Social Change and the Evolution of Ceramic Production and Distribution in a Maya Community (Mesoamerican Worlds: from the Olmecs to the Danzantes)

Social Change and the Evolution of Ceramic Production and Distribution in a Maya Community

Mesoamerican Worlds: From the Olmecs to the Danzantes General Editors: Davíd Carrasco and Eduardo Matos Moctezuma Editorial Board: Michio Araki, Alfredo López Austin, Anthony Aveni, Elizabeth Boone, Charles H. Long, and Eleanor Wake After Monte Albán, Jeffrey P. Blomster, editor The Apotheosis of Janaab’ Pakal, Gerardo Aldana Commoner Ritual and Ideology in Ancient Mesoamerica, Nancy Gonlin and Jon C. Lohse, editors Conquered Conquistadors, Florine Asselbergs Empires of Time: Calendars, Clocks, and Cultures, Revised Edition, Anthony Aveni Encounter with the Plumed Serpent, Maarten Jansen and Gabina Aurora Pérez Jiménez In the Realm of Nachan Kan, Marilyn A. Masson Invasion and Transformation, Rebecca P. Brienen and Margaret A. Jackson, editors The Kowoj, Prudence M. Rice and Don S. Rice, editors Life and Death in the Templo Mayor, Eduardo Matos Moctezuma Maya Daykeeping, John M. Weeks, Frauke Sachse, and Christian M. Prager The Madrid Codex, Gabrielle Vail and Anthony Aveni, editors Mesoamerican Ritual Economy, E. Christian Wells and Karla L. Davis-Salazar, editors Mesoamerica’s Classic Heritage, Davíd Carrasco, Lindsay Jones, and Scott Sessions, editors Mockeries and Metamorphoses of an Aztec God, Guilhem Olivier, translated by Michel Besson Rabinal Achi, Alain Breton, editor; translated by Teresa Lavender Fagan and Robert Schneider Representing Aztec Ritual, Eloise Quiñones Keber, editor Ruins of the Past, Travis W. Stanton and Aline Magnoni, editors Skywatching in the Ancient World, Clive Ruggles and Gary Urton, editors Social Change and the Evolution of Ceramic Production and Distribution in a Maya Community, Dean E. Arnold The Social Experience of Childhood in Mesoamerica, Traci Ardren and Scott R. Hutson, editors Stone Houses and Earth Lords, Keith M. Prufer and James E. Brady, editors The Sun-God and the Savior, Guy Stresser-Péan Tamoanchan, Tlalocan, Alfredo López Austin Thunder Doesn’t Live Here Anymore, Anath Ariel de Vidas; translated by Teresa Lavender Fagan Topiltzin Quetzalcoatl, H. B. Nicholson The World Below, Jacques Galinier

Social Change and the Evolution of Ceramic Production and Distribution in a Maya Community Dean E. Arnold

University Press of Colorado

© 2008 by the University Press of Colorado Published by the University Press of Colorado 5589 Arapahoe Avenue, Suite 206C Boulder, Colorado 80303 All rights reserved Printed in the United States of America The University Press of Colorado is a proud member of the Association of American University Presses. The University Press of Colorado is a cooperative publishing enterprise supported, in part, by Adams State College, Colorado State University, Fort Lewis College, Mesa State College, Metropolitan State College of Denver, University of Colorado, University of Northern Colorado, and Western State College of Colorado. The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials. ANSI Z39.48-1992 Library of Congress Cataloging-in-Publication Data Arnold, Dean E., 1942– Social change and the evolution of ceramic production and distribution in a Maya community / Dean E. Arnold. p. cm. — (Mesoamerican worlds) Includes bibliographical references and index. ISBN 978-0-87081-923-0 (alk. paper) 1. Maya pottery—Mexico—Ticul. 2. Maya pottery—Analysis. 3. Mayas—Mexico—Ticul— Social conditions. 4. Pottery industry—Mexico—Ticul. 5. Social change—Mexico—Ticul. 6. Ticul (Mexico)—Social conditions. I. Title. F1435.3.P8A76 2008 972'.65—dc22 2008033380 Design by Daniel Pratt 17 16 15 14 13 12 11 10 09 08

10 9 8 7 6 5 4 3 2 1

For June, for forty years of love, friendship, and companionship

Contents

List of Figures /

xi

List of Tables /

xvii

Foreword by Eleanor Wake / Preface /

xxi

xxiii

Acknowledgments /

xxvii

Chapter 1: Introduction / 1

Paradigms of Pottery and Social Change / The Limits of Ethnographic Analogy / 17 Collecting Data in the Field / 21 Data Reduction and Analysis / 24 The Plan of the Book / 27

vii

2

Contents

Chapter 2: How Have the Population and Organization of Potters Changed? / 31

The Social Context / 33 Changing Production Organization / 37 Forces of Social Continuity / 40 Forces of Social Change / 65 Social Change and Increased Production-unit Size / Conclusion / 88

78

Chapter 3: How Have Demand and Consumption Changed? / 93

Demand and Cultural Evolution / 94 Demand from Traditional Uses of Pottery / 95 New Demand and New Uses of Pottery / 113 Cycles of Demand and Their Changes / 119 Quantitative Measures of Change in Demand / Conclusion / 121

120

Chapter 4: How Has Distribution of the Pottery Changed? / 127

Changes in Transportation Infrastructure / Changes in the Types of Distribution / 133 Vertical Integration / 148 Conclusion / 150

128

Chapter 5: How Has Clay Procurement Changed? / 153

Under What Conditions Does Clay Procurement Change? / 154 How Do Changes in Clay Procurement Affect Procurement Organization? / 170 A Surrogate Measure of Production Intensity / 180 Procurement Intensity, Organization, and Production-unit Size / 183 Do Changing Clay Sources Reflect Evolutionary Social Change? / 184 Chapter 6: How Has Temper Procurement Changed? / 191

Changes in Temper for Cooking Pottery / 192 Changes in Temper for Non-cooking Pottery / A Surrogate Measure of Production Intensity / Control and Access to Temper Sources / 218 Conclusion / 218

viii

193 215

Contents

Chapter 7: How Has Composition of the Pottery Fabric Changed? / 221

Behavioral Changes in Paste Preparation / 222 Changes in Paste Composition over Time / 226 Conclusion / 227 Chapter 8: How Has the Forming Technology Changed? / 229

Why Were New Fabrication Techniques Adopted? / Changes in Forming Technology / 237 Choosing a Technique / 265 Changing Explanations of Dimensional Variability / Conclusion / 272

232

265

Chapter 9: How Has Firing Technology Changed? / 281

Changes in the Procurement and Use of Fuel / 282 Changes in Kiln-making Technology / 284 Task Segmentation and Specialization in Firing / 286 Changes in Kiln Sizes and Their Distribution among Potters / Conclusion / 304

302

Chapter 10: Conclusion / 309

Summary of Changes / 311 The Conservative Nature of Household Production / Efficiency / 318 Paradigms: Social Change and Specialization / 319 References Cited / Index /

314

327

345

ix

Figures

Figure 2.1. Trend line showing the exponential population growth in the muni-

cipio of Ticul, 1950–1990 / 32 Figure 2.2. Map of Yucatán showing major cities, towns, archaeological sites, and pottery-making communities between the late 1960s and 1994 / 34 Figure 2.3. Trend line for the total number of potters in each observation period from 1965 to 1997 / 35 Figure 2.4. Trend line showing the changes in the number of production units from 1965 to 1997 / 40 Figure 2.5. Trend lines showing the changes in the mean and median number of potters per production unit in Ticul from 1965 to 1997 / 41 Figure 2.6. Potters per production unit in 1965–1966, 1984, and 1997 / 41 Figure 2.7. Trend lines showing the number of the most common kin working as potters in production units from 1965 to 1997 / 44 Figure 2.8. Mixing paste is one of the first tasks children learn in making pottery / 46 xi

Figures

Figure 2.9. Trend lines of the changing percentages of the most common kin

types working as potters in production units from 1965 to 1997 / 50 Figure 2.10. Trend lines for the number of different types of relationships of potters to production-unit owners grouped by non-kin and lineal, collateral, and affinal relatives / 54 Figure 2.11. Trend lines for the percentages of potters related to productionunit owners grouped by lineal, collateral, and affinal relatives / 55 Figure 2.12. Trend line of the number of wage laborers who are not relatives of production-unit owners from 1965 to 1997 / 56 Figure 2.13. Bar graph of the frequencies of the distances between potters active in 1984 and their fathers, teachers, and nearest production unit / 58 Figure 2.14. Bar graph summarizing the changes in the locations in production units in Ticul since 1965 / 60 Figure 2.15. Bar graph summarizing the changes of production units from 1970 to 1997 compared to their location in the previous survey / 61 Figure 2.16a. Map showing the locations of potters’ production units in 1965–1966 / 62 Figure 2.16b. Map showing the locations of potters’ production units in 1968 / 62 Figure 2.16c. Map showing the locations of potters’ production units in 1970 / 63 Figure 2.16d. Map showing the locations of potters’ production units in 1984 / 63 Figure 2.16e. Map showing the locations of potters’ production units in 1988 / 64 Figure 2.16f. Map showing the locations of potters’ production units in 1994 / 64 Figure 2.16g. Map showing the locations of potters’ production units in 1997 / 65 Figure 2.17. The amount of kin-relatedness among production units in 1965– 1966 and 1997 based on traceable relationships / 66 Figure 2.18. Trend line showing the percentage of female potters from 1965 to 1997 / 74 Figure 2.19a. Trend lines showing the relationship between the number of units and the number of potters per unit in 1965–1966 / 84 Figure 2.19b. Trend line showing the relationship between the number of units and the number of potters per unit in 1968 / 84 Figure 2.19c. Trend line showing the relationship between the number of units and the number of potters per unit in 1970 / 85 xii

Figures

Trend line showing the relationship between the number of units and the number of potters per unit in 1984 / 85 Figure 2.19e. Trend line showing the relationship between the number of units and the number of potters per unit in 1988 / 86 Figure 2.19f. Trend lines showing the relationship between the number of units and the number of potters per unit in 1994 / 86 Figure 2.19g. Trend line showing the relationship between the number of units and the number of potters per unit in 1997 / 87 Figure 3.1. View of the hill ridge in 1968 looking east from the road between Ticul and Sacalum / 96 Figure 3.2. View from the top of the puuc ridge looking south toward the haystack-like hills of Yucatecan hill country in 1997 / 97 Figure 3.3. Bar graph showing the mean monthly temperatures in the town of Oxkutzcab / 98 Figure 3.4. Bar graph showing monthly precipitation in the town of Oxkutzcab / 99 Figure 3.5. Water-carrying pots (cántaros) that have just been removed from a kiln in 1965 / 100 Figure 3.6a and 3.6b.Two positions used for carrying a water pot / 101 Figure 3.7. The range of the materials used for containers in a Ticul kitchen in 1984 / 107 Figure 3.8. A household altar for the Day of the Dead rituals / 108 Figure 3.9. A pottery image of the Yucatecan mestiza used as a decoration at the Hacienda Uxmal / 114 Figure 3.10. Vessels with copies of ancient Maya designs in a Ticul pottery store in 1997 / 116 Figure 4.1. Map of the northwestern part of the Yucatán peninsula showing the principal railroad lines / 129 Figure 4.2. Map of the northwestern part of the Yucatán peninsula showing the principal roads between about 1970 and 1990 / 130 Figure 4.3. A potter in the Ticul market in 1968 selling jars for carrying and storing water / 134 Figure 4.4. Woman selling pottery in the Ticul market in 1984 / 135 Figure 4.5. Women in the Ticul market on October 31, 1984, selling pottery for the Day of the Dead rituals / 136 Figure 4.6. Painting workshop located in Ticul’s largest production unit in 1997 / 140 Figure 4.7. Loading a large truck with plant pots and figurines bound for an unknown destination in 1984 / 142 Figure 2.19d.

xiii

Figures

Figure 4.8. Loading a truck with cubano-shaped plant pots in 1984

/ 144 / 149 Figure 5.1. Sketch map of Hacienda Yo’ K’at showing clay sources surveyed in 1968 / 156 Figure 5.2. Sketch map of the area surrounding the Mejorada clay source in 1966 / 166 Figure 5.3. A generalized diagram of the profile of the marl quarries in Campeche that were used as clay sources in the 1990s / 168 Figure 5.4. The Dzitbalché clay source / 169 Figure 5.5. Plots showing the trends in the increasing specialization of clay mining between 1965 and 1988 / 171 Figure 5.6. Profile and plan view of the clay mine at Yo’ K’at used in November 1968 / 174 Figure 5.7. The surface portal to the Yo’ K’at clay mine in 1968 / 175 Figure 5.8. The narrow entrance to the Yo’ K’at clay mine at the base of the entrance shaft / 175 Figure 5.9. An underground excavated cavity in the Yo’ K’at clay mine in 1968 / 176 Figure 5.10. Generalized profile of the vertical shaft mines at Yo’ K’at in 1984 / 177 Figure 6.1. Delivering temper to a potter’s house in 1984 / 201 Figure 6.2. Plots showing trends in the increasing specialization of temper mining between 1965 and 1997 / 202 Figure 6.3. Mine opening at Yo’ Sah Kab in 1966 showing the depth of the palygorskite layer / 203 Figure 6.4. Temper preparation in 1965 showing crushing, bringing bags of nooy from the mines, and screening / 205 Figure 6.5. The interior of a mine at the Chapab source in 1997 showing a woman mining nooy for preparing temper / 206 Figure 6.6. Temper preparation area at the Chapab source in 1997 / 207 Figure 6.7. The shelter used to prepare temper at the Chapab mines in 1997 / 208 Figure 6.8. Bar graph showing the variation in the amount of clay minerals in temper samples collected in 1965–1966 / 213 Figure 6.9. A pile of palygorskite outside a potter’s house in 1988 / 214 Figure 7.1. Principal components plot of the INAA data of kiln wasters from Ticul, Tepakán, and Akil / 226 Figure 8.1. A Ticul potter using the traditional turntable (k’abal ) in 1970 / 233 Figure 4.9. Store along the highway south of Cancún

xiv

Figures

Figure 8.2. The base of a traditional turntable and the rotating platforms stored

on a pile of clay in a potter’s house in 1970 / 235 Figure 8.3. Potter making large tinajas in Ticul’s largest production unit in 1984 / 236 Figure 8.4. Daughter of a Ticul potter using the wheel in 1984 / 241 Figure 8.5. Molding a vessel using a traditional working position / 249 Figure 8.6. Obliterating the mold marks on a small cántaro in 1984 / 252 Figure 8.7. Shelves used to store molds in 1997 / 253 Figure 8.8. Trend line showing the relationship of the frequency and age of molds in one production unit in 1997 / 254 Figure 8.9. Bar graph showing the frequency of the maximum dimension of molds in one production unit in 1997 / 255 Figure 8.10. A ball-bearing turntable (tornete) used in 1984 / 257 Figure 8.11. A potter using a type of ball-bearing turntable anchored in a gallon can of cement in 1984 / 258 Figure 8.12. A potter using a type of ball-bearing turntable made with a machine pulley in 1984 / 259 Figure 8.13. Frequency of traditional and ball-bearing turntables among the 1984 production units / 260 Figure 8.14. Slip-casting molds filled with clay in the Ticul ceramics factory in 1997 / 263 Figure 8.15. The carousel in the Ticul ceramics factory in 1997 used for slip casting and drying molds / 264 Figure 8.16. Potter making a traditional water-carrying jar (cántaro or p’uul) in 1984 / 267 Figure 9.1. A traditional beehive kiln in 2002 / 285 Figure 9.2. Water-storage jars being removed from a kiln in 1965 / 288 Figure 9.3. Kiln that replaced the kiln shown in Figure 9.2 in 1984 / 289 Figure 9.4. Kiln shown in Figure 9.3 in 1997 / 290 Figure 9.5. The updraft kiln used in one production unit in 1997 / 292 Figure 9.6. An updraft kiln used by one potter in 2002 / 294 Figure 9.7. A square cement-block kiln used in 1997 / 295 Figure 9.8. A pot kiln used in one production unit in 1997 / 296 Figure 9.9. A gas kiln in the government-sponsored production unit in 1997 / 297 Figure 9.10. A hybrid kiln in 1997 that utilized both traditional and modern construction materials / 298 Figure 9.11. The profile of the hybrid kiln’s wall in 1997 / 299

xv

Tables

Table 2.1. Population of potters in Ticul, 1965–1997

/

39

Table 2.2. Learning styles of active male and female potters in 1984

/ 43 / 49 Table 2.4. Inventory of the number and types of kin who are potters, 1965–1997 / 53 Table 2.5. Number and percentage of non-relative wage laborers working as potters, 1965–1997 / 57 Table 2.6. Gender of production-unit owners, 1965–1997 / 74 Table 3.1. The different vessel shapes that one potter sold in the north and south of the State of Yucatán between 1965 and 1997 / 105 Table 3.2. Kiln inventories from two potters in 1965 and one in 1984 / 121 Table 3.3. The number of production units producing particular shape classes and vessel shapes in 1965–1966, 1970, 1984, and 1997 / 122 Table 2.3. Division of labor in the largest production unit in 1997

xvii

Tables

Table 3.4. Vessel classes and shapes produced by one production unit between

July and late November 1984 / 123 Table 4.1. Locations where pottery was sold between 1965 and 1997 / 132 Table 4.2. Types of marketing strategies used between 1965 and 1997 / 133 Table 4.3. Number of pottery stores, showing the continuity and changes in ownership and location from 1984 to 1997 / 139 Table 5.1. Clay sellers and clay sources used by production units in 1997 / 167 Table 5.2. The changing number of clay miners from 1965 to 1997 and their relationships to potters / 173 Table 5.3. Summary of changes in clay sources, their areal extent, and procurement organization between 1965 and 1997 / 179 Table 6.1. Comparison of temper miners at Chapab and Yo’ Sah Kab and their clients in 1997 / 198 Table 6.2. The changing number of temper miners from 1965 to 1997 and their relationships to potters / 199 Table 6.3. Number of temper miners and their clients in 1966 / 200 Table 6.4. Behavioral variation in sah kab temper preparation in 1966 / 209 Table 6.5. Changes in temper sources, their areal extent, and procurement organization between 1965 and 1997 / 216 Table 7.1. Summary of changing clay and temper sources used from 1965 to 1997 / 223 Table 8.1. Muscle groups required for using the traditional turntable and the wheel / 240 Table 8.2. Fabrication times for food bowls (cajetes) using the wheel, mold, and turntable / 244 Table 8.3. Summary of fabrication times for wheel-made plates (platos) from different production events / 245 Table 8.4. Principal steps in the behavioral chain for making mold-made vessels / 246 Table 8.5. The number of fabrication steps for mold-made vessels requiring more than one two-piece mold / 251 Table 8.6. The number of production units with traditional and ball-bearing turntables in 1984, and the mean number per unit / 260 Table 8.7. Comparison of the heights of the traditional and ball-bearing turntables in 1984 / 262 Table 8.8. The measurement units that potters use for making traditional vessels / 268 Table 8.9. Nomenclature for the size categories of traditional vessel shapes / 268 xviii

Tables

Table 8.10. Measurements used in the stages of the water-carrying jar

/

269

Table 8.11. Uses and measurements of the different size categories of

the apaste /

270

Table 8.12. Alternative measurements for making an apaste.

/

270

Table 8.13. Measurements used for each stage of the water-storage jar

/ 271 Table 8.14. The relationships of labor, skill, and capital required for different fabrication technologies / 273 Table 9.1. Changes in types of kilns from 1965–1966 to 1997 / 291 Table 9.2. The changing sizes of kilns in 1965–1966, 1984, and 1997 as measured by the number of firewood bundles used / 303 Table 9.3. Mean kilns per production unit and production units with two kilns in 1965, 1984, and 1997 / 303 Table 9.4. Summary of the changes in firing between 1965 and 1997 / 305

xix

Foreword

T

his latest publication from the Mesoamerican Worlds series focuses on the Maya community of Ticul, which lies some sixty miles south of Mérida in the heart of the Puuc region of the Yucatán peninsula, Mexico. Ticul has been home to generations of potters, some of whom undoubtedly lived through, and perhaps worked at the behest of, the great Maya civilizations. (Archaeological evidence suggests that pottery was produced there as early as 600 B.C.) Nevertheless, like so many of its neighbouring towns and villages, Ticul entered the second half of the twentieth century as a rather provincial community, plying its still preindustrial pottery trade at a local and principally household level. By the end of the century it had shot to fame as part of one of modern-day Mexico’s most successful commercial ventures: the Mexican Riviera, a popular tourist destination. Although still producing ceramics for traditional and religious usage, Ticul’s potter families now mainly respond to the demands of a booming tourist industry, specifically its souvenir and hotel-décor outlets. In Social Change and the Evolution of Ceramic Production and Distribution in a Maya Community, internationally renowned cultural anthropologist Dean E. Arnold employs data collected over more than thirty years’ field research (from xxi

Foreword

1965 to 1997) in this single community to explore the ways in which social and cultural change resulting from a comparatively rapid entry into the modern world can be measured and explained through Ticul’s pottery industry. What, Arnold asks, are the ethnographic realities of pottery production? In what ways do ceramics and ceramic production reflect history and the social, political, and economic changes in society? And, extending his proposed line of investigation to other scholarly fields, especially the discipline of archaeology, how can the methodologies and perspectives of this type of study assist in our understanding of the role of ceramics in societies in the distant historical past? Armed with his truly unique (not to mention enviable) sets of field data, which unquestionably were bolstered by the participant-observer model of anthropological enquiry undertaken and the personal rapport and friendships established over three decades of observing Ticul’s potters and their families—Ticul’s world—the author tempers his own “clay” by evaluating established and developing theoretical models and their variables that might form the theoretical base to explain social and cultural change in Ticul’s potter community. Recognizing the high complexity of a series of related factors that such explanation must also suppose, he rejects some models, or their paradigms, and elaborates on the potential of others. In so doing he moves away from the idea that efficiency, evolution, or choice alone can explain change in Ticul, opting rather for an analysis of the cultural and social embeddedness of making pottery. Western-world ideology and theory, he observes, cannot always be applied to populations of ancient or preindustrial potters for they fail to distinguish between different cultural values placed on such crucial factors as, for example, materials and their sources, the environment, the means and modes of learning and passing on skills, the role of kinship structures, and so on, all of which also affect change—where change is to be found, of course. In this context, the relatively new and developing model of material engagement on and in technological processes is probably the most appropriate to test his thesis. Whether or not the reader agrees with the author’s theoretical stance, Arnold’s book provides cogent and compelling reading thoroughly backed up by the generosity of several chapters packed with field results and thought-provoking commentary and analysis. Scholars working in the field of ceramics undoubtedly will find these data invaluable in their own right; but for all readers it is precisely such wealth of detail that accommodates and gels the opus, leading us finally to the concluding chapter in which other initially posed questions are answered and the overall social dimension of thirty-two years’ ceramic production and distribution at Ticul is demonstrated. This is anthropological discourse at its best. Enjoy! —Eleanor Wake xxii

Preface

A

rchaeologists in the 1970s turned to the study of modern societies to help understand the relationship between residues and nonmaterial patterns. Originally, two terms vied as a label for this activity: “living archaeology” (Gould 1980) and “ethnoarchaeology” (Oswalt and Van Stone 1967; Longacre and Skibo 1994). Pioneers in these approaches, Raymond Thompson (1958), William Longacre (1991), Nicholas David (David and Hennig 1972), Richard Gould (1980), Margaret Hardin (Friedrich 1970), and the late Carol Kramer (David and Kramer 2001; Kramer 1985, 1997), are well-known, and these scholars and their students have produced much excellent research that virtually defined the field. More recently, however, ethnoarchaeology has greatly expanded in content, areal extent, and methodology (P. Arnold 2000; Arthur 2006; Bowser 2000; Costin 2000; Gosselain 1992, 2000; Hegmon 2000; Neupert 2000; Stark 2003; Underhill 2003). It has moved beyond residues to encompass larger issues of social and economic adaptations and identify political and social groups. xxiii

Preface

This book was written in the latter genre of ethnoarchaeology and is largely independent of much of the early work in this field. Judging from the perspective of syntheses of this material (David and Kramer 2001; Donnan and Clewlow 1974; Gould 1978; Kramer 1985; Longacre 1991; van de Leeuw and Pritchard 1984), it is “outside the box” of ethnoarchaeology of the last third of the twentieth century. First, this research began in 1965, before most studies of ethnoarchaeology and before the term “ethnoarchaeology” was used to categorize this kind of research. Second, I did not come to the study of the ceramic production of living peoples from archaeology but rather from linguistics and cultural anthropology. I am largely an ethnographer who is intent on relating the study of contemporary ceramic production to significant archaeological questions and assumptions (e.g., Arnold 1985; Arnold et al. 1991, 1999, 2000). A third reason that this work differs from traditional ethnoarchaeology is that although the progression from manufacture and use through discard is extremely important to the archaeologist, I have done no work on the processes that have dominated so much recent research (e.g., Beck 2006). Rather, my focus has been on the ecological context of production (Arnold 1975a, 1978a, 1993), the community and social organization of potters (Arnold 1989a, 2003; Arnold and Nieves 1992), their indigenous knowledge (Arnold 1971), their raw materials (Arnold 1971, 1972a, 2000; Arnold et al. 1991, 1999, 2000), the nature of ceramic design (Arnold 1983, 1984), and the relationship of these phenomena to archaeology (Arnold 2005a, 2005b; Arnold et al. 2007). My fieldwork has covered more than four decades of research on contemporary potters in Peru, Mexico, and Guatemala. My first publication (Arnold 1967b) linked the practices of contemporary potters in Ticul to the ancient pigment Maya Blue, showing that a semantic category used by Ticul Maya potters corresponded to the clay mineral palygorskite (then called “attapulgite”), one of the critical components of Maya Blue. At least some of the ancient palygorskite used in Maya Blue appears to have come from a source known and used by Ticul potters (Arnold 2005b; Arnold and Bohor 1975, 1976; Arnold et al. 2007). One of the greatest problems that I see in the interpretation of archaeological ceramics is the disconnect between descriptions and interpretations of archaeological ceramics and the ethnographic realities of pottery production. Many reconstructions of the past based on archaeological ceramic data seem to bear little relationship to my own knowledge of the way potters make pots, why they do it, when they do it, how they are organized, and how they transmit their craft from generation to generation. Consequently, it seems that archaeological approaches to ceramics need to be supplemented with etic and emic approaches to ethnographic pottery making. Archaeologists need to bodily engage the techxxiv

preface

nology, developing the insight that one gains from seeing humans and their technology in a cultural and environmental context over a long period of time. Some of the failures to grasp generalizations of the constraints of climate on pottery production, for example, and the energy limits to resource procurement appear to be the result of never having experienced the seasonality of pottery production, the devastating effect of rain and cold on pottery making, or the difficulty of carrying fifty kilograms of clay or temper from a source to a production unit. How much can be learned about pottery, potters, or, in general, craft specialization in the past without actually studying it ethnographically and engaging it through participant-observation? This work uses an ethnographic approach to ceramic production and aims to relate more than thirty years of observations to basic interpretive questions about ancient ceramics. It is about people who make pots and the changes that have occurred in their craft over four or more generations. Although the focus, method, and data are ethnographic, this work is ethnoarchaeological in that it seeks to examine some basic hermeneutical and epistemological questions in archaeology about the relationship of broader patterns of cultural and ceramic production. This work continues to expand the ethnoarchaeological box beyond residues to the fabric of social and economic adaptation. A fuller range of the indigenous knowledge of Ticul potters will be presented in a subsequent monograph, and another monograph will be organized around pottery-making families and the continuity and changes of their production units. For now, however, this work focuses on the changes that have occurred in the population of potters and in the activities associated with the behavioral chain of pottery making between 1965 and 1997.

xxv

Acknowledgments

T

he research on which this volume is based was funded by a variety of organizations and I am very grateful for their support. In 1965, research was funded by the Social Psychology Laboratory of the University of Chicago and the Department of Anthropology at the University of Illinois. In 1966, field research was funded by the Department of Anthropology at the University of Illinois. In 1967, a brief visit upon my return from Peru was funded by an NDFL Title VI Fellowship. In 1968, the University of Illinois Research Board funded a trip to Yucatán with B. F. Bohor of the Illinois State Geological Survey. In 1970, stopovers to and from Guatemala were funded by a grant from the Pennsylvania State University College of Liberal Arts. An American Republics Research Grant awarded under the Fulbright Program funded my research in 1984. Field research in 1988 was supported by the Human Needs and Global Resources Program and the Norris Aldeen Fund of  Wheaton College. In 1994, fieldwork was made possible by a grant from the Wheaton College Alumni Association. Field research xxvii

acknowledgments

in 1997 was supported by the Wenner-Gren Foundation for Anthropological Research (Grant No. 6163), the National Endowment for the Humanities (Grant Number RK 20191-95), and the Wheaton College Alumni Association. I am particularly grateful to the Wheaton College administrators Ward Kriegbaum, Stanton Jones, Patricia Ward, and Dorothy Chappell for their support of this research and of the preparation of this manuscript through many awards from the Wheaton College Norris Aldeen Fund and Wheaton College Faculty Development funds. The National Endowment for the Humanities (Grant Number RK 20191-95) provided a two-year grant that supported the analysis and write-up of much of these data, in addition to the aforementioned 1997 field research. These funds released me from two-thirds of my teaching from 1995 to 1997 and made the preparation of the early stages of this book possible. My study of contemporary pottery began in 1965 when the late Duane Metzger sent me to Yucatán. He gave me total freedom to do my research and to go wherever my research took me. My academic advisor at the University of Illinois, the late Donald W. Lathrap, encouraged me immeasurably to continue that research and reinforced this freedom. Correspondence with Anna O. Shepard also provided encouragement to continue my research on pottery making in Yucatán and to take my research into deeper technical issues concerning the analysis of ceramic raw materials. A brief visit to her lab in Boulder in 1966 gave me a deeper understanding of the petrography of ceramics. I am also grateful to the late Fred Strodtbeck, formerly of the Social Psychology Laboratory at the University of Chicago, who provided some of the original funds for this research. This funding was part of a larger package that set up a research institute in Yucatán that greatly facilitated fieldwork between 1965 and 1970. The late Asael “Hans” Hansen and the late Herman Konrad, who were the institute’s directors during those years, provided logistical support and helped make short research trips extremely productive. I also thank Margaret Hardin, who after being asked whether I should return to Mexico, Peru, Bolivia, or Guatemala for my sabbatical in 1984, recognized the importance of long-term ethnoarchaeological restudy of pottery making and encouraged me to return to Ticul after an absence of fourteen years. Bruce F. Bohor, formerly of the Illinois State Geological Survey and the United States Geological Survey, did the original X-ray diffraction analyses of the samples of Yucatec potters’ raw materials. Our work together in the field helped me understand the geological context of the ceramic raw materials used by Maya potters. Ewan Russell, an exercise physiologist and Professor of Applied Health Science at Wheaton College, identified the muscular patterns used with different pottery-making techniques in Ticul. Lic. José Luis Sierra Villarreal, direcxxviii

acknowledgments

tor of the Centro Regional del Sureste del Instituto Nacional de Antropología e Historia; Professor Salvador Rodriguez, director of the Escuela de Ciencias Antropológicas, Universidad de Yucatán; and the entire staff of the Centro Regional del INAH and the Escuela de Ciencias Antropológicas provided collaboration and cooperation in facilitating and supporting this research. In 1987 and again in 1989, a small grant was received from the Alumni Association of Wheaton College to hire a student (Delores Ralph Yaccino) to put all of the field notes, surveys, and linguistic texts from this project into electronic form. Other teaching and research assistants over many years have helped me immeasurably in the analysis of these data and in preparing the illustrations for publication: Heidi Biddle, Helen Woodey, Charlie Shack, Lindsay Wiersma, Christy Reed, Sara Sywulka, Matt Wistrand, Susan Crickmore, Becky Seifried, and Danae Mullison Lauffer. They and others whose names I may have forgotten have helped in many ways, providing library assistance; preparing photos, charts, and graphs; and editing and checking. Finally, several artists and draftsmen worked on the maps and photos. Mike Anderson did some of the sketches and maps, George Pierce prepared the map of the locations of potters, and Bill Koechling created stunning digital images from sometimes less-than-ideal transparencies. Finally, this book would not have been possible without the kindness, help, and cooperation of all my potter friends in Ticul. I trust that this publication will provide an increased visibility of their craft that will ultimately benefit their economic well-being. I am also grateful for conversations with many individuals whose information enriched this book immensely. Robert Bede Clark, a studio potter from the University of Missouri, Columbia, provided anecdotal information about the medical problems of long-term professional studio potters. In 1984, I taught a course at the School of Anthropology at the University of Yucatán, and a student told me of a survey that he had done on artisan stores in Mérida, which I briefly summarize in Chapter 4. I am embarrassed that I did not record his name or the thesis name. The late Louana Lackey provided critical comments on the forming chapter. Lindy Scott of the Department of Foreign Languages at Wheaton College provided help with the subtle Mexican meanings of Spanish vocabulary in my field notes that had eluded my understanding. Elizabeth DeMarrais, Andrew Jones, Bill Sillar, and Anne Underhill had the patience and tenacity to read an expanded version of the manuscript at an early stage and made many helpful comments on it. During spring 2005, I used a draft of this manuscript as a text for my Ceramics and Culture course and required students to critique it to develop their critical-thinking skills. The results were phenomenal and I am grateful to the students in that class: Natalie Burris, Stephen Chu, Ashley Cofield, Joel Duncan, Megan Hamilton, Elizabeth Hartman, Stephanie Nelson, and Michelle xxix

acknowledgments

Villaume. Ann Underhill, Chris Pool, Anabel Ford, Eric Blinman, and Antonio Curet all read the manuscript at a final stage and helped refine it. I cannot forget the love, encouragement, and support of my late parents, Eldon and Reva Arnold of Dell Rapids, South Dakota. Although they never really understood much about anthropology and what I did for a living, they were always supportive and encouraging. My father died a few hours after reading my tribute to him in Ceramic Theory and Cultural Process (Arnold 1985). My memory of him continues to inspire me, and he would have been very proud of this and previous works. My mother died shortly after the final copy of this book was submitted. To my parents and to my wife, June, and my daughters, Michelle and Andrea, I am grateful, for without them and their patience and encouragement, this work could not have come to fruition. Finally, my daughters have helped immeasurably with this book. I took Andrea into the field with me in 1994, and she served as general gopher and assistant. Michelle went with me in 1997, and under my supervision, took photographs and made floor plans of the production units visited. Although the floor plans and many of the photographs will be used in a subsequent monograph, she took many of the photographs used in this book.

xxx

Social Change and the Evolution of Ceramic Production and Distribution in a Maya Community

Chapter

Introduction

one

U

nderstanding the relationship of pottery and society is fundamental to archaeology. Inevitably, pottery, its production, and its distribution change through time, and these changes provide fruitful sources of data for making inferences about an ancient society. But how precisely do changes in ceramics and ceramic production reflect history and the social, political, and economic changes in a society? How are social changes materialized in the pottery of a society? This book attempts to provide some answers to these questions by examining the relationship of sociocultural change to the production and distribution of pottery in Ticul, Yucatán, Mexico, over a period of thirty-two years. Although the study of modern preindustrial pottery production, distribution, use, and discard has become increasingly popular among archaeologists (D. Arnold 1993; P. Arnold 1988, 1991a, 1991b, 2000; Arthur 2006; David and Hennig 1972; David and Kramer 2001; Gould 1978, 1980; Hayden and Cannon 1984b; Kramer 1979, 1985, 1997; Longacre 1991; Longacre and Skibo 1994; 

Introduction

Miller 1985; Sillar 2000; M. Stark 1991a, 1991b, 2003; Underhill 2003; van der Leeuw and Pritchard 1984) and some have documented changes in modern ceramics over time (Arnold 1987; Arnold et al. 1999; Longacre and Skibo 1994; Thieme 2007), there has never been an ethnographic study of pottery production and distribution that spans a period of more than thirty years using data collected by a single investigator. This unique perspective is both a weakness and a strength. Its weakness is that a single investigator produces a much smaller amount of data than larger ethnoarchaeological projects. The strength of such an approach, however, lies in benefits that relate directly to the scientific questions of validity and reliability. Observation through time by a single set of ethnographic eyes provides a relatively objective perspective honed through repeated visits that takes one beyond one’s own culture into the culture of the potters and their technology. Consequently, such an approach provides a holistic perspective of the relationship of pottery and social change that is directly relevant to archaeologists’ study of the past. Paradigms of Pottery and Social Change Currently, several paradigms seek to explain social and technological change through time and relate that change to pottery. Using several paradigms provides a holistic perspective on change that transcends the limitations of any single perspective. Specialization and Evolving Complexity

One paradigm that relates pottery and social change involves the relationship of changes in technology to changes in production organization, and how those changes can be read from ancient ceramics to infer increasing social complexity. Such complexity involves an increase in the number of social groups in a society and their interconnections. Variously described as the evolution of specialization and the study of socioeconomic complexity, these studies have their roots in the work of eighteenth-century economic philosopher Adam Smith (1953 [1776]) and nineteenth-century sociologist Émile Durkheim (1933 [1893]). For both of these writers, the first major step in the evolution of socioeconomic complexity involves the division of labor not according to gender. Adam Smith (1953 [1776]:7–22) argued that labor was divided for three reasons: (1) the improvement of dexterity (skill), (2) time savings that resulted from the elimination of the time required to move from task to task, and (3) the labor-saving role of machinery. All of these factors were underlaid, he argued, by the human “propensity to truck, barter, and exchange one thing for another” (Smith 1953 

Introduction

[1776]:23). Durkheim (1933 [1893]) recognized the profound significance of the shift to specialized tasks in the evolution of society, but he was more concerned about the nature of the glue (i.e., its “solidarity”) that held these kinds of societies together. In order to sustain itself with a change to organic solidarity, a society had to engage in some kind of trade or exchange so that food could be obtained by nonagricultural specialists. In more recent years, some scholars have focused on series of social types or modes of production (Peacock 1982; van der Leeuw 1976) that relate to the organization of production, whereas others have been more concerned about the hypothetical material correlates of specialization, such as product standardization (Arnold and Nieves 1992; Benco 1987; Blackman et al. 1993; Crown 1995; Rice 1981, 1991; B. Stark 1995; Underhill 2003). Rice (1981, 1991) has proposed transition points of emerging socioeconomic complexity. The first transition point, she argued, like Adam Smith and Émile Durkheim, was the division of labor when some households became potters rather than farmers and exchanged their pots for food. Using Adam Smith’s explanation, she proposed that the second transition point occurred when pottery making became more efficient, technological changes made economies of scale possible, and pottery became more standardized. Ethnoarchaeological studies, however, have revealed that the dimensional standardization of pottery vessels is quite complicated and can be produced by a diverse set of causes. First, it is hard to assess standardization in antiquity given different size categories in an archaeological context (Longacre et al. 1988). Second, standardization is not necessarily an outcome of fabrication technology and does not take into account the agency of the potters; some vessels may deliberately be more standardized than others (P. Arnold 1991a; Arnold and Nieves 1992). Similarly, in a study of conical cups from the late Bronze Age on the islands of Kea and Melos in the Aegean, Berg (2004) has shown that their homogeneity may not be the result of economic or technological factors but rather is simply the result of potters trying to make perfect copies of prestigious vessels used in rituals associated with the Minoan culture on Crete. Intentional standardization, Berg proposed, has very different implications than accidental standardization. All of this research suggests that even in antiquity, the reasons for standardization are more complicated than one may think. Further, in ceramic pastes, many factors can account for standardization that may have nothing to do with specialization (D. Arnold 2000). Cathy Costin (1991, 2001, 2005, 2007), Christopher Pool (1992), and Chris­ to­pher Pool and George Bey (2007) have brought a great degree of coherency to much of the specialization literature. Costin (1991) presented four parameters of 

Introduction

specialization that consist of a range of variation of behavior between extremes. Her description emphasized degrees of changes on a gradual progressive scale rather than just the presence or absence of different features, types, or modes of production. Although she also proposed eight types using these different parameters, she argued that it is more important to describe specialization accurately, how it develops, and how these parameters are expressed differently in different environmental and cultural conditions (Costin 1991:9). Costin (2005) provides compelling reasons why the production of all crafts should be considered together. Nevertheless, pottery production is uniquely different from other crafts. These differences include the unique nature of clay minerals that require certain environmental conditions for fabrication, drying, and firing. Because Pool (1992) and Pool and Bey (2007) focus uniquely on ceramic specialization, their work resonates more clearly with the ethnographic realities of pottery production than with attempts to lump all crafts together. Rather than describe Costin’s parameters in terms of all crafts, I will summarize her parameters as they are applied to ceramics. This summary will also include some refinements based on my research in studying pottery making in Peru and Guatemala and on some of Pool and Bey’s critique (2007) of Costin. Detailed interactions with Pool (1992), Pool and Bey (2007), Costin’s more recent work (2001, 2005), and her critique of her previous syntheses of the study of specialization in general (Costin 2007) are far beyond the scope of this book. Context. Costin’s first parameter consists of the demand for the potter’s wares. At one end of the range are what Brumfiel and Earle (1987) called “independent specialists” who produce utilitarian vessels for ordinary consumers. Such vessels are used for food preparation, cooking, and serving and generally are vessels used for household sustenance. Production in this context, Costin (1991) proposed, is most often driven by profit or efficiency. Consumers, on the other hand, choose among alternative vessels based on cost, quality, or sociological factors, or some combination thereof. At the other end of the range of Costin’s context parameter, attached specialists produce vessels for limited demand by a highly restricted clientele. These vessels have critical importance within the political economy and for the status, power, or control structure of the society because they are symbols of wealth, power, and status. Consequently, access to these vessels is restricted to the elites who control their distribution by regulating their production. This kind of distribution thus restricts consumption because elite sponsorship controls the timing, availability, cost, quality, and kind of production of certain types of ceramic vessels and their ultimate distribution (Costin 1991:11–12). 

Introduction

Costin (1991) argued that economic factors underlie the evolution of independent specialists and differ from those that promote attached specialization. Sufficient demand must exist to support specialists economically (D. Arnold 1985:155–166), and Costin suggested that demand may be a consequence of a large population size and density. Population size and growth do provide deviation-amplifying feedback for the demand for ceramics and influence the development of specialization, but the relationship is more subtle and nuanced than one might think (D. Arnold 1985:155–166). Large populations provide a large market for pots, and a growing population increases the demand for pottery, resulting in an increase in the number of potters to supply the larger population of consumers. The demand for ritual pottery, however, probably provides the greatest deviation-amplifying effect on demand (D. Arnold 1985:158–165) and was elaborated by Spielmann (2002) with an extensive literature review. Further, trade and transportation networks extend the demand for ceramic products (D. Arnold 1985:165–166), and this extension may reflect higher levels of political integration (Costin 1991:11–12). Finally, Costin proposed that specialization may evolve under conditions of unequal resource distribution, especially when individuals and communities lack sufficient subsistence resources (agricultural land, water, or pastures) to sustain themselves. This explanation is affirmed by pottery-making communities in the Valley of Guatemala and in Quinua, Peru (D. Arnold 1975a; 1978b:330–334; 1985:168–196; 1993:52–71), where the existence of ceramic resources and lack of subsistence resources (limited, nonexistent, or poor-quality agricultural land) are complementary explanations for the development of specialized ceramic production. More specialized pottery production probably was selected because the eroded land exposed abundant ceramic resources and because potters lived near critical markets for their pottery. Potters in Quinua, Peru, for example, lived at a crossroad of prehistoric, historic, and modern routes through the south-central Andes and were thirty-one kilometers from Ayacucho, the regional capital (D. Arnold 1993:23, 39, 41–47). Similarly, potters in the Valley of Guatemala lived a few kilometers from Guatemala City, the country’s political and economic center, which provided a significant transportation hub for buses and trucks (D. Arnold 1985:165–166; Reina and Hill 1978). The resources, limited and poor-quality agricultural land near Quinua, and favorable weather and climate down the slope in the ancient city of Huari placed potters in a favorable position to intensify their craft by producing a greater variety of polychrome wares. When drought threatened agriculture during the Middle Horizon, potters were strategically located to intensify their craft and distribute 

Introduction

their polychrome wares widely utilizing socioeconomic and sociopolitical institutions to buffer their decreasing subsistence returns (Arnold 1993:209–217). More recently, Costin has used “demand” for this parameter, and in reality, it is a clearer way to describe it. In effect, this terminological change mitigates the problems of “phenomenological classification and lexical semantics” that Costin (2007) herself enumerates in her most recent work. Concentration. Costin’s second parameter concerns the spatial distribution of potters and their spatial relationship to one another and to the consumers that they supply with pottery. At one end of the range are potters that are evenly distributed across the landscape. At the other end, potters are aggregated in such a way that many production units are located in a single community and their products must be traded and exchanged for the products of other communities that do not have these specialists (Costin 1991:13). Within the more specialized portion of this range, the spatial arrangement of potters has multiple layers. First, the distribution of specialized potters in the regional landscape consists of spatially discrete (rather than continuous) populations of potters relative to non-pottery-making populations. In such cases, the greatest distance between production units within a local population is less than the distance of the aggregate of those production units to another population of potters. Second, this distribution of potters occurs relatively close to resources. Costin argued that independent specialization is often nucleated because production communities are close to resources, which are unequally distributed across the landscape (Costin 1991:14). Clay deposits, however, are often widespread and widely distributed. Although high-quality clay deposits are not so widespread, individual production units are seldom located more than a seven-kilometer walk from their resources (D. Arnold 1985:35–57; 2005a), except perhaps in the short term. Although Rice (1987:116) argued that other factors affect the distribution of pottery-making communities relative to their resources (such as markets and fuel), these factors are, in reality, secondary. In a survey of fuel resources of the Near East, Frederick R. Matson (1966) found that fuel resources for traditional crafts were varied and abundant and one type easily could be substituted for another. Often agriculture provided combustible by-products that were used for fuel or substituted if more desired fuel sources were not available. Further, markets and ceramic-distribution networks are culturally constructed. Potters’ distance to clay and temper resources, however, is partly the result of evolutionary forces selecting communities with small distances to resources rather than the distance itself influencing the location of production. 

Introduction

In order for ceramic resources to play such a role in the location of potters, there must be a push away from agricultural subsistence. This change occurs when potters live on or near marginal agricultural environments and ceramic specialization is selected because agriculture is insufficient for subsistence needs (D. Arnold 1975a; 1985:168–196). Finally, a third level of spatial arrangement consists of the distribution of production units within a local population. At this level, spatial distribution of potters is more complicated. In Yucatán, potters are located in nucleated communities of potters and non-potters. By way of contrast, in Quinua, Peru, they are disbursed over the rural landscape but also within a population of non-­potters (D. Arnold 1975a; 1993:49–51, 65). Similarly, in the northern Valley of Guatemala, potters’ production units are not totally dispersed or highly nucleated but occur both in agglutinated settlements and in dispersed settlements (D. Arnold 1978a, 1978b). Scale. Costin’s third parameter consists of scale and involves two interrelated variables: size of the production unit and the principles of labor recruitment. Size consists of the number of potters per unit, and labor recruitment consists of the composition of the unit and the way in which potters are brought into those units. At one end of the range are small, family-based units in which recruitment is based on kinship, whereas industrial production lies at the other end of the range where the recruitment is contractual and is based on skill and availability. Costin proposed that as production units grow, recruitment of close kin gives way to more distant kin, or fictive (or adoptive) kin, and ultimately, non-related individuals are added to the production unit (Costin 1991:16). Based on a visit to a ceramics factory in Ticul in 1997, recruitment for industrial production may be contractual, but it is not based on skill. Although the Ticul factory (as well as other factories) requires some skilled positions, skills may be acquired on-the-job. Furthermore, skills and knowledge required for a specialized position in a factory are less holistic and much less demanding than those of a traditional potter (see Arnold 1971, for example, for a potter’s knowledge of raw materials). Costin argues that the primary factor determining the scale of production for independent specialists is efficiency and is a function of the technology used and the level of production-unit output. She proposed that production-unit size will rise to take advantage of economies of scale if per-unit costs can be lowered through sharing expensive technology or by dividing tasks among many workers. Furthermore, larger units with greater output may be able to exploit certain marketing strategies (Costin 1991:16). As we will see, however, pottery production 

Introduction

in household-based production units has the advantage of household labor in the form of unskilled children and other relatives who can participate in production. These personnel resources can temporarily increase production-unit size when demand for their wares increases. More recently, Costin (2001) separates size of the units from the composition of the units and calls the composition their “constitution.” Similarly, Pool and Bey (2007) have challenged Costin’s conflation of production-unit size and labor recruitment into the same variable. They argue that these two components must be separated if one is to understand the degree to which they are related. The Ticul data show that Pool and Bey are correct in their argument for the importance of separating size and recruitment because they are very different phenomena; there are real limits to the size of production units that are kin-based. Further, as this study will demonstrate, overall output of a population for potters can be increased even with a very small increase in mean size of the production units in that population. Intensity. Costin’s last parameter consists of the amount of time that potters spend on their craft. The lower end of the intensity range consists of part-time specialization in which craft production supplements subsistence. At the other end of the range is full-time specialization where potters exchange their vessels for all required goods and services. Costin proposed three economic factors that determine whether production is part-time or full-time. First, efficiency affects intensity because routinizing production lowers per-unit costs. This change increases output and gives full-time potters a competitive edge over part-time potters. Capital investment in technology can be spread out over the production output, and per-unit costs are reduced by keeping tools and equipment operating as much as possible. Capital-intensive production thus requires full-time production to be cost-effective and eventually requires fewer full-time workers by requiring more skill and training (Costin 1991:16–17). The Ticul data, however, suggest that more capital-intensive production requires less skill and training (D. Arnold 1999). With some forming technologies, such as molding and slip casting, for example, capital investment replaces traditional skill. Second, risk affects the intensity of production in that part-time specialists who are also farmers can buffer the risk of producing unmarketable products by raising their own food (D. Arnold 1975a, 1985). Third, Costin argues that scheduling affects intensity because of agricultural demands. For ceramics, this explanation is illustrated in highland Peru (D. Arnold 1975a), where potters are farmers who make pottery during the part of the year that has fewer agricultural responsibilities; this situation is generally true worldwide (D. 

Introduction

Arnold 1985:99–108; Underhill 2003). Making pottery, however, also is affected by an additional scheduling factor that involves the limiting effect of weather and climate on production because of the chemical structure of clays (D. Arnold 1985:61–98; Underhill 2003). One way to solve scheduling problems with agriculture is by assigning conflicting tasks to different genders (D. Arnold 1985:99– 108), but the problems of weather and climate can be surmounted only by building structures to protect drying pottery or by greatly reducing production output so that pots can dry within domestic space (e.g., Arthur 2006:42–44). Discussion of Costin’s parameters. As important as these comprehensive categories are for the description of complexity, they do not really explain how ceramic production changes or how socioeconomic complexity develops. Rather, Costin and other theorists use efficiency as an explanation for changing complexity (e.g., Brumfiel and Earle 1987:1, 5; Costin 1991:15–16, 37–39; Pool 1992:278–279; Rice 1981, 1984:244–245, 1991). This view, of course, can be traced back to Leslie White (1949:368–369), who formulated his law of cultural development that “culture evolves as the amount of energy harnessed per capita per year is increased, or as the efficiency of the instrumental means of putting the energy to work is increased.” Specifically, he argued that “the degree of cultural development varies directly as the efficiency of the tools employed, other factors remaining constant” (White 1949:374–375). Efficiency usually concerns the relationship of input to output and can be described as occurring along at least five dimensions: time, labor (energy), personnel, space, and output. There are at least two strategies to developing efficiency. One strategy involves maximizing the output per unit of input, and a second strategy involves minimizing inputs per unit of output. In the archaeological literature of specialization, however, efficiency is often linked to the speed of the fabrication technique such that as the technique changes, the output (pots) per unit of input (labor or time) increases. A more efficient fabrication technique thus produces more vessels per unit of input (whether time or labor) than another, less efficient technique. As technology evolves and more efficient techniques become available, such techniques are presumably selected by the population and the result changes the organization of the craft. Consequently, sociocultural evolution is believed to be the product of a rational mechanistic process, and when faced with a choice, humans will choose alternatives of least effort and those with greater efficiency and cost-effectiveness (e.g., Brumfiel and Earle 1987; Costin 1991; Feinman et al. 1984:299–303; Rathje 1975; Rice 1991; Zubrow 1992). As a result, efficiency and economies of scale brought on by different fabrication technologies are believed to explain 

Introduction

greater degrees of ceramic specialization (Costin 1991:15–16, 37–39; Rathje 1975:430–434; Rice 1984:244–245). This rational-choice perspective, however, is based on a Western world view that has its roots in contemporary economic theory and may not, and indeed probably is not, characteristic of populations of ancient potters in the way that one might think. Rather, this view is tainted by a view of technological development that is profoundly affected by the American cultural value of the efficiency of time that permeates our industrial society. Most recently, Costin (2001, 2005) has retreated from her position on efficiency. She argued that the use of the concept of efficiency in the literature of craft production “is problematic and misleading” (Costin 2001:289), citing theoretical and methodological problems. Indeed, in an ethnographic study of a pottery-making community in northern Mexico, Estes (2003) has demonstrated just how complicated efficiency is. One expects efficiency to be highly valued in this community and more “contaminated” by the values of an industrial society, but Estes has shown that efficiency does not clearly exist in the way that archaeologists conceive of it. Although it may seem that efficiency is the driver of technological evolution, evaluating efficiency is complicated by the unique production sequence of making pottery and by its cultural and social embeddedness. By applying Costin’s parameters to the evolution of ceramic production in one community in the ethnographic present, it should be possible to evaluate their usefulness and universality. Further, this application should provide deeper insight into the details of the evolution of ceramic production and the process of specialization and determine if the evolution of ceramic production in the modern world follows a trajectory similar to that believed to occur in antiquity. Is the evolution of specialization described by archaeologists universal, or are its principles restricted to the unwritten past? Evolutionary Processes

A second paradigm for the study of social and technological change consists of an evolutionary paradigm and complements the limitations of some of the specialization literature just described. In many respects, it is probably the most powerful and comprehensive explanation for changing patterns of organization and the development of social complexity. Part of this power comes from its theoretical maturity for explaining biological change. Nevertheless, cultural evolution is not the same as biological evolution because, among other reasons, humans are intentional agents who try to influence their own behavior and evolution in spite of their inability to effect the kind of change that they intend (Kean 2006). Nevertheless, cultures and human behavior have inherent systemic properties, 10

Introduction

such as self-organization, that can be described by power laws that go far beyond the intentions and immediate understanding of human agents (Bentley and Maschner 2001; Bentley et al. 2004; Bentley and Shennan 2003). These properties suggest that processes such as random drift are at work, and that at least some aspects of decision-making processes are value neutral (Bentley et al. 2004). Some archaeologists go to great lengths to show the similarity, or lack thereof, of archaeology to the terminology of evolutionary theory. But without clear application to actualistic societies, this terminological fundamentalism obfuscates the usefulness of evolution for those who are trying to use the models. Shennan (2000), however, simply uses the Darwinian notion of “descent with modification” to explain culture change. Assuming that a parallel exists between biological and cultural evolution, the study of culture change then should address the descent mechanism in societies and how that mechanism relates to cultural change. Unlike biological organisms, however, cultures pass on their traditions through learning rather than through biological processes, and thus, one should address the learning mechanisms and learning contexts if one is to address culture change using this model. For social and technological change among potters, then, some attention should be given to the study of learning and learning contexts in which the craft is perpetuated. As we will see, learning and learning contexts are important considerations, but they are only two out of many factors. The evolutionary model is also based on using selection to explain change. Selection operates on two interrelated levels that affect both production and distribution. On the first level, the forces of selection act on the population as a whole to eliminate or favor individual potters (or specialists) just as in biological evolution. The second level involves the ceramic vessel itself and includes changing factors of demand that a population of consumers uses to acquire, or not acquire, a pot (Neff 1992, 1993). Consequently, changing values, functional considerations, and aesthetic preferences are powerful selective forces acting upon the marketing success of potters’ distributing, exchanging, trading, or selling their wares. If potters do not adapt to these changing preferences, they will not be able to use their craft to sustain themselves, and they will have to change to another occupation. Technological Choice

A third paradigm for explaining cultural change is technological choice (Lemonnier 1986, 1992; Loney 2000). This paradigm focuses on the participants who have a choice in the innovations that are accepted. Although technological reasons may explain why a particular selection or choice is made, social reasons 11

Introduction

also exist for making choices that may have no technological basis. The cultural and social contexts are critical in this paradigm, and it is much more culturally particularistic, requiring the archaeologist to reconstruct the choices available to the ancient potter. More effective for the study of modern material culture and cultural change, this approach appears to require full knowledge of the social, cultural, and environmental contexts in order to ascertain which choices are technological and which are not. Ironically, although the technical choices with a technological basis probably can be understood from the study of materials and climate of an area, those truly social choices cannot be ascertained from archaeological data alone except by technical/material criteria. Consequently, social choices are believed to be at work when technical criteria have been excluded. This approach seems to limit our understanding of the past. The notion of technological choice has been a part of Western civilization for at least 3,000 years. In the archetypical origin story of human beginnings in the first book of the Jewish Torah (the book of Genesis), the procurement of one subsistence resource was emphasized and involved two actors (Adam and Eve) who made a choice based on social rather than technological criteria. From this story, the notions of choice, human responsibility, and the consequences of choice have been a part of Western religious and moral thought for centuries. Given its deep roots in the ideology of the West (whether recognized or not), it is perhaps inevitable that this notion of human choice should become a part of anthropological and archaeological understanding of technology. It applies to technological and non-technological phenomena and is fundamental to any holistic understanding of human culture. Although decision making and the use of choices are part of the nature of the human brain (Koechlin and Hyafil 2007; Sanfey 2007) and are fundamental to understanding human culture, the distinction between technological and social choices has limited utility. Humans obviously make choices in technology, but it is equally obvious that those choices are not always technological—or even rational, for that matter. Since the ideology, social organization, and technology of a culture are interrelated, choices in technology (narrowly conceived) may have social, ideological, and religious bases that have nothing to do with technically advantageous properties. Potters thus make choices based on tradition, religion, market, social feedback, and other non-technological criteria that have nothing to do with rational choice or technological reasons. The issue about choice for potters has been described before, particularly in relationship to ceramic design (D. Arnold 1984; Krause 1984, 1985) and fabrication technology (D. Arnold 1972b; 1978b:349–351; 1993:73–100), although it was not emphasized as such. 12

Introduction

The argument for focusing on the particulars of the past using the technological-­choice approach to ceramics (rather than their commonalities with the present) is really a movement toward suppressing an awareness of the process of knowing and interpreting. It is a move away from understanding the integrity of the process of archaeological inference and retreats into unexamined, unintended unawareness of the social theory that is embedded in ceramic analysis. Can we abandon an examination of our implicit assumptions and presuppositions about social theory and its relationship to technology by arguing that we should only examine which choices ancient potters made? This technological-choice approach (Lemonnier 1992; Loney 2000; Roux 2007; van der Leeuw 1993), however, also can be seen as a reaction to the nomothetic concerns of the role of efficiency in the specialization literature, the rational-­choice theory embedded in selectionist and ecological approaches to change, and the perceived notion that other theories are deterministic. They are not, and they do allow choice. Although proponents of technological choice like to cite its exclusiveness and the social dimensions of choice, notions of choices are also embedded within the cultural-ecological approach (Steward 1955:36) and are “not deterministic” (D. Arnold 1975c:637, 1993). This combination of an ecological approach and social choice is represented in my own work with the variability of choices that a community of potters uses in their pottery (Arnold 1983, 1984, 1993). Constraints on design choices appear to be social and structural and may have a foundation in the local community and the way in which it has adjusted to the environment, although there is great variability. Human choices thus have multiple dimensions—social, economic, technological, and religious—with multiple layers of complexity. What varies, however, are the constraints for those choices, which may be environmental, social, political, or technological. Nevertheless, the notion of technological choice reminds us of the importance of the multidimensional causes of social change, and that the individual, and the choices that the individual makes, are important, just like the story from the first book of the Jewish Torah. Despite the problems with technological choice, understanding the importance of choice helps us to focus on those aspects that influence choices, such as the feedback from the environment, technology, society, and the influence of the learned semantic categories of a culture. Cognitive Anthropology and Engagement Theory

The data in this study also reflect two other implicit paradigms. One of these is cognitive anthropology (D’Andrade 1995), which focuses on the semantic categories of a culture and their structure as revealed in the definition and structure of the categories of its language. Originally, I approached the potter’s craft in 13

Introduction

Ticul through the language of the potters (Yucatec Maya) and elicited the Maya names and descriptions for ceramic technology. This approach was begun even before I knew Spanish and was accomplished through a technique then known as “ethnoscience” (Black 1963; Black and Metzger 1965; Frake 1964; Metzger and Williams 1963a, 1963b, 1966). “Ethnoscience” was a question/response technique using a field language that revealed semantic structures of informants who used that language. Eventually, I learned the Maya vocabulary of ceramic technology and the semantic distinctions made by Ticul potters. Although these data were foundational for this study, a full description of the ceramic technology using this paradigm will be the subject of a future monograph. A second implicit paradigm used in this work is one that has come to be known as “material engagement theory.” Still in its nascent stages, engagement theory has the potential to be a truly unifying theory for the study of material culture. According to Renfrew: Material engagement theory is concerned with the relationships between humans and the material world and focuses upon the use and status of material objects (mainly created objects or artefacts) which are employed to mediate in the interactions between human individuals, and between humans and their environment. Its purpose is to facilitate the analysis and understanding of culture change. . . . It seeks to overcome the mind/matter duality by stressing knowledge-based nature of human action. (Renfrew 2004:23)

Malafouris (2004) provides a more developed schema of engagement theory, and both Renfrew (2004) and Malafouris (2004) appear to be oriented more toward using it in describing the use of artifacts rather than their creation. Nevertheless, Malafouris (2004) uses the potter’s wheel as an example for describing the theory. Unfortunately, this example appears to rely more on theoretical understanding of the wheel than on actual practical engagement with it. Although useful, this example and the elaboration of the theory could be enriched. In some respects, material engagement theory is more useful in ethnoarchaeology than in archaeology and needs to be more rooted in the actual engagement with artifacts in the empirical world of ethnography before it is applied to the past. This approach is not always possible, but it is possible with technological processes such as making pottery, agriculture, and metallurgy. Nevertheless, material-engagement theory has the potential to draw together strands of cognitive anthropology, cultural ecology, practice theory, notions of habitus (including motor habits), as well as data from the environment and the inherent characteristics and constraints of the material used. With ceramics, for example, engagement theory can take into account the importance of the way in which potters categorize their raw materials, the characteristics that are derived from tradition, 14

Introduction

and the places from which they come (e.g., Arnold 1971). It also can take into account the role of the environment in providing choices for production (cultural ecology and technical choice theory) and the actual physical properties of raw materials learned by the potter. Third, engagement theory can take into account the habitual nature of human culture. Although part of this notion is habitus, there is a firm physiological basis for habitual muscle syntax (e.g., motor habits) that is stored in a different part of the brain than language. Finally, engagement theory has the potential to incorporate feedback (D. Arnold 1985) from aural, visual, and tactile percepta derived from the interaction with the raw materials, the behavioral chain of the pottery-making process, and the language of other humans. The notion of feedback developed in Ceramic Theory and Cultural Process (Arnold 1985), for example, was one way of describing the engagement of the potter with the social and natural environment that recognizes that potters are agents, that they are not oblivious to the social and natural world around them, and that they receive information (feedback) from it in a way that affects their behavior. The point of that book was to restore a neglected perspective to ceramic studies that potters live and work in a natural world, not just a social, or socially constructed, one. I also utilized engagement theory in this monograph because I was able to participate in the pottery-making process and understand the way in which potters engage the behavioral chain of pottery making. This embodiment and interaction have enriched this description greatly and illustrate why participant-­observation is so important in anthropological research. Because technology is artifact, activity, and knowledge, actual participation in the technological processes permits a degree of understanding beyond verbal interaction and observation. I learned how to select raw materials and then selected them myself. I learned how to fire pots by first eliciting descriptions of the process in Yucatec Maya and then fired them under the watchful eye of my informant. The result of this engagement aided me in understanding the nature of technological knowledge, how it is learned, maintained, changed, and passed on to others. After learning the categories of the Maya potter and firing five times by myself, I realized the complicated, multifaceted nature of human engagement with technological processes. Similarly, by engaging in some of the activities of mining raw materials, I learned lessons about the procurement process that I otherwise would not have understood as deeply. When geologist B. F. Bohor and I visited the clay mine at Hacienda Yo’ K’at in 1968, we crawled through an entrance tunnel that was barely fifty centimeters wide and twenty centimeters high (see Figure 5.8). It was so small that I had to move through it on my stomach with arms stretched out in front, propelling myself forward by the action of elbows and toes. As my toes 15

Introduction

dug into the bottom of the tunnel, my heels simultaneously scraped its ceiling. Although the tunnel opened up inside the mine into a large excavated room, I was profoundly aware of the challenges and dangers of clay mining. The air was bad and the recording I made there revealed rapid breathing. Reflecting on this experience afterward proved to be psychologically traumatic. Showing slides of the interior of the mine to my classes and playing the audiotape made there had devastating effects on my mental state. Nightmares about claustrophobia in the clay mine plagued me for years. Nevertheless, experiencing the embodiment of technology and partially engaging in the process were necessary to understand the problems of clay mining. Visiting the Ticul clay source again in 1984 also reflected the importance of understanding the embodiment of technology and engaging in at least some of the technological processes. By this time, the 1968 mine was abandoned and clay was extracted through a series of vertical shafts sunk five to eight meters into the ground to reach the clay layer. I lowered myself into one of these shafts, as miners had instructed me, by wrapping the rope around one hand, grabbing the rope with the other, and using the footholds on the shaft wall to provide support for changing hand positions on the rope. Climbing out of the shaft was much more difficult. Using the rope to raise myself from foothold to foothold was a daunting task, and I had to rest frequently by placing my back against the side of the shaft and pushing my feet against the opposite wall. Miners had insisted that I remove my shirt and jeans to descend into the shaft in order to keep them clean, but I had only removed my shirt. Forcing my back against the wall of the shaft loosened a considerable amount of marl behind me and deposited itself in my jeans and underwear. By the time I reached the top of the shaft, I was carrying considerable extra weight. The miners’ advice took on a new meaning after my own experience of descending into a clay mine. Rather than just a dirty body that could easily be brushed off, I had a dirty body and dirty clothes and had expended unnecessary energy carrying marl up the shaft in my clothes. The lack of actual engagement in the pottery-making process is one reason why some archaeologists have a difficult time understanding ethnographic perspectives of pottery making, such as the embodied flow of feedback from the senses (e.g., eyes, ears, touch, and taste) presented in works such as Ceramic Theory and Cultural Process (D. Arnold 1985). It perhaps is one reason why some archaeologists want to deny the relevance of the ethnography and ethnoarchaeology of pottery making to the past. This lack of engagement is probably also responsible for the lack of understanding of the effect of the constraints of the raw materials and the environment on the pottery-making process.

16

Introduction

Summary

None of these paradigms has an exclusive corner on explanatory validity. Furthermore, they are not, as some would have us believe, in competition with one another. Rather, they are complementary paradigms that explain different aspects of ceramic production, and like all paradigms, they are incommensurable. Sadly, ethnoarchaeology, like archaeology, is often practiced in a “context of theoretical atomism, interparadigmatic hostility and ignorance of alternative perspectives” (Fitzhugh 2002:789). The use of multiple paradigms in ceramic description depends on how interested one is in understanding human culture holistically rather than using the latest avant-garde theories. The study of the past, and the use of the present to interpret it, should be done with a commitment to understanding real people in real situations, not to try out the latest theories and perspectives to be “in style” with the current academic fashion of the times (D. Arnold 1991). Unfortunately, anthropological and archaeological theories have made dramatic swings from one extreme to the other, switching from particularistic concerns to nomothetic themes and then back again. Rather than building on previous work, investigators tend to justify new paradigms by stereotyping previous theories (such as the erroneous belief that ceramic ecology is deterministic) and then dismissing them as anachronistic because they do not fit the prevailing paradigms. Such condemnations are common in an academic culture where theoretical change and innovation seem to have a higher value than the truth value of holistic understanding (D. Arnold 1991). They appear to reflect a belief that science is propaganda, as philosopher of science Paul Feyerabend argued (Broad 1979). Because this work covers thirty-two years of rapid social and technological change, it provides a unique opportunity to explore and evaluate these paradigms and their ability to explain this change. Are the assumptions used to infer the development of ceramic specialization in antiquity valid and useful? For example, are Rice’s (1981) transition points in her trial model in the evolution of ceramic specialization and Costin’s (1991, 2001) explanations used in her parameters of context, concentration, scale, and intensity universal enough to be used in a contemporary context? Further, are efficiency and selection drivers for the change and stability of the craft? Do they explain production organization through time? Do production units grow in the way that Costin proposed that they do? The Limits of Ethnographic Analogy One of the problems of using data from the present to understand the past concerns the perceived limitation of the use of analogy. Cultures change and the 17

Introduction

conditions of the twentieth century are not the same as those that existed hundreds of years ago. Few modern societies can be related to ancient societies in a direct historical way, and it is often difficult, if not impossible, to determine how much a modern industrial cash economy and an extensive transportation and communication infrastructure have influenced a demand for pottery. All archaeological interpretation, however, is analogical, even that which eschews ethnographic analogy (Wylie 1985). Ancient societies are described and understood as a result of analogical thinking between what is known and what is not known. There is always some degree of ethnographic analogy in archaeology whether or not archaeologists realize it. In one sense, the prehistoric past is always incommensurable with the present. The real issue, however, is not incommensurability but rather the inexplicit role of one’s own presuppositions in understanding the past. All theories of the past come from our minds in the present and have been affected by often implicit and explicit social theories, personal assumptions, and academic tradition. To paraphrase Norwood Russell Hanson’s (1958) perspective, all data (even those that come from the past) are theory laden. Although it has been argued that the source of our testable hypotheses does not matter, in anthropology their source really does matter. The world of human culture is highly contextual with often limited possibilities, and archaeological interpretations cannot just be developed through hypothesis testing. Such hypotheses tend to be mono-casual with too little awareness of the theoretical and personal biases inherent in their selection. One of the ways to avoid some of the limitations of analogy and the problems of a direct historical approach to studying the past is to build a theory that focuses on the common linkages cross-culturally between ceramic production and behavior. By comparing these links with those in other societies, it is possible to formulate cross-cultural regularities that can be applied to both the present and the past. If the interpretation of ceramics is ever going to get beyond its tradition-bound categories, naïve inductivism, and culturally myopic interpretations, archaeologists need to use a cross-cultural theory that focuses on the commonalities of ceramic production worldwide. I have already formulated such a theory (D. Arnold 1985) built on the presupposition that societies that produce ceramics share common adaptive processes by virtue of the chemical similarities of clays. In other words, the behavioral chain of ceramic production has cross-cultural commonalities because of the similar molecular structure of clay minerals. These commonalities provide universal problems to which potters must creatively adapt if they are to make pottery. In some cultural and environmental contexts, there is more latitude in the choices to adjust to such problems than in other contexts (Steward 1955:36). 18

Introduction

I have applied this theory to one community of potters in a single point in time (D. Arnold 1993) using both universal adaptive processes of ceramic production (developed in D. Arnold 1985) and those more humanistic factors (like social patterns and religion) that affect the patterning of human technological choice in production, decoration, and distribution. I argued that the community of potters, rather than the pots, is the unit of adaptation and evolution, and that this unit has specific expressions in the design structure of the pottery that are different from those of neighboring communities, even though there is great choice in design within the limits of the design structure (D. Arnold 1983, 1984, 1993). This approach is consonant with the community-level analysis of Kolb and Snead (1997) and Yeager and Canuto (2000). The community of potters is thus not only a unit of adaptation, evolution, and production, but its material expression in the ceramics suggests that it should also be the unit of analysis and interpretation in the study of ancient pottery (D. Arnold 2005b). When the community of potters is understood in relationship to the local environment, distance to raw materials, subsistence scheduling, and settlement patterns, it is possible to infer the location, paste variability, settlement pattern, and intensity of ceramic production in antiquity. In the case of the Ayacucho Valley, Peru, understanding the contextual factors of pottery production around the village of Quinua provides some understanding of location, paste variability, scheduling, and intensity of pottery production within the great pre-Inca city of Huari, which flourished between A.D. 600 and A.D. 800, less than three kilometers down the slope from the location of most of the Quinua potters (D. Arnold 1975a, 1993:204–226). Ceramic production is, of course, far more complicated and needs to be understood through other paradigms, but using an ecological approach and the notion that the community of potters is an adaptive response does provide a basis for developing ethnographic analogies that have comparability with the past. Other scholars have successfully examined ethnographic cases of pottery production and distribution cross-culturally and provided a framework for interpreting the past. Pool (1992) synthesized a variety of production and distribution data for ceramics and developed some principles with which to understand production and distribution issues as well as the meaning of material residue patterns of ceramics. Similarly, Costin (1991, 2000, 2005) focuses more generally on issues of specialization cross-culturally, even though increasing ceramic specialization provides some unique adaptive problems that cannot be dealt with in a onemodel-fits-all format. Spielmann (2002) examines the role of “social demand” in the development of specialization, elaborating on my argument that utilitarian demand was insufficient to generate deviation-amplifying feedback for the development of specialization (Arnold 1985:158–166). Tourist demand today is 19

Introduction

much like ritual demand in that it is open-ended and has the potential to provide deviation-amplifying feedback for the development of increased intensification and specialization of ceramic production, much as it has in Ticul since 1965. The ethnographic study of social and technological change in a modern peasant community is not unrelated to understanding the past. Although there are many differences between the present and the past, there are also many similarities. The processes of social and cultural evolution do not change, and in a real sense, the Yucatec Maya, like other Indian peasants of Latin America, are products of a complex blend of forces operating on a society that had its foundation in the pre-Columbian past. By studying this evolution in the present, it is possible to understand some of the processes that were responsible for continuity and social and technological change in ceramic production and distribution more generally. Indeed, this was the point of the generalizations developed in Ceramic Theory and Cultural Process (Arnold 1985) and also the refinements of the threshold model of ceramic resources (Arnold 2005a). Finally, I was reminded about the universality of evolutionary processes that are shared by the present and the past when I was perusing a copy of World Trade: The Journal of International Logistics. In an editorial, the journal’s editorial director responded to a colleague who had emphasized the downside of increased international trade and globalization. [T]he dynamics of global flows—capital, labor, materials, technology, markets—are transforming the world from one economic order (industrial/ national) to another (informational/multinational). This is not the first time such transformation has occurred. Human society went from rural agrarian feudalism to mercantile kingdoms to private commerce and nation states. One might as well order the seas to stop reshaping the shoreline as to curtail this social law of nature. (Shister 2007)

Although Shister uses considerable hyperbole (e.g., “social law”) and seems callous about the social and personal costs of globalization, his comments do serve to remind us that cultural evolution is a continuing process that began in the remote past. Ethnoarchaeology makes it possible for us to learn about some of the forces and generalizing principles that have led to accelerating social and economic change and are rapidly reforming ceramic production and distribution. On the contrary, to argue that the current world has no relevance to the non-­globalized past is to believe that there is an evolutionary discontinuity of the present and the past—an epistemology that I have already argued was naïve. It is reminiscent of a “creationist epistemology” that denies that the processes observed in the present have any relevance to understanding the past. In reality, cultural evolution in the present is accelerating at a rapid pace, collapsing cen20

Introduction

turies of cultural evolution into the span of a scholar’s lifetime. A thirty-twoyear diachronic study of ceramic production thus has the potential of helping us understand ancient cultural evolution, which with the proper cautions can help us understand the evolution of ceramic technology in the past. Collecting Data in the Field The bulk of the field research for this book took place during ten visits to Yucatán, Mexico, over a period of thirty-two years. I first went to Yucatán in 1965, when I spent six months studying potters in the city of Ticul. Between 1965 and 1970, I returned five times and then went back again in 1984, 1988, 1994, and 1997 (D. Arnold 1967a, 1967b, 1971, 1987, 1989a, 1989b, 1991, 1997, 1998, 1999, 2000, 2006; Arnold and Bohor 1975, 1976, 1977; Arnold and Nieves 1992; Arnold et al. 1999, 2000; Ralph and Arnold 1988). I also returned for a brief visit of a few hours in 2002. Although it was not a formal research visit, it brought additional time depth, data, and insight to this study. The scope of this study thus spans 37.5 years and provides an unusual ethnoarchaeological perspective that details continuity and change in the production, organization, and distribution of pottery through time. One advantage of long-term research in the same location is its high degree of validity. Because I have returned to the same community again and again for more than thirty years, I have become well acquainted with potters, their relatives, their residence locations, and their technology. Many potters have become my friends. Each visit built on the rapport and knowledge of previous visits, and as a result, I could assess, even with a brief visit, the veracity of informants’ statements and the validity and continuity of long-term patterns. Consequently, I can detect both deliberate and involuntary deception and can easily verify and cross-check data. I can easily identify changes from previous visits, and because I have known most of the potters in the community, I could ascertain, often from observation alone, who is making pottery and where they are making it relative to the patterns of residence and social organization of the community. The relationships established with Ticul potters thus were a major contribution to the success of my research, and the information from them was validated again and again through the three decades of this research. From a methodological perspective, participant-observation was foundational. This classic anthropological methodology provides a holistic perspective of human behavior that helps the investigator get beyond the cultural, theoretical, and paradigmatic myopia of interpreting ceramics. Through participantobservation, one comes to understand pottery production and distribution from 21

Introduction

both the inside and the outside. As a result, one begins to comprehend linkages between phenomena that may never have been envisioned previously. At the same time, one’s objectivity is never completely lost because the investigator is still an external observer, not a native (Arnold 1991). Such a holistic approach is complementary to a rigorous scientific methodology and analysis both in the field and in the laboratory. Just as scientific and technical studies of ceramics (Arnold et al. 1991, 1999, 2000; Glowacki and Neff 2002; Pinto et al. 1987; Rice 1987; Rye 1981; Skibo and Schiffer 1987; Skibo 1992, 1994) are essential for understanding the past, so a holistic approach to ceramic production is necessary for inferring the links between ancient ceramics and the nonmaterial behavior patterns that produced them. Such an approach is foundational for understanding how pots relate to people. Similarly, there is no other way to learn the semantic categories, the motor habits, and the choices of the potter than through a personal engagement of the investigator with the people and their craft production. All ethnographic work also is socially embedded, and ethnographers, like potters, learn their craft in a social context. To chronicle this research experience, detailed field notes were written to supplement data obtained by surveys, photographs, and a question/response technique (called “ethnoscience”) used early in the research (Black 1963; Black and Metzger 1965; Frake 1964; Metzger and Williams 1963a, 1963b, 1966). As the research progressed, I began to see the importance of field notes as an independent data source and eventually came to use them as a description about everything that I learned about pottery production, whether or not it was relevant to the goals of the research at the time. Eventually, I discovered the importance of emptying my brain into my typewriter or computer until everything that I had learned on a given day had been written down. The ideal was not to return to observing and talking to informants until the recording process was complete. This ideal was not always realized because in some situations, two or more days elapsed before field notes were transcribed. During some days in 1965, for example, no written field notes existed because I did not have the time to do it. To solve this problem in 1968, I experimented with the use of a tape recorder. Although it provided a useful way to record field notes in the backseat of a Volkswagen driven by a colleague, lying in a hammock, or deep within a clay mine, getting the information off the tape and placing it into its appropriate context proved to be a difficult task, and one with which I still struggle more than thirty years later. Not the least of my difficulties was the occasional unintelligible gibberish I uttered into the recorder while lying in my hammock, exhausted after a hard day’s work. Thinking that I could record my field notes before I went to sleep, I soon discovered that my consciousness usually shut down before my mouth did. Recording 22

Introduction

field notes had to be more deliberate, conscious, and intentional with a fuller description of the context of the day’s experience. So I abandoned the use of a tape recorder after 1968 and considered my use of it for field notes as a failure. My visits to households often jogged informants’ memories of incidents that I had long forgotten and provided almost instant rapport. In 1997, a widower I had not seen for thirteen years invited me to his house for lunch. He was living with his only daughter and her husband. During our conversation, he proudly showed me a tinted 11 × 14–inch enlargement of a photograph of a young couple and their baby that I had taken thirty-one years earlier. It was one of many Polaroid photos that I had provided to informants in 1966 in order to build rapport with them when I had surveyed potters’ households, collected temper samples, and asked for information (D. Arnold 1967a, 1967b, 1971). The enlargement process, however, had not only amplified the image but had exacerbated the testimony of many years of handling of the original photograph and ragged wear around its edges. I was puzzled why an enlargement of my photograph was such a cherished memento until I returned home and examined my genealogical database (see below). At the time when the photo was taken, the potter had been married for two years and his wife had given birth to a daughter. A year later, however, his wife had suddenly died. My photograph appeared to be the only image that the widower had of his deceased wife and was the only visual record that he and his daughter had of her. Comparison of the data from Ticul with other pottery-making communities in Latin America has provided a comparative perspective that highlights significant insights in this work for archaeology. Having been a participant-observer in other pottery-making communities in Guatemala (D. Arnold 1978a, 1978b; Arnold et al. 1991) and Peru (D. Arnold 1972a, 1972b, 1975a, 1983, 1984, 1993) since the beginning of this study in 1965, I come to this book with diverse ethnographic experiences in the study of preindustrial ceramic production. These experiences have enhanced my relative objectivity in approaching the data presented here because I understand them within a comparative framework. This perspective increases the effectiveness of the archaeological application of this work because my experiences studying pottery-making communities in two other areas of the Americas serve to de-emphasize those data that do not have cross­cultural relevance or application. This book joins a rich legacy of ethnographic and ethnoarchaeological descriptions of pottery production in Mesoamerica (D. Arnold 1978a, 1978b; P. Arnold 1991a, 1991b; Deal 1988, 1998; Druc 2000; Foster 1948, 1955, 1960a, 1960b; Friedrich 1970; Hayden and Cannon 1984b; Kaplan 1974, 1980; Krotser 1974; Lackey 1982; Papousek 1974, 1981; Pastron 1974; Reina and Hill 1978; 23

Introduction

Thompson 1958; Weigand 1969; Williams 1992, 1994a, 1994b, 2006; Williams and Weigand 2001). Although the most direct application of this work will be in the Maya area and in Mesoamerica, this work should provide understanding of some basic processes of pottery production and distribution that have wide­ranging application to the present and the past. Data Reduction and Analysis From all of this research, three electronic databases were assembled. The major purpose of these databases was to compare the data from all ten of my visits and trace individual potters and production units through the thirty-two years of this project. The databases thus provided a means to organize information, facilitate writing the text, and support the points in the text with evidence. The Genealogical Database

Genealogies of the entire population of potters in Ticul provide the data for assessing change in social organization. My primary purpose in developing this database was to graphically represent the relationships among potters across all of the generations and provide links among the seventy-one kinship charts elicited in 1984. These links were complex and it was difficult (almost impossible in some cases) to see patterns from chart to chart. In order to understand the kin structure of the community, I needed to represent the genealogical data holistically and from the perspective of several different families. So the data from the 1984 kin charts were entered into a genealogy program (Parsons’ Family Origins program). The resulting compilation consisted of 1,024 individuals, 287 nuclear families (e.g., mother, father, and children), and 659 events. The events category consists of a group of fields that includes birth date, marriage date, death date, residence, place of origin, occupation, and other information. The events data, however, are uneven because some individuals have many such events and other individuals (such as those deceased for more than one generation) have no entries in the events fields. This database has proven to be a great advantage for analyzing the data, and for writing this book. First, it provided an easy way to correct and update the kin relationships with new data collected in 1988, 1994, and 1997. Second, the genealogical data can be accessed in a variety of formats, including ancestors, descendants, family groups, and standard family trees. Third, it made quantification of the data much easier. Fourth, it provided a quick and easy way to answer specific questions about kinship relationships. Each time a question arose about the relationship of one potter to another, I went to this database and answered 24

Introduction

questions such as, Who were the descendants of Potter X? Who were the ancestors of Potter Y? How many children did Potter Z have, and what were their names? How are the five generations of Family A related to the five generations of Family B? The data in the genealogical database were supplemented with data from microfilmed records of marriages from the Ticul church. Early in my research I had learned that a colleague was using microfilmed birth and marriage records from Ticul that were available at the genealogical library in Salt Lake City. I had always wanted to consult these records and expand my genealogical data. During more than thirty years of research in Ticul, all of my observations seemed to fit with my genealogical data. There were, however, some ambiguities that always seemed to slip through the cracks, and I was anxious to resolve them. Furthermore, I was hopeful that the microfilmed records might provide more precision on birth and marriage dates and also more data on potters than I had collected in the field. Much to my delight, I discovered that my genealogical data proved to be accurate, and the church records succeeded in resolving ambiguities that had puzzled me for almost fifteen years. The data also aided me in filling in some missing details, such as birth and marriage dates. The Production-unit Database

The second database used in preparing this book was compiled from data collected during visits to production units. Each potter who was making pottery during each visit has a record (N = 300) in the database and each record contains a set of fields for each visit from 1965 to 1997. In total, there are eighty-eight fields for each record. The set of fields includes the type of potter (an owner of the production unit, worker, or relative of the owner), production location, type of production, its address, the type of pottery produced, and the names of other potters working there. Three other data fields were used to record the data about any pottery store that was associated with a production unit. After the 1997 fieldwork, fields such as “helpers” and “painters” were added because some production units were becoming increasingly specialized with workers who were not potters. Additional summary fields were added to provide quantitative comparison between the different visits. Some fields have few data (such as those from the 1967 and 1968 visits), whereas other fields (such as those for the 1965, 1966, 1984, 1988, 1994, and 1997 visits) have many data. Since the 1965 and 1966 visits were only six months apart and the data were complementary, the data from these visits were considered contemporary and were combined into a new set of fields, the 1965–1966 fields. 25

Introduction

The production-unit database facilitates tracking individual potters and production units through the many years of my research. These data revealed changes in the composition and size of production units over time, changes in pottery produced by each production unit, and the growth of stores used to sell the pottery. I have also used this database to write a historical narrative of each production unit showing the change of each through the three decades of my research, but these narratives will be a subsequent monograph. The Potters Database

This database has a different structure than the genealogical database and the production-unit database. It consists of 451 individuals who learned pottery making sometime during their lifetime and organizes data about learners and active, inactive, and deceased potters compiled from the seventy-one genealogical diagrams collected in 1984. In this database, each potter (or learner) has a record. These potters include active and inactive potters, potters learning the craft in 1984, and those elicited from informants to the limits of their memory; sometimes this memory included deceased individuals living as much as four or five generations previously. Each record consists of twenty-nine fields that include the potter’s name, gender, relative age, marital status, and the years that the potter was active. In addition, this database contains fields that indicate the type of individual from whom the potter learned, why they learned, and the intergenerational learning lineage (e.g., FaFaMoFa). Fields also include the name of the household in which the potter lived, the type of household in which he or she lived (e.g., Fa, MoBr), names of other potters in the extended family, and the kind of fabrication technique that the potter used. This database includes distance data (in blocks) that consist of the distance of a potter’s residence to that of the individual from whom a potter learned the craft, the distance from a potter’s residence to his/her father, and the distance of a potter’s residence to the nearest pottery-making family. A parallel set of fields was set up with distances to their work location. Several fields focus on finding selective factors that affected the potters’ residence, work location, and the perpetuation of the craft. During the preparation of the kin diagrams in the field, for example, I noticed that the children of single mothers seemed to become potters more often than children in other potters’ households. So I added a field in this database to flag single mothers and the type of single mother (e.g., unmarried, widowed, or abandoned). A related field includes the percentage of a potter’s children who learned the craft, the percentage of children who became potters, and the percentage of one’s children who did not become potters and why. A final field was added that listed the reason why a potter had abandoned the craft if he or she was no longer active. 26

Introduction

The principal limitation of this database is that it does not contain much information from 1988 and 1994 and no data from 1997, because the fieldwork since that time was not primarily oriented to obtaining information in the fields in this database. It thus does not include individuals who have become potters since 1984. The Plan of the Book The Social Dimension of Production

The first section of the book focuses on the social dimension of production and emphasizes the social embeddedness (Sillar and Tite 2000) of technology, distribution, and agency (Dobres and Hoffman 1994). Although apparently new to archaeology and materials science, these perspectives are rather obvious to those of us who are ethnographers and who have spent all of our professional lives focusing on the ethnography of technology. We understand that technology is produced by human beings who creatively adjust to environmental, social, and infrastructural circumstances to solve problems, make choices, and interact with one another to pass on their knowledge. This knowledge is cognitive, but it is also embodied in the muscle patterns that may, or may not be, conscious. Further, ceramic technology is community-based within a population of potters that differs from that of other populations of potters nearby (D. Arnold 1971, 1978a, 1978b, 1991, 1993:9–12, 233–236, 2005a). This section emphasizes how the social dimension of production plays a significant role in the craft apart from the engagement of the potter with socially embedded knowledge, tools, and techniques. The first chapter of this section (Chapter 2) focuses on the population of potters, details its composition and change from 1965 to 1997, and shows how the craft is passed down from generation to generation. It describes those mechanisms that operate over time and select for, or against, certain members of the population and causes them to aggregate into larger production units. Besides production, distribution is a critical feature of any ceramic production system and also has a significant social component. Distribution consists of two different dimensions. First, demand is critical because producers must exchange their products for food or some commodity (like money) that can be readily exchanged for food. Consumers must desire the pots, and demand thus relates closely to the values of the consuming population. Like production itself, demand is also socially embedded and socially embodied because pottery shapes must be congruent with the values and the habitus of the consuming population, which includes its motor habits, carrying patterns, and furniture configurations 27

Introduction

(D. Arnold 1985:151–166). In some contexts, however, pluralistic populations of consumers may have varying uses for pottery vessels, and potters must respond to the container desires of these varying constituents. Chapter 3 thus details the demand of different consuming populations and how that demand has changed over time. In addition to the demand for ceramic vessels, distribution involves the way in which potters place their pottery into the hands of consumers. Although demand is embedded in human populations according to the values and motor habits of the consumers, actual distribution of pottery is embedded in a different way and often on a different level and is closely related to the economic infrastructure, such as transportation and trade networks. This infrastructure requires that the potters develop strategies to successfully market their wares. Chapter 4 thus details the methods of distribution by which potters meet the demand and how this system has changed since 1965. The Production Sequence

The second major section of this book describes the change in each of the major segments of the behavioral chain (Schiffer 1975) of pottery production through time. Probably one of the most fundamental problems with using ceramics in archaeology is failure to understand how the ceramic production sequence is limited by the molecular structure of clay minerals and how the embodiment of the craft comes from the potters’ engagement of their learned patterns with the visual, tactile, and aural feedback from the production process. Because many of the technological realities of the behavioral chain are isomorphic across time and space, the five chapters in the second section of the book are organized around that sequence. No matter when or where production occurs, potters must first procure clays, prepare the paste, form the vessels, dry them, and then fire them—in that sequence. For archaeologists without experience in real-world ceramic production, however, the emphasis on the behavioral chain is an important reminder that ceramic production is not totally responsive to culture but has important constraints because of the nature of clay minerals and the limits that those minerals place on the production sequence. The basic behavioral chain of making pottery is thus culturally universal and accounts for some of the cross-cultural regularities of ceramic production across time and space. Understanding this sequence permits inferences about ceramic production in the remote past that are based on processual analogies derived from the fundamental chemical processes of pottery production (D. Arnold 1985). In the study presented here, it is possible to see the changes in different parts of the ceramic production sequence that have relevance to the past and those that do not. 28

Introduction

Chapter 5 describes the changes in clay procurement and shows how these changes relate to landownership, micro-political factors, and procurement organization. Then Chapter 6 does the same with temper procurement. Chapter 7 focuses on the changes in the preparation of the paste and assesses whether the changes in procurement in the clay and temper can be seen in the chemical composition of pottery. Chapter 8 details the changes in the forming technology and evaluates theories of how technological change affects the fabrication technology. This chapter concludes with an evaluation of the oft-repeated assumption that efficiency is the driver of technological changes of independent specialists. Chapter 9 examines the firing technology and its changes through time. It concludes with another evaluation of the role of efficiency in changes in this technology over time. Finally, the concluding chapter of the book (Chapter 10) examines the larger picture of change in the system of production and distribution of pottery in Ticul from 1965 to 1997. First, it answers the questions posed at the beginning of this book by providing a summary of the relationship of pottery to the nonmaterial social world. Second, it describes the relationship of time and change by evaluating those aspects of production and distribution that change most rapidly and those that do not. Finally, the conclusion answers the question “What does pottery tell archaeologists about social change?” and elucidates the insights that the book provides that are relevant to archaeological interpretation. It also reviews the contributions of this work to the themes of the development of craft specialization and use of ceramics as a surrogate index of social change. Finally, the chapter summarizes the processes responsible for these changes and the contributions that the data can make to understanding the development of ceramic specialization in antiquity.

29

Chapter

How Have the Population and Organization of Potters Changed?

two

F

rom 1965 to 1997, Ticul has undergone many social changes. The population has almost doubled (Figure 2.1) and the transportation infrastructure has expanded, facilitating travel to Mérida and to communities in the interior of the peninsula. In 1965, travel to villages in the interior of Yucatán was difficult and lengthy given the geographical distance and quality of the roads. By 1997, many of the formerly unimproved roads were asphalt, which facilitated easy and quick access to these communities. During the same period, the Mexican government invested in additional infrastructure, such as schools, piped potable water, and electricity. In 1965, Yucatán was isolated from the Mexican heartland, and its citizens greatly valued their independent ethnic and regional identity. Although they recognized that they were part of the Mexican republic, they did not have high regard for their compatriots from the heartland, calling them huachis, a word that mimics the sound made by the sandals that Mexican troops had worn during 31

How Have the Population and Organization of Potters Changed?

Figure 2.1. Trend line showing the exponential population growth in the municipio

of Ticul, 1950–1990 (data from INEGI 1996:13). The municipio includes the rural population as well as the villages of Yotholin and Pustunich, but 86 percent of the 1990 population was concentrated in the city of Ticul. No data on percentages of population in Ticul are available for other years.

their marches through their communities decades previously. By way of contrast, they had a great affection for gringos from the United States, which reflected their geographical isolation from central Mexico, their independence as a republic during the first half of the nineteenth century, and their desire to come under the sovereignty of the United States (as a state) in 1848 (Orosa 1994:157–167, 178). By 1984, a greater Mexican presence was evident with more government offices and more people from highland Mexico. Economically, Yucatán’s most visible industry in 1965 was the growing and processing of henequen for the manufacture of rope and twine. Outside of the major urban areas, much of the population was either employed in the henequen agribusiness or depended on slash-and-burn agriculture for subsistence. In the mid-1970s, the construction of Cancún initiated a massive development of resort communities along the east coast of the peninsula. This tourist infrastructure became so extensive, and the Caribbean coast so alluring, that the region became known as the Maya Riviera and emerged as a valued vacation destination for North Americans and Europeans. From 1988 onward, large multinational corporations invested in hotel, restaurant, and other retail chains so that by 1994, Mérida had its first suburban 32

How Have the Population and Organization of Potters Changed?

air-conditioned indoor shopping center complete with parking garage. A massive Carrefor (called superlatively a “hypermarket”) had also been built that included a bank, fast-food court, and other amenities that would dwarf any Wal-Mart. In the northern part of Mérida, the road to Progreso passed a Burger King, a McDonald’s, a TGI Friday’s, and a KFC, all of which did not exist in 1984. Along with these changes, the henequen industry in Yucatán had largely disappeared. In 1965, Yucatec Maya was widely spoken in Yucatán. Although many people knew at least some Spanish, Maya was the language of choice. By 1997, Spanish became more dominant than Yucatec Maya, and the young had more education and more choices in employment. These transformations can be illustrated in the lives of two potters whose histories I have followed closely since 1965. Both were born in the 1920s, and each went to school for two years or less. They both knew enough Spanish to communicate well, but their language of choice was Yucatec Maya. Together they had ten children who learned Spanish in school and only understood, but did not speak, Yucatec Maya. None of them followed in their fathers’ footsteps as potters, instead becoming professionals such as teachers, technicians, accountants, and secretaries. It may seem that these social changes would have caused traditional pottery making to disappear. On the contrary, the craft has evolved and the population of potters has adapted to these changes. The Social Context The social context of production exists at different levels. The most inclusive level is the region and in this case consists of the population living in the northern part of the Yucatán peninsula. It includes both producers and consumers of pottery. Within this region, Ticul is only one of several pottery-producing communities (Figure 2.2). The second level of the social context of production consists of the local population of which the potters are a part. In this case, the local population consists of the city of Ticul, one of the largest cities in the southern part of the state of Yucatán (Figure 2.2). It is the administrative center of the municipality that includes the rural areas around it. Although Ticul was formally designated a city as late as 1867, a settlement has existed in this location since pre-Hispanic times. Ticul was mentioned in the pre-Conquest narrative Book of Chilam Balam of Chumayel (Roys 1933:70– 73), and a large archaeological site dating to the Terminal Classic period (A.D. 800–1000) lies just north of the city (Stephens 1996 [1843]:71–77; Velázquez 33

How Have the Population and Organization of Potters Changed?

Figure 2.2. Map of Yucatán showing major cities, towns, archaeological sites, and pot-

tery-making communities between the late 1960s and 1994. (Map drawn by George Pierce)

and López de la Rosa 1988). Smaller sites of this same period also exist nearby (Brainerd 1958). The Population of Potters

The third level of the social context of production consists of the actual population of pottery producers. This population is comparable to the local community level of Kolb and Snead (1997), a socially constituted community (Yaeger and Canuto 2000), and embodies a “community of practice.” As is true of potters elsewhere in the world, the potters in Ticul use different semantic categories of raw materials (Arnold 1971) and have different practices of paste preparation (Arnold 2000) than other populations of potters in Yucatán (see also Thompson 1958). Communities elsewhere that produce decorated pottery may have design structures that differ from one another even though the paint colors and design elements may be the same (Arnold 1993:140–196). The population of potters has ancient roots. The semantic structure of raw material acquisition (Arnold 1971), potters’ choices of raw materials, and the

34

How Have the Population and Organization of Potters Changed?

Figure 2.3. Trend line for the total number of potters in each observation period from 1965 to 1997. (This graph includes potters working at Uxmal; see Table 2.1.)

potters’ sense of place associated with sources of raw materials suggest that pottery production dates at least to the Terminal Classic period (A.D. 800–1000). The clay mine (at Hacienda Yo’ K’at) was used during the same period (Arnold and Bohor 1977), and Terminal Classic occupation also exists on top of a portion of the temper source at Yo’ Sah Kab (Arnold 2005b). During the last half of the twentieth century, Ticul had the largest number of potters in the northern Yucatán peninsula and was the most important producer of pottery in Yucatán. Although the number of potters increased from 1965 to 1997 (Table 2.1; Figure 2.3), the growth did not match the exponential growth of the population of the municipality of Ticul during this time (Figure 2.1). This growth of the population of potters in Ticul contrasts with the decline of other communities of potters elsewhere in Yucatán. In 1951, potters existed in Tepakán, Mama, Becal, Maxcanú, and Uayma (Thompson 1958) and were also reported in Akil (Steggerda 1943:232), Tizimín, Chikindzonot, and Duzununcán (Thompson 1958:13). By 1968, potters had ceased to exist in Becal, and the potters in Maxcanú were few and were getting old. Potters still remained in Tepakán and Mama, but only three potters were found in Akil. Three potters also had worked in Peto, but they said that they had abandoned the craft in 1962. By 1978, potters still remained in Uayma, Maxcanú, Akil, and Mama (Terán 1981:17), and by 1987–1988, potters still worked in Tepakán and Maxcanú and one potter worked in Becal (Varela Torrecilla 1990:193–201). In 1994, many potters still

35

How Have the Population and Organization of Potters Changed?

worked in Tepakán, although they were far fewer than those in Ticul. Only two potters were found in Mama, and only one was located in Akil. A similar kind of consolidation also occurred among the Kalinga of the Philippines (M. Stark 1994) and the Ibibio of Nigeria (1981:177). Infra-population Organization

Little significant organization exists between the level of the population of potters and the individual household. Many potters are related by kinship, and although there is a sense of kin-relatedness, familiarity, and some trust between members of an extended family, no corporate kin group exists above the level of the household. This pattern also exists elsewhere in Mesoamerica and may have existed in antiquity as well (Gillespie 2000). Social Class. Pottery making is one of the most lowly occupations in Ticul. Only the traditional milpero (corn farmer) and workers in the henequen fields ranked lower on the occupational prestige hierarchy in the 1960s (Thompson 1974:122–123). Potters’ social position was reinforced by their language and clothing. In 1965, all potters spoke Yucatec Maya and many had limited knowledge of Spanish. Most dressed like traditional Maya peasants. Potters’ social position has changed since 1965 as the craft has attracted non-Maya-speaking entrepreneurs. A few of these entrepreneurs were painting artisans, but none were traditional potters. These entrepreneurs bought fired vessels from potters or hired potters in their own production facilities to make the vessels that they painted. Although the social position of some potters has risen, pottery making still appears to be a lower-class activity even though some potters have become wealthy by Ticul standards. Recognizing economic disparities and the poor condition of many potters, the government has made repeated attempts over the years to organize the potters into workshops and/or cooperatives above the level of the individual production unit. All of these attempts, however, have failed. The Potters’ Gremio. The only corporate group and sense of organization that does exist above the level of the household is the potters’ gremio, one of many semi-religious occupational groups in Ticul organized to honor the blackened and blistered Holy Christ of the Blisters, a crucifix that survived a devastating fire in the town of Ichmul (near Peto) during the seventeenth century (Arnold 2006; Fernández and Negore 1994). Few data exist about this organization through time (Rendón 1947; R. Thompson 1974), but it appears to have changed little since 1965 except for the 36

How Have the Population and Organization of Potters Changed?

officers elected at the end of each year’s festivities. Their responsibilities include planning and organizing the annual gremio festivities (October 12) and raising the funds to finance these activities (Arnold 2006). Gremio membership is voluntary and based on those who contribute financially to meet the expenses of gremio festivities. The gremio organization has little power or influence beyond the religious celebration, but on occasion, the president of the organization is called on to represent the potters in times of crisis, as he did in the conflict over clay procurement in 1988 (see p. 162). Changing Production Organization The Evolution of Full-time Specialists

One of the principal ways in which the social context of pottery production has changed is the increase in its intensity from part-time to full-time specialists. Up until the 1960s, many potters were part-time specialists and cultivated maize using slash-and-burn agriculture. Slash-and-burn agriculture was compatible with pottery making for four reasons. First, it provided potters with their basic subsistence crop, maize, and provided a significant buffer from the vicissitudes of the demand for pottery because a potter could always feed his family from his maize plot. Second, it provided potters with a by-product that was crucial for their craft: fuel for firing. Since wood is left in the field after burning, each trip to the field for clearing, planting, and weeding was used to transport one or two bundles of firewood back to the household. Third, the amount of labor required by slash-and-burn agriculture allows the potter to grow maize and still make pottery on a part-time basis. In a study of five different towns north and east of Ticul, Steggerda (1943:125–126) found that only 190 eight-hour days were required to cultivate the average field of 39,692 square meters (99.23 mecates) to feed a Maya family. This amount of time, however, may be excessive for the Ticul area because south of the nearby hill (puuc) ridge, the soils are more fertile and potters reported that crop returns were doubled compared to those grown on land closer to Ticul. A subsistence agriculturalist thus could cut his agricultural labor in half by cultivating a field in this more productive area. Fourth, activities such as cutting the forest, burning, planting, cultivating, and harvesting can be scheduled so that they can complement, rather than compete with, pottery-making activities. First, the potter can work in his swidden plot during the early morning when fog and moisture may damage newly formed pottery and slow its drying (Arnold 1985:66–99). Then, after returning to his house in the late morning, he can make pottery when sunshine and heat are 37

How Have the Population and Organization of Potters Changed?

required to dry clay, dry pottery, and fire. Second, the scheduling of activities for slash-and-burn agriculture is flexible. Cutting and clearing the forest must be done early enough in the agricultural year so that the cleared field can be burned before the rainy season starts in late May or early June; planting needs to be done immediately before or immediately after the rains start. Weeding usually occurs at least four times during the growing season and its scheduling is also flexible. Finally, harvesting occurs after the rainy season passes when the ears are dry and can be stored without spoiling. The importance of slash-and-burn agriculture for potters has declined greatly since 1965 as the craft has become increasingly full-time. In 1965 and 1966, many swidden agriculturalists still existed in Ticul, but even with the complementary nature of swidden agriculture and pottery making, only a few potters planted maize plots. Nevertheless, two men who knew how to make pottery preferred cultivating maize to making pottery and devoted themselves entirely to agriculture. Swidden agriculture was also carried out by three part-time clay miners and by one part-time temper miner. Although the clay miners knew how to make pottery, they preferred subsistence agriculture to making pottery; the temper miner, however, had limited knowledge of the craft. By 1984, potters’ use of slash-and-burn agriculture had declined greatly and only one elderly potter practiced it. By 1997, however, one potter recognized the benefits of slash-and-burn agriculture for resource procurement and cleared a forested plot at the temper mines. He used the wood from clearing for firing and then planted maize. Finally, he dug a mine in the field to obtain temper. The Social Units of Production

The smallest unit of production organization consists of a group of cooperating potters who share facilities at a specific physical location (usually a houselot) where pottery is actually produced. Such groups (called “production units” here) have undergone great changes in number, size, composition, and spatial organization. In 1965, all production units (N = 29) except one consisted of households in which the members were related by descent, co-descent, and marriage. Household members slept in one house and cooked in a smaller house to the rear. Some households consisted of one extended family made up of multiple nuclear families in a “resident corporate group” (Hayden and Cannon 1982), and in some of these households, each nuclear family had its own house for sleeping. Each nuclear family that made pottery, however, did so in its own house and controlled its own production, but it usually shared the use of the kiln with others in the houselot. Physically separate households sometimes shared the use of a kiln. 38

How Have the Population and Organization of Potters Changed?

Table 2.1. Population of potters in Ticul, 1965–1997. The totals do not include minor children unless they are considered knowledgeable potters. The identification of a potter was by the potters themselves, who classified those who knew how to fabricate pottery. Usually, these individuals were adults. Total Potters (including Uxmal) Potters working at Uxmal Hacienda Uxmal Hotel Principe Total potters working in Ticul Number of female potters Percentage of female potters (N = All potters including Uxmal) Total number of production units (N = All potters including Uxmal) Total number of production units in Ticul Mean potters per unit (of total) Mean potters per unit in Ticul Median potters per unit (of total) Median potters per unit in Ticul

1965–1966 1968a 1970a 1984 1988 1994

1997

85 8 8 — 77 29 34%

29 3 3 — 26 7 24%

57 10 4 6 47 13 23%

135 — — — 135 45 33%

30

16

27

50

39

35

48

29

15

25

50

39

35

48

2.8 2.7 2.5 2

1.8 1.7 1 1

2.1 1.9 2 1

2.7 2.7 2 2

1.9 1.9 1 1

2.3 2.3 2 2

3.2 3.2 2 2

b

75 80 — — — — — — 75 80 23 12 31% 15%

153 — — — 153 25 16%

Notes: a. Data from 1968 and 1970 are incomplete because the list of potters working at the time was not collected systematically using a survey. Rather, the list was elicited from a group of several informants. Based on histories of individual production units observed and elicited separately, these lists are approximately 85 to 90 percent complete. b. Includes one unit that only makes pottery prior to the Day of the Dead rituals.

One way to describe the social context of production and its change through time encompasses Costin’s (1991:15–16) parameters of “scale” and “concentration.” To Costin, “scale” consists of the composition of the production units and the way in which labor is recruited for production. One extreme of her “scale” parameter consists of small family-based units in which labor recruitment is based on kinship. At the other extreme is large-scale industrial production where the labor recruitment is contractual and is based on skill and the availability of labor. As production units grow, Costin (1991:15–16) argues, recruitment of close kin is replaced by recruitment of more distant kin or fictive or adoptive kin, and ultimately, non-related individuals are added to the production unit. Costin’s (1991) parameter of “concentration” concerns the distribution of specialists and their spatial relationship to one another. According to Costin, one extreme of the range of this variable consists of highly dispersed production units, whereas the other extreme consists of a community of highly concentrated specialists (Costin

39

How Have the Population and Organization of Potters Changed?

Figure 2.4. Trend line showing the changes in the number of production units from 1965 to 1997. (This graph includes production units at Uxmal; see Table 2.1.)

1991). What happens to these parameters through time? Do production units grow in the way that Costin proposed that they do? Since 1965, the production units have undergone significant changes. First, the raw number of production units has increased dramatically (Table 2.1; Figure 2.4). Second, although the mean number of potters per production unit has fluctuated, a trend line reveals that the mean and median numbers of potters per production unit have increased slightly from 1965 to 1997 (Figure 2.5) but still reveal a small number of potters in most production units. This small size is also reflected in the distribution of the number of potters per production unit in 1965–1966, 1984, and 1997 (Figure 2.6). Although a few production units have gotten larger, the size of most units remained from one to three potters throughout the period. Forces of Social Continuity The factors that have affected the population of potters from 1965 to 1997 reflect a blend of the forces of continuity and the forces of change. Continuity involves the successful transmission and reproduction of cultural information from generation to generation. The principal way in which this information is reproduced involves learning. Understanding learning processes, then, is critical 40

Figure 2.5. Trend lines showing the changes in the mean and median number of potters per production unit in Ticul from 1965 to 1997. (Data from Uxmal not included; see Table 2.1.)

Figure 2.6. Potters per production unit in 1965–1966, 1984, and 1997 (see Table 2.1).

How Have the Population and Organization of Potters Changed?

to understanding the social continuity of pottery production and its change from generation to generation. Shennan (2000) called this process of culture change over time “descent with modification,” and although the application of his evolutionary analogy to cultural behavior is controversial and challenged by terminological nit-picking and hair-splitting, evolutionary theory does provide one way for archaeologists to understand cultural change through time. Challenges to the evolutionary analogy are described elsewhere in the literature, but changes in the transmission of culture certainly will have implications in the next generation. Other forces, however, affect culture change, and the relationship of these forces to pottery production will be reviewed later in the chapter. Social Continuity and Learning: The Context

Traditionally, learning how to make pottery involves learning the motor habits; indigenous knowledge about clays, tempers, and firewood; and how motor habits, measurements, and firing techniques are combined to produce a wide range of vessels. This knowledge and these skills are most effectively learned during childhood while muscles and motor-habit patterns are developing (Arnold 1989a; Hayden and Cannon 1984a:328). Children’s residence in their parents’ household is long enough for them to learn all that is necessary to make pots. Information and skills (such as the motor habits that consist of the grammar of muscle use) can be reinforced during the years before adulthood. If young children begin learning the craft, they will know how to make pots by the age of ten or eleven. Learning in a household context is important for other reasons. Skilled potters can support learners economically when the economic returns produced by neophytes may not be viable or sufficient for subsistence. Learning the craft as a child is also efficient because learning to make pots does not compete with activities for subsistence as it does when adults learn the craft. On the other hand, having children learn the craft creates risks of damaged or poorly made pottery, but children reside in a household for reasons other than economic ones, and damaged pottery can be tolerated because the long-term goal of making pots outweighs short-term losses (Arnold 1989a). Although children may learn how to make pottery from anyone in the household, they usually learn from their father. Even with different types of kin listed as sources of learning, 35 percent of the active potters in 1984 learned the craft from their fathers (Arnold 1989a; Table 2.2). More often, however, any member of the household can be the source of learning, and 81 percent of the potters learned from their immediate family (i.e., father, mother, husband, and wife). Most of the 42

Workshop Workshop Step Fa Total

Workshop

1 4 80

0.7 2.9 57.9

1 0 58

Note: Br = brother; Ch = children; Da = daughter; Fa = father; Hu = husband; Mo = mother; Si = sister; So = son; Wi = wife.

Not related No data Total

0.7 0 42

2 4 138

1.4 2.9 100

26.8 37

23 2 1 1 1 1 29

16.6 1.4 0.7 0.7 0.7 0.7 21 Hu Hu HuBr HuFa HuFa HuMo HuBr Hu (HuFaBr) ? Hu Fa Hu HuFaFaSiSoDa Total

7.2 0.7 7.9

Total %

10 1 11

Total N

0 7.2 0 0.7 0 Total 7.2 0

%

35.5 12.3 5.7 1.4 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7 60.8

N 49 17 8 2 1 1 1 1 1 1 1 1 84

Females learned from:

Fa 23.1 17 12.3 5.7 Mo 6.5 8 1.4 Fa and Mo 4.3 2 Mo MoBr 0.7 1 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7 20.2 Total 40.5 28

%

2.9 Postnuptial patterns Wi 4 1 BrWi 0.7 SiHu 1 0.7 0.7 Wi WiBr 1 0.7 WiMo WiMoBr 1 5.7 Total 8

10 1 11

Fa 32 Mo 9 6 Fa and Mo Mo MoBr 1 FaBr 1 1 FaFa FaMoBr 1 FaSi FaBrSo Not Related 1 Fa and Mo MoSi (workshop) 1 Fa and Mo MoSi Wi 1 Mo FaMo 1 MoBr 1 Total 56

Natal family patterns

N

Males learned from:



Table 2.2. Learning styles of active male and female potters in 1984 derived from kinship diagrams elicited from genealogical data. Some potters learned from more than one individual and they are listed on the same line with a space between. The total number is 138 potters (100%).

How Have the Population and Organization of Potters Changed?

Figure 2.7. Trend lines showing the number of the most common kin working as potters

in production units from 1965 to 1997. (Based on data in Table 2.4)

remaining 19 percent learned the craft from other relatives who lived in the same household. Some learned from more than one person (Arnold 1989a; Table 2.2). The importance of the household as the social unit of craft transmission is also reflected in the actual composition of the production units from 1965 to 1997. The predominant types of production-unit personnel reveal that potters largely were sons, wives, daughters, or mothers of production-unit owners (Figure 2.7). These data indicate that the household is the principal social unit for the transmission of the craft, and it provides a strong conservative force in the continuity of production. How Potters Learn

Some potters learned the craft on their own whereas others were taught by someone else. For those who learned on their own, visual example and imitation were important techniques, but those who learned in this way were not able to reflect much on the process. Two older potters, for example, simply said that they had the ability to look at any shape and copy it. This ability requires considerable acuity and skill in converting visual images into a sequence of behaviors to create a finished pot. Another potter said that she learned the craft by watching her mother and imitating her. Her two brothers learned the same way. Another said

44

How Have the Population and Organization of Potters Changed?

that no one taught him to make pottery, but he learned by observing his father at home. Independently, his brother volunteered that he learned in the same manner. Other potters, however, learned the craft through direct teaching in which learners were given specific tasks in each of the stages of learning the craft, instructed how to perform them, and then reprimanded when they made a mistake. These data are consistent with learning of the craft in other societies. In a sample of 100 pre-state societies, Patricia Crown found that data about learning pottery making were available only for 25, but 48 percent of these learned to make pottery by observation alone. Twenty-four percent received some verbal instruction from adults in domestic contexts, and 28 percent learned through more formal apprenticeship. Among ethnographic Pueblo societies of the Southwest, however, girls learned by observation and imitation of their mothers, aunts, grandmothers, and other adult females (Crown 2001:455). As for the age at which learning began, the ethnographic Pueblo Indians begin learning the craft as young as five or six (Crown 2001:455). Similarly, in a cross section of twenty-eight pre-state societies in the Human Relations Area Files, Crown found that learning the craft among most of these societies began in childhood, and most (27/28) learned by age sixteen, although for most of the societies the learning age was considerably younger. Similar data for learning pottery making exist for other Maya communities. In his monograph about pottery ethnoarchaeology in the Central Maya Highlands, Deal (1998:30) found that the age of learning the craft varied greatly just as in Ticul but also varied among the four communities for which he had data. Most potters learned before marriage and before the age of sixteen, but a few learned as young as six years old. In the community of Yolakitak, for example, 93 percent of the potters learned before the age of sixteen and before marriage. In the community of Chanal, however, 58 percent learned the craft after marriage at an age older than sixteen. The ages at which potters learned in the other two communities that he studied were between the extremes of Chanal and Yolakitak, with 58 percent learning before marriage in Aguacatenango and 60 percent learning before marriage in San Mateo Ixtatan. Stages of Learning. Learning the craft occurs in several stages that appear to be the same for children and adults. These stages, however, do not follow the behavioral chain of the production sequence but rather begin with tasks at the lowest skill level. Then, as skills and knowledge are developed, the learner moves to another stage that requires a higher skill level. This learning sequence, however, contrasts with the data from Crown (2001:455) in which learning largely followed the production sequence. 45

How Have the Population and Organization of Potters Changed?

Figure 2.8. Mixing paste is one of the first tasks children learn in making pottery (1984).

The first stage consists of several relatively unskilled tasks. During this stage, learners (e.g., children) are responsible for mixing the clay and temper and wedging the paste (Figure 2.8). They may also assist skilled potters by removing a vessel from the turntable for drying and replacing it with another vessel. Children are also responsible for moving pottery outside to dry in the sun and back inside again when inclement weather threatens. They also help with loading and unloading the kiln. Their assistance at this stage not only increases output by involving additional labor in the process, but also maximizes production by minimizing the amount of time that experienced potters need to spend in performing tasks at a lower skill level. This kind of task segmentation is a kind of nascent specialization. The next stage of learning, and the first experience in actually fabricating pottery, consists of making objects using vertical half molding. This technique can be learned quickly and requires little skill and little investment of learning time (Arnold 1999). The third stage involves making food bowls on the turntable. It requires more skill than molding, but compared to the complexities of forming larger traditional vessels, the knowledge and skills required are rudimentary. When food bowls are not needed, simple vessels such as ashtrays can also be made at this skill level. Learners at this stage may also slip pottery before firing.

46

How Have the Population and Organization of Potters Changed?

During the fourth stage, the learner may produce vertical-sided vessels, such as plant pots (called rectos), using the turntable. Such vessels do not require much skill because the potter does not need as much hand-eye coordination to shape the vessel on the outside with a scraper in one hand and use the other hand as a brace on the inside of the vessel. The brace on the inside must occur at precisely the point where the potter is scraping on the outside. The next stage of learning consists of making traditional multistage vessels that require careful measurements. For this stage, learners must know the measurements of the vessels to be made, the measurements of their component parts (see Tables 8.8–8.13), and how to thin the vessel walls by scraping in order to adjust vessel size and shape (see Figure 8.12). Finally, firing is a skill learned only by the most experienced potters. It is usually done by those potters who learned the craft in their natal household. Learning as an Adult. Besides learning as a child in one’s natal household, a second pattern involves learning the craft as an adult, and may occur after a non-potter marries into a pottery-making household. Adult learners, however, seldom acquire the amount of knowledge and skill as those who learned in their natal households because adults cannot take the time to learn all that is necessary to make pottery because they must perform the other tasks required to support their families. Such potters seldom are considered to be as skilled or as knowledgeable as those who have learned the craft when they were children. This pattern is reflected in the composition of the production units in that most potters are children or parents of the production owner (Figure 2.7), although 26.8 percent appear to have learned the craft after marriage (Table 2.2). Most of those from whom adults have learned the craft are from pottery-making families and have learned the craft as children. Changes in Learning Patterns

The time necessary to learn the craft has varied and has changed from 1965 to 1997, but changes in learning are more complex than just a change in the amount of time necessary to learn how to be an economically viable potter. When the repertoire of vessel shapes was more varied and the shapes were more complex, learning the craft required more time than it did after the early 1980s. In 1965, for example, one potter remarked that it took him three years to learn how to make pots. Because of the change in demand, the context of learning the craft changed between 1965 and 1997. With the development of large production units by 1984 to meet this demand, the learning context expanded to facilities outside of 47

How Have the Population and Organization of Potters Changed?

one’s own household, and adults and late adolescents were learning the craft. This change resulted in three significant modifications in learning patterns. First, those that learned pottery making outside of their own households did not learn as much as those who had learned the craft in a household context. Because of the specialization of tasks in the 1980s, potters did not need to know how to select raw materials and were dependent on the raw materials supplied by specialists. Furthermore, they did not need to know how to fire the pottery because they could work in a production unit in which someone else fired the pottery. Similarly, if potters learned the craft outside of their natal household but worked independently, they tended not to fire their pottery but sold it unfired to potters who did. Second, production after the late 1970s focused on plant pots that were easier to make than the vessels used for carrying, transporting, and serving water that were produced in the late 1960s. With water vessels, such as tinajas and cántaros, potters needed to use the proper set of measurements to form the different stages of a vessel and then align each stage to the appropriate size (see Tables 8.8–8.13). These vessels also required thin and concave vessel walls. By way of contrast, when plant pots were the predominant shape in the 1980s, less knowledge and skill were required to make them because they had simpler profiles and thicker walls than vessels made in the late 1960s. Plant pots were thus easier to make than the vessels for carrying, storing, and serving water. Third, potters who work in large production units can learn the requisite skills quickly because of task segmentation and mass production of the same forms. A laborer in a large production unit, for example, can learn how to make one kind of vessel in about a week and then continue making that same vessel day after day. But his overall skill is limited and he cannot make other vessel shapes that traditional potters learned in their natal household. As a consequence of these changes, those who have learned all of the skills of making pottery have decreased in number between 1965 and 1997. By 1997, relatively few potters remained who knew all of the skills for making pottery, from selecting raw materials to firing. Potters who learned the craft outside their own households after the late 1970s generally did not acquire this knowledge because they learned the craft in locations where the production sequence was segmented into different tasks carried out by different individual specialists (Table 2.3). The loss of the aggregate knowledge and skills is thus the result of increased specialization that organizes separate tasks and skills into a more complex production organization. This specialization has diminished the number of skills that a potter needs to participate in production and decreased the amount of time required to learn the craft. 48

How Have the Population and Organization of Potters Changed?

Table 2.3. Division of labor in the largest production unit in 1997 and the number of individuals who performed each task according to the classifications by informants. This unit has two physical locations. (Not included are four relatives who helped in the afternoons when school was not in session and during vacation.) Occupations ranked from low to high skill (1) Helpers (mix clay and slip pottery) (2) Those who use molds (3) Those who use the turntable (“potters”) (4) Painters (5) Those who fire

Number of individuals (highway unit)

Number of individuals (household unit)

1 3 7 — 1

5 — 9 4 1

Laborers who learn the craft in large production units, however, do not usually pass on the craft to their children. If their children do become potters, they learn the craft in a non-household context. Demand thus has feedback with learning skills and learning time; learning time is reduced when demand requires many vessels of the same shape and production tasks are more specialized. Social Continuity and Acquiring Household Personnel

Since the transmission of the craft from generation to generation involves learning, and learning is a social process that occurs in a social context, those factors that create and maintain that social context provide the most important means of stability in the craft. For making pottery, the traditional social context of learning in Ticul is the household. Transmission of the craft from generation to generation thus tended to coincide with the same factors that define, create, and perpetuate household composition. Processes of Personnel Acquisition

The first set of processes that contribute to the stability of production units consists of how new members are acquired. In the following discussion, the way in which individuals become members of a household is often not active, conscious, or deliberate but secondary to other factors, such as procreation and marriage. Procreation. Children who are born into a household become members of that household and often (but not always) learn to make pottery. At least some of these children eventually become adult potters. Indeed, the most frequent categories of potters in a production unit between 1965 and 1997 were sons and daughters of the owner (Figures 2.7 and 2.9), and most potters in 1984 had learned the craft from someone in the household (Table 2.2). These data suggest 49

How Have the Population and Organization of Potters Changed?

Figure 2.9. Trend lines of the changing percentages of the most common kin types working as potters in production units from 1965 to 1997. Types are defined by their relationship to the production-unit owner (see Table 2.4). All of the trend lines are weakly correlated with the data; the trend line of the wives of potters is the strongest of these and has the highest R2.

that procreation is the principal mechanism for acquisition of personnel for the craft, and this process has changed little over time. Inheritance of Household Land. Besides procreation, the second method of acquiring personnel consists of the inheritance of household land, and this pattern is also partially responsible for the composition of the production unit. Up until relatively recently, only men could inherit land. Consequently, the sons of production-unit owners constitute the highest percentage of potters in production units (Table 2.2; Figure 2.9). In order for a male to maintain his inheritance rights in his father’s household, he must reside there until his father dies. If a man has more than one son, problems may arise concerning which son inherits the father’s houselot. These problems can be alleviated if a lot has enough space to accommodate one or more new houses, and the father can subdivide his land and allocate it to the sons who remain with him. Each son, then, has his own lot on which to build his house if he did not inherit the portion with his father’s house. Household land is thus a valued possession and tends to stay within the patrilineally extended family. Household land that is not inherited patrilineally may also be purchased by collateral relatives and thus still remains in a larger patrilineally extended family. 50

How Have the Population and Organization of Potters Changed?

When my principal informant’s wife had a difficult pregnancy in 1983, he did not have the financial resources to pay the medical costs. In order to meet these costs, he sold part of his houselot to his father’s sister next door, but he had also asked relatives in his patrilineally extended family to buy it because it was part of his grandfather’s (his FaFa) land. When another potter in the same extended family needed money for a major operation in 1984, his family wanted him to sell his land to a cousin’s (MoBrDaHu) husband. When this man did not buy it and they could not find another buyer, a brother bought the land. Sales of land to relatives in times of crisis thus has the effect of keeping households of lineal and collateral relatives close to one another. One possible explanation why household land is kept within the family is that family members were formerly buried in the rear of houselots. This practice maintained a strong link with the past and was reinforced by the Day of the Dead rituals, when the spirits of the dead relatives are believed to return to the land of the living. Although informants never explicitly linked burial practices with keeping household land in the family, these practices were mutually reinforcing. Patrilineal land inheritance, the value of land, and perhaps former burial practices thus have the effect of keeping households of potters in the extended patrilineal family relatively close together. If a number of potters in a community are descendants from the same male, the effect over time is to have a concentration of pottery-making households in the same geographic area. Postnuptial Residence. The third way of acquiring personnel for production units consists of postnuptial residence behavior. From 1965 to 1970, a newly married couple lived in the household of the groom for at least several months. If the relationship between the new bride and her in-laws was good, the couple could remain permanently with the groom’s family or in a new residence on the groom’s parents’ houselot. Then, after the death of his father, the son would inherit the land. Richard Thompson (1974:32–33), an anthropologist who worked in Ticul in 1968, noted that newly married couples almost always lived with the family of the groom’s father. This pattern is not only typical of potters but was also common among the Yucatec Maya elsewhere in the northern Yucatán. It also occurred in the village of Pustunich just down the road from Ticul (Press 1975:129), in Chan Kom in eastern Yucatán (Redfield and Villa Rojas 1962 [1934]:92), and in the village of Cobá in the State of Quintana Roo (Kintz 1990:51, 59). Although a newly married couple was expected to live patrilocally at least temporarily, a new couple may also live in or near the bride’s parents’ household. This pattern was a form of bride service in the past (Kintz 1990:51). During the 51

How Have the Population and Organization of Potters Changed?

thirty-two years of this study, however, this practice occurred in only four circumstances: (1) when the bride’s father gave land to his daughter, (2) when conflict occurred between the bride and her new in-laws, (3) when the bride was treated poorly by the groom’s parents, or (4) when sickness or an accident forced the sale of house and land near the father’s home to pay medical costs. A newly married couple may move away from both parents but usually not until some time is spent living with the groom’s parents immediately after marriage. True neolocal postnuptial residence immediately after marriage was usually impossible unless one member of the couple had secure employment with a good salary (e.g., as a schoolteacher) or the couple had the financial resources to buy, rent, or construct a house. The importance of postnuptial residence patterns for acquiring personnel for making pottery can be demonstrated using two behavioral measures. Because of patrilocal residence, one expects to find that the wives of production-unit owners are potters. Although the number and percentage of wives who were potters fluctuated over time, regression lines of their number (Figure 2.7) and percentage in the total population of potters (Figure 2.9) show that the percentage declined between 1965 and 1997 (Table 2.4). Some of these women were already potters before marriage, but others learned after they married and moved into their husbands’ households. Throughout the period of this study, however, only three spouses of the children of production-unit owners were potters (Table 2.4), suggesting that postnuptial residence was not a way that production units acquired personnel to learn the craft. Second, the learning patterns also indicate that 26.8 percent of the potters in 1984 learned the craft from relatives acquired through marriage (Table 2.2). These data indicate that postnuptial residence pattern did play some role in providing personnel for learning the craft. Between 1970 and 1984, patrilocal postnuptial residence changed in a way that was more consistent with a virilocal pattern. Sons of potters were no longer bringing their wives into their fathers’ households after marriage (a patrilocal pattern) but rather were moving into a new household near their fathers’ households (a virilocal pattern). Sometimes this change resulted from the subdividing of the father’s houselot into separate houses, and sometimes the father would purchase land nearby for his sons. If a house was on the property, it was rented until the son married. If there was no house, then the father might build one for his son. In at least two cases, the father acquired land nearby for his sons because his production facilities had expanded to fill his houselot, and it could not be subdivided. A father may also buy land nearby for a daughter so that when she marries, she can bring her husband to live there. 52

Table 2.4. Inventory of the number and types of kin who are potters, 1965–1997. The kin types are reckoned from their relationships to the owner of the production unit. Type of affinal relative

1965–1966

1968

1970

1984

1988

1994

WiFaWiDa 2 2 1 DaHu 1 1 1 Wi 15 5 3 13 11 5 WiDa 2 WiMo 1 1 WiSi 1 WiSo 1 WiBr 1 SiSoWi SiDaHu 1 SiHu 1 WiBrSo SoWi 1 Total affinal relatives 22 6 6 18 13 5 Type of collateral relative 1965–1966

1968

1970

1984

1988

1994

MoFaSiSoSo BrSo 1 1 SiSoDa FaBr 1 1 FaBrSo 1 2 Si 1 2 1 MoFaWiDa FaFaBrSoSo 1 Cousin Br 3 1 1 MoFaSiSo 1 FaWiDa 1 MoBr 1 SiSo 1 0 Total collateral relatives 8 1 4 2 6 Type of lineal relative

1965–1966

1968

1970

1984

1988

1994

DaSo Mo 1 2 1 1 3 Fa 3 2 1 Da 6 3 16 5 1 ChCh 3 So 4 5 7 20 6 18 SoSo 1 Total lineal relatives 14 5 14 39 12 25 Non-kin Laborers Totals

6 50

4 15

11 32

30 91

14 41

10 46

1997 1 4

1 1 7 1997 1 2 1 2 1 1 6

1 15 1997 2 3 8 32 45 38 105

How Have the Population and Organization of Potters Changed?

Figure 2.10. Trend lines for the number of different types of relationships of potters to

production-unit owners grouped by non-kin and lineal, collateral, and affinal relatives. (Based on data in Table 2.4)

Hiring Non-household Personnel. Although household members always form the core of production personnel (Figures 2.7 and 2.9), production units may also recruit potters from outside the houselot to assist in production. Historically, this practice was temporary and occurred during peak demand. Sometimes these potters were relatives from adjacent or nearby households, but often they were not. Hiring extra-household personnel has a long history in Ticul beginning before 1965. The survey data from 1965 and 1966 indicate that except for the workshop at Uxmal, 7 percent (6/86) of the potters were working outside of their own households (Table 2.4). Three of these potters were not consistently associated with any one production unit but moved from household to household depending on the demand for their services. Indeed, one of the potters in the 1965 survey was listed as a laborer in five different households. Even in the six months between the research visits in 1965 and 1966, the composition of production units was different because laborers moved from one production unit in 1965 to another in 1966. The numbers of types of relatives grouped by their general relationship (lineal, collateral, affinal, and non-kin) to the production-unit owner indicate that the number of these types of relationships varied over the thirty-two years of this 54

How Have the Population and Organization of Potters Changed?

Figure 2.11. Trend lines for the percentages of potters related to production-unit owners

grouped by lineal, collateral, and affinal relatives. (Based on data in Table 2.4)

study (Table 2.4, Figure 2.10). The clearest trend occurred with the number of lineal kin types, which increased over time. As for the percentage of potters in each of these general relationships, lineal kin have predominated, followed by affinal and collateral kin (Figure 2.11). Trend lines of the percentages over time indicate that lineal kin have increased whereas collateral kin have remained the same or increased slightly. The clearest trend, however, indicates a dramatic drop in potters that are affinal kin of the production-unit owner (Table 2.4; Figure 2.11). The use of contractual non-household, non-kin labor also increased between 1965 and 1997 (Table 2.5). In the late 1960s and early 1970s, laborers not related to production-unit owners were not initially a significant component of production units, and those individuals in the non-relative category were largely employed by the workshops at the tourist hotels at Uxmal. Over time, however, the number and percentage of non-kin laborers increased (Table 2.5). When the potters attached to the workshops at Uxmal (from 1965 to 1970) are removed from the non-kin category and the number of wage laborers is expressed as a percentage of the total population, the percentage of potters who were wage laborers increased exponentially between 1965 and 1997 (Figure 2.12). This change first was evident in 1984, when owners of some units had hired many non-relatives to make pottery.

55

How Have the Population and Organization of Potters Changed?

Figure 2.12. Trend line of the number of wage laborers who are not relatives of produc-

tion-­unit owners from 1965 to 1997. These data do not include the laborers working in Uxmal between 1965 and 1970. (Based on data in Table 2.5.) The trend line is an exponential curve and the R2 represents the best fit of the different options. Measures of Continuity

Since learning is a social process, it requires social contact. Learning, however, takes place not only through verbal interaction but also through repeated and prolonged contact that enables learners to acquire the skills visually and by imitation. Individuals who live closer together would be expected to have more interaction than those who live more distant. The most intense social interaction obviously occurs within houselots, and it is not unusual that most of the potters learned pottery making within the nuclear family (Table 2.2; Figure 2.7). Some interaction, however, may take place between houselots. If the physically discrete houses of an extended family are contiguous, they often share a common patio to the rear of the houses. If adjacent lots belong to families that are more distantly related, a fence exists between them and limits the social interaction. Non-contiguous households, however, may have no social contact whatsoever. Distances between households thus provide a plausible surrogate measure of social interaction. Simple straight-line distance, however, may not necessarily reflect interaction frequency. Two houselots that share a fence at the back of a lot, for example, would never interact because the rear portions of houselots are seldom utilized 56

How Have the Population and Organization of Potters Changed?

Table 2.5. Number and percentage of non-relative wage laborers working as potters, 1965–1997. (Differences between laborers in this table and Table 2.4 concern ambiguities in how laborers in Uxmal were counted. Some worked in Uxmal during the week and in their own production units in Ticul on the weekends.) Total potters Total laborers Percentage of laborers (N = total potters including production-unit owners) Potters working as laborers in Uxmal Percentage of total potters working in Uxmal Number of laborers working in Ticul Percentage of total potters working as laborers in Ticul

1965–1966

1968

1970

1984

1988

1994 1997

85 15 18%

29 4 14%

57 11 19%

135 29 21%

75 14 19%

80 10 13%

153 39 25%

8

3

9

—

—

—

—

9%

10%

16%

—

—

—

—

7

1

2

29

14

10

39

8%

3%

4%

21%

19%

13%

25%

for anything other than latrines, and no interaction occurs between households in such private locations. If the members of such contiguous households ever interacted socially, they would have to travel around the block to the front of the lot. Consequently, simple straight-line distance between households is not a valid surrogate measure of social interaction except for adjacent lots that face the street on either side of a household. In order to measure social interaction in a way that quantified real patterns, I utilized a measure called “interaction distance,” which consists of the distance that a person in one household must travel to interact with someone in another household (Arnold 1989a). In a grid pattern such as that which exists in Ticul, interaction distance consists of the distance along a street from one household to another. First, an interaction distance value of 0.1 block was assigned to a distance between individuals who were members of the same household. Second, an interaction distance of 0.2 block was assigned to those individuals who lived across the street from one another and to those who lived in contiguous households. Third, an interaction distance greater than 0.2 block was assigned on the basis of an informed estimate of actual distances between households based on a map of the community. These estimated distances varied from the true distances by no more than 0.12 block (see Arnold 1989a for an explanation). Plotting these interaction distances by frequency revealed that 0.1 block was the most frequent distance between a potter and the person from whom he/she learned the craft and between a potter and the distance to the household of his/ 57

How Have the Population and Organization of Potters Changed?

Figure 2.13. Bar graph of the frequencies of the distances between potters active in 1984

and their fathers, teachers, and nearest production unit (N = 167).

her father (Figure 2.13). Since 0.1 block was the distance assigned to individuals living in the same household, potters most frequently learned the craft in their own household. This distance is consistent with the frequency data of learning types (Table 2.2), the patrilocal postnuptial residence pattern, and the patrilineal inheritance of land. The next most frequent inter-household distance occurred at 0.2 block and indicated that potters learned from those who lived in adjacent lots or across the street from one another. This pattern is consistent with a viri­ local postnuptial residence pattern or with patrilineal inheritance of household land if such land was subdivided after it was inherited from one’s father (Arnold 1989a). In order to formally test the hypothesis that a patrilineal/patrilocal model accounts for the learning patterns of Ticul potters, a correlation coefficient was calculated between the interaction distances of potters to their fathers and to those from whom they learned the craft. The results (R = 0.656, N = 116, significance = <0.001) reveal a strong relationship between the interaction distance to a potter’s father and to the person from whom the potter learned the craft (Figure 2.13; Arnold 1989a). Since potters’ households clustered together, residence proximity, rather than postnuptial residence, may also be a factor in learning the craft. To formally test this hypothesis, a correlation coefficient was calculated for the distance between the person from whom the potter learned the craft and that of the nearest pottery-­ 58

How Have the Population and Organization of Potters Changed?

making household (R = 0.608, N = 129, significance = <0.001). Simple residence proximity apart from land inheritance and postnuptial residence behavior, however, cannot be independently assessed. So it appears that the strength of the correlation also reflects patrilineal land inheritance and a virilocal postnuptial residence pattern (Figure 2.13). By 1997, very few married couples reportedly lived with the groom’s parents after marriage. Rather, many lived separately from both parents or with the bride’s parents. One possible reason for this change was that informants said that women now could legally inherit property, and this change provided more residence options for newly married couples. Spatial Continuity. The first way to assess the change and continuity in production-unit location was to trace each unit through the thirty-two years using the production-unit database. Then, the changes, or lack thereof, were placed into one of five categories. The first category (“same”) consisted of those units whose location did not change. The second category (“segmented”) consisted of those units that had internally segmented into different nuclear families that consisted of married children who brought their spouses to live in their parents’ houselots. The third category (“fissioned”) consisted of individuals who had been previously part of another unit, but the space there was too small to accommodate their production and they moved to another location. A fourth category (“continuing”) included those units that had made pottery during a previous survey and were missed because of a methodological bias or had temporarily stopped making pottery and then began again during a subsequent period of observation. This category also included potters who had made pottery during a previous survey but had moved to a new location for reasons other than fissioning of households. A fifth category (“new”) consisted of new units that had been established since the previous survey. When all production units were tracked, compared, counted, and graphed into these five categories, several patterns emerged. First, comparing the 1984 and 1997 surveys with 1965–1966 data, great household continuity occurred throughout the thirty-two-year period of this study (Figure 2.14). In 1984, 35.5 percent of the production units were located in exactly the same position as those in 1965–1966, and 29 percent were units derived from existing production units through segmentation and fissioning. By 1997, 26.7 percent of the production units were in the same location as they were in 1965, and another 33.3 percent were derived from those units through segmentation and fissioning. The development of new production units is also evident in these data (Figure 2.14). In 1984, 36 percent of the production units were new without 59

How Have the Population and Organization of Potters Changed?

Figure 2.14. Bar graph summarizing the changes in the locations in production units

in Ticul since 1965.

any continuity with units in 1965–1966. By 1997, the number of totally new production units had increased in absolute number and in the percentage (40 percent) of the total. These new units were created by entrepreneurs who had established production facilities to take advantage of the tourist market. They were usually painters, not traditional potters, and may have learned to make some pottery themselves, but they usually hired potters as wage laborers to make it and/or bought fired pottery from traditional potters for painting. When one looks at the changes in location relative to each successive period of observation, a slightly different picture emerges (Figure 2.15). The data still reveal a dramatic increase in new units in 1984 but also indicate a strong continuity of production units from 1965–1966 and a diminishing number of new production units since 1984. These new units show the development of the entrepreneurial workshops that were established in Ticul in the 1970s and yet demonstrate the continuity of production units of traditional potters. These patterns also suggest that most of the growth in the number of production units comes from new units rather than from segmentation or the fissioning of traditional units. The second way to trace production-unit location and continuity is to show the location of production units spatially in a map for each period of observation (Figures 2.16a–g). These maps document the clustering of households caused 60

How Have the Population and Organization of Potters Changed?

Figure 2.15. Bar graph summarizing the changes of production units from 1970 to 1997 compared to their location in the previous survey. The 1970 data are compared to the 1965–1966 data.

largely by patrilineal land inheritance, virilocal residence, and the desire of some potters to keep their children close to home. They also document the increased dispersion of production units over time created by the new potters who have learned the craft since 1970. Kinship Continuity. The population of potters in Ticul is largely kin-based and largely (but not exclusively) consists of members of eleven extended families whose ancestors can be traced four to six generations into the past. Six of these families are represented by more than one production unit. Did this strong kin emphasis change over time? One way to express the kinrelatedness of potters through time is to plot the number of production-unit owners related by kinship. When these data were grouped and graphed at the beginning of this study (1965–1966) and in 1997 (Figure 2.17), they revealed a very strong continuity of the kin-relatedness of production-unit owners, both across the entire population of potters and through the entire period of this study. This strong kin-relatedness indicates great social continuity in the craft between 1965 and 1997.

61

Figure 2.16a. Map showing the locations of potters’ production units in 1965–1966.

(Map drawn by George Pierce)

Figure 2.16b. Map showing the locations of potters’ production units in 1968. (Map drawn by George Pierce)

Figure 2.16c. Map showing the locations of potters’ production units in 1970. (Map drawn by George Pierce)

Figure 2.16d. Map showing the locations of potters’ production units in 1984. (Map drawn by George Pierce)

Figure 2.16e. Map showing the locations of potters’ production units in 1988. (Map drawn by George Pierce)

Figure 2.16f. Map showing the locations of potters’ production units in 1994. (Map

drawn by George Pierce)

How Have the Population and Organization of Potters Changed?

Figure 2.16g. Map showing the locations of potters’ production units in 1997. (Map drawn by George Pierce)

Forces of Social Change In contrast to the forces of continuity, the forces of change show how procreation and the patterns of inheritance and residence are modified and limit the learning of the craft. Although acquiring personnel for the production units appears to follow largely a kin-based model of procreation, patrilineal land inheritance, and virilocal postnuptial residence, learning the craft does not itself obey these patterns of household composition, perpetuation, and locus, but these patterns merely provide the personnel for that learning. It does not mean that all such personnel will become potters. Although the transmission of the craft to a new generation can be described by these patterns, it does not ensure that all eligible household members will become potters. Not all children of potters learn how to make pottery, and not all of those who learn the craft actually become potters. In reality, only a fraction of those children raised in a potter’s household end up practicing the craft as adults. If all the children of potters became potters in the next generation, the number of potters would grow at the same rate as the population (Figure 2.1). The population of potters, however, has grown much more slowly and sporadically and is subject to fluctuations in the demand for pot65

How Have the Population and Organization of Potters Changed?

Figure 2.17. The amount of kin-relatedness among production unit owners in 1965– 1966 and 1997 based on traceable relationships. Two owners were considered to be kinrelated if their relationship was traceable through descent or co-descent and/or across no more than one marriage. Relationships acquired through certain kinds of affinal ties were not considered to be kin-related, specifically those through marriages of affinal relatives and second marriages (or cohabitation after a first marriage) when children did not result from such unions.

tery, but shows a trend in growth over time (Figure 2.3). Procreation, inheritance rules, and residence rules produce a potential pool of learners, but these patterns in themselves do not ensure that all household members will learn the craft and become potters. Patterns of the acquisition of household personnel and their variants thus partially describe but do not explain the perpetuation of the craft. Why do some children of potters become potters and other children do not? Why did some non-potters who married potters become potters and others did not? The answer to these questions involves a series of mediating variables that both select for and select against the learning of the craft and also influence its perpetuation from generation to generation. These mediating variables constrain the growth of the population of potters over time when procreation and patterns of inheritance and postnuptial residence alone are insufficient to explain the perpetuation of the craft. These factors also affect the size and number of production units and are much more complicated than the behavioral patterns of land inheritance and postnuptial residence.

66

How Have the Population and Organization of Potters Changed?

This approach to the changes in the population of potters reflects Shennan’s (2000) use of the Darwinian “descent with modification.” Although one can quibble with whether Shennan’s notions precisely parallel biological evolution, as some of the comments in his Current Anthropology article indicate, they do express an important point that, since culture is transmitted by learning, modifications in this process can produce cultural change. Large-scale Social, Political, and Environmental Forces

Probably the most important factors that exert selective pressure on the population of potters consist of the political, social, and environmental forces from the region and/or the nation. These factors include state and national laws, national policies concerning labor and capital, large-scale conflicts, and epidemics. These factors affect all of the population, not just the potters, but can have a dramatic selective pressure on pottery making by removing potters from production through disease, death, or military conscription or by requiring them to work in a non-pottery-making capacity. The first recorded large-scale selective factor that affected the population was disease. Dumond and Dumond (1982) have synthesized a massive amount of primary source information about the demography of Yucatán in the nineteenth century and provided some information on the effect of disease on the population of Yucatán. From these data, mortality from disease was widespread in Yucatán communities during the nineteenth century. A document from Ticul dated June 30, 1836, provides an “[a]ccount of those who have died in this parish of epidemics, infants as well as adults, from the 1st of February of this year of 1826 in which measles began, to the thirtieth of June of the same year, notwithstanding that smallpox continues in great force” (Dumond and Dumond 1982:353). During these five months, 1,142 individuals died from smallpox and measles combined (Dumond and Dumond 1982:353). Besides disease, the War of the Castes also exerted selective pressure on potters during the late nineteenth and early twentieth centuries. Although this conflict began in the 1840s and ended in 1901 (Dumond 1997; Güémez Pineda 1997; Reed 1964), its most significant effect on potters occurred in 1848 when rebel forces overran Ticul and advanced to Sacalum. Before this event, potters’ oral history indicates that fear motivated the ancestors of some of the potters to escape the rebels and seek shelter in Ticul. The third factor that provided major selective pressure occurred during what potters call the Epoch of Slavery, which began with the presidency of Porfirio Díaz in late November of 1876 and continued through 1915 when the Mexican Revolution arrived in Yucatán. This period was characterized by abuses of the 67

How Have the Population and Organization of Potters Changed?

peasants because of forced peonage labor. Originally established to produce cattle and maize during the Colonial period, this feudal system changed in the nineteenth century to growing sugarcane in the southern portion of Yucatán (Meyers 2004; Rejón 1981, 1993) and cultivating a species of agave cactus called henequen, which was grown for the fibers extracted from its leaves, in the north (Dumond 1997:351–352). Large tracts of land were planted with this crop, and during the late nineteenth and early twentieth centuries, massive amounts of the fiber were exported to the United States, where it was twisted into ropes and twine for maritime and agricultural uses (Baklanoff 1980:208; Dumond 1997:351). Henequen production required a substantial amount of infrastructure for its success. First, it required large amounts of land to grow the henequen plants. Second, large amounts of capital were necessary to buy the machinery to transport and crush the leaves and remove the fibers. Capital was also needed to build structures to house the processing machinery, for quartering the horses and mules used for transporting the leaves along the narrow-gauge tramway (the tranvía), and for storing the bales of processed fibers (Baklanoff 1980). Last, the henequen industry required large amounts of labor for planting, weeding, cutting, transporting, and processing the henequen leaves and for drying and packing the processed fibers. Workers weeded the fields, cut the leaves, loaded them on the tramway platforms, and then used mules to pull these platforms to the crushing machinery. Workers operated the machinery, loaded the processed fibers onto the platforms, and then drove the mules to transport the platforms to the drying area. Labor was also required to cut wood for the boiler that provided steam for the engine that powered the machinery. The forced labor on the haciendas created much wealth during the late nineteenth and early twentieth centuries and fueled a glittering opulence that is still evident in the grand houses along the Paseo de Montejo and Avenida Colón in Mérida, the capital city of the State of Yucatán (Rasmusen 1994). Henequen production, however, was built on the backs of thousands of peasants. Work was hard; foremen and hacienda owners were demanding, and working conditions were reportedly the worst in the Mexican republic (Meyer et al. 1999:447). At one hacienda near Ticul (Yo’ K’at), informants reported that stocks were used to punish those who disobeyed the manager’s orders. The excesses of this era were not just restricted to the henequen haciendas. At the finca of Ya’ax Che’ near Bolonchén, workers had to pay the owner a portion of their crops and the owner also determined whom one would marry. Life was so regimented that in the morning, the bell of the hacienda would ring, signaling that the men had to report to the owner’s house for their work assignment for the day. 68

How Have the Population and Organization of Potters Changed?

The fourth macro-factor that affected the population of potters took place during the Mexican Revolution. Although it began in 1910 in central Mexico, the revolution did not affect Yucatán until 1915, when locals were forcibly recruited for small reactionary bands to resist the army from central Mexico ( Joseph 1980:145). By late 1915, however, workers abandoned the oppressive sugar plantations (such as Finca Tabí; see Meyers 2004), but the peasants still had little or no land. Yucatecans probably were forcibly conscripted again prior to the military violence in 1923 and early 1924, when land reform began ( Joseph 1980:145, 153). Land reform was not widely implemented, however, until 1937, when Mexico’s President Cárdenas landed at the port of Progreso and forcibly formed communal ejidos out of the haciendas ( Joseph 1980:158–161). Apart from the general principle that such macro-factors affected the population of Yucatán, the only account of how they actually had an impact on the potters comes from an oral history of the Tzum family. During the late nineteenth and early twentieth centuries, the population of potters suffered disease and experienced dislocations because of the War of the Castes. They survived forced labor during the Epoch of Slavery and military conscription during the years after the revolution. Because of the forced labor during the Epoch of Slavery, most of the men could not make pottery because they were working long hours at the Tabí sugar plantation. During the military conscription, able-bodied men were serving in the military away from home and many of them died. In fact, very little was known about many males of the Tzum family that lived in the early twentieth century except that they were said to have “died in the revolution,” probably a vague reference to the forced military conscription and violence of 1915 and 1923. Only one man in this family, Eusevio Tzum Dzul (ca. 1869–1959), was able to survive the devastating demographic effects of the War of the Castes, the Epoch of Slavery, the epidemics of disease, and the Mexican Revolution to make pottery and pass it on to his progeny. Although Eusevio survived the smallpox epidemics in the nineteenth century, he contracted polio that left him a partial invalid. His right arm was affected and he could not extend or fully bend it. One of his leg muscles was also affected and he walked with a limp. Ironically, Eusevio’s disability permitted him to transcend the devastating effects of the Epoch of Slavery and forced military conscription; he was considered to be useless (inutil) for forced labor and fighting. As a result, he remained at home when the other boys went to the sugarcane and henequen fields with their fathers, and he learned to make pottery from his mother. He married and five of his children survived to adulthood. All learned pottery making from their father and four of them produced eighteen offspring of whom seventeen became 69

How Have the Population and Organization of Potters Changed?

potters for at least some portion of their adult life. These eighteen grandchildren produced 106 children and most of them learned how to make pottery. Of these, twenty-one made pottery at some time during their adult life. In addition to the direct biological descendants of Eusevio, many of their spouses and several non-relatives learned the craft from his descendants. Indeed, most of the potters working between 1965 and 1997 could trace their learning of the craft to Eusevio. Consequently, the transmission of the craft to the current generations of potters was the result of a “bottleneck effect,” because this one man was not subject to the demographic pressures that eliminated other men from making pottery during the late nineteenth and early twentieth centuries. For other families, the craft was practiced by women and transmitted to subsequent generations through them because they were not subject to the same selective pressures as men. Pottery-making Infrastructure

The second selective factor for the learning and perpetuation of making pottery consists of the presence of social and material infrastructures for the learning context of the craft. First, learning occurs within a social and material context that requires the physical coexistence of those who learn and those with the requisite knowledge and skill. Second, facilities must be available to store raw materials and to make, dry, and store pottery. Third, equipment such as turntables, forming tools, and a kiln must be present in order to fabricate and fire the pottery. The social and material infrastructure can affect the perpetuation of the craft both positively and negatively, and it can provide deviation-amplifying feedback (Arnold 1985:17–18) for the perpetuation of the craft. The individuals who live in households with the appropriate infrastructure have the opportunity to participate in production. Access to this infrastructure is thus one reason why the young learn the craft. A non-potter who moves into a household with such infrastructure after marriage also has an incentive to learn the craft. Pottery production thus tends to remain in households that have the infrastructure for production, and this is documented from continuity of production units over time (Figures 2.14–2.15, Figures 2.16a–g). Patrilineal inheritance of household land and postnuptial residence patterns thus not only recruit new members for the household but also provide potential learners access to the social and material infrastructure of pottery making. Second, and conversely, the lack of pottery-making infrastructure provides deviation-counteracting feedback for the learning of the craft. Without this infrastructure, it is difficult, if not impossible, to make pots. Those who move away from that material infrastructure for production usually cease to make pot70

How Have the Population and Organization of Potters Changed?

tery entirely. If they want to make pots, they must create this infrastructure using capital to obtain the tools, expand household space, and build a kiln. Storing some raw materials and making pottery does not require much capital or space, but once pottery is fabricated, drying and storing it both before and after firing increases the spatial requirements for the craft and spatially constrains production and its growth. Indeed, drying pottery requires more space than any other part of the production process, just as Kramer (1985) suggested, because of the constraints on drying pottery that result from a combination of the molecular structure of clay minerals and weather and climate (Arnold 1985:61–98). If the size of pottery that was made is small, however, the spatial constraints on production are fewer. The farther away that the potter moves from pottery-making infrastructure, the more difficult it is to produce pottery. Cessation of the craft may not be immediate but occurs eventually, certainly in the next generation. Daughters of potters who move out of their parents’ houses after marriage because of a virilocal postnuptial residence pattern thus tend not to make pottery after marriage. They may, however, make small food bowls for the Day of the Dead, sell unfired pottery, or work in their father’s production unit. But when a woman with no access to the pottery infrastructure bears children, she may even stop making pottery for the Day of the Dead. If a potter moves away from Ticul, it is unlikely that he/she will make pottery elsewhere. On occasion, a few potters have moved away and made pottery in another location using either local raw materials or those from Ticul, but these nonlocal production units do not appear to persist through time. At the most, they are limited to the lifetime of the migrant potter. In 1951, a Ticul potter (named Timoteo Cobo) had married a woman from Mama and then migrated there. He was importing raw materials from Ticul to make pottery even though Mama had its own clay and temper sources (Thompson 1958:21). Genealogical data elicited in 1984, however, revealed that he eventually returned to Ticul and died there. By 1984, another Ticul potter had moved to Mérida and established a production unit there that used Ticul clay, but the fate of this production unit is unknown. Finally, another potter migrated to Pisté after 1988 and established a production unit there. By 1997, he had sold part of his interest in a tourist souvenir shop and was concentrating more on making pottery. Other reasons also exist for the demise of those few potters who leave Ticul and establish their own production units. The Ticul fabrication techniques and shapes are tied to the unusual local Ticul clay (see Chapter 5) and temper (see Chapter 6), and these raw materials occur in very few locations elsewhere in Yucatán. Both the palygorskite in the temper (Arnold 1967a, 1967b, 1971, 71

How Have the Population and Organization of Potters Changed?

2005b) and the random mixed layering of kaolinite-montmorillonite in the clay (Schultz et al. 1971) appear to be greatly restricted in distribution and do not occur elsewhere in quite the same abundance or accessibility. Although Isphording and Wilson (1974) proposed that both clays are widespread in Yucatán, the occurrence of a particular mineral category is not the same as a culturally appropriate kind of clay that is associated with a sense of place for that clay. Actually, this sense of place is based on the technological uniqueness of the Ticul clay, because the Ticul mixed layered clay is less plastic and more suitable for making large vessels than the other, more common montmorillonite clay. Consequently, considerable experimentation with new materials is necessary to make pottery away from Ticul unless the migrant makes small vessels that permit more variability in the clays used. Nevertheless, it is now more feasible to import raw materials from a greater distance because of the new mining and procurement organization that has developed since 1992 and the existence of trucks and an extensive highway infrastructure. Amount of Exposure to the Craft

A third selective factor involved in learning the craft is the amount of time that an individual spends in a pottery-making household and the amount of time available to learn the craft. Children have a definite time advantage. Similarly, a non-potter who marries into a family of potters may or may not learn how to make pottery, depending on the amount of time he/she spends in the household and whether he/she has an alternative means of subsistence. Although husbands may learn to make pottery from their wives (2.8 percent, Table 2.2), it is usually the wives who learn from their husbands (16.6 percent) and their husbands’ families (4.3 percent) if the traditional patrilocal postnuptial residence pattern puts them in a household with a pottery-making infrastructure. This pattern is expected because men may already have an occupation before they marry and their occupational responsibilities preclude being able to spend the time necessary to learn the craft. If a male non-potter marries a female potter, then he may learn some aspects of the craft from his wife. He usually does not learn much, however, nor does he learn it very well if he already has an occupation and cannot take the time to learn another. He will not risk learning a new occupation if his own occupation is secure and he can adequately provide for his new family. Other evidence that supports the influence of this selective factor consists of data regarding those potters who actually learned how to make pottery after moving into the households of their spouses after marriage. Between 1965 and 1997, daughters and sons of production-unit owners were the predominant kinds of potters in production units (Table 2.4; Figure 2.7). Very few of their 72

How Have the Population and Organization of Potters Changed?

spouses, however, were potters (SoWi [N = 1] and DaHu [N = 4]). These data suggest that even though the spatial distribution of potters is consistent with a patrilocal and virilocal postnuptial residence pattern, such patterns are changing and becoming less important as a means of acquiring new potters for production units than other factors, such as procreation and employing non-kin as wage laborers. Gender and Household Responsibilities

In some societies, females are potters, whereas in others, males are potters (Arnold 1985:99–108). In Ticul, however, both males and females are potters, although potters in most of the other communities in Yucatán are exclusively females. In 1951, Thompson (1958) found that women were potters in Mama, Maxcanú, Tepakán, and Uayma. Women were still potters in Mama in 1967, 1968, and 1994; in Akil in 1968 and 1994; and in Maxcanú in 1967, 1968, and 1987–1988 (although two men were also listed by Varela Torrecilla [1990:196– 201]). Women also had made pottery in Peto up until 1962. By way of contrast, both men and women have been potters in Ticul (1965–1997) and in Tepakán (1967, 1968, 1994). The gender pattern of pottery production, however, is less important than the reasons why it exists and reveals a fourth selective factor regarding the role of gender in making pottery. This consists of the role conflict of scheduling gender-based responsibilities of child care and household duties. Because only women can bear and nurse children, the tasks of making pottery compete with a woman’s household and child-care responsibilities for full-time production. Such tasks, however, are greatly compatible with these activities on a part-time basis if women have access to the pottery-making infrastructure (Arnold 1985:99–108; Duncan 1996). Women thus can make pottery on a part-time basis between nursing, cooking, child care, water fetching, and other household tasks, and pottery making can be adopted by women to supplement subsistence returns, a pattern that also exists in the northern Valley of Guatemala (Arnold 1978a, 1978b, 1985:99–108). Unlike Byrne (1994), however, who has demonstrated from cross-cultural data that the change in pottery making from a female activity to a male activity occurs with subsistence stress and the scarcity of agricultural land, the change in the allocation of pottery-making tasks from females to males in Ticul appears to be affected by changes in intensity and production locus (Arnold 1985:106– 108). As the craft evolved into a full-time activity and moved outside the household (or the immediate living quarters), child care and household activities interfered with making pots. Consequently, fewer women than men made pottery 73

How Have the Population and Organization of Potters Changed?

Table 2.6. Gender of production-unit owners, 1965–1997. In some cases, these may not be the actual “owners,” but they are the principal (or only) pottery producers in the unit. Males Females

1965–1966

1968

1970

1984

1988

1994

1997

25 2

2 1

23 2

36 9

31 3

33 2

40 5

a

Note: a. The addition of the number of male owners and female owners after 1984 is always one less than the total number of production units because one potter owned and operated two such units from 1984 to 1997.

Figure 2.18. Trend line showing the percentage of female potters from 1965 to 1997. (Data includes potters in Uxmal, see Table 2.1.)

(Table 2.6), and those women who did practice the craft made it less frequently. Consequently, the percentage of female potters has declined from 1965 to 1997 (Figure 2.18). This pattern is borne out by observations in all of the households that I have observed intensively between 1965 and 1997. In 1984, I visited four potters’ households at least once a week for five months, and the female potters in these households often did not have time for sustained production. Although my principal informant’s wife knew how to make pottery using molds, I seldom saw her making pottery. She was taking care of her new baby, washing clothes, and cooking each time I was there. The wife of another potter and her married daughter both knew how to make pottery, but they were busy with the daughter’s young children or were working in the kitchen. They never made pottery while I was

74

How Have the Population and Organization of Potters Changed?

there, even though equipment and space were available. On another occasion, a woman in one of these households said that she could not make pottery that day because she and her daughter were making tamales for a novena that evening in honor of her deceased mother. Another said that she could not make pottery because she was taking care of a sick grandchild. Production by women is complicated by other factors. If other crafts are available, women may prefer them. Further, the requirements of another craft may conflict with those of pottery making. The wife of another informant, for example, occasionally participated in the craft but did not like to make pottery. She preferred to embroider traditional Yucatecan women’s dresses (huipils). Because she needed clean hands both to embroider and to prepare meals, she did not make pottery very often. If a female non-potter marries a potter and moves into a household of potters, she may, or may not, learn how to make pottery. Even though she is exposed to the pottery-making infrastructure, she has household responsibilities (such as cooking) and may also have nursing and child-care responsibilities (Arnold 1985:99–108). Nevertheless, she may begin the process of making pots and have her husband finish them. Because of household and child-care responsibilities, women thus appear to make pottery less often. This pattern is reflected in the potters’ database from 1984 (Table 2.2); fewer potters learn from females (16.6 percent) than from males (65.6 percent, including those who learned outside of their households in large production units). Further, fewer potters learned the craft from their mothers (12.3 percent) than from their fathers (35.5 percent). Forty-eight percent of those who learned from their mothers, however, had no father in the household. Economically Vulnerable Individuals

Another selective factor that influences whether an individual will become a potter consists of the relationship of the craft to economic marginality. Pottery making may be the means for making a living by economically vulnerable individuals who have no other means of support. This factor is a variant of a principle developed elsewhere that pottery making (and crafts in general) is often the refuge of those who are unable to earn their living by agriculture alone, are often landless, or are farming agriculturally limited or insufficient land (Arnold 1975a, 1978a, 1985; Byrne 1994; Duncan 1996:49–51). In these cases, craft production (not just ceramic production) is selected because crafts supplement agricultural subsistence. Some kind of trade or exchange, however, must occur in order to obtain food for craft products. 75

How Have the Population and Organization of Potters Changed?

Because poor or insufficient agricultural land is often eroded, ceramic resources may be exposed and thus ceramic production may be selected over time if pottery is already being made in a community (Arnold 1975a, 1978a, 1985). Women. Although making pottery full-time competes with pregnancy, nursing, child care, and household duties for women, part-time production provides a strong selective advantage among economically vulnerable women and can provide some subsistence returns. Consequently, widows and unmarried, divorced, or abandoned mothers may take up the craft because they have no other means of economic support. Pottery making is thus an economically advantageous activity for such women who already possess pottery-making skill. Consequently, when the gender division of labor for subsistence excludes women, pottery making provides a critical economic pursuit for those who have no man to support them. For a woman who has learned pottery making during her youth, the craft is often a source of sustenance when she has no other means of economic support. The potters’ database from 1984 reveals that the percentages of female potters who were abandoned (3.7 percent, 16/423) or divorced, widowed, or single mothers (4.0 percent, 17/423) were very low. It does appear, however, that these women found pottery making an important economic means to support themselves and their children. A single woman, for example, could support herself and her ailing parents within their household. In 1965, there were two sisters who never married but had learned pottery making from their stepmother. After their father died, they went to live with their half sister and her husband and continued to make pottery to support themselves. By 1997, they still had not married and were living in the households of their half sister’s children and continuing to make pottery. Similarly, a woman who has learned pottery making when she was young and moved away from her parents’ household may return to make pottery when she experiences an economic crisis. One female potter, for example, learned to make pottery in her father’s household but lived with a succession of five different men and had five children with three of them. After she left each man’s household, she returned to her father’s house and made pottery to support herself. Another woman learned pottery making in her natal household, married, and then left to live with her husband. When she and her husband separated, she returned to her late father’s household, occupied by her brother, and made pottery to support herself and her five children. Her sister followed a similar pattern when she separated from her husband. Another female potter, who had a child out of wedlock, remained in her father’s household and made pottery to support herself and her child. She eventually married a non-potter, taught him to make pottery, and 76

How Have the Population and Organization of Potters Changed?

with their children eventually developed one of the largest production units in Ticul. When another woman potter was divorced from her husband, she returned to the home of her father to make pottery to support herself and her son. Eventually, her son learned the craft. She was given a piece of the family houselot by her father, and she became an entrepreneur, making and selling pottery during the expansion of her father’s business. Then, her father died, leaving the production unit in the care of one of her married brothers. When that brother died six years later, she managed the production unit, assisted by her mother, and continued making pots along with her son and several workers. Since the brother’s death left his wife and his three sons without economic support, the three sons learned how to make pottery. By 1997, this production unit was operated by two economically vulnerable women: a widow and her divorced daughter, assisted by the daughter’s son and the sons of her widowed sister-in-law. Although it seems logical that women who have learned to make pottery in their youth turn to the craft because of personal crises, women who have not learned the craft may also become potters in a time of crisis. This change can be facilitated by using a technique that requires little skill (such as vertical half molding; Arnold 1999). In the late 1960s, a mother of two had been abandoned by her husband, and she and her oldest son learned to make pottery using molds lent by another potter who sold the completed vessels for her. Children. When pottery making is located in the household, the craft can provide a selective advantage for other economically vulnerable individuals, such as children, if they have no other means of support. In one family, a woman abandoned her husband and four children. When her husband died, she returned to reclaim her children. They refused to go with her and therefore were left alone in the father’s houselot. Because they had learned to make pottery from their father, they were able to support themselves with no adults in the household. In another pottery-making household, one of the daughters was partially crippled and was unable to walk effectively. Early in her life she had learned to make pottery using molds and supported herself in her father’s household. She never married, and after her father died and the household had passed into the hands of her oldest brother, she remained in the household and continued producing mold-made pottery to support herself. Immigrants. Sometimes pottery making is advantageous for other types of economically vulnerable individuals besides women and children. Immigrants and their children, in particular, may be drawn into the craft when other economic 77

How Have the Population and Organization of Potters Changed?

choices may not be available. After the Mexican Revolution came to Yucatán (1915), two Mexican army officers named Loreto Bak and Espadas remained behind in nearby Santa Elena. They had the reputation of killing those who disagreed with them and this perception created fear in the community. When Julian Cob, one of Bak’s followers, fought with him, Cob fled to Ticul in 1924 with his wife, Teresa, and their eight-year-old daughter Candita because he feared retribution. Both Julian and his daughter learned how to make pottery to support themselves. When the other officer in Santa Elena, Espadas, was killed in his swidden field about the same time, his wife (Ildefonsa Huchim) and her son (Mariano [b. 1921]) fled to Ticul. After her son married, he learned how to make pottery from a potter who lived next door, and his daughter married a potter from an old pottery-making family. Generally, economically vulnerable women appear to have taken the craft into the next generation in disproportionately large numbers because their children have no other vocational choice. On the other hand, some vulnerable individuals did not pass the craft to the children in a subsequent generation. None of the immigrants mentioned above passed the craft to their children, and by 1997 none of their descendants were making pottery. In addition, the widow and her son who had learned how to use molds simply left Ticul in 1975 and went to Cozumel, where the son became a mason. These examples validate the importance of other selective factors (such as infrastructure and exposure to the craft) in the continuation of the craft from generation to generation. Others who make pots outside of pottery-making families may practice the craft for a generation or two, but ultimately, they are deselected from its perpetuation. Social Change and Increased Production-unit Size The second aspect of scale (or workgroup composition) according to Costin (1991, 2005) is the size of the production units. Between 1965 and 1997, the size of production units did not increase in a simple linear manner. Much like the comparison of the population of potters with the increase in the general population, the increase in the number of potters (Figure 2.3) parallels the increases in the population but fluctuates more. By way of contrast, mean production-unit size has not changed much over time (Table 2.1; Figure 2.5). This size, however, may vary seasonally and may vary over the lifetime of a potter, so that production-unit size may have short-term increases and decreases that are sometimes temporary and sometimes permanent. Although the principles for the acquisition of personnel appear to be relatively simple, they are highly nuanced and also are signifi78

How Have the Population and Organization of Potters Changed?

cantly influenced by other factors that affect short-term changes in production­unit size. Factors for Temporary Increases

Child Labor. Probably the most important factor that temporarily increases the size of production units is the use of child labor. This practice is one benefit of producing pottery independently in one’s household rather than making pots as an employee of someone else. If children learn at least some of the craft, they provide a labor resource that can bring increased economic returns. As a result, a household prospers economically. After my principal informant quit making pots as a laborer in the late 1970s and enlisted his children to make pottery at home, his weekly income increased 56 percent to a self-reported 9,000 pesos. It then rose another 100 percent to 18,000 pesos, and finally, it increased 778 to 1,000 percent to 70,000 to 90,000 pesos. These increases do not take into account inflation and the devaluation of the peso that also occurred during this period. But it was the increase in prices without an increase in wages that motivated him to leave his employer. He still faced inflation in the cost of raw materials, but working independently allowed him to prosper, even with inflation. The use of children for labor usually ends when they marry and move out of the household. As a result, production declines, and a potter may abandon the craft entirely because his supply of free labor is gone. In 1994, two potters whose children made pottery in 1984 had abandoned the craft because their children had grown and left the household. Such temporary increases in production-unit size may also have long-term effects. With the accumulation of capital obtained from child labor, production units may also sustain a permanent increase in size by using the capital generated to increase the number of workers and the amount of production space. Child labor was part of the great economic success of two potters who developed the largest production units in Ticul. Negative Effects of Education and Vocational Choice. As is common in many parts of the world (Arnold 1985:196–198, 228–229; Haaland 1978:57), making pottery is a low-status occupation in Ticul, and in the late 1960s, potters were near the bottom of the social scale (Thompson 1974:122–123). This social position was reinforced by the poor educational background of many potters. Several potters said that they were illiterate, whereas others had gone to school no more than one or two years. One surrogate indicator of education is facility in Spanish. Lack of facility in Spanish was linked to limited education and was characteristic 79

How Have the Population and Organization of Potters Changed?

of the adult population in Ticul in the 1960s. Indeed, in 1968, a random sample of Ticuleños who were thirty-six years of age and older had a mean of 3.43 years of education; among those who had an age of less than thirty-six, it was still low, with 4.33 years of education (Thompson 1974:99). During the late 1960s, Yucatec Maya was the language of choice and potters had limited facility in Spanish. By this standard, almost all potters who practiced the craft had very limited formal education. Yucatec Maya continued to be used among potters into the 1980s and 1990s, except for the Spanish-speaking entrepreneurs from outside of Ticul who started their own production units. Some potters recognize their low status and see education as a way for them to improve their children’s lives. Sending their children to school, however, creates both short-term and long-term implications for production-unit size. Although all children in a potter’s household may have learned how to make pottery in the past, such a pattern no longer existed between 1965 and 1997. Some simply refused to learn and eventually chose another occupation, such as shoemaking. Or, even if they learned to make pots, they refused to make them. When the children do not make pottery or do not participate in the process, the household income suffers significantly. This negative effect was verified by informants and observed in several households. Education selects against the perpetuation of the craft and an increase in production unit size in four ways. First, schooling removes children from the family work force for most of the day. As a result, they are unable to contribute labor to making pottery as they would if they had stayed at home. This practice inhibits temporary increase in production-unit size and production returns. Second, attending school requires money for supplies and uniforms and thus drains capital from other household goals, such as intensifying production or increasing the size of the production unit. Potters who invest in their children’s futures thus do so at great economic risk to their present livelihood and have traded long-term investment in their children’s well-being (and their well-being) for their own short-term economic gain. As a result, some children of active but poor and uneducated potters have become schoolteachers, bookkeepers, and skilled technicians rather than potters. This problem was illustrated by one potter who said that he was poor because of the cost of sending his children to school. This cost, he said, prevented him from buying household items, making improvements to the house, and building a kiln. It also precluded constructing more space to protect drying pottery from wind and rain. Third, education also acts as a selective force against increased productionunit size because poor education and speaking Yucatec Maya limit economic 80

How Have the Population and Organization of Potters Changed?

returns. When the ceramics were distributed exclusively to Yucatec Maya–­speaking populations, the lack of facility in Spanish was not a liability. Once the primary demand for ceramics moved outside of the Yucatec population to shop owners, brokers, and hotel managers who do not speak Yucatec Maya, potters who had limited facility in Spanish were at a disadvantage in commercial transactions relative to those who had been to school and had a good command of Spanish. Fourth, formal education delays learning many of the skills of pottery making and virtually ensures that children will not become potters. Even with some formal education, children often choose not to make pottery because they become aware of the low status of the craft and choose another occupation with a higher social status, more financial returns, and greater security. Between 1965 and the early 1990s, becoming a shoemaker was the most attractive alternative to making pots and was one of the most frequent alternative occupations that the children of potters pursued. But potters’ children also have become masons, waiters, and orchard workers or just moved away into unknown occupations. At least three adults who have learned the craft as children, however, have returned to it and set up production facilities after having gone to post-­secondary school. The son of one potter, for example, went to school in Oxkutzcab and then went to Mérida for one year of technical school. By 1984, he had returned to Ticul and was helping his father make pottery. Vocational choice and its effect on making pottery are not just limited to the children of potters but may affect adult potters as well. Some potters have opened a store on their property to sell various types of dry goods. Although no systematic data were collected about the success and failure of these stores, at least five pottery-making households had such stores at one time or another between 1965 and 1997, and a few potters who successfully established such stores permanently left the craft. Others operated a store cyclically with pottery making, and either supplemented their returns from the store by making pottery, made pottery seasonally for the Day of the Dead rituals, or returned to pottery making when their store failed. Factors for Permanent Increases

Besides factors that select for temporary increases in production-unit sizes, other factors select for the permanent increase in size. In some cases, production-unit owners employ unskilled individuals and teach them the basic skills of the craft. In other cases, well-established potters who could work independently choose to work for another production unit. Owners are often dependent on these experienced potters because new potters do not possess the ability necessary to undertake highly skilled tasks such as forming vessels with complex profiles, 81

How Have the Population and Organization of Potters Changed?

making molds, and firing. One way of seeing the evolutionary change to permanent increases in production-unit size is to see the addition of wage employees as a more reliable substitute for the potter’s children. So with wage laborers, potters can respond quickly to demand for new vessels. Why would an experienced potter want to work for someone else when he could make pottery independently? Another way to ask this question is, Why do some production units get larger and employ skilled potters from outside of their own family? What are the factors responsible for the aggregation of personnel (both potters and non-potters) that lead to permanent increase in the size of production units? The lack of capital is one of the most important reasons why an experienced potter works as an employee. This problem is a recent one because capital was not always necessary to make pottery. In the late 1960s, potters could mine clay and temper themselves with little cost except their labor and the transportation of these raw materials to Ticul. If a potter also practiced subsistence agriculture, he could feed his family and provide fuel for firing at no cost except his labor. With the increasing task segmentation that occurred between 1965 and 1997, potters had to buy their clay, temper, and firewood. Accumulating sufficient capital for purchasing raw materials and then managing that capital for food, household requirements, and pottery production have been difficult for some potters. Increasing the size of the production unit and increasing its intensity by adding more personnel requires two capital-intensive changes. First, increased production requires extra space for drying pottery and for protecting it from the vicissitudes of weather, and construction of this space requires capital. Second, capital is also required to fire pottery, and the cost of building, maintaining, and using a kiln may be considerable. Even if a potter (or a member of his family) has the skill to build a kiln, it still requires substantial capital for raw materials and fuel, and potters who are poor do not have the capital available to do so. Third, paying workers on a daily or weekly basis also requires capital, even though some employers do not pay their workers regularly and carry an accounts payable balance of wages owed. Maintaining a constant supply of capital is no easy task given the variations in the income from pottery sales. So in 1997, I mentioned to the owner of the largest production unit that I realized he required a great amount of capital to build and maintain a production facility, and I then asked him whether he had received a loan from the bank or used his own money. He said that he had used his own money. Such self-capitalization projects are only possible because capital returns from production are invested gradually over time and are well managed. 82

How Have the Population and Organization of Potters Changed?

A second reason that potters hire themselves out to another production unit concerns the delay between the investment of labor in forming, drying, and firing pots and cash returns for that labor. This delay can drive potters into other occupations, such as shoemaking or operating a platform tricycle as a taxi for goods and people. Such activities can either supplement or replace pottery production because they provide more immediate cash returns than the pottery industry, which requires waiting one to two weeks for the production sequence to run its course before returns can be received from the process. As a result, many potters prefer to work outside their own household and receive their wages on a weekly or daily basis. No investment in raw materials is necessary. The potter is often paid for each item made on the day that he made it; he does not have to wait for payment as he would if he had made pottery on his own. Consequently, he does not have to manage capital for ceramic production and balance those requirements with food and household necessities. A third reason that potters want to work for a production unit outside their own household is the difficulty in finding clients. When they do find clients, they may not get paid for their pottery immediately and are asked to provide credit to their buyers. Some potters refuse to give credit, but others said that credit is essential for them to compete in a market with an abundant supply of pottery. Even then, returns may not be immediate because some buyers ask potters to carry a credit balance, and the potters may never get paid for their pots—a problem cited again and again by independent potters. Indeed, one potter in 1997 voluntarily complained that he had an accounts receivable balance of more than a million pesos. Production-unit Size and the Number of Units

Although the size and number of production units have increased through time, the relationship between them has not changed. Indeed, one of the most surprising findings of this research was the consistent power law (log-log) relationship between the number of potters in each production unit and the number of production units (Figure 2.19a–g). In order to examine this relationship, a scatter plot of production-unit sizes and their frequency was created for each observation period. The data for some observation periods were more complete (1965–1966, 1984, 1997) than for the other periods of observation (1968, 1970, 1988, 1994). Trend lines were added for each plot with an attempt to find the trend line with the greatest conformity to the data as reflected in the highest value of R2. In all of the observations except two (1965–1966 and 1994), a power law relationship had the best fit, but these two exceptions had an R2 only slightly higher than the power law relationship. In 83

Figure 2.19a. Trend lines showing the relationship between the number of units and the number of potters per unit in 1965–1966. The log-log trend line is the solid line and its R2 was slightly lower than the exponential curve that was the best fit.

Figure 2.19b. Trend line showing the relationship between the number of units and the number of potters per unit in 1968.

How Have the Population and Organization of Potters Changed?

Figure 2.19c. Trend line showing the relationship between the number of units and the

number of potters per unit in 1970.

Figure 2.19d. Trend line showing the relationship between the number of units and the number of potters per unit in 1984.

all cases, however, the R2 of all of the power law trend lines was greater than 0.72. Although the number of data points is small (because of the small size of the production units), there is a consistent power law relationship between the number of production units and their size across the thirty-two years of this study. 85

How Have the Population and Organization of Potters Changed?

Figure 2.19e. Trend line showing the relationship between the number of units and the

number of potters per unit in 1988.

Figure 2.19f. Trend lines showing the relationship between the number of units and the

number of potters per unit in 1994. The log-log trend line is the solid line and its R2 was slightly lower than that for the exponential curve that was the best fit.

This relationship suggests that the number of production units and the number of potters in each unit are a kind of scale-free, self-organizing system found in a wide range of phenomena (Bentley and Maschner 2001; Bentley et al. 2004;

86

How Have the Population and Organization of Potters Changed?

Figure 2.19g. Trend line showing the relationship between the number of units and the

number of potters per unit in 1997.

Bentley and Shennan 2003). The reason for this relationship appears to be associated with the demand and the market, such as the rich-get-richer phenomenon found in social network analysis (Heyman 2006:605); some production-unit owners benefit economically from the larger production units and some household-trained potters prefer to work for larger production units because they receive a consistent wage. Conversely, as other potters work for the larger units and learn the skills of production, they leave these larger units to form their own small independent units. Costin (1991:15–16, 2001, 2005; see also Rathje 1975) has argued that as craft production evolves, production units increase in size to take advantage of economies of scale. The Ticul data suggest that growth in a production unit is more complicated. Why do most of the production units in Ticul continue to be small household units? Smaller units can take advantage of members of the household as labor and thus have reduced costs. Further, at least some small production units have been formed by those who have worked for larger production units and then formed their own production facility. It is almost as if workers gain experience in the large production facilities, learn about the process of making and marketing pottery, and then take this knowledge to work on their own and make pottery independently. Further, small production units also provide a better context for learning all aspects of the craft and provide personnel for the larger units. Consequently, there appears to be an important symbiotic relationship

87

How Have the Population and Organization of Potters Changed?

between larger units and smaller units, which is illustrated in the mathematical relationship between larger and smaller units over time. This relationship appears to be independent of the great social changes that have occurred and the changes in demand, market, shapes, and technology that occurred in pottery production in Ticul. Conclusion This chapter provides insights into the composition, continuity, and growth of the population of potters and their production units. First, the strong continuity in the location of production units through time is the result of the use of the nuclear family (So, Da, Mo, Wi) as the primary production personnel and reveals that most (but not all) production units are household-based. The small size of most units (<3) is also consistent with this finding. Second, although some production units have grown in size and have added wage laborers, most production units are small, remain household-based, and use household personnel. Third, most of the stability in the location of production units through time results from the patrilineal inheritance of household land. Consequently, even with great social changes, pottery production largely has remained household-based between 1965 and 1997. The household was the social context for the learning and perpetuation of the craft, and it is a location where the learners can be supported economically by others when production by neophytes may not be economically viable or sufficient for sustenance. Children’s residence in their natal households is long enough to learn all the aspects of the craft that are required to produce a broad range of vessels. When craft learning is centered in the household, its transmission corresponds to the behavior of acquiring household personnel. This behavior largely consists of the patrilineal inheritance of household land and the patrilocal or viri­local postnuptial residence and their accepted variants (Arnold 1989a). These variants may bring single females, widows, and unmarried or abandoned mothers into a household. Such individuals are almost always collateral relatives, but they never inherit household land. Although land inheritance and postnuptial residence do appear to describe the perpetuation of the craft, learning may take place outside of these patterns, but almost always individuals learn from those who occupy the same household—no matter how infrequent or seemingly bizarre the kinship relationship. Postnuptial residence behavior, however, has become much less important as a way of acquiring new production-unit personnel since 1965. It was much more important for acquiring potential female potters than for acquiring male potters, but its declining importance is reflected 88

How Have the Population and Organization of Potters Changed?

in several measures, not the least of which is the decrease in the percentage of female potters. One of the common complaints against using explanations of patterns of inheritance and postnuptial residence (Deetz 1965; Hill 1970; Longacre 1970) is that they are ideal cognitive rules that may be worked out differently in actual behavior (Allan and Richardson 1971). Others (see, e.g., Stanislawski 1977; Stanislawski and Stanislawski 1978) simply challenge the validity of using any kin-based model. The data reported here, however, consist of actual behaviors (not rules) based on participant-observation, maps, and independently derived and verified genealogical data. Most of these data are based on actual genealogical knowledge of the potters and their residence locations obtained during the more than thirty-two years of this research. Many of these data have been verified using the marriage records of the Ticul church. These patterns of inheritance and residence, however, change. By 1997, for example, living near one’s father was not simply the result of patrilineal inheritance of household land and patrilocal residence but rather resulted from the desire of parents to have their children live relatively near to them. To ensure this proximity, a father bought land nearby for his sons (most frequently) and/or his daughters (less frequently). Spatially, this behavior looks like virilocal residence since the postnuptial household is nearer the groom’s parents’ house than that of the bride. Postnuptial residence behaviors thus are based on more than just postnuptial residence rules. Although patterns of land inheritance and postnuptial residence appear to describe intergenerational transmission of the craft, they do not fully explain why some individuals become potters and others do not; kin-based patterns do not fully account for potters’ choices to learn the craft. Although the behavioral patterns of land inheritance and postnuptial residence provide personnel and expose individuals to ceramic production, the composition of potters’ households is never just the outcome of such patterns but includes factors that select for or against the practice of the craft. These factors serve to constrain the size of the population of potters to a level well below that which would occur with simple demographic growth. How do these data square with Costin’s description of scale and its usefulness as a parameter of defining and organizing principles of craft specialization? Costin provides two extremes of scale: (1) small, individual, or familybased production units; and (2) wage labor forces of the industrial West (Costin 1991:15). Although this dichotomy is a truism in many respects, it is also true that when family-based production units meet the industrialized West as we see in Ticul, a mix of an abundance of small household-based production units can 89

How Have the Population and Organization of Potters Changed?

coexist with much larger units that are also household-based but include wage labor. In the evolution of specialization in Ticul, all of the production units do not grow. Costin (1991) argues that the facilities of independent specialists (like those described here) can be both large and small. In Ticul, large production units co-occur with a predominance of small units. Large units, however, are still family- and household-based but also reveal an increased use of wage laborers over time, including members of the extended family and individuals related to production-unit owners by lineal, collateral, and affinal ties. This observation is similar to that which Costin drew from Kleinberg’s (1979) work with village pottery production in Japan (Costin 1991:15). As production units grow in size, labor is recruited first among distant, fictive, and adoptive kin and then nonrelated individuals are added (Costin 1991:15). The Ticul data are much more complicated because the composition of production units through time reveals a slight increase in the number of types of lineal kin and growth in the number and percentage of non-relatives. Furthermore, the percentages of collateral relatives in production units have remained the same, whereas the percentages of affinal kin, women, wives, and daughters of production-unit owners have decreased. Analytically speaking, it initially appeared that the size of production units should be decoupled from the number of production units. The Ticul data, however, reveal that although a few production units increased in size between 1965 and 1997, most remained small with a mean and median size that grew slightly since 1965–1966. Most important, however, is that the size and number of production units appear to be related. It is thus possible to have highly specialized, intensified production without an increase in production-unit size, even though a mutual relationship appears to exist between the number of production units and their size. The number of production units in Ticul, however, is a better measure of production intensity than production-unit size. Costin (1991:15–16) argues that for independent specialists, the primary factor determining the scale of production is the efficiency of production. This efficiency, she proposed, is a function of the technology used and the level of production-unit output. Workshop size, she says, will rise to take advantage of economies of scale if per-unit costs can be lowered through sharing expensive technology or by dividing tasks among many workers. Furthermore, she argues that larger workshops with larger output may be able to exploit certain marketing strategies (Costin 1991:15–16). In Ticul, production-unit size does not need to increase to take advantage of economies of scale because of the availability of unskilled household labor. Children and spouses can perform unskilled tasks to supplement production in times of high demand. Further, task segmenta90

How Have the Population and Organization of Potters Changed?

tion—such as raw materials procurement, firing, and marketing—is the domain of specialists, allowing vessel fabrication to be the only production activity for households. As a result, production units do not have to increase in size to take advantage of increased specialization. Rather, unskilled individuals can perform tasks as specialists and can be sellers of finished pottery. Larger production units have individuals who specialize in particular tasks, and this certainly creates economies of scale. But using the push toward economies of scale as an explanation for increased production and production-unit size does not work well because potters choose to join larger production units rather than work in their own households because of their lack of ability to manage resources, like skill and capital, and the time constraints on the returns from the ceramic production sequence. These factors may be viewed as efficiency but only in a way that is contextualized within the technological limitations of the ceramic production sequence and in the economic context in Ticul during the last decade of the twentieth century.

91

Chapter

How Have Demand and Consumption Changed?

three

P

otters may be able to mine raw materials; mix them effectively; and fabricate, dry, and fire pottery successfully. But if they cannot turn their pots into food or into a commodity that can be exchanged for food (like money), they cannot provide for their basic needs and their economic future, as pottery is in jeopardy. The process by which potters obtain food for their pottery has two different dimensions. First, there must be a consuming population that provides a demand for vessels. This demand is tied to the uses of the vessels and to the sociocultural values and the behavioral patterns that influence those uses. These values and behavior do not just consist of practical benefits usually associated with ceramic vessels, such as taste, aesthetic preferences, cost, and porosity (see Arnold 1985:127–151), but may also involve the symbolic role of pottery in ritual and in marking social position (Arnold 1985:158–164). The second dimension of how potters obtain food for pottery consists of the way in which pottery is 93

How Have Demand and Consumption Changed?

distributed to the consuming population. So after responding to demand by producing vessels, potters must make them available to consumers by selling them for money or exchanging them for food or some other commodity. The demand for and distribution of pottery thus constitute two of the most significant feedback mechanisms affecting pottery production that can limit or stimulate production (see Arnold 1985:127–167). They are two aspects of production organization that Costin (1991:11–12) calls the “context” of specialization. This chapter describes changes in the demand of consuming populations that use Ticul pottery and in the patterns of pottery consumption that have occurred since 1965. In a wide-ranging study of contemporary pottery making in Crete and the Aegean, Day (2004) showed that pottery production must be understood within a wider context of regional demand. Other scholars have also described ethnographic pottery production in regional contexts that relate to demand and distribution (Nicklin 1981; Reina and Hill 1978; Rye and Evans 1976; Stark 1994), and still others have recognized the importance of integrating issues of production with those of distribution (Bey and Pool 1992; Howard and Morris 1981). Similarly, ceramic production in Ticul cannot be adequately understood unless it is placed within a wider context of regional demand for pottery vessels. The demand for pottery exists within a consuming population that embodies values and behavior related to pottery and provides the impetus to acquire it. Between 1965 and 1997 three such populations have provided demand for Ticul pottery: (1) the native Yucatec Maya population, (2) tourists, and (3) urban dwellers and hotel managers. The demand for pottery among these populations has changed greatly during this period, as have the vessels that they consume. Each population constitutes a distinctive market niche and is targeted by potters when they make their pottery. Generally speaking, each niche provides a unique demand for particular kinds of vessels. Shapes produced for one kind of consuming population generally are not distributed and sold to another, even though individual potters may make pottery for multiple niches. Demand and Cultural Evolution In Ceramic Theory and Cultural Process, I argued that demand was a critical feedback mechanism for the development of specialized ceramic production (Arnold 1985:127–167). Although population growth does provide some deviationamplifying feedback for increased intensity of production and the development of full-time specialists, I argued that it was the forging of new links between ceramics and ideology and ritual that provided the greatest deviation-­amplifying 94

How Have Demand and Consumption Changed?

feedback for production intensity and specialization. This same theme was elaborated by Katherine Spielmann, who surveyed a massive ethnographic literature on ritual (Spielmann 2002) and showed the deviation-amplifying feedback provided for the development of increased specialization. Demand from Traditional Uses of Pottery The Yucatec Maya population has been the most significant long-term consumers of Ticul pottery and has provided the most enduring demand for it. Traditionally, the Yucatec Maya have used pottery as containers for carrying, storing, and serving water; for cooking, processing, and serving food; and for household ritual during the Day of the Dead ceremonies. As a segment of overall demand, however, the kinds and numbers of vessels produced for the Maya population have declined dramatically between 1965 and 1997 except for vessels for the Day of the Dead rituals. In 1965, all potters produced pottery for the Yucatec Maya population, but social change had already occurred with increased access to factory-made goods and with changes in ritual and consumer preferences. As a result, potters had already abandoned making a wide variety of traditional vessels that were no longer used. Potters still knew how to make these vessels, however, and at least one potter was able to fabricate 180 named categories of such vessels even though he produced no more than 5 percent of that repertoire regularly. Water Vessels: The Collapse of Demand

In 1965, the most important demand for pottery vessels came from their use for water transport and storage. Even though the use of pottery as a channel for water is widespread throughout the world, the demand for water vessels in Yucatán is particularly unique because it is mediated by powerful environmental forces that involve the geology, topography, and hydrology of the peninsula and the patterns of weather and climate. Geology, Topography, and Hydrology. The Yucatán peninsula is a vast limestone platform that stretches 400 to 600 km northward from the North American continent (Figure 2.3). The northern part of the peninsula is relatively flat, but as one moves south from the northern coast (Isphording 1975:240; Wilson 1980), the elevation increases gradually with an average gradient of one meter per three kilometers. Approximately eighty kilometers from the north coast and fifty kilometers inland from the western shore of the peninsula, the topography changes and a ridge (called the puuc in Yucatec Maya, or hill ridge) rises abruptly to an 95

How Have Demand and Consumption Changed?

Figure 3.1. View of the hill ridge in 1968 looking east from the road between Ticul and

Sacalum. This area has been cleared of forest for agricultural purposes.

elevation of 100 m (Isphording 1975:251; Figure 3.1). Beginning just east of the town of Maxcanú, this ridge (also called the Sierrita de Ticul) extends 160 km to the southeast (Isphording 1975:25; West 1964:71; Wilson 1980). Immediately south of this ridge, the terrain flattens out again with large areas of deep soils and little surface limestone (Figure 3.2; Dunning 1994). Beyond these flat areas, the undulating terrain begins again until approximately 16 to 20 km south of the hill ridge, where a series of haystack-like hills (Figure 3.2) dominates the landscape (Doehring and Butler 1974; Isphording 1975:255; Wilson 1980). All of these areas consist of a karst topography with no surface lakes or streams. Drainage is subsurface with water disappearing into the ground within a few hours after heavy rain (Doehring and Butler 1974:562). The only permanent natural sources of water are sinkholes (called cenotes), where the surface limestone has collapsed down to the level of groundwater, and natural depressions (called aguadas) that retain water from seasonal rainfall (Doehring and Butler 1974). Weather and Climate. The northern Yucatán peninsula consists of a tropical climate (the Köppen Aw type) in which (1) the mean monthly temperature exceeds 18°C (Figure 3.3), (2) the amount of annual rainfall is greater than 750

96

How Have Demand and Consumption Changed?

Figure 3.2. View from the top of the puuc ridge looking south toward the haystack-like hills of Yucatecan hill country in 1997. The highway is the road from Ticul to Santa Elena, and the light rectangular feature in the upper center right of the photograph is the roof of the church in Santa Elena. This photo shows portions of kakab, ya’ash kash, and the wits ethno-ecological zones (see p. 285).

mm, and (3) precipitation falls principally in the summer (García 1973:33–37; INEGI 1996:3; Tamayo 1962:166, 171–172). The rainy season begins abruptly in late May or early June and rain falls almost daily, with heavy downpours nearly every afternoon. The rains continue until October except for a brief break (canícula) in some areas during July or August (Figure 3.4). Relative humidity during the rainy season is very high. Besides this seasonal pattern, the Yucatán peninsula lies in the path of frequent tropical storms that sometimes develop into devastating hurricanes. During the late summer and fall, massive low-pressure cells arise out of the Caribbean or western Atlantic and move west and north. Some of these cells develop hurricane-strength winds, but most consist of days of cloudiness and rain. Although rainfall may be more intense, heavy, and prolonged during the hurricane season, the rains from October through January are less frequent, more erratic, and more unpredictable than in the rainy season. By January, the rains have usually ended, except for an occasional storm that comes from the north

97

How Have Demand and Consumption Changed?

Figure 3.3. Bar graph showing the mean monthly temperatures in the town of

Oxkutzcab (see Figure 3.4), located fifteen kilometers from Ticul. The data shown here are the means of seventeen years of temperature data collected between 1921 and 1960 (García 1973:73, 203).

across the Gulf of Mexico. From February to May, little precipitation falls until the rainy season begins again in late May or early June (Figure 3.4). Regional Variation and Water Availability. The geology, topography, and hydrology of northern Yucatán and the seasonal scarcity of rainfall have created a significant adaptive problem of water availability for human populations (Dunning 1994; Isphording 1975:244, 246, 251; McAnany 1990). With little rainfall from January through April, and without permanent sources of water, no water is available for human settlements. This seasonal water scarcity is chronic for the area along the north side of the hill ridge, where natural sources of water are nonexistent to rare because the water table is too far below ground level. Natural sinkholes are not deep enough to reach the water table and very few permanent natural sources of water exist. As one proceeds southeast along the hill ridge, the water table gets progressively deeper. At the northwestern end of the ridge at Muna (Figure 2.2), for example, water lies fifteen to twenty meters below the surface. In Ticul, it is twenty-four to twenty-five meters deep, and in Peto, eighty kilometers southeast of Ticul, the water table is seventy-five meters underground. South of the hill ridge (Figure 3.2), water scarcity becomes even more chronic because the water table is much deeper (Dunning 1994; Isphording 1975:244,

98

How Have Demand and Consumption Changed?

Figure 3.4. Bar graph showing monthly precipitation in the town of Oxkutzcab (elevation 32 m, 20° 5' North and 89° 17' W), located fifteen kilometers from Ticul. The data shown here are the means of eighteen years of precipitation data collected between 1921 and 1960 (García 1973:73, 203).

246, 251; McAnany 1990). In such areas, rains provide the only water available when it collects in seasonal pools of semi-artificial origin (aguadas) and in holes in impervious rock (haltuns; see Thompson 1895:7). With very little rainfall during the dry season from January through most of May, the lack of a permanent source of water creates an even greater adaptive problem here than north of the hill ridge. As a consequence, few major settlements existed along the hill ridge and southward until the beginning of the late prehistoric period (the Terminal Classic period, ca. A.D. 800) when the Maya dug cisterns (called chultuns) to hold water and constructed plazas to channel rainwater into them (Thompson 1895; Zapata Peraza 1989). This new technology increased the carrying capacity of the land, and the populations expanded into areas that had been previously uninhabited because of the lack of reliable sources of domestic water (McAnany 1990). Wells and Demand for Pottery. Beginning in the Colonial period, wells were dug to obtain water, and before the widespread availability of electricity, windmills, animal traction, or human muscle provided the energy to raise the water to the surface. Today, the only reliable sources of water along the hill ridge and southward are deep wells that must be dug through many meters of soil, rocks, and clay using metal tools and explosives. To understand the magnitude of this

99

How Have Demand and Consumption Changed?

Figure 3.5. Water-carrying pots (cántaros) that have just been removed from a kiln in

1965. One potter is holding a jar and the other potter is pointing out damage to the base of a vessel.

task, one only needs to consider the depth of the water table in the town of Santa Elena, ten kilometers south of the hill ridge (Figures 2.3 and 3.2). From the time that the town was established until a piped water system was installed, all of the water had to be drawn from a single well that was sixty to seventy-five meters deep. This task was so daunting that draft animals were employed to raise it to the surface using a waterwheel (noria). Water availability is so important in this area that it has played a significant role in political conflict. In a region with no surface water, right of access to subsurface water is critical to survival, and water rights played a significant role in the Rebellion of Nohcacab (now called Santa Elena) that led to the Caste War (1847–1901). In his account of this conflict, Güémez Pineda (1997) describes two incidents that involve access to water. In one incident (1832), the mayor of Nohcacab was accused of shortchanging the amount of water given to an Indian by closing the gate to the waterwheel before the assigned hour. Since this well was the only source of water in the community, its closure had serious consequences, and many others also had their access to water curtailed. In another incident in 1841, the owner of Hacienda Ya’ax Che’ (south of Santa Elena) took control of a well located on common property and charged local inhabitants a fee for

100

How Have Demand and Consumption Changed?

Figures 3.6a and 3.6b. Two positions used for carrying a water pot as demonstrated by a woman in 1984. One position places the base of the vessel on the hip (left) and the other places the horizontal handle on the hip (right). (Figure 3.6b from Dean E. Arnold, Ecology and Ceramic Production in an Andean Community, Cambridge University Press, 1993, p. 124, used with permission)

using it (Güémez Pineda 1997). In an area with abundant sources of water, these actions would have little significance, but in an area with no surface water, a water table deep below the surface, and very few wells, these incidents provoked anger; these two wells were probably the only reliable sources of water over many square kilometers. Water-carrying Vessels as a Channel. In many cultures, ceramic vessels are usually the principal channel for transporting water from its natural source to the household. In the northern Yucatán peninsula, water was carried for some distance because of limited access to permanent sources, and ceramic vessels offered significant advantages compared to other vessels, such as metal pails. Consequently, potters produced shapes with unique features that facilitated the transport of liquid. From 1965 to 1970, potters produced a shape called the cántaro ( p’uul in Yucatec Maya; Thompson 1958:34–37, 125; Figure 3.5), which was designed to carry water comfortably and efficiently. First, the narrow orifice, high neck, and recurved lip reduce spillage during transport. Second, the neck 101

How Have Demand and Consumption Changed?

is high enough to permit the carrier’s arm to go around it and clutch it to the body. Third, the shape is versatile enough to be carried using two different patterns. With one pattern, the angular edge of the vessel’s flat base rests on the hip shelf (Figure 3.6a). With the other pattern, the flat base of the horizontal strap handles rests on the hip (Figure 3.6b; Arnold 1993:124; Thompson 1958:104c, d, e). Both of these positions transfer part of the weight of the vessel to the hip so that water can be carried with ease over some distance without putting the entire weight of the vessel and its contents on the arms or head. All of these attributes (a recurved rim, a tall neck, handles with a flat horizontal plane) feed back into production and require separate steps during the fabrication of the vessel. The flat base is also critical and is largely the result of a forming technique that begins with a flat clay pancake. This kind of beginning creates a sharp angle between the base of the vessel and the vessel wall and facilitates ease of carrying because the angle rests on the hip shelf. These vessel attributes are different from those found in other parts of the Maya area. In central Guatemala, for example, water-carrying vessels are carried on the head on a ring of cloth (see Arnold 1985:145; Reina and Hill 1978:230), and this pattern requires a round base. Potters in Chinautla, Sacojito, and Durazno create this feature by molding the base of the vessel over an old jar (Arnold 1978b, 1985:203). Although the plasticity of the clay permits the congruency of a vessel shape with the motor habits for carrying water, clay vessels have a different set of advantages for storing water. Unlike metal vessels, ceramic vessels give water a pleasant taste (Arnold 1985:139). Furthermore, they are porous and permit water to move through the vessel walls (Arnold 1985:139–140; Rice 1987:230–231, 350–354; Shepard 1956:126–127). When the water reaches the external surface, it evaporates, removing heat from the vessel and cooling the liquid stored in it. Potters and consumers recognize these characteristics and potters produced a variety of vessels used for this purpose (see Thompson 1958:118, 120–126, 129–133). Water-storage vessels in the potters’ repertoire included the tall jar (tinaja; Thompson 1958:40 [fig. 7d]) and the neckless jar (apaste; Thompson 1958:45 [fig. 11e], 47 [fig. 13d], 116 [fig. 36l]). Other vessels for storing water consisted of a barrel-shaped vessel (barril; Thompson 1958:50 [fig. 15f ]) and deposito) and pitchers ( jarrias; Thompson 1958:42 [fig. 9e]) for storing and serving it. Demand and Calcite Inclusions. The use of Ticul vessels for anything involving water hastens their disintegration and creates a continuing demand for them. The cause of this disintegration lies with the calcite in the temper. From acid dis102

How Have Demand and Consumption Changed?

solution analyses done in 1966, the amount of carbonates (including calcite) in the temper for non-cooking pottery is 65.2 percent (N = 3). Since temper and clay are mixed using a ratio of two parts temper to one part clay, at least 57.1 percent of the paste consists of carbonates. Depending on the firing atmosphere, calcite decomposes into calcium oxide and carbon dioxide between 650°C and 900°C in a process called “calcination” (Rice 1987:98). The calcium oxide then combines with water in the air or in the vessel walls and creates calcium hydroxide in the paste. This molecule occupies more physical space than the original calcium carbonate and therefore creates stresses that cause spalling, cracking, and breakage. This danger is compounded by the porosity of the pottery that allows water to penetrate the interior of the fabric. When these vessels are used for water, or for plant pots and the plants in them are watered, their demise is inevitable, creating a continuing demand for these vessels. Spatial Variability of Demand: Depth of the Water Table. The depth and distribution of water sources in Yucatán also affected the consumer demand for vessels for transporting and storing water. Potters were aware of these differences, and in the late 1960s, they adjusted their production and distribution of these vessels accordingly. In the area north of Tecoh and Acanceh (Figure 2.2), the water table lies nearer the surface than it does in communities to the south of these towns. As a result, the sinkholes in the north usually have water in them. Wells are relatively shallow and can be dug easily and cheaply. Sources of water are thus abundant in the north, and water for households does not need to be carried over long distances. Consequently, the primary demand in the north consisted of vessels for storing water (such as the tinaja and apaste; see Thompson 1958:116, 118, 122), with much less demand for water-carrying vessels (such as the cántaro; see Thompson 1958:124–125). In the south near the hill ridge, however, the water table is much deeper than it is farther north. Because wells were the source of water here, cultural factors, such as capital costs, affected the number of sources of water. Digging a well required capital (for metal tools and explosives) and knowledge about where to place the explosives and how to set off a blast underwater. So the deeper the water table, the greater the cost of digging a well. Few wells were dug because few people could afford them. This pattern of few wells existed along the hill ridge and persisted until relatively recently. As recently as the late nineteenth and early twentieth centuries, for example, there were only eight wells in Ticul. Even as late as 1965, the cost of digging a well was still beyond the reach of many families. At a cost of 2,000 103

How Have Demand and Consumption Changed?

to 3,000 pesos (US $160–$240) per well and a daily wage of less than thirty-five pesos (US $2.80) a day, a family would have to accumulate money for a minimum of fifty-seven to eighty-six working days in order to dig a well.* With other household expenses, it would probably take years to accumulate sufficient capital necessary to dig a well. Further, accumulating this amount of capital would be an impossible task for a subsistence agriculturalist with limited access to cash. Consequently, very few households could afford to dig a well, and several shared one. The relative scarcity of wells in the south thus had implications for the distance that water was carried to the house. Since pottery vessels brought water to the household from a greater distance in the south than they did in the north, the greater use of water-carrying vessels in the south resulted in greater risk of breakage there. The amount of consumer demand for water-carrying vessels thus varied directly with the depth of the water table, and the greatest demand for such vessels existed in the cities along the hill ridge in the south, such as Tekax, Tzucacab, Peto, and in neighboring communities (Table 3.1; Figure 2.2). Changing Demand: The Collapse of Water-vessel Production. Since the 1970s, the demand for pottery for water transport and storage has decreased greatly. The major change that affected this decline was the introduction of piped water. In 1965, piped water systems had been installed in some of the cities in Yucatán, and digging the trenches for the pipes had already begun in Ticul. Realizing that piped water would bring great changes to a livelihood that was largely based on the production of vessels for carrying and storing water, I asked several potters about the anticipated effect of piped water on the demand for their pottery. One potter said that he thought that he would not be able to sell any water-carrying vessels but that the demand for water-storage vessels (tinajas, apastes, and barriles) would continue. Since pipes were laid near the ground surface, he said, the sun heated the water in the pipes, and because ceramic vessels cooled water stored in them, the demand for those vessels, he argued, would continue. The demand had already ceased, he noted, for the large three-handled jar used to carry water to cornfields (Thompson 1958:Fig. 6d, Fig. 39e) and the demand for water-serving vessels, such as pitchers (  jaras and jarrias), had slowed considerably. Another potter who regularly peddled water-storage and -carrying pots to many parts of Yucatán also believed that demand for water-storage vessels would continue. He noted that although Valladolid and Tizimín had piped *Where known, all dollar/peso equivalents are provided for the month and the year mentioned.

104

Type of pottery sold in the south Type of pottery sold in the north

Frequency sold No data No data Spring Every 8 days Every 8 days Every 8 days No data July Fiesta, Dec. 8 Fiesta, Jan. 6 No data Fiesta, Dec. 8 No data Fiesta, end of January Prior to Day of the Dead

City where sold Peto Valladolid Tekaxa Peto Peto Peto Sotuta Valladolid Izamal Tizimín Valladolid Izamal Valladolid Valladolid Acanceh

Notes: a. In spring 1965, I traveled with this potter to Tekax to sell a load of these pots. b. Information is based on a telephone conversation with the adult daughter in this household on December 24, 1994, 5:10 PM (CST).

Water-carrying pots (cántaros) 7/9/1965 7/9/1965 Water storage pots (tinajas) Water-carrying pots (cántaros) 1965 6/1970, 9/1970 Water-carrying pots (cántaros) Water-storage pots (apastes) 1970 Ritual pottery 1970 1970 Not known 7/13/1984 Plant pots (maceteros) 10/27/1984 Plant pots 11/29/1984 Plant pots (including cónicos, risados, and rectos) 8/11/1988 Plant pots (including cónicos, cubanos, apastes, and tinajas used as plant pots) and a few coin banks 12/24/94b Plant pots 12/24/1994 Not known 12/24/1994 Not known 10/29/1995 Food bowls (cajetes)

Date observed

Table 3.1. The different vessel shapes that one potter sold in the north and south of the State of Yucatán between 1965 and 1997.

How Have Demand and Consumption Changed?

water systems, their inhabitants continued to purchase many water-storage vessels (such as tinajas). Even though Peto had a piped water system, he said, the demand for water-carrying vessels there was still strong. After 1965, changes in demand for water vessels did not occur immediately because the installation was slow; implementation was resisted in some communities (Tizimín) or was suspended in others (Becal, Izamal). The reason for these reactions is unclear, but in Ticul at least, the piped water service was not free, and the cost was onerous for the poor. By June 1970, two informants had discontinued the service because it was too expensive. In the meantime, potters continued to sell the same kinds of vessels that they sold before the piped water was installed. Tekax had piped water by 1970, but many of the small towns around it did not, and Tekax remained as a significant market for water-carrying vessels. Potters thus shipped water-carrying vessels to Tekax and other communities that did not have access to piped water, such as Peto and Santa Elena (Figure 2.3). By 1980, however, informants reported that the demand for water vessels had declined precipitously. By 1984, piped water systems were widespread in Yucatán except in a few small, very isolated communities. A few water-carrying jars were still made because some residual demand existed in communities without piped water, but generally the demand for such vessels had almost totally collapsed. In November, an elderly Ticul potter had gone to Peto to sell food bowls for the Day of the Dead rituals, and clients from the nearby communities of Tixmehuac, Chacsinkín, and Shoy had requested vessels for carrying and storing water. These communities did not have piped water because the water table there was seventy-five meters below the surface and digging a well was very expensive. Consequently, a single well served each community and water had to be carried a considerable distance. To meet this demand, the potter made the jars for storing water, asked her son to make jars for carrying water, and then returned to Peto to sell them. In spite of the demise of the demand for water-carrying vessels in all but a very few communities in the south near Peto, the demand for water-storage vessels continued in the north. The potter who had produced storage vessels for this area in the 1960s continued to produce them for Tizimín, Sotuta, and Valladolid in 1984 (Table 3.1). Water-storage vessels continued to be used in Ticul at this time to store water at the pila for washing clothes and to cool drinking water from the piped water system (Figure 3.7). Some demand for water-carrying jars continued from 1984 to 1997. One or two water-carrying jars were observed being sold in the Ticul market in 1984 and 1988 just as they were in 1965 and 1968. Similarly, the same elderly potter who had made water-carrying jars in 1984 continued to do so in 1988 and 1994. 106

How Have Demand and Consumption Changed?

Figure 3.7. The range of the materials used for containers in a Ticul kitchen in 1984.

Different materials have different advantages and disadvantages. Metal vessels are in the left foreground and the center background. Above the metal pot on the left is a ceramic plate leaning up against a pottery water jar (with a metal cover) used for cooling water. In the center foreground is a large gourd bowl (left) and a smaller gourd bowl (right). Behind them is a thermos bottle for keeping water hot. To the right of the small gourd bowl is a stone bowl on which are a square piece of cement tile and a ceramic dinner plate. On the far right is a pottery candle holder.

By 1997, one of her sons continued to receive requests for such vessels from the pottery stores along the highway. Ritual Vessels: Continuity and Conservatism

Unlike the changes that have occurred with water-carrying and waters­ torage vessels, the demand for ritual vessels has continued throughout the years from 1965 to 1997. This continuity is related to the Day of the Dead rituals, and demand for these vessels has remained strong. Furthermore, the shapes made for these rituals have changed little during the last half of the twentieth century, except that in 1951, whistles were made in Becal and Maxcanú (Thompson 1958:135–137 [fig. 47]), whereas in 1984, they were made in Ticul and painted differently. The Day of the Dead rituals take place on October 31, November 1, and November 6 when the Yucatec Maya remember their dead relatives. Although these rituals occur on feast days in the traditional Christian calendar, they are not really Christian but reflect pre-Christian roots. 107

How Have Demand and Consumption Changed?

Figure 3.8. A household altar for the Day of the Dead rituals for the dead children

(October 31, 1984). This altar used pottery food bowls for offering food and pottery candle holders. Modern ceramic plates and bowls also contain food, and plastic vessels are used for flowerpots. A gourd bowl (out of sight to the left) holds water.

The offering of food is central to these rituals, but pottery objects such as bowls (cajetes; Thompson 1958:106 [fig. 30b]), incense burners (Thompson 1958:49 [fig. 14a], 110 [fig. 32c]), candle holders, and flower vases are also used (Figure 3.8). Pottery in these rituals appears to be an important symbol of the departed ancestors because it links the present to the past through vessels made of materials that come from the location of their departed ancestors, the earth. The first day (October 31, or All Souls Day in the Christian calendar) honors the household’s dead children and is symbolized by serving food in small bowls made from gourds (  jicaras) or pottery. Sometimes pottery whistles are given to the children of the family on this day to remember their dead siblings. The second day of the ceremonies occurs on November 1 (All Saints Day in the Christian calendar) when the dead adults are honored. For this ritual, food is served in large pottery food bowls and large gourds. The last day of these ceremonies occurs on the seventh-day anniversary (November 6) of the first day of the rituals, when the household may serve a special variety of tamales (called espelón) on large clay plates.

108

How Have Demand and Consumption Changed?

Tradition dictates that a family should buy new objects of pottery annually for these rituals. A family does not need to buy all their objects each year, but they should at least buy bowls (Figure 3.8). Each family usually requires six to eight bowls that consist of two different sizes. After the rituals are over, some households continue to use the bowls and others give them away. As a result of the value that requires new pottery for these rituals, a great seasonal demand exists for food bowls each year and, to a lesser extent, for whistles, candle holders, incense burners, and flowerpots. Some potters begin making bowls as early as July and most begin production by September. The demand for these vessels is so great that any potter who produces food bowls can be assured of clients to buy them. Consequently, potters produce as many vessels as they can. One potter said that his household usually produced about 600 food bowls for these rituals each year. In 1984, however, his actual production consisted of 80 to 95 incense burners, 100 candle holders, 175 whistles, 520 food bowls, and 20 to 37 flowerpots of three different varieties. The seasonal demand for food bowls and other pottery accouterments for these rituals is so strong that some potters who have otherwise abandoned the craft also make food bowls during this time. Daughters of potters who have married non-potters and moved out of the household may also participate in production. The demand for ritual vessels thus stimulates a significant, but seasonal, increase in production intensity and scale. Making ritual pottery for the Day of the Dead has several advantages for nascent potters who make pottery seasonally. First, pottery such as food bowls, incense burners, and flowerpots requires only minimum skill. Second, since these vessels are small, they can be stacked and nested and thus have a limited drying and storage footprint. Further, these vessels can be transported easily by bus or train and can be sold directly to the consumers with immediate cash returns. Potters thus can control the marketing of these vessels themselves. The demand for pottery for the Day of the Dead in towns outside of Ticul can be especially strong. In 1970, for example, a food bowl could be sold elsewhere for 20 to 400 percent more than its Ticul price. This markup, of course, does not take the cost of transportation into account, but even so, increased prices make selling food bowls outside of Ticul lucrative. Probably the most important demand for the Day of the Dead pottery exists in the Oxkutzcab market, located twenty kilometers southeast along the highway to Peto. When the demand in Ticul is too poor, selling ritual pottery in Oxkutzcab can bring lucrative returns. Most of the demand there is for food bowls, but in 1984, one potter sold a few flowerpots there as well. During the first week of October 1984, one potter’s children sold about sixty food bowls 109

How Have Demand and Consumption Changed?

there each day. By October 8, sales of pottery had slackened because of competition among sellers, but by October 29, his children and mother-in-law sold another sixty bowls that they had taken there. By the first day of the Day of the Dead rituals (October 31), eight potters were selling their food bowls in Oxkutzcab. Other distant markets also have a strong seasonal demand for food bowls. One lies beyond Oxkutzcab in Tekax. In 1984, one potter reportedly sold seventy­five food bowls there in only fifteen to thirty minutes, and from his reports, four other potters went there to sell food bowls. When another potter tried to sell bowls in Tekax in 1996, however, he discovered that he needed a license. As a result, he did not sell many. Peto is another distant market with strong demand for Day of the Dead pottery, but it is so distant that few potters traveled there. In 1984, however, one potter had gone to Peto on October 28 and sold bowls to her long-term clients. She returned through Tzucacab and sold some there as well. Even though the demand for the Day of the Dead ritual pottery changed the least between 1965 and 1997, it has changed nevertheless. In 1970, one household purchased six small food bowls, six large food bowls, one or two candle holders, an incense burner, and a cooking pot. By 1984, however, cooking pots were no longer used because they were no longer being made. Second, the demand for ceramic whistles and flowerpots had developed; neither of these items were mentioned as important in 1970 for the Day of the Dead. Finally, in the late 1960s, only two sizes of food bowls were used. By 1997, however, as many as ten sizes of such bowls were made for the Day of the Dead rituals. One potter said that his clients in Oxkutzcab, Peto, and Tekax often wanted smaller sizes than they did previously—particularly if a child had died; consumers in cities such as Tizimín and Valladolid also wanted more varieties of food bowls than they did previously. Cooking Pottery: Replacement

Between 1965 and 1997, metal vessels replaced clay cooking pots. This replacement resulted from some of the massive cultural changes that began in the early twentieth century. As Mexico industrialized and aluminum cooking vessels became widely available, they were quickly adopted because they were more durable than clay vessels. Clay vessels were less expensive than metal ones, but potters said that metal pots were still widely adopted. As a result, the demand for cooking pots dropped and many potters stopped making them. In 1951, Raymond Thompson (1958:19–20, 115) reported that twelve potters made cooking pottery. By 1965, however, the production of cooking pottery was 110

How Have Demand and Consumption Changed?

almost nonexistent and there were only three potters who made any cooking pottery at all. The two makers of cooking pots that Thompson observed (Thompson 1958:19–20) were still active, but they made little of this ware. By 1971, both had died. In spite of the demise of cooking pottery between 1951 and 1965, a seasonal demand for clay cooking pots remained in the late 1960s because they were used to prepare food for the Day of the Dead rituals. Traditionally, these rituals required clay vessels, and although the demand for clay bowls and incense burners continued unabated, metal cooking vessels eventually replaced clay ones for preparing the food for these rituals. Some residual demand for clay cooking pots continued during the late 1960s, however, and sporadically throughout the thirty-two years of this study. To make up for the loss of demand, some makers of cooking pottery turned to making non-cooking pottery. In one family of cooking-pot makers, the oldest son had switched to making non-cooking pottery as early as 1950, and the youngest son switched by 1965. Another son changed to making bowls of noncooking pottery seasonally for the Day of the Dead rituals, but otherwise abandoned the craft. In 1966, the fourth son was still making cooking pottery in his house, although he spent most of his time making non-cooking pottery as a hired laborer in other production units. He eventually abandoned making cooking vessels entirely. By the late 1960s, all other makers of cooking pots had turned to other occupations. Four such potters sustained themselves through traditional slash-andburn agriculture, and two of these supplemented their subsistence agriculture by mining and selling clay. By 1984, there was no evidence that any cooking pottery was made in Ticul at all—even for the Day of the Dead rituals. The abandonment of cooking pottery over time can be seen in the changes in the number of potters making this ware. In a database of 389 adult potters from five generations assembled from 1984 data (the Potters’ Database), forty-seven (12 percent) had made cooking pottery at some time during their lifetime. By 1984, however, almost all of these were deceased and most of the remainder made no pottery at all. By 1988, the demand for clay cooking vessels revived a bit and one or two potters made them for special orders. One of these potters said that he had made a few cooking pots, tortilla griddles, and casserole vessels (cazuelas) for merchants who asked for them. This small demand continued in 1994 when two former makers of cooking pottery were making perforated griddles (called anáfres) for heating food. Both were working in one of the large production units in Ticul but made cooking pottery in their home on evenings and weekends. 111

How Have Demand and Consumption Changed?

This slight demand for cooking vessels continued in 1997, but they were usually made only on special order. The only producer of this ware was a potter who regularly made non-cooking pottery. He bought the temper for cooking pottery from one of the former makers of cooking pottery and his son, and made vessels for cooking turkey in an underground oven. He complained, however, that he could not get enough temper to satisfy the demand for these vessels. He had sold all of them at a fiesta in Hocabá and had requests for more. The great decline in the production of cooking pottery is also reflected in the kiln surveys carried out between 1965 and 1997. A different kiln is used for cooking pottery than that used for non-cooking pottery because the door must be larger in order to place pots in the kiln during firing and then remove them rather than waiting until the firing process is finished. In 1965, a survey of kilns revealed only two out of twenty kilns in Ticul were made to fire cooking pottery. When the owners of these two kilns died in the early 1970s, their kilns were abandoned and never used again. By 1988, one of these kilns (which had long since collapsed) was dismantled and the rocks were used to build a kiln for noncooking pottery. No kilns for cooking pottery were found in the kiln surveys of 1984 and 1997. The occasional cooking pottery made between 1965 and 1997 thus appeared to be fired in a kiln for non-cooking pottery. Coin Banks: Continuity

A fourth kind of pottery consumed by the local Yucatec population consists of coin banks (alcancías) that were introduced in the 1940s along with the vertical half molding technology used to make them. These shapes consisted of cartoon characters (e.g., Mighty Mouse), animals (e.g., swans, ducks, horses, pigs, and bulls), and barrels of habanero chili peppers. By 1965, these banks had become a significant component of pottery with fifteen potters in twelve households devoted almost exclusively to their production. Although figurines in the shape of banks were produced during the entire period between 1965 and 1997, their popularity appears to have diminished greatly relative to other types of pottery. It is difficult to know the reason for this change, but assuming that coin banks were used to save money, the high inflation and the peso devaluation since late 1976 (Bureau of Census 1978; Meyer et al. 1999:649–651) would have greatly diminished the buying power of savings, eliminated cash accumulation as a way to accrue capital, and perhaps lessened the demand for coin banks. Some new demand, however, emerged for coin banks as tourist curios and as an advertising medium. In 1984, for example, it was reported that the Banco Internacional in Mérida gave a coin bank made of pottery to anyone who opened an account. 112

How Have Demand and Consumption Changed?

New Demand and New Uses of Pottery Pottery for Tourists

A second consuming population for Ticul pottery consists of the tourists who visit the Colonial city of Mérida, archaeological sites such as Uxmal and Chichén Itzá, and the Caribbean resorts of Cancún, Isla Mujeres, and Cozumel. The demand from this population stemmed from new links between pottery and social values that began in the 1970s. One such link was the materiality of the exotic Other; tourists wanted to take home a symbol from their trip to the Land of the Ancient Maya. Such a symbol, however, had to be small and fit into a suitcase. Consequently, by 1984, tourists had become one of the greatest sources of demand for Ticul pottery, and this demand continued through 1997. The second link between pottery and values was also related to tourism but marked a shift in demand from the Yucatec Maya populations to both local and nonlocal elites who wanted to use pots as containers for plants. Managers of tourist hotels also used plant pots for decorative accents in the lobbies, restaurants, and hallways of their facilities. History of Tourist Demand. Tourists were first tapped as a market in 1956 when potters were hired for a workshop at the tourist hotel at the archaeological site of Uxmal (Hacienda Uxmal). They were supervised by an American woman who managed the hotel’s souvenir shop. Formerly a teacher in an art school, she was the source of several innovations that made Ticul pottery more attractive for tourists. Although some vessels were produced using the natural color of the traditional slips, she introduced post-firing, water-based matte paints and taught the potters to use these paints in innovative ways, guiding their use of color and design (Figure 3.9). Potters also learned to make new shapes by modeling and molding. One such innovation consisted of a miniature crèche set with Maya-looking figurines and angels dressed in the traditional Yucatecan costume. Other objects included salt and pepper shakers molded in the shape of a Maya-looking man and woman and painted with the traditional Yucatec Maya costume and objects made in blackware in the form of the Maya guardians of the fields (called the alux; see Redfield and Villa Rojas 1962:119–121). Plates for wall decorations were also produced using innovative colors and designs. In 1969, potters were hired at another workshop at a new Uxmal hotel (Hotel Principe). Production, however, was not geared to tourists but rather to the needs of the hotel itself and to other hotels in its corporate chain. Demand for vessels from its sister hotel on Cozumel became particularly critical because of its growth as a tourist resort and the difficulty of transporting pottery there. 113

Figure 3.9. A pottery image of the Yucatecan mestiza used as a decoration at the

Hacienda Uxmal, a hotel at the archaeological site of Uxmal in 1997. Potters learned to fabricate and paint this kind of vessel when a workshop was attached to the hotel from 1956 to about 1982. Miniatures of male and female Yucatecans painted in this style are sold to tourists as salt and pepper shakers. (Photo by Michelle R. Arnold)

How Have Demand and Consumption Changed?

Vessels often arrived broken. So the corporation sent three potters to Cozumel to make vessels to meet the hotel’s container needs, and the potters worked there for four months. During the twenty-three-year duration of the workshops at the Hacienda Uxmal between about 1959 and late 1982 and at the Hotel Principe between 1969 and 1971, approximately nineteen Ticul potters cycled through them, learning new shapes and new decorative techniques. When they left, they took their new knowledge and skill back to Ticul, and their experience at Uxmal preadapted them to produce many new shapes for the tourist demand that emerged in the late 1970s. One potter, for example, who learned to paint pottery using the new designs during the seven years that he worked in Uxmal during the 1960s, then worked in a production unit in Ticul, using the knowledge and skills learned at Uxmal. By 1994, he and his son had established their own store. By 1997, their production unit was thriving and he had acquired a new store along the highway to sell his wares. Over the thirty-two years of this study, seven other potters who worked at Uxmal formed their own independent production units, using the knowledge and skills gained in the Uxmal workshops. Apart from the pottery made for tourists at the hotels at Uxmal, no tourist demand existed for Ticul pottery between 1965 and 1970. In June 1970, the Fund for the Promotion of Tourism Infrastructure (INFRTUR) officially chose Cancún, a sand peninsula on the eastern coast of the Yucatán peninsula, as the site of Mexico’s resort of the future. Formerly, the only resorts on the Caribbean coast were located on the islands of Isla Mujeres and Cozumel (Alisky 1980:259; Baklanoff 1980:230–232). By the early 1970s, the market for water-carrying and -storage vessels had collapsed, and the government established a workshop in Ticul to help the potters. Two kinds of innovative techniques were introduced. One of these built on the mold-making technology that was introduced in the 1940s but focused on making replicas of ancient Maya masks, figurines, and vessel shapes. A second innovation consisted of painting copies of ancient Maya designs on pottery, but this task required an investment of capital to purchase the paints and books from which the designs were copied (Figure 3.10). These innovations formed the basis of the potters’ subsequent engagement with, and response to, tourist demand. The first facilities in Cancún were operating by 1975 (Alisky 1980:259) and Cancún began to emerge as a new market for native Mexican craft products. By 1983, a student at the School of Anthropology at the University of Yucatán reported that there were eighty-three artisan stores in Cancún, and according to the Bank of Mexico, tourists spent an estimated $165 million there. Eighteen 115

Figure 3.10. Vessels with copies of ancient Maya designs in a Ticul pottery store in 1997.

Most of the vessels painted with these designs are small vessels for tourists to take back home as souvenirs. (Photo by Michelle R. Arnold)

How Have Demand and Consumption Changed?

percent of this amount was reportedly spent for artisan craft objects (Field Notes, July 20, 1984). By 1984, painting pottery with Maya designs was already entrenched in the hands of specialists who generally were not potters but rather were those who could trace their learning back to those who began the original painting workshops in the early 1970s. The execution of the designs was so detailed, however, and required so much control and organization that the skill development and time required were divergent from making the pottery. Consequently, few skilled potters actually learned to paint in this way. Those who did learn this skill tended to abandon the fabrication portion of the process. Those few traditional potters who did learn how to paint ancient designs spent more time making pottery than painting it. Because some production-unit owners did not have the expertise to make pottery themselves, they bought fired unpainted blanks from traditional potters. Some production facilities, however, employed both potters and painters. Innovative Uses of Ceramics for Ritual

Although the demand for ritual pottery for the Day of the Dead ritual has endured and the ritual vessels made for it were relatively unchanged, a new type of ritual demand has emerged using the techniques and shapes for tourist pottery to make souvenirs for local weddings and baptisms. These vessels are relatively small and consist of earring holders, small flowerpots, or vessels to hold rice for throwing at the bride and groom at weddings. They are sometimes painted with copies of ancient Maya images but are lettered with the date of the event and the name of the individual (for a baptism) or the first names of the bride and groom (for a wedding). Plant Pots

One of the greatest changes in demand since 1965 comes from two populations that provided no demand prior to 1970. One of these populations consists of urban homeowners and apartment dwellers who purchased large vessels for holding potted plants in houses, porches, patios, and apartments. The second population comprises managers who purchase plant pots for restaurants and for the lobbies, restaurants, and hallways of hotels. This new demand for plant pots resulted from an innovative link between pottery and the desire to have large pieces of vegetation in living spaces. In the late 1960s, potters made an occasional plant pot, but they were small and a very insignificant ceramic product. When the market for water-carrying and water­storage vessels collapsed in the 1970s, potters turned their attention to making 117

How Have Demand and Consumption Changed?

coin banks and large plant pots. The market for coin banks did not grow, but plant pots had emerged as a major product so that by 1984, they were the predominant shapes made by Ticul potters. Rather than producing only one shape and using only one word (macetero) to refer to flowerpots, as potters did in 1965, they developed an entirely new repertoire of shapes and associated terminology. These vessels had exclusively Spanish (rather than Maya) names and were totally unknown among Ticul potters in the late 1960s. These names were descriptive and usually related to some attribute of the vessels. For example, a cónico is shaped like a truncated cone. A recto (“straight”) is a vessel with vertical sides. A risado (from risar, “to ruffle”) is a neckless jar with an outflaring ruffled rim and a tripod base. A cilindro is a cylindrically shaped vessel. A cubano has an ovoid cross-section with a rim thickened with a bolster that has a semicircular cross section. In addition, some vessels that potters produced in the late 1960s as water-storage pots (such as tinajas, apastes, and depositos) were made as plant pots in the 1980s and 1990s. The demand for plant pots continued in 1988, 1994, and 1997. Although hotels were not systematically surveyed, Ticul plant pots were used in at least two hotels in Mérida in 1988. In 1997, a brief reconnaissance along the Paseo de Montejo in Mérida revealed Ticul plant pots in abundance in three hotels, two bank lobbies and patios, and one house. The calcite in the temper and resulting calcination from firing also hastened the demise of plant pots, like water pots, and created a sustained demand for them. Some hotels solved this problem by painting the pots or covering them with lacquer. This practice prevented the penetration of the water and roots into the fabric of the vessel, inhibited the absorption of water, and temporarily prevented the rehydration of any calcium oxide produced from calcination. Even so, it appears that replacing plant pots was a less labor- and capital-intensive method of adjusting to calcination than painting, and potters reported that the Cancún hotels, at least, replaced their plant pots every eighteen months. Besides changes in demand from water-carrying and water-storage vessels to plant pots, more subtle changes have also taken place in what could best be called the “style” of the shape of plant pots. When I returned to Yucatán in 1997, I walked down one of the grand old streets (the Paseo de Montejo) in Mérida, and I noticed many different shapes of plant pots in banks, hotels, front yards, and porches. Since they were not similar to any shape that I had seen being made in Ticul previously, I assumed that they had been imported from elsewhere and not made locally. When I went to Ticul a day later, however, I saw many of the same vessels being sold there. Potters had been producing new styles of plant pots that were different from those produced even three years previously. 118

How Have Demand and Consumption Changed?

Hotel managers have also provided a demand for ashtrays and their production also has increased from 1965 to 1997. Ashtrays were made in the 1960s, but they were produced in abundance in 1997 because tourists had a habit of taking ashtrays from hotel rooms as souvenirs of their trip. As a result, one potter painted the name of the hotel on ashtrays in 1988 so that they provided advertising for the hotel when their guests pilfered them. By 1997, another potter specialized in making mold-made ashtrays with the logo of the hotel embossed on them. Other potters made ashtrays with an image of an ancient Maya figure from Chichén Itzá (the reclining Chaac Mool) in low relief. Cycles of Demand and Their Changes Changes in demand did not just increase during the period from 1965 to 1997 but also fluctuated in at least two kinds of cycles. One cycle was an annual cycle that was seasonal and predictable. The first kind of annual cycle consisted of the demand from the local Yucatec population for vessels for the Day of the Dead rituals. This demand peaked in the weeks before the first week in November but was anticipated as early as July or August when potters began making these vessels. This kind of cycle has been documented as occurring consistently since 1965 except for lack of demand that results from the consequences of hurricanes. A second kind of annual cycle began in the late 1970s or early 1980s and consisted of the demand from tourists and hotel managers, with peak demand from November to January and again during Holy Week. Plant pots are in demand at this time because hotels and homes change these vessels before Christmas. Many Mexicans have fifteen days of vacation at this time and it is also the high season for the tourist industry when tourists come to Mexico to escape winter in North America and Europe. A second cycle of demand was unpredictable, was tied to the hurricanes, and occurred in multiyear cycles. After a devastating hurricane in Cancún, the demand for pottery can drop severely. Such a downturn occurred after Hurricane Gilbert severely damaged Cancún in 1988. Potters said that clients would take pottery on consignment and then not pay for it. By 1994, the tourist industry in Cancún was still recovering from storms during the period from 1988 to 1990. Potters could not find clients and many potters temporarily abandoned the craft. After Hurricanes Opal and Roxanne passed over Yucatán in fall 1995, I called one of my informants by telephone to ask how they had fared. His daughter said that the storm did not affect Ticul directly, but it had affected Tekax and Cancún and there had been a lot of destruction—particularly farther to the south and east of Ticul. The roads to the south had been cut by the storms and trucks were 119

How Have Demand and Consumption Changed?

not getting through from central Mexico. As a result, the price of merchandise had risen. Besides creating problems in transportation infrastructure, the storm had caused a lot of destruction in the agricultural sector, and the harvest of fruit, vegetables, and maize had been poor. Therefore, the cost of food had risen and there was little money. During my visit two years later, it appeared that the economy had rebounded and there were no signs of a lack of demand for pottery. The number of production units had grown from 1994 and pottery was being made in abundance. Besides tourist demand, hurricanes directly affect the demand for other pottery as well. The Day of the Dead rituals, for example, occur near the end of the hurricane season, and the effect of hurricanes on both agriculture and making pottery can be devastating. During the phone call in 1995 mentioned previously, I learned that a potter had taken 1,000 food bowls to sell in Acanceh for the Day of the Dead rituals, but he had sold none of them because no one wanted to spend the money to buy them; consumers had to forego the tradition of buying new bowls for the ritual. The hurricane and the resulting poor harvest had resulted in no demand for pottery—even for an event as important as the Day of the Dead. Similarly, in September 2002, Hurricane Isadore virtually eliminated the demand for food bowls for the Day of the Dead rituals that year. First, the maize fields had been destroyed. The ears had not matured, and since maize is the most important subsistence crop, the effect was devastating for swidden farmers. Second, the hurricane had devastated the citrus crop, causing the fruit to fall before it was ready to pick. Since picking fruit is labor-intensive and provides employment for many, no jobs were available in the citrus sector of the economy. As a result, one informant said that he did not make many food bowls for the Day of the Dead because people could not afford to buy them. Quantitative Measures of Change in Demand Although the most holistic way to describe the change in demand for pottery vessels from 1965 to 1997 consists of the narrative just recounted, some quantitative measures illustrate these changes. The first of these measures consists of the inventories of potters’ kilns in 1965 and 1984 (Table 3.2). In 1965, inventories were made of the fired vessels from the kilns of two different production units and inventories were made of one of these kilns at two different times. In 1984, an inventory was made of the kiln in one of these same households. These data indicate that the only shapes made in 1965 that were similar to those made in 1984 were coin banks. By way of contrast, the inventory data indicate that the kinds 120

How Have Demand and Consumption Changed?

Table 3.2. Kiln inventories from two potters in 1965 and one in 1984. Kiln inventories from Potter B were counted in both years. Potter B Potter A Potter A Potter B (3/26/65) (4/2/65) (4/21/65) (8/31/84)

Class of vessels Vessel shape Water storage

apaste

12



tinaja

1



barril

3

13

7

Water carrying

cántaro ( p’uul)

67

53

8

Water serving

pitcher ( jarría)

7

35



pila (for animals)

14

4

3

14

2

5

119

152

Plant pots

17

Ashtray Ritual (serving)

cajete (food bowl)

7

2

33

Uncertain shapes



5

Coin banks

various shapes

5

113

Souvenir vessel

triple flowerpot

220



cantarito

82



chocolatero (cylindrical vessel)

34



bola (ball-shaped vessel)

31



alajero (to store earrings)

10

of pottery made in 1965 and 1984 were radically different, with almost a total replacement of traditional utilitarian vessels with those produced for tourists. A second quantitative measure of the changes in demand is illustrated by the kinds of vessels fabricated in 1965, 1966, 1970, 1984, and 1997 (Table 3.3). Although the data are not complete relative to all of the production units for each period of observation, they do document the shift from water-carrying vessels made before 1970 to the production of plant pots and small tourist vessels in 1984 and 1997. Third, a complete inventory of the vessels of one production unit was made over a period of five months in 1984 (Table 3.4). This unit was one of the same units in which a kiln inventory was made in 1965 (Potter A in Table 3.2). These data also document the almost complete shift of production from vessels for water storage, transport, and serving to plant pots and vessels for tourists. Conclusion Social change has dramatically affected the demand for pottery from Ticul. Originally, the geology, hydrology, and climate of Yucatán affected the demand 121

Table 3.3. The number of production units producing particular shape classes and vessel shapes in 1965–1966, 1970, 1984, and 1997, based on surveys, informal interviews, observations, photographs, and field notes. Most production units produce more than one shape and class of vessel (e.g., Tables 3.2 and 3.4). Therefore, the total households for each class of vessels do not necessarily equal the sums of the number of all households that make the vessels in that category. Water-­storage vessels made in 1984 and 1997 were probably used as plant pots (see text). Vessel class

Vessel shape

Water-carrying vessels

Total production units cántaro

1965–1966

1970

1984

1997

11 11

7 7

a

4 4

2 2

Water-storage vessels

Total production units 7 tinaja 7 apaste 5 deposito 1 barril 7 tinajita suptido (small cántaro) cantarilla 2

7 5 4 3 1 1 1

10 6 3 1

2 2

Water-serving vessels

Total production units pila pitcher (  jarra, jarria, and jarron)

4 2 2

2 2

Ritual vessels (Day of the Dead)

Total production units 7 3 1 3 incense burner (incensario) bowl (cajete) 6 1 2 candle holder (candelero) whistle

12 2 8 2 1

Cooking vessels

Total production units

3

4

1

Coin banks

Total production units

7

2

4

Other mold- made figurines and objects

Total production units

3

13

Flowerpots (  floreros)

Total production units 4 1 florero 4 1 risado bola tibor

4 3 3 1 1

Ashtrays Plant pots (maceteros)

5 5 4

1

2

Total production units 3 columna 1 apaste cubano recto cilindro cónico

14 3 3 4 7 1 2

1

3 2 2 1

1

2 2

6

continued on next page

Table 3.3—continued Vessel class

Vessel shape

1984

1997

Souvenir vessels

Total production units plato tiborón chocolatero cantarito vaso cubanito cónico

9 2 1 1 2 1 1 1

4

Pre-Hispanic copies

3

2

38

16

Total production units with data

1965–1966

16

1970

13

Note: a. Although the vessels listed here are those formerly used for carrying water, only one of these units produced these vessels for that purpose. Rather, these vessels were either used as piñatas at Christmas or painted with ancient Maya figures and sold as tourist curios.

Table 3.4. Vessel classes and shapes produced by one production unit between July and late November 1984, including most—but not all (see p. 109)—vessels produced during the period. Water-storage vessels are used as plant pots in this case. Flowerpots and souvenirs were unslipped “blanks” to be painted with copies of ancient Maya designs by the broker who purchased these vessels. Vessel class Vessel shape Water-storage deposito vessels (used as plant pots) tinaja apaste

Reference for shape (age and figure numbers from Thompson 1958)

Number of vessels produced

122 (fig. 38i) 50 (fig. 15f ) 40 (fig. 7d) 116 (fig. 36l)

4

187

1 1

Ritual vessels

cajete (food bowl)

46 (fig. 12h)

Coin banks

alcancia de barril (“barrel shaped”)

138 (fig. 48e)

Flowerpots

risado bola tibor bulbous flowerpot (without tripod) flowerpot (with tripod)

5 53 168 47

Souvenir vessels

cónico cantarito 36 (fig. 5d, e) chocolatero plato vaso

6 117 101 611 78

6

10

How Have Demand and Consumption Changed?

for vessels for carrying and storing water. With the development of municipal piped water in the 1960s, however, the consumer demand for water-carrying and -storage vessels collapsed. After 1980, some demand for such water vessels continued in the remote portions of the peninsula that did not have piped water but only at a tiny fraction of its demand previously. Meanwhile, the use of ceramics for rituals during the Day of the Dead ceremonies managed to sustain demand for ritual pottery, such as food bowls, incense burners, whistles, and flowerpots, because food bowls, at least, needed to be replaced each year. The industrialization of Mexico and the availability of metal cooking vessels in the first two-thirds of the twentieth century led to the replacement of clay cooking vessels with metal ones. Even the value of using cooking vessels to prepare food for the Day of the Dead, however, was not enough to sustain demand for this kind of ware, and it has largely disappeared. Another way to describe this change in demand is to look at the changes in the populations who have consumed pottery between 1965 and 1997. During this period, consumption moved from the local Yucatec Maya population to urban dwellers and tourists. Vessels used for cooking, water storage, and coin banks for the local population have shifted to plant pots and small vessels painted with copies of ancient Maya designs. Even with massive social change, however, a consistent and continuing seasonal demand still existed among the native Yucatec Maya population for ritual pottery for the Day of the Dead, and such ritual vessels were the only vessels that changed little across the thirty-two-year evolution of production. The same ritual vessels produced in the late 1960s were also produced in the 1990s. Costin (1991) describes demand for craft products as a part of the “context” of production. For independent specialists, such as those who live in Ticul, the use of “context” for demand is appropriate because in the late 1960s, vessels were made for the consumption of the local Maya population. But Costin’s definition of “context” should also include geological and climate considerations because the scarcity of water produced a critical demand for carrying, storing, and serving water. These uses are covered by Costin’s general description of “context” for independent specialists, but the intensive use of water vessels shows the importance of local environmental conditions affecting the demand, production, and use of pottery. Furthermore, notions of context need to take into consideration the technological factors (such as calcination and color) of the vessels themselves that also stimulate demand. Potters in Sacojito, Guatemala, will alter the color of the surface of their pots by spreading volcanic ash on them during firing so that buyers will not think that the pots are unfired. Similarly, potters in Ticul generally avoid producing blackware for the same reason. Notably absent from 124

How Have Demand and Consumption Changed?

Costin’s definition of “context” is a discussion of the production of ritual vessels by independent specialists, although she included it in later syntheses. Local production of ritual vessels by independent specialists also occurs in the Terminal Classic and Early Postclassic periods at the ancient Maya site of Lamani (HowieLangs 2006). In many respects, the changes in demand for pottery vessels clearly reflect social change. The replacement of pottery cooking pots with metal ones reflects the industrialization of the Mexican economy. Furthermore, the increased modernization that comes with piped water is reflected in the almost total demise of pots for carrying and storing water. Finally, the economic importance of tourism in Yucatán is reflected in the predominant demand for souvenir vessels that fit into suitcases as material symbols for tourists who visit the Maya area. The demand by elites for plant pots to hold tropical vegetation in restaurants and hotels reflects the value of bringing this vegetation into the comfortable surroundings of hotels, restaurants, houses, and patios. This value represents a kind of domestication of the jungle from its perceived “wild” state. Finally, counter to this trend of great change in demand, those pots used in religious rituals and the demand for them persist among the local Yucatec Maya population on a seasonal basis from year to year. All of these changes reflect the role of demand as a feedback mechanism in the choice of vessels that the potters produce. They illustrate the importance of the feedback from demand in general in ceramic production and in technological change, as I elaborated in Ceramic Theory and Cultural Process (Arnold 1985:127–167) and was further elaborated by Costin (1991, 2005) and others. The second dimension of demand—putting vessels into the hands of consumers—consists of patterns of distribution. This dimension is also part of Costin’s context parameter and will be the subject of the next chapter.

125

Chapter

How Has Distribution of the Pottery Changed?

four

D

ecisions about making pottery are affected not only by demand but also by the patterns of distribution. So in addition to assessing the demand for ceramic vessels among populations that value, desire, and consume those vessels, the potter must get them into the hands of these populations. This task is one of the most significant problems for the potter who is dependent on his craft for a livelihood. It is one problem that persisted through the duration of this study. Again and again, potters talked about the difficulty of selling their pottery; it is the single most significant reason that potters abandon the craft, with excessive competition and the difficulty of finding clients given as instrumental reasons. This chapter details those mechanisms that place pots into the hands of consumers. Just as the demand for pottery has changed, so have the methods of distributing it. First, in this chapter I describe the transportation infrastructure and its role in the distribution of pottery. Then, I enumerate the changes in that 127

How Has Distribution of the Pottery Changed?

infrastructure that occurred between 1965 and 1997. Finally, I detail the ways in which the distribution of pottery has changed from 1965 to 1997. Changes in Transportation Infrastructure Between 1965 and 1997, the methods of distributing Ticul pottery were closely linked to the transportation infrastructure of Yucatán. Just as this infrastructure has changed, so has the distribution of pottery. These changes have provided access to new consuming populations that are linked to the expansion of the highway system in Yucatán and to the availability of capital necessary to acquire the services of a truck. The expansion of the transportation infrastructure has not only provided a means to get pottery into the hands of new consumers but has provided deviation-amplifying feedback for production that has stimulated technological and decorative innovation, increased production scale and intensity, and provided selective pressures for certain kinds of vessels. In short, the transportation infrastructure has provided one of the most significant stimuli for the great changes and expansion of Ticul ceramic production and distribution during the last half of the twentieth century. Since transportation is so critical for the movement of quantities of pottery, a historical review of its changes provides insight into changes in its distribution. Up until the development of the railroads and highways, travel was limited, difficult, and much more time-consuming than it is today. In the nineteenth century, the major transportation arteries were unimproved wagon roads (Dumond 1997:128) traversed by horse-drawn carts that required a change of horses every forty kilometers. This limited transportation existed in some places up until the beginning of the twentieth century, placing great restraints on the distribution of pottery and thus constraining production because of limited access to nonlocal consuming populations. Railroads provided the first public mass transportation. Originally built to transport processed henequen fibers to the ports of Progreso and Sisal, all of the major cities of Yucatán were linked by railway lines by the beginning of the twentieth century (Figure 4.1). With this development, potters had access to large consuming populations, and the market for the pottery expanded because potters could ship their wares to distant markets and sell it themselves. As recently as 1951, the railroad was the most important form of transportation in Yucatán (Thompson 1958:10). Because of the extensive network of railroads, paved highways did not become a significant form of transportation for pottery until relatively recently. The first highway was built from Mérida to nearby Kanasin in 1922. Subsequently, 128

How Has Distribution of the Pottery Changed?

Figure 4.1. Map of the northwestern part of the Yucatán peninsula showing the princi-

pal railroad lines. Most of these routes remained intact until the 1990s except for the line from Mérida through Muna to Ticul, which was torn up shortly before I arrived in 1965. The line from Acanceh to Sotuta was abandoned by the late 1990s according to a Web site of the Yucatán railroad club. (Map drawn by George Pierce)

highways linked Mérida and archaeological sites such as Chichén Itzá (Orosa 1994:268). In the 1940s, the highway system began to expand and employed many workers, including some Ticul potters. By 1951, paved roads were still very limited, with highways only between Mérida and the port of Progreso, northeast to Motul and Izamal, and east to the ruins of Chichén Itzá (Thompson 1958:10). By this time, the road to Uxmal had been extended to the town of Santa Elena and through the Yucatán hill country to Bolonchén, Hopelchén, and Campeche, where it linked up with the coastal route to central Mexico (Thompson 1958:10). Even with some highways and extensive railroads, however, Yucatán was relatively isolated from the Mexican heartland until the 1970s. In 1965, my trip to Yucatán by road required five days of travel from Mexico City. After leaving the mountains from central Mexico, the road followed the gulf coast, crossing rivers, estuaries, and coastal lakes by ferry. Traffic was frequently backed up at ferry crossings and often one had to wait for hours to cross. 129

How Has Distribution of the Pottery Changed?

Figure 4.2. Map of the northwestern part of the Yucatán peninsula showing the princi-

pal roads between about 1970 and 1990 and the towns connected by them. Since 1990, the Mexican government has made many changes in the highway infrastructure, such as adding a major highway between Muna and Maxcanú and changing the route of the road from Uman to Dzitbalché and from Muna to Uxmal. (Map drawn by George Pierce)

In 1965, the railroad system was declining in importance and was being replaced by asphalt highways that had become the principal transportation infrastructure (Figure 4.2). Highways linked Mérida, the economic and political center of the region, to other major cities. The highway from Mérida to Muna had been extended southeast along the north side of the hill ridge, parallel to the existing rail line connecting all of the cities from Muna to Peto. Beyond Peto, a road went to Chetumal near the Caribbean coast, but only part of it was paved. Apart from highways to the major cities, paved roadways were limited and much of the interior was only accessible by poor, unimproved roads. In April 1965, I traveled by bus from Ticul to the village of Tekit, northeast of Ticul. The road first went through Chapab and Mama. Beyond Mama, it was barely a track for a single vehicle. No attempt had been made to remove the rock outcrops or smooth out the undulating topography. Passengers often had to get out and walk so that the bus could climb steep hillocks without faltering. A measure of 130

How Has Distribution of the Pottery Changed?

the quality of the road was the three hours that were required to traverse a mere twenty kilometers. Only in retrospect did I realize that I could have walked to Tekit in a shorter amount of time. In the late 1960s, the government began to invest heavily in improving the roads and highways. Although the interior roads in the southern part of the state were still not paved, roads to many towns away from the main highways were easily accessible by relatively flat graded roads that had vastly improved since 1965. In 1967 and 1968, I drove to several interior towns, such as Tekit, Mama, Sacalum, and Maní, without any of the difficulty that plagued my bus trip to Tekit in April 1965. With the presidency of Luis Echeverria in 1970s, Mexico embarked on a program to improve the rural road system (Meyer et al. 1999:647). The highway from Mérida south to Maxcanú and Calkiní in the State of Campeche was improved, cutting off the longer journey that passed the ruins of Uxmal. This new bypass ultimately became the principal highway between Mérida and the Mexican heartland. Similarly, the road east from Mérida to Yucatán’s second largest city, Valladolid, became much more important in the 1970s with the opening of the resort of Cancún on the peninsula’s east coast. During this same period, the inland route from Villa Hermosa in the State of Tabasco to Campeche was completed, cutting off the time-consuming ferry crossings over rivers, lakes, and estuaries. Travel time to Mexico City was thus shortened, but a trip from Mexico City to Mérida in 1978 still required twenty-six hours of nonstop driving. By 1994, many of the roads into the interior were paved and trips to formerly isolated and remote interior towns could be made quickly. In contrast to the hours required to travel by bus from Ticul to the interior village of Tekit in 1965, the same distance required only about thirty minutes in 1994. Great changes came to the highway infrastructure in the early 1990s when a limited access, divided, four-lane toll road was built from a point sixty-eight kilometers east of Mérida to the resort city of Cancún. This toll road made Cancún and the resorts along the east coast of the peninsula even more accessible and provided potters with a market for anyone with access to a vehicle. With development of highways since the 1940s and their improvement in the last quarter of the twentieth century, potters had access to additional markets that had been relatively inaccessible previously. The increased use of buses and trucks on these highways provided access to new consuming populations and stimulated the growth of the craft. The most significant of these additional distant markets was Cancún, and in 1984, production-unit owners transported many loads of pottery there. One owner took pottery there every eight days. By 131

How Has Distribution of the Pottery Changed?

Table 4.1. Locations where pottery was sold between 1965 and 1997 based on surveys, observation, informal interviews, and field notes. The data do not represent a complete sample of production units. Pottery from any one production unit usually ended up in more than one sales venue. Categories and subcategories in the “Location sold” column are ranked in order of the distance from Ticul and the difficulty in transporting pottery to them. Location sold Total with data

1965–1966

1970

1984

1997

13

10

14

27

4

Local Ticul market Brokers Kin Non-kin

3

2

2

2 6

0 8

3 5

Villages of unspecified location

4

1

Cities accessible by truck and/or rail to the south or southeast Santa Elena Oxkutzcab 1 1 Tekax 1 2 Tzucacab 1 Peto 1 5 Chacsinkín (13 km from Peto by road) 1

1 1

1

2

1

Cities accessible by rail, bus, or train to the north Acanceh 1 Kanasin Mérida 1 1 1

1 12

Cities accessible by rail from Mérida Izamal 1 Sotuta 1 Valladolid 1 1 Tizimín 1 1

1 1

1

Distant cities accessible only by truck or bus to the east and south Chetumal 1 1 Cancún Cozumel Ciudad del Carmen (Tabasco)

1 9 4 1

1994, Cancún continued to be a significant and very lucrative market with gross receipts of as much as 400 percent of the cost of the pottery. By 1997, a survey of production units revealed that pottery was transported longer distances than it was in 1965. When sales destinations for pottery are ranked by increasing distance from Ticul (Table 4.1) and compared with the number of production units selling pottery at those destinations between 1965 132

How Has Distribution of the Pottery Changed?

Table 4.2. Types of marketing strategies used between 1965 and 1997. Data based on field notes and not on a complete survey or census. Type of pattern Number of production units that sold to brokers Number of production units that sold to local population at: Markets Fiestas Number of brokers mentioned Local Nonlocal Number of nonlocal locations of brokers/sales Production units for which there are data

1965

1966

1994

1997

2

2

8

23

4 4

0 1

0 0

4 0

0 0 0 9

2 0 0 3

7 3 1 8

9 11a 10 28

Note: a. Includes five painting workshops in Cancún, two painting workshops in Mérida, and the FONOPAS store in Mérida.

and 1997, it is clear that potters were selling their pottery in much more distant destinations than they did in 1965. This change reflects the importance of the improved regional transportation infrastructure. Changes in the Types of Distribution Although there are exceptions, the types of distribution of Ticul pottery have changed radically from selling directly to the local populations of the peninsula to using brokers or middlemen as intermediaries for selling pots to their consumers. Instrumental in this change has been the decrease in the pottery used by local populations and the increase in the demand for plant pots and tourist pottery. The increased distance to the consumers of their pottery and the lack of public transportation prevented potters from transporting their pottery to these new consumers themselves as they had done previously. Markets

The most persistent method of distributing pottery consists of selling directly to consumers in the markets of the cities and villages of Yucatán. With this method, potters transport their pottery themselves, using trains, buses, and trucks. In the late 1960s, it was the principal means by which potters sold their wares (Tables 4.1 and 4.2). The Ticul Market. The most readily available market to potters is the Ticul municipal market, but it was only a minor outlet for Ticul pottery and only a few

133

How Has Distribution of the Pottery Changed?

Figure 4.3. A potter in the Ticul market in 1968 selling jars for carrying and storing water. Water-carrying jars (cántaros) are at the left and center and water-storage jars (tinajas) are at the left and right background. Another water storage jar (apaste) is in foreground, center, and far right.

potters have used it consistently (Figures 4.3 and 4.4). Nevertheless, the importance of the Ticul market rises dramatically during the days preceding the Day of the Dead ceremonies. In contrast to the one or two potters who sold pottery in the Ticul market (in 1965, 1968, 1984, 1988, and 1997), the number of sellers of ritual pottery increased to nine on the morning of the first day (October 31) of the Day of the Dead rituals in 1984 (Figure 4.5). Markets outside of Ticul. Selling pottery to local consumers in the markets in Yucatán continued throughout the period from 1965 to 1997. Up until 1970, potters who sold pottery in these distant markets reported traveling every eight or fifteen days, or when a fiesta occurred in these communities. Although some potters continued to sell their pottery in this way from 1984 to 1997 (Tables 4.1 and 4.2), its importance as a method of distribution has diminished considerably since 1965. Because cities along the rail lines were easily accessible and larger amounts of pottery could be transported by rail than by bus, the railroad has served as the enduring method for distributing pottery. Railroads were used to transport pottery long before 1965 (Thompson 1958:105), and this role persisted between 1965 and 1995 (Figure 4.1). 134

How Has Distribution of the Pottery Changed?

Figure 4.4. Woman selling pottery in the Ticul market in 1984. Compare the vessels

offered to vessels being sold in 1968 (Figure 4.3). Two vessels for carrying water and two vessels for storing water are offered, but coin banks in the form of figures predominate. This woman also sold pottery in the Ticul market in 1968 and then offered more water-carrying vessels for sale. (From Dean E. Arnold, “Advantages and Disadvantages of Vertical-Half Molding Technology: Implications for Production Organization,” in Pottery and People: A Dynamic Interaction, edited by James M. Skibo and Gary M. Feinman, p. 63 [University of Utah Press, Salt Lake City]. Used with permission)

Because the cities along the northern edge of the Puuc hill ridge (Oxkutzcab, Tekax, Tzucacab, and Peto) are connected to Ticul by both railroad and highway, they have been the most accessible market niches to potters. In 1984, for example, one potter reported seeing thirty bundles of pottery being shipped to Peto, an amount impossible to ship by bus. Potters also utilized trucks to transport their pottery in the late 1960s, but trucks became more important in 1984 and later. In order to maximize marketing opportunities, potters did not always distribute their pottery themselves, but often sent a close relative to sell it. This practice helped maximize production for full-time producers by having non-potters market it. Potters also used this method to sell pottery in two different locations at the same time. One potter in 1970, for example, sent his son to sell pottery in distant markets while he made pottery at home or sold it in a different location. In another case, a man peddled pottery that his wife and his adult daughter made. He also sold pottery made by his wife’s brother, his wife’s father, and his sister. 135

How Has Distribution of the Pottery Changed?

Figure 4.5. Women in the Ticul market on October 31, 1984, selling pottery for the Day of the Dead rituals.

During sales of pottery prior to the Day of the Dead rituals in 1984, one potter sent his older children and his wife’s mother to sell the pottery in more distant towns (such as Oxkutzcab), while he stayed home to make more of it. The importance of rail transportation as a means of distributing pottery and its effect on changes in distribution can also be seen when rail services cease. In the 1940s and 1950s, the beach communities of Progreso and Chicxulub had no piped water and the wells were reportedly contaminated. Consequently, the populace drank only rainwater, creating a demand for vessels for storing water (tinajas, depositos, and apastes) and for serving water (  jarras). Plant pots (maceteros) were also in demand there. So every Friday during the months of June and July, two potters took their vessels to Progreso as people from Mérida flocked to beaches on the weekends. The potters returned to Ticul each Tuesday and then left again on Friday with more vessels. Sometime before 1965, however, train service to Progreso ceased, making transportation costs too great, and their sale of pottery in Progreso ceased. By 1997, rail transportation ceased to be an option for transporting pottery. Potters said that a Japanese company had purchased the railroad and no longer accepted pottery for shipment. This change was devastating to those potters who normally shipped their pots by rail. As a result, potters who usually shipped their vessels by train had to find another way to transport them. One potter who usually sold pots in Valladolid was able to link up with a truck owner. When this

136

How Has Distribution of the Pottery Changed?

potter needed transportation, he telephoned his son, who lives near Valladolid, and his son informed a local truck driver. The next time the driver traveled to Oxkutzcab to buy fruit, he drove to Ticul, loaded up the pots, and took them to Valladolid. For some potters, long-term consistent distribution in a particular market has resulted in the development of relationships with regular clients. Such relationships are valued by potters because they dislike soliciting new clients and prefer having a continuing and reliable outlet for their pottery. One potter successfully used such long-term relationships between 1965 and 1997. During the early part of this period (1965–1970), he sold water-storage pots in Valladolid, Tizimín, and Sotuta. By 1988, he had changed to producing plant pots and still retained a set of regular clients in Valladolid. These relationships continued into 1997 and he still traveled regularly to Valladolid to sell his vessels. In 1984, at least two other potters had relationships with regular clients. One sent pottery regularly by train to a client in Peto who had been a former resident of Ticul, and another had established relationships with other clients in Peto for selling water-carrying jars long after the market for these vessels elsewhere in Yucatán had disappeared. These relationships continued as long as clients paid for their pots. But some clients may ask that potters give them credit for the pots delivered. In distant markets, this strategy may be successful if the potter does not want to ship the pots back to Ticul, but some clients may consistently carry a credit balance and potters may ultimately refuse to sell them more pots. Fiestas

A second way that potters distribute their pottery to the local population is through traveling to local and regional fiestas. Potters began using this method by the 1950s after the emergence of the highway infrastructure when one potter, his son, and his daughter-in-law started using their pottery as prizes for games of chance that they set up at fiestas. People paid for the chance to win a prize by playing a game such as throwing a ring around a peg. By the late 1950s, two brothers in another family were also using games of chance in the same way but focused on those fiestas that were distant from Ticul. One stopped this practice in 1970 but the other brother continued until about 1980. In the late 1960s, other potters utilized games of chance to market their pottery. One active potter in 1966 had abandoned the craft entirely by 1978. By 1984, he had left Ticul and became a professional fiestero in Tabasco, traveling to fiestas with games of chance. Another potter active in 1965 also traveled to fiestas, but by 1984, he too was a fiestero and had abandoned making pottery. By 137

How Has Distribution of the Pottery Changed?

1984, other potters who had distributed pottery in this way had been so successful that they also abandoned the craft in order to travel from fiesta to fiesta using pottery purchased from others as prizes. The Ticul Fair

Potters may also use the annual Ticul fair as a means to sell their pottery. In July of every year, Ticul celebrates its elevation to the status of a city (1867) with festivities every evening and commercial booths selling food and other items. During the fair in 1967, one booth featured a potter who produced water­carrying jars to show that Ticul was a town with potters, but there were no booths selling Ticul pottery. In 1994, no potters had a booth in the fair, but by 1997, two potters had booths selling pottery and were taking orders for souvenir pottery for weddings and baptisms. Stores

An additional method for marketing pottery consists of potters selling their wares through stores. Such stores are located both in Ticul and elsewhere, and they sell pottery to the local population and tourists. Some are operated by the owners of production units and others are not. Local Stores. From 1965 to 1970, no local pottery store existed and only one potter (Enrique Garma) had any exposure to passing vehicular traffic along the highway. He did not have a store, but his production unit was located next to his house and he placed pottery on the sidewalk to attract customers. All other production units were located away from the highway in the barrios of San Enrique and Mejorada. In the late 1970s, potters began to recognize that many people passed through Ticul on their way to other communities to the south, and they realized that these people were potential consumers and constituted an important market. As a result, three production units opened pottery stores along the main highway through town. Such stores provide three important advantages to potters. First, they provide access to an important and previously untapped market niche because Ticul lies on one of the two main highways from Mérida to the east coast of the peninsula. Second, stores can provide a lucrative return. One potter said that his store can bring a 100 percent markup for his wares. Finally, potters can deal directly with customers and do not need to sell to brokers (see below), who take a substantial portion of the profit for selling the pottery. Since the late 1970s, the number of local stores selling pottery has grown dramatically, and this system has become one of the important ways by which 138

How Has Distribution of the Pottery Changed?

Table 4.3. Number of pottery stores, showing the continuity and changes in ownership and location from 1984 to 1997. The numbers in parentheses indicate the number of traditional potters who operated pottery stores or the number of individuals who married into families of traditional potters, became potters, and established pottery stores. “Traditional potters” are those who made pots in the late 1960s and come from families who have been potters for more than two generations.

1984

1988

1994

1997

Total stores

9 (5a)

8 (3)

12 (6)

16 (7)

Stores continuing from 1984

6 (3)

6 (3)

6 (3)

New stores since 1984

2 (0)

5 (3)

5 (2)

5 (3)

8 (4)

New stores since 1988

New stores since 1994

5 (2)

Continuing stores that changed location since previous survey

1 (0)

3 (2)

2 (1)

3 (1)

1 (0)

Stores for which the owner died but continued with other family members

Note: a. One of these stores did not have an inventory but was simply a potter making vessels in a traditional wattleand-daub house. It did not fit the pattern of the other stores, but was included in the total because it was located along the highway, its purpose was to sell pottery, and it was not a potter’s residence.

Ticul potters have sold their wares locally. About 1977, four stores (in addition to Garma’s production unit) were established along the highway by non-potter entrepreneurs who formerly had worked for a government workshop in the early 1970s. They realized the potential of the consumer demand by travelers who were passing through Ticul and seized the opportunity to establish a business. By 1984, the number of these stores had doubled (N = 9) and included stores operated by traditional potters who came from pottery-making families (Table 4.3). Except for Garma’s production unit, these stores also fronted workshops for painting pottery (Figure 4.6) or were outlets for painted pottery from production units elsewhere in Ticul. In most cases, pottery was fabricated in locations that were physically removed from the stores. By 1994, the number of these stores had tripled since 1977. By 1997, the number of stores had grown by 33 percent (N = 16) and were four times the number present in 1977. Although only about half of these stores were in the hands of traditional potters, most had a long continuity (Table 4.3). The development of stores required at least two kinds of resources. First, establishing such a business requires capital to either purchase or rent a facility and then money to sustain it. If the potter’s (or broker’s) financial situation becomes precarious, he must abandon the property. In July 1997, for example, one potter had recently purchased a building for a store near the Plaza of San 139

How Has Distribution of the Pottery Changed?

Figure 4.6. Painting workshop located in Ticul’s largest production unit in 1997. (Photo

by Michelle R. Arnold)

Enrique, but by 2002 he had abandoned the facility, built another west of Ticul, and then abandoned it, too. Second, stores require management ability and/or experience to manage capital and labor. One potter who worked for others for many years and began to work independently shortly before my visit in 1997 recognized this challenge when he explained that he did not have much experience as an independent seller. So he chose to sell his pottery to local brokers rather than opening a store or becoming a broker himself. Stores in Mérida. A second significant market for distributing pottery consists of selling pottery to stores in Mérida. This market segment has a long history in Ticul, but it became more important since plant pots and tourist vessels became more important in the 1970s. From the 1930s to the 1950s, five Ticul potters sold their wares to stores located in the arcade near the municipal market in Mérida, and these sales, potters say, sustained them. One of these stores bought everything that potters wanted to sell. After one proprietor died about 1945, his store was taken over by his niece and she continued to buy Ticul pottery until 1960. In the late 1960s, Ticul pottery appeared to have no presence in Mérida, and no demand existed for Ticul pottery as a tourist souvenir except at the store in the 140

How Has Distribution of the Pottery Changed?

tourist hotel at the Hacienda Uxmal. By the early 1980s, however, there were 80 to 100 artisan stores in Mérida and most of these sold pottery. A survey of twenty of these stores by a student in the School of Anthropology at the University of Yucatán found that only one had an association with artisan production. Most of them did not exist before 1974 and their development coincided with the increased tourism associated with the emergence of Cancún as a tourist resort (Field Notes, July 20, 1984). By 1984, souvenir shops in Mérida became important outlets for Ticul pottery. It was sold in stores in the central part of the city as well as in the shops in the hotels along the Paseo de Montejo. In the 1990s, a new outlet for Ticul ceramics emerged in the form of the government craft store (FONOPAS). Located in the historic center of Mérida, it sold craft products from the entire Mexican republic and was largely oriented toward a tourist clientele. Some Ticul potters sold their pottery there directly. The buyers asked potters for examples of their work and then only bought objects that they judged to be of good quality. Even so, the store was not an outlet for all of the vessels potters produced; they still needed other outlets for their wares. By 1994, the tourist market had expanded still further. Although no systematic survey was made, several stores selling Ticul pottery had appeared in locations where none had existed previously. Two new stores appeared in central Mérida. Another had appeared in the northern part of the city along the highway to Progreso, and one appeared along the principal route into the city from the south. This latter store joined another across the street from the Mérida Zoo that had been there in 1984. The new Hyatt Hotel was also selling copies of ancient Maya pottery in its shop. Most of the stores that sold Ticul pottery acquired it through brokers, or the store owners were brokers themselves. Stores Elsewhere. By 1984, Cancún was the most important market for Ticul pottery and several owners of production units sold their pottery there. By 1994, the number of stores outside of Ticul that sold its pottery increased. Such stores appeared along the highway in Uman, on the bypass around Becal, and at the service plaza along the toll road to Cancún. In addition, changes in access to the archaeological sites of Uxmal and Chichén Itzá created souvenir stores there and many of these sold Ticul pottery. Besides the more common mass-produced vessels with copies of ancient designs, carefully executed replicas of ancient vessels were also available. By 1997, stores for selling Ticul tourist pottery had expanded into more locations. At Pisté, the town adjacent to the ancient ruins of Chichén Itzá, vessels from Ticul were present in many of the tourist shops. In addition, a pottery 141

How Has Distribution of the Pottery Changed?

Figure 4.7. Loading a large truck with plant pots and figurines bound for an unknown destination in 1984.

store selling Ticul plant pots and hanging pots was visible two blocks north of the plaza in Uman. No systematic survey of stores was made in Cancún in 1997, but five stores had masses of Ticul plant pots for sale along the highway between Cancún and Puerto Morelos. Ticul pottery was also sold in the tourist shops in the Cancún airport. Brokers

Even with stores along the highway and with access to the hundreds of travelers who come through the city each year, Ticul is not on a major tourist route and potters needed another way to sell their pottery. Brokers provided this service. Brokers are middlemen who are pottery resellers. They often sell the pottery to another broker before it ends up in the hands of consumers. The advantage of selling to brokers is that potters can be assured of a buyer for their pottery. The principal means by which brokers sold their pottery was to transport it to clients using their own vehicle or the services of a truck owner. In the late 1960s, no potter owned a motor vehicle, but by 1984, at least four productionunit owners had small pickup trucks to transport pottery to distant markets. By 1997, at least five production-unit owners owned a van or a truck and one had

142

How Has Distribution of the Pottery Changed?

two such vehicles. An owner may also rent the services of a larger truck in order to increase the amount of pottery sold in distant markets and reduce his per-unit transportation costs (Figure 4.7). Selling pottery to brokers has a long history in Ticul. Although its beginning was small, its importance as an outlet for ceramics has changed greatly during the twentieth century. From the 1930s to the 1950s, three brokers came from Mérida by train to buy pottery. During the last half of the 1960s, however, brokers were a relatively minor way to sell vessels and few brokers were involved except for pottery used for the Day of the Dead rituals. Some potters sold their pottery to other potters who were brokers, but most sold their pottery directly to consumers. Since 1970, however, the use of brokers has increased greatly. Between 1970 and 1984, consuming populations became more distant from Ticul and public transportation that could carry pots was limited; thus, most potters could not transport pottery to these populations themselves. Brokers provided the necessary transportation and became agents in distributing pottery to new market segments: hotels, urban residents, and craft stores who catered to tourists. By 1984, Cancún had become the most significant market for Ticul pottery, but it was relatively inaccessible by public transportation. Buses were the only form of public transport available, and unlike trains, only a limited amount of pottery could be transported. Because brokers had trucks, they became the most significant purchasers and distributors of pottery destined for Cancún. As a consequence, brokers had a significant effect on pottery production; by 1984 the demand for pottery was almost entirely driven by broker requests (Table 4.2). Types of Brokers. Brokers fall into three general categories. One type consists of Ticul production-unit owners who are potters themselves and purchase fired or unfired vessels from smaller production units. They then fire and paint the vessels for resale. A second type of broker buys only fired pottery, comes from outside of Ticul (such as Mérida), is not a potter, and usually has a painting workshop. In some cases, they sell the painted ware in the stores associated with their workshops, or they may sell it to other stores. They may also buy other kinds of pottery (such as plant pots; see Figure 4.8). One of these brokers was formerly employed by the government painting workshop in Ticul in the 1970s but had since developed his own painting workshop in Mérida and contracted with Ticul potters to produce specific kinds of vessels. These first two categories of brokers are not mutually exclusive because a few brokers do not easily fit into either category. A few production-unit owners in Ticul, for example, are not potters and only paint pottery made by others. 143

How Has Distribution of the Pottery Changed?

Figure 4.8. Loading a truck with cubano-shaped plant pots in 1984. The buyer is the owner of a store and painting workshop in Mérida.

Potters will sell to both types of brokers. In 1984, my principal informant produced most of his pottery by request from two such brokers: one in Ticul and another in Mérida. The broker in Ticul owned a production unit that largely specialized in painting vessels with copies of ancient Maya designs. He bought blank vessels from as many as fifteen potters, painted them, and then took them to the hotels and shops in Cancún every five to six weeks. A third kind of broker comes from outside of Ticul and purchases pottery for the Day of the Dead rituals. This pattern of distribution has a long history in Ticul. Sometimes sales are made to specific brokers, and the brokers may sell the pottery directly either to consumers or to another broker. This kind of broker may also contract for a quantity of certain vessels, such as food bowls, whistles, candle holders, flowerpots, and incense burners. Other brokers may buy these vessels outright. In 1970, for example, one broker from Mérida bought as many as 200 bowls at a time. The price difference between selling Day of the Dead pottery to brokers and selling it directly to consumers can be considerable. For assuming the risks of selling to consumers and the costs of transportation, a broker will pay 25 percent less for a food bowl than the potter receives on the open market. In 1984, for example, a broker paid fifteen pesos for a food bowl, but the potter could sell 144

How Has Distribution of the Pottery Changed?

it in the Oxkutzcab market for twenty pesos. Similarly, potters may receive 50 percent less from a local broker than they would from a nonlocal broker. In 1984, for example, a local broker from Ticul paid seven pesos for a whistle, whereas a nonlocal broker paid fifteen pesos for the same type of whistle. If the potter sold the whistle in the market in Oxkutzcab, however, he would receive twenty pesos for it. Brokers and Demand. Using brokers as an outlet for selling pottery creates a different demand structure on pottery distribution, and potters no longer have to predict the kind of consumer demand that they might find when they sell their pots. Rather, brokers provide precise requests for the kind, shape, and size of pottery they want. Brokers request three kinds of pottery. First, they can order virtually any kind of unfired vessel, but they usually buy plant pots and small molded items. Only brokers who are production-unit owners in Ticul buy unfired pottery. Second, brokers buy fired but unslipped and unpainted blanks for painting. These blanks are generally small vessels in a variety of shapes that include bulbous vessels, cylindrical vessels, and ashtrays, but brokers occasionally purchase traditional vessels as blanks, such as full-sized water-carrying jars (cántaros). The blanks are usually painted in the brokers’ workshop with imitations of ancient Maya designs and then sold to tourist shops in Mérida and Cancún. Less frequently, blanks are painted with floral motifs. Third, brokers may order fired plant pots that are slipped with either the traditional red (k’an kab) or cream slips. In order to assess the shapes that might be in demand, a potter makes a variety of unfired shapes and then uses them as proposals to a broker. If the broker thinks the shapes will sell, the potter fires them, and then sells them to the broker. If the broker sells the vessels, he asks for more copies of that shape. If, on the other hand, the broker did not like the new shape, the potter may show it to another broker or break it up and reuse the clay. A broker may also provide a template from which the potter makes a mold to produce identical copies of the desired vessel. Production Scheduling and Broker Sales. Production scheduling is critical to the execution of broker requests because orders cannot be filled instantly but require at least two weeks to prepare the paste and fabricate, dry, and fire the vessel. Production scheduling may be accommodated in several ways. First, a potter can adopt a strict schedule and not accept orders that do not fit the schedule. When I visited one potter in 1997, a client arrived asking for fifty flowerpots of a particular size for delivery in thirteen days. The potter replied 145

How Has Distribution of the Pottery Changed?

that he was too busy and he could not accept the order because he required a minimum of twenty days to complete an order. More lead time was required for larger quantities. Eighty to 100 items, for example, require a lead time of 1 to 1.5 months, whereas 200 items require a lead time of 2 to 2.5 months. The potter and his wife then discussed whether they could meet the deadline for the order. Accepting but not completing an order has adverse consequences; the potter may be left with unsalable pots and no financial return. In 1984, for example, two potters had made bulls to sell to members of the cattleman’s gremio but did not fire them in time to meet the delivery date and had to dispose of them in other ways. If a potter fails to fulfill an order by the promised delivery date, he will not only lose the cash from the request but may also lose the opportunity for future orders as well. A second way to allocate production scheduling is to accept the order but subcontract the vessels requested to other potters. This strategy favors the large workshop and the extended family with multiple production units. The primary contractor can accept the order, satisfy the client, lay the basis for future orders, and also provide work for one’s relatives, but some owners may also subcontract vessels to non-relatives. In 1997, for example, one potter accepted an order for flowerpots from a client in Cancún but then subcontracted them to her brother. This strategy is also used when clients ask for traditional water-carrying and water-storage vessels that most potters can no longer make. A production-unit owner may accept the order even though his workers cannot make the vessels, but he will subcontract it to a traditional potter. The owner of the largest production unit in Ticul subcontracted production to both relatives and non-relatives but gave them clay from his own source to make the pots (see Chapter 5). Problems of Sales to Brokers. Selling pottery to brokers has both advantages and disadvantages. The most obvious advantage consists of the potter’s assurance of a regular client to buy his pottery. Selling to a broker reduces the potters’ risk of producing vessels that no one will buy and transfers that risk to the broker. Between 1965 and 1970, brokers were only a minor outlet for ceramics, but potters recognized the problems of selling their pottery to them. They understood that by selling to brokers, they lose control over their sales in a competitive market and receive lower prices. In 1966, for example, one informant reported that one of his fellow craftsmen was poor because he sold all of his pottery to brokers rather than selling it himself. As a result, this potter procured all of his raw materials (clay, temper, and firewood) himself, rather than buying them from specialists, in order to cut costs and increase profitability. In this way, he increased his monetary returns by controlling the costs of his inputs and conserving his capital. 146

How Has Distribution of the Pottery Changed?

The same potter who made this observation in 1966, however, found himself in an identical situation in 1984. A related disadvantage of selling pottery to brokers is that they often require the potters to sell to them on credit so that the potter shares the risk of poor sales. This practice is an obvious disadvantage because, unlike selling pottery directly to consumers, the potter feels that he must accept the money he is given for the pottery with no alternative except to stop selling pottery to the broker in the future. Because of the competition for clients, however, potters often believe that they have no alternative and must accept partial payment from a broker with promises to pay more later in order to ensure future sales. This creates a cash-flow problem because potters must have the capital to sustain themselves until they get paid. In many cases, potters have long-term debts owed them by brokers and often never get the contract price for their wares. If they do, payment may be spread out over months. Credit and unpaid debts from pottery was a persistent problem with my principal informant in 1984. Brokers did not pay for the pottery on delivery, and if the potter wanted a cash payment of the total amount, the broker reduced the price that he would pay. In August 1984, he was having trouble finding clients who would pay cash for his pottery. In one instance, a broker had paid him 5,000 pesos for a load of pottery but still owed him 10,000 pesos. He had gone to collect the remainder but received only 2,000 pesos. Similar problems occurred in 1988, and he had not yet received payment for pottery delivered weeks before. These problems can be devastating to potters and can keep them poor, but brokers say that these problems occur because of low sales. In some cases, brokers may pay higher prices for vessels if the potter pays the freight to have them shipped. In 1984, one Mérida workshop owner paid more for vessels if the potter paid the shipping costs. Sometimes these prices were 50 percent more than the price of what other brokers paid. This particular broker reportedly was not the best client, however, because he spread out his payment for the pottery over a long period of time. This problem can be illustrated by an incident that occurred in 1984. When I arrived at my informant’s house one morning, he made the unusual request for an advance of his wages for the day. His children had not had breakfast because he had no money to buy any food. When I asked why, he said that he was waiting to be paid by the broker to whom he had sold his pots and he had no cash. So, I advanced him the money to buy the food and gave him 500 pesos to pay for the trip to Mérida to try to get some of the money that was owed him by the broker. In 1994, another potter said that he could not afford to sell his pottery to a client in Mérida because the client wanted to buy it on credit. The potter said that 147

How Has Distribution of the Pottery Changed?

he had to have cash immediately and that those who have the capital to accept credit have an advantage over those who do not because such credit is part of the competition that one must accept when one makes pottery. In 1997, another potter said that clients who want to buy pottery on credit can take delivery of the pottery for half the cost plus a promise to pay the remainder later. If the broker does not pay, the potter does not sell any more pottery to him. If the person was reliable, he said, he would extend them credit. Even so, this potter says that he had many “promises to pay” tens of thousands of pesos, with a total outstanding accounts receivable debt of 4 million pesos. As a result of the problems with brokers, one potter wanted to sell his vessels directly to consumers in August 1984. To that end, he had started making moldmade figurines to sell in the villages. Such a distribution strategy incurred great risk, he said, because he had no assurance that if he went to a village, its inhabitants would buy his pottery. Even though potters who use brokers have an outlet for their pottery, they may not sell all of it to brokers. In August 1984, for example, one potter told me that although he sold some of his pottery to a store near the market in Mérida that sells piñatas, he had other pottery to sell without any clients to buy it. As a result, he said, he was going to Cancún to find clients. Problems with brokers may cause potters to abandon the craft. By 1994, for example, a number of potters who had made pottery in 1988 had stopped making pottery. One potter said that one reason was that the competition among potters was fierce. As a result, he had started to make vessels for the Day of the Dead festivities in mid-July. By producing pottery for the Day of the Dead, potters can sell directly to the consumer, be assured of a demand for their pottery, and get cash for it immediately. Some of these same potters resumed making pottery once the demand for pottery improved and were making pottery again in 1997. Vertical Integration Although the predominant evolutionary processes of pottery production in Ticul between 1965 and 1997 have consisted of a movement toward separation of production and distribution, the processes have operated in the opposite manner for one potter who had vertically integrated his production and distribution. He owned his own clay source and had the largest production facility in Ticul, which included two production units: one behind his house and a second along the highway. He owned two trucks to transport raw materials and finished pottery, and he sold pottery in a store in front of one of his workshops in Ticul near Cancún (Figure 4.9) and in another along the main highway to Mexico City on 148

How Has Distribution of the Pottery Changed?

Figure 4.9. Store along the highway south of Cancún owned by a Ticul potter who had

vertically integrated his production and distribution in 1997. All of the vessels on the shelves are souvenir vessels with ancient Maya designs painted on them. A large inventory of plant pots is located in front of this structure. (Photo by Michelle R. Arnold)

the bypass around the town of Dzitbalché. The Dzitbalché store is located on the property that includes his clay source, and when his workers came to dig clay, they brought pottery to sell there. His store in Cancún was operated by one of his sons, who had studied chemical engineering and had worked in Cancún for twelve years. When he lost his job, his father wanted him to return to Ticul, but his wife liked Cancún and wanted to stay there. So, his father encouraged him to set up a shop to sell his pottery there. By 1997, the Cancún store was located along the highway to Puerto Morelos. It was selling plant pots and pottery decorated with copies of ancient Maya designs, and its sales were quite lucrative with as much as a 100 percent markup. Besides selling pottery, the store painted pottery, and future plans called for the construction of a structure to make pottery and a kiln to fire it. It is not unusual that vertical integration is the exception rather than the rule in the evolution of ceramic production in Ticul. Vertically integrated production, as in many modern corporations, requires capital to acquire raw materials, raw material sources, and sales outlets. Further, like modern corporations, vertical

149

How Has Distribution of the Pottery Changed?

integration requires effective management skills. In this case, almost all of the production and sales facilities were managed by family members. It is perhaps because of the lack of management that, by the time of my visit in 2002, the store near Dzitbalché had closed. Conclusion The patterns of the distribution of Ticul pottery have changed greatly during the period of 1965 to 1997. In 1965, most potters sold their own pottery at the markets and fiestas of Yucatán. Since that time, the distribution choices for the Day of the Dead rituals have changed little whereas the organization of distribution of other kinds of pottery has changed greatly. The geographical area of distribution has greatly expanded, has moved farther away from Ticul, and has tapped the tourist markets of Mérida and Cancún, which were not available previously. The main factor responsible for these changes was the transformation of the highway infrastructure that gave potters access to consuming populations that were previously too remote for access by public transportation. Access to the railroads and highways provided the initial stimulus for marketing pottery outside of Ticul as well as a continuing market for new and innovative vessels. With the growth of the tourist industry beginning in the late 1970s, potters found an almost infinite market for pottery. Cancún became the most significant market for Ticul potters even though they did not have the means to go and sell their pottery there. Along with the changing markets and patterns of distribution, great changes have taken place in the organization of that distribution. In 1965, most potters distributed their pottery themselves and sold it directly to consumers. Beginning in the 1970s, however, local stores developed along the highway to sell pottery to travelers who passed through the city. At the same time, potters began to sell their wares to brokers who had access to trucks and could take pottery to distant markets. By 1997, a few potters began displaying their wares in the annual Ticul fair. All of these forms of distribution represent a different, more complex, and more specialized type of distribution than that of the late 1960s. Although almost all of these forms of distribution represent a segmentation of tasks that separates production from distribution, one potter developed vertical integration in which he has combined, rather than segmented, portions of the pottery-making process by acquiring a clay deposit and controlling distribution and sales of his pottery in Ticul as well as in Cancún and elsewhere. Does the marketing of Ticul pottery reflect social change? Indeed it does. Probably more than any other link in the behavioral chain of pottery produc150

How Has Distribution of the Pottery Changed?

tion and distribution, the changing distribution of pottery reflects the changes in the economy and in the realities of the transportation infrastructure of modern Mexico. Consequently, it may seem that studies of the distribution of modern pottery production are not helpful in understanding ancient economic realities of ceramic production. Nevertheless, the Ticul data for 1965 to 1997 reveal just how critical distribution is for the growth and evolution of ceramic production, and it is easy to see how critical the study of ancient patterns of economic distribution and trade are for understanding the change in ancient ceramics and in the development of socioeconomic complexity. Once social change eliminates some uses for pottery, expansion of patterns of distribution can serve to maintain production and even increase it.

151

Chapter

How Has Clay Procurement Changed?

five

T

he uneven distribution of resources across the landscape is often regarded as critical in the development of craft specialization and the evolution of socioeconomic complexity. If resources are localized and restricted, it is argued, access to these resources is unequal and this differential access stimulates the development of craft specialization and economic interdependency (Costin 1991:14; Sanders and Price 1968:10). Applying this same explanation to ceramics, it seems that the existence of localized clay sources across the landscape with differential access by a population is essential for the development of specialized ceramic production. As access to clay sources became restricted and fell under elite control, specialization intensified, and more complex levels of ceramic specialization developed and resulted in the reduced variability and greater uniformity of ceramic pastes (Rice 1981:221). This increased paste homogeneity thus was considered to be a surrogate measure of a more evolved, more tightly controlled, and more specialized ceramic production (Rice 1981:222–223). 153

How Has Clay Procurement Changed?

Is this explanation comprehensive and universal enough to explain ethnographic cases? Clay, of course, is essential to making pots, and its procurement is one of the most critical stages in the behavioral chain of ceramic production. Does clay procurement change over time, and if so, how does this change relate to the development of increased specialization and thus socioeconomic complexity? For archaeologists using the Rice model, understanding the relationship of changing clay sources to social and political organization is critical to ascertaining whether the composition of pottery reflects changes in elite control of clay resources. This relationship, however, begs several prior questions. Under what situations does clay procurement change? What factors affect the changes in the organization of that procurement? How do these changes relate to a society’s social organization and elite control of resources? Do changing clay sources reflect evolutionary social change? Examining contemporary clay resource utilization may help answer these questions. Since clay procurement in Ticul changed greatly between 1965 and 1997, describing and evaluating the history of clay procurement provides an opportunity to examine the causes that contribute to the change. This chapter thus describes the changing relationship of clay procurement to social and political factors and reveals how and why these factors affect that procurement. This history reveals that although high-quality clay is restricted in distribution, clays are sufficiently ubiquitous that they can be obtained from any number of locations. Furthermore, changes in clay procurement have led to the development of mining specialists and are reflected in the increased amount of clay consumed between 1965 and 1984. Finally, geological resources have finite limits and the history of clay procurement reveals that access to clay deposits can have a significant interpersonal component that has political overtones. Access to clay deposits may be mediated by powerful social relationships and conflicts. It may also be mediated by the constraints of elite control. Under What Conditions Does Clay Procurement Change? Clay occurs in many locations around Ticul, but potters’ oral history and published descriptions of Ticul pottery production throughout most of the twentieth century reveal that potters have preferred to use clay from Hacienda Yo’ K’at, located five kilometers from Ticul (Barrera Vásquez 1937:164; Brainerd 1958; Rendón 1947; Terán 1981:17; Thompson 1958:66; Varela Torrecilla 1990:202– 203). Yo’ K’at means “over clay” in Yucatec Maya. Based on surveys of accessible clay deposits in 1968 by geologist B. F. Bohor (Illinois Geological Survey) and 154

How Has Clay Procurement Changed?

me, the Yo’ K’at deposit is unusual and is the only known large deposit of this clay accessible to potters in the immediate area of Ticul (Arnold 1967a, 1967b, 1971, 2005b; Bohor 1975; Isphording and Wilson 1974:486). Analyses by Schultz and colleagues (1971:56), Isphording and Wilson (1974), and Bohor (1975:121) indicate that this clay consists of a random mixed layering of kaolinite and montmorillonite and only occurs in three known deposits in the northern Yucatán: Yo’ K’at, Tepakán, and Becal. For making large vessels, it is technologically superior to other clays (montmorillonite) found in Yucatán, and the relative scarcity and high quality of the Yo’ K’at clay were probably important reasons for naming the hacienda “over clay” (Arnold 1971). Clay can be found in several locations in Yo’ K’at, and over the years potters have procured their clay from different places within its confines (Figure 5.1). Informants reported that two mines were used at different times during the first half of the twentieth century, and although the portal of a mine utilized in 1934 was still evident in 1968, neither of these sources showed any obvious evidence of mining that suggested extensive mining underground, such as spoil piles or surface subsidence. Another mine was located in a henequen field 300 meters east of the gate at the rear of the hacienda, but mining there was abandoned when clay procurement was moved to the mine utilized in 1965. Clay was first mined at Yo’ K’at at least a millennium ago. The earliest evidence for clay mining at the hacienda was discovered in 1968 and comes from deep within the interior of the mine where a fragment of a bolster rim basin was associated with a collapsed mine tunnel (Arnold and Bohor 1977). The triangular shape of the sherd and its in situ occurrence over a lump of clay in a wall profile suggested that the sherd was a mining tool used to gather clay during the Terminal Classic period, between A.D. 800 and A.D. 1100 (Arnold and Bohor 1977). Since 1968, miners have continued to find evidence of ancient mining in the clay mines of Yo K’at. In 1984, two miners reported finding ancient pottery during mining, and another said that he had found an old food bowl in the clay deposit. In contrast to the clay deposit at Yo’ K’at, other clay deposits around Ticul consist of clays of inferior quality. These deposits are widespread and exist as small pockets in marl deposits or as massive deposits in the base of quarries that were used to mine the marl for construction purposes (cf. Isphording and Wilson 1974:486). Clays from many of these deposits were analyzed and they all consisted of montmorillonite, a clay that is inferior to the Yo’ K’at clay for making pots. It cannot be used to make large vessels because it causes them to crack and break (Arnold 1971). 155

How Has Clay Procurement Changed?

Figure 5.1. Sketch map of Hacienda Yo’ K’at showing clay sources surveyed in 1968, the extent of clay mining from 1965 to 1988, and some of the other clay sources. The clay mine used from 1965 to 1970 was approximately 565 meters north of the Ticul-Muna highway. (Drawn by Michael Anderson)

Changing Elite Ownership and Clay Procurement

Although potters appear to have used the Yo’ K’at deposit for mining clay for at least a thousand years, potters’ oral history indicates that the hacienda was 156

How Has Clay Procurement Changed?

owned by elites for as long as potters can recount. From historical documents, this kind of land tenure at Yo’ K’at extends at least to the early nineteenth century when the name Yo’ K’at appeared in the 1828 census as a hacienda in the municipio of Ticul (Dumond and Dumond 1982:257, 440). Even though its name implies its importance as a clay source, the clay was not important to the elites. Rather, from the late nineteenth century up until the late 1960s, the land was used to grow henequen. Since then, it was used to grow maize and raise cattle. Even so, its ownership changed over the years and this change had varying effects on the organization of clay procurement. Sometimes the effect was great and other times it was not. Over the years, the manner in which the elites have controlled the clay source has varied greatly. But generally, this control has been in the hands of the local managers/foremen who were empowered by the nonresident owners to make the decisions about the day-to-day operations of the hacienda. The decisions of these managers profoundly affected the potters’ access to the clay. Potters thus recount the oral history of clay procurement in terms of their relationship with the hacienda’s managers, which varied according to the degree of potters’ access to the clay. Potters have no recollection of the managers who served under the first owner, Juaquín Espejo, but they can recount incidents with the managers under the next owners, the Montes family, who owned the hacienda for more than fifty years. The first owner from this family was Avelino, who was prominent in Mérida politics. He is on record as a henequen businessman who purchased one of the grand old houses (Villa Beatriz) along the Paseo Montejo in Mérida (Rasmussen 1994:104). Avelino married the daughter of Oligario Molino, who was a politically prominent figure in Yucatán throughout the late nineteenth and very early twentieth centuries (Orosa Díaz 1994:194–195, 203). Molino was a federal deputy from 1869 to 1871 and from 1873 to 1875 (Orosa Díaz 1994:203). From 1902 to 1906, he was governor of the State of Yucatán near the end of the Epoch of Slavery, when the notorious Porfirio Díaz was president of Mexico (Orosa Díaz 1994:203). When Avelino died, the land was inherited by his daughter Josefina. The Montes family was not interested in the clay and entrusted the business of mining and selling it to the manager of the hacienda. The first two managers (Pepe Peniche and Felix Marqués) paid for digging the shaft for the mine, and potters only paid the cost for transporting the clay to Ticul on the tramway (tranvía). After the clay was mined, potters put their names on the bags and the bags were placed on the tramway’s horse-drawn platforms, taken to Ticul, and delivered to potters’ households by horse‑drawn cart. Little information exists about the third manager (Perfirio Alvarado) under the Montes family except that potters said he was a bad manager. It was likely that 157

How Has Clay Procurement Changed?

his administration occurred in the 1930s when three individuals from the village of Mama died when the mine collapsed. As a result, he closed the mine and potters had to get their clay elsewhere. The next manager (Elario Medina) was employed in the late 1930s and early 1940s. Because the hacienda needed labor, he wanted to discourage potters from digging the clay, so that they would become henequen workers. He did not want to be bothered with having potters digging clay and transporting it to Ticul on the tranvía. So he charged potters a price that was approximately two-thirds of a potter’s daily wage (fifty centavos) for each sack of clay and then limited the amount extracted to two sacks per potter unless they had special permission. Potters considered this cost excessive and they did not have the cash available to pay it. To counter the manager’s clay policy, the potters organized themselves into a union (sindicato) to defend their rights. They consulted their legislative deputy (Sostenes Carrillo) from the nearby town of Muna, and he asked them to prepare an official request (oficio) to present to the governor of Yucatán. In that request, the potters argued that although the land belonged to the hacienda, the subsurface rights belonged to the Mexican people. As a result of their appeal to the governor, the potters won their case and obtained a signed order stating that the hacienda only owned the surface of the ground. Those resources that were underground (such as clay) were the property of the state and the owner of the land had no control over them. Potters, the governor ruled, thus should be allowed to dig clay at Yo’ K’at and should do so without paying for it. Following this outcome, the potters only paid a freight charge of twenty-five centavos to bring the clay to Ticul. This outcome took place during a period when Mexico was nationalizing its mining and mineral companies (such as petroleum) to redress the abrogation of the nation’s ownership of subsurface resources that had occurred during the nineteenth century when the proprietors of the surface land had obtained ownership of all bituminous and mineral fuels in order to stimulate foreign investment (Meyer et al. 1999:31). This political climate thus contributed positively to the potters’ access to their traditional clay deposit. During the administration of the next manager, Humberto Herrera, the clay mine yielded only poor-quality clay and rocks (shishk’at), creating a crisis for potters. Herrera proposed a solution to one of the potters (Emilio Tzum) and invited him to pay for one of the nine nights of ritual prayers (novena) for the hacienda’s patron saint (San Pedro). By financially sponsoring a novena, Emilio was making a promise to the saint and establishing a patron-client type of dyadic contract (Foster 1963, 1967) with him. It was hoped that the saint would reciprocate by restoring the quality of the clay. 158

How Has Clay Procurement Changed?

After the promise was made, the potters dug two new mines nearer the buildings and discovered good-quality clay. As a result of this outcome, potters participated more directly in the adoration of the saint by sponsoring one night of the novena at the hacienda. About this time, the potters’ union dissolved. It is unclear why it ceased, but during the administration of the next manager (Enrique Valadez, from about 1952 to 1962), potters again were required to pay for clay from Yo’ K’at. The last manager for the Montes family was Hermilio Beltrán, who served for about twenty years (ca. 1962–1982). He changed the policy and no longer required potters to pay for the right to dig clay. Sometime between 1965 and 1970, the hacienda was sold, but Beltrán continued to be the manager. This change of ownership had a profound effect on the organization of clay procurement and led to irreversible changes. During this period, all of the portable assets of the hacienda were liquidated: machinery, rails of the tramway, the twenty-five mules used to pull the tramway platforms, and roof tiles from the houses. As a result, the hacienda transportation system could no longer be used to bring clay to Ticul. About 1982, the hacienda was sold again. This change moved ownership away from regional elites toward local elites. Rather than an absentee owner who lived far away and put the operation of the hacienda under a local manager, the new owner was a businessman from Ticul. The new owner instituted a number of changes in clay procurement that provided a unique kind of control that directly affected potters’ access to the clay. First, he imposed a 12.5 percent fee (fifteen pesos in 1984) on each bag of clay for the “right” (derecho) to mine it. Second, he required miners to use his truck to transport the clay to Ticul. Both the transportation cost and the right-to-mine fee were charged to the potters on delivery. Even with these changes restricting procurement, clay from the Yo’ K’at mines was becoming scarce by the late 1980s as the increased demand for pottery began to strain clay availability. In controlling the clay resource at Yo’ K’at, the owner had allocated a fixed amount of space for digging clay mines and by 1988 the miners had reached the limits of this space. Clay Scarcity and Interpersonal Politics

In the 1960s, a few potters anticipated the exhaustion of the Yo’ K’at clay and began searching for other clay sources, such as those used by Akil and Tepakán potters. One potter (Pepe, a pseudonym) had already investigated other clay sources when Hacienda Yo’ K’at was sold in the late 1960s. At the time, Josefina Montes had owned the hacienda and she wanted to sell it. Because 159

How Has Clay Procurement Changed?

Ticul potters had obtained their clay on the property and were “humble and simple folk,” she wanted to provide for the potters’ interests. Pepe had convinced three or four others to buy the hacienda with him, but they eventually backed out of the agreement. When the Yo’ K’at sources began showing signs of exhaustion in the 1980s, one potter (Pepe’s WiSiHu) began going elsewhere to obtain clay. He exchanged pottery for it and tried to buy land for a permanent source. About the same time, two other potters began to procure clay in Campeche also by exchanging pottery for it. They approached another potter (the late Enrique Garma) in order to convince him to participate in developing another source. Enrique was concerned about the diminishing amount of clay from Yo’ K’at and the need for another source, but he had already stockpiled 100 bags of the Yo’ K’at clay and wanted his own source. Meanwhile, Pepe continued to buy Yo’ K’at clay, consuming forty to fifty bags of it every two weeks. This amount was excessive, according to the owner of Yo’ K’at, and he believed that Pepe’s rate of use would deplete the clay deposit quickly. Consequently, those potters who were poor and had no alternative sources would have no clay at all. So the owner refused to sell Pepe any more clay. Pepe then started obtaining his clay from the communal (ejido) land of Dzitbalché, located eighty kilometers away in the State of Campeche, while he continued to search for a more permanent source. Even with his new source, Pepe was upset that the owner of Yo’ K’at would not sell him any clay. So he went to the township (municipio) officials to protest. He argued that the potters had a legal right to obtain clay at Yo’ K’at because they had done so for decades. To underscore his protest, he brought a lawsuit ( parijuda, which is a type of petition) against the hacienda owner that cited the earlier precedent that the owner of Yo’ K’at only owned the surface of the ground whereas the Mexican people owned the underground resources, which should be used for the benefit of all. Pepe tried to rally the potters around his cause, but they refused to become involved. They believed that if the lawsuit was decided in his favor, Pepe’s massive use of clay would deplete the Yo’ K’at source more rapidly, and they would have no clay at all. Since Pepe was obtaining his clay from Campeche at that time, he already had an alternative source. Others, however, had no such source, and the potters feared the complete loss of their livelihood. Although the lawsuit seemed to be straightforward, its roots lay deeper in local sociopolitical dynamics than just the refusal of the owner to sell clay to Pepe. The owner of Yo’ K’at also owned a ceramics factory, and a year earlier he had started buying white earth (sak lu’um) from two temper miners to ship to 160

How Has Clay Procurement Changed?

Monterrey in exchange for the clay used in his factory. Selling white earth was lucrative for the miners because they were paid 125 pesos per sack for what was essentially a by-product of temper mining (see Chapter 6). The white earth, however, was a critical resource for pottery temper and could be converted into temper with weathering and crushing at a future time. Potters thus were concerned that the sale of white earth would deplete this critical resource for them in the future. Furthermore, the white earth occurred on communal land and was owned by the people, but the only people who were benefitting from the sale of this resource were the miners and the owner of the ceramics factory. So Pepe complained to the township authorities, arguing that too few people within the municipality benefitted from the sale of white earth. As a result, the authorities intervened and ordered the owner of Yo’ K’at to stop buying the white earth from communal land and shipping it outside the township. This incident deepened the conflict between Pepe and the owner of Yo’ K’at. About this time, a Spaniard established a pottery workshop in Cancún. He hired fifteen Ticul potters and began to look for clay sources. He discovered that Ticul had the best clay, and according to rumor, he made a contract with the owner of Yo’ K’at to buy clay from him. When Pepe heard this rumor, he complained to the township authorities again because the Yo’ K’at clay was a valuable resource and was being sold outside the township. He believed that the alleged contract between the owner of Yo’ K’at and the pottery workshop in Cancún provided additional evidence for his lawsuit that the sale of clay would further deplete this resource at the hacienda. Pepe thus filed a complaint against the owner of Yo’ K’at and asked for the potters’ support until the original lawsuit could be decided. The potters, however, were afraid that if Pepe was paying the legal fees, he would control the outcome. So the potters did not support him. Even so, the township officials moved against the owner of Yo’ K’at because the resource below the ground belonged to the Mexican people; thus, he was legally restrained from selling clay to all persons outside of the township. Meanwhile, the owner of Yo’ K’at had stockpiled clay in the abandoned structures at the hacienda. One day the owner of a Cancún pottery workshop was in Ticul with a tractor trailer full of clay. Hearing about the truck, Pepe was convinced that the owner of Yo’ K’at had violated the restraining order and was selling Yo’ K’at clay outside of the township. Pepe believed that if township authorities got a warrant to search the owner’s warehouses, he could prove that the clay on the Cancún truck came from Yo’ K’at and that the owner had disobeyed the restraining order. The owner of Yo’ K’at, however, willingly opened his storehouses and showed the potters that he was not selling clay from the hacienda. 161

How Has Clay Procurement Changed?

Rather, the Cancún workshop owner had obtained his clay from Dzitbalché in the State of Campeche. Meanwhile, Pepe was procuring his clay from the source used by Tepakán potters in the communal land near Dzitbalché. He was digging clay next to the highway, and eventually, his mining moved under the highway. When the Ministry of Transport told him not to undermine the highway, he started to search for another source. Six months after he brought the lawsuit against the owner of Yo’ K’at, Pepe purchased an abandoned marl quarry 1.6 kilometers from the Tepakán source (Arnold et al. 2000) and started to obtain his clay there. Pepe, however, still wanted a local supply of clay. So with the cooperation of one of the clay miners, he hired workers from Chapab to open his own mine at Yo’ K’at without the owner’s permission. They dug a shaft into the clay deposit near the shafts of other miners but did not tell them that they were using explosives to break up the rock. When Pepe’s crew detonated the explosives, a miner working ten meters away feared that his mine would collapse on top of him and he became very angry. Other miners were concerned about their mines as well and complained to the hacienda’s manager. As a result, the owner halted the digging by Pepe’s miners and prohibited their return. Pepe’s attempt at opening his own mine at Yo’ K’at not only exacerbated the conflict with the owner of the hacienda but jeopardized the immediate future of clay mining there. As a result of Pepe’s action, the hacienda owner told the potters that if the lawsuit went forward, he would stop all clay mining at Yo’ K’at until it was resolved. This created a crisis for the other potters because they did not have the resources to go to Campeche to get their clay. For them Yo’ K’at was their only source. The potters did not know what to do. So they went directly to the owner of Yo’ K’at to talk frankly about the situation. The owner tried to assure them that he was willing to sell clay to the poor but not to the rich. In return, he simply wanted to use his own truck and be compensated for the cost of transportation, which was 14 percent of the cost of a bag of clay. This issue and the declining clay resources at Hacienda Yo’ K’at were the subjects of a meeting between the potters and the owner of the hacienda on July 9, 1987 (Diario de Yucatán, June 10, 1987:1, 4). The principal purpose of the meeting was to “establish the truth of the rumors that there was a scarcity of clay at Yo’ K’at” (Diario de Yucatán, June 10, 1987:1). Previously, potters had complained about the lack of clay because of over-exploitation. At the meeting, the president of the potter’s gremio asked his fellow craftsmen to openly express their concerns about the problem. One miner affirmed that there was an alarming scarcity of clay at Yo’ K’at. During the previous week, he had opened a new mine, and after 162

How Has Clay Procurement Changed?

digging for more than seven meters, he filled only five sacks of clay weighing ten kilograms each. Another miner said that clay was present, but it was of minimum quality. A third miner expressed himself similarly (Diario de Yucatán, June 10, 1987:1, 4). In answer to such complaints, the owner of the hacienda had already permitted the miners to open more shafts to search for new clay deposits. He reiterated his offer and said that miners could expand the area of clay exploitation but only on the condition that the clay be distributed equitably among the potters and not exported to other communities. As a result of this meeting and the continuing scarcity of clay, the president of the potters’ gremio and his cousin (MoBrSo) were commissioned to investigate the clay resources at Yo’ K’at and its alleged scarcity. One of these potters traveled to the clay source in Mama with the owner to determine whether the clay was easier to mine there than at Yo’ K’at. Clay Exhaustion and the Constraints of Elite Control

One of those potters commissioned to investigate other clay sources believed that the problem was not the lack of clay at Yo’ K’at but its scarcity within the limits of the tract allocated for mining (approximately 10,000 square meters). This scarcity had been verified by miners who had dug new shafts within this plot, but none of them found more clay. Nevertheless, one of the potters commissioned by the meeting to investigate other clay sources recommended that clay mining be moved closer to the buildings of the hacienda; the owner, however, would not permit miners to dig clay outside of the area that he had allotted for clay mining, because he did not want the mine shafts to pose a danger to his cattle or interfere with his fields of maize. By 1988, clay mining continued to intensify and the clay was becoming increasingly scarce. At this time, one clay miner reported that there were about ten vertical shafts that had been dug into the clay deposit, and by 1989, all of the mines had become interconnected. The last productive mine was dug about 1986 and yielded 10,000 sacks of clay, but it was on the edge of the section set aside for clay mining. The two men who dug this shaft had mined out all of the clay and then removed the supporting columns. They were the first miners to abandon mining at Yo’ K’at. Another miner had also dug a mine near the edge of the area allocated for clay mining. He found some clay, but it, too, was soon gone. By this time, the owner of Yo’ K’at had given permission for some exploratory mines outside of the zone allocated for mining, and one miner and his son tried sinking some new shafts there. They dug one shaft and found no clay. Then they dug another well beyond the depth that they should find clay (ten meters), 163

How Has Clay Procurement Changed?

but they found none. They ultimately dug four new shafts around the outside edge of the designated zone for mining but found no clay. It was thus clear to the miners that the clay within the designated mining area was exhausted. They said that it was possible that there was more clay at the hacienda, but they could not afford to sink more mines and come up with nothing for their efforts. Consequently, by January 1992, all clay mining at Yo’ K’at was abandoned, but not before a total of approximately forty shafts had been sunk into the designated mining area. None of these shafts, however, had produced more clay. In the context of my thirty-two-year ethnographic study in Ticul, it is clear that clay exhaustion at Yo’ K’at was artificial and resulted from the constraints that the owner placed on clay mining. From a survey in and around the hacienda in 1968, geologist B. F. Bohor and I found that clay is abundant at the hacienda but in locations different from those that miners have been allowed to dig. In 1968, the clay deposit was 1.5 meters thick, and although its areal extent was unknown, massive amounts had been removed from the mines there. Furthermore, observations and measurements by Bohor in the mine interior indicated that the clay deposit sloped upward toward the hill ridge south of the hacienda. Indeed, we found one such deposit within a meter of the surface about fifty meters east of the road leading from the highway to the main house of the hacienda and much closer to the hill ridge (Figure 5.1). Another deposit occurred on the floor of a large marl (sah kab) quarry between the hacienda and the hill ridge. This deposit had been exposed after the marl had been removed for building the adjacent highway. According to informants, however, the clay from these sources was undesirable for making pottery because it caused large vessels to crack, but it could be used for small objects. It was probably no worse than some of the clay imported from Campeche in the 1990s and was much closer to Ticul. Besides these sources, three other clay mines on the hacienda had been used at various times during the first half of the twentieth century (Figure 5.1). Clay in these mines, however, also became exhausted, or potters had been prohibited from mining clay there. Eventually, potters were allowed to return to obtain clay, and new sources of clay were discovered. If pottery production continues in Ticul, clay mining at Yo’ K’at will probably be restored in due time with new sources uncovered there. Elite Control and Alternative Sources

The elite control over clay resources at Yo’ K’at resulted in a history of clay procurement that is parallel to the history of the Yo’ K’at source when clay from there was unavailable or too costly. When the mine at Yo’ K’at yielded poor­quality clay or became exhausted, potters had to obtain their clay from other 164

How Has Clay Procurement Changed?

sources. Indeed, in 1978, Silvia Terán (1981:17) noted that there were several other clay sources in the municipio of Ticul besides Yo’ K’at. Mejorada Source. In the nineteenth century, both potters’ oral history and Mercer (1896:162) suggested that Ticul potters were obtaining their clay from elsewhere besides Yo’ K’at. Unlike the twentieth century, the clay mined at Yo’ K’at during the nineteenth century was not good for making pottery. So potters obtained their clay from a source located one-half block west of the Plaza of Mejorada, located three to six blocks northeast of where most of the potters lived (Figure 5.2). One elderly potter said that he had procured clay there for many years during the early 1900s and when the manager of Yo’ K’at (at that time, Elario Medina) prevented potters from mining clay at Yo’ K’at between 1924 and 1934. The clay from the Mejorada source, however, was used for small vessels because potters considered its quality to be inferior to that of the Yo’ K’at clay. Although other potters said the owner had used the Mejorada clay to make cooking pottery, in 1965, 1966, and 1968, the owner himself said that it was only used for making coin banks, figurines, and small food bowls. To my knowledge, this clay was never used by any other potter after 1965. A visit to the Mejorada clay source in 1997 revealed that it had been abandoned since 1982. Only a slight depression (less than 5 cm deep by 10 cm by 30 cm) remained where a massive hole had exposed the clay source in 1965, 1966, and 1968 (Figure 5.2). Surface soil covered the depression, and there was no evidence of any mining, spoil piles, or subsidence to indicate the existence of a clay mine in the past. Had I not seen the mine repeatedly in the late 1960s, I would not have believed the owner when he said that the slight depression had been the site of a former clay mine. Why was the Mejorada source abandoned? With the abundance and availability of the superior clay from Yo’ K’at in 1982, it was probably easier and more convenient to buy that clay than to mine one’s own, even with a mine on one’s own houselot. Because the Yo’ K’at clay was delivered by specialists, the owner of the land over the Mejorada source could devote all of his time to making pottery. Other Ticul Sources. Over the years, Ticul potters have obtained clay from several other sources besides the Yo’ K’at and Mejorada sources. A survey of clay sources that Bohor and I made in 1968 revealed that clay pockets sometimes occur in marl mines and these pockets may have been sources for potters’ clays. We also found extensive clay deposits in the floors of large marl quarries in nearby San Juaquín and in more distant Santa Rosa, along the highway to Muna. 165

How Has Clay Procurement Changed?

Figure 5.2. Sketch map of the area surrounding the Mejorada clay source in 1966. By

1997, no evidence of this source remained. (Drawn by Michael Anderson)

Three other clay sources reportedly were used by potters when the Yo’ K’at clay was unavailable. One occurred on the property of a man with a surname of Bracamonte in a houselot located at the northwest corner of Calles 23 and 30. A second source was located on the property of a man surnamed Cuevas near the base of the hills on the south edge of Ticul. In the 1930s, when the manager prohibited clay extraction from Yo’ K’at, clay was reportedly obtained from a mine on the west edge of Ticul north of the tramway line to Yo’ K’at. Potters had to procure it surreptitiously, however, and called it “clandestine clay.” The land 166

How Has Clay Procurement Changed?

Table 5.1. Clay sellers and clay sources used by production units in 1997. The number of production units (N = 22) indicates the number for which data are available. Dzitbalché source

Unspecified source/seller

Seller: Carlos Beto Number of pro- 15 5 1 (Dzitbalché) duction units 1 (Campeche) using clay from source Number of pro- duction units using clays from two sources

3

2

Number of pro- duction units using clays from three sources

1

1

Tepakán source Gil 3

Rosalino 1

2

1

Pepe’s source Clay not sold 2

belonged to Hacienda Yo’ K’at at that time and its manager did not want potters taking clay from the mine. So he broke glass bottles and dumped them into the mine along with large rocks so that the potters would not steal the clay. Expanded Sources in the 1990s. During the process leading to the abandonment of clay mining at Yo’ K’at from 1989 to 1991, potters had to rely on clay from their stockpiles or obtain clay from other sources. About this time, entrepreneurs from the State of Campeche started to bring clay to sell to potters. By 1994, these sources had expanded to three different locations. Two of these locations were the private sources of two different potters, but clay from a third source was supplied by an entrepreneur who brought a truckload of clay to Ticul every month from the source used by the Tepakán potters on communal land. By 1997, the number of sources had expanded to four. The two potters still used their own private sources, but the remaining potters obtained their clay from two other sources (Table 5.1). Unlike the deep mines in Yo’ K’at, the clay sources in Campeche occurred in open marl quarries in which the marl and rocks had been removed for construction purposes (Figure 5.3). Small clay pockets often occur in the marl deposits, but the floor of all of these quarries consisted of extensive clay deposits. Furthermore, two of the four quarries had sides with extensive clay deposits reaching within a meter or two of the ground surface.

167

How Has Clay Procurement Changed?

Figure 5.3. A generalized diagram of the profile of the marl quarries in Campeche that

were used as clay sources in the 1990s. After the rocks and marl were removed for construction purposes, a thick layer of clay was exposed at the base of the quarry. Width of the quarry in the diagram is not to scale; the size and extent of these quarries vary greatly and may be as large as several hectares. (Drawn by Michael Anderson)

The Tepakán Source. Located along the west edge of the old highway between Calkiní and Dzitbalché, this source was used by potters from Tepakán (Arnold et al. 2000:303). In 1994, the quarry was fenced and locked, but by 1997, it was open and clay was being mined in an area stretching about forty meters along the highway and thirty meters from it. Pepe’s Source. After being denied access to the Yo’ K’at source in 1986, Pepe mined clay at the Tepakán source and then purchased fifty hectares of land that included a large abandoned marl quarry near Dzitbalché (Arnold et al. 2000:303). Like many such quarries, extensive clay deposits occurred on its sides and floor. Although he used this clay primarily for his two production units, he sold some of it to other potters who were relatives. In other cases, he provided clay to potters who made vessels for him but who were not otherwise employed by him. By 1997, he had removed massive amounts of clay, including 5,000 bags dug in one area alone. By 2002, Pepe was still mining clay there. Jose’s Source. During 1993, a second potter ( Jose, a pseudonym) also acquired his own private clay source. He had purchased rights to obtain clay from another marl quarry near Calkiní that lay directly across the highway from the Tepakán source (Arnold et al. 2000:303–304). The clay had been excavated from the clay layer that was exposed after the marl had been removed for construction pur168

How Has Clay Procurement Changed?

Figure 5.4. The Dzitbalché clay source located at the base of a marl quarry east of

Dzitbalché in the State of Campeche in 1997. Because the quality of this clay was poorer than that from the Tepakán source, this source had recently been abandoned when this visit was made in 1997. (Photo by Michelle R. Arnold)

poses. By 1994, eight pits had been dug into the clay bed in an area ten to fifteen meters from the highway. By 1997, the quarry had been greatly expanded and again used for mining marl. The rocks and red earth (k’an kab) on the surface had been moved with a bulldozer and pushed into the area with the clay pits, and access to the clay deposit was restricted to a corner of the quarry within five meters of the highway. A few holes had been dug in the clay layer here, but they did not appear to have been used recently and were too small to provide much clay. During my first visit to Jose’s production unit in 1997, his son confirmed that Jose had had his own clay mine for the previous seven to ten years and had mined clay there every six months. When I visited his production unit a second time, however, Jose insisted that he did not have his own clay source but bought clay from the entrepreneurs who brought it to Ticul. From the evidence of my visit to his source in 1997, he might have recently abandoned it because the renewed marl mining had covered his clay pits on the quarry floor. The Dzitbalché Source. The Dzitbalché source is also located in a privately owned marl quarry east of the bypass around Dzitbalché (Arnold et al. 2000:303). 169

How Has Clay Procurement Changed?

In 1997, clay had been dug from the clay layer at the base of the quarry (Figure 5.4) and along some of its sides at its east end. By 2002, however, this source had been abandoned and was a pasture for cattle. In 2002, the owner of the quarry had built a restaurant nearby along the highway and he said that he had abandoned selling clay to Ticul potters about six years previously (1996). At first, this seemed inaccurate because I had visited the quarry in 1997 and it seemed to be still in use. In retrospect, however, I remembered that I had seen bags filled with clay in the clay pits in 1997 (Figure 5.4). Some were partially full; others were split open. These observations puzzled me at the time since I had never seen such bags around active clay sources, but I associated them simply with a pause in mining because of the rainy season. Nevertheless, such evidence was also consistent with abandonment behavior, particularly in light of the owner’s comments in 2002. How Do Changes in Clay Procurement Affect Procurement Organization? Besides the changes in the source of clay, one of the most dramatic changes in clay procurement consisted of the development of first part-time, and then fulltime, mining specialists. This change involved three major transitions. The first transition consisted of the development of part-time specialization when potters stopped getting their own clay and began relying on part-time specialists to mine it for them (Figure 5.5). The second transition involved the development of full-time mining specialists from part-time specialists (Figure 5.5). The third transition occurred when clay sources moved out of the Ticul area and miners were no longer a part of the community. Rather, miners were hired in the State of Campeche by nonlocal entrepreneurs who delivered the clay to Ticul potters. Emergence of Part-time Specialists

Before the 1940s, male potters went to Yo’ K’at to dig their own clay. Husbands procurred the clay if their wives were potters. Beginning in the early 1940s, however, mining specialists emerged who were potters or former potters who had abandoned the craft (Table 5.2). Some potters still continued to mine their own clay (Thompson 1958:66) and a few mined their own up through the late 1960s. But as specialists began mining and selling clay, potters began buying their clay from specialists rather than mining it themselves, and this change is supported by the trend lines showing the decrease in the number of potters mining clay (Figure 5.5). This change to part-time mining specialists was influenced by at least two factors: distance and agricultural marginality. 170

How Has Clay Procurement Changed?

Figure 5.5. Plots showing the trends in the increasing specialization of clay mining

between 1965 and 1988. Although the number of clay miners who were potters declined, the number of miners increased along with the number of miners who were relatives of miners. As potters ceased mining clay, some of the miners’ kin became clay mining specialists.

Distance to Clay Resources. The first factor that influenced the development of part-time miners was the distance from Yo’ K’at to the potters’ houses. Whether potters or specialists mined the clay, it still had to be transported five kilometers to the edge of Ticul and then an additional one to two kilometers to the potters’ houses. Since the boundaries of the hacienda stretched to the edge of town, the rails of the hacienda’s tramway extended into Ticul and clay was transported to the potters via the tramway. Beginning about 1943, the first specialist to sell clay to potters was a hauler named Serfín Canto, who had a horse-drawn cart. He hired one of the workers on the hacienda to dig the clay, put it into sacks, drag them through the length of the mine, and lift them to the surface. The sacks were loaded on horses, taken to the tramway, brought to Ticul, and delivered by Canto from the end of the tranvía. Economic Marginality. A second factor that influenced the development of part-time mining specialists was emerging agricultural marginality. In the 1940s, potters were also maize subsistence agriculturalists (milperos), but sometime during this period, locusts destroyed the maize crop and caused much poverty. As a result, one swidden agriculturalist who had married a potter stopped cultivating

171

How Has Clay Procurement Changed?

maize, began to mine clay and temper, and sell them to his in-laws. Subsequently, two other potters began to mine and sell clay. Later, they were joined by two more potters who were brothers. About 1965, the brothers’ cousin (MoBrSo) also became a clay miner along with the sons of the man who switched from being a peasant farmer to a clay miner in the 1940s. By 1973, another potter reportedly mined and sold clay for a brief period. All miners during this period, however, were part-time specialists and none relied exclusively on clay mining to make a living. Because these mining specialists tended to be potters rather than haulers, they still had to make provisions for transporting the clay to the potters’ houses. In 1965, clay was delivered to houses in a horse cart but was still brought to the edge of Ticul by the tranvía. By 1970, however, the rails for the tranvía had been removed and miners contracted with the owner of a pickup truck to deliver the clay. Thus, the former two-stage process of using the tranvía and a horse-drawn cart was combined into the use of a single delivery vehicle. Emergence of Full-time Specialists

Between 1970 and 1984, great changes occurred in the organization of clay procurement. First, the part-time miners of the late 1960s had evolved into eight full-time mining specialists. Second, individual miners changed from being potters, or former potters, to miners who were not potters at all. Five of these miners, however, were related to potters (Table 5.2). Why did full-time mining specialists emerge and why did mining shift from potters to non-potters? Why did potters stop mining clay? Shift in Mining Technology. The shift from part-time to full-time miners coincides with a shift in mining technology from that of a single mine (Figures 5.6–5.9) to using multiple vertical shafts. From 1965 to 1970, the clay mine consisted of a single underground horizontal tunnel (Figure 5.6) and anyone could enter the mine and dig clay—individual potter and clay seller alike (Arnold 1971:32–33). Clay was mined from the sides of the tunnel, creating mined-out cavities (Figure 5.9). As one cavity collapsed, mining moved into another area and created another cavity. As time progressed, clay extraction moved deeper and deeper into the mine and became more and more dangerous. By 1968, the mine extended sixty-three meters from the entrance shaft and terminated in a large excavated cavity (Arnold 1971:32–33; 1985:63; Arnold and Bohor 1977; Bohor 1975:115; Figure 5.6). Removing clay from the mine involved dragging sacks of clay through the mine to the surface. Near the portal, the access tunnel narrowed substantially and required a miner to take the rope through the opening first and 172

1940s 1965

1966

1970

Year 1984

1988

Not Ticul Tepakán Dzitbalché No data No data 0 Probably none Probably none No data Probably all No data Probably all

1997

Not Ticul Tepakán Dzitbalché No data 0 0 Probably none Probably none

1994

Note: a. These kin indicate the principal relationships. Other relationships outside the nuclear family, but accessible through these individuals, include WiMo, WiFa, WiBr, WiSi, Si, MoBr, and MoSi. In all cases, these relationships tie into large extended families of potters, although some overlap exists in the families indicated here.

Ticul Ticul Ticul Ticul Home community Ticul Ticul Yo’ K’at Yo’ K’at Clay source used Yo’ K’at Yo’ K’at Yo’ K’at Yo’ K’at 8 8 6 4 Number of miners 4 5 8 0 4 3 Miners continuing since previous observation 3 — 1 4 1 4 6 5 Miners who are Ticul potters 6 0 6 0 Non-potter miners related to Ticul potters 0 0 Relatives of miners who are pottersa Wi, Mo, Wi, Mo, DaHu, DaHu, SiHu SiHu 5 2 5 3 4 2 Parent, sibling, or cousin of another miner 8 New miners since 1965 8 2 1





Table 5.2. The changing number of clay miners from 1965 to 1997 and their relationships to potters.

How Has Clay Procurement Changed?

Figure 5.6. Profile and plan view of the clay mine at Yo’ K’at used in November 1968. Distances and depth are to scale, but angles of underground mining tunnels are approximate. There were several more collapsed mined-out cavities along the main mine tunnel that were not measured or sketched for safety reasons. For interior views of the mine, see Figures 5.7–5.9. (Drawn by Michael Anderson)

then pull the sack of clay through the tunnel (Figures 5.6 and 5.8). Without a helper, miners would have to go back into the mine again and again, reattaching a rope to each sack of clay to be removed. Needless to say, clay mining was difficult, and entering the mine, digging the clay, and bringing bags of it to the surface were a daunting task. Any miner could accomplish the task alone, but clay could not be extracted very rapidly or efficiently that way.

174

Figure 5.7. The surface portal to the Yo’ K’at clay mine in 1968. The entrance continues (Figure 5.8) at the base of this shaft through an opening partially visible in the right center of the photograph.

Figure 5.8. The narrow entrance to the Yo’ K’at clay mine at the base of the entrance shaft. This entrance is so small that in 1968 when this picture was taken, the toes and heels of one’s boots scraped the bottom and top of the tunnel simultaneously.

How Has Clay Procurement Changed?

Figure 5.9. An underground excavated cavity in the Yo’ K’at clay mine in 1968. After a distance of about five meters underground, the mine opens into a large excavated cavity before it descends again into the clay layer (Figure 5.6).

By 1984, the technology of clay mining had changed completely as a consequence of the increased intensity of mining. Rather than a single mine, clay was extracted at the base of six vertical shafts that were five to eight meters deep (Figure 5.10). At the base of each shaft, quarrying created large cavities that penetrated into the clay deposit. Sometimes mining connected one shaft to nearby shafts. Mining Technology and Changing Procurement Organization. This shift in mining technology from a single horizontal mine to multiple vertical shafts had several implications for the organization of clay procurement. First, this new technology required a different organization of miners, which resulted in the development of full-time mining specialists. Unlike clay procurement before 1970 that occurred in a single common mine, the use of vertical shafts required an initial investment of capital and a considerable amount of labor before the first sack of clay was removed. First, a miner had to dig through considerable overburden to access the clay deposit, requiring the purchase of specialized equipment such as explosives and a steel crowbar to dig through one-half meter of rock near the surface. Once the explosives had loosened the rocks, they had to be broken

176

How Has Clay Procurement Changed?

Figure 5.10. Generalized profile of the vertical shaft mines at Yo’ K’at in 1984. Distances

and depth are to scale. (Drawn by Michael Anderson)

up and removed. When the miner got below the rock, he had to dig through approximately four to five meters of friable marl and lift it up the shaft. When the miner reached the clay layer, it had to be mined and hoisted to the surface. As the mining face receded horizontally from the shaft, a new shaft was dug into the deposit. This strategy lessened the distance one would have to pull the full sacks of clay underground but required capital and sustained effort before any clay was removed. This investment was too great for a single potter who occasionally mined a few sacks of clay. Second, digging a vertical shaft required more than one miner and could not be done alone without constantly climbing up and down the shaft. At a depth of about five to eight meters, repeated climbing in and out of the shaft was no easy task. It made more sense for one person to dig in the shaft and fill the sacks while another at the top of the shaft lifted the sacks to the surface and emptied them. Two-man teams of miners were thus formed to dig a shaft and remove the clay from it. Each team had exclusive rights to extract clay from its shaft and these rights were respected by other miners. Third, mining is not easy and potters say that one who is not accustomed to the effort will soon tire and will not be able to mine much clay. Since the economic returns of mining are directly proportional to the amount of the clay extracted and the amount of clay extracted is directly proportional to the amount of time, effort, and stamina of the clay miner, excavating clay is not cost-effective for the occasional and infrequent miner. Indeed, one potter remarked in 1984 that after the emergence of full-time specialists, an inexperienced miner would not have the stamina to be able to mine much clay. It is easier, he said, to let a 177

How Has Clay Procurement Changed?

specialist extract clay because a potter could earn more money by making pottery than by digging his own clay. This response is consistent with observations by May Díaz (1970:141) that potters in Tonalá, Mexico, found it more economical to buy clay than to mine it. Finally, using a vertical shaft with a team of miners was safer than the single underground mine. With the vertical shaft technology, some underground mines had two access shafts that provided safety and security during mining. If one area caved in during mining or collapsed because of rainfall, a miner could exit via another shaft. If a cave-in occurred while a miner was absent, he still had access to the clay deposit through an alternative shaft. By way of contrast, a miner working alone in the single mine that existed prior to 1970 had to wait many hours to be rescued if a cave-in occurred. But even mining in a team involved risks of injury and death. When the first non-potter started mining clay, a portion of the overburden collapsed and buried him up to his waist, and the efforts of several men were required to free him. If a companion miner had not been present, he may not have been rescued. In spite of persistent cave-ins, no deaths of miners have occurred at Yo’ K’at between 1965 and the abandonment of mining there in 1992. The exhaustion of the clay at Hacienda Yo’ K’at marked the end of clay mining specialists in Ticul as one by one they abandoned mining there. Three miners switched to temper mining and two of these were still producing temper in 1997. The remainder abandoned mining completely and took up other occupations. New Procurement Organization

With the change in the location of clay sources to marl quarries in Campeche, the organization of clay procurement changed again and two new kinds of mining organization emerged (Table 5.3). For those sources owned or controlled by Ticul potters, workers took a truck to the sources several times a year when clay was needed. Since the sources were so distant, the truck was filled with pottery for sale in order to minimize the cost of the journey. One potter sold his pottery to clients in Tepakán and the other sold it in his store next to his clay source near Dzitbalché. After the pottery was unloaded, the trucks were loaded with clay and returned to Ticul. The second kind of new mining organization that emerged consisted of nonlocal entrepreneurs who sold clay to potters. They owned (or had access to) a truck, enlisted laborers to dig the clay, and then transported it to Ticul. Clay was brought from the Dzitbalché source by two brothers (Table 5.1). One of the two brothers exchanged half of the cost of clay for cash and the other half for pottery and then transported the pottery back to Dzitbalché. 178

Private Private

Multiple underground mines

Yo’ K’at 1988

Marl quarry N of Dzitbalché



Marl quarry N of Dzitbalché

Marl quarry E of Dzitbalché



Large

Small

Small

Large

Vertical integration Vertical integration Vertical integration Specialists hired by owner

Previously a marl quarry Expansion Reduced because of reuse of a marl quarry Previously a marl quarry

Tepakán potters and truck driver

Truck owner

Previously a marl quarry

Expanded Expanded Public (ejido land)

Private (Beto)

Very small Private (usufruct rights)

Private (Pepe)

Private (usufruct rights)

Marl quarry N of Dzitbalché



NE of Dzitbalché Marl quarry 1997

Public (ejido land)

Marl quarry N of Dzitbalché (Tepakán source)

Vertical integration

Previously a marl quarry Marl quarry NE of Unknown Private Dzitbalché

1994



Vertical integration

Specialists

Expansion

Unknown Marl quarry NE of Unknown Private Dzitbalché

Specialists Specialists

N/A Expansion

Kind of procurement organization

Change from previous observation



Unknown

Very small Larger

Single underground mine Private

Yo’ K’at

Multiple underground mines

1965–1970

Yo’ K’at 1984

Year of Type Source Areal extent Land tenure observation

Table 5.3. Summary of changes in clay sources, their areal extent, and procurement organization between 1965 and 1997.

How Has Clay Procurement Changed?

Clay sellers came from the Dzitbalché source more frequently than did those from the Tepakán source. In 1997, one brother came to sell clay from the Dzitbalché source every week and the other brother came three times a week. Another unknown seller came every fifteen days. Clay from the Tepakán source was delivered by two entrepreneurs, but the frequency was unknown, although it was less frequent than deliveries from Dzitbalché. A Surrogate Measure of Production Intensity With the changing organization of clay procurement between 1965 and 1997, it is possible to derive several surrogate indices that reflect the increased growth and intensity of ceramic production during this time. The first of these surrogate indices consists of the expansion of the number of clay sources from one to four. A second index is the change in mining technology that first involved a single underground mine, then used many vertical shafts, and finally evolved into mining in open quarries. A third surrogate index consists of the increase in the number of miners and their change from part-time potters to non-potters who were full-time mining specialists (Table 5.2). The number of miners increased from four to six part-time specialists between 1965 and 1970 and to eight full-time specialists in 1984. Another way to measure the change in production intensity is to compare the amount of clay used in the 1960s with the amount used in 1984. The number of specialists and the changing mode of transportation suggest that clay exploitation vastly increased and greatly intensified during this period. In 1965, a group of four miners dug eight to ten bags of clay at a time and each bag weighed from sixty to eighty kilograms. Clay was delivered to the potters’ houses in a horsedrawn cart that carried a maximum of ten to fifteen sacks weighing a total of 0.6 to 1.2 metric tons. By 1970, clay was brought to Ticul in a small pickup truck that probably carried no more than 0.5 to 1.0 metric ton. Although the frequency of clay delivery for 1965 and 1970 is unknown, it is reasonable to suggest that clay was brought to Ticul no more than once a week. The production process of mixing, forming, drying, and firing requires a minimum of one week but is usually on a two-week cycle. So the clay consumption of all potters between 1965 and 1970 probably was between 0.5 and 1.2 metric tons per week. By 1984, clay was delivered every one to two weeks in a truck that had a gross weight of 4.5 metric tons (10,000 pounds). Subtracting the weight of the truck (about 0.9 metric ton), the weight of clay carried to Ticul was 3.6 metric tons per trip. By comparing the capacity of the truck to the clay delivered between 1965 and 1970, the amount of clay consumed per week in 1984 was thus 3.0 to 7.2 times greater than what it was between 1965 and 1970. 180

How Has Clay Procurement Changed?

With the development of full-time mining specialists in 1984, the intensity of production can be calculated more precisely using the amount of clay as a surrogate because all clay came to Ticul via these specialists. To make these calculations, three variables must be taken into account. The first variable consists of the frequency with which specialists mined clay. Miners dig clay three to four days a week and use one day to deliver it. Clay miners also prepared temper and divided their time between mining clay and mining and preparing temper. Up until 1988, the Ticul paste required twice as much temper as clay. Translating these proportions into the time spent mining suggests that specialists spent one-third of their time mining clay. In 1984, for example, one miner and his son mined clay for a week and then mined temper for a week or two before returning to Yo’ K’at to mine clay again. If approximately four weeks a year are allowed for fiestas and religious holidays, such as Christmas, Day of the Dead, Holy Week, the potters’ gremio, and family birthdays, then miners spent approximately sixteen weeks annually mining clay. A second variable necessary for calculating the amount of clay mined in 1984 is the weight of each sack. In 1994, one miner estimated that before mining stopped at Yo’ K’at, each sack weighed between thirty and fifty kilograms; the weight of most sacks, he said, was in the middle of this range (thirty-five to forty kilograms). A third variable necessary to calculate the rate of clay production is the number of sacks mined per day. The miner cited above estimated that he could mine fifteen to twenty sacks of clay each day and that his son could mine thirty to forty sacks. If each sack weighed thirty-five to forty kilograms, then he alone mined 525 to 800 kilograms per day. This total is independently confirmed by the miner’s brother-in-law, who said that the miner and his son could each mine thirty sacks of clay per day, but that each sack weighed only twenty kilograms (see also Diario de Yucatán 1987:1, 4). This quantity of clay (600 kg/day) falls in the middle of the range of the miner’s own estimate of 525 to 800 kilograms per day and is within 10 percent of the median value of 662.5 kg/day. Using these variables, the amount of clay extracted in 1984 can be calculated in two ways. Using the value of 600 kg/day per miner and assuming that they were mining three to four days a week, one miner would thus dig 1,800 to 2,400 kilograms per week. If each miner spent one-third of his work year mining clay (sixteen weeks), then he would extract 28.8 to 38.4 metric tons a year. If each miner in all four two-man teams mined clay at this rate, miners would dig between 230.4 and 307.2 metric tons a year, assuming no seasonal adjustment for the rainy season when clay mines may collapse from moisture and may be dangerous or the clay may be temporarily inaccessible. 181

How Has Clay Procurement Changed?

A second way to calculate the amount of clay mined is to use the actual amount of clay mined for one truckload of clay (one delivery event). When miners have accumulated enough sacks to fill a truck, they request the owner’s truck and cooperate in loading and unloading the clay for delivery. A truckload of clay may be brought to Ticul as often as every week. From miners’ accounts, the amount of clay delivered by the truck varies from a low of about 155 sacks to a high of 160 sacks. If each delivery event consists of 3.6 metric tons of clay and two to four delivery events occur every month, the total clay extracted is 7.2 to 14.4 metric tons per month. If miners spend onethird of their work year mining clay (i.e., sixteen weeks), they produced an estimated 28.8 to 57.6 metric tons of clay (four times the monthly delivery amount) in 1984. When the calculations by the two methods are compared, they are quite disparate and fall into the range of 28.8 to 307.2 metric tons a year. By averaging the medians in the range of each method of calculation, the result is 156 metric tons a year. This quantity does not fit well with miners’ self-reported estimates of delivery frequency. If this average is divided by the truck capacity of 3.6 metric tons, the result is 43.3 trips, or delivery events, per year using the truck. If miners produced clay only sixteen weeks a year but spread the deliveries over an entire work year (forty-eight weeks after excluding holidays), the truck would make 0.9 trips per week, which is within the actual delivery frequency of 0.5 to 1 trip per week reported by the miners. After the abandonment of clay mining at Yo’ K’at, clay procurement moved to Campeche. Little is known about the amount of Campeche clay delivered to Ticul because mining was in the hands of nonlocal specialists. One of these quarries in Campeche, however, was owned by a Ticul potter. In 1997, workers traveled to the quarry about three times a year. Each worker reportedly could mine twenty sacks an hour. When a rest period and time for hauling the clay to the truck are taken into account, all of the workers mined 100 sacks of clay in five hours. The truck usually brought eighty sacks of clay at a time and might make two trips on the same day. At a weight of 35 to 40 kilograms per sack and a minimum of eighty sacks per trip to the mines three times a year, these miners produced 8.4 to 9.6 metric tons of clay per year, which was 5.4 to 6.2 percent of the estimate of seventy-two metric tons for all of the production units in Ticul in 1984. This amount is reasonable because this potter has two production units, and together these units are 4 percent of the production units in Ticul but employ 13 percent (20/153) of the potters. How does this compare with other production communities? In a study of one potting compound in the community of San Sebastian in the Valley of 182

How Has Clay Procurement Changed?

Mexico, Sheehy (1988) observed nineteen firings over a period of four months and recorded the amount of clay and fuel used in each firing to develop a clay/ fuel ratio that he then applied to the ancient potters of Tlajinga 33, a potter’s workshop that used the same clay in the nearby site of Teotihuacan. Based on his observations, Sheehy calculated a yearly clay consumption of 4.79 metric tons for the compound. In many respects, it is difficult to compare Sheehy’s data with the data presented here. Sheehy’s study approaches the amount of clay utilized from the point of view of the per-unit consumption of clay, and the data presented here approach clay consumption from the total amount of clay procured at the source. To make these two data sets compatible, the amount of estimated clay produced in Ticul in 1984 (28.8–307.2 metric tons) must be divided by the number of production units (50), yielding a range of 0.576 to 6.144 metric tons per production unit per year. The lower end of this range is probably an appropriate amount for small production units in Ticul, but the higher number is probably too low for the largest units. The amount of clay consumption in San Sebastian (4.79 metric tons), however, falls at the high end of this range, but it does suggest that in kinbased production units, such as those that exist in Ticul and San Sebastian, there is a limit to the increase in the intensity of the production of kin-based production units, and any subsequent increase in intensity for a community must occur through the increase in the number of production units rather than in their size. This same point was emphasized from a different perspective in Chapter 2. Procurement Intensity, Organization, and Production-unit Size Economies of scale in clay procurement organization had a feedback relationship with production-unit size and may have favored the increase in the size of production units. Up until 1970, potters could mine the clay themselves or purchase as little as one bag at a time from miners. By 1984, potters in small production units were at a disadvantage in buying one or two sacks of clay. By that time, clay production had intensified, clay had to be requested ahead of time, and the truck owner wanted to make as few stops as possible. So potters had to buy at least ten sacks in order to receive delivery directly to their house. The capital outlay for such a quantity was more than many small production units could afford. Their owners were limited by the amount of labor available, and the low production volume generated limited capital. Since economic returns occur only every two weeks because of the requirements of the pottery production cycle, the household potter may have limited capital available to buy ten sacks. 183

How Has Clay Procurement Changed?

Furthermore, ten sacks of clay required more storage space than many small production units had available, and they did not have the capital to build more. Consequently, if potters wanted only one or two sacks of clay, they had to go to one of the miners’ houses (or to another potter) to get the clay, pay a much higher price for it, and transport it to their own houses themselves. If the economic situation of a small production unit is precarious (as is often the case) and the owner does not have the cash to buy more than a few sacks, he is disadvantaged even further and may have to use otherwise economically productive time to buy clay. This problem was further complicated by the great demand for clay by the large production units. Such demand was so large that one potter said that if he went to a miner and asked for clay, the miner would respond that he could not sell it to him because the miner had been unable to fill an order of 300 to 400 sacks for a large workshop and could not take the trouble to sell him just a few sacks. If the miners will not sell any clay to him, he is further disadvantaged. If clay was limited and was becoming scarce, as it was in the late 1980s, then obtaining sufficient clay for making pottery could be a problem because it was sold to clients who bought the greatest quantities. Potters in small production units thus may prefer to hire themselves out as wage laborers in a large production unit where the economic returns are more regular and they do not have the constraints on clay delivery to small production units. Do Changing Clay Sources Reflect Evolutionary Social Change? How does clay procurement relate to social organization and elite control? Although potters must obviously select a technologically appropriate clay in order to make a pot successfully, their procurement choices are linked to a variety of other factors, such as the perception of appropriate clay resources, the distribution of these sources across the landscape, their sense of place, and a variety of socioeconomic and sociopolitical factors such as land tenure, organization of procurement, and local politics. Elite Control

The significance of resource restriction and paste standardization in the evolution of ceramic production in the Rice (1981) model rests on the meaning of “elite control.” Precisely what does it mean to control a resource like clay when, unlike precious metals and semiprecious stones, clay has no inherent value in its raw form, different qualities cannot be identified by inspection alone at the source, and generally massive amounts are required for any significant economic return? 184

How Has Clay Procurement Changed?

The Ticul data provide some context for understanding the meaning of elite control. It is clear that elite control of clay resources has existed for a long time in Ticul and is a product of land tenure. Even when mineral rights belong to a largescale political entity such as the state, land tenure is still in the hands of elites and clay procurement may be controlled. Nevertheless, much variability exists in how this control actually works in practice. First, elite control is not a presence/absence phenomenon as the Rice (1981) model implies. Rather elite control represents a range of behaviors. At one extreme, elite control consists of sponsorship and cooperation with potters, such as funding the digging of the clay mine. At the other extreme, however, access to the clay source is denied and mining is prohibited entirely. Between these two extremes lies a range of control behaviors, such as simply requiring permission to mine clay and assessing a right-to-mine fee. Or control may simply mean the control of the distribution of the clay once it has been mined so that elites control distribution and the potter pays transportation costs. In a capitalistic society such as Ticul, this type of control can have significant effects on poor potters because the efficiencies of economies of scale may mean denial of delivery to potters who use little clay. Second, clay is not a desirable resource to anyone except potters and it often occurs on eroded, marginal, or otherwise agriculturally undesirable land (Arnold 1985:183–196). Such poor eroded land occurs in the Northern Valley of Guatemala (Arnold 1978a, 1978b, 1985:183–196); around Quinua, Peru (Arnold 1975a, 1993:48–71); and around Ráquira, Colombia (Duncan 1996). Although clay mining still competes with agricultural use in Ticul, the clay resource is tertiary to agriculture and other more important resources, such as construction materials. With greater intensity of production, clay mining becomes more important, but it still comes from land whose primary use is for henequen, cattle, or maize, for example, or by way of contrast, from land that is useless for any other purpose (such as the base of marl quarries). Intensive elite control of clay resources that alter potters’ clay procurement behavior thus appears to occur when elite economic interests conflict with clay procurement in specific situations, such as (1) locations where mines (whether deep or shallow) destroy agricultural land (Arnold 1975a, 1993:62–66) or are hazards for agriculture (cattle and fields) and make farming difficult or impossible, (2) when the labor needs for extraction technologies (such as henequen production) compete for labor on the same land, (3) when the death of workers occurs in the clay mines, and (4) when micro-political conflict threatens the elite. Finally, the notion of elite control of clay sources has its limitations because it refers to the behavior of elites, not the behavior of potters. Clay resources 185

How Has Clay Procurement Changed?

are sufficiently ubiquitous that once elite control becomes excessive or sources become exhausted, potters obtain their clay from any number of other locations that, although not preferred, are acceptable and can provide resources until the preferred clay resource becomes available again (see also Arnold 2000). Procurement Technology

The growth and complexity of ceramic production in Ticul can be measured by the changing clay procurement technology, procurement organization, and the development of mining specialists. This evolution has followed two different tracks. The dominant track involves the development of task segmentation and specialization. The other track consists of the development of vertical integration in which individual potters have their own exclusive clay sources and control their own resource procurement, vessel production, and distribution of finished pottery. This evolution is also reflected in the increased amount of clay consumed between 1965 and 1984. Social Complexity

How do the changes enumerated in this chapter relate to the evolution of socioeconomic complexity? First, the processes involved with the growth and evolution of the craft do not necessarily relate to the control of ceramic resources. Even with access and distribution tightly controlled by elites, potters did not obtain their clay exclusively from a single restricted source (Table 5.3). When one source of clay became inaccessible or its quality declined, potters found other sources to meet their needs. This adjustment has occurred repeatedly in the past. Furthermore, increased elite control of resources has not narrowed the number of clay resources over time. Rather, because of the greater intensity of production, potters have obtained clay from additional locations and thus clay sources have expanded, becoming more dispersed rather than more restricted (Table 5.3). In the thirty-two-year period described here, the clay procurement area at Hacienda Yo’ K’at expanded from a single mine in 1965 to six mines in 1984, to ten in 1988, and to forty before mining was abandoned there in 1992 (Table 5.3). By 1994, all clay was coming from three different sources eighty kilometers away in the State of Campeche. Two potters had their own sources, and clay was delivered from a third source by an entrepreneur. By 1997, clay mining in Campeche intensified with clay coming from four different sources. Two of the sources were controlled by Ticul potters, but most clay was coming from two other sources not controlled by them (Table 5.1). One of these was a public source on ejido land and another was a private source. By 2002, one of these sources was abandoned because of the poor quality of its clay, and clay came from other sources. 186

How Has Clay Procurement Changed?

Efficiency

The change in clay procurement technology over time has affected the social organization of mining. Originally, clay was mined in a single deep underground mine, but mining specialists emerged in conjunction with two-man mining teams who dug vertical shafts to extract the clay. These shafts ensured the safety of the miners and protected them from being trapped by cave-ins. Other miners recognized the rights of those who invested time in digging shafts. The evolution of clay mining technology in Ticul strikingly parallels the evolution of mining coal and metal ores in North America. This similarly suggests that at least in part, efficiency drives the evolution of mining technology. The first stage of mining was the discovery of surface deposits. In Ticul, if any pottery clay ever did occur on the surface, it has disappeared long ago and this stage of mining has long passed. As surface deposits were exhausted, miners followed the mineral deposit underground, and this activity produced a mine shaft through which the mineral was extracted. In Ticul, miners first used a common shaft to extract their clay. During the next stage of the evolution of mining technology, miners dug multiple shafts into mineral deposits and mining law protected such claims once they were registered. In Ticul, miners who dug shafts in the clay deposit had usufruct rights from the owner of the source and from other miners that gave them control over the clay that they dug from their shafts. Eventually, many shaft mines became unprofitable, exhausted the mineral within them, or became inefficient to operate given the costs of mining versus the benefits of the extracted mineral. In Ticul, the clay extracted via shaft mines became exhausted and it became unproductive to use that technique to find more clay deposits. Finally, with changes in technology and transportation, some mineral deposits were exploited by open pit mining, which is used when it is more efficient to remove tons of overburden with new technology to recover the mineral below. Similarly, mining clay for Ticul potters shifted to open pit mining in marl quarries where the overburden (marl and rock) had been removed and used for construction purposes or was removed by clay miners to ensure safety and access to the clay. Mining thus occurred in open quarries rather than in underground shafts because it was easier, quicker, safer, and more productive than shaft mining. Task Specialization

The development of mining specialists followed its own trajectory over time and has occurred in several stages. Originally, potters mined their own clay. Then a few potters became part-time miners and sold their clay to other potters. This change was partially stimulated by the economic marginalization created by destruction of the maize crop by locusts in the 1940s and continued until the late 187

How Has Clay Procurement Changed?

1960s. By 1984, however, miners were no longer potters but full-time specialists dividing their time between mining clay and temper. Finally, as clay sources moved outside of the community, mining became the domain of specialists who were not related to potters at all, were from a different community, and had no contact with the potters except through the truck driver who delivered the clay. Although there are no data on these new mining specialists, they appeared to be contract laborers hired by the owner of private land to dig clay (the Dzitbalché source) or hired by an entrepreneur for digging clay on communal land (the Tepakán source). Finally, social change has led to the secularization of clay procurement. Through the 1980s, Ticul clay was technologically superior to clays from other locations. Clay is ubiquitous in Yucatán and is located under a layer of limestone and marl (Figures 5.3, 5.6, and 5.10), but potters recognized that the Yo’ K’at clay was different. Its technological superiority probably gave rise to the naming of the hacienda “over clay” (Yo’ K’at). Otherwise, why would a hacienda be named for a resource that is relatively useless to anyone besides potters? This association of location and the quality of the clay from there was reinforced through religious rituals (the novena for San Pedro, the hacienda’s patron saint; Arnold 1971). Eventually, clay was procured from Yo’ K’at not just because of its technological superiority but also because of its association with a particular sense of place with significant religious connotations. Potters thus responded to the ideology of the “place” of Yo’ K’at rather than the technological superiority of the clay, and clay was not obtained from elsewhere unless clay from Yo’ K’at was unavailable. When good-quality clay became temporarily exhausted at Yo’ K’at, one response was religious. Later, quality clay was rediscovered, and the association of good-quality clay with the potters’ sense of place reinforced the religious connotations of the hacienda as the only location of high-quality clay. Eventually, the Yo’ K’at clay source became exhausted, and clay came from elsewhere, but the religious and place associations of the traditional source were not transferred to the new sources. Task specialization can also be charted on a trend line that shows the decrease in the numbers of potters who were clay miners (Figure 5.5). Other trend lines show the nature of this process through the increasing use of individuals who were relatives of potters. These data demonstrate that clay procurement has its own evolutionary trajectory, and the development of the complexity of ceramic production can be measured by the changing mining technology, development of mining specialists, changing procurement organization, and development of exclusive clay sources for individual potters. These data directly challenge the notion that elite control 188

How Has Clay Procurement Changed?

results in greatly restricted clay sources. The evolution of clay procurement is far more complex. This chapter has only examined the relationship of social change to the behavior and organization of one aspect of raw material procurement—obtaining clay. Another part of the relationship between social change and raw materials consists of the procurement of temper and will be discussed in the next chapter.

189

Chapter

How Has Temper Procurement Changed?

six

C

lay is not the only critical resource necessary to make pottery. Suitable clay for making pottery must be plastic enough to form a pot but not so plastic that the newly formed vessel will lose its shape. To achieve a middle ground between these two very real material constraints, the potter may need to alter the properties of the raw clay in order to successfully fabricate a vessel. The potter changes these properties by adding non-plastics (such as sand, chaff, or ground potsherds) or by adding a raw clay that contains non-plastics (Arnold 1985:21–32, 2000; Rice 1987:74–75, 406–413; Rye 1981; Shepard 1956). Most naturally occurring clays contain some non-plastic materials, but adding more to the paste reduces its plasticity and changes its working, drying, and firing properties, such as reducing shrinkage (Rice 1987:74–75). These additional non-plastics have been called “temper” by archaeologists, but this material is often not simply non-plastics but rather a material that contains both plastics and non-plastics (Arnold 1971, 1975b, 2000; Rice 1981:406–413). 191

How Has Temper Procurement Changed?

In communities where temper is added to clay, obtaining temper is just as crucial as obtaining clay but potentially more variable since a variety of materials can produce the same effect on the paste and affect its performance characteristics. Because potters may consistently choose one option out of several available, temper choices may be more social than technological. Consequently, archaeologists often regard pottery temper as an indicator of tradition and cultural affiliation. Because the use of temper is often critical to preparing the paste, one expects to find important links between temper procurement and the organization of production, just like those developed in the last chapter with respect to clay. These links include (1) the distribution of temper resources across the landscape, (2) a sense of the appropriate place to obtain it, (3) land tenure, (4) the technology and organization of procurement, and (5) a variety of other social and political factors. More important for the purposes of this book are the interrelated questions, How does social change affect temper procurement? and What happens to temper procurement as ceramic production evolves and becomes more complex? Between 1965 and 1997, temper procurement in Ticul changed just as radically as clay procurement. First, since the great decline in the demand for cooking pottery, the fabrication of non-cooking pottery has dominated Ticul ceramic production. Second, the location of mining and preparation of temper for noncooking pottery has shifted and, along with it, its procurement technology and organization. Do the changes in temper procurement follow the same trajectory as that for clay described in the last chapter? Changes in Temper for Cooking Pottery Although the same clay is used for cooking pottery and non-cooking pottery, cooking pottery (i.e., pottery placed on the fire) is tempered with a different material (called hi’ ) that consists of a macrocrystalline calcite (Arnold 1971; Thompson 1958:69–71). Pottery with this kind of temper begins at least in the Terminal Classic period (Puuc Unslipped Ware) and is found in at least one site near Ticul (Yo’ Sah Kab; see Arnold 2005b). Although the production of hi’-tempered pottery has declined greatly since 1965, the procurement and mining location of hi’ temper have not changed much. In 1965, hi’ temper was mined in two caves (’aaktun) in the nearby hill ridge: ’Aktun Hi’ and ’Aktun Lara. Although little hi’ was used between 1965 and 1997, it was mined in ’Aktun Hi’ by two potters who formerly made cooking pottery. They either used it to make cooking pottery themselves or sold it to others who made the occasional cooking vessel on special order. Compared to the 192

How Has Temper Procurement Changed?

massive volume of the non-cooking pottery produced between 1965 and 1997, the production of hi’-tempered pottery is minuscule. Changes in Temper for Non-cooking Pottery Unlike the temper for cooking pottery, great changes have occurred in the location, organization, and intensity of mining temper for non-cooking pottery. The four most important changes are (1) the expansion of mining locations, (2) changes in procurement practices, (3) increasing task segmentation of temper preparation, and (4) a shift from potters mining their own temper to full-time specialists who were not potters. Changes in Procurement

Changes in temper procurement have been constrained by two major factors: the semantic categories of temper subclasses and the sense of place for temper procurement. Although these factors have a technological and physical basis, they are mediated and expressed by socially transmitted categories. These categories and associated beliefs limit the range of behaviors and have thus constrained potters’ technical choices and changes in procurement technology. Nevertheless, changes have still occurred. The semantic category that potters use for the temper used in non-cooking pottery is sah kab. This expression, however, is homophonous with a naturally occurring marl mined for mortars, plasters, and surfacing materials, but it is semantically different. Unlike the marl used for construction purposes, sah kab temper contains white earth (sak lu’um) as its most critical ingredient. Potters say that white earth provides strength (mu’uk’ ) to support the walls of large vessels so that they do not sag and crack (see Arnold 1967b, 1971:35–36) but is not as crucial for small vessels such as coin banks and figurines. White earth also helps the temper remain sticky and acts as a cement for binding the fabric of the vessel together. The ethnography of sak lu’um, its properties, and analyses of it by X-ray diffraction (Arnold 1967a, 1967b, 1971) reveal that sak lu’um is the clay mineral palygorskite (also known as attapulgite; Bailey et al. 1971). Potters recognize that the presence of sak lu’um sets Ticul pottery apart from pottery made in other communities. When potters made water-storage vessels in the late 1960s, they said that consumers believed that Ticul vessels were technologically superior to those made in other communities (such as Mama and Tepakán) because the Ticul vessels kept the water cooler and gave it a better taste. These properties, potters believe, were the result of the sak lu’um in the temper. 193

How Has Temper Procurement Changed?

The second factor that constrained changes in temper procurement was the potters’ connection to a specific place for temper procurement. Both white earth (sak lu’um) and the temper made with it come from a place called Yo’ Sah Kab, located 3.3 to 3.6 kilometers northeast of the Plaza of Guadalupe along the road to Chapab. Yo’ Sah Kab literally means “over sah kab,” but it can be freely and more appropriately translated as “the temper mines.” Potters’ oral history and descriptions by others have indicated that in 1965, Yo’ Sah Kab has been the preferred source of sah kab temper for most of the twentieth century (Barrera Vásquez 1937:164; Brainerd 1958; Rendón 1947; Thompson 1958:69; Varela Torrecilla 1990:205). Temper mining at Yo’ Sah Kab, however, occurs much earlier. In 1967, pottery dating to the Terminal Classic period (A.D. 800–1000) was discovered within one mining area, and an archaeological site dating to the same period occurred within fifty meters of it (Arnold 2005b). From 1965 until about 1990, Yo’ Sah Kab was the only location for mining and preparing sah kab temper because it was the only location where sak lu’um was found in the immediate area of Ticul. Yo’ Sah Kab encompasses approximately ten hectares (0.1 km2) and consists of four relatively distinct areas. Three of these areas were used for temper procurement between 1965 and 1997. One area consists of approximately three hectares of the communal land located on the south side of the Ticul-Chapab road. A second area is also part of the ejido and includes about four hectares on the north side of the road. A third area consists of three hectares owned by the Maya Cement Company (Cemento Maya), located on the south side of the road and adjacent to the communal land, but south and west of it. A fourth mining area was not used between 1965 and 1997 and occurred farther southwest, closer to the Finca Xtuk. This area was believed to be about 160 years old and was said to be the source of clay as well as temper used by a previous generation of potters. Informants were never able to locate the mines in this area. The habitual mining of temper at Yo’ Sah Kab, potters’ oral history, and their sense that Yo’ Sah Kab was the location to obtain temper are mutually reinforced by the existence of an archaeological site at the mines (Arnold 2005b). In the late 1960s, potters reported that the older generation of potters was afraid of mining temper on the north side of the road because of the archaeological site there. This fear was reinforced by local folklore. In the early 1900s, potters claimed that people from Chapab came to Yo’ Sah Kab to procure sak lu’um for medicinal purposes (Arnold 1967b; Arnold and Bohor 1975, 1976). They said that it was the place of many snakes, and one informant’s father had been bitten twice by snakes while he was mining temper there. When a visitor encountered a large snake that 194

How Has Temper Procurement Changed?

blew air and had a beard and hair on both sides of its head, the potters were afraid to return. They thus feared the area around the archaeological site and only one person returned to mine temper, but he gave offerings of the maize drink posole every Friday to placate the serpent (Arnold 2005b). Changes in Land Tenure and Expansion of Sources

During the last 100 years, temper mining moved from one area of Yo’ Sah Kab to another. Oral history reveals that these movements were related to changes in land tenure patterns and to the excessive exploitation of the raw materials used to prepare temper. As land tenure changed and temper resources became scarcer, procurement locations changed. The oldest mines at Yo’ Sah Kab are located on the communal land, but the private land and the communal land there have an interrelated history. Before the land reform after the Mexican Revolution, the communal land was part of Hacienda San Ignacio Xtuk (now called Finca Xtuk). The cement company property was formerly owned by Lucas Medina. At that time, most of the temper was prepared on the Xtuk property and only a little was mined on the Medina property, but potters had to pay fifty centavos per month for the right to mine temper in both locations. The ejido portion of Yo’ Sah Kab probably did not have communal ownership until 1937. After the Mexican Revolution of 1910, a new constitution (1917) provided for the restoration of lands to the peasants that had been taken from them in the 1850s (Meyer et al. 1999:364, 524–525). The restoration process, however, was implemented slowly, beginning with the presidency of Venustiano Carranza and continuing through the administration of President Lázaro Cárdenas in the late 1930s (Meyer et al. 1999:556, 563, 577–578). Land reform in Yucatán, however, was delayed until 1937, when President Cárdenas arrived at the port of Progreso to “preside over the largest single episode of agrarian reform ever carried out in Mexico” ( Joseph 1980:158–160). Within two weeks of his arrival on August 7, “the entourage of tecnicos [technicians] had effected the transfer tenure, consolidating unequal segments of hundreds of haciendas into 272 tracts of publicly-owned communal land called ejidos” ( Joseph 1980:160). The communal portion of Yo’ Sah Kab was apparently one of those ejidos, and as a result of this change, potters could obtain their temper at Yo’ Sah Kab without cost. About 1953, the Maya Cement Company bought the Lucas Medina property. The company wanted to use the white earth for making cement, so a caterpillar tractor removed the surface rock for a distance of 100 meters along a narrow strip. At some point, however, the company found that they could not use the 195

How Has Temper Procurement Changed?

white earth. They then stopped mining, abandoned large piles of the white earth, and opened the property to the potters, charging them fifty centavos for each bag they took. The potters then stopped mining temper in the communal land, purchased the white earth from the cement company, and brought it to Ticul to prepare their temper. By June 1965, two brothers acquired rights to use the cement company property as a rancho for their cattle. Because the property was comparatively unexploited for temper mining and the quality of the materials was excellent, the two men established a right-to-mine fee of fifty centavos for each sack of temper extracted. This fee was easy to collect because one of the two men owned a horse cart and transported the temper to Ticul. Consequently, almost all of the mining activity between 1965 and 1970 was concentrated on the cement company property. Sometime after 1965, the mining fee was raised to seventy-five centavos, per bag and then to one peso. At this point, all of the miners moved back to the communal land where they could mine their temper without paying a fee. The right-to-mine fee on the cement company land was then lowered to fifty centavos, where it remained from January 1966 through September 1970. Most of the miners then returned to the cement company land; they believed that the quality of the raw materials was better there than in the communal land because the cement company land was less exploited and the white earth was more abundant. Mining on the cement company land thus was worth a small right-to-mine fee. By 1984, almost all of the mining activity had moved away from the private and communal mining areas on the south side of the Chapab road into the communal land on its north side. This area included a Terminal Classic site that was completely destroyed in the process (Arnold 2005b). For a distance of 100 meters along the road and extending approximately 100 meters into the forest, miners had dug mines and piled up discarded tailings from temper preparation. Since this area had no evidence of active temper mining in the late 1960s, the extent of the mining north of the road indicated that temper mining had greatly expanded since that time. Most important, the mining activity that destroyed the Terminal Classic site suggested that the site had been built directly over a large deposit of palygorskite that probably had been used for pottery temper and for mining white earth used in the production of the ancient pigment Maya Blue (Arnold 2005b). Although some mining still occurred on the south side of the road, the former mining areas were almost entirely overgrown with forest. The Maya Cement Company still owned the land nearby, but no mining activity took place there.

196

How Has Temper Procurement Changed?

Depletion of Raw Materials and Changes in Sources

By 1984, the temper mines at Yo’ Sah Kab had become heavily exploited because miners had begun to sell the by-products of temper preparation. After the discarded tailings from the preparation process had been used three or four times, they were sold by the cartload for spreading on the floors of houses. Miners had also sold tons of white earth to a local entrepreneur who exchanged it for finished pottery at a factory in Monterrey. The heavy exploitation of temper and its by-products fueled potters’ concerns about the depletion of white earth, and they complained to township authorities. By 1988, authorities had responded that because the communal land was owned by the people, the white earth from that land could be sold by the miners but not shipped outside of the township for profit. The depletion of materials for temper preparation at Yo’ Sah Kab eventually led to a change in mining location. About 1983, one of the men who transported temper to Ticul (don Yoy Canto) told one of the miners that he had found white earth on his land five kilometers closer to Chapab and asked the miner to verify it. The miner found that white earth was indeed present on Canto’s land, and by 1988, Canto had begun to mine and prepare temper there and sell it to potters. By 1994, temper mining continued on the communal land at Yo’ Sah Kab; activity was concentrated along both sides of the road to Chapab for approximately 300 meters, beginning at a point 3.3 kilometers from the Plaza of Guadalupe in Ticul. No mining took place on the cement company land, but potters said that the remainder of Yo’ Sah Kab was very exploited. Mining was so difficult, potters said, that a miner could work all day and only produce four sacks of temper. Nevertheless, because Yo’ Sah Kab was public land, anyone could go there and mine temper without monetary cost. By way of contrast, mining was reportedly so easy at the Chapab mines that a miner could produce forty sacks of temper in one day. As a consequence, the Chapab source became the major source of temper by 1994 and supplied temper to the six largest production units in Ticul. By 1997, temper production had almost totally shifted to the Chapab source (Table 6.1). Mining there had intensified greatly, and the Chapab source was the source of temper for most (N = 23) of the production units in Ticul. The mining areas had expanded from a linear span of fifty meters in 1994 to a distance of approximately 125 meters west of the entrance road and fifty meters east of it. Mining still continued at Yo’ Sah Kab in 1997 but at a greatly diminished rate. For those potters who mined their own temper, Yo’ Sah Kab was still the source of choice. It was closer to Ticul, and because the mines were on communal land, anyone could go there and mine temper without paying a fee. 197

How Has Temper Procurement Changed?

Table 6.1. Comparison of temper miners at Chapab and Yo’ Sah Kab and their clients in 1997. The number of miners working at Yo’ Sah Kab includes one miner who did not work regularly. Two production units obtained temper from both sources and no data exist for the number of clients of two miners. Mining and sales of temper at the Chapab source are controlled by one man, and miners sell to any client. Number of miners Potters who mine their own temper and do not sell it to others Total number of clients Number of miners with one client Number of miners with two clients Number of miners with three clients

Chapab source Yo’ Sah Kab source 10 0 23 N/A N/A N/A

6 4 6 3 0 3

Development of Mining Specialists

During the first half of the twentieth century, most potters procured their own temper. Some went to the temper mines every other day, prepared a half sack of temper, and then carried it back to Ticul. Others went daily and prepared only enough temper to make pottery for that day. Two principal factors influenced the development of temper mining specialists. First, like those for mining clay, specialists emerged, in part, for transporting mined temper. Since potters could carry only one sack at a time to their houses (a total of 4.5–5.5 km), the amount of labor expended in mining, preparing, and carrying the temper required limited the amount of time a potter could spend making pottery. In the 1940s, potters began to transcend this constraint by hiring a horse cart to transport their temper. Because the hauler required a minimum load of twenty sacks to pay for the trip, six potters pooled their sacks so that they would have enough to pay for the cart. Four others did not like to pay transportation costs and traveled to the temper mines at night, prepared their temper, and then brought it back to Ticul on the bus from Chapab in the morning. Another potter used his horse to transport temper and sell it to potters. Even though transportation was critical in getting temper to Ticul, the first mining specialists were potters and not haulers. Although one informant said that no temper mining specialist existed before the 1940s, another informant said the first part-time temper mining specialist began in the 1920s and was a maker of cooking pots (Antonio Tilam) who perhaps chose to mine temper as the demand for cooking pottery abated. Then, three other potters began selling temper to others. By 1966, ten part-time specialists mined temper (Tables 6.2 and 6.3; Arnold 1967b, 1971). Six other potters still mined their own temper regularly and two

198

Ticul YSK 8 5 2 4 Wi, Mo, SiHu 2 7 —

Ticul YSK 5 1 0 5 Wi, Mo, SiHu 2 4 —

Year

Ticul Ticul Ticul Source community of miners YSK YSK YSK Temper source 4 10 2 Number of miners (does not include potters who mined their own temper) 4 2 — Miners continuing since previous observation 3 8 — Miners who are potters 1 2 2 Non-potter miners related to potters Kinds of relatives of non-potter Wi Wi, Mo Wi, Mo miners who are pottersa Miners related to another minerb 0 5 2 0 8 — New miners since 1965 — 6 — Potters who mined their own temper — 0 — 0 0 3 All 0 —

2 1 1 SiHu

Ticul/Chapab YSK/Chapab 3 —

1994

0 6 4

2 2 1 SiHu

8 All 0

— 0 0 0

Ticul/Chapab YSK/Chapab 10 6

1997

Notes: a. These relationships are the principal relationships. Other relationships outside the nuclear family include WiMo, WiFa, WiBr, WiSi, Si, MoBr, and MoSi. In all cases, these relationships tie into massive extended families of potters, although there is some overlap in the families indicated here. b. These relatives include parents, siblings, and cousins. In 1997, these relationships included brothers-in-law.

1966

1988

1965

1984



1970



Table 6.2. The changing number of temper miners from 1965 to 1997 and their relationships to potters. Trend lines for some of these data are shown in Figure 6.2. No data exist about miners for 1967 and 1968.

How Has Temper Procurement Changed?

Table 6.3. Number of temper miners and their clients in 1966 variously identified by buyers and sellers. Some clients buy temper from more than one miner. Each team consists of a father and his son. Total Prepared temper for themselves alone Miners who have one or more clients (seen as clients per miner)

Miners Clients 14 6 1 1 4 1 (team) 1 (team)

25 N/A 1 2 4 9 13

mined it occasionally. Some of these potters used the services of a local hauler who had a horse-drawn cart, although one potter preferred to bring his temper on the bus from Tekit that passed Yo’ Sah Kab. By 1970, many of the same miners were still active; with one exception, all were potters and mined temper part-time. By 1984, mining was in the hands of five full-time specialists who split their time between mining temper and mining clay (Table 6.2). Temper was more difficult to prepare than clay because temper required mining, crushing, mixing, and screening whereas clay only needed to be dug and placed in sacks. Temper was still delivered to potters by horse-drawn cart (Figure 6.1). A second reason why full-time mining specialists emerged was that potters found that making pottery was more productive than mining temper. This explanation is consonant with the development of the division of labor first articulated by Adam Smith (1953 [1776]), who argued that the segmentation of tasks enabled workers (in this case, potters) to be more efficient by saving the time required to change from one activity to another. This shift to specialization resulted in realizing economies of scale. When one potter, for example, was asked why he did not mine his own temper in 1984, he replied that he preferred to make pottery rather than prepare temper because he could only prepare three or four sacks of temper a day and would earn only 400 pesos per day. But he could earn 2.5 times that amount (1,000 pesos) by making pottery. Mining specialists, on the other hand, could mine more than twice (ten sacks) as much temper per day as a potter who was inexperienced at the task. Potters thus appeared to maximize their skills for making pottery rather than take time to mine temper in order to achieve greater financial benefit. By 1988, the number of miners increased to eight (Table 6.2) and the intensity of temper production at Yo’ Sah Kab reached its peak (Figure 6.2). Some temper miners only mined temper, but others mined temper and clay, just as most miners had done in 1984. 200

How Has Temper Procurement Changed?

Figure 6.1. Delivering temper to a potter’s house in 1984. Temper is mined by specialists and then delivered by haulers to the potters’ houses.

After the demise of the clay source at Yo’ K’at in 1991, two clay miners started to devote all of their time to mining temper at Yo’ Sah Kab. Because their father was a hauler, they had access to his horse cart and thus could profit by hauling temper as well as mining it. By 1994, the Chapab mines had become an important additional source of temper. Unlike mining clay at Yo’ K’at and mining temper at Yo’ Sah Kab, temper production at Chapab was in the hands of the owner who hired local workers. None of the Ticul miners worked at the Chapab mines, and none of the Chapab miners were potters. The increased intensity of temper production at the Chapab source created feedback with production scale. The high demand for temper and the increased amount of temper production achieved by economies of scale at the Chapab mines had evolved to the point that professional miners and haulers did not make temper deliveries to small production units that only needed a few sacks. A potter with a small production unit thus faced a choice of either paying an inflated price for temper purchased from another potter or preparing it himself at Yo’ Sah Kab. If he chose to buy it from a miner, he had to pay for it before delivery, the delivery date was uncertain, and often delivery did not occur when he needed it. Consequently, several potters mined their own temper and brought it to Ticul on their bicycles or on the bus from Chapab. 201

How Has Temper Procurement Changed?

Figure 6.2. Plots showing trends in the increasing specialization of temper mining

between 1965 and 1997. The clearest trend (the trend line with the highest R2) suggests that the number of temper miners who were potters declined. Less clear trends suggest that the total number of miners increased along with the number of miners who were related to another miner and the number of miners related to a potter. The decline in the number of temper miners who were potters is similar to the decline in the number of clay miners who were potters. The trend lines of the number of miners and the relatives of miners are not as clear as those for clay miners (Figure 5.5).

By 1997, mining at the Chapab source was supervised by the late owner’s son. Ten miners were working in five two-man teams. One member of each team mined the raw materials while the second person mixed them with the weathered tailings and then sifted the mixture. As in 1994, all miners came from Chapab and none were potters. Except for two workers (who may have been ritual coparents [compadres]), most of the workers were relatives of the late owner and included his brother, son, nephews, and brothers-in-law. Unlike the temper from Yo’ Sah Kab, temper from the Chapab source was sold on-site. Since the Chapab mines are 8.6 kilometers from Ticul, only those individuals who had access to a truck could buy temper at the source. Although two of the potters who owned trucks bought temper in this way in 1997, professional haulers also purchased the temper at the source, transported it to Ticul, and sold it to potters. One of these haulers was related to the owner of the Chapab source. Large production units purchased temper weekly. Meanwhile, as more and more potters bought their temper from the Chapab source, the number of specialists working regularly at Yo’ Sah Kab declined. Six 202

How Has Temper Procurement Changed?

Figure 6.3. Mine opening at Yo’ Sah Kab in 1966 showing the depth of the palygor-

skite layer below ground level. In most locations, the material at this depth was nonpalygorskite-bearing marl (sah kab, which was used for construction purposes).

miners (Table 6.1) worked at Yo’ Sah Kab, regularly selling their temper to potters who were not serviced by the Chapab source. Another potter occasionally mined temper and sold it to other potters. At least two potters still mined their own temper at Yo’ Sah Kab, and both transported it to Ticul themselves using a platform tricycle. Specialists, Mining Technology, and Mining Rights. The development of mining specialists in 1984 coincided with the development of usufruct mining rights for individual miners. In 1984, even though the mining occurred on communal land, each miner had his own mining and preparation area and the rights to work in that area were respected by other miners. These rights continued to be recognized at Yo’ Sah Kab in 1994 and 1997. Usufruct rights appeared to be a consequence of the mining techniques used because digging a mine requires substantial labor. Unlike clay mining, however, temper mining does not require teamwork because the mines are easily entered from the surface; thus, raw materials can be mined by one person because the white earth layer is one to two meters below ground level (Figure 6.3). Miners gain access to this layer through an existing depression, a collapsed mining area, abandoned mine, or, if necessary, digging through the surface rock. Then they 203

How Has Temper Procurement Changed?

tunnel horizontally into the white earth layer. As they remove the material, cavities are formed and pillars are left to support the cap rock above; some of these tunnels extended underground for as much as fifteen meters. As a result, miners have mutually respected excavation rights that ensure that only the miner who dug a mine benefits from the raw materials removed from it. Consequently, when potters mine their own temper, they avoid the mines and preparation areas of professional miners and either dig their own mines or use abandoned mines. Changes in Temper Variability

During the thirty-two years of this study, the behavioral variability of temper preparation has changed and existed within the boundaries of the semantic categories of raw materials that potters consider necessary to prepare temper. In some cases, the potters’ choices of these raw materials were technologically based; at other times, the choices were not technologically based. In all cases, the changes in potters’ choices of raw materials over time were dominated by the significance of white earth (sak lu’um), the crucial ingredient of the temper. Basics of Temper Preparation. The baseline against which change in temper preparation will be evaluated consists of the behavioral chain of temper preparation used in 1965. The first step involves selecting tailings (ta’achach) that cover the preparation areas. High-quality tailings contain white earth (sak lu’um) and have few rocks. The miner assesses the presence of white earth by crushing the tailings in his hand. If they can be crushed easily, white earth is present and he can proceed with the process of preparing temper (Arnold 1971). If, on the other hand, he cannot crush the tailings with his hand, sak lu’um is not present and he must use tailings from another preparation area. This test of friability is, in reality, a test for the presence of the clay mineral palygorskite. In its natural state, palygorskite is dry and very hard. When exposed to moisture, palygorskite falls apart and thus weathered tailings with palygorskite can easily be crushed with the hands. The practical ethnomineralogical knowledge of the potter thus corresponds to the scientific understanding of the mineral categories involved (Arnold 1967b, 1971). After the miner has selected suitable tailings, he gathers them into a small pile with his pick and crushes them by beating them with a log. Then, he discards rocks that do not break, mixes the pile with his pick, crushes the mixture again, and repeats these procedures as desired (Figure 6.4). The second major part of the preparation process involves obtaining raw material (nooy) from the mines (Figure 6.5). Miners select a mine carefully to be sure that the nooy from it contains white earth (sak lu’um). Once the sak lu’um is 204

Figure 6.4. Temper preparation in 1965 showing crushing, bringing bags of nooy from

the mines, and screening.

How Has Temper Procurement Changed?

Figure 6.5.The interior of a mine at the Chapab source in 1997 showing a woman min-

ing nooy for preparing temper. The excavated layer is relatively pure palygorskite. This woman is the only known female temper miner during the thirty-two years of research in Ticul. (Photo by Michelle R. Arnold)

identified in the wall or the floor of the mine, the miner digs the nooy, brings it to the preparation area, and dumps it on the pile of prepared tailings. The amount of nooy used is variable and is measured by the number of trips the miner uses to carry nooy to the preparation area (Figures 6.4 and 6.6). After he accumulates enough nooy, he mixes it with the ta’achach. After he crushes the mixture again, he props a screen over an empty sack and sifts the mixture, discarding the larger rocks that were not crushed (Figures 6.4 and 6.6). Behavioral Changes in the Variability of Temper Preparation. The patterns of temper preparation just elaborated are general patterns but with many variations that have changed through time. The categories of nooy in the mines and ta’achach from the preparation areas remained important semantic categories for influencing behavior, but between 1965 and 1997, the ratios of these two components in temper preparation changed. Equally important, however, is the variation in the preparation of temper at a particular point in time. In 1966, for example, great behavioral variation occurred in temper preparation because of the number of potters and miners preparing temper (Table 6.4). 206

Figure 6.6. Temper preparation area at the Chapab source in 1997. The excavated areas in the foreground are the source of the weathered screenings (ta’achach) from previous preparations, which are mixed with the freshly excavated nooy from the mines to prepare temper. (Photo by Michelle R. Arnold)

How Has Temper Procurement Changed?

Figure 6.7. The shelter used to prepare temper at the Chapab mines in 1997. (Photo by Michelle R. Arnold)

Potters provide several explanations why such synchronic and diachronic variation exists. First, mining sak lu’um and obtaining good-quality nooy are very difficult because of the effort required to break up the larger pieces of sak lu’um; only the smaller pieces can be crushed. Some miners thus deliberately mix an inferior raw material (called kut nooy) in the temper because it is easier to dig than other classes of nooy that were more appropriate for pottery making (Arnold 1971). As a result, temper prepared with kut nooy can be prepared more rapidly, allowing the miner to earn more money from his efforts. A second reason for the variation in temper preparation concerns the seasonal climate patterns. In the rainy season, the tailings cannot be screened easily because the clay minerals in them absorb water and the particles stick together. Some miners thus do not mix any tailings in their temper during the rainy season and instead prepare temper with only freshly mined nooy. The seasonal weather constraints can be transcended by screening temper in a shelter (Figure 6.7). Even though no structures were associated with temper mining at Yo’ Sah Kab, once mining moved to the privately owned mines near Chapab, the owner built a simple thatched-roof structure that provided shelter for workers to screen temper during rainy weather and protected sacks of 208

2

One pile

Use of ta’achach

4

(1) Crushes it first before mixing it with nooy; does not crush again before sifting (2) Used none

Observation of mining on 1/22/66

Description by miner

G

3 H

Mixes it with the crushed ta’achach of the nooy Uses one-half sack of kut nooy and one-half sack of regular nooy

None

Crushes it three times

5

2–3

F

contunied on next page

Another informant said that he used mostly ta’achach and only a little nooy; a small amount of the nooy was kut nooy

Only white nooy

Crushes and mixes it three times, then mixes it with nooy, crushes the mixture again, mixes, and then crushes again before sifting Communal (ejido) land

Only white nooy

Comment

One pile

Cement company land

Location

E Yes 3

1 D

C

None 2 B

A

Trips to obtain Miner nooy

Table 6.4. Behavioral variation in sah kab temper preparation in 1966.

Use of ta’achach Location

2

Yes

Crushes and mixes it with a pick three times Communal (ejido) land

Yes M Yes

2–3 L

K

J

3 Yes I

Trips to obtain Miner nooy

Table 6.4—continued

Comment

(1) Uses nooy only in the rainy season. (2) For figurines and cajetes, he brings ta’achach from Yo’ Sah Kab and mixes it with nooy from the sah kab mine in his yard in a ratio of three to two; another account indicated that the ratio was two to one

Uses nooy only in the rainy season; one informant thinks that he uses a lot of ta’achach that is mixed with dirt because his pottery is very dark

Says that the white clay from the mine in his houselot has sak lu’um in it and he uses it as sak lu’um. For making large pottery, he uses one part white clay from the mine and one part red marl (sah kab, for construction purposes) from this same mine. After the white clay is dried, he mixes it with marl and crushes the mixture. Then, he sifts it and adds one part sah kab temper from Yo’ Sah Kab to keep the pottery from cracking. For cajetes, figurines, maceteros, and other items not used for water, he uses a little clay from Yo’ K’at and white clay and the red marl from his mine. Then, he dries the materials, crushes and sifts them, and adds water to make the paste

Uses white nooy, although observation by informant indicates he uses kut nooy and nooy with dirt in it

How Has Temper Procurement Changed?

prepared temper. By 1997, the structure was larger than it was in 1994 and the thatched roof had been replaced with tar-impregnated cardboard sheets. A third reason for the variation in temper preparation concerns a miner’s desire to avoid the risk of using inferior raw materials. In order to avoid producing vessels that crack and break, some potters prefer to mix in more tailings than nooy (Table 6.4) in order to avoid inadvertently adding the inferior kut nooy. Others prefer to use tailings exclusively for the same reason. A fourth explanation for behavioral variation in temper preparation is related to the effect of time and the development of mining specialists. Between 1965 and 1984, miners changed from being potters and part-time mining specialists to being full-time miners. As a result of this change, preparation methods changed because miners were less concerned about quality. Full-time miners were not mining temper for their own use and thus were more concerned about increasing their comfort and decreasing the effort that was required to break up the tailings. Working in the mines was cooler and easier than working in the sun. As a result, some miners put fewer tailings in their temper and more nooy (including inferior kut nooy) than did potters who mined their own temper. Fifth, different potters/miners have different temper preparation recipes (Table 6.4). Some of this variation is related to technological and social explanations already described, but other variation is individual and has no such explanation. This type of variation could best be explained as “technological style” (Lechtman 1977) or the result of “technological choice” (Lemonnier 1986), in which behavioral variation occurs without technological basis and has no adverse technological consequences. Such variation, however, still occurs within hierarchical technological constraints already described. The behavioral variation in temper preparation observed in 1965 and 1966 appeared to continue through time with one significant difference: variation declined as the number of individuals mining temper declined. Individual variation was greater when potters were mining their own temper in 1965 and 1966 and mining specialists were part-time (Tables 6.2 and 6.3). Between 1965 and 1984, however, most individual potters stopped mining their own temper and full-time mining specialists emerged. As a result, fewer individuals mined temper in 1984 than in 1966 (Tables 6.2 and 6.3). One miner, for example, used more tailings than others. His son, however, used more dirt than tailings and this amount of dirt exceeded the amount safely allowed (one-half sack). Adding dirt and kut nooy diminishes temper quality, but they can be used in moderation with acceptable results. With the reduction of the number of specialist miners at Yo’ Sah Kab to three in 1994, the behavioral variation in temper preparation was further reduced even though the opening of the new mining area near Chapab provided a new source of variation. 211

How Has Temper Procurement Changed?

Temper preparation at the Chapab mines is basically the same as at Yo’ Sah Kab but with some minor differences. First, unlike the variability evident among miners who prepared temper at Yo’ Sah Kab, Chapab miners say that they mix nooy from the mines and ta’achach in roughly equal parts, even though they make no attempt to measure the amounts of nooy and ta’achach precisely. Second, at Yo’ Sah Kab, the nooy and ta’achach were crushed with a pole. At the Chapab mines, however, temper preparation occurs on the vehicle track that extends along the edge of the mining area that has been built up from the tailings. The Chapab miners make use of the truck movement along the track to crush the tailings. In so doing, they reduce the effort involved in temper preparation. Tailings are dug from the vehicle track, mixed with the nooy, and screened (Figure 6.6). The rocks in the mixture are discarded so that they will not end up in the screenings again. The screenings and large chunks of sak lu’um from the mines are then used to fill the holes in the road so that it can be crushed by the truck traffic and the rains can make it fall apart. A sixth source of variation in temper preparation consists of sifting prepared temper immediately before use in order to create a smaller particle size. This practice is used by some potters who make smaller mold-made objects and figurines with thin vessel walls. During the thirty-two years of this research, I have observed at least four potters following this practice. The screenings (shish) may be discarded or may be ground on a flat rock, mixed with an equal amount of sak lu’um, and then sifted and used as temper again. How does the behavioral variation in temper preparation actually affect the composition of the temper? In January 1966, 96.5 percent (28/29) of the potterymaking households in Ticul (Arnold 1967b, 1971) were surveyed. The analyses of temper samples collected from these production units revealed considerable qualitative variation in the mineral composition of the temper. Each sample contained calcite, dolomite, and palygorskite. Some samples also contained montmorillonite and a degraded material that was weathered montmorillonite. Sixty percent of the samples contained palygorskite as the only clay mineral and 74 percent contained no montmorillonite, whereas the remainder contained mixtures of palygorskite, montmorillonite, and degraded material (Figure 6.8). Specialists and Changes in Temper Quality. The development of mining specialists has affected temper quality. When potters prepared their own temper, the variation of temper preparation was inconsequential as long as the quality was good. Since potters cannot control the quality of the temper obtained from specialists, however, they must be sure that the temper quality is suitable for making pottery to avoid damaging their vessels. Variation in preparation could have dire 212

How Has Temper Procurement Changed?

Figure 6.8. Bar graph showing the variation in the amount of clay minerals in tem-

per samples collected in 1965–1966 (N = 35). “Degraded montmorillonite” refers to a weathered and poorly crystalline montmorillonite that is partially interlayered with hydroxy-alumina. The percentage of each clay mineral in these samples is based on the relative peak heights in the X-ray diffraction patterns as interpreted by clay mineralogist B. F. Bohor. Data condensed from Arnold (1971:32–33).

consequences and cause the pottery to sag and crack because of the clay mineral montmorillonite in the temper (Arnold 1971:36). In order to solve this problem, potters have developed several strategies for evaluating and controlling the quality of temper. First, they avoid buying temper from specialists with reputations for selling inferior temper. X-ray diffraction analyses of temper samples collected in 1965 and 1966 confirm this problem. Of the twenty-eight samples of temper analyzed, eleven contained montmorillonite (Arnold 1971:32–33). Although most of these samples were not evaluated by informants, one informant identified two miners who sold inferior temper. X­-ray diffraction analyses of their temper revealed that their temper did contain montmorillonite (Arnold 1971). Two other temper samples thought to contain inferior raw materials (kut nooy), however, did not contain montmorillonite. A second strategy for coping with temper acquired from specialists is to test its quality. Potters have developed two quality control tests to identify a sufficient amount of white earth in the temper (see Arnold 1971:35–36). 213

How Has Temper Procurement Changed?

Figure 6.9. A pile of palygorskite outside a potter’s house in 1988. When the quality of

temper is inferior, potters buy pure palygorskite (sak lu’um) to add to it.

Once potters determine that the temper they have purchased is inferior, they use several strategies to ameliorate the problem. First, they can try, if possible, to obtain temper from a different miner. This strategy, however, is more problematic to implement because it is difficult to change temper suppliers; each miner has regular clients whose requests are filled first. In some cases, miners are so far behind in supplying their regular customers that they will not supply a new client. A second strategy that potters use to avoid the problem of inferior temper is simply to prepare their own temper or send a trusted family member to do it. On October 8, 1984, a significant amount of breakage occurred during one of my informant’s firing episodes, and he blamed the losses on inferior temper. Since a client had given him an order for several large water-carrying vessels, he was afraid that using temper from the same miner would cause the pottery to break again. He did not have time to prepare temper himself and temper from a different miner was not available, so he sent his son to prepare the temper. Later, on October 23, he had the same problem again because a miner had sold him inferior temper. This time, however, he traveled to Yo’ Sah Kab to prepare his own temper. A third strategy of dealing with the inferior temper consists of mixing pure white earth with temper; several potters kept quantities of white earth especially for this purpose (Figure 6.9). The white earth is stored outside so that it can break

214

How Has Temper Procurement Changed?

apart in the elements, is ground up as needed, and then is mixed with inferior temper. A variation of this strategy is to mix tailings (ta’achach) with inferior temper. Since good-quality tailings contain white earth, adding tailings in effect increases its proportion in the temper. A Surrogate Measure of Production Intensity Like clay procurement, temper mining has intensified since 1965 and there are three surrogate measures of that change. The first measure consists of the increase in the number and size of areas of temper mining already described (Table 6.5). A second measure is the change from potters’ mining their own temper to the emergence of full-time specialists, who were not potters and had no relationship with potters except the economic transaction of selling temper to them (Table 6.2, Figure 6.2). A third measure consists of the increase in the amount of temper produced. Unfortunately, estimates of total temper production were not available for 1965–1970 because too many individuals mined temper and no data were collected about their mining frequency or the amount of temper they mined. With the reduction in the number of miners between 1965 and 1984 and the development of full-time mining specialists, it was possible to estimate the specialists’ temper production for 1984. First, an estimate must take account of the frequency of mining. In 1984, miners divided their time between mining clay at Hacienda Yo’ K’at and mining temper at Yo’ Sah Kab, and the ratio of the time spent mining these two resources corresponds to the clay-to-temper ratio (1:2) used in preparing the paste. One miner and his son, for example, said that they worked one week mining clay and two weeks mining temper. On a yearly basis, this ratio means that two-thirds of the year would be spent mining temper. If four weeks a year are allowed for family and religious celebrations, then full-time miners worked forty-eight weeks a year, and thirty-two weeks were devoted to mining temper. The rainy season does not appear to significantly affect the frequency of temper production. Rains do not come until early or midafternoon during the rainy season, and so miners and potters alike could travel to Yo’ Sah Kab in the early morning and spend five to six hours mining temper before the rains came. Second, calculations of temper production must include the number of sacks mined each day. Nine informant accounts of daily temper production rates at Yo’ Sah Kab in 1965, 1967, 1984, and 1994 provide a mean production rate of 8.05 sacks per miner per day and a median of 8.75 sacks per day. The average of these two values yields 8.4 sacks per miner per day and will be used for subsequent calculations. 215

Private Expansion toward the east and south Expansion 50 m by 350 m

Multiple surface mines; Public some deep mines

Multiple surface mines; Private some deep mines

Chapab source

10 m by 100 m

Yo’ Sah Kab 1997

Chapab source

Multiple surface mines



Public

Multiple surface mines; some deep mines

Yo’ Sah Kab 1994

Expansion

Expansion

Expansion

Expansion

No data

No data

Public

Multiple surface mines

Yo’ Sah Kab

1988 Expansion to the east

Expansion

N/A

East of ejido land Expansion north of road

Private Public

Multiple surface mines

Multiple surface mines

Yo’ Sah Kab

Yo’ Sah Kab



N/A

About 10,000 m2 along south side of road

Multiple surface mines Public 1 965–1970 Yo’ Sah Kab (ejido land)

1984

Change in area from previous observation

Year of Type Areal extent Land tenure observation Source

Specialists

Specialists and potters

Specialists

Specialists and potters

Specialists and potters

Specialists and potters

Specialists and potters

Specialists and potters

Kind of procurement organization

Table 6.5. Changes in temper sources, their areal extent, and procurement organization between 1965 and 1997. To compare with the changes in clay using the same categories, see Table 5.3.

How Has Temper Procurement Changed?

Third, mining estimates must take into account the weight of each sack. Informant accounts of sack weight are fairly consistent over time. In 1965, one potter said that the weight of a sack was fifty kilograms. In 1984, a miner said that each sack weighed between fifty and seventy kilograms. In 1994, another miner estimated the weight of each sack to be between fifty and sixty kilograms. If all of the median values for each range are averaged, the resulting value of fifty-five kilograms per sack represents a plausible weight to use for further calculations. Finally, temper production estimates need to be based on the number of days that miners work per week. Miners work five to six days a week with a pause on Sunday (and sometimes Saturday) for socializing and resting. The total weekly production of one miner is thus 42.0–50.4 bags per week. Assuming a weight of fifty-five kilograms per sack, the weekly production of one miner is thus 2.31– 2.77 metric tons. With one miner working thirty-two weeks a year, the annual production for one miner is 73.9–88.6 metric tons. With six miners working at that rate in 1984, the yearly temper production at Yo’ Sah Kab was 443.4–531.6 metric tons. At this rate, it is not surprising that potters thought that Yo’ Sah Kab was overexploited in the late 1980s. This estimate thus provides some empirical support for understanding why temper production moved from Yo’ Sah Kab to the mines near Chapab in the 1990s. By 1997, temper production at the Chapab source intensified while production at Yo’ Sah Kab declined. Yo’ Sah Kab continued to be the mining location of choice for those potters who mined their own temper and for six specialist miners (Table 6.1). Assuming that each miner at Yo’ Sah Kab produced 8.4 sacks of temper each day and each sack weighed fifty-five kilograms, these miners collectively would produce 16.6 metric tons per six-day week. Assuming that they were full-time specialists and worked forty-eight weeks a year, they would produce a total of 796.8 metric tons of temper a year. This number does not include the temper mined by potters. The private ownership of the Chapab source and the centralized organization of mining permit precise calculations about the amount of temper produced there. Since the owner of the mine collects one peso for every bag of temper mined, he monitors production amounts. He said that collectively the miners produced 800 bags of temper each week. Assuming a six-day work week, this quantity amounts to 13.3 sacks of temper per worker per day. This rate attests to the ease of mining temper at Chapab compared to the 8.4 sacks per day mined at Yo’ Sah Kab in 1984. Since each bag weighs about sixty kilograms according to the miners, the amount of temper produced at Chapab is forty-eight metric tons each week. 217

How Has Temper Procurement Changed?

Since Chapab miners are full-time specialists, and assuming that they take an estimated four weeks off for family and religious holidays, they work an estimated forty-eight weeks per year. Consequently, their aggregate annual temper production in 1997 was 2,304 metric tons. Like the increase in the amount of clay mined over time, it is also possible to use the amount of temper production as a rough measure of the increase in the intensity of pottery production between 1984 and 1997. Combining the total production of both the Chapab and Yo’ Sah Kab sources in 1997 results in a total production of 3,100.8 metric tons a year by full-time temper miners. This amount is 5.8 to 7.0 times the amount of temper produced at Yo’ Sah Kab in 1984 and is a surrogate measure of a greatly increased production intensity between 1984 and 1997. Control and Access to Temper Sources With clay sources, landownership rested with the elites, and a variety of behaviors controlled access to the mines and monitored procurement. With temper sources, however, land tenure was both public and private. Anyone could mine temper on public land with no control, although miners and potters recognized the rights of specialist miners to dig raw materials exclusively in their own mines. For private land, however, the same pattern developed as with clay mining. Those who managed the private land at Yo’ Sah Kab, for example, charged potters a right-to-mine fee like that at Yo’ K’at. When mining moved to the Chapab source, however, a slightly different pattern emerged. The owner controlled the exploitation of the temper and placed a right-to-mine surcharge on each sack removed, but he employed his own relatives in mining activities. Conclusion An examination of temper mining and procurement organization over time reveals patterns that parallel those of clay. First, changes in temper mining and procurement remain within the boundaries of the traditional component of temper called sak lu’um (“white earth,” palygorskite). Potters believe that this material imparts unique properties to the performance characteristics of the prepared paste as well as to the drying and firing characteristics of pottery. Until recently, this material was found only at Yo’ Sah Kab, and this location appears to have been the source of both white earth and pottery temper since the Terminal Classic period (Arnold 2005b). Although the religious associations of the temper mines have not been as strong as those of the clay source at Yo’ K’at, the ideology 218

How Has Temper Procurement Changed?

associated with the archaeological site at Yo’ Sah Kab appears to be pre-Conquest and formerly influenced miners to make offerings and avoid the site. By 1984, however, this ideology was no longer sufficient to keep miners away, and mining eventually destroyed all evidence of prehistoric occupation there. Whereas tradition and folklore gave Yo’ Sah Kab a unique sense of place as the location for mining temper, once sak lu’um was discovered near Chapab, potters were willing to use temper from that source as well. Second, like clay mining, the extent of temper mining increased over time. Between 1965 and 1970, temper mining alternated between the public and private locations within Yo’ Sah Kab. By 1984, temper mining had expanded over a much larger area. By 1994, mining occurred in two different locations: at Yo’ Sah Kab and on private property located five kilometers farther away near Chapab. Third, the development of specialists in temper mining parallels the development of specialists in clay mining (Figure 6.2). This process began with potters mining and preparing temper. Some potters then became part-time mining specialists selling temper to other potters. Then, full-time mining specialists developed; they were not potters at all but were related to potters. Eventually, the miners became full-time specialists who did not have any relationship whatsoever with potters except to sell temper to them. Like miners of the new sources of clay, these miners did not live in Ticul but rather lived in a neighboring village (Chapab) closer to the new temper mines. Fourth, the temper mining strategy also changed. When mining was occasional and part-time, temper mines were small and shallow. Any potter could use any mine. With the development of full-time specialists, however, mining techniques intensified with long underground passages of mined-out cavities, and miners recognized the exclusive rights of the miner who dug the mine to extract the raw materials from it. Fifth, considerable variation of temper procurement patterns occurred over time, but with an overall reduction in behavioral variability in how temper was prepared. The development of mining specialists between 1970 and 1984 had the effect of reducing the number of temper miners and the amount of behavioral variability in the preparation of temper. With an expansion of the number of sources and the spatial extent of temper exploitation (Table 6.5), any increased chemical and mineralogical homogeneity in the temper probably counterbalanced decreased behavioral variability of temper preparation. How this change was reflected in the actual chemical composition of paste will be discussed in the next chapter. Sixth, theorists in the technological choice school of thought want to separate choices that have a technological basis from those that do not have such a 219

How Has Temper Procurement Changed?

basis. The data from this chapter and from the previous chapter suggest that this distinction is a spurious one. All choices are social in that they result from potters and miners acting upon socially learned categories. Some of the choices of these categories, however, are technologically sound but are reinforced from the sensory feedback that comes from the physical properties of working with the categories that they choose (Arnold 1971). Since negative feedback from inferior raw materials primarily consists of the adverse effects (such as sagging and cracking) that they have on finished pottery, feedback has informed the categorization and selection of appropriate raw materials. As a result, tempering materials are selected because they come from socially constructed criteria (such as the appropriate place for such materials) and contain the appropriate category of materials (such as white earth). All the potter knows is that those materials (in this case, those that contain sak lu’um) come from socially recognized locations and result in good pots, whereas other materials from other locations do not. The technological significance of choices may be lower to the potter than to the archaeologist, but the technological basis of the choices still exists. Social choices cannot easily be separated from any technological basis for those choices. In many cases, the social criteria and the sensory feedback from the technology are mutually reinforcing. Choices are thus seldom exclusively social or technological. Seventh, like clay mining, temper mining provides three parallel, similar, and additional surrogate indices of the increase in production intensity between 1965 and 1997. One index consists of the dramatic increase in the extent of temper mining. A second index consists of the change from potters who mine their own temper to part-time specialists who are potters and, finally, to full-time mining specialists who are not potters (Figure 6.2). These specialists do not live in the same town as the potters and have no relationship with them except to sell temper to them. Even then, however, they may have no direct contact with potters but rather sell their temper to middlemen who transport it to Ticul and sell it to potters. A third surrogate index consists of the increased amount of temper produced between 1984 and 1997. Finally, one of the theories of the relationship of craft specialization to the evolution of socioeconomic complexity posits that increased elite control of resources results in increased uniformity of the fabric of the pottery. In the history of clay and temper procurement, sources may be controlled and may be restricted but these restrictions take on a variety of forms. Most often, control consists merely of assessing a right-to-mine fee for each individual who mines temper in a privately owned source. Potters, however, still have a number of choices for obtaining their clay and temper, even when sources are controlled.

220

Chapter

How Has Composition of the Pottery Fabric Changed?

Seven

T

he previous two chapters documented changes in the procurement of clay and temper. These changes indicate a common evolutionary trajectory toward increasing specialization, greater intensity of production, and more task segmentation. Can this trajectory be identified in the fired pottery? Are patterns of raw material procurement recoverable from the ceramics in the archaeological record? Is the expansion of the sources of raw materials over time reflected in changes in the composition of the pottery? This chapter answers these questions and attempts to assess whether the changes in the patterns of raw material procurement are reflected in the fired pottery. Since clay and temper must be mixed together to make pottery, this chapter first describes the behavioral changes of paste preparation through time. Then, it summarizes the changes in the chemical composition of kiln wasters collected in Ticul by Duane Metzger in 1964 and those of wasters collected in 1988, 1994, and 1997 and compares those changes to the ethnographic 221

How Has Composition of the Pottery Fabric Changed?

patterns described in previous chapters (Arnold 2000; Arnold et al. 1999, 2000). Behavioral Changes in Paste Preparation In order to prepare a paste necessary to fabricate pottery successfully, the potter must mix raw materials together in an appropriate ratio. Is this ratio based on tradition and immutable? Or does it change through time? If so, why does it change? As previous chapters have demonstrated, the sources of raw materials have changed between 1965 and 1997 (Table 7.1), and these changes have created new challenges for potters. As a result, potters have changed their pastepreparation techniques in order to produce a paste that can be used successfully with their fabrication technology. Changes in Preparing Clay

Before mixing clay with temper and water, it must be dried in the sun. Drying causes the clay to break up so that it will combine with the temper and water more thoroughly. If not, tiny balls of clay remain in the paste and cause the vessels to crack during forming, drying, and firing. The shift in clay sources between 1965 and 1997 has affected the amount of time that potters dry their clay. In 1984, potters believed that clay must be dried for one to three days (depending on the season) prior to mixing it with the temper. By 1997, clay was coming from four different sources in Campeche (Table 5.1). Clay from one of these sources (the Tepakán source) required three times longer to dry than the clay from the other source (the Dzitbalché source). The process of wetting the clay before mixing has also changed between 1965 and 1997. Up until the early 1990s, the dried clay was soaked in water and then mixed with the temper. In 1997, however, the clay from Campeche had so many rocks in it (as much as three kilograms per large jar of clay) that at least two production units had begun using levigation to prepare the clay. In this process, the clay was completely liquefied by adding more water than was necessary to make it plastic (Curtis 1962:492; Matson 1973:124–125; Rice 1987:118; Rye 1981:37; Shepard 1956:52). As the clay particles became suspended in the water, the rocks dropped to the bottom of the receptacle. The clay was then poured out and the rocks discarded. Changes in Clay Quality

When the sources and procurement organization of the clay changed, the quality of clay also changed. In 1965, the quality of clay was uniformly high 222

Multiple mines in single location Multiple mines in single location Multiple mines in two locations Multiple mines in two locations

Multiple mines Expansion 1 in two locations

Multiple mines in Expansion 2 multiple locations

Multiple mines in Expansion 2 multiple locations

2 1988

3 1994

4 1997

Temper

Multiple mines Expansion 1 in single location

Number of temper Extent sources

1 1984

Changes in mining area of each source from previous observation

Clay

Multiple mines in single location

Number of clay Extent sources

Single mine Little or no change 1 1965–1970 1

Year(s) of observation



Expansion

Expansion

Expansion

Expansion

Internal variation

Changes in mining area of each source from previous observation

Table 7.1. Summary of changing clay and temper sources used from 1965 to 1997. (From Dean E. Arnold, “Does the Standardization of Ceramic Pastes Really Mean Specialization?” Journal of Archaeological Method and Theory, 7 [2000]:347 [Table 2]. Copyright by Plenum Publishing Corporation. Reprinted with kind permission of Springer Science and Business Media)

How Has Composition of the Pottery Fabric Changed?

because clay choice was based on social criteria, such as tradition and a sense of the appropriate place from which it should come. This social choice was reinforced by the technological superiority of the clay, the religious association of the source location (Yo’ K’at) with a Christian saint (San Pedro), and the ease of transportation to Ticul (Arnold 1971). In addition, potters used two kinds of practical criteria for choosing high-quality clay when it was not available from Yo’ K’at: potters’ clay had a salty (ch’ooch’ ) taste and it remained solid and did not open up or crack as easily when it was dried in the sun (see Arnold 1971: 28–32). With changes in clay sources and clay procurement practices, potters’ assessments of clay quality changed completely. By 1984, the criteria used to distinguish clays in the 1960s were no longer used. Clay mining was in the hands of specialists who were not potters, and the demand for clay was so great that inferior clays could not be rejected but instead were modified with more temper or with pure palygorskite (sak lu’um), the most significant cultural component of temper. By 1997, different assessments of clay quality had emerged because the sources of clay had changed. At that time, all clay was coming from four sources in Campeche (Table 5.1), but most of the clay came from two of them. Each of these two sources produced clays of different qualities (Arnold et al. 1999). Potters considered the Tepakán clay better quality than the Dzitbalché clay because it was harder in texture, stronger, and more plastic. It also had fewer rocks and was less friable than the Dzitbalché clay. Just as the quality of the Tepakán clay was better, so were the products made from it. Potters said that the pottery made with the Tepakán clay was stronger than that made with the Dzitbalché clay, and they preferred to use the Tepakán clay for large vessels. If potters used the Dzitbalché clay to make large vessels, the pottery cracked during drying. So in order to compensate for the poorer-quality Dzitbalché clay and produce a better finished product, potters mixed the two clays in equal parts before adding the temper. Mixing the two clays, however, was a temporary measure, because by 2002, the owner of the Dzitbalché source had abandoned selling clay because potters were buying his clay on credit and then not paying for it. Furthermore, he realized that the Tepakán clay was superior. Changes in Mixing the Paste

After the clay has been dried and soaked with water, it is placed in a pile of temper. If potters want to make cooking pottery, they use ground crystalline calcite (hi’ ) and mix it with soaked clay (see Arnold 1971; Thompson 1958:72–73). If, on the other hand, they choose to make pottery that is not used for cooking, 224

How Has Composition of the Pottery Fabric Changed?

they mix sah kab temper with the clay. The amount of sah kab temper used, however, may vary depending on its quality, because if the quality is inferior, potters add pure palygorskite to the paste to compensate for poor-quality temper. The proper consistency of the paste occurs within a narrow range of proportions of temper and clay. If too much temper is added, the paste will be too stiff to work. If, on the other hand, the amount of temper is insufficient, the paste is too plastic and pots will not hold their shape. This range of workability is monitored by visual (and probably tactile) feedback that comes from potters’ perception of the performance characteristics of the paste. If during wedging, cracks (bilil k’at) appear in the paste near the basal working surface, then it must be wedged again. A second test consists of rolling the paste between the palms to form a vertical sausage-shaped coil. If part of the coil breaks off and falls during this process, it has not been mixed sufficiently and must be wedged again. Change in Paste Recipes

In order to maintain satisfactory workability and other performance characteristics of the paste for forming a vessel successfully, potters may change their paste recipes over time. Changes in raw materials, in particular, produce different performance characteristics in the paste (Arnold 2000). Between 1965 and 1984, potters reported a consistent paste recipe of two parts temper to one part clay. In 1988, however, clay miners had dug new mine shafts at Hacienda Yo’ K’at and had encountered a layer of white clay that was much nearer the surface (only three meters deep) than that mined between 1965 and 1984 (which was six meters deep). This new clay contained a naturally occurring marl (sah kab) and required 75 percent less temper than the clay obtained at a greater depth. By 1992, the Yo’ K’at source was abandoned and all clay was coming from new clay sources in the State of Campeche. Consequently, by 1997, potters had altered their paste preparation recipe and used a temper/clay ratio of one to one. When clay was levigated, however, the paste recipe required experimentation to see how much temper was necessary. The recipe for cooking pottery has also changed over time. In 1951, Thompson (1958:72) reported that potters used 3.5 parts temper to one part clay. By 1970, however, one potter said that equal parts temper and clay were mixed to make cooking pottery, and in 1988 the only potter still making the occasional cooking pot was mixing 1.25 parts temper with one part clay to make the paste. By 1997, one potter was using another method of paste preparation (see Chapter 8) that he learned at the local ceramics factory. Since this technique involved pouring a liquefied clay and temper mixture into a plaster mold, he had 225

How Has Composition of the Pottery Fabric Changed?

Figure 7.1. Principal components plot of the INAA data of kiln wasters from Ticul, Tepakán, and Akil. Ellipses represent a 90 percent level for membership in each group (plot from Arnold et al. 1999:74). The Ticul sherds collected in 1964 and in 1988 were distinct from those collected in Tepakán and Akil in 1994. By 1994, however, Ticul potters were obtaining their clay from the Tepakán sources. The 1994 sample from Ticul overlapped with both the Ticul sherds from 1964 and 1988 and the Tepakán sherds from 1994 because potters still had wasters made with the Yo’ K’at clay after the Yo’ K’at mines had been abandoned. (From Dean E. Arnold, Hector Neff, Ronald L. Bishop, and Michael D. Glascock, “Testing Interpretative Assumptions of Neutron Activation Analysis: Contemporary Pottery in Yucatán, 1964–1994,” in Material Meanings: Critical Approaches to the Interpretation of Material Culture, edited by Elizabeth S. Chilton, p. 74 [University of Utah Press, Salt Lake City]. Used with permission)

to develop his own recipe and used a combination of clay, sah kab temper, red earth (k’an kab), white earth, and lime. Changes in Paste Composition over Time Can changes in clay and temper procurement patterns, resource locations, and paste-preparation recipes be identified from the compositional analysis of the fired pottery? In order to answer this question, Instrumental Neutron Activation Analysis (INAA) was used to analyze pottery collected in 1964, 1988, 1994, and 1997 (Arnold et al. 1999, 2000). The results of these analyses revealed that the composition of the 1964 sherds from Ticul was almost identical to the composi-

226

How Has Composition of the Pottery Fabric Changed?

tion of those collected in 1988 (Figure 7.1). This combined sample was different from the pottery from other communities collected in 1994. Roughly half of the 1994 sherds were very similar in composition to the Ticul sherds collected in 1964 and 1988, whereas the composition of the other half of the 1994 pottery was consistent with the composition of the pottery from Tepakán (Figure 7.1). The composition of the 1994 pottery thus revealed a shift in the composition toward that of the Tepakán pottery. By 1997, the composition of the Ticul pottery was very similar (if not identical) to that of the Tepakán pottery. Why was there a shift in composition toward Tepakán pottery over time when the INAA data revealed that the compositions of the Ticul and Tepakán pottery were originally so distinct? These changing compositional patterns can be explained by the change in the clay sources used between 1988 and 1994. In the late 1980s, the clay mined at Hacienda Yo’ K’at had become increasingly scarce and potters began using clay from elsewhere. By early 1992, clay mining at Hacienda Yo’ K’at had ended, and by 1994, all potters were using clay from sources in the State of Campeche (Table 7.1). Some of the sherds collected in 1994, however, were made with the Yo’ K’at clay. Potters recognized this source difference by the color of the fired paste. Consequently, sherds of both colors were selected in roughly equal numbers so that the analysis included sherds made with each clay. Two other factors appeared to influence the divergent chemical pattern of Ticul pottery collected in 1964 and 1997 but were not recoverable from the INAA data. First, although some potters continued using temper from Yo’ Sah Kab in 1994, temper sources had expanded to a new location five kilometers farther away. Second, paste recipes were also adjusted to accommodate the characteristics of the clay from the new sources. This adjustment first occurred about 1988 when the Yo’ K’at clay contained an increased amount of naturally occurring non-plastics. Even so, the clear chemical similarity of some of the 1994 Ticul sherds to that of the Tepakán pottery suggests that the change in clay source overwhelmed all of these other sources of variation. Because Ticul pottery made after 1992 was made with Campeche clay, the composition of most of the pottery from Ticul during 1997 was more similar to that of the Tepakán pottery than it was to that of the Ticul pottery collected in 1964 and 1988 (Arnold et al. 1999, 2000). Conclusion Are patterns of raw material procurement and paste preparation recoverable in the archaeological record? Specifically, are the procurement patterns and their 227

How Has Composition of the Pottery Fabric Changed?

changes detailed in the previous two chapters encoded in the chemical composition of the ceramics? The answer to both questions is a highly qualified yes. The Ticul data demonstrate that clay procurement locations may shift over time as a result of exhaustion of clay deposits and micro-level sociopolitical factors. The compositional effect of such shifts can be profound. Although large-scale changes in clay sources can be detected, more subtle changes in paste recipes and in temper source were impossible to detect in the data using INAA. Changes in paste recipes and temper sources do not appear to affect the composition of pottery as much as changes in clay sources, even though changes in paste-preparation techniques are the potters’ adaptation to changing clay sources. Therefore, although the major compositional shifts in clay sources can be detected in the fired pottery, social and political reasons for those shifts cannot be inferred from these data alone. The reasons can only be inferred from a social theory that links the compositional data to social patterns. One such social theory links the ceramic resources of a population of potters with the spatial and energy limits of resource distances (Arnold 1985:32–60, 2000, 2005a). Other studies have now demonstrated the validity and usefulness of this model (e.g., Arthur 2006:52; Day 2004). In some cases, these distances are highly contextualized and thus are increased when clay is procured on the way to agricultural fields and where additional energy sources extend (e.g., boat travel) or supplement (e.g., pack animals, vehicles) human carriers. Other models, such as Rice’s model (1981) that the increasing intensity of production results in a narrowing of raw material sources, have already been challenged. What is needed now is a general theory that relates paste to social patterns of raw material procurement and preparation.

228

Chapter

How Has the Forming Technology Changed?

eight

F

orming a vessel is the most obvious step in changing a formless mass of clay into a useful object. Because clay is plastic, it would seem that the potter could form it into virtually any shape, using any number of techniques; indeed, archaeologists commonly assume that culture, ideas, and potters’ choices are freely expressed in the completed pot. This view of pottery has deep roots in American anthropology and archaeology (Arnold 1985:5–8) and is probably best expressed in Ruth Bunzel’s classic work, The Pueblo Potter: Owing to the plastic nature of the material, almost any conceivable form is possible. Setting aside modeling in the round as outside the scope of the present investigation and confining ourselves to practicable utensils, we find even in these an enormous range of possibilities. . . . Furthermore, there are no definite limits to the size of the vessels that can be made of clay. What limits there are, are set by the skill of the potter in mixing her materials, building the vessel and firing it rather than by the inherent nature of the material. (Bunzel 1929:2)

229

How Has the Forming Technology Changed?

This view of pottery has been partially reincarnated as the technological choice approach to ceramics. As Sander van der Leeuw put it: Virtually all known prehistoric techniques of pottery-making, and most ethnographically observed ones, have a rather wide tolerance of the clays and other raw materials needed, so that almost any of these techniques could probably be implemented almost anywhere, if need be[,] by introducing a few minor modifications. In pre-industrial societies, one must assume a considerable freedom of action for the potter. (van der Leeuw 1993:239)

Van der Leeuw is reacting to the notion (by anglophone archaeologists, he says) that the potter operates within a narrow range of physical, chemical, and economic constraints that dictate potters’ choices in vessel shapes (van der Leeuw 1993:238). By way of contrast, van der Leeuw’s counter notion seems to give the potter total choice in the way that a pot is made without the constraints of raw materials and technique, and appears to make the potter’s social choices the sole determinants of shape. Because technology is socially embedded, the vessel shapes produced thus become totally explainable by cultural preferences and social mediating factors that are purely, if not principally, historically relative (Lemonnier 1993; Loney 2000). Adherents of this technological choice approach to pottery want to separate technological from non-technological, or social, choices that consist of those options that are selected for reasons other than those based on the physical constraints of the materials and the forming technology (Lemonnier 1993; Loney 2000; van der Leeuw 1993). In reality, however, all choices are social choices because potters learn these options by social means. Forming a pot, however, is not predetermined and entirely technically based nor are choices exclusively based on non-technical (social) criteria. Potters’ knowledge of choices of different forming techniques and vessel shapes is acquired socially and provides the general categories for behavior. Further, potters learn about raw materials and their constraints socially, and these choices are reinforced by the feedback of visual and tactile experience with the raw materials that might require experimentation and modification (Arnold 1978b:347, 1993:80). Rice (1987:61) discusses these constraints in terms of a working range and the plastic limits. Finally, religious factors also affect choice of raw materials in Quinua, Peru (Arnold 1993:66–68), and, as already noted, at Yo’ K’at. Up until about 1969, clay procurement at Yo’ K’at was linked to the patron saint of the hacienda in the form of potters sponsoring nine nights of ritual prayers in his honor (a novena). Feedback, however, is not determinative but rather simply the flow of information from the raw materials, context, tradition, and the market through the potters’ senses that influences, but does not determine, their choices (Arnold 230

How Has the Forming Technology Changed?

1985). As they interact with the raw materials, paste, and emerging vessel, potters monitor and evaluate this information to ensure their success in making a pot (Arnold 2000). Sensory information about raw materials from feedback is critical for potters and it may modify their behavior in order to achieve the end product—successful fabrication of the vessel (Arnold 1985). Feedback simply recognizes that the relationship between materials and humans is not unidirectional. Feedback is the mirror image of materialization. Consequently, social choices (and culture) are not simply imprinted on the raw clay but are rather the product of the bodily engagement of the potter and the clay based on feedback that involves the interaction of the potter, his/her training and tradition, and the characteristics of the raw material, the environment, and the emerging pottery product. As this chapter will demonstrate, Ticul potters do have a choice in the forming technique that they use, but they may or may not have a wide latitude in what they make and how they make it. Nevertheless, the mineralogy of the clays used, the forming technique, and the vessel shape produced are all interdependent, and the potters’ choices in making vessels are not totally free nor are they predetermined. Rather, they are influenced by tradition, by the performance characteristics of the clay and other raw materials (Arnold 1971), by the feedback resulting from the interaction of these factors when the vessel is being formed, and by social information. The complex interaction between potters’ social and technical choices concerning clay, technique, and vessel shapes is illustrated by the advantages and disadvantages of the five different forming techniques available for Ticul potters. One technique is a traditional one, but potters have also experimented with four other techniques since the late 1930s and adopted two of them. Each of the nontraditional techniques was adopted (or not adopted) for a variety of reasons. Consequently, choosing one technique over another to make a particular shape is not just socially and culturally determined but is affected by the advantages and disadvantages of the techniques. One technique is chosen over another depending on the shape desired, the limitations of the technique, the limitations of the local clay, and the demand of the market. Besides reevaluating the nature of the potters’ choices, another implication of the fabrication techniques consists of their relationship to emerging socioeconomic complexity. Several scholars have argued that the development of the use of molds or the wheel reflects evolutionary change in technology that results in greater efficiency, economies of scale, and routinization of tasks, which leads to more uniform vessel shapes (Costin 1991; Rathje 1975; Rice 1981). Greater homogeneity of vessels thus becomes a surrogate measure of the emergence of 231

How Has the Forming Technology Changed?

higher levels of specialization and greater socioeconomic complexity. As this chapter will demonstrate, these links are complicated, and the relationships between standardization and forming technique are complex (Arnold and Nieves 1992). Unpacking these relationships is not just a matter of linking social and cultural change with changing forming technology. Rather, understanding the technologies themselves provides an analogical basis for applying knowledge of these techniques in the development (or lack thereof ) of socioeconomic complexity in antiquity. This chapter thus reveals that the potter’s use of a particular technique (such as molding) is not necessarily dictated by the desire to produce uniform vessels but rather reveals how technique is related to a complex set of factors such as clay, shape, skill, demand, and, in general, the organization of production. Why Were New Fabrication Techniques Adopted? Forming technology in Ticul has undergone great changes between 1965 and 1997. One technique is more traditional whereas the others consist of relatively new innovations that have been introduced since the late 1930s when a government workshop tried to teach the potters new methods of fabrication. Even the traditional forming technology, however, has changed since Raymond Thompson (1958) visited Ticul in 1951. An examination of all of these forming techniques reveals that they are not equal in utility and are chosen (or not chosen) for various reasons. It is a mistake to believe, however, that these new techniques simply replaced the traditional technique. Rather, potters have adopted them because the traditional technique was inadequate to produce some vessels. The new techniques thus supplement, rather than replace, the traditional technique. If the traditional technique had been adequate to form every shape with only a few minor modifications, as van der Leeuw (1993:238) suggested, potters would not have needed to experiment with new techniques and would not have adopted them, even temporarily. The explanations of the acceptance (and rejection) of these new techniques thus provide insight into why fabrication technique changes, whether it reflects social change, and what effect, if any, fabrication technology has on efficiency. In many respects, the traditional technique was adequate, if not superior, to the new techniques. But potters’ acceptance of some techniques (and rejection of others) reveals complex interactions of the new techniques with social, economic, and technological factors. Indeed, the advantages and disadvantages of these techniques demonstrate why they coexist and why one technique is preferred to another under different circumstances.

232

Figure 8.1. A Ticul potter using the traditional turntable (k’abal) in 1970. The potter is

seated on a low stool on the floor and may turn the device with his feet when he uses both hands to form the vessel. The large base of the k’abal keeps the device stationary while the upper platform rotates.

How Has the Forming Technology Changed?

The Baseline for Change: Modified Coiling

The baseline for comparing forming techniques is the traditional technique of modifying a slab coil on a turntable (k’abal). This device consists of a removable platform that rotates around a nail embedded in a thick piece of hardwood (Figure 8.1). The vessel is formed on this platform, and in order to facilitate movement, a circular metal disk is placed on top of the wood base. In order to reduce friction, oil is placed on the metal disk. The platform is removable so that when a vessel (or a portion of it) needs to dry, the potter can simply remove the platform with the vessel and replace it with another without risking damaging the vessel (or a portion of it) by lifting it, cutting it from the turntable, and centering it again on the turntable after partial drying in order to proceed with a subsequent stage (Figure 8.2). By having multiple platforms, a potter takes a fundamental step of mass-producing pottery by repeatedly forming the same segmented portion of a vessel before moving on to the next segment. The turntable can be viewed as a variant of a rotating disk called the molde that is widespread in Mexico and South America. According to Foster’s map of the distribution of pottery-fabrication devices in Mexico, the molde occurs in Oaxaca (Foster 1948) and Chiapas. It is also reported from Atzompa in Oaxaca (Henry 1992) and is found in Quinua, Peru (Arnold 1972b, 1993). Although the Atzompa device is a slab of wood that revolves on a fired pot and the potter works in a kneeling position, the use of large coils to make the vessel is virtually identical to the technique used in Ticul (Henry 1992:66, 76). One potter in Atzompa uses a device that turns around a fixed axis (Henry 1992:84) that appears to be identical in principle to the Ticul k’abal. Understanding the details of the modified coiling technique is crucial for understanding the constraints that raw materials place on the potters’ fabrication techniques and on changes in those techniques. The potter begins by making a pancake of clay and placing it on the turntable. Then, he or she takes one handful of paste, rolls it between the hands to form a sausage shape (approximately 5 cm by 25 cm), and then flattens it with the hands. The flattened coil is pushed onto the pancake against the palm of one hand placed inside of the vessel. Often this motion moves the turntable, but potters may also move it with their toes or feet in order to free both hands to form the vessel. The newly attached coil is then drawn up and thinned with a gourd scraper (Thompson 1958:75, 81–82), using the palm of one hand inside the vessel to support the vessel wall. Additional coils are prepared, attached, and drawn up in a similar way. When the vessel is completed, the potter forms the rim and smoothes the vessel with a piece of leather (Thompson 1958:86), using the foot or the other hand to turn the turntable. This behavior can produce a material signature that looks as if the vessel was wheel-made. 234

Figure 8.2. The base of a traditional turntable (background) and the rotating plat-

forms (foreground) stored on a pile of clay in a potter’s house in 1970. The dark stain on the base and on the turntables is from the use of oil to reduce friction and facilitate movement.

How Has the Forming Technology Changed?

Figure 8.3. Potter making large tinajas in Ticul’s largest production unit in 1984. Because of the highly plastic Ticul clay, they must be made in stages, and each stage must be measured. Traditionally, such measurements were hand measurements.

The technique of drawing up and thinning a slab coil is made possible by the highly plastic mineral montmorillonite in the clay (Rice 1987:49). This plasticity, however, places constraints on the vessel’s shape and size because only the smallest vessels can be made in one continuous sequence without the vessel sagging and collapsing. As a consequence, potters fabricate most vessels in more than one stage; each stage must dry sufficiently in order to support the weight of the next stage (Figure 8.3). Using modified (slab) coiling and building vessels in stages is thus potters’ adjustment to both the benefits and the constraints of a highly plastic paste that contains montmorillonite. As with many other traditional forming techniques, modified coiling on the turntable requires a set of specific motor habit patterns. Learning these patterns involves developing and strengthening the appropriate muscles that are used repeatedly in the fabrication process. An injury that damages or immobilizes these muscles seriously impairs a potter’s ability to make pottery. In March 1970, for example, two potters were injured in a truck accident while transporting their pottery to a fiesta. One potter broke an arm and the other bruised and dislocated his shoulder. At first, their injuries prevented them from making any pottery at all. Then, the inactivity nec236

How Has the Forming Technology Changed?

essary for healing caused the muscles to atrophy. As a result, these potters could not perform all of the activities necessary to make pots and could not make a living as independent household potters. In order to feed their families, they worked at the pottery workshop at the Hotel Principe in Uxmal, where they performed specialized tasks (such as painting) that did not require them to use the atrophied muscles. Except for a change in the turntable (see below), modified coiling on a turntable has been used continually for fabricating pottery since at least the late nineteenth century, when Henry Mercer (1897a, 1897b) first documented the technique in Yucatán. Of all fabricating techniques available, this technique is the most versatile, and potters have used it to produce all shapes that have a circular horizontal cross section. During the late 1960s, potters used it to produce traditional shapes, such as the water-storage pot (tinaja), the water-carrying jar (cántaro), and a vessel for soaking maize in lime water (apaste). Since then, potters have also used this technique to produce plant pots and most other vessels with a circular cross section. Changes in Forming Technology Although using modified coiling on the turntable is the traditional method of forming pottery, potters have also used four other techniques. Two of these techniques, the use of a potter’s wheel and molding, were introduced in the 1930s and 1940s by a government-sponsored program for helping potters improve their craft. A third technique is a modification of the traditional turntable and was introduced in the late 1970s. Finally, a fourth technique consists of slip casting, which was used in a local ceramics factory and adopted by one potter who worked there. Conservatism and Resistance: Rejection of the Potter’s Wheel

Many scholars have assumed that traditional potters were resistant to change (Foster 1965). In reality, however, as Rice (1984) points out, different aspects of pottery production change at different rates. But why are some aspects of pottery making more conservative than others? The failure of potters to accept the kick wheel provides some insight about technological conservatism among all craftsmen, not just potters. The potter’s wheel has its origin in the Old World and diffused across the Mediterranean during the Bronze Age (Rice 1987:134). The wheel uses the centrifugal force of a rotating disk to aid in forming a vessel such that a lump of clay can be placed on it and the momentum allows the potter to form a vessel simply by 237

How Has the Forming Technology Changed?

selectively squeezing and shaping the rotating ball (Rice 1987:134; Eric Blinman, personal communication). In the New World, the wheel was unknown before the Conquest and was introduced by the Spanish, although the production of pottery by indigenous peoples continued to involve various kinds of hand-forming technologies that usually were derived from pre-Columbian antecedents (e.g., Arnold 1978a, 1978b, 1993; Reina and Hill 1978). In the late 1930s, the government wanted to refine (  purificar) Ticul pottery in order to improve the quality and efficiency of production (Rendón 1947). So it established a workshop near the municipal market, hired a potter from Oaxaca, and installed five wheels with the intent of introducing them into the community. This workshop appeared to be the result of a development strategy that Lázaro Cárdenas implemented when he became president of Mexico (1934– 1940). Cárdenas was responsible for sending a mission of teachers in plastic arts, social work, home economics, and other specialties to Tzintzuntzan when he was governor of Michoacán (Foster 1967:26–29). Based on the success of that strategy in Michoacán, where the government had also tried to introduce the wheel, the introduction of the wheel was likely the result of an implementation of a similar plan on a more national scale. Another reason for instituting the government program in Ticul might have been related to earlier observations that the traditional turntable (the k’abal) was believed to be a primitive potter’s wheel (Mercer 1897a, 1897b; Rendón 1947; Thompson 1958:11–13, 76–81, 137–144). The proponents of the government workshop thus may have seen the introduction of the new wheel simply as an improvement on an already existing wheel. Although the workshop existed for more than ten years, the Oaxaca potter only stayed about four to five years. Several local individuals worked with him, but only one (Cesario Mex) purchased a wheel and used it to make pottery. By the time that Raymond Thompson (1958:20) visited the workshop in 1951, the Oaxaca potter had gone. Although several wheels remained in the workshop, Thompson reported that the potter there spent very little time using them. Since almost all of the potters failed to adopt the wheel, this discussion will begin with its disadvantages as the feedback that influenced the potters’ choosing to reject it. Disadvantages of Using the Wheel. The potters themselves provided the first disadvantage of using the wheel. The wheel, they said, was too expensive (500 pesos) and they did not have the money to buy it. This explanation is verified by the economic realities of the 1940s, when potters worked as laborers on the expanding highway system. Potters reportedly earned seventy-five centavos a day 238

How Has the Forming Technology Changed?

and only rarely did their daily income increase to one peso. Highway workers, however, were paid 1.5 pesos. As a result, at least six potters went to work on the highway. Assuming a six-day work week and four weeks off for religious holidays, the cost of a wheel required 231 percent of a potter’s yearly income (288 days × 0.75 peso) and 116 percent of a highway worker’s yearly wages. Because both potters and highway workers required money for food and other necessities, potters would find it difficult, if not impossible, to accumulate enough cash to purchase a wheel. Even some thirty years later, when daily wages had risen dramatically, potters still thought that the 500 pesos (US $40.00 in 1965) required to buy a wheel was too expensive. The excessive cost of the wheel relative to wages in the 1940s was corroborated in 1988 by a conversation with the president of the township where the workshop was operating. No money was available at that time, he said, and the potters did not want to participate in the workshop because they wanted to be paid every day. Given informant accounts that pottery making was a precarious way to make a living at the time and that potters were retreating from the craft into swidden agriculture and wage labor, it is understandable why potters wanted to be paid for their participation in the workshop; learning a new forming technique without wages to support a family during the transition was economically risky, if not disastrous. Potters thus did not have the time or the capital to risk adopting an untried innovation without regular compensation. A second disadvantage of using the wheel involved the kind of paste required. Clays used for wheel forming have different requirements than those used for hand forming. Potters said that the Ticul paste was too thick and too coarse and abraded the potters’ hands because of the speed of the wheel—a problem that was also noted in Temascalcingo in central Mexico (Papousek 1974:1024). As a consequence, the Oaxaca potter used ground rocks for temper. The machine used for crushing the rocks, however, came from Oaxaca and had to be purchased, and Ticul potters could not understand why they should use such a machine to produce temper when they had their own temper, which required no outlay of capital to produce. Even with the new temper, however, the wheel-made vessels did not turn out well because potters said the Ticul clay was different from that in Oaxaca. Eric Blinman (personal communication) found a similar phenomenon among the Pueblo potters of New Mexico. Some were using the wheel, and when they were given a clay used by the ancient Anazazi, they could not use it for wheel forming because it was too coarse for them. A third disadvantage of using the wheel comes from the effect of its use on the feet. Potters do not wear shoes; they wear Yucatecan sandals. Furthermore, they often make pottery barefoot so that they can move the turntable with their 239

How Has the Forming Technology Changed?

Table 8.1. Muscle groups required for using the traditional turntable and the wheel. These patterns were identified by Ewan Russell, an exercise physiologist in the Wheaton College Department of Applied Health Science. (Adapted from Dean E. Arnold, Jill Huttar Wilson, and Alvaro L. Nieves, “Why Was the Potter’s Wheel Rejected? Social Choice and Technological Change in Ticul, Yucatan, Mexico,” in Pottery Economics in Mesoamerica, ed. Christopher A. Pool and George J. Bey III, p. 70 [University of Arizona Press, Tucson, 2008]. Used by permission)

Traditional turntable

Wheel

Muscle groups required Individual muscles in group required Other muscles probably required

Quadriceps Rectus femoris Vastus medialis Vastus intermedius Vastus lateralis Ankle plantar flexion, using the gastrocnemius and the soleus muscles

Hamstrings Biceps femoris Semimembranosus Semitendinosus

feet. Using their bare feet (or the sandals) to propel the wheel, however, abraded and injured the feet. The final disadvantage of using the wheel is the relative incompatibility of the muscles, muscle syntax (motor habits), and associated muscle strength required by use of the wheel compared to that of the traditional turntable. The wheel requires a totally different set of motor habits than the traditional technique of modified coiling. With the traditional technique, the potter sits on the floor or on a low stool with one or both legs drawn up toward the body or with one or both legs outstretched. The forming is done on a turntable with a mean height of 9.7 centimeters (N = 19) from the floor. The muscles utilized in this position are the hamstrings. But most of the muscle strength and coordination required to make a pot involve the arms and hands; the feet are used only occasionally for rotating the turntable (Table 8.1). Using the wheel, on the other hand, requires that the potter sit on a bench and work on a surface that is eighty-seven centimeters above the floor. This position is crucial for the operation of the kick wheel because it leaves the legs extended with enough free space for a range of motion at the knee to propel the flywheel with the upper leg, lower leg, and foot. This pattern requires more strength and range of motion at the knee joint than does modified coiling (Figure 8.4) and requires consistent use of the quadriceps in the legs (Table 8.1). The motion of the flywheel is transferred by a shaft to a spinning platform on which the pots are formed. Shaping of the pottery occurs by using the hands to manipulate the revolving clay. The muscular patterns required by the wheel were too different for traditional potters to learn without a lengthy apprenticeship. Even if someone learns 240

Figure 8.4. Daughter of a Ticul potter using the wheel in 1984. Even though she used

the wheel, she still used modified coiling (shown here) rather than throwing to make pots.

How Has the Forming Technology Changed?

the new motor habits, the muscles required for those habits must be strengthened sufficiently to be able to use the wheel intensively. In order for the wheel to be a productive forming technique beyond occasional use, concentrated effort and persistent use are required. Potters typically work long days with few breaks. In Egypt, for example, the potter “only rarely ceases his work on the wheel” and then only to rest (Nicholson and Patterson 1992:30). Yet, the woman who was learning to use the wheel in Ticul became tired after only an hour of work, and this limitation indicates that the motor habits and muscle strength required for sustained wheel production were incompatible with traditional patterns. Consequently, the wheel could never become a consistent long-term fabrication technique without long periods of training and strengthening of the appropriate muscles. This explanation is borne out by the fact that the wheel was never adopted even though a few potters used it temporarily. Advantages of Using the Wheel. If the wheel has such obvious disadvantages, why was it used at all? Are there any advantages of using the wheel? The answers to these questions are embedded in the history of its brief use by one household in the 1980s. The unsuccessful attempt to introduce the wheel in the 1940s left behind an abandoned wheel in the municipal building. The mayor wanted to dispose of it and gave it to the president of the sindicato of Ticul potters. After it had been stored at his house for several years, one of his daughters began to experiment with it. She had already learned how to mold pottery, but she had never learned how to make pottery with the traditional turntable. She learned how to use the wheel, but only with the traditional technique of modified coiling, and eventually she taught her father the technique. By 1984, her father and her sisters were also making wheel-made pottery with modified coiling, but they were the only Ticul potters doing so. The production of wheel-made pottery by this household, however, was short-lived. The daughter who used the wheel went to the local normal school to become a teacher and eventually left home for her first teaching job. By 1988, she made pottery only when she returned for vacation. By 1994, however, she had married, was living in Cancún, and had abandoned pottery making entirely. By this time, her father had also left the craft. Although he still retained possession of the wheel, he did not use it. When he returned to making pottery in late 1994, he did not use the wheel. When I returned to Ticul again in 1997, no one was using the wheel. Interestingly, although this household temporarily adopted the wheel, their vessels made on the wheel were fabricated using modified coiling, just like vessels made on the traditional turntable; they were never 242

How Has the Forming Technology Changed?

thrown from a single formless lump of clay (“off the hump”). Thus, this family used this wheel for making pots in the traditional way, and the daughter who was the first to use it had not learned the turntable. She still used the traditional forming technique on the wheel, as a turntable, but powering it with a different set of muscles. Consequently, the centrifugal force of the wheel was not exploited to its fullest extent. Because the Ticul clay is too plastic to produce a large vessel in one continuous sequence without sagging or collapsing, large vessels still need to be fabricated in stages separated by periods of drying, just as they are on the turntable. This household thus still used slab coiling to make the vessels but used the wheel as a faster turntable. A second reason that the wheel was adopted temporarily was that once the family learned the motor habits to operate it and adapted its use to modified coiling, they discovered that wheel-made vessels required less forming time than other methods. Nevertheless, they could not spend much more than an hour a day working on the wheel because it was too tiring. In order to solve this problem, wheel production was reorganized by segmenting fabrication into a series of smaller tasks that combined the use of the wheel with modified coiling on the turntable. This change reduced the amount of time that one person worked on the wheel, but more important, it maximized production within the constraints of muscular stress and tiredness. In making the food bowl, for example, one person prepared a coil of clay, formed it into a crude bowl on the turntable, and then transferred it to the wheel, where another person finished it (Table 8.2, note c). After the bowl was cut from the wheel, the completed vessel was exchanged for another crudely formed bowl. In this way, two young women could complete about fifty large food bowls and ten small bowls in sixty to seventy-five minutes, and this amount of time was close to the stamina limit of the woman working on the wheel. Similarly, in making plates, one female adolescent made a disk of clay, passed it to another who added a ring of clay on the turntable, and then passed it to a third who finished it on the wheel. The completed vessel was then passed back to the first adolescent, who placed it in the sun to dry. Using this kind of task segmentation, three individuals made fiftythree plates in about an hour. The mean fabrication time per plate on the wheel from this production event was ninety-two seconds (N = 18 [approximately every third vessel was timed from the point a vessel was placed on the wheel until it was removed]; Table 8.3). The woman who formed these vessels on the wheel said that she can make fifty plates in forty minutes (forty-eight seconds each) with the help of her sister in preparing the crude vessels. If this same number of plates were made on the traditional turntable, she said, they would require 150 minutes to complete (180 seconds each). The wheel is thus 3.75 times faster than 243

How Has the Forming Technology Changed?

Table 8.2. Fabrication times for food bowls (cajetes) using the wheel, mold, and turntable. (Adapted from Dean E. Arnold, Jill Huttar Wilson, and Alvaro L. Nieves, “Why Was the Potter’s Wheel Rejected? Social Choice and Technological Change in Ticul, Yucatan, Mexico,” in Pottery Economics in Mesoamerica, ed. Christopher A. Pool and George J. Bey III, p. 66 [University of Arizona Press, Tucson, 2008]. Used by permission) Forming technique N Mean (sec) Range

Makera

wheel

15

47

wheel

2

wheel

Heightb (cm)

Mouthb (cm)

35–60

c

Da 1

6

13

122

111–134

Da 4

4

9

19

59

48–78

Da 1c

4

9

mold

3

211d

147–259

Fa

6.5

13.5

turntable

5

231

181–278

Si

—

14.5

Notes: a. The designation “Da” refers to “daughter” and the number following indicates the birth order of the daughters: “Da 1” is the oldest, and “Da 4” is the fourth oldest. b. Mouth and height diameters refer to modal measurements of the size of fired vessels except for the turntable vessels, which are unfired. Shrinkage of the unfired vessels from drying and firing should be about one centimeter, placing the widths of these vessels within the 13–13.5 cm range of mouth diameters for wheel- and mold-made vessels shown in this table. Although the height of the sister’s turntable vessels was not measured, cajetes of the same mouth diameter are roughly the same height. c. These times do not include preparing of a coil of clay by Da 2 before forming on the wheel. The preparation time was 13 seconds (N = 1). d. Mean values for forming each half of the mold for this vessel were first half = 114 sec, range = 73 –134; second half = 97 sec, range = 74–125. The mean time combined the mean forming times for both halves; the combined ranges consist of the addition of the ranges of the lowest values to those of the highest values. The times do not include the time involved in the steps described in Table 8.4.

the turntable according to this account, but it requires two persons to complete the task. Although the woman’s account of the speed of the wheel seems a bit overstated given the cooperative nature of her production on it, it does show that she perceived the wheel to be faster than the turntable. The effect of task segmentation can also be seen in the differences of fabrication time for identical shapes of the same size. The youngest daughter (Da 4, Table 8.2) is less experienced using the wheel than her older sister and takes longer to make wheel-made bowls by herself. Her speed of fabrication is slightly more than twice the time necessary to make the same shape with the task segmentation of two persons (Table 8.2). The use of an organizational solution of task segmentation for a technological problem underscores the social dimension of technology. Muscular patterns required for forming pottery are culturally determined and socially transmitted, and production intensity may be constrained by undeveloped and unstrength244

How Has the Forming Technology Changed?

Table 8.3. Summary of fabrication times for wheel-made plates (platos) from different production events. (Adapted from Dean E. Arnold, Jill Huttar Wilson, and Alvaro L. Nieves, “Why Was the Potter’s Wheel Rejected? Social Choice and Technological Change in Ticul, Yucatan, Mexico,” in Pottery Economics in Mesoamerica, ed. Christopher A. Pool and George J. Bey III, p. 75 [University of Arizona Press, Tucson, 2008]. Used by permission) Vessel shape plato 1 plato 2 plato 3

Forming technique N wheel wheel wheel

18b 3d 4 e

Mean (sec) Range (sec) Makera 92 91 67

59–115 63–118 36–94

Da 4, Da 2, Da 1 Da 1 Da 2, Da 1

Height (cm)

Mouth (cm)

— — —

24c 13 13

Notes: a. The designation “Da” refers to “daughter” and the number following indicates birth order: “Da 1” is the oldest, and “Da 4” is the fourth oldest. b. These were timed at a rate of one out of every three plates for the fifty-three plates produced(see p. 245). This forming time does not include time for scraping (mean time is 22 sec, N = 22) or smoothing (mean time is 61 sec, N = 18). c. This is a fired size. The unfired diameter for this one is 28 cm. d. One out of every three plates were measured in this production event. e. Timing began when the paste was removed from a lump and continued until the vessel was put aside to dry. In this production event, every second or third plate was measured. One daughter shaped the crude form of the vessel and the oldest daughter finished the vessel on the wheel. When working alone on the wheel, the oldest daughter completed a plate in 39.43 seconds.

ened muscles. As a result, social means are used to solve what appears to be a technological problem as potters reorganize production using task segmentation to increase productivity. Borrowing and Adoption: Molding

A third forming technique used by Ticul potters consists of using molds joined along a vertical axis. With this technique, a clay pancake is forced into each portion of a mold. After a brief drying period, the molds are joined, allowed to dry for a few minutes, and then the completed vessel is removed (Brainerd 1958:68; Foster 1948:357, 1955:6, 1967:115). This technique seems simple, but actually consists of a total of fifteen distinct and sequential behavioral steps for a two-part mold (Table 8.4). This technique was also introduced in the 1940s by the government­sponsored workshop. After the Oaxaca potter had worked for about four to five years, a man from Campeche ( Juan Bautista Chab) joined him and began producing pottery using molds made of plaster of Paris. The Campeche potter was not paid by the government but supported himself by selling the objects that he produced and remained in Ticul four to five years. When Raymond Thompson (1958:20) visited the workshop in 1951, he reported that Chab was in charge, but he used molds rather than the wheel. Eventually, Chab left Ticul sometime 245

How Has the Forming Technology Changed?

Table 8.4. Principal steps in the behavioral chain for making mold-made vessels. Steps 1, 2, 5, and 6 may occur in reverse order, but except for this variation, the steps represent a fixed sequence or “chain” of behaviors (adapted from Dean E. Arnold, “Advantages and Disadvantages of Vertical-half Molding Technology: Implications for Production Organization,” in Pottery and People: A Dynamic Interaction, ed. James M. Skibo and Gary M. Feinman, p. 62 [University of Utah Press, Salt Lake City, 1999]. Used with permission) Step 1: Dust the first half of the mold with sah kab temper to keep the clay from sticking to the mold. (The clay minerals in the temper absorb water from the clay so that the vessel can be removed easily.) Step 2: Remove a piece of clay from a large lump, roll it into a sausage shape, and flatten it into a pancake with a thickness of one-half centimeter, using one of two ways: a) Place it on a piece of cloth on the floor, flatten it with the palm, and then peel the clay pancake from the cloth (large molds); b) Flatten it between the hands (small molds). Step 3: Press the flattened clay in the first half of the mold, taking care that the clay is forced into all of the portions of the mold and has a consistent thickness. More clay may be added or subtracted, if necessary, where the molds are joined. Step 4: Set the mold and its contents aside to dry. Steps 5–8: Repeat steps 1–4 with the other portion of the mold. Step 9: Combine the parts of the mold. Step 10: Set aside the completed object in the mold to dry. Step 11: Remove the object from the mold. Step 12: Set the object aside to dry. Step 13: Place the mold in the sun to dry. Step 14: Obliterate the mold marks using a knife or gourd scraper. Step 15: Smooth the remaining portion of the mold marks with a hand dipped in water. (On circular vessels, the mold marks may be smoothed and finished on the turntable or the wheel.)

after Thompson’s visit in 1951, and a boy named Rene España continued working. When España died, the workshop closed. Six potters were involved in the new workshop, but only three (Antonio Chable, Carmelo Chan, and Cesario Mex) spent enough time there to learn the molding technology and how to paint objects using oil-based paint. None of these men came from pottery-making families, and none were really potters. Antonio, however, had married into a pottery-making family and Cesario Mex had learned how to make figurines (such as birds) from a member of one of the extended Chan families ( Juan Bautista Chan), but he did not make traditional Maya pottery. From the 1940s to the late 1960s, Carmelo and Antonio were the principal agents for introducing the molding and painting technologies to the community. After they left the workshop, Carmelo used the new technology to produce figurines in his house and Antonio went to work for another potter (Guadalupe 246

How Has the Forming Technology Changed?

Tzum). Then Antonio taught Guadalupe’s son, Manuel, how to use the new technology. By 1966, Carmelo, Antonio, and Cesario were still making figurines with this technique, but it had been adopted by 43 percent (12/28) of the production units in Ticul. Each of these units had someone who specialized in making figurines. By 1984, production using molds had expanded greatly to encompass many more potters who used the technique to produce copies of pre-Hispanic objects. Why Was the Molding Technique Adopted?

Advantages of the Molding Technique. Molding has several advantages that provide deviation-amplifying feedback for the adoption of the technique. First, molding can be used to make objects that cannot be fabricated in any other way. Most of the mold-made objects produced between 1965 and 1970, for example, were figures—such as swans, ducks, pigs, bulls, saints, cartoon characters, or barrels of a very spicy variety of chili pepper (habanero)—that were used as coin banks. Before molding was adopted, these shapes could not be formed using slab coiling on the turntable but could only be made by hand modeling. A second advantage of molding is that it requires much less skill and a less sophisticated development of motor habit patterns and muscle strength than virtually any other type of forming technology. In contrast to other fabricating techniques, molding does not require long periods of learning (see Arnold 1985:203–208, 1999). It is easy to learn, and almost anyone can produce a molded object with little practice; it is so easy to learn that it is the first forming technique learned by children and by those who learn the craft as adults. Anyone who learns this technique can be economically productive in a relatively brief period of time (see Arnold 1999). Because so little skill is required for making mold-made vessels, unskilled individuals can be drawn into production quickly without going through the more lengthy process of learning slab coiling on the turntable. Molding thus was an ideal technique to increase economic returns by involving unskilled children and relatives in production. They can easily supplement a potter’s production during peak demand and temporarily increase production intensity (Arnold 1999). The lack of elaborate motor habits required for molding means that this technique requires no particular working position and is thus compatible with virtually any number of existing muscular patterns associated with other fabricating techniques. Molding thus can be integrated easily into existing work postures and can be employed along with any number of other forming techniques. For a molding technique, one can sit, stand, or work in any number of positions. Most important, molding is compatible with sitting on a low stool or on the floor (Figure 8.5). This position was used most frequently for making pottery, resting, 247

How Has the Forming Technology Changed?

or doing any number of other activities. This advantage accounts, in part, for its adoption by Ticul potters and its popularity between 1965 and 1997. A third advantage of molding technology is that the molds for new, innovative shapes can be fabricated with minimum skill. Creating molds requires more skill than making objects from them, but it still requires fewer motor skills than modified coiling. To create a template for the mold, potters may model an object or may purchase an object to use as a template. To form the mold, potters smear a paste over the template and then place plaster of Paris over each half of it, taking care that the mold is thick enough (1.25–1.5 cm) that it will not break easily. When the plaster is partially dry, one-half is removed and laid out in the sun to dry. After that half is dry, the template is removed from the other half and dried. Molds for some vessels are easy to make, but for others the process is much more difficult. Creating a mold for a pre-Columbian figurine, for example, may require the skill of a master potter because one must first model a template, cut off the object’s hands and feet, and then make the molds for each part. Fourth, if potters do not know how to make molds or do not want to do so, they can often obtain them from others. So besides the low skill needed to produce mold-made vessels, the ease at which potters can acquire molds also explains the rapid adoption and dissemination of this technology. In 1966, one of the men (Carmelo Chan) who was one of the first to learn the molding technology sold molds to potters so that those who did not know how to make them could still use them. Molds also may be given away, borrowed from close relatives, and, on one occasion, copied surreptitiously by an employee. A fifth advantage of molding is that it creates a uniform product. This uniformity is inherent in the technology, and given the low level of skill required, it is a superior technique to produce vessel homogeneity compared to other techniques because the uniformity comes from the use of the mold, not from the skill of the potter. Identical vessels can be made with the turntable, but doing so requires practice, skill, and a measuring device. Although homogeneity may be important in pots, it is particularly critical in figurine production because molding can maintain the integrity of the image during repeated fabrication events. In figurine production, this iconographic integrity may be the single characteristic that the potters desire most because modeling may produce too much variability, something that consumers may not want (Arnold 1999). Disadvantages of the Molding Technique. The molding technique also has disadvantages, providing deviation-counteracting feedback for some uses. These disadvantages have limited its adoption for fabricating many vessels. 248

Figure 8.5. Molding a vessel using a traditional working position. Molding can be done by the unskilled, and this man was a priest vacationing in the home of his brother-inlaw, who was a potter in 1984. The cardboard box at the man’s feet holds the temper used to dust the mold to keep the clay from sticking.

How Has the Forming Technology Changed?

The first disadvantage of a molding technique is the amount of handling time required for fabrication. Molding requires fifteen steps to fabricate the vessel with a two-piece mold because the potter must repeatedly pick up a mold, perform an activity, and then set it down (Table 8.4; Arnold 1999). By way of contrast, if a vessel is small and does not require multiple stages of fabrication, it can be formed and finished in one step, using the wheel or turntable. With mold-made fabrication, production time thus does not just reflect the time of actually forming the vessel but also includes the length of the combined segments of preparation, partial drying, and finishing. Handling time, and therefore production time, for mold-made vessels increases dramatically when objects require more than one two-piece mold. Parts of figurines, for example, are often made in multiple molds, and then they are dried separately until they can be attached safely without sagging. The addition of each mold set required to make an object increases the length of the behavioral chain more than one might expect (Table 8.5). The number of steps can be reduced by using a single mold for multiple small parts of a figurine or by modeling those parts. Modeling, however, requires skill as well as time. Second, the production intensity of mold-made pottery is limited by the number of molds available because some drying must occur while the object is in the mold. Each vessel must dry partially in each section of the mold and again when the molds are joined. Production intensity thus is limited by the number of molds available because a mold cannot be used by more than one potter at a time. If one wants to increase the production of a particular shape beyond a single potter, each potter must have his/her own mold for each shape and size. A third disadvantage that produces feedback with production intensity consists of the amount of capital required for making mold-made objects. Using molds increases the amount of capital required for production because potters must purchase the plaster to make the molds. Consequently, potters must assess whether the returns from making a vessel with a mold are worth the cost. In 1984, for example, plaster of Paris cost twenty-five pesos per kilogram, and in order to produce a mold to make a miniature of the vessel used to carry water (cantarito), one potter said that three kilograms of plaster were necessary. So the mold cost seventy-five pesos to make. If the potter produced seventy-five of such vessels, the cost of the plaster would increase the overhead by one peso per vessel. Larger molds increase the per vessel cost, and if few vessels are produced, the initial costs of making the mold may never be recouped. Such costs, however, are counterbalanced by the use of unskilled labor to make the vessel and, of course, the hope that the mold may be used again in the future. Nevertheless, the cost of plaster to make molds can reduce potters’ returns. 250

How Has the Forming Technology Changed?

Table 8.5. The number of fabrication steps for mold-made vessels requiring more than one two-piece mold. Although the number of fabrication steps required for two-piece molds is fifteen (Table 8.4), the number of fabrication steps necessary for vessels that require more than two molds is greater and was calculated according to the following formula: Total number of molding steps = N [the number of portions of the mold] × 4 [Steps 1–4] + N – 1 [Step 9] + 6 [Steps 10–15]. If the parts of a molded figurine are small, potters may create a single mold for several component parts. In such cases, Steps 1 to 4 may not need to be repeated. Object

Total number of mold portions required Comment

Total number of steps required in the behavioral chain

Ancient Maya 6 image (ídolo)

One two-piece mold is necessary for the body, another is necessary for the arms and legs, and a third is necessary for the face. Some copies of ancient figurines require only four or five molds.

35

Angel 3

Each mold consists of one-third of the figure’s circumference.

20

Certain flowerpots

3

20

Pig

4

25

Horse

4

25

Bull (toro) 10

The body requires two molds; two additional molds are required for each leg.

55

6 Frog (zapo)

The body requires one set of molds; each front leg requires a set of two molds.

35

4 Swan (cisne)

The body requires one set of two molds and the neck and the head require another set of two molds.

25

Fourth, mold-made vessels still need to be finished using another technique. The joint marks where the molded portions come together must be obliterated (Figure 8.6). If the vessel is open and the interior of the vessel can be seen, the potter must also obliterate the finger impressions remaining from pressing the clay into the mold. If the molded vessel has a circular horizontal cross section, the potter usually smoothes these marks using the turntable. If a mold-made vessel requires finishing and smoothing on the turntable, however, why not just make the vessel with the turntable rather than with a mold? The turntable requires fewer steps of handling than molding and thus requires less time, even though the total amount of handling time was not measured (e.g., Table 8.2). A fifth disadvantage of the molding technique is that some vessels simply cannot be mold-made at all. In October 1984, I asked an informant whether 251

Figure 8.6. Obliterating the mold marks on a small cántaro in 1984. Larger sizes of this vessel were formerly used to carry water; small vessels of this shape are now made as souvenirs for tourists and painted with copies of ancient Maya figures.

How Has the Forming Technology Changed?

Figure 8.7. Shelves used to store molds in 1997. The requirement of a different mold for each size and shape puts pressure on storage space. One way to solve this problem is to use shelving; this innovation is one of the great changes in the use of space among production units since 1965.

fancy plant pots with tripod supports and a ruffled rim (risado) could be moldmade. He replied that they could be made with molds, but it would be difficult to produce the ruffled rims successfully. Even if some parts of a vessel are made with molds, other parts cannot be made with molds. Why, then, use a mold at all to make such vessels? A sixth disadvantage of using molds consists of the amount of space required to store them. Traditional slab coiling on the turntable only requires enough space for the potter to sit in front of the turntable. In this space, the potter can produce any number of different sizes and shapes. With molding, however, a different mold must be made for each size and shape the potter produces. If shapes and sizes change (as they often do), then the number of molds will increase; more molds require more storage space. This change increases the footprint of production and exerts pressure on the available space. Consequently, the more variability and innovation in the shapes made with molds, the greater the number of molds that are required and the greater the need for more space for storing them (Arnold 1999; Figure 8.7). This storage problem is reflected in the self-reported ages of the molds stored in one potter’s house in 1997. This potter told me that unused molds were thrown away, but after eliciting the ages of molds and plotting their ages on a graph, the graph revealed that some molds had been kept for as long as thirty years (Figure 8.8). These data underscore the point that molds are a capital-intensive aspect of 253

How Has the Forming Technology Changed?

Figure 8.8. Trend line showing the relationship of the frequency and age of molds in one production unit in 1997 (N = 60; mean = 8.7 years). This graph suggests that there is a “fall-off ” in the ages of molds. It also suggests either that there has been a great increase in diversity in the number of objects that were mold-made in the ten years before 1997 or that some molds have been discarded. Since some discarded molds were seen at this house in 1997, it is likely that both of these suggestions explain this trend. The lower number of older molds used suggests that the vessel shapes in demand are changing and that some older molds are discarded, just as the potter said. If pottery shapes are always changing and the potter made a new mold for each new shape, one would expect a very different kind of curve.

pottery production and potters do not appear to discard them unless they are damaged because they never know when they might need them again. Potters who keep their molds, however, must store them properly. Molds that are left outside in the weather, for example, deteriorate and cannot be used. The storage behavior can be illustrated by a graph of the ages of all the molds that one potter owned. The curve of the graph suggests that at least some molds must have been discarded because most of the molds were relatively new. If new molds were made regularly and usually stored, the curve of the graph would be very different. Seventh, like other techniques, potters cannot use molds to make every vessel size and shape because size and shape are constrained by the highly plastic character of the Ticul clay. This limitation is probably a consequence of the presence of the highly plastic clay mineral montmorillonite. With a molding technique, the clay must be plastic enough to be flattened, stretched, and forced into the mold 254

How Has the Forming Technology Changed?

Figure 8.9. Bar graph showing the frequency of the maximum dimension of each mold

in one production unit in 1997 (N = 63; mean = 16.27 cm). Since molds are a maximum of 1.25 to 1.5 centimeters thick, subtract 2.5 centimeters (1.25 × 2) from the dimensions shown here for the maximum length of the molded object.

without cracking, but not so plastic that the clay body will sag after removal from the mold (Louana Lackey, personal communication). Unfortunately, larger and more open vessels will sag if they are fabricated with molds. Consequently, only small vessels and larger vessels that are totally enclosed (like coin banks) can be made successfully with molds. One informant asserted that the largest vessel that can be made successfully with molds is a cylindrical vessel (vaso) about twenty centimeters tall. Such vessels can be removed from the molds and not sag or crack because the straight sides provide support for the clay above. This limit of twenty centimeters can be demonstrated by a frequency distribution of the sizes of the total inventory of molds from one production unit in 1997 (Figure 8.9). The distribution of these sizes of molds shows that some variability exists in this size limitation because larger vessels that are completely enclosed (such as a pig-shaped vessel) have an unbroken curvature that provides strength and support. Some small vessels, on the other hand, will sag after removal from the mold if they are not enclosed enough. If, for example, the potter uses a mold to make a vessel on which its neck lies directly above an inward slope of the wall, the wall below will sag because it does not provide enough support to bear the weight of the clay above (Arnold 1999:67–68). Some large figurines are still made with molds, but they have the problem of sagging. In order to prevent sagging, the figurine must remain in the mold for one 255

How Has the Forming Technology Changed?

day before it is removed. This constraint limits the speed that vessels can be made from that mold. Making more molds, however, increases the per-unit costs and places more demand on storage space. One potter solved this problem by placing newly molded pigs in a pile of temper to reduce the chance of deformation and cracking and allow the mold to be reused more quickly. In 1997, another potter solved this problem by wrapping each branch of mold-made cacti with cardboard, supporting the branches with a string attached to the rafters of the structure in which they were drying. In this way, the mold could be reused sooner than if the object dried in the mold. Innovation: The Ball-bearing Turntable

During the late 1970s, owners of larger production units began adopting a new kind of turntable that mechanics had assembled from metal parts. The new device operated in the same way as the traditional turntable, but it turned with ball bearings located either on top of, around, or below a stationary metal post on which a movable platform of metal was attached (Figure 8.10). The metal rod may be anchored in different ways: in cement placed in a gallon can (Figure 8.11), on a large wood block, or welded to a large piece of metal, such as a gear or pulley (Figure 8.12). Potters do not use the traditional Yucatec Maya word k’abal (“turntable”) for this new device but rather use the Spanish words tornete, tornet, and balero. This new device appears to be an internal innovation, but its ultimate origin is unknown. About 1981, one potter reportedly saw the new turntable in a few larger production units. He asked an automobile mechanic to make the device and then tried to sell it to potters in his extended family. By 1984, it had largely replaced the traditional turntable in 54 percent of the production units visited (Table 8.6; Ralph and Arnold 1988), including the larger production units (Figure 8.13). Although no systematic observations were made about the percentage of production units using the ball-bearing turntable in 1997, no traditional turntables were noted and the adoption of the new device appeared to be complete. Advantages of the Ball-bearing Turntable. Potters recognize several advantages of the new turntable. Both its ball-bearing operation and its height above the floor provided significant sensory deviation-amplifying feedback for its adoption. First, this new turntable uses the same motor habits as the traditional turntable but is easier to use because it requires less muscle strength. Each time the potter used the traditional turntable, he oiled the nail to make the turntable rotate 256

Figure 8.10. A ball-bearing turntable (tornete) used in 1984. There are many varieties of this turntable (see also Figure 8.11), but usually bases are made of something heavy enough to anchor the device and prevent its movement during the forming. In this case, the base of the turntable is anchored in a heavy block of wood. The single ball bearing can be seen on top of the shaft.

Figure 8.11. A potter using a type of ball-bearing turntable anchored in a gallon can of cement in 1984. Note that this same potter was using the k’abal in Figure 8.1.

Figure 8.12. A potter using a type of ball-bearing turntable made with a machine pulley in 1984. The potter is able to turn the vessel by the movements of his hands because of the ball bearing. The vessel wall is thinned and shaped to the desired dimensions using a gourd scraper, shown here (left). Although identical motor habit patterns are used for forming pottery on the traditional turntable and on the tornete, the platform of the tornete is higher largely because the pottery made in 1984 was generally smaller than that made from 1965 to 1970.

How Has the Forming Technology Changed?

Table 8.6. The number of production units with traditional and ball-bearing turntables in 1984, and the mean number per unit. The totals only represent the production units visited. Type of turntable Total Traditional Ball-bearing

20 34

Number of production units with them

Mean per production unit

17 14

1.2 2.4

Figure 8.13. Bar graph showing the distribution of traditional and ball-bearing turntables among the 1984 production units. The total number of production units for which turntable data exist is twenty-six. The traditional turntables tend to occur in larger production units with more than one turntable.

more easily. By way of contrast, the ball-bearing turntable can be moved quickly and easily with the force of the hand, and oil is needed only rarely. The foot is not required to turn the device even when both hands are used to form the pot. The ball-bearing turntable thus speeds production by ease of use, less maintenance, and less frequent use of oil. In 1984, potters did not get as tired using the ball-bearing turntable as they did with the traditional turntable. One potter said that he prefers the new device because it requires less strength and is easier to use; its operation is also smoother and allows more precision in forming, slipping, and finishing. Another potter said that he was calmer and more rested when he used the new turntable, and its operation is smoother and faster than the traditional device. A third potter said 260

How Has the Forming Technology Changed?

that he can turn the ball-bearing turntable more easily and he can smooth and finish the pottery better than he could using the traditional turntable. Furthermore, the new device speeds his work. The greater ease of using the ball-bearing turntable is affirmed by its longterm effect on a potter’s muscles. Potters say that once they use the new device consistently, they cannot return to making pottery with the traditional turntable. This observation suggests that using the new turntable is, in fact, easier because the muscles required for the traditional turntable are not being strengthened and tend to atrophy. As a result, the traditional turntable has been abandoned in households that use the new ball-bearing device. Second, the height of the working surface on the new turntable provides more flexibility in the working position and also makes fabricating pottery easier. Because the size and weight of the base are required to support the metal platform and ball bearings, the working surface is significantly higher than the height of the traditional turntable (18.4 cm vs. 9.7 cm; T = 8.21, p <0.00001, F = 41.7, p <0.001, N = 35; Table 8.7). The new device is heavily anchored, so the height of the working surface can be manipulated by placing blocks of cement or wood underneath it, raising it as much as thirty-eight centimeters above the floor. The potter can thus minimize the amount of bending from the waist that was necessary for the traditional turntable, and the new device thus puts less stress on the spine and back muscles. These advantages can be illustrated by a comment made by one potter in 1988. She had recently had an operation on her abdomen and could not make pottery with the traditional turntable that required sitting on a low stool. By using the ball-bearing turntable, she could sit in her hammock and make pottery because the new device was higher and easier for her to turn than the traditional turntable. Disadvantages of the Ball-bearing Turntable. The principal disadvantage of the ball-bearing turntable is its cost. Since local mechanics made this device and knew that the potters wanted it, they charged them a high price to make it. The ongoing demand for these devices had driven up their price in 1984, and by 1988, both inflation and demand had driven up their cost far beyond the 1984 price. In spite of the seemingly innovative quality of the ball-bearing turntable, it is just the latest development in the evolution of the turntable in Ticul. As one potter pointed out, the turntable has undergone a five-stage transformation. The first turntable was just a round piece of wood that was turned with the feet and hands, as noted by Thompson (1958:76–81) in 1951 in other communities in Yucatán. In the second stage, a square piece of wood was added to the circular block. Then, 261

How Has the Forming Technology Changed?

Table 8.7. Comparison of the heights of the traditional and ball-bearing turntables in 1984. These measurements do not include the height of the removable platform, which adds three to five centimeters to the height of the turntable. Range Mean height of working surface above floor Standard deviation Sample size

Traditional turntable

Ball-bearing turntable

6.5–12 cm 9.7 cm 1.6 19

11–36.5 cm 18.4 cm 5.9 35

the turntable was stabilized with a nail so that it would turn around a fixed point to prevent it from shifting while it was rotating. This change had already begun to occur when Thompson visited Ticul in 1951 (Thompson 1958:80). Then, sometime after the late 1960s, a circular metal plate was placed around the nail and potters used oil to facilitate the movement. Finally, by the 1990s, the ball-bearing turntable was fully adopted. An Unlikely Adoption: Slip Casting

By 1984, a ceramics factory was operating that utilized an industrial-level technology. In 1997, production consisted of a slip casting section for making white glazed porcelain and a mechanical extruder to produce earthenware tiles. The slip casting technique consists of filling plaster molds with liquefied clay. The clay is pumped directly from the mixer through flexible tubes that are used to fill molds that are open at one end and held together with large elastic bands (Figure 8.14). Because the mold absorbs some of the water from the clay, the clay near the mold hardens. Then, the molds are turned upside down and the remaining liquid is poured out and used again. Eventually, the mold is opened and the object is removed. The mold marks are obliterated and the object is smoothed, placed on a drying rack, and then dried, fired, glazed, and fired again. Slip casting reduces the time spent placing the clay in the molds that regular molding requires. But, just as drying time consumes much of the fabrication time with molding, drying time is also the critical time-consuming factor with slip casting, and the factory has the same problem as potters who use molds. With both techniques, molds of plaster are critical because the plaster adsorbs water from the clay, which allows the vessel to keep its shape when it is removed from the mold. Drying molds is thus critical for speeding the molding process. Although the amount of handling required for molding is reduced with slip casting, the major variable affecting production speed is the amount of time that the clay sits in the mold. So, the drier the mold, the less time it takes for a thin layer of clay to dry against it. Dry molds thus speed the molding process by reducing the drying time.

262

How Has the Forming Technology Changed?

Figure 8.14. Slip casting molds filled with clay in the Ticul ceramics factory in 1997. This highly capitalized and low-skill factory production exists alongside the more traditional pottery-making technology in Ticul, even though the two technologies are unrelated and have no influence on each other. (Photo by Michelle R. Arnold)

In order to speed production, the factory automated the molding process by using a large moving carousel along which workers performed specialized tasks in assembly line fashion (Figure 8.15). First, one worker binds a mold with a rubber band and places it on a carousel. Down the line, another worker fills the mold with a clay slurry, and farther along, a third worker dumps the excess clay from the mold into a receptacle below and places the mold upside down on the belt. Finally, a fourth worker removes the object from the mold and places the mold on the belt where it enters a drier. When it emerges from the drier, the process begins again. After partial drying, another worker trims the molded objects and sets them on a rack for further drying. The slip casting technique has all of the strengths and weaknesses of the molding technique that Ticul potters use. Storing molds creates great pressure on production space just as it does in the smaller production units. In addition to storing some molds on the factory floor, the owner built a large storeroom (approximately 50 m by 10 m) that contained hundreds, if not thousands, of molds stacked two meters high.

263

How Has the Forming Technology Changed?

Figure 8.15. The carousel in the Ticul ceramics factory in 1997 used for slip casting and

drying molds. (Photo by Michelle R. Arnold)

Making the molds is probably the most skilled task in the factory; skilled workers must create molds and then key them to fit together precisely. Some molds have three or more parts and the skill consists of knowing how many parts are necessary. Mold makers may use modeling clay to make a template, but they may also use plaster and then turn the template on a lathe to give it the proper shape. The finished template is then placed in a wooden box with two sides missing and the remainder of a box is built around the template. The void is then filled with plaster, and after the plaster has hardened, the template is removed and the mold is trimmed. As a highly capitalized facility, the factory represents the end product of the evolution of ceramic production that first began thousands of years ago. The factory, however, is not an evolutionary product of Maya pottery making but rather consists of a non-Maya industrial technology resulting from the owner’s access to capital. A few Ticul potters have worked in the factory, but generally the factory has not significantly influenced local pottery production. One potter, however, used his experience at the factory to develop a slip casting technique in his own household. He came from a family of potters, received a formal education, and 264

How Has the Forming Technology Changed?

then worked in the factory for eight years. In 1996, he quit his factory job and used his newly acquired knowledge for making slip cast vessels. This potter believed that slip casting produced vessels faster than molding. With enough molds, he says, he could make thirty vessels every half hour, whereas with the traditional molding technique, he could only make fifteen vessels in onehalf day with two-piece molds. With fifty molds, he said, he could keep the fabrication process going without any interruption in production. Consequently, the critical factor for the efficiency of this technique is the number of molds available. Since the potter must buy plaster to make the molds, the main obstacle to increasing his production and efficiency is the lack of capital. Choosing a Technique With so many technological and social constraints on the choice of fabrication technique, the only true choice that the potter has occurs with the production of food bowls, which can be fabricated with modified coiling, molding, or the wheel (Table 8.2). For the rest of the vessels, choosing a fabrication technique is more complicated. In 1984, one potter had a distinct plan for choosing a fabrication technique. He had abandoned the use of the traditional turntable but used three other techniques (molds, the wheel, and the ball-bearing turntable). He said that the wheel is faster than molding or the new turntable and that making vessels with the new turntable is easier than molding. He only used molds when he wanted to make small vessels of the same shape and size, or if a vessel could not be made with any other technique. Molding is not easier, he said, nor is it faster than the wheel or the new turntable. Changing Explanations of Dimensional Variability For archaeologists, ceramic variability through time has become one of the fundamental surrogates of social and technological change. Changing amphora shapes and domestic pottery from the House of the Vestals in Roman Pompeii, for example, mark both the changing trade relationships and the development of socioeconomic complexity in Pompeii between 150 B.C. and A.D. 79 (De Sena and Ikäheimo 2003). American archaeologists, however, have looked beyond simple changes in shapes and type frequency to dimensional variability of vessels. In particular, increased dimensional homogeneity of ceramic vessels through time has become a surrogate index of increased specialization (Benco 1987; Rice 1981; Underhill 2003). Such notions of homogeneity (standardization), however, 265

How Has the Forming Technology Changed?

have been criticized (Arnold and Nieves 1992; P. Arnold 1991a; Longacre et al. 1988; B. Stark 1995) because a variety of factors are responsible for increased dimensional homogeneity of ceramic vessels over time. The explanations of dimensional variability of Ticul vessels changed between 1965 and 1997. These explanations consisted of potters’ changing perceptions of variability and involve the ways in which the potter controls variability by measuring (or not measuring) vessels during production, how narrowly or broadly he/she conceives of the vessel shape categories, and how measurement techniques are expressed in the variability of the actual dimensions of these shapes. Seen from this perspective, vessel variability is socially constructed, and the potters, rather than the fabrication techniques, are the agents of creating that variability. Variability in Traditional Vessels

During the late 1960s, potters controlled the size and shape of vessels in three ways, and these methods are still used when they produce traditional vessels. First, they exercise limited control of the variability on bowls (cajetes), incense burners (incensarios), ashtrays (ceniceros), and candle holders (candeleros). The height of food bowls and incense burners, for example, is consciously measured using the width of the four fingers. The mouth diameter of all of these vessels is determined by the size of the gourd scraper used to shape and finish their interior. One size of scraper was used for a large bowl and another size for a small bowl. Although it is not as overt or conscious as other measuring techniques, this technique results in a certain size vessel regardless of whether it is made on the traditional turntable, the wheel, or the ball-bearing turntable. However, bowls, incense burners, ashtrays, and candle holders were (and still are) susceptible to size variations depending on interpersonal variability of finger width and the size of the gourd scraper used. Because scrapers eventually get smaller with wear, mouth diameter of vessels should change over time. Further, these measurement techniques affect the homogeneity of the bowls such that the statistical variance of their dimensions is significantly larger than that of other vessels (Arnold and Nieves 1992). The second method of controlling variability of traditional vessels consists of what potters call “calculating” (calcular), and they use this method to fabricate vessels such as the water-carrying jar (cántaro, Figure 8.16), the maize-soaking jar (apaste), and the water-storage jar (tinaja). This method is only used by older and more experienced potters in two ways. The first way that potters “calculate” vessel size involves using three measuring units based on the size of the hand (Table 8.8). These units are applied differently to different parts of the vessels to produce five different sizes (medidas) of these vessels (Tables 8.9–8.13). This technique is also used with pitchers (  jarras) that have a body height of one cuarta plus three or 266

Figure 8.16. Potter making a traditional water-carrying jar (cántaro or p’uul) in 1984.

Such vessels can only be made by traditional potters because of the skill required to thin the vessel walls and measure the different parts of the vessel.

How Has the Forming Technology Changed?

Table 8.8. The measurement units that potters use for making traditional vessels (such as cántaros, tinajas, and apastes). Measurement Definition

Metric equivalent for an adult male potter

jeme

The distance from the end of the extended thumb to the end of the extended index finger (forefinger)

15.5 cm

cuarta

The distance from the end of the extended thumb to the end of the extended little finger (a handspan)

20.5 cm

one dedo

The width of any one of the three longest fingers of the hand; often the width of the index finger is used

about 1.5 cm

three dedos

The combined width of the three longest fingers at the first joint from the palm

about 6.5 cm

four dedos

The width of the four fingers at the first joint from the palm

9 cm

four fingers (dedos). They are the same size as the water-carrying vessels (cántaros) that are used for piñatas. English Spanish Yucatec Maya In this case, the potters have defismall chico chichan nite social choices concerning the sizes mediano chan tan kelen or medium of vessels to make, and these sizes are chuumak dependent on several factors that are regular tup’is regulara related to the use of the vessels, their large grande nohoch social demand, and their intended gigante hach nohoch market. The size of the water-carrying a. A vessel larger than medium but not large. jars produced in the mid-1960s, for example, was related to two factors: age of the female carrying the vessel and the distance to the water source. Small jars were produced strictly for little girls, although there was some variability in the sizes and measurements (Table 8.10). The demand for large and medium jars, on the other hand, was dependent on the distance that one must walk from the water source while carrying a full vessel. The people in communities near the city of Tekax, for example, did not care about the size of their jars because water sources are relatively close to the houses. In Santa Elena on the other side of the puuc ridge, however, consumers wanted medium-size jars because the water table there was seventy meters deep and only one well existed in the community. Women had to walk some distance to obtain their water and thus wanted a smaller jar. A second way of “calculating” the size of a traditional vessel involves measuring a portion of a vessel by the number of coils of clay (vueltas) used to form it. Table 8.9. Nomenclature for the size categories of traditional vessel shapes (e.g., cántaros, tinajas, and apastes).

268

How Has the Forming Technology Changed?

Table 8.10. Measurements used in the stages of the water-carrying jar (cántaro in Spanish or p’uul in Yucatec Maya). Size of vessel

Vessel part

Small

Medium

Regular

Large

Base diameter

4 dedos or 1 jemea

1 jeme or 1 cuarta

1 cuarta

1 cuarta

Body height

1 cuarta or 4–5 dedosa

1 cuarta or 5 dedos

1 cuarta or 1 jeme

2 cuartasb

Neck heightc

2 vueltasd (a little A little less more than 8 than 1 jeme dedos, which is slightly less than 1 jeme)

A little less than 1 jeme

1 jeme (3 vueltas)

Mouth diameter

Depends on size of hande

Depends on size of hande

1 jeme

Depends on size of hande

Notes: a. The size of the base diameter and body height are used for making vessels for young girls. These vessels also have smaller mouths. b. The handles are placed exactly opposite of one another at an approximate height of one cuarta above the base. The same potter that provided these measurements in 1984 made twelve cántaros that had a body size of 1 cuarta plus 1 jeme on request from a workshop owner. c. The neck height for most sizes of the water-carrying vessels is approximately the same because the arm is placed around the neck while the vessel is carried on the hip. d. A vuelta (literally, a “turn”) is one full coil of clay and serves as another unit of measuring vessel sizes (see p. 271). e. In order to form the narrow neck of the cántaro, the hand must fit inside the mouth. The mouth diameter thus varies somewhat between potters. Some potters, however, make the mouth diameter smaller than their hands.

The experienced potter knows the appropriate size of each coil so that the number of coils becomes a rough surrogate measure of size. Both the hand-measurement method and the number-of-coils method may be combined on the same shape. The use of these measuring techniques does not create vessels that vary by the same amount across all sizes of a particular shape. Since particular motor habit patterns are required to carry some vessels, these patterns affect vessel dimensions, and some parts of the vessel need to be more uniform than others. The neck height of the water-carrying jar, for example, is very similar in all sizes (except the small size used for little girls) so that the arm can fit around it (Figure 3.6a–b). The dimensions of the other parts of this vessel, however, vary with the size of the vessel (Table 8.10) and according to what each potter deems as appropriate measurements for a given vessel. In contrast to these methods of controlling (or not controlling) variability for traditional vessels, potters only use precise measurements for making vessels 269

How Has the Forming Technology Changed?

Table 8.11. Uses and measurements of the different size categories of the apaste. The mouth diameter is not measured but must be smaller than the greatest diameter.

Size of vessel



Regular

Regular

Large

Very large

<1 cuarta (but more than 1 jeme)

1 cuarta + 2 dedos

2 cuartas + 1 jeme

3 cuartas

washing or soaking maize

bathing

no data

no data

nohoch homa’ kat

muy especial

Medium

Body height 1 cuarta + 5 dedos Use

soaking maize for preparing tortilla dough (nishtamal)

kati pook kum kat bo’h a’ kati ich kil Maya name

Table 8.12. Alternative measurements for making an apaste.

Small

Body height 1 jeme Height of second stage

2.5 dedos

Size of vessel Medium

Large

1 cuarta + 3 dedos

1 cuarta + 3 dedos

3 dedos

>4 dedos

when separate portions of a vessel must fit together. Pitchers (  jarrones), waterstorage jars (tinajas), and maize-soaking jars (apastes) with covers need to be made with precise measurements so that the cover will fit the vessel. Similarly, flowerpots with columnar pedestals also need to be measured precisely so that the pot will fit on top of the base. Learning to make pottery of the appropriate size thus involves learning the appropriate measurements for each shape. Since measurements are based on hand sizes, children cannot learn them effectively because measurements are based on the adult hand. If a potter asks a child to make a specific size, for example, then the potter has to specify the size more precisely because children do not know how to compensate for their small hands, and their hand sizes change as they grow. When children mature and know the proper sizes, however, a potter can ask them to make a particular vessel by specifying the hand measurements. Variability of Vessels Produced after the Late 1970s

Potters’ perception and control of dimensional variability changed by 1984, and they were much more concerned about producing uniform vessels. Using Berg’s (2004) terminology, Ticul potters began practicing “intentional standard270

How Has the Forming Technology Changed?

Table 8.13. Measurements used for each stage of the water-storage jar (tinaja). The body of the tinaja is the same as the body of the apaste. Vessel part

Small

Medium

Body height 1 cuarta 2 cuartas

Size of vessel Regular

Largea

2 cuartas + 4 dedos

2 cuartas + 1 jeme

Neck height <1 jeme 1 jeme 1 cuarta 1 cuarta Mouth diameter

<1 jeme

1 jeme

1 cuarta

1 cuarta

Very large 3 cuartas 1 cuarta + 2–3 dedos 1 cuarta

Note: a. This size holds sixty liters and corresponds to the volume of three cántaros of water. This size was more popular than the larger size, which consumers did not buy.

ization.” This change was largely related to the emergence of a different population of consumers. In the late 1960s, the population that bought the vessels actually used them so that the buyer was also the consumer and end user. Since the consuming unit was the household, producing dimensionally uniform vessels was not important because few, if any, households purchased more than one vessel from a potter’s production and sales event. With the development of intermediaries/brokers and demand from hotels and tourists, uniform sizes were required. Brokers contracted with potters for vessels of a particular size and shape (usually specifying the height and mouth diameter) and potters produced precisely measured vessels to fulfill these requests. Vessel homogeneity was required for two kinds of vessels for different reasons. First, plant pots had to be a specified size and shape because hotels wanted identical vessels throughout their facilities. Second, small vessels had to be made of a specified size and shape because they had to be small enough to fit into a tourist’s suitcase but also had to be identical because they were painted by specialists (not the potters who made them), and the designs were planned, formatted, and painted for a particular size. Design layout and formatting of the design are time-consuming and if all of the vessels are the same size, the time required to paint each vessel is less than that required for painting vessels of different sizes. Consequently, if one vessel is smaller, potters say, brokers will reduce the price they pay because the design has to be reformatted. Brokers thus requested certain shapes, sizes, and quantities and potters responded by making whatever the brokers wanted. Vessels made for brokers in each production event are measured precisely and are thus highly standardized for each sales event. The variability of vessels produced by the same potter from sales event to sales event, however, may vary and may also vary from potter to potter.

271

How Has the Forming Technology Changed?

In order to achieve this degree of uniformity and standardize the vessels for a broker, potters use different measurement techniques. The first of these techniques involves using a ruler to cut a stick of the appropriate length and then using it to measure the completed pot. A second technique consists of sizing the base of the vessel to the width of the platform (tabla) on which the vessel is formed. The width of the platforms may correspond exactly to the width of the base of the plant pots made on them so that the platform size is a guide to produce a uniform base diameter. This technique, of course, has no relevance for the heights and diameters of vessels; therefore, another measuring technique must also be used. A third technique consists of using a mold to fabricate the vessels. A template is made of the appropriate size (the desired size plus an allowance for shrinkage; see below). Then, a mold is made around it, and the mold is used to produce as many vessels as desired. The molding technique is usually restricted to small vessels, and it may be used with those small vessels that ordinarily would be produced using modified coiling. Producing vessels of a precise size and shape is not quite as simple as it might first appear. Potters must take into account the amount of shrinkage of the clay from drying and firing; they recognize that vessels shrink one centimeter for every ten centimeters of height. So they adjust their initial measurements accordingly so that the fired vessels are the appropriate size. Using molding to achieve uniform vessels is a great change in the use of that technique since the late 1960s. At that time, molding was only used for those figurines that could not be made by any other technique except hand modeling. Even though molding has been used in Ticul since the 1940s, it was not used deliberately or consciously to make vessels uniform until after the development of the tourist market in the late 1970s. In summary, as a result of the changes in demand and market, several innovations were necessary to produce vessels of uniform size. Using Schiffer’s terminology (2005), the development of uniform vessels required a “cascade of innovations.” Such innovations were not necessarily material but were behavioral and organizational. Conclusion For those archaeologists who see ceramics as largely the result of culture imprinting itself in the plastic clay, the change in the fabrication techniques in Ticul provides a contrary scenario. The repertoire of these techniques consists of a range from the traditional technique to the capital-intensive, mass-production techniques of the modern ceramics industry. Consequently, it would seem that 272

How Has the Forming Technology Changed?

Table 8.14. The relationships of labor, skill, and capital required for different fabrication technologies; their dates of introduction are noted in parentheses.

Low capital High capital

High labor Traditional turntable and skill (traditional k’abal )

Ball-bearing turntable (late 1970s)

Low labor and skill

Wheel (1930s)

Molding (1940s)

Slip casting (1980s)



Mechanical extrusion (1990s?)

these techniques could be arranged along an evolutionary continuum from the traditional high-skill, high-labor-intensive technique of slab coiling on the traditional turntable to the low-skill, high-capital-intensive slip casting and extrusion technique of the modern factory (Table 8.14). Such an assessment, however, glosses over the advantages and limitations of the different techniques and the nuances of their origin, adoption, and rejection. Further, such an evolutionary arrangement submerges insights gleaned from understanding the advantages and disadvantages of those techniques, and this understanding is useful for archaeologists in their interpretations of the past. The history of the use of fabrication techniques over time illustrates different processes of technological change. These processes consist of the replacement of the traditional technique with a new technique, the enhancement of the traditional technique, and the addition of totally new techniques. Mold making, the wheel, and slip casting have been added, and the ball-bearing turntable enhanced the traditional turntable and eventually replaced it. Describing technological change by these processes, however, overlooks the details of why the change occurred. Borrowing, diffusion, and innovation crosscut all three classifications. The notion of borrowing focuses on the action of the participants who adopt an item from elsewhere through social contact. Diffusion, however, is simply the movement of the technique from one social unit to another without specifying the actions of the participants. Innovation is a behavior in which someone invents a tool or recombines two or more disparate elements to create a new tool. Innovation may involve experimentation in which changes are tried and rejected before a device or technique is fully adopted.

273

How Has the Forming Technology Changed?

Innovation, borrowing, and diffusion of a fabrication technique can occur and the technique can still fail to be fully adopted. Adoption thus is a social process that builds on, and is different from, processes such as diffusion, borrowing, and innovation. Changes in the forming technology in Ticul illustrate all of these processes. The government’s action of trying to introduce the wheel was an attempt at diffusion. Although it was tried by several potters, it was not adopted and was eventually abandoned even by those who tried it. Even this adoption was minor because the potter did not really use the potential of the centrifugal force of the wheel but rather used it as a simple turntable for forming pottery with slab or modified coiling. The slip casting technique was recently borrowed from the local factory by a former employee, but it is unlikely that other potters will adopt the technique because the raw materials and the fabrication technology are so different from those that already exist in Ticul and require different knowledge and skill. Furthermore, the molds and clay require much more capital investment than most potters possess. The enhancement of the traditional turntable with a ball-bearing device was an innovation that developed in the late 1970s or early 1980s and appeared to be totally adopted by 1997. Some varieties of these turntables that were abandoned in some potters’ houses in 1984, however, indicate that welders, auto mechanics, and potters experimented with a number of different variations and some of them failed to be adopted. Consequently, even the adoption of the innovative ball-bearing turntable was not totally seamless; those features that turned out to be very important were selected by potters. What factors then led to the adoption and rejection of the new technology? In general, those techniques that are most compatible with the existing motor habits, work positions, and low or limited capital were adopted. Further, adoption of these new techniques occurred via social relationships. For example, even though the ball-bearing turntable was expensive, it was adopted by some large production units and then by the extended Tzum family because one potter in the family was convinced of its efficacy and encouraged his relatives to adopt it by using the trust of his kin ties (see Ralph and Arnold 1988). It was eventually adopted by all the potters in Ticul. Were the new fabrication techniques adopted because they were more efficient? It depends on the meaning of “efficient.” If “efficient” means that vessels were fabricated in less time than they were previously, then molding, for example, was not more efficient because the handling time required exceeds the fabrication time of making the same vessel on the turntable. If increased efficiency means that potters can increase their financial returns by fabricating vessels that they could not make previously, then the molding tech274

How Has the Forming Technology Changed?

nique is efficient because it permits potters to produce an innovative set of vessels for a different market. Molding also provides iconographic integrity for figurines such as animals and religious images. Achieving this integrity is not possible with modeling or with any other technique (Arnold 1999). Further, if by “efficient” one means that the size of production units can be temporarily increased because unskilled personnel can be drawn into the production process to fabricate and finish vessels that were previously restricted to skilled potters, then molding is a highly efficient technique. Molding is also efficient because it allows potters to allocate unskilled personnel to fabrication tasks that were heretofore impossible without skill, and it allows experienced potters to devote their time only to those tasks that require high skill. Molding is thus efficient in that it reorganizes production by reallocating the skilled labor involved in fabrication and by drawing unskilled labor into the process. Consequently, the adoption of a molding technique is indeed related to more efficient production, but in a socially and culturally embedded way. Even with the capital expenditure required to use a molding technique, potters were able to produce vessels that could not be made in any other way, and this change allowed them to increase the diversity of their shape repertoire and tap into a new demand and market for their vessels. This change permitted potters to adjust to changing market conditions. Molding thus was not competitive with the traditional technique but rather was complementary to it because it allowed potters to produce new kinds of vessels. Unlike modified coiling on the turntable, however, potters required capital to purchase the raw material (plaster of Paris) to make the molds. As the use of molds grew, capital was also required to create sheltered space to store the unused molds. Further, the size of molded vessels was limited by the constraints of the highly plastic Ticul clay. The adoption of the ball-bearing turntable was less complicated and also more efficient. Even though it was a capital-intensive item, potters recognized the ease of making pottery with the new device, and they were less tired when they used it. Potters who adopted the new turntable thus realized that it reduced the energy inputs for making pottery. At least some potters also believed that the device sped up production. This change was particularly important for the large production units that were the early adopters of the ball-bearing turntable. With greater ease of use, individuals learning how to make pottery could do so with less muscle strength. New potters could learn to use it with less development of new motor habit patterns than was required by the traditional turntable. Learning to use it thus was easier than learning the traditional turntable and resulted in greater productivity 275

How Has the Forming Technology Changed?

per individual. In this sense, the ball-bearing turntable was more efficient than the traditional turntable. Returning to the issue that began this chapter, do the forming techniques in Ticul “have wide tolerance of the clays and other raw materials . . . so that almost any of these techniques could probably be implemented almost anywhere” with “a few minor modifications,” as van der Leeuw (1993:239) proposed? If the traditional forming technique in Ticul did have a wide tolerance of clays and could be used to make any vessel shape, then why did potters experiment with and then adopt new forming techniques? It seems that the traditional technique should be adequate for making any vessel. Even though the Ticul clay is very plastic and can be formed into many different shapes, none of the techniques can produce any shape desired simply “by introducing a few minor modifications” (van der Leeuw 1993:239). Rather, the forming techniques and clay’s physical properties constrain the potters’ choices, and they must adjust to these constraints. In some cases, they are surmountable by using another forming technique or by combining two techniques. The history of fabrication techniques in Ticul thus demonstrates that the interrelationships among forming techniques, clay, and vessel shapes are much more complicated than van der Leeuw thought. From the point of view of technical choice, the Ticul potter has five choices to fabricate pottery. In reality, however, all of those choices are not equally viable. Although potters certainly do have a choice in the fabrication technique that they use, those choices are constrained by both technical and social factors. Potters create and adopt forming devices, but it is also true that the forming devices are subject to the limitations of the materials and those of the forming technologies themselves. Potters are constantly negotiating, and renegotiating, new solutions to technical problems that arise because some new shapes cannot be formed as effectively with one fabrication technology as with another. In some cases, vessels are made with two techniques to solve these problems. The existence of multiple techniques in Ticul (whether they were adopted or not) and their evolution thus suggest that a complex interdependence exists between the limitations of the raw materials, the techniques of fabrication, and the demand of the market. Not all techniques can be used to make every shape. Rather, under certain conditions, certain techniques are better suited than others to produce some shapes. Molds, for example, are best used to make small vessels less than twenty centimeters along their longest dimension. Molds are also ideal for making non-circular forms, such as figurines, that cannot be made in any other way, but molds cannot always be used to make uniform vessels. They can be used to make small uniform vessels but cannot be used to make large uniform vessels because the highly plastic Ticul clay causes larger vessels to sag and crack. 276

How Has the Forming Technology Changed?

Such vessels are best made in successive stages using modified coiling with drying periods between the stages and then measured in some way in order to ensure uniformity. The evolution of Ticul’s shaping techniques challenges some of the notions about the relationship of forming techniques and the development of specialization in antiquity. Some of the theoretical work dealing with the development of specialization has focused on the products of that specialization and their identification in the archaeological record (Benco 1987; Blackman et al. 1993; Rathje 1975; Rice 1981). One of these products is the presumable uniformity of vessels, which is regarded as a surrogate of higher levels of specialization resulting from more efficient technologies such as molding or the wheel. As the Ticul data show, it is not merely the presence of apparently more efficient technologies (such as molding) that is important. Rather, different forming technologies require different amounts of skill and social complexity. From the viewpoint of social agency, the adoption of molding in Ticul did not occur because it was a more efficient fabrication technique or because it produced uniform vessels. Rather, molding was first adopted because it allowed potters to make pottery that they had not made previously. It was not until the development of the demand for uniform vessels in the 1980s that molding was used to make such vessels. This change occurred forty years after the molding technique was adopted. Only then were the efficiencies of molding realized by temporarily drawing children and non-potters into making mold-made vessels because the technique did not require much skill. As for the production of uniform vessels using other techniques, what was important was not the development of a technique that led to vessel homogeneity but rather the way in which the techniques were utilized and organized. This point was made long ago in a parallel way by William Mitchell (1973) concerning the Wittfogel hypothesis. In response to the belief that irrigation required statelevel sociopolitical organization, Mitchell argued that according to Wittfogel, it was not the presence of irrigation that was important but rather the manner in which irrigation was organized. Many relatively small-scale community irrigation systems occurred in Peru without the need for higher levels of sociopolitical complexity (Mitchell 1976). Similarly, molding and wheel fabrication technologies do not necessarily indicate that potters are choosing them for reasons of efficiency or vessel homogeneity. Rather, as this chapter has demonstrated, potters choose fabrication technologies for a variety of reasons that are socially embedded (Sillar and Tite 2000) and technologically and economically based, such as the plasticity of the clay, the limitations of the technique, the amount of capital required, the kinds of vessels produced, and the skill of the maker. 277

How Has the Forming Technology Changed?

Fourth, the evolution of forming technology in Ticul reveals the complexity of agency in that evolution. Social agency in technological change is very obvious to anthropologists, but this agency is mediated through the constraints of the socially embodied and culturally determined motor habit patterns first noted and described by Marcel Mauss in 1935 (1976:97–123). These motor habits are largely unconscious, deeply embedded cultural patterns learned by repeated muscle actions (habitus in Mauss’s [1976:101] terminology) that are reinforced by furniture and working positions used (Arnold 1985:147–149; Spier 1967). Besides social agency, the fabrication techniques, skills, and materials used in technological production also have agency because certain techniques have certain universal consequences that crosscut cultural boundaries. These techniques limit choices and create new problems to which the potters must adapt. The Ticul data, for example, reveal the deeply conservative nature of work postures and motor habit patterns that constrain technological change. The devices for forming pottery may change, but the basic motor patterns and working positions involved in making pottery change very little, and then only to make work easier (e.g., the ball-bearing turntable). Even though the wheel seemed to be adopted, the different motor habits required different patterns of muscular movement, and the potter had to quit working on the wheel after 1.5 hours. The wheel was faster than the other techniques and modified coiling was still used as a forming technique on the wheel, but the motor habits required to propel the wheel used muscles that were weak and had not been strengthened. Work on the wheel thus could not be sustained and did not have any significant economic benefit because the time spent on the wheel was too short. The physical embodiment of the technology in the syntax of the muscular patterns thus also has agency. Similarly, the Ticul clay places constraints on the way in which vessels can be built. This constraint can be illustrated by the continued use of modified coiling for building vessels using the wheel and the new turntable. The plasticity of the Ticul clay is well suited to slab coiling and molding. Large vessels, however, must be made in stages with drying periods between the stages because otherwise they will sag. For this reason, large vessels with concave profiles can only be made with modified coiling and cannot be made with molds. Pottery-forming techniques thus are not just the products of sociocultural organization, but rather they exert causal pressure on other aspects of sociocultural phenomena, such as vessel shapes and the organization of production. Forming technologies and raw materials also have agency, just as humans do. Finally, in previous chapters I have described surrogate measures of increasing intensity and scale (Costin 1991). In this chapter, the surrogate measures of increased intensity consist of the adoption of the new ball-bearing turntable. It 278

How Has the Forming Technology Changed?

was first adopted by the larger production units (Figure 8.13), because only the larger production units could afford the capital outlay to purchase it. The return for this expense was ease of operation and presumably more rapid production. Further, workers who learned how to make vessels on the new turntable did not need as much muscle strength as potters who used the traditional turntable. The use of the molding technique also provides an indication of increasing intensity. First introduced in Ticul in the 1940s and used primarily to make vessels that could not be made in any other way, molding has become a preferred means to provide small uniform vessels ultimately destined for the tourist market. Painting workshops prefer uniform vessels to minimize the effort in formatting and painting the design. Equally important is the increased production intensity that potters can achieve by drawing unskilled labor into the craft. Usually, these increases are temporary and often consist of children, unskilled relatives, and other workers who may be hired temporarily to fulfill an order. Although the evolution of the fabrication technology over the thirty-two years of this study is complicated, evolutionary change has moved toward more efficient forming technologies, such as the adoption of the ball-bearing turntable. This change, however, did not result in the production of standardized vessels. Rather, it was the result of demand coming from middlemen/brokers that produced this outcome. It is likely that demand and market had similar effects on the production of vessels in antiquity.

279

Chapter

How Has Firing Technology Changed?

nine

I

n order to fix the shape of a clay vessel so that it will not revert again to a formless mass in the presence of moisture, the vessel must be fired. This process removes the remainder of the physically held water in the pottery after air drying and drives off the chemically held water in the molecular structure of the clay minerals (Arnold 1985:61–62; Kolb 1996; Rice 1987:63–65; Shepard 1956:81–91). At higher temperatures, the lattice structure of the clay minerals collapses and the particles fuse together, irreversibly fixing the form of the vessel (Rice 1987:90–94). In spite of its critical role in the behavioral chain of pottery making, however, firing has not figured prominently in theories or specialization, or in theories of the evolution of the craft. Differences in degree of firing have been inferred as the result of a more efficient firing technology (Costin and Hagstrum 1995), and several scholars have recognized that some firing techniques provide more control over the firing process than others (Arnold 1985:213–218; Rice 281

How Has Firing Technology Changed?

1987:152–163; Rye and Evans 1976:164–165; Sheehy 1988:212–214, 1992; Shepard 1956:81–88). Like the changes in production technology documented in previous chapters, firing technology has also changed between 1965 and 1997. Do these changes follow the same kind of trajectory as the changes in fabrication technology? Did specialists emerge for fuel procurement and firing just as they did in clay and temper mining? Was there a similar pattern in changes in efficiency? The answers to these questions reveal a pattern of task specialization parallel to clay procurement, temper procurement, and vessel fabrication, but with a slightly different implication for the organization of the craft. Changes in the Procurement and Use of Fuel Obtaining firewood is one of the most critical aspects of the firing process and is a convenient by-product of slash-and-burn agriculture. Prior to 1965, many potters practiced part-time slash-and-burn and possessed a knowledge of the forest and the types of trees that grew there. This knowledge was systematized into categories of different ethno-ecological zones that ranged from the northern coast of the peninsula to the tropical forest in the south. Some zones are more favorable for agriculture than others, but each zone has distinct characteristics with specific types of trees that grow there. Each type of tree produces wood with specific burning characteristics. Some trees, for example, produce wood that burns slowly or burns with a smoky flame, whereas other trees produce wood that burns with a strong, high, or smokeless flame. Knowing these burning characteristics helps potters control the firing process so that they can maximize the effects of the wood they use and not overfire or underfire their pottery. Knowledge of the ethno-ecological zones, the trees that grow in each zone, and the burning characteristics of the wood from each type of tree thus contributes to the success of firing and to the efficiency of the process. In the late 1960s, firewood was supplied by local slash-and-burn agriculturalists (milperos) who brought it from their fields in bundles (tercios). Sometimes milperos came to potters to sell their firewood, but at other times, potters went to milperos to purchase it. Since some potters were also milperos at various times between 1965 and 1997, they procured their fuel as a consequence of their own agricultural activities. Others simply cut their firewood from abandoned swidden plots. Specialization in the Procurement of Firewood

The kind of firewood available changed greatly from 1965 to 1997 because of the changing zones from which it came. This change was a consequence of at least two interrelated factors. 282

How Has Firing Technology Changed?

First, as previous chapters have shown, production of pottery increased after the 1970s. This change put pressure on the traditional firewood suppliers, the milperos, who brought a bundle or two of firewood each time they went to their fields. Meanwhile, fewer and fewer men were milperos, and those few could not supply the amount of firewood needed for the potters. With increased production, the demand for firewood increased and then outstripped supply. The fuel problem is common for preindustrial potters in other parts of the world as a result of deforestation (Arnold 1985:31–32, 36, 53, 54; Rice 1987:162–163) but is also related to the kind of resources used for fuel and population density. Some potters have adapted to this problem by using the byproducts of local industries, like sawdust, or the discards of an industrial society, such as tires (Lackey 1982). Other societies, like the Hohokam in the northeastern Sonoran Desert, where fuel is already scarce, used fuel-efficient strategies, such as firing at low temperatures and cooking with large heat-conserving communal roasting pits (Fish 2000:258, 261, 266, 274). Sheehy (1988, 1992), however, found that even with the use of fuel for cooking, for lime making for plaster, and for making pottery, one pottery workshop in Teotihuacan “did not require an inordinate amount of wood to complete their yearly ceramic production” (Sheehy 1988:221). The scarcity of fuel in Ticul may seem simply to be a byproduct of population growth and increased demand, but some ancient potters probably also faced a fuel problem, especially in areas where deforestation existed in antiquity, such as the southern Maya lowlands (Shaw 2003). In the modern setting, however, this problem is solved through energy extenders that come from modern transportation, just as it was in extending the distance required to procure clay and temper in Ticul. Second, to make up for the reduced amount of firewood, specialists emerged to provide potters a consistent supply. This organizational change produced an effect on the type and quality of fuel available. In 1965, firewood came from the ecological zones near Ticul—namely the kabalche’ zone and the nearby hill zones. By 1984, most potters no longer obtained firewood from local milperos because few men had fields and those who did could not supply enough to meet the potters’ demand. As a result, firewood was imported from the more remote ya’ash k’ash and wits zones south of the puuc ridge by entrepreneurs from Santa Elena and Bolonchén (see Figure 3.2). They contracted with milperos in those regions to cut 100 bundles at a time and then brought a truckload (usually about 200 bundles) of wood to Ticul every week, or whenever potters asked for it. The quality, however, was often poor and unpredictable. Consequently, some potters still purchased their firewood from milperos. It was more expensive, but its quality was better and it included a greater diversity of types than the firewood bought from specialists. 283

How Has Firing Technology Changed?

By 1988, the source of firewood had changed again and was brought by truck from nearby Chapab. By 1994, informants said that there was no longer any forest available for fields around Ticul and firewood could not be reliably obtained from local milperos. Rather, it came from other locations, such as Tekit and Chapab. By 1997, the demand for firewood had greatly expanded, and the number of trucks coming to Ticul to sell firewood had increased. As many as four individuals came from Chapab to sell it. Another one or two came from Tekit, another brought firewood from Ya’ax Che’ near Bolonchén, and others came from Mama and Santa Elena. A few potters were still swidden farmers and obtained their own firewood, and two potters still obtained it from local milperos. Changes in the Use and Amount of Firewood

Besides the procurement of fuel, changes have also occurred in its use. During the late 1960s, the potter sorted the firewood according to its burning characteristics. Each stage of the firing process had a different purpose and the potter needed to match the burning characteristics of the firewood with the purpose of each stage. This practice also had the latent, and perhaps unintended, consequence of conserving the amount of fuel used because the burning characteristics of one kind of wood may counteract the effects of another. One kind of fuel will burn slowly, another will burn with a smoky flame, and a third may burn quickly. By 1984, my principal informant had changed his practice of using specific types of firewood for each stage because his firewood came from more remote and less familiar ethno-ecological zones and he did not know the wood’s burning characteristics. Even when the types and their characteristics were known, however, its quality was poor because it had not dried sufficiently, burned slowly, or burned with a lot of smoke. Using such wood meant that he could not sort the wood and efficiently allocate its use during firing as he did in 1965. In order to avoid these problems, a few potters still fired in the old way, obtaining wood of known quality from milperos, carefully sorting it, and using the appropriate types for each part of the process. My principal informant, however, added any kind of wood at the end of the process, kept the flames high, and used visual feedback from the red glow (or lack thereof ) of the pottery when the flames subsided to evaluate whether the vessels had been fired sufficiently. Consequently, by 1984, potters recognized that firing required more wood than it did in 1965. Changes in Kiln-making Technology Traditionally, Ticul potters have used two different types of firing technology: one for cooking pottery and another for non-cooking pottery (Thompson 1958: 284

How Has Firing Technology Changed?

Figure 9.1. A traditional beehive kiln in 2002. Unlike many kilns with changes and modifications, this kiln is covered with sheet metal to protect it from rainfall, just as kilns were in 1965.

96, 99). Each type utilized a different beehive kiln of probable Moorish origin (Figure 9.1; Lister and Lister 1987). Since the production of cooking pots has been largely abandoned since 1965 (which is briefly discussed in the temper chapter), this chapter focuses on changes in firing technology that occurred with producing non-cooking pottery. Baseline for Change

In the 1960s, the knowledge and skill that a potter possessed enabled him/ her to build his/her own beehive kiln with no more cost than the purchase of the materials. Kilns are built using a unique type of rock called sakel bach tuunich (literally, “sakel bach rock” or “white birds’ eggs rock”) and a mixture of red earth (k’an kab) and pottery temper (sah kab) for making the mortar between the rocks and for the plaster on the kiln’s interior. Potters say that they use pottery temper for mortar and plaster because the white earth in the temper creates a hard plastered surface and is more fire resistant than the naturally occurring marl used for construction purposes. Nevertheless, this plaster does not resist the heat very well and must be replaced when it cracks and breaks. 285

How Has Firing Technology Changed?

Task Segmentation and Specialization in Firing The number of potters who had the knowledge and skill to build a traditional kiln has declined since 1965, so that by 1997, relatively few potters possessed the ability to build their own kilns. Consequently, the number of potters who do not own or use a kiln has increased. This increase has resulted in segmenting the tasks of forming from those of firing and in changes in the organization of production. Kiln Ownership and Task Segmentation

Between 1965 and 1997, changes have occurred in the organization of firing. In 1965, potters without a kiln, and who lived next door or across the street from a relative who owned one, asked the relative to fire their pottery for them. Because unfired pots are so fragile, moving them any distance increased the risk of breakage; thus, potters did not sell their unfired vessels to another potter. By 1984, however, some potters without a kiln paid a neighbor or relative to fire vessels for them or borrowed or rented a kiln. Many potters, however, preferred to sell their pottery unfired. Some kilns were used by more than one production unit, but it appeared that most potters without a kiln were selling their pottery unfired to potters who had small trucks and could easily transport the fragile pottery to their own production units for firing. What factors have contributed to separation of whose who fire pottery and those who fabricate it? First, a kiln is expensive to construct and requires costly materials and skilled knowledge; it is a potter’s most capital-intensive investment. Not all potters have the capital to buy the raw materials, and hiring someone else to build it is also expensive. In 1984, one potter said that a medium-sized beehive kiln required 10,000 pesos (about US $50.00) to purchase 400 rocks, a cartload of k’an kab, and a cartload of temper to make the mortar and plaster. A second reason that a potter may not own a kiln is the cost of maintaining it. A kiln must be maintained or it will deteriorate over time because of rainfall and repeated firing. Just as building a kiln requires knowledge, skill, and capital, maintaining it also requires these resources. If the rocks on the inside of the kiln are exposed to the full force of the fire, they will crumble because of calcination. So they must be protected by a fire-resistant plaster. If the plaster cracks and breaks, however, and the rocks are exposed to the flames, potters say, the pottery remains black. This result is probably a consequence of the production of carbon dioxide during calcination, which creates a reducing atmosphere during the final stage of firing. Consequently, a kiln with rocks exposed to the fire will eventually collapse. To avoid these problems, a potter must maintain the kiln by digging out the damaged plaster and the crumbling parts of the rocks on the inside of the kiln and then replastering it. 286

How Has Firing Technology Changed?

Potters must also maintain the outside of the kiln. The mortar and plaster on the outside are not water-repellant and will fall apart in the rain. The kiln is thus subject to the same kinds of environmental constraints as making and drying pottery, and potters must protect the kiln to keep the rain from seeping through the mortar. In 1965 and 1966, potters placed leaves of the huano palm on top of the kiln to protect it, but they also used tar-impregnated roofing sheets or large pieces of sheet metal for protection (Figure 9.2). One kiln was faced with a layer of cement. By 1984, however, two production units were protecting the outside of their kilns with cement, and in 1988, this innovation had extended to two more kilns. By 1994, one of these kilns had been totally covered with cement, and in 1997, 68 percent (32/47) of the kilns surveyed were at least partially covered with cement (Figure 9.3). This innovation protected the kiln from rainfall, extended its useful life, and reduced maintenance. Still, 21 percent (10/47) of the kilns had no cement covering at all (Figure 9.1); no data existed for the remaining 11 percent (5/47). These changes in kiln maintenance can be illustrated in the succession of kilns built in a single production unit between 1965 and 1997. In 1965, this unit had two kilns and both were covered with huano palm fronds and tar-impregnated roofing material (Figure 9.2). By 1984, the two kilns used in the late 1960s had been dismantled and the potter had constructed a new kiln, using cement around the door and palm fronds, sheet metal, and cardboard on the top for protection (Figure 9.3). By 1994, however, these portable materials had been replaced with a cement exterior, which still remained in 1997 (Figure 9.4). Not keeping a kiln in repair may have dire consequences. One kiln observed in 1984 had not been maintained and had fallen into disrepair. By 1988, the owner could no longer use it, and he had to rent a kiln from another potter. The rental, however, produced a new set of problems that required an even greater investment of time and money than buying firewood and using his own kiln; he not only had to pay for the rental of the kiln but also had to pay the cost of bringing the pottery from his house to the kiln site. At a cost of three pesos a trip, it cost twenty-four pesos just to carry the pottery to the kiln, and he lost one day of making pots in order to do it. If the potter was not paid well for his vessels, he gained nothing from renting a kiln. Consequently, by 1994, this same potter was selling his pottery unfired. A properly maintained kiln may last as long as thirty years. Although no systematic data were collected about kiln ages, none of the kilns observed in 1965 and 1966 remained in 1997 without major rebuilding. Based on these observations, thirty years is a reasonable maximum life for a kiln, but the median age appears to be much less. 287

How Has Firing Technology Changed?

Figure 9.2. Water-storage jars being removed from a kiln in 1965 as it was being loaded

with small vessels for providing water for animals (pilas). Kilns are made with rocks and a mixture of red earth (k’an kab) and can be damaged by precipitation. So in order to reduce maintenance and prevent collapse of the kiln, palm fronds, sheet metal, and tar-impregnated roofing material are placed on top of the kiln to protect it from rainfall. The white head coverings are protection from the heat of the sun and the heat of the kiln while unloading pottery.

A third factor contributing to task segmentation is that some potters do not have the knowledge and skill to fire pottery. As near as I can assess, all male potters who were heads of household in 1965 knew how to fire even though some did not have a kiln. By 1984, however, the frequency of those potters who knew how to fire in the traditional way was declining because many new potters (who had not come from pottery-making families) had not learned the skill. These potters had only a rudimentary knowledge of pottery production because they just learned how to use vertical half molds or how to make plant pots (or small vessels) using modified coiling. Firing required much more knowledge and skill. An unskilled potter who tried to fire did so at great risk of loss. If firing was unsuccessful and the pottery was damaged, then the potter lost not only the cost of the firewood but also the clay and labor invested in fabricating the vessels. Unlike unfired vessels that are damaged in rainfall or household activities, the clay from vessels damaged during firing cannot be reclaimed and used again. As a result, some potters do not fire their pottery at all but prefer to sell it unfired to an experienced potter. 288

Figure 9.3. Kiln that replaced the kiln shown in Figure 9.2 in 1984. Cement facing was

added to part of the kiln exterior, and tar-impregnated cardboard, sheet metal, and other materials were used to protect the kiln from rainfall.

How Has Firing Technology Changed?

Figure 9.4. Kiln shown in Figure 9.3 in 1997. The kiln had been totally faced with

cement to protect it from rainfall and reduce maintenance. By 2002, the kiln was unchanged. (Photo by Michelle R. Arnold)

A fourth factor contributing to task segmentation is that firing delays monetary returns from the craft. This delay occurs because potters must allow their vessels to dry, wait to accumulate a kiln load to fire, acquire and prepare firewood, spend a day firing, and spend another day allowing the kiln to cool before the pottery can be removed. These production steps can add as much as a week to the process and increase the risk of loss. Consequently, even experienced potters may sell their pottery unfired because they need financial returns quickly and do not want to wait until the pots are fired. A fifth factor contributing to the task segmentation of firing is the substantial capital outlay required to obtain firewood. Unless a potter has a swidden field (milpa), he does not have access to firewood and must buy it. If a potter has little or no capital available, he must sell his pottery unfired. Firing thus adds capital costs to making pottery. As one potter said, if he sold his pottery fired, then he would have to charge a higher price for it and be less competitive. Changes in the Types of Kilns

Since 1965, potters expanded the kinds of kilns that they use (Table 9.1). These changes are also related to increased intensity, scale, and the scheduling 290

How Has Firing Technology Changed?

Table 9.1. Changes in types of kilns from 1965–1966 to 1997. Number of traditional beehive kilns observed Total number of nontraditional kilns observed Pot kiln (Figure 9.8) Cylindrical updraft kiln (Figure 9.5) Gas kiln (Figure 9.9) Cement-block kiln (Figure 9.7) Square kiln

1965–1966

1984

1997

23 0 0 0 0 0 0

22 3 0 2 1 0 0

39 9 1 3 2 2 1

of production. By 1997, several innovative new kiln types had been adopted by Ticul potters. Although not necessarily more efficient than the traditional kiln, they required less skill for construction, required less maintenance, and helped solve scheduling problems that existed with firing in the beehive kilns. Small Updraft Kiln. The first new type of kiln adopted was a small cylindrical updraft kiln (Figure 9.5) with an open top. The door of the kiln was a stoking hole and a source of air for the updraft. Pottery was placed in the kiln from the top and was suspended above a firebox by some fire-resistant material. The updraft kiln has five advantages compared to the traditional beehive kiln, and these advantages were instrumental in potters’ choice to adopt it. First, a potter can fire a batch of small pots, use less fuel, and obtain economic returns more quickly without waiting to accumulate a large batch for a large kiln. One of the very few potters to build an updraft kiln did so to fire fewer pots, increase firing frequency, and reduce risk of breakage. Since he was living in his wife’s family houselot, he wanted to eliminate the time required to carry pottery to his regular kiln in his father’s houselot a block away. He thus hoped that the updraft kiln would be more convenient and allow him to sell his pottery more quickly than if he waited to fire a batch in a larger kiln. He also hoped that each load could be fired more economically than in his own kiln. Second, loading the updraft kiln was easier and required less effort because he could load it from the top. Third, the potter did not need to split the firewood necessary for the final stage of firing for placing behind the kiln furniture of the traditional beehive kiln. In that kiln, kiln furniture (usually rocks or old pots) raises vessels off the floor. Because the base of the bottom of the stack of pottery is shielded from the heat of the flames, the potter places firewood below the pots and behind the kiln furniture. When the kiln is hot enough, this wood undergoes spontaneous combustion. To be sure this combustion occurs at the proper time, the firewood placed here needs to be split. In the updraft kiln, however, all of the pottery was above the fire rather than

291

Figure 9.5. The updraft kiln used in one production unit in 1997. This potter had a brother who had used this kind of kiln in 1984. (Photo by Michelle R. Arnold)

How Has Firing Technology Changed?

at its side; thus, the wood did not need to be split to ensure that it would ignite at the proper time in the firing sequence. Fourth, the updraft kiln required less time (four hours) to fire than the traditional kiln (six hours). Finally, the updraft kiln required less fuel because it was smaller than the traditional kiln. As advantageous as the new kiln seemed to be, potters had three major problems with it. The first problem was the open top. Although the position of the pottery above the door kept the blowing rain from damaging the pottery, the open top allowed heat to escape and allowed rainfall to damage the pottery. So the top of the kiln was covered with a piece of sheet metal during firing. At the end of the process, however, the potter must lift the cover to see if the pottery is fired properly. Even if the pottery on the top has been fired, there is still a possibility that the vessels deep inside are underfired or overfired because he cannot see them. With a traditional kiln, on the other hand, the potter can see all of the pottery from the door and can determine whether it is properly fired by the presence of its red glow. A second problem with the updraft kiln concerns the need to use a refractory material to suspend the pottery above the fire. Pots should not have direct contact with the fire or they will get too hot and get “burned” (as potters say) by undergoing calcination. Rather, pottery should be separate from the fire so that the potter can add fuel without breaking the pots. Pieces of metal were tried as pottery supports, but they softened with the heat and collapsed, and the vessels fell into the firebox and broke. Another potter used broken wasters for support, but if they shattered, the pottery in the kiln suffered a similar fate. Third, even though the updraft kiln requires less fuel than the traditional kiln, the proportion of fuel consumed in relation to that of a traditional kiln was higher. Even though a smaller amount of pottery could be fired, the costs of using this kiln were excessive relative to the returns. One potter’s traditional kiln required about twenty bundles of firewood to fire pottery valued at 10,000 pesos, whereas his updraft kiln required five to seven bundles of firewood to fire pottery worth 2,000 pesos. Using these data, I calculated whether the new kiln was more efficient than the old one for an amount of pottery adjusted to the size of each kiln. The updraft kiln was not more fuel efficient and required 25 to 75 percent more fuel to fire an equivalent amount of pottery than a traditional kiln. One reason for this lack of efficiency was that the open top probably lost more heat than a traditional kiln. In 1984, two potters had adopted the updraft kiln to supplement their traditional kiln, but by 1988, one had abandoned it. Nine years later, however, two other potters adopted it. By 2002, the 1984 adopter who had abandoned it in 1988 had built another using cement, large sherds, and cement blocks rather than 293

How Has Firing Technology Changed?

Figure 9.6. An updraft kiln used by one potter in 2002. This kiln was built using large sherds, rocks, and a cement block embedded in cement. The triangular metal piece sticking out of the top of the kiln is used to cover the kiln during firing to retain heat. This kiln had replaced a traditional kiln that was abandoned.

the more traditional rocks and k’an kab mortar (Figure 9.6). The change was so successful that he abandoned the use of his beehive kiln. The advantages of firing a small amount of pottery and serving clients more quickly outweighed the disadvantages of proportionately consuming more firewood than that used by a traditional kiln.

294

How Has Firing Technology Changed?

Figure 9.7. A square cement-block kiln used in 1997. This production unit also used a larger kiln for firing large vessels. (Photo by Michelle R. Arnold)

Square Kiln. A second kiln innovation consisted of square kilns. One subvariety of this type was constructed with cement blocks (Figure 9.7), which had the effect of making it square. Since cement reportedly does not resist the heat of the fire, the inside of the kiln must be plastered entirely with a k’an kab/sah kab mixture. Cement-block kilns, however, are not as durable as beehive kilns and are said to last half as long. A second variety of square kiln was constructed with an unknown material, plastered on the inside with the k’an kab/sah kab mixture, and then covered on the outside with cement. Pot Kiln. A third type of kiln innovation consisted of a very small kiln made from a large plant pot turned upside down and placed on some rocks (Figure

295

How Has Firing Technology Changed?

Figure 9.8. A pot kiln used in one production unit in 1997. This production unit also

used a large kiln for firing plant pots. (Photo by Michelle R. Arnold)

9.8). A hole was made on the side of the vessel to serve as the door and stoking hole. Such a kiln was used to fire very small items of pottery (such as ashtrays) so that the potter did not need to wait to accumulate enough pottery to fill a large kiln. Gas Kiln. A fourth firing innovation was borrowed from the modern ceramics industry and consisted of a gas kiln. Only two of these kilns existed in Ticul 296

How Has Firing Technology Changed?

Figure 9.9. A gas kiln in the government-sponsored production unit in 1997. Unlike all other Ticul kilns, this kiln was built inside the workshop with a chimney that vented gases and smoke outside through the roof. (Photo by Michelle R. Arnold)

and both were built in government-sponsored, capital-intensive production units. One of these was established in the early 1970s, and when it failed, one of its managers took control of it and moved it to the edge of town. It is unclear whether the gas kiln was used after that change. By 1994, the government had financed construction of another large workshop and wanted to build a hightemperature kiln behind it. The attempt to build that kiln failed and the potters working there built a traditional kiln instead. By 1997, however, the traditional kiln was abandoned and pottery was fired inside the building in a brick gas-fired kiln complete with a chimney to vent the smoke and gases outside (Figure 9.9). One of the workers said that firing this kiln was only slightly more expensive than firing a traditional kiln. Furthermore, potters could fire whenever they wanted to and were not limited by rainy or cold weather, which ordinarily delayed the drying of firewood and firing with the traditional kiln. A third advantage of a gas kiln was extolled by another potter (who did not have one), who said that a gas kiln produced no color variations. Firing with a wood-burning kiln, he said, produced many color variations because pottery in one part of the kiln may be overfired and in another part may be underfired. During a visit to Ticul in 2002, however, the new government-sponsored workshop was abandoned along with its gas kiln.

297

How Has Firing Technology Changed?

Figure 9.10. A hybrid kiln in 1997 that utilized both traditional and modern construction materials. Cement mortar and ordinary rocks were used for the exterior shell and cement was used for the exterior facing. (Photo by Michelle R. Arnold)

Hybrid Kiln. A fifth kiln innovation consisted of a blend of traditional and modern technologies (Figure 9.10). This kiln had all the benefits of a Mediterranean beehive-shaped kiln but used modern innovations in order to minimize maintenance. The innovative character of this kiln consisted of its unique combination of two kilns, one nested inside the other (Figure 9.11). The inner shell utilized traditional technology with the sakel bach type of rock used for the bottom four rows of the interior where the wall is exposed to the fire. The mortar inside the kiln and the interior-facing plaster were made with the traditional k’an kab and pottery temper. The outer shell of the kiln, however, was constructed with cement and ordinary rocks because they do not touch the fire. Cement was used to cover the outside of the kiln to protect it from rainfall. The principal disadvantages with this kiln were its cost and the time to build it. Because the kiln was large and essentially two kilns, the walls were, of necessity, very thick and required a large number of stones. In July 1997, the potter had already spent 17,000 pesos on it, and he believed that it would ultimately cost 20,000 pesos because of the quantity and quality of the stones required.

298

Figure 9.11. The profile of the hybrid kiln’s wall in 1997, showing the inner wall made of traditional materials. (Photo by Michelle R. Arnold)

How Has Firing Technology Changed?

Why Did the Changes in Kiln Technology Occur? In the conclusion of the previous chapter, I argued that the adoption of new techniques occurred via social relationships and social means, but if an innovation is otherwise incompatible with existing technology, potters probably will not adopt it. The adopters of the small updraft kiln in 1997 were brothers of potters who used them in 1994. An examination of the list of the owners of the cement-block kilns in 1997, however, did not reveal any close kinship or social connection. Nevertheless, a variety of other reasons appear to account for the adoption of a new kiln technology. The first probable reason for adopting new kiln types is the relative visibility of many kilns in the rear of potters’ houselots. Unlike the fabrication technology that occurs inside buildings and away from casual observation, kilns are often visible from the street or a neighbor’s houselot and can invite curiosity and replication. In the last chapter, those innovations that were most compatible with existing motor habits, work positions, and low or limited capital were more likely to be widely adopted by the potters. In this case, the most widely adopted kiln innovation was the cement facing on the kiln. This change required the least capital and the least risk. Certainly, potters incurred less risk by using cement facing on their kilns than by trying a new kind of kiln. On the other hand, the highly capital-intensive gas kiln was never adopted outside of government-instituted production units. As for the compatibility of the other changes in kiln technology with traditional patterns, the answer is more complicated. As was expressed in the previous chapter, the concern with efficiency in firing technology is also culturally embedded and expressed in culturally patterned ways. One of the significant reasons given for the adoption of the pot kiln and the smaller updraft kiln, for example, was the need to produce smaller amounts of pottery in a more consistent product stream to meet brokers’ orders with a shorter turnaround time. With smaller kilns, potters did not need to wait to fill their traditional kiln and delay firing or to fire it when it was only partially full. So the development of smaller kilns was a response to a scheduling problem created by the new broker demand for pottery. With the new smaller kilns, potters could fill broker and client orders with small quantities more quickly, and their outlay of capital for firewood was lower for a firing cycle, although firing in the smaller kilns might be less efficient than in a larger kiln. This change in technology is not the result of efficiency brought on by economies of scale, but rather just the opposite. The adopter of a small kiln is concerned about scheduling his output in order to maintain relationships with clients and brokers to ensure that they will buy more pots in the future. This concern for sacrificing short-term efficiency for long-term sales might be construed as efficiency, but not as it has been understood by specialization theorists. 300

How Has Firing Technology Changed?

Development of Firing Specialists

Although the development of specialists in clay and temper procurement accounted for the earliest changes in the task segmentation of the production process, the development of firing specialists was one of the latest. All older traditional potters probably learned to fire. The development of task segmentation in the production process, however, has revealed that firing skill is not easily acquired, and over the years of this study, a decreasing number of potters have developed this skill. From my own experience, firing requires considerable knowledge and skills that are very different from the remainder of the pottery-making process. When I learned how to fire in 1965, I learned about firewood, how to add wood to the kiln during firing, and what kinds of wood to use for each segment of the process. In addition, I learned the parts of the kiln and the parts of the firing process and which type of wood was best for each stage. Knowing when the process is complete is based on the observational skill of inspecting the pottery inside the kiln, and often potters fire at night to be able to see the red glow of the pottery. Eventually, I fired five kiln loads of pottery, but not without the acquisition of the knowledge and skill of the process and considerable trust of the informants whose pots I fired. With the widespread use of low-skill activities such as molding and making plant pots, firing has become the most skilled task in the production sequence. Firing skill is particularly important when the potter must fire his pots in rainy weather because rain can delay firing, cause vessels to be irreparably damaged, or result in vessels that have to be refired. All of these problems increase firing costs. The development of firing specialists appeared to begin in the 1970s in larger production units where all workers were not equally skilled in the pottery-making process. In one such unit, one potter received extra compensation in addition to his regular salary for firing. By 1994, he was using his own kiln to fire for others, and one of those clients operated a production unit in Mérida. The client purchased raw materials in Ticul, took them to his Mérida workshop, and made the pots there. Then, the vessels were brought to Ticul to be fired. The potter furnished the firewood, the kiln, and the expertise and charged 150 pesos for his service. A second specialist rented his kiln to other potters. This practice antedates 1965 according to informants, and was particularly important for those who were beginning to produce pottery independently but had not yet built a kiln. By the mid- to late 1990s, three other potters had become part-time firing specialists. One, who had learned how to mold vessels from a potter in the government workshop in the 1940s, fired pottery for two production-unit owners. 301

How Has Firing Technology Changed?

In 1997, another traditional potter hired himself out as a firing specialist and fired for a large production unit with two kilns. Another traditional potter was hired temporarily for a production unit in which none of the permanent workers possessed the skill of firing. Changes in Kiln Sizes and Their Distribution among Potters The most dramatic change in firing technology that occurred between 1965 and 1997 was the increase in the size of beehive kilns, as measured by the amount of wood required to fire them. Although size data do not exist for all production units for all visits, the range, mean, and median bundles of firewood required per kiln (Table 9.2) have increased greatly since 1965. A second change in firing technology consisted of the changes in the distribution of kiln sizes among production units (Table 9.2). These data reveal that kilns have not only increased in size from 1965 but their sizes have become more varied. Third, the number of kilns per production unit was also increased. In 1965, the mean number of kilns per production unit was less than one. Only one unit had two kilns and one of these was not used. By 1984, three production units had two kilns and two of these units were owned by the same potter (Table 9.3). By 1997, forty-eight kilns were observed in thirty-nine production units (Table 9.2). Even with missing data from a few production units and for those units without a kiln, a mean of one kiln per production unit was calculated. If kilns per unit were calculated only from those units with kilns, each production unit had a mean of 1.5 kilns. Of those units with kilns, twelve units had two or more kilns. The changes in the size and number of kilns since 1965 are related to several factors. First, one potter explained the increase in the amount of firewood used by saying that the bundles have become smaller since 1965. In order to substantiate this explanation, the 1997 bundles would have to be about one-third the 1965 size if the bundle size was the only variable. Photographs of firewood bundles from 1965 and 1997, however, reveal that the change in bundle size is an insufficient explanation for the increased amount of firewood required per kiln. Further, comparison of photographs of kilns in 1965–1966 and in 1997 reveals that the 1997 kilns were indeed larger. Second, as this chapter has already demonstrated, potters used more wood for firing in 1997 because they no longer knew the burning characteristics of the wood being supplied and may have had to use wood of poor quality. The result is that firing has become less efficient than it was in 1965. 302

How Has Firing Technology Changed?

Table 9.2. The changing sizes of kilns in 1965–1966, 1984, and 1997 as measured by the number of firewood bundles used. Data for 1997 do not include the non-beehive kilns (see Table 9.1).

1965–1966

Number of beehive kilns observed Beehive kilns with size data

1984

1997

23 15

22 7

39 25

5–18

12–50

10–70

<10 bundles per kiln 10–19 bundles per kiln 20–29 bundles per kiln 30–39 bundles per kiln 40–49 bundles per kiln 50–59 bundles per kiln 60–69 bundles per kiln 70 bundles per kiln

3 12 0 0 0 0 0 0

0 5 0 0 2 0 0 0

0 4 3 7 5 2 1 3

Mean bundles of firewood per kiln Median bundles of firewood per kiln

15.2 10

Total range of bundles (tercios) of firewood per kiln

insufficient data insufficient data

39.4 35

Table 9.3. Mean kilns per production unit and production units with two kilns in 1965, 1984, and 1997. Mean per production unit Production units with two kilns

1965

1984

1997

<1 1

<1 3

1 12

Third, diversity of kiln sizes relates to the sizes and increased diversity of vessel shapes produced since 1965. Kilns are built in various sizes that are tied to the type and amount of pottery to be fired. Generally, smaller kilns are for smaller objects and larger kilns are for larger objects. Although large objects obviously may be too large to be fired in small kilns, small objects can be fired in large kilns, but doing so increases the amount of time that a potter must wait before he can obtain returns from his fired pots if he wants to fire a full kiln load. If, however, a potter faces a deadline for an order of small vessels and cannot take the time to fill a large kiln, his per-vessel firing cost will be greater with small vessels than if he filled the kiln because the amount of fuel required for firing will be the same whether or not the kiln is full. So, if a potter has a large kiln that is too large for the amount and size of the pottery to be fired, he may use one of three strategies. First, he may use a smaller kiln of a relative or neighbor so that he can fire a smaller amount of pottery. In 1984, for example, one potter wanted to fire twenty plant pots, but his kiln required forty bundles of firewood to fire it. Since his uncle’s kiln next door was smaller and required half

303

How Has Firing Technology Changed?

that amount, he asked to use it to save the cost of the extra fuel needed for his own kiln. Potters may also follow a second strategy: if they no longer make large pottery, they may tear down the larger kiln and build a smaller one in its place. A third strategy is to build an additional kiln; those potters who produce both large plant pots and small items of pottery recognize the advantages of having two kilns. Consequently, a trend has emerged in which some production units utilize two kilns: one for larger items such as plant pots, and another for smaller vessels such as figurines and tourist pottery. A fourth explanation for the increased number and diversity of kilns relates to the size of some production units. Since some small production units sell their pottery unfired to larger units, the larger units need more kiln space to fire this pottery. Similarly, as owners of some units no longer know how to build a kiln or fire pottery, owners of other production units build an additional kiln and hire firing specialists to fire unfired pottery purchased from potters who do not fire their pottery. The increased size and number of kilns in Ticul provide another surrogate measure for changing production intensity between 1965 and 1997 (Table 9.4). This measure reveals that those changes in increased intensity of production, such as the increased amount of pottery and increases in the sizes and numbers of production units, have an impact on the final production step of making pottery as well as on raw material procurement and vessel fabrication. Conclusion In general, the changes in firing technology since 1965 have paralleled those changes in other production steps in the behavioral chain of pottery making, but at a different rate. First, as a result of the segmentation of tasks in the production sequence, firewood procurement specialists and firing specialists have emerged parallel to the development of specialists in clay and temper mining. Second, although the Mediterranean beehive kiln is still the predominant type of kiln, potters have innovated and expanded their repertoire of kiln types, which range from the more traditional beehive kiln to the capital-intensive gas kiln that was introduced in the modern ceramics industry. As with fabrication techniques, it might seem that these types could be arranged along a simple evolutionary continuum from the high-skill, high-labor-intensive traditional kiln to the less labor-intensive, highcapital-intensive gas kiln of the modern factory. Again, as with the fabrication techniques, such an evolutionary arrangement submerges explanations for the changes and eliminates important insights useful in interpreting ancient production technology and organization. 304

How Has Firing Technology Changed?

Table 9.4. Summary of the changes in firing between 1965 and 1997. Numbers in parentheses are the number of production units with that kiln, if greater than one.

1965–1966

1984

Number of non- 0 3 traditional kilns updraft kiln (2) Kinds of non- traditional kilns gas kiln Mean bundles of 15.2 no data firewood per kiln Median bundles of 10 no data firewood per kiln Mean kilns per <1 <1 production unit Production units 1 3 with two kilns

1997 9 updraft kiln (3) gas kiln (2) cement-block kiln (2) square kiln pot kiln 39.4 35 1 12

Nevertheless, changes in kiln-building technology can be described in a manner similar to that for changes in fabrication techniques. Reasons vary concerning why they were adopted. First, the changes in kiln technology are multidimensional and consist of the replacement of the traditional kiln with a new type of kiln, the enhancement of the traditional kiln, and the addition of totally new kilns. The pot kiln, the small updraft kiln, the cement-block kiln, and the gas kiln have been added, whereas the nesting of a traditional kiln and a cement kiln has enhanced the traditional beehive kiln technology in order to make it more maintenance-free. The gas kiln and the updraft kiln appear to have temporarily replaced the traditional kiln in some production units, but the gas kiln was never really adopted by Ticul potters, and in at least one case, it was eventually abandoned. The updraft kiln, however, did replace the traditional kiln in two instances, but mainly because a smaller amount of pottery could be fired and produced in a more consistent and predictable product stream. Of these processes of change, enhancement has been the most widely adopted category of innovation. These enhancements include the use of cement facing on the outside of the traditional kiln to make it more maintenance-free. The use of cement blocks to construct kilns was not as widely adopted as the cement facing of kilns, but it was more widely adopted than the pot kiln or the small updraft kiln. Second, changes in kiln technology can also be classified by the processes described in the previous chapter (such as borrowing or diffusion, innovation, 305

How Has Firing Technology Changed?

and adoption) that crosscut the categories of replacement, enhancement, and addition. The updraft kiln and the gas kiln, for example, were introduced as the result of borrowing. One gas kiln was constructed for the government-sponsored production unit in the 1970s and another was built for the government production unit in the 1990s, but in each case, the new kiln type was not adopted by other production units. It is unclear what happened to the gas kiln in the 1970s government-sponsored production unit, but the gas kiln built in the 1990s was abandoned by 2002 along with the workshop. The origin of the updraft kiln is unclear, but at least one potter appears to have copied it from his brother. The originator of the updraft kiln, however, said he saw a kiln like it when he visited the United States. The most innovative kiln appears to be the pot kiln that was used only by one production unit in 1997. Third, the total number of kilns, their increased size, and the mean kilns per production unit are also surrogate measures of the increased scale and intensity of production. Nevertheless, other more subtle correlates also exist. First, the increased size of the kilns reflects increases in the size of the pottery produced since 1965. Second, the diversity of kiln sizes and the increased mean number of kilns per production unit reflect the increased diversity of vessels produced. Finally, since the sizes of kilns are measured by the number of bundles of firewood used, the increased use of fuel is also a result of the inefficient use of firewood. Because potters are no longer peasant-agriculturalists (milperos) and thus do not have fields for slash-and-burn agriculture, they no longer have the knowledge of the wood they are using and cannot match the burning characteristics of the wood to a particular stage of firing. As a result, they now have to use more firewood. A fourth way to describe changes in kilns and firing technology is to describe the consequences of those changes. First, because of the maintenance required for traditional kilns, potters have developed new techniques that reduce maintenance for their kilns. Second, increased task specialization separates those who fabricate the pots from those who fire them. These two consequences are interrelated. If innovative techniques to reduce kiln maintenance are not adopted, the maintenance tasks will become more onerous, fewer potters will have the skills to maintain them, and there will be fewer kilns and probably more firing specialists. Some authors (Brumfiel and Earle 1987; Costin 1991; Feinman et al. 1984) view efficiency as a driving force in the development of specialized production. Do the changes in firing from 1965 to 1997 support efficiency as a motivating factor in accepting change? The answer to this question is both yes and no. First, the movement of fuel sources farther and farther from Ticul has resulted in wood types that are unfamiliar to potters. This change, plus the existence of potters who are no longer familiar with firewood types, meant that potters had to use 306

How Has Firing Technology Changed?

more firewood in the firing process and use it less efficiently than it was used previously. Second, the use of cement facing on kilns reduced maintenance costs, prolonged kiln life, and reduced downtime for maintaining or rebuilding kilns. Cement-faced kilns thus could be used continually without pause to maintain them, and this increased production and presumably efficiency. Third, the use of cement facing on traditional kilns and the use of cement blocks to construct kilns eliminated much of the need for highly skilled labor to construct the kilns, and masons, rather than potters, could also be used for the construction and maintenance of kilns. This change increased the amount of time that the potter could use for production because he did not have to take time to maintain his kiln. This change thus probably increased efficiency. Fourth, the development of firewood procurement and firing specialists meant that the potter did not have to obtain the firewood himself, learn the burning characteristics, and then sort the wood accordingly. So the reduction in efficiency in firing is compensated by the decreasing time and effort necessary for the procurement and preparation of fuel for firing, and also for learning the process.

307

Chapter

Conclusion

ten

T

his chapter returns to the fundamental theme of this book concerning the relationship of pottery to social change. It summarizes the changes in the production and distribution of pottery since 1965 and reviews their implications for understanding the evolution of craft specialization. Finally, it assesses the use of ceramics as a surrogate index of social change and answers the question, What does pottery tell archaeologists about social change? To the skeptic who sees ethnoarchaeological data as not relevant for understanding the past, the Ticul data would seem to be a case in point. Now tied to capitalism and a cash economy, the production and distribution of pottery in Ticul since 1965 and the changes that have occurred in them may seem to have no relevance to understanding ancient societies. Such an uncritical dismissal of ethnoarchaeological data, however, reflects a naïve epistemology. No approach to the past exists apart from the present, and all archaeological interpretations are based on analogical reasoning with the known 309

Conclusion

that lies in the present. To refuse to recognize that the present has relevance to the past is to abdicate the commonsense notion that ideas, theories, and indeed all data about the past exist in the present. Archaeologists must take responsibility for being self-aware and use ethnoarchaeological data carefully, critically, and responsibly. How is that done? First, analogies should not come uncritically from isolated and remote “pristine” societies because these societies are still affected by larger complex societies. Some scholars have believed that in order to be relevant to the past, the groups that they study should be untouched, pristine, and uncontaminated by modern civilization. Human societies, however, have always been changing; even the most remote and pristine societies still have been affected by state-level societies (Headland 1997). The Lacandon, for example, traditionally regarded as the emblematic pristine and “untouched” Maya ethnic group, was the result of migration, acculturation, and assimilation (Palka 2005). Second, although the present is obviously different from the past, it is possible to sort out those patterns that are a product of globalization, capitalism, and a cash economy from those that are more relevant to ancient societies. Even so, the rapid social and cultural change going on today in the world’s societies provides a unique opportunity for ethnoarchaeologists to understand the processes of change and ascertain the forces that drive social and technological evolution. From the Ticul data, it is clear that contact with the modern world can drastically change pottery production and destroy traditional uses of pottery. On the positive side, however, it is also clear that demand is the driver of craft production and this demand can be understood more deeply. This perspective is, of course, hardly new. Modern corporations and advertising agencies recognize that without demand for a product, there are no sales. Similarly, the Ticul data reveal the importance of transportation infrastructure in articulating demand with production. Where the location of demand is separated from the location of production by hundreds of kilometers, the development of an infrastructure that links demand with production is vital. In Yucatán, this infrastructure consisted of highways and the accessibility of trucks to transport pottery to market. In antiquity, however, such material infrastructure was limited, but the roads of the ancient Maya probably did have some important economic role that facilitated transport of fragile pottery and tied centers to surrounding areas of production. Most important, however, the ancient infrastructure that distributed pottery was more social than material. Those patterns that the past and the present have in common consist of evolutionary, technological, and social processes that transcend time and space. One example of these kinds of processes consists of those that are directly tied to the 310

Conclusion

technology of production. The unique physical characteristics of clay minerals, for example, directly affect the behavioral chain (chaîne opératoire) of making pottery and have universal implications for making processual analogies about ancient ceramics (Arnold 1975a, 1985, 1993). Even though it is obvious that pottery production involves many more aspects that are cognitive, social, symbolic, and behavioral and are relative to space and time, the universal production processes tied to the molecular structure of clay minerals provide one kind of analogical foundation for making inferences about ancient ceramic production. Finally, another method of finding analogies to the past consists of using a comparative approach. With the data presented here, I have used comparison through time to identify long-term processes and patterns. Specifically, cultural evolution links the present with the past, and although the conditions of the present are different from those of the past, common evolutionary principles do exist. Identifying these common principles may be difficult, but they exist nevertheless. Summary of Changes Between 1965 and 1997, dramatic social changes occurred in Yucatán. During this time, the economy moved from one largely rooted in traditional subsistence agriculture to one largely based on cash. The Mexican government expanded and improved its highway infrastructure, and the resorts along the Maya Riviera became some of the most important tourist destinations in the Western Hemisphere. During this dynamic period of massive social change, great shifts occurred in the demand for pottery, and potters responded with a set of adjustments like those Schiffer (2005) called a “cascade of innovations.” Some of these adjustments were technological, whereas others were social and organizational. The first shift consisted of an almost total replacement of vessel shapes. This replacement reflected changing consumers of pottery. In the late 1960s, potters produced coin banks and vessels for carrying and storing water that they sold to the local Yucatec Maya population. By the early 1970s, however, piped water was widespread, which eliminated the need for water-carrying and water-storage vessels. As a result, the production of these vessels largely ceased. By the mid-1970s, the resort of Cancún was established and potters adjusted to this new consuming population by making two kinds of vessels: small vessels that symbolized and materialized the social memory of tourists’ visit to the land of the ancient Maya and plant pots to adorn the patios, lobbies, and restaurants of tourist hotels. By 1984, both kinds of vessels were potters’ predominant products. 311

Conclusion

Second, the distribution of pottery has shifted. Between 1965 and 1970, Ticul potters had easy access to consumers because they sold their wares to the local Yucatec populations. After the late 1970s, however, consumers were more distant from producers than they had been in 1965. No public transportation existed to move large numbers of pots to distant markets; thus, consumers were out of reach for most potters except those who had the capital to purchase a vehicle for transport. The railroad that formerly transported pottery did not go to the new consuming populations. Consequently, potters no longer distributed their vessels themselves but sold them to brokers who resold them to consumers or to other brokers or retailers. This change selected for those individuals with knowledge, skills, and capital to engage the Spanish-speaking world and resulted in the emergence of a group of brokers upon whom potters depended to sell their pottery. Some of these brokers were other Ticul potters who owned large production units, but others were not from Ticul at all. The third shift consisted of an overall increase in the number of potters and in the number of production units. Although the numbers fluctuated from observation period to observation period and the number of data points is limited, the correlation of the data and the trend line (R2) is moderately strong. A fourth shift involved the development of a more complex organization in which different tasks in the production and distribution of pottery were accomplished by different specialists. At one time, individual potters obtained raw materials and formed, decorated, and fired their own pottery. By 1997, however, the potter no longer controlled the entire sequence of production, and the behavioral chain of making pottery had become highly segmented. Potters formed and dried their vessels themselves, but they bought their raw materials from specialists, and many sold their products to a broker to be fired. With the development of painted decorations that imitated ancient Maya designs, painting specialists emerged as separate from those who made pottery, and potters supplied them with fired blanks to paint. With the increased segmentation of tasks, each segment of the behavioral chain followed different, but parallel, trajectories of specialization. The traditional potter who possessed the skills and knowledge to complete the entire pottery-making process was replaced by separate individuals performing highly specialized tasks. Over time, fewer and fewer potters had knowledge of the entire process. All of these changes, however, occurred at different rates. The consuming population, vessel shapes, and the patterns of distribution have changed relatively quickly, whereas the community of potters, the paste, and the technology of production have changed more slowly. Decoration was the most flexible and 312

Conclusion

most subject to changing demand, followed close behind by changes in vessel shapes. Even though the changes in procuring raw materials were a consequence of an increased demand for pottery, the raw materials and their sources did not change unless the sources became exhausted or over-exploited. Firing technology changed to cut kiln maintenance and prolong kiln life and through the use of nontraditional building materials to construct kilns. New types of kilns were also developed, and the kinds of firewood changed. Kilns have gotten larger over time, and the number of kilns per production unit has increased. Changes in design and vessel shape thus better reflect social change than changes in paste or fabrication technology, although changes in production intensity are reflected in changes in the size and number of kilns. Pottery produced for the annual Day of the Dead rituals, however, has remained unchanged since at least 1951, when Raymond Thompson (1958) visited Ticul. These rituals require earthenware vessels and this ideology translates into a strong seasonal demand for ceramic food bowls, whistles, incense burners, candlesticks, and flowerpots. Formerly, cooking pots were also produced and used in the rituals, but when metal cooking pots replaced ceramic ones, cooking pots were no longer made because they were difficult to produce, requiring a different temper and firing procedure, and the metal cooking pots were more durable. Nevertheless, the value of using ceramic vessels for the Day of the Dead rituals continues in every Maya household and provides a strong seasonal impetus for full-time and part-time potters to produce pottery. Ideology and religious values thus provide a strong conservative force in the production of pottery and in the kinds of vessel shapes made, especially those used for ritual purposes. Why has pottery production in Ticul been so successful and resilient given the great social changes that have occurred during the last fifty years? One answer is diversity. Ticul has a diversity of fabrication techniques, decorative techniques, paste types, and vessel shapes, which have provided flexibility for the selective forces of consumer demand. I first made this point at a conference in 1981 (Arnold 1989b) by comparing the number of fabrication techniques, pastes, and decorative techniques in the different communities in which I worked in Latin America. Those communities with the greatest diversity had the most potters and were the most viable. Consequently, if a community makes only tortilla griddles with only one paste, then the replacement of those griddles with metal ones will result in the disappearance of pottery making in that community. The town of Mixco, Guatemala, for example, had a long history of making pottery (Reina and Hill 1978:42–44), and in the summer of 1970, I found four potters making tortilla griddles (Arnold 1978b:337–345). They used an untempered paste with molds embedded in the floor. When Reina and Hill (1978:42) went to Mixco 313

Conclusion

in 1973, however, they found no potters there and called the disappearance an instance of “technological death.” In comparing Mixco (and other communities) with Ticul, it is clear that diversity was a significant factor in the survival of pottery production. The research of the original 1981 paper, however, was largely based on my work in Ticul from 1965 to 1970 and thus occurred before much of the research presented in this book was done (1984, 1988, 1994, 1997). Nevertheless, the importance of the principle of diversity proved to be a remarkably accurate prediction of the evolutionary success of Ticul pottery production and the demise of pottery making in Mama, Akil, and Tepakán. In 1970, I could not have predicted the development of Cancún as a primary market for Ticul potters, but the diversity of techniques, decoration, and shapes pre-adapted Ticul potters to respond to changing demand for their products, which resulted in the increase in the number of potters and number of production units. The Conservative Nature of Household Production In spite of changes in raw materials, procurement practices, vessel shapes, firing, decoration, demand, and market that have occurred since the late 1960s, pottery production is still household-based. This pattern persists because of the time necessary to learn the craft. Pottery making requires learning a series of motor habit patterns and learning the semantic categories of raw materials and those used in production processes (such as those of vessel shapes, their parts, and their measurements). These categories and the somatically embedded motor habit patterns are reinforced by actual engagement with the raw materials and pottery products, which provides feedback of sensory information (largely through sight and touch) that influences the choices potters make. As one might expect, the embodied motor patterns are largely unconscious and thus more resistant to change than the transmission of patterns of decoration. Motor habits require repetition over a period of time in order to be used effectively. They are more difficult to learn than the cognitive knowledge; thus, learning such patterns may require as much as several years. Consequently, the motor habits involved in the traditional fabrication technology (modified coiling) are a conservative force on the craft and its organization and favor learning in the household. Long training in the household favors learning by the young because of the amount of time they spend there. Given the length of the learning process, the advantages of learning during childhood, and the advantage of learning in the household, the household is not only the most important social and economic unit of production, but also the 314

Conclusion

most significant social unit responsible for the perpetuation of the craft. Indeed, the transmission of the knowledge and skill of making pottery can be described by the same behaviors responsible for the composition, location, and perpetuation of the group in which the learning takes place (Arnold 1989a:175–181). Since the household and its continuity are critical for the perpetuation of the society through procreation and socialization of new members, it is not surprising that the transmission of the craft can be described by the same processes that perpetuate the household and maintain its integrity over time. These patterns explain why learning patterns are still household- and kin‑based in spite of the orientation of the craft toward the production of tourist pottery. These points have been made before but greatly criticized. In the late 1960s and 1970s, Deetz (1965), Longacre (1970), and Hill (1970) argued that a kin‑based model is instrumental in transmitting ceramic style from generation to generation (see Hayden and Cannon 1984a and Kramer 1985 for a review). Several scholars challenged this model, and one such challenge concerned the nature of the kin‑based model itself (Allan and Richardson 1971; Hayden and Cannon 1984a; Stanislawski 1977; Stanislawski and Stanislawski 1978). Arnold (1989a), however, has shown that kin-based models indeed have validity in explaining the learning of the craft because actual residence proximity was used as a measure rather than abstract rules about what people say about how they behave or where they should live. Although some aspects of modern pottery production are not directly applicable to the past, it appears that the processes involved in the acquisition of household personnel, the perpetuation of the craft, and the residence locations of potters are not among them. Rather, when production is tied to the household, the perpetuation of the craft occurs through household processes in spite of massive social change. Indeed, Underhill (2003) also found that production in the household persisted in rural China in spite of social change. As pottery production continues to evolve in Ticul, kin‑based patterns of household composition and recruitment may no longer play a role in transmitting the craft from generation to generation (Arnold 1999) because the training is too time-intensive with few immediate returns. If learning the craft moves out of the household, learning traditional patterns of fabrication may be disrupted and change. This hypothetical projection illustrates Shennan’s (2000) application of the evolutionary notion of “descent with modification.” When an aspect of the descent mechanism changes, which in this case consists of the learning context, significant changes will occur in the technology. In such cases, fabrication techniques that required little skill and learning time (such as slip casting and vertical half molding) will have a selective advantage. The 315

Conclusion

limited skill required by vertical half molding means that increased demand for mold-made objects can be met quickly by drawing unskilled workers into the craft without a lengthy process of learning, like that required for modified coiling or modeling. Mold-made figurines, for example, can be produced easily by any type of production organization without a lengthy process of skill development. This explanation might explain the development of mold-made pottery in Tula (Bey 2005:16) and among the Moche and Chimú of ancient Peru. Molding, however, increases the space required for production (Arnold 1999) and creates a larger and more archaeologically visible spatial footprint, such as that found in the production sites of Moche pottery (Russell et al. 1998; Uceda and Armas 1998). Equally important, however, are the selection mechanisms that operate on the population of potters and constrain the number of potters below the level of population growth at large. These mechanisms can be macro factors that eliminate potters from the pottery-making population, like disease, military conscription, and forced labor; or they may involve the presence of pottery-making infrastructure, the amount of time a learner spends in a production unit, and conflict of roles between making pottery and birth, nursing, child care, and household responsibilities (Arnold 1985:100–108). Poverty, lack of education, and lack of economic opportunities also can be selective mechanisms. Changes in Production Units

During the period of this study, the size, composition, and location of the production units changed. Through it all, however, household organization did not disappear but was flexible enough to permit an increase in the number of potters within some units. Six households evolved into larger production units. Larger numbers of smaller units also developed. Small units have an important structural relationship with large units by providing them with skilled potters. Large units may also hire non-potters who learn the craft on the job, although their work is generally limited to a few highly specialized tasks. Over time, however, even large units are still family- and household-based, with an increased use of wage laborers that include non-relatives as well as members of the extended family. This trend is similar to the observation that Costin made from Kleinberg’s (1979) work with village pottery production in Japan, that as production units grow in size, distant, fictive, and adopted relatives are recruited first, and then non-relatives become workers as the production unit grows further (Costin 1991:15). In Ticul, the composition of the production units generally reflects these types of kin. Over time, however, all production units do not grow, but their composition does change. Except for an increase in 316

Conclusion

the amount and percentage of non-relatives, trend lines of all of these changes have a low correlation (R2) with the data. Nevertheless, an increase in the number of types of lineal relatives and a decrease in the number and percentage of potters who are affinal kin, women, daughters, and wives have the highest R2, even though these correlations are still not strong. Although the trends in the composition of production units over time parallel the changes in the composition of production units in Japan (Costin 1991; Kleinberg 1979), comparison of the size and number of units across the thirtytwo years of this study reveals an important relationship between small units and large units. This relationship can be expressed in a power-law (log-log) relationship between the number of potters in each production unit and the number of production units. These relationships reflect the kind of scale-free self-organizing system that is found in a wide range of phenomena (Bentley and Maschner 2001; Bentley et al. 2004; Bentley and Shennan 2003). Why does this relationship exist between the number and sizes of production units? One explanation appears to be related to a “rich-get-richer” explanation believed to be responsible for the patterns found in social network analysis (Heyman 2006:605). Owners of larger production units benefit from access to capital from increased production and from access to a vehicle for transporting their pottery. Consequently, they hire more workers who generate more capital, and this increase provides deviation-amplifying feedback for increased production-unit size. Household-trained potters are attracted to large production units because they receive a consistent wage and do not need to find clients to buy their pottery. As some production units increase in size, smaller production units cannot compete because owners have trouble finding clients to buy their pots, receive a price for them that is too low, may never receive returns for pots that they have sold on credit, or have limited access to clay delivery. Consequently, owners of small production units may abandon the craft. Other small units, however, continue to make pottery and sell it themselves or have found reliable brokers to buy it. By way of contrast, some of the workers employed by larger production units eventually leave those large units to form their own. This power-law relationship appears to be independent of the great social changes that have occurred in Ticul and the changes in the demand, market, shapes, and technology of the pottery. Rather, the power law appears to express an ongoing principle of technological evolution. In a common social system of a community of potters, a combination of large and small production units will persist through time because all units participate in the same economic system. Small units of production will not necessarily evolve into larger units. 317

Conclusion

Consequently, although production intensity has increased over time, it is not because of the uniform evolution of production-unit size. Some units have gotten larger, but most have not. Rather, the number of the smallest units continues to grow. Small household-based production thus can achieve increased intensity without growth in all production units because of the development of increased specialization. Finally, even though the craft is household-based and members of the nuclear family serve in the production units, no clear trend exists in the growth of the number and percentage of family members through time. Rather, the clearest trend in growth is the number of laborers who are not relatives of the production-unit owner. With the decline in the number and percentage of female relatives over time, it is clear that laborers are the source of most of the growth in the number of potters since 1965. Rather than replacing family members (except for females) in production units, non-relative laborers simply augment the number of relatives who are already working in the production unit. Efficiency The main question about using efficiency as a causal mechanism concerns the kind of efficiency to address (Costin 2001). Among the different kinds of efficiencies are space, time, effort, and personnel. The Ticul data suggest that efficiency, as expressed in fabrication time, is not necessarily an important consideration for preindustrial potters who have a choice between different forming technologies (e.g., Zubrow 1992:118). Efficiency is not necessarily the driver that results in a change in fabrication technique. The Ticul data also reveal that efficiency is not necessarily the prime mover for changes in production organization and production intensity. Efficiency, Costin (1991) argued, is a function of the technology used and the level of production unit output; it determines the scale of production by independent specialists. If per-unit costs can be lowered through sharing expensive technology or by dividing tasks among many workers, Costin (1991) proposed, productionunit size will rise to take advantage of economies of scale. To Costin, producers and consumers are best served by a production system that minimizes production and transaction costs. Furthermore, larger units with greater output may be able to exploit certain marketing strategies (Costin 1991:16). The Ticul data do not clearly fit Costin’s scenario. First, all production units do not grow. Yet, the overall intensity of production has increased. The mean number of potters per production unit did not change much until recently. Rather, it appears that specialization is itself the means by which potters have 318

Conclusion

increased production intensity as measured in production output rather than the amount of time worked, because potters have eliminated the time necessary to move from task to task. Because of the distance to their clay and temper resources, for example, buying raw materials from specialists can result in considerable savings in time, perhaps as much as one to two days a week. Increased specialization thus in itself increases production intensity because task segmentation increases output by eliminating the time required to move from task to task. Other changes in efficiency have occurred, such as the ball-bearing turntable. Its acceptance, however, took place within the context of existing motor habits and work positions. The ball-bearing turntable was easier to use than the traditional turntable because it required less effort and maintenance. Its initial acceptance, however, did not occur because it was easier, but because it was introduced by a trusted family member. The social dimension of the choice to adopt initially was just as important, if not more important, than the technical reasons for that choice (Ralph and Arnold 1988). The choice to adopt thus involved both social and technical explanations. Equally important, however, was the association of the ball-bearing turntable with the increased demand for pottery. In the late 1960s, there was no tourist market for pottery, no resort in Cancún, and no ball-bearing turntable. Production was oriented exclusively to a local Yucatec population. By 1984, however, most of the demand came from Cancún, and roughly 50 percent of the potters were using the ball-bearing turntable. By 1997, the ball-bearing turntable appeared to be totally adopted. Increased production thus also appears to be associated with the adoption of the ball-bearing turntable. Paradigms: Social Change and Specialization What, then, does this study reveal about the development of specialization? Does the evolution of ceramic production in the modern world appear to follow a trajectory similar to that believed to occur in antiquity? Indeed it does. First, specialization is an evolutionary process and not an event or series of events. It began in remote antiquity and continues into the present. In the case of Ticul, the evolution of ceramic production from 1965 to 1997 is a continuation of a process that began at least 1,000 years ago when potters discovered the superior quality of the clays from Yo’ K’at compared to other clays of the area and the advantages of substituting sak lu’um (palygorskite) for part of their temper. Palygorskite displaced some of the calcite of the temper, and even though it was a clay mineral, it functioned as a non-plastic and absorbed water from the paste. It probably lessened vessel shrinkage and reduced the amount of calcination during firing. 319

Conclusion

Potters perceived these characteristics as vessels that did not crack and break as much as those made without sak lu’um (Arnold 1971). Second, social change provides one of the stimuli for increased specialization and creates new circumstances to which potters must adapt. Such changes, however, may or may not be reflected in the ceramic assemblage. The replacement of clay cooking pots with metal ones, the development of piped water, the expansion of the highway infrastructure, the development of the tourist industry, and the demand for pottery in Cancún were all social changes to which potters had to adapt if they were to continue making pottery. The cessation of the production of clay cooking pots and vessels for carrying and storing water reflected the industrialization and modernization of Mexico. Furthermore, the massive production of plant pots and pots with ancient Maya designs reflected the rise of the tourist industry. Some potters adjusted to making these new kinds of vessels and others did not. Those who did adjust became more specialized, whether they did so by developing task specialization, by developing product specialization, or by working in a production unit outside of their households. Those who did not adjust abandoned the craft completely. Third, the evolution of specialization in the present and in the past appears to share certain processual similarities. Some decry the effect of globalization on traditional potters, but globalization continues an evolutionary process in which groups of specialists and their interconnections have expanded to include nationstates and multinational corporations. These interconnections reflect intercultural relationships rather than just intracultural ones. In this sense, globalization also can be seen as ancient, beginning millennia ago when cultural influences and trade expanded influence beyond the borders of ancient states (Lawler 2007). Fourth, the Ticul data reveal that the evolution of ceramic production can follow two overall trajectories. The first consists of the development of increasing specialization that segments the tasks of the behavioral chain of production and distribution. In Ticul, raw material procurement specialists developed first and emerged largely as a consequence of the distance to the resources of clay and temper sources. Similarly, firing by specialists was separated from the rest of the production process as some potters began selling their unfired pottery to others for firing. Some production units also segmented painting from the rest of the potterymaking process, and this kind of specialization probably occurred repeatedly in the past. Since design layout and painting are such highly specialized tasks and so different from making pottery, it is likely that the producers of painted vases among the ancient Maya were painters rather than potters. Potters could have supplied blanks to elite painters as a form of tribute. Painters required none of the 320

Conclusion

infrastructure necessary for making pottery and thus had a much smaller spatial footprint. The segmentation of painting from pottery making also occurred among the ancient Greeks (Clark et al. 2002:2; Oakley et al. 1997) and perhaps among the Moche, Nazca, and Huari potters in ancient Peru. For the Moche, Nazca, and Huari cultures, painting skill appears to be so divergent from the production of pots that more training and skill were necessary for painting the vessel than for fabricating it. This was particularly true for the mold-made vessels of the Moche, except perhaps for attaching the spout (Donnan 1965). Chimú potters, however, appeared to have reduced the amount of skill and effort in making their pottery from that of their Moche predecessors by investing no labor in painting it. The Chimú continued the technique of using molds to make their pottery, but they incorporated designs into the molds to produce designs in relief on the molded vessels and then fired them in a reducing atmosphere to make them black. The Chimú thus appeared to reduce their energy inputs for pottery from those of the Moche in order to increase production output, reduce the number of skilled potters, adjust to the lack of skilled painters, or adjust to the existence of a limited number of skilled potters. A second trajectory of technological evolution consists of vertical integration. This trajectory requires more resources and management skill than specialization because the production unit must acquire raw material sources, manage the fabrication of the pottery, and then manage distribution of it to sell directly to consumers. Only one Ticul potter successfully followed this trajectory because it required capital to purchase a clay source and to buy or rent property for stores in locations distant from Ticul. I suspect, however, that vertical integration is a very fragile and vulnerable kind of production organization and is subject to all kinds of perturbations not found in the trajectory of specialization. Personnel problems, for example, require much more time and management skill than one potter may possess. Indeed, the rarity of vertical integration in Ticul suggests that this trajectory is more a unique product of capitalism and individual entrepreneurial skill than specialization is. Nevertheless, there are other examples of vertical integration; an ancient Roman winery in France employed potters (Holden 2007) to produce vessels on site presumably to ferment, store, and transport wine. Thus, operations like this might be one kind of vertical integration that existed in antiquity. One way to get around the problems of vertical integration is to use kinship as an organizing principle, and the potter who developed integrative strategy in Ticul organized it on the basis of kinship. His wife and several of his ten children were potters, and he placed one in charge of each component of the operation. 321

Conclusion

He was in charge of the production unit along the highway, and his wife was in charge of the production unit in his house. One son was in charge of the store near Cancún. Another son was in charge of painting and still another son obtained clay for all the family production units from his father’s source. This son had his own production unit but sold pottery to his father. Although one non-family member was in charge of the store near Dzitbalché, he was an exception and was a weak link in the organization that may have led to the closing of the store by November 2002. Fifth, pottery making and the growth of ceramic specialization need to be understood holistically. The social dimension cannot be understood unless the relationship between the potters and their environment, raw materials, and the technology of production is also understood. Similarly, archaeological reconstructions of ceramic specialization should not be abstracted from the ceramic technology and environment because different production and extraction techniques have different effects on the social organization of production. Consequently, theory concerning the evolution of ceramic specialization should be understood in an empirical context of how people make pots, the resources that they use, and their spatial and environmental setting. Otherwise, a disconnection will continue between archaeological theories of ceramic specialization and the realities of ceramic ethnoarchaeology, limiting our understanding of the past. Social relations are not independent of the environment and technology, nor can such relationships be described as if variations in the environment and the technology do not exist or are constant from case to case or from craft to craft. Evaluation of Costin’s Parameters

One of the stated goals of this book is to evaluate Costin’s parameters of specialization and apply them to the evolution of ceramic production in Ticul. Do Costin’s parameters provide deeper insight into the details of the evolution of ceramic production and the process of specialization in Ticul? Indeed they do. Her parameters are very salient for the description and interpretation of ceramic production and the evolution of specialization. They have changed, however, and Costin (2001, 2005) has made several much-needed refinements in terminology, substituting “demand” for “context,” “spatial organization” for “concentration,” and “work group composition” for “scale.” Further refinements are necessary and one should be made in the intensity parameter. Costin’s intensity parameter consists of the amount of time that potters spend on their craft. The lower end of the intensity range consists of part-time specialization in which craft production supplements subsistence, and the other end of the range is full-time specialization 322

Conclusion

in which potters exchange their vessels for all required goods and services. The change from part-time to full-time specialists, however, is hard to identify, even in contemporary Ticul. Potters go back and forth from part-time specialization to full-time specialization, depending on the demand for pottery, and potters or members of their households often supplement their craft with subsistence agriculture, work as a tricycle taxi driver, or engage in other crafts (such as women’s weaving palm fronds or making huipils). Further, a number of potters work only seasonally, making pottery in the months preceding the Day of the Dead rituals, and this pattern occurs in the communities of Akil and Mama elsewhere in Yucatán. Assessing a meaning of the time spent making pottery and insisting on a rigid division between part-time and full-time ceramic production as an operational meaning of “intensity” in an archaeological context thus are impossible without some material index or surrogate. But what might such a surrogate be? Intensity is a critical factor in the evolutionary development of ceramic production, and from this study the intensity of production might best be described as the total output of pottery as measured by material surrogates, such as the number of vessels produced, the amount of clay and temper used, or the increased size of the kilns and the number of kilns per production unit. This material approach eliminates the materially elusive transition from part-time to full-time in the archaeological record. Nevertheless, this transition is important although it may be unknowable archaeologically. Probably the most obvious inference concerning this transition can be made using the nature of the ancient climate that influences the scheduling of production and the amount of covered space available to dry pottery in inclement weather (Arnold 1975a, 1985, 1993). Costin’s third parameter, scale, is defined as two interrelated variables: the size of the production unit and the principles of labor recruitment. This parameter has gone through two terminological changes: first as “constitution” (Costin 2001) and then as “work group composition” (Costin 2005). Size consists of the number of potters per unit, and labor recruitment consists of the composition of the unit and the way in which potential potters are acquired by those units. As Pool and Bey (2007) have suggested, however, the size of the production unit should be separated from its composition and how it acquires new potters. Although the composition of the unit and the acquisition of its personnel are interrelated, the size of a production unit is not necessarily related to its composition. Rather, the size of the production unit varies greatly across the population of potters, even in units with similar patterns of composition. Further, production units do not grow uniformly. Rather, the number of units and their size in the community have a systemic power-law relationship. 323

Conclusion

Given the insights from contemporary social and technological evolution, what material markers might be helpful in the development of specialized production in Ticul? It does appear that ceramic vessels became more homogeneous when brokers became the primary means of distributing pottery. When potters were selling their pottery directly to individual consumers, their vessels did not have to be uniform and potters provided many variants of traditional vessels based on hand measurements. When brokers bought many vessels at a time from single potters, however, they could specify what they wanted from the potters, and they wanted uniform vessels. Why did brokers want uniform vessels? In the case of flowerpots, hotel owners wanted uniform vessels to decorate their lobbies, hallways, and restaurants. In the case of vessels painted with ancient Maya designs, dimensional homogeneity was necessary because the design could be formatted in less time on uniform vessels than on vessels that were not uniform. Increased dimensional homogeneity thus was related to changing patterns of demand and distribution that involved middlemen buying many vessels from a single production unit. How potters produced this standardization, however, varied according to vessel shape. With plant pots, vessels were carefully measured with a ruler or a precut stick, taking the amount of shrinkage into consideration. With small vessels, however, potters simply used a preexisting fabrication technique (molding) to make uniform vessels. Molding had been adopted decades previously in order to fabricate vessels that could not be made easily with the traditional technique. With the dimensional homogeneity of both kinds of vessels, Ticul potters practiced intentional rather than accidental (or mechanical) standardization to meet broker requests (Berg 2004; Costin 2005). Technological Choice

A final paradigm used in this book was the notion of technological choice. Potters always have a choice in the way they make pots. Choices, however, are embedded in a contextual hierarchy. One can always narrow the analytical focus and make a point for the social dimensions of technological choice, but this narrowness is a product of analysis rather than cultural reality. Potters may make technological choices for what may appear to be technical reasons, but in reality, such choices are also made for social reasons even though the choices may have a technical basis. Potters use clay from Yo’ K’at and temper from Yo’ Sah Kab, for example, because they have traditionally used these sources and these locations have a sense of place for the potters, and the raw materials from these locations are technologically superior to materials from other locations nearby. Potters, however, make the choice of clays based on tradition and a sense of place but 324

Conclusion

do not always mention the superiority of the raw materials from these locations. Similarly, the ball-bearing turntable was accepted by some potters because of kin relations with one of the potters who tried to sell it, but it was ultimately adopted by all potters because it was easier to use than the traditional turntable. From the perspective of technical choice, the Ticul potter has five choices to fabricate pottery, but all of these are not equally viable and are constrained by both technical and social factors. Forming devices are subject to the limitations of the raw materials used, the forming technologies themselves, the organization of labor, and the demand for the vessels. Potters are constantly negotiating and renegotiating new solutions to technical problems that arise because some new shapes cannot be formed as effectively with a particular fabrication technology. In some cases, these solutions are technical. For example, a potter may make a homogeneous vessel with two techniques (molding and the wheel) in order to avoid the problem of sagging. In other cases, however, the solutions are social and involve the reorganization of the labor, such as bringing unskilled workers into the production process. The existence of multiple forming techniques in Ticul (whether or not they were adopted) suggests that a complex interdependence exists among the limitations of the raw materials, the techniques of fabrication, the organization of labor, and the demand of the market. Not all techniques can be used to make every shape. Rather, under certain conditions, certain techniques are better suited to produce certain shapes. Molds, for example, are best used to make small vessels less than twenty centimeters along their longest dimension. Molds are also ideal for making non-circular forms that cannot be made in any other way, such as figurines, but molds cannot be used to make every kind of uniform vessel. They can be used to make small vessels, but they cannot be used to make large vessels because the highly plastic Ticul clay causes these vessels to sag and crack. Consequently, large vessels are best made in successive stages, using modified coiling with drying periods between the stages, and then measured in order to ensure uniformity. There may be some reasons to analytically separate social dimensions of choice from technical ones, but in reality, humans are making choices for a variety of reasons that are interrelated and constrained by a combination of social, technical, and environmental factors. The subparts of culture are, after all, integrated and tied together, and emphasizing the distinction between technical and social choices is artificial. Humans have varied reasons for making the choices that they do, and these reasons are tied to the ideology, social structure, and technology of a culture as well as to the environment in which they occur. It is thus possible to understand those factors that are responsible for the continuity of the craft and those responsible for its great changes. The factors 325

Conclusion

inhibiting change are religious beliefs, raw materials, fabrication techniques, and household-based production. Those that stimulate change are demand, market, and methods of distribution. Although some aspects of the firing technology changed, the increases in kiln size and the increased number of kilns per production unit provide one kind of materialization of increased production intensity over time. This book describes the changes in pottery production and distribution with a broad brush that shows general patterns. But there is also a more social, more personal side of these changes in which production units develop, change, or abandon making pottery. These changes parallel the changes described here but reveal a different dimension to this change and provide a name and a face for each person affected by it. This perspective is the subject of a second monograph about ceramic change in Ticul.

326

References Cited

Alisky, Marvin. 1980. The Relations of the State of Yucatán and the Federal Government of Mexico, 1823–1978. In Yucatán: A World Apart, ed. E. H. Moseley and Edward D. Terry, 245–263. University of Alabama Press, Tuscaloosa. Allan, W. L., and J. B. Richardson III. 1971. The Reconstruction of Kinship from Archae‑ ological Data: The Concepts, the Methods, and the Feasibility. American Antiquity 36:41–53. Arnold, Dean E. 1967a. Maya Blue: A New Perspective. M.A. Thesis in Anthropology, University of Illinois, Urbana. ———. 1967b. Sak lu’um in Maya Culture: And Its Possible Relation to Maya Blue. University of Illinois, Department of Anthropology, Research Reports No. 2. Urbana. ———. 1971. Ethnomineralogy of Ticul, Yucatán Potters: Etics and Emics. American Antiquity 36:20–40. ———. 1972a. Mineralogical Analyses of Ceramic Materials from Quinua, Department of Ayacucho, Peru. Archaeometry 14:93–102. ———. 1972b. Native Pottery Making in Quinua, Peru. Anthropos 67:858–872.

327

References Cited

———. 1975a. Ceramic Ecology in the Ayacucho Basin, Peru: Implications for Prehis‑ tory. Current Anthropology 16(4):185–203. ———. 1975b. Principles of Paste Analysis: A Preliminary Formulation. Journal of the Steward Anthropological Society 6:33–47 (Fall 1974). ———. 1975c. Reply to Haaland and Browman. Current Anthropology 16:637–640. ———. 1978a. Ceramic Variability, Environment and Culture History among the Pokom in the Valley of Guatemala. In Spatial Organization of Culture, ed. Ian Hod‑ der, 39–59. Duckworth, London. ———. 1978b. The Ethnography of Pottery Making in the Valley of Guatemala. In The Ceramics of Kaminaljuyu, ed. Ronald K. Wetherington, 327–400. Pennsylvania State University Press, University Park. ———. 1983. Design Structure and Community Organization in Quinua, Peru. In Structure and Cognition in Art, ed. Dorothy Washburn, 56–73. Cambridge Univer‑ sity Press, Cambridge. ———. 1984. Social Interaction and Ceramic Design: Community‑wide Correlates in Quinua, Peru. In Pots and Potters: Current Approaches in Ceramic Archaeology, ed. Prudence M. Rice, 133–161. Monograph XXIV. Institute of Archaeology, Univer‑ sity of California, Los Angeles. ———. 1985. Ceramic Theory and Cultural Process. Cambridge University Press, Cambridge. ———. 1987. Maya Pottery after Twenty Years: Archaeological Implications. In Maya Ceramics: Papers from the 1985 Maya Ceramics Conference, ed. P. M. Rice and R. J. Sharer. BAR International Series 345, Part I, 545–561, British Archaeological Reports, Oxford. ———. 1989a. Patterns of Learning, Residence and Descent among Potters in Ticul, Yucatán, Mexico. In Archaeological Approaches to Cultural Identity, ed. S. Shennan, 174–184. George Allen and Unwin, London. ———. 1989b. Technological Diversity and Evolutionary Viability: A Comparison of Contemporary Pottery-making Technologies in Guatemala, Peru, and Mexico. In Ceramic Ecology, 1988: Current Research on Ceramic Materials, ed. Charles C. Kolb, 29–59. BAR International Series 513, British Archaeological Reports, Oxford. ———. 1991. Ethnoarchaeology and Investigations of Ceramic Production and Exchange: Can We Go beyond Cautionary Tales? In The Legacy of Anna O. Shepard, ed. R. L. Bishop and F. W. Lange, 321–345. University Press of Colorado, Boulder. ———. 1993. Ecology of Ceramic Production in an Andean Community. Cambridge Uni‑ versity Press, Cambridge. ———. 1997. Field Work in Mexico. La Tinaja: A Newsletter of Archaeological Ceramics 11(1):1–3. ———. 1998. Ceramic Ethnoarchaeology at Ticul, Yucatán, Mexico. The Society for Archaeological Sciences Bulletin 21(1–2):6–7. ———. 1999. Advantages and Disadvantages of Vertical-half Molding Technology: Implications for Production Organization. In Pottery and People: A Dynamic Interaction, ed. J. M. Skibo and G. M. Feinman, 50–80. University of Utah Press, Salt Lake City.

328

References Cited

———. 2000. Does the Standardization of Ceramic Pastes Really Mean Specialization? Journal of Archaeological Method and Theory 7:333–375. ———. 2003. Cambios de Aspectos Sociales de la Población de los Alfareros en Ticul, Yucatán, 1965–1997. Los Investigadores de la Cultura Maya 11 (2):494–503. Uni‑ versidad Autónoma de Campeche, Campeche, Mexico. ———. 2005a. Linking Society with the Compositional Analyses of Pottery: A Model from Comparative Ethnography. In Pottery Manufacturing Processes: Reconstitution and Interpretation, ed. Alexandre Livingstone Smith, Dominique Bosquet, and Rémi Martineau, 15–21. BAR International Series, British Archaeological Reports, Oxford. ———. 2005b. Maya Blue and Palygorskite: A Second Possible Pre-Columbian Source. Ancient Mesoamerica 16:51–62. ———. 2006. Las vasíjas como símbolos: El caso del gremio de alfareros de Ticul, Yuca‑ tán. In Ethnoarqueología: El contexto dinámico de la cultura material a través del tiempo, ed. Eduardo Williams, 105–125. El Colegio de Michoacán, Zamora, Micho‑ acán, Mexico. Arnold, Dean E., and Bruce F. Bohor. 1975. Attapulgite and Maya Blue: An Ancient Mine Comes to Light. Archaeology 28( January):23–29. ———. 1976. An Ancient Attapulgite Mine in Yucatán. Katunob 8(4):25–34 ( June 1974). ———. 1977. The Ancient Clay Mine at Yo’ K’at, Yucatán. American Antiquity 42:575– 582. Arnold, Dean E., Hector Neff, and Ronald L. Bishop. 1991. Compositional Analysis and “Sources” of Pottery: An Ethnoarchaeological Approach. American Anthropologist 93:70–90. Arnold, Dean E., Hector Neff, R. L. Bishop, and M. D. Glascock. 1999. Testing Interpre‑ tive Assumptions of Neutron Activation Analysis: Contemporary Pottery in Yuca‑ tán, 1964–1994. In Material Meanings: Critical Approaches to the Analysis of Material Culture, ed. E. Chilton, 61–84. University of Utah Press, Salt Lake City. Arnold, Dean E., Hector Neff, and Michael D. Glascock. 2000. Testing Assumptions of Neutron Activation Analysis: Communities, Workshops and Paste Preparation in Yucatán, Mexico. Archaeometry 42(2):301–316. Arnold, Dean E., Hector Neff, Michael D. Glascock, and Robert J. Speakman. 2007. Sourcing the Palygorskite Used in Maya Blue: A Pilot Study Comparing the Results of INAA and LA-ICP-MS. Latin American Antiquity 18(1):44–58. Arnold, Dean E., and A. L. Nieves. 1992. Factors Affecting Standardization. In Ceramic Production and Distribution: An Integrated Approach, ed. G. J. Bey III and C. A. Pool, 93–113. Westview Press, Boulder, Colorado. Arnold, Phillip J., III. 1988. Ceramic Production and Consumption in the Sierra de los Tuxtlas, Veracruz, Mexico, Research Paper Series No. 21. Latin American Institute, University of New Mexico Press, Albuquerque. ———. 1991a. Dimensional Standardization and Production Scale in Mesoamerican Ceramics. Latin American Antiquity 2:363–370.

329

References Cited

———. 1991b. Domestic Ceramic Production and Spatial Organization: A Mexican Case Study in Ethnoarchaeology. Cambridge University Press, Cambridge. ———. 2000. Working without a Net: Recent Trends in Ceramic Ethnoarchaeology. Journal of Archaeological Research 8(2):105–133. Arthur, John W. 2006. Living with Pottery: Ethnoarchaeology among the Gamo of Southwest Ethiopia. University of Utah Press, Salt Lake City. Bailey, S. W., G. W. Brindley, W. D. Johns, R. T. Martin, and M. Ross. 1971. Summary of National and International Recommendations on Clay Mineral Nomenclature: 1969–70 CMS Nomenclature Committee. Clays and Clay Minerals 19:129–132. Baklanoff, Eric N. 1980. The Diversification Quest: A Monocrop Export Economy in Transition. In Yucatán: A World Apart, ed. E. H. Moseley and Edward D. Terry, 202–244. University of Alabama Press, Tuscaloosa. Barrera Vásquez, A. 1937. Cerámica Maya. Obre, Organo de la Universidad Nacional del Sureste de México 1(3):162–164. Mérida, Yucatán. Beck, Margaret E. 2006. Midden Ceramic Assemblage Formation: A Case Study from Kalinga, Philippines. American Antiquity 71:27–51. Benco, N. L. 1987. The Early Medieval Pottery Industry at al-Basra, Morocco. BAR Inter‑ national Series 341, British Archaeological Reports, Oxford. Bentley, R. Alexander, Matthew W. Hahn, and Stephen J. Shennan. 2004. Random Drift and Culture Change. Proceedings of the Royal Society B: Biological Sciences 271(1547): 1443–1450. Bentley, R. Alexander, and Herbert D.G. Maschner. 2001. Stylistic Change as a Self­organized Critical Phenomenon: An Archaeological Study in Complexity. Journal of Archaeological Method and Theory 8(1):35–66. Bentley, R. Alexander, and Stephen J. Shennan. 2003. Cultural Transmission and Stochas‑ tic Network Growth. American Antiquity 68(3):459–485. Berg, Ina. 2004. The Meanings of Standardization: Conical Cups in the Late Bronze Aegean. Antiquity 78:74–85. Bey, George. 2005. Evidence of Change in Domestic Pottery Production and Distribu‑ tion in Tula, Hidalgo, Mexico. La Tinaja: A Newsletter of Archaeological Ceramics 16(1–2):10–21. Bey, George J., III, and Christopher A. Pool, eds. 1992. Ceramic Production and Distribution: An Integrated Approach. Westview Press, Boulder, Colorado. Black, M. 1963. On Formal Ethnographic Procedures. American Anthropologist 65(6): 1347–1351. Black, M., and D. Metzger. 1965. Ethnographic Description and the Study of Law. American Anthropologist 67(6)2:141–165. Blackman, J. M., G. J. Stein, and P. B. Vandiver. 1993. The Standardization Hypothesis and Ceramic Mass Production: Technological, Compositional, and Metric Indexes of Craft Specialization at Tell Leilan, Syria. American Antiquity 58:60–80. Bohor, Bruce F. 1975. Attapulgite in Yucatán. Guidebook to Field Trip No. 4. International Clay Conference, Mexico City, July.

330

References Cited

Bowser, Brenda J. 2000. From Pottery to Politics: An Ethnoarchaeological Study of Politi‑ cal Factionalism, Ethnicity, and Domestic Pottery Style in the Ecuadorian Amazon. Journal of Archaeological Method and Theory 7(3):219–248. Brainerd, G. W. 1958. The Archaeological Ceramics of Yucatán. Anthropological Records, 19. University of California, Berkeley. Broad, W. J. 1979. Paul Feyerabend: Science and the Anarchist. Science 206:534–537. Brumfiel, E. M., and T. K. Earle. 1987. Specialization, Exchange, and Complex Socie­ ties: An Introduction. In Specialization, Exchange, and Complex Societies, ed. E. M. Brumfiel and T. K. Earle, 1–9. Cambridge University Press, Cambridge. Bunzel, Ruth L. 1929. The Pueblo Potter: A Study of Creative Imagination in Primitive Art. Columbia University Press, New York. Bureau of the Census. 1978. Statistical Abstract of the United States. U.S. Department of Commerce, Bureau of the Census, Washington, D.C. Byrne, Bryan. 1994. Access to Subsistence Resources and the Sexual Division of Labor among Potters. Cross Cultural Research 28(3):225–250. Clark, Andrew J., Maya Elston, and Mary Louise Hart. 2002. Understanding Greek Vases: A Guide to Terms, Styles, and Techniques. J. Paul Getty Museum, Los Angeles. Costin, Cathy Lynne. 1991. Craft Specialization: Issues in Defining, Documenting and Explaining the Organization of Production. Archaeological Method and Theory 3:1–56. ———. 2000. The Use of Ethnoarchaeology for the Archaeological Study of Ceramic Production. Journal of Archaeological Method and Theory 7(4):377–403. ———. 2001. Craft Production Systems. In Archaeology at the Millennium: A Sourcebook, ed. Gary M. Feinman and T. Douglas Price, 273–327. Kluwer Academic/Ple‑ num Publishers, New York. ———. 2005. Craft Production. In Handbook of Archaeological Methods, ed. Herbert D.G. Maschner and Christopher Chippindale, 1034–1107. Altamira Press, Lan‑ ham, Maryland. ———. 2007. Thinking about Production: Phenomenological Classification and Lexi‑ cal Semantics. In Rethinking Craft Specialization in Complex Societies: Archaeological Analyses of the Social Meaning of Production, ed. Zachary X. Hruby and Rowan K. Flad, 143–162. Archaeological Papers of the American Anthropological Associa‑ tion, No. 17. American Anthropological Association, Arlington, Virginia. Costin, Cathy L., and Melissa Hagstrum. 1995. Standardization, Labor Investment, Skill and the Organization of Ceramic Production in Late Prehispanic Highland Peru. American Antiquity 60(4):619–639. Crown, Patricia L. 1995. The Production of the Salado Polychromes in the American Southwest. In Ceramic Production in the American Southwest, ed. Barbara J. Mills and Patricia L. Crown, 142–166. University of Arizona Press, Tucson. ———. 2001. Learning to Make Pottery in the Prehispanic American Southwest. Journal of Anthropological Research 57(4):451–469. Curtis, F. 1962. The Utility Pottery Industry of Bailén, Southern Spain. American Anthropologist 64(3):486–503.

331

References Cited

D’Andrade, R. 1995. The Development of Cognitive Anthropology. Cambridge University Press, Cambridge. David, Nicholas, and H. Hennig. 1972. The Ethnography of Pottery: A Fulani Case Seen in Archaeological Perspective. Addison Wesley Modular Publications 21:1–29. Addi‑ son Wesley, Reading, Massachusetts. David, Nicholas, and Carol Kramer. 2001. Ethnoarchaeology in Action. Cambridge Uni‑ versity Press, Cambridge. Day, Peter M. 2004. Marriage and Mobility: Traditions and the Dynamics of the Pottery System in Twentieth-century East Crete. In Pseira VIII: The Archaeological Survey of Pseira Island, Part I, ed. Philip P. Betancourt, Costis Davaras, and Richard Hope Simpson, 105–162. INSTAP Academic Press, Philadelphia. Deal, Michael. 1988. An Ethnoarchaeological Approach to the Identification of Maya Domestic Pottery Production. In Ceramic Ecology Revisited, 1987: The Technology and Socioeconomics of Pottery, ed. Charles C. Kolb, 1:111–142. BAR International Series 436, British Archaeological Reports, Oxford. ———. 1998. Pottery Ethnoarchaeology in the Central Maya Highlands. University of Utah Press, Salt Lake City. Deetz, J. 1965. The Dynamics of Stylistic Change in Arikara Ceramics. University of Illinois Press, Champaign. De Sena, Eric C., and Janne P. Ikäheimo. 2003. The Supply of Amphora-Borne Com‑ modities and Domestic Pottery in Pompeii 150 BC–AD 79: Preliminary Evidence from the House of the Vestals. European Journal of Archaeology 6(3):301–321. Diario de Yucatán. 1987. El barro de Ticul no podrá enviarse a otros municipios. Miercoles, June 10, pp. 1, 4. Mérida, Yucatán, Mexico. Díaz, May. 1970. Tonalá: Conservativism, Responsibility and Authority in a Mexican Town. University of California Press, Berkeley. Dobres, Marcia-Anne, and Christopher R. Hoffman. 1994. Social Agency and the Dynamics of Prehistoric Technology. Journal of Anthropological Method and Theory 1(3):211–258. Doehring, D. O., and J. H. Butler. 1974. Hydrogeologic Constraints on Yucatán’s Devel‑ opment. Science 186(4164):591–596. Donnan, Christopher B. 1965. Moche Ceramic Technology. Nawpa Pacha 3:115–134. Donnan, Christopher, and C. W. Clewlow Jr., eds. 1974. Ethnoarchaeology. Monograph 4. Institute of Archaeology, University of California, Los Angeles. Druc, Isabelle C. 2000. Ceramic Production in San Marcos Acteopan, Puebla, Mexico. Ancient Mesoamerica 11:77–89. Dumond, C. S., and Don Dumond, eds. 1982. Demography and Parish Affairs in Yucatán, 1797–1897: Documents from the Archivo de la Mitra Emeritense, Selected by Joaquín de Arrigunaga Peón. University of Oregon Anthropological Papers No. 27. Dumond, Don E. 1997. The Machete and the Cross: Campesino Rebellion in Yucatán. Uni‑ versity of Nebraska Press, Lincoln. Duncan, Roger J. 1996. The Ceramics of Ráquira, Colombia: Gender, Work, and Economic Change. University Press of Florida, Gainesville.

332

References Cited

Dunning, N. 1994. Puuc Ecology and Settlement Patterns. In Hidden among the Hills: Maya Archaeology of the Northwest Yucatán Peninsula: First Maler Symposium, Bonn, 1989, ed. Hanns J. Prem, 1–43. Verlag von Flemming, Anlage, Mochmuhl. Durkheim, Émile. 1933 [1893]. The Division of Labor in Society. Translated by George Simpson. Free Press, New York. Estes, John Robert. 2003. Middle Range Research into the Organization of Production and Product Standardization in Mata Ortiz, Chihuahua, Mexico. Ph.D. Disserta‑ tion, Department of Anthropology, University of New Mexico. Feinman, G., S. A. Kowalewski, and R. E. Blanton. 1984. Modeling Ceramic Produc‑ tion and Organizational Change in the Prehispanic Valley of Oaxaca, Mexico. In The Many Dimensions of Pottery, ed. S. E. van der Leeuw and A. C. Pritchard, 295–333. University of Amsterdam, Amsterdam. Fernández Repetto, F., and G. Negore Sierra. 1994. La Fiesta del Santo Cristo de Las Ampollas hacía finales del siglo XIX. Boletín de la Escuela de Ciencias Antropológicas de la Universidad de Yucatán 19(114):23–40. Fish, Suzanne K. 2000. Hohokam Impacts on Sonoran Desert Environment. In Imperfect Balance: Landscape Transformations in Pre-Columbian America, ed. David L. Lentz, 251–280. Columbia University Press, New York. Fitzhugh, Ben. 2002. Review of Social Theory in Archaeology and Anthropological Perspectives on Technology by Michael Brian Schiffer. American Antiquity 67(4):789–791. Foster, G. M. 1948. Some Implications of Modern Mexican Mold‑made Pottery. Southwestern Journal of Anthropology 4:356–370. ———. 1955. Contemporary Pottery Techniques in Southern and Central Mexico. Middle American Research Institute Publication No. 2. Tulane University, New Orleans, Louisiana. ———. 1960a. Archaeological Implications of the Modern Pottery of Acatlan, Puebla, Mexico. American Antiquity 26:205–214. ———. 1960b. Life Expectancy of Utilitarian Pottery in Tzintzuntzan, Michoacan, Mexico. American Antiquity 25:606–609. ———. 1963. The Dyadic Contract in Tzintzuntzan, II: Patron‑Client Relationship. American Anthropologist 65:1280–1294. ———. 1965. The Sociology of Pottery: Questions and Hypotheses Arising from Con‑ temporary Mexican Work. In Ceramics and Man, ed. F. R. Matson, 43–61. Aldine, Chicago. ———. 1967. Tzintzuntzan: Mexican Peasants in a Changing World. Little, Brown, Boston. ———, assisted by G. Ospina. 1948. Empire’s Children: The People of Tzintzuntzan, Institute of Social Anthropology Publication No. 6. Smithsonian Institution, Wash‑ ington, D.C. Frake, C. 1964. Notes on Queries in Ethnography. American Anthropologist 66 (3)2: 132–145. Friedrich, Margaret Hardin. 1970. Design Structure and Social Interaction: Archaeologi‑ cal Implications of an Ethnographic Analysis. American Antiquity 35(3):332–343.

333

References Cited

García, Enriqueta. 1973. Modificaciones al sistema de clasificación climática de Koppen (para adaptarlo a las condiciones de la República Mexicana). Universidad Nacional Autónoma de México, Ciudad Universitaria, Mexico, D.F. Gillespie, Susan D. 2000. Rethinking Ancient Maya Social Organization: Replacing “Lin‑ eage” with “House.” American Anthropologist 102:467–484. Glowacki, Donna M., and Hector Neff, eds. 2002. Ceramic Production and Circulation in the Greater Southwest: Source Determination by INAA and Complementary Mineralogical Investigations. Monograph 44. Cotsen Institute of Archaeology, Uni‑ versity of California, Los Angeles. Gosselain, Olivier P. 1992. Technology and Style: Potters and Pottery among Bafia of Cameroon. Man (New Series) 27(3):559–586. ———. 2000. Materializing Identities: An African Perspective. Journal of Archaeological Method and Theory 7(3):187–217. Gould, Richard A., ed. 1978. Explorations in Ethnoarchaeology. A School of American Research Book, University of New Mexico, Albuquerque. ———. 1980. Living Archaeology. Cambridge University Press, Cambridge. Güémez Pineda, Arturo. 1997. La rebelión de Nohcacab, prefacio inédito de la Guerra de Castas. Saastun: Revista de Cultura Maya Año 0(2):51–82. Haaland, Randi. 1978. Ethnographical Observations of Pottery-making in Darfur, West‑ ern Sudan, with some Reflections on Archaeological Interpretation. In New Directions in Scandinavian Archaeology, ed. Kristian Kristiansen and Carsten PaludanMüller, 47–61. National Museum of Denmark, Copenhagen. Hanson, N. R. 1958. Patterns of Discovery: An Inquiry into the Conceptual Foundations of Science. Cambridge University Press, Cambridge. Hayden, Brian, and Aubrey Cannon. 1982. The Corporate Group as an Archaeological Unit. Journal of Anthropological Archaeology 1:132–158. ———. 1984a. Interaction Inferences in Archaeology and Learning Frameworks of the Maya. Journal of Anthropological Archaeology 3:325–367. ———. 1984b. The Structure of Material Systems: Ethnoarchaeology in the Maya Highlands. SAA Papers No. 3. Society for American Archaeology, Washington, D.C. Headland, Thomas N. 1997. Revisionism in Ecological Anthropology. Current Anthropology 38(4):605–630. Hegmon, Michelle. 2000. Advances in Ceramic Ethnoarchaeology. Journal of Archaeological Method and Theory 7(3):129–137. Henry, Jean C. 1992. Atzompa: A Pottery-producing Village of Southern Mexico in the Mid1950s. Vanderbilt Publications in Anthropology, Vanderbilt University, Nashville. Heyman, Karen. 2006. Making Connections. Science 313:604–606. Hill, J. N. 1970. Broken K Pueblo: Prehistoric Social Organization in American Southwest. Anthropological Papers of the University of Arizona No. 18. University of Arizona Press, Tucson. Holden, Constance, ed. 2007. Random Samples: Very Old Wine. Science 317:301. Howard, Hilary, and Elaine L. Morris, eds. 1981. Production and Distribution: A Ceramic Viewpoint. BAR International Series 120, British Archaeological Reports, Oxford.

334

References Cited

Howie-Langs, L. 2006. Ceramic Production and Consumption in the Maya Lowlands during the Classic to Postclassic Transition: A Technological Study of Ceramics at Lamanai, Belize. Ph.D. Dissertation, Department of Archaeology, University of Sheffield, UK. INEGI (Instituto Nacional de Estadistica Geografía e Informatica). 1996. Ticul, Estado de Yucatán, Cuaderno Estadistico Municipal (Edición 1995). Instituto Nacional de Estadistica Geografía e Informatica. Aguascalientes, Mexico. Isphording, Wayne C. 1975. The Physical Geology of Yucatán. Transactions–Gulf Coast Association of Geological Societies 25:231–262. Isphording, Wayne C., and E. M. Wilson. 1974. The Relationship of “Volcanic Ash,” Sak Lu’um, and Palygorskite in Northern Yucatán Maya Ceramics. American Antiquity 39(3):483–488. Joseph, Gilbert M. 1980. Revolution from Without: The Mexican Revolution in Yucatán, 1910–1940. In Yucatán: A World Apart, ed. Edward H. Moseley and Edward D. Terry, 142–171. University of Alabama Press, Tuscaloosa. Kaplan, F. S. 1974. Learning and the Transmission of Style among Folk-urban Potters in Puebla, Mexico. International Congreso of Americanists Actas 3:243–248. Instituto Nacional de Antropología y Historia, Mexico, D.F. ———. 1980. Una tradición alfarera, conomiento y estilo: Un análsis basado en una tradición de alfarería Mexicana. Instituto Indigenista Nacional, Mexico, D.F. Kean, Sam. 2006. Nothing Succeeds Like Failure (Review of Why Most Things Fail: Evolution, Extinction and Economics by Paul Ormerod). Science 312:531. Kintz, Ellen R. 1990. Life under the Tropical Canopy: Tradition and Change among the Yucatec Maya. Holt, Rinehart and Winston, Forth Worth, Texas. Kleinberg, Mary Jill. 1979. Kinship and Economic Growth: Pottery Production in a Japa‑ nese Village. Ph.D. Dissertation, University of Michigan, Ann Arbor. Koechlin, Eienne, and Alexandre Hyafil. 2007. Anterior Prefrontal Function and the Limits of Human Decision-making. Science 318:594–598. Kolb, Charles C. 1996. “Firing” as the Critical Stage of Pottery Manufacture: A Ceramic Ecology and Material Science Perspective. In Archaeology Mesoamericana: Homenaje a William T. Sanders, ed. Jeffrey R. Parsons, Robert S. Santley, and Mari Carmen Serra Puche. Instituto Nacional de Antropología e Historia y Arqueología Mexicana 1:97–110. Kolb, Michael J., and James E. Snead. 1997. It’s a Small World after All: Comparative Analyses of Community Organization in Archaeology. American Antiquity 62(4): 609–628. Kramer, Carol, ed. 1979. Ethnoarchaeology: Implications of Ethnography for Archaeology. Columbia University Press, New York. ———. 1985. Ceramic Ethnoarchaeology. Annual Review of Anthropology 14:17–102. ———. 1997. Pottery in Rajasthan: Ethnoarchaeology in Two Indian Cities. Smithsonian Institution Press, Washington, D.C. Krause, R. A. 1984. Modeling the Making of Pots: An Ethnoarchaeological Approach. In The Many Dimensions of Pottery, ed. S. E. van der Leeuw and A. C. Pritchard,

335

References Cited

615–698. Universiteit van Amsterdam, Albert Egges van Giffen Insituut voor Praeen Protohistorie, Cingvla VII, Amsterdam. ———. 1985. The Clay Sleeps: An Ethnoarchaeological Study of Three African Potters. University of Alabama Press, Tuscaloosa. Krotser, P. H. 1974. Country Pottery of Veracruz, Mexico: Technological Survival and Culture Change. In Ethnoarchaeology, ed. C. B. Donnan and C. W. Clewlow Jr., 131–148. Monograph 4. Institute of Archaeology, University of California, Los Angeles. Lackey, L. M. 1982. The Pottery of Acatlán: A Changing Mexican Tradition. University of Oklahoma Press, Norman. Lawler, Andrew. 2007. Middle Asia Takes Center Stage. Science 317:586–590. Lechtman, Heather. 1977. Style in Technology—Some Early Thoughts. In Material Culture: Styles, Organization, and Dynamics of Technology. 1975 Proceedings of the American Ethnological Society, ed. Heather Lechtman and R. Merrill, 3–20. West Publishing, St. Paul, Minnesota. Lemonnier, Pierre. 1986. The Study of Material Culture Today: Toward an Anthropology of Technical Systems. Journal of Anthropological Archaeology 5(2):147–186. ———. 1992. Elements for an Anthropology of Technology. Anthropological Papers No. 88, Museum of Anthropology, University of Michigan, Ann Arbor. ———. 1993. Technological Choices: Transformation in Material Cultures since the Neolithic. Routledge, New York. Lister, Florence C., and Robert H. Lister. 1987. Andalusian Ceramics in Spain and New Spain: A Cultural Register from the Third Century B.C. to 1700. University of Ari‑ zona Press, Tucson. Loney, Helen L. 2000. Society and Technological Control: A Critical Review of Models of Technological Change in Ceramic Studies. American Antiquity 65(4):646–668. Longacre, William. 1970. Archaeology as Anthropology: A Case Study. Anthropological Papers of the University of Arizona No. 17. ———, ed. 1991. Ceramic Ethnoarchaeology. University of Arizona Press, Tucson. Longacre, William A., Kenneth L. Kvamme, and Masashi Kobayashi. 1988. Southwestern Pottery Standardization: An Ethnoarchaeological View from the Philippines. The Kiva 53:101–112. Longacre, W. A., and J. M. Skibo, eds. 1994. Kalinga Ethnoarchaeology: Expanding Archaeological Method and Theory. Smithsonian Institution Press, Washington, D.C. Malafouris, Lambros. 2004. The Cognitive Basis of Material Engagement: Where Brain, Body and Culture Conflate. In Rethinking Materiality: The Engagement of Mind with the Material World, ed. Elizabeth DeMarrais, Chris Gosden, and Colin Ren‑ frew, 53–62. McDonald Institute Monographs. McDonald Institute for Archaeo‑ logical Research, University of Cambridge, Cambridge. Matson, Frederick R. 1966. Power and Fuel Resources in the Ancient Near East. Advancement of Science 23: 146–153. ———. 1973. The Potters of Chalkis. In Classics and the Classical Tradition, ed. E. N. Borza and R. W. Carruba, 117–142. Pennsylvania State University Press, University Park.

336

References Cited

Mauss, Marcel. 1976 [1950]. Sociology and Psychology. Routledge and Kegan Paul, London. McAnany, P. A. 1990. Water Storage in the Puuc Region of the Northern Maya Low‑ lands: A Key to Population Estimates and Architectural Variability. In Precolumbian Population History in the Maya Lowlands, ed. T. P. Culbert and D. S. Rice, 263–284. University of New Mexico Press, Albuquerque. Mercer, H. C. 1896. Hill-caves of Yucatán: A Search for Evidence of Man’s Antiquity in the Caverns of Central America. J. B. Lippincott Company, Philadelphia. ———. 1897a. The Kabal, or Potter’s Wheel, of Yucatán. Bulletin of the Free Museum of Science and Art 1(2):63–70. University of Pennsylvania, Philadelphia. ———. 1897b. The Potter’s Wheel in Ancient America. Science 5(128):919–920. Metzger, Duane, and Gerald E. Williams. 1963a. A Formal Ethnographic Analysis of Tenejapa Ladino Weddings. American Anthropologist 65(5):1076–1101. ———. 1963b. Tenejapa Medicine 1: The Curer. Southwestern Journal of Anthropology 19(2):216–234. ———. 1966. Some Procedures and Results in the Study of Native Categories: Tzeltal “Firewood.” American Anthropologist 68(2)1:389–407. Meyer, M. C., W. L. Sherman, and S. M. Deeds. 1999. The Course of Mexican History (Sixth Edition). Oxford University Press, Oxford. Meyers, Allan. 2004. The Challenge and Promise of Hacienda Archaeology in Yucatán. The SAA Archaeological Record 4:20–23. Miller, Daniel. 1985. Artefacts as Categories: A Study of Ceramic Variability in Central India. Cambridge University Press, Cambridge. Mitchell, William P. 1973. The Hydraulic Hypothesis: A Reappraisal. Current Anthropology 14(5):532–534. ———. 1976. Irrigation and Community in the Central Peruvian Highlands. American Anthropologist 78:25–44. Neff, Hector. 1992. Ceramics and Evolution. Archaeological Method and Theory 4:141– 193. ———. 1993. Theory, Sampling, and Analytical Techniques in the Archaeological Study of Prehistoric Ceramics. American Antiquity 58(1):23–44. Neupert, Mark A. 2000. Clays of Contention: An Ethnoarchaeological Study of Fac‑ tionalism and Clay Composition. Journal of Archaeological Method and Theory 7(3): 249–272. Nicholson, Paul T., and Helen L. Patterson. 1992. The Ballás Pottery Project: Ethno­ archaeology in Upper Egypt. In Ceramic Production and Distribution: An Integrated Approach, ed. George J. Bey III and Christopher A. Pool, 25–47. Westview Press, Boulder, Colorado. Nicklin, Keith. 1981. Pottery Production and Distribution in Southeast Nigeria. In Production and Distribution: A Ceramic Viewpoint, ed. Hilary Howard and Elaine L. Morris, 169–186. BAR International Series 120, British Archaeological Reports, Oxford.

337

References Cited

Oakley, John H., William D.E. Coulson, and Olga Palagia, eds. 1997. Athenian Potters and Painters. Oxbow Monograph 67. Oxbow Books, Oxford. Orosa Díaz, Jaime. 1994. Historia de Yucatán (decimo tercera reimpresin). Universidad Autónoma de Yucatán, Mérida, Yucatán. Oswalt, W. H., and J. W. VanStone. 1967. The Ethnoarchaeology of Crow Village, Alaska. Smithsonian Institution Bureau of American Ethnology Bulletin No. 199, Wash‑ ington, D.C. Palka, Joel W. 2005. Unconquered Lacandon Maya: Ethnohistory and Archaeology of Indigenous Culture Change. University Press of Florida, Gainesville. Papousek, D. A. 1974. Manufactura de alfarería: En Temascalcingo, México, 1967. América Indígena 34(4):1009–1046. ———. 1981. The Peasant-potters of Los Pueblos: Stimulus Situation and Adaptive Processes in the Mazahua Region in Central Mexico. Van Gorcum, Assen, the Netherlands. Pastron, A. G. 1974. Preliminary Ethnoarchaeological Investigations among the Tara‑ humara. In Ethnoarchaeology, ed. C. B. Donnan and C. W. Clewlow Jr., 93–114. Monograph 4. Institute of Archaeology, University of California, Los Angeles. Peacock, D.P.S. 1982. Pottery in the Roman World: An Ethnoarchaeological Approach. Longman, London. Pinto, I. V., M. B. Schiffer, S. Smith, and J. M. Skibo. 1987. Effects of Temper on Ceramic Abrasion Resistance: A Preliminary Investigation. Archaeomaterials 1:119–134. Pool, Christopher A. 1992. Integrating Ceramic Production and Distribution. In Ceramic Production and Distribution: An Integrated Approach, ed. G. J. Bey III and C. A. Pool, 275–313. Westview Press, Boulder, Colorado. Pool, Christopher A., and George J. Bey III. 2007. Conceptual Issues in Mesoamerican Pottery Economics. In Pottery Economics in Mesoamerica, ed. Christopher A. Pool and George J. Bey III, 59–85. University of Arizona Press, Tucson. Press, Irwin. 1975. Tradition and Adaptation: Life in a Modern Yucatán Maya Village. Greenwood Press, Westport, Connecticut. Ralph, Delores, and Dean E. Arnold. 1988. Socioeconomic Status, Kinship, and Innova‑ tion: The Adoption of the Tornete in Ticul, Yucatán. In Ceramic Ecology Revisited, 1987: The Technology and Socio-economics of Pottery, ed. C. C. Kolb, 145–164. BAR International Series 436, British Archaeological Reports, Oxford. Rasmusen, C. H. 1994. (Elaboración de textos, Luis Millet.) Mérida en la época colonial y del oro verde. Universidad Autónoma de Yucatán, Mérida, Yucatán. Rathje, W. L. 1975. Last Tango in Mayapán: A Tentative Trajectory of Production­distribution Systems. In Ancient Civilization and Trade, ed. J. A. Sabloff and C. C. Lamberg-Karlovsky, 409–448. University of New Mexico Press, Albuquerque. Redfield, Robert, and Alfonso Villa Rojas. 1962. Chan Kom, a Maya Village (abridged edition). University of Chicago Press, Chicago. (Originally published in 1934, Carnegie Institution of Washington Publication 448, Washington, D.C.) Reed, Nelson. 1964. The Caste War of Yucatán. Stanford University Press, Stanford, California.

338

References Cited

Reina, Ruben E., and Robert M. Hill. 1978. The Traditional Pottery of Guatemala. Uni‑ versity of Texas Press, Austin. Rejón, Patrón, L. 1981. Tabí, una hacienda azucarera de Yucatán a fines del siglo XIX. In Yucatán: Peonaje y liberación, ed. Raul Maldonado Coello, 117–140. Comisión Editorial del Estado, INAH, Mérida, Yucatán. ———. 1993. Hacienda Tabí: Un capitulo en la historia de Yucatán. Cuadernos de Cultura Yucateca No. 3. Gobiernodel Estadode Yucatán, Mérida. Rendón, Silvia. 1947. Notas sobre la alfarería indígena de la Peninsula de Yucatán. Revista Mexicana de Estudios Antropológicos 9:106–123. Mexico, D.F. Renfrew, Colin. 2004. Towards a Theory of Material Engagement. In Rethinking Materiality: The Engagement of Mind with the Material World, ed. Elizabeth DeMarrais, Chris Gosden, and Colin Renfrew, 23–31. McDonald Institute Monographs. McDonald Institute for Archaeological Research, University of Cambridge, Cambridge. Rice, Prudence M. 1981. Evolution of Specialized Pottery Production: A Trial Model. Current Anthropology 22(3):219–240. ———. 1984. Change and Conservatism in Pottery-producing Systems. In The Many Dimensions of Pottery: Ceramics in Archaeology and Anthropology, ed. S. E. van der Leeuw and A. C. Pritchard, 231–288. Universiteit van Amsterdam, Albert Egges van Giffen Insituut voor Prae- en Protohistorie, Cingvla VII, Amsterdam. ———. 1987. Pottery Analysis: A Sourcebook. University of Chicago Press, Chicago. ———. 1991. Specialization, Standardization and Diversity: A Retrospective. In The Legacy of Anna O. Shepard, ed. Ronald L. Bishop and Frederick W. Lange, 257–279. University Press of Colorado, Boulder. Roux, Valentine. 2007. Ethnoarchaeology: A Non-historical Science of Reference Neces‑ sary for Interpreting the Past. Journal of Archaeological Method and Theory 14(2): 153–178. Roys, R. L. 1933. The Book of Chilam Balam of Chumayel. Carnegie Institution of Wash‑ ington Publication 438. Washington, D.C. Russell, Glenn S., Banks L. Leonard, and Jesus Breceño. 1998. The Cerro Mayal Work‑ shop: Addressing Issues of Craft Specialization in Moche Society. In Andean Ceramics: Technology, Organization, and Approaches, ed. Izumi Shimada, 63–89. Museum of Applied Science Center for Archaeology, University of Pennsylvania Museum of Archaeology and Anthropology, Philadelphia. Rye, O. S. 1981. Pottery Technology: Principles and Reconstruction. Taraxacum, Washing‑ ton, D.C. Rye, O. S., and Clifford Evans. 1976. Traditional Pottery Techniques of Pakistan: Field and Laboratory Studies. Smithsonian Contributions to Anthropology No. 21. Smithson‑ ian Institution Press, Washington, D.C. Sanders, W. T., and Barbara J. Price. 1968. Mesoamerica: The Evolution of a Civilization. Random House, New York. Sanfey, Alan G. 2007. Social Decision-making: Insights from Game Theory and Neuro‑ science. Science 318:598–602.

339

References Cited

Schiffer, Michael Brian. 1975. Behavioral Chain Analysis: Activities, Organization, and the Use of Space. Chapters in the Prehistory of Eastern Arizona, 4. Fieldiana: Anthropology 65:103–119. Field Museum of Natural History, Chicago. ———. 2005. The Devil Is in the Details: The Cascade Model of Invention Processes. American Antiquity 70(3):485–502. Schultz, L. G., A. O. Shepard, P. D. Blackmon, and H. C. Starkey. 1971. Mixed-layer Kaolinite-montmorillonite from the Yucatán Peninsula, Mexico. Clays and Clay Minerals 19:137–150. Shaw, Justine M. 2003. Climate Change and Deforestation: Implications for the Maya Collapse. Ancient Mesoamerica 14:157–167. Sheehy, James J. 1988. Ceramic Ecology and the Clay/Fuel Ratio: Modeling Fuel Con‑ sumption in Tlajinga 33, Teotihuacan, Mexico. In Ceramic Ecology Revisited, 1987: The Technology and Socioeconomics of Pottery, ed. Charles C. Kolb, 1:199–226. BAR International Series 436, British Archaeological Reports, Oxford. ———. 1992. Ceramic Production in Ancient Teotihuacan, Mexico: A Case Study of Tlajinga 33. 2 vols. Ph.D. Dissertation, Department of Anthropology, Pennsylvania State University, University Park. Shennan, Stephen. 2000. Population, Culture History and the Dynamics of Culture Change. Current Anthropology 41(5):811–835. Shepard, A. O. 1956. Ceramics for the Archaeologist. Carnegie Institution of Washington Publication 609. Washington, D.C. Shister, Neil. 2007. The Consequences of Integration. World Trade Magazine 20(3):7. Sillar, Bill. 2000. Shaping Culture, Making Pots and Constructing Households: An Eth‑ noarchaeological Study of Pottery Production, Trade and Use in the Andes. BAR International Series 883, British Archaeological Reports, Oxford. Sillar, Bill, and Michael Tite. 2000. The Challenge of “Technological Choices” for Materi‑ als Science Approaches in Archaeology. Archaeometry 42(1):2–20. Skibo, James M. 1992. Pottery Function: A Use Alteration Perspective. Plenum Press, New York. ———. 1994. The Kalinga Cooking Pot: An Ethnoarchaeological and Experimental Study of Technological Change. In Kalinga Ethnoarchaeology: Expanding Archaeological Method and Theory, ed. William A. Longacre and James M. Skibo, 113–126. Smithsonian Institution Press, Washington, D.C. Skibo, James M., and Michael B. Schiffer. 1987. The Effects of Water on Processes of Ceramic Abrasion. Journal of Archaeological Science 14:83–96. Smith, Adam. 1953 [1776]. An Inquiry into the Nature and Causes of the Wealth of Nations, Selections, Book I. Henry Gegnery Company, Chicago, Illinois. Spielmann, Katherine A. 2002. Feasting, Craft Specialization, and the Ritual Mode of Production in Small-scale Societies. American Anthropologist 104(1):195–207. Spier, R.F.G. 1967. Work Habits, Postures and Fixtures. In American Historical Anthropology: Essays in Honor of Leslie Spier, ed. C. L. Riley and W. W. Taylor, 197–220. Southern Illinois University Press, Carbondale.

340

References Cited

Stanislawski, M. B. 1977. Ethnoarchaeology of Hopi and Hopi‑Tewa Pottery Making: Styles of Learning. In Experimental Archaeology, ed. D. Ingersoll, J. E. Yellen, and W. MacDonald, 378–408. Columbia University Press, New York. Stanislawski, M. B., and B. B. Stanislawski. 1978. Hopi and Hopi‑Tewa Ceramic Tradi‑ tion Networks. In The Spatial Organization of Culture, ed. Ian Hodder, 61–76. Ger‑ ald Duckworth, London. Stark, Barbara. 1995. Problems in Analysis of Standardization and Specialization in Pot‑ tery. In Ceramic Production in the American Southwest, ed. Barbara J. Mills and Patri‑ cia L. Crown, 231–268. University of Arizona Press, Tucson. Stark, Miriam T. 1991a. Ceramic Change in Ethnoarchaeological Perspective: A Kalinga Case Study. Asian Perspectives 30(2):193–216. ———. 1991b. Ceramic Production and Community Specialization: A Kalinga Ethno‑ archaeological Study. World Archaeology 23:64–78. ———. 1994. Pottery Exchange and the Regional System. In Kalinga Ethnoarchaeology: Expanding Archaeological Method and Theory, ed. William A. Longacre and James M. Skibo, 169–197. Smithsonian Institution Press, Washington, D.C. ———. 2003. Current Issues in Ceramic Ethnoarchaeology. Journal of Archaeological Research 11(3):193–242. Steggerda, M. 1943. A Description of Thirty Towns in Yucatán, Mexico. Anthropological Papers No. 30. Bureau of American Ethnology Bulletin 136, 227–248. Washington, D.C. Stephens, John L. 1996 [1843]. Incidents of Travel in Yucatán (new edition by Karl Ack‑ erman). Smithsonian Institution Press, Washington, D.C. Steward, Julian H. 1955. Theory of Culture Change. University of Illinois Press, Champaign. Tamayo, J. L. 1962. Geografía Fisica, 2. In Geografía General de Mexico (Second Edition). Instituto Mexicano de Investigaciones Economicas, Mexico, D.F. Terán, Silvia, and Christian Rasmussen. 1981. Artesanías de Yucatán. Mérida, Yucatán, Mexico. PESIP/Arte y Comunicación, Dirección General de Culturales Populares/ SEP, Mérida, Yucatán. Thieme, Mary S. 2007. Changes in the Style, Production and Distibution of Pottery in Santa María Atzompa, Oaxaca, Mexico, during the 1990s. Museum Anthropology 30(2):125–140. Thompson, E. H. 1895. The Chultunes of Labná, Yucatán. Memoirs of the Peabody Museum of American Archaeology and Ethnology 1(3), Cambridge. Thompson, R. A. 1974. The Winds of Tomorrow: Social Change in a Maya Town. Univer‑ sity of Chicago Press, Chicago. Thompson, R. H. 1958. Modern Yucatán Pottery Making. Memoirs of the Society for American Archaeology No. 15, Salt Lake City. Uceda, Santiago, and José Armas. 1998. An Urban Pottery Workshop at the Site of Moche, North Coast of Peru. In Andean Ceramics: Technology, Organization, and Approaches, ed. Izumi Shimada, 91–110. Museum of Applied Science Center for Archae­­ology, University of Pennsylvania Museum of Archaeology and Anthropol‑ ogy, Philadelphia.

341

References Cited

Underhill, Anne. 2003. Investigating Variation in Organization of Ceramic Production: An Ethnoarchaeological Study in Guizhou, China. Journal of Archaeological Method and Theory 10(3):203–275. van der Leeuw, S. E. 1976. Studies in the Technology of Ancient Pottery. 2 vols. Ph.D. Dis‑ sertation, University of Amsterdam, Amsterdam. ———. 1993. Giving the Potter a Choice: Conceptual Aspects of Pottery Techniques. In Technological Choices: Transformation in Material Cultures since the Neolithic, ed. Pierre Lemonnier, 238–288. Routledge, London. van der Leeuw, S., and A. C. Pritchard. 1984. The Many Dimensions of Pottery: Ceramics in Archaeology and Anthropology. Universiteit van Amsterdam, Albert Egges van Giffen Insituut voor Prae- en Protohistorie, Cingvla VII, Amsterdam. Varela Torrecilla, Carmen. 1990. La producción alfarera artesanal del occidente de la Pen‑ insula del Yucatán: Un ejemplo de cambio cultural. Revista Española de Antropología Americana 20:183–220. Velázquez Morlet, Adriana, and Edmundo López de la Rosa. 1988. Atlas arqueológico de Yucatán. In Zonas Arqueológicos: Yucatán, ed. Adriana Velázquez Morlet and Edmundo López de la Rosa, 63–91. Instituto Nacional de Antropología e Historia, Mexico, D.F. Weigand, Phil C. 1969. Modern Huichol Ceramics: Mesoamerican Studies. Research Records of the University Museum, Southern Illinois University, Carbondale. West, R. C. 1964. Surface Configuration and Associated Geology of Middle America. Handbook of Middle America Indians 1:33–83. University of Texas Press, Austin. White, Leslie. 1949. The Science of Culture: A Study of Man and Civilization. Farrar, Straus and Giroux, New York. Williams, Eduardo. 1992. Pots, Pans and People: Ceramic Ecology in West Mexico. Papers from the Institute of Archaeology 3:44–51. University College, London. ———. 1994a. Ecología cerámica en Huáncito, Michoacán. In Arqueología del occidente de México, ed. Eduardo Williams y Robert Novella, 319–362. El Colegio de Micho‑ acán, Zamora, Michoacán, Mexico. ———. 1994b. Organización del espacio doméstico y produccíón cerámica en Huáncito, Michoacán. In Contribuciones a la arqueología y etnohistoria del occidental de México, ed. Eduardo Williams, 189–225. El Colegio de Michoacán, Zamora, Michoacán, Mexico. ———, ed. 2006. Etnoarqueología: El contexto dinámico de la cultura material a través del tiempo. El Colegio de Michoacán, Zamora, Michoacán, Mexico. Williams, Eduardo, and Phil C. Weigand, eds. 2001. Estudios cerámicos en el occidente y norte de México. Instituto Michoacano de Cultura, El Colegio de Michoacán, Zamora, Michoacán, Mexico. Wilson, E. M. 1980. Physical Geography of the Yucatán Peninsula. In Yucatán: A World Apart, ed. E. H. Moseley and Edward D. Terry, 5–40. University of Alabama Press, Tuscaloosa. Wylie, Allison. 1985. The Reaction against Analogy. In Advances in Archaeological Method and Theory, ed. Michael B. Schiffer, 8:63–111. Academic Press, New York.

342

References Cited

Yaeger, Jason, and Marcello A. Canuto. 2000. Introducing an Archaeology of Communi‑ ties. In The Archaeology of Communities: A New World Perspective, ed. Marcello A. Canuto and Jason Yaeger, 1–15. Routledge, London. Zapata Peraza, Renée Lorelei. 1989. Los Chultunes: Sistemas de captación y almacenamiento de agua pluvial. Colección Cientifica, Serie Arqueología, Instituto Nacio‑ nal de Antropología e Historia, Mexico, D.F. Zubrow, E.B.W. 1992. Formal Models of Ceramic Production. In Ceramic Production and Distribution, ed. G. J. Bey III and C. A. Pool, 115–129. Westview Press, Boulder, Colorado.

343

Index

Page numbers in italics indicate figures and illustrations

Agency, 27, 278 Agriculture, 6, 7, 8–9, 32, 82; marginality, 171–72; slash-and-burn, 37–38, 282 Aguacatenango, 45 Akil, 35, 36, 314, 323 Alcancías. See Coin banks Alvarado, Perfirio, 157–58 Analogy: ethnographic, 17–21; processual, 309–11 Apastes, 102, 103, 104, 237, 266, 270(tables) Archaeological sites: at Ticul, 33–34; at Yo’Sah Kab, 194–95, 196, 219 Archaeology, 265; and ethnographic analogy, 309–10 Artisans, in tourist industries, 115, 117, 141 Ashtrays (ceniceros), 119, 122(table), 266 Atzompa, 234 Ayacucho Valley, 19 Bak, Loreto, 78 Banco Internacional, 112 Baptisms, 117 Barriles, 102, 104 Becal, 35, 106, 107, 141, 155

Behavioral chains, 246(table), 311, 312; cross-cultural commonalities of, 18–19; of pottery making, 15, 28–29 Blackware, 113 Bolonchén, 129, 284 Book of Chilam Balam of Chumayel, 33 Borrowing, 273–74 Bowls, 26; food, 46, 71, 243, 244(table) Bracamonte family, 166 Brokers, 271, 324; sales to, 142–48 Cajetes, 266 Calcite, macrocrystalline, 192–93; in temper, 102–3 Calkini, 131 Campeche, 129; clay from, 160, 167, 168, 168–70, 178, 182, 186, 222, 225, 227 Cancún, 32, 143; pottery made for, 113, 118, 119, 131–32, 145, 150, 311, 314, 319, 320; stores in, 115, 117, 141, 142, 148, 149; workshops in, 161–62 Candle holders (candeleros), 266; for Day of the Dead rituals, 108, 109, 313 Cántaros, 100, 101, 101–2, 237, 266, 269(table)

345

Index

Canto, Yoy, 197 Capital, 8, 103, 183; for molds, 250, 253–54, 275; for pottery production, 82–83, 274, 286, 290; for potter’s wheel, 238–39; for store development, 139–40 Carbonates, in temper, 102–3 Cárdenas, Lázaro, 69, 195, 238 Carranza, Venustiano, 195 Carrefor, 33 Carrillo, Sostenes, 158 Casserole vessels (cazuelas), 111 Caste War, 67, 69, 100 Caves, calcite mining in, 192 Cement, in kiln construction, 287, 307 Ceniceros, 119, 122(table), 266 Chab, Juan Bautista, 245–46 Chable, Antonio, 246–47 Chacsinkín, 106 Chan, Carmelo, 246, 247 Chan, Juan Bautista, 246 Chanal, 45 Chan Kom, 51 Chapab source, 130, 162, 284; ownership of, 217–18; temper mining at, 196, 197, 198(table), 201–2, 206, 219; temper preparation at, 207, 208, 211–12 Chiapas, 234 Chichén Itzá, 113, 129, 141 Chicxulub, 136 Chikindzonor, 35 Children, 66, 80; Day of the Dead rituals for, 108, 110; labor provided by, 79, 90; learning by, 42, 45–47, 49, 72, 77, 314 Chimú, 316, 321 Chinautla, 102 Choice, 12; clay, 324–25; consumer, 4, 311–12; feedback in, 230–31; of potters, 220, 229–30, 265; vocational, 80–81. See also Technological choice Cisterns, Terminal Classic, 99 Clay, 6, 228, 239; choice of, 324–25; elite control of, 164–65, 184–86; mining of, 15–16, 29, 155–59, 170–72, 180–82, 187–89; for pottery workshops, 161–62; procurement and storage of, 183–84, 186–87; qualities of, 222–24; sources of, 71–72, 154–55, 167–70, 179(table), 223(table); Ticul, 35, 276, 278; from Yo’ K’at, 160–64, 166–67, 172–78, 319 Climate: and pottery production, 9, 124; Yucatán, 96–98

346

Cobá, 51 Cob family, 78 Cobo, Timoteo, 71 Cognitive anthropology, use of language in, 13–14 Coin banks (alcancías), 112, 117–18, 122(table), 123(table), 255, 311 Communities: pottery-making, 19, 23; production units, 39–40 Community of practice, 34 Concentration, 6–7 Conscription, during Mexican Revolution, 69 Consumers, consumption, 27, 95; choices made by, 4, 311–12; pottery, 93, 311, 312 Context, 4–6; of learning, 42, 44; of production, 33, 39–40, 124–25 Continuity, spatial, 59–60 Cooking pots, 122(table), 224, 284, 320; demand for, 110–12, 124; paste recipe, 225–26; temper used in, 192–93 Costin, Cathy L.: on efficiency, 318–19; on specialization, 3–10, 39–40, 124–25, 322–23 Cozumel, 113, 114, 115 Crèche sets, 113 Credit, 147 Cuevas family, 166 Culture, semantic categories of, 13–14 Culture change, 42; and technological choice, 11–13 Databases: genealogical, 24–25; potters, 26–27; production-unit, 25–26 Day of the Dead rituals, 51, 119, 120, 323; pottery sales for, 134, 136, 143; pottery used in, 71, 107–10, 124, 313; sales to brokers, 144–45 Decoration, changes in, 312–13 Deforestation, 283 Demand, 27, 93, 145; for cooking pots, 110–12; cycles of, 119–20; for Day of the Dead pottery, 107–10; as feedback mechanism, 94–95; and inventories, 120–21; for plant pots, 117–19; tourism, 113–17; types of, 19–20; for water vessels, 102–7 Depositos, 102 Diffusion, 273–74 Disease, mortality from, 67 Disk, rotating, 234 Distribution, 27–28, 29, 127, 150–51; by brokers, 142–48; to fiestas, 137–38; to markets,

Index

133–37; pottery, 94, 312; production and, 148–49; to stores, 138–42; and transportation infrastructure, 128, 131–33 Division of labor, 2, 49(table), 76 Durazno, 102 Durkheim, Émile, on socioeconomic complexity, 2, 3 Duzununcán, 35 Dzitbalché, 160; clay from, 162, 169, 169–70, 178, 180, 188, 222; store in, 149, 150 Earring holders, 117 Echeverria, Luis, 131 Economies of scale, 7, 201; efficiency and, 9–10; production-unit size and, 183–84 Economy, 153, 310; and craft production, 75–78; hurricanes and, 119–20; pottery’s role in, 93–94; Yucatán, 32–33; Yucatec Maya usage, 80–81 Education, effects of, 79–81 Efficiency, 8, 186, 274–75, 318–19; and economies of scale, 9–10; kiln technology, 293, 300 Ejidos, 69, 160, 195 Elites, 4; clay source control, 157–59, 164–65, 184–86, 188–89, 218 Engagement theory, 14–16 Epidemics, in Yucatán, 67 Epoch of Slavery, 67–68, 69 Espadas, 78 España, Rene, 246 Espejo, Juaquín, 157 Ethnoarchaeology, 14, 20 Ethnographic analogy, 309–10; limits of, 17–21 Evolution, sociocultural, 9–11, 20–21, 67 Fabrication techniques, 9, 315–16, 325; evolution of, 232–37, 273–76; intensity and scale in, 278–79; multiple, 276–77. See also by type Factories, 7; slip casting, 262–65 Fairs, pottery sales at, 138 Families, 38, 109; clay mining, 170, 171–72; land ownership, 50–51; learning in, 42, 44, 49–50; pottery making, 53(table), 69–70; pottery marketing, 135–36; production units, 316–17 Feedback mechanisms: demand as, 94–95; social choice and, 230–31

Fieldwork, methodology, 21–24 Fiestas, pottery marketing at, 137–38 Figurines, mold-made, 255–56, 316 Finishing marks, on mold-made vessels, 251, 252 Firewood: and kiln size, 303–4, 306–7; procurement of, 282–84, 290; in updraft kilns, 293–94 Firing, 47, 183, 281–82; specialization, 286–288, 290, 301–2, 320; in updraft kilns, 291, 293; vessel types and, 284–85 Flowerpots, 117, 122(table), 123(table), 324; for Day of the Dead ritual, 109, 313 Food bowls, Day of the Dead, 46, 71, 108, 109–10, 120, 313 Fuel resources, 6. See also Firewood Fund for the Promotion of Tourism Infrastructure (INFRTUR), 115 Garma, Enrique, 138, 139, 160 Gender, and pottery manufacturing, 73–75, 78, 88–89 Genealogical data, 24–25 Geology, 124; clay procurement, 154, 164; Yucatán, 95–96 Globalization, 20, 310, 320 Gourds, in Day of the Dead rituals, 108 Gremio, 36–37 Griddles, tortilla, 111, 313–14 Guardians of the fields (alux), 113 Guatemala, 4, 23, 102 Guatemala, Valley of, 5, 7, 73, 185 Haciendas, 69, 68 Hacienda Uxmal, pottery for, 113, 114, 115, 141 Hand-forming techniques, 238 Henequen industry, 32, 33, 68, 128, 157 Herrera, Humberto, 158 Hi’, in cooking pots, 192–93 Highways, 128–30, 130; pottery distribution and, 131–33 Hill ridge, 97; and water sources, 95–96, 99 Hocabá, 112 Holy Christ of the Blisters, 36 Holy Week, pottery demand for, 119 Hopelchén, 129 Hotel Principe, 113, 237 Hotels, pottery production for, 113, 115, 118, 119, 271, 324

347

Index

Households, 3, 59, 70; labor, 90–91; land acquisition, 50–51, 52; land inheritance, 60–61, 89; learning in, 42, 44, 314–15; pottery production, 7–8, 72–73, 314–16; as production units, 38, 87–88, 316; social interaction, 56–58; women’s role in, 73–75 Huari, 5–6, 19, 321 Huchim family, 78 Hurricanes, and pottery demand, 119–20 Hyatt Hotel, 141 Hydrology, Yucatán, 95–96 Ibibio, 36 Ichmul, 36 Identity, Yucatecan, 31–32 Imitation, learning through, 44–45 Immigrants, pottery manufacture, 77–78 Incense burners (incensarios), 108, 109, 266, 313 Inflation, 79 Infrastructure, pottery-making, 70–72 INFRTUR. See Fund for the Promotion of Tourism Infrastructure Inheritance, 70, 89; land, 50, 51, 60–61 Innovation, 273–74 Intensity, 8–9, 278–79, 323 Interaction distance, 57–58 Interaction frequency, 56–57 Interdependency, economic, 153 Inventories, 120–21 Isla Mujeres, 113, 115 Izamal, 106 Jarras, 102, 266, 268, 270 Jars. See by type José, 168–69 K’abal. See Turntables Kalinga, 36 Karst topography, 96 Kilns, 71, 100, 112, 285, 289; changes in types of, 290, 300, 304–6; gas, 296–97; hybrid, 298, 299; ownership and maintenance, 286–87, 313; and production units, 38, 302–4; small updraft, 291–94, 294; square and pot, 295, 296 Kinship, 7, 39, 41, 300; potters, 36, 53(table); pottery marketing, 135–36; pottery production, 72–73, 315, 321–22; and production units, 54–55, 61, 66, 316–17

348

Labor, 2, 79; forced, 67–68; hiring of, 54–55; household, 8, 90–91; in large production units, 48–49; non-household, 55–56, 56, 57; recruitment of, 7, 39, 323 Lacandon, 310 Lamanai, 125 Land tenure, 185; households and, 50–51, 52; inheritance and, 60–61; temper procurement and, 195–96, 217–18 Language, 33, 36; in cognitive anthropology, 13–14; Spanish vs. Yucatec Maya, 79–80 Learning, 11, 40, 42, 43(table), 52, 56, 72; family-based, 49–50; household-based, 314–15; process of, 44–47; and production-­unit size, 47–48 Maize, cultivation of, 171–72, 187 Mama, 130, 284; potters in, 35, 36, 71, 73, 314, 323 Markets, 150; brokers, 142–48; for Day of the Dead vessels, 109–10; pottery sales at, 133–37; stores as, 138–42 Marqués, Felix, 157 Marriage, 47; residence and, 51–52 Matrilocality, marriage and, 51–52 Maxcanú, 35, 73, 107, 131 Maya Blue, 196 Maya Cement Company, palygorskite site, 194–96 Maya Riviera, 32, 311 Measurements (calcular), of traditional vessels, 47, 266, 268–70 Medina, Elario, 158, 165 Medina, Lucas, 195 Mejorada, 138, 165, 166 Mérida, 31, 68, 71, 81; brokers from, 143, 144, 147; corporate development in, 32–33; highways, 128–29, 129, 130, 131; stores in, 140–41; tourist market, 113, 118, 136, 145, 150 Methodology, fieldwork, 21–24 Mex, Cesario, 238, 246, 247 Mexican Revolution, 69, 78 Mexico, pottery making in, 10 Mexico, Valley of, pottery production in, 182–83 Middle Horizona (Huari), 5–6 Milperos, firewood procurement, 282, 283 Miners, mining: clay, 15–16, 35, 155–59, 160–64, 166–67, 170–78, 180–82,

Index

186–89; Maya Cement Company, 195–96; temper, 194–95, 197–204, 215, 217, 219, 220; temper preparation by, 204–12 Mixco, 313–14 Moche, 316, 321 Modified coiling technique, 234, 236–37, 277; on potter’s wheel, 241, 242–43, 278 Molde, 234 Molds, molding, 46, 145, 249, 272, 276, 325; adoption of, 245–48, 277, 279, 315–16; disadvantages, 250–56; efficiency of, 274–75; slip casting, 262–65 Molino, Oligario, 157 Monterrey, 197 Montes, Avelino, 157 Montes, Josefina, 157, 159 Montes family, 157 Motor habits, 15, 42, 269, 278; modified coiling technique and, 236–37; use of potter’s wheel and, 239–40, 242, 244–45 Muna, 130 Muscle strength, potter’s wheel use, 240, 242, 244–45 Nationalization, 158 Nazca, 321 Neolocality, 52 Nohcacab, Rebellion of, 100 Oaxaca, moldes used in, 234 Observation, 45 Orders, broker-originated, 145–46 Oxkutzcab, 81, 98, 109–10, 135, 145 Painting, 145, 312, 320–21, 324 Palygorskite. See White earth Paste, 239; clay qualities, 222–24; composition changes, 226–27; mixing and wedging, 46, 224–25; preparation of, 221–22, 227–28; recipes, 225–26 Patrilineality, inheritance rights, 50, 70 Patrilocality, 51, 52, 89 Peasants, peonage labor of, 67–68 Peniche, Pepe, 157 Pepe, 159, 160, 161, 162, 168 Personnel, in production units, 54–55, 66 Peru, 4, 8 Peto, 104, 106, 110, 130; potters in, 35, 73; pottery markets at, 135, 137 Pisté, 71, 141

Pitchers (jarras), 102, 266, 268, 270 Plant pots (rectos), 47, 113, 117–19, 122(table), 124, 271, 320; markets for, 136, 137 Plates, 108, 113, 243–44 Playas, Terminal Classic, 99 Politics, of water, 100–101 Polychrome wares, Huari, 5–6 Population growth, 32 Potters, 71; access to clay sources, 157–63; choices exercised by, 220, 229–30, 265; databases on, 26–27; education of, 79–81; as firing specialists, 286–88, 301–2, 306–7; gender of, 73–77, 88–89; genealogical data, 24–25; gremio of, 36–37; kinship of, 53(table), 54; learning process, 42–47; motor habits, 236–37; population of, 34–36, 65–66, 83–88; production units, 38–40, 41, 62–65, 317; residence proximity, 58–59; selective pressure on, 67–70; social interaction, 56–58; as temper miners, 198, 200; and temper quality, 212–15 Procreation, and personnel acquisition, 49–50 Production, 2, 33, 71, 109, 250; capital in, 82–83, 275; cooking pots, 110–11; cycles of demand, 119–20; and distribution, 148–49; gender and, 73–75, 88–89; homogeneity of, 231–32, 324; immigrants in, 77–78; intensity of, 318–19; kinship, 72–73; location of, 6–7; organization of, 2, 277, 321–22; scheduling, 8–9; sequence of, 28–29, 45–47; slash-and-burn agriculture and, 37–38; social continuity of, 40–42, 313–14; social unit of, 38–40; on wheel, 243–44 Production units, 25–26, 41, 122–23(table), 323; ball-bearing turntables, 275–76; changes in, 316–18; clay procurement and storage by, 182–84; and distribution, 132–33; education and, 79–81; households as, 72–73, 314–16; kiln ownership and maintenance by, 286–87, 302–4; locations of, 59–61; personnel in, 49–56; potters, 62–65; sizes of, 47–48, 78–88, 90; social context of, 38–40; transportation means, 142–43 Progreso, 128, 129, 136, 141 Pueblo Indians, 45 Pustunich, 51 Quinua (Peru), 5, 7, 19, 185, 230, 234

349

Index

Railroads, 128, 130, 134, 136–37, 312 Rainy season, 97–98 Ráquira (Colombia), 185 Raw materials, 230, 312; acquisition of, 48, 82; categorization of, 14–15; mining, 15–16; sources of, 71–72; Ticul potters’ use of, 34–35. See also Clay; Firewood; Temper Rectos. See Plant pots Religion, 224, 230 Residence(s): postnuptial, 51–52, 89; potters’, 58–59 Resources, and clay production, 153–54 Ritual vessels, 117, 122(table); Day of the Dead, 71, 107–10, 124; demand, 19–20 Roads, 128–30; improvement of, 131–32 Sacojito, 102, 124 Sah kab, 224–25 Sakel bach tuunich, kilns made from, 285 Sak lu’um. See White earth Salt and pepper shakers, 113 San Enrique, 138 San Ignacio Xtuk, Hacienda, 195 San Juaquín, 165 San Mateo Ixtatan, 45 San Sebastian, 182–83 Santa Elena, 106, 129, 284; Rebellion of, 100 Santa Rosa, 165 Scale, 7–8, 278–79 Scheduling: and broker sales, 145–46; pottery production, 8–9 Selective pressure, on potters, 67–70 Self-capitalization projects, 82–83 Shopping center, in Mérida, 33 Shoy, 106 Sisal, 128 Slip casting, 262–65, 274 Slipping, 46 Smith, Adam, 2 Social change, 32–33, 65, 265, 309, 311–14, 326; and clay sources, 154–70; distribution and, 150–51 Social choice, 229; feedback in, 230–31 Social class, of potters, 36 Social continuity, 40–42, 61; household personnel and, 49–56; pottery manufacture, 313–14 Social embededness, 27 Social interaction, households, 56–58 Social processes, 310–11

350

Socioeconomic complexity, 2–3, 186 Sotuta, 106, 137 Souvenir industry, 113–17, 123(tables), 141–42 Spanish, speakers of, 79–80 Spatial distribution, 6–7 Specialization, 2, 19–20, 220, 265, 277, 324; in clay mining, 170–78, 180, 181, 187–89; Costin’s parameters of, 3–10, 39–40, 124–25, 322–23; craft, 90–91, 153; development of, 319–22; firewood procurement, 282–84; firing, 286, 288, 301–2, 306–7; full- or part-time, 37–38; painting, 320–21; and production intensity, 318–19; production-­unit size and, 48–49; scale of, 89–90; temper mining, 198–204, 217–18; temper preparation, 204–15; vessel homogeneity, 231–32 Standardization, 3, 265, 270–71, 272, 324 Storage, 184; of molds, 253–54, 263 Stores, 148–50; pottery sales at, 138–42 Task segmentation, 46. See also Specialization Teaching, learning through, 45 Technological change, 273–74; kilns, 289, 290, 300, 304–6 Technological choice, 211, 219–20, 265, 324–25; cultural change and, 11–13; in fabrication techniques, 232–37; and social choice, 230–31 Tekax, 104, 110, 119, 135 Tekit, 130, 131 Temper, 6, 29, 38, 191, 239; calcite in, 102–3; cooking pots, 112, 192–93; mining of, 194–95, 198–204, 220; preparation of, 204–12, 219; production estimates, 215, 217–18; quality of, 212–15; sources of, 216(table), 223(table); white earth, 160–61, 193–94, 195–98, 319–20 Tepakán, 178; clay quality, 222, 227; clay sources in, 155, 162, 167, 168, 180; potters in, 35, 36, 73, 314 Terminal Classic period, 99, 125, 155; temper sources, 192, 196; at Ticul, 33, 35 Ticul, 1, 7, 14, 20: fieldwork in, 21–24; occupational groups in, 36–37; potters in, 34–35, 39(table); pottery market, 133–134, 134, 135, 136; stores in, 138–40 Tilam, Antonio, 198 Tinajas, 102, 103, 104, 237, 266, 270

Index

Tixmehuac, 106 Tizimín, 35, 106, 110, 137 Topography, Yucatán, 95–96 Tourism, 112, 271, 311, 320; corporate development of, 32–33; production for, 19–20, 55, 113–17, 119–20, 131–32, 141, 145, 312 Transportation, 148, 284; brokers, 142–43, 144; and distribution, 131–33; infrastructure for, 128–31; and pottery markets, 133–37, 312; of temper, 198, 200, 201 Tula, 316 Turntables, 240, 244, 274, 319; ball-bearing, 256–62, 275–76, 278–79; use of, 46, 47, 233, 233–37, 238 Tzucacab, 104, 110, 135 Tzum, Emilio, 158 Tzum, Guadalupe, 246–47 Tzum, Manuel, 247 Tzum Dzul, Eusevio, 69–70 Tzum family, 69, 274 Uayma, potters in, 35, 73 Uman, stores in, 141, 142 Uniformity, 248, 279; post-1970s vessels, 270–72; traditional vessels, 266–70 Union, clay mining, 158 Usufruct rights, 203 Uxmal, 55, 56, 113, 129, 141 Valladolid, 106, 110, 131, 136–37 Variability: of post-1970s vessels, 270–72; of traditional vessels, 266–70 Vases, 108, 255 Vertical integration, of production, 148–50, 321 Vertical shaft technology, clay mining, 176–78 Vessels: ritual, 107–10; types of, 46, 47, 48, 122–23(tables); variability, 266–72; water, 100, 101, 102–7. See also by type Villa Hermosa, 131 Vocation, choice in, 80–81 War of the Castes, 67, 69, 100 Wasters, analysis of, 221, 226

Water: availability of, 95–96, 98–99, 124, 311, 320; storage of, 106–7; transportation of, 101–2; wells, 99–101, 103–4 Water vessels, 48, 95, 100, 122(table), 123(table), 124, 134, 237, 311; demand for, 103–7; for storage, 102, 118, 136, 193, 288; for transport, 101, 101–2, 268, 269(table) Weather, Yucatán, 96–98 Weddings, 117 Wells: depth of, 103–4; impact of, 99–101 Wheel, 237, 274; advantages of, 242–45; disadvantages of, 238–42; for modified coiling technique, 241, 278 Whistles, Day of the Dead, 107, 108, 109, 313 White earth (palygorskite; sak lu’um), 285; Maya Cement Company and, 195–96; mining of, 197–204; as temper source, 161, 193–94, 204–12, 214, 214–15, 218, 219, 224, 319–20 Women: household role of, 73–75; pottery manufacture by, 76–77, 78, 89–90 Workshops: in Cancún, 161–62; governmentsponsored, 238, 245–46; painting, 139, 140, 143, 279; tourism and, 55, 113–17 Xtuk, Finca, 195 Ya’ax Che’, Hacienda, 68, 100–101, 284 Yo’ K’at, Hacienda, 38, 224, 225; clay mining at, 15–16, 35, 154–59, 162–63, 171, 172, 174, 174–78, 181, 186, 188, 218, 227, 230, 324; clay quality at, 226, 319; exhaustion of clay at, 163–64; ownership of clay at, 160–62, 166–67; sale of, 159–60 Yolakitak, 45 Yo’ Sah Kab: ritual associated with, 218–19; temper mining at, 35, 194–95, 195, 196– 98, 200–201, 202–3, 211, 215, 217, 324 Yucatán, 1, 7, 34; forced labor in, 67–68; independent identity in, 31–32; socioeconomic change in, 32–33; topography and hydrology of, 95–96; weather and climate of, 96–98 Yucatec Maya, 14, 15, 20, 33, 36, 80–81, 95

351