is increasingly being utilized by geographers and allied researchers who wish to provide more ample descriptions of environmental debates from a range of actors. In this case, Q-method allowed me to show how Coxcatecos perceive what garbage they themselves generate and how one should deal with a particular type of garbage. These constructions stay congruent with and diverge from official discourses on garbage generation and appropriate management practices. There was some congruity, even if not statistically comparable, between the perceived and empirically quantified descriptions of the garbage that Coxcatecos generate. The “Mostly Foodies” factor, represented by a garbage assemblage dominated by foodstuffs and other organic fractions of garbage, points to a traditional, local consumption-oriented solid waste stream. The statistically different factor titled “Homebodies” represented a garbage assemblage that would be typical of a household that incorporates traditional consumption of locally-produced goods with mass-produced, processed goods from afar. This factor is situated between the “Most Foodies” factor and the third selected factor, “Packaged”. This last factor is represented by a garbage as- semblage consisting of packaging and inorganic components. “Packaged” refers to consumption preferences that engage mass consumerism, finished goods, and globalization at the expense of local, organic, and traditional consumption patterns. The summary of the garbage generation Q- survey point most directly to this tension between globalization and tradition. The second Q-survey investigated the subjectivities of Coxcatecos regarding appropriate management of particular classes of garbage. The intent behind this was to create understanding of how people believe types of garbage should be managed, directly by practice and indirectly by whom should manage it. The measurement scale used for this sort suggested a spectrum of agree- ment and disagreement between garbage management based in the home and that conducted by the municipio. The “Sanitary Citizens” factor weighed its priorities on the removal of dangerous, whether toxic or infectious, garbage through formal collection and disposal by the municipio. This perspective placed emphasis on garbage as a risk and its safe and sustainable control is beyond the moral and practical capabilities of the individual household. This factor illustrates a reliance on the garbage collection service and sees it as most appropriate. The “Home Recy- clers” factor also reflects a reliance on formal garbage management outside of the household. In 298 contrast to the “Sanitary Citizens” factor, it points to the management of recyclables rather than mitigating risk. Also, the “Home Recycler” factor does intimate a role for household manage- ment for organic and chemically neutral classes of garbage. This factor represents a viewpoint that mirrors the current mixed approach for garbage management for many homes in the valley. Contrasting with the previous two factors with regard to their specificity toward how garbage should be managed is the “Convenience Independents” factor. This factor conflates inorganic/or- ganic, toxic/safe, and recyclable/persistent garbage into similar categories. From this perspective garbage is garbage and not an assemblage of classes that each have associated preferable man- agement practices. The distinguishing characteristic for the “Convenience Independents” factor is its pragmatic view that garbage that takes up space and provides a hindrance at home, namely concrete and rubble, should be dealt with by the state. This study points out the usefulness of Q-methodology in drawing out the perspectives and viewpoints regarding a particular environmental issue. In this case, I identified three distinct viewpoints on how Coxcatecos see the materiality of garbage through what they generate in their homes. These viewpoints provide a participant-oriented, bottom-up public perspective on an issue largely dominated by official, managerial discourse. In the second Q-sort, I identified three factors that first represent the appropriateness of management practices for certain garbage classes but also define the roles of the household and the state in managing garbage. Part of the analytical power of Q-methodology is its ability to define unexpected perspectives. Intensive interviews, ethnography, and focus groups can elicit similar perspectives, but Q-method pro- vides a desirable statistical validated summary of these views. It does not make an appropriate central methodological base on which to approach a topic. Nevertheless, Q-methodology when employed correctly is a useful supplemental methodological tool to gain a nuanced analysis on a subject. The heterogeneity of opinion regarding garbage-related issues The previous Q-method study investigated the subjectivity of participants regarding two very important garbage-related issues. This study added a dynamic subjective perspective on garbage generation and garbage’s impact on the environment. Nevertheless, it still recognized a narrow definition of the garbage-society-environment nexus. The participatory risk mapping exercise described in Chapter 6 expands the definition of garbage and its management beyond 299 generation and disposal to represent a set of human-environmental issues. The intent of this research component was to describe how garbage and its social and environmental impacts affect a broad sample of residents from Coxcatlán municipio. It was an attempt to go beyond the gen - eration and disposal constructions of garbage as a social or environmental issue and sought to highlight what everyday residents see as legitimate garbage-related issues that affect their lives and communities. The participatory risk mapping methodology, while it is not fully participa- tory nor does it produce formal, geographical maps, is a valuable and expedient tool to gauge the prevalence and relative importance of an issue across a study population. In previous studies, the participatory risk mapping method was used to gauge the extent and severity of perceived risk. This instance of its use maintained the interview structure and statistical analytical methods, but substituted risks for garbage-related issues and severity regarding risks for relative importance of noted garbage-related issues. The numerical indices that the method produced allowed for statis- tical testing to see if the frequency of mentioning a particular issue correlated with a particular demographic or spatial categorical variable. The results show that 433 participants in this study listed 80 garbage-related issues. The issues with the highest incidence were problem burning, dumping, public cleanliness, street lit- tering, and service provision. Burning was the most frequently listed (highest incidence) issue across the entire study population exhibiting moderate to high importance as well. There were no significant statistical differences between any of the sample stratifications (women/men, valley/ sierra…) for burning frequency, but the issue of burning as a positive practice was mentioned frequently by sierra participants. Yet in spite of this other conceptualization of burning, problem burning still had high relative incidence and moderate importance for sierra participants. Dump - ing had the second highest incidence among all participants and had a higher importance value than burning. The subsample from the eight sierra communities listed dumping as a problem more frequently than any other issue. In the absence of formal garbage collection, dumping is the only practice to remove garbage and its residues outside of the home, perhaps adding to the frequency of dumping as a practice in the sierra. Women listed dumping significantly more times than men. Related to dumping, street littering had notable incidence and importance, with partici- pants from more urbanized contexts mentioning this issue more than those from smaller towns. The high incidence and importance of public cleanliness across the study population touches upon the political aspect of garbage. Poorly managed garbage and poor sanitation challenges no- 300 tions of public order and political competency, and is thus a notable issue, from the cabecera to the small towns of Coxcatlán municipio. The chi-square testing pointed out interesting spatial and demographic divisions regard- ing particular issues. The significantly greater incidence of the issues of service inadequacy and public garbage receptacle inadequacy by valley participants highlights their reliance on garbage management by actors outside of the home while pointing to the sense of self-reliance by sierra participants. The positive view of burning was almost exclusive to sierra participants. The issues of dumping and water contamination by dumping just missed the 0.05 confidence interval but have higher incidence in interviews with sierra participants. Dumping is likely more frequent in sierra communities, and due to their greater reliance on land resources to provide food and income from farming, it bears considerable importance there. Water is more scarce in the valley than in the sierra, but the greater reliance on land resources combined with minimal water deliv- ery infrastructure, maintaining the integrity of water resources is also likely greater in the sierra communities. In the cabecera municipal, recycling had significantly greater incidence by partici- pants from the town center than the peripheral colonias, where participation in the recycling pro- gram is greater. Street littering had significantly greater incidence with women participants than with men. In other research, women are less likely to litter (Cialdini, Reno and Kallgren 1990; Meeker 1997) and in the study area, women typically maintain the cleanliness of areas around their homes, making it a perceivable issue through their daily domestic work. Lastly, in a general sense, the issues taken from the participatory risk mapping data can be summarized into categories that point to how Coxcatecos conceive of waste disposal. The participatory risk mapping methodology allowed participants to construct the issues related to garbage as they saw them and to what degree they were important to them. When categorized, the issues showed how garbage is conceptualized as a hazard and as risk but also as everyday, in - evitable matter. Unmanaged garbage, in the form of dumping or as litter, can be conceived of as a challenge to formal governance. Litter exemplifies both the material and discursive properties of garbage, and as such, is the topic of the final chapter. Material realities and perception of litter in Coxcatlán In casual discussions and formal interviews during my stay in Coxcatlán, many residents would note that they believed Coxcatlán was an intensely littered place and one far more so than 301 where I was from in the United States. I was not sure about these anecdotal assessments and I thought it would be interesting to weigh Coxcatecos’ assumptions about litter and public cleanli - ness against quantifiable measurements of litter. As a fitting end to the dissertation, the last chap- ter centers on the tension between garbage as a material issue and a discursive or perceptional issue. To measure the degree to which Coxcatlán’s highways were littered, I borrowed and cus - tomized methods used by engineering firms and state departments of transportation to quantify and classify highway litter in the U.S. The methods centered on selecting random sites through - out a prescribed area, cleaning them of litter, and returning periodically to classify and measure the litter that had deposited in the meantime. To date, no such studies on highways have been conducted or made available to the public in Mexico to the best of my knowledge. After selecting ten, 5 by 50 meter sites and conducting an initial cleaning of them, two collections were made at 10-week intervals (August and November). During these collections, all litter was classed according to predetermined categories, then counted and weighed. One objective of this research was to classify and quantify litter in order to establish area-normalized (pieces/10m2) and municipio-wide estimates for highway litter in Coxcatlán municipio. The first estimate could be used to compare the relative abundance of litter in Coxcatlán to other places where such studies have been conducted with similar methodologies. The latter estimate pro- vides an absolute count or tonnage of litter in the municipio. Another objective was to determine what people are littering along the roadsides. Lastly, in order to test if there were any locational attributes related to litter accumulation, I conducted differences of means tests on litter abun - dance between sites classed as “urban” and “rural” as well as between sites on the major paved highway in the valley and those on the smaller, winding highway into the sierra. I also compared mean differences between collections to see if there is any seasonal effect on littering. Following both collections, this research estimates that there are 8.7 pieces weighing 130 grams for every 10 square meters of highway right-of-way in Coxcatlán municipio. In terms of weight, the 130 grams/10m2 in Coxcatlán are nearly identical to a similar study conducted across Ohio, where 132 grams/10m2 were estimated. When considering sheer abundance, Cox- catlán’s 8.7 pieces/10m2 were also similar to and less than the nationwide Keep America Beauti- ful survey’s 9.2 pieces/10m2. Thus, many Coxcatecos’ assessments were not congruent with the empirical measurements of litter along their highways. In fact, litter, in terms of its abundance 302 and mass in Coxcatlán, is the same or slightly less than in the U.S. These comparisons testify to the importance of perception and discourse in framing environmental issues. When these per area unit estimates are extrapolated to the entire easement area of Coxcatlan’s highways, there are an estimated 600,946 pieces of litter weighing 9,047 kilograms. When a qualitative contex- tual model is applied to this data, the estimates are less than the unclassified estimates. This even lower comparative estimate further emphasizes how public perception can be as powerful as empirical estimates. Plastic dominated the total litter assemblage, representing 63 percent of the total abun- dance (pieces) and 49 percent of the total weight. Paper and cardboard litter followed, with 12.5 percent of the total abundance and 11.3 percent of the total weight. Styrofoam or polystyrene litter followed with 8.6 percent of the abundance but because of its low per-item weight, it only comprised 1.7 percent of the total weight. These three most abundant classes point to what lit - ter really is beyond the typical materials-based classifications. Litter, in Coxcatlán, is mostly packaging. Disposable beverage containers, plastic bags, cups, plates, boxes, cartons, and cheap cutlery define what litter is and suggest the behavior and conditional circumstances of littering which center on eating, drinking, and convenient consumption on the go. The statistical comparison between site locational classifications and collection periods rendered no significant differences in terms of abundance or any of the indices borrowed from ecological methods. This is not to say that there are no spatial and temporal attributes associ- ated with littering, however. The lack of significant results between locational classes is likely due to there not being any real difference in locational attributes between areas listed as “urban” and those deemed “rural”. In a broader geographical sense, most sites in the study area would be deemed “rural” compared to highways that pass through metropolitan areas. Lastly, and perhaps most importantly, the small spatial frame for this study and the subsequent small sampling size also contribute to the lack of statistical significance between the sites and collection periods. The results of this study show that roadside litter accumulation in Coxcatlán municipio is lower than estimates from the United States, but it only sets the stage to see why litter remains a problem. This problem transcends litter generation and classification. Part of the intent of this dissertation is to expand the garbage issue in human-environmental terms beyond discussions re- garding generation. Litter is no different. In fact, the litter problem can be framed around its per - sistence, thus becoming largely a governance and management issue. Cleaning up litter along the 303 highway can be expensive and dangerous. Currently, there is a no formal litter cleanup program organized beyond that by the social welfare program Oportunidades, where recipients periodi - cally collect litter in and around their communities. This cleanup rarely extends to the highways. Some anti-litter signage is present around the municipio but its effectiveness is undeterminable. The almost unenforceable nature of anti-litter ordinances makes it impossible to determine their impact on litter prevention. The results of this study provide a useful, basic understanding of the material magnitude of the litter issue, but ultimately, understanding the litter problem will require more focused research into why and how litter remains and how it can prevented and managed in any sense. CONCLUSION Final commentary This dissertation is a regional geography of garbage in rural Mexico, adding to a grow- ing body of research on garbage in Mexico. Its use of multiple methodologies grounded in place by a descriptive regional approach seeks to go beyond the dominant epistemological themes of garbage generation and classification. Related to this literature, I intend this work to be an im- portant case study that offers different perspectives on garbage as a human-environmental issue from an often overlooked context. Topically, this dissertation’s focus on garbage management and the social construction of garbage-related issues expands this garbage research beyond the generation estimates and pepenador case studies that predominate. By examining garbage and its management in a rural, marginalized municipio like Coxcatlán, Puebla, this dissertation offers a unique case study outside of large cities. This research’s attention on the spatial scales of garbage management and its emphasis on households as sites of mixed garbage management practices within different management regimes are also important. By defining the scales of governance and management in Coxcatlán, the trajectory through which garbage passes from consumption to its final destination is made apparent. More importantly, this research on household garbage management in Coxcatlán draws attention to the persistence of traditional garbage management practices alongside contemporary garbage collection service. It further infers how uneven access to this garbage collection service creates two garbage management models, each with its own governance, spatiality, and social and environmental impacts. In essence, this research takes what were previously mere slivers on a pie chart, showing the proportions of Mexicans or Poblanos 304 with and without garbage collection, and applies the map to them. With this in mind, one can see spatial and demographic relationships with access to formal garbage infrastructure, tying garbage management with larger sociopolitical processes in Mexico regarding state engagement with the social sector in an era defined by neoliberal political economy. The garbage issue is much more than the generation of garbage and then finding a way to deal with it. In response, this dissertation also contends that garbage as a human-environmental issue is subjective and perceptional, in addition to it being constructed as a material one. The participatory risk mapping study shows that there is considerable diversity in how people con- struct garbage as a human-environmental issue and that these constructions are contingent upon place and demographic variables such as gender. Additionally, the Q-method study suggests that two given, well-known issues like garbage generation and management are not uncomplicated and are defined by different and incongruent perspectives. Lastly, the litter study is an example of how the material and perceptual aspects of a given issue such as litter may not run together, pointing to the power of discourses in the construction of environmental issues. Methodologically, this research incorporates three, not often used methods in human geographic research. Q-method is becoming an attractive methodological approach for research into human-environmental research, namely political ecology, and this dissertation gives another example of its useful employment. Participatory risk mapping, which I contend is not participa- tory, not mapping in the traditional sense, and not really confined to identifying risk, nonetheless holds immense potential to provide grassroots, ground-up perspectives with expedient fieldwork and statistically validated results on key issues in human geography. The litter study provided a useful empirical baseline to gauge litter’s subjective aspect and was also a successful pilot study for conducting a much more spatially expansive and temporally deep study in Mexico. Lastly, the regional geography that specifically comprises Chapter 2 and Appendix 2 but contextualizes the entire dissertation is a call to bring back detailed descriptions of society and the environment in places. While the positionality of the researcher-writer is inescapable in these sorts of descrip- tive regional geographies, they offer invaluable descriptions of the places that produce and are produced by larger social and environmental processes. 305 Coming back and the future of garbage in Coxcatlán? This addendum, written after a brief visit back to Coxcatlán town in August 2013, does not provide any definite certainty with regard to the future of garbage management in the muni- cipio, but it does attribute some measure of importance to this work. Since I left Coxcatlán just before Christmas 2011, many things have changed, just not many related to garbage. The north - ern section of the major paved highway was being completely overhauled and resurfaced. This is a much needed public work that will improve everyone’s lives a little. The construction did take out about a third of my litter study sites but it was all for the greater good. Now there is a direct, once daily ADO (1st class) bus service to the state capital. Also during my visit, I saw new con - struction all around the centro, including a new motel, and the streets around the parque were be- ing improved with new pavers. More lots looked like they had been developed in the peripheral colonias. The past sugarcane harvest was not very good and some cañeros I knew had not been paid by the refinery yet but they expected to be paid. The rainy season had not really arrived as a prolonged drought was affecting the region. There were new elections for municipal president and ejido comisariado. The former resulted in a big win for the PRI and the latter created what looks to be a divided and contentious struggle over the ejido’s future. My friends and acquain - tances I knew were well for the most part, although a few people I knew had passed away. The young man who was my research assistant had a full-time job and was engaged to be married. It had not been that long, only about 19 months, but things looked a little different. The garbage collection system, in contrast, was roughly the same as it was when I left it in December 2011. The same map I made outlining the collection service zone still applied. The same sanitation crew went on its weekly routes which all still ended in the municipal dump south of town. The recycling program remained in full effect. Up the mountain, the sierra communities kept on managing their garbage in their homes, community dumps, or in the monte, as they still lacked collection service. Yet, I do expect things to change. The just over year-and-a-half since I had left is not ample time to coordinate, plan, and execute the major effort of extending col - lection to the sierra communities. It would require doubling the capacity of the sanitation crews. Nor is 19 months long enough to plan and construct a sanitary landfill. The politics of getting nearby municipal governments to get involved in what would be an inter-municipally operated landfill would at least take that long. Combine that with the short, 3-year-long tenure of a mu- nicipal government and the current state of garbage disposal remains obvious. 306 The political pressure to extend service to a growing part of the municipio combined with the operational and budgetary capability of the state government will perhaps lead to some changes in the near future. The likely result will be an inter-municipal sanitary landfill shared with adjacent sierra municipios to the east or Coxcatlán joining some configuration of the exist- ing inter-municipal system based out of Ajalpan. Depending upon the success of this program, most of the municipio will follow the garbage collection model, along with its benefits and costs. These costs may eventually find their way to residents’ doorsteps as they have in other parts of the Tehuacán Valley where per bag fees and monthly charges exist. Nevertheless, for good and bad, which we have seen are highly contingent qualifiers when discussing garbage management, this change will signal the end for garbage disposal practices that have likely been in place for centuries if not millennia. Whatever happens or does not happen, this research offers itself as a baseline study. In the end, it provides a snapshot in time on how garbage was managed while Coxcatlán was a marginal-yet-globalizing and rural-but-urbanizing municipio. 307 308 References Ackerman, E. 1963. Where is a research frontier? Annals of the Association of American Geog - raphers 53: 429-440 Addams, H. 2000. Q methodology, in Social Discourse and Environmental Policy: An Applica - tion of Q Methodology , eds. H. Addams and J. Proops. Northampton, MA: Edward Elgar Pub - lishing. Agrawal, A. 1994. Mobility and control among nomadic shepherds: the case of the raikas, II. Hu - man Ecology 22(2): 131-144. Al-Khatiba, I., H. Arafat, R. Daoud, and H. Shwahneh. 2009. Enhanced solid waste management by understanding the effects of gender, income, marital status, and religious convictions on atti- tudes and practices related to street littering in Nablus – Palestinian Territory. Waste Management 29: 449-455. Alva-Alva, M. 2011. Personal communication, December 15, 2011. Anderson, B., J. Romani, H. Phillips, M. Wentzel, and K.Tlabela. 2007. Exploring environmental perceptions, behaviors and awareness: water and water pollution in South Africa. Population and Environment 28: 133-161 Anderson, E. 1952. Plants, Man and Life . NY: Little, Brown and Co. __________. 1954. Reflections on Certain Honduran Gardens. Landscape 4(1): 21-23 Appendi, K. 1998. Changing Agrarian Institutions: Interpreting the Contradictions. in The Transformation of Rural Mexico: Reforming the Ejido Sector , ed. W. Cornelius. San Diego, CA: UCSD Press. Arnold, P. 1990. The organization of refuse disposal and ceramic production within contempo - rary Mexican houselots. American Anthropologist 92: 915-932. Aschmann, H. 1959. The Central Desert of Baja California: Demography and Ecology. Iberoamericana 42. Berkeley, CA: University of California Press. Assetto, V., E. Hajba, and S. Mumme. 2003. Democratization, decentralization, and local envi - ronmental policy capacity: Hungary and Mexico. The Social Science Journal 40: 249-268. Assies, W. 2008. Land Tenure and Tenure Regimes in Mexico: An Overview. Journal of Agrar - ian Change 8(1): 33-63. Avendaño, P., A. Casas, P. Dávila, and R. Lira. 2006. Use forms, management and conservation of “pochote” Ceiba aesculifolia in the Tehucán Valley, central Mexico. Journal of Arid Environ - ments 67: 15-35. Ayala-Rodríguez, I. 2004. Acción Ambiental Universitaria: el reciclaje como una experiencia de educación ambiental. In La Basura en una perspectiva multidisciplinaria . Ed. J. Guevara Mar- tínez. Mexico, D.F.: Unlisted press. 309 Baird, T., P. Leslie, and J.T. McCabe. 2009. The effect of wildlife conservation on local percep - tions of risk and behavioral response. Human Ecology 37:463-474. Baitenmann, H. 1998. The Article 27 Reforms and the Promise of Local Democratization in Cen - tral Veracruz. Changing, In The Transformation of Rural Mexico: Reforming the Ejido Sector , ed. W. Cornelius. San Diego, CA: UCSD Press Baños Ramirez, O. 1998. PROCEDE: Gateway to Modernization of the Ejido? The Case of Yucatán, in The Future Role of the Ejido in Rural Mexico , eds. R. Snyder and G. Torres (La Jolla, CA: UCSD, Center for US-Mexican Studies). Barkin, D. 2006. La Gestión del Agua Urbana en México: Retos, Debates, y Bienestar . Guadala- jara: Universidad de Guadalajara. Barnes, G. 2009. The evolution and resilience of community-based land tenure in rural Mexico, Land Use Policy 26: 393-400. Barr, S. 2002. Household Waste in Social Perspective: Values, attitudes, situation, and behavior. Aldershot, U.K.: Ashgate. Barrows, H. 1923. Geography as human ecology. Annals of the Association of American Geogra - phers 13 (1): 1-14. Bartra, A. 2004. Rebellious Cornfields: Towards Food and Labour Self-Sufficiency, In Mexico in Transition . Ed. G. Otero. NY: Zed Books. Bassett, T. 1988. The political ecology of peasant-herder conflicts in Northern Ivory Coast. An - nals of the Association of American Geographers 78: 433-452. Bassett, T. and K. Zueli. 2000. Environmental discourses and the Ivorian Savanna. Annals of the Association of American Geographers 90 (1): 67-95 Beck, U. 1992. Risk Society: Towards a New Modernity . London: Sage. ______. 1996. The Reinvention of Politics: Rethinking Modernity in the Global Social Order . Cambridge, UK: Polity. ______. 1999. World Risk Society . Cambridge, UK: Polity. Beebe, J. 1987. Rapid appraisal: The evolution of the concept and the definition of issues. In Pro - ceedings of the 1985 International Conference on Rapid Rural Appraisal . Khon Kaen, Thailand: Khon Kaen University Press. Bennett, J. 1969. Northern Plainsmen: Adaptive Strategy and Agrarian Life . Chicago, IL: Aldine. Berglund, C. and Soderholm, P. 2003. Complementing empirical evidence on global recycling and trade of waste paper. World Development 31: 743–754. Bergman, R. 1980. Amazon Economics: The Simplicity of Shipibo Indian Wealth . Dellplain Latin American Studies 6. Ann Arbor, MI: University Microforms International. 310 Bernache, G. 2003. The environmental impact of municipal waste management: the case of Gua- dalajara metro area, Resources, Conservation, and Society 39: 223-237. Bernache, G., S, Sánchez-Colón, A.M. Garmendia, A. Dávila-Villarreal, M.E. Sánchez-Salazar. 2001. Solid waste characterization study in the Guadalajara Metropolitan Zone, Mexico, Waste Management and Research 19: 413-424. Bhuiyan S. 2010. A crisis in governance: Urban solid waste management in Bangladesh. Habitat International 34(1): 125-133. Bickerstaff, K. and G. Walker. 2003. The place(s) of matter: matter out of place – public under- standings of air pollution, Progress in Human Geography 27: 45-67. Blaikie, P. 1985. The Political Economy of Soil Erosion in Developing Countries . NY: Longman. Blaikie, P. and H. Brookfield. 1987. Land Degradation and Society . NY: Methuen. Blake, M. 1991. An Early Formative Chiefdom at Paso de la Amada, Chiapas, Mexico, In The Formation of Complex Society in Southern Mesoamerica , Ed. W.R. Fowler, Jr. Boca Raton, FL: CRC Press. Boege, E. 2008. El Patrimonio Biocultural de los Pueblos Indígenas de México . Mexico, D.F.: Instituto Nacional de Antropología e Historia y Comisión Nacional para el Desarrollo de los Pueblos Indígenas. México. Bonner, R. 2012. The cartel crackdown: winning the drug war and rebuilding Mexico in the pro - cess. Foreign Affairs 91(3): 12. Boserup, E. 1975. The impact of population growth on agricultural output. Quarterly Journal of Economics 89(2): 257-270. Brand, D. 1951. Quiroga: A Mexican Municipio . Washington, D.C.: US Government Printing Office. Brannstrom, C. 2011. A Q-method analysis of environmental governance discourses in Brazil’s northeastern soy frontier. The Professional Geographer 63(4): 531 – 549. Brannstrom, C., W. Jepson, and N. Persons. 2011. Social perspectives on wind-power develop - ment in West Texas. Annals of the Association of American Geographers 101(4): 839-851. Braun, B. and N. Castree. 1998. Remaking reality: nature at the millennium . NY: Routledge. Bray, D., L. Merino-Pérez and D. Barry. 2006. The Community Forests of Mexico: Managing for Sustainable Landscapes . Austin, TX: University of Texas Press. Bridge, G. and P. McManus. 2000. Sticks and Stones: Environmental narratives and discursive regulation in the forestry and mining sectors. Antipode 32(1): 10-47. Brookfield, H. 1964. Questions on the human frontiers of geography. Economic Geography 40: 283-303. ___________. 1969. On the Environment as Perceived. Progress in Geography 1: 51–80. 311 Brown, S. 1980. Political Subjectivity . New Haven, CT: Yale Press. Brown, S. and K. Mathewson. 1999. Sauer’s descent? Or, Berkeley roots forever? Yearbook of the Association of Pacific Coast Geographers 61: 137-157 Brunet, J. 1967. Geologic Studies. In The Prehistory of the Tehuacan Valley, Volume One, Envi - ronment and Subsistence , Ed. Douglas S. Byers, 66-90. Austin, TX: University of Texas Press. Bryant, R. 1992. Political Ecology: An emerging research agenda in Third World studies. Politi - cal Geography 11: 1-36. Buenrostro, O., G. Bocco, and G. Bernache. 2001. Urban solid waste generation and disposal in Mexico: a case study, Waste Management and Research 19: 169-176. Buenrostro, O., G. Bocco, and J. Vence. 2001. Forecasting generation of urban solid waste in developing countries--a case study in Mexico. Journal of the Air and Waste Management Asso - ciation 51: 86-93. Buenrostro, O., G. Bocco, and S. Cram. 2001. Classification of sources of municipal solid wastes in developing countries. Resources, Conservation, and Recycling 32: 29-41. Buenrostro, O. and G. Bocco. 2003. Solid waste management in municipalities in Mexico: goals and perspectives, Resources, Conservation, and Recycling 39: 251-263. Buenrostro, O., S. Ojeda Benítez, L. Márquez-Benavides. 2007. Comparative analysis of hazard - ous household waste in two Mexican regions, Waste Management and Research 27: 792-801. Buenrostro, O., L. Márquez, and F. Pinette. 2008a. Consumption patterns and household solid waste generation in an urban settlement in México. Waste Management 28: S2-S6. Buenrostro, O., M. Mendoza, E. López, D. Geneletti. 2008b. Analysis of land suitability for the siting of inter-municipal landfills in the Cuitzeo Lake Basin, Mexico. Waste Management 28: 1137-1146. Bulkeley, H and K. Askins. 2009. Waste interfaces: Biodegradable waste, municipal policy and everyday practice. Geographical Journal 175: 251–260. Bulkeley, H. and N. Gregson. 2009. Crossing the threshold: municipal waste policy and house- hold waste generation, Environment and Planning A 41(4): 929-945. Bullard, R. 1992. Environmental blackmail in minority communities. In Race and the incidence of environmental hazards . Eds. B. Bryant and P. Mohai. Westview Press, Boulder, CO. _________. 1994. Dumping In Dixie: Race, Class, and Environmental Quality . Boulder, CO: Westview Press. Burton, I., R. Kates, and G. White. 1978. The Environment as Hazard . NY: Oxford University Press. Butterfield, C. and D. Myers. 1999. The Florida Litter Study: Economic Impacts of Litter on Florida’s Businesses . Gainsville, FL: State University System of Florida. 312 Butzer, K.. 1976. Early Hydraulic Civilization in Egypt: A Study in Cultural Ecology . Chicago, IL: University of Chicago Press. _________. 1989. Cultural Ecology. In Geography in America , eds. C. Wilmott and G. Gaile. NY: Merrill. Byers, D. 1967. Climate and Hydrology. In The Prehistory of the Tehuacan Valley, Volume One, Environment and Subsistence , Ed. Douglas S. Byers, 48-65. Austin, TX: University of Texas Press. Byland, B. and J. Pohl. 1994. In The Realm of Eight Deer: The Archaeology of the Mixtec Codi - ces. Norman, OK: University of Oklahoma Press. Campbell, J. 2001. Participatory rural appraisal as qualitative research: Distinguishing method - ological issues from participatory claims. Human Organization 60(4): 380-389. Carneiro, R. 1960. Slash-and-burn Agriculture: A Closer Look at Its Implications for Settlement Patterns . Indianapolis, IN: Bobbs-Merrill. Carney, J. 1993. Converting the wetlands, engendering the environment: The intersection of gen - der with agrarian change in The Gambia, Economic Geography 69: 329–349. ________. 1996. Rice milling, gender, and slave labor in colonial South Carolina. Past and Pres - ent 153: 108-134 Carruthers, I. and R. Chambers. 1981. Rapid appraisal for rural development. Agricultural Ad - ministration 8(6): 407-422. Cascadia Consulting Group. 2005. Washington 2004 State Litter Study Litter Generation and Composition Report. Seattle, WA: Cascadia Consulting. Castillo-Berthier, H. 1983. La Sociedad del la Basura: Caciquismo en la Ciudad de México . Mexico, D.F.: UNAM. ________________. 2003. Garbage, work, and society. Resources, Conservation and Recycling 39: 193-210. Castree, N. and B. Braun. 2001. Social Nature: Theory, Practice, and Politics . Malden, MA: Blackwell Publishing. CDI (Comisión Nacional para el Desarollo de Pueblos Indígenas). Regiones Indígena de Méxi - co. México, D.F.: CDI. Chambers, R. 1980. Rapid rural appraisal: Rationale and repertoire. IDS Discussion Paper 155. __________. 1994a. The origins and practice of Participatory Rural Appraisal. World Develop - ment 22: 953-969. __________. 1994b. Participatory Rural Appraisal (PRA): Analysis of experience. World Devel - opment 22: 1253-1268. 313 __________. 1994c. Participatory Rural Appraisal (PRA): Challenges, Potentials, and Paradigm. World Development 22:1437-1454. Chance, J. 2003. Haciendas, Ranchos, and Indian Towns: A Case from the Late Colonial Valley of Puebla. Ethnohistory 50 (1): 15-45. Christie, M. 2008. Kitchenspace: Women, Fiestas, and Everyday Life in Central Mexico . Austin, TX: University of Texas Press. Cialdini, R. and D. Baumann. 1981. Littering: A new unobtrusive measure of attitude. Social Psychology Quarterly 44(3): 254-259. Cialdini, R., R. Reno, and C. Kallgren. 1990. A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychol - ogy 6: 1015-1026. Cialdini, R., C. Kallgren, and R. Reno. 1991. A focus theory of normative conduct: A theoretical refinement and reevaluation of the role of norms in human behavior. In Advances in Experimen - tal Social Psychology , Ed. M. Zanna. New York: Academic Press. Clark, J. 1991. The Beginnings of Mesoamerica: Apologia for the Soconusco Early Formative, In The Formation of Complex Society in Southern Mesoamerica . Ed. W.R. Fowler, Jr. Boca Raton, FL: CRC Press. Cockburn, A. and J. Ridgeway. 1979. Political Ecology . NY: New York Times Press. Cointreau, S. 1982. Environmental Management of Solid Wastes in Developing Countries. Wash - ington, D.C.: World Bank Paper. Corral-Verdugo V. 1996. A structural model of reuse and recycling in Mexico. Environment and Behavior 28: 665-696. ______________. 1997. Dual ‘realities’ of conservation behavior: self-reports vs. observations of reuse and recycling behavior. Journal of Environmental Psychology 17: 135-145. ______________. 2003. Situational and personal determinants of waste control practices in northern Mexico: a study of reuse and recycling behaviors. Resources, Conservation and Recy - cling 39 (3): 265-281. Corral-Verdugo V. and L. Armendariz. 2000. The ‘New Environmental Paradigm’ in a Mexican community. Journal of Environmental Education 31: 25-31. Corral-Verdugo, V. and A. Figueredo. 1999. Convergent and divergent validity of three measures of conservation behavior: the multitrate/multimethod approach. Environment and Behavior 31: 848-863. Cresswell, T. 2004. Place: A Short Introduction . Malden, MA: Blackwell. Danielson, S. 2009. Q Method and urveys: Three ways to combine Q and R. Field Methods 21: 219-237. 314 Dasgupta, P. 2005. Q methodology” for Mapping Stakeholder Perceptions in Participatory Forest Management, UK Department for International Development Report. Dávalos, M. 1997. Basura e Ilustración, la limpieza de la Ciudad de México a fines del Siglo XVII . Mexico, D.F.: INAH. Davies, A. 2008. The Geographies of Garbage Governance: Interventions, Interactions, and Outcomes . Aldershot, U.K.: Ashgate. ________. 2009. Clean and green? A governance analysis of waste management in New Zealand. Journal of Environmental Planning and Management 52:157–176. Davies, A and A. O’Callaghan-Platt. 2009. Does money talk? Waste charging in the Republic of Ireland: Government, governance and performance. Journal of Environmental Policy and Plan - ning 10: 271–287. Dávila, P., M. del Coro Arizmendi, A. Valiente-Banuet, J.L. Villaseñor, A. Casas, R. Lira. 2002. Biological diversity in the Tehuacán-Cuicatlán Valley, Mexico. Biodiversity and Conservation 11: 421-442. Davoudi, S .2009. Scalar tensions in the governance of waste: the resilience of state spatial Keynesianism. Journal of Environmental Planning and Management 52: 137–156. Davoudi, S. and N. Evans. 2005. The challenge of governance in regional waste planning. Envi - ronment and planning C: Government and Policy , 23(4): 493–519. De Ita, A. 2006. Land Concentration in Mexico after PROCEDE. In Promised Land: Competing Visions of Agrarian Reform , Eds. P. Rossett, R. Patel, and M. Courville. Oakland, CA: Food First Books. De Kort, Y., T. McCalley, and C. Midden. 2008. Persuasive trash cans: Activation of littering norms by design. Environment and Behaviour 40: 870-891. Deal, M. 1985. Household pottery disposal in the Maya Highlands: An ethnoarchaeological in - terpretation. Journal of Anthropological Archaeology 4: 243-291. Del Castillo V., G. 1996. NAFTA and the Struggle for Neoliberalism: Mexico’s Elusive Quest for First World Status, in Neoliberalism Revisited: Economic Restructuring and Mexico’s Political Future . Ed. G. Otero. Boulder, CO: Westview Press. Delgado-Wise, R. and J. Cypher. 2007. The Strategic Role of Mexican Labor under NAFTA: Critical Perspectives on Current Economic Integration. The Annals of the American Academy of Political and Social Science 610: 119-142 Demeritt, D. 1998. Science, social constructivism and nature. In Remaking Reality: Nature at the Millennium . Eds. B. Braun and N. Castree. NY: Routledge. __________. 2001. The construction of global warming and the politics of science. Annals of the Association of American Geographers 91(2): 307-337. 315 Denevan, W. 1961. The Upland Pine Forests of Nicaragua. A Study in Cultural Plant Geography. University of California Publications in Geography 12: 251-320. _________. 1983. Adaptation, variation, and cultural geography. Professional Geographer 35: 399-407. _________. 1992. The pristine myth. The landscape of the Americas in 1492. Annals of the As - sociation of American Geographers 82 (3): 369-385. Denevan, W. and K. Schwerin. 1978. Adaptive strategies in Karinya sunsistence, Venezuelan Llanos. Antropológica 50: 3-91. Dennis, P. 1987. Intervillage Conflict in Oaxaca. New Brunswick, NJ: Rutgers University Press. Diario Oficial de la Federación. 1988. Ley General del Equilibrio Ecológico y la Protección al Ambiente . Dillehay, T. 2000. Settlement of the Americas: A New Prehistory . NY: Basic Books. Doolittle, W. 1984. Agricultural Change As an Incremental Process. Annals of the Association of American Geographers 74(1): 124-137. __________. 1990. Canal Irrigation in Prehistoric Mexico: The Sequence of Technological Change . Austin, TX: University of Texas Press. __________. 1992. Agriculture in North America on the eve of contact: A reassessment. Annals of the Association of American Geographers 82 (3): 386–401. __________. 2004. Gardens are Us, we are nature: Transcending antiquity and modernity. Geo - graphical Review 94 (3): 391-405. Doolittle, W., A. Sluyter, E. Perramond, P. Crossley, and D. Lamber 2002. Feeding a Growing Population on an Increasingly Fragile Environment. In Latin America in the Twenty-First Cen - tury: Challenges and Solutions . Ed. G. Knapp. Austin, TX: Conference of Latin Americanist Geographers and the University of Texas Press. Doss, C., J. McPeak, and C.B. Barrett. 2008. Interpersonal, Intertemporal and Spatial Variation in Risk Perceptions: Evidence from East Africa. World Development 36: 81453–1468. Douglas, M. 1966. Purity and Danger: An Analysis of Concepts of Pollution and Taboo . NY: Routledge and Kegan Paul. Drackner, M. 2005. What is waste? To whom? – An anthropological perspective on garbage. Waste Management and Research 23: 175-181. Duncan, J. 1980. The Superorganic in American Cultural Geography. Annals of the Association of American Geographers 70: 181-198. Dunlap, R., Gallup, G., and Gallup, A. 1993. Global environmental concern: Results from an international public opinion survey. Environment 35 (7): 33-39. 316 Dunlap R. and R. Scarce. 1991. The polls, poll trends: environment problems and protection. Public Opinion Quarterly 55: 651–672. Dunning, N. and T. Beach. 1994. Soil erosion, slope management, and ancient terracing in the Maya Lowlands. Latin American Antiquity 5: 51-69. Durand, J., D. Massey, and F. Charvet. 2000. The changing geography of Mexican immigration to the United States: 1910-1996. Social Science Quarterly 81(1): 1-15. Durdan, C., G. Reeder, and P. Hecht. 1985. Litter in a university cafeteria: Demographic data and the use of prompts as an intervention strategy. Environment and Behavior 16: 387-404. Eden, S. 1998. Environmental issues: knowledge, uncertainty and the environment, Progress in Human Geography 22: 425-432. Eden, S., A. Donaldson, and G. Walker. 2005. Structuring subjectivities? Using Q-methodology in human geography. Area 37(4): 413-422. Eisenstadt, J. 2004. Catching the State Off Guard: Electoral Courts, Campaign Finance, and Mexico’s Separation of State and Ruling Party. Party Politics 10(6): 723-745. __________. 2007. Usos y Costumbres and Postelectoral conflicts in Oaxaca, Mexico, 1995- 2004: An empirical and normative assessment. Latin American Research Review 42(1): 52-77. Enge, K. and S. Whiteford. 1989. The Keepers of Water and Earth: Mexican Rural Social Orga - nization and Irrigation . Austin, TX: University of Texas Press. Ericson, J. 2006. A participatory approach to conservation in the Calakmul Biosphere Reserve, Campeche, Mexico. Landscape and Urban Planning 74 (3–4): 242–266. Eriksen, M., E., N. Maximenko, M. Thiel, A. Cummins, G. Lattin, S. Wilson, J. Hafner, A. Zellers, S. Rifman. 2013. Plastic pollution in the South Pacific subtropical gyre. Marine Pollution Bulletin 68: 71-76. Escobar, A. 1996. Constructing Nature: Elements for a poststructural political ecology, in Libera - tion Ecologies: Environment, Development, and Social Movements . Eds. R. Peet and M. Watts. NY: Routledge. _________. 1999. After Nature: Steps to an Antiessentialist Political Ecology. Current Anthro - pology 40 (1): 1-30 Espinosa, R.M., S. Turpin, G. Polanco, A. De la Torre, I. Delfin, I. Raygoza. 2008. Integral urban solid waste management program in a Mexican university, Waste Management and Research 28: S27-S32. Evans, S. 2004. Ancient Mexico and Central America: Archaeology and Culture History . NY: Thames and Hudson. Ewing, G. 2001. Altruistic, egoistic, and normative effects on curbside recycling. Environment and Behavior 33: 733–764. 317 Fairweather, J. and S. Swaffield. 2000. Q method Photographs to Study Perceptions of the En- vironment in New Zealand. In Social Discourse and Environmental Policy: An Application of Q Methodology . Eds. H. Addams and J. Proops. Northampton, MA: Edward Elgar Publishing. Fergutz, O., S. Dias, and D. Mitlin. 2011. Developing urban waste management in Brazil with waste picker organizations. Environment and Urbanization 23(2): 597-608. Fernández-Crispín, A. 2009. La construcción de una Cultura Ambiental mediante la Educación formal en Puebla (México). Utopía y Praxis Latinoamericana 44: 131-136. Fielding, W. 2008. Dogs: A continuing and common neighborhood nuisance of New Providence, The Bahamas. Society and Animals 16: 61-73. Finnie, W. 1973. Field experiments in litter control. Environment and Behavior 5: 123-144. Fitting E. 2006. Importing corn, exporting labor: the neoliberal corn regime, GMOs, and the ero- sion of Mexican biodiversity. Agriculture and Human Values 23: 15–26 Flaks, J. 2010. The Culture of the Good Death in Seventeenth-Century Mexico City . Unpublished Ph.D. Dissertation, University of Nevada-Reno. Flannery, K. 1967. The vertebrate fauna and hunting patterns. In The Prehistory of the Tehuacan Valley, Volume One, Environment and Subsistence , Ed. Douglas S. Byers, Austin, TX: University of Texas Press. Flynn, J., P. Slovic, and C. Mertz. 2006. Gender, race, and perception of environmental health Risks. Risk Analysis 14(6): 1101-1108. Foody, G., E. Ghoneim, N. Arnell. 2004. Predicting locations sensitive to flash flooding in an arid environment. Journal of Hydrology 292: 48-58. Forbes, G. 2003. Reducing Litter on Roadsides: A Synthesis of Highway Practice, NCHRP Syn - thesis 394 . Washington, D.C.: Transportation Research Board. Forsyth, T. 2005. Enhancing climate technology transfer through greater public-private coopera - tion: Lessons from Thailand and the Philippines. Natural Resources Forum 29: 165-176 Fosberg, F. 1963. The Island Ecosystem. In Man’s Place in the Island Ecosystem. Ed. F. Fosberg. Honolulu, HI: Bishop Museum. Funes Izaguirre, J.L. 2004. Una perspectiva jurídico ambiental de los residuos sólidos munici - pales. In La Basura en una perspectiva multidisciplinaria . Ed. J. Guevara Martínez. Mexico, D.F.: Unlisted press. Gade, D. 2008. Irreverent Musings on the Dissertation in Latin American Geography. In Eth - nogeographic Research in Latin America: Essays Honoring William V. Davidson . Eds. Peter H. Herlihy, Kent Mathewson, and Craig Revels. Baton Rouge, LA: Geoscience Publications. Galván-Meráz, J. 2008. El Prontuario sobre Legislación de Residuos en México . México D.F.: Ediciones Arlequín. 318 Gates, M. 1996. The Debt Crisis and Economic Restructuring: Prospects for Mexican Agricul- ture. In Neoliberalism Revisited: Economic Restructuring and Mexico’s Political Future , Ed. G. Otero, 43-62. Boulder, Colo.: Westview Press. Geertz, C. 1963. Agricultural Involution: The Processes of Ecological Change in Indonesia . Berkeley, CA: University of California Press. Geller, E., J. Witmer, and M. Tuso. 1977. Environmental interventions for litter control. Journal of Applied Psychology 62: 344-351. Geller, E. S., W. Brasted, and M. Mann. 1980. Waste receptacle designs as interventions for litter control. Journal of Environmental Systems 9: 145-160. Giddens, A., 1990. The Consequences of Modernity . Cambridge, UK: Polity Press. Girdner, E. and J. Smith. 2002. Killing Me Softly: Toxic Waste, Corporate Profit and the Struggle for Environmental Justice . NY: Monthly Review Press. Glasco, S. 2010. Constructing Mexico City: Colonial Conflicts over Culture, Space, and Author- ity . NY: Palgrave Macmillan. Gledhill, J. 1995. Neoliberalism, Transnationalization and Rural Poverty, A Case Study of Mi - choacán, Mexico . Boulder, CO: Westview Press. Gobierno Federal de los Estados Unidos Mexicanos. 1917. Constitución Política de los Estados Unidos Mexicanos . Goldring, L. 1996. The Changing Configuration of Property Rights Under Ejido Reform, in Re - forming Mexico’s Agrarian Reform , ed. L. Randall. Armonk, NY: M.E. Sharp Gómez, G., M. Meneses, L. Ballinas, and F. Castells. 2008. Characterization of urban solid waste in Chihuahua, Mexico, Waste Management 28: 2465-2471. Gómez-Pompa, A. and A. Kaus. 1992. Taming the wilderness myth. Bioscience 42(4): 271-279. Gonzales, M. 2002. The Mexican Revolution, 1910-1940 . Albuquerque, NM: University of New Mexico Press. González, R. 2001. Zapotec Science: Food and Farming in the Northern Sierra of Oaxaca . Berkeley, CA: University of California Press. Grandstaff, T. and S. Grandstaff. 1987. A conceptual basis for methodological development in rapid rural appraisal. In Proceedings of the 1985 International Conference on Rapid Rural Ap - praisal. Khon Kaen, Thailand: Khon Kaen University Press. Gravel, N. 2007. Mexico smallholders adrift: The urgent need for a new social contract in rural Mexico, Journal of Latin American Geography 6(2): 77-98. Green, R. 2005. Community perceptions of environmental and social change and tourism devel- opment on the island of Koh Samui, Thailand, Journal of Environmental Psychology 25: 37-56. 319 Grossman, L. 1977. Man-Environment relations in anthropology and geography. Annals of the Association of American Geographers 67(1): 126-144. __________. 1993. The political ecology of banana exports and local food production in St. Vin - cent, Eastern Caribbean. Annals of the Association of American Geographers 83(2): 347-367. Guarnaccia, P., T. Vivar, A.Bellows, and G. Alcaraz. ‘We eat meat every day’: ecology and economy of dietary change among Oaxacan migrants from Mexico to New Jersey. Ethnic and Racial Studies 35(1): 104-119. Guevara-Martínez, J. and A. Fernández-Crispín. 2009. Experiencias de investigación en Edu - cación Ambiental . Puebla, Mexico: UPAEP. Guevara-Martínez, J., F. Quintanar-Olguín, and C. Rodríguez-Álvarez. 2004. Sociedad orientada ambientalmente: actitudes Ambientales de habitants de un vecindario de Puebla. In La Basura en una perspectiva multidisciplinaria . Ed. J. Guevara Martínez. Mexico, D.F.: Unlisted press. Gúzman-Chávez, M. and C. Macías-Manzanares. 2012. El manejo de los residuos sólidos munic- ipales: un enfoque antropológico. El caso de San Luis Potosí, México. Estudios Sociales 20(39): 235-261. Haenn, N. 2006. The changing and enduring ejido: a state and regional examination of Mexico’s land tenure counter-reforms, Land Use Policy 23: 136-146. Hansmann, R., and R. Scholz. 2003. A two-step informational strategy for reducing littering be - haviour in a cinema. Environment and Behaviour 35 (6): 752-762. Hardy, K. 1996. The Preceramic sequence from the Tehuacán Valley: A Reevaluation. Current Anthropology 37(4): 700-716. Harrison, C. and J. Burgess. 1994. Social constructions of nature: A case study of conflicts over the Development of Rainham Marshes, Transactions of the Institute of British Geographers 19(3): 291-310. Hartmann, C. 2012. Uneven urban spaces: Accessing trash in Managua, Nicaragua. Journal of Latin American Geography 11(1): 143-163. Harvey, D. 2005. A Brief History of Neoliberalism . London: Oxford University Press. Hawkins, G. 2003. Down the Drain: Shit and the Politics of Disturbance. In Culture and Waste: The Creation and Destruction of Value . Eds. G. Hawkins and S. Muecke. Lanham, MD: Rowman and Littlefield. Hawthorne, T., J. Krygier, and M.P. Kwan. 2008. Mapping ambivalence: Exploring the geog - raphies of community change and rails-to-trails development using photo-based Q method and PPGIS, Geoforum 39: 1058-1078. Hayden, B. and A. Cannon. 1983. Where the garbage goes: refuse disposal in the Maya High - lands. Journal of Anthropological Archaeology 2: 117-163. 320 Hecht, S. and A. Cockburn. 1989. The Fate of the Forest: Developers, Destroyers and Defenders of the Amazon . London: Verso. Herlihy, P. 2003. Participatory Research Mapping of indigenous lands in Darién, Panama. Hu - man Organization 62(4): 315-331. Herlihy, P. and A. Leake. 1997. Participatory Research Mapping of Indigenous Lands in the Hon - duran Mosquitia. In Demographic Diversity and Change in the Central American Isthmus . A. R. Pebley and L. Rosero-Bixby, eds. Pp. 707-736. Santa Monica, CA: Rand Books. Herlihy, P. and G. Knapp. 2003. Maps of, by, and for the peoples of Latin America. Human Or - ganization 62(4): 303-314. Herlihy, P. and F. Wiseman. 2008. Ethnogeography of the Dooryard Orchard-Garden of the Yucatecan Maya, in Ethnogeographic Research in Latin America: Essays Honoring William V. Davidson . Eds. P. Herlihy, K. Mathewson, and C. Revels. Baton Rouge, LA: Geoscience Publi- cations. Hernández-Cendejas, G. 2012. Las Transformaciones Agrárias y el Impacto del PROCEDE en la Huasteca Potosina. Un Analisis Multiescalar . Unpublished Ph.D. Dissertation, Universidad Nacional Autónoma de México. Hernández, O., B. Rawlins, and R. Schwartz. 1999. Voluntary recycling in Quito: factors associ - ated with participation in a pilot programme. Environment and Urbanization 11(2): 145-160. Herrera-Rodríguez, M. 2012. Social change and land tenure regimes in Mexico. Geojournal 77(5): 633-649. Heynan, N., M. Kaika, and E. Swyngedouw. 2006. In the Nature of Cities: Urban political ecol - ogy and the politics of urban metabolism . NY: Routledge. Heywood, J. 2002. The cognitive and emotional components of behavior norms in outdoor recre - ation. Leisure Sciences 24: 271–281. Hill, A. 1964. The Changing Landscape of a Mexican Municipio Villa Las Rosas, Chiapas . Chi- cago, IL: University of Chicago Press. Hirth, K. 1993. The Household as an Analytical Unit: Problems in Method and Theory. In Pre - hispanic Domestic Units in Western Mesoamerica: Studies of the Household, Compound, and Residence . Eds. R. Santley and K. Hirth. Boca Raton, FL: CRC Press. Hooker-Clarke, A. 2001. Beliefs regarding society and nature: A framework for listening in for - est and environmental policy. Operant Subjectivity 24: 159-182 Horsley, A. 1988. The unintended effects of a posted sign on littering attitudes and stated inten - tions. Journal of Environmental Education 19(3): 10-14. Huffman, K., W. Grossnickle, J. Cope, and K. Huffman. 1995. Litter reduction: A review and integration of the literature. Environment and Behaviour 27(2): 153-183. Humes, E. 2012. Garbology: Our Dirty Love Affair with Trash . NY: Penguin Press. 321 Hunter, L., A. Hatch, and A. Johnson. 2004. Cross-national gender variation in environmental behaviors. Social Science Quarterly 85: 677-694. Huysman, M. 1994. Waste picking as a strategy for women in Indian cities. Environment and Urbanization 6: 155–174. INEGI (Instituto Nacional de Estadística, Geografía, e Informática). 1990, XI Censo General de Población y Vivienda . Aguascalientes, Mexico: Instituto Nacional de Estadística, Geografía, e Informática. _____________________________________________________. 1995. Conteo de Población y Vivienda . Aguascalientes, AGS: Instituto Nacional de Estadística, Geografía, e Informática. _____________________________________________________. 1999. Carta Estatal Geológi - ca – Puebla . Aguascalientes, AGS: Instituto Nacional de Estadística, Geografía, e Informática. _____________________________________________________. 2000. XII Censo General de Población y Vivienda . Instituto Nacional de Estadística, Geografía, e Informática. _____________________________________________________. 2002. Encuesta Nacional de Uso del Tiempo 2002 . Aguascalientes, AGS: Instituto Nacional de Estadística, Geografía, e Informática. _____________________________________________________. 2005. Conteo 2005 . Aguas- calientes, AGS: Instituto Nacional de Estadística, Geografía, e Informática. _____________________________________________________. 2007a. Diccionario de datos de Uso del Suelo y Vegetación 1:250,000 (Vectorial) . Aguascalientes, AGS: Instituto Nacional de Estadística, Geografía, e Informática. _____________________________________________________. 2007b. Censo Agropecuario y Ejidal 2007. Aguascalientes, AGS: Instituto Nacional de Estadística, Geografía, e Informática. ______________________________________________________. 2009. Conjunto de Datos Vectoriales de la Carta de Uso de Suelo y Vegetación, Escala 1:250,000 Serie IV . Aguascalien- tes, AGS: Instituto Nacional de Estadística, Geografía, e Informática. ______________________________________________________. 2010. Censo de Población y Vivienda 2010 . Aguascalientes, AGS: Instituto Nacional de Estadística, Geografía, e Informática. Inglehart, R. 1995. Public support for environmental protection: Objective problems and subjec - tive values in 43 societies, Political Science and Politics 28: 57-72. Ishiyama, N. 2003. Environmental justice and American Indian tribal sovereignty: Case study of a land-use conflict in Skull Valley, Utah. Antipode 35: 119–139. Ivar do Sul, J., and M. Costa. 2007. Marine debris review for Latin America and the wider Carib - bean Region: From the 1970s until now, and where do we go from here? Marine Pollution Bul - letin 54: 1087–1104. 322 Ivar do Sul, J., A. Spengler, and M. Costa. 2009. Here, there and everywhere. Small plastic frag - ments and pellets on beaches of Fernando de Noronha (Equatorial Western Atlantic). Marine Pollution Bulletin 58: 1236–1238. Ivar do Sul, J., I. Rodrigues-Santos, A. Friedrich, A. Mathiensen, G. Fillmann. 2011. Plastic Pol- lution at a Sea Turtle Conservation Area in NE Brazil: Contrasting Developed and Undeveloped Beaches. Estuaries and Coasts 34: 814-823. Jacobs, J. 2002. Community participation, the environment, and democracy: Brazil in compara - tive perspective. Latin American Politics and Society 44: 59-88. Jepson, W., C. Brannstrom, and N. Persons. 2012. ‘‘We Don’t Take the Pledge’’: Environmental - ity and environmental skepticism at the epicenter of US wind energy development. Geoforum 43: 851-863. Johannessen, C. 1963. Savannas of Interior Honduras. Iberoamericana 46: Los Angeles, CA: University of California Press. Johnson, F. and R.S. MacNeish. 1972. Chronometric dating. In The Prehistory of the Tehuacan Valley, Vol 4: Chronology and Irrigation . Ed. F. Johnson. Austin, TX: University of Texas Press. Joseph, K. 2006. Stakeholder participation for sustainable waste management. Habitat Interna - tional 30: 863-871. Kates, R. 1962. Hazard and Choice Perception in Flood Plain Management , University of Chi - cago, Department of Geography Research, Paper No. 78. _______. 1971. Natural hazard in human ecological perspective: Hypotheses and models. Eco - nomic Geography 47: 438-451. Keil, R. 2005. Urban political ecology. Urban Geography 26 (7): 640–651. Kelling, G. and C. Coles. 1996. Fixing Broken Windows: Restoring Order and Reducing Crime in Our Communities . NY: Touchstone. Kelly, J. 2013. Village-Scale Practices and Water Sources in Indigenous Mexico after the Neolib - eralizing of Social Property . Unpublished Ph.D. Dissertation, The University of Kansas. Kelly, J., P. Herlihy, A. Ramos Viera, A. Hilburn, D. Smith, and G. Hernández-.Cendejas. 2010. Indigenous territoriality at the end of the social property era in Mexico: The case of the Huasteca Potosina. Journal of Latin American Geography 9(3): 161-181. Kelly, K. 1973. Garbage: The History and Future of Garbage in America . NY: Saturday Review Press. Kennedy, G. 2007. An Ontology of Trash: The Disposable and Its Problematic Nature . Albany, NY: SUNY Press. Keys, E. 1999. Kaqchikel gardens: women, children, and multiple roles of gardens among the Highland Maya of Guatemala. Yearbook, Conference of Latin Americanist Geographers 25: 89- 100. 323 ______. 2005. Market intermediaries link farms to markets: southeastern Mexican examples. Geographical Review 95(1): 24-46 Killion, T. 1990. Cultivation intensity and residential site structure: An ethnoarchaeological ex - amination of peasant agriculture in the Sierra de los Tuxtlas, Veracruz, Mexico. Latin American Antiquity 1(3): 191-215. Kimber, C. 1966. Dooryard gardens of Martinique. Yearbook of the Association of Pacific Coast Geographers 28: 97-118. _________. 1973. Spatial patterning in the dooryard gardens of Puerto Rico. Geographical Re - view 63 (1): 6-26 _________. 2004. Gardens and dwelling: People in vernacular gardens. Geographical Review 94 (3): 263-283. Klimpel, S., J. Heukelbach, D. Pothmann, and S. Rückert. 2010. Gastrointestinal and ectopara- sites from urban stray dogs in Fortaleza (Brazil): high infection risk for humans? Parasitology Research 107:713-719. Klooster, D. 2002. Toward adaptive community forest management: Integrating local forest knowledge with scientific forestry. Economic Geography 78(1): 43-70. _________. 2003. Campesinos and Mexican forest policy during the 20th Century. Latin Ameri - can Research Review 38(2): 94-126. Knapp, G. 1991. Andean Ecology: Adaptive Dynamics in Ecuador . Boulder, CO: Westview Press. Krauze, E. 1998. Mexico: Biography of Power. A History of Modern Mexico, 1810-1996 . NY: HarperPerennial. Kristeva, J. 1982. Powers of Horror: An Essay on Abjection . New York: Columbia University Press. Kurtz, H. 2003. Scale frames and counter-scale frames: Constructing the problem of environ - mental injustice. Political Geography 22: 887–916. _______. 2005. Alternative visions for citizenship practice in an environmental justice dispute. Space and Polity 9: 77–91. Leatherman, T. and A. Goodman. 2005. Coca-colonization of diets in the Yucatan. Social Science and Medicine 61: 833-845. Lee, A. 2006. Economic Crisis and the Incorporation of New Migrant Sending Areas in Mexico: The Case of Zapotitlán Salinas, Puebla. Working Paper 136 : The Center for US/Mexican Studies and Center for Comparative Immigration Studies. _______. 2007. “Para salir adelante”: The emergence and acceleration of international migration in new sending areas in Puebla, Mexico. Journal of Latin American and Caribbean Anthropology 13(1): 48-78. 324 LeFebvre, H. 1991. The Production of Space . Oxford, U.K.: Blackwell. Leighly, J. 1987. Ecology as metaphor: Carl Sauer and human ecology. Professional Geographer 39: 405-412. Leitner, H., K. Sziarto, E. Sheppard, and A. Maringanti. 2007. Contesting urban futures: Decen - tering neoliberalism. In Contesting Neoliberalism: Urban Frontiers . Eds. H. Leitner, J. Peck, and E. Sheppard. New York: Guilford Press Lemieux, P., B. Gullett, C. Lutes, C. Winterrowd, and D. Winters. 2003. Variables affecting emis- sions of PCDD/Fs from uncontrolled combustion of household waste in barrels. Journal of the Air and Waste Management Association 53(5): 523-531 Lepawsky, J. and C. Mather. 2011. From beginnings and endings to boundaries and edges: re - thinking circulation and exchange through electronic waste. Area 43(3): 242-249. Lepawsky, J. and C. McNabb. 2010. Mapping international flows of electronic waste. The Cana - dian Geographer 54(2): 177-195. Lewis, J. 2002. Agrarian change and privatization of ejido land in Northern Mexico, Journal of Agrarian Change 2(3): 401-419. Liggett, R., A. Loukaitou-Sideris, and H. Iseki. 2001. Bus Stop – Environment Connection: Do Characteristics of the Built Environment Correlate with Bus Stop Crime? Washington, D.C.: Transportation Research Board of the National Academies. Liverman, D. 1990. Drought impacts in Mexico: Climate, agriculture, technology, and land ten - ure in Sonora and Puebla. Annals of the Association of American Geographers 80(1): 49-72. Liverman, D. and. S. Vilas. 2006. Neoliberalism and the environment in Latin America. Annual Review of Environmental Resources 31: 327-363. Lowenthal, D. 1967. Environmental Perception and Behavior . Chicago: University of Chicago, Department of Geography Research Paper No. 109. Luton, L. 1996. The Politics of Garbage: A Community Perspective on Solid Waste Policy Mak - ing. Pittsburgh, PA: University of Pittsburgh Press. Lynch, D. and C. Hutchinson. 1992. Environmental education. Proceedings of the National Acad - emy of Sciences 89: 864-867. Maantay, J. 2006. Power, justice, and the environment: A critical appraisal of the environmental justice movement. Urban Geography 27: 198–200. MacGregor Tan Research. 2012. Keep Australia Beautiful: National Litter Index 2011/2012 . Frewville, SA, Australia: MacGregor Tan Research. MacNeish, R. 1964. Ancient Mesoamerican civilization. Science 143 (3606): 531-537. 325 ___________. 1972. Summary of the Cultural Sequence and Its Implications in the Tehuacan Valley, in The Prehistory of the Tehuacan Valley, Volume 5 , ed R. MacNeil. Austin, TX: Univer - sity of Texas Press. Maldonado, L. 2006. The economics of urban solid waste reduction in educational institutions in Mexico: A 3-year experience. Resources, Conservation and Recycling 48: 41-55. Maldonado-Koerdell, M. 1964. Geohistory and Palaeogeography of Middle America. In Hand - book of Middle American Indians, Vol. 1. Natural Environment and Early Cultures , Eds. R. Wauchope and R.C. West, 3-32. Austin, TX: University of Texas Press. Maquila Solidarity. 2003. Tehuacan: blue jeans, blue waters and worker rights. Toronto, ON: Maquila Solidarity Network. Marcus, J. and K. Flannery. 1996. Zapotec Civilization: How Urban Society Evolved in Mexico’s Oaxaca Valley . NY: Thames and Hudson. Márquez, M., S. Ojeda, and H. Hidalgo. 2008. Identification of behavior patterns in household solid waste generation in Mexicali’s city: Study case. Resources, Conservation, and Recycling 52: 1299-1306. Martin, G. and P. James. 1993. All Possible Worlds, A History of Geographical Ideas . NY: Wiley and Sons. Mathewson, K. 1990. Rio Hondo Reflections: Notes on Puleston’s Place and the Archaeology of Maya Landscapes. In Ancient Maya Wetland Agriculture: Excavations on Albion Island, North - ern Belize . Ed. M. D. Pohl. Boulder, CO: Westview Press. McKeown, B. and D. Thomas. 1988. Q Methodology . London: Sage. McStay, J. and R. Dunlap. 1983. Male–female differences in concern for environmental quality. International Journal of Women’s Studies 6(4): 291-301. Medina, M. 1997. Supporting scavenger co-ops. BioCycle 38(6): 45-48. _________. 1998. Border scavenging: a case study of aluminum recycling in Laredo, TX and Nuevo Laredo, Mexico. Resources, Conservation and Recycling 23: 107-126. _________. 2000. Scavenger cooperatives in Asia and Latin America. Resources, Conservation and Recycling 31(1): 51-69. _________. 2005. Serving the unserved: informal refuse collection in Mexico, Waste Manage - ment Research 23: 390-397. Meeker, F. 1997. A comparison of table littering behavior in two settings: A case for a contex- tual research strategy. Journal of Environmental Psychology 17: 59-68. Melosi, M. 2005. Garbage in the Cities: Refuse Reform and the Environment, 2 nd Ed . Pittsburgh, PA: University of Pittsburgh Press. 326 ________. 2008. The Sanitary City: Environmental Services in Urban America from Colonial Times to the Present, 2 nd Ed . Pittsburgh, PA: University of Pittsburgh Press. Meltzer, D. 2010. First Peoples in a New World: Colonizing Ice Age America . Berkeley, CA: University of California Press. Moore, S. 2008. The Politics of Garbage in Oaxaca, Mexico, Society and Natural Resources 21: 597-610. _______. 2009. The excess of modernity: Garbage politics in Oaxaca, Mexico, The Professional Geographer 61(4): 426-437. _______. 2011. Global garbage: waste, trash trading, and local garbage politics. In Global Politi - cal Ecology . Eds. R. Peet, M. Watts, and P. Robbins, Pp. 133-144. _______. 2012. Garbage matters: Concepts in new geographies of waste. Progress in Human Geography 36 (6): 780-799. Moreno-Sánchez, R. and J. Maldonado. 2006. Surviving from garbage: the role of informal waste-pickers in a dynamic model of solid-waste management in developing countries. Environ - ment and Development Economics 11(3): 371-391. Moriwaki, H., S. Kitajima, and K. Katahira. 2009. Waste on the roadside, ‘poi-sute’ waste: Its distribution and elution potential of pollutants into environment. Waste Management 29: 1192- 1197. MSW Consultants. 2009. National Visible Litter Survey and Litter Cost Study, Final Report . Stamford, CT: MSW Consultants. Mundy, B. 1996. The Mapping of New Spain, Indigenous Cartography and the Maps of the Rela - ciones Geográficas. Chicago, IL: University of Chicago Press. Muñoz-Cadena, C.E., F.J. Arenas-Huertero, E. Ramón-Gallegos. 2008. Comparative analysis of the street generation of inorganic urban solid waste (IUSW) in two neighborhoods of Mexico City, Waste Management . Murray, M. 2009. Waste management in Ireland: discourses of domination in an (un)reflexive society. The Sociological Review 57(1): 81-101. Myers, G. 2005. Disposable Cities: Garbage, Governance and Sustainable Development in Ur - ban Africa . Burlington, VT: Ashgate Publishing. Nadal, A. 2001. The environmental and social impacts of economic liberalization on corn pro - duction in Mexico . London, UK: Oxfam. Nchito, W. and G. Myers. 2004. Four caveats for participatory solid waste management in Lu- saka, Zambia. Urban Forum 15(2): 109-122. Netting, R. 1968. Hill Farmers of Nigeria: Cultural Ecology of the Kofyar of the Jos Plateau . Seattle, WA: University of Washington Press. 327 Netting, R. 1986. Cultural Ecology . Prospect Heights, IL: Waveland Press. Neumann, R. 2010. Political ecology II: theorizing region. Progress in Human Geography 34(3): 368-374. Nietschmann, B. 1973. Between Land and Water: The Subsistence Ecology of the Miskito Indi - ans, Eastern Nicaragua . NY: Seminar Press. Njeru, J. 2006. The urban political ecology of plastic bag waste problem in Nairobi, Kenya. Geo - forum 37: 1046-1058. Nuitjen, M. 2003. Family Property and the Limits of Intervention: The Article 27 Reforms and the PROCEDE Programme in Mexico, Development and Change 34(3): 475-497. NuStats. 2009. Texas Department of Transportation 2009 Visible Litter Study. http://www.dont - messwithtexas.org/docs/DMWT_2009_Visible_Litter_Full_Report.pdf. Accessed Online: 6-3- 2013. OECD. 2012. OECD Environmental Date Compendium: Waste. Retrieved October 3, 2012. Ojeda-Benítez, S., C. Armijo de Vega, M. Ramírez-Barreto. 2000. The potential for recycling household waste: a case study from Mexicali, Mexico. Environment and Urbanization 12(2): 163-173. ________________________________________________. 2002. Formal and informal recov - ery of recyclables in Mexicali, Mexico: handling alternatives. Resources, Conservation, and Recycling 34(4): 273-288. ________________________________________________. 2003a. Characterization and quanti - fication of household solid wastes in a Mexican city, Resources, Conservation, and Recycling 39: 211-222. Ojeda-Benítez, S. and J. Beraud-Lozano. 2003. The municipal solid waste cycle in Mexico: final disposal. Resources, Conservation, and Recycling 39: 239-250. Ojeda-Benítez, S., C. Armijo de Vega, and M. Marquez-Montenegro. 2008. Household solid waste characterization by family socioeconomic profile as a unit of analysis, Resources Conser - vation, and Society 52: 992-999. Ojeda-Benítez, S., G. Lozano-Olvera, R. Morelos, and C. Armijo de Vega. 2008. Mathematical modeling to predict residential solid waste generation. Waste Management 28: S7-S13. Ojedokun, O. 2011. Attitude towards littering as a mediator of the relationship between personal - ity attributes and responsible environmental behavior. Waste Management 31(12): 2601-2611. Oosterveer, P. 2009. Urban environmental services and the state in East Africa; between neo- developmental and network governance approaches. Geoforum 40: 1061–1068. Otero, G. 1996. Neoliberal Reform and Politics in Mexico: An Overview. In Neoliberalism Rev - isted: Economic Restructuring and Mexico’s Political Future . Ed. G. Otero. Boulder, CO: West- view Press. 328 Pandey, J. 1990. The environment, culture, and behavior. In Applied cross-cultural psychology. Ed. R. Brislin. Thousand Oaks, CA: SAGE. Park, P. 1993. What is Participatory Research? A Theoretical and Methodological Perspective. In Voices of Change: Participatory Research in the United States and Canada . Peter Park, Mary Brydon-Miller, Budd Hall, and Ted Jackson, eds. Pp. 1-19. Westport, CT: Bergin and Garvey. Parker, W. and C. Pederzini. 2000. Gender Differences in Education in Mexico . NY: World Bank. Parsons, J. 1955. The Miskito Pine Savanna of Nicaragua and Honduras. Annals of the Associa - tion of American Geographers 45: 36-63. Parsons, J. and W. Denevan. 1967. Pre-Columbian Ridged Fields. Scientific American 217: 92- 101. Peck, J. and A. Tickell. 2002. Neoliberalizing Space, in Space of Neoliberalism: Urban Restruc - turing in North America and Europe . Eds. N. Brenner and N. Theodore. Malden, MA: Blackwell. Peet, R. and M. Watts. 1996. Liberation Ecologies: Environment, Development, Social Move - ments. London, U.K.: Routledge Press. Peet R., M. Watts, and P. Robbins. 2011. Global Political Ecology . NY: Routledge. Pedrero, M., T. Rendón, and A. Barrón. 1997. Segregación ocupacional por género en México . Cuernavaca, MOR., MX: UNAM-CRM. Pelling, M. 2003. The Vulnerability of Cities: Natural Disasters and Social Resistance . London: Earthscan. Pellow, D. 2002. Garbage Wars: The Struggle for Environmental Justice in Chicago . Cambridge, MA: MIT Press. Perramond, E. 2008. The rise, fall, and reconfiguration of the Mexican ejido, Geographical Re - view 98(3): 356-371. ___________. 2010. Political Ecologies of Cattle Ranching in Northern Mexico: Private Revolu - tions. Tucson, AZ: University of Arizona Press. Philips, D., I. Restrepo, and W. Rathje. 1984. El Proyecto Basura: The archaeology of industrial transformation in Mexico. The American Behavioral Scientist 28(1): 139-160. Powers, R., J. Osborne, and E. Anderson. 1973. Positive reinforcement of litter removal in the natural environment. Journal of Applied Behavior Analysis 6: 579-586. Pred, A. 1984: Place as historically contingent process: structuration and the time-geography of becoming places. Annals of the Association of American Geographers 74, 279–97. Prem, H. 1992. Spanish colonization and Indian property in Central Mexico, 1521-1620. Annals of the Association of American Geographers 82(3): 444-459. Previte, J., B. Pini, and F. Haslam-McKenzie. 2007. Q Methodology and rural research. Sociolo - gia Ruralis 47(2): 135-147. 329 Pulido, L. 2000. Rethinking environmental racism: White privilege and urban development in Southern California. Annals of the Association of American Geographers . 90(1): 12-40. Pulsipher, L. 1993. Changing Roles in the Life Cycles of Women in Traditional West Indian Houseyards. In Women and Change in the Caribbean: A Pan-Caribbean Perspective , Ed. J. Momsen, 50-64. London: James Currey Quinn, C., Huby, M., Kiwasila, H., and Lovett, J. C. 2003. Local perceptions of risk to livelihood in semi-arid Tanzania. Journal of Environmental Management 68: 1111–119. Raat, W. D. 2004. Mexico and the United States: Ambivalent Vistas, 3rd Ed. Athens, GA: Univer- sity of Georgia Press. Radcliffe, S. 2007. Latin American indigenous geographies of fear: Living in the shadow of rac - ism, lack of development, and antiterror measures. Annals of the Association of American Geog - raphers 97(2): 385-397. Rappaport, R. 1968. Pigs of the Ancestors: Ritual in the Ecology of a New Guinea People . New Haven, CT: Yale University Press. ___________. 1975. The Flow of Energy in an Agricultural Society. In Biological Anthropology . Ed. S. Katz. San Francisco: W.H. Freeman. Rathje, W. 1974. Garbage Project: A new way of looking at the problems of archaeology. Ar - chaeology 27(4): 236–241. Rathje, W. and C. Murphy. 2001. Rubbish! The Archeology of Garbage . Tucson, AZ: The Uni- versity of Arizona Press. Read, A. 2003. Delivering more sustainable waste management in Mexico. Resources, Conserva - tion, and Recycling 39: 187-191. Registro Agrario Nacional (RAN). 2009. Padron e Historial de Núcleos Agrarios. Online data- base. Accessed: July 2012. Reiter, S. and W. Samuel. 1980. Littering as a function of prior litter and the presence or absence of prohibitive signs. Journal of Applied Social Psychology 10(1): 45–55 Restrepo, I. and D. Philips. La basura: consume y desperdicio en el Distrito Federal . Mexico, D.F.: Centro de Ecodesarollo. Riley, M. 2008. From salvage to recycling – new agendas or same old rubbish? Area 40(1): 79- 89. Robbins, P. 2000. The Practical Politics of Knowing: State Environmental Knowledge and Local Political Economy. Economic Geography 76(2): 126-144. _________. 2004. Political Ecology . Malden, MA: Blackwell Publishing. _________. 2006. The politics of barstool biology: environmental knowledge and power in Greater Northern Yellowstone. Geoforum 37: 185-199. 330 _________. 2007. Lawn People: How Grasses, Weeds, and Chemicals Make Us Who We Are . Philadelphia, PA: Temple University Press. Robbins, P. and R. Krueger. 2000. Beyond bias? The promise and limits of Q methodology in Human Geography. The Professional Geographer 52(4): 636-648. Rocheleau, D. 1995. Maps, numbers, text and context: Mixing methods in feminist political ecol - ogy. Professional Geographer 47(4): 458-467. Rocheleau, D., B. Thomas-Slayter and E. Wangari. 1996. Feminist Political Ecology: Global Perspectives and Local Experiences . Routledge. London. Rocheleau, D. and D. Edmunds. 1997. Women, men and trees: gender, power and property in Forest and Agrarian Landscapes. World Development 25(8): 1351-1371. Rodrigues-Santos, I., A. Friedrich, M. Waldner-Kersanach, and G. Fillmann. 2005. Influence of socio-economic characteristics of beach users on litter generation. Ocean and Cultural Manage - ment 48: 742-752. Rodriguéz, V. 1987. The Politics of Decentralization in Mexico: Divergent Outcomes of Policy Implementation . Unpublished Ph.D. Dissertation, University of California, Berkeley. Rogers, H. 2005. Gone Tomorrow: The Hidden Life of Garbage . NY: The New Press. Rohrschneider, R. 1988. Citizens’ attitudes towards environmental issues: Selfish or selfless? Comparative Political Studies 21(3): 347-367 Royte, E. 2005. Garbage Land: On the Secret Trail of Trash. NY: Little, Brown and Company. R.W. Beck Consultants. 2007. A Review of Litter Studies, Attitude Surveys and Other Litter-relat - ed Literature . Stamford, CT: Keep America Beautiful. Saarinen, T. 1984. Environmental Planning: Perception and Behavior . Longrove, IL: Waveland Press. Sack, R. 1997. Homo Geographicus . Baltimore, MD: The Johns Hopkins University Press. Şahil, H. and S. Erkal. 2010. The attitudes of middle school students towards the environment. Social Behavior and Personality 38(8): 1061-1072. Saint-Germain, M. 1995. Similarities and differences in perception of public service among pub- lic administrators on the US-Mexico border. Public Administration Review 55(6): 507-516. Salzinger, L. 2003. Genders in Production: Making Workers in Mexico’s Global Factories . Berkeley, CA: The University of California Press. Sánchez-Mendez, M. and V. Olmos-Torres. 2009. Estudio de Generación y Caracterización de Residuos Sólidos Urbanos, Coxcatlán, Puebla . Unpublished municipal report. Sanders, W., J. Parsons, and R. Santley. 1979. The Basin of Mexico: Ecological Processes in the Evolution of a Civilization . New York: Academic Press. 331 Santos, I., A. Friedrich, and J. Ivar do Sul. 2009. Marine debris contamination along undeveloped tropical beaches from northeast Brazil. Environmental Monitoring and Assessment 148: 455–462 Sauer, C. 1963. The Morphology of Landscape. In Land and Life: Selected Writings of Carl Or - twin Sauer , Ed. John Leighly. Berkeley, CA: University of California Press. _______. 1950. Grassland climax, fire, and man. Journal of Range Management 3: 16-21. Scanlan, J. 2005. On Garbage . London: Reaktion Books. Schmieder, O. 1930, The Settlements of the Tzapotec and Mije Indians, State of Oaxaca, Mexico . Berkeley, CA: University of California Press. Schultz, P., R. Bator, L. Brown Large, C. Bruni and J. Tabaconi. 2013. Littering in context : Personal and environmental predictors of littering behavior. Environment and Behavior 45(1): 35-59. Schultz, P. and S. Stein. 2009. Executive summary: litter in America 2009 national litter research findings and recommendations. http://www.kab.org/site/DocServer/Executive_Summary_-_FI- NAL.pdf?docID=4601. Accessed online 6-3-2013. Schwaller, J. 1987. The Church and Clergy in Sixteenth-Century Mexico . Albuquerque, NM: University of New Mexico Press. Scott, J. 1976. The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia . New Haven, CT: Yale University Press. Seco Pon, J. and M. Becherucci. 2012. Spatial and temporal variations of urban litter in Mar del Plata, the major coastal city of Argentina. Waste Management 32: 343-348. SEMARNAT. 2005. Informe de la situación del medio ambiente en México, Compendio de es - tadístacas ambientales . Mexico, D.F.: Secretaría del Medio Ambiente y Recursos Naturales. SEMARNAT. 2012. SNIARN, Base de datos estadísticos, Módulo de consulta temática, Dimen - sión ambiental, junio, 2011. www.semarnat.gob.mx. Accessed Oct. 4, 2012. Sheridan, T. 1988. When the Dove Calls: The Political Ecology of a Peasant Corporate Commu - nity in Northwestern Mexico . Tucson, AZ: The University of Arizona Press. Sibley, C. and J. Liu. 2003. Differentiating active and passive littering: A two-stage process of modeling littering behavior in public places. Environment and Behaviour 35(3): 415-433. Sieber, R. 2006. Public Participation Geographic Information Systems: A literature review and framework. Annals of the Association of American Geography 96(3): 491-507. Silva-Iñiguez, L. and D. Foster. 2003. Quantification and classification of marine litter on the municipal beach of Ensenada, Baja California, Mexico. Marine Pollution Bulletin 46(1): 132- 138. Simonian, L. 1995. Defending the Land of the Jaguar: A History of Conservation on Mexico . Austin, TX: University of Texas Press. 332 Sisson, E. 1973. First Annual Report of the Coxcatlan Project . Andover, MA: R.S. Peabody Foundation for Archaeology. _______. 1974 Second Annual Report of the Coxcatlan Project . Andover, MA: R.S. Peabody Foundation for Archaeology. Sluyter, A. 1999. The making of myth in postcolonial development: Material-Conceptual land - scape transformation in Sixteenth Century Veracruz. Annals of the Association of American Geographers 89(3): 377-401. Smith, B. 2005. Reassessing Coxcatlan Cave and the early history of domesticated plants in Me - soamerica. Proceedings of the National Academy of Sciences 102(27): 9438-9445. Smith, C. 1965. Flora, Tehuacan Valley. Fieldiana 31(4): 101-144. Smith, C. and M. Cameron. 1977. Ethnobotany in the Puuc, Yucatan. Economic Botany 31(2): 93-110. Smith, D., P. Herlihy, J. Kelly, A. Ramos Viera. 2009. The certification and privatization of indig- enous lands in Mexico, Journal of Latin American Geography 8(2): 175-207. Smith, D., P. Herlihy, J. Kelly, A. Ramos, A. Hilburn, J. Dobson, and M. Aguilar-Robledo. 2012. Using Participatory Research Mapping and GIS to explore local geographic knowledge of indig- enous landscapes in Mexico. FOCUS on Geography 55(4): 119-124 Smith, K., Barrett, C. B., and Box, P. W. 2000. Participatory Risk Mapping for targeting research and assistance: With an example from East African pastoralists. World Development 28: 1945– 1959. Smith, K., Barrett, C. B., and Box, P. W. 2001. Not necessarily in the same boat: Heterogeneous risk assessment among East African pastoralists. The Journal of Development Studies 37: 51–30. Soja, E. 1996. Thirdspace: Journeys to Los Angeles and Other Real-and-Imagined Places . Mal- den, MA: Blackwell. Solorzano-Ochoa, G., D. de la Rosa, P. Maiz-Larralde, B. Gullett, D. Tabor, A. Touati, B. Wyrzykowska-Ceradini, H. Fiedler, T. Abel, and W. Carroll Jr. 2012. Open burning of household waste: Effect of experimental condition on combustion quality and emission of PCDD, PCDF and PCB. Chemosphere 87: 1003-1008. South, R. 2006. Spatial variations in Mexican maquiladora closure. Urban Geography 27(8): 734-756. Spores, R. 1967. The Mixtec Kings and Their People . Norman, OK: University of Oklahoma Press. Stanislawski, D. 1947. Early Spanish town planning in the New World. Geographical Review 37 (1): 95-105. Steelman, T. and L. Maguire. 1999. Understanding participant perspectives: Q-methodology in 333 National Forest management, Journal of Policy Analysis and Management 18(3): 361-388. Stein, S. and D. Syrek. 2005. New Jersey Litter Survey: 2004 . Fairfax, VA: Gershman, Brickner, and Bratton, Inc. Steinberg, M. 1998. Neotropical kitchen gardens as a potential research landscape for conserva - tion biologists. Conservation Biology 12(5): 1150-1152. Stephen, L. 1998. The Cultural and Political Dynamics of Agrarian Reform in Oaxaca and Chi - apas. In The Future Role of the Ejido in Rural Mexico . Eds. R. Snyder and G. Torres. La Jolla, CA: Center for U.S-Mexican Studies, University of California-San Diego. Stephenson, W. 1935. Technique of factor analysis. Nature 136: 297. Steward, J. 1955. Theory of Culture Change: The Methodology of Multilinear Evolution . Urbana, IL: University of Illinois Press. Strasser, S. 1999. Waste and Want: A Social History of Trash . NY: Metropolitan Books. Swaffield, S. and J. Fairweather. 1996. Investigation of attitudes towards the effects of land use change using image editing and Q sort method, Landscape and Urban Planning 35: 213-230. Swyngedouw, E. 2004. Globalisation or “glocalisation”? Networks, territories and rescaling. Cambridge Review of International Affairs 17(1): 25-48. _____________. 1997. Power, nature, and the city. The conquest of water and the political ecol - ogy of urbanization in Guayaquil, Ecuador: 1880-1990. Environment and Planning A 29(2): 311-332 Sutro, L. 1991. When the River Comes: Refuse Disposal in Díaz Ordaz, Oaxaca. In The Ethnoar - chaeology of Refuse Disposal . Eds. E. Staski and L. Sutro. Anthropological Research Papers 42. Tempe, AZ: Arizona State University. Szasz, A. 1994. EcoPopulism: Toxic Waste and the Movement for Environmental Justice . Minne- apolis, MN: University of Minnesota Press. Teichman, J. 1995. Privatization and Political Change in Mexico . Pittsburgh, PA: University of Pittsburgh Press. __________. 1997. Mexico and Argentina: Economic reform and technocratic decision making. Studies in Comparative International Development 32(1): 31-55. __________. 2004. The World Bank and policy reform in Mexico and Argentina. Latin American Politics and Society 46(1): 39-74. Ticante-Roldán, J., J. Ventura-Pérez, A. Moreno-Ortíz, and M. Valera-Pérez. 2009. Caracter - ización Física y Química delos Lixiviados del Relleno Sanitario del Municipio de Puebla. In La Basura en una perspectiva multidisciplinaria . Ed. J. Guevara Martínez. Mexico, D.F.: Unlisted press. Toledo, V. 1996. The Environmental Consequences of the 1992 Agrarian Law of Mexico. In 334 Mexico’s Agrarian Reform , ed. L. Randall. Armonk, NY: M.E. Sharp. Trejo-Vázquez, R. and R. Cespedes-Soto. 1989. The recycling from municipal solid wastes in Mexico. Journal of Resources Management and Technology 17:15-2 7. Tuan. Y. 1977. Space and Place: The Perspective of Experience . Minneapolis, MN: University of Minnesota Press. Turner, B. II. 1983. Once Beneath the Forest: Prehistoric Terracing in the Río Bec Region of the Maya Lowlands . Boulder, CO: Westview Press. United Nations. 2009. Environmental Indicators: Waste. Available online at http:/ /unstats.un .org/unsd/environment/wastet.reatment.htm (Accessed October 3, 2012). UNEP. 2005, 2006, 2007. Annual Reports, 2005, 2006, 2007. United Nations Environment Pro - gramme. Vayda, A. and B. McKay 1975. New directions in ecology and ecological anthropology. Annual Review of Anthropology 4: 293-306 Vayda, A. and R. Rappaport. 1968. Ecology: Cultural and Noncultural. In Introduction to Cul - tural Anthropology . Ed. A. Clifton. New York: Houghton Mifflin. Vázquez-Castillo, M. 2004. Land Privatization in Mexico: Urbanization, Formation of Regions, and Globalization in Ejidos . NY: Routledge. Vanderwood, P. 1998. The Power of God against the Guns of Government: Religious Upheaval in Mexico at the Turn of the Nineteenth Century . Palo Alto, CA: Stanford University Press. Van Eeten, M. 2000. Recasting Environmental Controversies: A Q Study of the Expansion of Amsterdam Airport. In Social Discourse and Environmental Policy: An Application of Q Meth - odology , eds. H. Addams and J. Proops. Northampton, MA: Edward Elgar Publishing. Veblen, T. 1994. Conspicuous Consumption . Ed. R. Lekachman. NY: Penguin Press. Vinet, F. 2008. Geographical analysis of damage due to flash floods in southern France: The cases of 12–13 November 1999 and 8–9 September 2002. Applied Geography 28: 323-336. Vining, J. and A. Ebreo. 1990. What makes a recycler? A comparison of recyclers and non-recy - clers. Environment and Behavior 22: 55-73. Volk, T. and J. Cheak. 2003. The effects of an environmental education on students, parents, and community. The Journal of Environmental Education 34(4): 12-25. Von Bertrab, E. 2003. Guadalajara’s water crisis and the fate of Lake Chapala: a reflection of poor water management in Mexico. Environment and Urbanization 15(2): 127-140. Wakefield, S. and S. Elliott. 2000. Environmental risk perception and well-being: effects of the landfill siting process in two southern Ontario communities. Social Science and Medicine 50: 1139-1154. 335 Ward, L. 2013. Eco-governmentality revisited: Mapping divergent subjectivities among Integrat - ed Water Resource Management experts in Paraguay. Geoforum 46: 91-102. Ward, P. 1998. From machine politics to the politics of technocracy: Charting changes in gover - nance in the Mexican municipality. Bulletin of Latin American Research 17(3): 341-365. Watts, M. 1983a. On the Poverty of Theory: Natural Hazards Research in Context. In Interpreta - tions of Calamity , ed. K. Hewitt. Boston, MA: Allen and Unwin. Watts, M. 1983b. Silent Violence: Food, Famine, and Peasantry in Northern Nigeria . Berkeley, CA: University of California Press. Watts, M. and R. Peet 1996. Towards a theory of liberation ecology. In Liberation Ecologies: Environment, Development, and Social Movements , eds. R. Peet and M. Watts. NY: Routledge. West, R. 1948. Cultural Geography of the Modern Tarascan Area . Washington, D.C.: US Gov- ernment Printing Office. ______. 1993. Sonora: Its Geographical Personality . Austin, TX: University of Texas Press. White, G. 1945. Human Adjustment to Floods . Department of Geography Research Paper No. 29. Chicago: University of Chicago Press. White, M. and L. Hunter. 2009. Public perception of environmental issues in a developing set - ting: Environmental concern in coastal Ghana, Social Science Quarterly 90(4): 960-980. Whitmore, T. M., and B. L. Turner II. 2001. Cultivated Landscapes of Middle America on the Eve of Conquest . Oxford: Oxford University Press Whyte, W., D. Greenwood, and P. Lazes. 1989. Participatory Action Research: Through practice to science in social research. American Behavioral Scientist 32:513-551. Wildcat, D. 2009. Red Alert!: Saving the Planet through Indigenous Knowledge . Golden, CO: Fulcrum Publishing. Wilder, M. and P. Romero-Lankao. 2006. Paradoxes of decentralization: Water reform in and its social implications in Mexico. World Development 34: 1977-1995 Wilk, R. 1983. Households in Process: Agriculture Change and Domestic Transition among the Kekchi Maya of Belize. In Households: Comparative and Historical Studies of the Domestic Group . Eds. R. McNetting, R. Wilk, and E. Arnould. Berkeley, CA: University of California Press. Wilken, G. 1987. Good Farmers: Traditional Agricultural Resource Management in Mexico and Central America . Berkeley: University of California Press. Williams, R. 1999. Environmental injustice in America and its politics of scale. Political Geog - raphy 18: 49–73. 336 Wokekoro, I. and M. Inyang. 2007. Waste disposal in low-income neighbourhoods and its impact on health: The case of Port Harcourt, Nigeria. Proceedings of the Conference Waste Manage - ment, Environmental Geotechnology and Global Sustainable Development (ICWMEGGSD’07 - GzO’07)” Ljubljana, SLOVENIA, Wolf, E. 1972. Ownership and political ecology. Anthropological Quarterly 45: 201-205 _______. 1982. Europe and the People Without History. Berkeley, CA: University of California Press. Woodbury, R. and J. Neely. 1972. Water control systems of the Tehuacan Valley. In The Prehis - tory of the Tehuacan Valley . Ed. R. S. MacNeish. Austin, TX: University of Texas Press Woods, W., W. Teixeira, J. Lehmann, C. Steiner, A. WinklerPrins, and L. Rebellato. 2009. Ama - zonian Dark Earths: Wim Sombroek’s Vision . Springer, Berlin. Works, M. 1990a. Continuity and conversion of historic dooryard gardens in western Amazonia. Yearbook of the Association of Pacific Coast Geographers 52: 31-64. ________. 1990b. Dooryard gardens in Moyobamba, Peru. Focus 40(2): 12-17. Yeich, S. and R. Levine. 1992. Participatory Research’s contribution to a conceptualization of empowerment. Journal of Applied Social Psychology 22: 1894-1908. Yepes, G. and T. Campbell. 1990. Assessment of Municipal Solid Waste Services in Latin Ameri - ca. Washington, D.C.: World Bank. Young, E. 1999a. Local people and conservation in Mexico’s El Vizcaíno Biosphere Reserve. Geographical Review 89: 364–390. _______. 1999b. Balancing conservation with development in small-scale fisheries: Is ecotour- ism an empty promise? Human Ecology 27(4): 581–620. Zabin, C. 1997. Free Markets and Forestry: The Prospects for Community-based Forestry in Mexico in the Era of Neoliberal Reform. In Reform in the Mexican Countryside . Ed. David Myhre. San Diego, CA: Center for U.S.-Mexican Studies, University of California, San Diego Zimmerer, K. 1993. Soil erosion and social (dis)courses in Cochabamba, Bolivia: Preceiving the nature of environmental degradation. Economic Geography 69(3): 312-327. __________. 1994. Human geography and the “New Ecology”: The prospect and promise of integration. Annals of the Association of American Geographers . 84(1): 108-125. __________. 1996. Ecology as Cornerstone and Chimera in Human Geography. In Concepts in Human Geography . Eds. C. Earle, K. Mathewson, and M. Kenzer. Lanham, MD: Rowman and Littlefield Publishers. __________. 2000. The reworking of conservation geographies: Nonequilibrium landscape and nature-society hybrids. Annals of the Association of American Geographers 90(2): 356-369. 337 __________. 2003. Environmental Zonation and Mountain Agriculture in Peru and Bolivia: So- cioenvironmental Dynamics of Overlapping Patchworks and Agrobiodiversity Conservation. In Political Ecology . Eds. K. Zimmerer and T. Bassett. NY: Guilford Press. __________. 2007. Cultural ecology (and political ecology) in the ‘environmental borderlands’: exploring the expanded connectivities within geography. Progress in Human Geography 31(2): 227-244. Zimmerer, K. and T. Bassett. 2003. Future Directions in Political Ecology: Nature-Society Fusions and Scales of Interaction. in Political Ecology. Eds. K. Zimmerer and T. Bassett. NY: Guilford Press. Appendix 1: Researching Garbage in Small Town Mexico INTRODUCTION A dissertation can be a large piece of graduate career-defining work (Gade 2008). At least it has been for me. First, the production of one requires a comprehensive graduate-level curricu- lum, where the dissertator follows a inter-disciplinary course to develop a skill set to help them accomplish not only their requirements as students but more importantly create a specialty that allows them a unique perspective and position to contribute to larger academic and professional debates. Approaching a dissertation also involves the comprehending of a diverse set of litera- tures, allowing she or he to understand where they can possibly contribute to the institutionalized production of knowledge. A grasp of the epistemologies and ontologies regarding one’s interest also informs how to contribute to the field, eventually leading to the formulation of a dissertation topic. The degree to how much an adviser guides a student in this process varies, where some advisees are almost given a topic to investigate within the context of the adviser’s larger interests to instances where the adviser only supervises the theoretical development and edits the final documents. In my case, I developed this project through classes, piles of reading, and long hours in front of the computer at KU. Thankfully, I was also given the opportunity to conduct applied research in the field in Mexico by my adviser over the course of my graduate career leading up to the dissertation, but the topic, methods, and study region were of my own discovery. After a topic is selected, the required plan of research often means extensive field research working somewhere besides home, which in my case was over a year and a half spent in various parts of southern Mexico. What follows may only to some degree approach the ideal definition noted above, but the following appendix is a description on how I came to research garbage manage- ment in Mexico and my research and life experiences while there. STUDYING GARBAGE IN MEXICO Before the dissertation My dissertation is a narrative of how the rural-but-urbanizing Mexican municipio of Coxcatlán, Puebla and its citizens manage their garbage and how they view this management as an environmental issue. Garbage is an important issue and potentially an environmental problem there. This is the case across Mexico and in Coxcatlán, where generation rates of a more per- 338 sistent and toxic wastes increase amid a progressively scarce land base to deposit it. However, other issues such as water availability, land tenure security, and global sugar prices may nornmal- ly loom larger in the minds of Coxcatecos (Coxcatlán residents) as they live at the end of the so- cial property era in Mexico in a semi-arid climate zone where the local economy is largely based on the cultivation of sugarcane. This research is not a plea to recognize a fixed course toward severe environmental degradation from garbage, nor is it a pressing call for attention to garbage- related environmental justice or to highlight a gripping struggle over the control of a resource. Creating a narrative on waste and society in Coxcatlán does inevitably touch on environmental degradation, potential health issues, and notions of governmentality. It also involves individual perceptions of what garbage is, how one should deal with it, and what precisely are the relevant environmental issues related to it. The inspiration for this dissertation on garbage first came to me in 2005 as a Master’s student at the University of Southern Mississippi. For my thesis research that summer, my ad- viser Joby Bass had convinced me to go to the sierra east of Oaxaca City, where various dialects of Mixe are spoken to document village landscape change and indigenous urbanization where only a few generations earlier most area residents lived in dispersed rancherías or small hamlets. In Ayutla Mixes, where I spent the most of my time, I observed a growing town of about 1,600 people that only 30 years before had just under 700 residents. Corresponding with this growth in urbanization was what I thought to be an enormous and growing garbage dump outside the town limits (Figure A1.1). Just from this small town were acres of bottles, diapers, broken plastic chairs, tires, bathroom items, and food waste all strewn out on a 4-5 percent slope off the road. I was blown away, somewhat to my own naivety, that such a fairly small and economically mar- ginalized town could produce so much garbage and that the municipal government could collect any of it. Yet the municipal government did collect this garbage to a regular degree, in spite of having old collection trucks and incredibly rugged terrain to cross. Garbage management looked like a worthwhile topic and was something I talked about when asked about what I had seen dur- ing my stay in Oaxaca that summer, but it was not my main research focus there. Furthermore, I did not know anything about garbage management, and more importantly I needed to finish my thesis, so I shelved it as an interest for some time. I graduated in 2006 from USM and made the move to Kansas to begin my doctoral research the same year. Over the next two years I returned to the state of Oaxaca twice, in the 339 summers of 2007 and 2008, while working on the American Geographical Society’s Bowman Expedition led by my adviser, Peter Herlihy in the Sierra Juárez in the northern part of the state. Here, we were investigating the effects of neoliberal land reforms on indigenous communities. Practically speaking, we conducted participatory research mapping in conjunction with the com- munities to produce maps that documented their existing land tenure systems. By the start of my second summer, I knew I wanted to investigate garbage management in rural Mexico but I was clumsily trying to place the issue in what I had learned over the past summer, mainly in the con- text of land reform at the scale of Mexico’s social properties (ejidos and comunidades agarias), which have much less to do with garbage than the municipal governments do. I had also been delayed by falling ill for extended periods during these stays. I was not producing the results and conceptual development that I needed to do my dissertation on garbage and its management in a non-urban place in Mexico. I needed to change my research focus to the municipio and I also needed to broaden my horizons beyond Oaxaca. At the time, I had no research experience outside Oaxaca’s Sierra Juárez and Sierra Mixe, but I had some ideas of where I wanted to work. On the trips to Oaxaca, I took a bus from Mexico City’s TAPO station to Oaxaca City. The seven-hour trip first headed east through the Figure A1.1. Garbage dump on the outskirts of Ayutla Mixes, Oaxaca. Photo by A. Hilburn, July 2005. 340 urban sprawl of the eastern arm of the State of Mexico, up and over the Iztaccíhuatl-Popocatépetl mass of volcanoes, and across the heavily-populated Puebla-Tlaxcala Valley before turning south towards Oaxaca. It was just after this turn into the northern edge of the Tehuacán Valley that the landscape changed enough to always catch my eye when I passed through it. The truck farming communities and mid-size cities of the eastern Puebla-Tlaxcala Valley gave way to large ex- panses of whitewashed rolling hills and travertine cliffs covered in a variety of xerophytic plants. Snowy Pico de Orizaba, Mexico’s tallest peak at 5,636 meters (18,491 feet) above sea level could be seen just to the east (Figure A1.2). Small farming villages and towns punctuated the landscape until the sprawl of the city of Tehuacán came into view at the valley’s center. Beyond Tehuacán to the south were smaller towns and cities amid a flatter valley framed by the Sierra Negra to the east and Mixteca Baja to the west. To my own casual assessment, it appeared to be a region unto itself and I wanted to know more about it. The weather was usually nice when I passed through in the summer. Across its north-south extent, the Tehuacán Valley slopes down in elevation in southerly direction and sits between the altitudinal zones of tierra fría and tierra caliente making it an agreeable place to be and do research (Figure A1.3). Furthermore, it was not chilly and rainy like the Oaxacan sierra nor was it hot and humid like my hometown of Mobile, Alabama. It was also just somewhere different. It was somewhere where I had no contacts, no mental map, and zero familiarity with it. I knew next to nothing about the Tehuacán Valley besides the early maize studies and the spar- kling water for which it is famous and I knew even less about how garbage was managed in areas around the city of Tehuacán. So I spent the next Christmas break in and around Tehuacán, learn- ing about how garbage is managed in municipios as far north of Tehuacán as Cañada Morelos and Tepanco de López, to Zapotitlán Salinas at the edge of the Mixteca Baja, all the way to the south end of the valley in Ajalpan and Coxcatlán. I interviewed municipal officials in charge of garbage management, garbage workers, and officials at the headquarters of the federal biosphere reserve that covers large portions of the valley. I also went to dumps, from open-air fields of gar- bage to lined and drained sanitary landfills (Figure A1.4). When I left, I felt I had a sense of who to talk to, where to collect information, and possibly where I would want to work for my dis- sertation research, but I still lacked a research question and an approach to address it, whatever it would be. What I did come to understand was that garbage and garbage management in rural and 341 Figure A1.2. Pico Orizaba or Citlaltépetl from the northern Tehuacán Valley near San José Ixtapa. Photo by A. Hilburn, January 2011. Figure A1.3. Another look at Pico Orizaba or Citlaltépetl from the northern Tehuacán Valley near San José Ixtapa. during the rainy season. Photo by A. Hilburn, June 2011. 342 peri-urban Mexico was a compelling and complex issue about a policy-governance-stakeholder nexus over the control of a non-resource. It also appeared to be much more than that due to the highly subjective nature of what garbage is, how it should be managed, and more importantly, what comprise the most pressing garbage-related issues. Back at KU in the fall of 2008, I ex- plored Q-methodology to systematically approach the perceptional aspect of garbage and its management. I also applied for and was awarded some small research grants, a Tinker Founda- tion Field Research Grant and a Conference of Latin Americanist Geographers (CLAG) Disser- tation Field Research Award, to do preliminary research on the Q-study and a garbage census. During June and July of 2009, I began working in and around Tehuacán looking to do a broad regional analysis of garbage management there, but came to the conclusion that it would be bet- ter to focus on one municipio and relate it to the larger region and Mexico. For a number of reasons, I chose to do my dissertation research in Coxcatlán, the most outlying municipio that I had visited the year before. Elsewhere, the municipios south to the south of Tehuacán share management of a sanitary landfill, have concentrated settlements across easily accessible valley landscapes, and have very extensive garbage collection service provi- Figure A1.4. Intermunicipal sanitary landfill outside Ajalpan, Puebla. The municipios of Ajalpan, San Sebastian Zi - nacatepec, San José Miahuatlán, Chilac, and Altepexi use and share management of this site. Photo by A. Hilburn, January 2009. 343 sions. They are also mostly urban and I wanted to research somewhere more sparsely settled as well as a place with a more ethnically diverse population. I also was looking for a municipio that did not have universal collection service and that still had places where people and communities have other ways of contending with their garbage. I found that Coxcatlán had these character- istics and seemed an ideal case study site. Half of the population lives in what one would call “towns” with most of the trappings of urban life and the rest living in villages centered on small scale agriculture. The municipio is also split between the Tehuacán Valley and the rugged Sierra Negra, lending an immense geographical challenge for the municipio to provide service. Also, Coxcatlán, both the town and the municipio, are really beautiful. The town still retains much of its old Revolution-era buildings in its center and the extensive canopies of fruit and shade trees make it a beautiful green oasis in a largely dry, dusty scrub forest landscape, something the larger settlements nearby cannot claim. The municipio itself has a diverse mix of stunning landscapes of sugarcane production, thorn-scrub cactus forest, and upland pine-oak forest. The contrast be- tween the valley and the sierra is stunning (Figure A1.5). The PRI-affiliated municipal government was also friendly, interested in my project, and helpful where they could be. They were also at the time drafting their garbage management plan in conjunction with the State government authorities to receive funding for a recycling program and a waste transferal system, so it was an opportune time to be there. The municipal president’s office drafted me a letter in support of my research. I also scouted the nearby larger town of Ajal- pan as a backup study municipio. Over the next year, I developed my research questions and submitted proposals for re- search fellowships to fund my field work. I ran the gauntlet , applying for the Fulbright-IIE, the Fulbright-Hays, and the NSF DDRI a and was fortunate to be selected for the Fulbright-Hays in the summer of 2010. I also passed my Ph.D. comprehensive exams at KU earlier in May. At the time, I had a research assistantship under my adviser so I postponed leaving for southern Puebla until January 2011. January to June 2011 The New Year rolled around and five days later I left Kansas City’s airport for Mexico City with my wife, Alison, who wanted to help me get settled in Mexico. I was very excited to be out on my own, doing my own research with no other plans other than completing my dis- 344 Figure A1.5. Looking east from Chilac across the southern Tehuacán Valley toward Calipan, Coxcatlán, and Sierra Negra. Photo by A. Hilburn, December 2011. Figure A1.6. The south side of the park in Coxcatlán’s centro. The church is to the right, the ayuntamiento build - ing is in front, and the park is on the left. Photo by A. Hilburn, Date unknown. 345 sertation field work. I was also bummed that I was going to be without my wife for the major- ity of that time. Alison, an archaeology doctoral student, was in the midst of preparing for her comprehensive exams while teaching introductory archaeology classes at KU, so she could not stay with me during the entire ten months of my field research. Living apart, though, was not new to us. We met and became a couple as undergrads at the University of South Alabama, but had also spent our Master’s degree careers at separate universities. I was at the University of Southern Mississippi and she attended the University of Kentucky before moving together to Lawrence. We had done the long distance relationship thing the whole time for that time and we practically did it every summer since we lived in Lawrence. Also, we had planned ahead and had saved money to buy flights for her to come down and visit when she had an open block of time. We knew it was all going to work out, which it did, but we both dreaded being apart for extended Figure A1.7. Arriving in the early morning to San Luis Potosí. Photo by A. Hilburn, January 2011. periods. We arrived in Mexico City and immediately traveled to the capital city of San Luis Potosí State of the same name. Miguel Aguilar-Robledo, a geographer and the Director of la Coordi- nación de Ciéncias Sociales y Humanidades at the Universidad Autónoma de San Luis Potosí 346 had graciously written a letter of support offering to be the domestic host of my Fulbright-Hays fellowship. It may seem peculiar that I was hosted by an academic institution outside of the state of Puebla, but Dr. Aguilar is one of the most admired Latin Americanist geographers in the field as well as being a member of the prestigious Academia Mexicana de Ciéncias. He was also the in-country collaborator investigator for the AGS Bowman Expedition to Mexico and a close col- league and personal friend of my adviser. There, I met with the faculty at the Coordinación for advice, data, contacts, and research. I also received an important research letter from Dr. Aguilar explaining my sponsorship by UASLP. After a little over a week, I left San Luis Potosí for Mex- ico City to get my student visa sponsorship and to meet the staff at the COMEXUS (Comisión México-Estados Unidos para el Intercambio Educativo y Cutlural), the institute that sponsors all Fulbright programs and other scholarly and cultural exchanges between Mexico and the United States. We then left for Tehuacán where we spent the next few days introducing myself and my project at regional government offices, while I finished the garbage route maps I planned to pres- ent to the Coxcatlán municipal government. With only a few days before Alison left for Kansas, we went to Coxcatlán’s municipal building. We were surprised to find that we had arrived at a time of complete transition of the municipio’s administration and my official contacts, had only about a week left in its office. Besides there not being anyone to meet with besides the secretary, there was barely even furni- ture in the building. I turned in my maps and left to ponder the next step. I spent the next few days talking to people about potentially working in nearby Ajalpan while looking into finding a more permanent place to stay and getting a phone or internet or any of those things before taking Alison back to Mexico City. When I got back, I would need new permission from the new PAN municipal administration in Coxcatlán before I could begin my research. Yet, the new admin- istration had a backlog of business to attend to before considering my project again. I really wanted to work in Coxcatlán but Ajalpan was less than 30 minutes away and I had met some new contacts there. I also needed somewhere to stay and I had met a nice family who ran a small mo- tel in Ajalpan, so I put a month’s rent down there to stay while I simultaneously planned to work in both Coxcatlán and Ajalpan. I also had to take care of my visa, which would require two more visits to the state capital, four hours away. Luckily, by the time my visa was granted, the new municipal President, Efigenio Atilano, graciously gave me a meeting where I explained to him and his cabinet who I was and described the intent of my project and what I hoped to achieve 347 while in Coxcatlán. The President then expressed his interest in my work, gave his approval, and officially welcomed me back to Coxcatlán. He planned to have one of his friends give me a tour of the town and sierra the next day, as well as set me up with contacts and place to work and stay. Although I was quite familiar with the place, I accepted. On the tour, the president’s friend, a man named Salvio Mier, asked me what my hob- bies were. After I shrugged and couldn’t give an answer, he asked me if I liked woodworking because he had a friend that was a carpenter. I said sure and we set off to basically kill time at the carpenter’s house and workshop in lieu of giving the remainder of what everyone came to see as an unnecessary tour. There I met Miguel Alva, the carpenter, who could be described as a late- 50s PE teacher/carpenter. Miguel had lived almost a decade in Berkeley, California and spoke English fairly well. He was a divorcee who had in recent years moved back from Puebla City to take care of his elderly mother and live in his hometown. Besides being intellectually curious and having an excellent library, Miguel was a Nahuatl tutor. When he was a boy, Miguel’s parents had moved the family from their Nahua community in the sierra about an hour away for more employment opportunities and educational opportunities for the kids. The now-deceased patri- arch was a schoolteacher and farmer who compiled and wrote one of the few Spanish-Nahuatl Figure A1.8. A political sign for the winning PAN candidate for state governor. Photo by A. Hilburn, January 2011. 348 dictionaries that represents the specific Nahuatl dialect of the adjacent Sierra Negra. It could be an overstatement to say we became instant friends and one doesn’t always pay attention to such things, but Miguel immediately opened his home to me and offered me its use as a home base or office or place to hang out. I began to spend my free time entirely at Miguel and his widowed mother, Doña Teresa’s place. For the next month I still kept my apartment in Ajalpan because I had already paid for it and I did not want to formally move in with the Alvas yet, even though they had offered me a room in their ample house. Besides, I was already spend- ing every day in Coxcatlán working and when I was not, I spent all my free time there. Some- times I slept there when work or some activity kept me beyond the last bus leaving town. I was also eating many of my meals with the Alva family. I had a home away from home. I also met other friends through Miguel. Pepe Armas, Miguel’s best friend and the owner of the town’s best internet cafe/telephone/Kinko’s and also a budding sugarcane grower from one of Coxcatlán’s most prominent families, was my age and his then-girlfriend Paty, who taught at the local high school (preparatoria) while getting her Master’s degree in education in Tehuacán also became great friends. Salvio, the friend of the president, who had recently moved back to Coxcatlán after his brother closed his border town bar where he worked due to tourism declines from the narco wars, also came by most days to chat. The Alva family compound was located about a half-acre in a colonia called Benito Juárez on the other side of the intermittent arroyo that ran through the center of town. The lot was at the ridgeline of a linear, one-street-wide hill that looked down to the center of town and was circled by a new galvanized chain-link fence. Running along the dirt street, the house, built in the 1960s, was a clean, bright five-room palimpsest of stuccoed cement block that had grown from just two rooms that were now the living/dining room and kitchen. A toilet and shower were housed in an outbuilding near the house beside the 1,500 liter water tank that was used to collect water for household use from the municipal tubes that were turned on for scheduled times each week. Miguel’s woodworking shop was just downslope of the bathroom. A concrete shed at the other end of the property was the television/DVD/library room. The dooryard, whose plantings would be an ethnobotanist’s dream, was where most evenings a group of friends and neighbors would talk, complain, and kill time while drinking coffee or beer. Mangoes, annonas, guayabas, and orange trees were the most common trees around the lot with a large tempesquixtle (Siderox - ylon palmerii ) tree providing shade in its center. Pots of plants, both ornamental and medicinal, 349 were placed in large groupings near the house. Besides being a recreational gardener and plant enthusiast, Doña Tere would often be consulted by neighbors and people from the sierra commu- nities for botanical-traditional medicines. At this early stage in my research, I was planning my Q-methodology survey while talk- ing to residents about garbage and what issues related to it affected their lives the most. I rode along with the garbage collection trucks to get a sense of the daily practice of garbage collection across the municipio. Besides working toward an understanding of research goals, I also spent a lot of time hanging out; meeting people, going to community events, participating in the rigor- ous fiesta schedule, and helping out. I spent quite a bit of time in Miguel’s workshop and became especially adept at planing and sanding pine wood from the sierra before it was turned into fur- niture, doors, or cabinets. The trips to install doors and cabinets in Coxcatlán and in other towns were particularly interesting, both for me and Miguel’s clients who were not accustomed to a gringo carpenter’s apprentice. In late March, I began my Q-method survey and was introduced to many of the participants through my friends and professional contacts. Having been offered a place to stay by the Alvas early on, I eventually moved in by the end of March. Despite numerous efforts to leave money for my room as well as account for Figure A1.9. View from the author’s residence in Coxcatlán. Photo by A. Hilburn, April 2011. 350 the food I was eating, neither Doña Teresa nor Miguel would accept any of it, which became a source of stress for me. I took it upon myself to buy groceries daily and had my parents who came to visit me in April as well as my wife who visited in the spring and later that summer bring nice gifts from the U.S.. I also ran errands and helped out with the housework. Miguel’s 19-year-old daughter, Teresa, came to live in Coxcatlán from the state capital for the late spring and summer before she began her first semester at the state normal college in the fall. During the first half of my stay, a typical day began around 6:30 a.m. when I would organize notes, read, or enter data. We shared a wi-fi connection with Miguel’s younger brother, Emiliano, a state health worker, who lived next door. I would huddle beside the wall separat- ing the lots where I could get a decent signal and check email or chat with Alison via Skype or Instant Messenger. The internet connection was not fast, highly intermittent, and at times out for hours. I would often sit on the hard concrete beneath the household water tank to not connect. On the other hand I was not trapped in my over-heated office in Lindley Hall on KU’s campus. After using the internet, I would then take a walk around town, finally picking up fruit or eggs or fresh tortillas on the way home. After breakfast, around 8:00 a.m., I would then prepare all my materials for whatever I research I had planned for that day and set out. If it was a slow day, I would come home for the comida or dinner around 4:00 p.m., but if I was busy I would eat junk wherever I was working and return around 6:00 p.m. After the comida or arriving home from a long day out, I would get back to work for a few more hours, either conducting or setting up more Q-surveys, entering data, taking down notes, or planning a new research project. If not, I would go for a walk, use the internet, read, or buy beers from the lady who ran a shop down the street and help out Miguel in the workshop. Around 9:00 p.m., we would eat bread and coffee or small servings of leftovers for the light cena before everyone went their way to prepare for bed around 10:30 or 11:00 p.m.. Days like this went on until the end of May, when my Q-surveys wrapped up and Alison came to live with me. For the five weeks that Alison would be around, I had planned to conduct the first two-thirds of my litter study. I also planned to prepare, organize, and test my garbage census and risk mapping exercise that would begin in July. Besides all of my work, I was happy to have Alison around. She did not speak Spanish, but knew enough to understand questions and make basic responses. Her days were kept busy by working on her field statements for her final comprehensive exams but I also had her help me with the litter study, in which she eagerly par- 351 ticipated. However, many friends and acquaintances would rib me about being abusive by mak- ing my wife pick up garbage in the heat on the side of the highway. After more than three weeks, we had selected the sites to be used for the study and had cleaned all the litter from them. Then we needed to wait a month before returning to them. My next focus was a project on garbage dumping on communal lands in the municipio. I drafted and submitted a proposal to the local ejido authority to map the tiraderos clandestinos or unauthorized dump sites on Coxcatlán’s ejido ’s lands. This proposal was initially rejected by the comisariado (ejido president) on the grounds that it would bring bad publicity to the ejido being dirty and contaminated. After reconsideration, the comisariado and his advisers accepted it with the ambitious condition that we map and then clean up the sites which could then perhaps be turned into a public event with an emphasis on the latter more than the former. Understand- ing their wishes and recognizing their authority but also trying to avoid being amid a political struggle between municipal and ejido authorities, I respectfully declined the offer and thanked them for their time. Figure A1.10. Walking about in Colonia Benito Juárez. Photo by Alison Hilburn, May 2011. 352 Late June: Work interruption I spent the most of Monday, June 20th, 2011 waiting for and in meetings at the ejido of- fice and I was not feeling well, both disappointed that my plan had failed and also feeling physi- cally icky, so I returned back to the Alvas’s house and spent the remainder of the afternoon and evening hanging out with Alison and reading. Later that night, I began to feel a burning sensation around my diaphragm. I initially ignored it and I had my coffee and bread for cena. Over the next hour or so, this burning sensation gradually shifted to an increasingly acute pain that first migrated to my stomach, which then forced me to vomit, and then localized itself on the lower right side of my abdomen. Alison and Miguel suggested that I go the local, state-run health clinic but I told them that I would go lie down in bed, thinking it was just stomach cramps. After all, I had not been ill since I had arrived in Mexico months earlier and I had thought myself adjusted to the local foodways. Yet the pain only intensified over the course of the next hour and finally Alison told Miguel that I needed to go to the clinic. Alison, Miguel, Salvio, and I piled in the Volkswagen Thing that Miguel, Pepe, and I had restored months earlier to go down the slope to the clinic around midnight. Pepe and Paty followed behind on his motorcycle. At the clinic, there was no one in the waiting room and Miguel and Pepe began calling for some attention. A nurse appeared and she immediately took me back to one of the rooms where a much more ill-looking young man was in a bed with an IV in his arm. The attending physician, a man in his 40s, came over to examine me as I described my symptoms. After dis- cussing details and what had transpired in the previous few hours, I think he knew immediately what my condition was. I was unsure or in denial. Nevertheless, and for due diligence, the doctor began a series of physical tests where he told me to tell him where it hurt as he prodded various locations on my abdomen. When he applied light pressure on the lower-right part of my belly, I yelped in pain. He knew it right away what it likely was, but gave me a hernia test anyway to make sure that it was appendicitis. Unfortunately, the ultrasound machine for the clinic was be- ing repaired at the time, so he told me that I would need to go to Tehuacán for a more definite diagnosis and perhaps, a subsequent appendectomy. I was then given anti-inflammatory drugs while Emiliano, Miguel’s brother who worked for the state health department, pulled strings to have an ambulance take me to Tehuacán general hospital. I was scared. This was not what I needed to happen during my stay but there I was in an ambulance on the road to Tehuacán with Alison and Miguel, who felt responsible for myself and Alison. 353 We arrived at Tehuacán’s general, state-run hospital in about 45 minutes which is about half the time it takes by car. I was admitted and was given a bed beside an immobilized elderly lady in a neck brace and a young woman who was vomiting every fifteen minutes while lightly crying. After blood was drawn and X-rays taken, Alison and I waited together while Miguel attempted to rest in the cramped waiting room. Comparing myself to my neighbors’ obvi- ously worse predicaments and now feeling the effects of the anti-inflammatory meds, I began to think that maybe this was not appendicitis and we would be back in Coxcatlán with a funny story. However, the two attending physicians doing rounds, one of whom was teaching medi- cal students at the time, read the x-rays, perused the test results, and performed physical tests. Both doctors independently came to the easy conclusion that I had a classic case of appendicitis and I would soon require an appendectomy. When consulting one of the rotating physicians, I asked about my grant-sponsored health insurance covering the bill. He seemed surprised that I would be at the public hospital if I had insurance and told me that although the care and surgery would be of a similar quality, one of the private hospitals in town would allow Alison space to relax beside me more comfortably while I recovered. I still needed to get an ultrasound and an expensive-looking private Catholic hospital called the Corazón Sagrado (Sacred Heart) was near a recommended radiologist. Around dawn, I was discharged from Tehuacán General under the doctor’s orders that I go directly to the other hospital for consultation and likely surgery. The ultrasound confirmed my inflamed appendix and I began to prepare my own frayed mentality for surgery. This decision was made easier after the consulting physician at Corazón Sagrado gently but firmly explained to me the differences between the invasive, more risky surgery needed to remove a burst appendix than the more routine one to remove it beforehand. I agreed and admit- ted myself to the hospital. The same very capable and reassuring surgeon I had met earlier at Tehuacán General was making his rounds at the Corazón Sagrado. I was nervous because I had never had surgery before but I felt like I was in good hands. Alison, who had slept even less than I had, was similarly anxious but reassured by the nice surroundings and my translation of the doctor’s assessment of my condition and treatment plan. We had not contacted any family as of yet, but Alison called her parents, who are an Registered Nurse and a Certified Registered Nurse Anesthaestist, to ask their opinion of the situation and they agreed that what we were doing was best. With just less than a bar of battery life, Alison called my father who happened to be driving my brother, sister-in-law, and 15-month-old nephew to the airport in New Orleans after a visit 354 from New York where they live. My brother, J.D., an artist in the medium of losing things, had brought his passport after having lost his driver’s license a week before, suggested that he come lend moral support in lieu of my father who would have to go back to Mobile and then likely leave from New Orleans the next day. He also did not have to return to work for five more days. He bought a ticket at the New Orleans airport for a late night flight to Puebla from Houston and was scheduled to take a bus to Tehuacán the next day. My surgery was scheduled for later that afternoon. Miguel was beyond exhausted and had earlier gone back to Coxcatlán to eat and take care of some business before coming back with Teresa before my surgery to wish me luck. The surgery was pretty simple and straight- forward. After getting prepped for surgery, I was wheeled to a small but clean operating room. There I was given an epidural anesthetic and a sedative while they hooked me up to a number of monitors. I was going to be awake during the surgery but they pulled a screen in front my chest as is done during Caesarian section births. The surgeon got to work. I felt pressure but no pain. The anesthesiologist sat beside my shoulders and kept me company. He had worked at the Cox- catlán clinic on a residency after medical school, so we chatted about Coxcatlán while the sur- geon operated and occasionally chimed in on what he was doing. After what seemed about thirty or forty minutes the surgeon looked at me and told me that it was done and everything was fine. Indefinable things happened as they readied me for recovery, and finally, the surgeon asked me if I would like to see my extracted appendix. I told him that I did and he showed me something that looked like a giant tadpole or fried egg with a tube attached to it. He proceeded to outline the normal size of an appendix which was less than half of mine. I was then taken back to my room to recover. Alison had a couch to hang out on and Miguel and Teresa kept us company until they could see we were both tired and they left later that evening. I spent the next two days in that room recovering with frequent nurse visits. I had never had surgery or hospital care before, but the quality of care that I received at the hospital was excellent. My brother arrived the day after my surgery and had rented the presidential suite at the Hotel México, a large colonial style hotel only three blocks away. The Alvas and friends came to visit during my stay but the next day or so was a blur of not-too-bad abdominal pain, bland food, and bad Televisa programming. On Thursday, I was released after J.D. and Alison withdrew cash from the ATM and paid the far-too-small sum of 20,000 pesos or about $1,700 US Dollars for the surgery, hospital stay, and fees. I walked with my brother and Alison the three 355 Figure A1.11. The author’s appendix in a jar of formaldehyde. Photo by A. Hilburn, April 2013. Figure A1.12. Going back to the Mexico City airport by bus for recovery. Photo by A. Hilburn, June 2011. 356 blocks to the hotel wearing basketball shorts, flip flops, and a ridiculous t-shirt that J.D. had bought in a nearby store with a quote that read “I feel the need for speed. Powered by awesome” with a vintage motorcycle on it. We settled in at the hotel and proceeded to relax for the next couple of days. We could not do much but we watched black market dvd’s and hang out. Throughout my recovery, I was wor- ried about continuing my work. After I described a typical work day, the surgeon recommended four-to-five weeks of recovery. I had corresponded with my grant officer at KU, adviser, and COMEXUS liaison who all but ordered me I would need to come home to recover. At the time, I did not want to as my work was really going well. Alison, however, would not accept me staying. Also, when I told them about staying for my recovery, Miguel and my Coxcateco friends sarcas- tically asked me how effective I would be at my work with an abdominal gash. The US Depart- ment of Education was extremely nice and allowed me to leave and reactivate my grant when I came back. The surgeon wanted to see me a week after surgery to remove the stitches and to see if everything was fine, so Alison and I stayed until Wednesday at the Hotel México after J.D. left on a bus back to Mexico City on Saturday. We booked tickets for a flight out of Mexico City, I had my stitches removed on Wednesday, and we left the next day for Kansas City for a month’s recovery. Getting back to it: August to December 2011 I returned home to Lawrence at the end of June and I stayed there until August 12, when I returned to Mexico to finish the four months of research I still had under my Fulbright-Hays fellowship. I came back to Coxcatlán with a sense of urgency and a ready purpose to finish my research strongly. I had achieved quite a bit already, but I still needed to conduct the household garbage census I had earlier planned and I needed to continue the follow-up collections of the litter study I started in early June. Coxcatlán was much the same as it had only been five weeks before except that it was far greener and more humid than when I had left. In late June before I fell ill, the heavy rains of the wet season had just begun but had not had the greening effect that I saw when I returned in August. Streams actually had water in them and all the house lots were covered in greenery. I was warmly welcomed back by the Alvas and all my friends that I had left and began to catch up. After a few days stay, I later found the house was going to be crowded due to family visits. I was invited to stay but I did not want to intrude and my friend Pepe had a 357 room in his house, so I moved in with him. I was still invited to basically live at the Alvas, so it was not a big move. Besides, I wanted a change of scenery. Pepe lived just across the street from the town plaza and life in the centro seemed a worthwhile different experience. Also, Pepe had a cool house, literally and figuratively. A few years earlier, he had converted half of his lot into a multi-tier bar, the only one in the entire small town, which he opened for holidays and weddings. The front of the house was his successful internet/telephone/copy/cell phone minutes business and his wireless signal connected to an unused adjacent room that I used as my office. It was like working next to a Kinko’s. My room was huge, about 25’ by 20’ with a 20’ ceiling. The walls were about two feet thick so the space was cool in the day and warm at night. I had no furniture beside a bed and a few tables borrowed from the bar that I used as dressers and a hanging mos- quito net. I also had my own bathroom which was nice. The home-y feeling of the Alvas was great for the first half of my stay when I was getting used to life in Coxcatlán but the bachelor pad feel of Pepe’s was perfect for my hurried last four months of work. Besides, I did not feel as I was imposing so much as Pepe was always either working in his shop or in his sugarcane fields or was with his girlfirend Paty. All of my friends still met three to four evenings a week for cof- fee or beers or light dinner. We began to rotate between Pepe’s and Miguel’s houses. We cooked far too infrequently and rarely did laundry because of our busy schedules, so Paty suggested we hire someone to help out around the house. We hired Lupita, a young lady who lived in Coxcat- lán but was originally from a sierra community up from Coxcatlán and had migrated to Mexico City for a few years. She came twice a week to straighten up and cook enough for leftovers. My first research objective when I got back was to collect and weigh the litter that I had hoped had accumulated on my study sites during the previous ten weeks. I had initially planned to have the interval between collections set at six weeks, but my appendix issue had thrown that out the door. Compared to the initial cleanup that I had done with Alison, the expectedly less lit- ter was manageable alone. After five days, I had finished. I then started my garbage census. For that exercise, I needed a research assistant. After discussing this with Pepe, who through his busi- ness that most everyone in town seemed to use, suggested a young man who had experience in giving surveys and would be a responsible researcher. When Ricardo, who also went by “Klever” due to his alleged likeness to a soccer player of the same nickname, showed up to our prescribed meeting I was a little unsure of this. Perhaps it was his low-profile purple Mohawk or his 1980s retro hip-hop style, but after a few minutes discussing my project I realized he was a capable and 358 intelligent guy. After discussing pay rates with Pepe beforehand, I decided to offer Ricardo twice the daily rate for similar work. Besides working for me, he was a part-time albañil or cement worker and a saxophonist in a cumbia band that played town fiestas and weddings on the week- ends. After some training and practicing we began testing the census questions and the appended risk mapping exercise. Following the revision of the survey document and the interview proce- dures, we began in late August. I settled into a daily routine where we would survey during the hours most residents were not busy with meals or childcare, usually after breakfast around 8:30 a.m. to 3:00 p.m. when most people were eating the large daily meal. During this time we worked in the valley settle- ments. First, we surveyed the two largest towns Coxcatlán and Calipan, then we moved on to Tilapa and the small valley settlements of Pueblo Nuevo and San Rafael. We used buses and cabs to get from Coxcatlán to where we worked for the day. Coxcatlán is only about 250 square kilometers and Tilapa, the farthest valley settlement form Coxcatlán, is only twenty minutes away. When possible, we worked five days a week in the field with Saturday spent entering data and discussing the results and notes of the previous week. When Ricardo had a gig or had to help his brothers with a construction job, we would not survey. I spent Sundays relaxing or going to the sugarcane fields with Pepe or Miguel. Occasionally, I would go to Tehuacán to see a movie, eat a good meal, drink a draft beer, and buy stuff. The busy town fiesta schedule, especially the September 15 and 16 Independence Day celebrations also provided welcome distractions. For that holiday, I volunteered my mixologist skills and made micheladas (beer, lime, and chili pow- der cocktails) in two almost all-night shifts at Pepe’s bar that he had temporarily opened for the holiday. By the beginning of October, we had conducted about 350 interviews in the valley and we were ready to survey the sierra communities. Beforehand, we revised the survey document to streamline the process for the sierra while keeping the data collection consistent. The survey- ing was much slower and laborious in these towns. First, getting to one of the towns in the sierra took far longer, usually an hour or more than the easily accessible valley towns. The schedules of the sierra-bound buses were a little less predictable, the ascent was slow, averaging around 10 mph on the twisting bends and switchbacks in refurbished school buses. Many sierra towns were only accessible on dirt roads that were a few kilometers walk below the ridgeline high- way. Second, we would meet with the officials of each town before starting our surveys. We had 359 a permission letter from the municipal president and no local official objected, but we wanted to make sure that everyone in charge knew our intentions and the basics of our research goals. Lastly, unlike the more compact and gridded settlements of the valley, the sierra communities were spread out amid small fields and an extensive network of dirt roads and paths. For the next five to six weeks, we surveyed seven sierra towns and wrapped up the census while I prepared another Q-methodology questionnaire and trained Ricardo to assist me with its administration. We conducted ten Q-sorts in Coxcatlán but left it alone while we conducted the second collection of litter. By the time we finished the litter project, Alison came down on Thanksgiving break and the November 20th commemoration of the Mexican Revolution. The ejido -sponsored fiesta and town rodeo/all-night bender were exceptionally notable. The last month of my stay was a blur. I spent my time filling in the gaps of my under- standing of garbage management there. I verified hanging questions in my field notes and con- ducted some final interviews with the regidora (regent) of ecology, the lady in charge of munici- pal garbage service in Coxcatlán, to confirm particular assumptions that I had made. For a little more than a week, I returned to the state capital, Puebla, to meet with officials at the Secretaría de Sustentabilidad Ambiental y Ordenamiento Territorial (SSAyOT) and request materials as- Figure A1.13. The view from my room and mosquito pabellón. Photo by A. Hilburn, August 2011. 360 sociated with the garbage management plan for Coxcatlán. I also had a “books and stationery” allotment in my grant to spend. I scoured bookstores near the city’s universities, finding a num- ber of excellent books on garbage, Mexican ecology, and reference books on plants, Nahuatl, and indigenous peoples that I could not have found in the US. The state INEGI office (Instituto Nacional de Estadística, Geografía, e Informática) was a worthwhile visit where I spent the re- mainder of my research money on maps, GIS data, census data, and published materials. I returned to Coxcatlán for the last two weeks to do the things that I had put aside for perceivedly more important tasks earlier in my stay. On a side project, I wanted to examine how the town of Coxcatlán has expanded in the past twenty years, so I performed a ground-truthing exercise and town mapping for a few days all over the town. The sugarcane zafra , or harvest, was beginning and I helped prepare Pepe’s as well as some of his siblings’ fields before and after they were harvested to understand this process. There were also a number of places that I wanted to photograph across the municipio for slides and figures. Lastly, I also wanted to spend time with my friends and what felt like my Coxcateco family before I left. Unfortunately, the penultimate week of my stay was punctuated by the passing of Pepe’s chronically ill mother. The deceased Figure A1.14. The rodeo for the 20 de noviembre celebration. Photo by A. Hilburn, November 2011. 361 was the matriarch of one of the most influential families in town and the mother of fourteen children including two previous municipal presidents. The resulting funerary proceedings were of considerable importance in the town and municipio. The week was filled with late-night vigils, church services, meals, and an immense funeral at the end. After the funeral, preparations for the annual town fiesta and Christmas celebration took over the town’s collective interest during the last week of my stay. Coxcatlán’s patron saint is St. John the Baptist, whose feast day is Christmas Eve, and business slows down as preparing for the holidays increases. The last few days of my stay were spent hanging out with the Alvas, eating at other friends and acquaintances’ houses, and finalizing my travel plans back to the US. On December 18, I said goodbye to the Alvas and my immediate close friends before Pepe and Paty drove me to the bus stop in Tehuacán for my trip to Puebla and then Mexico City. The next morning I left Mexico City for Houston and Kansas City. Alison picked me up at the airport and drove me back the 45 minutes to Lawrence. I was happy to be back with Alison with some sense of permanence but I also missed my life in Coxcatlán. I was back, missing a semi-vestigial organ but I had a mountain of information and data, and more importantly, invaluable experiences and close friends. 362 Appendix 2: A Prehistoric and historical regional geography of Coxcatlán To academics and interested outsiders, Coxcatlán, Puebla is known primarily for its archaeology. This is mainly the result of the lasting impact of the Peabody Foundation-sponsored Tehuacán Valley Project led by the Canadian archaeologist Richard “Scotty” MacNeish and a large interdisciplinary team of archaeologists, cultural anthropologists, geologists, geographers, and historians who worked in the Tehuacán Valley, including Coxcatlán, from 1959-1963. The project’s primary goals were to locate the origins of Mesoamerican agriculture and to estab - lish a sequence of occupancy for the Tehuacán Valley that would also serve as a broader model Figure A2.1. Tehuacán Valley Projec t excavation crew at the Cueva del M aíz; Richard M acNeish front-right. Photo scanned from a postcard given by M acNeish to crew members. Photo cour tesy of Oscar Nuñez. Date unk nown. of social and economic change for most of highland Mesoamerica. Coxcatlán figures into this story largely because of sediments found in a rock shelter approximately eight kilometers south of the cabecera municipal. Within these sediments the research team identified what were then believed to be the oldest corn cobs along with remnants of other predecessors of contemporary domesticates such as cotton, avocado, and squash. While recent analysis of radiocarbon dates from these famous artifacts diminishes their previously stated antiquity, Coxcatlán Cave, as it is 363 known in published material, remains an important site in the minds of archaeologists as well as the general public. The municipio of Coxcatlán proudly wears the name “Cuna de Maíz” (Corn’s Crib) on its signage. Archaeology is important there, and for that reason the following outline on the prehistory of Coxcatlán and the Tehuacán Valley is provided. More importantly and related to this research, the Tehuacán Valley Sequence is a deep biography of garbage and disposal in ancient Mesoamerica. Human prehistory in most any place in North America can go back as far as the first known occupations by Beringian migrants around 13,000 years ago with at least Clovis culture (Dillehay 2000; Meltzer 2009) hunters and gatherers and their likely predecessors. There are no Clovis sites in and around the Tehuacán Valley. The archaeological survey by Richard MacNeish in the 1960s provides a long sequence of occupancy and culture change in and around the Tehu - acán Valley (Figure A2.1). The following is a brief summary of MacNeish’s hypothetical sequent occupance from about 12,000 years ago to the arrival of the Spanish in 1520 near Tepeaca in the Puebla-Tlaxcala Valley. In any other context, the following detailed description of the study area’s prehistory would appear superfluous and overwrought, yet the MacNeish et al. project essentially used prehistoric waste to construct their narrative of cultural and social change in the Tehuacán Valley. The Tehuacán Valley sequence is an account about a place and its societies told through the garbage they generated and left there. It is also worth noting that recently the methodological integrity of the Tehuacán Val - ley project has come under substantive scrutiny (Hardy 1996 ; Smith 2005), yet these criticisms address methodological issues and do not entirely refute the descriptive capacity of MacNeish et al.’s model of prehistory in the Tehuacán Valley and Mesoamerica abroad. The population estimates are by no means determinate figures but are simply just that, estimates. It is important not to take these descriptions of the settlement history of the region too literally as to their exact chronologies and as precise suppositions of subsistence, settlement, and society. Rather, the Te - huacán Valley Sequence, as it is known, is best viewed as a hypothetical narrative of prehistoric lifeways and settlement in the region and across central Mesoamerica. Palaeoindian to Archaic The prehistory of the region begins with the Ajuereado phase, named for a highly eroded hill in southern Coxcatlán municpio, which is posited as the earliest known occupation of the 364 Tehuacán Valley beginning around 10,000 B.C. and lasting until about 7,000 B.C. (Johnson and MacNeish 1972: 16) . The majority of the artifact evidence from this phase comes from the excavations at nearby Coxcatlán Cave (Tc50), known locally as the Cueva de Maíz as well as others nearby and in the vicinity of present-day Tehuacán (Figure A2.2). Ajuereado phase would be classified as part of the Paleoindian period of North American prehistory (Evans 2004: 86). The climate of the Tehuacán Valley during this time was likely slightly cooler and drier during this time, with the valley floor perhaps representing a steppe-like vegetation of grasses and small shrubs. A comparable vegetation assemblage would be something like what is found on highland steppes in the State of Chihuahua (Flannery 1972: 144). The population of the entire valley was probably quite low, with only about a dozen to twice that amount living in 2-3 microbands whose mode of subsistence was mainly hunting with some gathering of wild plants and small biota. Due to the antiquity of the this phase, there is comparably less faunal evidence of what comprised the foodshed of the early Ajuereado phase but what does exist represents a fairly broad spectrum of fauna. Large mammals include now-extinct species such as wild horses, antelope, and a large fox. An extinct gopher tortoise as well as a large jackrabbit was likely hunted. An array of small mammals such as ground squirrels, wild rats, mice, small rabbits, skunks, and coyotes are also Figure A2.2 Cueva de M aíz. Coxcatlán municipio. Photo by A. Hilbur n, O c tober 2011 365 part of what is thought to have been hunted during the Ajuereado (Flannery 1967). The latter part of the Ajuereado marks a transition to the faunal assemblages seen today in the Tehuacán Val - ley (Flannery 1972) The Ajuereado toolkit found is composed primarily of chipped stone blades, projectile points, scrapers, and gravers, although this is likely only part of it as bone and wood tools were not preserved in the ten to twelve millennia since their manufacture (MacNeish 1964). Archaic Coxcatlán The El Riego Phase, named after a cave and springs near the present day city of Tehu- acán, marks the transition from the Paleoindian period to the Early Archaic. This phase begins around 7,000 B.C. and ends around 5,000 B.C. It is characterized by a quadrupling of population and the wet-season coalescence of microband camps into macrobands to supposedly optimize gathering. El Riego also marks a slight movement away from greater reliance on hunting with increasing subsistence on plant-based food sources. White-tailed deer ( Odocoileus virginiana ) and cottontail rabbits become more common in the faunal assemblages (Flannery 1967: 159). Wild versions of squash ( Cucurbita spp. ), chile ( Capsicum spp. ), and avocados ( Persea ameri - cana) are found in archaeological strata of this phase. Some textiles and basketry as well as a full suite of chipped stone tools were used by the El Riego residents and grinding tools suggesting the processing of seeds and plant fibers. Ceremonially cremated human remains as well prepared and ornamented burials are defined El Riego cultural traits (MacNeish 1964). The subsequent Coxcatlán Phase (5,000 – 3,400 B.C.), named after the study area mu- nicipio and the famous cave, illustrates a further movement towards subsistence on plant-based resources and the beginning of domestication. Manos and metates (mortar and pestle-like imple - ments) are more common as the need to process this increasingly plant-based diet became neces- sary. Early varities of maize (Zea spp. ), beans and tepary beans ( Phaseolus spp. ), amaranth ( Am - aranthus spp. ), chiles, squash, avocados, and yellow zapotes ( Pouteria spp. ) become apparent from the early point in the phase. Population likely increased to ten times the Ajuereado popula- tion. A macroband-microband settlement pattern existed, similar to that seen during El Riego, but with perhaps larger macrobands with longer occupation of wet-season camps (MacNeish 1964). Hunting patterns generally follow those of the previous phase with wet-season exploitation of deer and peccary on the high slopes of the eastern sierra and the dissected canyonlands of south- ern Coxcatlán municipo and the Cuicatlán Cañada with perennial trapping of rabbits and other 366 small game on the valley floor and western slopes of the valley (Flannery 1967). Preclassic beginnings The Abejas Phase (3,400 – 2,300 B.C.) represents a notable change in the settlement landscape of the region. For the early and middle parts of this phase, the micro-macroband pat - tern of seasonal encampments persists, but in the latter portion one can see a transition to seden- tism with what are believed to be pit houses in semi-permanent villages. The population at this time is approximately 40 times the original population at 500 - 1,000 people in the valley and surroundings. The subsistence model shows an increasing transition to domesticated or semi- domesticated plants with occasional camps for hunting and gathering (MacNeish 1964). Hunting still provided important sources of protein but became more focused on deer and rabbits hunted in seasonal camps. Domesticated dogs also arrive in the archaeological record at this time and likely provided another, more sedentary meat source (Flannery 1964). Stone bowls and ollas (pots) are noted from this phase. More intricate basketry and also the beginnings of the contem - porary forms of concave metates (mortars) appear as well (MacNeish 1964). Figure A2.3. The ancient Presa Pur rón or M equitongo, Coxcatlán municipio. Photo by A. Hilbur n, D ecember 2011 367 The Abejas Phase transitions into the Purrón Phase which began around 2,300 B.C. and ended roughly 1,500 B.C. The Purrón Phase is the least understood of the Tehuacán Valley sequence owing to the few artifacts available in the strata from dry caves around the valley. It is notable for the first ceramics in the region. These were modeled on pre-existing stone bowls and ollas or natural forms such as gourds called tecomates. A similar settlement pattern to the Abejas phase is hypothesized for the Purrón phase (MacNeish 1964). Sites from central Mexico (Sand- ers, Parsons, and Santley 1979) to the Mixteca Alta (Byland and Pohl 1994; Spores 1984) and Oaxaca Valley (Marcus and Flannery 1996) all the way to the Soconusco Coast (Blake 1991 ; Clark 1991) have contemporaneous archaeological components with ceramic artifacts and could prove to be comparable models for the settlements and their political systems. The Ajalpan phase (1,500 – 850 B.C.) officially notes the larger Mesoamerican Period known as the Formative or Preclassic. For the Tehuacán Valley, most of its residents at this time were full-time farmers planting a fairly comprehensive assortment of New World domesticates. They practiced wet-season agriculture in and around the barrancas. Some seasonal hunting and gathering added to the subsistence of Ajalpan Phase people (Flannery 1967) Ceramics become more present in the archaeological record and the typical settlements were small villages of 100- 300 residents with a total regional population is estimated at. The most common dwellings for this time were wattle-and-daub huts (MacNeish 1964). The Santa Maria Phase is contemporaneous with the larger late Preclassic or Late Forma - tive Periods (850 – 150 B.C.). MacNeish posits further urbanization in the valley with smaller permanent settlements oriented around larger villages with some non-utilitarian ceremonial buildings (MacNeish 1964). Irrigated agriculture arrives during this phase (Woodbury and Neely 1972; Doolittle 1990). One of the oldest instances of New World irrigation is the Presa Purrón and Mequitongo dam, located just 2.5 kilometers north of the town of Tilapa in Coxcatlán. The latter ancient edifice made a substantial reservoir by damming an intermittent stream in a strate- gically narrow point in its valley that was used for irrigating nearby fields (Woodbury and Neely 1972) (Figure A2.3). MacNeish (1964) hypothesizes that the larger settlements in the valley could be indicative of secular rather than ceremonial centers built around the control of irrigation infrastructure by secular rulers rather than a priestly elite that were believed to control the ele- ments supernaturally, which is the model in the Gulf lowlands and Yucatan Peninsula at the same time. Agriculture was the primary mode of subsistence. Ceramics became more complex with 368 bichromatic designs and complex incision methods. Hunting becomes much localized around permanent settlements and eventually becomes of trivial importance toward the end the Santa María Phase (Flannery 1967). Classic Coxcatlán and Sansuanchi The Palo Blanco phase (150 B.C. – A.D. 700) represents the Classic Period in the Te - huacán Valley. The settlement landscape of Palo Blanco is characterized by villages with some socio-political orientation around centers with full Mesoamerican ceremonial complexes includ- ing plazas, pyramidal buildings, and ball courts. Agriculture remained the dominant subsistence mode with a greater array of crops including tomatoes, lima beans, and peanuts. The domesti- cated turkey ( Meleagris gallopavo ) becomes another source of household protein for villagers living in wattle-and-daub huts. Irrigation becomes systematic and of greater importance during this time. The estimated population of the valley was around 11,000 inhabitants (MacNeish et al 1964: 535). The final phase corresponding to the Postclassic period of Mesoamerica is known locally Figure A2.4. Coxcatlán Viejo or S ansuanchi, the site of the postclassic cacicazgo situated at the easter n end of the cabecera municipal. Photo by Alison Hadley, M ay 2011 369 as Venta Salada (AD 700 – 1540). Cacicazgos or small kingdoms seated in hilltop ceremonial and secular centers surrounded by loyal hamlets and villages appear to be the sociopolitical sys- tem of this time7 . The secular ruler, known as a cacique or chief, likely based their power on the control of irrigation resources (Enge and Whiteford 1989: 61). Full time agriculture with exten- sive irrigation was the primary mode of subsistence for Venta Salada people. Craft specialization is apparent with artisans producing an array of goods for the growing population of the valley including cotton textiles and fine, mold-made ceramics. The site of Coxcatlán Viejo or Sansuan- chi8, was one of these centers. The estimated population for the entire valley was 55,000 inhabit - ants (MacNeish et al. 1964). Sansuanchi, the old cabecera of postclassic (Venta Salada Phase) Coxcatlán is uphill from the north bank of the Barranca Soyolapa that bisects the present-day cabecera of Coxcatlán mu- nicipio (Figure A2.4). This separation from the current town is much less distinguishable today than in decades past as Sansuanchi’s ruins have already or are becoming part of the town fabric as Coxcatecos build houses nearer and nearer to the central lomas that bear the ancient settle- ment’s principal pyramids and plazas. Portions of the colonias of Benito Juárez and Pala are now on top of the site’s easternmost flanks. Goat herders graze their animals there today and over 50 years ago, a few cycles of milpa were planted on the site and the last standing walls of the site were knocked down. Shrubs and giant cardón cacti dot the site and trees grow around and some - times through the 3-5 meter high pyramids, which from a distance look like mere vertical exten - sions of the hilltops. A closer inspection gives a better view of the terraces, plazas, and rectilinear buildings that gave the old town its form. Potsherds or tepelcates of a wide array of shapes, patterns, and polychromatic finishes are strewn everywhere on the ground. Chert flakes and the occasional piece of obsidian also litter the surface. A fair amount of looting is present and holes, some fresh and others partially filled in, can be seen near any obvious archaeological feature. Today, Sansuanchi is a ruin of a town. That was not the case in the Postclassic. The Peabody Foundation-funded Coxcatlán Archaeological Project of the early 1970s led by Edward Sisson provides the only archaeologi- cal narrative of life in the southern Tehuacán Valley in the period before the conquest. Postclassic Coxcatlán was then a regional center with its own hinterland of hamlets and villages over which it held political domain. The settlement itself was about one square kilometer in area and was laid out on five small ridges framed by the Barranca Soyolapa to the north and Barranca Atem- 370 pango to the south. It was a highly defensible site with the sierra to its back and a commanding view of the valley. Across the site there are ten pyramid plaza complexes of varying size and a number of rectangular buildings, which were perhaps workshops or residences (Sisson 1973). Besides locally produced ceramics, other types recovered from the site have origins in the Valley of Mexico, the Puebla-Tlaxcala Valley, the Gulf Coast, and the Mixteca Alta suggesting an ex - tensive trade network with important postclassic centers (Sisson 1973). Additonally, non-native lithics such as obsidian, which is igneous in origin, were recovered in relatively high frequency further suggesting vast networks of exchange (Sisson 1973). Salt produced from the numerous salinas in the valley was likely the most valuable export from the immediate area (Sisson 1973). The estimated population of Sansuanchi was anywhere from 800 – 100 (Smith 2005: 422; Sisson and Lilly 1994), although its colonial period population in 1548 is listed as 2,538 (Byers 1967). Sisson’s count appears more believable and it is based on taxation records from 1580. Although Sisson’s household population estimates appear too low, it should be noted that in the in the 32 years between counts, the population could have declined substantially from dispersal or disease. While Sansuanchi was the largest postclassic settlement in what is now the municipio of Coxcatlán, there were smaller contemporaneous sites throughout the valley and in the sierra above town. The Venta Salada site, located in the northern half of the municipio near the Río Salado was a similar size and substantial occupation, but lacking the more impressive ceremonial buildings. This site featured three distinct plazas, low platforms, and what are thought to have been workshops for salt processing. Another similar, more dispersed site of residences with a plaza and raised platforms sits just north of the present-day town of Pueblo Nuevo (Sisson 1974). In the intensely eroded hills and bluffs just north of Tilapa, hilltop “fortified” settlements were identified by the Tehuacán Valley Project led by MacNeish, but the later Coxcatlán Project attests to the ambiguity in the interpretation of these edifices, which could have been residential walls. The site of Postclassic Tilapa was identified nearby. Small hamlets of a few residences were the typical settlement in Coxcatlán’s sierra (Sisson 1973) The inhabitants of the larger settlements lived in residences that consisted of a group of slightly elevated rooms surrounding a sunken patio. Cylindrical columns helped support what were pole and beam roofs, helping lend the residential spaces an airiness and openness. All residences had hearths and debris caches. Stucco-lined pits were likely water basins. The nearly ubiquitous presence of spindle whorls in residences attests to the fact that weaving and textile 371 production were common household activities. Worship took place in nearly all the residences, evidenced by small shrines of censers and god figurines. The living typically lived with their deceased relatives as ollas of cremated individuals were found in depressions in most residences’ floors (Sisson 1974). Social stratification between residences likely did exist, but the results from the Coxcatlán Project do not point to any definitive answer showing clear material differences between elites and commoners. The ethnohistorical accounts of the nearby Cuicatec chiefdoms support this assertion of the relative equality of ancient Coxcatecos compared to the more popu- lated, politically significant, and more socially stratified Postclassic centers like Cacaxtla, Cholu- la, or Tenochtitlán (Hunt 1972). Conquest, Corregimiento, Haciendas, and Persistence The first contact between Coxcatlán and the Spanish was at a meeting in Tepeaca in 1520 where the leaders of Coxcatlán, then a tributary to the Aztec state at Tenochtitlán, sided with Cortés in his campaign to overthrow the Aztecs. After the fall of Tenochtitlán and the imposi - tion of the Spanish colonial state, Coxcatlán was made the head of a corregimiento (co-regency) in 1534 placing its administration, and taxation, under the direct rule of the virrey , or viceroy, who was the Crown’s agent in New Spain. Corregimientos were administered to by a corregidor (co-regent) based in Mexico City. The usual lack of presence of the corregidores in their own corregimiento meant that local forms of indigenous governance likely continued in Coxcatlán during the early colonial period. As a corregimiento, Coxcatlán had eleven subject towns stretch- ing from the Veracruz lowlands to its east to nearby portions of the Tehuacán Valley to the west (Sisson 1973). The ecclesiastic districting of New Spain placed Coxcatlán in the diocese of Tlaxcala in 1580. The church in Coxcatlán administered to the territory of its corregimiento as well as a few nearby towns. Its church was led by a local vicar who conducted service in the local Nahuatl. The corregimiento persisted until 1650 when Coxcatlán was absorbed into Tehuacán’s alguacil mayor (larger district) to whom it had always been suffragan (Gerhard 1972). Coxcatlán, although being a corregimiento heavily taxed by the Crown, was more fortu- nate in a sense, than other populations in the Tehuacán Valley and across New Spain. These other areas were subject to the encomienda system. This system “commended” or “entrusted” indig - enous lands and labor to an appointed New World “nobleman” who levied tribute and labor man - 372 dates in exchange for dubious notions of protection and the administration of Catholic religious services (Prem 1992; Chance 2000; Assies 2008). This extremely exploitive social and economic system took land tenure rights and natural resources away from indigenous communities, which had the effect of establishing highly unequal fiefdoms across New Spain. The abuses of encomienda eventually forced reform and in the New Laws of the Indies of 1542, it was abolished. Yet, through royal land grants (mercedes), squatting, and purchasing of land from desperate parties, wealthy elites in New Spain wrested even more land from indig - enous communities, establishing the hacienda or estate system, codified in 1622, which would persist until the Revolution of 1917 (Enge and Whiteford 1989; Assies 2008). Labor for the haciendas would be coerced through debt-peonage and the still-extant repartimiento laws where indigenous labor was conscripted for mandatory, but paid work on the haciendas . The hacienda system sought to replace subsistence agriculture of indigenous communties with large-scale com- modity production of cattle, wheat, minerals, and in the case of the southern Tehuacán Valley and Coxcatlán, sugarcane (Assies 2008). Through the colonial period, haciendas were established throughout the Tehuacán Valley, taking land and more importantly, water, from the still largely indigenous communities. In Cox - Figure A2.5. The ruins of Hacienda La S oledad. Photo by A. Hilbur n, August 2013 373 catlán, the irrigated portions around the Río Salado were and still are the prime areas for sugar - cane (Figure A2.5). In 1703, a small trapiche (mill) and associated sugar fields were established near Calipan. The indigenous residents there were further pushed off their lands in 1760, when the operation expanded (Enge and Whiteford 1989). Workers across the sugarcane lands of the valley were difficult to come by in the area owing to the brutal labor required in the cultivation, harvest, and processing of sugarcane. The hacienda owners, or hacendados , working with the colonial and subsequent Mexican state tried various coercive means in addition to repartimiento, such as taxes and the need for wages to pay them, to conscript labor. They also fought zealously for water rights, which in addition to labor is the main limiting factor for agricultural production there. The increased access to water by the hacendados meant less water for subsistence crops for the indigenous populations, further im- poverishing these communities (Enge and Whiteford 1989). The displacement and marginalization of indigenous populations in the region and across Mexico continued through the 19 th century, even after independence from Spain in 1823. The Lerdo Laws codified in Article 27 of the 1857 Constitution removed the inalienability of church, civil, and even peasant community lands to initiate investment and activity in what was errone- ously seen as economic and agrarian stagnation. The implied goal of this land reform was to inspire a yeoman class of private landowners to lead Mexico to prosperity. However, the primary result of this liberalization of the property regime was the expansion of latifundia or large private land holdings such as haciendas in many areas of Mexico (Assies 2008). At a national level, the peasantry as well as urban workers would take a back seat to large capital interests and development throughout the rest of the 19 th century. During the Porfiriato (1876-1910), a time of singular and relatively stable political control, globalizing forces expand- ed in Mexico. The country extended itself to foreign investment in order to industrialize, rein - forced by dictatorial control of the country by President Porfirio Díaz. Science and rationalism ruled the development ideology, leading to the establishment of standardized banking, scientific land survey, and modern infrastructure. Increased militarism to solidify the state and the devel - opment of a brutal police forces and militias to crush dissent ensured enough stability to entice foreign investment in Mexico (Krause 1998; Vanderwood 1998; Gonzáles 2002). These among other modernizations combined with lucrative government concessions, set the stage for invited American and British industrialists to invest in Mexico for valuable mineral resources like oil, 374 copper, and coal as well as cheap labor for export-oriented industry (Gonzáles 2002). Foreign capital also financed the development of Mexican infrastructure. Railroads connected Mexico’s peripheries, including a line running through the Tehuacán Valley, to Mexico City and then to the United States (Enge and Whiteford 1989; Raat 2004). Domestic producers, notably the Madero family of Coahuila, took a back seat to foreign enterprise during the Porfiriato, causing dissat- isfaction and disenfranchisement within the middle and upper tiers of Mexican society, giving way to grumblings of increased democracy. Rising food costs, poor working conditions, forbid - den labor organization, and extremely low wages opened the industrial and mining proletariat for change as well. Peasant land displacement increased during the during the Porfiriato (1876-1910), the 34-year long period when Porfirio Díaz ruled Mexico. In the last decade of the 19th and the first decade of the 20 th century, expanding global demand for sugar and a drop in global agricultural commodity prices to greater demand for land and water in the valley (Enge and Whiteford 1989; LaFrance 1989). In Calipan, the state governor near the turn of the century was granted owner- ship of the hacienda and used his political might to force laborers from the area to work under strict and brutal conditions. The Calipan hacienda was the largest but not the only one in present- day Coxcatlán and surrounding areas. One in the vicinity of Tilapa, another near Venta Salada called La Soledad, and another just to the north of Calipan were other important sugar haciendas whose ruins can still be found in the ejido lands across the valley. At the turn of the 20 th century after three centuries of Spanish colonialism, nearly a century of a neocolonial Mexican state, followed by the over two-decade rule of Porifirio Díaz, the situation for agricultural workers and most rural residents across the state of Puebla and in the study area was bleak. The average pay for a day laborer in the fields of a hacienda varied from 18-37 centavos per day when work was available, as mechanization of sugar haciendas cheapened and diminished the need for human labor. The advent of rail access to the Tehuacán Valley brought in cheaper grains and food staples from farther abroad. This had the dual effect of undermining local peasant agriculture while forcing the haciendas to expand on peasant commu- nity lands to keep up with ballooning economies of scale. Quality of life was very low. Over 80 percent of the state’s population lacked a primary education, the infant mortality rate was 492 for every 1,000 live births, and three-quarters of the state’s households lived in buildings classified as chozas or huts (La France 1989). 375 Figure A2.6. The S erdán family house. Today it is the M useo de la R evolución Casa de los Her manos S erdán in Puebla Cit y, Puebla. Notice the ex tant bullet mar ks from the Por firian attack . Photo by A. Hilbur n, Januar y 2011 Figure A2.7. Ruined R evolutionar y- era building in Coxcatlán town. Photo by A. Hilbur n, S eptember 2011 376 The social inequality, state militarism, and political disenfranchisement characteristic of Mexico’s colonial and early post-colonial history reached its apex at around the turn of the centu - ry under Porfirio Díaz. In Puebla, the Díaz-appointed state governor, Mucio Martínez maintained strict control over his state through a dedicated militia, close police contacts, and surveillance by telegraph and telephone (LaFrance 1989). The economic downturn that occurred in 1907- 1908 and the desperation felt by most Mexicans provided increasing impetus for social change (González 2002). As the first decade of the 20th century came to a close, the conditions were ripe for revolution in Puebla and across Mexico. Revolution and Institutions November 20, 1910 is typically commemorated as the day that marked the beginning of Francisco I. Madero’s revolution against President Porfirio Díaz. However, the struggle against Díaz had begun two days earlier on November 18t in Puebla de los Angeles, Puebla with a failed uprising led by the wealthy Serdán family (Figure A2.6). While the major battles and notable protracted armed struggles occurred elsewhere, mainly in the north-central part of Mexico, the state of Puebla and the study area participated in the overthrow of Díaz by Madero in 1910-11 and the drawn out civil conflict following Madero’s assassination in 1913 that lasted until 1920. While Díaz’s overthrow may have been well received enough for Madero to gain the presidency, there was still discontent, especially in and around the study area with what many saw as the im- position of an ineffective or similarly corrupt Maderista (Madero) government. Also, there were strong factions that still maintained loyalties to the old regime. Small armed conflicts punctu- ated the Tehuacán Valley during the contested Madero presidency from 1911-1912 and most of these concerned local political struggles regarding new jefes políticos or district-level (a former substate entity) political bosses. Just to the north of Coxcatlán, in Chilac, an anti-Maderista mob beat and killed a group of men selected to elect a new local political boss. Coxcatlán saw violent action when members of the town rose up against a district boss’s attempts to appoint friends and family to local authority positions instead of hosting elections for them (LaFrance 1989: 108- 109). This demand for local autonomy and reinforcement of community-level governance in the study area and across Mexico was not met by the actions of the abbreviated Madero presi- dency. The abandonment of support by the lower and middle classes of Madero combined with 377 the failures of its government to establish strong governance set the stage for opposition forces to ultimately depose the Revolutionary administration. Following the February 1913 overthrow and subsequent execution of Madero by Díaz loyalists led by Army General Victoriano Huerta, three major factions first allied themselves against Huerta’s federal forces in a violent struggle that would last over a year. After toppling Huerta by mid-1914, the agrarian base’s eponymously named Zapatista armies led by Emiliano Zapata in the south, the industrial proletariat forces led by Pancho Villa in the north, and the more gentrified mainstream Constitucionalista (Constitu - tionalist) forces led by Venustiano Carranza all vied for control of the country. The southern Te - huacán Valley was largely controlled by Constitucionalista forces for the time after Madero’s fall until the Revolution’s end around 1920. A regional strongman named Francisco J. Barbosa from Ajalpan was the predominant Constitutionalista in the area. Barbosa and his lieutenants used contacts with both the hacienda owners as well as the rural peasantry leadership to forge consen- sus and local power during this time (LaFrance 2003: 77-78). Likely the most famous Coxcateco in the town’s history rose to prominence then. Donato Bravo Izquierdo originally fought with Madero and eventually became a leader in the Constitucionalista faction led locally by Barbosa (LaFrance 2003:42). Following the Revolution, Bravo Izqueirdo became the state governor as well as other important political positions (Bravo Izquierdo 1964). By 1917, the Constitucionalistas had largely won supremacy of the country. Small pock - ets of resistance from Villa’s factions in the north and Zapatistas in southern Mexico plagued Carranza’s dominance but the newly-minted constitution was ratified and Carranza assumed the presidency. In the southern Tehuacán Valley, local caciques like Barbosa and Bravo Izqui - erdo ruled their own territories, checking any effective Revolutionary changes that were sought by the victorious Carranza faction and the Carranzista governor, Alfonso Cabrera. Eventually, factionalism and clashes with regional power structures caused the Carranza regime to fall and be replaced by former Constitucionalista Álvaro Obregón who, with his successor Plutarco Élias Calles, would plant the seeds for the 20 th Century Mexican state. After the fighting was over, the Mexican Revolution across Puebla and in the southern Tehuacán Valley was not the sweeping so - cial and political revolution that Madero promised. It had the effect of destroying the dictatorship of Porfirio Díaz. However, regional power based in local landholding elites and peasant repre- sentatives led by intermediaries loyal in name only to the Constitutionalista faction co-opted any wide sweeping social and political change. To most in Puebla, the years between 1910 and the 378 early 1920s were a time of sporadic factional violence and continued economic hardship punctu - ated by a global influenza pandemic (Figure A2.7). Neoliberal Era to the Present in Mexico and the Tehuacán Valley Following the Mexican Revolution, in the 1930s, the recovering Mexican state was slowly gaining in the process of socially reorganizing Mexico from the vestiges of colonialism while economically reorienting itself away from control by foreign industrialists. Mexico was becoming an inward-looking state that attempted to gain more self-determination by reducing foreign interests but also satisfying domestic politics. The petroleum and copper industries that were largely American and British-controlled were nationalized by the C árdenas administration in 1938. Also at this time, the land reform that was a major driving force of the Revolution was being enacted through the expropriation and redistribution of the large haciendas and Porfirian foreign-owned land grants to peasant communal groups through the ejido structure. In general this reform tried to return land tenure to a more native structure, albeit a somewhat idealized form. The redistribution of land to communally-held corporate entities composed of peasants also had the effect of garnering the support of the peasant class by the PRI party, which was to control Mexican politics until the 2000 elections. Following the ratification of Article 27 of the 1917 Constitution and the 1934 Código Agrario (Agrarian Code), land redistribution occurred across the Tehuacán Valley and ejidos subsequently came into being, with many reflecting extant community boundaries, both social and real. The scale and intensity of these distributions varied according to the presidential ad- ministration in power, with an apex during the Cárdenas administration (1934-1940) and a nadir during the subsequent Alemán sexenio (1940-1946). While the production of communally-held space through the redistribution of land into ejidos largely reset land tenure to the pre-existing indigenous system, the commodification of natural resources, mainly water, remained and was transferred into state ownership in the Tehuacán Valley and across Mexico (Enge and Whiteford 1989: 89-98). The Bracero Program of the 1940-1960s, essentially the widespread granting of temporary agricultural work visas to Mexican labor by the United States, provided an outlet for surplus Mexican labor (Calavita 1992). The limited capacities of the ejidos and their growing populations made the rural Tehuacán Valley a considerable sender region for braceros (Lee 2007; Fitting 2006). The result of this largely self-determined transmigration, as a whole, was the injec - 379 tion of cash into previously cash-poor villages but also the influence of el norte on local culture and aspirations. These following years of the “Mexican Miracle” (1940-1970) saw Mexico using its oil futures as collateral for loans used in developing more modern infrastructure such as high- ways, dams, electricity grids, and its petroleum industry further. Mexico was also developing a less foreign-dependent industrial base, employing an import substitution model (ISI). The state capital of Puebla, Puebla de los Angeles, became the Mexican “Detroit” by producing state- subsidized Volkswagens and an ISI heavy truck, the DINA. Mexican ISI, however, was not a true import substitution and still depended on foreign inputs such a tools and electronics. Aside from ISI, Mexico was also courting some export-oriented industry through the Border Industrialization Plan and the subsequent maquila program during the mid-1960s to create jobs for surplus labor when the Bracero program came to an end. These programs were highly regulated and followed strict governmental regulation in terms of quantity of tax-free imports and wages, where the value added to the goods was taxed, some domestic inputs were required, and all manufactures had to be exported (Delgado-Wise and Cypher 2007). The end of the “Mexican Miracle” was underway when the focus of the world came unto Mexico during the summer of 1968, when it hosted the Summer Olympic Games. Just weeks before the Olympics, growing discontent over the excessive costs of hosting the games amid economic troubles boiled over into widespread protest by many groups, primarily university stu- dents. The culminating event of this protest was a brutal massacre of student protestors allegedly perpetrated and immediately covered up by the federal government at Tlatelolco Plaza in Mexico City. The cracks appearing in the “Mexican Miracle” were left wide open for the entire world to see. Brutal crackdowns on persistent protest occurred in other cities that summer and years later as well, most notably in Monterrey (Raat 2004: 149). The following decades would foretell another era of globalization, one begun in economic crisis and resulting in the neoliberal reality that characterizes Mexico today. The global petro crises of the early 1970s caused a global economic crisis leading to the subsequent dismissal of the Bretton Woods Agreement which spread unprecedented monetary instability across the globe. The wildly fluctuating dollar to which the Mexican peso was infor- mally pegged led to a peso crash in 1972. Yet, this crash was relatively small and one from which Mexico recovered easily, even bringing about a resurgence of populism (Teichman 1996; Otero 380 1996). Across the globe and even in Mexican circles, though, the new ideology of neoliberalism was beginning to find political economic space following the perceived failures of Keynesian government involvement in economic policy. In fact, some claim that the discourse that envelops the concept of “globalization” stems from agendas that really promoted neoliberalism (Swynge - douw 2004; Leitner et al 2007). Owing to Mexico’s position of being an oil exporter, the country maintained its protectionist stance and remained economically and geopolitically self-determi- nant during the 1970s (Raat 2004: 162). Populism and the façade of progressivism also led the political discourse of the Mexican presidencies during this time as well. However, it was the global drop in petroleum prices in the early 1980s that quickly forced them to change their tune. It was 1982, when oil-backed Mexican protectionism could hold no more against the global forces of neoliberalism. After years of taking on credit based on unsustainable oil futures, Mexico found itself unable to make balances of payments to its creditors when the price of oil dropped substantially in 1982, leaving Mexico with little recourse. Mexican economic policy makers adopted a neoliberal strategy to satisfy their debts in the eyes of their global North credi - tors for the next twelve years. This strategy entailed some dismantling of its protectionist tariff structure and a move toward the privatization of its state-run industries and government-involved businesses. Domestic fiscal austerity along the lines of the “Washington Consensus” was de- manded by the IMF and US creditors. Although Mexican policy makers resisted these demands, the reality of extreme indebtedness with declining ability to pay forced the government to retract considerably from social service spending and infrastructural development all throughout the 1980s (Teichman 1996: 153-156). Mexico joined the General Agreement on Tariffs and Trade (GATT) in 1986 (Raat 2004: 163). Neoliberal ideologies were working their way into domestic political discourse in Mexico in the late 1980s and 1990s. Even the PRI, which, hinting at the “institutional” in its name had always appeared protectionist, selected a free-market fundamentalist in Carlos Salinas de Gor - tari for its presidential candidate in 1988. President Salinas and many of his top advisors were Ivy League-educated neoclassical economists who found political space among the already ripe economic conditions to further their own brand of neoliberalism (Teichman 1997). This political economic environment cultivated a growth in industrialization within the maquila sector in the late 1980s and early 1990s (Kopinak 1996). Part of the contingencies 381 embedded in GATT and IMF structural adjustment entailed the loosening of restrictions both in terms of import/export tariffs but also the diminishment of labor rights. Industrial jobs, previ - ously under the domain of men, were becoming feminized (Salzinger 2003). Parallel to feminiz - ing the industrial workforce was the cheapening of it, resulting in plummeting wages. Organized labor had long enjoyed immense political power within the dominant PRI party, but with the move away from ISI, so fled their protection from wage decreases and ultimately their influence diminished (Teichman 1997). Mexico’s industrial sector was being turned inside out; going from a holistic producer for domestic and regional markets with organized labor to becoming an as - sembler for global export markets with low wage labor. This process of industrial reorientation towards a more neoliberal ideal has been referred to as maquilization (Kopinak 1996: 13). It was not until the late-1990s and early 2000s that these effects were to be fully realized. Neoliberal change was not just limited to the industrial sector. In 1992, the long commit - ment of the PRI to agrarian reform ended with the revision to Article 27. This revision ended the government’s obligation towards land redistribution and the establishment of new ejidos . It also set up the infrastructure to internally survey, title, and ultimately privatize previously inalienable communal lands and goods (Gates 1996). This program (PROCEDE), along with a “roll back” (Peck and Tickell 2002) in agricultural subsidies (PROCAMPO), had the ostensible intent to make Mexican agriculture more efficient in terms of the global market, yet empirical data tells otherwise (Bartra 2004; Gates 1996; Appendi 1996; Stephen 1998; Perramond 2008). PROCEDE has just run its course, ending in 2006, leaving 97 percent of Mexico’s ejidos and comunidades agrarias surveyed. This gradual abandonment of the agrarian sector has caused Mexico, the land where corn was first cultivated, to move from a largely self-sufficient producer of corn to a sub- stantial net importerv (Nadal 2001; Bartra 2004; Fitting 2006). The final walls to neoliberal political economy fell in 1994, when the Mexican economy declined to unprecedented levels. A peso crash caused by the downturn of the US dollar to which it was pegged, diminished the prospects of Mexico’s ability to pay back its growing national debt, leading to massive capital flight. Amidst this crisis, Mexico ratified the terms of the North American Free Trade Agreement (NAFTA). The abolishment of trade barriers that were required by NAFTA fully realized and integrated Mexican industry, agriculture, and some services within global markets. The aforementioned changes in Mexican trade policy set the stage for Mexican labor and 382 industry to become what Delgado and Cypher (2007) call “subordinate[ly] integrat[ed]” to US industry. Transnational corporations in the de-industrializing United States vi thus optimized their position to import and export their inputs and manufactures freely across the international bound- ary while exploiting the cheap, captive, and unfree labor stuck on the other side. This cheap labor was reproduced by the lack of social sector support resulting in a growing informal sec- tor, both of which are caused by the fiscal austerity demanded by structural adjustment policies. Adding to this cheapening is the difficulty involved in transborder movement by Mexican labor. Cheap Mexican labor is not just embedded in exported manufactured goods, it is also “exported” cheaply through illegal migrants entering the United States. Their illegal status never truly allows migrants to escape the border, giving them little space for demanding adequate pay and worker rights (Delgado-Wise and Cypher 2007). Besides cheapening labor, the border provides Mexico with a comparative manufacturing advantage because of its limited capacity to enforce its environmental law. Although Mexico’s federal environmental code is far more progressive than that of the US, poor funding of enforce - ment, again stemming from government “roll backs”, allows transnational firms to conduct manufacturing operations that would either be too costly or illegal in the home country (Kopinak 1996). The result is an increasingly degraded environmental picture along the border as well as in Mexico’s rural peripheries where urban-to-rural industrialization iin Mexico is follwing a US model of industrial change (Van Dooren and Zarate-Hoyos 2008). Neoliberal logic contended that Mexico’s predominant minifundista (smallholder) system of agriculture would not be worth supporting due to its relative inefficiency in global food/com- modity markets (Gates 1996; Appendi 1998; Baños Ramírez 1998; many others). Ignoring the environmental value of smallholder agriculture (Toledo 1996; Wright 2002), neoliberal agricul - tural policy based on neoclassical economics sought to metabolize the Mexican countryside by ceasing support for the agrarian sector and putting ejidatarios with 2-10 hectare plots in competi - tion with 1000+ acre corporate farmers in places like Kansas or Iowa (Bartra 2004). The surplus agricultural labor would be used for industrial export truck farming (Bacon 2004), maquiladoras, or likely fill the swelling informal economic sector. Migrating to the United States became a likely and necessary option, also (Delgado-Wise and Cypher 2004). NAFTA, the watershed event of a fully neoliberal rural reality, combined with the earlier agrarian counter-reforms, was, not surprisingly, not welcomed by the Mexican peasantry. In fact, the catalyst for the 1994 Zapatista 383 uprising in Chiapas was NAFTA’s signing (Harvey 1994; EZLN 2002). The agrarian sector, especially its indigenous counterparts, justifiably felt that NAFTA’s triumph and its demands for the gradual “roll back” of government subsidies combined with a less-than-adequate rural poverty alleviation (PROCAMPO) program were failures of its own government and society to defend itself and its cultural patrimony from a global north invasion (Marcos 2002). Most of these neoliberal effects are manifest uniquely in the Tehuacán Valley. Additionally, the lure of a life outside of the social limitations, relative poverty, and general economic uncertainty on the ejido and in the village prove strong and many forsake their lands in search of more opportunity in the city or al norte, usually New York and New Jersey (Lee 2007). Southeastern Puebla and the city of Tehuac án experienced immense industrial develop- ment and urban growth in the late 1980s and 1990s as a result of the maquilization of Mexican industry. Industry jobs in the region totaled nearly 50,000 in 2000 (Maquila Solidarity Network 2003). Textile assembly plants, mostly ones assembling blue jeans, sprung up across the val - ley and have remained in spite of declines in the maquila sector in other parts of Mexico (South 2006). Additionally, the southern Tehuacán Valley became one of the largest poultry and egg producing areas in Mexico. Both industries have harnessed the cheap surplus labor in the valley and in other parts of Puebla as well as the States of Veracruz and Oaxaca. While the growth of the maquiladora industry in the Tehuacán Valley has brought many jobs to an economically de - pressed region, nearly as many complaints about the quality of these jobs have also arisen. Sex - ual harassment, illegal wage garnishments, underage employment, and low safety standards re- main persistent problems many of the approximately 400 registered maquilas, which are perhaps worse at the equally numerous, unregistered ones (Maquila Solidarity 2003). Attempts have been made at establishing more just working conditions and better wages at the larger maquiladoras by semi-formal unions and NGOs, but largely these have been largely successfully repressed by factory management through layoffs, firings, and allegedly, through violent threats and acts by hired thugs. Nevertheless, worker resistance and the publicizing of their plight has forced some fashion labels, like Guess?, to take a larger role in their assembly contracts (Cox 1998). Neoliberalism’s effects tend to magnify in or converge on cities (Brenner and Theodore 2002b; Leitner et al. 2007). The southern Tehuacán Valley is no exception. The growth of the maquila industry combined with the declining sustainanbility of rural economic opportuni- ties in tthe valley and sierra have paralleled a massive increase of the urban population in and 384 around Tehuacán, seeing an annual growth rate of about five percent. In 2000, its population had grown to nearly 230,000 (INEGI 2000). In 2010, the population hovered around 275,000. This extreme population growth is manifested in the explosion of colonias surrounding the city. This unplanned growth has proven to be problematic in terms of environmental sustainability, service provision, and urban planning. Even smaller towns nearby to the south such as Ajalpan, Alte- pexi, and San Sebastian Zinacatepec also struggle with these urban growth issues. Maquila industrialization in the Tehuacán Valley, as in other parts of Mexico, has brought net unemployment down since its arrival, yet it does not easily translate to increases in house- hold economic prosperity. For one, the aforementioned low wages, long hours, and few em - ployee rights provide little incentive to remain employed in the factories. People I knew who had worked in the maquilas recounted to me that they were glad they no longer did the difficult, mind-rotting work that paid “next to nothing”. Furthermore, the feminization of maquila work limits its appeal to economically active males who typically provide the greater portion of Mexi- can household incomes (Salzinger 2003; Fitting 2006: 22). In fact, most of the maquila workers I knew or observed were women or non-head-of-household men. Combined with decreasing profitability and opportunity in the agrarian sectors, migration to the United States has become an increasingly necessary option for many in the Tehuacán Val - ley, Coxcatlán, (Lee 2007; Fitting 2006) as well as Mexico overall (Rouse 1991; Martínez 2001; Wright 2002; Delgado-Wise and Cypher 2007). Migration to the United States was a familiar but varied experience for many of my friends and acquaintances in Coxcatlán. Some were success - ful and enriched their lives economically as well as experientially, others had about broken even, and for some it was a bust. I met people who lived fully bi-national lives, had married Ameri - cans, and had dual citizen children. There were people who had started successful businesses. Some just worked for a while doing jobs they never envisioned doing and sent money home before returning themselves or being deported. The recent economic downturn in 2008 forced one person I knew to move back from a decade long stay in the U.S. when he lost his job due to a layoff. For some, it cost them their lives. Toward the end of my stay, word came back to town that two Coxcateco brothers had died from exposure crossing into the United States. Migration to the U.S. is a part of everyday life in the study area. For many men, especially those in the sierra communities, it is a rite of passage and an escape from the town or village. For them and others, it is also a necessity that drives them far from home into an uncertain future. 385 386 The municipio and surrounding area are largely known to the outside world because a team of ar- chaeologists established an over ten thousand year sequence of human occupancy, including the origins of Mesoamerican agriculture. MacNeish and his team constructed a record of lifeways in and around the Tehuacán Valley from when only a small band of about 10-15 people lived there at the beginning of the Holocene to the regional network of señorios (fiefdoms) each with hundreds or a few thousand residents at the time of contact with the Spanish. The establishment of the colony of New Spain by the Spanish conquerors largely rewrote the political-economic landscape of the region under encomienda and reducciones. The efforts of the Catholic Church to Christianize the indigenous populations gave Coxcatlán special regional status as a corregimien- to, where Coxcatlán served as a local ecclesiastical center. This concession meant that Coxcatlán was subject to the viceregal government in Mexico City and not to a typically exploitive Span - ish colonist, granting a Coxcatlán a measure of autonomy due to its distance from the colonial capital. However, the abuses of the encomienda system led to its own demise decades later and the growth of Tehuacán and changes in the spatial hierarchies of the New Spain Catholic Church led to the dissolution of Coxcatlán’s special status. The arrival of haciendas shortly thereafter brought sugarcane cultivation and what would be centuries of struggle between the landed elite and local populations over land, labor, and water. The replacement of the Spanish colonial government with the new Mexican state changed life little for the majority of area residents, even through to the Revolution that began in 1910. The Mexican Revolution brought low-intensity violence, economic decline, and little social change initially. Eventually, the goals of greater social equality and redistribution of resources that were central to the Revolution were realized in the 1930s with the redistribution of land to social properties and the growth of the Mexican economy through petroleum-backed develop- ment and domestic industrialization. The middle portion of the 20 th century was a time of political stability for Mexico and the study area as the PRI created a centralized patron-client state. Periodic economic crises in the late 1970s, early 1980s, and 1990s led to the decline of this political system and opened the doors to globalizing neoliberal forces. This change, precipitated by demands for economic structural adjustment by Mexico’s debtors, largely involved the opening of previously protected domestic industry and markets to global ones. It also meant a retraction by the government from the social sector and the selling off of nationalized infrastructure such as communications and railways to 387 private capital. Around the study area, export-oriented, low wage manufacturing in textiles and foodstuffs grew as a major employment sector in the 1990s and 2000s and persists to this day. Also during this time, Mexico also ended its long-tenured agrarian reform and enacted a program to title and capitalize previously inalienable social properties affecting over one million square kilometers of land, comprising just over 50 percent percent of Mexico’s land surface and millions of social property members. In Coxcatlán, all but one ejido has participated in this program. By the end of the 20 th century, the PRI’s nearly 80-year dominance of Mexican politics ended with the Vicente Fox Presidency and the rise of his party, the center-right PAN, as well as the substan - tive contention of other political parties, notably the markedly leftist PRD that has seen popular - ity in major urban areas. Endnotes 1. The pochocuil takes its name from the Nahuatl “pochotl ” mean ng “pochote” and “cuilli” meaning “ wor m”. 2. S ee Hunt 1972 or Enge and Whiteford 1989 for a wonder ful summaries of these societies. 3. “S ansuanchi” is a Nahuatl cor ruption of S an Juan or S aint John the Baptist who is the patron saint of the colo - nial and present settlement of Coxcatlán town.