Page iii
Reconstructing Ancient Maya Diet Edited by Christine D. White THE UNIVERSITY OF UTAH PRESS Salt Lake City
...
190 downloads
1158 Views
6MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Page iii
Reconstructing Ancient Maya Diet Edited by Christine D. White THE UNIVERSITY OF UTAH PRESS Salt Lake City
Page iv
Disclaimer: This book contains characters with diacritics. When the characters can be represented using the ISO 88591 character set (http://www.w3.org/TR/images/latin1.gif), netLibrary will represent them as they appear in the original text, and most computers will be able to show the full characters correctly. In order to keep the text searchable and readable on most computers, characters with diacritics that are not part of the ISO 88591 list will be represented without their diacritical marks.
© 1999 by The University of Utah Press All rights reserved LIBRARY OF CONGRESS CATALOGINGINPUBLICATION DATA Reconstructing ancient Maya diet / edited by Christine D. White. p. cm. Includes bibliographical references and index. ISBN 087480602X (alk. paper) 1. Mayas—Food. 2. Mayas—Nutrition. 3. Central America— Antiquities. 4. Mexico—Antiquities. I. White, Christine, D., 1951 F1435.3.F7R43 1999 641.3'008997'4152—dc2l 9915542
Page v
To my father, Robert R. White, to whom I owe my curiosity and love of things ancient. I will remember him with all that I write. C.D.W.
Page vi
CONTENTS Acknowledgments
viii
Introduction: Ancient Maya Diet Christine D. White
ix
Part I: Botanical and Faunal Analyses
1. Plant Resources of the Ancient Maya: The Paleoethnobotanical Evidence David L. Lentz
3
2. Classification of Useful Plants by the Northern Petén Maya (Itzaj) Scott Atran
19
3. Continuity and Variability in Postclassic and Colonial Animal Use at Lamanai and Tipu, Belize Kitty F. Emery
61
4. Social and Ecological Aspects of Preclassic Maya Meat Consumption at Colha, Belize Leslie C. Shaw
83
Part II: Paleopathology
103
6. Land Use, Diet, and Their Effects on the Biology of the Prehistoric Maya of Northern Ambergris Cay, Belize David M. Glassman and James F. Garber
119
7. Dietary Change of the Lowland Maya Site of Kichpanha, Belize Ann L. Magennis
133
8. Caries and Antemortem Tooth Loss at Copán: Implications for Commoner Diet Stephen L. Whittington
151
9. Late Classic Nutrition and Skeletal Indicators at Copán, Honduras Rebecca Storey
169
5. Coming Up Short: Stature and Nutrition among the Ancient Maya of the Southern Lowlands Marie Elaine Danforth
Page vii
Part III: Bone Chemistry
183
11. The Elements of Maya Diets: Alkaline Earth Baselines and Paleodietary Reconstruction in the Pasión Region Lori E. Wright
197
12. Dietary Carbonate Analysis of Bone and Enamel for Two Sites in Belize Shannon Coyston, Christine D. White, and Henry P. Schwarcz
221
Glossary
245
Contributors
250
Index
251
10. Cuisine from HunNalYe David Millard Reed
Page viii
ACKNOWLEDGMENTS The production of this volume would not have been possible without the work and encouragement of many people, first and foremost the authors, who were punctual, patient, and always cheerful through its many stages. I would also like to thank Kim Law for her invaluable help in preparing the manuscript, and the Department of Anthropology of the University of Western Ontario for its support. And then there are the many Maya archaeologists who not only challenge us to elucidate the character of Maya society and the conditions of ancient life using human biology but also allow us to challenge archaeological theory in the search for an elusive "truth." Last, but not least, I am most grateful to Jeff Grathwohl, editor at the University of Utah Press, for believing that the study of diet has an important role to play in the understanding of a culture. CHRISTINE D. WHITE
Page ix
INTRODUCTION ANCIENT MAYA DIET Christine D. White Food is a vital component of both human biology and culture. It is necessity and pleasure, macronutrient and metaphor. It is nature nurturing nature, and it is nature transformed by culture into culture. Just as our dietary evolution characterizes our anatomy, so is the global diversity of foods we consume a reflection of our cultural adaptability. Culture is identified, among other things, by a shared repertoire of food choices, preparation techniques, and ritual and everyday food behaviors that have symbolic meaning. In addition to the broad distinctions that exist between cultures, there are, within cultures, distinct groups (e.g., socioeconomic, religious, occupational, gender, age) that may further express their identity through selective cuisine and food consumption behavior. At the most minute level of variation, there are individuals who have food preferences that serve as an expression of personal identity. Overlying this synchronic complexity in diet and food behavior is the inevitability of change—ideological, environmental, and technological—which produces new sets of food consumption patterns and habits. Because food behavior and the vital necessity of eating articulates with cultural identity and process, the reconstruction of systematic and idiosyncratic diversity from the archaeological record should be a primary objective for those who wish to understand both ancient cultural ideology and the relationship of culture, environment, and biology. Eating constitutes an act of belief transformed into meal, environment transformed into menu, and technology transformed into biology. Reconstructing the diet of any ancient culture, therefore, provides theoretical and methodological means of framing both synchronic snapshots and diachronic process. For these reasons alone, the reconstruction of Maya diet is a worthwhile pursuit. But the way we have come to understand ancient Maya foodways also provides a good demonstration of how both theory and data come to be resituated through processes of natural paradigm shift and new discovery. Initial interest in Maya subsistence can certainly be seen as a reflection of a generalized concern for reconstructing foodways which was rooted in the processual approach to archaeology arising in the 1960s and 1970s. Motivations for reconstructing Maya subsistence were particularly strong, however, because they were concomitantly propelled by the discovery of heavy population densities surrounding Maya ceremonial centers (Haviland 1969, 1970), densities that in many cases appear to have been greater than those existing today (Culbert and Rice 1990). Ethnohistoric accounts described slashandburn
Page x
horticulture as the primary subsistence technique (Landa 1566, in Tozzer 1941; Hellmuth 1977). These, combined with modern analogy, in which the same form of procurement is practiced, led archaeologists to believe that the continuity of procurement observed from the Colonial period to the present would also extend from the Colonial period backward in time. How such large ancient populations provisioned themselves posed two major problems of explanation. First, it was not believed that slashandburn horticulture produced enough food over the long term to support such large numbers of people. Second, how could the fragile tropical ecosystem survive the shifting and destructive nature of an extensive milpabased food economy? Thus, the reconstruction of subsistence practices became critical to understanding processes of development and decline in Maya culture. The emergence of these questions produced a duality in response: the production of theoretical models and a flurry of data collection. The dominant theoretical model of the time (and one that continues to have many adherents) posited an ecological explanation for the collapse of Classic Maya society (Willey and Shimkin 1973; Coe 1980; PodoLedezma 1985; Santley et al. 1986; Culbert 1988; Webster et al. 1992). Although the ecological model is contextualized within the contemporary preoccupation of Mayanists to explain the collapse, it takes its place among other explanatory models (Demarest 1992, 1993; Miller 1993; Suhler and Friedel 1992; Fash 1994). In the most simplistic form of this model the Maya were assumed to have outstripped their environment of sustaining food resources, the consequences of which were social, economic, and demographic decline. Set in the modern context of rising North American environmentalism, this research also took on a relevance to the continued survival of contemporary civilizations. The Maya were to teach us a lesson from the past. Meanwhile, new data on previously unrecognized environmental usage were elucidating alternate processes of Maya food production and changing our understanding of ecological relationships (Bronson 1966; Lange 1971; Wiseman 1972, 1985; Siemens and Puleston 1972; Turner 1974, 1978; Matheny 1976, 1982; Willey 1978; Healy et al. 1980, 1983; Turner and Harrison 1981, 1983; Puleston 1982). The agronomic status of the Maya quickly moved from horticultural to agricultural with the discovery of intensive production techniques such as raised fields, extensive hillside terracing, and water control and storage techniques. These subsistence methods had apparently gone into disuse by the time of the Spanish Conquest, and the technology was essentially lost to modern people in these areas (Hellmuth 1977; Lambert et al. 1984). The existence of intensive agriculture, even without the help of draft animals or plows, concomitantly implied the existence of population pressure (Sanders 1977; Turner et al. 1977). It illustrated an adaptive cultural response and altered the simplistic ecological explanation for the "collapse." Environmental degradation now needed to be contextualized within methods of intensive land use, some of which (e.g., terraces and raised fields) appeared indefinitely sustainable. Furthermore, the nature of the collapse as a geographically uniform experience had been questioned. Although the Maya clearly had the ability to produce
Page xi
much more food than previously thought, extensive research now suggests that intensive production techniques varied considerably by region (Pope and Dahlin 1989; Fedick and Ford 1990; Dunning and Beach 1994). This is consistent with the broad ecological diversity that exists in the Maya region. Although much effort has gone into the study of land use, environmental features, and cultural process relating to procurement, relatively less has been accomplished in the actual reconstruction of human diet. Unfortunately, procurement is often equated with consumption, but the leap from procurement methods to ingestion of specific foods is a big one. As useful as it is to reconstruct how the Maya used their land for food production, and the effect this may have had on ecology and political economy, the reality is that procurement studies tell us little about consumption patterns. To examine Maya culture from the inside out, we need to answer the basic questions of what was consumed by whom, when, where, and how. Such knowledge has been inhibited by a number of factors. First, tropical conditions generally result in poor preservation of data sources for diet reconstruction. Maya environments are full of destructive agents, such as heat and moisture that speed the rate of decay, heavy rainfall that flushes away tiny remains and promotes the breakdown of bone, insects and animals that feed on remains, and rapid root growth that breaks up and displaces plant and animal remains. Consequently, plant remains (macro and micro) are relatively rare, and bone remains (animal and human) are generally in poor condition and fragmentary. Second, archaeological research design in the past was largely focused on architecture, ceremonial centers, spatial distribution, and artifacts. Thus it tended to yield relatively little in biological sources for data. In spite of the difficulties in gleaning dietary data from Maya sites, greater effort has been made recently to do so. There are several reasons for movement in this direction. Some reflect the current state of physical anthropology and archaeological fashion in general. Other reasons relate specifically to changes in activity and theory in Maya archaeology. The last 25 years have seen a major shift in focus within skeletal biology which recognizes the importance of bone in reconstructing both life experiences and the effects of culture and environment on biology. Through this change in direction, archaeology has come to place greater value on the contribution of skeletal biology (human and faunal) to the understanding of ancient lifeways and cultural dynamics. For example, in Maya research most skeletal data were traditionally relegated to descriptive sections of site monographs. Where osteological work was used independently, it tended to be typological, emphasizing the description and categorization of cranial deformation and dental decoration or focusing on biological distance and/or epigenetic traits (see Buikstra 1997 for a quantitative historical review of topic trends). Nevertheless, it is significant that in spite of the preservation problems, Maya osteology was at the forefront of the movement away from traditional description and toward the healthrelated bioarchaeology that dominates North American research in skeletal biology. For example, with Haviland's (1967) work on
Page xii
stature at Tikal, we witness the emergent use of skeletal populations to reflect cultural change and complexity. In addition, Saul's classic work, The Human Skeletal Remains of Altar de Sacrificios (1972), was one of the earliest osteobiographies produced. More recently, techniques for collecting and analyzing data have become even more sophisticated. Furthermore, attempts have been made to standardize these processes (Buikstra and Ubelaker 1994), and interpretive theory is developing out of sampling and demographic issues (Wood et al. 1992). Paleodiet and paleopathology should be contextualized within this movement. The academic appeal of these fields is twofold. First, they encompass a diversity of sources and methodologies and thus, through multiple lines of evidence, can have great power of explanation. Second, they articulate with broader issues such as disease etiology, ecological change, demography, social relationships and complexity, and economy, issues that are also of interest to other Maya scholars. Paleodiet has been revitalized by the development of chemical techniques for analyzing bone (Gilbert 1977; DeNiro and Epstein 1978, 1981; Schoeninger 1979; Schwarcz and Schoeninger 1991). Both isotopic and trace element analysis of bone have several advantages over the more traditional techniques of faunal and floral analysis. They provide data on real rather than potential consumption (i.e., they give us the ''meal" rather than the "menu" [Bumsted 1985]), they obviate the need for individuals to be pathological before dietary assumptions can be made, and they can make use of fragmentary, nondiagnostic material. The lastmentioned is particularly significant for Central American skeletal populations. Although it is obvious that the ultimate goal of most archaeological dietary studies (including plant and animal analyses) is to reconstruct the diets of humans, it is only recently that we have been able to extract data directly from humans themselves. Because of some of the limitations of chemical data (e.g., see Sillen et al. 1989), however, their interpretation is best made in the context of either a priori knowledge of diet or multiple lines of evidence. The need to know diet before it can be reconstructed sets up an intuitively counterproductive "catch22" mode of explanation. Although methodology has an apparent weakness in this respect, the combined use of related methodology can ultimately result in greater depth of understanding. Data from discrete techniques will either be reinforcing or elucidate areas that need further study. Essentially, multiple lines of evidence end up testing one another and can dramatically improve the power of explanation. Reconstructing Ancient Maya Diet illustrates this process, in that floral and faunal analyses are integrated with human data from bone chemistry and paleopathology in addressing shared research issues. The nature of research in paleopathology has also changed over the last 15 years. Previously, it was characterized by description that may have been considered esoteric and by the "grocery list" appearance of case studies. As pathology was increasingly employed to address issues in anthropology and archaeology, the goal of using archaeological bone just to reconstruct pathogenic process was displaced, at least in part, by epidemiological and population
Page xiii
approaches. Significantly, paleopathology also played a key role in the development of a popular model for the Classic Maya collapse, one that continues to color our perceptions of ancient Maya living conditions. The discovery of a high frequency of the bony manifestation of irondeficiency anemia at the Cenote of Sacrifice (Chichén Itzá) spawned the idea that nutritional deficiency may have played a major role in the downfall of the Classic Maya (Hooton 1940). Cases of anemia subsequently found fueled ecological theories of collapse (Wright and White 1996), so that nutritional deficiency became equated with ecological degradation. Motivation increased to gather pathology data that would test the nutritional theory, but poor preservation continues to limit the potential of this avenue of research. In addition to the fact that there is a greater acceptance of the belief that biological data from plants, animals, and humans reflect cultural dynamics and issues, the level of activity in Maya archaeology has also increased in recent years (Buikstra 1997; Danforth et al. 1997). The romance and enigmatic nature of Maya culture has no doubt attracted researchers and created public popularity. However, the rise in productivity for Maya archaeology, and especially skeletal biology, may also coincide with the real and perceived loss of research opportunities associated with repatriation and reburial of native North American skeletal material. Furthermore, the development of Central American tourist industries that feature ancient history has promoted excavation (e.g., Ball 1993; Sheets 1992). Consequently, more material is now available for analysis.
New Paradigms and Paleodiet Fortunately for paleodiet studies, this amelioration of data gathering occurred at the same time that issues of subsistence in Mesoamerican environments became a focus of study. The positioning of these research contexts is further influenced by changes in the way archaeologists are thinking about the relationship between theory and data. During the process of reconstructing Maya culture, research design naturally frames interpretation and model building. There are four perceptual shifts in Maya archaeology to which dietary data, in particular, can make significant contributions. The first is the recognition that spatial variation exists in subsistence practice, demography, and sociopolitical and socioeconomic systems and that it needs to be understood as much as temporal variation does (Culbert and Rice 1990). Too often false analogies and assumptions led to the development of models that were weak or without depth. A case in point is the example of the unidimensional maize dependency model of Maya diet. Although there are some important exceptions that challenge the primacy of maize as the carbohydrate staple (Bronson 1966; Barerra et al. 1977; Puleston 1982; Atran 1993; McKillop 1994), maize is generally considered the foundation, and heart and soul, of Maya diet. But to assume that little variation existed over time and space in the absolute and relative quantities of maize consumed is to lose much definition of the character of Maya society. Maya sites differ enormously in
Page xiv
their biotic and abiotic environments, their demographic characteristics, their political economies, and consequently their food production, consumption choices and patterns, and resultant health status. This book urges the recognition and celebration of both geographic and temporal dietary variation among the Maya. The second is the move toward understanding cultural dynamics from the bottom up rather than from the top down (Sharer 1996; Fedick 1989; Lentz 1991). The main analytical unit has traditionally been monumental architecture (i.e., the top). While this was the case, perceptions of Maya culture could only be based on elite activities, thus creating a strong bias in our understanding of Maya life. Theoretically, change in food behavior of elites may or may not appear in that of other classes. Therefore, in diachronic perspectives particularly, foodways can constitute a measure of the social depth of cultural change. The movement to use households as analytical units allows a better reconstruction of daily life for what we might assume would constitute the bulk of the population and fits well with dietary measures of how the Maya filled the basic need for food. Thus, reconstructing consumption choices and patterns adds another stage of refinement to a "bottom up" perspective. It adds a different dimension to our understanding of temporal and spatial variability. The third shift is the recognition that social complexity needs to be defined more clearly and reconstructed more precisely for an understanding of the structure and function of Maya politics and economy (Chase and Chase 1992). The traditional view of Maya society as a twotiered system (Haviland 1970; Webster 1985) is now yielding to revisions that emphasize a multiclass culture (Chase and Chase 1992) that may not be distributed in a simple concentric hierarchical fashion around site epicenters. The identification of elites has largely been artifactually or architecturally based, but as Chase and Chase point out, these archaeological identifiers are less than consistent in their associations—for example, luxury goods may be found in burial contexts that do not represent increased energy expenditure. There has been some use of skeletal evidence in the identification of elites. This has been mainly limited to intentional and unintentional morphological alteration such as cranial deformation, dental modification, and stature (Haviland 1967, 1970). Based on the assumptions that status affects access to resources and that valued foods are culturally defined, food consumption is another useful means of determining social differentiation. Chemical analysis of bone, in particular, can directly characterize the diets of individuals and thus has the potential for identifying individuals or groups that have differential access to the most fundamental resources for life. In conjunction with other lines of biological evidence—for example, dental and physical health, faunal and floral remains, and archaeological evidence (e.g., grave goods, grave type, burial location)—dietary assessment can be an invaluable tool. Because of its independence of pathology or cultural alteration, diet has a potential to identify social patterning in a way no other line of evidence can, for example, sorting out status affiliation for multiple burials in elite contexts. The fourth shift is the postprocessual move to connect ideological with
Page xv
materialist or economic modes of explanation (e.g., Pyburn 1989). The study of food behavior is well situated to illustrate the integration of this apparent duality because of the association of food with metaphor and the fact that food is an aspect of material culture, often a major part of market and trade economies. The Maya themselves used food as metaphor. The Popul Vuh, a sacred Quiche Maya text, describes maize as the staple food from which humans were created and on which the Maya civilization was built (Béhar 1968). Food appears in Maya iconography, is depicted in art, and was an important part of ritual activity. It is unlikely coincidental that the Maya considered themselves religiously connected to food while the base of their economy was agriculture. Just as food was used by the Maya as a metaphor for their belief system, it can be used by us as a metaphor for understanding their culture. The selection of diet and the patterning of consumption through time, across space, and within populations can greatly extend our understanding of cultural behavior, values, and interactions. In terms of economic systems, the potential contribution of reconstructing foodways is mainly in understanding local and longdistance food distribution systems. Trade patterns have largely been understood through analysis of artifacts less perishable than foodstuffs. Nonetheless, food must have been an integrative aspect of ancient exchange, as it is today. And it may not have just been the material nature of food that moved across the landscape but its ideological representation as well.
Approaches to Maya Diet This book samples the breadth and depth of the contribution paleodiet research has made to Maya archaeology. In three sections—paleobotany and zooarchaeology, paleopathology, and bone chemistry—practical applications of diverse methodologies are used to address shared issues and show how the reconstruction of diet articulates with the new paradigms described above. Each chapter emphasizes its own methodological strengths and weaknesses. Because a primary goal of this book is to demonstrate levels of complexity in cultural behavior through dietary behavior, a deliberate attempt was also made to secure a variety of studies from different temporal and geographical contexts. Four main time periods are represented, periods for which dietrelated or subsistencerelated issues are of concern and for which dietary data exist in some form. They include the Preclassic (pre1000 B.C.A.D. 250), Classic (A.D. 250900), Postclassic (A.D. 9001500), and Historic (post1500) (chronology varies somewhat by site). Paleodiet research makes a contribution to two general concerns throughout this sequence. The first is the identification of trade networks. Chapters 3,4, 6,11, and 12 (Emery, Shaw, Glassman and Garber, Wright, Coyston et al.) deal with evidence of trade in a variety of time periods. The second is the characterization of social organization and complexity. In recognition of the importance of this kind of reconstruction, all the chapters (except one review chapter [Chap. 1, Lentz]) present data that represent degrees of dietary differentiation by socioeconomic status, gender, or age. In addition, each time period is associated with its own specific issues. Data
Page xvi
from the Preclassic period speak to both the rise of intensive agriculture to meet the demands of population increase, and the emergence and entrenchment of class differentiation. The dominant issues to be addressed by dietary data for the Classic period center on regional variability and the theories of ecological degradation and nutritional deficiency as explanations for the "collapse." Postclassic data provide a means of contrasting the effects of cultural process on diet with environmental interaction, as the political structures and economic bases of Maya culture undergo change. Postclassic populations represent cultural and biological survival and help define the conditions for cultural success. In the succeeding Historic period, issues include the effects of the Spanish presence on agricultural, biological, and cultural systems. Five chapters—3, 4, 7, 11, and 12 (Emery, Shaw, Magennis, Wright, Coyston et al.)—provide sitespecific diachronic comparisons of at least two time periods. Three chapters review data from multiple sites and time periods in an effort to differentiate general from specific patterns of nutrition (Chap. 5, Danforth) and plant use (Chaps. 1 and 2, Lentz, Atran). The sites studied and used for comparison in this text are dominantly located in the Maya Lowlands, that is, Guatemala, Belize, and Honduras (Figure I.1). The focus on the Lowlands is simply a reflection of the availability of bioarchaeological data. Although this region of Mesoamerica has a broadly defined tropical environment, within the Maya area more finely defined local environmental diversity exists. The selection of sites representing potential variation in food resources helps us understand the relative importance of panMaya cultural ideology in food consumption patterns versus food behavior on local polity and household levels. Future research will undoubtedly result in the ability to compare similarities and differences in highland and lowland areas as well. The book begins with the most traditional archaeological methods of diet reconstruction: botanical and faunal analyses. Both methods are absolutely fundamental to paleodiet research, creating a portrait of available biological resources. In essence, they provide us with ancient "menus" (Bumsted 1985), forming a comparative base for data derived from methods aimed at reconstructing "meals." It is appropriate that the opening chapter is a muchneeded and longawaited review of plant use among the Maya for more than 3,000 years. Significantly, David Lentz notes that we are now at the point in Maya archaeology where the techniques of paleoethnobotanists and the strategies of archaeologists have been synthesized sufficiently to provide artifactual (botanical) evidence to replace speculation about diet, ecology, and agriculture. While outlining general patterns of plant use, Lentz not only addresses some old issues, such as controversy over defining the staple of Maya diet, and ecological adaptation, but also makes suggestions for future research. His focus is on both agricultural and arboricultural dietary resources but also includes botanical resources that had other cultural uses or that may have been implicated in trade. Extending the use of ethnobotany in a more ideological and metaphoric way, Scott Atran provides us in Chapter 2 with linguistic and cognitive data that
Page xvii
Figure 1 Sites referred to in the text.
integrate with ethnohistorical and archaeological data on plant use among the Maya of the northern Petén in Guatemala. In a diachronic approach that reaches into modern times, Atran notes that the majority of edible and medicinal species used today were also used at the time of the Conquest and may have had continuity into more ancient times. He provides us with important reference tables of biologically useful plant species, their folk and scientific names, and their cultural uses. The data in this chapter are not only invaluable for their detail and completeness but also particularly important for posterity, as we are warned of loss of this knowledge in the very near future. In contrast, the two faunal chapters are site specific but make significant methodological and substantive contribution to our understanding of Maya diet. In Chapter 4 Leslie Shaw not only emphasizes the need to be aware of
Page xviii
sampling error in faunal analysis but also notes that analytical methods developed for cultures in temperate climates that have domesticated animals should not be applied to the Maya, who lived in tropical environments and appear to have subsisted on wild species and/or had very few domesticated species (e.g., dog, turkey). She emphasizes the importance of developing interpretive models to suit Maya circumstances. In keeping with the general shift in archaeology to examine culture from the bottom up, Shaw examines how faunal use at the household level reflects the emergence of trade and of social and economic inequality in the Preclassic period at Colha, Belize. With a similar awareness of how the complexity of tropical ecosystems might affect the analysis of faunal material, in Chapter 3 Kitty Emery offers a complex methodological approach using multiple measures that help control for sampling error while creating finer reconstructive detail. She analyzes faunal use diachronically at two related sites in Belize (Lamanai and Tipu), for which rare data spanning the Postclassic to Colonial transition are available. In contrasting the patterning at the two sites, Emery finds both continuity and variability in the use of animals. She analyzes animal use to elucidate the economic strategies of trade networks and the influence of the Spanish on the use of food as ethnic identification. Thus, she is able to connect ideological with economic and material considerations in her interpretation of data. The section on paleopathology is devoted to the reconstruction of diet through measures of physiological response in human skeletal remains. Here we are getting closer to actual food consumption by indirect assessment of nutritional quality of diets. Four out of five chapters in this section include data on dental pathology. The importance of teeth in diet reconstruction should not be underestimated. They are the most basic of all food processors. Everything we eat passes over them, and the evidence of this is left behind in dental disease and morphological alteration. But there is another reason for the dominance of dental analyses in Maya pathology. Because of preservation problems in the Maya area, teeth often provide the most reliable, and largest quantity of, data. Reflecting the methodological improvement in osteological research, each of the pathology chapters attempts to maximize the power of explanation in the data. This has been done by (1) using multiple lines of evidence from both teeth and bones (Chaps. 6 and 9, Glassman, Storey), (2) integrating data from one line of evidence with research done by others at a single site (the work at Copán is a good illustration of this kind of strategy—see Chaps. 8 and 9, Whittington, Storey), (3) using intersite comparison (Chaps. 5 and 6, Danforth, Glassman and Garber), (4) using intertemporal comparison (Chaps. 5, 6, 7, and 8, Danforth, Glassman and Garber, Magennis, Whittington), and (5) using powerful statistics (Chap. 8, Whittington). Sites appearing in this section vary in location, time period, size, and political importance, thus providing some crosssectional perspective. The longest time sequence for these sites is found at Kichpanha, Belize, stretching from the Protoclassic to the Late/Terminal Classic (250 B.C. to 900 A.D.) (Chap. 7, Magennis). Three Classic period phases are represented at Copán, Honduras (Acbi, Coner, and Late Coner) (Chap. 8, Whittington), and the Late and Termi
Page xix
nal Classic periods are represented at Ambergris Cay (Chap. 6, Glassman). Notably, all these sites were abandoned by the end of the Classic period. The data are skewed to periods preceding the Postclassic and Colonial periods because recent research on human skeletal material from these periods is as scarce as excavated samples. Environmental contexts vary from coastal (Ambergris Cay—Chap. 6, Glassman) to inland lacustrine with proximity to wetlands (Kichpanha—Chap. 7, Magennis) to river valley with restricted agricultural land (Copán—Chaps. 8 and 9, Whittington, Storey). Given the importance of ecological degradation as an explanation for the Classic collapse, a comparison of human biological response to diets derived from diverse physical environments is crucial to understanding cultural processes. Each site is a different size and appears to have a different political structure. Ambergris Cay (which actually includes the two sites of San Juan and Chac Balam) is the smallest of the three areas examined. Its physical structures are distinct from the other two sites, and its sociopolitical structure appears to have been based on trade (Chap. 6, Glassman). Kichpanha (Chap. 7, Magennis) is probably typical in size and political structure of the many mediumsized ceremonial centers of the tropical lowlands. Copán (Chaps. 8 and 9, Whittington, Storey) is a very large and powerful polity with a highly complex political structure. If we are going to understand similarities and differences in food behavior as they may relate to political structure and economy, we must work with contrasting samples such as these. The paleopathology section starts with a review of Maya stature data (Chap. 5, Danforth). It is fitting to begin with stature, as it historically represents one of the earliest attempts to use biology to illustrate social complexity and the effects of cultural and environmental change (Haviland 1967). Marie Elaine Danforth applies modern concepts concerning genetic and environmental factors involved in growth to stature as a measure of adaptability in the Maya. Using virtually all the published data currently available, she apprises us of areas of weakness in stature data, examines evidence for sexual dimorphism and discusses its implications, and identifies regional variation. Most significantly, Danforth warns us against using existing stature data to make broad generalizations, especially as they pertain to the Classic collapse. In Chapter 6 David Glassman and James Garber provide us with a good methodological model to apply to small samples. They reconstruct health and nutritional status at two closely related sites on Ambergris Cay in northern Belize using multiple lines of evidence that include mainly nonspecific pathology. Because it is not possible to assess chronological trends within their sample, Glassman and Garber compare their data with those of contemporaneous inland sites. They note a difference in the patterning of dental pathology, from which a distinctive diet is inferred. The Ambergris populations are set apart not only by diet, size, and sociopolitical structure but also in patterns of health. It appears that good health was not sufficient to maintain survival in a political economy based on trade, as the cay was abandoned after trade was restructured during the Postclassic.
Page xx
In a diachronic study of dental caries and calculus (Chap. 7), Ann Magennis investigates diet change at Kichpanha, Belize. While emphasizing the importance of multiple lines of evidence in improving explanation, she integrates the dental data with a variety of botanical sources of data for the site. Although the frequency of caries and calculus increases over time, patterns here are not completely consistent with those of contemporaneous sites. On the surface, this phenomenon may appear to be due to regional differences, but Magennis aptly warns us of the possible effect of status and/or processing as confounding variables. In Chapters 8 and 9, respectively, Stephen Whittington and Rebecca Storey provide us with complementary studies of paleopathology at the large and complex site of Copán, Honduras, over a sequence of Classic period phases. These chapters reflect the need to know how the different social segments of a population adapt biologically in articulation with a shared culture, as they also document levels of stress from the bottom up. Using dental caries and antemortem tooth loss, Whittington examines the diet of commoners, who would have been closest to the production of food but who, he assumes, would have been less buffered from biological stress because of the structure of political economies. With sophisticated statistical techniques that are applied to a large sample and that factor in variables such as time, residence type and location, tooth class, age, and sex, Whittington shows us that although there is a sex difference in caries, stress was pervasive among commoner households at Copán at the time of the collapse. Notably, the dental data support his previously published demographic data (Whittington 1991). Comparing these data with those of other sites, Whittington emphasizes the existence of dietary heterogeneity by geographic location but asserts that Copán may have been more dependent on maize, which would be consistent with its more restricted environmental context. The crosssectional perspective on social rank at Copán is provided by Storey's analysis of multiple nonspecific health indicators. Here we have a good example of how social complexity is reflected by biological response. It appears that all segments of Copán society were affected by stress but that higherstatus individuals may have survived childhood stress more often. Thus, although even the elite exhibit high levels of morbidity, differential nutrition and living conditions protected them at least somewhat from mortality. Storey's interpretation illustrates well the recent theoretical perspectives of the "osteological paradox" (Wood et al. 1992) and is integrated with supporting data from stable isotope analysis of human bone (Reed, this volume) and botanical analysis (Lentz, 1991; this volume). Differences between status groups seem to be more marked than gender differences, which suggests a good degree of gender equality. However, not only is there is ambiguity in gender data within Copán (see also Whittington and Reed, this volume), which probably reflects the social complexity of the site, but there is also variation in the degree of gender differences between sites. Paleopathological analyses, by definition, rely on biological responses recorded in bone. Although studies of ancient health and disease may be able to
Page xxi
reveal nutritional deficiencies, in which case they indirectly reconstruct diet, conversely they may indicate normality. Nothing could be said about the specific composition of diet from osteological analysis if nutrition was adequate. Unlike faunal and botanical analyses, both pathology and bone chemistry analyses operate at the level of the individual consumer and eventually build a population perspective. Bone chemistry, however, also moves the level of analysis from menu (botanical and faunal) and biological response to actual consumption. One might ask why we don't simply begin with chemical analysis of bone and dispense with indirect measures. The answer is that the other methods give bone chemistry an interpretive base, potentially provide depth to our understanding of biological responses, and increase the power of explanation. Without pathology, we cannot speak of nutrition; without bone chemistry, it is hard to speak of diet. The studies in the bone chemistry section represent several different regions in the Maya Lowlands. Chapters 11 and 12 (Wright, Coyston et al.) compare diet derived from distinct ecosystems, and Chapter 10 (Reed) provides further contrast. This regional crosssection sheds light on the degree to which local environment can circumscribe food consumption. Chapters 11 and 12 (Wright, Coyston et al.) work with long chronologies. Thus we can examine the degree to which cultural process overlies or interacts with environmental restrictions. Because all the studies encompass the period of the Classic collapse, we now have another means of evaluating ecological models of the collapse. Adding another layer of complexity to this reconstruction, each chapter also examines social status. Socially meaningful patterning is found at each site, some of which is not discernible from artifactual or archaeological data. Although issues of environmental interaction, chronological patterning, and social rank are addressed in each chapter, each study represents a different type of chemical analysis. All take their primary data from human bone. The organic portion of bone (collagen) is analyzed isotopically for Copán (Chap. 10, Reed), the inorganic portion (apatite) is analyzed isotopically for Lamanai and Pacbitun (Chap. 12, Coyston et al.), and the inorganic portion is analyzed elementally for Altar de Sacrificios, Seibal, and Dos Pilas (Chap. 11 Wright). In spite of the fact that slightly different constituents of diet are reconstructed, each method is able to quantify maize consumption. Although restricting bone chemistry studies to a single method would provide the most soundly controlled intersite comparison, there are advantages to using the methodological diversity in this volume. First, a comparison of chapters illustrates the strengths and limitations of each chemical technique. Second, because each technique has its own way of measuring maize consumption, in effect we are still able to make comparisons of relative quantities of foods consumed over time and by intrapopulational variables. Third, within each study there is methodological control or balance. For example, carbon isotopes in collagen are more validly interpreted when combined with nitrogen isotopes. Here we are looking at two different measures of protein consumption (Chap. 10 Reed). Carbon isotopes in apatite give depth to the interpretation of carbon isotopes in collagen (Chap.
Page xxii
12, Coyston et al.). Here the apatite allows quantification of the proportions of macronutrients (fat, carbohydrates, protein), thus adding depth to the interpretation of collagen data, which reflects only protein consumption. Barium and strontium are both measures of plant versus meat and marine food consumption (Chap. 11 Wright). Reed's analysis of stable carbon and nitrogen isotopes in bone collagen at Copán articulates with and supports the pathology research done by Storey and Whittington, particularly in terms of dietary differentiation by socioeconomic status. The isotope data especially lend credence to the etiological explanation that the patterning in nonspecific health indicators is, at least in some part, dietary. But Reed's analysis takes us one step further by finding patterning that has not been detected in any other analyses, specifically rural/urban differences and agespecific gender differences. The isotopic data, therefore, add to our appreciation of social complexity at the large center of Copán. Using a novel approach to chemical analysis, Wright brings out the methodological sophistication and interpretive complexity of elemental analysis in paleodiet interpretation. Human bone data are compared to data for baseline ecosystem components, and the effects of different maize processing techniques are quantified, thus linking biology to technology. Overlying patterns created by postmortem chemical change (diagenesis), Wright finds patterning between three Petén sites—Altar de Sacrificios, Seibal, and Dos Pilas—which has implications for models of agricultural trade. Variation within these sites also reflects chronological trends that address the relationship between diet and agricultural economy from Preclassic to Terminal Classic periods. Again, social rank is dietarily defined. In a study that complements previously published analyses of stable isotope data for bone collagen (White and Schwarcz 1989; White et al. 1994), Coyston et al. isotopically analyze the mineral portion (carbonate) of human bones and teeth. Two sites in Belize, Lamanai and Pacbitun, which had contrasting survival success after the Classic period, are used to compare the relationships between diet and ecology, trade, culture change, social status, and gender. The role of maize versus marine resources in particular is reassessed in light of the macronutrient reconstruction made possible with the use of isotopic analysis of carbonate. Significantly, shifts observed in the faunal data for the Postclassic and Historic periods at Lamanai (Chap. 3, Emery) are supported by the isotopic data and seem to reflect social, economic, or political factors.
What Do We Know about Maya Diets? Although there is a great need for more research, these chapters tell us that although maize was the dominant staple, ancient Maya diets were far from simple. Thus, although there exists a panMaya ideological grounding for dietary regimes, there is little to offer at present in terms of universal dietary patterning within the Maya sphere. The amount of maize consumed and the foods consumed with it varied by location, time period, site size, social status, gender, and age.
Page xxiii
The majority of sites investigated here which represent more than one time period show dietary shifts of some sort (i.e., in plant or terrestrial versus marine animal consumption), but no sound generalizations can be made yet about increasing or decreasing maize consumption, for example, or about the effects of animal exploitation. The extremity and nature of temporal change in diet appears to depend on resource availability created by either local environments or ability to trade. Local ecology probably had a profound effect on food consumption. In fact, this is one of the few areas in which one can make some generalizations. Not surprisingly, coastal sites and those sites with more heterogeneous environments appear to have offered the best nutrition to their inhabitants, whereas areas with more tightly circumscribed environments and less production potential (e.g., Copán) produced more generalized health stress. Environment contributed to the longterm survival and physical wellbeing of Maya populations but did not completely determine it. Socially meaningful patterning in food consumption is found at each site, whether based in status, gender, or age differences. Although status differences in diet exist at all sites, these differences do not take the same form, that is, there is no single apparent'' highstatus diet." In some cases elites eat more maize, and in some they eat less. Therefore, elite diets are probably better defined within the context of local resources than by any strict cultural prescription. By contrast, gender patterning in diet is not found at all sites. Where it does exist, however, it operates like the patterning in diets of socioeconomic status groups, that is, there do not appear to be any rules governing what males eat versus what females eat. Male diets may more closely approximate highstatus diets, however. It is clear from the chapters in this book that Maya diets need to be interpreted first at the household, site, and polity levels before the complexity of Maya foodsystems can be fully understood.
Prospects: Where Do We Go from Here? In recent years archaeologists have improved the quality of our understanding of the Maya by using multidisciplinary approaches, applying new analytical technologies, reconstructing written history, and developing regionspecific models for cultural and environmental phenomena. The level of research activity has also markedly increased. Therefore, there is much reason to be optimistic about the expansion and clarification of our picture of the Maya. Perhaps osteology can now take its turn in contributing to our understanding ancient Maya culture. This volume provides a starting point through the collection and analysis of data on food behavior and the use of food as a metaphor for culture. The chapters in this book have illustrated that diet can be associated with social, political, economic, ecological, and nutritional factors and that Maya foodsystems were complex. They justify paleodiet as a research area with enormous potential to test archaeological hypotheses and to connect biology with culture. Although the lack of universality in Maya foodways which exists in the face of a symbolically based foodsystem may seem distressing at first glance, it
Page xxiv
is precisely the variability that gives definition and meaning to our understanding of Maya life and culture. Quite simply, complex societies (such as that of the Maya) should have complex foodsystems and exhibit considerable dietary diversity (Gummerman 1997). Our current and future challenge is to use this diversity to establish patterns of Maya culture and to try to find reasons for the particular form they take. Methodological advances such as those presented in each section of this volume will aid in this endeavor, but there is still a need to seek and develop new ways of collecting and analyzing data. The ability to address both longstanding and emerging issues in Maya archaeology naturally follows from the move to understand variation. Expansion and integration of archaeological and skeletal data are needed to continue clarification of complex issues currently existing in Maya archaeology, such as the existence of a Late Preclassic collapse, population pressure and the intensification of agriculture, the rise and maintenance of social status, the role of human influence on the environment, sociopolitical restructuring in the Postclassic, and the effect of the Spanish presence. The chapters in this book underscore this need by giving us a glimpse of the potential of integrated paleodiet studies. Dietary data also have a potential to address other issues that are important to archaeologists everywhere, such as gender, the identification of lineages within sites, the identification of foreigners in local populations, the establishment of marriage patterns, the relationship between mortuary treatment and social status, and the relationship among ideology, economics, and material culture. Perhaps more important, paleodiet research has a potential to test hypotheses not defined by archaeologists but useful to them, and to discover inter and intrasite variability that is often not detectable from the archaeological data alone. These chapters are presented as an example of the specific contribution dietary research can make to Maya archaeology, but I hope they will also be used as a model for a different way of addressing common issues for archaeology everywhere.
References Cited Atran, S. (1993) Itza Maya tropical agroforestry. Current Anthropology 34:633700. Ball, J. W. (1993) Cahal Pech, the Ancient Maya, and Modern Belize: The Story of an Archaeological Park. San Diego: San Diego State University Press. Barrera, A.; GómezPompa, A.; and VázquezYanes, C. (1977) El manejo de las selvas por los Mayas: Sus implicaciones silvícolas y agrícolas. Biótica (Mexico) 2:4761. Béhar, M. (1968) Food and nutrition of the Maya before the Conquest and at the present time. In Biomedical Challenges Presented to the American Indians. Scientific Publication No. 165. Washington, D.C.: Pan American Health Organization, pp. 114119. Bronson, B. (1966) Roots and the subsistence of the ancient Maya. Southwestern Journal of Anthropology 22:251279. Buikstra, J. E. (1997) Studying Maya bioarchaeology. In S. L. Whittington and D. M. Reed (eds.): Bones of the Maya. Washington, D.C.: Smithsonian Institution Press, pp. 221228. Buikstra, J. E., and Ubelaker, D. H. (eds.) (1994) Standards: For Data Collection from Human Skeletal Remains. Arkansas Archaeological Survey Research Series No. 44. Fayetteville: Arkansas Archaeological Survey.
Page xxv
Bumsted, M. P. (1985) Past human behavior from bone chemical analysis: Respects and prospects. Journal of Human Evolution 14:539551. Chase, D. Z., and Chase, A. F. (eds.) (1992) Mesoamerican Elites: An Archaeological Assessment. Norman: University of Oklahoma Press. Coe, M. D. (1980) The Maya. London: Thames and Hudson. Culbert, T. P. (1988) The collapse of Classic Maya civilization. In N. Yoffee and G. L. Cowgill (eds.): The Collapse of Ancient States and Civilizations. Tucson: University of Arizona Press, pp. 69101. Culbert, T. P., and Rice, D. S. (eds.) (1990) Precolumbian Population History in the Maya Lowlands. Albuquerque: University of New Mexico Press. Danforth, M. E.; Whittington, S. L.; and Jacobi, K. P. (1997) Appendix: An indexed bibliography of prehistoric and early historic Maya human osteology: 1839 1994. In S. L. Whittington and D. M. Reed (eds.): Bones of the Maya. Washington, D.C.: Smithsonian Institution Press, pp. 229261. Demarest, A. A. (1992) Ideology in ancient Maya cultural evolution: The dynamics of galactic polities. In A. A. Demarest and G. Conrad (eds.): Ideology and Pre Columbian Civilizations. Santa Fe: School of American Research Seminar, School of American Research Press, pp. 135157. Demarest, A. A. (1993) War, peace, and the collapse of a native American civilization. In T. Gregor (ed.): What Do We Know about Peace? New York: H. F. Guggenheim Foundation. DeNiro, M. J., and Epstein, S. (1978) Influence of diet on the distribution of carbon isotopes in animals. Geochimica et Cosmochimica Acta 42:495506. DeNiro, M. J., and Epstein, S. (1981) Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta 45:341351. Dunning, N. P., and Beach, T. (1994) Soil erosion, slope management, and ancient terracing in the Maya Lowlands. Latin American Antiquity 5:5169. Fash, W. L. (1994) Changing perspectives on Maya civilization. Annual Review of Anthropology 23:181208. Fedick, S. L. (1989) The economics of agricultural land use and settlement in the Upper Belize Valley. In P. A. McAnany and B. L. Isaac (eds.): Research in Economic Anthropology: Prehistoric Maya Economies of Belize, Supplement 4. Greenwich, Conn.: JAI Press, pp. 215253. Fedick, S. L., and Ford, A. (1990) The prehistoric agricultural landscape of the central Maya Lowlands: An examination of local variability in a regional context. World Archaeology 22:1833. Gilbert, R. I. (1977) Applications of trace element research to problems in archaeology. In R. L. Blakely (ed.): Biocultural Adaptation in Prehistoric America. Southern Anthropological Proceedings, No. 11. Athens: University of Georgia Press, pp. 85100. Gummerman, G. (1997) Food and complex societies. Journal of Archaeological Method and Theory 4:105139. Harrison, P. D. (1977) The rise of the bajos and the fall of the Maya. In N. Hammond (ed.): Social Process in Maya Prehistory. London: Academic Press, pp. 470509. Harrison, P. D., and Turner, B. L. II (1978) PreHispanic Maya Agriculture. Albuquerque: University of New Mexico Press. Haviland, W. A. (1967) Stature at Tikal, Guatemala: Implications for ancient Maya demography and social organization. American Antiquity 32:316325. Haviland, W. A. (1969) A new population estimate for Tikal, Guatemala. American Antiquity 34:316325. Haviland, W. A. (1970) Tikal, Guatemala, and Mesoamerican urbanism. World Archaeology 2:186197. Healy, P. F.; Lambert, J. D.; Arnason, J. T.; and Hebda, R. J. (1983) Caracol, Belize: Evidence of ancient Maya agricultural terraces. Journal of Field Archaeology 10:397410.
Page xxvi
Healy, P. F.; van Waarden, C.; and Anderson, J. J. (1980) Nueva evidencia de antiquas terrazas Mayas en Belice. America Indígena 40:733796. Hellmuth, N. M. (1977) CholtiLacandon (Chiapas) and PetenYtza agriculture. In N. Hammond (ed.): Social Process in Maya Prehistory. London: Academic Press, pp. 421428. Hooton, E. A. (1940) Skeletons from the cenote of sacrifice at Chichén Itzá. In C. L. Hay (ed.): The Maya and Their Neighbors. New York: AppletonCentury, pp. 272280. Lambert, J. D. H.; Arnason, J. T.; and Siemens, A. H. (1984) Ancient Maya drained field agriculture: Its possible application today in the New River flood plain, Belize, C.A. Agriculture, Ecosystems, and Environment 11:6784. Lange, F. W. (1971) Marine resources: A viable subsistence alternative for the prehistoric lowland Maya. American Anthropologist 73:619639. Lentz, D. L. (1991) Maya diets of the rich and poor: Paleoethnobotanical evidence from Copán. Latin American Antiquity 2:269287. McKillop, H. (1994) Ancient Maya tree cropping: A viable subsistence adaptation for the Island Maya. Ancient Mesoamerica 5:129140. Matheny, R. T. (1976) Maya lowland hydraulic systems. Science 193:639646. Matheny, R. T. (1982) Ancient Lowland and Highland Maya water and soil conservation strategies. In K. V. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press, pp. 157178. Miller, M. E. (1993) On the eve of the collapse: Maya art of the eighth century. In J. A. Sabloff and J. S. Henderson (eds.): Lowland Maya Civilization in the Eighth Century A.D. Washington, D.C.: Dumbarton Oaks, pp. 355414. PodoLedezma, L. F. (1985) Enfermedades transmitidas por el agua el colapso de la civilización Maya Clásica. Mesoamérica 10:391410. Pope, K. O., and Dahlin, B. H. (1989) Ancient Maya wetland agriculture: New insights from ecological and remote sensing research. Journal of Field Archaeology 16:87106. Puleston, D. E. (1982) The role of ramon in Maya subsistence. In K. V. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press, pp. 353367. Pyburn, K.A. (1989) Maya cuisine: Hearths and the Lowland economy. In P. A. McAnany and B. L. Isaac (eds.): Research in Economic Anthropology: Prehistoric Maya Economies of Belize, Supplement 4. Greenwich, Conn.: JAI Press, pp. 325344. Sanders, W. T. (1977) Environmental heterogeneity and the evaluation of the Lowland Maya civilization. In R. E. W. Adams (ed.): The Origins of Maya Civilization. Albuquerque: University of New Mexico Press, pp. 287297. Santley, S. R.; Killion, T W.; and Lycett, M. T. (1986) On the Maya collapse. Journal of Anthropological Research 42:123159. Saul, F. P. (1972) The Human Skeletal Remains of Altar de Sacrificios: An Osteobiographic Analysis. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 63, No. 2. Cambridge: Harvard University. Schoeninger, M. J. (1979) Diet and status at Chalkatzingo: Some empirical and technical aspects of strontium analysis. American Journal of Physical Anthropology 51:295310. Schwarcz, H. P., and Schoeninger, M. J. (1991) Stable isotope analysis in human nutritional ecology. Yearbook of Physical Anthropology 34:283321. Sharer, R. J. (1996) Daily Life in Maya Civilization. Westport, Conn.: Greenwood Press. Sheets, P. D. (1992) The Ceren Site. Fort Worth: Harcourt Brace Jovanovich. Siemens, A. H., and Puleston, D. E. (1972) Ridged fields and associated features in southern Campeche: New perspectives on the Lowland Maya. American Antiquity 37:228239. Sillen, A.; Sealy, J. C.; and van der Merwe, N. J. (1989) Chemistry and paleodietary research: No more easy answers. American Antiquity 54:504512.
Page xxvii
Suhler, C., and Freidel, D. (1992) The Selz Foundation Yaxuna Project: Final Report of the 1992 Field Season. Dallas: Southern Methodist University. Tozzer, A. M. (trans.) (1941) Landa's Relación de los Cosas de Yucatán. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 18. Cambridge: Harvard University. Turner, B. L. II (1974) Prehistoric intensive agriculture in the Maya lowlands. Science 185:118124. Turner, B. L. II (1978) The development and demise of the swidden thesis of Maya agriculture. In P. D. Harrison and B. L. Turner II (eds.): PreHispanic Maya Agriculture. Albuquerque: University of New Mexico Press, pp. 1322. Turner, B. L. II; Hanham, R. Q.; and Portararo, A. V. (1977) Population pressure and agricultural intensity. Annals of the Association of American Geographers 67:384396. Turner, B. L. II, and Harrison, P. D. (1981) Prehistoric raisedfield agriculture in the Maya Lowlands. Science 213:399405. Turner, B. L. II, and Harrison, P. D. (1983) Pulltrouser Swamp: Ancient Maya Habitat, Agriculture, and Settlement in Northern Belize. Austin: University of Texas Press. Webster, D. (1985) Recent settlement survey in the Copán Valley, Honduras. Journal of New World Archaeology 5:3951. Webster, D.; Sanders, W. T.; and van Rossum, P. (1992) A simulation of Copán population history and its implications. Ancient Mesoamerica 3:185197. White, C., and Schwarcz, H. P. (1989) Ancient Maya diet: As inferred from isotopic and elemental analysis of bone. Journal of Archaeological Science 16:451 474. White, C.; Healy, P. F.; and Schwarcz, H. P. (1994) Intensive agriculture, social status, and diet at Pacbitun, Belize. Journal of Anthropological Research 49:347 375. Whittington, S. L. (1991) Detection of significant demographic differences between subpopulations of Prehispanic Maya from Copán, Honduras, by survival analysis. American Journal of Physical Anthropology 85:167184. Willey, G. R. (1978) PreHispanic Maya agriculture: A contemporary summation. In P. D. Harrison and B. L. Turner II (eds.): PreHispanic Maya Agriculture. Albuquerque: University of New Mexico Press. Willey, G. R., and Shimkin, D. B. (1973) The Maya Collapse: A summary view. In T. P. Culbert (ed.): The Classic Maya Collapse. Albuquerque: University of New Mexico Press, pp. 457502. Wiseman, F. M. (1972) A model for increased productivity in Lowland milpa agriculture. Journal of the Arizona Academy of Sciences Proceedings 7:14. Wiseman, F. M. (1985) Agriculture and vegetation dynamics of the Maya collapse in Central Peten, Guatemala. In M. Pohl (ed.): Prehistoric Lowland Maya Environment and Subsistence Economy. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 77. Cambridge: Harvard University, pp. 6371. Wood, J. W.; Milner, G. R.; Harpending, H. C.; and Weiss, K. M. (1992) The osteological paradox: Problems of inferring prehistoric health from skeletal samples. Current Anthropology 33:343370. Wright, L. E., and White, C. D. (1996) Human biology in the Classic Maya collapse: Evidence from paleopathology and paleodiet. Journal of World Prehistory 10:147198.
Page 1
PART I BOTANICAL AND FAUNAL ANALYSES
Page 3
Chapter 1 Plant Resources of the Ancient Maya The Paleoethnobotanical Evidence David L. Lentz During the last 15 years there have been tremendous strides in the documentation of the sources of sustenance for the ancient Maya. Many of these advances have been made by combining paleoethnobotanical techniques with archaeological excavation strategies, and the results have been fruitful. Data from across the Maya realm have been generated by paleoethnobotanists working in close cooperation with archaeologists, so that now a discussion can begin based on recovered artifacts rather than speculation concerning the dietary practices, past ecology, and agricultural adaptations of the ancient Maya. This chapter is an effort to summarize the currently available paleoethnobotanical data and synthesize the corpus into a meaningful framework that addresses some old questions and provides some direction for future research. Methodology Most paleoethnobotanical studies are conducted using a methodical collection scheme whereby soils from archaeological units are sampled uniformly to examine the plant remains found within. This generally involves the collection of measured units of soil that can be analyzed for their content of macro and microremains. Macroremains, fragments large enough to be detected with the unaided eye, usually are processed through some type of flotation device that separates carbonized materials that float to the surface of a suspending liquid (water in most cases) from soil particles, rocks, and other debris that sink to the bottom. Sometimes deflocculants are employed to free charred remains from the soil matrix. Many devices and techniques have been devised for flotation, and thorough discussions of the various options can be found elsewhere (Pearsall 1989; Struever 1968; Watson 1976). Soil samples for analysis of microremains (pollen and phytoliths) should be collected along with flotation samples, but these must be processed using chemical extraction techniques and carried out in properly equipped laboratories (Bohrer and Adams 1977; Pearsall 1989). A combination of these data sets can be a powerful tool in the effort to elucidate plant use practices of the past; however, most of the securely identified food remains thus far retrieved from Mesoamerican archaeological sites have been derived from the analysis of macroremains, and that is the focus of the discussion here.
Page 4
Ancient Maya Subsistence The earliest archaeological plant remains from the Maya area were found in 3,000yearold deposits in Belize at the Cuello site (Miksicek et al. 1991), followed closely by other sites with Formative components, such as Copán, Pulltrouser Swamp, and Cerros. Accordingly, the discussion of dietary patterns of the ancient Maya, at least for the moment, begins with the Formative period because archaeobotanical data from earlier times are absent. Beginning with the Formative period and extending through the Postclassic, we can observe a more or less fully formed subsistence pattern based on domesticated plants and a host of wild or partially cultivated plants. Paleoethnobotanical remains from several Classic period sites—Cobá, Cerén, Dos Pilas, Wild Cane Cay, Copán, and, to a lesser extent, Tikal and Río Azul—have been examined and can furnish the background for an understanding of the food procurement strategies that took place during the florescence of Mayan culture at civicceremonial centers with large populations. Postclassic sites of Cihuatan and Naco provide insights into the subsistence practices of satellite communities after the Maya collapse. These patterns are described in general terms, yet there certainly were special localized adaptations to diverse habitats, such as the marine environment modification seen at Wild Cane Cay (McKillop 1994). One of the earlier concepts of ancient Maya subsistence, a dietary pattern with heavy reliance on maize, beans, and squash, has indeed been borne out by the paleoethnobotanical record. Evidence for maize utilization was recorded at virtually every site where systematic archaeobotanical surveys were conducted (Table 1.1). Most of the maize fragments identified have a morphological resemblance to the ChapaloteNalTel complex (part of a cluster of races that Benz refers to as the Isthmian Alliance [1986]), and there are few references to other kinds of corn. ChapaloteNalTel is a complex that grows well in warm climates at low elevations and would have been well suited to the Maya Lowland areas. The complex has a wide range of adaptation in terms of soil requirements and matures quickly (Wellhausen et al. 1957). The ears are short, with 814 rows, and have smooth, rounded kernels. This complex of maize races has been described largely from existing land races, but several examples have been identified from designs or impressions on ceramic vessels in Late Classic Monte Albán in Oaxaca, Mexico, and at the Augustin site in Guatemala (Wellhausen et al. 1952). Archaic period examples of archaeological ChapaloteNalTel maize were recovered from La Perra Cave in Tamaulipas, Mexico (Mangelsdorf et al. 1956), and dry caves in Chihuahua and Sonora (Mangelsdorf and Lister 1956). Clearly, maize of this complex was widely propagated throughout ancient Mesoamerica from early times and likewise was grown as a mainstay of the Maya. Other races of maize surely were exploited as well but have yet to be unearthed and described. A pervasive problem in the Maya area is preservation: the ravages of the wet and dry tropics cause many archaeological plant parts to deteriorate rapidly. The result of this depredation is that we do not find whole cobs with the key characteristics needed to determine racial origin in the Maya area as often as in some other regions where conditions are
Page 5
more favorable to preservation. Furthermore, information about maize in the highlands is sparse to nonexistent, largely a result of inadequate recovery efforts, so data regarding interregional variability for this valuable food source are lacking. Continued investigations should resolve these difficulties, and eventually sufficient data will accumulate to more adequately elucidate the maize races used by the ancient Maya. A thorough knowledge of early maize races will not only help define the origins of this vital domesticate but direct us to the seed sources and agricultural antecedents of the first Maya farmers as well. The common bean (Phaseolus vulgaris L.), the second part of the Mesoamerican food triad, originally domesticated in two areas, Middle America and the Andes (Gepts and Debouck 1991), appeared in the Maya area sometime between 945 and 340 B.C. (Miksicek et al. 1991) and probably was introduced much earlier. Bean remains at Lowland Maya sites are not common, largely because they do not preserve well. The parts most often found are the seed cotyledons, which under normal circumstances would be consumed and not discarded, unlike corncobs. An exception to this unfortunate situation was found at the Cerén site in northcentral El Salvador.1 Rapid deposition of volcanic ash resulted in excellent preservation of plant parts at this small farming village, which was inundated sometime around A.D. 600 (Sheets 1994). Much like Pompeii, the site was occupied at the time of a volcanic eruption, so plant parts and other artifacts were preserved in situ, providing us with perhaps one of our best views of agricultural and subsistence practices in Mesoamerica. Large handfuls of beans, both common beans (Phaseolus vulgaris L.) and sieva beans (P. lunatus L.) plus some wild relatives, were found in ceramic vessels and other storage units. Because different kinds of beans were mixed together, it appears that the Cerén farmers were not very careful about separating varieties of the cultigen (Lawrence Kaplan, personal communication 1996). Presumably, these mixed collections would be cooked or sown together as well. The Cerén bean collection is perhaps the largest from Mesoamerica and will tell us much about this valuable domesticate. In any case, the combination of beans with maize formed the nucleus of the diet of the ancient Maya and other New World indigenes. The synergistic effects of a diet of corn and beans have been well publicized. When consumed together, they provide all the essential amino acids for human nutrition. Where maize is deficient, as in tryptophan, beans provide an adequate supplement (Kaplan 1973). Corn soaked in limewater provides an essential boost in dietary calcium. Additionally, beans, with the help of their nitrogenfixing Rhizobium spp. symbionts, can produce highquality proteins even when grown in nitrogendeficient soils. This is a real asset in many of the less fertile areas of Mesoamerica, and it undoubtedly helped see the Maya through many plantings on exhausted soils. Squash is the third component of the Mesoamerican food triad. It is a good carbohydrate source and provides substantial quantities of A and B vitamins, niacin, pantothenic acid, calcium, and potassium (Dunne 1990). Squash also works well with maize as a crop because it can be planted between the stalks of 1
There has been a question as to the ethnic affinity of the Cerén inhabitants; they may not have been Maya. In any case, they shared in the Maya lithic and ceramic traditions and undoubtedly their agricultural practices as well (Sheets 1992). The same question may be raised for the occupants of other sites in El Salvador and the Naco site in Honduras. If they were not Maya, at the very least the site occupants were nearest neighbors with the Maya, had active exchange networks with them, and would have shared their floral germplasm.
Page 6 Table 1.1. Plant Remains from Maya Sites. Parts Founda
Useb
Locationc
Timed
Referencee
Taxon
Common Name
Agavaceae
Agave sp.
agave
8,9
5
11
2
10
Aizoaceae
Mollugo verticillata
carpet weed
1
?
1
1
1
Anacardiaceae
Anacardium occidentale
cashew
5
1,3
2,14
1,3
2,13
Astronium graveolens
frijolillo
5
2,3
10
2
9
Spondias sp.
hog plum
2,5
13
13,7,9,13,16
1,2
13,6,8,12,15
Metopium brownei
chechem
1?
4?
13
2
12
Annonaceae
Annona sp.
soursop
5
1,2
2,9
1
2,8
Apocynaceae
Aspidosperma sp.
malady
5
2,3
11
2
10
Stemmadenia sp.
cojeton
5
2?
2
1
2
Thevetia gaumeri
chilidrón
1
4?
13
2
12
Arecaceae
Acrocomia aculeata
coyol
2
1,7
1,36,8,10,11
1,2
1,35,7,9,10
Attalea cohune
cohune
1,2
1,7
3,4,7,10
2
3,6,9
Bactris major
jaucote palm
5
2,3,7
2,3,4,9
1,2
2,3,8
Bactris sp.
huiscoyol
2
1,7
1,2,10,14
13
1,2,9,13
Crysophilia argentea
escoba palm
5
2,3
2,7
1,2
2,6
Sabal sp.
botan
1,5,8
3
5,7,10
1,2
4,6,9
Asteraceae
sunflowers
3
1?
1,2
1,2
1,2
Baltimora recta
flor amarilla
1
?
14
3
13
Helianthus annus
sunflower
3
1?
16
1
15
Melampodium sp.
flor amarilla
1
?
7
2
6
Tithonia rotundifolia
9
9
11
2
10
Bignoniaceae
Crescentia spp.
calabash
4,5
1,2,6
3,7,9,11,16
1,2
3,6,8,10,15
Cydista diversifolia
?
?
13
2
12
Bombacaceae
Ceiba pentandra
ceiba
5
2,3
16
1
15
Pachira aquatica
provision tree
5
2,3?
9
1
8
Boraginaceae
Cordia sp.
siricote
1,5
13
2,5,7,9,10
1,2
2,4,6,8,9
Brassicaceae
Brassica sp.
mustard
1
1?
1
1,2
1
Burseraceae
Bursera spp.
gumbolimbo
5
24,9
2,7,10,13,16
1,2
2,6,9,12,15
Protium copal
copal
9
4,9
5,9
1
4,8
Caricaceae
Carica papaya
wild papaya
1
1
7
2
6
Chenopodiaceae
Chenopodium sp.
goosefoot
1
1
1
1
1
Clusiaceae
Rheedia intermedia
caimito
5
2,3
1,10
2
1,9
Combretaceae
Bucida buceras
bullet tree
1,5
2,3
5,7,9
1,2
4,6,8
Terminalia sp.
nargusta
5
2,3
7,9
1
6,8
(table continued on next page)
Page 7
(table continued from previous page) Parts Founda
Useb
Locationc
Timed
Referencee
Taxon
Common Name
Convolvulaceae
Ipomoea sp.
morning glory
1
4,9
13,14
2,3
12,13
Cucurbitaceae
Cucurbita moschata
squash
1
1,7
1,2,6
1,2
1,2,5
Cucurbita sp.
squash
1,4
1
1,2,5,7,9,11,13,14
13
1,2,4,6,8,10,12,13
Lagenaria sp.
gourd
4
6
1,11
2
1,10
Sechium edule
pataste
1
1
1
2
1
Cyperaceae
3
?
2
1
2
Cladium jamaicense
razor grass
1,9
?
2,9
1
2,8
Cyperus canus
sedge
8
9
10
2
9
Scleria sp.
sawgrass
1
?
1,2
1,2
1,2
Dilleniaceae
Curatella americana
sandpaper tree
5
2,3
7
1
6
Ebenaceae
Diospyros sp.
persimmon
1
1
5,13
1,2
4,12
Euphorbiaceae
Euphorbia sp.
spurge
1
4?
13
2
12
Jatropha gaumeri
pomolché
1?
?
13
2
12
Manihot esculenta
manioc
5,9
1,2,
2,11
1,2
2,10
Sapium sp.
5
2,3?
1
1,2
1
Fabaceae (sensu lato)
Acacia sp.
cockspur
5,9
2,3
2,7,13
1,2
2,6,12
Albizzia sp.
5
2,3
1
1
1
Cassia sp.
1
4?
1
2
1
Crotalaria sp.
chinchin
1
1
2
1
Dalbergia sp.
rosewood
5
2,3
1
2
1
Desmodium sp.
beggar's lice
1
Enterolobium sp.
guanacaste
5
2,3
2,9
1
2,8
Haematoxylon sp.
logwood
5
2,3,9
7
1,2
6,8
Hymenea sp.
guapinol
5
13
1,2
1
1,2
Indigofera suffructicosa
indigo
5
9
2
1
5
Inga sp.
bribri
5
13
2,9
1
2,8
Lysiloma sp.
tsolam
1
4?
13
2
12
Mimosa sp.
sensitive plant
1?
4?
14
3
13
Phaseolus lunatus
sieva bean
1
1
11
2
10
P. vulgaris
common bean
1
1
1,9,11,12
13
1,8,10,11
Phaseolus sp.
bean
1
1
1,2,13,14
1,2
1,2,12,13
Pithecellobium sp.
turtlebone
5
2,3
2,9
1
2,8
Pterocarpus sp.
bloodwood
5
2?
1
2
1
Vigna sp.
frijol
1
1?
1
2
1
Fagaceae
Quercus sp.
oak
5
2,3
1
1,2
1
Flacourtiaceae
Casearia sp.
wild lime
5
2,3
11,13
2
10,12
Muntingia calabura
capulin
1
1
11,16
1,2
10,15
Lauraceae
Nectandra sp.
aguacatillo
5
2,3
10,12
2
9,11
Ocotea sp.
aguacatillo
5
2,3
1
2
1
Persea americana
avocado
2,5
13
13,69,11,16
1,2
13,58,10,15
(table continued on next page)
Page 8
(table continued from previous page) Parts Founda
Useb
Locationc
Timed
Referencee
Taxon
Common Name
Malpighiaceae
Byrsonima crassifolia
nance
2,5
1,2
13,5,6
1,2
15
Malvaceae
Gossypium hirsutum
cotton
1,9
1,5,7
2,5,11,14
13
2,4,10,13
Sida sp.
escobilla
1
9
2,7,14,
1,3
2,6,13
Meliaceae
Cedrela mexicana
Spanish cedar
5
2,3
2,11
1,2
2,10
Menispermaceae
Cissampelos pareira
peteltun
1
4?
5
1
4
Moraceae
Brosimum alicastrum
ramón
1?
1
13
2
12
Cercropia peltata
trumpet tree
5
2
2
1
2
Ficus sp.
wild fig
1,5
1?,2,3,9
2,3,79,11
1,2
2,3,68,10
Pseudolmedia oxyphyllaria
manax, cherry
5
2,3
2
1
2
Trophis racemosa
San Ramón
5
2,3
7
2
6
Myrtaceae
Pimenta diioica
allspice
5
2,3,9
2,7,9
1
2,6,8
Psidium guajava
guava
5
2,3,9
2,11
1
2,10
Nyctaginaceae
Pisonia sp.
uña de gato
1
?
5
1
4
Onagraceae
Oenothera sp.
evening primrose
1
?
12
2
11
Oxalidaceae
Oxalis sp.
wood sorrel
1
?
9
1
8
Passifloraceae
Passiflora sp.
passion flower
1
1
1,2,5
1,2
1,2,4
Phytolacaceae
Rivitia sp.
tropical pokeweed
1
?
2,7
1,2
2,6
Pinaceae
Pinus caribaea
Caribbean pine
5
2,3
2,7
1,2
2,6
P. oocarpa
ocote
5
2,3
10,1
1,2
9,10
Pinus sp.
pine
5
13
1,12,14
13
1,11,13
Piperaceae
Piper sp.
cordoncillo
5
24?
2
1
2
Poaceae
Echinocloa sp.
barnyard grass
1
?
7
2
6
Paspalum sp.
virgin grass
3
?
1,2,7
1,2
1,2,6
Trachypogon
8,9
3
11
2
10
Zea mays
maize
6,7,9
1
114
13
13,613
Rhizophoraceae
Rhizophora mangle
red mangrove
5
2,3
9
1
8
Rosaceae
Prunus sp.
cerezo
5
13
11
2
10
Rubiaceae
Randia sp.
tujé
1?
1?
13
2
12
Hamelia patens
redhead
5
2
2,7,16
1
2,6,15
Sapindaceae
Cupania sp.
pava
5
2,3
10
2
9
Talisia oliviformis
kinep
1
?
2,13
1,2
2,12
(table continued on next page)
Page 9
(table continued from previous page) Taxon
Parts Founda
Common Name
Useb
Locationc
Timed
Referencee
Sapotaceae
5
13
3
2
3
Calocarpum mammosum
mamey
1,5
13
13,5,8,10,16
1,2
14,7,9,15
Chrysophyllum sp.
star apple
1,2
1
2
1
2
Manilkara achras
sapodilla
1,5
1,2
2,69
1,2
2,3,58
Mastichodendron capiri
tsabak
1,5
2,3
1,5
1,2
1,4
Solanaceae
Capsicum annuum
chile pepper
1,9
1
2,10,11
1,2
2,9,10
Capsicum sp.
chile pepper
1
1
5
1
4
Solanum sp.
nightshade
1,5
?
2,7,9
1
2,6,8
Sterculiaceae
Guazuma ulmifolia
wild bay cedar
1,5
2,3
2,7,10
1,2
2,6,9
Melochia sp.
escobilla
1
5?
7
2
6
Theobroma cacao
cacao
1,4,5,9
13
1,2,5,7,11,14,15
13
1,2,4,6,10,13,14
Tiliaceae
Corchorus siliguosus
escobillo
1
?
1
2
2
Typhaceae
Typha sp.
cattail
8
3?
7,9
1,2
6,8
Ulmaceae
Celtis sp.
hackberry
2,5
1
1,2,5,11
1,2
1,2,4,10
Verbenaceae
Cornutia pyramidata
zopilote
5
2?
17
2
6
Vitex sp.
fiddlehead tree
5
2,3
1,10,13
2
1,9,12
Vitaceae
Vitis sp.
wild grape
1
1
1
2
1
a
1 = seed; 2 = pit; 3 = achene; 4 = rind; 5 = charcoal; 6 = kernel; 7 = cupule; 8 = leaf; 9 = other.
b
1 = food; 2 = firewood; 3 = construction; 4 = medicine; 5 = fiber; 6 = container; 7 = oil; 9 = other.
c
1 = Copán; 2 = Cuello; 3 = Wild Cane; 4 = Tiger Mound; 5 = Cerros; 6 = Tikal; 7 = Pulltrouser Swamp; 8 = Colha; 9 = Albion Island; 10 = Dos Pilas; 11 = Cerén; 12 = Naco; 13 = Cobá; 14 = Cihuatán; 15 = Río Azul; 16 = Santa Leticia. d
1 = Preclassic; 2 = Classic; 3 = Postclassic.
e
1 = Lentz 1991a; 2 = Miksicek et al. 1991;3 = McKillop 1994; 4 =Cliff & Crane 1989; 5 = Turner & Miksicek 1984; 6= Miksicek 1983; 7 = Caldwell 1980; 8 = Miksicek 1990; 9 = Lentz 1994; 10 = Lentz et al. 1996; 11 = Lentz 1991b; 12 = Beltrán Frias 1987; 13 = Miksicek 1988; 14 = Hurst et al. 1989; 15 = Miksicek 1986.
Page 10
maize as it is growing. This excellent ground cover fills in areas of exposed soil to help eliminate weedy competitors and reduce erosion at the same time. The intercropping of maize and squash is a common practice among the Paya of Honduras (Lentz 1993) and other contemporary indigenous groups of Central America. Most of the cucurbit remains retrieved from Mesoamerican archaeological sites have been from the rind of the fruit (Table 1.1), a tissue layer that generally cannot be identified to the species level. Peduncles, the stem attachments to the fruit, and seeds are diagnostic parts that are most likely to be recovered. Unfortunately, no peduncles have been found, but several seeds of Cucurbita moschata (Duch.) Duch. ex Poir. have been identified. No other evidence for Cucurbita species has been recorded other than C. pepo L. pollen from Edzna (Turner and Miksicek 1984) and several seeds of the same species from Cerén (Lentz et al. 1996). The Cerén C. pepo seeds may have been modern intrusions and therefore are not included in Table 1.1. Other species of Cucurbita probably were used by the Precolumbian Maya, and additional research in the area should reveal the full complement of squashes eaten in the past. One reason squash seeds are not common at Maya sites is that they were probably targeted as a food source; they are a good source of oil and are delicious when dried, so they are unlikely entrants into the trash pile. It seems reasonable to suggest that maize, beans, and squash were grown in swidden, or shifting cultivation, fields away from the house compounds as is the general practice in Mesoamerica today. Evidence from the Cerén site, however, where a cornfield planted in neat rows was growing directly adjacent to a house compound, indicates that even the most reasonable assumptions can be erroneous, and the ancient Mesoamericans continue to show more complexity in their adaptive patterns than our simple models can accommodate. Additional intensive Maya farming techniques included terracing (Beach and Dunning 1995; Turner 1974), raised fields (Siemens and Puleston 1972; Turner and Harrison 1983), check dams (Turner and Johnson 1979), drained fields (Pohl et al. 1996), and other forms of hydraulic agriculture (Bloom et al. 1983; Matheny 1976; Scarborough 1991). Until recently, cultivated chile peppers (Capsicum annuum L.) have been almost invisible in the paleoethnobotanical record. Perhaps the earliest example of chile came from Phase II (Formative period) deposits at Cuello (Miksicek et al. 1991). One seed was found and was of such small size it may have been from a wild variety, C. annuum var. aviculare (Dierb.) D'Arcy & Eshbaugh. Another chile seed was found at Late Formative Cerros (Cliff and Crane 1989), and carbonized peduncles were identified at Late Classic Dos Pilas (Lentz 199b). This meager evidence might lead us to believe that chile peppers were not important to the Precolumbian Maya, but recent studies at the Cerén site present quite a different picture. Carbonized chile seeds, peduncles, and rinds were found in great abundance, especially in storage rooms and a kitchen, where they were hung from the rafters in large clusters (Lentz et al. 1996). Several ceramic vessels contained seeds that possibly were in storage for subsequent planting or consumption. Chiles, with their abundant vitamin con
Page 11
tent, were clearly an important component of the diet at the Cerén site, at least by Middle Classic times. Probably these were grown as house garden plants as is the practice among the Kekchi Maya of Guatemala today. Manioc (Manihot esculenta Crantz) has been hypothesized as a major crop for the ancient Mesoamericans (Bronson 1966), but little documentation for this practice has been found in the Maya area. Some carbonized manioc stems were found in Late Formative deposits at Cuello (Miksicek et al. 1991), and manioc root casts were uncovered at Cerén (Lentz et al. 1996) in a house garden near one of the domiciliary structures. The limited evidence for manioc cultivation is at least in part due to the double problem of poor preservation of root crops in seasonally wet areas and the difficulty in identifying the fleshy parts, especially after they have been crushed or carbonized. Also, manioc is usually propagated vegetatively without the use of seeds, so there are few opportunities for seeds to become preserved by usual means. Hather and Hammond (1994) also report manioc from Cuello based on electron micrographs of charred parenchymatous tissue, but secure identification of root organs is complete only with clearly discernible vascular tissue embedded within the parenchyma. In addition to manioc, Bronson hypothesized Maya exploitation of other root crops, such as cocoyam (Xanthosoma sp.), sweet potato (Ipomoea batatas [L.] Lam.), and jícama (Pachyrhizus erosus [L.] Urb.), but substantive evidence has yet to be identified. Cotton (Gossypium hirsutum L.) may well have been an indigenous crop in the Maya area. Not only was it used as a source of fibers for clothing and other purposes, but the seeds were also used as a source of oil. Confirmation of this was found in a metate trough adjacent to a food storage area at Cerén where 74 cotton seeds were being ground for oil extraction (Lentz et al. 1996). The oil may have been used as a base for paints, as a liniment, or for other purposes, but the archaeological context suggests the seeds were being ground for cooking purposes. If the oil had been used for cooking, it would have added fats to a diet that is in other respects deficient. One drawback to unrefined cottonseed oil, however, is that it contains about 0.36 percent gossypol (Smallwood 1978), a polyphenolic binapthalenedialdehyde. At high concentrations gossypol is toxic. For example, the LD50 (lethal dose in 50 percent of the cases tested) in rats is 2,870 mg/kg (Telek and Martin 1981). In other experiments a 15 percent diet of cotton meal with 1.1 percent gossypol caused growth suppression in baby chickens (Krishnamurthi 1954). If these results can be converted to human equivalents, a child with a diet of 45 percent cottonseed oil would suffer from stunted growth, or a 50kg adult who ingested 40 liters of cottonseed oil in one sitting would receive a lethal dose of gossypol. These possibilities are, at best, highly unlikely. More realistically, individuals probably consumed far less oil. Accordingly, the use of small quantities of cottonseed oil for cooking purposes may have been plausible in Precolumbian times. These calculations, however, do not take into account possible longterm effects of gossypol consumption. Gossypol does have an interesting side effect: it lowers sperm counts and has been used as a reliable male contraceptive in China with only mild side effects (Telek and Martin 1981). One can only wonder at the impact this may have had
Page 12
on Precolumbian society. One final note on cotton: because its natural habit is to grow as a shrub, it seems likely that cotton was grown as a perennial at Cerén and elsewhere in Mesoamerica. Another fiber plant, agave (Agave spp.), currently cultivated widely throughout Central America, was grown as a prolific home garden plant at Cerén. Casts of many of these plants define the location of an extensive, carefully tended, infield garden that was directly adjacent to one of the house compounds (Lentz et al. 1996). This garden and others like it were the source of raw material for the many examples of agave fiber twine found at the site. To color textiles and other artifacts, the Maya may have used indigo (Indigofera suffructicosa Mill.) and inkwood (Haematoxylon campechianum L.) as dye plants (Turner and Miksicek 1984). Numerous authors have hypothesized, in various ways, how the ancient Maya relied extensively on tree cropping (Caballero 1989; Dahlin 1979; Folan et al. 1979; GómezPompa et al. 1987, 1990; Lange 1971; Lundell 1938; Miksicek 1990; McBryde 1947; McKillop 1994; Netting 1977; Pérez Romero and Cobos P. 1990; Puleston 1968, 1982; Turner and Miksicek 1984; Wilken 1971; Wiseman 1978), and the paleoethnobotanical data largely support these assertions. The remains of many fully domesticated fruit trees—for example, avocado (Persea americana Mill.), cacao (Theobroma cacao L.), and cashew (Anacardium occidentale L.)— have been found in two or more sites. Other fruitbearing trees that existed somewhere along the continuum of wild tree to completely domesticated crop included such native species as mamey (Calocarpum mammosum Pierre), calabash (Crescentia spp.), guava (Psidium guajava L.), hogplum (Spondias mombin L.), nance (Byrsonima crassifolia [L.] H.B.K.), capulín (Muntingia calabura L.), and papaya (Carica papaya L.). These are fruit trees that could have been gathered from forest stands or may have been grown in dooryard orchards. Once again, for insights as to how this was accomplished in Precolumbian times, we can turn to Cerén, where economically important trees were planted in household courtyards. Branches, leaves, and cotyledons of avocado; abundant casts of whole guava fruits; branches and rinds of calabash; and stems and fruits of nance were arrayed across the activity surfaces of several house compounds. Clearly, the citizens of Cerén grew much of their fruit in infield gardens or orchards directly adjacent to their houses where they could be fertilized with household waste and night soil. One aspect of the ancient Maya subsistence pattern revealed from the archaeobotanical record which was not predicted by early scholars was the widespread use of palms. The most widely used palms were the indigenous coyol (Acrocomia aculeata [Jacq.] Lodd. ex Mart.),2 huiscoyol (Bactris spp.), and cohune (Attalea cohune [Mart.] Henderson) (Lentz 1990). Remains of all these have been recovered from numerous sites throughout the region (Table 1.l), often with more than one species being exploited at any one site. The Maya may have been interested in palm fruits because of their high oil content. Fats are an essential part of the diet; nutritionists recommend that 2030 percent of the caloric intake consist of fats (Dunne 1990). In addition to providing energy, fats 2
Acrocomia aculeata is often listed in botanical and archaeological literature as Acrocomia mexicana.
Page 13
act as carriers for the fatsoluble vitamins A, D, E, and K; aid in the absorption of vitamin D; and help convert carotene to vitamin A (Dunne 1990). An examination of Table 1.1 reveals that fat sources from plants, with notable exceptions of mamey, squash seeds, and avocado, were limited for the Precolumbian Maya, a fact that was especially problematic in light of their meager assortment of domesticated animals. Although the Maya domesticated turkeys, Muscovy ducks, and dogs, most of their animal meat came from whitetailed deer, generally regarded as a lean meat source. All the palms listed in Table 1.1 are native to Mesoamerica, are good oil producers, and could have been exploited as wild plants. It seems likely, however, that at least one species, coyol, was cultivated during prehistoric times by peoples of central Mexico (Smith 1967) and the Maya, who were responsible for its introduction to the Copán region (Lentz 1991a). The Copán Maya recognized the value of the palm and cultivated it to enhance their nutritional options. Curiously, there appears to have been no extensive use of palms at Cerén as a food source. The explanation may be that the Cerén inhabitants had no need for palms because they obtained their oil from other sources. In short, each community in the ancient Maya realm needed a good source of fats, and the oil of palm could have filled this dietary need. Of course, when the Spaniards arrived in the sixteenth century, the Mesoamerican diet changed rapidly to a reliance on fats from a new source: the adipose tissues of Old World domesticated animals. In spite of the growing body of paleoethnobotanical data, a number of widely used New World domesticates have not been found at Mayan archaeological sites. Amaranth (Amaranthus spp.), tomato (Lycopersicon esculentum Mill.), and tobacco (Nicotiana spp.) have all escaped discovery. Wiseman (1983) claimed to have retrieved amaranth pollen from Pulltrouser Swamp, but his identifications were tentative. This is understandable, because the genera in the Amaranthaceae have morphology that is difficult to distinguish. Moreover, the seeds of amaranth, tomato, and tobacco are tiny and may simply have escaped notice by field collectors. Continued searching with the use of fine screens (mesh size < 0.5 mm) to collect flotation material should resolve this problem. Another food source of the ancient Maya that has been discussed for decades is ramón (Brosimum alicastrum Sw.). The concept that the fruit of this tropical forest tree was a staple for the prehistoric Maya was suggested first by Lundell (1938), later embellished by Puleston (1968, 1982), and recently resurrected by Atran (1993). Ramón is edible, and many indigenous people of Mesoamerica continue to consume it (Alcorn 1984; Lentz 1993), but more as a famine food and not as a major foodstuff. The paleoethnobotanical record reflects little use of the plant in the past. Of all the studies represented in Table 1.1, only one seed of ramón has been recorded (from the Cobá site). The seed of ramón is not small nor is it delicate, and it seems that if it were being used in any significant way, substantial evidence would have been uncovered by now. At the Cerén site, where preservation is as favorable as one can expect in Central America, there is no trace of ramón as of 1999. Although the ramón hypothesis
Page 14
has been the focus of many stimulating discussions, the paleoethnobotanical record simply does not support the use of ramón as an important food source for the ancient Maya. Environmental Indications Whereas carbonized seeds and fruit rinds can tell us much about the food consumption habits of the prehistoric Maya, the charcoal from ancient hearths and burned buildings of the past help define what wood products the site occupants were extracting from the surrounding environment. Charcoal records from Cuello (Miksicek et al. 1991) and Copán (Lentz 1991a) provide documentation of wood changes through time with an adequate chronological view. At these sites a diachronic shift from primary forest trees to second growth or understory trees has been recorded. This undoubtedly paralleled forest clearance and intensification of agriculture as populations grew in nucleated areas. Furthermore, a surprising trend revealed by the charcoal record is the amount of pine (Pinus spp.) found at sites where the trees are not found today, for example, Cuello (Miksicek et al. 1991), Dos Pilas (Lentz 1994), and Cerén (Lentz et al. 1996). The explanation for these observations includes a longdistance transport hypothesis or one of climate or humaninduced environmental change in which pine trees grew closer to habitation sites in prehistoric times. Continued research will help elucidate this phenomenon more fully. Summary Results of paleoethnobotanical studies in the Maya area clearly indicate that the prehistoric Maya, beginning at least by 1200 B.C., indeed had a maizebased system of food production along with squash, peppers, and several varieties of beans. These results are corroborated by isotopic data derived from ancient Maya bone collagen (Wright and White 1996). Likewise, manioc and probably other root crops, as valuable cultigens, contributed to the Precolumbian diet, although the degree of contribution is still open to question. Annual crops were most likely grown in a shifting cultivation technique with regional adaptations employing terracing, raised fields, check dams, drained fields, and other forms of hydraulic agriculture practiced to enhance productivity in areas where social and physical conditions were conducive to these developments. Perennials and tree crops were combined with the production from annuals to help meet the demands of expanding and increasingly nucleated populations, especially during the Late Classic period. Orchard farming, with tree crop infields grown adjacent to house compounds, was definitely a widespread practice directly observable at Cerén and indirectly at several other sites. Important broadleaf trees included avocado, hogplum, mamey, guava, nance, cacao, cashew, and calabash. Little evidence for ramón, however, has been recovered from archaeological deposits. Several species of palm were exploited by the ancient Maya, probably because of the oil content of the fruits, and at least one species, coyol, seems to have been actively cultivated. Cotton and agave supplied fiber, and cotton seeds, as by products, provided useful oil.
Page 15
Forest products used for construction material and fuel extracted from plant communities surrounding each site have been identified from charcoal remains. Results of some of these analyses have indicated that forest degradation accompanied prolonged human occupation. Pine appears at some sites where it does not grow now, suggesting longdistance transport of the wood, perhaps as pine charcoal. Alternatively, pines might have grown closer to the occupation sites in the past than they do today because of human activity or climatic change and therefore could have been harvested more conveniently. What is outlined here is a general pattern of plant use by a large group of diverse peoples that occupied the Maya area over a 3,000year time period, with little allowance for intraregional or diachronic variability. Certainly there were many variations on the themes described, resulting from environmental and cultural differences across the breadth of the Maya realm. Even within a community there can be variations in dietary patterns, as seen among the social status levels at Copán (Lentz 1991a). Nevertheless, the overall patterns portrayed in this account seem to be reflected in the archaeobotanical record at most of the areas studied. Future studies will help flesh out our understanding of ancient Maya subsistence and ecological adaptation through time and across regional and social boundaries. This will be accomplished if currently used techniques, and more advanced ones, are applied universally. The paleoethnobotany of the highlands of the ancient Maya realm remains a mystery, and work in this region should be encouraged. Tikal and other sites that had large, socially stratified populations have been examined archaeologically, yet we know little of how they derived their sustenance and what plant resources they exploited. As with all peoples, plants provided essential raw materials for the survival and growth of the Maya cultural manifestation. Continued study will help reveal how the Maya obtained their necessities and modified the environment in the process. References Cited Alcorn, J. B. (1984) Huastec Mayan Ethnobotany. Austin: University of Texas Press. Atran, S. (1993) Itza Maya tropical agroforestry. Current Anthropology 34:633700. Beach, T., and Dunning, N. P. (1995) Ancient Maya terracing and modern conservation in the Peten rain forest of Guatemala. Journal of Soil and Water Conservation 50:138145. Beltrán Frias, L. (1987) Subsistencia y aprovechamiento del medio. In L. Manzanilla (ed.): Coba, Quintana Roo análisis de dos unidades habitacionales Mayas. Mexico City: Universidad Nacional Autónoma, pp. 213232. Benz, B. F. (1986) ''Taxonomy and Evolution of Mexican Maize." Doctoral dissertation, University of Wisconsin, Madison. Bloom, P. R.; Pohl, M.; Buttleman, C.; Wiseman, F.; Covich, A.; Miksicek, C.; Ball, J.; and Stein, J. (1983) Prehistoric Maya wetland agriculture and the alluvial soils near San Antonio Rio Hondo, Belize. Nature 301:417419. Bohrer, V., and Adams, K. (1977) Ethnobotanical Techniques and Approaches at Salmon Ruin, New Mexico. Contributions in Anthropology, Vol. 8. Portales: Eastern New Mexico University. Bronson, B. (1966) Roots and the subsistence of the ancient Maya. Southwestern Journal of Anthropology 22:251279.
Page 16
Caballero, J. (1989) Modern Maya homegardens of the Yucatán peninsula. Paper presented at the 54th Annual Meeting of the Society for American Archaeology, Atlanta. Caldwell, J. R. (1980) Archaeobotanical aspects of the 1980 field season. In T. R. Hester, J. D. Eaton, and H. J. Shafer (eds.): The Colha Project, Second Season, 1980: Interim Report. San Antonio: Center for Archaeological Research, University of Texas, pp. 257268. Cliff, M. B., and Crane, C. J. (1989) Changing subsistence economy at a Late Preclassic Maya community. In P. A. McAnany and B. L. Isaac (eds.): Prehistoric Maya Economics of Belize. Research in Economic Anthropology, Supplement 4. Greenwich, Conn.: JAI Press, pp. 295324. Dahlin, B. H. (1979) Cropping cash in the Protoclassic: A cultural impact statement. In N. Hammond and G. R. Willey (eds.): Maya Archaeology and Ethnohistory. Austin: University of Texas Press, pp. 2137. Dunne, L. J. (1990) Nutrition Almanac. 3d ed. New York: McGrawHill. Folan, W. J.; Fletcher, L. A.; and Kintz, E. R. (1979) Fruit, fiber, bark, and resin: Social organization of a Maya urban center. Science 204:697701. Gepts, P., and Debouck, D. (1991) Origin, domestication, and evolution of the common bean (Phaseolus vulgaris L.). In A. van Schoonhoven and O. Voysest (eds.): Common Beans: Research for Crop Improvement. Wellingford, U.K.: C.A.B. International, pp.754. GómezPompa, A.; Flores, J. S.; and Fernández, M. A. (1990) The sacred cacao groves of the Maya. Latin American Antiquity 1:247257. GómezPompa, A.; Flores, J. S.; and Sosa, V. (1987) The "Pet Kot": A manmade tropical forest of the Maya. Interciencia 12:1015. Hather, J. G., and Hammond, N. (1994) Ancient Maya subsistence diversity: Root and tuber remains from Cuello, Belize. Antiquity 68:330335. Healy, P. F.; Lambert, J.; Arnason, T.; and Hebda, R. (1983) Caracol, Belize: Evidence of ancient Maya agricultural terraces. Journal of Field Archaeology 10:397410. Hurst,W. J.; Martin, R.A., Jr.; Tarka, S. M., Jr.; and Hall, G. D. (1989) Authentication of cacao in ancient Mayan vessels using HPLC techniques. Journal of Chromatography 466:279289. Kaplan, L. (1973) Ethnobotanical and nutritional factors in the domestication of American beans. In C. Earl Smith, Jr. (ed.): Man and His Food. Tuscaloosa: University of Alabama Press, pp. 7585. Krishnamurthi, M. N. (1954) Cottonseed and Its Products. New Delhi: Council of Scientific and Industrial Research. Lange, F. W. (1971) Marine resources: A viable subsistence alternative for the prehistoric lowland Maya. American Anthropology 73:619639. Lentz, D. L. (1990) Acrocomia mexicana: Palm of the ancient Mesoamericans. Journal of Ethnobiology 10:183194. Lentz, D. L. (1991a) Maya diets of the rich and poor: Paleoethnobotanical evidence from Copán. Latin American Antiquity 2:269287. Lentz, D. L. (1991b) Archaeobotanical remains from the Naco Valley, Honduras. Ms. on file, New York Botanical Garden Harding Laboratory. Lentz, D. L. (1993) Medicinal and other economic plants of the Paya of Honduras. Economic Botany 47:358370. Lentz, D. L. (1994) Paleoethnobotanical evidence for subsistence practices and other economic activities in the Petexbatun region during the Classic period. Paper presented at the 93d American Anthropological Association Meeting, Atlanta. Lentz, D. L.; BeaudryCorbett, M.; Reyna de Aguilar, M. L.; and Kaplan, L. (1996) Foodstuffs, forests, fields, and shelter: A paleoethnobotanical analysis of vessel contents from the Cerén Site, El Salvador. Latin American Antiquity 7:247262. Lundell, C. L. (1938) Plants probably utilized by the old empire Maya of Peten and adja
Page 17
cent lowlands. Papers of the Michigan Academy of Sciences, Arts, and Letters 24:3756. McBryde, F. W. (1947) Cultural and Historical Geography of Southwest Guatemala. Publication No. 4. Washington, D.C.: Smithsonian Institution of Social Anthropology. McKillop, H. (1994) Ancient Maya tree cropping, a viable subsistence adaptation for the island Maya. Ancient Mesoamerica 5:129140. Mangelsdorf, P. C., and Lister, R. H. (1956) Archaeological evidence of the evolution of maize in northwestern Mexico. Botanical Museum Leaflet, Harvard University 17:151178. Mangelsdorf, P. C.; MacNeish, R. S.; and Galinat, W. C. (1956) Archaeological evidence on the diffusion and evolution of maize in northeastern Mexico. Botanical Museum Leaflet, Harvard University 17:125150. Matheny, R. T. (1976) Maya Lowland hydraulic systems. Science 193:639646. Miksicek, C. H. (1983) Macrofloral remains of the Pulltrouser area: Settlements and fields. In B. L. Turner II and P. D. Harrison (eds.): Pulltrouser Swamp: Ancient Maya Habitat, Agriculture, and Settlement in Northern Belize. Austin: University of Texas Press, pp. 94104. Miksicek, C. H. (1986) Paleobotanical identifications. In A. A. Demarest (ed.): The Archaeology of Santa Leticia and the Rise of Maya Civilization. New Orleans: Tulane University Press, pp. 199200. Miksicek, C. H. (1988) Man and environment at Cihuatán. In J. H. Kelly (ed.): Cihuatán, El Salvador: A Study in Intrasite Variability. Nashville: Vanderbilt University Press, pp. 149155. Miksicek, C. H. (1990) Early wetland agriculture in the Maya lowlands: Clues from preserved plant remains. In M. Pohl (ed.): Ancient Maya Wetland Agriculture: Excavations on Albion Island, Northern Belize. Boulder, Colo.: Westview Press, pp. 295312. Miksicek, C. H.; Wing, E. S.; and Scudder, S. J. (1991) The ecology and economy of Cuello. In N. Hammond (ed.): Cuello: An Early Maya Community in Belize. Cambridge: Harvard University Press, pp. 7084. Netting, R. McD. (1977) Maya subsistence: Mythologies, analogies, possibilities. In R. E. W. Adams (ed.): The Origins of Maya Civilization. Albuquerque: University of New Mexico Press, pp. 299333. Pearsall, D. M. (1989) Paleoethnobotany: A Handbook of Procedures. San Diego: Academic Press. Pérez Romero, A., and Cobos P., R. (1990) Una nota arqueológica respecto a la presencia de Theobroma cacao en las tierras bajas mayas. Boletín de la Escuela de Ciencias Antropológicas de la Universidad de Yucatán 17:3366. Pohl, M.; Pope, K. O.; Jones, J. G.; Jacob, J. S.; Piperno, D. R.; deFrance, S. D.; Lentz, D. L.; Gifford, J. A.; Danforth, M. E.; and Josserand, J. K. (1996) Early agriculture in the Maya Lowlands. Latin American Antiquity 7(4):355372. Puleston, D. E. (1968) Brosimum alicastrum as Subsistence Alternative for the Classic Maya of the Central Southern Lowlands. Master's thesis, University of Pennsylvania, Philadelphia. Puleston, D. E. (1982) The role of ramón in Maya subsistence. In K. V. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press, pp. 353366. Scarborough, V. L. (1991) Water management adaptations in nonindustrial complex societies: An archaeological perspective. In M. B. Schiffer (ed.): Archaeological Method and Theory. Tucson: University of Arizona Press, pp. 101154. Sheets, P. (1992) The Cerén Site. Boulder: University of Colorado. Sheets, P. (1994) Tropical time capsule. Archaeology 47:3033. Siemens, A. H., and Puleston, E. (1972) Ridged fields and associated features in southern Campeche: New perspectives on the Lowland Maya. American Antiquity 37:228239.
Page 18
Smallwood, N. J. (1978) Significance of gland pigment in processing, storage, handling, and utilization of cottonseed oil. In G. Harper (ed.): Glandless Cotton: Its Significance, Status, and Prospects. Beltsville, Md.: Agricultural Research Service, U.S. Department of Agriculture, pp. 1330. Smith, C. E., Jr. (1967) Plant remains. In D. S. Byers (ed.): The Prehistory of the Tehuacan Valley. Vol. 1. Environment and Subsistence. Austin: University of Texas Press, pp. 220255. Struever, S. (1968) Flotation techniques for the recovery of smallscale archaeological remains. American Antiquity 33:352353. Telek, L., and Martin, F. W. (1981) Okra seed: A potential source for oil and protein in the humid Lowland tropics. In E. H. Pryde, L. H. Princen, and K. D. Mukherjee (eds.): New Sources of Fats and Oils. Champaign, Ill.: American Oil Chemists Society, pp. 3753. Turner, B. L. II (1974) Prehistoric intensive agriculture in the Mayan Lowlands. Science 185:118124 Turner, B. L. II, and Harrison, P. D. (eds.) (1983) Pulltrouser Swamp: Ancient Maya Habitat, Agriculture, and Settlement in Northern Belize. Austin: University of Texas Press. Turner, B. L. II, and Johnson, W. C. (1979) A Maya dam in the Copán Valley, Honduras. American Antiquity 44:299305. Turner, B. L. II, and Miksicek, C. H. (1984) Economic plant species associated with prehistoric agriculture in the Maya lowlands. Economic Botany 38:179193. Watson, P. J. (1976) In pursuit of prehistoric subsistence: A comparative account of some contemporary flotation techniques. MidContinental Journal of Archaeology 1:77100. Wellhausen, E. J.; Robert, L. M.; and Hernandez, X. E. (1952) Races of Maize in Mexico. Cambridge: Harvard University Press. Wellhausen, E. J.; Robert, L. M.; and Hernandez, X. E. (1957) Races of Maize in Central America. Washington, D.C.: National Academy of Sciences. Wilken, G. C. (1971) Foodproducing systems available to the ancient Maya. American Antiquity 36:432448. Wiseman, F. M. (1978) Agricultural and historical ecology of the Maya Lowlands. In P. D. Harrison and B. L. Turner II (eds.): PreHispanic Maya Agriculture. Albuquerque: University of New Mexico Press, pp. 63115. Wiseman, F. M. (1983) Analysis of pollen from the fields at Pulltrouser swamp. In B. L. Turner II and P. D. Harrison (eds.): Pulltrouser Swamp: Ancient Maya Habitat, Agriculture, and Settlement in Northern Belize. Austin: University of Texas Press, pp. 105119. Wright, L. W., and White, C. D. (1996) Human biology in the Classic Maya Collapse: Evidence from paleopathology and paleodiet. Journal of World Prehistory 10:147198.
Page 19
Chapter 2 Classification of Useful Plants by the Northern Petén Maya (Itzaj) Scott Atran and Edilberto Ucan Ek' Kinukuch ch'ib'al kuxlajoo' taanil ti dyoos. "Our great ancestry lived before God." Yun Domingo Chayax Suntecun, Itzaj exorcist and prioste or Aj Nojil K'unaj
This chapter is a digest of how contemporary Itzaj Maya classify and use more than 400 plant species in the tropical forest region of northcentral Petén, Guatemala. The Itzaj are the region's last Lowland Maya with demonstrable historical ties to Precolumbian Lowland Maya civilization (Atran 1993). Itzaj classification of useful plants reveals a stable cognitive structure, a predominance of indigenous species, and a longstanding cultural framework that carries information about the ecological, medicinal, and nutritional dimensions of that ancient past. Given the alarming rate of deforestation and the moribund state of the Itzaj language, these data may also be a timely aid to understanding how the region once supported a population exceeding the currently destructive level by an order of magnitude. Our data base for the Itzaj includes herbaria (on deposit with the Comité BioItza of San José, Petén, and the University of Michigan); an Itzaj grammar and Itzaj/Spanish dictionary of natural history (Lois and Vapnarsky forthcoming); psychological studies on how Itzaj categorize and reason about the natural world (Atran 1999a; Atran et al. 1997; Coley et al. 1997; López et al. 1997); and comparative studies of ecological cognition and agroforestry practice among Itzaj and other Lowland Maya (Yukatek, Lakantun) versus Spanishspeaking Ladinos and Highland Maya (Q'eqchi', Tzeltal) who have immigrated to the Lowlands (Atran 1999b; Atran and Medin 1997; Atran et al. in press). Structural aspects of the classification system presented here reflect a statistical "cultural consensus" of at least twelve Itzaj speakers (six men and six women) based on factor analysis (Medin et al. 1997; Atran 1999a; cf. Romney et al. 1986). The nutritional, medicinal, and cultural contents of the classification represent a synthesis of information elicited from (up to) twenty Itzaj informants for each species. Tables 2.1 to 2.3 provide a summary of the names and functions that Itzaj Maya assign to their principal useful plants (and immediate folkbotanical allies). These tables, which appear at the end of this chapter, are broadly illustra
Page 20
tive rather than exhaustive: many useful plants are excluded (because not yet scientifically identified), and many more are named but unused (although they may have had uses in the past). Where possible, the corresponding names are furnished for other Lowland Maya languages since the time of the Conquest.1 Historical Background In Petén, topographic and microclimatic variation allows for a dramatic range of vegetation over relatively small areas, and sustaining both this diversity and people's livelihood over the past two millennia has likely required correspondingly flexible agroforestry regimes.2 The Precolumbian economy of central Petén may have been biointensive, based on systematic management of micro and mesoscale environmental variation in the form of agroforestry (Puleston 1982; C. Earle Smith in Turner and Miksicek 1984). Only recently has evidence come to light at Bajo la Justa (between Yaxha and Nakun) of geointensive production systems in central Petén that involved landscape modification through agroengineering works (Culbert et al. 1995); evidence is more compelling, however, for largescale production at outlying Classic sites in Belize and the Yucatán (Puleston 1978; Siemens 1982; Sluyter 1994). Paleolimnological analysis of sediments in the central lakes region of Petén associates demise of Classic Maya civilization toward the end of the first millennium with geometrically increasing rates of deforestation (Rice 1993). There is evidence of spiraling population growth (Culbert and Rice 1990), unrelenting warfare (Chase and Chase 1989; Demarest 1993), and nutrient deficiency (Santley et al. 1986). Economic infrastructure supporting a population of perhaps 3 million people collapsed. Transport and communication links apparently disintegrated between central Petén cities (e.g., Tikal) and outlying production centers (e.g., Caracol in Belize). Although the central lakes region may have suffered less drastic population loss than neighboring Tikal, resettlement of Petén seems never to have surpassed the few hundred thousand people estimated for the immediate Preconquest era. Dense forest cover reappears during the Late Postclassic period, which precedes a brutal Spanish conquest in 1697 (Wiseman 1978). By and large, this cover endured through the midtwentieth century. Since 1959, when the military government first opened up Petén to "colonization and development," more than half the forest cover of Petén has been destroyed or degraded (AHG/APESA 1992, Anexo 6 and 13). The rate of deforestation, which averaged nearly 300 km2 between 1962 and 1987, increased to more than 500 km2 by 1990, as population rose from barely 20,000 to over 300,000 (Schwartz 1990:11). Most of southern Petén's rainforest has vanished, leaving the bulk of Petén's remaining forest north of 17°10' latitude. In a debtfornature swap engineered by the Agency for International Development (USAID) in 1990, the Guatemalan government set aside most of the area north of this latitude as a United Nationssponsored "Maya Biosphere Reserve." Yet even within the biosphere, forest continues to burn apace ("S.O.S: Se muere biosfera Maya," Prensa Libre, Guatemala daily, 22 April 1998). Because legislation associated with the Maya 1
The spelling of all Petén Maya words (Itzaj and Mopan) follows the conventions of the Academia de las Lenguas Mayas de Guatemala (1988). This also applies to twentieth century terms of the Yukatek (Yucatec) and Lakantun (Lacandon) Maya of Mexico, who along with the Petén Maya think of themselves as "The Maya" (ajmaayajoo(b)'). Morpheme breaks within Maya words are indicated by a hyphen () and compounds with a tilde (~). Other notational conventions presented below reflect the cognitive structure of Itzaj folk taxonomy The etymologies given for Itzaj folk taxa often represent a synthesis of folk etymologies elicited from informants rather than "true" etymologies (see Paso y Troncoso 1988/1886 for a similar approach to etymologies of Nahuatal folkbiological taxa). 2
At the height of the growing season, July rainfall in Flores (site of the Precolumbian Itzaj capital, Tayasal) went from 121 mm in 1993 to 335 mm in 1996, and in nearby Tikal from 58 mm to 137 mm; in May, when crops are planted, there was no rainfall in Tikal in 1993 for 23 days, then 130 mm in 3 days, etc. Guatemala Govt. Institute of Meteorology, INSIVUMEH.
Page 21
Biosphere bans or severely restricts felling and burning of trees, loggers, ranchers, and farmers raze more territory than they can currently use in anticipation of harsher measures. The major cause of deforestation, however, remains population pressure from the overcrowded and tired lands of southern Guatemala (cf. Schwartz 1995). There some 9 million people live in an area roughly twice the size of Petén, where nearly twothirds of the land is controlled by about 2 percent of the people. The intense scholarly attention that Preconquest Itzaj attract contrasts markedly with lack of interest in the Postconquest Itzaj. This is curious, since research and speculation about Precolumbian Maya civilization often extrapolates from modern ethnographic sources. The reasons for this studious avoidance may have to do with the sorry state of Petén Maya society today compared with its illustrious past. This contrast in cultural fortunes—perhaps more glaring than elsewhere in Mesoamerica—has been wrongly interpreted as a complete culture gap. For example, Colonel Modesto Méndez, on "discovering" Tikal, wrote in his diary (2 March 1848) that the native Itzaj inhabiting the region, who guided him to the site they called "the place of the wind [spirits]" (tiik'al), could not possibly be of the "race . . . descended from those who wanted to immortalize their names" at Tikal (Soza 1970, 1:246). Similar peremptory claims by influential historians (Means 1917) and archaeologists (Hellmuth 1977) that modern Itzaj knowledge and culture bear no significant relation to that of Precolumbian Petén Maya is supported by no empirical evidence or firsthand knowledge of modern Itzaj language and culture. On 22 March 1697, nine days after the destruction at Lake Petén Itza of the last independent Maya confederacy, the conquistador Don Martin de Ursua y Aresmendi wrote to the king of Spain of the finish to an abhorrent culture—a culture that nevertheless reflected the "ingenious abilities" of its native folk to create "exquisite" statues, temples, and glyphs" wherein much was found to be seen and admired," abilities that boded well for the use of corvée labor to build new Spanish cities (Sevilla, Archivo General de Indias, Audiencia de Guatemala legajo 343, folios 70 recto71 verso):3 When the infidels heard our arms, and experienced the valor of those advancing, they began to flee in such a vile manner, men and women threw themselves into the water and filled it all the way to the mainland. And I do not doubt that many would be imperiled. . . . There were found twentyone sanctuaries full of horrible and deformed idols, and among these the shin bone of a horse which, according to an old Indian woman, was the horse of Don Fernando Cortes, who passed these lands on the way to Honduras. Homes were also found to be full of idols, with whose destruction [my] people were occupied from eight in the morning until five in the afternoon. . . . These miserable waywards of the devil, deprived of the true light, must not have had any other activity than idolatry owing to their not having an economic form of settlement, but rather having all kin live together barbarously in one house. The pleasantness and fertility of the land, the delightful beauty of this lake—the breadth and length of its waters on all sides, its inlets and 3
Relying on Ursua's account, Villagutierre (1701/1985), the Spanish Crown's archivist, who never set foot in the New World, makes it seem as if the Spanish gave the Itzaj every opportunity to honor prior commitments to accept Catholic faith and government. Although Villagutierre cites reports of mass Itzaj suicide as the Spanish made their final assault, he intimates that "the Itza in effect invited the Spaniards to conquer them" (Bricker 1981:24),and this is presumably why "the Itzá gave up with barely a whimper on March 13,1697" (Schele and Freidel 1990:399). But other testimony indicates the Spanish were not so peacefully disposed. Diego de Rivas, a Mercedarian friar accompanying Ursua's expedition, later testified that "de la entrada que hizo el Señor Don Martin de Urzua consta que era tanto el número de los que se le opusieron que habiendo muerto innumerables las balas que dispararon los nuestros parecia isla en la laguna la que se formó con los cuerpos muertos de los indios" ("of the entry made by Don Martin Urzua [I] state that so great was the number of those who opposed it, so innumerably many were those killed by the bullets we shot, it seemed like an island in the lake was formed by the bodies of the dead Indians") (15 noviembre 1698, AGI Guatemala 345, folio 389 recto). In the Spanish census of 1714 the native population of Petén had fallen to 3,000 (Schwartz 1990:11), far from the tens of thousands that the Itzaj chieftain kan~ek' governed on the eve of the Conquest (Avendaño y Loyola 1696/1987:4748; Villagutierre 1701/1985:318). Current opinion is that the Itzaj had resigned themselves
(footnote continued on next page)
Page 22 streams, with its continuous waves giving it the appearance of the sea—is extraordinary, as is the highest quality of woven cotton fabrics in delicacy and dye. The land yields two consecutive harvests of produce yearly; at present new maize is being gathered, the ears and kernels are extremely thick, and everywhere very adequate. There are gathered wild fruits, indigo, vanilla, cacao, anotta, cotton, wax, honey and every kind of vegetable and bean from [this] land and from Castile. And although the males seem lazy, because they have small milpas that are expected to yield continual harvests, females to the contrary are known to work from sunup to sundown, without speaking a word.
After conquest many Itzaj were moved around Petén to service cattle ranches. Spanish interest in native agriculture concerned only maize and beans. Because of this, and because until a generation ago Itzaj were punished for speaking their language in public, reports on modern Itzaj subsistence have been limited to farming (cf. Cowgill 1962; Reina 1967). Not that the colonizers were altogether unaware of a dazzling array of local crops, game, and uses of forest resources. But this merely underscored a conviction that absence of great fields and herds in the soilpoor tropical forest was owing to savage sloth (Villagutierre 1701/1985 X, xixii): "Meat, wheat and other things were not raised on the land; not because it wasn't fertile enough to produce them, but because of barbarity, poor upbringing, political and economic insufficiency. . . . These vast lands were quite suitable for Spanish towns . . . among the richest, most productive and advantageous one could imagine. . . .The Indians had everything in their milpas, but little, because they did not cultivate them well." Forest plants with nutritive value comparable to maize and wheat, such as breadnut (ramón) and palmnut (corozo), were supposedly used more "to sustain pack animals and other stock" (VII, xii) than people (cf. Hellmuth 1977:434). Itzaj were forced to overextend maize cropping to sustain Spanish overreliance on cereal. Production often fell short of demand. Spaniards cried "famine," bewailing idle barbarian custom—such as relying on root crops and seeking escape from hunger and exploitation in the fruits, game, and cover of the forest (Schwartz 1990:55): "At times the Spanish and the rest of the population were 'forced' (as the documents put it) to eat such foods as ramón . . . camote (sweet potato), yuca (manioc), ñame and macal (yams), green plantains, and mamey and sapote fruit. Although this list may indicate that the Indians persisted in producing a diversity of crops other than grains despite the Spanish attempt to get them to concentrate on maize and beans . . . the Spanish had culturally defined nutritional standards that hardly made these considerations good news." In fact, Itzaj have survived conquest and colonization with many Precolumbian dietary and medicinal strategies, many of which arguably date to Classic times. Itzaj FolkBiological Taxonomy Itzaj folk biology supports generalizations not only about the characteristic taxonomic structure that delimits the universal cognitive domain of folk biology but also for the profound influence of local ecology and culture. In any (footnote continued from previous page) to conquest because their religion fated that the time had come to selfdestruct (see Puleston 1979). The year of conquest coincided with end of waxak ajaw, the last 20year ka'~tun of the 256year may, or calendrical cycle. That cycle began with the descent (or return) of the Itzaj to Petén from chi'~ch'e'en during the may~apan (Yucatán) civil wars between the Itzaj and Xiw. Over the course of the seventeenth century Franciscan friars negotiated with Itzaj chieftains of the kan~ek' clan for conversion and submission on the basis of Maya prophetic cycles. But the Itzaj themselves evidently used their prophecies as a negotiating ploy to stall the Spanish and to obtain trade. Not only was the Itzaj Sun Priest, kan~ek's kinsman ajk'in kan~ek', in favor of continuing attacks on the Spanish, but kanek' himself was ambivalent. Ursua interrogated his prisoner kanek' a month after the conquest: [Ursua] asking what motive [did kan~ek' have] in sending ambassadors and requesting said fathers, if it was out of fear of the Spanish or some other reason. [Kan~ek'] said that what had motivated him was the necessity of commerce and to have axes and machetes and to ask the fathers to baptize [the Itzaj]. . . . [Ursua] asking how, if he wanted peace, did he have the island fortified with trenches, and on the morning that I came with my men to this Petén, all [Itzaj] were ready for war. [Kan~ek'] said that he had lost the obedience of his Indians. (16 abril 1696, AGI Guatemala 343, folios 331 verso333 recto)
Page 23
Plants of the Northern Peten Maya (Itzaj) in Use Since The Spanish Conquest
Page 24
Table 2.1 Staple Milpa Crops and Their Close Folkbotanical Allies among Preconquest CholtiLakantun, Yukatek, PresentDay Lakantun, Mopán, Yukatek, Itzaj.
(table continued on next page)
Page 25
(table continued from previous page)
Page 26
Table 2.2 Plants Commonly Intercropped in the Milpa (kol, petex), Kitchen Garden (päk'aal~jol~naj) and Orchard ''Plantation" (päk'aal); Trees "Tended" (amb'äl) in the Forest Useful Untended Plants (k'aax), Close Folkbotanical "Companions" (et'~ok), Noxious Plants.
(table continued on next page)
Page 27
(table continued from previous page)
(table continued on next page)
Page 28
(table continued from previous page)
(table continued on next page)
Page 29
(table continued from previous page)
(table continued on next page)
Page 30
(table continued from previous page)
(table continued on next page)
Page 31
(table continued from previous page)
(table continued on next page)
Page 32
(table continued from previous page)
(table continued on next page)
Page 33
(table continued from previous page)
(table continued on next page)
Page 34
(table continued from previous page)
(table continued on next page)
Page 35
(table continued from previous page)
(table continued on next page)
Page 36
(table continued from previous page)
(table continued on next page)
Page 37
(table continued from previous page)
(table continued on next page)
Page 38
(table continued from previous page)
(table continued on next page)
Page 39
(table continued from previous page)
(table continued on next page)
Page 40
(table continued from previous page)
(table continued on next page)
Page 41
(table continued from previous page)
(table continued on next page)
Page 42
(table continued from previous page)
(table continued on next page)
Page 43
(table continued from previous page)
(table continued on next page)
Page 44
(table continued from previous page)
Page 45
Table 2.3 Petén Maya Palms (Arecaceae) and Their Close Folkbotanical Allies (Zingiberales).
(table continued on next page)
Page 46
(table continued from previous page)
Page 47
culture there may be several different sorts of ''specialpurpose" folkbiological classifications that are organized by particular interests for particular uses (e.g., beneficial versus noxious, domestic versus wild, edible versus inedible). For Itzaj, one widespread specialpurpose classification is based on habitat and edibility: land or water or air animals that are either edible or inedible (Lois 1998). Another is based on wild versus domestic plants and animals (such as the classification most readily expressed in one Itzaj informant's idiolect and presented as "Itzaj Taxonomy" in Hofling 1997:6180). Nevertheless, in every culture there is one kind of recognizable "generalpurpose" taxonomy (Atran 1990; Berlin 1992) which supports systematic reasoning about living kinds and properties of living kinds (López et al. 1997; Atran et al. 1997) and which may be a specific adaptation of hominid evolution (Atran 1998). The relatively "culturefree" organization of folkbiological taxonomy is common not only to all Maya groups but to people across the world. In addition to the spontaneous arrangement of local fauna and flora into specieslike groupings, these basic groupings have "from the most remote period in . . . history . . . been classed in groups under groups. This classification is not arbitrary like the grouping of stars in constellations" (Darwin 1859:431). But this does not mean that folkbiological categories are culturally irrelevant. On the contrary, insofar as they reflect a cognitively biased, phenomenal appreciation of the surrounding environment, they help set the constraints on life that make a forest culture possible. It is little wonder, then, that folkbiological taxonomies tend to be among the most stable, widely distributed, and conservative cognitive structures in any culture. Once set into place, such a structure would likely survive even catastrophic historical upheaval to a recognizable degree. Ancient and contemporary Maya societies would be no exception. Even with the social order and cosmological system sundered, folkbiological structure would persist as a cognitive basis for cultural survival, under two conditions: first, there must be significant biological continuity in the ecological distribution of species; and second, there must be significant linguistic continuity with the dialect that first encoded the knowledge. Tables 2.1 to 2.3 suggest that both conditions have long endured. Across the world, it appears, folkbiological taxonomy is composed of a small number of distinct hierarchical levels, or ranks (Berlin 1992): the levels of folk kingdom (e.g., animal, plant),4 life form (e.g., bug, fish, bird, mammal, tree, grass, bush),5 generic species (e.g., gnat, shark, robin, dog, oak, wheat, holly, toadstool),folk specific (poodle, retriever; white oak, red oak), and folk varietal (hunting poodle, toy poodle; swamp white oak, stunted white oak). Ranking is a cognitive mapping that projects living kind categories onto a structure of absolute levels, that is, fundamentally different levels of reality. Ranks, not taxa, are apparently universal. Generic Species The rank of generic species—the level at which Itzaj classify organisms such as ajb'aatz' (howler monkey) or put (papaya)—is where morphological, behav 4.
It makes no difference if these groups are named. English speakers ambiguously use the term "animal" to refer to at least three distinct classes of living things: nonhuman animals, animals including humans, and mammals (the prototypical animals). The term "beast" seems to pick out nonhuman animals in English but is seldom used today. The term "plant" also refers ambiguously to the plant kingdom or to members of that kingdom that are not trees. 5.
Life forms may differ from culture to culture. For example, cultures such as ancient Hebrew or modern Rangi (Tanzania) include the herpetofauna (reptiles and amphibians) with insects, worms, and other "creeping crawlers" (Kesby 1979). Other cultures, such as Itzaj and most Western folk cultures, include the herpetofauna with mammals as "quadrupeds." Some cultures, like Itzaj, place phenomenally isolated mammals like the bat with birds, just as Rofaifo (New Guinea) place phenomenally isolated birds like the cassowary with mammals (Dwyer 1976). Whatever the content of lifeform groupings, or taxa, the lifeform level, or rank, universally partitions the living world into broadly equivalent divisions.
Page 48
ioral, and ecological relationships between organisms maximally covary. By far, the majority of taxa in any folkbiological system belong to this level. In all societies people privilege this level to understand biological discontinuities. When an organism is encountered that is not readily identifiable as belonging to a named generic species, it is still expected to belong to one. The organism is usually assimilated to one of the named taxa it resembles, although at times it is assigned an "empty" genericspecies slot pending further scrutiny (e.g., "suchandsuch a plant is some [genericspecies] kind of tree"; see Berlin 1999). Generic species represent cuts in nature that Maya children first name and form an image of (Stross 1973) and that Maya adults most readily use in speech, recall in memory, and communicate to others about (Berlin et al. 1974; Hunn 1977).6 The genericspecies rank is also that at which Maya are most likely to attribute biological properties. This includes characteristic patterns of inheritance growth and physiological function as well as more "hidden" properties, such as hitherto unknown organic processes, organs, and diseases (Atran et al. 1997). Correspondence of generic species to scientific species or genera is not isomorphic and varies according to patterns of species distribution within biological families and other factors. Still, generic species usually encompass single biological species and usually do not extend beyond biological genera for larger vertebrates and flowering plants. Specifics and Varietals Subdivision of generic species into lowerorder folk categories may also presumptively reflect deep biological relationships among the subordinate taxa. Nevertheless, the phenomenal boundaries delimiting folkspecific taxa (more so for folkvarietal taxa) are often culturally determined and fail to reflect scientifically relevant distinctions (Berlin 1992). Rather, differences in cultural use highlight a contrast among two or more biologically related subordinate taxa (i.e., a contrast set) along some salient phenomenal dimension (color, "sex," habitat, shape, size, analogy with another object, etc.). Differences are customarily marked as a binomial contrast: that is, each subordinate taxon is linguistically denoted by the use of an attributive that modifies the superordinate stem and points to some value of the distinguishing phenomenal dimension (see Table 2.4). Only specifics and varietals that reflect intense, longstanding cultural use tend to be labeled with a single lexeme: for example, ixjoob'oj, k'inim, and other subtypes of the hogplum, 'ab'äl (Spondias spp.). Foreign organisms introduced into a local environment are often initially assimilated to generic species as folkspecifics. For example, Lowland Maya originally called the Spanish pig "village peccary" (k'ek'en[+]kaj) or "Castilian peccary" (k'ek'en[+]kastil).7 The Spanish referred to indigenous pacas and agoutis as "bastard hares," just as they denoted the Maya breadnut tree "Indian fig'' (Beltrán 1742/1859). Subordinate folkspecific and varietal ranks correspond to ranges of perceptible natural variation that humans are most apt to appropriate and manipulate as a function of cultural interests (Tables 2.12.3). 6.
Botanists and ethnobotanists emphasize morphological criteria in identifying the basic folkbiological level with the scientific genus (Bartlett 1940; Berlin 1992), whereas zoologists and ethnozoologists stress behavioral (especially reproductive) criteria and identify with the species (Diamond 1966; Bulmer 1970). We use "generic species," instead of "folk generic" or "folk specieme," because a distinction between biological genus and species is not consistently relevant for folk. Phenomenally salient species for humans (e.g., canopy trees and large vertebrates) often belong to monospecific genera in a locale. Closely related species of a local polytypic genus are often hard to distinguish, because no morphological or ecological "gap" is readily apparent between them. Historically, the distinction did not appear until an influx of newly discovered species from all over the world compelled European naturalists to mnemonically manage them in a worldwide system of genera built around (mainly European) species types (Atran 1990 in press a). 7.
Today the pig is the primary referent of k'ek'en, which can be optionally labeled k'ek'en(+kastil) or k'ek'en(+kaaj). By contrast, the peccary is labeled with the obligatory qualification k'ek'en(+)che' ("forest k'ek'en").
Page 49 Table 2.4. Divisions of Culturally Important Itzaj FolkGeneric Species into Subordinate Taxa. Mono/Poly types
Tree
NonTree
TOTAL
MONOSPECIFICS
146 (73%)
119(72%)
265 (73%)
POLYSPECIFICS
54 (27%)
46 (28%)
100 (28%)
TOTAL GENERIC SPECIES
200
165
365 (representing 437 scientific species & 100 families)
Subordinate Types
Tree
NonTree
TOTAL
FOLK SPECIFICS
126
137
263
FOLK VARIETALS
27
45
72
SUBVARIETALS TOTAL SUBDIVISIONS
2
6
8
155
188
343
Note: Parenthetical numbers indicate dual dimensions: e.g., ixpäpäj[+]p'ak ("sour tomato") = ixpech'ek[+]p'ak = ("flat tomato")
Phenomenal Dimension
FOLK SPECIFICS Tree
Color
NonTree 87 (+3)
Habitat
FOLK VARIETALS Tree
46(+10)
ALL SUBORDINATES
NonTree
TOTAL
% TOTAL
14(+2)
29+(1)
176(+15)
+ 2 subvar.
+ 6 subvar.
+ 8 subvar.
59% (58%)
15 (+2)
18 (+2)
0
0
33 (+4)
11% (11%)
"Sex"
4
16 (+2)
0
4
24 (+2)
8% (8%)
Shape
2 (+1)
8 (+8)
(2)
4 (+2)
14 (+13)
5% (8%)
4
5
2
2(+1)
11 (+1)
4%(4%)
2 (+1)
4 (+15)
(2)
(4)
6 (+22)
2% (8%)
Taste
4
1 (+3)
0
1
6 (+3)
2% (3%)
Texture
1
3
0
1
5
2% (1%)
3 (+1)
9 (+1)
4
0
16 (+2)
5% (5%)
0
2 (+1)
3
0
5
(+1)
Size Analogy
Other analyzable Not analyzable 2% (2%)
Foreign name TOTAL
0
4
1 (+1)
2
7(+1)
2%(2%)
122 (+8/2)
116 (+42/2)
26 (+6/2)
47 (+8/2)
311 (+64/2)
%/of 311
=126
= 137
=29
=51
=343
(% of 343)
Kingdoms and Life Forms Superordinate levels of Itzaj folk taxonomy provide further evidence for a universal cognitive structure in a Maya idiom. There is no common lexical entry for the plant kingdom in Itzaj; however, the numeral classifier teek is used with all and only plants. Plants generally fall under one of four mutually exclusive life forms: che' (trees), pok~che' (herbs, shrubs = undergrowth), ak' (vines), and su'uk (grasses). Each life form conforms to a distinct stem habit. Some intensively cultivated plants are unaffiliated with any of these life forms and are
Page 50
simply denoted jun~teek (lit. "one plant," e.g., jun~teek ixi'im = a maize plant). This is also true of many phylogenetically isolated plants, such as the cacti. All informants agree that mushrooms (xikin~che' lit. "treeear") have no pusik'al (lit. "heart'') and are not plants but take life away from the trees that host them. Lichens and bryophytes (mosses and liverworts) are not considered to be plants, to have an essence, or to live. In Itzaj the term for animals (b'a'al~che' = "forestthing") polysemously refers to (1) the animal kingdom as a whole (including invertebrates, birds, and fish); (2) a more restricted grouping of quadrupeds (i.e., b'a'al~che'+kusiit' = amphibians, or "jumping animals"; b'a'al~che'+kujiltik ub'aj = reptiles, or "slithering animals"; or b'a'alche'+kuxi'mal = mammals, or "walking animals"); (3) typically the mammals alone. Birds (ch'iich' including ajsotz' = bats) and fish (käy) exhibit patterns of internal structure similar to the "unnamed" mammal life form as well as the plant life forms (Atran et al. 1997). Mammals and herpetofauna are also joined and embedded under the mutually exclusive category quadruped (i.e., b'a'alche' sense 2), which can be explicitly rendered a'b'a'al~che'yan uyok ("animals having feet") or käntaach uyok ("fourfooted"). More often, käntaach uyok refers exclusively to herpetofauna, much as the old Yukatek terms xaknal or xakatnal were translated as cuadrúpedo but often only denoted herps (Beltrán 1742/1859:228). Snakes are said to have "hidden" feet "only the speechless can see" (chen ch'uch'kucha'antik uyok kan). No overarching ecological or morphobehavioral framework seems to encompass all and only herpetofauna or quadrupeds. Like the invertebrate life form (mejen+b'a'al~che' = "small animal"), herpetofauna seem to form a "residual" life form that lacks a conceptually distinctive role in "the economy of nature." This contrasts with other plant and life forms, which have mutually defined ecological roles: birds and trees in the air (ik') and upper forest tier; mammals and herbs on the ground (lu'um) in the forest understory; vines in the connecting "middle" (tanchumuk) tiers; grasses in open lands (chäk'an); fish in water (ja'). True, boundaries between these "adaptive zones" are permeable by members of other life forms, but each life form has its respective habitat, or "home" (otoch). Thus, because the chicken (ajkax) has its home exclusively on the ground with people and cannot live in the air like other birds, it is not a bird, nor is it included under any other life form.8 Maize (ixi'im), which has no "natural" allies or place in the forest but only a culturally defined existence in the manmade milpa (kol), is unlike other plants and so is also not included under any life form. Intermediate Taxa Intermediate levels also exist between the genericspecies and lifeform levels. Taxa at these levels usually have no explicit common name (e.g., frogs and toads), although sometimes they may (e.g., felines, palms). Such taxa—especially unnamed "covert" ones—tend not to be as clearly delimited as generic species or life forms, nor does any one intermediate level always constitute a fixed taxonomic rank that partitions local fauna and flora into a mutually ex 8
For other Maya, such as Tzeltal, the chicken is the prototypical bird (Hunn 1977).
Page 51
clusive and virtually exhaustive set of broadly equivalent taxa. Those having names usually take on the name of a prototypical generic species: for example, in Itzaj, b'alum (large felines, and the jaguar in particular) and xa'an (palm trees, and guano in particular). Intermediate taxa, whether named or unnamed, are often composed of biological groups limited to local families or orders (Atran 1990). Like many biological families and orders, they may cut across life forms. Table 2.3 provides an example with the intermediate Itzaj category of palms and their folkbotanical allies, the broadleaf zingiberales. Although there is a significant statistical correlation between folk and scientific taxonomies, there is also significant divergence. This divergence, however, does not tend to be culturally driven by specialpurpose interests. The folk taxa seem to be constituted by a more generalpurpose concern for comprehending ecological relationships relative to human perception. With palms, for example, the morphology of the stem and leaf appear to be paramount (Atran 1999a). Aspects of this morphological standpoint correspond to evolutionary relationships that exist independently of cognitive biases in human perception; others, such as relative size, are crucial to everyday human understanding of the surrounding natural environment but of only secondary importance to the more minute and extensive "cosmic" vantage of science (Hunn 1999). Maya Nomenclature and Notation Reliance on nomenclature alone to indicate taxonomic status can be misleading. For example, the systematic qualification of folkbiological categories by attributives often points to binomial folkspecific taxa of cultural importance, but not always. To better understand the cognitive distinctions between superficially similar expressions, a set of nomenclatural marks is introduced. These notations represent "hidden" cognitive features of folk categorization which are not apparent from spoken linguistic forms and which help in the appreciation of the cognitive structure and content of Tables 2.12.3. Terms that express taxonomic ranking are composite expressions rather than compounds or descriptive phrases (cf. Conklin 1962). Composites consist of a qualifier plus a stem. The stem designates a category immediately superordinate to the category in question. The relationship between stem and qualifier is indicated by a plus sign, "+." In Table 2.3, for example, ajb'on(+)xa'an ("the cabbagepalm guano") refers exclusively to the generic species Sabal mexicana, which is the closest taxonomic ally of S. mauriitiformis. A minority of informants consider the composite stem optional and simply refer to ajb'on or b'on. But for all other generic species of the intermediate palm category, xa'an, inclusion of the composite stem is always optional: for example, ajkuum (+xa'an) = Crysophilia staurocauta), tuk'(+xa'an) = Acrocomia mexicana, and so on. Terms that are intermediatelevel composite expressions are indicated by a plus sign in parentheses,"(+).'' Optional composite expressions are indicated by enclosing a plus sign within parentheses together with the stem or qualifier. By contrast, the relationship between terms in a descriptive phrase is
Page 52
indicated by a blank space between terms. For example, jach xa'an ("true guano") describes the protypical status of Sabal mauriitiformis among the intermediate category of (usually) taller palms, xa'an. On occasion it can describe a specific kind of S. mauriitiformis as the prototypical folkspecific, ajb'äyäl[+]xa'an. Most folkspecifics are composites of a genericspecies stem plus a qualifier. This relationship is indicated by a plus sign in brackets,"[+]."9 Varietals are nearly always composites whose superordinate specific is itself a composite. The embedded composite relationship is marked by a plus sign in double brackets, "[[+]]." The introduced ixya'ax[[+]]kab'al[+]kookoj ("the green dwarf coconut'') is thus subordinate to the folkspecific ixkab'al[+]kookoj ("the dwarf coconut"). Cognitive appreciation of such prized Postconquest plants differs little from appreciation of such "traditional" subordinate taxa as ixnoj[[+]]ch'uuk[+]'ik ("the big sweet chile") and ixch'uuk[+]'ik ("the sweet chile"). A compound—as opposed to a composite or description—is formed by uniting two terms whose different meanings may or may not be related, in order to form a single new meaning. The relationship between compounded terms is indicated by a tilde, "~." For example, pok~che' ("burnt tree") is a compound that denotes the life form herbaceous (under)growth. Although the constituent morphemes give a descriptive meaning that is etymologically related to the nature of the life form, they are fused into a new meaning to denote the life form. Now let us consider some representative folkbiological expressions from Tables 2.1 to 2.3 to see how the notation helps clarify their cognitive status (for further details, see Lois 1998). 1. ixchäkä~ja'as = "reddish plantain." This expression refers exclusively to the red mamey tree (Pouteria mammosa), which has no folktaxonomic relationship nowadays with plantains and bananas. Historically, however, the native mamey was originally labeled ja'as (Roys 1931:228). It was initially perceived as related to the introduced plantains and bananas in much the way that the tapir and horse were perceived to be related. When the Spanish introduced the horse (a perissodactyl), the Maya classified it as a specific kind of tapir (the only native perissodactyl). Over time the importance of the horse in the Maya vision of "the economy of nature" came to outweigh the tapir's. The original unmarked term for tapir, tzimin, was passed on to the horse, and the tapir acquired the obligatory marking tzimin(+)che' ("forest tzimin"). Still, the intermediatelevel taxonomic connection persists among the Itzaj and other Yukatekan Maya (Lakantun, Mopán, Yukatek), indicating awareness of a significant biological relationship between the tapir and horse. By contrast, Maya ultimately recognized the initial morphological analogy between the native mamey fruit and the introduced plantains and bananas to be biologically superficial and thus taxonomically insignificant. 2. chäkäl~te' = "reddish tree" (but possibly from chäk~k'ul~te' = "red god tree," a companion of tropical cedar; see next example). This compound expression refers exclusively to the mahogany tree (Swietania macrophylla). The morpheme te', which is Southern Mayan for "tree/forest/wood," often func 9
Contrast this with b'äyäl(+ak') = "basket whist vine" (Desmoncus spp.). These spiny palm climbers belong to the lifeform category ak' (VINE) and optionally include the life form stem when referring to the generic species. In some contexts, Itzaj use this composite form to distinguish basket whist from the prototypical guano folkspecific, ajb'äyäl[+] xa'an. Generic species that include the lifeform stem may be considered composites, rather than compounds, if the stem is strictly optional.
Page 53
tions linguistically like the Northern Maya morpheme che'. In such cases, the morpheme functions as a productive meaningconstituent of a compound rather than as a lifeform stem that indicates taxonomic partitioning. Nor is the qualifier used simply to describe a salient morphological aspect of the mahogany tree (i.e., its reddish wood). 3. k'u~che' = "god tree." This compound refers exclusively to the generic species tropical cedar (Cedrela mexicana). The morpheme che' in this case has a semantic role equivalent to that of te'. Traditionally, tropical cedar was a sacred tree, and the etymological significance of the compound name is thus apparent on inspection. But few presentday Itzaj are spontaneously aware of the constituent meanings; no more, say, than most Americans automatically think of the compound term "eggplant" as, first of all, describing an eggy plant. 4. ajsäk[+]k'u~che' = "white god tree." This composite refers to one of the two specifics of k'u~che'; the other is ajchäk[+]k'u~che' ("red god tree"). Although Itzaj use color qualifiers to refer to the character of the wood, as often as not the differences signaled reflect distinctions in woodgrain rather than color per se. Frequently, when there is a specific contrast between chäk and säk, the specific qualified by chäk is considered the "true" (jach) type of the generic species. The majority of subordinate Itzaj folk taxa reflect color contrast, and the most habitual contrast is that between chäk and säk (see Table 2.5). This is true despite the fact that distinctions involving "green," "yellow,'' or "black" may be no less obvious to the naked eye. Use of contrasting color specifics, which are generally restricted to the five primary colors, is perhaps related to the overriding importance of these colors in Maya cosmology (see Bruce 1968 for Lakantun; Barrera Marín et al. 1976 for Yukatek). This cosmology traditionally associates red with the true wind of the East, which brings rain and bounty, and white with the false wind of the North, which brings drizzle and deception. This does not deny that color contrasts signal perceptible distinctions among folkspecifics. It suggests only that color perception alone may underdetermine whether, say, "red" versus "white" is really more apparent in a given case than "black" versus "yellow."10 Cultural Survival The most abundant tree in the northern Petén forest is breadnut (Brosimum alicastrum, Sp = ramón, M = 'oox), followed by chicle (Manilkara achras, Sp = chicozapote, M = ya') (AHG/APESA 1992). Breadnut (89 percent of tree total) and chicle (67 percent) belong to the middle tree tier and dominate the vegetative associations that Bartlett (1936) and Lundell (1937) referred to by their local Spanish names, "ramonal" (M = uk'aaxil 'oox) and "zapotal" (M = uk'aaxil ya'). For Itzaj, 'oox is "the animal milpa" (ukolil b'a'al~che'), whose fruits and foliage sustain multiple species of deer, peccary, monkey, procyonid, large rodent, gallinaceous bird, tinamou, parrot, and toucan. It is nature's frame and complement to a Maya garden (Atran 1993:684). In ecologically degraded areas, where overburning by immigrants has reduced forest cover to the point that birds hesitate to fly over, the fireresistant fruit kernel (M = jab'en~tun) of 10
Contrast this with säk(+)k'u~che' = "white god tree." This composite refers to the cedrillo trees (Guarea excelsa, Trichilia hirta), which are considered distant relatives of the tropical cedar. The three meliaceous generic species—k'u~che', chäkäl~te', and säk(+)k'u~che'—comprise an unnamed intermediatelevel folk category. Intermediatelevel groupings are often recognized as "companions" (et'~ok, lit. "together~foot") of the same "lineage" (ch'ib'al).
Page 54 Table 2.5. Contrasting Divisions of Itzaj Plant Generic Species. Number of Divisions in each Contrast Set
FOLK SPECIFIC
FOLK VARIETAL
TOTAL Contrast Sets
% TOTAL Contrast Sets
NonTree
2
42
28
7 (+1 subvarietal)
8 (+3 subvarietal)
89
67%
3
10
10
1
7
28
21%
4
3
3
1
2
9
7%
6
0
2
1
0
3
2%
7
0
1
0
0
1
1%
9
0
1
0
0
1
1%
11
0
1
0
0
1
TOTAL Sets and Subordinate Taxa
55 sets =125 taxa
46 sets =137 taxa
11 sets =29 taxa
20 sets = 51 taxa
132 sets =343 taxa
Color Contrasts
chäk/säk
Tree
Tree
FOLK SPECIFIC
Tree
NonTree
FOLK VARIETAL
NonTree
Tree
19 (+5)
16
chäk/b'ox
3
1 (+1)
0
0
chäk/k'än
4 (+3)
3
2(+1 subvarietal)
7(+1 subvarietal)
chäk/ya'ax
TOTAL of 121 Basic Color Contrast Sets
NonTree 0
1%
6 (+1 subvarietal)
47 = 39% 5 = 4% 21 = 17%
1
0
0
0
1 = 1%
säk/b'ox
7 (+2)
2 (+2)
0
0
13 = 11%
säk/k'än
5 (+2)
2(+1)
1
3
14 = 11%
säk/ya'ax
1
2
0
0
3 = 2%
b'ox/k'än
3 (+2)
(2)
0
0
7 =6%
b'ox/ya'ax
0
0
1
1
2 = 2%
k'än/ya'ax
4
1
2
1(+1 subvarietal)
9 = 7%
the corozo palm (tutz) leads to a "dominant corozal" (M = uk'aaxil tutz). The degree to which human intervention has favored the characteristic forest associations remains moot (Lundell 1938; Puleston 1982; Lambert and Arnason 1982). Before the Conquest, breadnut, chicle, and palms—such as corozo and cocoyol (tuk')—had a history of value to the Maya. Early Spaniards reported use of palms' hearts, fruits, oil, and thatch leaves, of chicle wood and fruit, and of breadnut: "tasty figs they call Ox" (Landa 1566/1985:174; Avendaño y Loyola 1696/1987:61). As late as the 1940s, Itzaj report that the margin of survival often depended on consumption of the fruits of 'oox, ya', tutz, and tuk'. Today, newer marketdriven economic values, such as 'oox foliage for cattle fodder and ya' sap for chicle gum (cha'), have largely supplanted the old ones related to subsistence and religion. But vestiges of older values persist in memories of the recent past. Although there may be debate about whether oox was a garden tree in Classic Petén, modern Itzaj have no doubt that their ancestors,
Page 55
whom they believe inhabited Tikal, tended the tree in the forest as some Itzaj still do: making drinks and cooked dishes from it, offering it to religious functionaries, and storing it against famine. Indeed, until the arux, or wood fairies, were recently scared off by tourists and the introduction of noise from radio and television, they protected the forest by healing the ya' after it was tapped and guarding the 'oox of Tikal. The grandson of Na' Ix Justa Chayax, the last monolingual Itzaj speaker, relates what she told him about breadnut at Tikal: Te'lo'ti tiik'al kutz'ikoo' ujanaloo' nukuch winikoo' uchij There at Tikal they guarded their food great men (ancestors) in the past mentikej ujol chaltunej layti'b'in uyotoch aruxej. that's why the hole of the chaltun (it is said) is the home of the arux. B'aylo'ej mak ma'tanuman uch'e'ma'te'ulaak'jol yok'l ucha'ik maak saakil Like that nobody goes spying into another hole so as to leave a person scared men kutz'ikoo' uchijej kulik'sikoo'uyich'oox because they put it in the past to guard breadnut men uyich'ooxej ma'tulakalk'in yan. because breadnut not every day there is. Mentikej layti'oo'ej kumolikoo'kutasikoo'b'aylo'ej yan ujanaloo', That's why they collected it brought it like that they had their food men uchi'ej chen uyich 'oox ujanaloo' nukuch maakoo'. because in the past only breadnut was the food of the great people. Concern with breadnut is also evident from psychological experiments aimed at eliciting mental models of folk ecology. Of all plant species, breadnut is seen by the Itzaj as interrelating most with animals of the forest and as most deserving respect and protection (Atran 1999b; Atran et al. forthcoming). Breadnut is just one of many examples of Itzaj survival within the Precolumbian folkbiological framework of PeténMaya forest culture, despite SpanishCreole preoccupation with cattle and corn. This is reflected in the tables: there are 437 biological species distributed among 100 scientific families (Tables 2.1 to 2.3). These represent 365 folkgeneric species and 343 subordinate taxa (see Table 2.4). Twothirds of these generic species (247) are edible or medicinal, of which the overwhelming majority (209) are native species (i.e., excluding also Precolumbian introductions, such as manioc = tz'iim). Conclusions The idea that modern Itzaj exhibit a measure of continuity with their Precolumbian forebears in knowledge and use of Petén plants (and animals) seems plausible for three reasons. First, the general higherorder structure of the local folk taxonomy appears to have universal aspects which any predecessors would also likely have and which other contemporary Maya groups certainly do have (Berlin et al. 1974; Barrera Marín et al. 1976; Hunn 1977). Second, the majority of local generic species that serve as the empirical basis for this generalpurpose folk taxonomy were almost surely present before the
Page 56
Conquest: archives and reports indicate that most were being used shortly after the Spanish arrived, and paleobotanical evidence shows that at least some were in use in ancient times (e.g., Turner and Miksicek 1984; Lentz 1991, this volume). Third, ethnographic, ethnohistorical, and ethnolinguistic analyses suggest that many of the names, referents, and uses of culturally specific subordinate taxa are the same, at least since historical records began. This is not to say that there is a timeless underlying nature to Lowland Maya culture or that contemporary Petén Maya natural history directly mirrors Precolumbian knowledge and use of the forest. Demise of Maya commerce and ritual has led to extinction of cultural knowledge and, in some cases, awareness of plants for which Itzaj were once renowned: for example, Amaranthus spp. (ixtes), Vanilla fragrans (siis~b'ik), and Indigofera anil (ch'oj). Even the Petén Maya word for "moon" ('uj) is all but forgotten. Nevertheless, evidence suggests that the Itzaj present and two millennia of the Maya past continue to be mutually and profoundly informative. Acknowledgments This research was supported by the Centre National de la Recherche Scientifique (Contrat 92C0758) and the National Science Foundation (Grant SBR 94 22587). We wish to thank Ximena Lois (CREAEcole Polytechnique, Paris) for help with linguistic analyses, José María Aguilar (Escuela Nacional Central de Agricultura, Guatemala City) for help with species identifications, and David Taylor (Department of Biology, University of Michigan) for help with botanical collections. References Cited Academia de las Lenguas Mayas de Guatemala (1988) Lenguas Mayas de Guatemala: Documento de referencia para la pronunciación de los nuevos alfabetos oficiales. Documento No. 1. Guatemala City: Instituto Indigenista Nacional. AHG/APESA (1992) Plan de desarollo integrado de Petén: Inventario forestal del Departamento del Petén (Convenio Gobiernos Alemania Guatemala). Santa Elena, Petén: AGRARUN HYDROTECHNIK GMBH and Asesoría y Promoción Económica. Atran, S. (1990) Cognitive Foundations of Natural History. New York: Cambridge University Press. Atran, S. (1993) Itza Maya tropical agroforestry. Current Anthropology 34:633700. Atran, S. (1998) Folkbiology and the anthropology of science: Cognitive universals and cultural particulars. Behavioral and Brain Sciences 21:547568,593609. Atran, S. (1999a) Itzaj Maya folkbiological taxonomy. In D. Medin and S. Atran (eds.): Folkbiology. Cambridge: MIT Press, pp. 119203. Atran, S. (1999b) Managing the Maya commons: The value of local knowledge. In V. Sandoval (ed.): Ethnoecology: Situated Knowledge/Located Lives. Tucson: University of Arizona Press, pp. 190214. Atran, S.; Estin, P.; Coley, J.; and Medin, D. (1997) Generic species and basic levels: Essence and appearance in folk biology. Journal of Ethnobiology 17:2245. Atran, S., and Medin, D. (1997) Knowledge and action: Cultural models of nature and resource management in Mesoamerica. In M. Bazerman, D. Messick, A.Tinbrunsel,
Page 57
and K. WaydeBenzoni (eds.): Environment, Ethics, and Behavior. San Francisco: New Lexington Press. Atran, S.; Medin, D.; Ross, N.; Lynch, E.; Ucan Ek', E.; and Vapnarsky, V. (1999) Folkecology and commons management in the Maya Lowlands. Proceedings of the National Academy of Sciences U.S.A., Vol. 96. Avendaño y Loyola, Fray Andrés de (1696/1987) Relation of Two Trips to Petén Made for the Conversion of the Heathen Ytzaex and Cehaches Trans. C. Bowditch and G. Rivera. Ed. F. Comparato. Culver City, Calif.: Labyrinthos. Barrera Marín, A.; Barrera Vásquez, A.; and López Franco, R. (1976) Nomenclatura etnobotánica Maya. Mexico City: INAH. Barrera Vásquez, A. (1991) Diccionario Maya. 2d ed. Mexico City: Porrúa. Bartlett, H. (1936) A Method of Procedure for Field Work in Tropical American Phytogeography Based on a Botanical Reconnaissance in Parts of British Honduras and the Petén Forest of Guatemala. Botany of the Maya Area, Miscellaneous Papers No. 1. Washington, D.C.: Carnegie Institution of Washington, Publication 461. Bartlett, H. (1940) History of the generic concept in botany. Bulletin of the Torrey Botanical Club 47:319362. Beltrán, Fray Pedro de (1742/1859) Arte del Idoma Maya. 2d ed. Mérida: J. D. Espinosa. Berlin, B. (1992) Ethnobiological Classification. Princeton: Princeton University Press. Berlin, B. (1999) One Maya Indian's view of the plant world. In D. Medin and S. Atran (eds.): Folkbiology. Cambridge: MIT Press, pp. 7199. Berlin, B.; Breedlove, D.; and Raven, P. (1974) Principles of Tzeltal Plant Classification. New York: Academic Press. Berlin, E., and Berlin, B. (1996) Medical Ethnobiology of the Highland Maya of Chiapas, Mexico: The Gastrointestinal Illnesses. Princeton: Princeton University Press. Bricker,V. (1981) The Indian Christ, the Indian King. Austin: University of Texas Press. Bruce, R. (1968) Gramática Lacandón. Mexico City: INAH. Bulmer, R. (1970) Which came first, the chicken or the egghead? In J. Pouillon and P. Maranda (eds.): Echanges et communications: Mélanges offerts à Claude LéviStrauss. The Hague: Mouton, pp. 10691091. Cáceres, A. (1996) Plantas de uso medicinal en Guatemala. Guatemala: Editorial Universitaria, San Carlos. Chase, A., and Chase, D. (1989) The investigation of Classic period Maya warfare at Caracol, Belize. Mayab 5:518. Cowgill, U. 1962. An agricultural study of the southern Maya Lowlands. American Anthropologist 64:273286. Coley, J.; Medin, D.; and Atran, S. (1997) Does privilege have its rank? Inductive inferences within folkbiological taxonomies. Cognition 63:73112. Conklin, H. (1962) Lexicographical treatment of folk taxonomies. International Journal of American Linguistics 28:119141. Culbert, T. P.; Levi, L.; McKee, B.; and Kunen, J. (1995) Investigaciones arqueológicas en el Bajo La Justa, Entre Yaxha y Nakun. Paper presented to the Simposio de Arqueologia Guatemalteca, Guatemala City. Culbert, T. P., and Rice, D. (1990) (eds.) Precolumbian Population History in the Maya Lowlands. Albuquerque: University of New Mexico Press. Darwin, C. (1859) On the Origins of Species by Natural Selection. London: Murray. Demarest, A. (1993) The violent saga of a Maya kingdom. National Geographic 183:95111. Diamond, J. (1966) Zoological classification of a primitive people. Science 151:11021104. Dwyer, P. (1976) An analysis of Rofaifo mammal taxonomy. American Ethnologist 3:425445. Hellmuth, N. (1977) CholtiLacandon (Chiapas) and PeténYtzá agriculture, settlement pattern, and population. In N. Hammond (ed.): Social Process in Maya Prehistory. London: Academic Press.
Page 58
Herrera Castro, N. (1994) Los huertos familiares Mayas en el oriente de Yucatán. Etnoflora Yucatense, fasículo 9. Mérida: Universidad Autónoma de Yucatán. Hofling, A. (1997) ItzajMayaSpanishEnglish Dictionary. Salt Lake City: University of Utah Press. Hunn, E. (1977) Tzeltal Folk Zoology. New York: Academic Press. Hunn, E. (1999) Size as limiting the recognition of biodiversity in folk biological classifications. In D. Medin and S. Atran (eds.): Folkbiology. Cambridge: MIT Press, pp. 4769. Kesby, J. (1979) Range classification of animals and plants. In R. Reason and D. Ellen (eds.): Classifications in Their Social Contexts. New York: Academic Press. Lambert, J., and Arnason, J. T. (1982) Ramón and Maya ruins: An ecological, not an economic, relation. Science 216:298299. Landa, Diego de (1566/1985) Relación de la cosas de Yucatán. Ed. M. Rivera Dorado. Crónicas de America, No. 7. Madrid: Historia 16. Lentz, D. (1991) Maya diets of the rich and poor: Paleoethnobotanical evidence from Copán. Latin American Antiquity 2:269287. Lois, X. (1998) Gender markers as "rigid determiners" of the Itzaj Maya world. International Journal of American Linguistics 64:224282. Lois, X., and Vapnarsky, V. (forthcoming) Diccionario ethnolingüístico del Itzaj Maya. López, A.; Atran, S.; Coley, J.; Medin, D.; and Smith, E. (1997) The tree of life: Universals of folkbiological taxonomies and inductions. Cognitive Psychology 32:251295. Lundell, C. (1937) The Vegetation of Petén. Washington, D.C.: Carnegie Institution of Washington, Publication 478. Lundell, C. (1938) Plants probably utilized by the Old Empire Maya of Petén and adjacent lowlands. Papers of the Michigan Academy of Science, Arts, and Letters 24:3756. Marcus, J. (1982) The plant world of the sixteenth and seventeenthcentury Lowland Maya. In K. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press. Martinez Hernández, J. (1929) Diccionario de Motul. Mérida. Means, P. (1917) History of the Spanish Conquest of Yucatan and of the Itzas. Papers of the Peabody Museum, Vol. 7. Cambridge: Harvard University. Medin, D., Lynch, E., Coley, J., Atran, S. (1997) Categorization and reasoning among tree experts: Do all roads lead to Rome? Cognitive Psychology 32:4996. Morton, J. (1981) Atlas of Medicinal Plants of Middle America. 2 vols. Springfield, Ill.: Charles C Thomas. Nations, J., and Chan K'in, J. V. (1992) Vocabulario Lacandón: Plantas y animales. Paper prepared for Conservation International, April, Washington, D.C. Nations, J., and Nigh, R. (1980) Evolutionary potential of Lacandon Maya sustainedyield tropical forest agriculture. Journal of Anthropological Research 36:130. Paso y Troncoso, F. (1988/1886) La botánica entre los Nahuas y otros estudios. Mexico City: Consejo Nacional de Fomento Educativo. Puleston, D. (1978) Terracing, raised fields, and tree cropping in the Maya Lowlands. In P. Harrison and B. Turner (eds.): PreHispanic Maya Agriculture. Albuquerque: University of New Mexico Press, pp. 225245. Puleston, D. (1979) An epistemological pathology and the collapse, or why the Maya kept the short count. In N. Hammond and G. Willey (eds.): Maya Archaeology and Ethnohistory. Austin: University of Texas Press. Puleston, D. (1982) The role of ramón in Maya subsistence. In K. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press. Reina, R. (1967) Milpa and milperos: Implications for prehistoric times. American Anthropologist 69:120. Rice, D. (1993) Eighthcentury physical geography, environment, and natural resources in the Maya Lowlands. In J. Sabloff and J. Henderson (eds.): Lowland Maya Civilization in the Eighth Century A.D. Washington, D.C.: Dumbarton Oaks.
Page 59
Romney, A. K.; Weller, S.; and Batchelder, W. (1986) Culture as consensus: A theory of culture and informant accuracy. American Anthropologist 88:313338. Roys, R. (1931) Ethnobotany of the Maya. Middle America Research Institution Publication No. 2. New Orleans: Tulane University Press. Santley, R.; Killion, T.; and Lycett, M. (1986) On the Maya collapse. Journal of Anthropological Research 42:123159. Schele, L., and Freidel, D. (1990) A Forest of Kings: The Untold Story of the Ancient Maya. New York: William Morrow. Schwartz, N. (1990) Forest Society: A Social History of Petén, Guatemala. Philadelphia: University of Pennsylvania Press. Schwartz, N. (1995) Colonization, development, and deforestation in Petén, northern Guatemala. In M. Painter and W. Durham (eds.): The Social Causes of Environmental Destruction in Latin America. Ann Arbor: University of Michigan Press, pp. 101130. Siemens, A. (1982) Prehispanic agricultural use of the wetlands of northern Belize. In K. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press, pp. 205225. Sluyter, A. (1994) Intensive wetland agriculture in Mesoamerica: Space, time, and form. Annals of the Association of American Geographers 84:557584. Sosa, V., Salvador Flores, J., RicoGray, V., Lira, R., and Ortiz, J. (1985) Etnoflora Yucatense, fascículo 1: Lista florística y sinónima Maya. Xalapa, Mexico: Instituto Nacional de Investigaciones sobre Recursos Bioticos. Soza, J. M. (1970) Monografía del Departamento de El Petén. 2 vols. Guatemala: Editorial José de Pineda Ibarra. Standley, P., and Steyermark, J. (194677) Flora of Guatemala. Vol. 24, Parts 113. Chicago: Field Museum of Natural History. Stross, B. (1973) Acquisition of botanical terminology by Tzeltal children. In M. Edmonson (ed.): Meaning in Mayan Languages. The Hague: Mouton, pp. 109 141. Turner, B., and Miksicek, C. (1984) Economic plant species associated with prehistoric agriculture in the Maya Lowlands. Economic Botany 38:179193. Villagutierre SotoMayor, Juan de (1701/1985) Historia de la conquista de Itzá. Ed. J. García Añoveros. Crónicas de America, No. 13. Madrid: Historia 16. Wiseman, F. (1978) Agricultural and historical archaeology of the Maya Lowlands. In P. Harrison and B. Turner (eds.): Prehispanic Maya Agriculture. Albuquerque: University of New Mexico Press, pp. 63115.
Page 61
Chapter 3 Continuity and Variability in Postclassic and Colonial Animal Use at Lamanai and Tipu, Belize Kitty F. Emery Spanish arrival in the Americas precipitated dramatic changes in the political, economic, and ceremonial lives of many indigenous populations and often irrevocably altered even fundamental patterns of animal, plant, and other resource use (e.g., Reitz and Cumbaa 1983; Ruhl 1990). Recent evidence from the southern Maya Lowlands, however, suggests that in this region the Spanish faced a considerable and highly successful opposition. Here the Postclassic period was a time of great intellectual, political, and commercial activity, and initial reaction to the Spanish presence was not one of immediate submission or wholehearted acceptance. Although Spanish religious beliefs and economic and political authority were eventually accepted in the Maya Lowlands, commercial and missionizing contacts were initially brief, and the Maya retained political control for much longer than in the fully administered northern regions (Graham 1991; Jones 1989; Pendergast 1986a). To what extent was the conflict between acceptance of and resistance to the Spanish presence reflected in basic patterns of resource use in this frontier zone of resistance? A zooarchaeological investigation of animal bone assemblages from Postclassic and early Colonial archaeological deposits at the sites of Lamanai and Tipu, Belize, provides an analysis of the effects of early Spanish contact, a dramatic source of change in other regions, on fundamental patterns of animal use at these two politically and economically important southern lowland centers. A high degree of chronological contemporaneity and environmental similarity between the two sites, as well as contextual and preservational similarity between the archaeological deposits examined, allows a direct comparison of patterns of change in the consumption of animal species and the use of ecosystems in the region. The results provide a valuable opportunity to expand on a limited body of literature discussing the effects of Spanish contact on indigenous Postclassic resource use in the southern Maya Lowlands. These two important sites have been the focus of extensive archaeological and ethnohistoric research (Graham 1987, 1991; Graham et al. 1985; Jones 1985, 1989; Pendergast 1986a, 1986b, 1990, 1991) that documents not only extraordinarily long occupation histories (Preclassic to Colonial) but also the effects of initial contact between Maya and Spanish peoples at these sites. Both centers were active participants in Postclassic mercantile spheres and as such were powerful entities in
Page 62
the political scenario of the period. Evidence from ethnohistoric Spanish documents often conflicts with the archaeological records of these sites, revealing an intriguing complexity of acceptance and rejection of both religious and secular Spanish norms. Analyses of the faunal remains from Postclassic and Colonial deposits at Lamanai and Tipu were undertaken to answer three basic questions. Does chronological variation exist in patterns of animal use during these periods of extreme internal and external cultural stress? Are patterns of dietary continuity and/or change comparable at the two sites and can they be used to suggest any panregional adaptation in animal use practices at the time of Spanish contact? And if variability exists, is it representative of a continuation of Postclassic Maya practices, or are the changes wholly without historical precedent and thereby attributable to the effects of Spanish contact? Ecological analyses of animal community diversity reflected in faunal assemblages from the sites of Lamanai and Tipu were combined with traditional zooarchaeological analyses of chronological change in species and ecosystem utilization at each site to provide evidence for both regional and chronological patterning.1 Ecological community statistics are most commonly used to quantify the distribution of, and the relationships between, living taxa in a natural ecosystem. As analytical tools, they are valuable additions to the techniques of faunal analysis, as they provide more detailed information on changing temporal and spatial patterns of resource use than is generally available through the simple analysis of taxa recovery frequencies and discussions of emic values of utilized species. The use of these combined measures on comparable zooarchaeological assemblages from Lamanai and Tipu allows the documentation of the variability of responses to the stresses of cultural change through the description of two successful strategies of animal use during the important transition from the Postclassic to early Colonial period. The results of these analyses point strongly to several conclusions. Patterns of animal use at both Lamanai and Tipu changed both during the period between the Middle and Late Postclassic and during the transition to the Colonial period. There is little evidence for any relationship between the patterns seen at the two sites, and the temporal variations are more pronounced at Lamanai. Despite the regional and temporal discontinuity, however, the changes seen in the Colonial period appear to have strong roots in the Postclassic period patterns at each site. There is no appearance of a strictly Colonial animal use pattern at either Lamanai or Tipu. Methods The Zooarchaeological Samples The Lamanai sample, a total of 5,737 analyzed zooarchaeological remains, is derived from four midden lots originating from Postclassic and Colonial period settlements and spanning collectively the dates from A.D. 1150 to A.D. post1641 (Table 3.1). The four middens are securely dated and are directly associated 1
Full community descriptions are available in tabular form in Emery 1990a.
Page 63 Table 3.1. Chronological Listing of Zooarchaeological Assemblages. Date
Period
Lamanai
Tipu
A.D. 1520+
Colonial
N1118
H126,7,8,12,13,14
A.D. 14001520
Late Postclassic
N113
H126,8,12,13
A.D. 11501400
Middle Postclassic
N109
H1214
with three different structures, N109, N113, and N1118. All these structures were most likely the residences of elite members of the Lamanai community, and there is no evidence for a permanent Spanish residency at the site despite the association of several European objects with the Colonial period structure (Pendergast, personal communication 1989). The larger Tipu faunal assemblage consists of 24,590 analyzed specimens from excavations at seven separate structures in and around the main ceremonial complex of the site. Included in the sample are remains from undisturbed Middle and Late Postclassic period deposits sealed beneath later Colonial residences (structures H126, H1212, H1214), and Colonial period deposits from both these Colonial period constructions as well as Colonial residences unassociated with earlier structures (structures H128, H1213). Although disturbance of stratified occupation levels is common at Tipu, zooarchaeological samples for this analysis were selected only from unmixed deposits. European artifactual material and ethnohistoric records suggest that two of the Colonial residences may have been used on a temporary basis by members of the Spanish religious order (Graham, personal communication 1986). As Tipu was a small rural community at all times, its architectural emphasis was on domestic structures. The location of the sampled structures within the site core, however, argues for the relatively high status and wealth of their inhabitants. Assuming Sample Comparability Basic to this analysis of variability in animal use patterns is the assumption that the sites of Lamanai and Tipu are characterized by a high degree of cultural and environmental similarity. That these sites were occupied simultaneously and were affected by the same cultural trends during the Postclassic and Colonial periods has been established through extensive archaeological analysis by Pendergast and Graham (Graham 1991; Graham et al. 1985,1989; Pendergast 1986a, 1986b, 1991). The sites differ in their community layout, and as discussed above, in the distribution of architecture representing elite residences or ceremonial buildings. Nevertheless, the deposits examined are comparable in terms of both function and status. Although the excavated samples from Lamanai and Tipu are generally comparable, investigation at both sites was oriented toward Postclassic and Colonial structures which have some architectural complexity and which reflect the activities of the upper echelon of these periods. The faunal assemblages are not, therefore, representative of all time periods or of all members of the ancient population. The sample bias affects both assemblages,
Page 64
however, and although it does not affect these analyses, it affects comparisons with other sites. The sites are also ecologically comparable, located in the subtropical moist life zone dominated by highcanopy broadleaf forest (Hartshorn et al. 1984), and are characterized by the same general conditions of geology and climate. Both sites are also located on strategic freshwater tributaries, Lamanai on the northwest shore of the New River Lagoon, and Tipu on the west bank of the Macal River.2 Terrestrial associations at both sites would have included the built environment of human habitation, cultivated lands, secondary growth or high bush, and undisturbed pristine canopy forest. Swampy, lowlying bajo environments were within easy reach of Maya hunters at both sites (closer at Lamanai than at Tipu), as was the pine ridge zone of the Maya Mountains. Because of the similar availability and diversity of resources, differences observed between patterns of animal exploitation cannot be explained by any difference in ecological parameters. Finally, any analysis of variability in patterns of animal use may be confounded by biases introduced through the effects of variability in archaeological sampling strategies, taphonomy, or zooarchaeological sample sizes. Archaeological and zooarchaeological sampling strategies were identical at the two sites, although at both Lamanai and Tipu recovery methods differed between deposit types, as features, middens, and occupation surfaces were more frequently screened (1/4'') or water screened (1/16") than other deposits (Pendergast, personal communication 1988; Graham, personal communication 1987). Analysis of the taphonomic conditions of the Lamanai and Tipu samples (Emery 1990) indicate that bone preservation did not vary significantly between deposits or sites. And although the Tipu zooarchaeological assemblage is considerably larger than the Lamanai sample, regression analyses (Emery 1990) also show that the ecological trends and measures derived here do not result from sample size variability. These analyses are more fully described in Emery 1990 and 1990, where the effects of intra and intersite variability in sample composition and size were evaluated and shown to be negligible or easily circumvented by simple analytical manipulation. Zooarchaeological Analysis and Community Statistics The following statistical analyses are based on minimum number of individual (MNI) estimations, as the independence of this measure from discrepancies in both natural and introduced skeletal part frequencies, and its independence from the degree of fragmentation of skeletal parts, ensures a high degree of comparability between samples (CruzUribe 1988). Frequencies were derived from original identification lists produced by various researchers following standardized procedures that mitigated researcher variability during initial identification.3 MNI calculations for both the Lamanai and Tipu assemblages were based on provenience units defined by the excavators as chronologically associated occupation levels within structures (Emery 1990). Although the use of singlestructure assemblages may obscure the effects of food sharing be 2
The New River Lagoon system differs ecologically from the smaller, more riverine Macal. Located 70 kilometers inland from Chetumal Bay on the Caribbean coast, the lagoon is a primarily estuarine system in which certain peripheral and brackishwater species occur that are not found close to Tipu (Lovisek, personal communication 1986). 3.
Original identifications by P. Baker, C. Cathcart, K. Emery, C. Neill, and C. Yasui are on file at the University of Toronto Faunal ArchaeoOsteology Laboratory.
Page 65
tween larger residential units, it does not bias the analysis by assuming habitation associations where none have been proven. MNI frequencies for the provenience units were grouped into three chronologically defined zooarchaeological communities at each site for the purposes of statistical description and comparison (Table 3.1). For each temporal zooarchaeological community, relative species MNI frequencies were calculated and larger taxonomic and community associations were used to describe ecosystem resource utilization patterns. The accurate analysis of ecosystem use is based on the degree to which each animal species is associated with, or fidelic to, one or more ecological zones. To reduce the effects of natural variability in species fidelity to any single biotic community, the archaeological taxa were each allocated a percentage fidelity value for one or more of eight ecosystem associations on the basis of current knowledge of habitat preferences.4 These modified counts (termed eMNI here) were then used to determine proportionate representation of the various ecosystems in the zooarchaeological assemblage. A combination of measures was used to quantify the diversity of each zooarchaeological community. These included species heterogeneity, richness, evenness, and dominance. The ecological community statistic known as diversity or heterogeneity cannot be accurately derived without reference to its two distinct components: species richness or the number of taxa present in a collection containing a specified number of individuals; and species evenness, calculated as the similarity in abundance of several taxa in a sample. The distinctions between these measures allow consideration of the different properties of the structure of the ecological community. There has been considerable debate as to the effectiveness of the various measures of diversity in archaeological studies (CruzUribe 1988; Leonard and Jones 1989). An earlier study evaluated the potential of the variously proposed heterogeneity, richness, and evenness measures with respect to the Lamanai and Tipu zooarchaeological samples (Emery 1990b) and provided evidence for a lack of sample size dependency with the use of Simpson's index of heterogeneity and Odum's richness measure. Indices of general heterogeneity and richness were therefore derived using these measures. An index of species evenness was calculated using simple variance based on relative frequencies of species for each community as recommended by Bobrowsky and Ball (1989). Finally, the relative dominance of the primary (most frequently occurring) species in each community was defined as the difference between relative frequencies of occurrence of these species within the community. Once described, the taxonomic similarities between the zooarchaeological communities were quantified using Spearman's correlation coefficient tests of sample correlation based on unranked relative species frequencies for each chronological and spatial community to provide an analysis of locational variability and changing animal use over time. A chronological analysis of change over time was made for each of the descriptives derived for the zooarchaeological communities, and changes in community diversity, species frequencies, and ecosystem resource use were all evaluated. In all cases where the 4
A detailed discussion of the derivation of ecosystem fidelity measures and the numeric equivalents used in this and other analyses can be found in Emery 1990a and 1997. Species habitat fidelities are derived from various zoological sources (e.g., Emmons 1990; Leopold 1959).
Page 66
communities are compared, analyses are based on relative as opposed to absolute frequencies in order to negate the effects of sample size variability. Results Regional Disjunction—Chronological Continuity Analyses of community similarity (Table 3.2), taxonomic characteristics (Tables 3.3 and 3.4), ecological community statistics (Table 3.5), and environmental use patterns (Table 3.6) combine to provide information on both diachronic and synchronic patterning in animal use at Lamanai and Tipu. Spearman's tests of zooarchaeological community correlation, used to compare the Lamanai and Tipu chronological assemblages, reveal variability in species distributions between all the samples examined (Table 3.2). More important, however, the results indicate a significantly greater degree of variability between the sites of Lamanai and Tipu, both overall and during any one time period, than exists between the chronological communities at either site. The differences between overall patterns of animal use at these two sites are greater than the internal or temporal differences at either site. The most closely correlated communities are found at Lamanai, where the highest degree of relatedness is found between the Middle and Late Postclassic assemblages. The Tipu communities are almost equally well internally correlated overall, and the relationship between the Postclassic and Colonial communities here is also strong. The greatest divergences are found between any of the Tipu time periods and the Lamanai Colonial period, where both correlation and significance are minimal. Overall, the Spearman's tests reveal an intersite variability that is surprisingly high considering the cultural and environmental similarities between the two sites. As well, despite evidence for variability between dietary patterns over time at both Lamanai and Tipu, these results suggest a high degree of internal consistency at each site. Analysis of the changing patterns of animal and ecosystem resource use seen over the three periods surveyed suggests that despite obvious chronological variability, there are several significant overlying trends of continuity at each site. There is evidence for a generalized stability over time at Tipu in terms of not only specific species (Figures 3.1 and 3.2) but also overall diversity of species used (Figure 3.3). This is not the rule at Lamanai, where changes in heterogeneity (Figure 3.4), community associations (Figure 3.5), and species used (Figures 3.1 and 3.2) are more dramatic while remaining directionally consistent over time. There is some evidence for a similarity of changes occurring at the two sites during the transition from Middle to Late Postclassic, and there is considerable evidence that the more dramatic changes seen at both sites during the Colonial period are direct reflections of earlier Postclassic patterns. At both Lamanai and Tipu the Middle Postclassic period is characterized by very similar animal use patterns. High heterogeneity levels, the comparability of which is offset only partially by a lower species richness at Lamanai, are accompanied by a concentration on large mammals as well as secondary and
Page 67 Table 3.2. Spearman Correlation Coefficients for Lamanai and Tipu Assemblages.
Lamanai Middle Postclassic
Lamanai
0.7902
**
Colonial
0.6301
**
Tipu
Late Postclassic
Middle Postclassic Late Postclassic Colonial
Late Postclassic
0.6906
**
0.4660
**
0.4220
**
0.4874
**
Tipu Middle Postclassic
Colonial
Late Postclassic
0.4519
**
0.2343
0.4558
**
0.0797
0.4853
**
*
0.1603
0.5267
**
0.4963
**
0.5826**
*
Significance level .05.
**
Significance level .01 Table 3.3. Taxonomic Distribution of Lamanai Fauna.
Middle Postclassic
Lamanai Carcharhinidae
MNI
Late Postclassic
%
MNI
Colonial
%
MNI
Total %
MNI
%
—
—
—
—
1.00
0.50
1.00
0.32
Osteichthyes (5 spp.)
2.00
11.76
6.00
6.59
114.00
57.00
122.00
39.61
Crocodylus moreleti
1.00
5.88
1.00
1.10
2.00
1.00
4.00
1.30
Dermatemydidae
1.00
5.88
1.00
1.10
2.00
1.00
4.00
1.30
—
—
—
—
3.00
1.50
3.00
0.97
Crax rubra
1.00
5.88
8.00
8.79
8.00
4.00
17.00
5.52
Phasianidae
—
—
—
—
6.00
3.00
6.00
1.95
Meleagris spp.
—
—
13.00
14.29
38.00
19.00
51.00
16.56
Dasypus novemcinctus
1.00
5.88
2.00
2.20
2.00
1.00
5.00
1.62
Canis familiaris
1.00
5.88
7.00
7.69
1.00
0.50
9.00
2.92
Felis concolor
1.00
5.88
2.00
2.20
1.00
0.50
4.00
1.30
Anatidae
Panthera onca
—
—
2.00
2.20
—
—
2.00
0.65
Tapirus bairdii
1.00
5.88
1.00
1.10
—
—
2.00
0.65
Tayassuidae
1.00
5.88
5.00
5.49
5.00
2.50
11.00
3.57
Mazama americana
3.00
17.65
13.00
14.29
9.00
4.50
25.00
8.12
Odocoileus virginianus
2.00
11.76
27.00
29.67
8.00
4.00
37.00
12.01
Agouti paca
1.00
5.88
3.00
3.30
—
—
4.00
1.30
Dasyprocta punctata
1.00
5.88
—
—
—
—
1.00
0.32
17.00
100.00
91.00
100.00
200.00
100.00
308.00
100.00
TOTAL
Note: MNI values for families have been proportionately divided into species categories where these already exist, to reduce the effects of diversity inflation.
cultivated ecosystem resources, the result in part of the importance of brocket deer at both sites and a tertiary importance of reptiles, birds, and fishes. The two sites differ only in the secondary concentration on armadillo in addition to brocket deer at Tipu and the greater importance of pristine canopy forest species here than at Lamanai. Overall, the pattern is reminiscent of that found at many other Maya lowland sites, particularly during the Terminal and Postclassic periods (Hamblin 1984; Masson 1995; Pohl 1990,1994; Scott 1982).
Page 68 Table 3.4. Taxonomic Distribution of Tipu Fauna.
Middle Postclassic
Tipu
MNI
Rajiformes Lamnidae Carcharhinidae
—
Late Postclassic
%
MNI
—
Colonial
%
MNI
Total %
MNI
%
—
—
2.00
0.59
2.00
0.30
1.00
0.75
1.00
0.30
2.75
0.41
—
—
—
—
1.00
0.30
1.00
0.15
1.00
2.13
1.00
0.75
6.00
1.78
10.88
1.61
—
—
—
—
1.00
0.30
1.00
0.15
1.00
2.13
2.00
1.50
3.00
0.89
9.63
1.42
Petenia splendida
—
—
1.00
0.75
3.00
0.89
4.75
0.70
Cyprinodontiformes
—
—
—
—
1.00
0.30
1.00
0.15
Carangidae
1.00
2.13
—
—
—
—
3.13
0.46
Sparisomatinae
1.00
2.13
—
—
1.00
0.30
4.13
0.61
Scaridae
Siluriformes Ictaluridae Cichlidae
1.00
2.13
1.00
0.75
2.00
0.59
6.88
1.02
cf. Serpentes
—
—
—
—
1.00
0.30
1.00
0.15
Ctenosaura similis
—
—
1.50
1.13
4.66
1.38
6.16
0.91
Iguana iguana
2.00
4.26
1.50
1.13
2.33
0.69
5.83
0.86
Crocodylus moreleti
1.00
2.13
1.00
0.75
5.00
1.48
9.88
1.46
Dermatemys mawii
2.00
4.26
2.00
1.50
13.20
3.92
17.20
2.54
—
—
2.00
1.50
9.90
2.94
11.90
1.76
Emydidae (3 spp.)
2.00
4.26
5.00
3.76
8.78
2.61
23.80
3.52
Chelydra serpentina
1.00
2.13
—
—
—
—
3.13
0.46
Galliformes
—
—
1.00
0.75
3.00
0.89
4.75
0.70
Crax rubra
2.00
4.26
5.45
4.10
8.00
2.37
15.45
2.28
Penelope purpuracens
—
—
1.09
0.82
—
—
1.09
0.16
Ortalis vetula
—
—
1.09
0.82
—
—
1.09
0.16
Meleagris sp.
2.00
4.26
4.36
3.28
16.00
4.75
22.36
3.30
Rallidae
—
—
1.00
0.75
—
—
1.75
0.26
Psittacidae
—
—
1.00
0.75
1.00
0.30
2.75
0.41
1.00
2.13
5.00
3.76
8.00
2.37
19.89
2.94
Kinosternidae (2 spp.)
Didelphidae Tamandua sp.
—
—
1.00
0.75
1.00
0.30
2.75
0.41
6.00
12.77
12.00
9.02
46.00
13.66
85.79
12.68
Alouatta villosa
—
—
1.50
1.13
9.00
2.67
10.50
1.55
Ateles geoffroyi
—
—
1.50
1.13
—
—
1.50
0.22
Canis familiaris
1.00
2.13
3.00
2.26
7.00
2.08
15.38
2.27
Dasypus novemcinctus
cf. Canis latrans
—
—
1.00
0.75
1.00
0.30
2.75
0.41
Urcyon cineoargentus
—
—
1.00
0.75
5.00
1.48
6.75
1.00
cf. Felis concolor
—
—
2.00
1.50
3.50
1.04
5.50
0.81
cf. Panthera onca
—
—
3.00
2.26
1.50
0.45
4.50
0.67
cf. Felis pardalis
—
—
3.00
2.26
6.00
1.78
11.26
1.66
Felis wiedii
—
—
1.00
0.75
—
—
1.75
0.26
Felis yagouaroundi
—
—
1.00
0.75
—
—
1.75
0.26
Mustelidae
—
—
—
—
1.00
0.30
1.00
0.15
Procyon lotor
0.50
1.06
1.00
0.75
3.00
0.89
4.50
0.67
Nasua narica
0.50
1.06
—
—
3.00
0.89
3.50
0.52
Tapirus bairdii
1.00
2.13
2.00
1.50
9.00
2.67
15.63
2.31
—
—
2.00
1.50
26.00
7.72
29.50
4.36
Tayassu tajacu
3.00
6.38
4.00
3.01
—
—
16.39
2.42
Tayassu pecari
1.00
2.13
6.00
4.51
1.00
0.30
14.64
2.16
Mazama americana
7.00
14.89
10.00
7.52
31.00
9.20
70.41
10.41
Odocoileus virginianus
4.00
8.51
9.00
6.77
24.00
7.12
52.28
7.73
Tayassuidae
(table continued on next page)
Page 69
(table continued from previous page) Table 3.4 (continued) Taxonomic Distribution of Tipu Fauna.
Middle Postclassic
Late Postclassic
Colonial
Total
MNI
%
MNI
%
MNI
%
MNI
%
Sciuridae
—
—
1.00
0.75
2.00
0.59
3.75
0.55
Geomyidae (2 spp.)
—
—
3.00
2.25
8.00
2.37
13.25
1.96
Heteromyidae (2 spp.)
—
—
1.00
0.75
1.00
0.30
2.75
0.41
Cricetidae (5 spp.)
—
—
6.00
4.50
9.00
2.67
19.50
2.89
Coendou mexicana
—
—
1.00
0.75
—
—
1.75
0.26
Agouti paca
4.00
8.51
11.00
8.27
24.00
7.12
55.78
8.24
Dasyprocta punctata
1.00
2.13
6.00
4.51
11.00
3.27
24.64
3.64
Sylilagus brasiliensis
—
—
—
—
2.00
0.59
2.00
0.30
47.00
100.00
132.99
100.00
336.87
100.00
676.66
100.00
TOTAL
Note: MNI values for families have been proportionately divided into species categories where these already exist, to reduce the effects of diversity inflation. Table 3.5. Ecological Community Statistics for the Lamanai and Tipu Assemblages.
Middle Postclassic
Lamanai
Late Postclassic
Colonial
Total
N specimens
16.00
94.00
192.00
300
No. taxa
13.00
17.00
17.00
21.00
Heterogeneity (Simpson's)
30.00
7.50
4.55
6.89
Evenness (1/variance*100)
12.94
49.28
96.90
53.81
Richness (s—l/log N)
13.64
14.78
16.81
20.82
6.25
14.89
22.11
12.33
Dominance (species a—species b)
Tipu
N specimens
47.00
132.99
336.87
516.86
No. taxa
24.00
49.00
49.00
64.00
Heterogeneity (Simpson's)
19.39
29.55
19.10
21.19
*
Evenness (1/variance 100)
12.54
4.22
7.09
5.22
Richness (s—1/log N)
23.40
48.53
48.60
63.63
2.12
0.75
4.46
3.09
Dominance (species a—species b)
The Late Postclassic period shows an increasing disparity between the two sites in terms of both diversity of species represented and actual resources used. During the two hundred years between the Middle Postclassic period and the Late Postclassic, overall species heterogeneity at Lamanai drops dramatically as the result of a decreasing evenness of species representation and increasing species dominance as whitetailed deer becomes the most commonly used species at this site. Although mammals continue to dominate the sample, there is a slight reduction in their prominence and a generalized shift away from all smaller mammals with the exception of the domestic dog. Replacing the small mammals in importance are the avians, and in particular the turkey5 and curassow. The fishes also rise in importance, as do the turtles and riverine 5
The accurate identification of the ocellated (Meleagris ocellata) and common (Meleagris gallopavo) turkey is difficult and often possible only through detailed osteometric analysis. For this reason, all turkeys have been identified here as Meleagris spp.
Page 70 Table 3.6. Ecosystem Use Distributions at Lamanai and Tipu.
Middle Postclassic
eMNI
Lamanai
Late Postclassic
%
eMNI
Colonial %
eMNI
Total %
eMNI
%
Canopy forest
0.50
3.85
25.00
23.72
10.50
5.22
36.00
11.27
Riverine
1.50
11.54
9.50
9.01
116.00
57.71
127.00
39.76
Shoreline
2.00
15.38
2.00
1.90
6.00
2.99
10.00
3.13
High bush
3.40
26.15
25.40
24.10
11.50
5.72
40.30
12.62
Cultivated land
1.90
14.62
26.30
24.95
51.00
25.37
79.20
24.80
Habitation
1.00
7.69
7.00
6.64
1.50
0.75
9.50
2.97
Pine ridge
2.20
16.92
9.70
9.20
4.50
2.24
16.40
5.13
Bajo
0.50
3.85
0.50
0.47
0.00
0.00
1.00
0.31
Tipu Canopy forest
8.00
16.67
Riverine
6.00
12.50
Shoreline
9.40
19.58
High bush
40.00
29.01
74.00
21.96
122.00
9.00
6.53
18.00
13.05
23.33
32.00
9.50
47.00
8.99
41.80
12.40
69.20
13.23
11.00
22.92
24.80
17.98
73.00
21.66
108.80
20.81
Cultivated land
7.60
15.83
22.60
16.39
52.30
15.52
82.50
15.78
Habitation
0.50
1.04
10.60
7.69
22.00
6.53
33.10
6.33
Pine ridge
4.50
9.38
11.90
8.63
35.40
10.50
51.80
9.91
Bajo
1.00
2.08
1.00
0.73
6.50
1.93
8.50
1.6/3
Figure 3.1 Frequency distribution of mammalian remains at Lamanai and Tipu (percentage of total MNI).
molluscs.6 Species with a high fidelity for secondary forest, cultivated land, and canopy forests occur with much greater frequency during the Late Postclassic at Lamanai, while resources from other ecosystems almost disappear. In contrast, at the site of Tipu, overall species heterogeneity, richness, and evenness all rise to peak levels, while primary species dominance drops sharply. The brocket deer is replaced in dominance by a quartet of species—the armadillo, agouti, brocket, and whitetailed deer—and the prominence of agoutis and pacas suggests a much greater dependency on small mammals. As at Lamanai, there is a rise in species with high fidelity for secondary forest, cul 6
As a result of sporadic collection and inclusion (Pendergast, personal communication 1988), neither turtle carapace nor molluscan shell has been analyzed here, but large quantities of both turtle and Pomocea flagellata remains have been reported from Late Postclassic deposits at Lamanai (Pendergast 1986a; Emery 1990a), and undoubtedly their numerical inclusion in the analysis would provide evidence of a higher on riverine resources than is reflected.
Page 71
Figure 3.2 Frequency distribution of nonmammalian remains at Lamanai and Tipu (percentage of total MNI).
Figure 3.3 Tipu ecological community statistics of heterogeneity.
Figure 3.4 Lamanai ecological community statistics of heterogeneity.
Page 72
Figure 3.5 Ecosystem community associations at Lamanai and Tipu.
tivated land, and canopy forests; in contrast to the pattern at Lamanai, at Tipu the use of these resources does not indicate the exclusion of resources from other ecosystems. During the Colonial period the faunal assemblages at Lamanai and Tipu reflect even more divergent subsistence patterns. At Lamanai the Postclassic trend of lowered diversity continues as heterogeneity and species evenness continue to fall. At the same time, species richness continues to rise despite the increasing dominance of a small number of species. Directly reminiscent of their Postclassic abundance, fish rise to an overwhelming dominance during the Colonial period. This rise is accompanied by a simultaneous increase in the importance of birds, again foreshadowed by their Postclassic frequency. Turkeys and curassows are more frequent, and several avian species appear for the first time. As a result of these two changes, riverine resources are overwhelmingly abundant while cultivated land is the only other ecosystem used to any extent. Animal use at Tipu during the Colonial period is reminiscent of the earliest Postclassic pattern of generalized resource use. Heterogeneity and species evenness fall as the armadillo becomes the focus of a clear concentration. Species richness, however, is higher than at any other time. Reflecting patterns from the Middle Postclassic, turtles rise in importance again, although this is not the case for other reptiles whose abundance dropped during the Late Postclassic. In an interesting parallel to Lamanai, turkeys rise to become the dominant avian species. At Tipu all the ecosystems are still used to a similar extent as in previous periods, and there is evidence for a rebound to a Middle Postclassic distribution in the use of species with high fidelity for secondary growth, riverine, and canopy forest ecosystems. Trends in patterns of animal use over time at Lamanai and Tipu indicate that despite comparable heterogeneity at the two sites during the early phases of the Postclassic period, the diversity of utilized faunal species at Tipu remained high over time, whereas diversity at Lamanai dropped significantly during the Late Postclassic and continued to do so into the Colonial period. Stability over time at Tipu is reflected most prominently by the continuity in frequency of appearance of both avian species and the large mammals, particularly the peccaries and whitetailed deer, through time. However, an analysis of change in species richness and evenness over time shows that variation,
Page 73
though slight, does occur at Tipu. In combination with an overall increase in both species richness and evenness, shifts in species dominance are obvious. The slight Middle Postclassic dominance of the armadillo and brocket deer shifts during the Late Postclassic to a lower primary species dominance as represented by a concentration on four codominant species. During the Colonial period there is a clear return, though, to the Middle Postclassic concentration on the armadillo as a dominant species. For the most part, during the Colonial period the Tipuans seem to have returned to an earlier pattern of concentration on a wide diversity of species, emphasizing small mammals and reptiles and, to a limited extent, fish. Abundances remain relatively stable over time at Tipu, but they change quickly at Lamanai. The most dramatic differences in animal use appear in the Colonial period, characterized by a very high level of primary species dominance and a concentration on fish and birds. However, all the changes seen in the Colonial period at Lamanai are reflected by similar trends during the transition from the Middle to Late Postclassic period. Despite the fact that the overwhelming use of fish and the reduction in the use of large mammals at Lamanai seems to be a phenomenon of the Colonial period, this trend is also evident during the transition from the Middle to Late Postclassic period. Although the frequency of fish is much higher in the Colonial period, it has already begun to rise in the Late Postclassic period. As well, the increase in use of birds and decrease in use of reptiles and small mammals at Lamanai has its roots in the Postclassic period. Although the turkeys are much more frequent in the Colonial period than in any other period, both the turkeys and the curassow first appear at Lamanai in the Late Postclassic period. An examination of the trends of change in the use of ecosystem resources over time emphasizes the lack of change at Tipu in most aspects, compared with the more dramatic nature of the shifts occurring at Lamanai. At Lamanai the use of resources from cultivated land and from the rivers increases over time, while the use of every other type of ecosystem decreases in frequency of occurrence in the archaeological record. The Lamanaians began to use the river systems and cultivated land resources to the virtual exclusion of all other resources. This is not the case at Tipu, where the use of the majority of resources does not change significantly and generalization of ecosystem use remains the rule. Discussion Patterns of both change and continuity are apparent in the Postclassic and Colonial period faunal assemblages from Lamanai and Tipu. Although there is chronological variability in the patterns of species and ecosystem use at both sites, there is clear evidence for continuity of animal use practices between the Postclassic and Colonial periods. The observed changes apparently are not the result of direct Spanish contact but are instead related to Postclassic patterns. Furthermore, although Lamanai and Tipu are environmentally similar, chronologically contemporaneous, and located within easy access of each other, the
Page 74
faunal assemblages from the sites do not exhibit any similarity of diachronic variation. Tipu is characterized by a marked heterogeneity of animal species use in combination with a highly generalized use of available ecosystems in all time periods. The frequent use of the most abundant species of the Tipu faunal assemblages (armadillo, deer, agouti, and paca), in combination with high species richness and a consistently low level of primary species dominance, is reminiscent of most lowland Mesoamerican sites. The patterns of species use are significantly different at Lamanai, where a steady decrease in heterogeneity of species use over time and a dramatic increase in primary species dominance are observed. During the Late Postclassic period at this site the primary species is the whitetailed deer, but this shifts to a clear concentration on fish and turkeys during the Colonial period. Changes in animal and plant products use during times of societal stress have received relatively little theoretical consideration in Mesoamerican studies. Although these patterns are often analyzed with respect to the nutritional benefits and ecological adaptive value of varying strategies of the use of these products, their importance in the maintenance of external relationships through trade and tribute, and as a method of defining internal social relationships, cannot be ignored. The explanation for the chronological changes and regional variability seen in the Lamanai and Tipu zooarchaeological assemblages lies in an understanding of both the dietary and the social importance of the animal use strategies in each time period. The Postclassic period saw the dramatic increase in Classic period patterns of maritime trade (Andrews 1990) and, coincident with it, a shift in Postclassic settlement focus toward coastal, riverine, and lacustrine shorelines throughout the southern Mesoamerican lowlands (Chase and Chase 1985). Considerable archaeological and ethnohistoric evidence shows that Lamanai and Tipu were active players in the increasing commercialism of the Postclassic period as a whole, and particularly the Late Postclassic. Both Lamanai and Tipu were also affected by the Postclassic population movements caused first by the Itza expansion and then by its eventual collapse, and both had been the recipients of large groups of Yucatecan Maya refugees during the Middle Postclassic and in the later half of the sixteenth century during the period of first Spanish contact in the northern peninsula (Graham 1991). These groups perhaps carried ideas and goods that had a major impact on the Postclassic economies of Lamanai and Tipu, causing changes in animal use practices during the transition between the Middle and Late Postclassic periods. The appearance in Postclassic Lamanai deposits of trade goods from the Yucatán, Mayapanstyle vessels, Tulumstyle architectural constructions, and a new lithic technology characterized by small, sidenotched arrowheads (Pendergast 1990) suggests the movement of both goods and ideas from the north. Although a secure identification of the Lamanai turkeys as wild or domestic was not possible, their frequency in the Late Postclassic deposits at Tipu and Lamanai and their later rise to dominance at Lamanai suggests that their presence may be the effect either of direct importation of the animals or of the idea of their importance by Yucatecan peoples.
Page 75
A more significant arrival, though, is the bow and arrow technology, probably introduced at Lamanai in the Late Postclassic period, possibly in association with the arrival of Yucatecan refugees in the last half of the sixteenth century (Graham 1991). An interesting study of Lacandon Indian bow and arrow use indicates that several arrow types are specifically designed to shoot fish underwater and to kill birds in the tallcanopy forest (Nations and Clark 1983). The arrival of a new and more effective technology for the hunting of fishes and birds would certainly have affected the frequency of their appearance in the archaeological record. In contrast to the situation at Lamanai, the introduction of the bow and arrow seems to have been a Colonial phenomenon at Tipu (Graham 1991), and therefore the lack of a similar emphasis on these species at Tipu is not surprising. That Lamanai was not just the passive receptacle for the diffusion of Postclassic northern traditions is emphasized by several important technological innovations at the site. Evidence for copper manufacturing and working (Pendergast 1990) and for the in situ development of new ceramic traditions (Graham 1987; Pendergast 1986a) suggests that Lamanai may have offered valuable commodities of its own to its trading partners. The Late Postclassic rise in fish suggests that this may have been another of those commodities. In many cases the Postclassic settlement shift toward coast and shorelines was accompanied by a corresponding subsistence focus on aquatic resources as indicated by the presence of both aquatic faunal remains (e.g., Carr 1986; Pohl 1994; Scott 1982) and probable net weights in many archaeological deposits (e.g., Chase and Chase 1985). Several authors have suggested the importance of aquatic products as the primary protein source in the Late Postclassic period (Lange 1971; Wing 1977), and at the site of Lamanai, analyses of human bone isotope chemistry by Coyston et al. (this volume) indicate that fish were indeed an important resource during the Late Postclassic and Colonial periods. In association with their importance as subsistence resources during the Late Postclassic, the possibility of trade of piscine resources both out of sites (Hamblin 1984; Miller 1977; Mock 1997) and into them (Pollock and Ray 1957) has also been suggested by many authors. Located on the banks of an estuarine inland waterway teeming with freshwater and peripheral species, Lamanai would have been ideally situated for maximum access to both resources and trading systems. The rising frequency of fish in the archaeological record, in combination with the appearance in the Postclassic period of large numbers of standardized, firedclay, ball net sinkers replacing the Classic period notched, reused pottery sherds (Pendergast, personal communication 1990), supports the probability of a change in the importance of fishing technologies at Lamanai during the Late Postclassic period. Certainly, the combination of human dietary studies (Coyston et al., this volume) and zooarchaeological evidence suggests that fish began to play a more significant subsistence role during this period. It has also been suggested that one of the main foci of Postclassic commerce was an increasing demand for certain animal species for tribute and use as sacrificial victims (Pohl and Feldman 1982; White and Schwarcz 1989). Among these species are the turkey, cervids, turtles, and domestic dog. The evidence
Page 76
for increasing frequencies of these species at Lamanai during the Late Postclassic period suggests they may have been traded as well, either out of or perhaps into Lamanai. Interestingly, there is no correlative evidence of rising frequencies of any of these possible tribute species at Tipu (Emery 1990a). At the site of Lamanai, archaeological evidence for strong trade links with the politically vibrant northern Yucatán combines with evidence for increasing frequencies of animal species potentially valuable for subsistence and trade to suggest an active participation in the commercial network of the period that was dependent in part on these animals. At Tipu, where dietary heterogeneity seems to rise in the Late Postclassic, there is no evidence for any increase in the frequency of either fishes or the deer, dog, turkey, and turtles potentially important as trade items. At the same time, there is evidence at the site for a generalized reduction in all exotic species during the Late Postclassic period (Emery 1990a), indicating a reduction in contact with external, and particularly coastal, sources. The residents of Tipu were not participating in the trade of animal resources to the same extent as those at Lamanai. Possibly because of their stronger ties with the inland southern lowland polity, the residents of Tipu were not as active participants in the Yucatánbased trading systems; in fact, they reduced their trading circle to the extent that exotic resources were limited, relying instead on a continuing and even increasing diversity of local resources. Both Lamanai and Tipu were occupied well into the period of Colonial Spanish contact. The first physical evidence for contact at Lamanai was the demolition in 1544 of a small Late Postclassic temple and the construction of a Christian church in its place (Pendergast 1986b, 1991). It was also at this time that Lamanai was brought under encomienda control (Jones 1985). A similar church was built at Tipu some 20 to 25 years later (Graham, personal communication 1987). The later construction at Lamanai of a second church on a much more European plan suggests an increase in both population (Pendergast 1986b) and Spanish presence at the site, which is supported by the ethnohistoric evidence for an increased use of the site for reducciones during this period (Pendergast 1991). At both sites, however, the Spanish presence probably did not exceed the periodic visitations characteristic of the Spanish visita (circuit riding) system. The strong affiliation of the Lamanai and Tipu dietary practices with earlier Postclassic practices and the corresponding lack of discontinuity associated with the Colonial period are supported by isotopic (White and Schwarcz 1989), paleopathological (White et al. 1994), and archaeological evidence from the sites, suggesting that none of the basic patterns of existence at Lamanai or Tipu were significantly affected by the Spanish presence. Patterns of community organization (Pendergast 1986b), architectural construction, refuse handling, and ceramic and lithic production remained similar to earlier Postclassic patterns at Lamanai (Pendergast 1991). The ethnohistoric record defines an early and relatively peaceful incorporation of Christianity into the Tipu belief system, and this is supported by evidence for relatively strict Spanish control over community layout at this site (Graham 1991) as well as an incorporation of
Page 77
Christian burial practices into the local system. Nevertheless, the considerable archaeological evidence for continuity with Postclassic patterns of building construction and local pottery and lithic production and use suggest that basic lifeways were relatively unaffected by the Spanish presence (Graham 1987). The lack of any zooarchaeological remains from European animals at either site is also consistent with the relative dearth of European goods in any Colonial deposits (Pendergast 1991). Despite ethnohistoric evidence suggesting the strict censuring and manipulation of traditional foods and eating habits with the arrival of the Spanish missionaries (Landa 1941), there is no evidence for a discontinuity in animal use patterns at either site. In total, the zooarchaeological evidence argues that direct Spanish influences on diet and other animal product use were transitory or nonexistent. Changes in patterns of animal utilization occurred at both sites, however, and are particularly apparent at Lamanai. Zooarchaeological evidence from this site indicates a dependency on species with a high fidelity for riverine and cultivated land ecozones and the lack of the expected diversity of animal resources. These trends may be better understood with reference to the specifics of Colonial Tipu and Lamanai than as the result of direct Spanish intervention. Lamanai, located close to the Spanish center of control at Bacalar, functioned as a center for reducciones throughout its colonial history (Pendergast 1991) and was probably subject to the stresses of both population changes and unanticipated growth during this time. Zooarchaeological evidence from Lamanai indicating an expanded use of animal resources from agricultural land coincides well with the image of a population stretched to the limits of its resource availability, particularly in view of the limited experience of the newcomers in the manipulation of the Lamanai environment. The availability of both bow and arrow and fish netting technology, introduced during the Postclassic period, would have made the availability of the lagoon fishes an even more attractive dietary possibility. Finally, evidence from studies of skeletal health at Lamanai (White et al. 1994) indicating a decrease in overall health of the population during the Colonial period also supports the suggestion of Colonial period stress. There is considerable evidence that the arrival of the Spanish signaled a disruption of the coastal trading systems so important during the Postclassic (Graham 1991), both as an effect of simple presence and through active attempts to disrupt the contacts between the Yucatán and the inland Itza peoples (Graham et al. 1985; Jones 1989). Lamanai, maintaining its vitality and rising in importance as a dominant center in the Chetumal province (Pendergast 1986b), undoubtedly remained active in the trade of important resources. It is during this period that fish remains are most abundant in the deposits at Lamanai. The inland location of Lamanai, well protected from the disruptive influence of the Spanish presence, and its continued access to peripheral fish species, left its residents ideally situated for the continuation and intensification of Postclassic patterns of fish use as a dietary species as well as in trade. Tipu, always the more remote site, and one with less active Yucatán contacts even during the Postclassic period, remained a relatively stable community into which reducción populations were only infrequently introduced and at
Page 78
which the patterns of life remained relatively uninterrupted (Graham 1991). The relative dietary stability at Tipu is reflected in studies of the Tipu skeletal population that indicate the maintenance of a high level of health even in the Colonial period at this site (Cohen et al. 1994). Finally, some authors have suggested that choice of foods consumed may be less a passive response to external change than an active response to perceived societal changes (Kalcik 1984). Under Colonial conditions responses may be mediated by the extent to which prestige or even survival is attached either to the adoption of new ethnic or societal markers or to the conservation of old ones. The process of religious conversion in a missionizing situation is implicitly tied to changes in basic patterns of life involved in the maintenance of social cohesion, such as sexual practices, eating habits, and even village construction. Village pattern was often manipulated to conform to Spanish ideals, family affiliations were undermined through the outlawing of extended family residences and multiple marriages, and food and eating habits were strictly censured in many ways (Landa 1941). Kalcik's (1984) suggestion that food preferences are an active signifier of cultural and social identity is applicable to any situation in which that identity is threatened. Contact with missionizing Spaniards was a situation that severely threatened the Maya identity. The maintenance of Late Postclassic traditions of animal use at Lamanai, and the apparent return to earlier Postclassic subsistence patterns at Tipu, may signify a conscious return to traditional Maya foodways and identity. Conclusions The survival of both Lamanai and Tipu well into the Colonial period and the maintenance of their cultural integrity is evidence for the success of their very different strategies of animal utilization in the face of the extreme cultural stress of both the Postclassic and the Colonial periods in the southern Mesoamerican lowlands. The benefits of dietary stability at Tipu and of variation in patterns of animal use at Lamanai may well have outweighed the costs accrued by the maintenance of these strategies. Seagraves (1974) has emphasized the advantages of environmental generalization for the preservation of cultural stability in times of external stress. At Tipu, where the diversity of species use is high at all times, the benefits of maintaining this stability in the face of several periods of dramatic cultural stress are clear. At Lamanai economic benefits may have been the prime consideration. Maintaining and increasing a position of commercial strength provides a measure of safety during periods of cultural change, population movements, and resource stress. Cultural instability and flexibility may have been the rule throughout the Postclassic at Lamanai. This is indicated by the shifting use of both species and ecosystems, by the dramatic drop in overall heterogeneity of animal species used, and by the increasing reliance on primary dominant species. Seagraves (1974) would suggest that, in a situation of conflict with another culture, Lamanai would bear the losses. However, perhaps a position of
Page 79
economic and possibly political strength would have counterbalanced the lack of resource generalization and buffered the effects of confrontation. As well, control of one of the most important routes of access between the coast and the inland jungles would have provided the Lamanaians with a strong bargaining position. In terms of cost, economic strength may have been a more effective adaptive strategy for the inhabitants of Lamanai than cultural stability through resource generalization. Acknowledgments This research is based on a master's thesis completed at the University of Toronto (Emery 1990a) and financially supported by that institution, an Ontario Graduate Scholarship, and various grants in the names of David Pendergast and Elizabeth Graham. A large proportion of the preliminary Tipu identifications were generously provided by Cathy Yasui and Polydora Baker. My thanks go to David Pendergast and Elizabeth Graham, who provided excavation opportunities at the sites of Lamanai and Tipu, access to the zooarchaeological samples, and most important, guidance and enthusiasm throughout the years of this research. Thanks also go to the three anonymous reviewers and Christine White, who provided both constructive criticism and editorial assistance. References Cited Andrews, A. P. (1990) The role of trading ports in Maya civilization. In F S. Clancy and P. D. Harrison (eds.): Vision and Revision in Maya Studies. Albuquerque: University of New Mexico Press, pp. 159167. Bobrowksy, P. T., and Ball, B. F. (1989) The theory and mechanics of ecological diversity in archaeology. In R. D. Leonard and G. T. Jones (eds.): Quantifying Diversity in Archaeology. Cambridge: Cambridge University Press. Carr, H. S. (1986) Faunal Utilization in a Late Preclassic Maya Community at Cerros, Belize. Unpublished Ph.D. dissertation, Department of Anthropology, Tulane University. Chase, A. F., and Chase, D. Z. (1985) Postclassic temporal and spatial frames for the Lowland Maya: A background. In A. F. Chase and P. M. Rice (eds.): The Lowland Maya Postclassic. Austin: University of Texas Press, pp. 122. Cohen, M. N.; O'Connor, K.; Danforth, M.; Jacobi, K.; and Armstrong, C. (1994) Health and death at Tipu. In C. S. Larsen and G. R. Milner (eds.): In the Wake of Contact: Biological Responses to Conquest. New York: WileyLiss, pp. 121133. CruzUribe, K. (1988) The use and meaning of species diversity and richness in archaeological faunas. Journal of Archaeological Science 15(2):179196. Emery, K. F. (1990a) Postclassic and Colonial Period Subsistence Strategies in the Southern Maya Lowlands: Faunal Analyses from Lamanai and Tipu, Belize. Unpublished M.A. thesis, University of Toronto. Emery, K. F. (1990b) Manipulation of resource diversity: Strategy choices in the development of tropical urbanism. Paper presented at the Annual Meeting of the American Anthropological Association, New Orleans, December. Emery, K. F. (1997) The Maya Collapse: A Zooarchaeological Investigation. Unpublished Ph.D. dissertation, Cornell University. Emmons, L. H. (1990) Neotropical Rainforest Mammals: A Field Guide. Chicago: University of Chicago Press.
Page 80
Graham, E. (1987) Terminal Classic to Early Historic period vessel forms from Belize. In P. M. Rice and R. J. Sharer (eds.): Maya Ceramics. BAR International Series 345(i). Graham, E. (1991) Archaeological insights into Colonial period Maya life at Tipu, Belize. In D. H. Thomas (ed.): Columbian Consequences. Vol. 3: The Spanish Borderlands in PanAmerican Perspective. Washington, D.C.: Smithsonian Institution Press, pp. 319335. Graham, E.; Jones, G. D.; and Kautz, R. R. (1985) Archaeology and ethnohistory on a Spanish colonial frontier: An interim report on the MacalTipu Project in western Belize. In A. Chase and P. Rice (eds.): The Lowland Maya Postclassic. Austin: University of Texas Press, pp. 206214. Graham, E.; Pendergast D. M.; and Jones, G. D. (1989) On the fringes of conquest: MayaSpanish contact in colonial Belize. Science 246:12541259. Hamblin, N. L. (1984) Animal Use by the Cozumel Maya. Tucson: University of Arizona Press. Hartshorn, G.; Nicolait, L.; Hartshorn, L.; Bevier, G.; and Brightman, R. (1984) Belize Country Environmental Profile: A Field Study. Belize City: R. Nicolait and Associates. Jones, G. (1985) MayaSpanish relations in sixteenthcentury Belize. Belcast Journal of Belizean Affairs 1(1):2840. Jones, G. (1989) Maya Resistance to Spanish Rule: Time and History on a Colonial Frontier. Albuquerque: University of New Mexico Press. Kalcik, S. (1984) Ethnic foodways in America: Symbol and the performance of identity. In L. K. Brown and K. Mussell (eds.): Ethnic and Regional Foodways in the United States: The performance of group identity. Knoxville: University of Tennessee Press, pp. 3765. Landa, D. de (1941) Landa's Relación de las cosas de Yucatán. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 28. Cambridge: Harvard University. Lange, F. W. (1971) Marine resources: A viable subsistence alternative for the prehistoric Lowland Maya. American Anthropologist 73(3):619639. Leonard, R., and Jones, G. (1989) Quantifying Diversity in Archaeology. Cambridge: Cambridge University Press. Leopold, A. S. (1959) Wildlife of Mexico: The Game Birds and Mammals. Berkeley: University of California Press. Masson, M. A. (1995) Community feasting rituals and Postclassic Maya village political structure: Evidence from archaeofaunal remains. Paper presented at the Annual Meeting of the Society for American Archaeology, Minneapolis, May. Miller, A. G. (1977) The Maya and the sea: Trade and cult at Tancah and Tulum, Quintana Roo, Mexico. In E. P. Benson (ed.): The Sea in the PreColumbian World. Washington, D.C.: Dumbarton Oaks Research Library and Collections, pp. 97138. Mock, S. B. (1997) Monkey business at Northern River Lagoon: A coastalinland interaction sphere in northern Belize. Ancient Mesoamerica 8:165183. Nations, J. D., and Clark, J. E. (1983) The bows and arrows of the Lacandon Maya. Archaeology 36(1):3643. Pendergast, D. M. (1986a) Stability through change: Lamanai, Belize, from the ninth to seventeenth century. In J. A. Sabloff and E. W. Andrews V (eds.): Late Lowland Maya Civilization: Classic to Postclassic. Albuquerque: University of New Mexico Press, pp. 223249. Pendergast, D. M. (1986b) Under Spanish rule: The final chapter in Lamanai's Maya history. Belcast Journal of Belizean Affairs 3(1/2):17. Pendergast, D. M. (1990) Up from the dust: The central lowlands Postclassic as seen from Lamanai and Marco Gonzalez, Belize. In F. S. Clancy and P. D. Harrison (eds.): Vision and Revision in Maya Studies. Albuquerque: University of New Mexico Press, pp. 169177.
Page 81
Pendergast, D. M. (1991) The southern Maya Lowlands contact experience: The view from Lamanai, Belize. In D. H. Thomas (ed.): Columbian Consequences. Vol. 3: The Spanish Borderlands in PanAmerican Perspective. Washington, D.C.: Smithsonian Institution Press, pp. 337354. Pohl, M. D. (1990) Ethnozoology of the Maya: Faunal remains from five sites in the Petén, Guatemala. In G. R. Willey (ed.): Excavations at Seibal, Guatemala. Peabody Museum Monographs, Vol. 17, No. 3, pp. 142174. Cambridge: Harvard University Press. Pohl, M. D. (1994) The economics and politics of Maya meat eating. In E. M. Brumfiel (ed.): The Economic Anthropology of the State. New York: University Press of America, pp. 119148. Pohl, M. D., and Feldman, L. H. (1982) The traditional role of women and animals in Lowland Maya economy. In K. V. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press, pp. 295312. Pollock, H. E. D., and Ray, C. E. (1957) Notes on Vertebrate Animal Remains from Mayapan. Carnegie Institution Report No. 41. Washington, D.C.: Carnegie Institution. Reitz, E. J., and Cumbaa, S. L. (1983) Diet and foodways in eighteenthcentury Spanish St. Augustine. In K. Deagan (ed.): Spanish St. Augustine. New York: Academic Press, pp. 151185. Ruhl, D. (1990) Spanish mission palaeoethnobotany and culture change. In D. H. Thomas (ed.): Columbian Consequences. Vol. 2: Archaeological and Historical Perspectives on the Spanish Borderlands East. Washington, D.C.: Smithsonian Institution Press, pp. 555580. Scott, R. F. (1982) Notes on the continuing faunal analysis for the site of Colha, Belize: Data from the Early Postclassic. In T. R. Hester, H. J. Shafer, J. D. Eaton (eds.): Archaeology of Colha, Belize: The 1981 Interim Report. San Antonio: Center for Archaeological Research, University of Texas, pp. 203288. Seagraves, B. A. (1974) Ecological generalization and structural transformation of sociocultural systems. American Anthropologist 76:531552. White, C. D., and Schwarcz, H. P. (1989) Ancient Maya diet: As inferred from isotopic and elemental analysis of human bone. Journal of Archaeological Science 16:451474. White, C. D.; Wright, L. E.; and Pendergast, D. M. (1994) Biological disruption in the early Colonial period at Lamanai. In C. S. Larsen and G. R. Milner (eds.): In the Wake of Contact: Biological Responses to Conquest. New York: WileyLiss, pp. 135145. Wing, E. S. (1977) Factors influencing the exploitation of marine resources. In E. Bronson (ed.): The Sea in the PreColumbian World. Washington, D.C.: Dumbarton Oaks Research Library and College.
Page 83
Chapter 4 Social and Ecological Aspects of Preclassic Maya Meat Consumption at Colha, Belize Leslie C. Shaw The last few decades have seen great strides in the study of faunal resource use by the Preconquest Lowland Maya. Although quantitative approaches, retrieval methods, and dietary reconstruction have all been dramatically improved, we have moved more slowly in connecting this new information to the dynamic social forces that shaped so much of Maya life. The early Maya faced a particular challenge in their growth to a statelevel society because of the absence of large domesticated animals like those that supported Old World civilizations. The picture now emerging for the Preclassic period Maya (250 B.C.A.D. 250) is one in which there was considerable experimentation and variation in the strategies used to acquire meat for food. From this experimentation, I argue, came the strategies that dominated meat procurement and resource control used to support the large populations of the Late Classic period. The recognition of this experimentation and variability in Preclassic faunal use has come about because of the attention to faunal recovery at several site excavations that focused on Preclassic occupations. In particular, the excavation of residential areas at these sites has begun to provide the ''social crosssection" needed for the recognition of coeval differences in procurement strategies and distribution parameters. The data used here to identify these strategies come primarily from the site of Colha in northern Belize (Figure I.1), a site with a long and continuous record of Preclassic period occupation (Hester and Shafer 1984; Shafer and Hester 1983). The Colha data are considered against several other key studies, most notably those from the sites of Cuello (Wing and Scudder 1991), Cerros (Carr 1985,1986), and Cahal Pech (Stanchly 1995). An understanding of the significance of patterning in faunal remains requires that faunal use be viewed within the larger structure of society. This is perhaps especially true for complex, agricultural societies. Many of the models used to project subsistence strategies are derived from studies of hunters and gatherers (e.g., catchment analysis, optimal foraging). These models see the procurement of faunal resources as a primary and largely independent activity. Simply stated, the models assume that people often leave their residential areas with the primary or sole purpose of procuring meat for food. Such models then highlight the articulation between faunal availability on the one hand and the efficiency of human procurement strategies on the other. This dichotomy is not
Page 84
as clearcut for sedentary, agricultural communities. For agricultural households, the primary reason for leaving the residential area is likely to be to attend to gardens, with faunal procurement being a secondary, although clearly complementary, activity. This agricultural pattern has been illustrated before, most notably by Linares (1976; see also Redford and Robinson 1987), but it is rarely addressed explicitly in interpretations of Maya faunal data. In addition, when garden hunting is proposed by archaeologists, it is often expressed behaviorally as the passive collection of what was available, as opposed to an active form of management of meat resources within the garden area. As societies become more complex, there is a greater potential for the introduction of surplus subsistence goods into regional market exchange, thus allowing some households to reduce their own food production significantly. The identification of food traded into or out of a community is challenging to identify archaeologically. The presence of exotic species in a faunal assemblage can signal farreaching exchange relationships, even if their numbers are small in relation to more readily available meat resources. An important clue to the exchange of faunal resources is the recognition of largescale processing areas at a site which may indicate the preparation of a surplus for transport, as suggested by Carr (1986) for a fishprocessing area at Cerros. Interpretive Methodology One of the most important considerations in Maya faunal studies today is interpretive methodology. The use of intensive recovery methods, such as flotation and fine screening, in Maya site excavations is now almost standard. The means by which we arrive at the interpretations based on these faunal assemblages remain the limiting factor in connecting meat resources with the social conditions under which they were used. I have argued in greater detail elsewhere that sample context is particularly critical in the interpretation of quantitative results (Shaw 1991), and not simply temporal context but functional and social contexts as well. The Maya were clearly great managers of domestic waste, as seen by evidence for its use as fill in construction or as garden fertilizer. It is important to consider how a deposit containing faunal remains was created and manipulated and to recognize that disposal practices may have been dictated by a variety of both practical and symbolic parameters. The temporal context of a faunal assemblage is often determined by the archaeologist, but all too often it is the only contextual variable addressed. The emphasis on temporal associations is due in part to the common interest by archaeologists in changes in subsistence practices through time. Although temporal variation is important for many research problems, the use of time as the only contextual trajectory along which to evaluate similarities and/or differences between samples fails to address the full range of variation in the data. An overemphasis on time as the comparative delimiter imposes two embedded assumptions: that all members of a group (or social subgroup) had the same diet during a given time period, and that the faunal assemblage recovered through excavations reflects that diet. Clearly, this is rarely the case, especially when one
Page 85
is dealing with complex, sedentary societies (Crabtree 1990). The variability introduced by socially dictated access to resources and functionally different consumption behaviors acts to shape faunal use. The temporal contexts of a faunal sample must be evaluated in conjunction with a consideration of both functional and social contexts if there is to be an accurate identification of changes in faunal use through time. The functional context of a faunal sample relates to the behavioral processes that created the deposit. This may include such humanly made contexts as middens, hearths, pits, construction fill, and surface scatter or materials in their primary context of use. All too often, different functional contexts will have been lumped by the archaeologist, usually according to temporal associations, thus obscuring variations that may have been due to differences in consumption and/or disposal behavior. We cannot assume that the Maya disposed of their food waste randomly between middens and pits or, even if they did, that the different contexts would be equally affected by postdepositional disturbances. When samples from various functional contexts are lumped together, any cultural or natural factors that had influenced sample composition are masked. Few studies have focused directly on variation between functional contexts, but those that have (e.g., Styles 1981) have identified differences in the places where people disposed of various types of faunal waste. The social contexts of faunal samples include such factors as status, occupation, and/or kinship. It is likely that elite and nonelite meat consumption would not have been equivalent in the Late Preclassic period, if the variability seen in Late Classic and Postclassic contexts can be used as a guide (Hamblin 1984; Pohl 1976). More refined comparisons are needed, however, to move beyond this basic distinction, especially comparisons that consider in detail the status, occupation, and lineage information from the structures associated with a particular faunal sample. This might allow more specific statements to be made about differences in faunal use between craft specialists and agriculturalists or between a ruling elite and a merchant elite. The method used here for the quantification of the faunal data is bone counts (number of identified specimens, or NISP). The minimum number of individuals (MNI) was calculated, and this information is presented elsewhere (Shaw 1991, in press). The MNIS were generally low for the midden contexts under discussion, probably because of sample size (e.g., Grayson 1984), and do not alter the patterns presented using the NISP data. The Preclassic faunal data from Colha came from only two areas in the site and therefore offer only a selective view when considered contextually. The focus on these areas does allow for careful control on the functional and social contexts, in which the strategies for faunal procurement could be evaluated across the 1,250 years of domestic occupation within a restricted area. The evaluation of the data that are available is approached here as a building block on which other studies of Colha fauna may be placed. In addition, the growing number of detailed faunal studies from Central Lowland sites will soon allow for better controls on all three contextual variables. Through this, a more complete picture will emerge of Maya faunal use through time and across space.
Page 86
The Site of Colha The Maya site of Colha (Figure I.1), located in northern Belize along Rancho Creek, has provided the necessary conditions for exploring this relationship between changing social conditions and meat resources. Colha was a moderatesized community that supported the production of lithic utilitarian tools and prestige items, most of which were moved through northern Belize in extensive regional exchange networks (Hester and Shafer 1984; Shafer and Hester 1983). The multiyear excavation project conducted at Colha, primarily directed by Thomas Hester and Harry Shafer, identified settlement at the site by the early Middle Preclassic (1,000600 B.C.), with earlier occupation suggested by pollen data (Jones 1991). Occupation continued until the end of the Early Postclassic period (roughly A.D. 1250), but, like most Maya centers, the community experienced some rather dramatic fluctuations in population. Colha was the major producer of lithic tools for northern Belize during the Late Preclassic period (300 B.C.A.D. 250), with the initial production of stone tools for distribution to sites beyond Colha probably dating to the Middle Preclassic (Hester and Shafer 1984). Many Late Preclassic lithic workshops have been identified at Colha, and these represent the activities of craft specialists who produced standardized tools that included both utilitarian and ceremonial/prestige types (Hester 1985; Hester and Shafer 1984; Shafer and Hester 1983,1986). Based on the volumes of debitage found in the workshops, it is estimated that millions of utilitarian tools were produced, most of which were the oval bifaces and tranchetbit implements used in land clearing and agriculture (Shafer and Hester 1986). These massproduced utilitarian tools were traded throughout the northern Belize region. They have been found at numerous sites, including Cerros, Cuello, El Posito, Kichpanha, Nohmul, and the wetland agricultural fields of Pulltrouser Swamp and Albion Island (Shafer and Hester 1990). The quantities of lithic tools that were moved across northern Belize attest to the magnitude of the regional exchange networks. The clearing of wetland agricultural fields in northern Belize relied heavily on tools produced at Colha (McAnany 1989; Shafer and Hester 1990). This involvement in the exchange of utilitarian items suggests that subsistence goods produced in the wetland areas were included in the regional exchange networks (e.g., Pohl 1990; Turner and Harrison 1983). The faunal assemblage evaluated here comes from two discrete occupation areas at Colha. The excavations at Operation 2031 (Figure 4.1) encountered 1,250 years of residential activity that began in the early Middle Preclassic and continued with multiple remodeling as a residence until the end of the Preclassic, when the area was incorporated into a ceremonial space (Anthony and Black 1994). Operation 2012, located nearby (Figure 4.1), also began in this early period as a residence but had shifted in use by the Late Preclassic to ceremonial purposes (Potter 1982). Although this faunal assemblage represents only a small fraction of what was consumed at the growing community of Colha, the focus on changes over time at the household level provides exceptional controls over functional and social contexts. The evaluation of how emerging
Page 87
Figure 4.1 Map of the ceremonial center at Colha indicating Operations 2012 and 2031 (from Anthony and Black 1994).
social inequalities might have affected access to meat resources requires a finegrained look at change over many generations. As the data bear out, this strategy supports an interconnection between social status and access to meat resources that would not have been apparent in a more broadly grouped data set. The Colha Preclassic vertebrate faunal assemblage (n = 14,553) included samples from 68 different contexts within Operations 2012 and 2031 dating from the early Middle Preclassic (900600 B.C.) through the Terminal Preclassic period (A.D. 100250) (Shaw 1991). Although all the faunal remains recovered from these operations were analyzed, the samples from a total of six residential midden contexts in Operations 2012 and 2031 proved to be the most useful for crosscomparison in the evaluation of the social and environmental impacts on meat diet through time. This is because the samples were of like
Page 88
size, they were from similar functional contexts, they had comparable postdepositional histories, and each included a flotation sample for the recovery of small remains. These six samples offer an opportunity for comparison through time with a relatively tight control on functional and social influences. Interpreting the Faunal Record The Preclassic faunal assemblage from Colha has been evaluated along several lines of inquiry, and the primary data and statistical manipulations of these data can be found in detail elsewhere (Shaw 1991, in press). The effort presented here pulls together the results of this comprehensive study, particularly in how environmental and social factors impacted the strategies used to acquire meat resources during the 1,250 years of occupation of these Preclassic household groups. The patterns of faunal use identified indicate a change in the procurement strategies used, with strong support for the position that the social and economic networks that connected all of northern Belize by the Late Preclassic included the exchange of faunal resources. The focus on a limited area within the site of Colha restricts the interpretations to the household level, but the patterns that were identified in this context provide an excellent situation for detailed comparisons, at both the community and the regional level. Garden Hunting in the Early Middle Preclassic The early Middle Preclassic (1000600 B.C.) faunal samples from the two midden contexts discussed here were recovered from the lower strata in both Operations 2012 and 2031 (Table 4.1; samples 2012eMP and 2031eMP). At this time, domestic structures were placed either directly on the ground surface, as at Operation 2012, or on low, circular stone platforms, as identified at 2031 (Anthony and Black 1994; Potter 1982; Shaw 1991). The two middens associated with this settlement appear to have been deposited around these structures and were covered and preserved below subsequent construction. The faunal data for this time period suggest the use of a wide variety of species from both terrestrial and aquatic habitats, with a dominance of wetland species (turtles and fish) if bone count is the primary measure (Table 4.1). If considered by biomass, terrestrial mammals would probably dominate the sample. The two middens are in fact quite different in terms of the proportional frequency of classes represented. If all the faunal remains that were identified at least to class are considered, a picture of diversified hunting, fishing, and collecting in the immediate habitat emerges as the dominant procurement strategy (Figure 4.2). An evaluation of the habitats exploited (based on the two midden contexts) indicates that the early Middle Preclassic population heavily exploited the wetlands that dominate the region. In the midden associated with Operation 2031, the lowbush terrestrial species were represented in slightly greater proportion than in the other midden (Figure 4.3). The generalized fish category is used because it is difficult to determine species from the small vertebrae, spines, and
Page 89 Table 4.1. Frequency of Faunal Remains in Six Primary Middens at Colha (by bone count and percent). OP2012eMP*
TAXA
Osteicthyes
OP2031eMP
NISP
%
354
41.21
1
0.12
NISP 100
OP20311MP
OP2031eLP
%
NISP
%
14.95
179
50.85
3
0.85
NISP
OP2031ILP %
321
NISP
61.26
112
OP2031TP %
NISP
%
23.68
88
12.29
1
0.14
Crotalinae Bothrops asper
Small snake
Medium lizard
Crocodylus sp.
Dermatemys mawii
3
0.45
3
0.45
11
1.64
1
0.28
Kinosternon sp.
86
10.01
57
8.52
17
4.83
?Kinosternon
16
1.86
6
0.90
1
0.15
Staurotypus triporcatus
1
0.28
2
2
0.42
0.28
22
4.20
10
1
0.19
1.40
1
0.14
3
0.57
?Staurotypus
2
0.23
1
0.19
Chelydra serpentina
6
0.70
3
0.57
Pseudemys scripta
3
0.35
8
1.20
4
1.14
3
0.57
1
0.21
11
1.54
26
3.03
13
1.94
9
2.56
1
0.19
10
2.11
15
2.09
4
0.47
1
0.15
2
0.38
1
0.21
1
0.14
?Pseudemys sp. Rhinoclemys areolata Emydidaeindeterminate
55
6.40
95
14.20
10
2.84
8
1.53
5
1.06
11
1.54
Small turtle
142
16.53
61
9.12
53
15.06
41
7.82
29
6.13
54
7.54
Large turtle
4
0.47
12
1.79
3
0.42
Reptileindeterminate
9
1.05
7
1.05
14
1.96
10
2.84
9
1.72
2
0.38
8
1.69
Casmerodius albus Anhingidae
Avessmall
1
Aveslarge
Avesindeterminate
0.12
3
1
0.14
1
0.15
2
0.28
4
0.60
0.35
3
0.45
3
0.57
1
0.19
3
0.63
2
0.28
2
0.42
2
0.28
Didelphis marsupialis ?Didelphis
1
0.15
?Tamandua tetradactyla
1
0.15
5
0.75
1
0.28
2
0.38
1
0.28
2
0.38
Dasypus novemcinctus
5
0.58
Sylvilagus sp.
1
0.15
Orthogeomys hispidus
5
0.75
(table continued on next page)
1
0.21
9
1.26
2
0.28
1
0.14
Page 90
(table continued from previous page) Table 4.1 continued Frequency of Faunal Remains in Six Primary Middens at Colha (by bone count and percent). OP2012eMP*
TAXA
NISP
OP2031eMP %
Dasyprocta punctata
Agouti paca
OP2031IMP
NISP
%
3
0.45
NISP
OP2031eLP %
NISP
OP2031ILP %
NISP
%
NISP
%
1
0.21
2
0.28
1
0.14
OP2031TP
Dasyproctidae
1
0.12
Canis familiaris
6
0.70
28
4.19
14
2.67
31
6.55
69
9.64
?Canis
5
0.58
7
1.05
3
0.57
6
1.27
11
1.54
1
0.14
Urocyon cinereoargenteus
?Urocyon
Nasua nasua
Mustela frenata
Felis pardalis
Carnivoraindeterminate
Tayassu tajacu
?Tayassu
Odocoileus virginianus
1
18
2.10
?Odocoileus
8
0.93
Mazama americana
1
0.12
Artiodactyla
0.15
0.15 6.73
5
1.42
2.54
2
0.57
1
0.28
17
1
0.21
1
0.14
1
0.14
2
0.28
2
0.28
1
1
45
0.19
1
0.19
13
2.48
37
7.82
56
7.82
0.38
4.47
5
1.06
32
2
5
1.06
4
10.21
0.56
Mammallarge
59
6.87
82
12.26
31
8.81
34
6.49
81
17.12
148
20.67
Mammalmedium/large
19
2.21
56
8.37
14
3.98
19
3.63
44
9.30
40
5.59
6
0.70
6
0.90
3
0.85
3
0.57
43
9.09
25
3.49
17
1.98
22
3.29
7
1.99
9
1.72
37
7.82
62
8.66
2
0.23
2
0.30
5
1.06
16
2.23
Mammalmedium Mammalsmall/medium Mammalsmall Mammalindeterminate Total
859
100.00
669
100.00
352
100.00
524
100.00
2
0.42
13
1.82
473
100.00
716
100.00
Page 91
Figure 4.2 Distribution of species by taxonomic class for the six Preclassic midden deposits discussed in the text (based on data in Table 4.1).
Figure 4.3 Distribution of species by habitat category for the six Preclassic midden deposits discussed in text.
ribs that make up the vast majority of the fish material. Therefore, the distinction between freshwater and marine fish could not be conclusively determined for most specimens, although it is expected that the majority are small, freshwater species. The representation of terrestrial species would increase somewhat if mammal bones indeterminate to genus were included in the frequency counts used to generate Figure 4.3 (see Table 4.1). Most of the species identified could have been found in the immediate vicinity of Colha, but several species from distant habitats were also recovered. These include Dermatemys mawii, a turtle found predominantly in large rivers, and parrotfish (Sparisoma sp.), which would have been found in marine reef habitats some distance from Colha. The parrotfish was identified in an Operation 2031 sample from a primary midden other than those considered here (see Shaw 1991). Marine fish remains were also found in early Middle Preclassic deposits at the inland sites of Cuello (Wing and Scudder 1991) and Cahal Pech (Stanchly 1995). This could reflect longdistance fishing trips to the coast or may be the beginnings of exchange of basic goods between communities in differing habitats. The rise in sea level (Folan et al. 1983) may have submerged
Page 92
many of the early coastal sites, leaving the intensity of marine resource exploitation underrepresented for the early Middle Preclassic period. The heavy use of wetland species indicated by the Colha data, tied with the recognition that agriculturalists often procure much of their meat from their garden areas, suggests that the early Maya were focusing on wetland margins for agriculture as early as the early Middle Preclassic. Evidence for wetland agriculture has been identified in Cobweb Swamp just to the north of Colha, although the dating of this activity has not been fully established (Jacob and Hallmark 1987). Pohl and others (Pohl et al. 1990) have proposed a pattern of agricultural use of the Río Hondo wetlands in which the Middle Preclassic settlers intentionally selected areas for their gardens adjacent to wetlands that were prone to seasonal flooding. The use of both wetland margins and upland milpa gardens would have allowed the early Maya to avoid some of the risks associated with dependence on the productivity of a single strategy, especially when annual variations would be expected (e.g., Miksicek 1991; Wilk 1985). The need for more laborintensive wetland agriculture techniques, such as channeling and ridged fields, would not have been necessary until the Late Preclassic, when the naturally rising water table necessitated draining to maintain productivity in the agricultural fields (Pohl et al. 1990). The faunal data from Colha for the early Middle Preclassic support a model of garden hunting in which both milpa and wetland margin agricultural areas were maintained. This strategy allowed for a reasonably high degree of diversity in the animal species used and had the potential to incorporate seasonal variation in species abundance and ease of capture into the annual procurement strategy. The appearance of marine fish in early Middle Preclassic contexts at Colha and several other interior sites in Belize hints at the beginnings of the regional exchange network that was to become so important in the Late Preclassic. Intensification during the Late Middle and Early Late Preclassic Taken together, the late Middle Preclassic (600300 B.C.) and early Late Preclassic (300100 B.C.) period faunal samples from two primary middens at Colha exhibit a trend toward a decrease in the use of mammals and an increase in the use of small fish (Table 4.1, samples 2031lLP and 2031eLP). The increased use of fish during this time may relate to a general degradation of the Colha environment due to the combined effects of a longterm exploitation of the area for meat resources and a growing population in the community of Colha. For people who were still focusing much of their faunal procurement activities in the areas immediately around Colha, and especially in the garden areas, the strategies for dealing with declining animal populations may have been to concentrate exploitation efforts on those resources that would yield a higher biomass per unit of area, such as fish and turtles. Garson (1980) has noted the seasonally high densities of smallsized fish in reduced wetland areas of river floodplains and seasonally inundated savannas during the dry season in the tropical lowlands of South America (see also Limp
Page 93
and Reidhead 1979). Garson notes that the flooding of the lowlying savannas and bajos during the rainy season provides habitats for small freshwater and brackish water fish to spawn. As these wetlands shrink during the dry season, the fish become increasingly concentrated in the remaining surface waterbodies. These pools would then be densely packed with fish as well as turtles and, with the use of massretrieval collection methods such as nets, traps, or poisons (Garson 1980; Wing and Reitz 1982), could have provided a highyield protein resource. The conditions observed by Garson in the South American lowlands parallel the situation in the wetlands of northern Belize. The occurrence of a marked dry season in Belize (Wright et al. 1959) could create similar seasonal fluctuations in wetland habitat productivity. In addition, the late Middle and early Late Preclassic periods may have been a time of higher water tables in this region, making the wetland habitat around Colha even more extensive. Miksicek (1991:84) has identified a hydrological peak in water table levels between roughly 550 and 250 B.C. at Cuello, and the sea level rise along the Belize coast (Folan et al. 1983) may also have affected the inland water table. A similar increase in the use of fish during the late Middle Preclassic has been identified by Wing and Scudder (1991) at the site of Cuello, whose residents would also have had access to extensive wetlands. Although most of the fish remains from Cuello, like those from Colha, could not be identified to genus, the majority are probably from small freshwater species. Marine reef fish remains were also identified for this time at both Cuello (Wing and Scudder 1991) and Cahal Pech (Stanchly 1995) as well as Colha (Shaw 1991). This suggests that the access, whether direct or indirect, to marine resources noted during the early Middle Preclassic continued at similar levels during the following period. The use of proportional frequencies to quantify the faunal data may be skewing this picture of faunal use somewhat, since a decrease in one faunal category causes an increase in another. However, a qualitative review of the faunal samples from the primary middens associated with these time periods supports, by the overall low sample size and the heavily fragmented condition, a dramatic decline in the availability of meat resources, particularly from mammals. The evidence for the late Middle Preclassic through the early Late Preclassic at Colha suggests a period of low faunal availability and consequently greater food stress, with the residents of Colha coping with these conditions by intensifying previous strategies of management and procurement in garden areas. A solution to this foodrelated stress was found, at least by the household under study, toward the end of the Late Preclassic through substantive changes in the socioeconomics of food procurement, which had the net effect of expanding the catchment area. Social Solutions during the Late Preclassic The Late Preclassic period at Colha was a dynamic time when the community burgeoned as the major stone tool producer in the region, with estimates of
Page 94
perhaps millions of tools being produced each year (Shafer and Hester 1986). The late Late Preclassic (100 B.C.A.D. 100) and Terminal Preclassic (A.D. 100250) are represented at Operation 2031 by two primary middens (Table 4.1, samples 20311LP and 2031TP). The household contexts under consideration were probably of an emerging highstatus family, although not particularly of the ruling elite. The faunal evidence indicates a significant shift in the organization of food procurement strategies, with a decrease in household autonomy in food production but an increase in the potential quantity and diversity of meat resources available. This is seen in the marked changes in the types of fauna used, the quantities of meat available, and the range of habitats exploited. The evidence from the preceding periods suggests a gradual degradation of the faunal population around Colha, due probably to both human and natural impacts. The solution adopted by the Colha household was a change in the organization of procurement rather than a greater intensification of strategies already in use. A look at the faunal samples from the late Late and Terminal Preclassic primary middens, as well as the site data as a whole (Shaw 1991), shows a shift to a greater use of mammals in the late Late Preclassic, with this shift becoming more marked during the Terminal Preclassic (Figure 4.2). Although whitetailed deer and domestic dog account for much of the trend, these samples also include a greater variety of mammals, such as brocket deer (Mazama americana), peccary (Tayassu tajacu), and large rodents (Dasyprocta punctata and Agouti paca) (Table 4.1). The brocket deer and paca are found most commonly in highforest habitats, which suggests that the household residents had access to resources from a greater distance than did those in early periods. There is also a higher use of the emydid turtles than in the preceding periods, coupled with a slight decrease in the use of the smaller mud turtles (kinosternids). Bird remains, which are very rare throughout the Preclassic at most sites (e.g., Carr 1986; Wing and Scudder 1991), are at least represented in a greater number in the later centuries of the Late Preclassic (Figure 4.2). Second only to deer, domestic dog was the most commonly identified mammal in the Colha assemblage. The pattern of dog remains through the Preclassic period shows a relatively low use from the early Middle Preclassic through the early Late Preclassic, with an increased use in the latter half of the Late Preclassic (Table 4.1). This is seen in the frequency of dog remains in the primary midden contexts, in which 47 percent of all dog remains came from the Terminal Preclassic deposit. When the entire site assemblage is looked at by time period, irrespective of functional context, this temporal pattern does not change (Shaw 1991). Pohl (1976) has identified a general trend in data from Petén sites, in which dog tends to be most prevalent in Preclassic and Postclassic contexts. It is unclear why the use of dog should decrease in the Late Classic if it had been viewed simply as a meat source. Sample error may be a factor, or it may relate to a decrease in the use of dog by the elite, who are represented most heavily in the Petén assemblages investigated by Pohl (1976, 1985). Another consideration is that dog, as the only significant domesticated meat resource in Mesoamerica
Page 95
until the introduction of the domesticated turkey in the Postclassic (Hamblin 1984; Shaw and Mangan 1994), may have been saved and fattened for ritual feasting (Shaw 1995). Ritual feasting as a social mechanism to legitimate and solidify power may have been used most intensively during periods of social instability and change (Clark and Blake 1994; Hayden 1995), such as would be expected during the period of social change in the Late Preclassic. Late Classic elites may have relied on institutional forms of control, and feasting could have been more social (i.e., between elite families from various sites) than competitive; deer and small, fatrich rodents seem to be the most common meat consumed. The emphasis on garden hunting strategies for meat procurement proposed for the early occupations at the Colha households changes notably in the late Late Preclassic. At this time a broadening of the resource base away from a focus on wetland garden areas has occurred (Figure 4.3) and is replaced by a situation in which a variety of procurement strategies might have been used concurrently. The regional evidence (e.g., Pohl et al. 1990) indicates a continued use of wetland agriculture in conjunction with upland milpas in the Late Preclassic, although the use of wetlands might have required more intensive modification such as the maintenance of raised fields. It would be expected that garden hunting would have continued as the primary strategy for agricultural households, but the household represented in Operation 2031 under study here appears to have shifted its strategy in the Late Preclassic, becoming more involved in the administrative and/or economic aspects of the emerging lithic production community at Colha. The household at Operation 2012 is remodeled at around this time for use in ceremonial activities, and there are no middens associated with this activity (Potter 1994). It is proposed that indirect strategies of meat procurement were becoming available and economically accessible to households such as the one represented at Operation 2031, including the participation in the emerging regional exchange networks. The market envisioned for the Lowland Preclassic Maya communities is one that is generally decentralized (following closely Fry 1980 and Feinman et al. 1984) and would have functioned primarily to disseminate utilitarian goods at the regional level. The extensive exchange of utilitarian goods during the Late Preclassic period is supported strongly by the site of Colha as a mass producer of stone tools. It is also indicated in northern Belize by areas of extensive agricultural production (McAnany 1989; Turner and Harrison 1983) and the rapid growth of port communities, such as Cerros at the mouth of the New River (Carr 1986; Freidel 1986). In addition to what is recognizable archaeologically, more perishable resources such as cotton cloth or salt may also have moved through these networks (e.g., Graham 1987; Voorhies 1982). The movement of meat resources within this regional market is suggested here as one strategy that the late Late and Terminal Preclassic Colha residents used to offset the reduced access to meat resources caused by the logistical difficulty faced by nonagriculturalist households in procuring faunal resources. The problem of meat procurement by nonagriculturalists would have been compounded by decreasing faunal productivity within the Colha
Page 96
environment owing to heavy human predation. Meat could have entered the market exchange system through many avenues, including locally generated surpluses (e.g., community specialists in meat procurement, agricultural household collection of garden fauna, etc.), community specialization in procurement and processing of selected resources (e.g., marine or highforest), or an expedient collection of meat resources for market exchange by merchants transporting other goods. The raising and fattening of domestic dogs at the household level could also have been used to generate a surplus, a strategy even lowstatus households could use to participate in the exchange network. In this proposal for the market exchange of meat, it should be noted that it is not being suggested here that food stress linked to meat shortages was a causal factor in the development of the extensive Late Preclassic exchange networks. It is more likely that the exchange of meat would have been recognized as a viable strategy in light of the developing exchange of utilitarian goods. What is most significant here is that the household under study, and likely many others that were moving away from an agricultural livelihood, used a social solution to cope with decreasing faunal availability, as opposed to simply intensifying an existing strategy. The use of a variety of direct and indirect means of acquiring meat, including the building of multiple social and economic relationships that crosscut lineage lines and community boundaries, probably set in place the mechanisms that allowed Classic period communities to supply meat to the expanding nonagriculturalist segment of society. Conclusions My approach in the analysis of the Preclassic faunal assemblage from Colha has been to weigh heavily the social and functional contexts in the interpretation of faunal use through time. One of the important shifts in approach has been to recognize that models of faunal procurement and use which have emerged from studies of hunters and gatherers tend to obscure and homogenize results if they are applied uncritically to complex societies. The development of models for the interpretation of faunal use within complex societies has largely been slanted to Old World societies that developed an increasing dependence on large domesticated species. The rise of a civilization in the tropical jungles of lowland Central America, which used predominantly wild animal species, must be considered from a new and somewhat different perspective. An important first step in the development of this new perspective is to recognize the importance of sample context in the interpretations of faunal data. The analyses of the relatively few Maya faunal assemblages to date have suffered from small sample size, a focus of site excavations on elite areas and monumental architecture, and poor data recovery methods. A few thousand bones from a site that is over 6 km2, such as is projected for Colha, will not tell us ''what the Maya ate" at any given time period. All too often, faunal samples from a site excavation are lumped by time period, and reconstructions of faunal use are then presented as "the pattern for that time period." What we are actually looking at is much more modest. The financial and time constraints that affect any archaeological excavation of a Maya site will be such that relatively small and dis
Page 97
crete areas are excavated, and the research questions that direct what areas will be investigated rarely consider faunal data retrieval as a primary concern. Although we are in no position at any site to say what the communitywide patterns of consumption were for a given time period, what we are able to say is very interesting. If contexts of faunal samples are used to define and frame the interpretations, the Colha data reflect the use of animals at the household level over the course of about 1,250 years. As the residents of this household changed with each generation, the people were faced with the task of procuring meat for consumption and ritual use against the dynamics of a changing social and economic world. The analysis of the Colha data with reference to both context and social and economic parameters has highlighted several changes in faunal procurement strategies at the two households represented in the assemblage, changes that are proposed as articulating with the emergence of social and economic inequality. The early settlers of Colha probably used a modified form of garden hunting (e.g., Linares 1976) that included both milpa and wetland margin agricultural areas. On the basis of what is known about other early settlements in the region, such as Cuello (Wing and Scudder 1991) and Albion Island (Pohl et al. 1990), we may conclude that most households of this period were probably largely selfsufficient in their acquisition of meat resources. The growth of the Colha population and the longterm exploitation of the immediate Colha environment probably resulted in a significantly degraded biomass in the area of the site by the late Middle Preclassic. During this time, and continuing into the early Late Preclassic period, the household represented at Operation 2031 appears to have intensified the garden hunting strategy with a particular focus on the wetland environment. Terrestrial resources, especially large mammals, were significantly underrepresented in comparison with earlier and later centuries. There are indications that this was a relatively cool and moist period in Belize (Folan et al. 1983; Miksicek 1991; Pohl et al. 1990), and wetland areas and aquatic habitats may have increased significantly. A shift in faunal procurement strategies has been identified in the Colha data beginning in the late Late Preclassic, a trend that continues to intensify until the end of the Preclassic period. The data from the Colha household indicate a much greater emphasis on the use of mammals for food, with a broadening of the number and habitat ranges of the species used. This household may have been less involved in agricultural activity and, therefore, unable to procure meat easily from garden areas. The shift toward an increased reliance on mammals may mean that this household had the ability, through social and economic means, to remove itself from the relative subsistence autonomy of earlier periods and move to a strategy of indirect procurement through trade or tribute. The greater use of more distant habitat resources, noticeable at both Colha and Cuello (Wing and Scudder 1991), suggests that meat was one of the types of basicneed resources moving through the regional exchange networks identified for the Late Preclassic. It has been an explicit tenet of this study that a look at faunal use at the Preclassic households at Colha can provide only a part of what will be needed in a much larger effort to identify the strategies the Maya used to acquire meat
Page 98
resources against the backdrop of dramatic changes in the social and economic spheres. The explicit recognition of contextual parameters within the Colha assemblage has improved the insights that can be made concerning the impacts of emerging social and economic inequality on daily life. What this chapter illustrates is that there are many empirical ways to approach social behaviors, including the analysis of faunal remains. The Maya Lowlands provide an excellent location for the consideration of such social questions, and I expect the next decade to hold great promise in the use of faunal data to explore these issues. Acknowledgments The Colha faunal data and contextual information were generously provided by Thomas Hester, Harry Shafer, Daniel Potter, Fred Valdez, Jr., Dana Anthony, and Stephen Black. The Belize Department of Archaeology, Belmopan, is gratefully acknowledged for the loan of the faunal materials from Colha for analysis. The primary data for this chapter are drawn from my doctoral research at the University of Massachusetts, and I gratefully acknowledge the support of my committee: H. Martin Wobst, George Armelagos, and David Klingener. This research has benefited from comments offered by the people mentioned above as well as H. Sorayya Carr, John Cross, Tonya Largy, Christine White, and four anonymous reviewers. I am also grateful to José Rosado and Maria Rutzmoser for providing access to the reference collection at the Museum of Comparative Zoology at Harvard University. Early drafts of this chapter were completed while I was a fellow at the Mary Ingraham Bunting Institute at Radcliffe College, and I offer my appreciation for that opportunity to focus on my research in such a supportive environment. References Cited Anthony, D., and Black, S. (1994) Excavations at Operation 2031. In T. R. Hester, H. J. Shafer, and J. D). Eaton (eds.): Continuing Archeology at Colha, Belize. Austin: Texas Archeological Research Laboratory, University of Texas at Austin, Studies in Archeology No. 16, pp. 3958. Carr, H. S. (1985) Subsistence and ceremony: Faunal utilization in a Late Preclassic community at Cerros, Belize. In M. Pohl (ed.): Prehistoric Lowland Maya Environment and Subsistence Economy. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 77. Cambridge: Harvard University, pp. 115132. Carr, H. S. (1986) Faunal Utilization in a Late Preclassic Community at Cerros, Belize. Unpublished Ph.D. dissertation, Tulane University, New Orleans. Clark, J. E., and Blake, M. (1994) The power of prestige: Competitive generosity and the emergence of rank societies in Lowland Mesoamerica. In E. M. Brumfiel and J. W. Fox (eds.): Factional Competition and Political Development in the New World. Cambridge: Cambridge University Press, pp. 1730. Crabtree, P. J. (1990) Zooarchaeology and complex societies: Some uses of faunal analysis for the study of trade, social status, and ethnicity. In M. B. Schiffer (ed.): Archaeological Method and Theory. Tucson: University of Arizona Press, pp. 155205. Feinman, G.; Blanton, R.; and Kowalewski, S. (1984) Market system development in the Prehispanic valley of Oaxaca. In K. G. Hirth (ed.): Trade and Exchange in Early Mesoamerica. Albuquerque: University of New Mexico Press, pp. 157178.
Page 99
Folan, W. J.; Gunn, J.; Eaton, J. D.; and Patch, R. W. (1983) Paleoclimatological patterning in southern Mesoamerica. Journal of Field Archaeology 10:453468. Freidel, D. A. (1986) The monumental architecture. In R. A. Robertson and D. A. Freidel (eds.): Archaeology at Cerros Belize, Central America. Dallas: Southern Methodist University Press. Fry, R. E. (1980) Models of exchange for major shape classes of Lowland Maya pottery. In R. E. Fry (ed.): Models and Methods of Regional Exchange. Society for American Archaeology Papers, No. 1. Washington, D.C.: Society for American Archaeology, pp. 318. Garson, A. G. (1980) Comment upon the economic potential of fish utilization in riverine environments and potential archaeological biases. American Antiquity 45:562567. Graham, E. (1987) Resource diversity in Belize and its implications for models of Lowland trade. American Antiquity 52:753767. Grayson, D. K. (1984) Quantitative Zooarchaeology. Orlando: Academic Press. Hamblin, N. L. (1984) Animal Use by the Cozumel Maya. Tucson: University of Arizona Press. Hayden, B. (1995) Pathways to power: Principles for creating socioeconomic inequalities. In T. D. Price and G. M. Feinman (eds.): Foundations of Social Inequality. New York: Plenum Press. Hester, T. R. (1985) The Maya lithic sequence in northern Belize. In M. G. Plew, J. C. Woods, and M. G. Pavesic (eds.): Stone Tool Analysis: Essays in Honor of Don E. Crabtree. Albuquerque: University of New Mexico Press, pp. 187210. Hester, T. R., and Shafer, H. J. (1984) Exploitation of chert resources by the ancient Maya of northern Belize, Central America. World Archaeology 16:157173. Jacob, J. S., and Hallmark, C. T. (1987) Characterization of raisedfield and upland soils in the vicinity of Colha. Ms. on file, Department of Anthropology, University of Texas at Austin. Jones, J. G. ( 1991) Pollen Evidence of Prehistoric Forest Modification and Maya Cultivation in Belize. Ph.D. dissertation, Texas A&M University, College Station. Limp, W. F., and Reidhead, V. A. (1979) An economic evaluation of the potential of fish utilization in riverine environments. American Antiquity 44:7078. Linares, O. (1976) "Garden hunting" in the American tropics. Human Ecology 4:331349. McAnany, P. A. (1989) Stonetool production and exchange in the eastern Maya Lowlands: The consumer perspective from Pulltrouser Swamp, Belize. American Antiquity 54:332346. Miksicek, C. H. (1991) The ecology and economy of Cuello. In N. Hammond (ed.): Cuello: An Early Maya Community in Belize. Cambridge: Cambridge University Press, pp. 7084. Pohl, M. D (1976) Ethnozoology of the Maya: An Analysis of Fauna from Five Sites in Petén, Guatemala. Ph.D. dissertation, Harvard University, Cambridge. Pohl, M. D. (1985) The privileges of Maya elites: Prehistoric vertebrate fauna from Seibal. In M. Pohl (ed.): Prehistoric Lowland Maya Environment and Subsistence Economy. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 77. Cambridge: Harvard University, pp. 133145. Pohl, M. D. (1990) Summary and proposals for future research. In M. D. Pohl (ed.): Ancient Wetland Agriculture: Excavations on Albion Island, Northern Belize. Boulder, Colo.: Westview Press, pp. 397439. Pohl, M. D.; Bloom, P. R.; and Pope, K. O. (1990) Interpretations of wetland farming in northern Belize: Excavations at San Antonio Rio Hondo. In M. D. Pohl (ed.): Ancient Maya Wetland Agriculture: Excavations on Albion Island, Northern Belize. Boulder, Colo.: Westview Press, pp. 187254. Potter, D. R. (1982) Some results of the second year of excavation at Operation 2012. In T. R. Hester, H. J. Shafer, and J. D. Eaton (eds.): Archaeology at Colha, Belize: The 1981 Interim Report. Center for Archaeological Research, University of Texas at San Antonio; Centro Studi e Ricerche Ligabue, Venezia, pp. 98122.
Page 100
Potter, D. R. (1994) Strat 55, Operation 2012, and comments on Lowland Maya blood ritual. In T. R. Hester, H. J. Shafer, and J. D. Eaton (eds.): Continuing Archeology at Colha, Belize. Studies in Archeology, No. 16. Austin: Texas Archeological Research Laboratory, University of Texas, pp. 3137. Redford, K. H., and Robinson, J. G. (1987) The game of choice: Patterns of Indian and colonial hunting in the Neotropics. American Anthropologist 69:120. Shafer, H. J., and Hester, T. R. (1983) Ancient Maya chert workshops in northern Belize, Central America. American Antiquity 48:519543. Shafer, H. J., and Hester, T. R. (1986) Maya stonetool craft specialization and production at Colha, Belize: Reply to Mallory. American Antiquity 51:158166. Shafer, H. J., and Hester, T. R. (1990) The Puleston Axe: A Late Preclassic Maya hafted tool from northern Belize in northern Belize: Excavations at San Antonio Rio Hondo. In M. D. Pohl (ed.): Ancient Maya Wetland Agriculture: Excavations on Albion Island, Northern Belize. Boulder, Colo.: Westview Press, pp. 279 294. Shaw, L. C. (1991) The Articulation of Social Inequality and Faunal Resource Use in the Preclassic Community of Colha, northern Belize. Ph.D. dissertation, University of Massachusetts, Amherst. Shaw, L. C. (1995) The importance of dog in ritual feasting in the Maya Preclassic. Paper presented at the 60th Annual Meeting of the Society for American Archaeology, Minneapolis, May. Shaw, L. C. (in press) Animal resources and emerging inequality: The Preclassic faunal record at Colha, Belize. Austin: Texas Archeological Research Laboratory, University of Texas at Austin, Studies in Archeology. Shaw, L. C., and Mangan, P. H. (1994) Faunal analysis of an Early Postclassic midden, Operation 2032, Colha, Belize. In T. R. Hester, H. J. Shafer, and J. D. Eaton (eds.): Continuing Archeology at Colha, Belize. Studies in Archeology, No. 16. Austin: Texas Archeological Research Laboratory University of Texas, pp. 6978. Stanchly, N. (1995) Formative period Maya faunal utilization at Cahal Pech, Belize: Preliminary analysis of the animal remains from the 1994 field season. In P. F. Healy and J. J. Awe (eds.): Belize Valley Preclassic Maya Project: Report on the 1994 Field Season. Occasional Papers in Anthropology, No. 10. Peterborough: Department of Anthropology, Trent University. Styles, B. W. (1981) Faunal Exploitation and Resource Selection: Early Late Woodland Subsistence in the Lower Illinois Valley. Northwestern University Archaeological Program Scientific Papers, No. 3. Evanston: Northwestern University. Turner, B. L., and Harrison, P. D. (1983) Pulltrouser Swamp and Maya raised fields: A summation. In B. L. Turner and P. D. Harrison (eds.): Pulltrouser Swamp: Ancient Maya Habitat, Agriculture, and Settlement in Northern Belize. Austin: University of Texas Press, pp. 246270. Voorhies, B. (1982) An ecological model of the early Maya in the central Lowlands. In K. V. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press, pp. 6595. Wilk, R. R. (1985) Dry season agriculture among the Kekchi Maya and its implications for prehistory. In M. D. Pohl (ed.): Prehistoric Lowland Maya Environments and Subsistence Economy. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 77. Cambridge: Harvard University. Wing, E. S., and Reitz, E. J. (1982) Prehistoric fishing economies of the Caribbean. Journal of New World Archaeology 5:1332. Wing, E. S., and Scudder, S. J. (1991) The exploitation of animals. In N. Hammond (ed.): Cuello: An Early Maya Community in Belize. Cambridge: Cambridge University Press, pp. 8497. Wright, A. C. S.; Romney, D. H.; Arbuckle, R. H.; and Vail, V. E. (1959) Land in British Honduras: Report of the British Honduras Land Use Survey Team. Colonial Research Publication No. 24. London: Her Majesty's Stationery Office.
Page 101
PART II PALEOPATHOLOGY
Page 103
Chapter 5 Coming Up Short Stature and Nutrition among the Ancient Maya of the Southern Lowlands Marie Elaine Danforth Although the estimation of stature is a routine part of the analysis of human remains, it has received particular attention among the prehistoric Maya because of its possible association with the Late Classic "collapse" of the tenth century A.D. Many ecologically based models for the collapse have cited a decline in height from Preclassic times as a strong indication for a health crisis (Lowe 1985; Santley et al. 1986; Willey and Shimkin 1973). Since then, the decrease in Maya stature has become a minor part of the folklore of Mesoamerican archaeology and in other contexts is often cited as an example of the effects of nutritional inadequacy and the potential enormity of these effects. For example, one recent medical anthropology text reported that "the skeletal remains of Mayans who lived during this [Classic] period show that people of the common class were shorter on average with each successive generation (Haviland 1967)" (McElroy and Townsend 1989:38). The evidence for such a decline, however, is compromised by small sample sizes and limited opportunity for intertemporal comparison. Furthermore, interpretation of stature data is complicated by the complex interaction of a number of factors—from higher mortality of short individuals to catchup growth—that can affect a population's mean values. This chapter presents a review of stature patterns in Maya of the southern Lowlands using all the published data available. It addresses modern studies concerning genetic and environmental factors affecting growth and then attempts to apply these concepts to an analysis of stature change in Precolumbian populations of the region. Factors Affecting Adult Stature The modern Maya are among the shortest populations in the world, with studies consistently reporting mean stature value of about 155 cm for males and 143 cm for females (e.g., Starr 1902; Steggerda 1932:1112; Russell 1976), although one study does report a mean stature value of 167 cm for males in Guatemala (Méndez and Behrhorst 1963). Genetics plays a role in determination of adult stature in all populations, but its particular contribution is uncertain. It has been argued that the impact of environmental factors increases variation in growth patterns in less advantageous populations compared with their more
Page 104
economically developed counterparts because of the greater nutritional and disease challenges typically endured (Mueller 1976; Bogin 1991). This may be one factor causing the relatively low parentchild height correlations, ranging between 0.22 and 0.32, seen among the rural Maya of Guatemala (Russell 1976). Recently, however, researchers (e.g., Frisancho 1993) have begun to suggest that genetics plays only a minor role in adaptation to stressors. In fact, Martorell and Habicht (1986) note similar growth patterns, and presumably similar adult statures, among wellnourished individuals from a variety of populations worldwide, including samples from Europe, Africa, India, and the Middle East. Such observations, combined with his own work, have led Bogin (1995) to argue that given adequate access to resources, mean Maya stature could attain similar levels. All these findings strongly indicate that the most productive focus for analysis concerning explanations of stature change over time in genetically continuous populations involves environmental factors. Most discussions concerning environmental causes of short stature ultimately concentrate on nutrition, particularly adequacy of protein levels. Traditionally, evaluations of the maizebased diet of the Maya have stressed the relatively low levels of protein, a conclusion strengthened by the suggestion that up to 40 percent of early childhood mortality in Guatemala during the 1950s was related to kwashiorkor, a nutritional disease caused by protein deficiency (Scrimshaw and Behar 1961). Some researchers, however, have begun to question whether too often a protein deficiency is assumed when, in fact, the cause of the malnutrition is simply too few calories. Waterlow and Payne (1975) have concluded that the modern Maya diet meets over 95 percent of protein requirements for young children, although this includes certain food sources, especially nonhuman milk, that would not have been available to precontact populations. It also may be misleading to focus on protein adequacy alone, since nutritional deficiencies almost never occur in isolation. Essentially, deficiencies of any essential nutrients potentially can stunt growth as resources are diverted to maintain basic physiological functions of the body (Bogin 1988). Furthermore, the consequent health conditions associated with malnutrition, regardless of the particular deficiency, are similar. For example, antibody production is inhibited, making various infections more likely to occur. Energy levels are quite low. Even when food becomes available, the patient may exhibit anorexia. Other environmental factors, such as parasitic infection associated with poor sanitation, are also frequently in operation. Thus, a number of causes typically act synergistically to produce stunting of growth, and it can be almost impossible to tease them apart. Some have argued that reduction in stature is actually adaptive, since little bodies require fewer resources and thus may be better suited to more limited environments (e.g., Stini 1971; Seckler 1982). Indeed, this argument has been used as a possible explanation for the reduced height of the rural Maya today (Márquez Morfín 1984; Saul and Saul 1989). As Saul and Saul (1989:300) note: "It was the shorter Maya, for the most part, whose characteristics were passed on. The more recent (and modern) Maya represent survivors—their small size being the result of a process of successful adaptation through microevolution."
Page 105
The "small but healthy" hypothesis has been challenged on several points, however (Pelto and Pelto 1988; Bogin 1995). Studies of living populations have consistently shown that individuals experiencing stunted growth also suffer from higher levels of morbidity and mortality since they do not have sufficient reserves to provide a safety net during times of environmental insult. The shorter individuals also show deficits in cognitive development as well as reduced energy levels. Ultimately, all these factors compromise reproductive capacity, making it unlikely that individuals suffering from stunted growth would be passing their genes on in greater numbers than would taller members of the population. The degree of stature depression affecting the modern Maya is easily observed when their mean height values are compared to those of populations experiencing greater levels of access to resources. The more wealthy ladino groups in Guatemala average 7.71 cm taller for males and 11.14 cm taller for females (Bogin et al. 1992). In another study Bogin (1995) observed an average of 5.5 cm increase in stature between fullblooded Maya living in the United States as compared to those in rural Guatemala. More favorable environments also might have an impact on body proportions. Tanner and colleagues (1982) found that most of the secular trend for greater stature recently seen among the Japanese is related to an increase in leg length with virtually no increase seen in trunk length. This finding may have particular relevancy to the Maya, and perhaps Mesoamericans in general, since it has been observed that they have proportionately longer arms and shorter legs compared with many other populations in the world (Steggerda 1932; Genovés 1967). Examination of stature patterns between males and females provides yet another means of evaluating the effects of environmental stress on height. Females have traditionally been thought to enjoy a genetic buffering against environmental insult (Stini 1969; Stinson 1994). Thus, during times of depressed nutritional resources their stature is less affected than that of males. Although such a pattern has frequently been observed in living populations (e.g., Dettwyler 1992), the picture presented is actually quite complex. For example, Bogin and colleagues (1992) found that in rural Guatemala sexual maturation was less delayed in girls than in boys when compared to their more affluent ladino counterparts, but that ultimately the Maya girls ended up relatively shorter, possibly because they progressed through puberty more quickly. Russell (1976) also failed to find the expected pattern of sexual dimorphism in her analysis of stature patterns among the Maya, suggesting that early childbirth before growth is completed might be an additional factor stunting female stature. Other factors that can offset the predicted decrease in sexual dimorphism in compromised environments include higher mortality among short males and male children being given preferential treatment in medical care and access to nutritional resources. Thus, patterns of sexual dimorphism must be interpreted carefully. In summary, consideration of stature patterns among the modern Maya may offer guidelines for evaluation of patterns among their Precolumbian ancestors. First, the role of the environment appears to be extremely important, if not predominant, in determining adult stature, and thus we cannot assume
Page 106
that small stature is the result of natural selection. Second, it suggests that stature levels in a population can rebound quickly if adequate resources become available and that much of the stature increase might be related to longer legs rather than longer trunks. Finally, examination of differences in stature patterns between the sexes may potentially be informative in the assessment of the severity of environmental insults. Previous Analysis of Prehistoric Maya Stature The earliest suggestion of a decline in mean stature of Maya populations from the prehistoric to the modern periods was made by Stewart (1949) in his analysis of skeletal remains from several sites in highland Guatemala. He considered various possible explanations for the decrease, including replacement by migration, differential access to resources due to social status, and climatic influence. Stewart concluded, however, that the most likely cause was poor nutrition, possibly related to a deficiency in iodine. Two other studies made more indirect speculations concerning a stature decline. At Copán, Longyear (1952) noted that three Early Classic individuals were bigger than virtually all their Late Classic counterparts at San José. He suggested that migration might be at work, noting that one particular burial might "represent a blend of massive, robust, indigenous ingredients and more fragile and gracile Classic Maya traits" (Longyear 1952:87). In contrast, data from long bones at Barton Ramie led Willey (1965:538) to state that "increasingly meager nutrition or the result of some other environmental force" might be responsible for increasing gracility and decreasing ruggedness in the Maya from Preclassic to Postclassic periods. Certain methodological problems, however, are present in both investigations. The Copán study (Longyear 1952) obviously suffers from small sample size (N = 3); in addition, two of the stature estimates were based on in situ burial measurements. The sample size at Barton Ramie (Willey 1965) appears to be larger than at Copán, but the published mean long bone dimensions were not separated by sex. Interestingly, despite the fact that a decrease in height over time has been reported as occurring in the sample (e.g., Santley et al. 1986:142), the entire Barton Ramie series was never evaluated for stature patterns since the analysts felt the number of individuals available was too low (Willey 1965:536). The skeletal sample from the site, however, was reevaluated a few years ago. Although the trend for decreasing size and increasing roundness of long bone dimensions no longer is nearly as strong as in the earlier data, a similar trend still may be seen for a few measurements, particularly in the subtrochanteric region of the femur (Danforth et al. 1991). Studies of skeletal series excavated at Tikal and Altar de Sacrificios brought the suggested stature decline and its implications to the forefront of discussion. At Tikal, Haviland (1967) found a statistically significant decrease of nearly 10 cm between Early Classic and Late Classic males. Haviland also observed that the decrease was much larger in nontomb burials, which he concluded might
Page 107
reflect differential access to nutritional resources, especially protein. Interestingly, female stature at Tikal appeared to have been more consistent over time. A few years later at Altar de Sacrificios, Saul (1972) not only observed a ''striking" decrease in stature among males but also noted that it appeared to correlate with high rates of anemia, scurvy, and childhood growth disruption. Saul's investigation remains one of the most comprehensive skeletal studies of Prehispanic Maya skeletal series, but the study was hampered by poor preservation. The maximum sample size for stature reconstruction for males for any time period was four (Saul 1972:93). Based on these studies, however, words such as "malnutrition" and "disease burden" became a part of many models for the fall of the Late Classic Maya (e.g.,Willey and Shimkin 1973). Two other major considerations of stature change over time have also emerged. Using data from all over Mesoamerica as well as the Maya region, Nickens (1976) argued that the health consequences associated with agriculture, such as a less varied diet and greater sedentism, explain the reduction in stature over time. In the only research project to systematically study stature patterns in fifteen prehistoric populations from the northern Lowlands, Márquez Morfín (1984) similarly found a decrease of several centimeters in mean stature from the Preclassic to the Classic period, especially among males. In contrast, there appeared to be little change between Classic and Postclassic populations. Again, however, her study was hampered by small sample size; for example, a total of five males provided data for the Preclassic. In summary, previous studies concerning Maya stature show several similar findings. First, the greatest decline in height appears to occur between the Preclassic and Classic periods; the transition from the Classic to the Postclassic enjoyed relatively greater stability. Second, decreases in male stature are nearly always markedly greater than those seen in females. Third, environmental factors are usually cited as the cause of the stature reduction. Finally, all the studies involve small sample sizes with only one (Haviland 1967) addressing statistical testing of the data. Current Analysis of Prehistoric Maya Stature Few would argue with the statement that little skeletal material exists for the prehistoric Maya. Table 5.1 lists all the stature estimates produced from an exhaustive review of the published literature in the human osteology of the southern Lowlands (Danforth et al. 1997) and represents more than 125 southern Lowland sites with reported skeletal remains. Estimates calculated by the original (cited) author were used whenever possible, but stature was calculated specifically for this study when only raw measurements were reported. Stature estimates based on leg bones were preferred over those based on arm bones because of their generally greater reliability (Trotter 1970). For a few sites, if formulae other than Genovés 1967 were used, the estimates were converted to Genovés when it was expressly stated that maximum measurements from a single bone (e.g., "femur" as opposed to "leg bones") were used. Furthermore,
Page 108
since such a conversion potentially creates another level of miscalculation, the estimates as originally given are also reported and used whenever appropriate, as in intrasite comparison by time period. The small sample sizes seen in Table 5.1 attest to one of the problems that has always plagued those studying Maya human remains, namely that tropical soils unfortunately are not conducive to bone preservation. Only about 15 percent of the sites with osteological information yielded reliable stature estimates, and relatively few of these offer any sort of opportunity for intertemporal comparison. Also discouraging is that the modal sample size is one individual. Even if stature estimates based on arm bones were included, there would be little improvement. The situation is even poorer for females, since they are outnumbered by males nearly four to one in the sample; it is unclear why this occurs, although possible reasons include the fact that it is less likely that gracile bones will be preserved and more likely that excavated tombs will contain males (Welsh 1988). For all these reasons, the data base for prehistoric stature patterns among the Maya is a marginal one. When all available data are considered, regardless of time period, the mean male stature is 160.1 cm with a range of 146 to 176 cm (N = 293). The female mean is 147.8 cm with a range of 135 to 157 cm (N = 77). When these results are compared with those previously discussed for the modern Maya, a 56cm decrease in mean stature is seen for both males and females. The reduction is especially evident when it is noted that Steggerda (1932:11) reports that among rural Guatemalans only 3 of 100 males measured were over 165 cm in height and only 1 of 25 females was taller than 155 cm. Thus, Stewart's (1949) observation that the ancient Maya were bigger than their modern descendants is supported by this preliminary analysis. Of course, temporal comparison of stature is crucial to our ability to evaluate suggestions of decreasing nutritional resources and/or increasing disease load during the collapse. The data roughly support a dip in male stature during the Late Classic with possibly a modest recovery by the early Historic period (Figure 5.1). Females were similarly tallest during the Preclassic, but their stature appears to have been relatively stable from the Late Classic on. When those sites with samples from more than one time period are analyzed for intertemporal change in height, the pattern expressed is less consistent (Figure 5.2). Stature reconstructions at the majority of the six sites show decline, but these differences are statistically significant only for the Tikal males using MannWhitney nonparametric testing. The general pattern of stature change seen in the evaluation of raw data shows the expected variation by sex in that male stature appears to fluctuate slightly more than female stature. When levels of sexual dimorphism at individual sites are considered, only two, Altar de Sacrificios and Tikal, show the predicted increase from the Preclassic to the Late Classic. In contrast, two other sites, Seibal and Barton Ramie, show a decrease. The change in sexual dimorphism is achieved in a very curious manner in three of the four cases, however. Generally, it would be expected that the mean stature of both sexes would change in the same direction, although the magnitude of change would be
Page 109
Figure 5.1. Mean stature estimates by time period.
Figure 5.2. Change in mean stature estimates by individual site: (a) Preclassic to Late Classic; (b) Early Classic to Late Classic. Note: Sample sizes given in Table 5.1.
Page 110 Table 5.1. Stature estimates for Southern Lowland Maya Sites. Site
N
Mean
Range
MALES
Bones Used
Formula Used
Preclassic All sites
150.9175
4 166.6
163.5173.0
Leg
G
2 154.3
153.8154.8
Fe
G
6 163.17
158174
Lb
P
151.50167.13
Fe
G
34 161.83
Altar deSacrificios
1
Barton Ramie2 3
Chiapa de Corzo 4
11 161.47
5
1 154.9
Fe
G
2
Seibal
2 153.1
150.9155.4
Fe
G
Tikal7
6 164.3
161169
Fe
TGW
158.6166.1
Fe
G
Ti
G
Cuello Colha
= 6 163.1
Uaxactún
8
1 164.9
Early Classic All sites Altar de Sacrificios Barton Ramie2 3
Chiapa de Corzo Copán
9
Rio Azul
155.9176
2 159.4
158.8160.0
Fe
G
3 154.0
146.4161.1
Fe
G
1 164
P
1 162.89
Fe
P
= 1 161.96
Fe
G
1 172
Fe
G
158176
Fe
TGM
155.9172.5
Fe
G
Ti
G
23 162.14 1
11
Tikal7
15 167
= 15 164.2
Uaxactún
12
1 159.36
Late Classic All sites Altar de Sacrificios Baking Pot13 Barton Ramie
2
Chiapa de Corzo 2 l4
Lamanai
Palenque15 Palenque
16
Piedras Negras 18
San Agustin San José
19
Seibal2 Tikal
Ti
G
1 162.5
Hu
G
153.6162.6
Fe
G
159170
Lb
P
10 158.0
Fe
G
1 169.4
Fe
G
1 157.5
Lb
G
3 158.83
s = 3.88
Fe
G
1 160.0
Ul
G
3 158.9
157.0162.6
Fe
G
1 156.6
Fi
G
18 159.1
148.6172.4
Fe
G
21 157.4
144176
Fe
TGM
143.0172.6
Fe
G
10 156.6 3
Copán
143.0176
1 162.0
82 158.3
7
1
17
5 163.8
= 21 155.3 20
4 158.2
155162
Lb
P
Uaxactún
21
1 164.9
Ti, Fe
G
Uaxactún
12
1 168.5
Fi
G
6 159.38
153.44167.39
1 159.1
Fe
P
= 1 157.4
Fe
G
153.44167.39
Fe, Ti
G
Toniná
Late Classic/Early Postclassic All sites Copán
10
22
Lagatero
5 159.78
(table continued on next page)
Page 111
(table continued from previous page) Site
N Mean
Bones Used
Formula Used
All sites
5 163.48
159.6166
Baking Pot13
1 162.4
Fe
G
1 166
Leg
P
FEMALES
Postclassic
Toniná
20
Preclassic
All sites Altar de Sacrificios Barton Ramie
1
2
143.8163.25
2 148.3
147.5149.0
Fe
G
2 148.9
143.8153.9
Fe
G
1 149
Lb
P
9 153.67
146.50163.25
Fe
G
2
2 148.5
145.1151.9
Fe
G
4 147.0
145148
Fe
TGW
144.2147.4
Fe
G
Cuello Seibal 6
= 4 146.3
Early Classic All sites Altar de Sacrificios Tikal
20 150.43
4
Chiapa de Corzo3
Tikal
Range
1
6
6 142.2
136.9149.2
1 149.5
Fe
G
4 141.1
138145
Fe
TGW
136.9144.2
Fe
G
Fe
G
140.0157
1 147.5
Fe
G
6 145.7
141.3150.8
Fe
G
16 147.2
Fe
G
1 148.0
Fe
G
1 149.96
Leg
P
= 4 141.1
1 139.3
Uaxactún 7
Late Classic
Altar de Sacrificios Barton Ramie Copán
43 147.4
All sites 1
2
8
Palenque
15
Piedras Negras16
= 1 147.7
Fe
G
3 148.4
146.7150.3
bFe
G
Seibal
4 145.7
143.1150.8
Fe
G
6
11 149.3
142157
Fe
TGW
140.0156.8
Fe
G
17
San Agustin 2
Tikal
= 11 148.7
Classic/Postclassic
All sites Altar de Sacrificios Copán
1
9
Lagatero
Postclassic All sites
Sarteneja22 Tikal
6
140.2153.60
1 146.5
Fe
G
1 142
Fe
P
= 1 140.2
Fe
G
145.39153.60
Fe, Ti
G
3 147.17
146.5149
1 146.5
Fe
G
1 146.6
Fe
G
1 149
Fe
TGW
= 1 148.4
Fe
G
3 149.81
21
Altar de Sacrificios
5 147.23
1
Key to table 5.1 1 Saul 1972 2 Cohen et al. 1989 3 Jaen Esquivel 1966 4 Saul and Saul 1991 5 Young 1994 6 Haviland 1967 7 Ricketson and Ricketson 1937 8 Longyear 1940 9 Longyear 1952 10 Steele 1986 11 Smith 1937 12 Bullard and Bullard 1965 13 Helmuth and Pendergast 198687 14 Serrano Sánchez 1973 15 Marquez Morfin 1984 16 Coe 1959 17 Kidder 1937 18 Thompson 1939 19 Romano Pacheco 1979 20 Wauchope 1934 21 Matheny 1988 22 Kennedy 1983 Bone abbreviations: Fe = Femur Ti = Tibia bFe = Bicondylar femur length (5 mm were added to use Genovés 1967) Fi = Fibula Hu = Humerus Ul = Ulna Lb = Long bone Formula abbreviations: G = Genovés 1967 P= Pearson 1899 TGM = TrotterGleser (1958) Mexican TGW = TrotterGleser (1958) White as modified by Haviland 1967
Page 112
larger in males since it has been argued that their growth is more likely to deviate from the normal curve (Tanner 1962). Such a pattern occurs only at Altar de Sacrificios. At Tikal, while males are becoming significantly shorter between the Preclassic and Late Classic, females actually gain 2.3 cm in height. Under periods of duress they would not necessarily be expected to get taller. In contrast, at Seibal and Barton Ramie, while males are becoming taller over time, the females are becoming shorter. Very likely, these mixed results reflect the complex nature of the causes of sexual dimorphism. The role of leg length in revealing information concerning the effects of the environment on stature is difficult to assess in the prehistoric Maya. Unfortunately, the study among the modern Japanese (Tanner et al. 1982) demonstrated that changes in body proportions, namely increases in leg length, were associated with recovery from dietary stress, a situation not seen among the Maya. If the reverse corollary of legs becoming shorter during times of stunting may be assumed, then analysis of limb proportions might be informative in the present study. Using stature estimates for arm versus leg bones shows that the Preclassic inhabitants of Altar de Sacrificios indeed had relatively longer legs than did their Early Classic counterparts. Unfortunately, this is the only instance in which comparisons using sample sizes of more than one is possible. At the rest of the sites, stature estimates based on leg bones are generally longer than those on arm bones or the two estimates are very similar. Therefore, no specific conclusions may be drawn. Discussion and Conclusions Overall, the pattern of stature change in the prehistoric Maya is one from which few conclusive statements can be made. The modern Maya indeed appear to be shorter than the Precolumbian Maya, and most of the reduction seems to have taken place in the last 500 years. Such a finding would not surprise individuals familiar with the modern living conditions of the Maya in rural Guatemala. The stature decline, however, is of greatest interest to archaeology in terms of its association with the Late Classic "collapse' and the aggregated data adhere to expectations only at the very broadest of levels. Although it might be argued that skeletal samples from different sites should not be combined since they potentially represent distinct gene pools, this may not be of great concern. Very likely, the role of genetics in stature determination in these populations is fairly minimal compared with that of the environment. Probably none of the populations involved are reaching their genetic potential for height. When the data for individual sites are evaluated, however, the fact that they do not consistently display a decline during the Late Classic suggests that the "collapse" may not have been a ubiquitous homogeneous phenomenon across the southern Lowlands. Thus, generalizing the stature decline to all Maya populations may cause us to miss potentially important and informative variation. In considering the reliability of these conclusions, we must address the dubious quality of much of the data base. Obviously, the biggest hindrance in
Page 113
analysis of stature patterns of the southern Lowlands is the relatively scanty amount of skeletal material available. The problem lies not only in sheer numbers but in how those individuals present are distributed, since it makes both statistical testing and betweengroup comparisons nearly impossible. First, as was previously mentioned, females are dramatically underrepresented compared with males. Second, the sample is unevenly distributed by time period, so that the number of individuals dating to the Late Classic period is larger than the sum of the other three periods combined. Especially lacking is an adequate sample from the Postclassic, which would allow analysis for possible recovery after the collapse. The situation is also somewhat analogous to that which Márquez Morfin (1984) encountered in the northern Lowlands, where skeletal material from the Preclassic is rare. Third, exceedingly few sites have skeletal material dating to more than one time period; thus, it is difficult to follow the interrelationship between health patterns and a site's particular history. Additionally, certain sites with large samples dating to a single period, such as Cuello and the Preclassic, often overwhelm the data from smaller sites in their influence on the mean stature values for the period. For example, eliminating Tikal from the Late Classic male sample raises the average height for the period by a full centimeter. Another variable that must be considered in stature evaluation is social status. Although rarely addressed, its influence in height determination in both the modern Maya in Guatemala (e.g., Bogin et al. 1992) and especially the prehistoric Maya at Tikal (Haviland 1967) is strongly evident. Perhaps, then, it should not be unexpected that the modern groups are shorter than their ancestors; the living group is composed of lowstatus peasants whereas the archaeological sample is more likely to include a large segment of individuals who are of elite status, since grand tombs are excavated more often than hut floors. Large ceremonial centers were also far more frequently excavated than were more outlying regions, thereby once again increasing the probability that the prehistoric Maya skeletal sample includes a disproportionate number of wealthier individuals. It was only starting in the mid1960s with the Belize River survey (Willey et al. 1965) that smaller communities began to be systematically investigated. Interestingly, the mean male stature values for the modern Maya (Steggerda 1932) and the nontomb burials at Late Classic Tikal (Haviland 1967) are almost identical. Also unanticipated was the fact that the results using the aggregated data generally adhered to expectations for variation by sex, despite the number of variables that can affect such dimorphism. Studies of other populations in which sexual dimorphism was similarly expected to decrease also failed to show differences (e.g., Lallo 1973), leading some to suggest that there may be "difficulty in demonstrating changes in dimorphism in an archaeological population undergoing rapid changes in diet" (HussAshmore et al. 1984:414). Interestingly, the results also suggest that male preference in certain cultural practices, such as protein access or medical attention, was not offsetting any advantages females had through genetic buffering. Male preference would be ex
Page 114
pected in patrilineal societies such as the Maya, but it was not seen in analysis of other childhood health indicators, including porotic hyperostosis and linear enamel hypopolasia, in the southern Lowlands (Danforth et al. 1991). In conclusion, the truism so often heard in Mesoamerican archaeology that "the Maya got short" during the collapse has been given far more emphasis than is warranted according to analysis of the available data. Although a slight decline in stature may have been associated with males during the Late Classic, the sample sizes for most sites are exceedingly small and the differences in height are not statistically significant. Similarly, it does not appear to have been a uniform pattern across the Petén. Broad generalizations concerning Maya stature also may be doing disservice by potentially masking informative patterns of variation. The variation in stature very much correlates with the apparent variation seen in other health and dietary indicators during the Late and Terminal Classic periods (Wright and White 1996). In this manner, the biological evidence supports the archaeological evidence that the collapse was a complex phenomenon with many patterns of manifestation across the southern Lowlands and that its explanations are likely to be equally diverse. References Cited Bogin, B. (1988) Patterns of Human Growth. Cambridge: Cambridge University Press. Bogin, B. (1991) Measurement of growth variability and environmental quality in Guatemalan children. Annals of Human Biology 18:285294. Bogin, B. (1995) Plasticity in the growth of Mayan refugee children living in the United States. In C. G. N. MascieTaylor and B. Bogin (eds.): Human Variability and Plasticity. Cambridge: Cambridge University Press, pp. 4674. Bogin, B.; Wall, M.; and MacVean, R. B. (1992) Longitudinal analysis of adolescent growth of Ladino and Mayan school children in Guatemala: Effects of environment and sex. American Journal of Physical Anthropology 89:441446. Bullard, W. R., and Bullard, M. R. (1965) Late Classic Finds at Baking Pot, British Honduras. Art and Archaeology Occasional Paper No. 8. Toronto: Royal Ontario Museum. Coe, W. R. (1959) Piedras Negras Archaeology: Artifacts, Caches, and Burials. Museum Monographs, University Museum. Philadelphia: University of Pennsylvania. Cohen, M. N.; Bennett, S. L.; and Armstrong, C. W. (1989) Final report to the National Science Foundation on grants BNS 8506785 (Health and genetic relationships in a colonial Maya population) and BNS 8303693 (Excavation of a colonial Maya cemetery in Belize). Ms. on file, Department of Anthropology, State University of New York at Plattsburgh. Danforth, M. E.; Jacobi, K. P.; and Cohen, M. N. (1991) Gender and health in the prehistoric and early contact period Lowland Maya. Paper presented at the annual meeting of the American Anthropological Association, Washington, D.C., April. Danforth, M. E.; Whittington, S. L.; and Jacobi, K. P. (1997) An indexed bibliography of Maya human osteology, 18391994. In S. L. Whittington and D. M. Reed (eds.): Bones of the Maya: Skeletal Studies of an Ancient People. Washington, D.C.: Smithsonian Institution Press, pp. 229260. Dettwyler, K. A. (1992) Nutritional status of adults in rural Mali. American Journal of Physical Anthropology 88:309322. Frisancho, A. R. (1993) Human Adaptation and Accommodation. Ann Arbor: University of Michigan Press.
Page 115
Genovés, S. (1967) Proportionality of the long bones and their relation to stature among Mesoamericans. American Journal of Physical Anthropology 26:6979. Hammond, N.; Pring, D.; Wilk, R.; Donaghey, S.; Saul, F. P.; Wing, E. S.; Miller, A. V.; and Feldman, L. N. (1979) The earliest Lowland Maya? Definition of the Swasey Phase. American Antiquity 44:92110. Haviland, W. A. (1967) Stature at Tikal, Guatemala: Implications for ancient Maya demography and social organization. American Antiquity 32:317325. Helmuth, H., and Pendergast, D. M. (198687) Lamanai tomb N958/1: Analysis of the skeletal evidence. Ossa 13:109118. HussAshmore, R.; Goodman, A. H.; and Armelagos, J. (1984) Nutritional inference from paleopathology. In M. Schiffer (ed.): Archaeological Methods and Theory, vol. 5. New York: Academic Press, pp. 395474. Jaen Esquivel, M. T. (1966) El material osteológico de Chiapa de Corzo, Chiapas. Anales del Instituto Nacional de Antropologia e Historia 19:6577. Kennedy, G. E. (1983) Skeletal remains from Sarteneja, Belize. In R. V. Sidrys (ed.): Archaeological Excavations in Northern Belize. Central America Monograph No. 17. Los Angeles: Institute of Archaeology, University of California at Los Angeles, pp. 353372. Kidder, A.V. (1937) Notes on the Ruins of San Agustin Acasaguastlán, Guatemala. Contributions to American Archaeology, Vol. 3, No. 15. Washington, D.C.: Carnegie Institution, Publication 456. Lallo, J. (1973) The Skeletal Biology of Three Prehistoric American Indian Populations from Dickson Mound. Unpublished Ph.D. dissertation, University of Massachusetts, Amherst. Longyear, J. M. (1940) An Old Empire skeleton from Copán, Honduras. American Journal of Physical Anthropology 27:151154. Longyear, J. M. (1952) Copán Ceramics. Washington, D.C.: Carnegie Institution, Publication 597. Lowe, J.W. G. (1985) The Dynamics of Apocalypse. Albuquerque: University of New Mexico Press. McElroy, A., and Townsend, J. (1989) Medical Anthropology in Ecological Perspective. Boulder, Colo.: Westview Press. Márquez Morfín, L. (1984) Distribución de la estatura en colecciones oseas Mayas prehispánicas. In R. Ramos Galván and R. M. Ramos Rodriguez (eds.): Estudios de antropologia biológica (II Colegio de Antropología Física Juan Comas, 1982). Mexico City: Estudios de Antropología Biología, Universidad Nacional Autónoma de México, pp. 253271. Martorell, R., and Habicht, J. P. (1986) Growth in early childhood in developing countries. In F. Falkner and J. M. Tanner (eds.): Human Growth: A Comprehensive Treatise, vol. 3. New York: Plenum Press, pp. 241262. Matheny, D. G. (1988) The Northwest Plaza Burials at Lagartero, Chiapas, Mexico. Unpublished Ph.D. dissertation, University of Utah, Salt Lake City. Méndez, J., and Behrhorst, C. (1963) The anthropometric characteristics of Indians and urban Guatemalans. Human Biology 36:457469. Mueller, W. H. (1976) Parentchild correlations for stature and weight among school aged children: A review of 24 studies. Human Biology 48:379397. Nickens, P. R. (1976) Stature reduction as an adaptive response to food production in Mesoamerica. Journal of Archaeological Science 3:3141. Pearson, K. (1899) Mathematical contributions to the theory of evolution. V. On the reconstruction of the stature of prehistoric races. Philosophical Transactions of the Royal Society of London 192:169244. Pelto, G. H., and Pelto, P. J. (1988) Small but healthy: An anthropological perspective. Human Organization 48:1152.
Page 116
Ricketson, O. G., and Ricketson, E. B. (1937) Uaxactún, Guatemala, Group E, 192631. Washington, D.C.: Carnegie Institution, Publication 477. Romano Pacheco, A. (1979) El material osteológico humano de Toniná, Chiapas: Estudio morfológico, descriptivo y comparativo. In P. Becquelin and C. F. Baudez (eds.): Toniná: Une cité Maya du Chiapas (Mexique). Etudes Mesoamericaines,Vol. 6, No. 1. Mexico City: Archeologique et Ethnologique Française au Mexique, pp. 179192. Russell, M. (1976) Parentchild and siblingsibling correlations of height and weight in a rural Guatemalan population of preschool children. Human Biology 48:501 515. Santley, R. S.; Killon, T. W.; and Lycett, M. T. (1986) On the Maya collapse. Journal of Anthropological Research 42:123159. Saul, F. P. (1972) The Human Skeletal Remains from Altar de Sacrificios, Guatemala: An Osteobiographic Analysis. Papers of the Peabody Museum, Vol. 63, No. 2. Cambridge: Harvard University. Saul, F. P., and Saul, J. M. (1989) Osteobiography: A Maya example. In M. Y. Iscan and K. A. R. Kennedy (eds.): Reconstruction of Life from the Skeleton. New York: Alan R. Liss, pp. 287302. Saul, F. P., and Saul, J. M. (1991) The Preclassic population of Cuello. In N. Hammond (ed.): Cuello: An Early Maya Community in Belize. Cambridge: Cambridge University Press, pp. 134158. Scrimshaw, N. S., and Behar, M. (1961) Protein malnutrition in young children. Science 133:20392047. Seckler, D. (1982) Small but healthy: A basic hypothesis in theory, measurement, and policy of malnutrition. In P. V Sukhatme (ed.): Newer Concepts in Nutrition and Their Implications for Policy. Pune: Maharashtra Association for the Cultivation of Science, pp. 127137. Serrano Sánchez, C. (1973) La lesión suprainiana en Mesoamérica: Implicaciones arqueológicas. Estudios de Cultura Maya 9:2943. Smith, R. E. (1937) A Study of the Structure A1 Complex at Uaxactún, Petén, Guatemala. Contributions to American Archaeology, No. 19. Washington, D.C.: Carnegie Institution, Publication 456. Starr, F. (1902) The Physical Characteristics of the Indians of Southern Mexico. Chicago: University of Chicago Press. Steele, D. G. (1986) The skeletal material from Río Azul, 1984 season. In D. G. Steele (ed.): Río Azul Reports Number 2, The 1984 Season. San Antonio: Center for Archaeological Research, University of Texas at San Antonio, pp. 111116. Steggerda, M. (1932) Anthropometry of Adult Maya Indians. Washington, D.C.: Carnegie Institution, Publication 434. Stewart, T. D. (1949) Notas sobre esqueletos humanos prehistóricos hallados en Guatemala. Antropología e Historia de Guatemala 1:2334. Stini, W. A. (1969) Nutritional stress and growth: Sex differences in adaptive response. American Journal of Physical Anthropology 31:417426. Stini, W. A. (1971) Evolutionary implications in human populations. American Anthropology 73:10191030. Stinson, S. (1994) Are females more buffered than males during postnatal growth? American Journal of Physical Anthropology, Suppl. 18, p. 188. Tanner, J. M. (1962) Growth in Adolescence. 2d ed. Oxford: Blackwell. Tanner, J. M.; Hayashi, T.; Preece, M. A.; and Cameron, N. (1982) Increase in length of leg relative to trunk in Japanese children and adults from 1957 to 1977: Comparison with British and with Japanese Americans. Annals of Human Biology 9:411423. Thompson, J. E. S. (1939) Excavations at San José, British Honduras. Washington, D.C.: Carnegie Institution, Publication 506. Trotter, M. (1970) Estimation of stature from intact limb bones. In T. D. Stewart (ed.): Personal Identification in Mass Disasters. Washington, D.C.: National Museum of Natural History, pp. 7183.
Page 117
Trotter, M., and Gleser, G. C. (1958) A reevaluation of estimation of stature based on measurements taken during life and of long bones after death. American Journal of Physical Anthropology 16:79123. Waterlow, J. C., and Payne, P. R. (1975) The protein gap. Nature 258:113117. Wauchope, R. (1934) House Mounds at Uaxactún, Guatemala. Washington, D.C.: Carnegie Institution, Publication 436. Welsh, W. B. M. (1988) An Analysis of Classic Lowland Maya Burials. International Series 409. London: British Archaeological Reports. Willey, G. R. (1965) Human burials. In G. R. Willey, W. R. Bullard, J. B. Glass, and J. C. Gifford (eds.): Prehistoric Settlement Patterns in the Belize Valley. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 54. Cambridge: Harvard University, pp. 530558. Willey, G. R.; Bullard, W. R.; Glass, J. B.; and Gifford, J. C. (eds.) (1965) Prehistoric Settlement Patterns in the Belize Valley. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 54. Cambridge: Harvard University. Willey, G. R., and Shimkin, D. B. (1973) The Maya collapse: A summary view. In T. P. Culbert (ed.): The Classic Maya Collapse. Albuquerque: University of New Mexico Press, pp. 457501. Wright, L. E., and White, C. D. (1996) Human biology in the Classic Maya collapse: Evidence from paleopathology and paleodiet. Journal of World Prehistory 10:147198. Young, D. (1994) Analysis of human skeletal remains from Operation 2031, Colha, Belize. In T. R. Hester, H. J. Shafer, and J. D. Eaton (eds.): Continuing Archeology at Colha, Belize. Studies in Archeology No. 16. Austin: Texas Archeological Research Laboratory, University of Texas, pp. 5963.
Page 119
Chapter 6 Land Use, Diet, and Their Effects on the Biology of the Prehistoric Maya of Northern Ambergris Cay, Belize David M. Glassman and James F. Garber The Ambergris Cay Archaeological Project was initiated in 1985 and continued survey and excavation work until 1989. Directorship responsibilities were shared among Thomas Guderjan, Herman Smith, James Garber, and David Glassman, representing the collaboration of the Institute of Texan Cultures, Corpus Christi Museum, and Southwest Texas State University. The central focus of the project concentrated on identifying the role of settlements on the northern end of Ambergris Cay in the prehistoric Maya maritime trade system. Subsequent research was conducted on the impact of trade on settlement patterns, relationships between sites, socioecological relationships of island habitation, and the skeletal biology of the inhabitants. The island habitat provided an ecological challenge that differed from the agriculturally based subsistence patterns of mainland populations. Data for the study were collected from two sites, San Juan and Chac Balam, located on the leeward side of the island. Both sites provided substantial evidence of participation in prehistoric maritime trade. Geography of Ambergris Cay Ambergris Cay is a large offshore island in northern Belize stretching approximately 39 km in length and varying in width between 1.5 and 7.5 km. The island is oriented approximately northsouth. The barrier reef, a source of considerable marine resources, runs parallel to the island at a general distance of 100 to 200 m from the coastline. Waters inside the reef are calm and shallow, generally less than 2 m in depth, and would have provided easy, safe passage for canoe travel involved in marine subsistence practices, site interaction, and maritime trade. Ambergris Cay was formed during the Pleistocene from a limestone shelf foundation as an extension of the Xcalac Peninsula of Quintana Roo, Mexico. It has been postulated that by A.D. 600, a narrow channel may have been excavated by the Maya resulting in the present island configuration (Guderjan 1988). This intentionally constructed channel, known as the Boca Bacalar Chico Passage, would have greatly facilitated canoe travel by increasing safety and decreasing time and energy requirements for movement between the Yucatán coast and the Chetumal Bay.
Page 120
Guderjan (1995a) recognized five distinct environmental zones on the island exclusive of the reef itself. The zones were defined by their location (coastline beach versus inland) and their position relative to the height of the underlying limestone plate. For example, in areas where the plate is lowest, the limestone is submerged and interior island lagoons surrounded by mangrove swamps occur. At slightly higher elevations, zones consisting of savanna vegetation are recognized. Finally, at the highest elevations, which exceed 3 meters above sea level, the savanna vegetation graduates into a low scrub forest environment. Site Descriptions The two sites, San Juan and Chac Balam, were located on the leeward side of the island. Both sites were constructed near the Bacalar Chico Canal in locations that were ideal for servicing the longdistance maritime trade network that made use of the canal. Archaeological evidence suggested that these sites had multiple functions as habitation, religious, and administrative centers. During its peak development the site plan of San Juan consisted of eight or nine substructural mounds that were informally arranged atop artificially raised platforms. The site covered approximately 9,000 m2. Initial construction of the site took place before approximately A.D. 600. Occupation and site development continued into the Terminal Classic. Several of the San Juan structures exhibited multiple construction phases. By the end of the Terminal Classic the site was abandoned. More detailed descriptions of the excavations and architecture of the San Juan site have been reported by Guderjan (1995b). The site of Chac Balam was surrounded on three sides by mangrove swamp. The site core consisted of a single plazuela group. Chac Balam covered approximately the same area as San Juan at 9,000 m2. In contrast to San Juan, the four major substructural mounds of Chac Balam were formally arranged, creating an intentional area of public space. Access to the courtyard was restricted to entrances on the northern and southern sides. Occupation of the site began during the Late Classic and continued through the Terminal Classic and into the Early Postclassic. A detailed description of the excavations and architecture of Chac Balam has been reported by Driver (1995). Analysis of the San Juan and Chac Balam sites indicated their role in the prehistoric Maya maritime system. Artifacts recovered from both sites included exotic items often made from raw materials not indigenous to the island habitat, such as jade and obsidian. Furthermore, the relative size of the mounds was inconsistent with solely domestic activity and indicated a socially stratified settlement. Population of the sites appeared to have been relatively large and concentrated. Surveys of the surrounding areas revealed mostly mangrove swamp and did not indicate any peripheral housemound complexes. It appeared that the inhabitants of northern Ambergris were restricted to habitation within, and immediately surrounding, the substructural mound complexes. The size of the sites would have allowed for the positioning of multiple residential units, in
Page 121
cluding elite patio groups and domestic housing. Although a population estimate of San Juan is not yet available, population size of Chac Balam has been estimated at 240 persons (Driver 1995). Land Use and Diet Soil conditions of northern Ambergris are highly acidic and considered poor for agricultural pursuits. Even if soil conditions were more conducive to cultivation, the amount of land available for food production was extremely limited by lowelevation lagoons and mangrove swamp. Surveys of the areas surrounding the San Juan and Chac Balam sites did not reveal any raised fields or dark soil concentration, indicative of prehistoric agriculture. It is logical to assume that the northern Ambergris Maya exploited the marine resources and wild fauna and flora for their nutritional base. Supplementary to local resources, shelled maize and other foods may have been imported from the agriculturally rich mainland. Marine resources were easily accessible and provided a stable, highprotein dietary source. Several species of marine gastropods were recovered at both sites. Differing species indicated the exploitation of shallow waters between the coastline and reef, the reef, and brackish water environments of the lagoons. A detailed analysis of faunal materials recovered at the northern Ambergris site of Ek Luum has been reported by Shaw (1995). Although this site was not chosen for the present study because of the lack of human osteological remains, it provided a framework for understanding the diet of the island Maya. The quantity of molluscs recorded at San Juan and Chac Balam suggested that they did not make a major contribution to the diet but were supplemental. Recovered species of mollusc that were probably used as a food source were Melongena melongena and Cittarium pica. The Melongena species is associated with shallow environments as would occur at the reef and between the reef and coast. The Cittarium species is most commonly found in intertidal environments. The largest sample of faunal materials recovered at the two sites was from bony fish. This was consistent with the faunal frequencies identified for Ek Luum (Shaw 1995). Terrestrial resources, although used, appear to have been much less relied on than marine resources, as based on the recovery of a comparatively small sample of reptilian and mammalian bones at San Juan and Chac Balam. Although no agricultural fields were identified for the northern Ambergris inhabitants, the presence of manos and metates at the two sites indicated maize preparation activities. Nineteen manos and 9 metates were excavated at San Juan, and 20 manos and 17 metates were recovered at Chac Balam (Garber 1995). The occurrence of these grinding artifacts may indicate that small gardens were maintained in the very limited areas that had arable soil adjacent to the sites. Alternatively, shelled maize may have been traded to the site rulers as partial payment for use of the Bacalar Chico passage by maritime traders. More likely, however, a combination of in situ production and importation of maize probably occurred.
Page 122
Overall, the northern Ambergris Maya exploited various habitats around the site—terrestrial as well as marine localities—for food resources. Nevertheless, the archaeological evidence clearly pointed to marine resources as the dependent dietary staple, supplemented by wild game and maize. The incorporation of maize in the diet is inferred from the presence of manos and metates at the sites. In the case of the marine environments, Shaw's (1995) data for Ek Luum suggested reliance on resources associated with the shallow waters near the coastline as well as the waters of the nearby reef, which would have provided a variety of fish, molluscs, and crustaceans. San Juan and Chac Balam Burials Thirtyeight burials were excavated from the sites of San Juan and Chac Balam, yielding the remains of 43 individuals. All burials from San Juan were single interments and accounted for 9 individuals. The remaining 34 individuals at Chac Balam were recovered from 29 burials, of which 3 were multiple interments of 2 individuals each and 1 burial included 3 individuals. Several of the remains were examined and measured in situ for osteologic data before removal. All remains were taken to the physical anthropology laboratory at Southwest Texas State University for processing and analysis. Each individual was examined for sex, age, stature, and pathological conditions according to standard methods and techniques (e.g., Bass 1979; Brothwell 1981; Krogman 1962; Trotter and Gleser, 1958; Ubelaker 1978). Cranial and postcranial metric data were collected when possible. Finally, data indicative of developmental stress, including dental enamel hypoplasias, porotic hyperostosis, and radiographic Harris lines (San Juan sample only), were collected. The samples from the two sites were combined because of small sample size and similar Late Classic to Terminal Classic occupation. Age and gender composition is presented in Table 6.1. The mean age of an individual's estimated age range was used for assigning that individual to one of the age categories. Eighteen individuals could be identified by sex, with an equal distribution of 9 males and 9 females. Thirtynine of the 43 individuals could be placed in an age category; however, the largest group included adult individuals whose ages could not be further specified owing to limited data. All age categories were represented by the sample, and near equal sample sizes occurred for the categories between birth and 40 years. A social position was subjectively assigned to each burial to test for effects of disproportionate access to resources on stature, childhood developmental stress, and incidence of other health parameters such as trauma and infection. One of the four social positions (low, middle, high, and elite) was assigned on the basis of the quantity, type (ritual versus utilitarian), and source (import versus local) of grave furniture and the burial location within the sites. Assignment was made by the project archaeologist and junior author (Garber). We acknowledge that cultural status follows a continuum and discrete categories are oversimplifications of a much more complex system. Nevertheless, the placement of individuals into these categories permitted an examination of broad
Page 123 Table 6.1. Age and Gender Composition of the Ambergris Prehistoric Maya Burials.
B2
26
618
1830
3040
40+
Adult (age unknown)
Indet.
Total
%
Male
—
—
—
2
4
1
2
—
9
20.9
Female
—
—
1
4
—
—
4
—
9
20.9
5
7
6
—
—
—
—
—
18
41.9
—
—
—
—
—
—
3
4
7
16.3
5
7
7
6
4
1
9
4
43
100.0
11.6
16.3
16.3
14.0
9.3
2.3
20.9
—
9.3
100.0
Subadult Indeterminate Total Percent
generalizations potentially revealing of the interaction between biology and culture. Skeletal Biology of the Ambergris Burials The total morphological pattern of the Ambergris sample suggested a population of relatively muscular, smallstature individuals exhibiting good health. The practice of intentional cranial and dental deformation was common and appeared in 7 and 5 individuals, respectively. The practice of deformation was not associated with a specific sex or social position (Glassman 1995). Stature could be estimated for eight males and three females. Estimates were based on limb bone measurements according to formulae reported by Trotter and Gleser (1958) for Mexicans. When multiple limb bones were available for measurement, lower limb bones were given priority in stature estimation over arm long bones. The Trotter and Gleser formulae were chosen over other formulae devised for Mesoamericans (e.g., Genovés 1967) because of their application to long bones of both the leg and arm and for comparability to stature estimates of Late Classic populations reported for Tikal (Haviland 1967) and Altar de Sacrificios (Saul 1972). Danforth (this volume) outlines previous research that has been conducted on prehistoric Maya stature and provides a discussion on the question of changing stature over time. Ambergris males ranged in average stature between 159.7 cm and 169.3 cm with an overall average of 165.2 cm. The overall average stature for females was 156.4 cm with an average stature range between 155.4 cm and 158.0 cm. An association between stature and status level was examined for males but not for females because of small sample size. Although not statistically significant, a pattern emerged in which individuals assigned to the high/elite group (N = 5) averaged 167.1 cm compared to the average 162.0 cm for individuals assigned to the middle and low social status groups (N = 3). Little can be made of this pattern given the small sample size and methodology in assigning social position. It is interesting to note, however, that the two male individuals assigned to the elite social position were among the three tallest individuals in the sample. This finding supports the notion that for highly stratified societies, individuals in higher social position were healthier and had a better diet than those of lower social position and therefore were more likely to reach their stature potential.
Page 124
Dental Pathology Dental data were collected for attrition and antemortem loss. In addition, rates were calculated for carious lesions, abscesses, calculus deposits, and alveolar bone loss. The emerging dental pattern was subsequently interpreted relative to the marine diet attributed to the Ambergris inhabitants. Dental attrition was evaluated using the methodology and scoring systems of Lovejoy (1985) for the dental arch as a whole; Smith (1984) for the incisors, canines, and premolars; and Scott (1979) for the first and second molars. Overall, the Ambergris Maya exhibited a pattern of minimal dietary dental attrition as reflected in molar morphology. Differential degrees of attrition were noted between the anterior and posterior dentition with relatively increased amounts of dental attrition occurring to the anterior occlusal surfaces. For all individuals in which complete, or mostly complete, dental arches could be examined (N = 9), scores were restricted between the B2 and D categories according to Lovejoy (1985). Minimal attrition to the molars was further supported by examining the additional teeth that belonged to individuals whose complete dental arches could not be evaluated. No first or second molar exceeded an attrition score of greater than 20 using the Scott (1979) scale, and only one individual exceeded a score of 15, generally characterized as a tooth that is worn flat with no, or minimal, dentin exposure. Examination of all incisors and canines suggested a mosaic pattern in which the anterior dentition for individuals varied between little and severe attrition (score ranging from 2 to 7 on Scott's 1979 scale of 1 to 8). Although the sample size was small, the degree of anterior dental attrition did not appear to follow an agerelated gradient. Antemortem dental loss was difficult to document in the Ambergris sample owing to incomplete and fragmentary remains. Only 2 individuals, both estimated to be between 25 and 45 years old, had complete jaws for observation, and in both cases a complete dentition was present. Of the remaining 14 adults with partial alveolar processes that could be examined for dental loss, 5 demonstrated antemortem loss of 2 to 11 teeth. It appears that the loss of dental elements was not uncommon among this island population; the degree, and any agerelated pattern, to which dental loss was characteristic of the Ambergris Maya could not be determined, however. Each recovered tooth from all individuals was inspected for carious lesions and calculus deposits. A rate of incidence was calculated as the number of teeth exhibiting at least one carious lesion divided by the number of all recovered teeth for that individual. In addition, an overall population rate was calculated as the number of total carious teeth divided by the total sample of teeth recovered. The analysis was restricted to the permanent dentition of dentally adult individuals. The sample produced 16 teeth with at least one carious lesion among the 324 teeth available for scoring. This produced an overall caries rate statistic of .049. No individual that had at least 10 teeth recovered (N = 11) had an individual rate that exceeded .318, and only 4 of these 11 individuals (36.4 percent) had any carious lesions at all. It would appear that although dental caries were present among the Ambergris Maya, they were not a significant
Page 125
problem. It is reasonable to conclude that the relatively low caries rate among the Ambergris Maya was due to the concentration of marine resources in the diet, which would have been less progressive in the decalcification of enamel than sticky, highcarbohydrate plant foods (Cohen 1989). Calculus deposits were scored as present or absent. The data were then converted to rates following the procedure used for dental caries. The extent of the deposits ranged from small concretion flakes to a continuous band surrounding the entire tooth crown. The presence of at least one tooth having calculus was found for 9 (56.3 percent) out of the 16 adult individuals who had teeth available for study. A population total of 71 teeth out of 298 exhibited calculus deposits, for an overall rate of 23.8 percent. The highest rate for any single individual was .963, in which 26 of 27 recovered teeth displayed deposits. The calculated rates suggested that the presence of calculus deposits was a common characteristic among the Ambergris Maya. This presents a pattern consistent with that reported for other Maya populations at Lamanai and Tayasal, characterized by an inverse relationship between a relatively high incidence of calculus and a relatively low incidence of dental caries (Evans 1973; White 1986). Increased rates of dental calculus and caries have both been associated with high carbohydrate (sucrose) diets (Goodman et al. 1984; Schwartz 1995). If it is assumed that the Ambergris population was relying primarily on noncarbohydrate calories for subsistence, an alternative explanation is necessary to explain the high incidence of dental calculus. Perhaps the ingestion of particular marine resources was responsible for increasing pH levels of the saliva necessary for mineral crystals to form on the teeth but was not conducive to subsequent fermentation of bacteria and lowering of pH levels required for the acid related decalcification of enamel. Periapical abscesses were recorded for all recovered alveolar portions whether or not the associated tooth was present. To be counted as an abscess, the hole needed to be of characteristic morphology for drainage, including a round shape and defined margins (Schwartz 1995). Seventeen individuals were represented by alveolar portions, of varying completeness, representing the associated area of 296 teeth. Only three individuals exhibited periapical abscesses, with a frequency of one each. This is equivalent to a rate of .01 as calculated from the total population number of periapical abscesses divided by the total number of alveolar dental sockets available for examination. As periapical abscesses are often a subsequent condition in the continual process of dental destruction beginning with the development of carious lesions, the low rate of abscesses in the Ambergris population was consistent with, and related to, the low incidence of dental caries. The degenerative loss of alveolar bone was the final dental characteristic examined for the population. Alveolar loss was recorded as present or absent based on whether resorption had occurred to an extent greater than 1.5 mm between the alveolar margin and the cementoenamel joint to a tooth. Clarke and Hirsch (1991) suggest caution in applying this methodology, as increased distance may be the result of continued facial growth, including incremental eruption of the teeth, and not the result of loss of bone. Thirteen dentally
Page 126
adult individuals were available for data observation. Ten of them (76.9 percent) exhibited resorption of alveolar bone associated with at least one tooth. An overall rate of alveolar loss was calculated from dividing the total number of teeth with alveolar loss by the number of teeth scored. The population alveolar loss rate was .406, suggesting that the resorption of bone was a common degenerative condition. An agerelated pattern was identified in which individuals younger than 21 years of age exhibited no, or minimal, alveolar loss. The onset of loss occurred during the midtwenties and for individuals of age greater than or equal to 35 years; resorption was common throughout the upper and lower alveolar processes, generating alveolar loss rates between .833 and 1.0. Alveolar bone loss has been generally considered the result of progressive periodontitis. The data indicated that the diet and dental care practices of the Ambergris Maya did not preclude the condition of progressive inflammation of the gum tissue and subsequent destruction of the surrounding bone. This condition was consistent with, and perhaps related to, the common occurrence of calculus deposits within the population. Pathology Indicative of Nutritional and Disease Stress One of the most informative methods of determining the impact of diet on the skeletal biology of past populations is the evaluation of specific gross morphology attributed to vitamin and mineral and/or growth stress resulting from disease. The morphological indicators of stress evaluated in the Ambergris population include Harris lines of arrested growth, dental enamel hypoplasias and microstructure defects, and porotic hyperostosis. Harris lines are probably the most controversial indicator of stress (Harris 1926). The radiographic lines are created by transverse deposits of bone being laid down as a result of interruptions in the growth process. These interruptions have been thought to be related to either dietary or disease stress placed on the individual (e.g., Cook 1979; McHenry 1968; Wells 1967). Because normal processes of remodeling can obliterate these deposits, the strength of association between the presence of Harris lines and stress remains equivocal (Hummert and Van Gerven 1985; Marshall 1968). Because of the limitations of Harris lines in depicting dietary and disease stress, the decision was made to restrict the analysis to a subsample of the Ambergris burials. All recovered long bones of six individuals from the San Juan site were subjected to radiographic interpretation. The small subsample included two individuals under 12 years of age, two individuals determined to be young adults, and two adult individuals of age unknown. The two subadults both exhibited several Harris lines, although these lines tended to be restricted to only one or two pairs of long bones each. One of the young adults retained a single Harris line, and the remaining adult individuals showed no Harris lines for any bone element. The lack of Harris lines in the adults may be predicted by the normal pattern of Harris line obliteration through the remodeling process
Page 127
and does not necessarily indicate that these individuals did not have Harris lines at younger ages. The results of this subsample analysis, although cautiously interpreted because of very small sample size, suggested that Harris lines were not uncommon among the Ambergris population; their presence may reflect a disease load that affected growth during childhood and was episodic. A variety of dental defects, most specifically enamel hypoplasias and opacities, have been examined from archaeological populations to assess diet quality during the dental development ages. The relationship of enamel defects to metabolic stress has been considered substantially stronger than that between Harris lines and metabolic stress, primarily because defects formed in the tooth enamel during childhood are not subject to normal alteration or obliteration by bone remodeling as are Harris lines and therefore are preserved into the adult years (Goodman and Rose 1990). Ability to observe them is, however, affected by trauma, attrition, calculus, and dental loss. All deciduous and permanent teeth recovered from the Ambergris sample were examined for enamel defects. This sample included 280 teeth representing 30 different individuals. Only 31 teeth (11 percent), distributed among 6 individuals, exhibited any kind of enamel defect; and 1 individual solely accounted for 19 (61.3 percent) of the 31 defects. These relatively low percentages, as compared with other inland sites (Saul 1972), of both the overall rate of enamel defects and the incidence of individuals characterized by their presence, were a strong indicator that the marinebased diet of the northern Ambergris Maya was stable and of high quality and the disease load was probably low. Porotic hyperostosis has also been implicated as an indicator of iron deficiency, which can be caused by diet, parasites, or genetic disorders ( ElNajjar et al. 1976; Hengen 1971; Hooton 1940; Saul 1972). The locational pattern for porotic hyperostosis is commonly symmetrical on the parietal and occipital bones and adjacent to the major neurocranial sutures. In younger individuals porotic hyperostosis is occasionally found in the upper margin of the interior orbital surface, where it is known as cribra orbitalia. Examination of the Ambergris crania revealed two individuals characterized by porotic hyperostosis. One was a child between the ages of 3 and 5 years with the condition of cribra orbitalia, and the other was a young adult female exhibiting porotic hyperostosis in the area adjacent to the lambdoidal suture on the occipital. Although small sample size places caution on any generalization, the two individuals had been assigned to the low cultural position, perhaps indicating differential access to resources by this stratified society. The low frequency of porotic hyperostosis in the Ambergris population was consistent with the paucity of dental defects in suggesting that the marinebased diet was sufficient in nutrients and minerals to promote normal skeleton and dental growth and development. A high frequency of porotic hyperostosis, related primarily to parasitic exposure, has been reported for at least one marinedependent population (Walker 1986). Apparently, the Ambergris population was not affected by parasites in a similar manner.
Page 128
Other Pathological Conditions: Infectious Disease and Degenerative Joint Disease Abnormal bone formations attributed to infectious disease were identified for 13 individuals (31 percent) of the Ambergris population. These observations were limited to periosteal reactions and periapical abscesses. No abnormal endosteal bone formation or reduction in the size of the medullary cavity was noted. In each case in which abnormal bone formation was represented, the microorganism responsible for reactions was nonspecific. Of the 13 individuals with abnormal bone formation, 8 exhibited periosteal reactions at multiple sites. Of the remaining 5 individuals, there were 3 with one periapical abscess each, 1 with periosteal reaction in association to trauma, and 1 with periostitis in the locality of the sacroiliac joint. The fairly high frequency of abnormal bone formation suggested that infectious disease was a health risk factor for the Ambergris Maya. No association was found between the incidence of infection and social position. A precise estimate of the infectious disease load was not determined because of small sample size and, for some individuals, incomplete data. It might be expected, however, that a relatively large, sedentary population occupying a small, circumscribed island space would favor the transmission of disease. This may be further enhanced with the maritime trade activities that placed these inhabitants in continuous contact with other mainland groups. Analysis of degenerative joint disease may identify patterns of occupational stress including that related to subsistence activities. Data indicative of joint disease, including osteophyte development, porosity of the articular surfaces, and eburnation, were recorded for the Ambergris sample. Seven individuals representing five males and three females exhibited at least one site of degenerative joint disease, with four of the individuals (57.1 percent) exhibiting multiple sites of pathology. Joint disease occurred at several articular surfaces, including the shoulder, elbow, wrist, hip, knee, and ankle. The most common site of joint disease was the proximal ulna. Four individuals, all male, showed degeneration at this site. This pattern may be related to subsistence activities of males, in which repetitious hyperextension of the arms is involved in the canoeing activities necessary for collecting marine foods. Discussion and Conclusions The Late and Terminal Classic periods of the Maya in the Petén have been associated with increased disease loads and nutritional stress in concordance with the pressures of increasing population size and, subsequently, the collapse of the sociopolitical structure (Haviland 1967; Puleston and Callender 1967). It has been hypothesized that stature may have decreased and that other indicators of nutritional stress, such as the frequency of porotic hyperostosis, increased during these same time periods (Haviland 1967; Hooton 1940; Longyear 1952; Saul 1972). An analysis of stressrelated temporal change among the Ambergris Maya could not be completed since all burials have been dated to
Page 129 Table 6.2. Mean Stature for Terminal Classic Maya Populations (in cm).
Male
Female
Tikal
Copan
Altar de Sacrificios
157.4
159.1
165.8
165.2
(N = 21)
(N = 1)
(N = 2)
(N = 8)
147.0a
142.0
(N = 19)
(N = 1)
Ambergris
156.4 (N = 3)
Note: Method for calculating stature was Trotter and Gleser Mexican formulae for Tikal (Haviland 1967), Altar de Sacrificios (Saul 1972), and Ambergris; Pearson formulae and direct measurement for Copán (Longyear 1952). a Mean stature derived from individuals of all cultural periods.
the Late and Terminal Classic periods. Comparisons can be made, however, between the Ambergris population and other contemporary Maya groups. The average stature estimated for Ambergris males was 165.2 cm and for females was 156.4 cm (Table 6.2). These values are greater than average stature estimates from Tikal (Haviland 1967) and Copán (Longyear 1952) and only slightly smaller than the average of the two males from Altar de Sacrificios (Saul 1972). Ambergris males averaged 7.8 cm greater height than Tikal males, which represented the largest sample size population. The increased height for Ambergris females was more substantial, averaging between 9.4 cm and 14.4 cm. Although the sample sizes available for comparing stature are extremely small, and the use of differing stature estimation formulae by different researchers (e.g., Longyear for Copán) may hinder comparability, it would appear that relative to stature, the Ambergris Maya were more likely to reach their biological height potential. The stature estimates of the Ambergris, Tikal, and Altar de Sacrificios samples were based on similar formulae and therefore considered comparable. Recently, the hypothesis of decreasing stature during the Late and Terminal Classic periods has been critiqued from a methodological point of view and reevaluated relative to differing causal factors associated with stature reduction (Danforth 1994). Relatively high rates of porotic hyperostosis have been reported during the Late Classic or Terminal Classic periods for Altar de Sacrificios (Saul 1972) and Chichén Itzá (Hooton 1940) (see also Storey, this volume, for data from Copán), and, as is the case for stature reduction, the causal factors have been related to some form of nutritional stress and/or increased disease loads. In comparison, the Ambergris sample deviates from this pattern, having exhibited a low frequency of porotic hyperostosis with only two observable cases. Similarly, the incidence of stressrelated dental defects, such as hypoplasias, also had a low frequency rate. Only the relatively high rate of infectious disease at Ambergris was consistent with the nutrition and disease stress model hypothesized for the Late and Terminal Classic periods. It appeared that the Maya living on the northern end of Ambergris Cay were not subjected to the same decreased diet quality or caloric instability that has been proposed for other Maya populations owing to increased population
Page 130
pressure during the Late and Terminal Classic. In contrast, the island would have provided the inhabitants with a continuously stable, highprotein diet of marine resources. Within the Ambergris population, however, social position appeared to have an effect on nutritional health, as evidenced by the increased average stature of higher socially positioned individuals and the presence of two cases of porotic hyperostosis in individuals assigned to the lower social category. Social position did not appear to affect the rate of dental hypoplasias, carious lesions, or abscesses between groups. No pattern of differences was noted in the rates of dental and skeletal pathology conditions between males and females. Inland Maya diets that became increasingly dependent on high sticky carbohydrates, such as maize, have been implicated in the increased incidence of dental caries, dental calculus, periapical abscesses, and dental attrition in these groups. The Ambergris diet resulted in a varied pattern of high and low frequencies of dental pathology. Consistent with a low dependency on sticky carbohydrates, the Ambergris Maya exhibited relatively low rates of dental caries, periapical abscesses, and minimal molar attrition. Conversely, the diet supported the buildup of dental calculus. Although the sites of San Juan and Chac Balam were eventually abandoned during the Terminal Classic period, the biological evidence from skeletal material suggests that the inhabitants were of relatively good health and supported by a nutritious diet. Population pressure did not seem to affect the diet and lives of these inhabitants adversely. The abandonment of their settlements was probably a reaction to the breakdown of the sociopolitical environment of inland populations, which resulted in major changes in maritime trade activities, as seen in the Postclassic. With the Postclassic restructuring of the trading system, the trading functions of San Juan and Chac Balam would have been eliminated and the maintenance of their sociopolitical structure impractical. Acknowledgments The authors wish to thank Trent Stockton and Stephen Schubert for their assistance in data collection; our colleagues Tom Guderjan and Herman Smith for their contributions to the excavations of the northern Ambergris sites; our field assistants, David Driver and Lisa BrodyFoley; and the numerous student volunteers who worked in the field and in the laboratory. Finally, we gratefully acknowledge the Belize Department of Archaeology for granting authorization of this project. This research was supported, in part, by the Texas Higher Education Coordinating Board Advanced Research Program (Grant No. 3645) and the Corpus Christi Museum. References Cited Bass, W. M. (1979) Human Osteology: A Laboratory and Field Manual of the Human Skeleton. 2d ed. Columbia: Missouri Archaeological Society, Special Publication No. 2.
Page 131
Brothwell, D. R. (1981) Digging Up Bones. 2d ed. Ithaca, N.Y.: Cornell University Press. Clarke, N. G., and Hirsch, R. S. (1991) Physiological, pulpal, and periodontal factors influencing alveolar bone. In M. A. Kelly and C. S. Larsen (eds.): Advances in Dental Anthropology. New York: WileyLiss, pp. 241266. Cohen, M. N. (1989) Health and the Rise of Civilization. New Haven: Yale University Press. Cook, D. C. (1979) Subsistence base and health in prehistoric Illinois valley: Evidence from the human skeleton. Medical Anthropology 3:109124. Danforth, M. E. (1994) Stature change in prehistoric Maya of the Southern Lowlands. Latin American Anthropology 5:206211. Driver, W. D. (1995) Chac Balam: Excavations and architecture of a formal plaza group. In T. H. Guderjan and J. F. Garber (eds.): Maya Maritime Trade, Settlement, and Populations on Ambergris Cay, Belize. Lancaster, Calif.: Labyrinthos, pp. 4365. ElNajjar, M.; Ryan, D. J.; Turner, C.; and Lozoff, B. (1976) The etiology of porotic hyperostosis among the historic and prehistoric Anasazi Indians of southwestern United States. American Journal of Physical Anthropology 44:477487. Evans, D. T. (1973) A preliminary evaluation of tooth tartar among the Preconquest Maya of the Tayasal area, El Petén. American Antiquity 50:102116. Garber, J. F. (1995) The artifacts. In T. H. Guderjan and J. F. Garber (eds.): Maya Maritime Trade, Settlement, and Populations on Ambergris Cay, Belize. Lancaster, Calif.: Labyrinthos, pp. 113137. Genovés, S. C. (1967) Proportionality of long bones and their relation to stature among Mesoamericans. American Journal of Physical Anthropology 26:6778. Glassman, D. M. (1995) Skeletal biology of the prehistoric Maya of northern Ambergris Cay. In T. H. Guderjan and J. F. Garber (eds.): Maya Maritime Trade, Settlement, and Populations on Ambergris Cay, Belize. Lancaster, Calif.: Labyrinthos, pp. 7393. Goodman, A. H.; Martin, D. L.; Armelagos, G. J.; and Clark, G. (1984) Indications of stress from bone and teeth. In M. N. Cohen and G. J. Armelagos (eds.): Paleopathology at the Origins of Agriculture. Orlando: Academic Press, pp. 1349. Goodman, A. H., and Rose, J. C. (1990) Assessment of systematic physiological perturbations from dental enamel hypoplasias and associated histological structures. Yearbook of Physical Anthropology 33:59110. Guderjan, T. H. (1988) Maya Maritime Trade at San Juan, Ambergris Cay, Belize. Unpublished Ph.D. dissertation, Southern Methodist University, Dallas. Guderjan, T. H. (1995a) The setting and Maya maritime trade. In T. H. Guderjan and J. F. Garber (eds.): Maya Maritime Trade, Settlement, and Populations on Ambergris Cay, Belize. Lancaster, Calif.: Labyrinthos, pp. 18. Guderjan, T. H. (1995b) Excavations and architecture at San Juan. In T. H. Guderjan and J. F. Garber (eds.): Maya Maritime Trade, Settlement, and Populations on Ambergris Cay, Belize. Lancaster, Calif.: Labyrinthos, pp. 3142. Harris, H. A. (1926) The growth of long bones in children, with special reference to certain bony striations of the metaphysis and to the role of the vitamins. Archives of Internal Medicine 38:785806. Haviland, W. A. (1967) Stature at Tikal, Guatemala: Implications for ancient Maya demography and social organization. American Antiquity 32:316325. Hengen, O. P. (1971) Cribra orbitalia: Pathogenesis and probable etiology. Homo 22:5775. Hooton, E. A. (1940) Skeletons from the Cenote of Sacrifice at Chichen Itza. In C. L. Hay, R. Linton, S. K. Lothrop, H. Shapiro, and G. C. Vaillant (eds.): The Maya and Their Neighbors. New York: AppletonCentury, pp. 272280. Hummert, J. R., and Van Gerven, D. P. (1985) Observation on the formation and persistence of radiopaque transverse lines. American Journal of Physical Anthropology 66:297306.
Page 132
Krogman, W. M. (1962) The Human Skeleton in Forensic Medicine. Springfield, Ill.: Charles C Thomas. Longyear, J. M. III (1952) Copan Ceramics: A Study of Southwestern Maya Pottery. Washington, D.C.: Carnegie Institution of Washington, Publication 597. Lovejoy, C. O. (1985) Dental wear in the Libben population: Its functional pattern and role in the determination of adult skeletal age at death. American Journal of Physical Anthropology 68:4756. McHenry, H. (1968) Transverse lines in long bones of prehistoric California Indians. American Journal of Physical Anthropology 29:117. Marshall, W. A. (1968) Problems in relating the presence of transverse lines in the radius to the occurrence of disease. In D. R. Brothwell (ed.): The Skeletal Biology of Earlier Human Populations. Oxford: Pergamon Press, pp. 245261. Puleston, D. E., and Callender, D. (1967) Defensive earthworks at Tikal. Expedition 9:4048. Saul, F. P. (1972) The Human Skeletal Remains from Altar de Sacrificios, Guatemala: An Osteobiographical Analysis. Papers of the Peabody Museum, Vol. 63, No. 2. Cambridge: Harvard University. Schwartz, J. H. (1995) Skeletal Keys: An Introduction to Human Skeletal Morphology, Development, and Analysis. New York: Oxford University Press. Scott, E. C. (1979) Dental wear scoring technique. American Journal of Physical Anthropology 51:213218. Shaw, L. C. (1995) Analysis of faunal materials from Ek Luum. In T. H. Guderjan and J. F. Garber (eds.): Maya Maritime Trade, Settlement, and Populations on Ambergris Cay Belize. Lancaster, Calif.: Labyrinthos, pp. 175181. Smith, B. H. (1984) Patterns of molar wear in huntergatherers and agriculturalists. American Journal of Physical Anthropology 63:3956. Trotter, M., and Gleser, G. C. (1958) A reevaluation of estimation of stature based on measurements of stature taken during life and of long bones after death. American Journal of Physical Anthropology 16:79123. Ubelaker, D. H. (1978) Human Skeletal Remains: Excavation, Analysis, Interpretation. Chicago: Aldine. Walker, P. L. (1986) Porotic hyperostosis in a marinedependent California Indian population. American Journal of Physical Anthropology 69:345354. Wells, C. (1967) A new approach to paleopathology: Harris lines. In D. R. Brothwell and S. T. Sandison (eds.): Disease in Antiquity. Springfield, Ill.: Charles C Thomas, pp. 390404. White, C. D. (1986) Paleodiet and Nutrition of the Ancient Maya at Lamanai, Belize: A Study of Trace Elements, Stable Isotopes, Nutritional and Dental Pathology. Unpublished M.A. thesis, Trent University, Peterborough, Ontario.
Page 133
Chapter 7 Dietary Change at the Lowland Maya Site of Kichpanha, Belize Ann L. Magennis Prehispanic diet and subsistence practices in the Maya Lowlands engenders a great deal of debate among Mesoamerican archaeologists. A number of subsistence models have been proposed, but all of them remain relatively controversial. Initially, swidden agriculture of predominantly maize, beans, and squash was proposed as the most likely subsistence model for the ancient Maya, much like the milpa system practiced in the region today. More recent research, however, has demonstrated that intensive agricultural techniques were practiced, including raised and ditched fields (Harrison and Turner 1978; Pohl 1990; Turner and Harrison 1983) and terracing (Fedick 1994; Healey et al. 1983; White et al. 1993) as a means of managing water and the landscape and increasing agricultural productivity. Further, it appears that these agricultural intensification practices started as early as the Preclassic in some areas (Hammond and Miksicek 1981; Pohl 1990). Research over the past decade or so has also shown that a number of economic tree species have been used (e.g., Cliff and Crane 1989; Hammond and Miksicek 1981; McKillop 1994; Miksicek 1991; Lentz, this volume; Pohl et al. 1990; Wiseman 1983), although it is still controversial as to whether these tree species are simply utilized or intentionally managed and manipulated as part of the subsistence regime. Also featuring prominently in these subsistence debates is the importance of aquatic and terrestrial animal resources and the extent to which some or any of these may have been managed or harvested, as well as relative importance of aquatic or terrestrial protein sources in the overall subsistence mix. Although there is still much to be learned about Maya subsistence practices, much also remains to be learned about Maya diet. Only in recent years have archaeologists specifically included in their research designs questions about diet, incorporated the appropriate excavation and analytical techniques such as screening and flotation, and then integrated the analyses of pollen, phytoliths, macrobotanical remains, animal bone, human bone, molluscs, soils, stratigraphy, and contemporary ecology to address those questions. Although all these types of information are very important for subsistence reconstruction and these remains tell us something about the ''menu" or resource potential, they do not identify specifically what is actually consumed. The analysis of human skeletal remains has played an increasingly important role in addressing questions of ancient Maya subsistence, social organization, economics, and diet. Particularly illustrative with respect to reconstructing Maya diet have
Page 134
been analyses of stable isotopes and trace elements in human bone (e.g., Reed 1994; Tykot et al. 1996; White 1988,1994; White and Schwarcz 1989; White et al. 1993; Wright, this volume). These analyses can provide broad indicators of diet and may allow estimates of relative proportions of major food groups in the diet. Also important are analyses of skeletal and dental pathological conditions such as porotic hyperostosis, dental enamel hypoplasias, periostitis, or patterns of childhood growth and development (e.g., Glassman and Garber, this volume; Storey 1992a, 1992b; White et al. 1994; Whittington 1992; Wright and White 1996; Wright 1990,1997; see also Goodman et al. 1984; HussAshmore et al. 1982; and Larsen 1987 for general discussion), because they provide indications of dietary and nutritional adequacy and as such are partly measuring the consequences of consuming a given diet. Dental disease, specifically dental caries and calculus, provides general information about type of diet and dietary consistency. Considered alone, any of these measures of food resource potential, nutritional adequacy, or dietary intake supplies only a part of the picture. With the integration and synthesis of all these sources of information we may begin to develop a clearer view of ancient Maya diet. The purpose of this chapter is to investigate types of diet and dietary change from the Late Preclassic to the Terminal Classic period using the human skeletal sample from Kichpanha, a moderatesized ceremonial center in northcentral Belize. Specifically, I examine and report on dental caries and calculus frequency and patterning. Using these data, I make inferences about diet and offer hypotheses about specific changes in diet. Diet and Dental Pathology The dentition can be especially informative about diet, food preparation techniques, and food consumption. In particular, the patterning and frequency of dental caries in prehistoric skeletal samples have been used to make inferences about diet and behavior. Dental caries is a multifactorial disease, and Newbrun (1982) cites three principal factors that are involved in its promotion: (1) the host, including the teeth and the saliva, (2) the presence of plaque, which adheres to the tooth surface and which includes the microflora, or bacteria, and (3) the diet. Larsen and coworkers (1991) consider these to be "essential factors" in the occurrence of caries. Larsen et al. (1991), citing the work of Rowe (1975), also list a number of "modifying factors" that can affect the site and speed of carious lesion development. These include such things as tooth crown morphology and size, degree of occlusal attrition, developmental enamel defects, age, periodontal disease, traceelement composition of enamel, food texture and processing techniques, systemic disease, and the presence of fluorides and other local geochemical factors (Calcagno and Gibson 1988; Hildebolt et al. 1988,1989; Larsen 1987; Milner 1984; Powell 1985; Schneider 1986). Given the complex interaction of essential and modifying factors in caries formation, it is important to understand these factors to interpret caries frequency and patterning in a population. Once a tooth emerges in the oral environment, it is coated with a thin layer
Page 135
of protein from the saliva. Shortly thereafter bacteria, mostly streptococci, become adsorbed onto the protein layer. The bacteria on the tooth surface are embedded in a matrix that is created partly by the bacteria themselves and partly from proteins in the saliva (Hillson 1986:28387). As the bacterial community and the matrix build, they form dental plaque deposits. Particularly important for plaque development in humans are polysaccharides synthesized from sucrose. Nutrients—sugars and amino acids—from the diet diffuse into the plaque, where they are fermented by the bacteria to produce energy. The process of fermentation produces acid as a waste product. This acid production reduces the plaque pH, which, if lowered sufficiently, demineralizes enamel, cement, and dentine. Dental caries is the process of localized destruction, or demineralization, of dental tissue by these plaque microorganisms. Although carbohydrates in general are clearly the critical dietary factor in dental decay, the form of the carbohydrate is also important. The role of sugars in promoting caries has been studied extensively (see Newbrun 1982). Diets high in sugar, especially sucrose, are particularly cariogenic. Complex carbohydrates or carbohydrates in solution that do not adhere to the tooth surface have little negative effect. Although the role of carbohydrates in caries formation has been studied thoroughly, less is known about other dietary constituents and their impact on dental caries formation. For example, starches have often been implicated as an important dietary component in caries formation. Starches, however, are relatively large molecules that do not readily diffuse into the plaque, and hence the bacteria do not have access to them for fermentation. Starches first have to be broken down by salivary enzymes, amylases, to yield sugars, which the plaque bacteria can then ferment (Hillson 1986; Newbrun 1982). Newbrun (1982) also points out that studies of people with hereditary fructose intolerance show that they consume very little sucrose but do consume foods with other sugars and starch such as dairy products, pasta, rice, breads, and potatoes. He notes that these individuals are essentially caries free and concludes that starchy foods per se do not produce dental decay; sugary foods produce decay. Dietary proteins are noncariogenic because they are not broken down by salivary enzymes and thus they are not available for plaque bacteria to ferment. Hillson (1986:286), citing the work of others, states that an increase in the proportion of protein in the diet is also associated with a decrease in acidproducing bacteria in the plaque. Like protein, fats are not broken down in the mouth, and the nutrients are largely unavailable to oral bacteria. Hillson also points out that fat in the diet, while providing little or nothing for the bacteria to act on, may act to reduce caries indirectly by reducing the adherence of food deposits to plaque (Hillson 1986:286). Given the relationship between the consumption of dietary carbohydrates and dental caries, researchers have used the prevalence of the lesion, along with a consideration of patterning of sites of dental decay, to make inferences about the relative amount of carbohydrates in the diet. Numerous studies have documented an increase in dental caries and other dental pathology with increased consumption of dietary carbohydrates, particularly sucrose, in contemporary (Mayhall 1970; Price 1936) and prehistoric or historic populations (Corbett and
Page 136
Moore 1976; Moore and Corbett 1971, 1973,1975; Kelley et al. 1987; Cohen and Armelagos 1984; Hillson 1979; Larsen et al. 1991; Milner 1984; Powell 1985; Turner 1979; among others). Changes in dental caries frequency have also been used to evaluate the effects of changing subsistence strategy from foraging to agriculture. Although that subsistence shift is not of relevance to this study, nor is the inclusion of refined sugar in the diet an issue, the general pattern of increasing caries prevalence with increasing consumption of carbohydrates is a useful indicator of dietary change. Dental calculus, calcified plaque, has been considered infrequently in paleodietary studies. Nevertheless, it may be useful for making inferences about dietary consistency as well as overall oral health. Recently, analysis of food and nonfood debris adhering to the calculus has been shown to provide direct evidence of certain vegetable foods consumed in the diet. Calculus itself is "dead" material but is generally covered by a layer of active plaque (Hillson 1986). Plaque bacteria on the surface of the calculus may ferment carbohydrates and ultimately lead to the focal demineralization of the underlying substrate, but in this case it is calculus, not tooth enamel. If these episodes of low pH are insufficient to cause the calculus to redissolve, then caries are less likely to result. It is uncertain precisely what triggers calculus formation. Fibrous foods and those that are relatively abrasive exert a cleansing action on the tooth surface during chewing. Increased saliva flow also exerts a cleansing action that diminishes calculus formation (Powell 1985). Although the factors that contribute to calculus formation are complex, it seems that a diet that is refined, processed, and generally soft results in increased calculus deposits (Dr. O. Sheridan, personal communication 1996). The relationship between diet and calculus frequency, severity, and patterning is not entirely straightforward. Hillson (1979,1986) observes that there is an inverse relationship between the occurrence of dental calculus and caries. Presumably, it is because calculus can cover at least part of the surface of the tooth enamel that it apparently exerts a prophylactic effect against caries. The precise manner in which calculus formation, caries, and diet are related remains unclear, however. Excessive calculus deposits on the teeth can contribute to poor dental health, despite their apparent cariostatic role. As calculus deposits increase in size, they can lead to periodontal disease and ultimately tooth loss. The process of calculus formation has been described and strategies for recording its location and abundance on the teeth have been devised, but calculus frequency, severity, and patterning are not commonly studied. The paucity of studies, at least in archaeological contexts, may in part reflect the fact that it is uncertain precisely what triggers calculus formation and deposition and therefore it is not known whether and to what extent it may reflect dietary composition. Occasionally, food particles and debris can become trapped in or adhere to plaque and later become mineralized in the calculus. Although this is uncommon, studies have shown that plant phytoliths, pollen grains, starch granules, and other food and nonfood particles can be recovered and identified from dental calculus (e.g., Capasso et al. 1995; Dobney and Brothwell
Page 137
1986, 1987; Fox et al. 1996; Magennis and ScottCummings 1996; ScottCummings and Magennis 1997). Recovery of such food items from dental calculus can provide the opportunity to identify some of the vegetable foods that were consumed by at least some members of the population. The Site Kichpanha is a moderatesized site located in northern Belize about 35 km from the Caribbean coast (see Figure I.1). The site is located on the edge of a freshwater lake, Kate's Lagoon, which is connected to a large area of wetlands and marshes that ultimately drain into the sea. The site is situated on a low, welldrained limestone ridge that is one of the few areas of higher ground in the near vicinity. Survey and testing of the site were conducted from 1973 to 1982 (Gibson 1994; Hammond 1973); more intensive research efforts began in 1985 and were subsequently limited to five short seasons of excavation (Gibson et al. 1986; Shaw 1995, in press). Kichpanha was first occupied during the early Middle Preclassic (about 1000850 B.C.), when settlement was focused along the margins of the lake and associated wetlands (Shaw 1995). Population size at the site remained relatively modest until the Late Preclassic, when by about 100 B.C. there is a pronounced growth in the population. This rapid population growth is marked by the construction of the monumental center as well as evidence of social inequality and the establishment of an elite class by the end of the Late Preclassic (250 B.C.A.D. 250), as indicated in both burial and domestic contexts (Shaw 1995, in press). A strong presence of an elite class continued to be maintained at Kichpanha into the Early Classic (A.D. 250550). By the end of that period, Shaw (in press) argues, there was a marked decline in population and activity at the site as the elite families moved away, leaving behind a small community of agriculturalists who may have operated as a satellite of one of the regional centers. The presence of a small agricultural community at Kichpanha continued for some time during the Late Classic period, but the site, like many others in the region, was abandoned by the end of the Terminal Classic period (A.D. 700900). The natural environmental setting for Kichpanha includes lagoon, marsh, and pine ridge resources, all environments that would provide a variety of food resources. There is no evidence of raised or ditched fields associated with Kichpanha, but the aerial and ground survey necessary to document them has not yet been done. Although there is minimal arable land around Kichpanha, wetlandedge and dooryard gardens were probably used to intensify production. The diversity of environments would provide a broad array of potential food items available for consumption. Preliminary study of faunal remains from Kichpanha indicates the presence of terrestrial species such as deer, peccary, dog, armadillo, and agouti and aquatic species such as crocodilians, turtles, snails, and fish (Shaw 1995). It has also been suggested that Kichpanha participated in a regional network in which local food resources were traded (Shaw, this volume).
Page 138
Diet and Subsistence at Kichpanha: Preclassic to Late Classic Ideally, the pattern and frequency of dental pathology would be used in conjunction with other information about the diet for Kichpanha. Since the faunal and macrobotanical studies have yet to be completed at the site, we have only a very general picture of the possible dietary items, drawn largely from studies on other Lowland Maya sites. Increasingly, evidence points to diverse diets in the Lowland Maya area of Belize, at least during the Protoclassic. Macrobotanical, faunal, and pollen analyses from other contemporaneous sites in northern Belize point to the utilization not only of maize but of other cultigens as well, including chilies, beans, and squash (Cliff and Crane 1989; Crane 1986; Crane and Carr 1994; Hammond and Miksicek 1981; Miksicek 1991). The use of manioc, whether gathered from wild plants or cultivated, has been argued previously (Bronson 1966; Stone 1984), although the presence of manioc and other root crops has not been identified commonly in the Preclassic, or any other period for that matter. Recently, the presence of manioc in the botanical remains from Preclassic Cuello has been confirmed (Hather and Hammond 1994). Its importance in the diet is still uncertain, however. The utilization of tree crops, possibly as part of small dooryard gardens, also seems evident from the macrobotanical remains recovered in various archaeological contexts, including the Preclassic and Classic periods (Cliff and Crane 1989; Lentz, this volume; McKillop 1994; Miksicek 1991; Pohl and Miksicek 1985; Pohl et al. 1990). Animal protein in the diet in this region apparently comes from terrestrial mammals as well as various fish, shellfish, and reptiles from freshwater lakes and marshes. In a recent study of carbon and nitrogen isotopes in human bone collagen and apatite from the Cuello burials, Tykot and coworkers (1996) conclude that during the Preclassic, maize did not constitute the overwhelming percentage of the diet. They suggest that the relatively enriched collagen carbon isotope signature can be explained by the possible consumption of dog and armadillo, animals that likely consumed maize. Echoing the point made by Wright and White (1996), these authors also note that residents at sites in this region of Belize, though still dependent on maize, had access to a wide range of ecozones from which to obtain dietary items (Tykot et al. 1996). Whether marine fish and shellfish were a part of the diet at Kichpanha is unknown, but that protein source is important at other inland sites such as Lamanai (White and Schwarcz 1989) and certainly was an important food source at sites close to the Caribbean. Although possible food items in the diet can be enumerated, the importance of each one is unclear. Diet and subsistence during the Early Classic in the area of Kichpanha is less well known. Based on their analysis of carbon and nitrogen isotopes from bone collagen at Lamanai, White (1988) and later White and Schwarcz (1989) suggest that during the Preclassic and the Early Classic, maize was quite important and made up approximately 50 percent of the diet. During the Late Classic at Lamanai maize began to decrease in importance, and during the Terminal Classic there was a rather marked decline in maize consumption. Presumably, maize continued to be the important staple grain crop at Kichpanha, but
Page 139
whether it constituted similar proportions of the diet as determined at Lamanai is unknown. Similarly, at Pacbitun, White and coworkers (1993) suggest that maize consumption reached its peak during the Early Classic and into the Classic. It remains to be determined whether the appearance of the elite class at the end of the Preclassic and its continuation during the Early Classic at Kichpanha is accompanied by intensification of maize agriculture. Since more Classic period Maya sites have been excavated and reported, there is more known about diet and subsistence during this period in general. In a recent review of paleopathological and isotopic studies primarily of Classic period Maya skeletons, Wright and White (1996) note several trends in diet that can be observed in the Maya Lowlands. Results of the isotopic analysis of skeletons aimed at dietary reconstruction from 14 sites all generally show that Maya diets relied heavily on maize agriculture from Preclassic times on. They note, however, that there is a great deal of social, temporal, and geographic variation in the reported isotopic results, which they interpret as considerable diversity in dietary strategies. They also found that the Belizian sites show lighter carbon isotope signatures in bone collagen. They interpret this to mean that maize agriculture was less important to local Belizian diets as compared with sites farther inland in the Petén and inland areas to the south, specifically Copán. Although arguments for differences in maize consumption were traditionally advanced to explain the observed isotope variation, the authors caution that variation in fish or meat consumption may be involved in producing the resultant isotope values, even with equal maize consumption. Finally, Wright and White (1996) observe that there is a great deal of variation in diet by social status. Relative contribution of maize to the diet by status varies from place to place, apparently as does consumption of animal foods. The authors conclude that diet of the elite is "independently constituted across the Maya lowlands" (Wright and White 1996:185). Further confounding the interpretation of temporal trends in dietary change at Kichpanha is the undoubted change through time in the site's social, political, and economic relationships to other sites in the region, especially after the Early Classic, when the site is largely abandoned by the ruling elite. Now that the temporal, geographic, and social variation in diet in the Maya Lowlands has been highlighted, let us examine diet at Kichpanha. The Dental Sample and Methods The sample of dentitions and teeth used for this study included 40 adults from Protoclassic, Early Classic, and Late/Terminal Classic contexts at the site (Table 7.1).All the individuals from the Protoclassic period (N = 18) appear to be from the newly formed elite class. All but three of these burials were recovered from a single mound (Op 3003). Three of the burials from this mound included multiple individuals, a primary burial associated with one or more secondary burials. Various grave goods were also associated with these burials including numerous ceramic vessels, jade beads, marine bivalve pendants, chert bifaces, and carved bone objects (Shaw 1995, in press). One of the bone objects has a
Page 140 Table 7.1. The Kichpanha Dental Sample. Age Group
Protoclassic No. of Individuals
Early Classic No. of Individuals
Late/Terminal Classic No. of Individuals
2030
5
4
4
3040
9
5
7
4050
4
0
2
Total
18
9
13
sequence of glyphs and may be the oldest recovered to date from an excavated context (Gibson et al. 1986). The majority of burials from the Early Classic (N = 9) also represent the elite at the site, with most of them recovered from a single residential area (Op 3001). All burials from this period were single interments from burial pits and included one or two vessels and commonly marine shell ornaments. Even though the grave goods and the residential and ceremonial construction associated with the burials suggest elite status, Shaw (in press) points out that these families were not coeval with highstatus families at growing centers such as Altun Ha or Lamanai. The Late/Terminal Classic occupation at Kichpanha appears to represent a small community of agriculturalists with some status differences, but there is no clear evidence for the presence of highstatus nobility at Kichpanha at this time (Shaw 1995). Excavated burials (N = 13) associated with plaza groups indicate relatively modest households when measured against the wealth of highstatus families in large centers. In general, bone preservation at Kichpanha is poor. Attempts were made to determine sex of all the adults. In the majority of cases this could not be done except for those individuals that were at either end of the range of body and tooth size dimorphism, particularly those individuals that were large and robust and hence identifiable as male. Because of the difficulty of determining sex of most of the individuals, it cannot be considered in this analysis. Age was determined by seriating all the dentitions based on degree of attrition. The entire Kichpanha series was arranged in order of increasing dental wear, and published standards (Lovejoy 1985) were used to establish 10year age groups. The obvious difficulty with using published standards for agegroup determination is that environment, diet, and tooth use are undoubtedly different between Kichpanha and the reference population—differences that introduce an unknown bias into the estimates. There are insufficient numbers of subadults and young adults to establish molar wear gradients relevant for Kichpanha (Miles 1963,1978). Since the sample size is small, particularly that from the Early Classic, further partitioning of the sample by age is unwarranted. Obviously missing in the sample too are those individuals who experienced great amounts of antemortem tooth loss. These individuals also more than likely represent the older age groups at Kichpanha. All teeth were examined for the presence of dental calculus and caries. Calculus was scored according to a fourpoint scale modified after Dobney and Brothwell (1987). Each tooth was observed for the presence of caries lesions.
Page 141
Caries were identified as an irregular pit, sometimes darkly stained. Caries were recorded as to their location on the tooth, including occlusal, buccal fissure, smooth surface, interproximal contact facet, and those that occurred at the cementoenamel junction (CEJ). There is only one case of caries lesions in a molar fissure, no cases of interproximal contact facet lesions, and no smooth surface lesions in this series. Therefore, caries location categories were combined to include only occlusal or CEJ caries. Whenever possible, antemortem tooth loss and the occurrence of dental abscesses were recorded. It is recognized that failure to estimate antemortem tooth loss due to decay can distort the real caries rate in a population (e.g., Lukacs 1995). Because preservation of mandibular and maxillary alveolar bone was inconsistent, at best, it was not possible to control for tooth loss due to excessive decay. Therefore, the caries frequency reported here should be considered a minimum estimate. Results Caries are relatively common in the Kichpanha dental samples. As a general trend, the number of individuals with caries during the Protoclassic is moderate, but caries rate actually declines slightly during the subsequent Early Classic. During the Late/Terminal Classic the number of individuals with at least one caries lesion increases. In the Protoclassic sample only 3 of the 18 individuals did not exhibit any carious teeth. In the Early Classic 4 of 9 individuals failed to exhibit caries, and in the Late/Terminal Classic only 1 of the 13 individuals was cariesfree. When caries rate is examined on the basis of percentage of carious teeth, the same general trend is evident. First, if we look at the occurrence of at least one lesion on any tooth, regardless of location of the lesion, we see that durng the Protoclassic there is a 14.3 percent frequency of carious teeth (Table 7.2). The percentage of carious teeth declines slightly in the Early Classic (11.1 percent) but more than doubles in the Late/Terminal Classic to 28.5 percent of the teeth. The decline in caries frequency during the Early Classic is actually more dramatic than these numbers would indicate, because a single individual contributes disproportionately to the caries frequency. That individual shows lesions on 9/26 teeth, which is half the number of carious teeth in the entire Early Classic sample. If that individual is removed from the analysis, caries frequency is only 6.7 percent during the Early Classic at Kichpanha. Caries were also recorded as to their location on the tooth surface. In general, in all three periods, caries occur predominantly at the CEJ (Table 7.2). Occlusal caries decline in frequency during the Early Classic, but it must be cautioned that small sample size may be affecting the results. In general, however, based on this sample, occlusal caries decline slightly from the Protoclassic to the Late/Terminal Classic while caries at the CEJ increase slightly. Caries is age progressive, so examining lesion frequency while controlling for age is a clearer indicator of change in caries frequency. When the sample is divided into groups by age (2030,3040, and 4050 years) and culture period, the increase in caries frequency during the Late/Terminal Classic is marked, at
Page 142 Table 7.2. Caries and Calculus Frequency in the Kichpanha Dental Sample.
Protoclassic
Early Classic
Late Classic
34/238(14.3%)
18/162(11.1)
51/179(28.5%)
Occlusal
12/34(35.3%)
1/18(5.6%)
13/51 (25.5%)
Nonocclusalc
17/34 (50.0%)
16/18 (88.9%)
32/51 (62.7%)
Both
5/34(14.7%)
1/18(5.6%)
6/51 (11.8%)
Calculuse
40/228 (17.5%)
41/160 (25.6%)
82/176 (46.6%)
Relative risk estimatef
1.54255
0.56783
1.65780
Caries Anya b
d
a
Teeth with any caries lesion, regardless of its location on the tooth.
b
Teeth with caries lesions that occur on the occlusal surface only.
c
Teeth with caries lesions that occur on the neck of the tooth, almost exclusively at the cementoenamel junction. This includes mesial, distal, buccal, and lingual surfaces. d
Teeth with caries lesions that occur on the occlusal surface as well as the neck of the tooth. e
Presence of calculus without regard to quantity or location.
f
Percentage of teeth with caries among those with no calculus / Percentage of teeth with caries among those with calculus. See text for discussion. Table 7.3. Frequency of Carious Teeth by Age and Cultural Period.
Protoclassic
Age
n a/nb
Early Classic
%c
n a/nb
%c
Late/Terminal Classic n a/nb
%c
2030
7/84
8.3
7/75
9.3
23/73
31.5
3040
16/101
15.8
11/87
12.6
24/83
28.9
4050
11/53
20.8
—
—
4/23
17.4
a
Number of teeth with a caries lesion.
b
Total number of teeth observed.
c
Percentage of teeth that are carious.
least in those individuals less than 40 years of age (Table 7.3). In the younger two age groups there is little difference in caries rate between the Proto and Early Classic. Among 2030yearolds, however, caries frequency is at least double and nearly triple in the Late/Terminal Classic compared with that seen earlier. Among the 4050yearolds there is no apparent change in caries rate between the Protoclassic and Late/Terminal Classic, although sample size in this age group is small. Overall, there is an increase in caries frequency during the terminal occupation at Kichpanha that is not related to the age structure of the sample. The presence of calculus on the teeth can be indicative of dietary consistency. For this study, even though calculus was recorded on a fourpoint scale (none, trace, slight, and heavy), those teeth with only a "trace" amount of calculus were included with those teeth that showed no evidence of calculus. Since there was generally little variation in the expression of calculus on the remaining teeth, calculus is considered either present or absent. During these three periods the frequency of teeth with at least some calculus increases steadily from the Protoclassic to the Late/Terminal Classic (Table 7.2).
Page 143
The cooccurrence of calculus and caries for each period was also examined. As discussed previously, research has shown there to be an inverse relationship between caries and calculus. As the occurrence of calculus increases, caries frequency would be predicted to decline. Based on the cooccurrence of calculus and caries, a relative risk estimate for caries in the absence of calculus can be calculated. In other words, does having calculus increase or decrease the risk of having caries? The relative risk estimates are presented in Table 7.2. Relative risk values greater than 1 would mean that having no calculus is associated with an increased risk of having caries (Table 7.2). It can be seen that in both the Protoclassic and the Late/Terminal Classic, having no calculus on the tooth increases the risk of caries by a factor of slightly more than 1.5. By contrast, in the Early Classic, having no calculus actually decreases the risk of having caries. Dental Caries and Calculus: Implications for Diet at Kichpanha Results of this study can be summarized as follows. First, changes in caries prevalence in the Kichpanha dental sample are not unidirectional. From the Protoclassic to the Early Classic there is a slight decline in the frequency of dental caries. In the subsequent Late/Terminal Classic period caries prevalence increases rather markedly. The observed increase in caries frequency through time is accentuated when we control for age at death of the samples. Clearly, a more cariogenic diet is implied by the observed trend. Second, there is a steady increase in the prevalence of dental calculus from the Protoclassic to the Late/Terminal Classic at Kichpanha. This would suggest that the dietary consistency changed through time to a diet that was softer and less abrasive. Third, during the Protoclassic and the Late/Terminal Classic, having no calculus increases the risk of caries by a factor of slightly more than 1.5. This relationship is not observed during the Early Classic. Thus, in general, even though the consistency of the diet was softer, leading to greater calculus and likely increasing the risk of periodontal disease with the possible greater risk of subsequent tooth loss, the increased calculus deposition also lends protection against the development of caries. At Kichpanha during the Protoclassic only about 14 percent of the teeth are carious. During the Early Classic the frequency of caries declines even further, to approximately 11 percent of the teeth. These are not particularly high dental caries lesion frequencies for agriculturally based populations. Comparison with published summaries of caries frequency for a number of North American Late Woodland and Mississippian horticulture and agricultural populations shows that the Protoclassic and Early Classic dental lesion rates at Kichpanha fit comfortably within the lower range of reported values (Larsen et al. 1991; Milner 1984). In the Maya Lowlands White (1988,1994) reports a frequency of caries for a small Preclassic dental sample at Lamanai of around 20 percent. This increases slightly during the Early Classic to around 24 percent of the teeth. There is a slight decline in caries frequency during the Late Classic (approximately 18 percent of the teeth) but a dramatic drop in caries frequency during the Terminal Classic (only about 1.8 percent of the teeth), reflecting a
Page 144
change in diet interpreted as a marked decline in maize consumption (White 1994; White and Schwarcz 1989). The relatively low frequencies of dental caries at Kichpanha during the Protoclassic and Early Classic suggest that sugary carbohydrates such as maize may not have been the dominant component of the diet. The caries frequencies during these two periods instead suggest consumption of a mixed diet including maize as well as other plant foods and a variety of animal protein sources. At Lamanai, White and Schwarcz (1989) argue for the possibility of backdoor gardens or a multispecies horticulture model of food production to explain the chemical data and low frequency of dental caries during the Terminal Classic (White and Schwarcz 1989:468). This dietary model may also characterize Kichpanha during the Protoclassic and Early Classic in light of the relatively low caries frequency, although those frequencies are not as low as that reported for Lamanai. At Kichpanha the rather sharp increase in caries frequency during the Late/Terminal Classic is in marked contrast to that reported at Lamanai (White 1988, 1994). This increase in caries frequency indicates that refined carbohydrates were more important in the diet than in previous periods. In particular, these data suggest that maize constituted a greater proportion of the diet at Kichpanha. It has also been suggested that increased consumption of other starchy foods in the diet, specifically root crops such as manioc, may have contributed to increased caries frequencies (e.g., White 1988). Although this is an attractive suggestion, manioc is a starch, and starchy foods do not necessarily elevate caries lesion frequencies, at least as far as studies of contemporary populations suggest (Newbrun 1982) and as reviewed above. Preliminary results of a study of plant remains and other debris observed in dental calculus samples from Kichpanha indicate that maize starch granules are the most abundant single remain from each of the three periods: Protoclassic, Early Classic, and Late/Terminal Classic (Magennis and ScottCummings 1996). Contrary to expectations, no phytoliths of maize paleas or lemmas were identified. Phytoliths from Palmae were noted in calculus from Protoclassic and Early Classic individuals. Also common were phytolith forms from a grass other than maize, most likely a C3 or festucoid grass. Whether this grass was consumed as a dietary item or whether the phytoliths became embedded in the dental calculus as a consequence of working or holding fibers in the teeth is unknown. Also identified was pollen from Cheno Am in individuals from the Protoclassic. Finally, manioc starch granules were provisionally identified in calculus samples from each of the three periods. Although this study is ongoing, these preliminary results show that a variety of food items were apparently included in the diet. One difficulty with interpreting these results is that although specific food items may be identified, it is uncertain how important any of them are in the total diet. Overall, the patterning and frequency of caries and dental calculus in the Kichpanha skeletal series suggest that diet changed over time at the site. The modest caries frequency observed in the Protoclassic sample most likely reflects a mixed diet. Maize was undoubtedly important but was not the major
Page 145
part of the whole diet. This is also reflected by the variety of edible vegetable phytoliths, pollen, and starch granules identified in the dental calculus. It also seems clear that carbohydrate consumption, probably maize, increased at Kichpanha during the Late/Terminal Classic, as reflected in the approximate doubling of the caries frequency. This shift in diet is apparently accompanied by a change in dietary consistency to one that is softer, given the increased frequency and severity of dental calculus observed in the sample. It may be that the increase in calculus reflects not so much a change in the proportion of various food types in the diet but instead a change in the maize processing techniques. Presumably, phytate removal from the maize would result in smaller, smoother, finer flour particles. This softer carbohydrate would fail to cleanse the teeth with abrasive action and would more easily adhere to the tooth surfaces. RuggGunn (1981) also suggests that phytate is cariostatic, and the increase in caries during the Late/Terminal classic may also reflect that change in maize processing technique. White and Schwarcz (1989) suggest that at Lamanai phytate removal from maize was either initiated or more effective beginning in the Postclassic and continued thereafter. Whether maize processing techniques changed at Kichpanha during the Late and Terminal Classic remains to be explored. Conclusions Analysis of dental caries and calculus can yield insights into diet and dietary consistency. Results of this study show there to be an increase in caries frequency from a rather modest 14 percent of the teeth during the Protoclassic to a frequency double that during the Late/Terminal Classic at Kichpanha. The increased frequency of dental calculus signals a dietary change, specifically an increase in maize or other carbohydrate consumption and possibly a concomitant change in maize processing techniques. Alternatively, there may simply have been a change in dietary consistency, although a consideration of calculus frequency together with caries argues for dietary change as well. Another hypothesis that remains to be tested is whether the observed change in caries frequency reflects dietary change per se or socially determined differences in diet. At Kichpanha the presence of an elite class during the Protoclassic and through the Early Classic seems evident, and that elite class is reflected in the skeletal sample from those two periods. At the end of the Early Classic, however, Kichpanha was largely depopulated as the elite families moved away and a small agricultural community remained at the site. It has been argued that the community of agriculturalists remaining at Kichpanha may have operated as a satellite of one of the regional centers. The modest caries frequency during the Protoclassic and Early Classic may indicate elite access to a varied diet, and the increase in caries frequency during the late phases of the site's occupation may reflect a more narrow diet of those who lack access to the powerful elite diet of the major centers in the region. Although the observed patterns of dental pathological conditions reflect dietary change, these data considered alone still leave many questions unanswered.
Page 146
The interpretation of caries and calculus data is strengthened when considered in the context of known subsistence practices and food resource potential for a given site and region. The integration of specific theory, methods, and analysis of pollen, phytolith, macrobotanical remains, animal bone, aquatic vertebrate and invertebrate remains, human skeletal remains, soils, and stratigraphy needs to be a continuing priority in future Maya archaeology. Ongoing analyses of other skeletal pathologies indicative of nutritional adequacy and disease stress are under way on the Kichpanha series and will add to the interpretative potential of the analyses already completed. Taken together, all these data are essential for dietary reconstruction and understanding the adaptation of past human populations. Acknowledgments I would like to thank Leslie Shaw, project director, for inviting me to participate in the excavations and skeletal analysis at Kichpanha. The late Francis Meskill provided invaluable assistance with contextual information and ceramic identification critical to the skeletal analysis. Mike Lacy provided valuable guidance in the relative risk assessment for this chapter. I would also especially like to thank the late Harriot Topsey, Department of Archaeology of Belize, for his continued support of the Kichpanha Project. Financial support for the Kichpanha Project from the Andover Foundation for Archaeological Research, Andover, Mass., is greatly appreciated. I am also indebted to the Department of Anthropology and the College of Liberal Arts at the Colorado State University for monetary support. References Cited Bronson, B. (1966) Roots and the subsistence of the ancient Maya. Southwestern Journal of Anthropology 22:251279. Calcagno, J. M., and Gibson, K. R. (1988) Human dental reduction: Natural selection or the probable mutation effect. American Journal of Physical Anthropology 77:505517. Capasso, L.; DiTota, G.; Jones, K. W.; and Tuniz, C. (1995) Synchrotron radiation microprobe analysis of human dental calculi from an archaeological site: A new possible perspective in palaeonutrition studies. International Journal of Osteoarchaeology 5:282288. Cliff, M. B., and Crane, C. J. (1989) Changing subsistence economy at a Late Preclassic Maya community. In P. A. McAnany and B. L. Isaac (eds.): Prehistoric Maya Economics of Belize. Research in Economic Anthropology, Supplement 4. Greenwich, Conn.: JAI Press, pp. 295324. Cohen, M. N., and Armelagos, G. J. (eds.) (1984) Paleopathology at the Origins of Agriculture. Orlando, Fla.: Academic Press. Corbett, M. E., and Moore, W. J. (1976) The distribution of dental caries in ancient British populations: The nineteenth century. Caries Research 10:401412. Crane, C. J. (1986) Late Preclassic Maya agriculture, wild plant utilization, and landuse practices. In R. A. Robertson and D. A. Freidel (eds.): Archaeology at Cerros, Belize, Central America. Vol. 1 and Interim Report. Dallas: Southern Methodist University Press, pp. 147151. Crane, C. J., and Carr, H. S. (1994) The integration and quantification of economic data from a Late Preclassic Maya community in Belize. In K. D. Sobolik (ed.): Paleonutri
Page 147
tion: The Diet and Health of Prehistoric Americans. Center for Archaeological Investigations, Occasional Paper No. 22. Carbondale: Southern Illinois University, pp. 6679. Dobney, K., and Brothwell, D. (1986) Dental calculus: Its relevance to ancient diet and oral ecology. In E. Cruwys and R. A. Foley (eds.): Teeth and Anthropology Oxford: British Archaeological Reports International Series 291, pp. 5581. Dobney, K., and Brothwell, D. (1987) A method for evaluating the amount of dental calculus on teeth from archaeological sites. Journal of Archaeological Science 14:343351. Fedick, S. L. (1994) Ancient Maya agricultural terracing in the upper Belize River area. Ancient Mesoamerica 5:107127. Fox, C. L.; Juan, J.; and Albert, R. M. (1996) Phytolith analysis on dental calculus, enamel surface, and burial soil: Information about diet and paleoenvironment. American Journal of Physical Anthropology 101:101113. Gibson, E. C. (1994) Archeology at Kichpanha, a minor ceremonial center in northern Belize (19811983). In T. R. Hester, H. J. Shafer, and J. D. Eaton (eds.): Continuing Archeology at Colha, Belize. Studies in Archeology, No. 16. Austin: Texas Archeological Research Laboratory, University of Texas, pp. 277284. Gibson, E. C.; Shaw, L. C.; and Finamore, D. R. (1986) Early Evidence of Maya Hieroglyphic Writing at Kichpanha, Belize. Working Papers in Archaeology, No. 2. Center for Archaeological Research, University of Texas at San Antonio. Goodman, A. H.; Martin, D. L.; Armelagos, G. J.; and Clark, G. A. (1984) Indications of stress from bone and teeth. In M. N. Cohen and G. J. Armelagos (eds.): Paleopathology at the Origins of Agriculture. Orlando, Fla.: Academic Press, pp. 1349. Hammond, N. (1973) British MuseumCambridge University Corozal Project 1973 Interim Report. Cambridge: Center for Latin American Studies, Cambridge University. Hammond, N., and Miksicek, C. H. (1981) Ecology and economy of a formative Maya site at Cuello, Belize. Journal of Field Archaeology 8:259269. Harrison, P. D., and Turner, B. L. II (eds.) (1978) PreHispanic Maya Agriculture. Albuquerque: University of New Mexico Press. Hather, J. G., and Hammond, N. (1994) Ancient Maya subsistence diversity: Root and tuber remains from Cuello, Belize. Antiquity 68:330335. Healy, P. F.; Lambert, J. D. H.; Aranson, J. T; and Hebda, R. J. (1983) Caracol, Belize: Evidence of ancient Maya agricultural terraces. Journal of Field Archaeology 10:397410. Hildebolt, C. F.; ElvinLewis, M.; Molnar, S.; McKee, J. K.; Perkins, M. D.; and Young, K. L. (1989) Caries prevalences among geochemical regions of Missouri. American Journal of Physical Anthropology 78:7992. Hildebolt, C. F.; Molnar, A.; ElvinLewis, M.; and McKee, J. K. (1988) The effect of geochemical factors on prevalences of dental diseases for prehistoric inhabitants of the state of Missouri. American Journal of Physical Anthropology 75:114. Hillson, S. W. (1979) Diet and dental disease. World Archaeology 11:147162. Hillson, S. W. (1986) Teeth. Cambridge: Cambridge University Press. HussAshmore, R.; Goodman, A. H.; and Armelagos, G. J. (1982) Nutritional inference from paleopathology. In M. B. Schiffer (ed.): Advances in Archeological Method and Theory, vol. 5. New York: Academic Press, pp. 395474. Kelley, M. A.; Barrett, T. G.; and Saunders, S. D. (1987) Diet, dental disease, and transition in northeastern Native Americans. Man in the Northeast 33:113125. Larsen, C. S. (1987) Bioarchaeological interpretations of subsistence economy and behavior from human skeletal remains. In M. B. Schiffer (ed.): Advances in Archaeological Method and Theory, vol. 10. New York: Academic Press, pp. 339445. Larsen, C. S.; Shavit, R.; and Griffin, M. C. (1991) Dental caries evidence for dietary change: An archaeological context. In M. A. Kelley and C. S. Larsen (eds.): Advances in Dental Anthropology. New York: WileyLiss, pp. 179202. Lovejoy, C. O. (1985) Dental wear in the Libben population: Its functional pattern and
Page 148
role in the determination of adult skeletal age at death. American Journal of Physical Anthropology 68:4756. Lukacs, J. R. (1995) The ''caries correction factor": A new method for calibrating dental caries rates to compensate for antemortem loss of teeth. International Journal of Osteoarchaeology 5:151156. McKillop, H. (1994) Ancient Maya tree cropping: A viable subsistence adaptation for the island Maya. Ancient Mesoamerica 5:129140. Magennis, A. L., and ScottCummings, L. S. (1996) A record of food and grit in human dental calculus at Kichpanha, Belize. American Journal of Physical Anthropology, Suppl. 22:155 (abstract). Mayhall, J. T (1970) The effect of culture change upon the Eskimo dentition. Arctic Anthropology 7:117121. Miksicek, C. H. (1991) The ecology and economy of Cuello. In N. Hammond (ed.): Cuello: An Early Maya Community in Belize. Cambridge: Cambridge University Press, pp. 7097. Miles, A. E.W. (1963) The dentition and its assessment of individual age in skeletal material. In D. R. Brothwell (ed.): Dental Anthropology. New York: Pergamon Press, pp. 191209. Miles, A. E. W. (1978) Teeth as an indicator of age in man. In P. M. Butler and K. A. Josey (eds.): Development, Function, and Evolution of Teeth. New York: Academic Press, pp. 455464. Milner, G. R. (1984) Dental caries in the permanent dentition of a Mississippian period population from the American Midwest. Collegium Antropologicum 8:77 91. Moore, W. J., and Corbett, M. E. (1971) The distribution of dental caries in ancient British populations: AngloSaxon period. Caries Research 5:151161. Moore, W. J., and Corbett, M. E. (1973) The distribution of dental caries in ancient British populations: Iron Age, Romans, British, and Medieval periods. Caries Research 7:139151. Moore, W. J., and Corbett, M. E. (1975) The distribution of dental caries in ancient British populations: The seventeenth century. Caries Research 9:163174. Newbrun, E. (1982) Sugar and dental caries: A review of human studies. Science 217:418423. Pohl, M. D.). (1990) The Rio Hondo project in Northern Belize. In M. D. Pohl (ed.): Ancient Maya Wetland Agriculture: Excavations on Albion Island, Northern Belize. Boulder, Colo.: Westview Press, pp. 119. Pohl, M. D.; Bloom, P. R.; and Pope, K. O. (1990) Interpretation of wetland farming in northern Belize: Excavations at San Antonio Rio Hondo. In M. D. Pohl (ed.): Ancient Maya Wetland Agriculture: Excavations on Albion Island, Northern Belize. Boulder, Colo.: Westview Press, pp. 187254. Pohl, M. D., and Miksicek, C. H. (1985) Cultivation techniques and crops. In M. Pohl (ed.): Prehistoric Lowland Maya Environment and Subsistence Economy. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 77. Cambridge: Harvard University, pp. 920. Powell, M. L. (1985) The analysis of dental wear and caries for dietary reconstruction. In R. I. Gilbert, Jr., and J. H. Mielke (eds.): The Analysis of Prehistoric Diet. Orlando, Fla.: Academic Press, pp. 307338. Price, W. A. (1936) Eskimo and Indian field studies in Alaska and Canada. Journal of the American Dental Association 23:417437. Reed, D. M. (1994) Ancient Maya diet at Copán, Honduras, as determined through the analysis of stable carbon and nitrogen isotopes. In K. D. Sobolik (ed.): Paleonutrition: The Diet and Health of Prehistoric Americans. Center for Archaeological Investigations, Occasional Paper No. 22. Carbondale: Southern Illinois University, pp. 210221. Rowe, N. H. (1975) Dental caries. In P. F. Steele (ed.): Dimensions of Dental Hygiene. 2d ed. Philadelphia: Lea and Febiger, pp. 198222.
Page 149
RuggGunn, A. J. (1981) Diet and dental caries. Frontiers in Oral Physiology 3:5365. ScottCummings, L., and Magennis, A. (1997) A phytolith and starch record of food and grit in Mayan human tooth tartar. In A. Pinilla, J. JuanTresserras, and M. J. Machado (eds.): The StateoftheArt of Phytoliths in Soils and Plants. Monograph 4. Madrid: Centro de Ciencias Medioambientales, pp. 211218. Schneider, K. N. (1986) Dental caries, enamel composition, and subsistence among prehistoric Amerindians of Ohio. American Journal of Physical Anthropology 71:95102. Shaw, L. C. (1995) Boom and bust: The growth and decline of Kichpanha, Belize. Paper presented at the 1st International Symposium on Maya Archaeology, San Ignacio, Belize. Shaw, L. C. (In press) Kichpanha: Excavations at a moderatesized Maya center in Belize. In Proceedings of the First International Maya Conference. Belmopan, Belize: Belize Department of Archaeology. Stone, D. (1984) PreColumbian migration of Theobroma cacao Linnaeus and Manihot esculenta Crantz from northern South America into Mesoamerica: A partially hypothetical view. In D. Stone (ed.): PreColumbian Plant Migration. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 76. Cambridge: Harvard University, pp. 6783. Storey, R. (1992a) Patterns of susceptibility to dental defects in the deciduous dentition of a Precolumbian skeletal population. In A. H. Goodman and L. L. Capasso (eds.): Recent Contributions to the Study of Enamel Developmental Defects. Chieti, Italy: Edigrafital Teramo. Journal of Paleopathology, Monographic Publications, No. 2, pp. 171183. Storey, R. (1992b) The children of Copán: Issues in paleopathology and paleodemography. Ancient Mesoamerica 3:161167. Turner, C. G. II (1979) Dental anthropological indications of agriculture among Jomon people of central Japan. American Journal of Physical Anthropology 51:619636. Turner, B. L. II, and Harrison, P. D. (1983) Pulltrouser Swamp and Maya raised fields: A summation. In B. L. Turner II and P. D. Harrison (eds.): Pulltrouser Swamp: Ancient Maya Habitat, Agriculture, and Settlement in Northern Belize. Austin: University of Texas Press, pp. 246289. Tykot, R. H.; van der Merwe, N. J.; and Hammond, N. (1996) Stable isotope analysis of bone collagen, bone apatite, and tooth enamel in the reconstruction of human diet: A case study from Cuello, Belize. In M. V. Orna (ed.): Archaeological Chemistry, 5. Washington, D.C.: American Chemical Society, pp. 355365. White, C. D. (1988) Diet and health in the ancient Maya at Lamanai, Belize. In B. V. Kennedy and G. M. LeMoine (eds.): Diet and Subsistence: Current Archaeological Perspectives. Calgary: University of Calgary Archaeological Association, pp. 288296. White, C. D. (1994) Dietary and dental pathology and cultural change in the Maya. In A. Herring and L. Chan (eds.): Strength in Diversity: A Reader in Physical Anthropology. Toronto: Canadian Scholar's Press, pp. 279302. White, C. D.; Healy, P. F.; and Schwarcz, H. P. (1993) Intensive agriculture, social status, and Maya diet at Pacbitun, Belize. Journal of Anthropological Research 49:347375. White, C. D., and Schwarcz, H. P. (1989) Ancient Maya diet: As inferred from isotopic and chemical analysis of human bone. Journal of Archaeological Science 16:451474. White, C. D.; Wright, L. E.; and Pendergast, D. M. (1994) Biological disruption in the Early Colonial period at Lamanai. In C. S. Larsen and G. Milner (eds.): In the Wake of Contact: Biological Responses to Conquest. New York: WileyLiss, pp. 135145. Whittington, S. L. (1992) Enamel hypoplasia in the lowstatus Maya population of Prehispanic Copán, Honduras. In A. H. Goodman and L. L. Capasso (eds.): Recent Contributions to the Study of Enamel Developmental Defects. Chieti, Italy: Edigrafital Teramo. Journal of Paleopathology, Monographic Publications, No. 2, pp. 185205.
Page 150
Wiseman, F. M. (1983) Subsistence and complex societies: The case of the Maya. In M. B. Schiffer (ed.): Advances in Archaeological Method and Theory, vol. 8. New York: Academic Press, pp. 143189. Wright, L. E. (1990) Stresses of conquest: A study of Wilson bands and enamel hypoplasias in the Maya of Lamanai, Belize. American Journal of Human Biology 2:2535. Wright, L. E. (1997) Biological perspectives on the collapse of the Pasión Maya. Ancient Mesoamerica 8:267273. Wright, L. E., and White, C. D. (1996) Human biology in the classic Maya collapse: Evidence from paleopathology and paleodiet. Journal of World Prehistory 10:147198.
Page 151
Chapter 8 Caries and Antemortem Tooth Loss at Copán Implications for Commoner Diet Stephen L. Whittington One of the major Classic period Maya centers was located at Copán, in western Honduras. The city apparently was involved in whatever series of events caused the collapse of Classic Maya civilization throughout much of the southern Lowlands. The last dated monument at Copán was erected in the early ninth century (Schele and Freidel 1990), and it appears that centralized political authority collapsed soon thereafter. The abandonment of the Copán Valley was a long process, however, and did not culminate until after A.D. 1200 (Webster and Freter 1990). Rue (1987) and Abrams and Rue (1988) interpret evidence from pollen cores to mean that environmental degradation was associated with the collapse at Copán. Paleodemographic and paleopathological studies based on the skeletons of commoners also seem to support an environmental basis for the collapse at Copán. Significant demographic changes, reflected in the sample of commoner skeletons, were occurring around the time of the collapse (Whittington 1991). High frequencies of skeletal indicators of subadult stress and anemia in commoner skeletons are also suggestive (Whittington 1992; Whittington and Reed 1997). This chapter describes dental caries and antemortem tooth loss among members of the lowest socioeconomic level of Maya society at Copán around the time of the Classic collapse. Aspects of commoner diet at that time can be inferred from caries and tooth loss. The lowstatus segment of Maya society frequently has been ignored by archaeologists, despite the fact that most of the population belonged to it. Only during the last two decades have archaeological studies concentrated specifically on commoners (e.g., Webster and Gonlin 1988). In general, the advantage to studying commoners is that they grew the food and performed the labor that supported Maya civilization and its relatively small number of elite individuals. They presumably were not buffered from stresses by redistribution networks and reciprocal arrangements the way the elite were, and their health should, therefore, reflect societal and environmental conditions more closely than that of the elite. The advantage to studying commoners at Copán in particular is that extensive archaeological work at their households (Webster and Freter 1990) has produced a relatively large, internally heterogeneous sample, whose analysis has the potential to expose important factors related to the collapse at Copán and elsewhere.
Page 152
Dental Caries and Antemortem Loss Dental caries is an infectious, transmissible disease that involves the progressive destruction of tooth structure initiated by the activity of microbes on the tooth surface (Pindborg 1970). Susceptibility of teeth to caries depends, among other things, on morphology and location in the dental arch. Tooth surfaces that are not well cleansed by saliva flow and the mechanical actions of the tongue and cheeks are more susceptible to plaque formation, which is a prerequisite for development of caries (Powell 1985). Posterior teeth are more prone to be carious than anterior teeth, and maxillary teeth are more frequently attacked than mandibular teeth (Patterson 1984). Caries has been found to be positively correlated with dietary carbohydrate intake, but the amount of carbohydrates consumed is a less important factor in caries incidence than are consistency and frequency (Bibby 1961; Powell 1985). Refined carbohydrates and cooked foods tend to be stickier and to remain on the teeth longer than unrefined and raw foods, thereby exposing the enamel to attack from caries for much longer. Except with highly retentive foods, however, caries attack is more closely related to the frequency with which carbohydrates enter the mouth than to the total amount consumed. Overall caries susceptibility is related to various individual factors. Age affects the incidence of carious lesions. Caries is a disease of children and young adults in most populations (Carlos and Gittelsohn 1965). Although older individuals have their teeth exposed longer to the conditions likely to produce lesions, older enamel is more resistant than newly erupted enamel to dissolution by bacterial acids (Powell 1985). Patterson (1984) cites various clinical studies showing that caries incidence normally is higher for males than for females in deciduous teeth, and higher for females than for males in permanent teeth, especially during adolescence. Environmental and cultural factors also influence overall caries susceptibility. Incidence of caries is affected by the amount of environmental stress to which an individual is exposed, since stress can cause defective tooth structure to be produced, which is more susceptible to caries (Brown 1981; Wing and Brown 1979). Caries occurs earlier in lower socioeconomic groups than in higher ones (Steggerda and Hill 193536). Agriculturalists generally have more carious teeth than hunter gatherers or those relying on mixed subsistence (Patterson 1984; Turner 1979). Antemortem tooth loss is often, but not always, the result of caries exposing the pulp chamber of the tooth. Once the pulp is exposed, there is high risk of infection, which can result in abscessing and destruction in alveolar bone, exfoliation of the tooth, and eventual filling in and repair of the affected bone (Ortner and Putschar 1985). Rapid, heavy dental attrition also can expose the pulp chamber and lead to abscessing and tooth loss, but this generally is a problem in older individuals and in populations consuming large amounts of grit along with food, whereas caries causes much antemortem loss in early adulthood (Brown 1981; Ortner and Putschar 1985). Periodontal disease related to mechanical irritation, infection, or tissue breakdown of scurvy is another cause of tooth loss (Saul 1972).
Page 153
Materials The skeletons of 163 individuals excavated at Copán during the 1970s and 1980s were determined to be members of the lowstatus segment of the population. Definition of status was based on the type of site in which an individual was buried, since archaeological and ethnographic evidence indicates that Maya were typically buried beneath, immediately around, or near their place of residence during life (Haviland 1972; Ruz 1965; Willey 1965; Wisdom 1940). At the time of the collapse the Copán polity consisted of a regalritual center with palaces and monumental architecture known as the Main Group, surrounded by urban zones of densely packed residential compounds that, in turn, were surrounded by rural residential compounds scattered throughout the remainder of the Copán Valley. Although the entire valley comprised a single political entity, archaeologists typically treat each compound as an individual site. Willey and Leventhal (1979) defined Type I through Type IV residential sites, which they equated with different socioeconomic status levels. Type I sites have one or two small plazas surrounded by three to five mounds, 0.25 to 1.25 m high, constructed of earth fill and small to mediumsized stone rubble. Type II sites have as many as six or eight mounds up to 2.5 to 3.0 m high surrounding one or two plazas. Stone, including some dressed blocks, is more abundant. Willey and Leventhal suggested that Type I sites were commoner households and Type II sites belonged to lowlevel elites. Webster and Freter (1990) later modified the Willey and Leventhal scheme. They defined a new class of site, Aggregate, having two or more mounds not organized around a plaza but probably functionally equivalent to Type I sites. They also suggested that the line separating commoner from elite households actually lies between Type II and even larger Type III sites. A definitive study of the social aspects of burials still has not been completed for Copán, but it is apparent that many factors influenced quality and quantity of associated goods and treatment. The situation within each residential compound is complex, with burial characteristics reflecting each individual's sex, age, and, probably, closeness of relationship to the head of the compound. Viel and Cheek (1983) also reported a relationship with archaeological phase. Given such variability, the type of site in which burial occurred appears to be a more accurate gauge of overall status within Copán society than burial goods and treatment. In the present analysis, individuals buried in Aggregate, Type I, Type II, and nonmound sites are considered commoners. Burials from Aggregate and Type I sites are grouped together because the sites are similar in size and probably function. Type I/Aggregate sites are analyzed separately from Type II sites to confirm that no major differences in lifestyle, as reflected in pathological lesions, exist. Skeletons from urban locations came from the crowded residential zones surrounding the Main Group, which have a structure density of 1484/km2. Outside these zones structure density decreases to between 30/km2 and 139/km2 and settlement takes on a rural character (Webster and Freter 1990).
Page 154
All skeletons for which temporal parameters can be determined are from one of three archaeological phases, as defined by Webster and Freter (1990). The Acbi phase (A.D. 400700) was a period of population growth and the development of Copán's political power. The early Coner phase (700800) was when the population growth rate peaked and population size was approaching its maximum just before the collapse of centralized political power. The late Coner phase (post 800) was after political collapse when population size began to decrease (Webster et al. 1992). Skeletons whose burial context indicates that they were from the Coner phase, but which lack specific indications of being from the early Coner phase, are by default assigned to the late Coner phase, which lasted much longer. The skeletal remains of 148 of the 163 lowstatus individuals were available for osteological analysis (Table 8.1). Detailed studies of 145 individuals and cursory examinations of 3 more produced data on pathological lesions, age at death, and sex, which are the basis of the present study. Problems with the sample include poor bone preservation, evidence that the ageatdeath distribution does not accurately reflect some aspects of the structure of the living population, and lack of precision for some contextual data, especially phase. Some subpopulations are represented by only a few individuals, whereas others are represented by relatively many, meaning that groups with variances of different sizes are being compared in many cases. Despite these limitations, there is value in analyzing this sample in detail because of its potential to reveal important factors about the Classic collapse not obtainable by other means. Methods Evidence of caries was recorded for all of each individual's tooth fragments. At the same time, the amount preserved of each tooth was recorded. A tooth was scored as not having caries only when 100 percent of the crown was preserved and no lesion was present. If less than 100 percent was preserved and no lesion was present, the tooth was considered to have missing data for caries. A tooth was scored as being lost antemortem only if there was no evidence of an unremodeled socket in the alveolar bone at the tooth's normal location. For simplification, tooth classes are used for most analyses. Exploratory data analysis validates use of tooth classes rather than individual teeth. Caries frequencies at Copán are similar in the mandible and maxilla, as well as in the left and right sides of the dental arcade (Whittington 1989). Although loss appears to occur somewhat earlier and more frequently in the mandible, particularly in the posterior teeth, there is no difference between the sides of the mouth. Use of tooth classes results in some loss of precision in defining patterns but produces more comprehensible patterns. Each individual was assigned a single age at death so that interactions between presence of pathological lesions and age could be explored. This single age was based for subadults on dental development and tooth eruption (Ubelaker 1978) or size seriation (Storey 1986). For adults, it was based on auricular surface seriation (Lovejoy et al. 1985), tooth wear (Molnar 1971), or rough field
Page 155 Table 8.1. Characteristics of Copán LowStatus Skeletons Studied by the Author (N = 148).
Site type
N
Type I/Aggregate
48
Type II
93
Nonmound or unknown
7
Site location Rural
41
Urban
107
Phase Acbi
11
Early Coner
17
Late Coner
103
Unknown
17
Sex Male
42
Female
45
Unknown
61
or lab estimations when nothing else was available. Simple or weighted averages of ranges derived from the various methods were taken to produce a single ageat death estimate for each individual. As age at death for different individuals is based on different criteria, each having its own inherent degree of error, it is not possible to characterize the overall imprecision in the estimates for the entire sample. For some analyses, individuals are grouped into broad age classes roughly corresponding to subadults (younger than 20 years of age), young adults (20 to 34), and old adults (35 and older). The Wilcoxon rank sum test (Snedecor and Cochran 1967) is used to compare groups based on particular traits for statistically significant differences in age at death. Sex was determined through both osteological analysis and multivariate statistics. Physical traits of the pelvis and cranium (Bass 1971) were used when possible to sex individuals. When visual determination of sex was impossible, discriminant functions similar to those published by Ditch and Rose (1972) and Giles (1970), but tailored to this population, were applied to bone and tooth measurements. Loglinear models of independence (Knoke and Burke 1980) were constructed for multiway contingency tables to uncover significant interactions between caries frequencies and subpopulations based on site type (Type I/Aggregate vs. Type II), location (rural vs. urban), phase (Acbi vs. early Coner vs. late Coner), or sex. Four separate fourfactor contingency tables based on caries data were analyzed with the BMDP statistical software package (Dixon 1985). The factors in each contingency table were subpopulation, tooth class,
Page 156
age class, and presence/absence of caries. Various researchers have used loglinear models to describe relationships between variables for anthropological data (e.g., Burns 1979; Muller and Mayhall 1971). Other robust statistical techniques that require few assumptions about sample composition, such as calculation of the Pearson coefficient of correlation, are used to extract the maximum amount of information; to separate significant from nonsignificant patterns related to age, sex, time period, and burial location; and to strengthen the conclusions that can be drawn. Statistical analysis is important for any study of ancient disease, but especially for one that concerns subpopulations represented by few individuals and a poorly preserved sample, as at Copán. Results Caries appears in at least one permanent tooth in 68.2 percent of individuals (58 of 85) having any erupted permanent teeth preserved and in at least one deciduous tooth in 43.5 percent of subadults (10 of 23) having any erupted deciduous teeth preserved. Antemortem loss occurs in 37.2 percent of individuals (32 of 86) with either erupted permanent teeth or alveolar bone preserved. These figures give only rough estimates of the frequency of caries and antemortem loss, since individuals of widely differing ages are included, without regard for the number or state of teeth present or the condition of the alveolar bone. Even so, it is apparent both that antemortem loss is a less common condition than caries and that caries is more common in adults than subadults in the lowstatus population. Frequencies of caries and teeth lost antemortem are presented in Table 8.2. All caries frequencies are higher than corresponding loss frequencies, except for permanent incisors. Antemortem loss is most frequent in the molars and decreases sequentially in the incisors, premolars, and canines. Caries and antemortem loss frequencies are broken down by tooth class and age class in Figure 8.1. Old individuals have higher frequencies of caries and tooth loss than young individuals in all tooth classes with one exception: subadult caries frequency is higher than that for young adults in incisors. With increasing age, the area of highest caries frequency shifts from the front to the back of the mouth. At all ages loss occurs at approximately equivalent frequencies in incisors and molars and at lower frequencies in premolars and especially canines. Caries frequencies exceed loss frequencies for all tooth classes and at all ages except for old adult incisors. Average age at death for individuals with and without caries and antemortem loss in each left mandibular tooth is plotted in Figure 8.2, where teeth are arranged from left to right in order of eruption. Exploratory data analysis indicates that patterns are similar in each quadrant of the mouth, although more age differences are statistically significant for the left mandible than for any other quadrant (Whittington 1989). To maximize available data, average age for individuals with caries in a tooth, regardless of its completeness, is compared to average age for individuals
Page 157 Table 8.2. Caries and Antemortem Loss Frequencies for Copán Commoners. Caries
Na
Loss
Nb
Permanent incisors
0.077
272
0.115
383
Permanent canines
0.142
183
0.030
235
Permanent premolars
0.225
346
0.062
451
Permanent molars
0.226
394
0.130
579
All permanent
0.179
1195
0.093
1648
Deciduous incisors
0.209
67
Deciduous canines
0.000
Deciduous molars All deciduous
Teeth
16
0.129
70
0.150
153
a
Complete, erupted, identifiable teeth only.
b
Including partial, erupted, identifiable teeth, or tooth locations in jaws.
Figure 8.1. Frequencies of caries and antemortem loss in tooth classes for Copán commoners. Solid shapes represent caries, and open shapes represent loss. Circles are subadults (< 20 years old), squares are young adults (2034 years old), and diamonds are old adults (35+ years old).
with that tooth intact and no caries. Average age for individuals with caries is higher than average age for individuals without caries for all teeth except the canine. The pattern for the canine is probably a random fluctuation without statistical importance. Statistically significant differences, according to the Wilcoxon rank sum test, occur for the first premolar, second molar, and third molar (p < .005). For individuals aged 20 or older, average age is higher for those with antemortem loss than for those without loss for all teeth. The Wilcoxon rank sum test shows significant differences for the first, second, and third molars (p < 0.05). Average age is higher for individuals with tooth loss than for individuals with caries for all teeth except the second and third molars. None of the differences in average age is statistically significant according to the Wilcoxon rank sum test, however, which suggests that the age patterns for caries and loss are quite similar in the population of commoners. For this reason, further results are summarized only for caries.
Page 158
Figure 8.2. Average age at death for Copán commoners with caries and antemortem loss in teeth of the left mandible. Solid shapes represent caries, and open shapes represent loss. Presence of caries or loss is shown by circles, and absence is shown by squares. Teeth are listed in order of eruption.
Fourfactor contingency tables based on caries data and their complete loglinear analysis appear in Whittington (1989), and only results pertaining to relationships between caries and subpopulation are summarized here. One significant interaction for the contingency table in which subpopulation is based on phase is between phase, age, and presence of caries (Figure 8.3). Early Coner subadults and young adults are associated with absence of caries, and old adults are associated with presence. Late Coner subadults are associated with absence of caries, and young and old adults are associated with presence. Acbi young adults are associated with presence of caries, and old adults are associated with absence. Acbi subadults are associated with presence of caries in the model, a paradoxical result that is probably an artifact of the total absence of Acbi subadult teeth. Although the threeway interaction itself is statistically significant and belongs in the loglinear model based on the contingency table, the particular associations are not and cannot be given much interpretive weight. In loglinear modeling a significant interaction between three factors implies that all interactions between any two of them are also statistically significant and belong in the model. The significant interaction between phase and presence of caries for the contingency table is summarized in Figure 8.4. Acbi and late Coner teeth are associated with presence of caries, and early Coner teeth are associated with absence of caries. In relative terms, caries frequency is moderate (11.3 percent) for the Acbi phase, low (6.8 percent) for the early Coner, and high (21.7 percent) for the late Coner. Interpretation of this pattern, however, must be tempered by the fact that none of the associations between particular phases and presence of caries is statistically significant. A statistically significant interaction between sex and presence of caries occurs in the loglinear model for the contingency table in which subpopulation is based on sex. Females are significantly associated with presence of caries, and males are significantly associated with absence of caries (Figure 8.5). Caries frequency is 26.0 percent for females but only 14.3 percent for males. Loglinear modeling reveals no statistically significant interactions be
Page 159
Figure 8.3. Frequencies of caries in age classes for Copán commoners. Circles are Acbi phase (AD. 400700), squares are early Coner phase (A.D. 700800), and diamonds are late Coner phase (postA.D. 800).
Figure 8.4. Frequencies of caries in archaeological phases for Copán commoners. Differences between frequencies are not statistically significant.
tween presence of caries and either location within the Copán Valley or type of site (Figure 8.5). Caries frequency is higher for the rural area (20.5 percent) than for the urban area (16.7 percent), but not significantly so. Caries frequency is higher for Type I/Aggregate sites (19.9 percent) than for Type II sites (17.5 percent), but the difference again is not significant. Discussion Patterson (1984) presented caries frequencies in teeth from individuals older than 16 years of age in a worldwide sample of skeletons from populations using different subsistence strategies, from hunting and gathering through mixed horticulture/hunting and gathering to horticulture. This sample was modified from an earlier study by Turner (1979). In general, populations relying heavily on horticulture had the highest frequencies of carious teeth. The caries rate in permanent teeth for Copán's low status population is 17.9 percent, higher than for any of the 15 mixedsubsistence populations in the sample and higher than for all but 4 of the 19 horticultural populations. This suggests a heavy reliance on horticulture at Copán, as do the results of stable isotope analysis of bones
Page 160
Figure 8.5. Frequencies of caries for subpopulations of Copán commoners. The difference between male and female frequencies is statistically significant. Differences between rural and urban frequencies and Type I/Aggregate and Type II site frequencies are not statistically significant.
from Copán, which indicate that maize made up a large proportion of the lowstatus diet (Whittington and Reed 1997). Few of the populations in the worldwide sample practiced maizebased horticulture, however. Potentially more valuable comparisons can be made between Copán's rates of caries and antemortem loss and rates for skeletons from prehistoric, protohistoric, and historic aboriginal populations in southern Ontario and Ohio which relied on maizebased horticulture. In comparison with samples from 14 Ontario sites presented by Patterson (1984:313), Copán's 17.9 percent caries rate for permanent teeth is most similar to the rates for 3 horticultural populations relying to a fair degree on hunting and gathering or fishing (range 10.830.2 percent). However, Copán's 15.0 percent rate for caries in deciduous teeth places it among 4 populations relying heavily on horticulture (range 13.929.2 percent). Compared with samples from Ohio sites presented by Schneider (1986), Copán's rate for caries in permanent teeth falls between rates for 3 populations practicing mixed horticulture/hunting and gathering (range 9.710.9 percent) and the rate for another population relying more heavily on horticulture but still engaging in some hunting and gathering (24.8 percent). Copán's antemortem loss rate for permanent teeth (9.3 percent) is about half that of most Ontario horticultural populations (range 10.424.0 percent) and approximates the 8.0 percent rate at a nonhorticultural site (Patterson 1984:308). Frequencies of caries and antemortem loss calculated for some samples of skeletons from Classic period Maya sites are presented in Table 8.3. There is some variability in how the frequencies in the table were calculated, and they are at best rough estimates that must be compared with caution. The frequency of permanent teeth with caries in Copán commoners is higher than the frequency of teeth with caries in adults in 2 other samples. The frequency of Copán commoners with caries in permanent teeth is higher than the frequency of adults with caries in 9 of 10 samples. Seemingly at odds with the other patterns, the frequency of Copán commoners affected by antemortem loss of at least one permanent tooth is lower than the frequency of adults with loss in 2 other samples. This may be due to the poor preservation of the alveolar area in many Copán specimens, which does not allow for determination of whether
Page 161 Table 8.3. Frequencies of Caries and Antemortem Loss at Some Classic Period Maya Sites.
Site
Age
Period
Teeth with Caries
Reference
Indiv. with Caries
Teeth with Loss
Indiv. with Loss
Altar de Sacrificios
Adults
Classic
Saul 1972
0.622
0.378
Chichén Itzá
Adults
Classic
Peña 1985
0.167
Chichén Itzá (Chultun)
Adults
Clas sic
Peña 1985
0.000
Colha
Children
Term. Classic
Massey 1989
0.700
Colha
Adults
Term. Classic
Massey 1989
0.750
0.550
Copán (including elite)
Subadults
Classic
Hodges 1985
0.079
0.273
Copán (including elite)
Adults
Classic
Hodges 1985
0.122
0.657
Copán (commoners)
Deciduous
Classic
This study
0.150
0.435
Copán (commoners)
Permanent
Classic
This study
0.179
0.682
0.093
0.372
Cozumel
Adults
Classic
Peña 1985
0.273
Komchén
Adults
Classic
Peña 1985
0.500
Lamanai
Adults
Classic
White 1994
0.130
Lubaantun
Adults
Late Classic
Saul 1975
0.435
Tancah
Children
Classic
Saul 1982
0.167
Tancah
Adults
Classic
Saul 1982
0.200
Xcan
Adults
Classic
Peña 1985
0.364
loss occurred before or after death. The frequency of caries in deciduous teeth from Copán commoners is higher than the frequency of teeth affected by caries in children in 2 of 3 other samples. Some remarkable patterns are revealed through comparison of Copán's caries rates for tooth classes with rates for 7 southern Ontario populations (Patterson 1984:311). Copán commoners have rates for caries in permanent incisors, canines, and premolars that are comparable to those among Ontario horticulturalists, but the rate for molars at Copán is about half those for Ontario. Caries rates for deciduous canines and molars at Copán seem low but reasonable in comparison with those for Ontario horticultural populations, but the rate for incisors at Copán is approximately 10 times higher than those in Ontario. Some factor at Copán seems to have reversed the typical pattern of higher caries rates in posterior teeth. A good candidate is environmental stress, as reflected in enamel defects, which have increased susceptibility to attack by caries. The enamel hypoplasia rate in individuals with permanent dentitions appears to be 100 percent (Whittington 1992). Defective enamel occurs at high rates for anterior teeth, especially canines. The Pearson coefficient of correlation (r) between frequency of hypoplasia for incisors, canines, and premolars and frequency of caries for all teeth is 0.53, which is significantly different from zero (p < 0.025). The shift in highest caries frequency from anterior to posterior teeth with increasing age may reflect a transition from defectrelated caries in the young to pitandfissurerelated caries in the elderly. The antemortem loss rate for permanent incisors at Copán is similar to the rate for Ontario horticultural populations, but the rates for Copán canines, premolars, and molars are less than half the Ontario rates (Patterson 1984:308).
Page 162
Considering the Copán pattern another way, the rate of loss for incisors is elevated compared with the rates for the other teeth. Is this pattern due to scurvy, which Saul (1972) suggested was present at Altar de Sacrificios? Antemortem tooth loss in individuals old enough to have permanent teeth appears to have been rarely, if ever, caused by scurvy at Copán. Scurvy would cause the highest frequency of loss to occur among the incisors and, perhaps, the canines and premolars, all teeth with single roots, and anterior tooth loss would be expected to occur in subadults or young adults with almost as much frequency as in old adults. These are not the patterns at Copán. Lack of significant differences between their age patterns indicates there is an intimate relationship between caries and antemortem loss in Copán commoners. As with caries, environmental stress is reflected in the correlation between frequency of antemortem loss for molars and frequency of enamel hypoplasia for incisors, canines, and premolars (r = 0.54, p < 0.025). Caries almost invariably occurs earlier and at a higher frequency than antemortem loss, further bolstering the argument that much tooth loss appears to be a result of caries. Tooth wear at Copán is not heavy in comparison with that in other prehistoric Amerindian populations, so pulp exposure followed by alveolar abscessing does not seem to have been a significant factor in antemortem tooth loss. Loss cannot be assumed to be equivalent to caries in this population, however, especially among old adults, where frequency of antemortem loss exceeds frequency of caries in the incisors. Another factor, probably periodontal disease, enters the equation. Hodges (1985) found periodontal disease to be present in 88.2 percent of adult skulls from Copán. The overall pattern of loss apparently reflects the combination of dental decay commencing in posterior teeth at relatively young ages and periodontal disease becoming acute and causing anterior teeth to drop out in older adults. Some aspect of commoner lifestyle at Copán made females significantly more susceptible than males to caries, which fits the normal pattern for permanent teeth from clinical studies, sites in the Petén and Yucatán (Peña 1985), and Lamanai (White 1994). However, this pattern is contrary to what was found at Barton Ramie (Willey 1965) and Sarteneja (Kennedy 1983). In response to criticism by Schoeninger (1998) of a previous interpretation of stable isotope data divided by sex for Copán commoners (Whittington and Reed 1997), it seems prudent to offer a reinterpretation here. Independent, separate variances ttests, without any assumption of equal variance (Wilkinson et al. 1992), reveal that there is no significant difference between mean carbon stable isotope values for males (9.17) and females (9.42),but there is a significant difference (p < 0.05) between mean nitrogen stable isotope values (7.37 for females vs. 7.82 for males). The range of carbon values for males (10.27 to 8.62) is greater than that for females (9.94 to 8.79), even though the means are essentially equivalent. Both male and female diets consisted of a high proportion of maize, but females had more restricted diets (reflected in carbon values) and ate foods with significantly different (probably less) protein content (reflected in nitrogen values) than did males. The significantly higher female caries rate is
Page 163
therefore not surprising. In addition, since frequency of carbohydrate consumption outweighs amount consumed in producing caries, sexbased occupational differences among commoners may have allowed females to eat maize more frequently during the course of the day's activities than males. In relation to the collapse, the frequency of caries drops as it approaches, then increases greatly after it occurs. The 11.3 percent Acbi rate and 21.7 percent late Coner rate for caries in permanent teeth are most comparable to horticultural populations partially dependent on hunting and gathering. The 6.8 percent rate places the early Coner group among populations practicing a hunting and gathering strategy. The different rates may reflect a changing emphasis on gathered foods as part of the diet. The rates are not significantly different, however, and must be interpreted with caution. The peak in caries frequency apparently occurred after A.D. 800 at Copán but preceded this date at other sites for which temporal trends in caries frequencies can be determined. At Lamanai, White (1988,1994) determined that the mean percentage of caries peaked during the Early Classic (A.D. 250600). Evans (1973) found a peak in the mean frequency of caries during the Middle Classic (A.D. 400700) in the Tayasal area. At Altar de Sacrificios the mean number of carious lesions per mouth peaked in the Early Classic (Saul 1972), whereas at Seibal the number of teeth per mouth affected by either caries or tooth loss for young and middle adults was highest in the Preclassic (preA.D. 250; Saul 1973). Peña (1985) also noted a Preclassic peak in the mean percentage of caries in a combined sample from various Petén and Yucatán sites. Factors causing caries appear to have been about the same for commoners regardless of population density. People in lightly populated areas of the Copán Valley led a lifestyle that made them only slightly more susceptible to caries than people living in more crowded conditions. Population density may not have had a significant impact on diet at Copán because no one was far from agricultural land. People in the urban zone, in fact, lived adjacent to the prime bottomlands near the river. Cariogenic factors were probably similar in both the smaller and the larger sites considered in this study. The slightly higher caries rate for Type I/Aggregate sites may indicate that residents had slightly lower status than residents of Type II sites, but the difference is not large enough to reflect major lifestyle differences. Both Type I/Aggregate and Type II sites appear to have been inhabited by commoners, as suggested by Webster and Freter (1990). Conclusions Analysis of dental caries and antemortem tooth loss in the skeletons of lowstatus inhabitants of Copán has revealed important information about diet, heterogeneity within the population, and the collapse of political authority at Copán after A.D. 800. Comparison with data from other Maya sites has revealed heterogeneity between populations, which has implications for characterizing overall Maya diet as well as the course of the collapse throughout the southern Maya Lowlands.
Page 164
A large proportion of the food eaten by Copán commoners consisted of maize, but an important component of the diet apparently came from hunted and gathered foods, or at least from horticultural produce that was not as soft, sticky, and carbohydrateladen as cooked maize. Nutrients in these other foods could be part of the reason the pattern of tooth loss within the mouth and the accumulation of antemortem loss with increasing age make it unlikely that scurvy was present at Copán, at least on the scale suggested by Saul (1972) for Altar de Sacrificios. Although the commoner population can be subdivided along various lines, most sources of heterogeneity did not produce significant differences in diet. With the exception of sex, associations between caries and subpopulations are nonsignificant, leading to the conclusion that diet among commoners was about the same regardless of temporal, populationdensity, or residencesize differences. Analysis of enamel hypoplasia, porotic hyperostosis, and stable isotopes lead to similar conclusions about health, stress, and diet (Whittington 1992; Whittington and Reed 1997). Correlations between enamel hypoplasia frequency and frequencies of caries and antemortem loss suggest that environmental stress was affecting the entire lowstatus segment of Copán's population around the time of the political collapse. This is in agreement with other evidence of stress from the demographic profile, frequency and severity of enamel hypoplasia, and frequency of evidence of anemia (Whittington 1991,1992; Whittington and Reed 1997). Comparison of Copán's dental caries and antemortem loss rates with those at other Classic period Maya sites seems to show that much dietary variation existed between different geographic locations and that hunted and gathered foods may have made up an even greater proportion of the diet elsewhere in the Maya Lowlands. Temporal and sex patterns in caries are not consistent between sites. Heterogeneity between populations of Maya living in different locations apparently was more important than that within a single population at one site. Thus, evidence of an environmental stimulus to the collapse at Copán may mean little for distant sites, where other factors may have been preeminent. It is clear that more indepth studies of skeletons from a variety of sites are needed before it will be possible to make a final assessment of the importance of environmental stress in an overall model of the collapse. Acknowledgments This research was supported by Dissertation Improvement Grant BNS8314234 from the National Science Foundation and by a Hill Foundation Fellowship through The Pennsylvania State University. Excavations and laboratory research were performed with the kind permission of the Instituto Hondureño de Antropología e Historia. My gratitude goes to Richard Leventhal, Denise Hodges, Frank Saul, George Armelagos, Rebecca Storey, and the late Clifford Clogg for their help. Special thanks are due William Sanders, director of the Proyecto Arqueológico Copán, Fase 2. Feedback from Christine White and anonymous reviewers greatly improved this chapter.
Page 165
References Cited Abrams, E. M., and Rue, D. J. (1988) The causes and consequences of deforestation among the prehistoric Maya. Human Ecology 16:377395. Bass, W. M. (1971) Human Osteology: A Laboratory and Field Manual of the Human Skeleton. Columbia: Missouri Archaeological Society. Bibby, B. G. (1961) Cariogenicity of foods. Journal of the American Medical Association 177:316321. Brown, A. B. (1981) Assessment of paleonutrition from skeletal remains. Annals of the New York Academy of Science 376:405415. Burns, P. E. (1979) Loglinear analysis of dental caries occurrence in four skeletal series. American Journal of Physical Anthropology 51:637648. Carlos, J. P., and Gittelsohn, A. M. (1965) Longitudinal studies of the natural history of caries. II. A life table study of caries incidence in the permanent teeth. Archives of Oral Biology 10:739751. Ditch, L. E., and Rose, J. C. (1972) A multivariate dental sexing technique. American Journal of Physical Anthropology 37:6164. Dixon, W. J. (ed.) (1985) BMDP Statistical Software Manual. Berkeley: University of California Press. Evans, D. T. (1973) A preliminary evaluation of tooth tartar among the Preconquest Maya of the Tayasal area, El Petén, Guatemala. American Antiquity 38:489 493. Giles, E. (1970) Discriminant function sexing of the human skeleton. In T. D. Stewart (ed.): Personal Identification in Mass Disasters. Washington, D.C.: Smithsonian Institution, pp. 99109. Haviland, W. A. (1972) Estimates of Maya population: Comments on Thompson's comments. American Antiquity 37:261262. Hodges, D. C. (1985) Ms. in possession of the author. Kennedy, G. E. (1983) Skeletal remains from Sarteneja, Belize. In R. V. Sidrys (ed.): Archaeological Excavations in Northern Belize, Central America. Monograph No. 17. Los Angeles: Institute of Archaeology, University of California, pp. 353372. Knoke, D., and Burke, P. J. (1980) LogLinear Models. Sage University Papers Series on Quantitative Applications in the Social Sciences, No. 07020. Beverly Hills, Calif.: Sage Publications. Lovejoy, C. O.; Meindl, R. S.; Pryzbeck, T. R.; and Mensforth, R. P. (1985) Chronological metamorphosis of the auricular surface of the ilium: A new method for the determination of adult skeletal age at death. American Journal of Physical Anthropology 68:1528. Massey, V. K. (1989) The Human Skeletal Remains from a Terminal Classic Skull Pit at Colha, Belize. Papers of the Colha Project, vol. 3. Austin: Texas Archeological Research Laboratory, University of Texas; College Station: Department of Anthropology, Texas A&M University. Molnar, S. (1971) Human tooth wear, tooth function, and cultural variability. American Journal of Physical Anthropology 34:175190. Muller, T. P., and Mayhall, J. T. (1971) Analysis of contingency data on torus mandibularis using a log linear model. American Journal of Physical Anthropology 34:149154. Ortner, D. J., and Putschar, W. G. (1985) Identification of Pathological Conditions in Human Skeletal Remains. Smithsonian Contributions to Anthropology, No. 28. Washington, D.C.: Smithsonian Institution. Originally published 1981. Patterson, D. K., Jr. (1984) A Diachronic Study of Dental Paleopathology and Attritional Status of Prehistoric Ontario PreIroquois and Iroquois Populations. Archaeological Survey of Canada. National Museum of Man Mercury Series, Paper No. 122. Ottawa: National Museums of Canada. Peña Saint Martin, F. (1985) Nutrición entre los mayas prehispánicos: Un estudio osteobiográfico. Cuicuilco 4(16):516.
Page 166
Pindborg, J. J. (1970) Pathology of the Dental Hard Tissues. Philadelphia: W. B. Saunders. Powell, M. L. (1985) The analysis of dental wear and caries for dietary reconstruction. In R. I. Gilbert, Jr., and J. H. Mielke (eds.): The Analysis of Prehistoric Diets. Orlando, Fla.: Academic Press, pp. 307338. Rue, D. J. (1987) Early agriculture and Early Postclassic Maya occupation in western Honduras. Nature 326:285286. Ruz Lhuillier, A. (1965) Tombs and funerary practices of the Maya Lowlands. In G. R. Willey (ed.): Archaeology of Southern Mesoamerica: Handbook of Middle American Indians, vol. 2, part 1. Austin: University of Texas Press, pp. 441460. Saul, F. P. (1972) The Human Skeletal Remains of Altar de Sacrificios: An Osteobiographic Analysis. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 63, No. 2. Cambridge: Harvard University. Saul, F. P. (1973) Diseases in the Maya area. In T. P. Culbert (ed.): The Classic Maya Collapse. Albuquerque: University of New Mexico Press, pp. 301324. Saul, F. P (1975) Human remains from Lubaantun. In N. Hammond (ed.): Lubaantun. Monograph No. 2. Papers of the Peabody Museum of Archaeology and Ethnology. Cambridge: Harvard University, pp. 389410. Saul, F. P. (1982) The human skeletal remains of Tancah, Mexico. In A. G. Miller (ed.): On the Edge of the Sea: Mural Paintings at TancahTulum. Washington, D.C.: Dumbarton Oaks, pp. 115128. Schele, L., and Freidel, D. (1990) A Forest of Kings: The Untold Story of the Ancient Maya. New York: William Morrow. Schneider, K. M. (1986) Dental caries, enamel composition, and subsistence among prehistoric Amerindians of Ohio. American Journal of Physical Anthropology 71:95102. Schoeninger, M. J. (1998) Bones of the Maya: Studies of Ancient Skeletons edited by Stephen L. Whittington and David M. Reed. Latin American Antiquity 9:87 88. Snedecor, G. W., and Cochran, W. G. (1967) Statistical Methods. 6th ed. Ames: Iowa State University Press. Steggerda, M., and Hill, T. J. (193536) Incidence of dental caries among Maya and Navajo Indians. Journal of Dental Research 15:233242. Storey, R. (1986) Perinatal mortality at preColumbian Teotihuacan. American Journal of Physical Anthropology 69:541548. Turner, G. C. II (1979) Dental anthropological indications of agriculture among the Jomon people of central Japan. X. Peopling of the Pacific. American Journal of Physical Anthropology 51:619636. Ubelaker, D. H. (1978) Human skeletal remains. Washington, D.C.: Taraxacum. Viel, R., and Cheek, C. D. (1983) Sepulturas. In Introducción a la arqueología de Copán, Honduras, vol. 1. Tegucigalpa: Instituto Hondureño de Antropología e Historia, pp. 551609. Webster, D., and Freter, A. (1990) The demography of Late Classic Copán. In T. P. Culbert and D. Rice (eds.): Precolumbian Population History in the Maya Lowlands. Albuquerque: University of New Mexico Press, pp. 3762. Webster, D., and Gonlin, N. (1988) Household remains of the humblest Maya. Journal of Field Archaeology 15:169190. Webster, D.; Sanders, W. T.; and van Rossum, P. (1992) A simulation of Copán population history and its implications. Ancient Mesoamerica 3:185197. White, C. D. (1988) Diet and health in the ancient Maya at Lamanai, Belize. In B. V. Kennedy and G. M. LeMoine (eds.): Diet and Subsistence: Current Archaeological Perspectives. Calgary: University of Calgary Archaeological Association, pp. 288296. White, C. D. (1994) Dietary dental pathology and culture change in the Maya. In A. Herring and L. Chan (eds.): Strength in Diversity: A Reader in Physical Anthropology. Toronto: Canadian Scholar's Press, pp. 279302.
Page 167
Whittington, S. L. (1989) Characteristics of Demography and Disease in LowStatus Maya from Classic Period Copán, Honduras. Ph.D. dissertation, Pennsylvania State University. Ann Arbor: University Microfilms. Whittington, S. L. (1991) Detection of significant demographic differences between subpopulations of Prehispanic Maya from Copán, Honduras, by survival analysis. American Journal of Physical Anthropology 85:167184. Whittington, S. L. (1992) Enamel hypoplasia in the lowstatus Maya population of Prehispanic Copán, Honduras. In A. H. Goodman and L. L. Capasso (eds.): Recent Contributions to the Study of Enamel Developmental Defects. Chieti: Associazione Antropologica Abruzze. Journal of Paleopathology, Monographic Publications, No. 2, pp. 185205. Whittington, S. L., and Reed, D. M. (1997) Commoner diet at Copán: Insights from stable isotopes and porotic hyperostosis. In S. L. Whittington and D. L. Reed (eds.): Bones of the Maya: Studies of Ancient Skeletons. Washington, D.C.: Smithsonian Institution Press, pp. 157170. Wilkinson, L.; Hill, M. A.; and Vang, E. (1992) SYSTAT: Statistics. Version 5.2 ed. Evanston, Ill.: SYSTAT. Willey, G. R. (1965) Human burials. In G. R. Willey, W. R. Bullard, Jr., J. B. Glass, and J. C. Gifford (eds.): Prehistoric Maya Settlements in the Belize Valley. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 54. Cambridge: Harvard University, pp. 530558. Willey, G. R., and Leventhal, R. M. (1979) Prehistoric settlement at Copán. In N. Hammond and G. R. Willey (eds.): Maya Archaeology and Ethnohistory. Austin: University of Texas Press, pp. 75102. Wing, E. S., and Brown, A. B. (1979) Paleonutrition: Method and Theory in Prehistoric Foodways. New York: Academic Press. Wisdom, C. (1940) The Chorti Indians of Guatemala. Chicago: University of Chicago Press.
Page 169
Chapter 9 Late Classic Nutrition and Skeletal Indicators at Copán, Honduras Rebecca Storey The site of Copán, Honduras, was one of the major centers of the Classic Maya civilization. Located on the southeastern periphery of the Maya world, it is in a 24 km2 mountain valley nearly 600 m above sea level, distinguishing it geographically from other Lowland Maya centers in Guatemala, Yucatán, and Belize, which are somewhat higher in elevation. Although clearly Maya in its cultural florescence during the Classic period from circa A.D. 400 to 900, Copán is also distinctive because it has cooler daily temperatures and more marked seasons than the lowland areas (Fash 1991). Thus, in any discussion of diet and nutritional status among Maya populations, Copán will most likely not provide a typical situation. The concentrated archaeological investigations at the site and its environs since 1975 have provided some of the best information currently available about the history and characteristics of a Mayan kingdom. One characteristic that Copán shares with other Classic Maya centers is that it was one of those that ''collapsed" after the Terminal Classic. Copán started to decline in population around A.D. 900, after reaching a peak of perhaps 27,000. Half the population was apparently lost quickly, within 75 years, with only 29 percent left by A.D. 1000 (Webster et al. 1992). The valley was not finally abandoned until around A.D. 1200. Thus, although the "collapse" was not as sudden or as catastrophic as archaeologists had thought, it nevertheless provides a dramatic case of population loss, and loss of cultural influence. Copán disappears from influence and presence in the Maya world after the Terminal Classic at about A.D. 1000. The time of highest cultural florescence was the Late Classic. During this period the urban core around the Acropolis also had the densest population recorded for a Maya site (Webster et al. 1992). Archaeological excavations have concentrated on the Late Classic period, and the Copán skeletal collection is dominated by individuals dated to the Late and Terminal Classic. Thus, the skeletons can provide information about life during this crucial period when the site peaks and declines. A portion of the Copán skeletal sample is used here to look at evidence about nutritional status. Since ecological degradation resulting from the effects of agricultural intensification to feed the dense population has been suggested as an important cause in the collapse of Copán (for example, see Fash 1991), nutritional status is a pertinent subject. Did poor nutrition contribute to the decline of this society?
Page 170
In the future, because of the fine chronological control provided for the valley by the use of obsidianhydration dating (Freter 1992), it should be possible to divide the sample into those individuals living and dying when the population was at its peak and those coming from the decline phase. At this point it is not yet possible to date each skeleton reliably. However, preliminary indications are that the bulk of the sample to be discussed here dates to after A.D. 900 and will be informative about the declining conditions. For now, all skeletal individuals must be treated as Late/Terminal Classic and studied for evidence bearing on nutritional status. Much information on diet and nutrition reconstructed from ancient contexts is somewhat indirect. Archaeologists can recover evidence of plant and animal remains that were probably consumed but cannot tell what the daily or yearly intakes of various items were. Assessments determining that resources were available to provide wellbalanced diets does not, of course, mean that such were eaten by every member of the society. Skeletons can provide more direct evidence on dietary intake by the use of chemical analyses, especially isotopic analyses, as seen later in this volume (Reed, Chap. 10). Chemical analysis on its own cannot, however, inform us about how the diet might have influenced individual morbidity and adaptation. On the other hand, certain paleopathological indicators on skeletons can provide some information on health status but do not inform directly about dietary intake. Thus, as pointed out in Sobolik 1994, a good understanding about paleonutrition really involves contributions from a variety of data and researchers. Although the knowledge base for Copán which integrates archaeological data and chemical analysis is still being built, there is already information available to complement the data from pathological indicators. Unfortunately, there is no faunal analysis yet, but a study of plant remains indicates that a variety of items were available to the Copán residents (Lentz 1991). Maize, beans, and squash (the usual Mesoamerican cultigens) were present, but so were several tree species and a great deal of coyol palm, which was one of the most common remains in the Late Classic. Lentz hypothesizes that coyol, which can be used for food, oil, and wine (1991:277), became a desperation food, its use an index of agricultural production problems. Also, plant remains by status were compared. The diversity of plants recovered from elite Type IV households was greater than that for lowerstatus Type I, II, and III households. Because of the fats and calories available from coyol, the Copán elite, at least, might have been able to get better nutrition from their diet than other individuals. Stable isotope analysis of some Copán skeletons has also been done (Reed 1994,1995, this volume). Although Reed has not yet given a detailed interpretation of the dietary differences that may underlie the isotopic differences, he has found that the diet was dominated by maize and varied by age, sex, and social status. Notably, the elite appeared to have greater dietary variability. The Copán Skeletons and Paleopathology Paleopathological indicators are an indirect reflection of diet and nutrition. Such indicators are the result of chronic or severe stress that overwhelms phys
Page 171
iological buffering ability, and they leave evidence in human bone (Goodman et al. 1988). Because of the limited response of bone to stress, much pathology is nonspecific, caused by different diseases and conditions and likely to be the outcome of complex interactions of nutritional status and illness (Goodman et al. 1988). To add to this complexity, some indicators form only during childhood or growth and development, whereas others can happen at any point in the life cycle. When several pathological indicators are examined at the same time, evidence of stress, or its lack, is strengthened. At Copán, indicators of stress during growth (i.e., enamel hypoplasias, porotic hyperostosis, adult stature) and adulthood (i.e., infectious lesions) are investigated. The intent is to relate the common occurrence of these indicators to possible nutritional stress during the Late Classic at Copán. The Copán skeletal population is a fairly large one, of which only a portion has been analyzed to date. Here only information from adults is described, as they have all survived childhood and their indicators from that time can be compared. Children often die during infancy and early childhood and thus represent a biased mortality sample from those years. A recent article has attacked the use of paleopathology as an indicator of past health because skeletons represent a selective mortality sample (Wood et al. 1992). That is, skeletons represent a sample of individuals dying at various ages and are no guide to those that survive those ages. In fact, one should expect the dead to appear sicker and in poorer health than the living population. That is certainly likely to be true of children and is the reason for their separate analysis. For adults, however, evidence of less stress during childhood should indicate better social and physiological buffering during that time and an easier passage to adulthood than those with evidence of more stress. The presence of infectious lesions reflects conditions more proximate to the time of death and should indicate lifestyle during adulthood. Sex and age at death are two pieces of basic information that can be determined from skeletons, although each has its own possible sources of error. The Copán adults were sexed using standard morphological traits of the cranium and pelvis, where possible; others were done on the basis of size and robusticity using discriminant functions. Individuals of uncertain age and sex are still undergoing statistical analysis and are not included here. Because of poor preservation, it was possible only to determine whether the individual had died over age 20 or whether he/she was an older adult, that is, over age 50 at death. The Classic Maya centers were characterized by inequalities in status, wealth, and power (Chase and Chase 1992). This was certainly true of Copán, where individuals lived in residences varying in materials, size, and location. The elites, as defined primarily by elaborateness of residence, were congregated near the Acropolis, the main ceremonial/civic buildings, in the Copán pocket; the poorer individuals were more dispersed around the countryside (see Webster 1992; Webster and Gonlin 1988). Thus differences in social status must be considered. Most of the sample is from the 9N8 compound, the "House of the Bacabs" (Webster 1989), which was the largest outside the Main Group, containing eleven adjoining patio groups with more than 50 structures, housing around
Page 172
200 individuals at one time. The compound is believed to have been occupied by families belonging to a prominent noble lineage. However, mortuary characteristics varied within the compound, from individuals with tombs and various offerings that included exotic shell and jade/serpentine objects, to simple earth pits with no offerings. Thus, there was probably a range of statuses within the compound, and only a few individuals actually had a recognized elite rank. Therefore, this sample was divided into three groups: status group 1 comprised individuals with tombs or other grave construction such as cobble cists and capstones or the presence of grave offerings; status group 2 was made up of individuals who had only earthen pits and no grave offerings; status group 3 was composed of individuals recovered from the modest dwellings found in rural areas outside the urban core of the site. These are likely to be relatively poor commoners compared with the residents of the palatial compound of 9N8. The status group 2 individuals are distinguished from those of 3, because although there were likely to be commoners in the Copán social system, they were living in what appears to have been one of the wealthiest compounds and should have benefited from that setting. Comparisons of pathological indicators are made among these three different status groups, contrasting males and females. This should allow for exploration of the variability by status and sex that has been suggested by paleobotanical and stable isotope analyses. Methods for Paleopathological Indicators Infectious lesions and the indicators of stress during childhood (porotic hyperostosis, enamel hypoplasia, stunted adult stature) reflect chronic stressors for which nutritional status is implicated as either an underlying cause or a predisposition but is not necessarily the proximate cause of the morbidity. This is especially true of the childhood indicators, which are associated with mild to moderate undernutrition in contemporary public health studies (Goodman 1994). Poor resistance to disease also is associated with impaired nutritional status (Goodman et al. 1984). Therefore, it is assumed that those with more stress indicators probably had a more impoverished nutritional status than those with fewer stress indicators. Infectious lesions on bone, periosteal reactions on the surface of long bones, and periapical abscesses around the teeth are the result of chronic bacterial infections (Ortner and Putschar 1985). These are scored as present or absent for abscesses, where at least 10 crypts were available for scoring. Infection was scored for all the available skeletons as absent, slight, and systematic, the last requiring clear evidence of extensive periosteal reactions on several bones. The percentage of individuals with periosteal reactions and abscesses in each status group is compared. Porotic hyperostosis is evidence of iron deficiency anemia during childhood (StuartMacadam 1985) and appears as a spongy bone on the outer surface of the skull. Such anemia would result from various interactions among diseases, parasites, and dietary deficiencies, but obviously, whatever the iron content of the diet, an individual with anemia lacks sufficient iron for physio
Page 173
logical needs. The adults at Copán all have healed lesions, indicating that they survived the childhood stress of anemia. Because porotic hyperostosis occurs symmetrically on parietal bones, the indicator is scored as present or absent if at least one parietal bone was present, and the percentage of individuals is calculated for status and sex groupings. Dental enamel hypoplasias are deficiencies in the thickness of enamel and result from physiological stress sufficient to interfere with the normal enamel development of teeth during early childhood (Goodman and Rose 1991). Such dental defects appear to be sensitive to nutritional status, with mild and moderately malnourished children having more hypoplasias in contemporary populations (May et al. 1993). The maxillary central incisors and all canines are the teeth most susceptible to hypoplasia, so only individuals with these teeth are analyzed here. The scoring scale contrasts those with evidence of none, one, or more than one hypoplasia on the susceptible teeth, as the lastnamed is evidence of multiple stress episodes during early childhood. Adult stature can be affected by illness and nutrition during childhood, such as poor growth status resulting in short adult stature, and is common even among children with moderate malnutrition (Goodman 1994). Stature is calculated from the length of the femur and tibia, using Genovés's formula (1967), providing a preliminary estimate of adult stature. The relative difference in stature estimates for males and females and the three status groups should provide a relative measure of growth status. Males are normally expected to be taller than females but are more affected by growth stress than females (Stini 1971). Although there is obviously a genetic component to height which causes some variation, it is assumed that within this Classic Maya population, individuals would tend to have the same mean height. Differences within the status subgroups are more likely to be due to environment (i.e., nutrition) than genetics (Bogin 1988). Results The maximum sample size at present available for the analysis is 128 individuals distributed as presented in Table 9.1, although differences in preservation mean not all individuals can be scored for all indicators. Using the chisquare test, we found no statistical significance between different numbers of males and females in the different status groups, except for Group 2, which has twice as many females as males. Thus the best comparisons here, those not affected by significant differences in the sample sizes in the categories, are between the status groups within each sex. Table 9.2 compares the age breakdown of males and females and also lists the individuals that cannot be scored for each of the pathological indicators. Both sexes show an increased proportion of old adults in status group 3, which is statistically significant ( 2 = 6.5, df = 2, p < .05) for the females only. This age difference could affect the incidence of infectious lesions (i.e., older individuals might have more) but should not affect the childhood indicators. The pattern of missing values for indicators seems to be variable, as would be hoped for an
Page 174 Table 9.1. Sex and Status Distribution of Copán Individuals.
Male
Female
Status 1
25
17
a
35
Sig.
N.S.
128
N.S.
Result
75
35
53
c2
N.S.
18
Totals
24
58
11
Status 3
Resulta
33
Status 2
2
Totals
N.S.
N.S.
Sig.
a
2
N.S. means p > 0.05; Sig. means p < 0.05. For all six categories, = 2.24, df = 2, p > 0.05. Table 9.2. Characteristics of the Male and Female Samples.
Status 1
Status 2
Status 3
N(%)
N(%)
N(%)
Females
Under 50
24(73)
17(71)
7(39)
Over 50
9(27)
7(29)
11(61)
No Teeth
4(12)
3(13)
3(17)
No Skull
7(21)
3(13)
1( 6)
No Crypts
7 (21)
7 (29)
4(22)
No Stature
7(21)
8(33)
5(28)
Males Under 50
16(64)
7(64)
8(47)
Over 50
9(36)
4(36)
9(53)
No Teeth
2( 8)
2(18)
2(12)
No Skull
8 (32)
3 (27)
3(18)
No Crypts
3(12)
5 (45)
8 (47)
No Stature
8 (32)
4 (36)
5(28)
essentially random effect. No one status has consistently more missing values, although there are fewer males that can be scored for abscesses. Because of variable preservation, some individuals were missing various skeletal elements. Only three individuals had a missing value for three indicators, and fifteen were missing two out of the sample of 128. Missing values are present but do not represent a factor that will bias trends in the data. The presence of indicators among females by status group is presented in Figure 9.1 and Table 9.3. The pathological indicators are common in the female skeletons, whatever their status. With regard to adult indicators of infection, for both males and females, the pattern for abscesses reveals that they are most common in the status 3 group. For other infectious lesions, all status groups have more slight than systematic ones. The status 2 group is the most distinct, however, with fewer individuals having no evidence of lesions and far more
Page 175
Figure 9.1. Incidence of paleopathological indicators among females of the three status groups (by percentage of status group).
having only slight lesions. In spite of this apparent difference, there is no significant difference in the distribution of abscesses or infectious lesions between status groups, according to the chisquare, and so all status groups are more or less afflicted equally. For the indicators from childhood, porotic hyperostosis has a definite trend for both males and females (see Figure 9.1), where status group 1 is little affected and status group 3 has the majority of lesions. This difference is statistically significant by chisquare and indicates differential susceptibility to anemia during childhood. For hypoplasias, status group 3 is again distinctive. There was only one female (in status group 1) in the scorable sample that had no evidence of this stress. Otherwise, status groups 1 and 2 have approximately equal distributions of single and multiple episodes of enamel hypoplasia. Although status group 3 has more individuals with one than with multiple hypoplasias, this difference is not statistically significant. Stature was fairly similar between all three status groups, as seen in Table 9.3. Thus, no one group of females appears to be particularly stunted, although the status 3 females are the shortest on average. A oneway ANOVA found no significant differences in stature between the three groups, although the largest standard deviation and range was in status group 1 For this particular indicator, nutritional status does not appear to have consistently stunted stature growth of females between status groups during childhood. The presence of indicators among male skeletons is shown in Figure 9.2 and Table 9.3. It is not possible to use chisquare to look for differences in the incidence of indicators between males, however, because the few numbers in status group 2 cause the test to violate expected frequency requirements. Nevertheless, the patterns can be compared with those of the females. Status group 3 individuals have a higher incidence of abscesses and of porotic hyperostosis and a pattern in which having only one hypoplasia is much more common than having more than one than in the other status groups. Status group 1 has the lowest incidence of abscesses and more porotic hyperostosis than the status 1 females. Males in status group 1 also have a higher incidence of slight infections than males in the other status groups. This is the only case in which status
Page 176 Table 9.3. Distribution of Paleopathological Indicators among Status Groups at Copán.
Status 1 N (%)
Status 2 N (%)
Status 3 N (%)
Females
No abscesses
17 (65)
11 (65)
6 (43)
Abscesses
9 (35)
6 (35)
8 (57)
No infection
15 (45)
5 (21)
8 (44)
Slight infection
11 (33)
16 (67)
6 (33)
Systemic infection
7 (22)
3 (12)
4 (23)
No porotic hyperostosis
23 (88)
14 (67)
7 (41)
Porotic hyperostosis
3 (12)
7 (33)
10 (59)
No hypoplasias
1 (3)
0
0
1 Hypoplasia
13 (45)
11 (52)
11 (73)
>1 Hypoplasia
14 (52)
10 (48)
4 (27)
Stature
155.2 cm ± 4.2
155.4 cm ± 2.2
154.9 cm ± 3.4
Males
No abscesses
16 (73)
4 (67)
3 (33)
Abscesses
6 (27)
2 (33)
3 (33)
No infection
7 (28)
6 (55)
8 (47)
Slight infection
14 (56)
2 (18)
6 (35)
Systemic infection
4 (16)
3 (27)
3 (18)
No porotic hyperostosis
11 (65)
8 (100)
1 (7)
Porotic hyperostosis
6 (35)
0
13 (93)
No hypoplasias
0
0
0
1 Hypoplasia
11 (48)
3 (33)
11 (73)
>1 Hypoplasia
12 (52)
6 (67)
4 (27)
Stature
163.4 cm ± 3.0
163.4 cm ± 2.5
160.1 cm ± 2.9
group 1 has the highest incidence of a stress marker. The status group 2 males have no cases of porotic hyperostosis and the highest percentage of multiple hypoplasias. In general, from Figure 9.2, the contrast between the males of status group 3 and the others is greater for most indicators than among the females. This pattern is fairly dramatically supported and illustrated by the stature estimates (Table 9.3). There is definitely a difference between the rural commoners and the individuals living in the 9N8 elite compound, with the lower status group 3 individuals revealing stunting. A oneway ANOVA found that there was a significant difference in mean stature among the status groups, and a multiple comparison test (using Tukey's b at p < 0.05) found that the only pairwise significant difference was between status groups 1 and 3. As groups 1 and 2 are very close, perhaps only the small status group 2 sample prevented a significant difference between groups 2 and 3 as well. This finding replicates that at Tikal (Haviland 1967), where elite males were significantly taller than commoners as well. The stature data provide evidence of impaired nutritional status during childhood in status group 3 and of the likelihood that males from the elite compound had better nutrition to support growth.
Page 177
Figure 9.2. Incidence of paleopathological indicators among males of the three status groups (by percentage of status group).
Conclusions Evidence of pathology is ubiquitous among this skeletal sample whatever the status. Interestingly, status group 1, the elites, does not consistently have the lowest incidence of pathoses. Status group 3, the rural commoners, is definitely more affected by abscesses and porotic hyperostosis for both males and females, but not in either slight or systemic infection or in multiple hypoplasias. Status group 3 males were also definitely more stunted in adult stature than the other status groups, although females did not differ in stature by status. Thus, looking at infectious conditions during adulthood and the stress markers from childhood, we find that no status group seems to have been able to consistently buffer individuals from their environment or to have had a sound nutritional status. The evidence from Copán suggests that males and females generally had similar experiences of impaired nutritional status, and all individuals but one (a status group 1 female) have at least one skeletal lesion of stress during childhood. The pattern of incidence of lesions is also interesting. Although all indicators are associated with impaired nutritional status, the exact physiological effect varied by status. For example, status group 3, the poorest commoners, were more susceptible to severe anemia during childhood and shorter stature for males but generally had only one hypoplasia. These individuals could have been subjected generally to only one severe morbidity episode during childhood, resulting in the two conditions of porotic hyperostosis and hypoplasia. Poorerquality diet all through childhood, however, resulted in shorter adult stature in males but not in females. On the other hand, status group 1 individuals had more multiple hypoplasias than single ones in both males and females but little porotic hyperostosis. Thus, for status group 1 the nutritional status was sufficient to protect against anemia and stunting of adult stature but not against multiple episodes of stress sufficient to cause hypoplasias. Also, it may be that in the poorer individuals, only those that suffered just one serious stress episode survived to adulthood, whereas the status group 1 and 2 individuals
Page 178
had sufficient buffering and quality of diet to survive more than one serious episode. The evidence from the skeletons of Copán suggests that social status had an effect on susceptibility of disease, with the higheststatus individuals being the least affected. Only the differences in porotic hyperostosis in females between status groups and the differences in mean adult stature in males between status groups were statistically significant, however. These indicators reflect impaired nutritional status and support paleobotanical and stable isotope studies (Reed, this volume), suggesting that the higherstatus individuals probably had a more varied and nutritionally sufficient diet, which better buffered them as both children and adults from morbidity. Because status group 1 individuals generally had one or more indicators as well, however, it seems that in the overall Copán polity no one was fully buffered and that during periods of childhood and adulthood all individuals had impaired nutritional status. In the Late/Terminal Classic periods Copán society was thus experiencing generalized nutritional stress, which was likely one of the factors underlying its collapse. References Cited Bogin, B. A. (1988) Patterns of Human Growth. Cambridge: Cambridge University Press. Chase, D. Z., and Chase, A. F. (eds.) (1992) Mesoamerican Elites: An Archaeological Assessment. Norman: University of Oklahoma Press. Fash, W. L. (1991) Scribes, Warriors, and Kings: The City of Copán and the Ancient Maya. London: Thames and Hudson. Freter, A. C. (1992) Chronological research at Copán: Methods and implications. Ancient Mesoamerica 3:117134. Genovés, S. (1967) Proportionality of the long bones and their relation to status among Mesoamericans. American Journal of Physical Anthropology 26:6777. Goodman, A. H. (1994) Cartesian reductionism and vulgar adaptationism: Issues in the interpretation of nutritional status in prehistory. In K. D. Sobolik (ed.): Paleonutrition: The Diet and Health of Prehistoric Americans. Center for Archaeological Investigations, Occasional Paper No. 22. Carbondale: Southern Illinois University, pp. 163177. Goodman, A. H., and Rose, J. C. (1991) Dental enamel hypoplasias as indicators of nutritional status. In M.A. Kelley and C. S. Larsen (eds.): Advances in Dental Anthropology. New York: WileyLiss, pp. 279294. Goodman, A. H.; Martin, D. L.; Armelagos, G. J.; and Clark, G. (1984) Indications of stress from bone and teeth. In M. N. Cohen and G. J. Armelagos (eds.): Paleopathology at the Origins of Agriculture. Orlando, Fla.: Academic Press, pp. 1350. Goodman, A. H.; Thomas, R. B.; Swedlund, A. C.; and Armelagos, G. J. (1988) Biocultural perspectives on stress in prehistoric, historical, and contemporary population research. Yearbook of Physical Anthropology 31:169202. Haviland, W. A. (1967) Stature at Tikal, Guatemala: Implications for ancient Maya demography and social organization. American Antiquity 32:316325. Lentz, D. L. (1991) Maya diets of the rich and poor: Paleoethnobotanica l evidence from Copán. Latin American Antiquity 2:269287. May, R. L.; Goodman, A. H.; and Meindl, R. S. (1993) Response of bone and enamel formation to nutritional supplementation and morbidity among malnourished Guatemalan children. American Journal of Physical Anthropology 92:3751.
Page 179
Ortner, D. J., and Putschar, W. C. J. (1985) Identification of Pathological Conditions in Human Skeletal Remains. Smithsonian Contributions to Anthropology, No. 28. Washington, D.C.: Smithsonian Institution. Reed, D. M. (1994) Ancient Maya diet at Copán, as determined through the analysis of stable carbon and nitrogen isotopes. In K. D. Sobolik (ed.): Paleonutrition: The Diet and Health of Prehistoric Americans. Center for Archaeological Investigations, Occasional Paper No. 22. Carbondale: Southern Illinois University, pp. 210221. Reed, D. M. (1995) Maya diets at Late Classic Copán. Paper presented at the 60th Annual Meeting of the Society for American Archaeology, Minneapolis, April. Sobolik, K. D. (ed.) (1994) Paleonutrition: The Diet and Health of Prehistoric Americans. Center for Archaeological Investigations, Occasional Paper No. 22. Carbondale: Southern Illinois University. Stini, W. (1971) Evolutionary implications of changing nutritional patterns in human populations. American Anthropologist 73:10191030. StuartMacadam, P. (1985) Porotic hyperostosis: Representative of a childhood condition. American Journal of Physical Anthropology 66:391398. Webster, D. L. (ed.) (1989) The House of the Bacabs. Studies in PreColumbian Art and Archaeology, No. 29. Washington, D.C.: Dumbarton Oaks. Webster, D. L. (1992) Maya elites: The perspective from Copán. In D. Z. Chase and A. F. Chase (eds.): Mesoamerican Elites: An Archaeological Assessment. Norman: University of Oklahoma Press, pp. 135156. Webster, D. L., and Gonlin, N. (1988) Household remains of the humblest Maya. Journal of Field Archaeology 15:169190. Webster, D. L.; Sanders, W. T.; and van Rossum, P. (1992) A simulation of Copán population history and its implications. Ancient Mesoamerica 2:185198. Wood, J. W.; Milner, G. R.; Harpending, H. C.; and Weiss, K. M. (1992) The osteological paradox: Problems of inferring prehistoric health from skeletal samples. Current Anthropology 33:343370.
Page 181
PART III BONE CHEMISTRY
Page 183
Chapter 10 Cuisine from HunNalYe David Millard Reed Intrasocial patterns of food preparation, distribution, and consumption reflect status distinctions and social distance within societies. Burials excavated from over a decade of research at Copán, Honduras, by the Pennsylvania State University's Copán Archaeological Project provide a resource for the analysis of human dietary variability along the dimensions of age, sex, social position, and residential zone within a complexly organized ancient polity. The stable carbon and nitrogen isotopic composition of bone collagen from 82 humans (9 subadults, 70 adults, and 3 probable adults), 7 deer, and 1 jaguar contribute evidence for a principally maize diet at Copán. Inferred diet varied by age and sex with clear differences between male adults and female adults. A greater range of diets is inferred for urban residents relative to their nonurban counterparts. These intrasocial patterns assist us in our quest to reconstruct ancient Maya social structure at the household, community, and polity levels. Copán In the Copán Valley of western Honduras, Maya settlement began with sedentary farmers circa 1000 B.C. (Fash 1991; Freter 1992; Rue 1987,1989).1 Population was sparse until A.D. 400, and both population size and density increased rapidly after A.D. 600. At the zenith of Copán kingship (circa A.D. 820), population size has been estimated at a maximum of 27,500 (Webster and Freter 1990a; Webster et al. 1992). Afterward population size remained stable for another century with a rapid demographic decline and an abandonment of the valley from A.D. 900 through 1250 (Webster et al. 1992). In the Copán ceramic sequence, Coner ceramics date from A.D. 600 to 1250, with a subphase division between early and late Coner at A.D. 900 (Freter 1992; Webster and Freter 1990b).2 Therefore, the Coner period encompasses the growth, centralization, and collapse of elite control of the Late Classic and Postclassic Copán polity. Seven types of residential sites have been identified at Copán, and examples of each have been excavated (Freter 1992,1994; Webster and Freter 1990a). Architecture, energy expenditure for construction, and richness measures of artifact diversity have proved more practical at distinguishing social rank at Copán than other measures (Abrams 1994; Gonlin 1994; Reed et al. 1993). Site ranking ranges from the ruling elite royal compounds (the Main Group) to the few, but impressive, Type IV residences of the subroyal, through the less impressive Type III, down to the more numerous Type I, Type II, singlemound, 1
Maize has been identified as present before 1700 B.C. in recently analyzed pollen cores (Webster, personal communication 1996).
2.
Viel (1993) places the dating of the Coner phase from A.D. 650 to 950.
Page 184
and nonmound sites of the commoners (Fash 1991; Freter 1994; Webster and Freter 1990b; Webster and Gonlin 1988). Residential artifact assemblages at all social levels are similar; therefore, domestic artifacts alone are poor discriminators of status at Copán (Gonlin 1993; Webster et al. 1997). Although all individuals interred at the lowerranked Type I, Type II, and smaller sites were undoubtedly commoners, it cannot be inferred that all people buried at higherranked sites were members of royal and subroyal groups (e.g., kin or lineage occupants of the patio complexes). Some interments at the higherranked sites probably included lowerranked individuals (e.g., retainers and servants). Demographic patterns can be distinguished based on the location of residential units relative to the urban core of occupation. The longestoccupied sites lie in the alluvial pocket near the Main Group. This core of settlement has the highest structure density of all Classic Maya centers (Webster and Freter 1990a). An intrapopulation distinction between the urban and nonurban sectors may provide discrimination between privileged and commoner members of Copán society better than associated site type. Stable Isotopes in Paleodietary Research Because categories of food resources have distinct stable carbon and nitrogen isotopic ratios and dietary isotopic composition is reflected in bone collagen, it is possible to reconstruct diet from osseous remains (DeNiro and Epstein 1978, 1981). Extensive reviews of stable isotope paleodietary research have been presented by Ambrose (1993), DeNiro (1987), Katzenberg (1992), Keegan (1989), Norr (1995), Schoeninger and Moore (1992), Pate (1994), Schwarcz and Schoeninger (1991), and van der Merwe (1982). The stable isotope ratios of carbon (13C to 12C) and nitrogen (15N to 14N) are measured in units per mil (0/00) as deviations of the isotope ratio in the sample from a reference. The carbon reference is the rostrum from the Cretaceous Pee Dee belemnite formation (PDB), and the nitrogen reference is ambient air (AIR). The notation is expressed in per mil (0/00) for carbon as 13CPDB = [(13C/12C)SAMPLE/(13C/12C)PDB 1]1000, and for nitrogen as 15NAIR = [(15N/14N)SAMPLE/(15N/14N)AIR 1]1000. Terrestrial plants can be categorized by their carbon isotope composition and photosynthetic type as either Calvin (C3based), HatchSlack (C4based), or Crassul0/00acean acid metabolism (CAM) (Bender 1968,1971; Bender et al. 1973; Calvin and Bassham 1962; Hatch 1976). On average, C3 plants have a 13CPDB value of 270/00, C4 plants have a 13CPDB value of 12.50/00, and CAM plants have 13CPDB values between C3 and C4 plants. Maize, a C4 plant, has a 13CPDB value near 11.50/00 (O'Leary 1988; DeNiro and Hastorf 1985); most other plants consumed by the Maya, such as ramón (27.40/00), beans (25.50/00), squash (24.5 0/00), cacao (34.10/00), chile pepper (29.00/00), and manioc (25.70/00), are C3based (DeNiro and Hastorf 1985; Norr 1990; Wright 1994; Yoshinaga et al. 1991). Nitrogen isotopes have been used to distinguish between marine animals,
Page 185 15
most of which have NAIR values greater than 12 /00, and terrestrial plant sources, which have values between 0 /00 and 10 /00. Stable nitrogen isotope ratios have also been used to separate legumes, all of which are C3based, and nonlegumes (DeNiro 1987; Keegan 1989).A stepwise increase of 15NAIR by 10/00 to 50/00 with successively higher trophic levels has been observed (Minagawa and Wada 1984). In humans, nursing infants have shown a 20/00 to 40/00 more positive 15NAIR value than weaned children and adults (Fogel et al. 1997; Fogel et al. 1989; Katzenberg et al. 1993; Tuross and Fogel 1994). 0
0
0
Determining the specific dietary source signaled by the isotopic composition of various bone fractions has become a leading issue in stable isotope paleodietary research. In the prevailing view, the carbon isotopic composition of dietary protein is represented to a higher degree in collagen than other diet fractions, whereas the whole diet is best reflected in the carbon isotopic value of bioapatite (Ambrose and Norr 1993; Krueger and Sullivan 1984; Tieszen and Fagre 1993). Previous Studies of Ancient Maya Diet and Health Most previous isotopic research in Mesoamerica has focused on temporal or ecological variation. Other Late Classic Maya sites have been studied by Coyston (1995), Gerry (1993), White and Schwarcz (1989), White, Healy, and Schwarcz (1993), and Wright (1994). Few samples from individual sites or polities have been large enough for intrapopulation pattern analysis. Results of my previous Copán research highlight the importance of sample size when addressing questions of intrapopulation and interpopulation variances. In my initial presentation of 25 specimens, no differences were discerned, except for a juvenile of nursing age (Reed 1991). That analysis of 12 male and 7 female adults yielded results that were interpreted as equal male and female diets (Reed 1991). In further studies, patterns of sex differences and social status variability emerged with an increase in sample size (Reed 1992,1994). Archaeobotanical research at Copán has furnished evidence for the C4 plant maize (Zea mays) and the C3 plants bean (Phaseolus vulgaris), squash (Cucurbita moschata), nance (Byrsonima crassifolia), and wild grape (Vitis sp.) (Lentz 1991). Other archaeobotanic remains identified from Copán included chayote (Sechium edule), bottle gourd (Lagenaria sp.), palm or coyol (Acrocomia mexicana), ciruela (Spondias sp.), avocado (Persea americana), zapote (Pouteria sp.), hackberry (Celtis sp.), and frijolillo (Cassia occidentalis). Many of these plants were likely to have been supplemental or famine foods (Marcus 1982). None of the Neotropical C3 cultivars—yam (Dioscorea trifida), manioc (Manihot esculenta), malanga (Xanthosoma sp.), sweet potato (Ipomoea batatas), breadnut or ramón (Brosimum alicastrum), chile peppers (Capsicum annuum), or cacao (Theobroma cacao)—have been identified among the archaeobotanical remains from Copán, although chile pepper, cacao, and palm have been identified at other ancient Maya sites (Lentz 1991, this volume). In addition, tools for processing maize, mostly rhyolite manos and metates, are ubiquitous throughout the Copán Valley in Coner phase contexts (Spink 1983).
Page 186 Table 10.1. Sample Sizes for Categories of Sites and Individuals from Copán. Sexa Group Status
Totals
F
Age at Deathb
M
U
I
J
D
YA
MA
OA
A
U
Type I
11
7
4
—
—
—
—
1
4
6
—
—
Type II
15
6
9
—
—
—
1
6
4
4
—
—
Type III
10
1
4
5
—
2
—
—
3
1
1
3
Type IV
46
23
20
3
1
1
4
10
15
14
1
—
Locale
Urban
63
29
28
6
1
1
5
15
18
19
1
3
Nonurban
19
8
9
2
—
2
—
2
8
6
1
—
Totals
82
37
37
8
1
3
5
17
26
25
2
3
a
F: female, M: male, U: unknown.
I: infant (01 years), J: juvenile (214 years), D: adolescent ( 1519 years), YA: young adult (2034 years), MA: middle adult (3550 years), OA: old adult (over 50 years), A: adult (over 20 years), U: unknown age. b
Bones or shells from several animals have been identified: deer (Odocoileus virginianus), peccary (Tayassu sp.), dog (Canis familiaris), puma or cougar (Felis concolor), jaguar (Felis onca), paca (Cuniculus paca), freshwater snail (Pachychilus corvinus or P. largillierti), sea urchin, and mussel (Feldman 1994; Gerry 1993; Pohl 1994; Zeleznik, personal communication 1996). Whittington (1989,1992) and Storey (1992, this volume) report indications of physiological stress among the elite and lowstatus subpopulations of Copán, particularly during the Coner phase. Subadults suffered extended chronic illnesses and experienced episodes of acute stress as inferred from porotic hyperostosis and enamel hypoplasia. Based on an association between high enamel hypoplasia frequencies and high mortality, Whittington (1989) suggests that the age of weaning extended until children were nearly four years old. Nutritionally related stress was experienced by all members of society regardless of social status or sex. Materials For analysis, the human bone specimens were grouped by sex, age at death, urban or nonurban site location, and social status inferred from associated site type (Table 10.1). The typology of residential architecture serves as an index of coresidential wealth and social status. It is assumed that coresidents shared relative wealth in term of living conditions, socioeconomic ties, and access to resources, including food. Age and sex assignments were based on information provided by Rebecca Storey (personal communication 1995) and Stephen Whittington (1989). Age categories were defined as infant (ages 0 to 1 years),juvenile (2 to 14), adolescent (15 to 19), young adult (20 to 34), middle adult (35 to 50), and old adult (over 50). From a collection of more than 600 individuals, 350 human bone specimens were taken from ribs when available. Deer longbone fragments were
Page 187
taken from remains excavated at site 9M,22A, and jaguar longbone fragments were obtained from a cache in the Main Group. One hundred thirtytwo Coner phase human samples with reliable age and sex identifications were analyzed for the stable carbon and nitrogen isotopic composition of their collagen. Approximately one third of the specimens yielded poorly preserved collagen and were therefore excluded from the isotopic analysis. Wellpreserved collagen was obtained from 82 humans, 7 deer, and 1 jaguar, a larger sample than presented in Reed (1992,1994) and Whittington and Reed (1997). In fact, it is the largest published Maya sample from a single polity. Methods A collagen preparation protocol for isotopic and preservation analyses was developed based on the widely used acid and base washing procedure (Ambrose 1990; DeNiro and Weiner 1988; Schoeninger and DeNiro 1984). Approximately 1 g of crushed bone was treated with 1 N HCl for 20 minutes and .125 N NaOH for 20 hours to remove acid and base soluble contaminants. An extract was produced from the washed bone by solubilizing collagen at 90°C for 15 hours and lyophilizing the filtered solution. Collagen preservation for each specimen extract was assessed by infrared spectral analysis and dry weight percent of the extract. Only specimens with an infrared spectrum similar to a modern collagen standard and a dry weight greater than 2 percent were used for isotopic analysis (Ambrose 1990; DeNiro and Weiner 1988; Goldberg 1993). A gas mixture was produced by combustion of approximately 8 mg of collagen extract in a sealed, evacuated quartz tube with cupric oxide, granular copper, and silver for 3 hours at 900°C. Dinitrogen and carbon dioxide were cryogenically separated for mass spectrometric analysis of nitrogen, carbon, and oxygen stable isotopes. The analytical reproducibility for the isotopic measurements, based on 13 samples of a collagen standard, was ±.040/00 for 13CPDB, and ±.120/00 for 15NAIR. Results Carbon and nitrogen isotopic measurements were made on 44 individuals associated with the nine patio groups of site 9N8, 38 interments associated with another 14 sites throughout the Copán polity, 7 deer, and 1 jaguar. In Figure 10.1 these isotopic values are illustrated along with those from Gerry 1993. Nitrogen values from both studies indicate that people at Copán were eating a terrestrial diet. Carbon values indicate a maize staple diet. Lack of evidence for processing tools and archaeobotanical remains for root crops, ramón, or staple cultivars other than maize further supports the inference of a maizebased diet (Lentz 1991). As shown in Figure 10.1, dogs, pacas, and peccaries were eating C4based diets. These animals could have contributed to the C4 signature in human bone. Too little faunal evidence exists to predict more than a minor contribu
Page 188
Figure 10.1. Stable carbon and nitrogen isotopic values for male and female adults from the Copán Valley with comparison to data presented by Gerry (1993).
tion of meat to the diets of the Copán Late Classic human population. For all social levels, deer have been reported as the most common faunal remains and dogs as the second most frequent (Pohl 1994). The 41 deer analyzed by myself and Gerry (1993) were C3 browsers. If deer were a major food source, as inferred from the faunal assemblage by Pohl (1994), then humans should show a stronger C3 signature than observed through isotopic analysis. One infant and one young child show more positive nitrogen isotope ratios than the other individuals in my study (Figure 10.1). They were probably nursing before death, and their isotopic values correspond to a 20/00 to 40/00 trophicrelated isotopic shift observed in nursing infants relative to weaned children (Fogel et al. 1997; Fogel et al. 1989). The juvenile was 2 to 3 years old, whereas the older two juveniles in this study, with nitrogen values similar to those of the adults, were 12 to 15 years old. These isotopic measurements indicate a dietary transition after 3 years of age, consistent with Whittington's (1992) estimate of weaning between 3.5 and 4.5 years of age. For adults of ages 35 to 50 years (MA) and over 50 years old (OA), a Student's ttest yields a statistically significant difference between the carbon isotope means of male and female adults (Table 10.2). Average carbon isotope values become increasingly negative from youngest to oldest for female adults; that is, these women ate progressively less maize (Figure 10.2). In the young adult category, three of the four female adults 25 years of age or younger have 10/00 or more positive carbon isotope ratios than the two female adults older than 25 years of age. Thus, the trend for female adults toward more negative carbon isotope values with increasing age holds within the young adult category. The more positive mean for female young adults can be accounted for by the more positive values for the youngest ones (Figure 10.2). In Figure 10.3 the urbannonurban dichotomy is illustrated with divisions by male and female. Although no statistically significant difference between the urban and nonurban sectors in mean carbon values exists (t = 1.942, p = .06), a
Page 189 Table 10.2. Stable Carbon and Nitrogen Isotope Means and One Standard Deviation for Sex and Age at Death Categories with Pooled TwoSample Student's TTest Results (total a = .05) for the Mean Differences between Male and Female Results within Age at Death Categories.
Sample Sizea
Male
Female
Age
Mean ±1 sd
Mean ±1 sd
Male
Female
Difference
p
YA
8.9 ± .5
9.0 ± .8
11
6
.1
.87
MA
8.9 ± .6
9.5 ±.6
13
12
.6
.02*
OA
9.0 ± .6
9.7 ± .5
9
16
.7
.01*
YA
7.7 ± .4
7.2 ± .3
11
6
.5
.02*
MA
7.5 ±.6
7.5 ± .5
13
12
0
.90
OA
7.6 ± .4
7.4 ±.6
9
16
.2
.40
b
13
CPDB(0/ 00)
15
NAIR(0/ 00)
a
The exclusion of missing and extreme values makes some sample sizes smaller than those found in Table 10.1.
b
Age at death. YA: young adult (2034 years), MA: middle adult (3550 years), and OA: old adult (over 50 years).
*
Statistically significant difference at a .05 level.
Figure 10.2. Average stable carbon and nitrogen isotopic values with one standard deviation bars for male and female adults in the young (YA), middle (MA), and old (OA) adult categories.
difference in spread is apparent in the scatter plots. Several individuals from the urban portion show more positive carbon values. The difference in carbon isotope values between male adults is statistically significant for urban sites (Table 10.3) but not for either sex at nonurban sites. In Figure 10.4 male and female adults from each site type are displayed in
Page 190 Table 10.3. Stable Carbon and Nitrogen Isotope Means and One Standard Deviation for Urban and Nonurban Categories with Pooled TwoSample Student's TTest Results (total a = 0.05) for the Mean Differences between Male and Female Results within Locale Types.
13
CPDB (0/ 00)
0
NAIR ( / 00)
Mean ±1 sd
Mean ±1 sd
Male
Female
Difference
p
urban
8.9 ± .6
9.4 ± .6
27
29
.5
<.01*
nonurban
9.2 ± .5
9.6 ± .3
7
8
.4
.10
.3
.2
.07
.27
urban
7.6 ± .4
7.4 ± .5
27
29
.2
.052
nonurban
7.3 ± .7
7.4 ± .5
7
8
.1
.86
within sex difference
Locale
pvalue 15
Female
within sex difference
Sample Sizea
Male
pvalue
.3
0
.20
.85
a
The exlusion of missing and extreme values makes some sample sizes smaller than those found in Table 10.1.
*
Statistically significant difference at a .05 level.
Figure 10.3. Stable carbon and nitrogen isotopic values and means with one standard deviation bars for male and female adults from urban and nonurban sites.
Page 191 Table 10.4. Stable Carbon and Nitrogen Isotope Means and One Standard Deviation for Sex and Site Type Categories with Pooled TwoSample Student's TTest Results (total a = 0.05) for the Mean Differences between Male and Female Results within Site Types.
Male
Female
Mean ±1 sd
Mean ±1 sd
I
9.0 ± .2
II
III
b
Sample Sizea
Difference
p
7
.6
.01*
8
5
0
.96
3
1
.5
—
9.5 ± .7
17
21
.8
<.01*
7.5 ± .3
7.4 ± .6
4
7
.1
.83
II
7.9 ± .5
7.3 ±.4
8
5
.6
<.01*
III
6.8 ± .6
7.0
3
1
.2
—
IV
7.6 ± .4
7.4 ± .5
17
21
.2
.24
Male
Female
9.6 + .3
4
9.2 ± .6
9.2± .4
9.5 ± .9
9.0
IV
8.7 ± .4
I
Type 13
CPDB (0/ 00)
15
NAIR(0/ 00)
The exclusion of missing and extreme values makes some sample sizes smaller than those found in Table 10.1. Social rank associated with site type from lowest Type I to highest Type IV.
a
b *
Statistically significant difference at a .05 level.
scatter plots, and mean values with standard deviations are shown in the adjacent plots. The difference between male and female adults in carbon isotope values is most distinct and statistically significant in individuals from Type I and Type IV sites. Type II sites show a statistically significant difference in nitrogen isotope values. Category statistics are listed in Table 10.4 with statistical test results for male and female mean differences. Conclusions People in the Copán Valley during the Late Classic consumed a terrestrial, primarily vegetarian, maizebased diet, an inference supported by the material remains for maize processing, archaeobotanical remains, ecological reconstructions, and the stable isotopic composition of bone collagen. The dental caries rate for those of low status at Copán is high and correlates with other populations relying on a horticultural diet (Whittington, this volume). Many of the edible fauna were C3 browsers and were unlikely contributors to the C4 signature of the humans. Animals with C4 signatures have been found only in small quantities, were poor meat sources, and therefore were unlikely contributors to the positive carbon values. Supplements from deer and other meat sources evidently were irregular and minimal during the Coner phase for the vast majority of people. Deer, the most likely meat source, appear to have been C3plant eaters, as no C4planteating deer have been noted by myself or Gerry (1993). Dietary differences are most marked in subgroups with larger sample sizes. Average isotopic subgroup differences were typically small, up to 0.8 0/00 for carbon and 0.60/00 for nitrogen. Statistically significant differences in the carbon isotope ratios between male and female adults exist for those over the age of 35 (Table 10.2), for the urban location (Table 10.3), and for the lowest and highest
Page 192
Figure 10.4. Stable carbon and nitrogen isotopic values and means with one standard deviation bars for male and female adults from Type I, II, III, and IV sites.
ranked social groups (Table 10.4). Diet for female adults shifts toward more negative carbon isotope values, or less maize consumption, with increasing age, whereas male adults maintain similar diets for all age categories. For female adults, increasing age translated into fewer C4 foods, or reduced maize consumption. Presumably, some aspect of the social system at Copán led to age and sexbased differences, which may represent differential social behavior, a result undetected from other archaeological data.
Page 193
Future Directions Future studies should incorporate more samples from the poorly represented lowerstatus and rural (nonurban) sectors of Late Classic Copán society. Alternative measures of the social status of individuals would allow for additional, and possibly more refined, analysis of intrasocial dietary variability. An isotopic study of specimens from earlier and later settlements at Copán should be undertaken to examine trends relative to agricultural intensification and demographic changes. Intrasocial differences in diet reflect patterns in aspects of stratified societies. Archaeological reconstructions of ancient societies need to incorporate large, comprehensive skeletal studies that are founded on extensive knowledge of settlement systems, ecological reconstructions, and household excavations. Small samples, although often representative of the basic diet, fail to capture the variability in social behavior and organization observed in larger samples because of the narrow separations for group differences in isotopic values. Acknowledgments The research was performed in the Department of Geosciences Mass Spectroscopy of Minerals Laboratory at The Pennsylvania State University, University Park Campus, under the supervision of Peter Deines (Department of Geosciences) and with the guidance of George Milner, David Webster, Henry Harpending, William Sanders (Department of Anthropology), and Rebecca Storey (Department of Anthropology, University of Houston). Bone specimens were removed for analysis and the results presented with the kind permission of the Instituto Hondureño de Antropología e Historia. Research funds were provided by the National Institutes of Health through a Biomedical Research Support Grant, the Hill Fellowship Fund through the Department of Anthropology, and by grants from the College of Liberal Arts, The Pennsylvania State University. Instrumentation was partially funded by National Science Foundation EAR 85 11549 grant to Peter Deines. References Cited Abrams, E. M. (1994) How the Maya Built Their World: Energetics and Ancient Architecture. Austin: University of Texas Press. Ambrose, S. H. (1990) Preparation and characterization of bone and tooth collagen for isotopic analysis. Journal of Archaeological Science 17:431451. Ambrose, S. H. (1993) Isotopic analysis of paleodiets: Methodological and interpretative considerations. In M. K. Sandford (ed.): Investigations of Ancient Human Tissue: Chemical Analyses in Anthropology. Langhorne, Pa.: Gordon and Breach, pp. 59130. Ambrose, S. H., and Norr, L. (1993) Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In J. B. Lambert and G. Grupe (eds.): Prehistoric Human Bone: Archaeology at the Molecular Level. New York: Springer, pp. 133.
Page 194
Bender, M. M. (1968) Mass spectrometric studies of carbon 13 variations in corn and other grasses. Radiocarbon 10:468472. Bender, M. M. (1971) Variations in the 13C/12C ratios of plants in relation to the pathway of photosynthetic carbon dioxide fixation. Phytochemistry 10:12391244. Bender, M. M.; Rouhani, I.; Vines, H. M.; and Black, C. C. Jr. (1973) 13C/12C ratio changes in crassulacean acid metabolism plants. Plant Physiology 52:427430. Calvin, M., and Bassham, J. A. (1962) The Photosynthesis of Carbon Compounds. New York: Benjamin. Coyston, S. L. (1995) An Application of Carbon Isotopic Analysis of Bone Apatite to the Study of Maya Diets and Subsistence at Pacbitun and Lamanai, Belize. Unpublished M.A. thesis, Trent University, Peterborough, Ontario. DeNiro, M. J. (1987) Stable isotopy and archaeology. American Scientist 75:182191. DeNiro, M. J., and Epstein, S. (1978) Influence of diet on the distribution of carbon isotopes in animals. Geochimica et Cosmochimica Acta 42:495506. DeNiro, M. J., and Epstein, S. (1981) Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta 45:341351. DeNiro, M. J., and Hastorf, C. A. (1985) Alteration of 15N/14N and 13C/12C ratios of plant matter during the initial stages of diagenesis: Studies utilizing archaeological specimens from Peru. Geochimica et Cosmochimica Acta 49:97115. DeNiro, M. J., and Weiner, S. (1988) Chemical, enzymatic, and spectroscopic characterization of collagen and other organic fractions from prehistoric bones. Geochimica et Cosmochimica Acta 52:21972206. Fash, W. L. (1991) Scribes, Warriors, and Kings: The City of Copán and the Ancient Maya. London: Thames and Hudson. Feldman, L. H. (1994) Appendix E: The mollusks of Copán. In G. R. Willey, R. M. Leventhal, A. A. Demarest, and W. L. Fash, Jr. (eds.): Ceramics and Artifacts from Excavations in the Copán Residential Zone. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 80. Cambridge: Harvard University, pp. 477479. Fogel, M. L.; Tuross, N.; Johnson, B. J.; and Miller, G. H. (1997) Biogeochemical record of ancient humans. Organic Geochemistry 27:275287. Fogel, M. L.; Tuross, N.; and Owsley, D. W. (1989) Nitrogen isotope tracers of human lactation in modern and archaeological populations. Annual Report of the Director Geophysical Laboratory of the Carnegie Institution of Washington 89:111117. Freter, A. (1992) Chronological research at Copán: Methods and implications. Ancient Mesoamerica 3:117133. Freter, A. (1994) The Classic Maya collapse at Copán, Honduras: An analysis of Maya rural settlement trends. In G. M. Schwarcz and S. E. Falconer (eds.): Archaeological View from the Countryside: Village Communities in Early Complex Societies. Washington, D.C.: Smithsonian Institution Press, pp. 160176. Gerry, J. P. (1993) Diet and Status among the Classic Maya: An Isotopic Perspective. Unpublished Ph.D. dissertation, Harvard University. Goldberg, C. (1993) The Application of Stable Carbon and Nitrogen Isotope Analysis to Human Dietary Reconstruction in Prehistoric Southern California. Unpublished Ph.D. dissertation, University of California, Los Angeles. Gonlin, N. (1993) Rural Household Archaeology at Copán, Honduras. Unpublished Ph.D. dissertation, Pennsylvania State University, University Park. Gonlin, N. (1994) Rural household diversity in Late Classic Copán, Honduras. In G. M. Schwarcz and S. E. Falconer (eds.): Archaeological Views from the Countryside: Village Communities in Early, Complex Societies. Washington, D.C.: Smithsonian Institution Press, pp. 177197. Hatch, M. D. (1976) Photosynthesis: The path of carbon. In J. Bonner and J. E. Varner (eds.): Plant Biochemistry. New York: Academic Press. Katzenberg, M. A. (1992) Advances in stable isotope analysis of prehistoric bones. In
Page 195
S. R. Saunders and M. A. Katzenberg (eds.): Skeletal Biology of Past Peoples: Research Methods. New York: WileyLiss, pp. 105119. Katzenberg, M.A.; Saunders, S. R.; and Fitzgerald, W. R. (1993) Age differences in stable carbon and nitrogen isotope ratios in a population of prehistoric maize horticulturists. American Journal of Physical Anthropology 90:267281. Keegan, W. F. (1989) Stable isotope analysis of prehistoric diet. In M. Y. Isçan and K. A. R. Kennedy (eds.): Reconstruction of Life from the Skeleton. New York: Alan R. Liss, pp. 223236. Krueger, H. W., and Sullivan, C. H. (1984) Models for carbon isotope fractionation between diet and bone. In J. R. Turnlund and P. E. Johnson (eds.): Stable Isotopes in Nutrition. Washington, D.C.: American Chemical Society, pp. 205220. Lentz, D. L. (1991) Maya diets of the rich and poor: Paleoethnobotanical evidence from Copán. Latin American Antiquity 2:269287. Marcus, J. (1982) The plant world of the sixteenth and seventeenthcentury Lowland Maya. In K. V. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. New York: Academic Press, pp. 239273. Minagawa, M., and Wada, E. (1984) Stepwise enrichment of 15N along food chains: Further evidence and the relation between Cosmochimica Acta 48:11351140.
15
N and animal age. Geochimica et
Norr, L. C. (1990) Nutritional Consequences of Prehistoric Subsistence Strategies in Lower Central America. Unpublished Ph.D. dissertation, University of Illinois, UrbanaChampaign. Norr, L. (1995) Interpreting dietary maize from bone stable isotopes in the American tropics: The state of the art. In P. W. Stahl (ed.): Archaeology in the Lowland American Tropics: Current Analytical Methods and Recent Applications. Cambridge: Cambridge University Press, pp. 198223. O'Leary, M. H. (1988) Carbon isotopes in photosynthesis. BioScience 38:328336. Pate, F. D. (1994) Bone chemistry and paleodiet. Journal of Archaeological Method and Theory 1(2):161209. Pohl, M. D. (1994) Appendix D: Late Classic Maya fauna from settlement in the Copán Valley, Honduras: Assertion of social status through animal consumption. In G. R. Willey, R. M. Leventhal, A. A. Demarest, and W. L. Fash, Jr. (eds.): Ceramics and Artifacts from Excavations in the Copán Residential Zone. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 80. Cambridge: Harvard University, pp. 459476. Reed, D. (1991) Stable isotopes and ancient Maya diet at Copán, Honduras. Paper presented at the 56th Annual Meeting of the Society for American Archaeology, New Orleans, April. Reed, D. (1992) Ancient Copán diet through stable carbon and nitrogen isotopic analysis. Paper presented at the 57th Annual Meeting of the Society for American Archaeology, Pittsburgh, April. Reed, D. M. (1994) Ancient Maya diet at Copán, Honduras, as determined through the analysis of stable carbon and nitrogen isotopes. In K. D. Sobolik (ed.): Paleonutrition: The Diet and Health of Prehistoric Americans. Center for Archaeological Investigations, Occasional Paper No. 22. Carbondale: Southern Illinois University, pp. 210221. Reed, D.; Gonlin, N.; and Webster, D. (1993) Health and wealth of Classic Maya commoners. Paper presented at the 92nd Annual Meeting of the American Anthropological Association, Washington, D.C., November. Rue, D. (1987) Early agriculture and early Postclassic Maya occupation in western Honduras. Nature 326:285286. Rue, D. J. (1989) Archaic Middle American agriculture and settlement: Recent pollen data from Honduras. Journal of Field Archaeology 16:177184.
Page 196
Schoeninger, M. J., and DeNiro, M. J. (1984) Nitrogen and carbon isotopic composition of bone collagen from marine and terrestrial animals. Geochimica et Cosmochimica Acta 48:625639. Schoeninger, M. J., and Moore, K. (1992) Bone stable isotope studies in archaeology. Journal of World Prehistory 6:247296. Schwarcz, H. P., and Schoeninger, M. J. (1991) Stable isotope analyses in human nutritional ecology. Yearbook of Physical Anthropology 34:283321. Spink, M. L. (1983) Metates as Socioeconomic Indicators during the Classic Period at Copán, Honduras. Unpublished Ph.D. dissertation, Pennsylvania State University, University Park. Storey, R. (1992) The children of Copán: Issues in paleopathology and paleodemography. Ancient Mesoamerica 3:161167. Tieszen, L. L., and Fagre, T. (1993) Effect of diet quality and composition on the isotopic composition of respiratory CO2, bone collagen, bioapatite, and soft tissues. In J. B. Lambert and G. Grupe (eds.): Prehistoric Human Bone: Archaeology at the Molecular Level. New York: Springer, pp. 121155. Tuross, N., and Fogel, M. L. (1994) Stable isotope analysis and subsistence patterns at the Sully site. In D. W. Owsley and R. L. Jantz (eds.): Skeletal Biology in the Great Plains: Migration, Warfare, Health, and Subsistence. Washington, D.C.: Smithsonian Institution Press, pp. 283289. van der Merwe, N. J. (1982) Carbon isotopes, photosynthesis, and archaeology. American Scientist 70:596606. Viel, R. (1993) Copán Valley. In J. S. Henderson and M. BeaudryCorbett (eds.): Pottery of Prehistoric Honduras: Regional Classification and Analysis. Los Angeles: UCLA Institute of Archaeology, pp. 1218. Webster, D., and Freter, A. (1990a) The demography of Late Classic Copán. In T. P. Culbert and D. S. Rice (eds.): Precolumbian Population History in the Maya Lowlands. Albuquerque: University of New Mexico Press, pp. 3761. Webster, D., and Freter, A. (1990b) Settlement history and the Classic collapse at Copán: A redefined chronological perspective. Latin American Antiquity 1:66 85. Webster, D., and Gonlin, N. (1988) Household remains of the humblest Maya. Journal of Field Archaeology 15:169190. Webster, D.; Gonlin, N.; and Sheets, P. (1997) Copán and Ceren: Two perspectives on ancient Mesoamerican households. Ancient Mesoamerica 8:4361. Webster, D.; Sanders, W. T.; and van Rossum, P. (1992) A simulation of Copán population history and its implications. Ancient Mesoamerica 3:185197. White, C. D.; Healy, P. F.; and Schwarcz, H. P. (1993) Intensive agriculture, social status, and Maya diet at Pacbitun, Belize. Journal of Anthropological Research 49:347375. White, C. D., and Schwarcz, H. P. (1989) Ancient Maya diet: As inferred from isotopic and elemental analysis of human bone. Journal of Archaeological Science 16:451474. Whittington, S. L. (1989) Characteristics of Demography and Disease in LowStatus Maya from Classic Period Copán, Honduras. Unpublished Ph.D. dissertation, Pennsylvania State University. Whittington, S. L. (1992) Enamel hypoplasia in the lowstatus Maya population of Prehispanic Copán, Honduras. Journal of Paleopathology 2:185205. Whittington, S. L., and Reed, D. M. (1997) Commoner diet at Copán: Insights from stable isotopes and porotic hyperostosis. In S. L. Whittington and D. M. Reed (eds.): Bones of the Maya: Studies of Ancient Skeletons. Washington, D.C.: Smithsonian Institution Press, pp. 157170. Wright, L. E. (1994) The Sacrifice of the Earth? Diet, Health, and Inequality in the Pasión Maya Lowlands. Unpublished Ph.D. dissertation, University of Chicago. Yoshinaga, J; Minagawa, M; Suzuki, T; Ohtsuka, R; Kawabe, T; Hongo, T; Inaoka, T; and Akimichi, T. (1991) Carbon and nitrogen isotopic characterization for Papua New Guinea foods. Ecology of Food and Nutrition 26:1725.
Page 197
Chapter 11 The Elements of Maya Diets Alkaline Earth Baselines and Paleodietary Reconstruction in the Pasión Region Lori E. Wright The investigation of ancient Maya subsistence has been a vigorous field of study, spurred on by the seeming anomaly of social complexity and high population density in a lowland tropical environment. Paleoagronomic, paleobotanical, and archaeozoological studies have now documented a diversity of wild and cultivated food resources that were exploited by the Lowland Maya and provide insight into the complex tapestry of agricultural intensification and land tenure systems that supported this culture. With few exceptions, these investigations address Maya subsistence on a large scale. Innovations in bone chemistry over the past two decades have provided new tools for the study of ancient Maya diets through chemical analysis of archaeological skeletons. These techniques permit a more detailed view of dietary behavior at the level of the individual consumer, which can be used to evaluate changing subsistence strategies and their role in Maya history. Trace elemental analysis was the first bone chemical technique used to study paleodiets, and it found one of its earliest applications in Mesoamerica. Schoeninger (1979a, 1979b) documented a systematic difference in strontium content of skeletons buried with and without jade artifacts at Preclassic Chalcatzingo. Although Schoeninger's specific dietary conclusions may require revision in light of recent reinterpretations and possible diagenetic effects, this landmark study demonstrated the value of paleodietary chemistry for addressing issues of cultural significance in Mesoamerica. Among the Maya, elemental analyses have been attempted only at Lamanai (White 1986; White and Schwarcz 1989) and Tipu (Bennett 1985). This chapter explores paleodietary reconstruction with alkaline earth elements in bone from three sites in the Pasi6n region of the Petén, Guatemala. Based on the assumption that dietary intake determines the abundance of bone elements, the method was originally developed with reference to strontium (Sr), which was hailed as an indicator of trophic level (Brown 1973) but was soon expanded to other elements (Gilbert 1975). Dietary interpretation from zinc and magnesium levels in bone is now in question (Ezzo 1994; Klepinger 1990), largely because levels of these elements are biologically mediated for cellular functions. Recent emphasis has refocused on Sr and on barium (Ba), another alkaline earth element. Both are incorporated into bone in place
Page 198
of calcium (Ca) and serve no metabolic function. Intestinal absorption discriminates against Ba and Sr in favor of Ca; therefore, relative to Ca these elements occur in lower concentrations in the tissues of animals than in their foods (Comar 1963; Comar et al. 1957; Elias et al. 1982; Price et al. 1985; Schroeder et al. 1972). This principle of ''biopurification" is the basis of the postulated use of alkaline earths as trophic indicators. Perhaps due to the larger ionic radius of Ba, which hinders substitution for Ca, this discrimination is greater against Ba than Sr, amplifying the magnitude of trophic differences. In addition, Ba is useful for distinguishing the consumption of marine foods because of its near absence in marine waters (Burton and Price 1990). Recent modeling illustrates the fact that interpretation of alkaline earth levels in bone is more complicated than first recognized (Burton and Wright 1995). Problems with the trophic level model were noted early on with respect to maize consumption by North American agriculturalists (Katzenberg 1984). As maize contains little Sr, its consumption would confound with carnivory if absolute bone Sr values are taken as the unit of analysis. But since Ba and Sr are incorporated into bone as "accidental" substitutions for Ca, focus on Ba/Ca and Sr/Ca ratios is critical. Bone Ba and Sr levels are biased by the relative contribution of each food to total alkaline earth intake (Ca + Sr + Ba). That is, bone values mirror Ba/Ca and Sr/Ca of the predominant source of dietary Ca. Differences in consumption of lowCa foods, such as meat, are often masked by highCa diet items, such as plants. Rather than a simplistic index of trophic level, Ba/Ca and Sr/Ca ratios in bone provide a means to examine the origin of Ca in prehistoric diets (Burton and Wright 1995; Runia 1987). Calcium is the most abundant mineral in the human body. In addition to its obvious role as a principal constituent of the skeleton, maintenance of Ca levels in blood is critical to nerve and muscle function (thus keeping our hearts beating), blood coagulation, and many enzymatic processes. Absorbed into the bloodstream in digestion, most dietary Ca is ultimately stored in the skeleton, which in turn buffers shortterm dietary imbalances. Changing dietary and culinary practices can disturb this homeostatic system and may affect human health. For instance, declining milk consumption by adults in North America has led to an increase in osteoporosis (Goulder and Lutwak 1988:27). Important dietary sources of Ca include dairy products, eggs, and leafy greens. In some cultures culinary mineral supplements are crucial to adequate Ca intake. By reorienting our interpretive model to focus on Ca, analysis of alkaline earth ratios in bone becomes an important avenue to examine past dietary mineral ecology. Paleodietary reconstruction with alkaline earths is dependent on accurate documentation of Ba/Ca, Sr/Ca, and, most critically, Ca concentrations of all possible dietary constituents (Burton and Wright 1995). Since alkaline earth dynamics vary widely between soil types and ecosystems, the collection of dietary baseline data for each study population is critical to the success of paleodietary reconstruction. In addition to the natural parameters that determine elemental abundances, culinary practices may also have a dramatic effect on the elemental content of diets (e.g., Kuhnlein 1981; Krause, Solomons, et al.
Page 199
1992; Krause, Tucker, et al. 1992) and thereby bone, so appropriate ethnographic analogy regarding cooking methods is important to realistic modeling of prehistoric diets. A major obstacle to the application of elemental analysis has been the evaluation of postdepositional changes to bone mineral composition and structure. Although certainly diagenesis can never be excluded as a factor contributing to elemental abundances in archaeological bone, and in some cases may be undefeatable (Tuross et al. 1989), a number of techniques have been developed to evaluate and minimize the effects of soil contamination. Multielemental analysis, bone to soil comparisons, modern to archaeological faunal comparisons, and acid leaching permit the identification of biogenic mineral composition in many instances (Price et al. 1992). This chapter presents ecosystem baseline data on alkaline earths in the Pasión region of the Petén, Guatemala, and uses these data to interpret chronological changes in human Ba/Ca and Sr/Ca at the archaeological sites of Altar de Sacrificios, Seibal, and Dos Pilas. Although the revisions to the basic principles of elemental analysis necessitate that this work be largely methodological in nature, the observed trends indicate substantial shifts in mineral ecology both spatially and in concert with major transitions in Pasión prehistory. The Pasión Maya Lowlands The Pasión region of the Guatemalan Petén has been a frequent focus of archaeological investigation that seeks to address the nature of the ninthcentury Classic Maya collapse. In this region political factors are commonly assigned priority in discussions of collapse (Demarest 1996, 1997; Sabloff and Willey 1967), although environmental and subsistence pressures have been cited as factors contributing to interpolity competition and collapse (Adams 1983; Sharer 1977; Willey and Shimkin 1973). The region is distinguished by an abundance of deciphered hieroglyphic monuments that allow detailed reconstruction of political histories at individual sites (Houston 1993; Houston and Matthews 1985; Mathews and Willey 1991; Willey 1990; Demarest 1997). The site of Altar de Sacrificios, excavated by Gordon Willey and researchers from Harvard University in the early 1960s, lies on a knoll on the south shore floodplain of the Río Pasión near its confluence with the Río Chixoy, with which it joins to form the Río Usumacinta. Excavations at Altar, which was occupied from Middle Preclassic times on, produced burials spanning a long occupational history. Maximal population density occurred in the Late Classic period. Largescale architectural construction and monument dedication came to an end near A.D. 800, presumably signaling the decline of elite dynastic governance and reduced participation in the regional political scene. Nonetheless, a large community remained at the site into Terminal Classic times until nearly A.D. 950, when the location was completely abandoned (Mathews and Willey 1991; Willey 1973). Likewise, Seibal, also excavated by Willey and colleagues in the late 1960s, has a long settlement history, from Preclassic through Terminal Classic times.
Page 200
Seibal is located on the uplifted limestone horst of the west bank of the Río Pasión, where the river takes a sharp bend to the west toward Altar and the Usumacinta downstream. A relatively small site for much of its history, Seibal had its maximal occupation during Terminal Classic times, when dynastic authority was recorded in hieroglyphic inscriptions that postdate the collapse of neighboring polities. This population growth, probably due to intraregional migration from less fortunate sites, was shortlived, however, as Seibal too was abandoned by the end of the first millennium A.D. (Mathews and Willey 1991; Willey 1990). By contrast, Dos Pilas is located inland in the center of the limestone horst that lies between the Río Salinas and Riachuelo Petexbatún, south of the Pasión. Recently excavated by Arthur Demarest and colleagues, Dos Pilas was founded only in Late Classic times, possibly by elites from Tikal. The Petexbatún dynasty periodically gained sway over several neighboring sites, including Seibal, through militaristic encounters. Ultimately, warfare against nearby sites appears to have led to the destruction and abandonment of most of the city around A.D. 760, although a small population remained behind defensive walls and near the main plaza into Terminal Classic times (Demarest 1996, 1997; Demarest et al. 1997; Palka 1995, 1997). The Pasión region is dominated today by humid broadleaf forest overlying a karstic Cretaceous limestone. This terrain is dissected by broad surface rivers that rise dramatically with rainy season runoff. Soils are primarily rendzinas, derived from limestone parent material. An important exception, however, is the immediate vicinity of Altar, where PaleoceneEocene sandstone and highland alluvial sediments predominate. Although these sediments were once postulated to have made the region a breadbasket (Adams 1983), raised fields are absent and would not have been feasible given the dramatic annual fluctuation in water tables (Dunning et al. 1997). Nonetheless, alluvial deposits along the river floodplains and in sinkholes are extremely fertile. Terracing of slopelands and walls demarcating agricultural plots indicate a complex patchwork of intensive agricultural exploitation during the Classic period (Dunning and Beach 1994; Dunning et al. 1997). Methods Sampling Strategy Investigation of elemental parameters in the Pasión ecosystem necessitated collection of a broad spectrum of ecological samples. Since elemental abundances are initially governed by soil values, soil samples were collected during excavation of all Dos Pilas burials in 199092. These also serve to evaluate diagenetic exchange between soil and archaeological bone mineral. A number of soil samples were obtained from the marrow cavities of long bones from Altar de Sacrificios skeletons. Soil had been more fully cleaned from Seibal bones, but a few samples were collected at the site. I collected edible portions of wild and cultivated plants near Dos Pilas, at Altar, at Aguateca, and from milperos in
Page 201
the Petexbatún area in 1992. These were sundried and packed in silica gel for transport. Modern animal bones were acquired from local hunters, and fish and snails were caught in streams and lagoons. Since mammalian carnivore bones are difficult to obtain, bone from several snakes was collected. In addition, faunal remains from Vanderbilt University Petexbatún Project excavations were sampled to broaden the species range examined. Moreover, comparisons of archaeological with modern faunal bones permit evaluation of diagenetic change. Samples of femoral cortical bone that appeared to be well preserved were taken from all skeletons at Altar de Sacrificios, Seibal, and Dos Pilas. Laboratory Procedures All elemental analyses were conducted by J. H. Burton at the Laboratory for Archaeological Chemistry at the University of WisconsinMadison. For archaeological bone samples, discolored surface bone was removed by mechanical abrasion with a Dremel "mototool" in order to minimize the effect of surface contaminating elements (Lambert et al. 1990). For modern samples, surface bone was not removed. Bones were then rinsed with deionized water in an ultrasonic water bath and soaked overnight in 1molar acetic acid to dissolve diagenetic carbonates (Price et al. 1992; Sillen 1989, 1990). This acid treatment would also eliminate any unlikely contamination from the rotary tool. Bone samples were ashed at 725ºC for eight hours. For each sample, 50 mg of bone ash was dissolved in 1 ml of concentrated nitric acid at 120°C for one hour, then diluted to a total volume of 17 ml (1/340 dilution) and analyzed by Inductively Coupled Plasma Emission spectroscopy (ICP AES) for Al, Ba, Ca, Fe, K, Mg, Mn, Na, P, Sr, and Zn. Detection limits for the ICP are well below the concentrations measured in bone for these elements. The N.I.S.T. bone standard, H5, and two internal laboratory standards were analyzed with each sample run to monitor accuracy and precision. Modern shell samples were cleaned with deionized water and dried. Fifty milligrams of each shell sample was dissolved in 2 ml of hot HNO3, diluted to a final volume of 17 ml (1/340 dilution) and analyzed by ICP. Plant samples were ovendried and macerated. Initially, 50 mg of each plant was digested in 2 ml of HNO3, then diluted (1/340) and analyzed for the same elements as the bone samples. Since most flora had Ba and Sr levels below the limits of reliable detection, a small subset of plants was subsequently reanalyzed in more concentrated solution as follows: 500 mg of tissue was digested in 2 ml of HNO3 at room temperature for 24 hours, then heated to 115°C for 1 hour. Diluted to a total volume of 17 ml, this solution was analyzed by ICP. Background intensities on both sides of the peak for each element were measured and subtracted from the signal intensity, allowing detection of Ba and Sr at lower concentrations. Soil samples were analyzed with a method derived from soilphosphate analysis (Burton and Simon 1993), which measures only the ions that are soluble and available for uptake by plants or interchange with bone mineral. After oven drying, 200mg samples of each sample were weighed into polypropylene
Page 202
vials. Twenty ml of 1molar hydrochloric acid was added to each vial, and the samples were kept at 26°C for two weeks with intermittent agitation. This solution was decanted and analyzed by ICP for the same set of elements. Raw elemental data are presented in detail in Wright (1994, n.d.). Because of space limitations, only patterns in the data can be addressed here. Although not a focus of this chapter, stable carbon and nitrogen isotopic analyses of collagen from the same human skeletons aid in the interpretation of elemental data presented here (Wright 1994, 1997). Statistical trends in elemental ratios in human bone are evaluated with the Mann Whitney U test because samples are small and the data are not normally distributed. Alkaline Earth Baselines for the Pasión Region The distribution of alkaline earths in any ecosystem is dependent on their abundance in underlying bedrock and the soils derived from it. In general, Pasión soils are very high in soluble Ca. Relative to Ca, both Sr and Ba are low, approximately an order of magnitude lower than in soils of the American Midwest (Burton, personal communication 1992). Yet, within the Pasión region, soils vary substantially. On average, soluble elements in Dos Pilas soils average 11 percent Ca, whereas Altar de Sacrificios soils average only 3 percent Ca. Relative to Ca, Altar soils contain more soluble Sr (mean 1000Sr/Ca = 1.17) and Ba (mean 1000Ba/Ca = 6.83) than Dos Pilas soils (Sr mean = 0.46, Ba mean = 1.61). In part, these distinctions may be due to the older Paleocene/Eocene marine sandstones that predominate at Altar, in contrast to Cretaceous limestones at Dos Pilas. Moreover, the floodplains of the Chixoy, Usumacinta, and Pasión Rivers receive a substantial contribution of alluvial sediments derived from highland volcanic soils upstream. These sediments may be more significant to Altar soil chemistry than is bedrock geology. At Seibal, soil from the site center resembles Dos Pilas soils, as we would expect from the underlying limestone. A sample from the riverbank, however, is more like Altar soils, although not so enriched in Ba, perhaps because less highland alluvium reaches the floodplain at Seibal. Although significant erosion of local hillslopes occurred during the Preclassic era, later erosion appears to have been successfully checked by agricultural terracing (Dunning and Beach 1994), so geomorphological conditions were similar in the Classic period to the modern state. Therefore, I make the uniformitarian assumption that modern values are useful analogues to past elemental dynamics. Since such differences in the base abundance of alkaline earths between sites are carried through the food web, it is evident that dietary inferences must be made on a sitespecific basis. Given these soil patterns, we expect flora, fauna, and human bone at Altar to be more enriched in Ba than at Dos Pilas. Seibal had access to agricultural soils on both these geomorphologic types, so intermediate values are anticipated. The observed alkaline earth composition of human bone ash broadly follows these expectations (see below). Calcium content, Ba/Ca, and Sr/Ca of the 24 plant species analyzed are given in Table 11.1. For many plants, Ba and Sr were below the analytic limits of
Page 203 Table 11.1. Alkaline Earth Ratios and Ca Concentrations in Pasión Plant Tissues. All values of Ca are in ppm. Scientific Name
Common Name
Location
Cultivated species
Bixa orellana
achiote
Sepens
Capsicum sp.
chile
Aguateca
Capsicum sp.
chile
Nacimiento
Caa
Cab
Ba/Cab
Sr/Cab
1640
4185
788
Cucurbita sp.
squash
Altar
Cucurbita sp.
pumpkin seeds
Nacimiento
Dioscorea alata
macal
Dos Pilas
3038
Ipomoea batatas
camote
Sepens
2505
Phaseolus vulgaris
black beans
Paso de Caribe
869
7609 573
Phaseolus vulgaris
black beans
Sepens
657
Theobroma cacao
cacao
Sepens
2013
a
1.418a
.666c
.695c
2.024 1021
c
1442
4.602a
.617c
.485
Zea mays
maize
Altar
96
238
2.899
2.689c
Zea mays
maize cob
Altar
886
490
2.245
1.510c
Zea mays
maize
Nacimiento
146
Fruits and forest species
Annona muricata
soursop fruit
Sepens
2891
Annona sp.
anona fruit
Sepens
2534
Bromelia karatas
piñuela leaf
Dos Pilas
6525
Brosimum alicastrum
ramon
Aguateca
Brosimum alicastrum
ramon
Aguateca
Brosimum alicastrum
ramon
Aguateca
1753
Brosimum alicastrum
ramon
Aguateca
2209
Brosimum alicastrum
ramon
Dos Pilas
795
Brosimum alicastrum
ramon tortilla
Aguateca
1938
Byrsonima crassifolia
nance fruit
Nacimiento
3652
Carica sp.
papaya fruit
Sepens
Dialium guianese
wild tamarind
Dos Pilas
Licania platypus
succotz fruit
Opuntia sp.
nopal cactus
Orbignya cohune
c
9072
1.330
2.304
1064
c
.611
.498c
1877
.464c
.400c
.212c
3781
.476
23565
665
Dos Pilas
1095
Aguateca
152939
corozo palm
Dos Pilas
1210
Orbignya cohune
corozo palm
Dos Pilas
1374
2405
.432
.520
Persea americana
avocado
Dos Pilas
1252
1426
.610c
.975
Pouteria mamosa
zapote fruit
Dos Pilas
1464
Pouteria mamosa
zapote fruit
Dos Pilas
2627
Pouteria mamosa
zapote nut
Dos Pilas
854
Pouteria mamosa
zapote nut
Dos Pilas
4444
Psidium guajava
guava
Dos Pilas
640
?
chapay palm
Dos Pilas
519
?
chapay palm
Dos Pilas
664
?
junco palm leaf
Dos Pilas
6284
Nondietary plants
?
waterlily
L. Petexbatún
First dilution using 50 mg of plant tissue.
b
Second dilution using 500 mg of plant tisssue, except where noted.
Values were below the detection limits, so these ratios indicate maximum values.
.431
14124
a
c
92843
1.860a
16529
.444
.672
5.318
Page 204
the ICP even when the second preparation was used, so these data should be taken as maximal estimates of the ratios. Also, Ca content is quite variable in those species for which multiple samples were analyzed. Accordingly, I interpret differences among them with some caution. Most of the plants sampled in this study were collected at Dos Pilas or in nearby villages with comparable soils. These plants fall into the Ba/Ca range of soils at Dos Pilas but show a broad spread of Sr/Ca. This may be due to slight variation in soil drainage and watertable levels. A waterlily collected in the Laguna Petexbatún contains the most Sr of all plants sampled. Water in rivers and lagoons of the Pasión ecosystem is enriched in Sr relative to Ba, due to differential leaching of Sr from soils. This pattern is especially evident in faunal alkaline earth chemistry (discussed below) but is foreshadowed in the floral data. The macal root, junco palm, and piñuela are enriched in Sr/Ca relative to other flora. These were collected adjacent to the Dos Pilas springs and may be enriched in Sr from that water in comparison with plants collected elsewhere that relied primarily on rainwater. In addition to hydrology, the effect of local soil chemistry on floral ratios is evident in the very high Ba and Sr of the squash seeds from Altar in comparison with those from Nacimiento, a village near Dos Pilas. The high Ba/Ca and Sr/Ca of maize is likely also due to the soil chemistry at Altar, where both measurable samples were grown. Such environmental variability overshadows any patterns in elemental ratios that might occur among plant groups. Analysis of a much larger series of plants would be necessary to define the full extent of variability but was not possible with the resources available. Nonetheless, the data illustrate significant variability in Ca content among foods. Maize contains much less Ca than any other plant sampled. By contrast, nopal and piñuela contain much more Ca than most plants, and their consumption would disproportionately affect total dietary Ca. Fruits generally contain little Ca; the high Ca of the wild papaya may be a result of the small green state of this sampled fruit. Moderately high Ca levels occur in chiles, achiote, ramón, and palm nuts. Other Ca rich plants that were likely consumed, but not sampled, include pacaya palm (Chamaedorea spp.), chipilin (Crotalaria longirostra), chaya (Cnidoscolus aconitifolius), and epazote (Chenopodium ambrosoides) (I.N.C.A.P. 1961). These greens may provide a significant proportion of dietary minerals because they are frequently used as seasonings in the preparation of beans and other lowCa plant foods. As with the plants, most modern faunal specimens were collected near Dos Pilas and thus are biased toward elemental distinctions of this environment. However, the shifting ranges of many animals, especially deer, peccary, felids, and birds, may even out spatial heterogeneity in soils. For instance, at Altar, faunal values may not be substantially different than at Dos Pilas or Seibal if local fauna graze beyond the floodplains or if hunting was practiced farther afield. Values of smaller fauna with fixed dens, such as agouti and armadillo, may be more locally specific. Figure 11.1 shows the alkaline earth composition of modern and archaeological faunal remains from the Pasión. Given the small sample sizes, actual data points are shown (rather than box plots) to facilitate comparison of modern and archaeological samples.
Page 205
Figure 11.1. Ba/Ca and Sr/Ca in modern ( ) and archaeological (·) faunal remains from the Pasión region.
Faunal ratios average substantially lower than ratios of flora, owing to intestinal discrimination against Ba and Sr uptake. Interspecific differences are greater for Ba than Sr, presumably because of greater discrimination against its larger ionic radius. However, the expected trophic separation is not clearly visible for either ratio. Sealy and Sillen (1988) have previously noted that trophic differences are most marked when specific predatorprey relationships are considered and may be less evident when broad behavioral classes are compared, an observation supported by the present study. The only modern large carnivore sampled, a small cat (likely ocelot, Felis pardalis), does show lower Ba/Ca than the deer and agouti. Likewise, carnivorous snakes have lower alkaline earth ratios than do these herbivores. Although often classified as herbivores, peccaries are omnivorous (Kiltie 1981) and have Ba/Ca values that are even lower than those of these carnivores. The peccary results underscore the idea that trophic differences may be obscured by consumption of highCa foods of anomalous origin. The results suggest that peccaries consume some
Page 206
Badeficient foods that are not eaten by true herbivores. Although some carnivory would contribute to these lower levels, consumption of Carich plant foods is more likely responsible. Insectivorous armadillos have Ba/Ca like that of herbivores but are somewhat Sr enriched. Two birds sampled, a domestic turkey and a tinamou eggshell (Tinamus major), are low in both ratios. Another useful distinction for paleodietary reconstruction is the composition of local fish and snails. The Ba/Ca values of freshwater fauna are comparable to those of terrestrial fauna, but mean Sr/Ca is higher. As noted above, this pattern is due to disproportionate leaching of Sr relative to Ba from soils into groundwater, which in turn enriches watershed Sr/Ca. In soil, Sr is more soluble than Ba, which is bound up in barite and soil sulfates. Cowgill and Hutchinson (1963) also documented differential leaching of Sr relative to Ba in the Bajo de Santa Fe at Tikal. Likewise, Elias et al. (1982) found higher Sr/Ca in alpine streamwater than in soil moisture, although these were similar in Ba/Ca. Unlike mammals, which obtain Ca dietarily, fish and snails take up Ca through the gills and mantle, so their values are similar and reflect water composition instead of dietary behavior. Turtle and cayman also have similar values because of consumption of aquatic flora and fauna. Although alkaline earth differences among terrestrial fauna are unlikely to be useful for the interpretation of specific meat sources in prehispanic human diets in the Pasión region, these data illustrate that animal foods have lower Ba/Ca and Sr/Ca values than plant foods do. Since consumption of terrestrial meat is unlikely to provide a significant proportion of dietary Ca, trophic effects (i.e., biopurification of Ca) maybe of limited use to paleodietary reconstruction. By contrast, the elemental composition of freshwater fauna may be more distinctive, especially if culinary methods of fish preparation result in dissolution of bone mineral or consumption of bones (e.g., Burton and Wright 1995). Mollusc consumption could provide a measurable Ca contribution since molluscs store quantities of Ca in reserves within the foot muscle for use in shell regeneration (Sminia et al. 1977; Watabe et al. 1976). Alkaline Processing of Maize: An Experimental Approach Like other cultures that consume large quantities of maize, today's Maya first process their maize in a lime solution (Katz et al. 1974). This treatment improves the nutritional quality of the grain. Interference by maize phytate with intestinal Fe and Zn absorption is effectively reduced by removal of the pericarp during alkaline processing. Proteins in maize glutelin are liberated, thereby improving the isoleucine/leucine ratio and increasing available lysine, tryptophan, and niacin (Bressani et al. 1958; Bressani and Scrimshaw 1958). Without alkaline processing, amino acid imbalances in maize protein would lead to pellagra if maize was an important dietary source of protein. Alkaline processing also produces a dramatic increase in the calcium content of maize, which is otherwise Capoor. Limetreated maize is the predominant dietary source of Ca in modern Maya diets (Krause, Solomons, et al. 1992; Krause, Tucker, et al. 1992), and lime Ca is readily absorbed from tortillas in di
Page 207
gestion (SernaSaldivar et al. 1991, 1992). We can assume that the ancient Maya also treated maize with lime solutions (e.g., Nations 1979), as do their modern descendants. A variety of treatment methods have been reported for the Maya area which may differ in effect on dietary Ca and Ba/Ca and Sr/Ca ratios and are critical considerations for paleodietary reconstruction with alkaline earths. A variety of maize products are prepared with different treatment methods and with lime from different sources, and certain ritual maize foods are not treated with lime (GuiterasHolmes 1961; Redfield and Villa Rojas 1934; Tozzer 1907). White and Schwarcz (1989) note the presence of limeencrusted colander pots at Lamanai and suggest these were used for maize processing. Although this ceramic form is extremely rare in the Petexbatún assemblage (Foias 1996), this does not imply an absence of alkaline treatment since colanders are not requisite for this procedure. To evaluate the effect of alkaline processing on dietary Ca intake among the Maya, I processed three varieties of maize with lime using two traditional methods. The varieties are white maize grown on the Guatemalan Pacific coastal plain, yellow maize grown near Chimaltenango, and black maize grown near San Juan Comalapa. I prepared each variety with traditional Guatemalan "nixtamal" methods for tortillas or tamales (Bressani and Scrimshaw 1958) by boiling it in a 0.5 percent solution of lime made from burned limestone (sold for this purpose in Antigua, Guatemala) for one hour. I left this mixture to soak overnight, after which the nixtamal was rinsed clean. Using a second, Yucatec method for preparation of posole (Redfield and Villa Rojas 1934:3839), I boiled maize in a 0.5 percent lime solution for 1 hour, then reboiled it in clean water for two hours. I also boiled each variety in clear tap water to control for the effects of water boiling alone. The Lacandon Maya prefer slaked Pachychilus snail shells as a source of lime (Nations 1979; Tozzer 1907). In lieu of Pachychilus, I slaked a Pomacea shell at 900°C and treated yellow maize with this lime using the two methods described above. Treated maize samples were ovendried and analyzed by ICP spectroscopy. Both alkaline processing methods produced a dramatic increase in maize Ca content (Table 11.2), although boiling in hard Chicago tap water alone also raised Ca substantially. A twofactor ANOVA on maize variety and treatment with burned limestone (water, nixtamal, posole) indicates a significant difference in Ca content between the two treatment methods (F = 10.17, p = .008), three maize varieties (F = 3.53, p = .062), and the interaction of variety and treatment (F = 3.47, p = .065). Nixtamal preparation raised maize Ca by 9 to 20 times, whereas preparation for posole raised Ca by only 7 to 11 times. These two treatment methods have similar effects on black maize, which absorbs the least Ca of the varieties tested. Yellow and white maize absorb much more Ca in nixtamal preparation, but most of this rinsed out from yellow maize in posole preparation, whereas substantial Ca is retained by white maize. Alkaline processing with Pomacea lime raised the Ca content of yellow maize by approximately 6 times when prepared for posole and by 10 times when prepared for nixtamal. These increments are comparable to the results with limestone and probably within the range of idiosyncratic variation. Thus
Page 208 Table 11.2. Effects of Alkaline Processing on the Ca Content of Maize. Maize
Treatment
Lime Type
N
Mean Ca (ppm)
white
unprocessed
—
3
a
white
water boiled
—
2
608
white
lime soaked
limestone
3
2027
white
lime/water
limestone
3
1144
black
unprocessed
—
3
a
black
water boiled
—
2
680
black
lime soaked
limestone
3
863
black
lime/water
limestone
3
952
yellow
unprocessed
—
3
a
yellow
water boiled
—
2
572
yellow
lime soaked
limestone
3
1890
yellow
lime/water
limestone
3
687
yellow
water boiledb
—
1
492
yellow
lime soaked
b
Pomacea
1
2059
yellow
b
Pomacea
1
1327
lime/water
a
Ca levels were below the ICP analytical detection limits of 100 ppm.
b
Boiled in Hamilton municipal water instead of Chicago water.
we can assume that the effect of alkaline processing on Ca content is comparable whether slaked limestone or snail shells are used as the source of lime. These results illustrate that alkaline processing has a dramatic impact on the interpretation of maize consumption using alkaline earth levels in bone. Nixtamal preparation of maize for tamales would raise maize to the Ca level of other Pasión plants or slightly higher, whereas posole preparation would have a lesser effect. Regardless of the exact Ca increment, Ba/Ca and Sr/Ca of processed maize become those of the lime used. With lime from burned limestone, alkaline processing would increase absolute Ba/Ca and Sr/Ca ratios to values higher than those in other dietary components. For lime made from slaked snail shells, a differential increase in Sr/Ca is expected relative to Ba/Ca, as discussed above for the freshwater ecosystem. A third possible source of lime, which remains unexplored, is plant ash. The Lacandon use mahogany bark lye only when cooking a specific maize drink, saqnum (Tozzer 1907:51). Since wood ash alkali is typically potassium hydroxide (KOH), not calcium oxide (CaO) (Katz et al. 1974:770), I hypothesize that this lye would not affect alkaline earth ratios of treated maize substantially. It is important to note also that culinary practices that do not involve mineral additives may also change elemental composition of foods. Water boiling produced a substantial change in maize chemistry as well. Differential use of water from sources of varied composition (e.g., rainwater versus river or ritual cave water) might also contribute to variability among human bone elemental ratios. Mineral grit from grinding stones could also affect bone ratios, although the magnitude of this effect is likely to be much smaller than that from alkaline treatment. Although a few sandstone metates are reported from Altar, utilitar
Page 209
ian metates in the Pasión region are limestone and would affect maize Ca in a manner indistinguishable from that of alkaline processing. A few small metates of imported volcanic stone were probably used to grind spices or pigments, not maize, and would contribute less Ca to the diet because of their greater hardness and low Ca content. Moreover, dental attrition is slight in these skeletons, indicating a relatively lowgrit diet. Evaluating Diagenesis Before we proceed to interpret Pasión human diets, the possibility of diagenetic alteration of archaeological bone mineral must be evaluated. An extended treatment of this critical issue is beyond the scope of this chapter; I refer readers to more detailed discussion in Wright (1994:239254 and n.d.). First, calcium/phosphorus ratios (Ca/P) were scrutinized to evaluate the overall integrity of bone mineral (Sillen 1989; White and Hannus 1983). Pasión bones have Ca/P ratios ranging from 2.08 to 2.32, very close to the theoretical value for hydroxyapatite, 2.15. Moreover, absolute content of Ca (37.140.6 percent) and P (17.218.6 percent) are close to theoretical values (Ca: 39.9 percent; P: 18.5 percent). Second, diagenesis was evaluated by comparing elemental values in bone with adjacent soil at Altar and Dos Pilas (Nelson and Sauer 1984). At both sites Spearman's rank order correlation coefficient is low and not statistically significant for all elements, although rho for Ba approaches significance at Altar (p = .08). At Dos Pilas soil was collected at 0, 10, and 20 cm from sampled bones so soil exchange could be studied in greater detail (Lambert et al. 1984). Mann Whitney U tests find no differences in elemental composition of soils at 0 versus 10 cm or at 0 versus 20 cm from bone, indicating that elemental transport between bone and soil was not sufficient to affect soil concentrations or that bone/soil exchange has reached an equilibrium state at Dos Pilas. For faunal bone, comparisons between modern and archaeological specimens provide a third avenue to investigate diagenesis (Ezzo 1992; Sillen 1981). Figure 11.1 shows similar Ba and Sr values in archaeological deer, agouti, and turtle to those found in modern specimens. Archaeological felids broadly bracket modern cat Ba/Ca but are comparable in Sr. Peccary show the most conclusive evidence for diagenesis, with both archaeological specimens enriched in Ba relative to modern ones. One of these peccaries and the highBa felid show high levels of Mn, Fe, and Al, indicating significant contamination from soil. These results illustrate that levels of soil elements may herald diagenetic changes in alkaline earths. To examine elemental patterning in human bone, I conducted principle factoring on Pearson's correlation matrices for elemental data at each site (Buikstra et al. 1989). These analyses demonstrate the strong relationship between the soil contaminants Al, Fe, K, and Mn at Altar and Dos Pilas. At neither site do alkaline earths correlate with these elements. Rather, they load on different factors. At Seibal the first principle factor indicates a relationship between Ba, Sr, and Mn that reflects both the basic chemical similarity of the alkaline
Page 210
earths and incorporation of Ba into microbial manganese oxides on bone surfaces. Since high levels of soil contaminants identify faunal bone with anomalous Ba, I eliminated 11 samples with high levels of these elements from further consideration (Al > 900 ppm, Fe > 300 ppm, K > 400 ppm, Mn > 100 ppm) and repeated the factor analysis. Elimination of these contaminated samples removed the BaMn correlation at Seibal. At both Dos Pilas and Seibal the alkaline earths load on the first factors, accounting for most variability in these rotated matrices. At Altar Al and Fe contamination remains the first factor but is independent of relationships among dietarily informative elements. In no sample does a BaMn correlation document oxide contamination. Diagenetic change of bone mineral is documented in the Pasión samples, but the worst cases are easily identified by high levels of soil contaminants not normally found in bone. Exclusion of these samples minimizes the role of contamination in structuring the elemental data. Nonetheless, diagenesis remains a possible factor in the ensuing paleodietary interpretation of the alkaline earths in Pasión bones. Paleodiet in the Pasión Figure 11.2 illustrates the alkaline earth composition of human bone from the three sites in comparison with soil samples from each. Strictly speaking, this comparison between soil and bone is not valid because the amount of Sr and Ba leached from soil using the acid extraction protocol does not necessarily correspond to the amount of these elements that would be leached in soil water and therefore available for incorporation into plants. It is evident, however, that intersite patterning in human bone alkaline earth ratios parallels these soil patterns. Like Altar soils, Altar human bone mineral is enriched in Sr and Ba relative to bones at Seibal and Dos Pilas. Since stable isotope ratios indicate very similar diets between Late Classic sites (Wright 1994,1997), the differences in elemental composition of bone between sites are not due to dramatic dietary differences. Instead, they reflect environmental differences that are carried through the foodweb. Moreover, the fact that human bone parallels the soil patterning suggests that that most foodstuffs were consumed near their source of production. The alkaline earth composition of human skeletons from Altar de Sacrificios is illustrated in Figure 11.3. A chronological trend in Ba/Ca ratios is evident. The highest Ba/Ca ratios occur in the Preclassic, and the lowest occur in the Early and Late Classic periods. Terminal Classic Ba/Ca is lower than Preclassic Ba/Ca but appears to show a slight rise over Late Classic values. The Mann Whitney U test finds that only the Preclassic Ba/Ca is statistically different from the Terminal Classic mean (p < .004). However, the Preclassic to Early Classic decline approaches significance (p = .11). If the Early and Late Classic samples are considered together, they differ significantly from the Terminal Classic mean (p = .01) and approach a significant difference when compared to the Preclassic mean (p = .06). Sr/Ca do not show a particularly strong trend over time, and no phase
Page 211
Figure 11.2. Ba/Ca and Sr/Ca in soils and human bone from the Pasión region.
Figure 11.3. Chronological patterns in elemental ratios at Altar de Sacrificios.
differs significantly from others. In contrast to Preclassic and Terminal Classic periods, which show distinct clustering of both alkaline earth ratios, Early and Late Classic bones are broadly scattered across the graph. In particular, Late Classic samples scatter toward high Sr/Ca. Ba/Ca at Altar is negatively correlated with stable carbon isotopic composition or shifts (Wright 1994). Since 13C changes are best attributed to
13
C (rho = 0.33, p = 0.07), and these chronological shifts in Ba/Ca track
13
C
Page 212
changes in maize consumption, Ba/Ca shifts must likewise reflect the changing role of maize in Altar diets. Ordinarily, if bone Ba/Ca was disproportionately derived from lime for alkaline processing of maize, and burned limestone was used, we would expect a positive correlation instead. At first glance, this result argues against the use of limestone in maize preparation at Altar de Sacrificios. Note that limestone is not locally available at Altar, although it may be obtained a few kilometers away. Since imported limestone is expected to have lower Ba/Ca than soils at Altar, however, we cannot exclude the possibility that the lower Ba/Ca of high maize diets at Altar is due to alkaline processing with lime, and that high 13C corresponds directly to a low Ba/Ca of lime with increasing maize consumption in comparison with the high Ba/Ca of other locally grown flora. Alternate sources of lime are from burned shells—perhaps the Amblema clam used in mortar for Altar constructions (Smith 1972)—or plant ashes. Mollusc lime would not raise Ba/Ca to such an extent but should produce elevated Sr/Ca. This may be seen in some Late Classic bones, but the overall lack of Sr/Ca correlation with 13C makes it unlikely that a use of slaked shells explains the observed pattern. If maize at Altar was not processed with alkali or if plant ashes were used (i.e., potassium hydroxide), then Ba/Ca shifts over time could be due to changes in the relative visibility of meat signals in the total dietary Ba/Ca with changing proportions of maize versus other Carich plants. If fauna consumed at Altar were hunted beyond the floodplain, meat would have disproportionately low Ba/Ca relative to plant foods grown on the Barich flood plain. This divergence might accentuate the magnitude of ''trophic effect" on Ba/Ca more so than Sr/Ca, which otherwise is likely to be obscured by dietary Ca from plant foods (see Burton and Wright 1995). However, the change would be due to variation in plant consumption, not to changes in the importance of hunting. Stability in 15N ratios over time implies that meat consumption did not fluctuate (Wright 1994,1997). The greater variability in elemental ratios of the Early and Late Classic periods may be due to a change in methods of alkaline treatment of maize or to increased heterogeneity in diet among subgroups of the more socially diverse occupation at this time. The use of different methods of food preparation between social groups is also likely. Diagenesis cannot be excluded either. It is interesting to note that the two most elaborate Late Classic burials (Burials 88 and 128) show the highest 15N value and high elemental ratios, implying dietary distinctions with social status (Wright 1994). In contrast to the situation at Altar, neither Ba/Ca nor Sr/Ca is correlated with 13C at Seibal (Wright 1994). This suggests that lime for preparation of maize products was produced from locally abundant limestone. Moreover, although cultivation of the Barich floodplain soils results in higher mean human bone values at Seibal than at Dos Pilas, average plant and animal values would differ less than at Altar because of the much narrower floodplain and the cultivation of upland (lowBa) slopes. Dunning (personal communication 1991) has recently identified hillslope terraces at Seibal. At Seibal, Preclassic and Terminal Classic alkaline earth ratios in human
Page 213
Figure 11.4. Chronological patterns in elemental ratios at Seibal.
bone are very similar (Figure 11.4). As at Altar, Late Classic bones show a broader elemental spread, especially in Sr/Ca, whereas Ba/Ca and Sr/Ca are highly correlated in Preclassic and Terminal Classic skeletons. The higher Sr/Ca of Late Classic bones is statistically different from Terminal Classic Sr/Ca as tested with the Mann Whitney U test (p = 0.03). A shift to use of slaked snail lime for alkaline processing or a dietary shift to consumption of greater quantities of freshwater versus terrestrial fauna might contribute to this pattern. As at Altar, diagenetic enrichment of Sr cannot be ruled out. Greater fish consumption is supported by the heavier 15 N values among Late Classic Seibal burials (Wright 1994). At Dos Pilas, bone mineral of four Terminal Classic skeletons shows higher Sr/Ca and Ba/Ca than in Late Classic skeletons (Figure 11.5). The difference is statistically significant for both elements (both p = .03). Terminal Classic bone is also isotopically lighter in carbon (though not statistically so), but alkaline earth ratios are not correlated with isotopic composition at this site (Wright 1994). If the elemental ratios reflect the same dietary shift, this may imply that maize was untreated or was treated with lye, not lime, and so contained little Ca. If so, a decline in maize consumption resulted in increased ratios from consumption of Carich C3 plants. But since the isotopic changes are not significant, a shift in relative consumption of other plant foods might have brought about this result. Given the heterogeneity of elemental ratios in floral samples collected at Dos Pilas, it is not possible to specify dietary shifts responsible for this pattern. The tight correlation between Ba and Sr in the dataset implies little variation in exploitation of aquatic foods and that slaked shells were not used as a source of lime. Variation in consumption of plant foods among social groups at Late Classic Dos Pilas is indicated by clustering of elemental ratios among burials
Page 214
Figure 11.5. Chronological and social patterns in elemental ratios at Dos Pilas.
differing in grave facility, skeletal position, orientation, and accompaniments (Wright 1994). As at Altar, the most elite ("royal") burials have high elemental ratios. These are the hieroglyphically identified Ruler 2 (Burial 30) and the "Woman of Cancuen" (Burial 20). Curiously, lowstatus burials are also somewhat high, whereas intermediatestatus skeletons have lower Ba/Ca and Sr/Ca. If only the lowerstatus population remained at Dos Pilas after its military defeat in A.D. 760, this change in population composition might account for the observed elemental shift without any true dietary change. The precise nature of social distinctions in diet is more difficult to identify, but a heavier reliance on cultivated staples by the lower strata of society is probable, with maize lime contributing to the high elemental ratios. Conclusions In this chapter I have explored the potential of alkaline earth elements for paleodietary reconstruction in the Pasión region of El Petén, Guatemala. This goal has proved more challenging than I had hoped. When the interpretation of Ba and Sr levels in bone is reoriented to focus on dietary sources of Ca, paleodietary inferences are more realistic than when framed by a trophic level model, but the endeavor is also more complicated. In particular, paleodietary reconstruction with alkaline earths in the Maya area is hindered by the extremely low natural abundance of Ba and Sr in this ecosystem. These low levels are presumably due to the originally low content of Ba and Sr in marine sedimentary carbonates, from which most lowland soils are derived, in combination with the high degree of leaching by heavy tropical rains. Compounding this is the technical difficulty of measuring such low concentrations of these elements. In the Pasión the interpretive power of this ap
Page 215
proach was compromised by the fact that most plants have Ba and Sr levels below the detection limits of ICP, which is perhaps the most sensitive analytical tool available. This difficulty might be circumvented by use of a different sample dilution protocol and is probably not insurmountable. These caveats apply only to the floral data; for bone, the methods function well. Baseline data collected in this study show considerable scatter. The interpretive power of these data would be substantially improved by a larger, more representative series of samples from each faunal and floral species. It is difficult to differentiate idiosyncratic from speciesspecific variability in these small samples. Additionally, replicate sampling of flora and fauna across distinct soil zones would be beneficial. The evidence for substantial variation in baseline data between geomorphologic zones documented here underscores the need to study elemental ratios in bone at the local site level and to collect local baseline data. In the present case, local variability in soil chemistry is paralleled in human bone data. Diagenesis does not provide an adequate explanation for this intersite patterning. The covariation among elemental and isotopic data at Altar as contrasted with Dos Pilas and Seibal provides support for the idea that most foods consumed were locally grown and not traded over long distances. This finding has implications for models of Maya agronomy that see trade of staple foods as a key factor permitting dense population in the Maya Lowlands. Paleodietary reconstruction with alkaline earths in the Pasión region is most challenged by the menu complexity of ancient Maya diets. Unlike isotopic analysis, which is able to target narrow classes of foods, discrimination among a large variety of potential sources of dietary calcium is difficult. In complex diets there are too many possible dietary variables to identify with confidence the menu changes responsible for shifts in elemental composition. The study of elemental ratios in conjunction with isotopic analysis is beneficial to such interpretation. Insight from material culture into culinary practices would also improve the success of dietary inference. Nonetheless, alkaline earths have potential to reveal past dietary behavior. Among the prehispanic Maya remains studied here, a number of statistically confirmed trends were documented during the occupations of the three sites. Diagenesis does not provide a clear explanation for these changes, although it cannot be definitively excluded from consideration. Socially meaningful patterning in the elemental composition of diets is evident, and changes in the overall structure of the elemental datasets at each site imply changes in the nature of social partitioning of foods over time. Although the precision of interpretation about food choice is limited, these results enhance our understanding of ancient subsistence behavior in the Pasión Maya Lowlands. Acknowledgments The research reported here was funded by an NSF dissertation improvement grant (BNS9112561) and a Wenner Gren Foundation predoctoral grant (#5447). I thank A. Demarest for the opportunity to participate in Vanderbilt
Page 216
University's Petexbatún Regional Archaeological Project, and the Instituto de Antropología e Historia de Guatemala and the Peabody Museum of Harvard University for permission to analyze bone samples. Jim Burton has been a guiding force in this research, and I thank him for his generous advice and for conducting the laboratory analyses. References Cited Adams, R. E. W. (1983) Ancient land use and culture history in the Pasión River region. In E. Z. Vogt and R. M. Leventhal (eds.): Prehistoric Settlement Patterns. Albuquerque: University of New Mexico Press, pp. 319335. Bennett, S. L. (1985) Diet, Health, and Status at Tipu. Unpublished M.A. thesis, State University of New York, Plattsburgh. Bressani, R.; Paz y Paz, R.; and Scrimshaw, N. S. (1958) Chemical changes in corn during preparation of tortillas. Agricultural and Food Chemistry 6(10): 770 774. Bressani, R., and Scrimshaw, N. (1958) Effect of lime treatment on in vitro availability of essential amino acids and solubility of protein fractions in corn. Agricultural and Food Chemistry 6(10): 774778. Brown, A. B. (1973) Bone Strontium Content as a Dietary Indicator in Human Skeletal Populations. Unpublished Ph.D. dissertation, University of Michigan, Ann Arbor. Buikstra, J. E.; Frankenberg, S.; Lambert, J. B.; and Xue, L. (1989) Multiple elements: Multiple expectations. In T. D. Price (ed.): The Chemistry of Prehistoric Human Bone. Cambridge: Cambridge University Press, pp. 155210. Burton, J. H., and Price, T. D. (1990) The ratio of barium to strontium as a paleodietary indicator of consumption of marine resources. Journal of Archaeological Science 17:547557. Burton, J. H., and Simon, A. W. (1993) Acid extraction as a simple and inexpensive method for compositional characterization of archaeological ceramics. American Antiquity 58(1):4559. Burton, J. H., and Wright, L. E. (1995) Nonlinearity in the relationship between bone Sr/Ca and diet: Paleodietary implications. American Journal of Physical Anthropology 96(3):273282. Comar, C. L. (1963) Some overall aspects of strontiumcalcium discrimination. In R. H. Wasserman (ed.): The Transfer of Calcium and Strontium across Biological Membranes. New York: Academic Press, pp. 405418. Comar, C. L.; Wasserman, R. H.; Ulberg, S.; and Andrews, G. A. (1957) Strontium metabolism and strontium/calcium discrimination in man. Proceedings of the Society for Experimental Biology and Medicine 95:386391. Cowgill, U. M., and Hutchinson, G. E. (1963) El Bajo de Santa Fé. Transactions of the American Philosophical Society 53(7):150. Demarest, A. A. (1996) War, peace, and the collapse of a native American civilization. In T. Gregor (ed.): A Natural History of Peace. Nashville: Vanderbilt University Press, pp. 215248. Demarest, A. A. (1997) The Vanderbilt Petexbatún Regional Archaeological Project, 19891994: Overview, history, and major results of a multidisciplinary study of the Classic Maya collapse. Ancient Mesoamerica 8:209227. Demarest, A. A.; O'Mansky, M.; Wolley, C.; Van Tuerenhout, D.; Inomata, T.; Palka, J.; and Escobedo, H. (1997) Classic Maya defensive systems and warfare in the Petexbatún region: Archaeological evidence and interpretations. Ancient Mesoamerica 8:229253. Dunning, N. P., and Beach, T. (1994) Soil erosion, slope management, and ancient terracing in the Maya Lowlands. Latin American Antiquity 5:5169.
Page 217
Dunning, N.; Beach, T.; and Rue, D. (1997) The paleoecology and ancient settlement of the Petexbatún region, Guatemala. Ancient Mesoamerica 8:25566. Elias, R. W.; Hirao, Y.; and Patterson, C. C. (1982) The circumvention of the natural biopurification of calcium along nutrient pathways by atmospheric inputs of industrial lead. Geochimica et Cosmochimica Acta 46:25612580. Ezzo, J. A. (1992) A test of diet versus diagenesis at Ventana Cave, Arizona. Journal of Archaeological Science 19:2337. Ezzo, J. A. (1994) Zinc as a paleodietary indicator: An issue of theoretical validity in bonechemistry analysis. American Antiquity 59(4):606621. Foias, A. (1996) Changing Ceramic Production and Exchange Systems and the Classic Maya Collapse in the Petexbatún Region. Unpublished Ph.D. dissertation, Vanderbilt University, Nashville. Gilbert, R. I. (1975) Trace Element Analysis of Three Skeletal Amerindian Populations at Dickson Mounds. Unpublished Ph.D. dissertation, University of Massachusetts, Amherst. Goulder, L., and Lutwak, L. (1988) The Strong Bones Diet. Gainesville, Fla.: Triad. GuiterasHolmes, C. (1961) Perils of the Soul: The World View of a Tzotzil Indian. New York: Free Press. Houston, S. D. (1993) Hieroglyphs and History at Dos Pilas: Dynastic Politics of the Classic Maya. Austin: University of Texas Press. Houston, S. D., and Mathews, P. (1985) The Dynastic Sequence of Dos Pilas, Guatemala. Monograph No. 1 San Francisco: PreColumbian Art Research Institute. I.N.C.A.P. (1961) Food Composition Table for Use in Latin America. Bethesda, Md.: National Institutes of Health. Katz, S. H.; Hediger, M. L.; and Valleroy, L. A. (1974) Traditional maize processing techniques in the New World. Science 184:765773. Katzenberg, M. A. (1984) Chemical Analysis of Prehistoric Human Bone from Five Temporally Distinct Populations in Southern Ontario. Mercury Series, Archaeological Survey of Canada, Paper No. 129. Ottawa: National Museum of Man. Kiltie, R. A. (1981) Stomach contents of rain forest peccaries (Tayassu tajacu and T pecari). Biotropica 13:234236. Klepinger, L. L. (1990) Magnesium ingestion and bone magnesium concentration in paleodietary reconstruction: Cautionary evidence from an animal model. Journal of Archaeological Science 17:513517. Krause, V. M.; Solomons, N. W.; Tucker, K. L.; LopezPalacios, C. Y.; Ruz, M.; and Kuhnlein, H. V. (1992) Ruralurban variation in the calcium, iron, zinc, and copper content of tortillas and intake of these minerals from tortillas by women in Guatemala. Ecology of Food and Nutrition 28:279288. Krause, V. M.; Tucker, K. L.; Kuhnlein, H. V.; LopezPalacios, C. Y.; Ruz, M.; and Solomons, N. W. (1992) Ruralurban variation in limed maize use and tortilla consumption by women in Guatemala. Ecology of Food and Nutrition 28:289297. Kuhnlein, H. V. (1981) Dietary mineral ecology of the Hopi. Journal of Ethnobiology 1(1):8494. Lambert, J. B.; Simpson, S. V.; Buikstra, J. E.; and Charles, D. K. (1984) Analysis of soil associated with Woodland burials. In J. B. Lambert (ed.): Archaeological Chemistry—III. Advances in Chemistry Series, No. 205. Washington, D.C.: American Chemical Society, pp. 97113. Lambert, J. B.; Weydert, J. M.; Williams, S. R.; and Buikstra, J. E. (1990) Comparison of methods for the removal of diagenetic material in buried bone. Journal of Archaeological Science 17:453468. Mathews, P., and Willey, G. R. (1991) Prehistoric polities of the Pasión region: Hieroglyphic texts and their archaeological settings. In T. P. Culbert (ed.): Classic Maya Political History: Hieroglyphic and Archaeological Evidence. New York: Cambridge University Press, pp. 3071.
Page 218
Nations, J. D. (1979) Snail shells and maize preparation: A Lacandon Maya analogy. American Antiquity 44:568571. Nelson, D. A., and Sauer, N. J. (1984) An evaluation of postdepositional changes in the trace element content of human bone. American Antiquity 49(1):144147. Palka, J. W. (1995) Classic Maya Social Inequality and the Collapse at Dos Pilas, Petén, Guatemala. Unpublished Ph.D. dissertation, Vanderbilt University, Nashville. Palka, J. W. (1997) Reconstructing Classic Maya socioeconomic differentiation and the collapse at Dos Pilas, Petén, Guatemala. Ancient Mesoamerica 8:293306. Price, T. D.; Blitz, J.; Burton, J.; and Ezzo, J. A. (1992) Diagenesis in prehistoric bone: Problems and solutions. Journal of Archaeological Science 19:513529. Price, T. D.; Connor, M.; and Parsen, J. D. (1985) Bone chemistry and the reconstruction of diet: Strontium discrimination in whitetailed deer. Journal of Archaeological Science 12:419442. Redfield, R., and Villa Rojas, A. (1934) Chan Kom: A Maya Village. Chicago: University of Chicago Press. Runia, L. T. (1987) Strontium and calcium distribution in plants: effect on paleodietary studies. Journal of Archaeological Science 14:599608. Sabloff, J. A., and Willey, G. R. (1967) The collapse of Maya civilization in the Southern Lowlands: A consideration of history and process. Southwest Journal of Anthropology 23:311336. Schoeninger, M. (1979a) Dietary Reconstruction at Chalcatzingo, a Formative Period Site in Morelos, Mexico. Museum of Anthropology, Technical Reports, No. 9. Ann Arbor: University of Michigan. Schoeninger, M. J. (1979b) Diet and status at Chalcatzingo: Some empirical and technical aspects of strontium analysis. American Journal of Physical Anthropology 51:295310. Schroeder, H. A.; Tipton, I. H.; and Nason, A. P. (1972) Trace metals in man: Strontium and barium. Journal of Chronic Diseases 25:491517. Sealy, J. C., and Sillen, A. (1988) Sr and Sr/Ca in marine and terrestrial foodwebs in the Southwestern Cape, South Africa. Journal of Archaeological Science 15:425438. SernaSaldivar, S. O.; Rooney, L. W.; and Greene, L. W. (1991) Effect of lime treatment on the availability of calcium in diets of tortillas and beans: Rat growth and balance studies. Cereal Chemistry 68(6):565570. SernaSaldivar, S. O.; Rooney, L. W.; and Greene, L. W. (1992) Effects of lime treatment on the bioavailability of calcium in diets of tortillas and beans: Bone and plasma composition in rats. Cereal Chemistry 69(1):7881. Sharer, R. J. (1977) The Maya collapse revisited: Internal and external perspectives. In N. Hammond (ed.): Social Process in Maya Prehistory. London: Academic Press, pp. 532552. Sillen, A. (1981) Strontium and diet at Hayonim Cave. American Journal of Physical Anthropology 56:131137. Sillen, A. (1989) Diagenesis of the inorganic phase of cortical bone. In T. D. Price (ed.): The Chemistry of Prehistoric Human Bone. Cambridge: Cambridge University Press, pp. 211229. Sillen, A. (1990) Response to N. Tuross, A. K. Behrensmeyer, and E. D. Eanes. Journal of Archaeological Science 17:595596. Sminia, T.; De With, N. D.; Bos, J. L.; van Nieuwmegen, M. E.; Witter, M. P.; and Wondergem, J. (1977) Structure and function of the calcium cells of the freshwater pulmonate snail Lymnaea stagnalis. Netherlands Journal of Zoology 27(2):195208. Smith, A. L. (ed.) (1972) Excavations at Altar de Sacrificios: Architecture, Settlement, Burials, and Caches. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 62, No. 2. Cambridge: Harvard University. Tozzer, A. M. (1907) A Comparative Study of the Maya and the Lacandon. New York: Macmillan.
Page 219
Tuross, N.; Behrensmeyer, A. K.; and Eanes, E. D. (1989) Strontium increases and crystallinity changes in taphonomic and archaeological bone. Journal of Archaeological Science 16:661672. Valdés, J. A.; Foias, A.; Inomata, T.; Escobedo, H.; and Demarest, A. A. (1993) Proyecto Arqueológico Regional Petexbatún. Informe Preliminar No. 5. Quinta Temporada. On file at the Instituto de Antropología e Historia de Guatemala. Watabe, N.; Meenakashi,V. R.; Blackwelder, P. L.; Kurtz, E. M.; and Dunkelberger, D. G. (1976) Calcareous spherules in the gastropod Pomacea paludosa. In N. Watabe and K. M. Wilbur (eds.): The Mechanisms of Mineralization in the Invertebrates and Plants. Columbia: University of South Carolina Press, pp. 283 308. White, C. D. (1986) Paleodiet and Nutrition of the Ancient Maya at Lamanai, Belize: A Study of Trace Elements, Stable Isotopes, Nutritional and Dental Pathologies. Unpublished M.A. thesis, Trent University, Peterborough, Ontario. White, C. D., and Schwarcz, H. P. (1989) Ancient Maya diet: As inferred from isotopic and chemical analysis of human bone. Journal of Archaeological Science 16:451474. White, E. M., and Hannus, L. A. (1983) Chemical weathering of bone in archaeological soils. American Antiquity 48:316322. Willey, G. R. (1973) The Altar de Sacrificios Excavations: General Summary and Conclusions. Papers of the Peabody Museum of Archaeology and Ethnology, Vol. 64, No. 3. Cambridge: Harvard University. Willey, G. R. (1990) Excavations at Seibal. IV. General Summary and Conclusions. Memoirs of the Peabody Museum of Archaeology and Ethnology, Vol. 17. Cambridge: Harvard University Press. Willey, G. R., and Shimkin, D. B. (1973) The Maya Collapse: A summary view. In T. P. Culbert (ed.): The Classic Maya Collapse. Albuquerque: University of New Mexico Press, pp. 457502. Wright, L. E. (1994) The Sacrifice of the Earth? Diet, Health, and Inequality in the Pasión Maya Lowlands. Ph.D. dissertation, Department of Anthropology, University of Chicago. Wright, L. E. (1997) Ecology or society? Paleodiet and the collapse of the Pasión Maya Lowlands. In S. Whittington and D. Reed (eds.): Bones of the Maya: Skeletal Studies of an Ancient People. Washington, D.C.: Smithsonian Institution Press, 18195. Wright, L. E. (n.d.) Diet, health, and status among the Pasión Maya: A reappraisal of the collapse. Nashville: Institute of Mesoamerican Archaeology, Vanderbilt University Press. Monograph in preparation.
Page 221
Chapter 12 Dietary Carbonate Analysis of Bone and Enamel for Two Sites in Belize Shannon Coyston, Christine D. White, and Henry P. Schwarcz Subsistence practices and dietary choices are influenced by a complex set of factors that include resource availability, cultural traditions, and economic, social, and political considerations (Bryant et al. 1985; Fajans 1988). Consequently, archaeological evidence of paleodiet and subsistence not only can inform us about how people related to their environment but may also suggest how economic, social, and political structures operated and changed in the past (Hastorf and Johannessen 1993; Welch and Scarry 1995). In this chapter paleodietary reconstructions based on isotopic analyses of human bone are used, first, to interpret the different responses of two Maya communities in Belize to changes that occurred at the end of the Classic period (A.D. 250900) and, second, to investigate how food was used by the Maya to distinguish between groups of differing social status. Lamanai and Pacbitun, Belize, were both relatively large centers that were established during the Preclassic (2000 B.C.A.D. 250) and grew rapidly through the Late Preclassic (300 B.C.A.D. 250) and Classic (A.D. 250900) periods. At the end of this time, however, Pacbitun was abandoned while Lamanai continued to be occupied into the Historic period (A.D. 15321675). Isotopic analyses reveal that the diets consumed at these sites underwent changes just before the end of the Classic period. The nature of the dietary changes may provide insights into the very different trajectories of the two settlements. Already there are indications that greater resource diversity at Lamanai and different opportunities to participate in coastal trade and political networks contributed to this center's longterm success. Previous isotopic studies of bone collagen (White and Schwarcz 1989; White et al. 1993) suggest that Lamanai was always less reliant on maize agriculture than was Pacbitun. Pacbitun's decline has been associated with ecological and economic stresses brought about in part by intensive maize cultivation during the Late and Terminal Classic. In this study the role of maize agriculture and maize consumption in each site's history is reconsidered. Recent experiments (Ambrose and Norr 1993; Tieszen and Fagre 1993) have demonstrated that the isotopic composition of collagen may largely reflect the 13C/12C ratio of dietary proteins, as carbon from protein is preferentially routed to collagen. Dietary studies before 1993 assumed that the isotopic composition of collagen was determined by the whole
Page 222
diet and may not provide accurate estimates of the proportions of C3 and C4 foods consumed. The isotopic composition of carbonate in bone apatite, which is the mineral phase of bone, may provide a more reasonable estimate of the composition of the whole diet. Although the additional carbonate isotopic information provides different insights into the composition of the diets, conclusions drawn from collagen studies regarding the relative importance of maize agriculture at each site are still tenable. The Sites: Their Settings and Histories Lamanai Lamanai is a relatively large (4.5 km2) ceremonial center situated on the west shore of the New River Lagoon in northern Belize (Figure I.1). Excavations directed by David Pendergast, of the Royal Ontario Museum, revealed that the settlement was continuously occupied for more than 20 centuries, beginning in the Middle Preclassic (1250400 B.C.). By the Late Preclassic (400 B.C.A.D. 250) Lamanai had become a relatively large center with regional importance. The site continued to grow through the Classic period (A.D. 2501000), and although there are indications of decline during the Postclassic (A.D. 10001520), Lamanai was still a thriving community. In the early part of the Historic period only a small population remained at the site. Lamanai was finally abandoned around A.D. 1675. The site's prosperity and longevity have been attributed to its location, which afforded opportunities for trade and abundant resources (Pendergast 1981, 1986). A diverse set of habitats, including tropical forest, pine ridge, savanna, bajo, estuarine, lacustrine, and riverine zones, exist near the site. Therefore, a wide variety of plant and animal foods would have been at hand. Coastal habitats may also have been considered accessible, as Lamanai is only 70 km from the Caribbean coast via the New River. Agricultural zones were also extremely important. Remnants of raised fields exist north of the site, but it is not known at what time these fields were constructed or used. Nonetheless, at some point in the site's history a combination of intensive cultivation of forest milpas and raised fields likely reduced production risks and ensured adequate crop yields for the residents of Lamanai. Paleobotanical studies were not done at Lamanai; faunal remains from the site, however, reflect the diversity of the site's setting (Emery 1990 and this volume). Pacbitun Pacbitun is located in the foothills of the Maya Mountains, 80 km southwest of Lamanai (Figure I.1). Paul Healy, of Trent University, directed excavations at the site (Healy 1990). Like Lamanai, Pacbitun was established in the Middle Preclassic (ca. 900 B.C.) as a small farming village and by the Late Preclassic (300 B.C.A.D. 250) had become an urban center (ca. 4 km2) with regional influence. The proliferation of carved monuments, extensive construction of monumen
Page 223
tal architecture, and an abundance of exotic materials in elite burials attest to the growth and prosperity of Pacbitun during the Classic period. Settlement studies indicate that population levels peaked at the LateTerminal Classic transition. Then, in the Terminal Classic (A.D. 700900), Pacbitun seems to have experienced an economic and cultural decline. It was abandoned by A.D. 900. Pacbitun's temporal stability has also been attributed, in part, to the ecological diversity of its setting (Healy 1990). Pacbitun sits at the juncture of the lowland tropical forest ecozone and the upland pine ridge ecozone of the Maya Mountains. A variety of resources could have been gathered from the sparsely settled uplands, including plants, game, fuel, and lithic raw materials for ground stone implements. Additional game and plant resources would have existed in the lowland forests, and aquatic resources were available from several rivers and small streams near the site. The relatively fertile soils of the forest also provided the settlement with adequate agricultural produce for many centuries. As the Pacbitun population continued to grow, it apparently became necessary to expand agricultural land by constructing terraces in the hills surrounding the site (Healy 1990). Paleobotanical remains have not been found at Pacbitun (Healy, personal communication 1994). Faunal remains, however, indicate use of both terrestrial and aquatic zones and include, among others, deer (Mazama americana, Odocoileus virginianus), peccary (Tayassu tajacu, T. pecari), tapir (Tapirus bairdii), rabbit (Sylvilagus floridanus, S. brasiliensis), domestic dog (Canis familiaris), turkey (Meleagris ocellata, M. gallopavo), curassow (Crax rubra), and jute snail (Pachychilus) (Emery 1991; Healy et al. 1990). Theoretical Principles: Sources of Carbon and Nitrogen in Maya Diets Ratios of 13C/12C and 15N/14N are used to separate foods into isotopically distinct categories. Most food plants are C3 plants, which have an average 13C ratio of 26.50/00 (Smith and Epstein 1971). C4 plants have higher 13C values that average 12.50/00 (Smith and Epstein 1971), although maize is typically heavier (11.50/00; Tieszen and Fagre 1993). Since the industrial revolution, however, use of fossil fuels has lowered the 13C of atmospheric CO2 by approximately 1.50/00 (Marino and McElroy 1991). Therefore, it is expected that 13C values of modern plants are also lower than they were in the past. Tieszen and Fagre (1993) have demonstrated such a difference in 13C values of modern and archaeological samples of maize, and Schwarcz et al. (1985) found that archaeological maize had a carbon isotopic composition of 90/00. Based on this information, and on a measurement of 10.70/00 for modern maize grown at the site of Lamanai (White and Schwarcz 1989), the average 13C ratio of maize in prehistory is assigned a value of 90/00. Plants are also categorized by their nitrogen isotope ratios. Legumes have 15N values of +10/00, and nonlegumes average +2 to +40/00. Together, 13C and 15N values separate plants into nonleguminous C3 plants, nonleguminous C4 plants, and C3 legumes. The Maya consumed plant foods belonging to all three of these categories, including beans (C3 legumes),
Page 224
maize (C4 nonlegume), and root crops, chiles, tomatoes, and squashes (C3 nonlegumes). The 13C and 15N values measured in bone and enamel can be used to establish the relative importance of isotopically distinct foods in paleodiets. The isotopic signatures of plant foods, plus some fractionation factor, are passed on to the tissues of consumers (DeNiro and Epstein 1978,1981). In humans, carbon is fractionated by roughly +50/00 between diet and collagen (van der Merwe and Vogel 1978). Estimates ranging from +9.80/00 to +130/00 reflect the uncertainty surrounding the degree to which carbon is fractionated between diet and carbonate (DeNiro and Epstein 1978; Krueger and Sullivan 1984; LeeThorp et al. 1989). 15N ratios are enriched by +3 to +40/ relative to the diet at each trophic level (DeNiro and Epstein 1981; Schwarcz and Schoeninger 1991). 00 Carbon isotope studies of collagen have been used to assess the importance of maize in Maya diets and agricultural economies at several lowland sites (Gerry 1993; Reed 1992; Tykot et al. 1996; White 1986; White and Schwarcz 1989; White et al. 1993; Wright 1994). However, meat consumption is also reflected in the 13C composition of collagen and carbonate. Consuming tropical reef fish and shellfish, or maizefed domestic animals such as dog, peccary, or turkey, results in elevated 13C values, similar to those produced by eating maize. Marine resources may have been accessible at Lamanai (Emery 1990) but are not expected to have been an important part of the diet at Pacbitun. Freshwater riverine fish were available at both sites, and jute snails (Pachychilus) were a secondary source of animal protein at Pacbitun (Healy et al. 1990). The effect of eating these aquatic resources, or terrestrial game that fed on C3 plants, is depleted 13C values that overlap those produced by consuming C3 plants. The nitrogen isotope composition of collagen can be used to identify the source of dietary protein (Schoeninger 1985). Because 15N is enriched in a consumer's tissues, eating meat produces higher 15N values in collagen than plant consumption does. Terrestrial herbivores have 15N values that are less than or equal to 70/00. 15N values greater than or equal to 90/ are typical for carnivores but may also indicate the use of freshwater fish. Tropical reef resources and jute snails are depleted 00 in 15N compared with temperate marine resources, terrestrial animals, and even some plant foods (Wright 1994; van der Merwe et al. 1994). A theoretical model depicting the carbon and nitrogen isotopic composition of the resources available at Lamanai and Pacbitun is shown in Figure 12.1. Until recently, a linear mixing model (Schwarcz 1991) has been used to describe the uniform distribution of carbon from all ingested macronutrients (carbohydrates, lipids, and proteins) to both collagen and carbonate (Sullivan and Krueger 1981; van der Merwe 1982). The isotopic composition of both phases of bone was believed to reflect the isotopic composition of the whole diet. Subsequently, it was proposed (Sullivan and Krueger 1983; Krueger and Sullivan 1984; LeeThorp et al. 1989) that carbon from the growth portion of the diet (proteins) is preferentially routed to collagen while carbon from the energy fraction (primarily lipids and carbohydrates) is incorporated into carbonate. Recent experimental studies (Ambrose and Norr 1993; Tieszen and Fa
Page 225
Figure 12.1. Theoretical model of the carbon and nitrogen isotopic composition of the foods available at Lamanai and Pacbitun (van der Merwe et al. 1994; White 1986; White et al. 1993; Wright 1994).
gre 1993) have demonstrated that carbon derived from proteins is, in fact, preferentially routed to collagen. Therefore, 13C values of collagen ( 13CCO) underestimate the energy portion of the diet. The isotopic composition of carbonate has been shown to be influenced, to a greater extent, by carbon from all the macronutrients. Consequently, the 13C values of carbonate ( 13CCA) should be a better indicator of the isotopic composition of the total diet than is collagen (Tieszen and Fagre 1993). Admittedly, the process whereby carbon is distributed to collagen and carbonate is still poorly understood. However, since 13CCO ratios are influenced primarily by carbon from proteins, and are not uniformly influenced by carbon from all the macronutrients, elevated 13CCO values of ancient Maya individuals may be a product of consuming both maize and maizefed animals. Previous estimates of the amount of maize directly consumed may be exaggerated, although the overall importance of maize in food production systems and the economy is not. Sample Description Determinations of 13CCA were made on samples of bone from 47 burials representing the Preclassic to Historic cultural sequence at Lamanai (Table 12.1). Reliable 13C values had previously been obtained from collagen for 41 of these burials (White 1986; White and Schwarcz 1989). The sample includes 42 adults (22 males, 14 females, 6 undetermined) and 5 juveniles. All these individuals are thought to have been members of the elite class, on the basis of burial location
Page 226
in the ceremonial precinct and grave contents (Pendergast 1986). A male (N956/1) and a female (N953/1) interred in a tomb during the Early Classic have been assigned to a higher ''royal" status. Samples of bone carbonate were obtained for 20 burials from Pacbitun (Table 12.2), all of which had been included in the earlier collagen study (White et al. 1993). There are 18 adults (8 males, 9 females, 1 undetermined) and 2 juveniles. The sample includes burials from the Early Classic to Terminal Classic period, as well as 1 adult female interred at some unknown point after the site was abandoned. Early Classic (Tzul phase) and Late Classic (Coc phase) burials were all located within the site core. Terminal Classic (Tzib phase) burials were obtained from the core and from housemounds in the site periphery and are assumed to represent individuals of different (i.e., lower) socioeconomic status. Status distinctions are also indicated by differences in grave types; these include elaborate crypts and cists with limestone roof slabs, simple crypts and cists without roof slabs, simple pits, and child burials in urns. To minimize problems of diagenesis, only dense cortical sections of long bones were sampled (LeeThorp and van der Merwe 1987,1991). Biological apatite is similar to mineral hydroxyapatite in its inorganic components and crystalline structure. Bone apatite, however, is more poorly crystallized, more soluble, and therefore subject to greater amounts of ionic exchange (Glimcher et al. 1981). Precipitation of calcite from the burial matrix onto bone surfaces and exchange with groundwater carbonate readily occur in bone apatite (Hassan et al. 1977; LeeThorp and van der Merwe 1987) and obscure the original dietary signal. Several studies, however, suggest that chemical pretreatment methods can remove these diagenetic carbonates (Hassan et al. 1977; Krueger and Sullivan 1984; LeeThorp and van der Merwe 1991). Tooth enamel apatite is more resistant to diagenesis (LeeThorp and van der Merwe 1991). Therefore, 13CCA values of enamel were also measured in order to assess the extent to which bone apatite may be affected by diagenesis. The individuals were selected from the same sample of burials on the basis of which molars were available for analysis. The isotopic ratios of third molars are most likely to be like the isotopic composition of an adult diet. Only four third molars were available from Lamanai and three from Pacbitun. A second molar was also included from Lamanai, even though the composition reflects a younger, childhood diet (37 years). The germ of a molar, belonging to one of the juveniles from Pacbitun, was also analyzed. Methods Before chemical pretreatment, samples of bone and enamel were cleaned and discolored surfaces were removed from enamel using a dental drill. Chemical pretreatment procedures used are similar to those outlined by Sullivan and Krueger (1981) and LeeThorp and van der Merwe (1987). Diagenetic surface carbonates were removed by soaking the bone in 1 M acetic acid for 1 hour. To remove organic material and to isolate the carbonate as much as possible, samples were then soaked in sodium hypochlorite (NaOCl) for 24 hours. The
Page 227 Table 12.1. Carbon and Nitrogen Isotope Values of Carbonate and Collagen, and CO2 Yields and Crystallinity Indices of Carbonate, of Lamanai Burials by Context and Phase. 13
CI
13
CCO
15
0
0
Burial
Age
Sex
0
( / 00)
( / 00)
( / 00)
/ 00 yield
CCA
0
N
Preclassic (300 B.C.A.D. 250) P89/1
no information
2.58
4.04
6.97
N/A
N/A
P89/2
adult
M
3.51
3.37
6.67
12.67
10.6
P8103/1
adult
M
2.56
3.82
7.10
12.66
9.8
10.0
Early Classic (A.D. 250400) N933/1
3040
M
3.74
3.31
8.19
10.62
N933/2
3040
M
3.33
3.75
7.20
11.22
10.4
N933/5
no information
4.55
2.96
9.39
N/A
N/A
N953/1
40
F
4.58
2.86
5.71
13.16
10.0
N956/1
45
M
3.10
3.45
4.56
14.11
13.2
Late Classic (A.D. 650900) N115/5
adult
M
4.40
3.32
6.45
14.01
10.3
N119/3
adult
F
2.41
4.05
6.20
15.32
N/A
N933/6
30
M
2.80
3.44
6.96
13.12
10.4
10.2
Terminal Classic (A.D. 9001000) N1066/2
3040
M
4.70
2.82
7.55
15.15
N1066/8
adult
M
4.15
3.16
5.43
17.18
9.4
N1068/3
adult
M
4.55
3.03
8.27
15.34
10.0
N1068/4
1521
M
2.40
3.80
8.35
14.70
9.8
Postclassic (A.D. 10001520) N959/1
child
—
3.64
3.03
5.71
9.45
9.0
N101/12
no information
2.82
3.65
6.14
N/A
N/A
N102/7
45
M
3.98
3.29
8.06
11.96
10.7
N102/23
adult
M
3.35
3.38
4.90
9.01
10.2
N102/24
40+
M
4.74
2.78
7.89
9.97
9.7
N102/40
adult
F
3.65
4.35
9.03
10.1
N102/42
adult
F
3.18
9.12
8.9
N102/50
8 or 9
—
3.60
2.91
6.37
9.33
8.8
N103/4
adult
M
4.55
3.01
5.64
9.50
8.0
N107/1
no information
4.82
2.93
4.77
N/A
N/A
N109/10
40
M
4.93
2.78
5.54
8.87
9.7
N104/11
30+
F
3.83
3.34
5.06
9.03
8.6
N104/19
adult
F
5.05
2.74
6.31
9.16
9.2
N104/21
adult
M
3.15
3.31
9.28
10.2
N104/31
18+
F
3.00
3.38
5.24
9.94
8.5
N104/42
20
F
3.00
N/A
10.21
8.39
9.0
N104/45
30+
M
3.80
3.54
4.98
7.97
9.5
N109/1
9 mo1 yr
—
3.54
2.88
10.66
9.09
8.5
N109/2
2
—
5.52
3.08
6.22
9.80
11.4
N114/1
10
—
3.94
3.41
4.88
9.53
9.9
N115/7A
adult
M
3.83
3.20
7.13
10.14
9.6
N 115/7B
adult
F
4.39
3.61
5.13
8.95
9.2
(table continued on next page)
Page 228
(table continued from previous page) Burial
Age
Sex
/ 00 yield
13
CI
0
CCA
(0/ 00)
13
CCO
15
(0/ 00)
(0/ 00)
N
Historic (A.D. 15201675)
N1211/2
no information
4.98
3.01
6.36
N/A
N/A
N1211/GP2
35+
M
4.82
3.10
5.45
8.63
9.3
N1211/4
4050
F
4.65
3.08
6.05
9.54
9.7
N1211/5A
adult
F
3.12
3.54
4.19
9.89
10.8
N1211/19
no information
4.37
3.18
6.64
N/A
N/A
YDL85/25
adult
F
3.22
4.72
10.19
9.8
YDL85/35
adult
F
3.83
3.15
5.13
9.67
9.9
YDL85/64
adult
F
4.92
2.92
5.80
9.73
8.5
YDL85/73
adult
M
5.60
2.86
9.13
9.9
YDL85/81
3545
M
5.12
3.02
10.16
9.6
5.77
Note: Carbon and nitrogen isotope data for collagen from White 1986, Appendix H.
chemically treated bone and enamel were reacted with 100 percent phosphoric acid (P2O5) under vacuum. CO2 produced during the reaction was collected by cryogenic distillation on a vacuum line. The volume of CO2 gas was measured manometrically in order to calculate CO2 yields. Ratios of 13C/12C were measured using the McMaster University V.G. 602D Micromass mass spectrometer. The precision of analysis is ±0.10/00. Sample reproducibility, based on 10 duplicates, is less than 0.20/00. Results are expressed as 13C values relative to the Pee Dee Belemnite standard and are calculated by: 13
C = [(13C/12C)sample/(13C/12C)standard 1] x 1000
Testing for Diagenesis Carbonate has been used infrequently in paleodietary studies because of the controversy surrounding the extent to which postmortem diagenesis obscures the original dietary signal (DeNiro and Epstein 1978; Sullivan and Krueger 1981,1983; Schoeninger and DeNiro 1982). Therefore, several different attempts were made to identify the effects of diagenesis in the Lamanai and Pacbitun bones. Fourier transform infrared spectroscopy (FTIR) was used to examine the crystal structure of the bone apatite. Crystallinity indices (CI) generated after the method used by Shemesh (1990) reflect the relative size of the crystals and the ordering of atoms in the lattice (Shemesh 1990; Weiner and Bar Yosef 1990). During diagenesis archaeological bone may recrystallize, resulting in larger crystals, a more ordered crystal structure, and higher CI values. Recrystallization also results in a change in isotopic composition due to fractionation or incorporation of soil carbonate. CI values greater than those observed in modern bone (2.7 to 3.2) (Sillen, Kolodny, personal communications to Schwarcz, 1993) would indicate that archaeological bone had recrystallized and that 13CCA values may be altered. In the Lamanai sample CI values range from 2.7 to 4.1 (N = 46, x = 3.3 ±
Page 229 Table 12.2. Carbon and Nitrogen Isotope Values of Carbonate and Collagen, and CO2 Yields and Crystallinity Indices of Carbonate, of Pacbitun Burials by Context and Phase. Burial
Site Core
Age
Sex
Grave type
13
/ 00 yield CI
0
13
CCA
15 N (0/ 00)
CCO
0
0
( / 00)
( / 00)
crypt
2.00
3.91
5.10
9.17
8.12
Tzul Phase (A.D. 300550) BU 16
5060
M
Coc Phase (A.D. 550700) BU 11
2040
F
cist
2.30
3.65
4.88
9.84
10.01
BU 24
40+
M
cist
2.32
3.63
5.45
7.28
9.14
BU 42
adult
M
crypt
2.63
3.45
3.89
8.25
8.85
Tzib Phase (A.D. 700900) BU 17
adult
M
cist
2.93
3.57
4.49
9.12
10.11
BU 21
adult
F
crypt
2.65
3.41
5.36
9.27
9.59
BU 22
3540
M
crypt
2.88
3.20
5.67
7.85
9.84
BU 25
6 to 7
—
urn
4.38
2.82
6.74
11.83
8.04
in fill
4.09
3.10
9.16
13.53
7.59
Postabandonment (date unknown) BU 43
adult
F
Site Periphery Tzib Phase (A.D. 700900) Lot 301
adult
F
—
3.56
3.17
4.92
9.61
8.37
Lot 301
child
—
crypt
3.53
2.58
5.49
11.48
8.40
Lot 304
adult
M
cist
3.01
3.34
5.64
9.87
8.65
Lot 305
adult
M
cist
3.57
7.19
9.57
8.83
Lot 340
adult
F
cist
1.17
3.43
12.13
8.75
Lot 415
adult
F
pit
4.34
3.10
6.09
11.14
9.67
Lot 472
adult
F
pit
2.67
3.74
8.08
12.68
10.64
Lot 484
adult
F
—
3.67
3.27
6.42
10.44
9.33
Lot 485
adult
—
pit
2.35
3.70
4.69
10.30
9.56
Lot 486
adult
M
—
3.42
3.23
6.11
10.85
9.21
Lot 487
adult
F
crypt
2.69
3.59
5.01
10.36
9.75
13
Note: Bone samples for Lots 304 and 305 may have been labeled incorrectly, and CCA values for these two individuals may be reversed. Carbon and nitrogen isotope data for collagen from White et al. 1993.
0.33). At Pacbitun values range from 2.8 to 3.9 (N = 18, x = 3.4 ± 0.34). There is no correlation between CI values and 13CCA values at either site whether the samples are considered as a whole or subdivided by phase. Statistical tests failed to define a threshold for CI values that would indicate diagenesis, despite CI values as high as 4.05 at Lamanai and 3.91 at Pacbitun. This may be due to the very small number of bone samples that have CI values at the high end of the range. Therefore, a decision was made not to include samples with CI values 3.8 (a value established for diagenesis of sedimentary hydroxyapatites [Shemesh 1990]) in statistical analyses. Four samples from Lamanai and one from Pacbitun were eliminated from the samples.
Page 230
Carbon dioxide yields were also used to evaluate the reliability of the carbonate data. High yields may reflect the addition of diagenetic carbonates (Wright and Schwarcz 1996), whereas extremely low yields may reflect the loss of carbonate. No correlation occurs between CO2 yields and 13CCA values, at either site, once statistical outliers are eliminated from the sample. Individuals identified as outliers (BU 43, N1068/4, N109/1) were not used in any statistical analyses. A negative linear correlation of CO2 yields and sample age at both Pacbitun (N = 19, df = 17, r = .625, a = 0.05) and Lamanai (N= 44, df = 42, r = .585, a = 0.05) indicates that CO2 yield decreases as sample age increases. Therefore, there has been some loss of natural carbonate but no significant addition of diagenetic carbonate. Burials dating to the Historic period at Lamanai have high CO2 yields, however, which may reflect carbonate enrichment in this burial context. Rather than being buried beneath house floors or within monumental structures, these burials were interred in a cemetery dug into a Postclassic mound. Bone from other burial contexts seems to have been less affected by diagenesis, and acetic acid treatment appears to have been effective in removing diagenetic surface carbonates from these burials. Therefore, diagenesis may be more dependent on burial context than sample age. Overall, crystallinity indices and CO2 yields suggest that bone carbonate is fairly well preserved at both sites. In total only five samples were excluded from statistical analyses on the basis of these diagenetic criteria. Results from the analysis of enamel carbonate (Table 12.3) suggest that bone carbonate is fairly well preserved at Pacbitun, where 13C values for bone and enamel generally differ by less than 20/00. At Lamanai, however, there are indications that bone carbonate from Postclassic and Historic contexts may have undergone postmortem changes. 13C values of two third molars from these periods are shifted markedly in the C4 direction compared to those of two teeth from earlier periods. This shift is much smaller in samples of bone carbonate from the same individuals. Additional C3 carbon, possibly dissolved bicarbonate derived from C3 plant detritus in burial sediments, may have been absorbed onto bone carbonate crystals and was not completely removed with chemical pretreatment. Much more work needs to be done in terms of finding ways to recognize diagenesis in bone carbonate. Recently, Wright and Schwarcz (1996) have established a more rigorous protocol than the one employed here for the use of infrared spectroscopy (to measure CI and carbonate content), CO2 yields, and 18O values to identify diagenesis. Therefore, it is necessary to reexamine the carbonate using their criteria. Some of the conclusions drawn from this study should be viewed as tentative. Results and Discussion of Carbonate Analysis The Role of Maize Agriculture Lamanai As expected, the isotopic composition of carbonate reflects the consumption of a mixture of C3 and C4 foods, in all periods, at both sites. The
Page 231 Table 12.3. Carbon Isotope Values of Bone and Enamel Carbonate for Selected Individuals from Lamanai and Pacbitun, Belize. Lamanai / 00CO2
13
13 diet
13
13
Burial
Age
Sex
Tooth
0
3
2.51
6.97
18.97
8.81
20.81
M
3
M
2.33
5.43
17.43
7.32
19.32
CAP
C
CAP
bone (0/ 00)
Cdiet
enamel (0/ 00)
Preclassic (300 B.C.A.D. 250) P89/1
no information
M
Terminal Classic (A.D. 9001000) N1066/8
adult
Postclassic (A.D. 10001520) N 104/19
adult
F
M3
1.89
6.31
18.31
1.42
13.42
N104/45
30+
M
M2
1.93
4.98
16.98
2.65
14.65
2.20
5.80
17.80
1.81
13.81
Historic (a.d. 15201675) YDL85/64
adult
F
3
Burial
Age
Sex
Tooth
0
M
Pacbitun / 00CO2
13
CAP
13
Cdiet
bone (0/ 00)
13
CAP
13
Cdiet
enamel (0/ 00)
Tzib Phase (A.D. 700900) BU 17
adult
M
BU 25
67
—
Lot 472
adult
F
M
2.54
4.49
16.49
2.72
14.72
M2
2.40
6.74
18.74
5.20
17.20
M
2.06
8.08
20.08
6.02
18.02
M
2.24
9.16
17.16
9.95
21.95
3
3
Postabandonment (not dated) BU43
adult
F
2
isotopic composition of the diets changed over time, however. Qualifying these changes may help clarify why the dietary changes occurred and whether they were related to other environmental, social, or political processes that were taking place at each site. At Lamanai temporal variation in 13C values is less marked in carbonate than it is in collagen (Figure 12.2a). Cultural periods were grouped into Preclassic/Early Classic (x = 6.95 ± 1.730/00, N = 6), Late/Terminal Classic (x = 6.81 ± 1.010/00, N = 6), Postclassic (x = 6.03 ± 1.440/00, N = 19), and Historic (x = 5.57 ± 0.790/00, N = 9) periods in order to permit an analysis of variance among mean 13CCA values for different time periods. Mean 13CCA values for each cultural period do not differ significantly (ANOVA, F = 1.881, a 0.05). However, the mean 13CCA values of burials dating from the Preclassic through Terminal Classic (Early period = 6.88 ± 1.350/00, N = 12) are significantly lower than the mean 13CCA values of burials from Postclassic and Historic contexts (Late period = 5.88 ± 1.270/00, N = 28) (ttest, t = 4.67, a 0.05, df = 38). Although there is considerable overlap in 13CCA values between the two periods, Figure 12.2b clearly illustrates the separation between the two periods. Significantly higher 13CCO ratios (ttest, t = 9.62, a 0.05, df = 35) also indicate that marked changes to the Lamanai diet occurred at the end of the Terminal Classic period. The decision to regroup the burials by an Early and Late period was based on what seemed to be a real temporal shift in the 15N data (see below) at the end of
Page 232
the Late Classic, as well as other types of archaeological evidence indicating that significant social, economic, and political changes took place at Lamanai at the end of the Classic period. Several lines of evidence suggest that changes in the Early to Late periods involved proteins. First, the shift is most apparent in the 13C values of collagen, which are primarily influenced by dietary proteins. Second, the average 15N ratio of Early period burials (10.34 ± 1.280/00, N = 12) is 10/00 lower (ttest, t = 2.909, a 0.05, df = 35) in the Late period (9.51 ± 0.780/00, N = 28). Further, the declining 15N values are negatively correlated with 13CCO values (Pearson's Correlation Coefficient, r = .404, a 05, df = 38), indicating that it was the source of nitrogen (i.e., protein) that also resulted in changes in the carbon isotopic composition of collagen. The isotopic shifts in carbonate and collagen can be explained several ways. Consumption of maize may have increased in the Late period (White and Schwarcz 1989), replacing some portion of the C3fed terrestrial animals in the diet. However, maize probably did not comprise 70 percent of the diet, given the effect of macronutrient routing. Alternatively, reef fish and shellfish may have been substituted for some of the terrestrial animals. Finally, the Lamanai Maya may have eaten both more maize and more marine animals. The recovery of fish remains, which are abundant in Postclassic middens and increase in Historic contexts (Emery 1990 and this volume), supports the idea of increasing use of reef resources during these periods. The fish have been broadly classified as riverine and peripheral or estuarine species (Emery, this volume). The estuarine species are primarily marine types. Some migrate between the lagoon and the coast but most are adapted to the estuarine waters of the lagoon. Emery (this volume) believes that most fish were probably obtained from the lagoon. The isotopic composition of fish that migrate between marine and estuarine environments may be influenced by marine systems and may be similar to those of reef species. Even those fish that were entirely estuarine may have isotopic ratios that are more like those of reef fish than those of freshwater fish, which are lower in 13C and higher in 15N. The isotopic composition of the fish remains from Lamanai have yet to be determined. Therefore, this interpretation remains tentative. The timing of this dietary change at Lamanai is intriguing. Since fish were no less abundant in the lagoon during earlier periods, there may be some sort of social or cultural explanation for the greater focus on fish following the Classic period. It is at this point that Lamanai established stronger political and economic ties with centers in the Yucatán and along coastal trade routes—centers where marine fish were probably an important part of the diet. This group of Lamanai elites was perhaps identifying its alliance with a new sociopolitical network by increasing its consumption of fish. Greater participation in coastal trade networks at this time may also have provided traders from Lamanai with opportunities to return to the site with fish that they had obtained through trade or caught themselves. Pacbitun At Pacbitun apparent temporal changes in
13
CCA track changes in
13
CCA values for each phase more closely than they do at Lamanai,
Page 233
Figure 12.2. Temporal trends in the relationship between 13C ratios of carbonate ( ) of Lamanai burials from (a) Pre/Early Classic to Historic periods and (b) Early ( ) periods.
Page 234
Figure 12.3. Temporal changes in 13C ratios of carbonate and collagen of Pacbitun burials.
although the shifts are still more apparent in collagen (Figure 12.3; Table 12.2). The agreement could be random, however, as Early Classic Tzul phase and Late Classic COC phase samples are extremely small. The 13C composition of both phases of bone indicates that C4 resources (maize and probably maizefed domestic animals) comprised a significant portion of the Pacbitun diet by the Tzul phase (N = 1, 13CCA = 5.100/00; N = 1, 13CCA = 9.170/00). Then in the COC phase, 13C values in bone peak (N = 3, 13CCA = 4.74 ± 0.790/00; N = 3, 13CCO = 8.46 ± 1.290/00), indicating that reliance on C4 resources was greatest during this time. One explanation for the isotopic shift is consumption of greater amounts of maize. Alternatively, temporal shifts in 13C values may reflect differences in the major source of dietary protein, as the change is more apparent in collagen than in carbonate. Relatively unchanging 15N values (Figure 12.4) indicate that terrestrial herbivores were the major source of protein during all periods (White et al. 1993) and that the proportions of meat and plant foods in the diet remained fairly stable. Therefore, it may be that the Pacbitun diet included more maize fed domestic animals, such as dog, peccary, or turkey, during the Coc phase. The single dog represented in the Pacbitun faunal recoveries ate a diet that included a large proportion of maize (White et al. 1993). Increased consumption of a combination of maize and C4 consumers could also explain the Coc phase data. Regardless of the exact nature of dietary change, it is clear that the importance of C4 foods, and ultimately maize agriculture, increased during the Coc
Page 235
Figure 12.4. Temporal changes in 15N ratios of collagen in burials from (a) Lamanai and (b) Pacbitun.
Page 236
phase. Coincident with the dietary change is evidence for the initial construction of agricultural terraces at the site (Healy 1990). These efforts to increase agricultural production likely were made in response to a growing population (Healy 1990) and apparently were successful in view of the food consumption data reported here. Intensive cultivation and construction of terraces continued in the Tzib phase, but it appears that levels of maize production did not increase. This is implied from 13C values in bone (N = 14, 13CCA = 5.85 ± 1.010/00; N = 3, 13CCO = 9.91 ± 1.360/00), which indicate that consumption of C4 foods decreased by roughly 6 percent during this Terminal Classic phase. A portion of maize may have been replaced by C3 plants in the diet. Root crops may have been important among the C3 alternatives. Root crops grow in a variety of soil types, are fairly drought resistant, and tend to flourish even under conditions that cause other food crops to fail (Bradbury and Holloway 1988). Evidence for early use of root crops by the Maya has been found in Preclassic contexts at Cuello, Belize (Hather and Hammond 1994), and at Joya de Cerén, El Salvador, on the southeast periphery of the Maya area (Lentz et al. 1996). Alternatively, the changes seen in 13C value may reflect the substitution of some of the domestic animals in the diet by game. Pacbitun had reached its population maximum by the Terminal Classic (Healy 1990), and it may have become increasingly difficult to meet production demands and/or too costly to keep the same number of domestic animals. Unfortunately, because botanical remains have not been found and faunal recoveries are few (Emery 1991), these types of data cannot be used to clarify or corroborate trends observed in the isotopic data. The temporal sequence at Pacbitun ends with the Terminal Classic sample. However, one burial dating to sometime after the site was abandoned was found within the fill of a core zone structure. This adult female is unusual with respect to not only her late interment but also her isotopic data, which are anomalous. The carbon isotopic composition of both phases of bone ( 13CCA = 9.160/00, 13CCO = 13.530/00) suggest that her diet contained substantially more C3 foods than the diets consumed when Pacbitun was occupied. Considerably lower 15N values (7.590/00) may indicate that the bulk of these C3 foods were plants. Social Differentiation through Diet Lamanai In stratified societies differences in food consumption and access to food are often used to establish, maintain, and legitimize inequalities in socioeconomic status (Hayden 1990) or simply to mark boundaries between social groups (Bryant et al. 1985). The distinctive isotopic data of the highestranking male (N956/1), buried during the Early Classic period, indicates that Lamanai elites also used this strategy to mark their privileged status. Compared with his contemporaries, this individual has a much higher 15N value (13.20/00), his collagen is depleted in 13C, and the 13C value of his carbonate is elevated (Figure 12.2a). Exclusive access to imported seafood (White and Schwarcz 1989) can no longer explain this patterning since it has been recently
Page 237 Table 12.4. Comparison of Means by Age and Sex. 13
N
Group
Lamanai Males
13
N
CCA (0/ 00)
15
N
CCO (0/ 00)
N (0/ 00)
9
6.15 ± 1.22
11
9.51 ± 1.05
11
9.62 ± 0.68
Females
11
5.65 ± 1.65
12
9.39 ± 0.52
11
9.27 ± 0.59
Adults
20
5.88 ± 1.45
23
9.45 ± 0.80
23
9.50 ± 0.70
4
5.79 ± 0.67
4
9.55 ± 0.19
4
Juveniles
9.78 ± 1.18
Pacbitun Males
7
5.49 ± 1.07
8
8.99 ± 1.15
8
9.09 ± 0.64
Females
6
5.98 ± 1.89
8
10.68 ± 1.21
8
9.51 ± 0.71
Adults
15
4.30 ± 1.31
17
9.86 ± 1.39
16
9.32 ± 0.67
2
5.54 ± 0.64
3
12.32 ± 1.18
2
8.22 ± 0.25
Children Note:
13
CCO and
15
N values are from White 1986; White and Schwarcz 1989; White et al. 1993.
established that fish and shellfish from reefs off the Belize coast have lower 15N values and more positive 13C values than do temperate marine animals (van der Merwe et al. 1994). The 13C and 15N values of his collagen indicate that the royal male ate more C3fed game and/or freshwater fish than other Early Classic elites. Both deer and peccaries had important roles in Maya ritual and ceremony (Pohl and Feldman 1982). Their consumption by elites could reasonably have been a symbolic representation of status. In order to investigate the relationship between diet and status based on sex or age groups, we separated the sample of burials from Lamanai into Early and Late periods, as the diets differ between the two periods. Only one juvenile and only one female with wellpreserved carbonate date to the Early period, however, precluding any statistical analyses. In the Late period, 13CCA ratios of males versus females, and of adults versus juveniles, are not statistically different (Table 12.4). This agrees with statistical tests using collagen data (White and Schwarcz 1989). If the Lamanai Maya used dietary differences to distinguish status differences along the lines of age and sex, it cannot be discriminated isotopically. Pacbitun The carbon isotopic composition of bone carbonate does not indicate any association between diet and either sex or age at Pacbitun (Table 12.4). Although the mean 13C ratio of the two children is somewhat lighter than that of the two adults, values of the two groups overlap. To some extent, this corroborates collagen data that CA show that children ate much smaller quantities of C4 foods (White et al. 1993). The mean 13CCA ratio of males is slightly heavier than that of females, but the difference is not significant. The collagen, however, suggests that males ate significantly more C4 foods (White et al. 1993). White et al. (1993) distinguished between elite and commoner burials based on elaborateness of grave construction, distance of the burial from the site core, and quantity and type of grave contents. Similar criteria were used in this study. We recognize that identifications of elite and nonelite made on the
Page 238 Table 12.5. Comparison of Means by Burial Type at Pacbitun. 13
13
N
Crypt
4
4.98 ± 0.78
—
—
4
9.51 ± 0.45
Cist
5
5.53 ± 1.03
—
—
6
9.25 ± 0.65
Pit
3
6.29 ± 1.70
—
—
3
9.96 ± 0.59
Urn
1
6.74
—
—
1
8.04
CCA (0/ 00)
N
15
Grave Type
N
CCO (0/ 00)
N (0/ 00)
Crypt/Cist
—
—
12
9.52 ± 1.39
Pit
—
—
3
11.37 ± 1.20
Urn
—
—
2
12.75 ± 1.37
Note: Crypts do not include the child burial, Lot 302.
13
CCO and
15
N data from White et al. 1993.
basis of material remains and spatial context are not always straightforward (Chase and Chase 1992) and acknowledge that our assignments of status may eventually need to be reconsidered as criteria for identifying socioeconomic status are revised. The lack of any significant difference in the mean 13CCA values of individuals buried in crypts, cists, and pits (Table 12.5) indicates that the diets consumed by groups of varying social status, at Pacbitun, were not isotopically distinct. This interpretation contradicts the collagen study, which demonstrated statistically that higherranking individuals buried in crypts and cists ate more C4 foods than lowerstatus individuals interred in pits and urns. Although the same trend occurs in the carbonate, it is not as obvious. 15N values also do not differ by grave type (White et al. 1993). The 13C composition of carbonate is not statistically correlated with distance of the burial from the site core, although 13CCA values appear to decrease in burials farther away from the core (Table 12.6). The 13CCO ratios of individuals buried at greater distances from the site core are lower, indicating that they consumed fewer C4 foods than higherranking people buried within the core. So again, the statistical significance that appears in the collagen study does not appear in carbonate. Finally, the isotopic composition of bone from several individuals designated ''royal' based on elaborate grave goods, was compared with that of lowerstatus burials. A royal male (BU 24) buried in the Coc phase has a depleted 13CCA value (5.450/00) compared with two lowerranking elites from this time (4.390/00, N = 2). His 13C value (7.280/ ), however, is roughly 20/ higher. A royal female (BU 21) interred during the Tzib phase was also found to have a 13C value higher than that CA 00 00 CO of nonelites buried in the site periphery. Her 13CCA value (5.630/00), however, probably does not differ significantly from that of these commoners (mean 13CCA = 5.960/00 ± 1.060/00). Overall, the analysis of bone carbonate reveals that the diets consumed by residents of Pacbitun, regardless of their status, were relatively uniform in carbon isotopic composition. Although elevated 13CCA values indicate that higherstatus groups (i.e., males, elites buried in crypts and cists or within the
Page 239 Table 12.6. Comparison of Means by Distance of Burials from the Site Core at Pacbitun. N
Distance
13
CCA (0/ 00)
N
13
CCO (0/ 00)
N
15
N (0/ 00)
0 m
7
5.21 ± 0.91
7
9.37 ± 0.74
5
9.29 ± 0.63
110 m
4
5.69 ± 1.03
4
8.56 ± 0.22
5
9.25 ± 0.65
6
6.07 ± 1.20
7
9.56 ± 0.58
4
340 m
9.48 ± 1.07
0 m
—
—
3
8.75 ± 0.77
110 m
—
—
3
9.68 ±0.16
340 m
—
—
4
10.49 ± 0.25
—
—
3
11.98 ± 0.78
425445 m Note:
13
CCO and
15
N data from White et al. 1993.
ceremonial precinct) consumed slightly larger proportions of C4 foods than lowerstatus groups (i.e., females, juveniles, nonelites buried in pits and urns or in the site periphery), the difference is not statistically significant. However, 13CCO data imply that there was a significant difference in the major source of proteins consumed by different status levels. Higherstatus groups typically have more positive 13CCO values than their lowerstatus counterparts. Thus, they may have consumed larger proportions of maizefattened domestic animals, which may have been highly valued by people whose diet was otherwise low in fats. Dogs and turkeys were also important in agricultural ceremonies and were raised in order to meet tribute obligations (Pohl and Feldman 1982). Therefore, these maizefed domesticates may have been more accessible to higherranking individuals. 15N values, which do not vary among different status groups (White et al. 1993), indicate that the difference was not in the amount of meat consumed. Gerry (1993) has also argued that the difference between the diets of elites and commoners was in the types of meat consumed rather than the amount of meat. If this is true, it would indicate that the Classic Maya created a pattern of differential meat consumption, relatively uncommon within highly stratified societies, to demarcate differences in status (see, for example, Fiddes 1991). Conclusions Isotopic analyses reveal that maize made up a considerable part of the diets consumed at Pacbitun and Lamanai. From this it can be inferred that maize agriculture was an important part of the economy at both settlements. Even with the additional insights provided by the analysis of bone carbonate, however, it remains apparent that Pacbitun was always more reliant on maize agriculture than was Lamanai. It is conceivable that this circumstance contributed to the very different trajectories of the two sites following the end of the Classic period. Evidence presented above shows that, when confronted with the demands of feeding a growing population during the Coc phase, Pacbitun opted to
Page 240
intensify maize production. The fact that residents of this center persisted in their attempts to increase agricultural production in the Tzib phase, which is indicated by continued construction of terraces, suggests that other options may not have existed. Their attempts apparently were unsuccessful, however, as declining proportions of maize in the diet imply that production dropped off. Presumably, this was one factor that ultimately led to the abandonment of the site at the end of this phase. Lamanai's population also grew steadily through the Classic period, and it, too, must have been challenged to find ways of increasing food production. The stability in the isotopic composition of bone right through this time implies that maize production kept pace with population increases. The amount of maize in the diet may have even increased in the Postclassic. However, the residents of Lamanai appear to have also taken advantage of the aquatic and marine resources available from the lagoon and possibly the coast. This option was not open to Pacbitun. The incentive to increase the amount of fish in Postclassic and Historic diets may have involved social and political factors rather than population pressures. New perspectives on how the Maya at Pacbitun and Lamanai used food to demarcate differences in social status have also been developed. Locally available foods, valued for their ritual importance, rather than imported, exotic foods, appear to have been designated as highstatus foods at both Lamanai and Pacbitun. Animals such as deer, peccary, dog, and turkey likely were more accessible to elites through their greater participation in ritual ceremonies and also through tribute payments made to them. Males, at Pacbitun, may also have acquired a larger proportion of dog and/or turkey in their diet than females through larger involvement in ritual. It is also possible, however, that they were simply given a larger share of relatively fatty meat because of a higher status that derived from being male. The relatively higher cost of producing or obtaining certain meats, determined according to local patterns of resource availability, may also have contributed to their prestigious place in the diet. Acknowledgments This project was completed while Shannon Coyston was a master's student in the Department of Anthropology, Trent University, under the supervision of Hermann Helmuth, Christine White, and Henry Schwarcz. The research was supported by two Ontario Graduate Scholarships and by Social Sciences and Humanities Research Council grants to Henry Schwarcz and Christine White and to Paul Healy (Trent University). The authors are grateful to David Pendergast (Royal Ontario Museum) and to Paul Healy and Hermann Helmuth for allowing us the opportunity to analyze the skeletal material from Lamanai and Pacbitun, respectively. Ken Fowler, Wayne King, and Al Slavin helped design and set up a vacuum line at Trent University. Dean Ostrander and John LaPlante provided access to lab space and equipment. Special thanks to Martin Knyf (McMaster University), who demonstrated the laboratory procedures and ran the samples.
Page 241
References Cited Ambrose, S. H., and Norr, L. (1993) Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In J. Lambert and G. Grupe (eds.): Molecular Archaeology of Prehistoric Human Bone. Berlin: Springer, pp. 142. Bradbury, J. H., and Holloway, W. D. (1988) Chemistry of Tropical Root Crops: Significance for Nutrition and Agriculture in the Pacific. Canberra: Australian Centre for International Research. Bryant, C. A., Courtney, A., Markesbery, B. A., and De Walt, K. M. (1985) The Cultural Feast: An Introduction to Food and Society. New York: West. Chase, D. Z., and Chase A. F. (eds.) (1992) Mesoamerican Elites: An Archaeological Assessment. Norman: University of Oklahoma Press. DeNiro, M. J., and Epstein, S. (1978) Influence of diet on the distribution of carbon isotopes in animals. Geochimica et Cosmochimica Acta 42:495506. DeNiro, M. J., and Epstein, S. (1981) Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta 45:341351. Emery, K. (1990) Postclassic and Colonial Period Subsistence Strategies in the Southern Maya Lowlands: Faunal Analyses from Lamanai and Tipu, Belize. Unpublished M.A. thesis, University of Toronto. Emery, K. (1991) The secular/ritual dichotomy in animal use: Final faunal analysis, Pacbitun, Belize. Ms. on file, Department of Anthropology, Trent University, Peterborough, Ontario. Fajans, J. (1988) The transformative value of food: A review essay. Food and Foodways 3:143166. Fiddes, N. (1991) Meat: A Natural Symbol. New York: Routledge. Gerry, J. P (1993) Diet and Status among the Classic Maya: An Isotopic Perspective. Unpublished Ph.D. dissertation, Department of Anthropology, Harvard University, Cambridge. Glimcher, M. J., Bonar, L. C., Grynpas, M. D., Landis, W. J., and Roufosse, A. H. (1981) Recent studies of bone mineral: Is the amorphous calcium phosphate theory valid? Journal of Crystal Growth 53:100119. Hassan, A. A., Termine, J. D., and Haynes, V. C., Jr. (1977) Mineralogical studies on bone apatite and their implications for radiocarbon dating. Radiocarbon 19(3): 364374. Hastorf, C. A., and Johannessen, S. (1993) PreHispanic political change and the role of maize in the central Andes of Peru. American Anthropologist 95(1):115 138. Hather, J. G., and Hammond, N. (1994) Ancient Maya subsistence diversity: Root and tuber remains from Cuello, Belize. Antiquity 68:330335. Hayden, B (1990) Nimrods, piscators, pluckers, and planters: The emergence of food production. Journal of Anthropological Archaeology 9:3169. Healy, P. F. (1990) Excavations at Pacbitun, Belize: Preliminary report on the 1986 and 1987 investigations. Journal of Field Archaeology 17:247262. Healy, P. F., Emery, K., and Wright, L. (1990) Ancient and modern Maya exploitation of the Jute snail (Pachychilus). Latin American Antiquity 1(2):170183. Krueger, H. W., and Sullivan, C. H. (1984) Models for carbon isotope fractionation between diet and bone. In J. E. Turnland and P. E. Johnson (eds.): Stable Isotopes in Nutrition. Washington, D.C.: American Chemical Society, Symposium Series 258, pp. 205222. LeeThorp, J. A., Sealy, J. C., and van der Merwe, N. J. (1989) Stable carbon isotope ratio differences between bone collagen and bone apatite, and their relationship to diet. Journal of Archaeological Science 16:585599. LeeThorp, J., and van der Merwe, N. J. (1987) Carbon isotope analysis of fossil bone apatite. South African Journal of Science 83:712715.
Page 242
LeeThorp, J., and van der Merwe, N. J. (1991) Aspects of the chemistry of modern and fossil biological apatites. Journal of Archaeological Science 18:343354. Lentz, D. L., BeaudryCorbett, M. P., Reyna de Aguilar, M. L., and Kaplan, L. (1996) Foodstuffs, forests, fields, and shelter: A paleoethnobotanical analysis of vessel contents from the Ceren Site, El Salvador. Latin American Antiquity 7(3):247262. Marino, B. D., and McElroy, M. B. (1991) Isotopic composition of atmospheric CO2 inferred from carbon in C4 plant cellulose. Nature 349:127131. Pendergast, D. M. (1981) Lamanai, Belize: Summary of excavation results, 19741980. Journal of Field Archaeology 8:2953. Pendergast, D. M. (1986) Stability through change: Lamanai, Belize, from the ninth to the seventeenth century. In J. A. Sabloff and E. W. Andrews (eds.): Late Lowland Maya Civilization: Classic to Postclassic. Albuquerque: University of New Mexico Press, pp. 223249. Pohl, M., and Feldman, L. H. (1982) The traditional role of women and animals in Lowland Maya economy. In K. V. Flannery (ed.): Maya Subsistence: Studies in Memory of Dennis E. Puleston. Toronto: Academic Press, pp. 295312. Reed, D. M. (1992) Ancient Copán diet through stable carbon and nitrogen isotopic analysis. Paper presented at the 57th Annual Meeting, Society for American Archaeology, Pittsburgh, April 812. Schoeninger, M. J. (1985) Trophic level effects on 15N/15N and 13C/12C ratios in bone collagen and strontium levels in bone mineral. Journal of Human Evolution 14:515525. Schoeninger, M. J., and DeNiro, M. J. (1982) Carbon isotope ratios of apatite from fossil bone cannot be used to reconstruct diets of animals. Nature 297:577578. Schwarcz, H. P. (1991) Some theoretical aspects of isotope paleodiet studies. Journal of Archaeological Science 18:261275. Schwarcz, H. P., Melbye, J., Katzenberg, M. A., and Knyf, M. (1985) Stable isotopes in human skeletons of southern Ontario. Journal of Archaeological Science 12:187206. Schwarcz, H. P., and Schoeninger, M. J. (1991) Stable isotope analyses in human nutritional ecology. Yearbook of Physical Anthropology 34:283321. Shemesh, A. (1990) Crystallinity and diagenesis of sedimentary apatites. Geochimica et Cosmochimica Acta 54:24332438. Smith, B. N., and Epstein, S. (1971) Two categories of 13C/12C ratios for higher plants. Plant Physiology 47:380384. Sullivan, C. H., and Krueger, H. W. (1981) Carbon isotope analysis of separate chemical phases in modern bone and fossil bone. Nature 292:333335. Sullivan, C. H., and Krueger, H. W. (1983) Carbon isotope ratios of bone apatite and animal diet reconstruction. Nature 301(13):177. Tieszen, L. L., and Fagre, T. (1993) Carbon isotopic variability in modern and archaeological maize. Journal of Archaeological Science 20:2540. Tieszen, L. L., and Fagre, T. (1993) Effect of diet quality and composition on the isotopic composition of respiratory CO2, bone collagen, bioapatite, and soft tissues. In J. Lambert and G. Grupe (eds.): Molecular Archaeology of Prehistoric Human Bone. Berlin: Springer, pp. 121155. Tykot, R. H., van der Merwe, N. J., and Hammond, N. (1996) Stable isotope analysis of bone collagen, bone apatite, and tooth enamel in the reconstruction of human diet: A case study from Cuello, Belize. In M. J. Orna (ed.): Archaeological Chemistry: Organic, Inorganic, and Biochemical Analysis. Washington, D.C.: American Chemical Society, pp. 353365. van der Merwe, N. J (1982) Carbon isotopes, photosynthesis, and archaeology. American Scientist 70:596606. van der Merwe, N. J., Tykot, R. H., and Hammond, N. (1994) Diet and animal husbandry of the Preclassic Maya at Cuello, Belize. Paper presented at the Wenner Gren Fourth Advanced Seminar in Bone Chemistry, Banff, September.
Page 243 13
van der Merwe, N. J., and Vogel, J. C. (1978) C content of human collagen as a measure of prehistoric diet in woodland North America. Nature 276:815816. Weiner, S., and BarYosef, O. (1990) States of preservation of bones from prehistoric sites in the Near East: A survey. Journal of Archaeological Science 17:187 196. Welch, P. D., and Scarry, C. M. (1995) Statusrelated variation in foodways in the Moundville chiefdom. American Antiquity 60(3):397419. White, C. D. (1986) Paleodiet and Nutrition of the Ancient Maya at Lamanai, Belize: A Study of Trace Elements, Stable Isotopes, Nutritional and Dental Pathologies. Unpublished M.A. thesis, Department of Anthropology, Trent University, Peterborough, Ontario. White, C. D., Healy, P. F., and Schwarcz, H. P. (1993) Intensive agriculture, social status, and Maya diet at Pacbitun, Belize. Journal of Anthropological Research 49:347375. White, C. D., and Schwarcz, H. P. (1989) Ancient Maya diet: As inferred from isotopic and elemental analysis of human bone. Journal of Archaeological Science 16:451474. Wright, L. E. (1994) The Sacrifice of the Earth? Diet, Health, and Inequality in the Pasión Maya Lowlands. 2 vols. Unpublished Ph.D. dissertation, University of Chicago. Wright, L. E., and Schwarcz, H. P. (1996) Infrared and isotopic evidence for diagenesis of bone apatite at Dos Pilas, Guatemala: Palaeodietary implications. Journal of Archaeological Science 23:933944.
Page 245
GLOSSARY A alluvial: Of, relating to, or found in alluvium, the sediment deposited by flowing water, as in a riverbed, floodplain, or delta. abscess: A gathering of pus due to infection in any part of the body. adipose: Of fat, fatty. alveolar: Relating to the bony part of the dentition that contains the sockets for the teeth. ambient: Surrounding, encircling. anthropogenic: Caused by humans. antemortem: Before death. apatite: See bioapatite or bone carbonate. archaeobotanic, archaeobotanical: Remains of plants found in archaeological deposits. assemblage: A collection of related artifacts. auricular surface: The earshaped surface of bone on the iliac bone which articulates with the sacrum. avian: Of birds. avocado (Persea americana): A mediumtall tree (1020 m) with a strong root system having oval or pearshaped edible fruit with leathery skin, yellowishgreen flesh, and a large seed. B bajo: Lowlying ground. bean (Phaseolus vulgaris): Any of various Neotropical twining herbs in the pea family having leaves with three leaflets, variously colored flowers, and edible pods and seeds. belemnite: a coneshaped fossilized internal shell of any of an extinct genus of cephalopods related to the cuttlefish. biface: Having two surfaces. bioapatite: Also apatite or hydroxylapatite. More properly identified as dahllite, a carbonate hydroxylapatite, Ca5(OH){[PO4]0.50.9>[(CO3(OH)]0.10.5}3, which is contained within the collagen matrix as thin tablets 33 nm in thickness. The elements lead, sodium, and strontium may substitute for calcium. Fluoride and chloride may substitute for a hydroxyl ion. biomass: The total mass or amount of living organisms found in a particular area. bivalve: A class of molluscs (Bivalvia) with two shells hinged together; includes mussels and clams. bone carbonate: The inorganic component of bones and teeth, it is a calcium phosphate [Ca10(PO4)6(CO3)(OH)2] similar to mineral hydroxyapatite in its organic components and crystalline structure. bottle gourd (Lagenaria sp.): A largeseeded Cucurbitaceae cultivar. The seeds are eaten dried or green, and the shells may be used as bottles, bowls, and fishnet floats. breadnut or ramón (Brosimum alicastrum): A large Neotropical tree (2440 m) having yellow fruits each with a large, edible seed and milky juice that exudes from the bark when cut. C cacao (Theobroma cacao): A small evergreen Neotropical tree (810m) that produces leathery, ellipsoid, tenribbed, stimulating fruits of yellow, green, red, or dark purple color. Cocoa refers to the seed kernel and the beverage prepared from the roasted and ground beans. calabash: A tropical American tree (Crescentia cujete) of the bignonia family which bears large, gourdlike fruit. carbonate: Salt of carbonic acid containing the divalent negative radical CO3. caries: A disease process characterized by focal demineralization of dental hard tissues by organic acids produced by bacterial fermentation of carbohydrates. cariogenic: Producing caries. cariostatic: Inhibiting caries production. cassava: See manioc. catchment analysis: Analysis of resources available within a circumscribed area to a particular group. CEJ (cementoenameljunction): The place where the cementum tissue of the root of a tooth meets the enamel tissue of the crown. cervid: See deer. chayote (Sechum edule): A perennial herbaceous vine with thickened roots and slender, branching stems up to 10m long. The pearshaped fruit, stems, young leaves, and tuberized portions of the roots are eaten as a vegetable. chile peppers (Capsicum annuum): The pungent fresh or dried fruit of any of several cultivated varieties of capsicum, used especially as a spicy flavoring in cooking. ciruela (Spondias sp.): Small (37 m) Neotropical fruit tree with edible ellipsoidal drupes with brilliant red epicarps. The pulp has a sweetsour flavor and is prepared fresh, dried, as an atole, mixed with maize flour and sugar, and as chicha (maize liquor). cist: Tomb made of stone slabs or hollowed out of rock. collagen: The fibrous protein constituent of bone,
Page 246
cartilage, tendon, and other connective tissue. In paleodietary research collagen refers to Type I, the only one in bone and tendon.
cotyledon: The first single leaf or one of the first pair of leaves produced by the embryo of a flowering plant.
cougar: See puma.
coyol: See palm.
cribra orbitalia: See porotic hyperostosis.
crypt: An underground chamber or vault, as under the floor of place of worship. crystallinity index (CI): Used to characterize the crystal structure of carbonate, it is a measure of the degree of phosphate (PO43) band splitting at wave numbers 565, 595, and 605 in the infrared spectra of a sample of carbonate.
cucurbit: A tendrilbearing plant having fleshy edible fruit with a leathery rind and unisexual flowers. The ripe fruit is eaten as a vegetable. The seeds are eaten whole or ground, roasted or toasted, and have a high oil content. D debitage: The waste created by the manufacture of stone tools. decalcification: Removal of calcium or calcium compounds from bone. deer (Odocoileus virginianus): A whitetailed hoofed ruminant animal of the Cervidae family, common to the Maya area. deflocculant: An agent activating the process by which large particles present in a suspension break up into fine particles. demography: Statistical study of population structure. DEH (dental enamel hypoplasia): A condition of decreased or arrested growth in enamel of teeth that are still in the process of growth or formation which causes enamel defects. dental arcade (arch): The curved row of teeth that is formed in the maxilla or mandible. dental attrition: The wearing down of teeth. dental caries: See caries. diachronic: Changes occurring over a period of time. diagenesis: Alteration of original chemical composition, structure occurring in a material (e.g., bone) after it has been deposited. discriminant function: A statistic that allows one to determine category. dog (Canis familiaris): A carnivorous mammal related to the foxes and wolves and domesticated in a variety of breeds. E eburnation: An abnormal condition of bone or cartilage in which it becomes very dense and smooth, or polished, often caused by wear. ecosystem: A system made up of communities of plants and animals and their relationship to the physical environment. emic: The view of the insider. enamel defect: See DEH. encomienda: A repressive system used by the Spanish to extract labor or tribute from the natives. endosteal: Referring to the connective tissue lining the marrow cavity of bones. et'ok: Literally ''companion." A group of generic species that are related to one another in Itzaj folkbiological taxonomy. Itzaj usually recognize such relationships as intermediate taxa that fall within the same scientific family by virtue of a salient aspect of common morphology and/or use. F flotation: A method of recovery of botanical remains in which they rise to the surface of water and are removed from there. folkbiological taxa: The biological categories that appear in a folkbiological taxonomy, such as che' (tree), put (papaya tree), and putil (forest papaya) in Itzaj Maya. folkbiological taxonomy: A hierarchical system of biological classification whose universal structure is found in all cultures. folk kingdom: The highest level of folkbiological taxonomy. In all cultures there are two folk kingdoms corresponding to the categories "plant" and "animal." Itzaj refer to all and only animals as b'al'~che' but have no single name to refer to the plant kingdom as a whole: nevertheless, Itzaj have a numerical quantifier, teek, that quantifies all and only plants (e.g., jun~teek ixi'im = "oneplant maize," that is, a maize plant). folkspecific: The level of folkbiological taxonomy immediately subordinate to the genericspecies level. Taxa at this level usually occur as categories that contrast with one another along some perceptible dimension that is culturally salient: for example, Itzaj contrast putil[+]k'aax ("forest papaya," Carica mexicanum) with put il[+]kaj ("village papaya," Carica papaya). folkvarietal: The level of folkbiological taxonomy immediately subordinate to the folkspecific level. Taxa at this level usually occur as categories that contrast with one another along some perceptible dimension that is culturally salient: for example, Itzaj contrast ixk'än[[+]]putil[+]kaj (yellow village papaya) with ixsäk[[+]] putil[+]kaj ("white village papaya"). fractionation: The selection for or against one or more isotopes of an element during a chemical reaction which results in a measurable difference in the isotope ratios between a reaction product and its substrate. freshwater snail (Pachychilus crovinus or P. largillierti): An edible mollusc also known as jute, the shell of which can be processed for lime.
Page 247
frijolillo (Cassia occidentalis): A legume. G gastropod: Any of a large class of molluscs (Gastropoda) having onepiece, straight, or spiral shells, such as snails, or having no shells or greatly reduced shells, such as slugs. generic species: The fundamental level of classification in folkbiological taxonomy. At this level folk taxa often correspond to scientific genera or species: for example, "dog" and "tomato" in folkbiological classification of ordinary English speakers or, equivalently, pek' and p'ak for Itzaj Maya. germplasm: Hereditary material, often enclosed within a seed. H hackberry (Celtis sp.): Any of various fruit trees or shrubs of the genus Celtis, having inconspicuous flowers and small usually ovoid drupes. Harris' Line: Linear areas of increased density in the ends of long bones which are viewed with xray and created by recovery from a period of growth arrest. HCL: Hydrochloric acid, a clear, colorless, fuming, poisonous, highly acidic aqueous solution of hydrogen chloride. horticulture: Gardening. hydraulic agriculture: Irrigation agriculture. hyperextension: Movement or extension beyond that normally allowed, as in joints of the body. I in situ: Literally "in place." intermediate taxa: The folkbiological taxa that appear between the genericspecies and lifeform levels. Unlike the genericspecies and lifeform levels, intermediate taxa do not fully partition the locally recognized biota, and they are often perceived as related to a prototypical generic species rather than explicitly named: for example, Itzaj recognize uyet'ok b'alum ("companions of the jaguar" = felines), uyet'ok ya' ("companions of the chicle tress" = Sapotaceae family), and uyet'ok xa'an ("companions of the Sabal palm = palms and broadleafed zingiberales). isotope: one of two or more atoms having the same atomic number but different mass numbers. A difference in the number of neutrons in the nuclei of an atom leads to differing atomic mass or atomic weight. Nuclides that transform into different nuclides only through an external agent are considered stable. Radioactive isotopes are nuclides that spontaneously transform into one or more different nuclides. isotope mass spectrometry: a measurement technique that relies on the principle that electrically charged atoms can be separated by their masses based on their masscharge ratio when passed through a magnetic field. Itzaj: The last surviving group of Lowland Maya speakers native to the northern Petén rainforest of Guatemala. J jaguar (Felis onca): A large feline mammal having a tawny coat spotted with black rosettes. L ladino: In Spanish America a person of mixed ancestry. lemma: The outer or lower of the two bracts or scales surrounding the flower of a grass. lesion: Injury, injurious change in texture or action of an organ of the body. life form: The level of folkbiological taxonomy immediately subordinate to the folkkingdom level. In every culture there are a handful of plant and animal life forms that include all, or most, subordinate generic species. For example, Itzaj recognize the life forms che' (tree), pok'~che' (herb/bush), su'uk (grass), ak' (vine), Käy (fish), ch'iich' (bird including bats), b'a'al~che'+kuxi'mal ("walking animal" = mammal), b'a'al~che'+kujiltikub'aj ("slithering animal" = reptile), and mejen+b'a'al~che' ("small animal'' = invertebrate). linear mixing model: A model that predicts (1) that carbon atoms from all the macronutrients ingested as foods are evenly distributed to bone collagen (organic phase) and to bone and enamel carbonate (inorganic phase) and (2) that the isotopic composition of both collagen and carbonate is a product of the isotopic composition of all carbons consumed (i.e., the whole diet). lithic: Of stone. loglinear: Logarithmicbased representation. lyophilize: To freezedry. M macrobotanical: Plant remains observable with the naked eye. macronutrients: The three major nutritional elements in foods: proteins, carbohydrates, and lipids. maize (Zea mays): Any of numerous cultivated forms of usually tall annual cereal grassbearing grains or kernels on large ears. malanga (Xanthosoma sp.): A herbaceous Neotropical perennial with edible subterranean tuberous stems comparable in nutritional value to the potato. mandible: The jaw. manioc (Manihot esculenta): Cassava. A shrubby Neotropical plant grown for its edible, large, tuberous, starchy roots. It is eaten after leaching and drying to remove cyanide. Cassava starch is also the source of tapioca. mano and metate: Manos are stones used to grind food against flat stone metates. maxilla: The bony complex that holds the upper teeth. medullary cavity: The hollow portion in the middle of long bones. midden: A refuse heap.
Page 248
milpa: A maize field created by slashandburn techniques. mollusc: Any of numerous invertebrates of the phylum Mollusca, typically having a soft unsegmented body, a mantle, and a protective calcareous shell. multivariate statistics: Quantitative analysis involving many variables. mussel: Any of several edible marine or freshwater bivalve molluscs. N NaOH: Sodium hydroxide, an alkaline compound. nance (Byrsonima crassifolia): Small trees or shrubs with edible sweet yellow fruits. nutrient routing model: A model that predicts (1) that carbon atoms from the three macronutrients ingested as foods are differentially distributed to bone collagen and to bone and enamel carbonate; (2) that carbon atoms from proteins are preferentially routed to collagen while carbon from all macronutrients is incorporated into carbonate; and (3) that the isotopic composition of collagen reflects that of proteins consumed whereas the isotopic composition of carbonate reflects that of the whole diet. O optimal foraging: A subsistence pattern in which advantage is taken of available resources without much patterning or advance planning. orbit: The bony structure that encases the eye. osseous: Composed of, containing, or resembling bone. osteobiography: The reconstruction of individual life history through the study of the skeleton. osteology: The study of the skeleton. osteophyte: A bony outgrowth. P paca (Cuniculus paca): A large, nocturnal, burrowing, spotted rodent that lives on plants and fruit and is hunted for its edible flesh. palea: The upper or inner, thin, membranous tract enclosing the flower in grasses. paleodiet: The study of diet in extinct systems. paleoethnobotany: The study of plant remains and their uses in extinct systems. paleonutrition: The study of nutrition in extinct systems. paleopathology: The study of health, disease, and nutrition from skeletal remains. palm or coyol (Acrocomia mexicana): An evergreen tree often used for producing an alcoholic beverage. palmae: Tropical or subtropical monocotyledonous trees or shrubs having a woody, usually unbranched trunk and large evergreen featherlike or fanshaped leaves growing in a bunch at the top. parenchyma (parenchymatous tissue): A soft tissue made up of thinwalled, undifferentiated living cells with air spaces between them, constituting the chief substance of plant leaves and roots, the pulp of fruits, the central portion of stems, etc. patrilineal: Designation of descent, kinship through the father instead of the mother. pathoses: Abnormal conditions. peccary (Tayassu sp.): Any of several piglike hoofed mammals of the family Tayassuidae having long, dark, dense bristles. peduncle: The stalk of a solitary flower. periapical abscess: An abscess, or infection in which pus is produced, that occurs near the tip of a tooth. periodontitis (periodontal disease): Inflammation of the periodontal or gum tissue. periosteal: On the outside surface of bone. periostitis: Inflammation of the periosteum, the layer of tissue that is found on the outside surface of bone. per mil: A unit of onethousandth. perennial: Having a life span of more than two years. photosynthesis: A process by which carbonates are synthesized from carbon dioxide and water using light as an energy source, most notably in green plants. phytolith: A small opaline or silicate inclusion in plant cells. piscine: Of fishes. plazuela: A raised quadrangular court with several small square or oblong architectural structures grouped around it. postcranial: Relating to the portion of the skeleton that does not include the head, i.e., the torso and limbs. postmortem: After death. porotic hyperostosis: Skeletal lesions associated with irondeficiency anemia involving the outer table of cranial vault bones and the roof areas of the eye orbits (cribra orbitalia). puma or cougar (Felis concolor): A large, powerful, wild cat found in mountainous regions and having an unmarked, tawny body. R ramón: See breadnut. reducción: A reserve, or area where natives who had been converted to Christianity by the Spanish were placed to live. rhyolite: A finegrained extrusive volcanic rock, similar to granite in composition. ridged fields: A type of selfirrigating agriculture in which earth (usually swampy) is piled into ridges that are planted. rind: A tough outer covering, the skin of some fruits. rostrum: An elevated platform, often beakshaped. S sacroiliac joint: The area in the pelvis where the sacrum articulates with the iliac bone. savanna: Extensive open grassy plain. sea urchin: Any of various echinoderms of the class Echinoidea, having a soft body enclosed in a round, symmetrical, calcareous shell covered with long spines. seriation: A typological arrangement according to series.
Page 249
sexual dimorphism: A difference in size or shape that exists between males and females, often involving body height. socioeconomic: Of or involving both social and economic factors. squash (Cucurbita moshata): See cucurbit. stratified: Layered, or in the case of status, differences based on wealth. stratigraphy: The arrangement of rocks, or geological deposits, in layers or strata. subpopulation: A subdivision of a population that has common, distinguishing characteristics. subtrochanteric: An area on the proximal end of the femur below a feature called the trochanter. sweet potato (Ipomoea batatas): A Neotropical vine with roseviolet or pale pink funnelshaped flowers and cultivated for its fleshy, tuberous orangecolored roots, which are eaten cooked as a vegetable. swidden: Slashandburn. symbiont: Dissimilar organisms living together in a close association that may be, but is not necessarily, of benefit to each. synchronic: Occurring at the same time. synergistic: The simultaneous operation of separate agencies, which when taken together create a greater effect than would the sum of their individual effects. T taxa: Units of biological classification, e.g., kingdom, phylum, class, order, family, genus, and species. tranchet bit: A slicing tool. trophic: Of or involving the feeding habits or food relationship of organisms at different levels in the food chain. U ubiquitous: Being or seeming to be everywhere at the same time, omnipresent. V vascular tissue: Characterized by or containing vessels for the transmission or circulation of plant fluids. W wild grape (Vitis sp.): A plant with woody vines bearing clusters of edible berries and widely cultivated in many species and varieties. Y yam (Dioscorea trifida): Any of numerous tropical vines of the genus Dioscorea, many of which have edible starchy, tuberous roots. Z zapote (Pouteria sp.): A tree up to 24 m high with a dense leaf canopy that bears edible, sweet, spherical fruits with a single large seed. zooarchaeology: A branch of archaeology that involves the study of animal remains.
Page 250
CONTRIBUTORS Scott Atran Institute for Social Research University of Michigan Ann Arbor, MI 481061248, USA Centre National de la Recherche Scientifique CREAEcole Polytechnique) Port Vendres, France Shannon Coyston Department of Anthropology McMaster University Hamilton, ON L8S 4L9, Canada Marie Elaine Danforth Department of Anthropology and Sociology University of Southern Mississippi Hattiesburg, MS 394065074, USA Kitty F. Emery Department of Anthropology State University of New York Potsdam, NY 13676, USA James F. Garber Department of Anthropology Southwest Texas State University San Marcos, TX 78666, USA David Glassman Department of Anthropology Southwest Texas State University San Marcos, TX 78666, USA David L. Lentz New York Botanical Garden Bronx, NY 10458, USA Ann L. Magennis Department of Anthropology Colorado State University Fort Collins, CO 80523, USA David Millard Reed Research Fellow in Biostatistics Department of Anthropology The University of Michigan 1020 LSA, 500 S. State St. Ann Arbor, MI 481091382, USA Henry P. Schwarcz Department of Geology McMaster University Hamilton, ON L8S 4L9, Canada Leslie C. Shaw Department of Sociology/Anthropology Bowden College 7000 College Station Brunswick, ME 04011, USA Rebecca Storey Department of Anthropology University of Houston Houston, TX 77204, USA Edilberto Ucan Ek' Director Herbolaria Maya Uman, Yucatan Christine D. White Department of Anthropology University of Western Ontario London, ON N6A 5C2, Canada Steven L. Whittington Hudson Museum University of Maine, 5746 Maine Center for the Arts Orono, ME 044695746, USA Lori E. Wright Department of Anthropology Texas A&M University College Station, TX 77802, USA
Page 251
INDEX A abnormal bone formation due to infectious disease at Ambergris, 128 Abrams and Rue (1988) use pollen cores to associate Copán collapse with environmental degradation, 151 Acbi phase at Copán, 154 achiote, moderately high Ca levels in, 204 Acropolis at Copán had densest recorded population for a Late Classic Maya site, 169 adult stature calculations at Copán, 173 agave grown as prolific home garden plant at Cerén, 12 agricultural trade model implications of Peten sites, xxii Albion Island Colha stone tools at, 86 early selfsufficiency in meat resources acquisition, 97 alkaline earth ratios in human bone at Altar de Sacrificios, 212213 Dos Pilas, 213 Seibal, 212213 alkaline processing of maize calcium content of maize greatly increases with, 206 pellagra could result without, 206 Altar de Sacrificios alkaline earth ratios in human bone at, 212213 Amblema clam unlikely as source of lime at, 212 average stature slightly higher than at Ambergris, 129 description of site of, 199 diagenetic alteration of archeological bone mineral at, 209 differences in elemental composition of bone at, 210212 inorganic portion of bone analyzed elementally for, xxi leg bones longer in Preclassic than in Early Classic, 112 peak in caries frequency at, 163 porotic hyperostosis relatively high rates, 129 reasons for variability in elemental ratios at, 212 Saul (1972) says scurvy present at, 164 Saul's classic osteobiography on, xii scurvy as a cause for loss of teeth, 162 shows predicted increase in sexual dimorphism, 108 soils, 202 stature estimation at, 123 striking decrease in stature among males at, 107 alveolar loss in Ambergris teeth sample, 125126 suggest caution suggested in applying measurements, 125 Amaranth not recovered archaeologically from the Maya area, 13 Ambergris Cay (See also sites of Chac Balam, Ek Luum and San Juan) abnormal bone formation due to infectious disease at, 128 alveolar loss in teeth sample of, 125126 Archaeological Project, 119130 average stature at, 129 carious lesions in teeth sample, 124125 description of, 119 enamel defects at, 127 fish use at. See fish gastropods recovered archaeologically from, 121 infectious disease rate relatively high at, 129 low frequency of porotic hyperostosis at, 129 maize at. See maize males with elite social position among tallest individuals at, 123 porotic hyperostosis at, 127. See also porotic hyperostosis time sequence at, xviiixix Amblema clam unlikely as source of lime at Altar de Sacrificios, 212 Ambrose (1993), review of stable isotope paleodietary research, 184 animal use change patterns in Postclassic and transition to Colonial, 62, 73 antemortem tooth loss, causes of, 152 aquatic resources increasing importance, 75 armadillos have Ba/Calike herbivores but are somewhat Sr enriched, 206 arux (wood fairies) as guardians of breadnut and chicozapote trees, 55 Atran, Scott. On extending the use of ethnobotany, xvixvii Augustin site in Guatemala ChapaloteNalTel complex, 4 avocado (Persea americana) presence at Copán, 185 B Barium (Ba) extremely low natural abundance hinders paleodietary reconstruction, 214215 peccary (Tayassu sp.) eat some foods not eaten by true herbivores that are deficient in, 205206 squash seeds from Altar de Sacrificios very high in, 204 Ba/Ca (Barium/Calcium) animal foods have lower values than plant foods, 206 armadillos somewhat Sr enriched although Ba/Calike herbivores, 206 Bajo de Santa Fe at Tikal, 206
Page 252
Bajo la Justa geointensive production systems, 20 Barton Ramie females not more susceptible than males to caries, 162 shows decrease in sexual dimorphism, 108 while males becoming taller the females become shorter, 112 Willey (1965) suggests increasingly meager nutrition at, 106 bean (Phaseolus vulgaris) evidence for presence at Copán, 185 biointensive economy of central Petén, 20 "biopurification" principle as basis of postulated use of alkaline earths as trophic indicators, 198 black maize processing from near San Juan Comalapa, 207 Bogin (1995) on Maya average increase in stature when living in United States, 105 Bogin and colleagues (1992) discovered Guatemalan sexual maturation less delayed in girls than in boys, 105 bone chemistry studies, xxixxii, chapters 10, 11, 12 bones counts as a method for the quantification of faunal data, 85 elemental analysis for, 201 isotropic analysis for, chapters 10, 12 paleopathology for, chapters 5, 6, 9 botanical and faunal analyses absolutely fundamental to paleodiet research, xvi bottle gourd (Lagenaria sp.) evidence for presence at Copán, 185 bow and arrow technology importation, 75 breadnut for Itzaj (See also ramon) as plant species most deserving respect and protection, 55 have no doubt that ancestors tended and used, 55 Bronson hypothesis of Maya use of root crops, 11 C C3 plants are most other than maize consumed by the Maya, 184 C4based diets at Copán of dogs, pacas, and peccaries, 187 Cahal Pech, 83 marine fish remains at, 91, 93 calcium (Ca) content of maize greatly increases with alkaline processing, 206 fish and snail values reflect water composition not dietary behavior, 206 importance of, 198 moderately high levels of, 204 mollusc consumption could provide a measurable contribution of, 206 Pasión soils high in soluble Ca, 202 rich plants, 204 calculus increase over time at Kichpanha, 142 scored according to a fourpoint scale, 140 Calvin (C3based) terrestrial plants categorized by carbon isotope composition and photosynthetic type, 184 camote, Itzaj use of, 21 carbon isotope collagen studies used to assess maize importance, 224 caries (archaeological). See also dental caries; caries lesions Copán frequency of, 156157 frequency increase during terminal occupation at Kichpanha, 142 study at Kichpanha of, 141 caries lesions in Ambergris teeth sample, 124125 modifying factors that can affect the site and speed of, 134 Cerén agave grown as prolific home garden plant at, 12 bean remains preservation at, 5 chile peppers present at, 1011 Classic period Mayan archaeological plant remains from, 4 cotton use at, 11 fruit trees grown in household courtyards at, 12 manioc presence at, 11 oil extracted from cotton seeds at, 11 sieva beans at, 5 Cerros, 83 Colha stone tools at, 86 Formative period Mayan archaeological plant remains from, 4 presence of chile peppers, 10 rapid growth as port in Late Preclassic, 95 Chac Balam on Ambergris Cay. See also Ambergris Cay burials at, 122 description of, 120 molluscs recovered from site of, 121 Chalcatzingo systematic differences in strontium content of skeletons, 197 ChapaloteNalTel complex (Isthmian Alliance) most maize fragments of, 4 characterization of social organization and complexity paleodiet research contribution to, xv charcoal records provide documentation of wood changes over time, 14 chaya Carich plant, 204 chayote (Sechium edule) evidence for presence at Copán, 185 chemical techniques for analyzing bone revitalized paleodiet, xii Chichen Itza, relatively high rates of porotic hyperostosis at, 129 chicken, Itzaj did not classify as bird, 50 Chihuahua and Sonora dry caves, ChapaloteNalTel complex at, 4 children remains sicker and in poorer health than the living population, 171 chile peppers Cuello earliest example of, 10 moderately high Ca levels in, 204 Chimaltenango, processing of maize from, 207 chipilin, Carich plant, 204 Cihuatan Maya archaeological plant remains, 4 ciruela (Spondias sp.), evidence for presence at Copán, 185 Cittarium pica, 121 Clarke and Hirsch (1991) suggest caution in applying alveolar loss measurements, 125
Page 253
Cobá, Classic period Maya archaeological plant remains from, 4 Cobweb Swamp, evidence for wetland agriculture in, 92 Colha, 83 deer use at, 94 dog use at, 94 limitations regarding contextual data faunal information, 85 marine fish remains at, 91 marine reef fish remains in late Middle Preclassic, 93 mass producer of stone tools in Late Preclassic, 95 meat resources selfsufficiency of early settlers, 97 Preclassic vertebrate faunal assemblage, 8788 shift to use of mammals in Late and Terminal Preclassic, 93 site of, 8688 stone tools production at, 86 turtles at, 94 collagen carbon 13 values underestimate energy portion of the diet, 225 preservation and preparation techniques, 187 collapse of Classic Maya society ecological explanation, x colonial changes in animal usage have Postclassic roots, 62 Lamanai utilization of species in, 72 Tipu utilization of species in, 72 common bean in Maya area, 5 Copán, animals of C4based diets of dogs, pacas, and peccaries at, 187 deer at. See deer dogs at. See dog freshwater snail. See Pachychilus freshwater snail jaguar (Felis onca) identified at, 186 pacas (Cuniculus paca) at. See pacas (Cuniculus paca) Pachychilus freshwater snail at, 186 peccary (Tayassu sp.) C4based diets at, 187 puma or cougar (Felis concolor) identified at, 186 Copán, disease at different disease incidence among status groups, 175176 multiple hypoplasias at, 176 nutrient deficiency associated with demise of Classic at, xiii, 20 periodontal disease present in 88.2 percent of adult skulls, 162 porotic hyperostosis at, 129, 175 scurvy at, 162, 164 study of paleopathology at, xx Copán, people of adult stature calculations at, 173 average stature lower than at Ambergris, 129 collapse associated with environmental degradation, 151 densest Late Classic Maya site population at Acropolis of, 169 Early Classic people bigger than Late Classic counterparts, 106 environmental degradation associated with collapse at, 151 females at. See females at Copán human bone specimens description, 186187 Late/Terminal Classic generalized nutritional stress, 178 males associated with the absence of caries, 159 nutrient generalized stress experienced by society of, 178 organic portion of bone analyzed isotopically for, xxi physiological stress among subpopulations of, 186 sexing and aging human remains at, 154155 study of commoners at, 151 Copán, plants of avocado (Persea americana) presence at, 185 bean (Phaseolus vulgaris) evidence for presence at, 185 bottle gourd (Lagenaria sp.) evidence for presence at, 185 charcoal documents wood changes through time, 14 ciruela (Spondias sp.), evidence for presence at, 185 Classic period Maya archaeological plant remains from, 4 complementary studies of paleopathology at, xx coyol palm (Acrocomia mexicana) at, 13, 170, 185 Formative period Maya archaeological plant remains from, 4 frijolillo (Crassia occidentalis) present at, 185 grape (wild) evidence for presence at, 185 maize very important for low status diet, 159160. See also maize nance (Byrsonima crassifolia) presence at, 185 pollen core interpretation of collapse, 151 zapote (Pouteria sp.) at, 185 Copán, site of Acbi phase at, 154 Acropolis at, 169 description and history of occupation of site of, 183184 early Coner phase at, 154, 159 House of the Bacabs, 171172 new class of site at, 153 stable isotope data, criticism of previous interpretation of, 162 time sequence at, xviii Type I through Type IV residential sites at, 153 type of sites at, 153 Copán, teeth of caries frequency at, 156157, 159, 163. See also caries (archaeological); dental caries horticulture associated with presence of caries at, 159 males associated with the absence of caries, 159 methods used to study teeth at, 155156 recording of caries at, 154 Cotton use in the Maya area, 1112 Cowgill and Hutchinson (1963) documented differential leaching of Sr relative to Ba, 206 coyol palm (Acrocomia mexicana) archaeobotanical evidence for presence at Copán, 185 cultivated in prehistoric times, 13 introduced by Maya into Copán, 13 one of most common remains in Late Classic Copán, 170 Crassulacean acid metabolism (CAM) terrestrial plants categorized by
Page 254
carbon isotope composition and photosynthetic type, 184 Cretaceous Pee Dee belemnite formation (PDB) carbon reference for stable isotope ratio, 184 Cuello, 83 charcoal records document wood changes through time, 14 chile peppers earliest example at, 10 Colha stone tools at, 86 fish use increases during late Middle Preclassic at, 93 hydrological peak in water table levels at, 93 manioc presence at, 11 marine fish remains at, 91 marine reef fish remains in late Middle Preclassic, 93 meat acquisition selfsufficiency of early settlers, 97 plant remains of Mayan Formative period earliest from, 4 sample size overwhelms data from smaller sites, 113 cultural dynamics understanding from the bottom up, move towards, xiv cultural process overlapping with environmental restrictions, xxi curassow at Pacbitun, 223 D Danforth, Marie Elaine growth compared to stature as a measure of adaptability, xix deer Colha use at, 94 Copán archaeological identification, 186 Copán C3based diets, 188 Copán irregular and minimal use during Coner phase, 191 Pacbitun presence, 223 deforestation associated with demise of Classic Maya civilization, 20 since 1962, 20 deformation both cranial and dental common at Ambergris, 123 degenerative joint disease at Ambergris, 128 demineralization, dental caries as, 135 DeNiro (1987) review of stable isotope paleodietary research, 184 dental calculus recovery of food particles from, 136137 useful for making inferences abou dietary consistency, 136 dental caries. See also caries (archaeological) essential factors in the occurrence of, 134 infectious, transmissible disease, 152 inverse relationship with calculus, 143 three principal factors involved in promotion of, 134 dental enamel hypoplasias, 172173 dental plaque deposits formation, 135 dentition value for study, 134 diagenesis alteration of archaeological bone mineral possibility, 209 more dependent on burial context than sample age, 230 dietary patterning in nonspecific health indicators, xxii dietary baseline data importance, 198 dietary inferences must be made on sitespecific basis, 202 dietary protein, isotopic composition of collagen may largely reflect carbon isotope ratio of, 221 disease burden part of many Maya Late Classic decline models, 107 diversity of each zooarchaeological community methods used to quantify, 65 Dobney and Bothwell (1987), calculus scored according to a fourpoint scale of, 140 dog (Canis familiaris) Colha use, 94 Copán archeological presence, 186 Copán C4based diets, 187 Pacbitun presence, 223 Petén Preclassic and Postclassic importance, 94 Dos Pilas agricultural trade model implications, xxii alkaline earth ratios in human bone at, 213 bone elemental composition differences with Seibal reflect environmental differences, 210 bone inorganic portion analyzed elementally for, xxi chile peppers at, 10 consumption variation of plant among social groups, 213214 diagenetic alteration of archaeological bone mineral at, 209210 plant collection at, 204 site description, 200 soils, 202 E earliest Maya archaeological plant remains from Cuello, 4 early Coner phase at Copán, 154 teeth associated with absence of caries, 159 early Middle Preclassic, 9192 garden hunting in, 88 ecological explanation for collapse of Classic Maya society, x economic infrastructure collapse associated with demise of Classic Maya civilization, 20 ecosystem, basis of accurate analysis of, 65 Ek Luum (Ambergris). See also Ambergris Cay analysis of faunal materials, 121 reliance on shallow water and reef waters environment at, 122 Elias et al. (1982) Sr/Ca in alpine stream water higher than in soil moisture, 206 elite males taller than commoners at Tikal and Copán, 176 El Posito, Colha stone tools at, 86 Emery, Kitty. how complexity of tropical ecosystems might affect the analysis of faunal material, xviii enamel defects at Ambergris, 127 environmental degradation, Copán collapse associated with, 151 environmental factors significant in causing change in stature, 104 environmental restrictions, cultural process overlapping with, xxi environment determining adult Maya stature, 105106 epazote, Carich plant, 204
Page 255
Evans (1973) found caries mean frequency peak during Tayasal area Middle Classic, 163 F fats may reduce caries, 135 faunal studies at Postclassic and Colonial lowland Maya sites, reasons for, 62 females at Copán caries associated with, 159 maize eaten less with increasing age, 192 Petén or Yucatan more susceptible than males to caries, 162 fish Ca values reflect water composition not dietary behavior, 206 increasing Preclassic use of small, 9293 largest sample of faunal material from Ambergris sites, 121 seasonal high densities in river flood plains wetlands, 9293 flotation use in sampling plant remains, 3 folkbiological taxonomies most stable, widely distributed and conservative cognitive structures in any culture, 47 folkspecifics, foreign organisms often initially assimilated to generic species as, 48 foreign organisms, assimilation of, 48 Formative period Maya archaeological plant remains, 4 Fourier transform infrared spectroscopy (FTIR), 228 freshwater snail. See also Pachychilus freshwater snail identified archaeologically at Copán, 186 frijolillo (Crassia occidentalis) present at Copán, 185 fructose intolerance people are essentially caries free, 135 fruit trees grown in household courtyards at Cerén, 12 functional and social contexts importance in study of archaeological faunal material, 8485 G Garber, James and David Glassman provide methodological model to apply to small samples, xix garden hunting in early Middle Preclassic, 88, 9192 Garson (1980) on seasonally high densities of small fish, 9293 gastropods recovered archaeologically from Ambergris sites, 121 generic species not isomorphic with scientific species or genera, 48 where relationships between organisms maximally covary, 4748 Genovés formula, 107, 173 Gleser formula for stature estimation, 123 Gossypol as male contraceptive, 11 grape (wild) evidence for presence at Copán, 185 Guatemalan Pacific coast, processing of maize from, 207 Guatemalan sexual maturation less delayed in girls than in boys, 105 H hackberry (Celtis sp.) evidence for presence at Copán, 185 Harris lines, 126127 HatchSlack (C4based) terrestrial plants categorized by carbon isotope composition and photosynthetic type, 184 Haviland (1967) found statistically significant decrease in stature at Tikal, 106107 work on stature at Tikal, xixii health agriculture consequences explain reduction in stature, 107 not sufficient for survival, xix herpetofauna as a "residual" Itzaj life form that lacks a conceptually distinctive role, 50 hill slope terraces at Seibal, 212 Hillson (1986) increase in diet protein associated with decrease in acidproducing bacteria, 135 Hodges (1985) found periodontal disease present in 88.2 percent of adult Copán skulls, 162 horticulture associated with presence of caries at Copán, 159 House of the Bacabs provided most of Copán skeletal sample, 171172 human bone specimens from Copán description, 186187 human diet in Maya lowlands reconstruction problems, xi I ideological with materialist economic modes of explanation postprocessual move to connect, xivxv infectious disease at Ambergris, abnormal bone formation due to, 128 infectious lesions on bone, 172 infrared spectroscopy, rigorous protocol for the use of, 230 intensive agriculture with environmental degradation as explanation for Classic Maya collapse, x intermediate taxa functioning, 5051 irondeficiency anemia, xiii Itzaj data base for study of useful plants of, 19 FolkGeneric Species subordinate taxa, 49 lack of knowledge of prehispanic culture, claims for, 21 prophecies as negotiating ploy, 21 J jaguar (Felis onca) identified archaeologically at Copán, 186 junco palm enriched in Sr/Ca relative to other flora, 204 jute at Pacbitun, 223. See also Pachychilus freshwater snail K Kate's Lagoon, 137 Katzenberg (1992) review of stable isotope paleodietary research, 184 Keegan (1989) review of stable isotope paleodietary research, 184 Kichpanha calculus increase over time at, 142 caries frequency increase during terminal occupation, 142 Classic importance of maize as percentage of the diet at, 138 Colha stone tools at, 86 description of site of, 137 dietary change at the Lowland Maya site of, 133146 investigates diet change at, xx longest time sequence at, xviii
Page 256
maize did not constitute the overwhelming percentage of the diet, 138 methods used to study bones and teeth, 140 Preclassic maize not overwhelming percentage of the diet, 138 kwashiorkor presence in Guatemala, 104 L Lacandon Maya use of snail shells as source of lime, 207. See also Pachychilus freshwater snail Laguna Petexbatún. See Petexbatún, Laguna Lamanai, xviii, 222 burial sample size, 225 caries frequency peak at, 163 caries reports at, 143144 Colonial era, reasons for changes, 77 diagenesis testing at, 228230 elemental analyses attempted at, 197 females significantly more susceptible than males to caries, 162 heterogeneity changes are more dramatic than Tipu over time, 66 importance of work at, 61 increasing use of reef resources during Postclassic and Historic, 232 inorganic portion of bone analyzed isotopically for, xxi isotopic analysis of mineral portion of human bones and teeth, xxii limeencrusted colander pots, 207 maize and maritime increase with population increase, 239240 phytate removal from maize effective beginning in Postclassic, 145 status differences based upon land animal consumed by elite, 236 survival rationale of, 221 Lamanai, plant and animal species at avian Late Postclassic importance at, 69 backdoor gardens or multispecies horticulture at, 144 Colonial utilization of species at, 72 importance of pristine canopy forest species at, 67 Late Postclassic utilization of species at, 6970, 72 Middle Postclassic utilization of species at, 6667 La Perra Cave in Tamaulipas ChapaloteNalTel complex, 4 Larsen et al. (1991) essential factors in the occurrence of caries, 134 Late and Terminal Preclassic shift to use of mammals at Colha, 93 Late Classic decline in stature not ubiquitous homogeneous phenomenon, 112 late Coner phase at Copán, 154 teeth associated with presence of caries, 159 Late Postclassic utilization of species at Lamanai and Tipu, 6970, 72 Lentz, David. review of plant use among the Maya, xvi lichens not classified as Itzaj plants, 49 life forms may differ from culture to culture, 47 Linares (1976) gives primary reason for leaving the residential area, 83 linear mixing model to describe uniform distribution of carbon, 224 local ecology had a profound effect on food consumption, xxiii local meats of high prestige as elite markers, 240 Longyear (1952) noted Early Classic individuals bigger than Late Classic counterparts at Copán, 106 Lovejoy (1985), dental attribution evaluated using system of, 124 lower status males of Late Classic similar in stature to modern Maya, 113114 M macal root enriched in Sr/Ca relative to other flora, 204 Itzaj use of, 21 Macal river system, 64 Magennis, Ann. investigates diet change at Kichpanha, xx maize Copán archaeobotanical evidence for presence, 185 C4 plant, 184 unlike other plants and so not included with other plants, 50 Ambergris importation of, 121 Pacbitun consumption reached peak during Classic, 139 caries promotes, 144 not overwhelming percentage of the diet at Kichpanha, 138 domesticates fed maize more accessible to elite at Pacbitun, 239 low status diet importance at Copán, 159160 processing technique change as promoted caries, 145 staple diet at Copán indicated by carbon values, 187 variations in Maya Lowland use, 139 marine resources comparison with, xxii males at Copán associated with the absence of caries, 159 preference not seen in analysis of childhood health indicators, 113114 with elite social position among tallest individuals at Ambergris, 123 malnutrition part of many models for the fall of the Late Classic Maya, 107 mamey fruit use by Itzaj, 21 manioc Itzaj use of, 21 presence of, 11 as promoting caries, 144 use of, 138 MannWhitney tests, 108, 202 manos and metates ubiquitous throughout the Copán Valley, 185 marine fish remains at Belize sites, 91 marine reef fish remains identified at Cuello, Cahal Pech and Colha in late Middle Preclassic, 93 market in Preclassic Maya communities, 95 Márquez Morfín (1984) systematically studied stature patterns in fifteen prehistoric populations over time, 107 Maya average increase in stature when living in United States, 105
Page 257
Biosphere Reserve, 20 collapse shortening given far more emphasis than is warranted, 114 diet reconstruction a worthwhile pursuit, ix modern diet meeting young children protein requirements, 104 past bigger than their modern descendants, support for, 108 stature decrease as folklore of archaeologists, 103 stature reports of modern rural Guatemalan, 108 today among shortest populations in the world, 103 Melongena melongena, 121 Middle Postclassic utilization of species at Lamanai and Tipu, 6667 Miksicek (1991) identified hydrological peak in Cuello water table level, 93 mineral grit from grinding stones can affect bone ratios, 208209 minimum number of individual estimations (MNI) , 64 mollusc consumption could provide a measurable Ca contribution, 206 Monte Albán in Oaxaca ChapaloteNalTel complex, 4 morphological indicators of stress in Ambergris population, 126 mosses and liverworts not classified as Itzaj plants, 50 multiple hypoplasias, status group 2 males at Copán have highest percent, 176 mushrooms not classified as Itzaj plants, 50 N Naco, Postclassic period Mayan archaeological plant remains from, 4 nance (Byrsonima crassifolia) presence at Copán, 185 Newbrun (1982) fructose intolerance people are essentially caries free, 135 three principal factors involved in promotion of dental caries, 134 New River Lagoon system, 64 Nickens (1976) health consequences associated with agriculture explain reduction in stature over time, 107 NISP (number of identified specimens), 85 nitrogen isotopes composition of collagen used to identify source of dietary protein, 224 distinguish between marine animals and terrestrial plants, 184185 nitrogen reference for stable isotope ratio ambient air (AIR), 184 ''nixtamal" methods for tortillas and tamales, 207 Nohmul, Colha stone tools at, 86 nomenclature used alone to indicate taxonomic status can be misleading, 51 nopal contains much more Ca than most plants, 204 Norr (1995) review of stable isotope paleodietary research, 184 nutrient deficiency associated with demise of Classic Maya civilization, xiii, 20 generalized stress, Copán society experiencing, 178 O Odum's richness measure, 65 Ontario populations of horticulturists, caries comparison with Copán, 161 outside sources for changes in animal use practices, 74 P pacas (Cuniculus paca) archaeologically identified at Copán, 186 C4based diets at Copán of, 187 pacaya palm Carich plant, 204 Pacbitun abandonment rationale, 221 burial sample size, 226 curassow at, 223 deer at, 223 description of site of, 222223 diagenesis testing at, 228230 diet was relatively uniform in carbon isotopic composition, 238 dog remains at, 223 inorganic portion of bone analyzed isotopically for, xxi isotopic analysis of mineral portion of bones and teeth, xxii jute at, 223. See also Pachychilus freshwater snail maize consumption reached peak during Classic at, 139 maizefed domesticates more accessible to elite at, 239 maize increasing dependence until Terminal Classic, 236 maize production increase with population growth, 239240 post occupation internment with substantially more C3 foods, 236 rabbit at, 223 turkey at, 223 Pachychilus freshwater snail Copán archaeological presence, 186 Lacandon use as source of lime, 207 Pacbitun jute, 223 paleodietary reconstruction with alkaline earths hindered by extremely low natural abundance of Ba and Sr in this ecosystem, 214215 paleodiet revitalized by chemical techniques for analyzing bone, xii paleolimnological analysis of sediments in central lakes region of Petén, 20 paleopathological indicators as indirect reflection of diet and nutrition, 170171 research changes, xiixiii palms nuts moderately high Ca levels in, 204 widespread use of, 1213 papaya high Ca may be result of small green state of sampled fruit, 204 parrotfish remains found at Colha, 91 Pasión soils high in soluble Ca, 202 Pate (1994) review of stable isotope paleodietary research, 184 Patterson (1984) worldwide sample of caries frequencies, 159 peak in caries frequency at various Mayan sites, 163 peccary (Tayassu sp.) C4based diets at Copán of, 187
Page 258
eat some Badeficient foods not eaten by true herbivores, 205206 at Pacbitun, 223 pellagra could result without alkaline processing of maize, 206 Peña(1985) Preclassic peak in mean percentage of caries, 163 periapical abscesses around the teeth, 172 periodontal disease present in 88.2 percent of adult skulls from Copán, 162 periosteal reactions on surface of long bones, 172 Petén trend in Preclassic and Postclassic, importance of dog remains in, 94 Petexbatún, Laguna waterlily contains the most Sr of all sampled plants, 204 physiological stress among subpopulations of Copán, 186 pine where it does not now grow suggests longdistance transport, 15 piñuela contain much more Ca than most plants, 204 enriched in Sr/Ca relative to other flora, 204 plantains, Itzaj use of, 21 plants ash as a possible source of lime, 208 mutually exclusive life form groupings of, 49 plant samples elemental analysis for, 201 Pohl (1976) identified Petén trend in Preclassic and Postclassic importance of dog remains, 94 Pomacea snail shells as source of lime, 207 Popol Vuh connection of ideological with materialist, xv population growth associated with demise of Classic Maya civilization, 20 porotic hyperostosis, 172173 at Ambergris, 127 at Copán statistically significant distribution, 175 posole, Yucatec method for preparation of, 207 postprocessual move to connect ideological with materialist economic modes of explanation, xivxv primary goal of book, xv protein proportion in diet increase associated with decrease in acidproducing bacteria, 135 provisioning of ancient Maya populations, problems in understanding, x Pulltrouser Swamp Colha stone tools at, 86 Formative period Mayan archaeological plant remains from, 4 puma or cougar (Felis concolor) identified archaeologically at Copán, 186 R rabbit at Pacbitun, 223 ramon (See also breadnut for Itzaj) Itzaj use of, 21 little archaeological evidence for use as food, 1314 moderately high Ca levels in, 204 Rio Azalea Classic period Mayan archaeological plant remains, 4 Rio Honda wetlands proposed pattern of agricultural use, 92 Owe (1975) on modifying factors that can affect the site and speed of carious lesion development, 134 Rue (1987) interpretation of pollen core evidence that environmental degradation associated with collapse at Copán, 151 RugGnu (1981) suggest change in maize processing technique change increases caries, 145 Russell (1976) didn't find expected Maya pattern of sexual dimorphism, 105 S San Juan on Ambergris Cay. See also Ambergris Cay burials at, 122 description of, 120 molluscs recovered from site of, 121 San Juan Comalapa, processing of maize from, 207 sapote. See zapote. Sarteneja females as susceptible to caries as males, 162 Saul (1972) classic osteobiography on Altar de Sacrificios, xii decrease in stature among males at Altar de Sacrificios, 107 scurvy present at Altar de Sacrificios, 164 Saul and Saul (1989) small size as a successful adaptation among the Maya, 104 Schoeninger (1979a, 1979b) documents systematic differences in strontium content of skeletons, 197 Schoeninger (1998) criticism of previous interpretation of stable isotope data for Copán by, 162 Schoeninger and Moore (1992) review of stable isotope paleodietary research, 184 Schwarcz and Schoeninger (1991) review of stable isotope paleodietary research, 184 Scott (1979) methodology and scoring system evaluated dental attribution, 124 scurvy at Copán as a cause for loss of teeth, 162 unlikely to be present, 164 Seagraves (1974) emphasized environmental generalization advantages in preserving cultural stability in times of stress, 78 Sealy and Sillen (1988) trophic differences most marked for specific predatorprey relationships, 205 Seibal alkaline earth ratios in human bone at, 212213 description of site, 199200 diagenetic alteration of archaeological bone mineral at, 209210 environmental differences with Dos Pilas reflected in elemental composition of bone, 210 hill slope terraces at, 212 inorganic portion of bone analyzed elementally, xxi males becoming taller while females become shorter, 112 soils, 202 sexual dimorphism growth patterns need careful interpretation, 105
Page 259
Shaw, Leslie temperate climate analytic methods don't apply to Maya, xviixviii (1995) analysis of Ek Luum Ambergris faunal materials, 121 shell (modern) samples, elemental analysis for, 201 sieva beans at Cerén, 5 Simpson index of diversity, 65 slight infections, highest status have highest incidence, 175 small size as a successful adaptation among the Maya, 104 why would not be a successful adaptation among the Maya, 105 Smith (1984) methodology and scoring system used in dental attribution evaluation, 124 snails Ca values reflect water composition instead of dietary behavior, 206 snakes have "hidden" feet that only the speechless can see, 50 Sobolik (1994): paleonutrition understanding involves contributions from variety of data, 170 social complexity of Maya needs better definition and reconstructed more precisely, xiv soil samples elemental analysis, 201202 Spanish over reliance on cereal forced Itzaj over extention of maize cropping, 21 spatial variation in subsistence practice, demography and sociopolitical and socioeconomic systems, xiii Spearman's correlation coefficient tests of sample correlation, 65 tests of zooarchaeological community correlation, 66 squash growing in Maya area, 5, 10 squash (Cucurbita moschata) evidence for presence at Copán, 185 squash seeds from Altar de Sacrificios very high Ba and Sr of, 204 Strontium (Sr) extremely low natural abundance hinders paleodietary reconstruction, 214215 relative to Ba, documented differential leaching of, 206 squash seeds from Altar de Sacrificios very high in, 204 waterlily of Laguna Petexbatún contains the most Sr, 204 Sr/Ca (Strontium/Calcium) animal foods have lower values than plant foods, 206 in alpine stream water higher than in soil moisture, 206 plants enriched relative to other flora in, 204 stable isotope analysis of Copán bones indicates maize made up a large proportion of low status diet, 159160 stable isotopes in paleodietary research, 184185 stature patterns analysis in southern lowlands have scanty skeletal sample, 113 over time in fifteen northern Maya prehistoric populations, 107 status differences based upon type of land animal consumed by elite, 237 Steggerda (1932) reports of modern rural Guatemalan Maya stature, 108 Stewart (1949) earliest suggestion of Maya populations mean stature decline, 106 observation that ancient Maya bigger than their modern descendants supported by preliminary analysis, 108 Storey, Rebecca study of paleopathology at Copán, xx (1992): physiological stress among subpopulations of Copán, 186 Student ttests at Copán, 188191 Sullivan and Krueger (1981) chemical pretreatment procedures similar to those of, 226 T Tanner and colleagues (1982) trend toward greater stature related to increase in leg length, 105 tapir at Pacbitun, 223 Tayasal area, peak in caries frequency at, 163 teeth, study of, xviii temporal context of a faunal assemblage should not be the only contextual variable addressed, 84 terraces at Seibal, 212 Tikal average stature lower than at Ambergris, 129 Bajo de Santa Fe, 206 overwhelms data from smaller sites, 113 Classic period Mayan archaeological plant remains from, 4 Colonel Modesto Méndez discovery of, 21 Haviland (1967) work on stature at, xixii, 106107 nontomb males of Late Classic similar in stature to modern Maya, 113 shows predicted increase in sexual dimorphism, 108 stature estimation at, 123 while males becoming shorter the females gain in height, 112 Tikal and Copán elite males significantly taller than commoners, 176 time periods represented in book, xv Tipu, xviii Colonial utilization of species at, 72 elemental analyses attempted at, 197 generalized stability over time in overall diversity of species used, 66 importance of work at, 61 Middle Postclassic utilization of species at, 6667 reasons for Colonial era stability, 7778 secondary concentration on armadillo at, 67 tobacco not recovered archaeologically from the Maya area, 13 tomato not recovered archaeologically from the Maya area, 13 trade networks identification, paleodiet research contribution to, xv transport and communication links disintegration associated with demise of Classic Maya civilization, 20
Page 260
trend toward greater stature related to increase in leg length, 105 trophic differences most marked when specific predatorprey relationships are considered, 205 Trotter formula for stature estimation, 123 turkey at Pacbitun, 223 turtles use at Colha, 94 Tykot and coworkers (1996): maize did not constitute the overwhelming percentage of the diet at Kichpanha, 138 U Ursua y Aresmendi, Don Martin de. reflections on prehispanic culture, 2122 V van der Merwe (1982) review of stable isotope paleodietary research, 184 Villagutierre, errors in account of, 21 W warfare associated with demise of Classic Maya civilization, 20 waterlily collected in the Laguna Petexbatún contains the most Sr of all sampled plants, 204 Waterlow and Payne (1975): modern Maya diet meets over 95 percent of protein requirements for young children, 104 water table levels at Cuello hydrological peak, 93 Webster and Freter (1990) new class of site at Copán, 153 White (1988, 1994) determination of mean percentage of caries frequencies, 163 reports on caries at Lamanai, 143144 White and coworkers (1993) suggest maize consumption reached peak during Classic Pacbitun, 139 White and Schwarcz (1989) backdoor gardens or multispecies horticulture at Lamanai, 144 Lamanai limeencrusted colander pots, 207 suggest effective phytate removal from maize beginning in Postclassic Lamanai, 145 white maize processing, 207 Whittington, Stephen study of paleopathology at Copán, xx (1989, 1992) physiological stress among Copán subpopulations, 186 wild animals, rise of civilization using gives new perspective, 96 Wild Cane Cay Classic period Mayan archaeological plant remains from, 4 marine environment modification seen at, 4 Willey (1965) suggests increasingly meager nutrition responsibility for less rugged skeletons at Barton Ramie, 106 Willey and Leventhal (1979) Type I through Type IV residential sites at Copán, 153 Wing and Scudder (1991) increase in fish use during Cuello late Middle Preclassic, 93 "Woman of Cancuen," 214 worldwide sample of caries frequencies, 159 Wright (1994) raw elemental data presented by, 202 discussion of diagenetic alteration of archeological bone mineral, 209 Wright and Schwarcz (1996) rigorous protocol for the use of infrared spectroscopy, 230 Wright and White (1996) note diet trends that can be observed in the Maya Lowlands, 139 Y yellow maize of Chimaltenango processing, 207 Yucatan and Petén sites peak in caries frequency, 163 Yucatec method for preparation of posole, 207 Z zapote (Pouteria sp.) at Copán, 185 fruit, Itzaj use of, 21