COMMUNITY HEALTH AT NEMEA, GREECE: A BIOARCHAEOLOGICAL APPROACH TO THE IMPACT OF SOCIOPOLITICAL CHANGE IN BYZANTIUM By Jared Scott Beatrice A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Anthropology 2012 ABSTRACT COMMUNITY HEALTH AT NEMEA, GREECE: A BIOARCHAEOLOGICAL APPROACH TO THE IMPACT OF SOCIOPOLITICAL CHANGE IN BYZANTIUM By Jared Scott Beatrice This dissertation investigates the biological consequences of sociopolitical transformation in Byzantine Greece using a comparative analysis of two human skeletal samples (N=259) from the site of ancient Nemea. The skeletal samples from Nemea represent agricultural communities th th that lived at the site during the Early Christian (5 -6 centuries AD) and Middle to Late th th Byzantine (12 -13 centuries AD) periods. In the latter period, southern Greece was invaded by western Europeans and the Byzantine Empire experienced changes in political administration that had lasting, disruptive effects. Such events tend to dominate historical narratives and have influenced the interpretation of archaeological patterns, yet few studies have utilized human skeletal remains as an independent line of evidence to explore their impact on local Greek communities. The bioarchaeological approach employed in this research integrates osteological, archaeological, and historical datasets in the reconstruction of patterns of physiological stress and activity in order to test the following hypotheses related to the biological and social consequences of alterations to Byzantine Greek society: (1) political instability and invasions diminished the quality of life of the Middle to Late Byzantine community at Nemea; (2) different spheres of activity for men and women resulted in sex-based differences in health within the Early Christian and Byzantine communities; and (3) burial location at Nemea was influenced by the social hierarchy. The results of the skeletal analysis demonstrate that, despite living through a period of administrative problems and Frankish conquest, the individuals from the Middle to Late Byzantine period exhibit prevalence rates of paleopathological conditions and mortality patterns that are markedly similar to those of their Early Christian counterparts. While few skeletal health disparities were found between males and females, significant differences in the prevalence rates of cribra orbitalia and osteoarthritis in the Early Christian sample lend support to arguments concerning gendered expectations of behavior and a gendered division of labor in Byzantium. Finally, analyses of the spatial distribution of paleopathological conditions across the mortuary space of Nemea reveal limited evidence of patterns consistent with the organization of burials according to the social hierarchy. Instead, the organizing principles of the cemeteries of each period can be linked to changes in Byzantine socioreligious notions of the community and the family. With few exceptions, the results of this research demonstrate that the health and wellbeing of each community at Nemea was more dependent on local conditions than on the statelevel sociopolitical changes that figure prominently in Byzantine historical narratives. Copyright by JARED SCOTT BEATRICE 2012 To my parents, Mark and Jonelle, and my grandmother, Virginia v ACKNOWLEDGMENTS The successful completion of a doctoral dissertation is the culmination of years of coursework, research, and perseverance. It cannot be accomplished on one’s own. I have been extremely fortunate in the individuals who shaped my academic development and who provided much needed assistance throughout the writing process. I wish to thank Dr. Nancy Tatarek for igniting my interest in and passion for the discipline of anthropology. My advisor and dissertation committee chair, Dr. Todd Fenton, was responsible for cultivating and channeling that passion toward a professional career as a human osteologist. I will be forever grateful for the opportunities, personal interactions, learning experiences, and fun offered by our field seasons in Albania. I would also like to thank the additional members of my dissertation committee, Dr. Norman Sauer, Dr. William Lovis, Dr. Jon Frey, and Dr. Denise Demetriou, for their guidance and for the many hours of careful scrutiny spent on my dissertation. The final product is improved substantially because of your efforts. Locating a suitable skeletal collection is one of the most basic components of producing a dissertation focusing on human skeletal biology, but it is also frequently the most challenging. For access to the human skeletal samples from Nemea, I am grateful to Dr. Kim Shelton and to the Nemea Center for Classical Archaeology. During my field research I also benefitted enormously from advice and support from Dr. Jon Frey, Dr. Elizabeth Langridge-Noti, Leslie Lemire, Dr. Effie Athanassopoulos, and Christian Cloke. In order to expand the scope of my research, I relied on help from an outside source. I owe a sincere thank you to Dr. Voula Tritsaroli, who permitted me to use paleopathological data vi collected in her dissertation research for comparative purposes in my own. It is my hope that the present research can serve as a jumping off point for future interdisciplinary investigations of health and burial practices in Byzantine Greece. Inevitably the goals of a dissertation demand that one work to acquire new skills, however successfully. I am indebted to Sarah AcMoody and Dr. Jon Frey for their assistance building a functional GIS for the analysis of the burials at Nemea. It is impossible to estimate the number of frustrated hours, late nights, and cups of coffee that were associated with this task. It would also have been impossible for me to complete it without you. Perhaps the most important factor contributing to the successful completion of a doctoral dissertation is one’s support system. The graduate students (and former graduate students!) with whom I had the pleasure of being acquainted at Michigan State University are, I believe, as supportive as they come. Ideas and suggestions related to the content and presentation of this research were generated in discussions with Dr. Colleen Milligan, Dr. Lindsey Jenny, Dr. Michael Koot, Carolyn Hurst, Cate Bird, and Jen Vollner. I owe most of all, however, to Dr. Angela Soler, who provided me with advice, encouragement, and unconditional support when I needed it most. Last but certainly not least, I would like to thank my parents and my grandmother. As family members and as the individuals most intimately involved with my progress in graduate school, no one had more patience and understanding. You made this possible with every form of support imaginable. It is to you that I dedicate this dissertation. vii TABLE OF CONTENTS LIST OF TABLES xii LIST OF FIGURES xxi CHAPTER 1: INTRODUCTION Sociopolitical Change and the Byzantine Greek Countryside The Contribution of Bioarchaeology Bioarchaeological Research in the Eastern Mediterranean Nemea and its Environs Phases of Occupation and Activity Research Goals Outline of the Dissertation 1 1 4 5 10 14 18 20 CHAPTER 2: BIOARCHAEOLOGY AND MORTUARY ANALYSIS Physiological Stress and the Skeleton Enamel Hypoplasias Porotic Hyperostosis Cribra Orbitalia Periosteal Reactions Osteoarthritis Trauma Issues of Interpretation: “The Osteological Paradox” Mortuary Analysis: Historical Development and Theoretical Background Burial and Social Organization Postprocessual Approaches Gender and Mortuary Analysis Spatial Analysis Summary 21 21 23 24 26 28 29 31 34 35 36 39 43 45 47 CHAPTER 3: GREECE IN THE EARLY CHRISTIAN AND MIDDLE TO LATE BYZANTINE PERIODS: PLACING THE NEMEA COMMUNITIES IN CONTEXT 49 Greece and its Place in the Mediterranean World of Late Antiquity 50 Cities, Towns, and the Countryside 50 Religious Life 55 Gender 58 Health and Health Care 60 The Early Christian Community at Nemea 63 The Abandonment of Early Christian Nemea 65 Greece and Byzantium in the Middle to Late Byzantine Period 68 Prosperity under the Komnenoi 68 Factors Precipitating Fragmentation 69 The Fourth Crusade and the Conquest of the Peloponnese 71 Frankish Greece 72 viii Social and Religious Change in the Middle to Late Byzantine Period Religion Gender and the Family Medicine and Public Health The Middle to Late Byzantine Community at Nemea Summary 74 74 76 77 79 83 CHAPTER 4: MORTUARY PRACTICE IN BYZANTIUM Early Christian and Byzantine Funerary Ritual Burial in Byzantium Grave Construction, Arrangement of the Body, and Grave Contents Collective Burial and Secondary Burial The Spatial Organization of Byzantine Cemeteries Burial Location and Religious Beliefs Burial Location and the Social Hierarchy Summary 84 85 88 89 90 91 91 92 95 CHAPTER 5: RESEARCH QUESTIONS, HYPOTHESES, AND EXPECTATIONS Investigating the Biological Impact of Sociopolitical Change Research Hypotheses and Expectations Hypothesis One Hypothesis Two Hypothesis Three Summary 96 96 97 97 99 101 104 CHAPTER 6: MATERIALS AND METHODS Research Materials History of the Excavation of Burials at Nemea The Early Christian Burials The Middle to Late Byzantine Burials The Human Skeletal Samples from Nemea Research Methods Skeletal Analysis Demographic Data Collection Measuring Physiological Stress at Nemea Statistical Analysis: Indicators of Physiological Stress Mortuary Analysis Assessing Spatial and Temporal Variation in Burial Patterns at Nemea Examining Differences in Physiological Stress across Space Statistical Analysis: Spatial Analysis Comparative Skeletal Samples Summary 106 106 106 107 110 111 113 113 114 118 124 125 125 126 127 128 131 CHAPTER 7: RESULTS OF THE SKELETAL ANALYSIS The Early Christian Skeletal Sample Demographic Patterns: Adults 133 134 135 ix Demographic Patterns: Subadults Summary of Early Christian Demography Physiological Stress Indicators: Intraphase Results The Middle to Late Byzantine Skeletal Sample Demographic Patterns: Adults Demographic Patterns: Subadults Summary of Middle to Late Byzantine Demography Physiological Stress Indicators: Intraphase Results Summary of Significant Results Pertinent to Hypothesis Two Interphase Comparison Demography Indicators of Physiological Stress Summary of Significant Results Pertinent to Hypothesis One A Regional Perspective on Health in Byzantine Greece: Nemea, Akraiphnio, Thebes, and Spata Summary 137 138 139 152 153 154 155 156 169 171 172 174 187 CHAPTER 8: RESULTS OF THE MORTUARY ANALYSIS Variability in Mortuary Treatment at Nemea: Graves and Their Contents Sources of Data Location of the Graves Mortuary Variables Grave Type Bodies Present/Burial Type Minimum Number of Individuals (MNI) Grave Orientation Head Treatment Grave Goods Spatial Analysis of Demography and Physiological Stress Indicators Demography and Burial Location Physiological Stress and Burial Location Summary of Significant Results Pertinent to Hypothesis Three Summary 200 200 200 201 204 204 214 222 231 235 243 257 258 267 299 301 CHAPTER 9: DISCUSSION Reconstructing Health and Living Conditions at Byzantine Nemea Sociopolitical Change and Physiological Stress at Nemea Hypothesis One: Discussion and Implications Gender and Physiological Stress at Nemea Hypothesis Two: Discussion and Implications A Regional Perspective on Physiological Stress in the Byzantine Period: Nemea and Central Greece Reconstructing Mortuary Practice at Byzantine Nemea Death and Burial at Early Christian Nemea Death and Burial at Middle to Late Byzantine Nemea Burial Location and Physiological Stress at Nemea 302 303 322 322 328 328 x 189 197 331 336 336 343 346 Hypothesis Three: Discussion and Implications Summary 346 350 CHAPTER 10: CONCLUSIONS Contributions of this Dissertation Sociopolitical Change and Skeletal Stress Women and Men in the Byzantine Countryside The Organization of Byzantine Burials Limitations of this Study Future Research Directions 353 353 354 356 357 359 360 APPENDICES Appendix A: Data Coding Scheme Appendix B: Demarking Points for the Nemea Skeletal Samples Appendix C: Permission to Reprint Figures 364 367 369 BIBLIOGRAPHY 375 xi LIST OF TABLES Table 6.1 Minimum Number of Individuals in Each Period 112 Table 6.2 Number of Adults and Subadults in Each Period 113 Table 6.3 Number of Early Christian Individuals by Burial Location 113 Table 6.4 Skeletal Elements Chosen to Represent Each Joint Examined for Osteoarthritis in Commingled Burials 123 Table 6.5 Demographic Profiles of Comparative Skeletal Samples (Adapted from Tritsaroli [2006: Table 1]) 129 Table 7.1 Number and Proportion of Early Christian Individuals in Each Adult Age Category 136 Table 7.2 Number of Early Christian Femora and Demarking Point for Midshaft Circumference 136 Table 7.3 Distribution of Sex among Adults in the Early Christian Sample 137 Table 7.4 Number and Proportion of Early Christian Individuals in Each Subadult Age Category 137 Table 7.5 Age and Sex Demographics of the Early Christian Skeletal Sample 139 Table 7.6 LEH Prevalence by Tooth in the Early Christian Period 140 Table 7.7 LEH Prevalence by Tooth in Early Christian Adults and Subadults 141 Table 7.8 LEH Prevalence by Tooth in Early Christian Males and Females 142 Table 7.9 Prevalence Rates of Porotic Hyperostosis in Early Christian Adults and Subadults 143 Table 7.10 Prevalence Rates of Porotic Hyperostosis in Early Christian Males and Females 143 Table 7.11 Prevalence Rates of Cribra Orbitalia in Early Christian Adults and Subadults 144 Table 7.12 Prevalence Rates of Cribra Orbitalia in Early Christian Males and Females 144 xii Table 7.13 Prevalence Rates of Periosteal Reactions in Early Christian Adults and Subadults 145 Table 7.14 Expression of Periosteal Reactions in Early Christian Adults and Subadults 146 Table 7.15 Activity of Periosteal Reactions in Early Christian Adults and Subadults 146 Table 7.16 Prevalence Rates of Periosteal Reactions in Early Christian Males and Females 146 Table 7.17 Expression of Periosteal Reactions in Early Christian Males and Females 147 Table 7.18 Activity of Periosteal Reactions in Early Christian Males and Females 147 Table 7.19 Osteoarthritis Prevalence by Joint in the Early Christian Period 148 Table 7.20 Osteoarthritis Prevalence by Joint in Early Christian Males and Females 148 Table 7.21 Fracture Prevalence by Cranial Bone in the Early Christian Period (Adults Only) 150 Table 7.22 Type and Healing Status of Early Christian Cranial Fractures 150 Table 7.23 Fracture Prevalence by Postcranial Bone in the Early Christian Period (Adults Only) 151 Table 7.24 Type, Location, and Healing Status of Early Christian Postcranial Fractures 151 Table 7.25 Fracture Prevalence by Bone in Early Christian Males and Females 152 Table 7.26 Number and Proportion of Middle to Late Byzantine Individuals in Each Adult Age Category 153 Table 7.27 Distribution of Sex among Adults in the Middle to Late Byzantine Sample 154 Table 7.28 Number and Proportion of Middle to Late Byzantine Individuals in Each Subadult Age Category 155 Table 7.29 Age and Sex Demographics of the Middle to Late Byzantine Skeletal Sample 155 Table 7.30 Sex Distribution among Non-commingled Middle to Late Byzantine Burials 156 Table 7.31 LEH Prevalence by Tooth in the Middle to Late Byzantine Period 157 xiii Table 7.32 LEH Prevalence by Tooth in Middle to Late Byzantine Adults and Subadults 158 Table 7.33 LEH Prevalence by Tooth in Middle to Late Byzantine Males and Females 159 Table 7.34 Prevalence Rates of Porotic Hyperostosis in Middle to Late Byzantine Adults and Subadults 160 Table 7.35 Prevalence Rates of Porotic Hyperostosis in Middle to Late Byzantine Males and Females 160 Table 7.36 Prevalence Rates of Cribra Orbitalia in Middle to Late Byzantine Adults and Subadults 161 Table 7.37 Prevalence Rates of Cribra Orbitalia in Middle to Late Byzantine Males and Females 161 Table 7.38 Prevalence Rates of Periosteal Reactions in Middle to Late Byzantine Adults and Subadults 162 Table 7.39 Expression of Periosteal Reactions in Middle to Late Byzantine Adults and Subadults 162 Table 7.40 Activity of Periosteal Reactions in Middle to Late Byzantine Adults and Subadults 163 Table 7.41 Prevalence Rates of Periosteal Reactions in Middle to Late Byzantine Males and Females 163 Table 7.42 Expression of Periosteal Reactions in Middle to Late Byzantine Males and Females 163 Table 7.43 Activity of Periosteal Reactions in Middle to Late Byzantine Males and Females 164 Table 7.44 Osteoarthritis Prevalence by Joint in the Middle to Late Byzantine Period 165 Table 7.45 Osteoarthritis Prevalence by Joint in Middle to Late Byzantine Males and Females 166 Table 7.46 Fracture Prevalence by Cranial Bone in the Middle to Late Byzantine Period (Adults Only) 167 Table 7.47 Type and Healing Status of Middle to Late Byzantine Cranial Fractures 167 xiv Table 7.48 Fracture Prevalence by Postcranial Bone in the Middle to Late Byzantine Period (Adults Only) 168 Table 7.49 Type, Location, and Healing Status of Middle to Late Byzantine Postcranial Fractures 168 Table 7.50 Fracture Prevalence by Bone in Middle to Late Byzantine Males and Females 169 Table 7.51 Summary of Statistically Significant Results Pertinent to Hypothesis Two 171 Table 7.52 Age and Sex Demographics of the Early Christian (EC) and Middle to Late Byzantine (B) Samples 172 Table 7.53 Frequency and Prevalence Rates of Physiological Stress Indicators at Nemea 175 Table 7.54 Frequency and Prevalence Rates of Physiological Stress Indicators in Adults 177 Table 7.55 Frequency and Prevalence Rates of Physiological Stress Indicators in Subadults 178 Table 7.56 Frequency and Prevalence Rates of Physiological Stress Indicators in Males 180 Table 7.57 Frequency and Prevalence Rates of Physiological Stress Indicators in Females 181 Table 7.58 Frequency and Prevalence Rates of Osteoarthritis in Adults 183 Table 7.59 Frequency and Prevalence Rates of Osteoarthritis in Males 184 Table 7.60 Frequency and Prevalence Rates of Osteoarthritis in Females 185 Table 7.61 Frequency and Prevalence Rates of Fractures by Bone in Adults 186 Table 7.62 Frequency and Prevalence Rates of Fractures by Bone in Males 187 Table 7.63 Frequency and Prevalence Rates of Fractures by Bone in Females 187 Table 7.64 Summary of Statistically Significant Results Pertinent to Hypothesis One 189 Table 7.65 Comparison of Prevalence Rates of Physiological Stress Indicators in Adults at Early Christian Nemea and Akraiphnio 191 Table 7.66 Comparison of Prevalence Rates of Physiological Stress Indicators in Subadults at Early Christian Nemea and Akraiphnio 192 xv Table 7.67 Prevalence Rates of Fractures at Early Christian Nemea and Akraiphnio 192 Table 7.68 Comparison of Prevalence Rates of Physiological Stress Indicators in Adults at Middle to Late Byzantine Nemea and Thebes 194 Table 7.69 Comparison of Prevalence Rates of Physiological Stress Indicators in Subadults at Middle to Late Byzantine Nemea and Thebes 194 Table 7.70 Prevalence Rates of Fractures at Middle to Late Byzantine Nemea and Thebes 195 Table 7.71 Comparison of Prevalence Rates of Physiological Stress Indicators in Adults at Middle to Late Byzantine Nemea and Spata 196 Table 7.72 Comparison of Prevalence Rates of Physiological Stress Indicators in Subadults at Middle to Late Byzantine Nemea and Spata 196 Table 7.73 Prevalence Rates of Fractures at Middle to Late Byzantine Nemea and Spata 197 Table 8.1 Frequency of Grave Types in Each Period 205 Table 8.2 Spatial Distribution of Early Christian Grave Types 206 Table 8.3 Frequency of Early Christian Adults and Subadults in Each Grave Type 210 Table 8.4 Early Christian Adults and Subadults in Roof Tile and Field Stone Covered Graves 211 Table 8.5 Frequency of Middle to Late Byzantine Adults and Subadults in Each Grave Type 211 Table 8.6 Frequency of Adults and Subadults in Each Grave Type by Period 212 Table 8.7 Frequency of Early Christian Males and Females in Each Grave Type 213 Table 8.8 Frequency of Middle to Late Byzantine Males and Females in Each Grave Type 213 Table 8.9 Frequency of Males and Females in Each Grave Type by Period 214 Table 8.10 Frequency of Burial Types in Each Period 215 Table 8.11 Spatial Distribution of Early Christian Burial Types 218 Table 8.12 Spatial Distribution of Single Inhumations and Empty Graves in the Early Christian Period 218 xvi Table 8.13 Frequency of Early Christian Adults and Subadults in Each Burial Type 220 Table 8.14 Frequency of Middle to Late Byzantine Adults and Subadults in Each Burial Type 220 Table 8.15 Frequency of Adults and Subadults in Each Burial Type by Period 221 Table 8.16 Frequency of Early Christian Males and Females in Each Grave Type 221 Table 8.17 Frequency of Middle to Late Byzantine Males and Females in Each Grave Type 222 Table 8.18 Frequency of Males and Females in Each Burial Type by Period 222 Table 8.19 Frequency of MNI for Early Christian and Middle to Late Byzantine Burials 224 Table 8.20 Spatial Distribution of MNI by Burial Area in the Early Christian Period 225 Table 8.21 Frequency of Early Christian Adults and Subadults by Grave MNI 229 Table 8.22 Frequency of Middle to Late Byzantine Adults and Subadults by Grave MNI 229 Table 8.23 Adults and Subadults in Single versus Multiple Burials in the Middle to Late Byzantine Period 230 Table 8.24 Frequency of Early Christian Males and Females by Grave MNI 230 Table 8.25 Frequency of Middle to Late Byzantine Males and Females by Grave MNI 231 Table 8.26 Males and Females in Single versus Multiple Burials in the Middle to Late Byzantine Period 231 Table 8.27 Frequency of Head Treatment Types in Middle to Late Byzantine Burials 241 Table 8.28 Presence or Absence of Graves Goods in Each Period 244 Table 8.29 Frequency and Distribution of Grave Goods in Early Christian Burials 245 Table 8.30 Frequency and Distribution of Grave Good Types in Early Christian Burials 249 Table 8.31 Frequency of Grave Good Types in Middle to Late Byzantine Burials 252 Table 8.32 Presence of Grave Goods in Early Christian Adult and Subadult Burials 254 xvii Table 8.33 Presence of Grave Goods in Middle to Late Byzantine Adult and Subadult Burials 254 Table 8.34 Presence of Grave Goods in All Middle to Late Byzantine Burials by Age 254 Table 8.35 Frequency of Early Christian Adults and Subadults by Grave Good Type 255 Table 8.36 Presence of Grave Goods in Early Christian Male and Female Burials 256 Table 8.37 Presence of Grave Goods in Middle to Late Byzantine Male and Female Burials 256 Table 8.38 Frequency of Early Christian Males and Females by Grave Good Type 257 Table 8.39 Frequency and Distribution of Adults and Subadults in Early Christian Burials 258 Table 8.40 Frequency of Adults and Subadults in Temple versus Bath Burials 258 Table 8.41 Frequency of Adults and Subadults in Temple versus Basilica Burials 259 Table 8.42 Frequency of Adults and Subadults Inside versus Outside of the Basilica 262 Table 8.43 Spatial Distribution of Sex in Early Christian Burials 263 Table 8.44 Frequency of Males and Females Inside versus Outside of the Basilica 267 Table 8.45 Spatial Distribution of Linear Enamel Hypoplasias (LEH) on Left Mandibular Canines (#22) among Early Christian Burials 272 Table 8.46 Comparison of Linear Enamel Hypoplasias (LEH) on Left Mandibular Canines (#22) between Bath and Temple/Basilica Burials 272 Table 8.47 Spatial Distribution of Linear Enamel Hypoplasias (LEH) on Right Mandibular Canines (#27) among Early Christian Burials 273 Table 8.48 Comparison of Linear Enamel Hypoplasias (LEH) on Right Mandibular Canines (#27) between Bath and Temple Burials 273 Table 8.49 Comparison of Linear Enamel Hypoplasias (LEH) on Right Mandibular Canines (#27) between Bath and Basilica Burials 273 Table 8.50 Comparison of Linear Enamel Hypoplasias (LEH) on Left Mandibular Canines (#22) Inside versus Outside of the Basilica 275 xviii Table 8.51 Comparison of Linear Enamel Hypoplasias (LEH) on Right Mandibular Canines (#27) Inside versus Outside of the Basilica 275 Table 8.52 Spatial Distribution of Porotic Hyperostosis (PH) among Early Christian Burials 281 Table 8.53 Comparison of Porotic Hyperostosis (PH) between Temple and Bath Burials 281 Table 8.54 Comparison of Porotic Hyperostosis (PH) between Temple and Basilica Burials 281 Table 8.55 Presence of Porotic Hyperostosis (PH) Inside versus Outside of the Basilica 282 Table 8.56 Spatial Distribution of Cribra Orbitalia (CO) among Early Christian Burials 286 Table 8.57 Comparison of Cribra Orbitalia (CO) between Basilica and Temple Burials 286 Table 8.58 Comparison of Cribra Orbitalia (CO) between Basilica and Bath Burials 286 Table 8.59 Presence of Cribra Orbitalia (CO) Inside versus Outside of the Basilica 288 Table 8.60 Spatial Distribution of Periosteal Reactions (PR) on Left Tibiae among Early Christian Burials 289 Table 8.61 Spatial Distribution of Periosteal Reactions (PR) on Right Tibiae among Early Christian Burials 289 Table 8.62 Presence of Periosteal Reactions (PR) on Left Tibiae Inside versus Outside of the Basilica 290 Table 8.63 Presence of Periosteal Reactions (PR) on Right Tibiae Inside versus Outside of the Basilica 290 Table 8.64 Spatial Distribution of Osteoarthritis (OA) among Early Christian Burials 291 Table 8.65 Results of High/low Cluster Analyses of Osteoarthritis (OA) Presence by Joint among Middle to Late Byzantine Burials 292 Table 8.66 Presence of Osteoarthritis (OA) Inside versus Outside of the Basilica 293 Table 8.67 Spatial Distribution of Early Christian Graves Containing Individuals with Fractures 297 Table 8.68 Comparison of Fractures between Temple and Basilica Burials 297 xix Table 8.69 Comparison of Fractures between Bath and Basilica Burials 297 Table 8.70 Comparison of Fractures between Temple and Bath Burials 297 Table 8.71 Presence of Fractures in Graves Inside versus Outside of the Basilica 299 Table 8.72 Summary of Statistically Significant Results Pertinent to Hypothesis Three 300 Table 9.1 Summary of Statistically Significant Results: Early Christian Nemea versus Akraiphnio 333 Table 9.2 Summary of Statistically Significant Results: Middle to Late Byzantine Nemea versus Thebes 335 Table 9.3 Summary of Statistically Significant Results: Middle to Late Byzantine Nemea versus Spata 336 Table A.1 Data Codes 364 Table B.1 Number of Early Christian Femora and Demarking Point for Midshaft Circumference 367 Table B.2 Number of Middle to Late Byzantine Femora and Demarking Point for Maximum Head Diameter 367 Table B.3 Number of Middle to Late Byzantine Femora and Demarking Point for Midshaft Circumference 367 Table B.4 Number of Middle to Late Byzantine Tibiae and Demarking Point for Circumference at the Nutrient Foramen 367 Table B.5 Number of Middle to Late Byzantine Humeri and Demarking Point for Epicondylar Breadth 368 Table B.6 Number of Middle to Late Byzantine Humeri and Demarking Point for Vertical Head Diameter 368 Table B.7 Number of Middle to Late Byzantine Humeri and Demarking Point for Minimum Circumference 368 Table B.8 Number of Middle to Late Byzantine Radii and Demarking Point for Tuberosity Circumference 368 xx LIST OF FIGURES Figure 1.1 Location of Nemea in the Northeastern Peloponnese. Adapted from Miller (1990: Figure 1) 13 Figure 1.2 Site Plan of Nemea Showing the Sanctuary of Zeus and the Tsoungiza Hill. Courtesy Kim Shelton 15 Figure 1.3 Detail of the Sanctuary of Nemean Zeus. Adapted from Miller (1988: Figure 1) 17 Figure 3.1 Map Showing Medieval Sites from the Nemea Valley Survey (Athanassopoulos [2010: Figure 1, Rosemary Robertson]) 82 Figure 6.1 Spatial Distribution of the Nemea Burials by Period 109 Figure 6.2 Map Showing the Location of Comparative Sites (Adapted from Tritsaroli [2006: Appendix 1, Figure 1]) 129 Figure 7.1 Number of Individuals in Each Age Category by Period 173 Figure 7.2 Number of Individuals in Each Sex Category by Period 174 Figure 7.3 Prevalence Rates of Physiological Stress Indicators at Nemea 176 Figure 7.4 Prevalence Rates of Physiological Stress Indicators in Adults 177 Figure 7.5 Prevalence Rates of Physiological Stress Indicators in Subadults 179 Figure 7.6 Prevalence Rates of Physiological Stress Indicators in Males 180 Figure 7.7 Prevalence Rates of Physiological Stress Indicators in Females 182 Figure 8.1 Spatial Distribution of the Nemea Burials 202 Figure 8.2 Spatial Distribution of the Burials by Period 203 Figure 8.3 Spatial Distribution of Grave Types in the Early Christian Period 207 Figure 8.4 Spatial Distribution of Grave Types in the Middle to Late Byzantine Period 209 Figure 8.5 Spatial Distribution of Burial Types in the Early Christian Period 217 xxi Figure 8.6 Spatial Distribution of Burial Types in the Middle to Late Byzantine Period 219 Figure 8.7 Spatial Distribution of MNI among Early Christian Burials 226 Figure 8.8 Spatial Distribution of MNI among Middle to Late Byzantine Burials 228 Figure 8.9 Grave Orientation of Early Christian Burials 233 Figure 8.10 Grave Orientation of Middle to Late Byzantine Burials 234 Figure 8.11 Presence or Absence of Head Treatment in Early Christian Burials 237 Figure 8.12 Spatial Distribution of Head Treatment Types in Early Christian Burials 239 Figure 8.13 Presence or Absence of Head Treatment in Middle to Late Byzantine Burials 240 Figure 8.14 Spatial Distribution of Head Treatment Types in Middle to Late Byzantine Burials 242 Figure 8.15 Spatial Distribution of the Presence of Grave Goods in Early Christian Burials 246 Figure 8.16 Spatial Distribution of the Presence of Grave Goods in Middle to Late Byzantine Burials 248 Figure 8.17 Spatial Distribution of Grave Good Types in Early Christian Burials 251 Figure 8.18 Spatial Distribution of Grave Good Types in Middle to Late Byzantine Burials 253 Figure 8.19 Spatial Distribution of Age at Death among Temple Burials 259 Figure 8.20 Spatial Distribution of Age at Death among Bath Burials 260 Figure 8.21 Spatial Distribution of Age at Death among Middle to Late Byzantine Burials 261 Figure 8.22 Detail of Age Distribution among Middle to Late Byzantine Burials in the Nave 262 Figure 8.23 Spatial Distribution of Sex among Temple Burials 264 Figure 8.24 Spatial Distribution of Sex among Bath Burials 265 Figure 8.25 Spatial Distribution of Sex among Middle to Late Byzantine Burials 266 xxii Figure 8.26 Detail of Sex Distribution among Middle to Late Byzantine Burials in the Nave 267 Figure 8.27 Spatial Distribution of Linear Enamel Hypoplasias on Left Mandibular Canines among Temple Burials 269 Figure 8.28 Spatial Distribution of Linear Enamel Hypoplasias on Left Mandibular Canines among Bath Burials 270 Figure 8.29 Spatial Distribution of Linear Enamel Hypoplasias on Left Mandibular Canines among Early Christian Basilica Burials 271 Figure 8.30 Spatial Distribution of Linear Enamel Hypoplasias on Left Mandibular Canines among Middle to Late Byzantine Burials 274 Figure 8.31 Detail of the Spatial Distribution of Linear Enamel Hypoplasias on Right Mandibular Canines among Middle to Late Byzantine Burials in the Nave 276 Figure 8.32 Spatial Distribution of Porotic Hyperostosis among Temple Burials 278 Figure 8.33 Spatial Distribution of Porotic Hyperostosis among Bath Burials 279 Figure 8.34 Spatial Distribution of Porotic Hyperostosis among Early Christian Basilica Burials 280 Figure 8.35 Spatial Distribution of Porotic Hyperostosis among Middle to Late Byzantine Burials 282 Figure 8.36 Spatial Distribution of Cribra Orbitalia among Temple Burials 283 Figure 8.37 Spatial Distribution of Cribra Orbitalia among Bath Burials 284 Figure 8.38 Spatial Distribution of Cribra Orbitalia among Early Christian Basilica Burials 285 Figure 8.39 Spatial Distribution of Cribra Orbitalia among Middle to Late Byzantine Burials 288 Figure 8.40 Spatial Distribution of Early Christian Temple Graves Containing Individuals with Fractures 294 Figure 8.41 Spatial Distribution of Early Christian Bath Graves Containing Individuals with Fractures 295 xxiii Figure 8.42 Spatial Distribution of Early Christian Basilica Graves Containing Individuals with Fractures 296 Figure 8.43 Spatial Distribution of Middle to Late Byzantine Graves Containing Individuals with Fractures 298 Figure 9.1 Mortality Curves of the Nemea Skeletal Samples 305 Figure 9.2 Mortality Profile of Early Christian Adults 309 Figure 9.3 Mortality Profile of Middle to Late Byzantine Adults 309 Figure 9.4 Prevalence Rates of Physiological Stress Indicators in Early Christian Adults and Subadults 314 Figure 9.5 Prevalence Rates of Physiological Stress Indicators in Middle to Late Byzantine Adults and Subadults 315 Figure 9.6 Accentuated Striations on the Medial Aspect of the Left Tibia Near Midshaft. SK 066 316 Figure 9.7 Endocranial View of Healed Depression Fracture on Left Parietal. SK 076 321 Figure 9.8 View of the Mountainside of Polyphengi. The Arrow Shows the Location of the Late Byzantine Settlement 325 Figure 9.9 Anterior View of Right (a.) and Left (b.) Distal Humeri Showing Displacement and Remodeling of the Trochlea and Capitulum 327 Figure 9.10 Spatial Distribution of Early Christian Grave Types 341 xxiv CHAPTER 1: INTRODUCTION Sociopolitical Change and the Byzantine Greek Countryside A desire on the part of scholars to investigate the daily lives of all members of Byzantine society has come slowly and even reluctantly. From the perspective of archaeology, the study of the post-Roman eastern Mediterranean in general developed more slowly than that of the Classical period as a result of intellectual traditions in the field of classical archaeology. In Greece, the Byzantine period has in the past been viewed as an unfortunate chapter in the nation’s history, separating modern Greece from its rich classical heritage (Athanassopoulos 2008; Bowden 2009). Even when archaeologists began to pay greater attention to the Byzantine th components of classical sites during the 20 century, their focus was firmly on the study of art and architecture rather than on the study of the daily lives of ordinary people (Athanassopoulos 2008). Only through recent efforts involving regional archaeological survey has a clearer picture of the Byzantine Greek countryside begun to take shape (Athanassopoulos 1993, 2008, 2010; Kardulias 1997). Achieving an understanding of daily life for average people in the Byzantine Empire through written sources has also proven difficult. Byzantine historians, both ancient and modern, tend to place considerable emphasis on emperors, imperial politics, wars, and the effects of such events on Constantinople or other major urban centers. The problem for modern historians is that the voices that have survived in the historical record are often those of the privileged, who wrote in many cases to promote particular interests. The details that emerge from those sources are, unsurprisingly, largely concerned with elite life. Even broad chronicles of the Byzantine period are typically focused on large-scale political and military events (Athanassopoulos 1993). This myopic view of Byzantine society is somewhat paradoxical given that peasants formed the 1 backbone of Byzantium. It was, in fact, individuals residing in the countryside who made up the greatest proportion of the Byzantine Empire’s population, workforce, and tax base (Kazhdan 1997). In Greece human skeletal remains and, to a much greater degree, their mortuary contexts have been underutilized as sources of evidence capable of shedding light on the health, living conditions, and social organization of this neglected segment of Byzantine society. This dissertation aims to contribute to our current understanding of life and death in the Early Christian and Byzantine Greek countryside through a bioarchaeological analysis of the burials and human skeletal samples from the site of ancient Nemea. The individuals buried within the Sanctuary of Zeus at Nemea represent members of two distinct farming communities that th th occupied the site during the Early Christian (5 -6 centuries AD) and Middle to Late Byzantine th th (12 -13 centuries AD) periods. Each of those communities existed during periods of significant sociopolitical change in the Byzantine Empire. This dissertation, however, explores the possibility that the events of the Middle to Late Byzantine period had significant, negative consequences for the living conditions of the inhabitants of the Nemea Valley. th During the late 12 century AD, marginal areas of the empire such as the countryside of southern Greece experienced a deterioration of the provincial administration that exposed local populations to excessive taxation and a lack of services (Herrin 1975, 1985). Even more serious, however, were the results of the Fourth Crusade, which abandoned the goal of conquering the Holy Land and sacked Constantinople, the capital of the Byzantine Empire in 1204 AD. This disaster ushered in a period of political fragmentation and social transformation under western powers. The population of the Peloponnese experienced some of the most direct consequences of 2 the Fourth Crusade when Frankish knights invaded and conquered territory from Corinth to th Monemvasia during the first half of the 13 century. In the years that followed this conquest, the Franks imposed a western-style feudal system on the local Greek population and, as a result, the position of local peasants at the bottom of the social hierarchy became both legal and hereditary (Jacoby 1973, 1989, 2008; Lock 1995). A number of questions related to the events of the Middle to Late Byzantine transition lend themselves to further investigation. Some of those questions are related to the possibility that living conditions were affected. For example, in what ways (if any) did the strained th relationships between the central and provincial administrations during the late 12 century impact individuals living in the Greek countryside? Did the imposition of a western-style feudal system following Frankish conquest worsen living conditions and increase workload for the local Greek population? Did dietary practices change with western influence? Other questions are more focused on what the material evidence, such as information obtained from burials, can reveal about social structure and sociocultural change. For example, what do differences in burial practices suggest about diachronic changes in the nature of Byzantine Christianity? Can western influence be detected among those burial practices? What is the significance of the spatial organization of Byzantine burials? What can burial practices tell us about diachronic changes in the structure of the Byzantine family? Were gender roles and gender relationships affected? While it is beyond the scope of this dissertation to answer all of these questions in a complete manner, it is possible to address many of them in a general way using a multidimensional bioarchaeological approach. 3 The Contribution of Bioarchaeology Bioarchaeologists are uniquely positioned to use skeletal analysis in conjunction with archaeological and historical data to test hypotheses about ancient communities. The human skeleton is dynamic in that it is capable of responding to changing biomechanical demands and physiological stressors. If episodes of stress are not mitigated by factors such as an individual’s biological resistance and/or cultural mechanisms, they may trigger a response observable in skeletal and dental tissue (Goodman et al. 1984, 1988). In this way, the prevalence rates of skeletal indicators of physiological disruption such as enamel defects, porotic cranial lesions, and reactive bone formation in ancient skeletal samples may be extremely informative about health and living conditions in the past. Analyses of skeletal indicators of health and disease become even more powerful when combined with a thorough understanding of the archaeological and historical context of human remains. This was demonstrated by Jane Buikstra in the 1970s when she outlined a “bioarchaeological” research program that focused on the integration of archaeological and osteological datasets in the examination of biocultural change in the Illinois River valley during the Middle to Late Woodland transition (1977:69). Buikstra recognized the advantages of an interdisciplinary study of cemeteries and their contexts, such as controlling (at least to the extent possible) for sampling bias resulting from the mortuary program and identifying examples of differential susceptibility to disease. Goldstein (1976, 1981) has argued convincingly that spatial organization is one of the most important aspects of the archaeological context to consider in the analysis of mortuary sites. The arrangement and ultimate form of cemeteries are the result of conscious choices made by groups of people and are therefore unlikely to be random. For example, spatial relationships 4 between burials or groups of burials may reflect social relationships during life. With this in mind, the investigation of the spatial component of cemeteries can potentially provide information on status distinctions, the organization of the family, and other characteristics of the social structure of a society (Goldstein 1981:57). The incorporation of skeletal analysis offers an independent line of evidence that can be used to test hypotheses about status differentiation or family groupings made based on the spatial data. Bioarchaeology can make similar contributions when applied to mortuary sites dating to historical periods. In this case, data on skeletal health can be carefully integrated with archaeological and textual data to fill in gaps in the historical record and test commonly held assumptions about historical processes and events (Perry 2002, 2007). In this way, combining skeletal data with archaeological and historical context in an interdisciplinary bioarchaeological approach results in a richer analysis—one that is capable of addressing the types of broader questions concerning the living conditions of Greek populations under Frankish rule. Bioarchaeological Research in the Eastern Mediterranean Until relatively recently, it has been uncommon for studies of human skeletal samples recovered from archaeological sites in the eastern Mediterranean to employ an integrative bioarchaeological approach. The disconnect between the study of human skeletons and the analysis of their contexts in this region can be traced to the excavations of classical sites during th th the 19 and early 20 century, which were generally unconcerned with human remains. If skeletons were analyzed at all, they were typically subjected to craniometric assessments for the purposes of establishing racial typologies or tracing population origins (MacKinnon 2007; 5 1 Roberts et al. 2005). The emphasis on “race” was characteristic of physical anthropology at the time and the incentive to demonstrate biological relationships between ancient and modern populations in Europe was driven by the culture historical approach and colonialist ideas (Armelagos 2003; Armelagos and Van Gerven 2003; MacKinnon 2007). The sole emphasis on osteological measurements and extended descriptions also meant that a large amount of data was collected without substantial reference to the archaeological and historical contexts of the remains (Buikstra and Lagia 2009). This not only minimized communication between physical anthropologists and other researchers, but also created a model in which skeletal analysis was viewed as a “post-excavation” specialization (MacKinnon 2007:496) and osteological findings were routinely relegated to site report appendices (Larsen 2006; Roberts et al. 2005). th Around the mid-20 century, studies of skeletal samples from the ancient Mediterranean began to address health-related questions more frequently. The work of J. Lawrence Angel figured most prominently in this transition. While some of Angel’s early research did focus on “race,” his major contributions were in the areas of health, disease, and adaptation. For example, Angel’s work in Greece and Turkey was focused on the relationships between porotic hyperostosis, inherited anemias, and falciparum malaria. He observed that the geographical distributions of malaria and genes for abnormal hemoglobin overlapped. Furthermore, he demonstrated ancient populations living in marshy areas within malarial regions and during periods when agricultural techniques were poor tended to exhibit greater frequencies of porotic hyperostosis. Based on this evidence, Angel argued that porotic hyperostosis in the eastern 1 Unfortunately it was also common in this period to discard both human and animal bone discovered during excavation. Nemea is a case in point as the 1920s excavations recorded a number of burials, mostly in and around the Early Christian basilica, containing human remains that were not saved. Burials containing poorly preserved skeletons were likewise discarded in the 1960s excavations. 6 Mediterranean was probably a skeletal manifestation of one of the genetic anemias, which provided heterozygous individuals with a degree of protection against malaria (Angel 1966, 1967, 1978). His conclusions remain relevant to studies of health in the region as the precise etiology of cranial lesions continues to be an area of active research and debate. Angel’s emphasis on health in the eastern Mediterranean has expanded in a number of directions in recent years. In Greece, numerous studies that employ a paleopathological methodology emphasizing multiple indicators of stress and disease have been carried out on skeletal samples from a variety of time periods (e.g., Agelarakis 1997; Barnes 2003; Bourbou 2003; Bourbou and Tsilipakou 2009; Fox-Leonard 1997; Fox 2005; Papageorgopoulou and Xirotiris 2009; Papathanasiou et al. 2000, 2009). Many of these publications have taken into account the historical context of the remains and the possible effects of historical events on the overall health of the represented communities. For example, Fox’s (2005) comparative assessment of Hellenistic and Roman samples from Paphos and Corinth found that the latter was characterized by greater childhood mortality, an elevated prevalence rate of enamel hypoplasias, and shorter stature. She suggests that those differences may have been related to more unstable living conditions at Corinth, which followed its sack by the Romans in 146 BC. Approaches that combine more traditional bioarchaeological methods with new analytical techniques represent another research trend in the bioarchaeological analysis of Greek skeletal material. For example, trace element and stable isotope studies have provided new insights into ancient dietary practices (Bourbou and Richards 2007; Bourbou et al. 2011; Garvie-Lok 2001; Papageorgopoulou and Xirotiris 2009; Papathanasiou et al. 2000, 2009; Petroutsa et al. 2009) as well as population migration (Garvie-Lok 2009). Garvie-Lok’s (2001) work is particularly relevant to the present research as she included twenty-nine Early Christian and twenty-five 7 Middle to Late Byzantine individuals from Nemea in her isotopic study of diet in Medieval Greece. Her results suggest that dietary composition in both periods was quite similar and focused on grains and terrestrial animal proteins. Millet may also have been consumed at Nemea—an interesting finding given that Byzantine dieticians viewed it unfavorably (Kazhdan 1997). Garvie-Lok (2009) has also provided confirmation of immigration to the northeastern Peloponnese during the Frankish period through an analysis of stable oxygen isotope ratios from 17 individuals buried at Corinth. Perhaps in response to the recent call for researchers to use osteological data as an independent line of evidence with which to evaluate historical narratives (e.g., Perry 2002, 2007), studies of skeletal paleopathology in Greece are increasingly focused on examining the ability of ancient communities to adapt to cultural and environmental change. For example, the effects of natural disasters and invasions have been evaluated for proto-Byzantine samples from Crete and the southern Peloponnese (Bourbou 2003, 2010), as well as from northern Greece (Bourbou and Tsilipakou 2009). Bourbou (2003) found that those external factors had a greater th th impact on health at Eleutherna, Crete than at Messene in southern Greece during the 6 -7 centuries AD. Children at Eleutherna seem to have been particularly affected by a decline in nutrition and living conditions and display greater mortality along with elevated prevalence rates of anemia and scurvy. Bourbou and Tsilipakou’s (2009) assessment of a contemporaneous skeletal sample from Sourtara in northern Greece reveals low levels of infections and general good health and nutritional status. Together these studies suggest that the health consequences of disruptive events for Greek communities during this period depended largely upon local conditions. 8 The present study contributes to this body of research an investigation of the impact of sociopolitical changes suggested by traditional historical narratives of Late Antiquity and the Middle to Late Byzantine transition on small-scale agricultural communities in the northeastern Peloponnese. It is also an overarching goal of this dissertation to expand upon recent research models by using an approach that explicitly tests hypotheses about mortuary practice in Byzantium. In this way, I follow Buikstra and Lagia (2009) in their recommendation of a problem-oriented approach to the study of cemeteries and human remains. As discussed above, the analysis of human skeletons in conjunction with their archaeological context represents an important break with tradition—one that is now being pursued more frequently (see, for example, Morris [1992], Paine et al. [2007], Perry [2002, 2007], Schepartz et al. [2009], Triantaphyllou [2001], Tritsaroli [2006], and Tritsaroli and Valentin [2008]). Recognition of the interdependence of osteological and archaeological datasets is crucial for the continued development of our understanding of cemetery organization in the Byzantine eastern Mediterranean. The burials and samples of human skeletons from Nemea, which will be presented in detail in Chapter Six, are in many ways ideal for the examination of the impact of sociopolitical transformation in southern Greece during the Byzantine period. Although they represent two different communities from distinct periods of time, the similarities between those communities are marked. Based on the archaeological evidence, which is outlined in detail in Chapter Three, each community at Nemea was agricultural in character and neither was particularly wealthy. Because both groups of people lived and worked at the same site, it is unlikely that they experienced dramatic contrasts in environmental conditions or access to local resources. The 9 2 communities at Nemea also generally shared a Christian worldview. The fact that these basic features of the two communities held more or less constant through time makes any apparent differences, such as diachronic changes in burial practices, all the more revealing. Variability in the prevalence rates of paleopathological conditions and in the mortuary domain provides perhaps the best sources of data with which to evaluate the effects of cultural transformation at Nemea. Nemea and its Environs Ancient Nemea is located near the southern end of a narrow valley in the northeastern Peloponnese to the southwest of Corinth and north of Mycenae and Argos (Figure 1.1). At roughly 330 m above sea level, the site is also the point of origin for the Nemea River, which drains northward out of the valley toward the Corinthian Gulf (Miller 1990). The contours of the valley along the river’s course are well defined by the surrounding hills. To the southeast is Evangelistria Hill, on which a modern church and the remains of an Early Christian basilica are located. The ridge that forms the eastern boundary of the valley culminates at Mount Phoukas, while the western hills give rise to Mount Prophitis Ilias closer to the middle of the valley’s north-south length (Wright et al. 1990). Access to neighboring valleys and the maintenance of connections with settlements in adjacent areas have played a significant role in Nemea’s history. The Nemea Valley is located near to various passes linking ancient cities and facilitating travel in the northeastern Peloponnese (Wright 1982). A minor pass in the southeast corner of the valley near the sanctuary 2 Each community at Nemea utilized at least one Christian basilica, albeit in different ways. The Middle to Late Byzantine community buried its dead within and outside of the ruins of the basilica constructed in the former athletic sanctuary. There are very few examples of burials that exhibited non-Christian features or perhaps a mix of Christian and pagan themes. These exceptions are discussed in Chapters Eight and Nine. 10 of Zeus leads east into the Longopotamos Valley, in which lies ancient Kleonai. From there one gains access south to the Tretos pass connecting the Corinthia and the Argolid. Exiting the Nemea Valley to the west, one enters the Phliasian Plain, which contains the modern town of New Nemea and the ancient city-state of Phlious. Another important route south into the Argive Plain, the Kelossa Pass, is accessible from this point as well. The significance of the location of Nemea and its neighbors has recently been demonstrated by Marchand (2009), who points out that the Corinth-Argos road, which passed by Kleonai and utilized the Tretos pass, functioned not only as the major north-south link between those two cities, but also as an access point to other routes heading in various directions through the northeastern Peloponnese in antiquity. Nemea’s location is thus probably one of the major reasons for which periods of occupation of the site have consistently coincided with periods of higher population and settlement densities throughout the region (Wright et al. 1990). The unique geographical and geological conditions that characterize the Nemea Valley make it ideal for agriculture when properly drained. Alluvium from the encircling hills has accumulated on the valley floor since the Neolithic period and the excellent fertility of this fill is illustrated by the presence of numerous vineyards, olive groves, and fruit trees in addition to crops such as wheat and barley (Miller 1990; Wright 1982; Wright et al. 1990). Nemea’s height above sea level and exposure to mountain winds also provide it with a unique microclimate. Summers tend to be somewhat cooler than in other areas of Greece, while winters bring a significant amount of moisture and precipitation that tapers off by early summer (Miller 1990). Water is also available locally in the form of springs located along the margins of the valley and aquifers underneath the valley floor (Wright 1982). The flow of the Nemea River has always been crucial to drainage of the valley for agricultural purposes and the promotion of more 11 th habitable conditions in general. Early modern travelers visiting Nemea in the 19 century characterized the Nemea Valley, which was blocked at the time, as a sparsely occupied and unhealthy marshland (Athanassopoulos 1993). The challenge of maintaining adequate drainage is such that Wright et al. (1990) suggest adequate manpower has been a prerequisite to permanent settlement at Nemea. 12 Figure 1.1: Location of Nemea in the Northeastern Peloponnese. Adapted from Miller (1990: Figure 1). Perachora Sikyon Isthmia Titane Stimanga Corinth Stymphalos Petri Aidonia Kleonai Phlious Tenea Nemea Mycenae 13 Phases of Occupation and Activity The earliest clear evidence of habitation in the Nemea Valley comes from the Tsoungiza Hill located west of the Sanctuary of Zeus (Figure 1.2). A number of refuse pits containing, among other finds, substantial quantities of Early Neolithic pottery, obsidian and stone tools, and animal bone fragments have been excavated by Blegen (1927) and Miller (1975, 1976, 1980). Although the precise location of the Early Neolithic settlement has not been determined, the quantity of material recovered from various areas on the hill suggests that it must have been extensive (Wright et al. 1990). Regardless of the location of the Neolithic settlement, Tsoungiza was certainly occupied by the Early Bronze Age. Blegen (1927) identified house foundations along with associated pottery and storage vessels dating to the Early, Middle, and Late Helladic periods. A reexcavation of the area explored by Blegen and Harland (unpublished) in the 1920s revealed additional Early Helladic ceramics and architectural remains (Pullen 1986; Wright et al. 1990). The Early Helladic settlement appears to have been concentrated on the crest of the hill. While the Middle Bronze Age is not well represented, Miller’s (1975, 1976, 1980) excavations produced significant amounts of high quality Late Helladic as well as Middle Helladic pottery. Adding to this evidence, Wright (1982) identified a burned structure containing domestic pottery and a hearth dating to the end of the Middle Helladic or Early Late Helladic period. The Nemea Valley Archaeological Project survey has since also identified a number of sites including Tsoungiza as well as areas within the Sanctuary of Zeus that contained Late Bronze Age finds (Cherry et al. 2000; Wright et al. 1990). The distribution and quantity of material recovered suggests that the Mycenaean settlement on and associated with Tsoungiza 14 was the dominant site within a hierarchy of settlements in the survey area and must have been of some regional importance. Figure 1.2: Site Plan of Nemea Showing the Sanctuary of Zeus and the Tsoungiza Hill. Courtesy Kim Shelton. 15 Significant human presence in the Nemea Valley following the Bronze Age took on a very different character as permanent settlement at Nemea was replaced by the sporadic habitation of an athletic sanctuary. The Nemean Games were established in 573 BC, joining those held at Olympia, Delphi, and Isthmia in the pan-Hellenic cycle. While the schedule of the games operated on a four-year cycle, those held at Nemea took place every two years. It was only during perhaps the weeks leading up to the games and during the festival itself that the Nemea Valley would have seen great activity as athletes and visitors arrived at the site. At all other times, only a small group of caretakers and priests would have been present (Frey 1998; Miller 1990). rd Nemea functioned as an athletic sanctuary into the 3 century BC. However, the games were moved to Argos around 400 BC following a violent conflict that resulted in damage throughout the sanctuary and destruction of the Archaic Temple of Zeus (Miller 1977, 1979, th 1990, 2006). A period of new construction in the last third of the 4 century BC marked the return of the games to Nemea, although they continued to be held periodically at Argos depending on the political situation between the Argives and the Macedonians (Miller 1978, 1982). Much of the architecture that can be observed at the site today including the Temple of Zeus, the Bath, the Xenon, and the Stadium (Figure 1.3) dates to that time (Birge et al. 1992; Miller 1977). The games moved permanently to Argos by 271 BC (Miller 2006). 16 Figure 1.3: Detail of the Sanctuary of Nemean Zeus. Adapted from Miller (1988: Figure 1). It is with the next two phases of activity at Nemea that the present study is primarily th th concerned. In the 5 and then again in the 12 century AD, Nemea was occupied on a permanent basis by agricultural communities. In each case, the area of the former athletic th th sanctuary was converted to farming plots and, especially during the 5 -6 century, portions of its classical buildings were reused in the construction of ecclesiastical and domestic architecture. Importantly, the sanctuary served as the focal point of religious activity for the Early Christian 17 th and Byzantine communities. A basilica constructed around the 6 century AD (Miller 1981) on top of the Xenon that centuries earlier housed competitors in the games would have been the most prominent artificial component of the landscape. It was within and adjacent to this basilica th th th th that many of the nearly 300 burials at Nemea were placed during the 5 -6 and 12 -13 centuries AD. 3 Research Goals The aim of this research is to examine the ways in which sociopolitical transformations affected the health, living conditions, and socioreligious behavior of rural Greek communities during the Byzantine period. This is achieved through a multidimensional bioarchaeological approach that considers evidence of skeletal stress in conjunction with mortuary archaeology and historical data. While the research hypotheses and expectations presented in subsequent chapters are based on arguments presented by historians as well as by archaeologists whose interpretations relied on written evidence, my research has been conducted independently of traditional historical narratives and, in this way, my results also serve as a means of evaluating those narratives. The possibility that the impact of sociopolitical changes and disruptive events has been exaggerated in the historical record is always considered. The following are the principal goals of this research: 1. This study compares the prevalence rates of skeletal indicators of physiological stress and activity between the Early Christian and Middle to Late Byzantine communities at Nemea. The purpose of this research component is to determine whether or not the dramatic events of the 3 This figure represents the number of Early Christian and Byzantine burials that have been excavated to date at Nemea. It is possible that additional burials exist in unexcavated areas of the site. 18 latter period, such as the Frankish conquest of southern Greece during the Fourth Crusade, created living conditions that were comparatively worse than those of the Early Christian community. 2. This study compares the prevalence rates of skeletal indicators of physiological stress and activity between males and females within each period. Historical data suggests that women were a marginalized group in Byzantine society. The extent to which sex-based health disparities during this period have been examined using biological data, however, is limited. The results of this research shed light on gender roles and relationships in the Byzantine Greek countryside. 3. This study compares the prevalence rates of skeletal indicators of physiological stress and activity between the skeletal samples from Nemea and contemporaneous samples from central Greece. This research component places the results from Nemea into a broader, regional context and also allows for an examination of differences in health status between rural and urban sites. 4. This study tests archaeologically derived hypotheses about the relationship between burial location and social status in the Byzantine Empire. The mortuary space at Nemea is examined both visually and using statistical analysis for spatial concentrations of the presence or absence of physiological stress indicators that might be indicative of status or group differentiation expressed in burial location. The specific research questions, hypotheses, and expectations of this research are outlined in Chapter Five. 19 Outline of the Dissertation The remaining chapters of this dissertation are organized in the following manner: Chapter Two outlines the theoretical perspectives and models that I draw upon in my analysis of physiological stress and mortuary practice at Nemea. Chapter Three places the site of Nemea into the larger context of Greece and the eastern Mediterranean in both Late Antiquity and the Middle to Late Byzantine period. It also provides a brief overview of the archaeology of Nemea with an emphasis on its settlements and on the organization and character of the burials within the Sanctuary of Zeus. Funerary ritual and burial practices in Late Antiquity and Byzantium are the subjects of Chapter Four. In chapters Five and Six I set forth my research design. There I outline the research questions and hypotheses tested, introduce the samples of burials and human skeletons from Nemea, and review the paleopathological variables and recording system used in the assessment of physiological stress. The results of the skeletal and mortuary analyses are outlined in chapters Seven and Eight, respectively. Chapter Nine synthesizes the results of this research and discusses their implications for health and living conditions, mortuary behavior, and the effects of sociopolitical change in the Byzantine Greek countryside. My conclusions and the directions for future research at Nemea and in the broader region of Greece are presented in Chapter Ten. 20 CHAPTER 2: BIOARCHAEOLOGY AND MORTUARY ANALYSIS This chapter introduces the concept of physiological stress and discusses the biological mechanisms responsible for producing the skeletal and dental paleopathologies analyzed in this research. I also provide theoretical background necessary for the study and interpretation of aspects of burial customs that may be observable in the archaeological record. Although fundamental principles of skeletal and mortuary analysis are treated separately here, I argue for and later employ an approach that integrates the two datasets. When osteological data on stress is studied in conjunction with archaeological evidence from burials, the results are uniquely informative about fundamental aspects of past societies such as changes in the nature of religious practice through time, status differences, patterns of kinship and descent, and even patterns of gender. Physiological Stress and the Skeleton Bioarchaeologists generally conceptualize stress as a “biobehavioral response” to external stimuli that are either perceived or part of the physical environment (Goodman et al. 1988:173). Because skeletal tissue is dynamic and continuously responds to changing physiological demands, episodes of physiological stress often leave characteristic indicators on both skeletal and dental tissues (Martin et al. 1998). In this way, human skeletons recovered from ancient burials may provide a record of the life stresses experienced by the individuals they represent. Goodman et al. (1984, 1988) have proposed a generalized model of the potential for physiological disruption to produce skeletal changes. The skeletal manifestation of stress is dependent upon three factors: (1) environmental constraints, (2) cultural systems, and (3) host resistance or individual susceptibility. Environmental constraints include external stressors as 21 well as the availability of resources. Human cultural systems, such as agriculture, may act as buffering mechanisms against problems with the living environment, but are also capable of creating additional stressors. Bush (1991), for example, points out that an unstable social environment is an important potential source of psychological stress. The impact of stressors that are not alleviated through external means depends on the biological resistance of the individual. Prolonged episodes of stress that are unmitigated by cultural mechanisms or individual biological resistance have functional consequences that can impact quality and duration of life (Larsen 1997). Such episodes are the most likely to disrupt normal processes of growth and development and leave indicators on the skeleton. Recent large-scale projects have used multiple stress indicators in conjunction to standardize the assessment of health from a skeletal sample. Steckel et al. (2002a, 2002b, 2006) have even developed a health index that attempts to quantify quality of life, a generally agreed upon component of health. In this method, an index value for a skeletal sample is calculated based on the presence and expression of certain paleopathological variables as well as on the estimated duration for which those conditions were experienced. The variables chosen for the index are aimed at identifying physiological stress during all life stages. In the health index approach, health is viewed as a composite of an individual’s life experience with factors such as nutrition, disease, and occupational stress. However, there is some debate among physical anthropologists and paleopathologists as to whether the analysis of physiological stress indicators actually measures the “health” of individuals or groups. In particular, it is unclear whether or not combining variables that each measure something different into a singular value representing health is a useful approach for understanding or reconstructing the experiences of a given community in the past. Furthermore, defining health is problematic in the first place. Most 22 definitions, whether ancient or modern, include components related to mental and social wellbeing that cannot be measured from skeletal remains (King 2005; Waldron 2009). Because of the shortcomings of the health index approach, especially given the goals of this dissertation, this research uses indicators of physiological stress suggested by Steckel et al. (2002a, 2002b, 2006) independently as opposed to calculating composite values representative of the health of the two communities from Nemea. This approach makes possible general statements regarding the health status of the Nemea communities but, more importantly, allows detailed conclusions about specific aspects of health to be drawn from the prevalence and expression of each independent skeletal marker of stress. The paleopathological conditions examined in this study include linear enamel hypoplasias, porotic hyperostosis, cribra orbitalia, osteoarthritis, periosteal reactions, and trauma. The following sections examine each of those conditions in turn. Enamel Hypoplasias Enamel hypoplasias are areas of reduced enamel thickness that appear as horizontal lines, pits, or patches on the teeth. They are created when physiological stressors such as nutritional deficiencies and infections interrupt the formation of the enamel (Goodman and Rose 1990; Rose 1977; Steckel et al. 2002b). Tooth enamel is laid down incrementally by cells called ameloblasts that secrete a protein matrix. The progress of enamel layering is recorded within the tooth by internal structures called brown striae of Retzius. Brown striae that run along the sides of tooth crowns also have surface manifestations that take the form of circumferential grooves called perikymata. Most macroscopic defects that appear on the external surfaces of the teeth are linear enamel hypoplasias, which are essentially an abnormal spacing of successive perikyma grooves resulting from premature cessation of ameloblast activity (Hillson 1992, 1996; Hillson and Bond 23 1997). The activity of ameloblasts is particularly sensitive to metabolic disturbances, making enamel hypoplasias excellent indicators of physiological stress (Larsen 1997). Furthermore, because enamel is not remodeled after being laid down, hypoplastic defects constitute a permanent record of stresses experienced during childhood. Enamel hypoplasias have been associated with a wide variety of physiological disturbances and must be considered generalized indicators of stress. Roberts and Manchester (2005:75) suggest that the etiologies of dental enamel defects may fall into the categories of systemic metabolic stress, congenital anomalies, and localized trauma. In most cases, enamel hypoplasias visible in archaeologically derived human teeth are interpreted as reflecting acute episodes of disease or malnutrition from which individuals recovered and subsequently continued normal development. Differences have been demonstrated in the susceptibility of specific teeth to hypoplasia formation (Larsen 1997). In general, anterior teeth are affected more frequently than posterior teeth. Goodman and Rose (1990) further suggest that the permanent maxillary central incisors and the mandibular canines are most susceptible. Porotic Hyperostosis Porotic hyperostosis refers to porous lesions appearing on the external surface of the cranial vault. Angel (1966, 1967) suggested this term as a more accurate description of the condition “osteoporosis symmetrica,” previously offered by Hrdlička. The lesions of porotic hyperostosis appear most commonly on the parietal and occipital bones and range in expression from isolated, pinpoint pores to large, coalescing apertures (Stuart-Macadam 1989). It is usually assumed that porotic hyperostosis is associated primarily with stress experienced during 24 childhood, as lesions characterized as active at the time of death are found more frequently in children than adults (Larsen 1997; Stuart-Macadam 1985; Walker et al. 2009). Traditionally, porotic hyperostosis and cribra orbitalia (similar lesions appearing in the superior portions of the eye orbits) were viewed as having a common etiology, each being a manifestation of iron-deficiency anemia (e.g., Larsen 1997; Roberts 2005; Stuart-Macadam 1989). The appearance of the condition was normally explained as an expansion of the diploë or marrow containing layer of the skull with concomitant thinning of the outer table in response to the increased need for red blood cell production. While hypotheses on the precise etiology of porotic hyperostosis included such factors as diet (El-Najjar et al. 1982), parasites (Ubelaker 1992), and response to pathogen load (Stuart-Macadam 1992), iron deficiency was understood to be the underlying mechanism. More recently, however, the link with iron deficiency has been called into question, as has the relationship between porotic hyperostosis and cribra orbitalia. Walker et al. (2009) have argued persuasively that iron deficiency inhibits rather than facilitates the hypertrophic marrow 4 response that may at times be responsible for the appearance of porotic hyperostosis. However, it should be noted that other researchers have previously argued for the abandonment of an association between iron-deficiency anemia and porotic hyperostosis (e.g., Rothschild 2002a, 2002b; Rothschild and Martin 2006; Waldron 2009). Current research suggests that marrow hypertrophy is likely caused by one of two mechanisms. The first is a hemolytic anemia, such as thalassemia or sickle cell anemia, whereby the destruction of red blood cells occurs at a faster rate than they can be produced by hemopoietic marrow. The second possibility is a megaloblastic 4 It is increasingly being recognized that other conditions such as scurvy, rickets, scalp infections, etc. may result in porous cranial lesions that appear similar to those produced by diploic expansion and thinning of the outer table traditionally associated with porotic hyperostosis. 25 anemia, which is characterized by an enlargement of hemopoietic cells resulting in dyserythropoiesis and subsequent destruction of poorly formed red blood cells. Megaloblastic anemia is commonly caused by deficiencies in folate and/or vitamin B12 (Allen and CasterlineSabel 2001; Walker et al. 2009; Weatherall and Kwiatkowski 2003). Based on those findings, Angel’s (1966, 1967, 1978) attribution of porotic hyperostosis to a genetic anemia may be correct in some cases. When evaluating porotic hyperostosis in human skeletal samples, especially in the eastern Mediterranean, it is important to distinguish between genetic anemias and those that are related to either problems with nutritional quality or malabsorption. Congenital anemias such as thalassemia have additional skeletal manifestations that include hypertrophy of the facial bones and displacement of dental structures, generalized osteopenia, enlargement of vascular foramina, and premature fusion of epiphyses (Lagia et al. 2007; Tayles 1996; Ortner 2003). Sickle cell anemia has its own unique features such as a relative lack of involvement of the facial bones, depression of vertebral endplates resulting in a “fish vertebra appearance,” calcaneal and metacarpal lesions, and localized cranial expansion or “ballooning” (Hershkovitz et al. 1997; Ortner 2003:368). Cribra Orbitalia Porous lesions appearing in the eye orbits (cribra orbitalia) have traditionally been given similar treatment as porotic hyperostosis. It has been argued that the two conditions share a common etiology (usually iron-deficiency anemia) and that cribra orbitalia may represent an earlier manifestation of the disease process (Stuart-Macadam 1989). Both parts of this argument are problematic. As discussed above, the link between porous hypertrophic lesions and iron deficiency has been called into question through a better understanding of hematologic 26 processes. Furthermore, crania often exhibit porotic hyperostosis independently of cribra orbitalia (Walker et al. 2009). The idea that cribra orbitalia may in fact be caused by a number of different mechanisms, some of which have no necessary association with porotic hyperostosis, is not new (see Ortner and Erickson 1997; Ortner et al. 1999; Schultz 2001). However, this view has gained due attention only quite recently. Of the conditions capable of producing orbital lesions, scurvy (vitamin C deficiency) deserves special mention. Vitamin C must be obtained from the diet in humans and prolonged deficiency impairs collagen formation, leading to weakened blood vessels (Brown and Ortner 2011; Fain 2005). Hemorrhaging can result from even normal movement and may stimulate an osteogenic response if it occurs in vessels located either adjacent to bone or within the periosteum (Ortner and Ericksen 1997). Especially in children, among whom the periosteum is less firmly attached to bone and tissue formation is more rapid, subperiosteal hemorrhages ossify and appear as deposited layers of vascular bone (Brickley and Ives 2006; Brown and Ortner 2011; Schultz 2001; Walker et al. 2009). Walker et al. (2009) demonstrate that this scenario often occurs in the eye orbits and that the resultant lesions may appear similar to those produced by marrow hypertrophy associated with an entirely different process (anemia). It remains possible that the cause of some orbital lesions is anemia of either megaloblastic or hemolytic origin (Walker et al. 2009). However, differential diagnosis should also include scurvy, rickets, infectious diseases, and eye infections along with potentially associated inflammatory processes affecting the sinuses and lacrimal glands (Brickley and Ives 2008; Schultz 2001; Wapler et al. 2004). It is often overlooked but certainly worth noting here that eye problems make up a sizeable portion of the complaints cited in ancient medical treatises and inscriptions at healing shrines (Edelstein and Edelstein 1945a, 1945b; Cruse 2004; Jackson 27 2000; LiDonnici 1995). It is even possible that some cases of cribra orbitalia are due to taphonomic factors such as erosion (Wapler et al. 2004). It may be not be possible to suggest a specific etiology for orbital lesions, although a pattern of porous lesions throughout the skeleton may be suggestive of a systemic disease such as scurvy. In the absence of a clear pattern of skeletal involvement, however, cribra orbitalia should be viewed more generally as an indicator of nutritional problems, infections, and their synergistic interaction. Periosteal Reactions Periosteal reactions are areas of irregular new bone formation that are often present on the long bones of archaeologically derived human skeletons. They are produced when osteoblasts in the inner layer of the periosteum are stimulated by factors such as systemic disease or trauma (Larsen 1997; Ortner 2003). During the active phase of the inflammatory response, the deposited bone is woven in appearance. It is gradually remodeled during the healing process and will take on a sclerotic appearance as the new bone is incorporated into the original cortex. While periosteal reactions are considered non-specific indicators of stress, they are often interpreted as evidence for infection (Goodman et al. 1984, 1988; Steckel et al. 2002a, 2002b). In some cases, it may be possible to link reactive bone with specific diseases or disease syndromes. For example, treponemal infections may result in characteristic alterations of the lower leg bones that include substantial deposition of periosteal new bone on the anterior and medial surfaces of the tibiae (Ortner 2003; Rothschild and Rothschild 1995). However, more typical periosteal reactions may reflect an osseous response to a variety of stimuli. Waldron (2009) includes infections, metabolic diseases, tumors, congenital anomalies, venous stasis, and soft tissue lesions among the causes of periosteal reactions. 28 Weston (2008) underscores the difficulty of attributing periosteal reactions to any specific etiology. The problem is due mainly to the fact that bone tissue responds in a similar way to a variety of insults. Despite this lack of diagnostic potential, periosteal reactions are useful as indicators of non-specific stress, as generalized interpretations can be made based on their skeletal distribution (Gladykowska-Rzeczycka 1998). Diffuse or bilaterally expressed changes are more likely manifestations of infection, while localized periosteal new bone formation is more suggestive of minor trauma. The tibia is the most common site for the appearance of periosteal reactions, perhaps because of its location close to the surface of the skin (Roberts and Manchester 2005; Steckel et al. 2002a). Osteoarthritis Osteoarthritis is characterized by the gradual breakdown of synovial joint surfaces subjected to repeated mechanical stress. Osteoarthritis is often referred to in the paleopathological literature as a degenerative process, although that description is not entirely accurate. As the articular cartilage of joint surfaces breaks down, the response of cartilage and the subchondral bone is actually proliferative and reparative in nature. The formation of osteophytes reflects adaptive remodeling in order to mitigate the stresses placed on the joint (Roberts and Manchester 2005; Rothschild and Martin 2006; Waldron 2009). In addition to the presence of osteophytes, eburnation is considered highly diagnostic of osteoarthritis and occurs when movement at the joint continues subsequent to the destruction of the articular cartilage (Roberts and Manchester 2005). In the vertebral joints, both the process of aging and the impact of mechanical stresses over time cause degenerative changes in the intervertebral discs that lead to the production of osteophytes. Osteophytic lipping usually occurs along the anterior margins of the vertebral 29 bodies and, as in diarthrodial joints, the overgrowth of bone represents an attempt to adjust to modified and/or increased forces (Adler 2000; Roberts and Manchester 2005). The development of osteoarthritis is not straightforward. The breakdown of joint surfaces is dependent upon multiple factors such as sex, age, weight, genetics, nutrition, disease, trauma, and movement (Roberts and Manchester 2005; Waldron 2009; Weiss and Jurmain 2007). Of all of those factors, movement and associated “biomechanical wear and tear” have traditionally been considered to have the greatest influence on the expression of osteoarthritis (Goodman et al. 1984:35; Larsen 1997). In a recent review, Weiss and Jurmain (2007) point out that, at present, evidence indicates that high levels of activity beginning early in life seem to play a significant role. Weight-bearing joints such as the hip and knee are the most commonly affected, although joints of the upper limbs may be more frequently affected in ancient skeletal samples, presumably due to differences in activity patterns (Roberts and Manchester 2005). The specificity with which osteoarthritis may be used to reconstruct patterns of activity in past human populations has been debated. Jurmain and Kilgore (1995) argue that the lack of consistent patterns of expression among ancient groups suggests that assigning specific etiologies to the appearance of the condition is inappropriate. The most consistent patterns are found in the vertebral column, where alterations to the intervertebral and apophyseal joints probably in large part reflect biomechanical factors associated with normal spinal curvature and bipedal locomotion (Knüsel et al. 1997; Weiss and Jurmain 2007). This complicates the interpretation of osteoarthritis in the spine and it has been suggested that appendicular joints— especially the non-weight-bearing joints of the upper limbs—should be preferred in assessments of past activity (Knüsel et al. 1997). Another consideration is the argument that the factors contributing to osteoarthritis act on different areas of joint surfaces. For example, Jurmain (1991) 30 has suggested that certain changes such as pitting and eburnation are more associated with biomechanical stress, while age may play a larger role in the development of marginal osteophytes (Jurmain 1991). With these caveats in mind, the severity and skeletal distribution of osteoarthritis should be understood as providing only general measurements of levels and patterns of activity (Bridges 1994; Jurmain and Kilgore 1995; Steckel et al. 2002a, 2002b). Trauma The analysis of trauma provides a measurement of activity patterns that is informative about the interaction between a population and its physical and social environments. In most cases, trauma observed in archaeological human skeletal samples is the result of either accidents or interpersonal violence (Larsen 1997). Regardless of the etiology, even relatively minor traumatic episodes from which individuals recover have the potential to impair normal functioning and reduce quality of life. More serious injuries such as open fractures that penetrate soft tissue may become infected and lead to complications with healing and even death. A fracture can be defined as “an incomplete or complete break in the continuity of a bone” (Lovell 1997:141). Goodman et al. (1984) point out that fractures are the most common type of injury seen in ancient skeletal samples. When a discontinuity occurs in bone, ruptured blood vessels result in the formation of a hematoma. This process stimulates osteoblasts in the periosteum and endosteal space, which lay down woven bone to form the periosteal and medullary callus, respectively (Martin 1998). In dry bone, the presence of an osteogenic reaction demonstrates that healing had begun to take place and allows one to distinguish between injuries that occurred before the time of death (antemortem) from those occurring either around (perimortem) or after death (postmortem) (Sauer 1998). If healing continues, the primary callus of woven bone is gradually replaced by lamellar bone. Remodeling may ultimately restore 31 marrow activity and normal bone contour, but the process can take years (Roberts and Manchester 2005; Waldron 2009). Healing is likely to proceed regularly provided the broken ends remain in contact, immobile, and retain an adequate blood supply (Adler 2000). The healing process is also largely dependent on factors such as age, nutritional status, and the anatomical location of the fracture. While trauma associated with accidents and violence account for most fractures in the archaeological record, it is important to distinguish between those very different etiologies. In this regard, it is critical in any assessment of trauma to note the type of observed fracture. Lovell (1997) provides a review of common types of fractures and the forces with which each is associated. Understanding the underlying mechanism of injury helps accurately reconstruct activity patterns on both an individual and a population level. It is also important to record the location of the fracture both in the skeleton (bone and side affected) and on the individual bone or bones affected (position of the fracture on the bone) (Judd and Roberts 1999; Roberts 1991). Certain patterns of fractures are more typical of injuries sustained as a result of occupational hazards or accidental trauma. For example, frequent travel or work carried out over difficult terrain is likely to result in a high incidence of fractures due to falls. Colles’ fractures (distal radius) are often cited in this regard as they are frequently produced by forces involved with a fall onto an outstretched hand (Mays 2006). Fractures of the ulna and clavicle may result from similar indirect forces (Lovell 1997). The tibia and fibula are also frequently affected in populations that engage in intensive agricultural activities (Larsen 1997). Relatively high frequencies of these types of fractures have been observed in studies of ancient farming communities (e.g., Judd and Roberts 1999) and provide support for the notion that agriculture was a dangerous activity in the past. It should be noted here that fractures sustained while 32 carrying out activities such as farming not only have immediate health implications, but may also preclude an individual from normal participation in occupational and social activities within his or her community. The presence of other forms of skeletal trauma such as parry fractures (occurring on the distal third of an isolated ulna) and cranial depressed fractures are more likely to reflect interpersonal violence (Lovell 1997; Milner 1995; Steckel et al. 2002a). Once again, noting the type and location of the fracture is critical because it allows for the assessment of the direction of force and may provide clues as to the position of the assailant in violent confrontations. However, as Judd (2008) points out, the circumstances surrounding even injuries that likely resulted from direct blows such as parry fractures can never be completely known. The possibility exists that the individual who received the injury could have been an attacker, a victim, or even a participant in an agreed upon activity. The assessment of trauma across age categories may shed light on differences in the treatment of adults and subadults in a society. For example, the identification of trauma in children and adolescents can potentially be suggestive of the age at which they begin to participate in adult occupational and social roles. Certain patterns of childhood trauma such as multiple injuries exhibiting different stages of healing may be indicative of abuse (Walker et al. 1997). Evidence of trauma occurring during the subadult years may be impossible to discern in adult remains due to the process of remodeling. However, this is not always the case. Glencross and Stuart-Macadam (2000) point out that healed trauma discernible as subtle alterations of normal bone morphology can be diagnosed more confidently with the help of clinical data on age-based patterns of injury types and locations. 33 Issues of Interpretation: “The Osteological Paradox” In an important critical paper, Wood et al. (1992) caution that using frequencies of skeletal lesions to estimate the health of past populations may be much less straightforward than is usually assumed. They outline three factors—changes in demographic patterns, problems with the representativeness of samples, and individual variation in susceptibility to disease—that cannot generally be known and thus potentially confound the interpretation of lesion prevalence rates and age at death distributions. As a result of those factors, paleopathological and paleodemographic data are open to opposing yet equally plausible interpretations. For example, a community represented by a skeletal sample exhibiting high lesion prevalence rates and a low mean age at death might be interpreted as having experienced high levels of frailty and mortality. Alternatively, this pattern may reflect a community characterized by high fertility and long-term survival of stressful episodes such that skeletal changes were frequently manifest. The arguments set forth by Wood et al. (1992) strongly suggest that the assessment of health from human skeletal remains presents unique challenges. However, there are approaches and techniques that may help clarify the relationship between lesion frequency and health in a given sample. For example, Goodman (1993) and Cohen (1994, 1997) have pointed out that differential survival has less influence on the appearance of skeletal indicators that reflect chronic stress, such as osteoarthritis, periosteal reactions, and porotic hyperostosis, than on acute conditions such as infections. The former indicators, then, can be interpreted in a more straightforward manner. The use of multiple indicators of stress in conjunction and comparisons of their prevalence rates across age classes can also assist in the interpretation of observed lesion patterns (Pechenkina and Delgado 2006). Finally, the incorporation of contextual evidence is critical. In most cases, ethnographic and epidemiological data support the association of higher 34 levels of physiological stress indicators such as enamel hypoplasias with disadvantaged groups rather than with advantaged groups (Cohen 1994, 1997; Goodman 1993). The analysis of the burial context is also valuable for the identification of differences in status, gender relationships, or kin groups that may direct hypotheses about differential susceptibility to disease (Wright and Yoder 2003). Mortuary Analysis: Historical Development and Theoretical Background The development of bioarchaeology in the second half of the twentieth century involved a process of recognizing the advantages of integrating techniques and data from both skeletal biology and archaeology (Armelagos 2003; Zuckerman and Armelagos 2011). More recently, however, there has been a tendency to emphasize one side of the discipline or the other in research approaches and trajectories. At present the most popular definition of bioarchaeology follows Larsen (1997) and conceptualizes the field as the study of human skeletons from archaeological sites. Jane Buikstra’s (1977) original definition was more encompassing, emphasizing the study of human skeletons and their contexts. Although the implementation of Buikstra’s original model is not always practical, its major principles should not be abandoned. Goldstein (2006) has demonstrated that, while the complete integration of data from skeletal biology and archaeology presents significant challenges, ignoring data from one perspective or the other presents inherent problems. It is difficult, for example, to draw conclusions about the impact of cultural change in a community if features of the burial program are unknown. The analysis of human burials therefore offers important opportunities for collaboration between physical anthropologists and archaeologists. While not all aspects of funerary rituals are represented in the archaeological record, the mortuary domain may contain a wealth of information about the religious beliefs, cultural practices, and social organization of a group of 35 people. When utilized in bioarchaeological studies, this type of data allows anthropologists to move beyond descriptive studies and test broader hypotheses about the relationship between cultural behavior and health in the past (Armelagos 2003). Rakita and Buikstra (2005) point out that intensive cross-cultural study of mortuary practices in American anthropology was generated primarily by two publications: Arnold van Gennep’s The Rites of Passage and Robert Hertz’s “A Contribution to the Study of the Collective Representation of Death.” In his work, van Gennep (1960 [1908]) explored an underlying structure common to ceremonies marking life transitions such as birth, marriage, and death. Passages from one life stage to another, he argued, usually involve a process characterized by rites of separation, rites of liminality, and rites of reintegration. Van Gennep’s study is useful for understanding particular aspects of rituals and the reasons for which certain themes such as birth and death commonly appear together (Metcalf and Huntington 1991). Hertz (1960 [1907]) also stressed the idea of liminality in his paper focusing on secondary burial in Indonesia. He suggested that the act of allowing the flesh of a body to decay parallels the transitional state of the soul. Moving the bones to a final, family burial location represents the soul’s integration into the world of his or her ancestors. Burial and Social Organization Mortuary studies subsequent to the contributions of Hertz and van Gennep were increasingly guided by the principles of the New Archaeology, popularized by Binford in the 1960s. With the emphasis of processualism on understanding culture as a system and on a more scientific approach, archaeology became more rigorous in the search for models that explained cultural processes (Johnson 1999; Trigger 1989). It was during this period that Arthur Saxe completed his dissertation research, which proved to be an extremely important contribution to 36 the analysis of mortuary practices. Saxe (1970) developed and tested eight hypotheses which together aimed to reconstruct social organization from aspects of burial practice. While most of those hypotheses focused on the relationship between mortuary variables and social identity, Hypothesis 8 dealt with the spatial component of burials and has since been given the most attention. Saxe’s (1970:199) original hypothesis states that groups will reserve formal areas for the burial of the dead based on the degree to which control over restricted resources is derived from lineal descent from the dead. Goldstein (1976, 1981) has reformulated that hypothesis to include related sub-hypotheses, arguing that permanent, bounded burial areas are likely to represent corporate groups and that the maintenance of those areas are one way in which corporate groups may legitimize rights to restricted resources. The association between burial practices and status differentiation was a significant emphasis of mortuary studies in the New Archaeology. Binford (1971) used ethnographic data to demonstrate that the number of social dimensions related to status increased with the development of settled agriculture. Working from that data, he argued that burial practices should increase in complexity with agricultural intensification and sociopolitical complexity. Looking more specifically at grave contents, Brown (1971:101) related the presence of “important and precious” mortuary artifacts to higher rank in his formal analysis of aspects of the burial program at the Mississippian site of Spiro. The ideas of Saxe, Binford, and Brown regarding the elaboration of the mortuary domain figure prominently in later studies concerned with developing models to identify certain forms of sociopolitical organization using archaeological evidence (e.g., Peebles and Kus 1977; O’Shea 1981; Tainter 1977, 1978). Those studies, however, had important contributions of their own. Peebles and Kus (1977:431) explored two different dimensions of the social persona: 37 “subordinate,” based on age, sex, and life achievement, and “superordinate,” which cannot be attributed to those factors. They argued that ranked societies can be identified by a burial program that includes both dimensions. Tainter (1977, 1978) focused on the role of energy expenditure in funerary ritual and the construction of the interment facility. He argued that energy expenditure can to a degree act as an independent source of data on rank because it can be evaluated across a cemetery even in situations where associations between graves are unclear. Tainter’s work suggested that there is a relatively direct correlation between the amount of energy expended in the mortuary ritual and the social rank of the individual. While continuing to work within the framework of the “Saxe-Binford research program” (Brown 1995), archaeologists studying funerary remains in the 1980s were increasingly cognizant of the fact that reconstructing social organization from burial evidence is not as straightforward as it had been made to appear. O’Shea’s (1981) study of the effects of formation processes on funerary remains among the Arikara, Pawnee, and Omaha achieved a more nuanced picture of vertical and horizontal differentiation as well as energy expenditure. In his case studies, vertical social distinctions (rank-based) were found to be more visible in the archaeological record than horizontal social distinctions (kin-based). O’Shea attributed this to the fact that vertical distinctions were symbolized in the mortuary ritual by features such as elaborate grave construction, while horizontal distinctions tended to be expressed using organic materials, such as clothing and furniture, that are less likely to be preserved. Brown (1981) pointed out that distinctions can be made between rank, power, and authority and that those distinctions might be represented in different ways in the mortuary domain. Prehistoric societies may also demonstrate gradations of social ranking (e.g., ranking without centralized leadership) that further complicate the interpretation of burial data. Brown 38 also called attention to the fact that the ritual process preceding burial needs to be carefully considered. This is because complicated, multiphase rituals may give the appearance of groups having different statuses. Outlining these potential difficulties, Brown stressed the importance of developing site-specific models for the interpretation of social behavior from mortuary data. The Saxe-Binford approach has not been without its critics though. Responding mostly to the research of Tainter (1977, 1978), which focused on social change in the Middle to Late Woodland transition, Braun (1981) demonstrated how compounding error during data manipulation procedures may bias results of analyses that associate features of a burial program with social rank. Braun argued that Tainter’s analyses were problematic first on the basis of his choice and execution of statistical tests. He also pointed out inconsistencies in the classification of locally produced and imported grave goods that biased the final assessments of complexity for each period in question. Finally, Braun identified flaws in Tainter’s theoretical assumptions. For example, in his argument that differences in energy expenditure indicate differences in social rank, Tainter fails to address the possibility that differences in energy expenditure may occur between individuals of the same rank. Rakita and Buikstra (2005) add to Braun’s critique of the processual approach the idea that certain elements of a burial program, such as spatial relationships, are not easily quantified. Postprocessual Approaches Reactions to processual studies on both methodological and theoretical grounds gradually coalesced into a new paradigm. In contrast to the major assumptions of the Saxe-Binford program, postprocessual approaches maintain that funerary rituals and their material remains need not be a direct reflection of the social hierarchy or of any other aspect of sociopolitical organization. For example, there is no necessary connection in postprocessual models between 39 social status and factors such as energy expenditure and the presence and type of grave goods. Instead, funerary rituals and burials themselves are viewed as areas of active contestation and negotiation by the living (Parker Pearson 1999). Rather than reproducing the social hierarchy, funerary rituals can be employed as mechanisms to manipulate or even transform it (Rakita and Buikstra 2005). While postprocessual studies are similar in their criticism of the Saxe-Binford approach, they represent a diversification of research perspectives which consider the roles of gender, agency, and emotion in the formation of the mortuary domain. They have also added to preexisting analytical frameworks. For example, analyses of the spatial component of cemeteries have expanded to include investigations of the relationship between cemeteries and their associated regional landscapes. A critical point of emphasis in these postprocessual critiques is that care must be taken to examine the larger social and historical context in the interpretation of material remains. Hodder (1982) pointed out that reconstructing the functional relationships of past societies addresses only part of a larger problem. This is because those relationships are based on symbolic rules that are presented in ways that may change as individuals or groups negotiate statuses and roles. As the act of burial is one arena in which negotiation may occur, burial patterns cannot be taken as direct reflections of social differentiation. Furthermore, because material remains are imbued with symbolic meaning that is unfixed, objects may have different meanings in different contexts. This is a fundamental obstacle to the development of models that attempt to translate material remains or energy expenditure into social rank. Similarly, Shanks and Tilley (1982:132) argued that the social order is legitimized using ideology, which is meant to “conceal real social relations”. Like Hodder, they underscored the fact that material culture is created within a symbolic context and, when used in mortuary rituals, 40 may function to maintain the social order through the misrepresentation of social divisions. Shanks and Tilley also demonstrated that the human body and its manipulation have great symbolic significance. In their analysis of English and Swedish Neolithic barrows, they point to the arrangement of disarticulated skeletal elements during secondary burial as an indication of the denial of structured social differences and the assertion of collective identity. The practice of exhumation and reburial in village ossuaries is common today in areas of rural Greece. In this modern context, secondary rituals have been interpreted in a manner consistent with van Gennep’s (1960 [1908]) rites of passage. They mark the end of the liminal mourning period and the incorporation of the deceased into the world of the dead. Storage of the remains in an ossuary also emphasizes the collective identity of the village dead (Alexiou 2002; Danforth and Tsiaras 1982). Brown (2003) argued that ossuary burials can be understood as expressing collective identity when disarticulation is consistent across all individuals. In such cases, the treatment of each of the deceased in the same manner is likely carried out in an effort to replace individual identities with those of a community or group. His analysis of the Cahokia Mound 72-Sub 1 burials makes clear that interpretations involving collective representation are possible even in instances where there appear to be obvious status differences. For example, burials might be arranged as a performance in which certain individuals are given symbolic roles that could resemble status differences. Secondary burial involving mortuary ceremonies may also have political objectives. Chesson (1999) has pointed out that the visibility of secondary ceremonies, which may involve the transportation or alteration of the remains and/or the grave structure, provides an ideal opportunity to either reinforce or renegotiate social and political associations. At Bab edh-Dhra’, 41 secondary burial in large charnel houses is suggestive of an expansion of formal kinship relations and probably served to strengthen the connections between kin groups and their ancestors. The use of secondary funerary rituals to create and reinforce ancestral ties in order to legitimize the status of the living has also been suggested for the Classic Maya (Gillespie 2001) and at Cahokia (Goldstein 2000; Porubcan 2000). Although there is certainly utility in being able to identify aspects of social organization using mortuary data, postprocessual studies emphasize that burial patterns are influenced by other factors. Decisions regulating the burial of adults versus non-adults, for example, might be influenced by factors other than age or status such as beliefs about attaining a state of humanness or the development of the soul. The need to examine belief systems is emphasized by Carr (1995) in his analysis of ethnographic data from the Human Relations Area Files. Although he found associations between social organization and both the form of cemeteries and energy expenditure, he demonstrated that aspects of mortuary treatment such as body position and orientation were more likely to reflect philosophical-religious beliefs. Cannon (1989) has argued that mortuary expression is largely influenced by competitive display and changes through time in a similar manner as fashion in dress. In this model, increasing elaboration by elites is gradually adopted by lower status groups. To maintain their social distinction, elites subsequently scale back the intensity of their funerary expression. Using cross-cultural examples, Canon showed how historical cycles of elaboration and restraint in mortuary behavior might be explained by competitive display. The implication of Cannon’s hypothesis is that changes in mortuary expression do not necessarily correspond with changes in social structure. They may instead signal only changes in context-specific patterns of display and emulation. 42 Gender and Mortuary Analysis The archaeological study of gender as a social construction is another important development associated largely with postprocessual approaches (Parker Pearson 1999). In 1984 Conkey and Spector forcefully demonstrated that assumptions about gender were pervasive in the archaeological literature. Especially prior to their work, there was a widespread tendency to impose modern, western cultural ideas about gender on past societies during the interpretation of archaeological deposits. This was especially apparent when mortuary remains were considered. Conkey and Spector’s critical review identified instances in which the same objects found in the burials of males and females had been interpreted differently to conform to western gender stereotypes. For example, trade items found in the burials of males were assumed to indicate active participation in trade, whereas similar items found in female burials were suggested only to have been acquired. O’Gorman (2001) points out that mortuary studies in particular are important for the investigation of gender in the past because they may incorporate skeletal analysis, which can provide reliable estimates of biological sex. The use of osteological techniques is critical in the avoidance of traditional approaches in which sex identifications were made based on associated mortuary artifacts (Weglian 2001). A large body of research demonstrates clearly that associations between males and females and the presence of certain types of adornment or artifacts are variable both cross-culturally and within a single society (Conkey and Spector 1984; Crass 2000, 2001; Crown and Fish 1996; Doucette 2001; Hamlin 2001; Hollimon 2011; Parker Pearson 1999; Weglian 2001). At the same time, Hollimon (2011) points out that bioarchaeologists using techniques from mortuary analysis and skeletal biology are in an 43 exceptional position to identify third gender categories in the archaeological record, although this must be done carefully using multiple lines of evidence from both approaches. Mortuary analyses that focus on gender may be informative about the roles of men and women in past societies as well as changes in divisions of labor through time (Crass 2000, 2001; Hamlin 2001). O’Gorman’s (2001) examination of Oneota domestic and mortuary contexts suggests that women exerted control over household production and storage, in addition to having a diversity of inter- and intra-household social relationships that were reflected in greater variability in their mortuary patterns relative to men. In an analysis of the Pre-Classic to Classic transition among the Hohokam, Crown and Fish (1996) argue that women’s workload increased through time while their public presence declined. However, they suggest that women derived prestige at the level of the household or household group from their productive activities, which became increasingly important as social differentiation increased. Reconstruction of the activities of men and women is another area in which osteological data may be successfully integrated. Indicators of occupational stress such as osteoarthritis, enthesopathies, and variation in bone cross-sectional geometry are especially useful in the identification of sex-based patterns of activity (Hollimon 2000, 2011; Larsen 1997; Peterson 2000). Joyce (2001) offered a unique approach to the analysis of gender in mortuary studies. She argued that burials represent sites at which social memories of the dead are constructed by the living. While the statistical analysis of mortuary data can identify certain patterns, they cannot capture the emotions involved in the performance of burial rites and in the reactions of those connected to deceased. To the extent possible, it is useful to combine more traditional analyses with more personal, biographical accounts of individual burials. Joyce’s approach to the burials at Tlatilco allowed her to focus on subtleties that make each woman’s grave distinctive. 44 Spatial Analysis The analysis of the organization of burial space is critical because the use of space is often meaningful on a number of levels. Decisions about the location of the dead are important and are usually made very deliberately (Parker Pearson 1999). Ashmore and Geller (2005) have pointed out that spatial patterns in cemeteries may contain information about the relationships between individuals and groups of individuals, as well as insight into the broader cosmological concerns of past societies. The multidimensional nature of spatial patterns has also been discussed by Goldstein (1981). She demonstrated that many ideas can be expressed simultaneously in the location of burials and in the relationships within the grave between mortuary artifacts and the body. Ideally, spatial analysis should include assessments of the arrangement of burials relative to each other, their relationship with the surrounding landscape, and their individual contexts (e.g., aspects of grave construction, internal organization, and biological attributes of the deceased) (Goldstein 1981:59). As discussed previously, Saxe’s (1970) Hypothesis 8, which has been modified by Goldstein (1976) and later Charles (1995), has generated much discussion concerning the association between spatially bounded cemeteries and control over access to resources by corporate groups. Hypothesis 8 has been criticized on the grounds that it reduces the treatment of the dead and the creation of cemeteries to economic considerations and ignores other ideas that social groups might have been attempting to communicate (Morris 1991; Parker Pearson 1999). However, keeping in mind that the presence and form of cemeteries may be the result of processes other than attempts to establish control over property, the connection between spatially distinct disposal areas and corporate groups is still a useful concept in many contexts (Brown 1995; Morris 1991; Parker Pearson 1999). 45 Buikstra and Charles (1999), for example, relied on the Saxe/Goldstein hypothesis in their analysis of mortuary sites in the lower Illinois valley. They showed how landscapes, both natural and constructed, were utilized over millennia for negotiating relationships between kin groups. The monumental structures of Mississippian elites drew upon but also transformed the much earlier Archaic tradition of placing burials strategically in the landscape. Buikstra and Charles also demonstrated that it may be possible to correlate changes in the location of competitive mortuary rituals and the burial of the dead with changing sociopolitical conditions (Buikstra and Charles 1990; Charles and Buikstra 2002). The idea that the mortuary landscape is constructed within and thus influenced by a dynamic social context has also been illustrated using modern cemeteries, for example in Peru (Silverman 2002) and Greece (TzortzopoulouGregory 2008). In some cases, interpretations of the meaning of mortuary sites can be reached only through their analysis within a larger regional context. For example, Goldstein (1995) expanded upon a number of previous interpretations of Effigy Mound sites in southeastern Wisconsin by examining their locations relative to the distribution of regional resources. Since diverse forms of Effigy Mounds concentrated in areas of diverse resources, she argued that, in addition to places of burial, the mounds functioned as symbolic maps indicating directions to group resources. Recent studies of the organization of cemeteries have demonstrated the utility of incorporating more sophisticated spatial-analytic techniques such as GIS. The mapping of burials and human remains accurately in space has been successfully applied to diverse goals such as illustrating changes in patterns of commemoration through time (Tzortzopoulou-Gregory 2008), elucidating spatial relationships in ossuaries and commingled burials (Herrmann 2002), investigating artifact distributions in graves and patterns of grave construction, and identifying 46 spatial patterns in the distribution of skeletal paleopathologies that may be suggestive of social differentiation reflected in burial location (Jenny 2011; Soler 2011). Building on these studies, GIS is employed in the present research in order to test hypotheses about the organization of the mortuary space at Early Christian and Byzantine Nemea. As Goldstein (2006) demonstrates, the spatial component of mortuary sites is one of many areas in which increased collaboration between archaeologists and physical anthropologists would prove fruitful. There is additional room for collaboration with historians as well. Working with periods for which written evidence is available presents its own opportunities and challenges. While firsthand or at least contemporary accounts of historical processes and events are invaluable, it must be recognized that they are subjective and produced by people who often had agendas other than the faithful recording of those events. Morris (1992) has underscored the importance of using mortuary evidence in the study of sociopolitical change in the complex societies of the classical ancient world. The analysis of burials gives access to a much broader segment of society than written sources, which were produced primarily by and were directed mainly toward the elite. In this way, mortuary analysis and the bioarchaeological study of human skeletal remains can provide much more than conformational evidence set against an historical record that is taken to be accurate. Instead, they can function as independent sources of information that can be used to supplement and even challenge traditional historical narratives (Morris 1992; Perry 2002, 2007). Summary This chapter has focused on two related approaches that are employed together in this research: bioarchaeology and mortuary analysis. In the first section, I discussed the concept of physiological stress as commonly used in the assessment of health from human skeletal remains 47 recovered from archaeological contexts. The skeletal indicators of stress introduced in this chapter, while generally understood as non-specific, each provide a different perspective on stress in past populations. For example, linear enamel hypoplasias are informative about acute episodes of stress experienced during early childhood, while periosteal reactions may be present at any age. Porotic hyperostosis and cribra orbitalia are likely to reflect physiological disruption caused by nutritional or metabolic disturbances, while osteoarthritis is due largely to activityrelated stress. The assessment of multiple stress indicators increases the likelihood of avoiding the pitfalls described by Wood et al. (1992) in reference to the interpretation of skeletal lesions. In the second section, I traced the historical development of the analysis mortuary sites and, in so doing, introduced multiple theoretical approaches for their interpretation. I placed particular emphasis on considerations of the expression of status differentiation and gender in the archaeological record, as well as on the investigation of the spatial component of cemeteries. Those areas are the most pertinent to the present research at Nemea. I argue that the evidence from skeletal biology should be used in conjunction with archaeological and historical data not only to place the biological evidence in context, but also so that assumptions about the other two datasets may be tested. The historical and archaeological contexts of the Early Christian and Middle to Late Byzantine communities at Nemea are the subject of the next chapter. 48 CHAPTER 3: GREECE IN THE EARLY CHRISTIAN AND MIDDLE TO LATE BYZANTINE PERIODS: PLACING THE NEMEA COMMUNITIES IN CONTEXT Central to a bioarchaeological approach, as conceived by Buikstra (1977), is the idea that human remains cannot be studied in isolation. By themselves, skeletal remains have the potential to shed light on patterns of biological factors such as stress, disease, and activity. However, it is only when those biological datasets are combined with the archaeological and cultural context of the remains that they may contribute in a meaningful way to anthropological investigations of past societies. Thus, the most effective anthropological study of the human skeletal remains from Nemea first requires a careful examination of the social and political context of each community that permanently occupied the site. The Early Christian and Middle to Late Byzantine inhabitants of Nemea lived and died during periods characterized by major social and cultural transformations. In Late Antiquity (around 200-700 AD) the transition from paganism to Christianity and a shift in the political center of power from Rome to Constantinople altered the social and political structures that had th previously dominated the Mediterranean. During the 6 century AD, the impact of wars, invasions, disease, and even climactic shifts resulted in a period of relative decline that lasted for centuries in parts of southeastern Europe including Greece (Gregory 2010). While the Middle Byzantine period (717-1204 AD) saw renewed prosperity in the Byzantine Empire, conditions in th the countryside gradually deteriorated in the second half of the 12 century as a result of increasingly dysfunctional provincial administration. This seems to have been a precursor to the disaster of the Fourth Crusade and the Frankish invasion of the Peloponnese. This chapter considers in detail the nature of the post-classical communities at Nemea and situates them within the broader context of sociopolitical change in Greece and the Byzantine world. 49 Greece and its Place in the Mediterranean World of Late Antiquity Cities, Towns, and the Countryside For Greece and, indeed for most of the Mediterranean world, the post-classical period was characterized by transition. While scholars continue to debate the competing influences of continuity and change, there can be little doubt that both processes were at work in the shaping 5 of the social and physical landscapes of the late antique Mediterranean. In Greece cities continued to be the centers of population and cultural life, but a noticeable transformation of their character gradually took place (Gregory 1984). Cameron (1985) observes that traditional elements of classical cities such as open marketplaces and baths were frequently replaced by shops and other spaces that were more confined. It was also at this time that burials began to be placed in the fora of urban sites like Corinth, a practice forbidden during the Roman period (Ivison 1996; Sanders 2002). These developments together demonstrate that late antique cities were functioning in a manner that differed from their classical past. The alterations to the physical appearance of cities have been attributed to a number of factors. Saradi-Mendelovici (1988) has pointed out that those changes should be attributed more to Christianization and administrative processes than to natural disasters or external threats. By th th the 5 -6 centuries, the influence of Christianity was becoming increasingly apparent as basilicas were constructed in cities and towns throughout the Greek East (Cameron 1985; Gregory 2010; Sanders 2002). Christian attitudes and ways of life in some ways contrasted greatly with those of pagans—a fact that resulted in the open criticism and avoidance of a 5 The concept of “modified continuity” is a useful way to understand how a middle ground may exist between the extreme views of continuity and discontinuity of classical ways of life into Late Antiquity. In this perspective, cultural change is understood as the result of a pragmatic selection process in which certain cultural elements are changed and others are maintained. For more detailed discussions of this concept see Kardulias (1995) and Frey (2006). 50 number of classical institutions such as the theatre, public baths, and the agora. Christianity also fundamentally changed the classical system of wealth redistribution. While affluent citizens traditionally supported the construction of public monuments, they began to redirect their wealth toward the Church instead (Leyerle 1994; Spieser 2001a, 2001b). This ideological shift and its consequent economic effects explain patterns visible in the late antique phases of sites such as Thessalonike and Butrint, where impressive ecclesiastical architecture was constructed and maintained at the same time as classical urban spaces were neglected (Bowden 1997, 1999; Spieser 2001b). Another outcome of this process was that local administrative elites, finding their posts increasingly less fruitful, began to abandon them for imperial or ecclesiastical positions (Curta 2006, Saradi-Mendelovici 1988, Spieser 2001a, 2001b). The power and influence of the clergy in local communities grew substantially as a result. At least some of the changes to the form of towns and cities across southeastern Europe th th around the 5 -6 century AD can also be explained as a result of the growing insecurity of the Early Byzantine State and the execution of a new and ambitious defensive strategy. Curta (2001a, 2006) has demonstrated, for example, that the contraction and, in many cases, abandonment of Roman rural sites in the Balkans in favor of sites that were largely hilltop fortifications was related in large part to the emperor Justinian’s efforts to protect the Empire against barbarian incursions. Justinian was nearly successful in an attempt to reunite the Roman Empire by reconquering Italy. However, the drawn out war in the West along with his building program, which included both the refurbishment of old fortifications and the construction of new ones, placed a great deal of financial and logistical strain on the Empire (Cameron 1985; Curta 2001a). 51 In the period immediately following Justinian’s reign, invasions became more difficult to th control. The collapse of the Danube frontier in the late 6 century allowed groups of Bulgars and Slavs to enter Greece, the latter group managing to cross the isthmus of Corinth and invade the Peloponnese in the 580s (Gregory 2010; Herrin 1985; Whitby 2000). Archaeological evidence reveals a further decline in the use of public buildings, the contraction of major cities, and the abandonment of many rural sites at that time (Cameron 1985; Curta 2001a; Herrin 1985; Whitby 2000). The exact nature and impact of the so-called “Slavic invasions” of Greece during the late th 6 century has been the subject of considerable debate. In the traditional narrative, which is based on written sources such as the Miracles of St. Demetrios and the Chronicle of Monemvasia, groups of Slavs migrated south across the Danube and carried out massive, sustained invasions of Byzantine territory in the Balkans and Greece. In many cases, archaeological evidence of destruction and abandonment of settlements dating to this period has been linked to those activities. The Slavs are also suggested to have occupied southern Greece for more than two centuries (Gregory 2010). Recent syntheses of historical and archaeological data, however, suggest that the accepted narrative paints an unrealistic picture of the Slavic presence in the Balkans. For example, Curta (2001b) has argued persuasively that the Slavs gained political and military relevance largely as a result of conditions created by Justinian’s fortification program. In this way, the Slavic ethnicity and political identity were each created by the Byzantines (Curta 2008). Other groups such as the Goths and the Avars were implicated in raids of Greece and it is possible that groups of Slavs were operating under Avar control (Curta 2004; Heather 1996). In 52 fact, Curta (2001b, 2004) points out that there is a great deal of confusion even among th contemporary sources as to the identity of those invading Greece in the late 6 century. Another problem with the traditional narrative of the Slavic invasions concerns both their immediate and long-term ramifications. Not all evidence of Slavic presence in southern Greece is associated with devastation. For example, Gregory (1993) identifies a possible Slavic settlement at Isthmia based on ceramic deposits that are consistent with permanent habitation rather than destruction and conquest. The conception of Greece as overrun with Slavic invaders has also led some researchers to interpret the late antique occupation of small islands around the Greek mainland as evidence of groups of people fleeing for safety (Hood 1970). More recently, however, Kardulias et al. (1995) have presented alternatives to this model. They demonstrate that at least some of those island settlements were of a permanent nature and that daily life would have necessitated continuous interaction with the mainland. It is more likely, then, that such settlements reflect the exploitation of marginal areas as strategy of coping with a problem very different from continuous invasion: increasing population pressure. The reconsideration of the Slavic invasions is only one line of evidence that has been used to question the extent to which the late antique countryside experienced universal decline (Kazhdan 1997). Archaeological surface surveys have produced surprising data that suggest at least some areas of Greece were experiencing population growth and expansion. For example, th th Bintliff (1996) has demonstrated that parts of central Greece in the 5 -7 centuries were not only more populated than contemporary accounts suggest, but appear to have been nearly as prosperous as they were during classical times. Increased activity and settlement density in the Greek countryside during Late Antiquity has also been noted generally by Alcock (1993) and, more specifically, in the southern Argolid (Runnels and van Andel 1987) and the Nemea Valley 53 (Wright et al. 1990). Runnels and van Andel (1987:319) go as far as to characterize the period of rd th the 3 through 6 centuries AD as “generally an era of prosperity and peace.” While survey data at first glance clearly indicate a rise in the number of settlements during the Late Roman period, other interpretations are possible. Pettegrew (2007, 2010) cautions that the large number of settlements identified in the region of Corinth and elsewhere in Greece may be due to the ease with which Late Roman pottery is recognized in surface surveys when compared with Early Roman pottery. Adjusting for differential visibility, there is much less contrast between the two periods, suggesting more continuous economic activity as opposed to a pattern of boom and bust. This need not, however, mean that the late antique countryside of Greece was stagnating. Even if the model of marked settlement expansion is incorrect, the amount of identifiable Late Roman ceramics, especially imported pottery and fine ware in the case of the Corinthia, is much more consistent with prosperity than decline (Pettegrew 2007). The ability of settlements to flourish in the Greek East during Late Antiquity was likely due in part to the strength of the Eastern Empire relative to the West during this period. Cameron (1985) reports that Constantinople was able to buy off invading groups at least temporarily. The location of Greece also probably played a role as it was, for the most part, buffered from the th turbulent events in the northern Balkans until the late 6 century (Gregory 1984; Whitby 2000). The fact that settlements like Nemea were able to succeed in valleys as opposed to fortified hilltops is further evidence that the Early Christian period in southern Greece was perhaps not as turbulent as previously believed (Rautman 2006). The most recent reassessments of the data from regional archaeological surveys in Greece th suggest that it is not until the 7 century AD that the number of settlements drops off 54 significantly (Pettegrew 2007, 2010). Even this pattern, however, could be explained by a shift toward local pottery of relatively poor quality that was less likely to be preserved (Pettegrew 2010). In this way, the extent to which the countryside was actually abandoned remains open for debate. While the pattern of contracting settlements demonstrated by Curta (2001a, 2006) for most of the Balkans may be clearer, the survey data in Greece, together with a revised understanding of the Slavic presence, have produced a more optimistic view of the late antique countryside. As will be outlined in greater detail below, the Early Christian community which th th occupied Nemea during the 5 -6 centuries was an example of the flourishing of rural Greece up to the abandonment of the site in the 580s AD. Religious Life The rise of Christianity, visible in the physical landscape of late antique cities and towns in the form of religious architecture, was a reflection of what Brown (1989:49) describes as a “new mood” that developed in the later Roman Empire. The worldview of the Empire in Late Antiquity was increasingly personal. In contrast to the pagan emphasis on objects such as temples and oracles, attention was turned to individuals as intermediaries between God and human beings. The emperor Constantine, who famously converted to Christianity after his victory over Maxentius at the battle of the Milvian Bridge in 312 AD, was a relatively early example of this process (Brown 1989). The growth of monasticism and the prominence of holy men in Late Antiquity also attest to a need to fill the void created by the absence of classical institutions and the neglect of religious buildings such as temples, which were more often in a state of disrepair (Brown 1971). Along with the shift in focus from things to people was a movement away from concern with religious practice toward concern with religious belief. Garnsey and Humfress (2001) point 55 out that there continued to be a psychological connection between the happiness of the gods and the prosperity of Roman society. What changed, however, was that the happiness of the Christian God was understood to depend on the correctness of the belief system as opposed to the precision with which rituals were carried out in the pagan tradition. Contemporary accounts of hostility between pagans and Christians and the persecution of the latter during the early Christian centuries give the impression that conflict between the two religious groups was the norm. While such opposition cannot be ignored in a discussion of the religious climate of Late Antiquity, there is plenty of evidence to suggest that the Early Christian period, perhaps especially in Greece, was replete with examples of coexistence and even syncretism. The utilization of pagan sacred space for religious activity by Christians was 6 extremely common, albeit the motivations behind such examples are not always clear. The site of Nemea, which features a relatively large Early Christian basilica and Christian cemeteries placed directly within what would have remained an obvious former pagan sanctuary illustrates 7 this scenario well. In addition to the continuity of religious space, Gregory (1986) points out that both religious traditions addressed broad life concerns such as healing, death, and the family. Selectively retaining elements of paganism would thus have been useful in understanding Christian concepts during conversion. Trombley (1985) echoes that sentiment and adds that the process of Christianization in the countryside of Greece and Anatolia was characterized by the integration of Christian themes as opposed to the wholesale elimination of pagan practice. 6 While constructing churches near the boundaries or at the former sites of pagan temples suggests continuity of religious practice, these actions may often have been deliberate attempts by Christians to intimidate pagans and/or eliminate cult activity. 7 nd Writing in the 2 century AD, Pausanias (2.15.2) reports that the Hellenistic Temple at Nemea was still present, although its roof had collapsed and its cult-statue was missing. 56 th By the late 4 century, legislation outlawing aspects of pagan worship such as public sacrifice was put into place. It was also around this time that many instances of the destruction of pagan temples and looting of shrines, especially in rural areas, were recorded (Caseau 2004; Garnsey and Humfress 2001). It is important to keep in mind, however, that there is no necessary connection between the destruction of temples and the end of paganism. In many cases, temples were destroyed only after they ceased to function as the sites of pagan ritual (Spieser 2001c). More importantly, Garnsey and Humfress (2001) point out that paganism was traditionally a flexible system and individuals could draw upon local cults, imported traditions, or simply employ alternative forms of worship in order to renegotiate their religious identities within a pagan framework. The many examples of syncretic cult activity in the historical and archaeological records illustrate a diversity of religious practice in Late Antiquity and confirm that paganism not only survived, but continued to evolve (Bowersock 1990; Garnsey and Humfress 2001; Trombley 1985). Archaeological evidence suggests that, in southern Greece, pagan cult activity may have th continued, perhaps alongside Christian worship, well into the 6 century AD Rothaus (2000) argues for this situation at Corinth, where votive lamps must have been offered during the 6 th century at the site of the Temple of Asklepios, which had been destroyed perhaps a century earlier. Additionally, while Christian burials encroached upon the site of the former temple, they never occurred directly on it. This may indicate that Christians were using burial space to intimidate pagans, but in a manner that respected those still carrying out rituals. Rothaus (1996, 2000) also argues that an abandoned bath structure near the Asklepieion was the site of ritual th th deposition of votive lamps by both Christians and non-Christians during the 4 to 6 centuries. 57 If correct, this demonstrates that Christians and non-Christians interacted in ways that involved performing the same act at a site that had mutual ritual significance—the only difference was the 8 deity worshipped. Examples of shared forms of religious behavior also show that the distinction between the categories “pagan” and “Christian” was in reality often ambiguous (Rothaus 1996). th It is clear that small pockets of pagan activity survived beyond the 6 century. Trombley (1985) identifies a number of minor pagan rituals from rural Greece and Anatolia between the th th 6 and the 10 century, recorded largely in hagiographic texts. However, the Byzantine Empire was, for the most part, thoroughly Christian in religious orientation (Rautman 2006). Based on th th the Christian architecture as well as on the burial practices that will be described later, the 5 -6 century community that developed at Nemea in plain view of the monuments of the former pagan athletic sanctuary shared this worldview. Gender The growth of the Christian Church had mixed results for the status of women, who were generally expected to have a lesser public presence than men in the Greco-Roman tradition. In the early Church, women were able to take on positions of leadership. Women had public speaking roles in churches at Corinth and at Kenchreai the church featured a female minister (Snyder 1999). As Christianity grew in popularity and worship became more public, however, women were more often marginalized. They were not permitted to join the priesthood and their roles in the church were usually limited. For example, women might serve as deaconesses, 8 Sanders (2005) has more recently argued that archaeological evidence for Hellenic worship at th Corinth is difficult to discern after the 4 century AD. While he acknowledges that the presence of vessels appearing in apparent Christian graves in the Asklepieion cemetery is reminiscent of pre-Christian practice and may indicate syncretism, he favors a later date for the graves and disagrees with the assertion that Christian burial there was meant to discourage pagan ritual. 58 chosen from among virgins and widows to assist other women in the ritual of baptism and during times of illness (Alexandre 1992). One option that offered significantly more freedom to women was participation in the monastic movement that increased greatly in popularity during Late Antiquity. In addition to the spiritual benefits, joining a female monastic house could provide intellectual stimulation, opportunities for travel, and a way to avoid the health risks associated with pregnancy and childbirth (Alexandre 1992; Garnsey and Humfress 2001). While widows were in a unique position to lead an ascetic lifestyle, this was also a possibility available to young women who chose to renounce marriage and sexuality (Elm 2000 [1994]). Women’s subordinate position in the Church extended to other facets of life. Views of women drew upon scriptural themes depicting them as both weak and prone to leading men into temptation (Garnsey and Humfress 2001). Ideas regarding the suppression of such supposedly innate qualities influenced early Christian texts, which included prescriptive recommendations even for such mundane activities as eating. Grimm (1995) points out that young Christian women in Late Antiquity were urged to maintain restricted diets in order to minimize physical desires such as sexuality. The extent to which such recommendations were followed is unclear but, if adopted, they may very well have had negative consequences for the nutritional status and overall health of women during this period. The public/private dichotomy of men and women seems to have become stricter over time. Rautman (2006) describes a number of ways in which late antique women participated in public life. In addition to attending religious services and festivals, women frequented public institutions such as the theater and baths and played a crucial role in the fate of socio-religious movements such as iconoclasm. He suggests that later periods of Byzantine history were 59 characterized by greater restrictions on women’s activities outside of the home, which may have been limited to the church. It should be noted that non-elite women, especially in the countryside, probably participated in a wider range of activities than their aristocratic urban counterparts. Health and Health Care A number of options were available in Late Antiquity to individuals suffering from medical problems, although access to treatment outside of the care of the family was not always easy and its quality varied tremendously. The treatment of the sick during this period still drew th th largely on the Hippocratic tradition (5 and 4 centuries BC) and, to an even greater degree, on nd the work of Galen of Pergamon, a prolific doctor practicing and writing in the 2 century AD (Nutton 1984). The Christianization of the Empire also significantly influenced health care, perhaps even creating greater demand for it by focusing more attention on bodily suffering as well as on categories of people such as the sick and the poor (Horden 2008; Nutton 1984; Perkins 1995). Ultimately it was a Christian religious institution—the monastery—that created a new form of treatment center that would become the first hospitals. In classical antiquity, cult sites dedicated to the god Asclepius were prominent centers of healing throughout the Mediterranean. The cult of Asclepius serves, however, as another example of the continuity of pagan traditions in the late antique world. Sanctuaries are reported th to have functioned until around the 6 century AD (Edelstein and Edelstein 1945; Wickkiser 2008). Asclepius was held to have healed or advised his patients by visiting them in a dream as they slept in a particular part of the sanctuary. Inscriptions that record the work of Asclepius suggest that his cures were miraculous. However, it is interesting to note that he is described as 60 performing them in ways that often resembled the work of human physicians (Edelstein and Edelstein 1945; LiDonnici 1995). Temple medicine had certain advantages. One of those was the cost. Individuals requesting the services of the god typically needed to make an offering, but this could be as little as whatever the poorest visitor could spare. Another benefit of the Asclepius cult was the possibility of receiving treatment for chronic illnesses for which Hippocratic physicians were reluctant to provide care (Wickkiser 2008). On the other hand, visiting a sanctuary of Asclepius could require one to travel great distances, especially in Late Antiquity when fewer temples were functioning. For many, including the very ill, visiting the site of a healing cult may not have been a reasonable option. Doctors, like healing sanctuaries dedicated to Asclepius, continued to function in the traditional ways in Late Antiquity. Public physicians were appointed in towns and cities by the nd 2 th th century AD and their status perhaps even rose in the 5 -6 centuries (Nutton 1984; Rosen 1958). Doctors of the Hippocratic tradition working during the Roman period had already expanded into a number of diverse specializations such as surgery and ophthalmology (Jackson 1988). As in the church, women were not given the full range of opportunities as men in medicine. They could not receive formal training, but they commonly acted as midwives and worked alongside doctors in that capacity (Clark 1993). While doctors were probably widely available, they could be expensive. Furthermore, there was a considerable range of variation in the training and expertise of individuals who claimed to be physicians. Nutton (1984) and Horden (2008) point out that ecclesiastical texts often express disdain for doctors and reference their ineptitude. 61 The dialectical relationship that developed between Christianity and other forms of medical care in Late Antiquity is noteworthy and created a lasting association. Nutton (1984) and Avalos (1999) have drawn attention to the fact that the conception of Jesus as a healer has not received due attention in discussions of Christianity’s success. Christianity contained its own healing system, which provided a number of advantages, mostly related to access, over GrecoRoman medicine. The emphasis on faith, invoking the name of Jesus, and laying on of hands presented the sick with a system that avoided complex rituals and eliminated issues such as extended travel and expense. Interestingly, like the cult of Asclepius, Christianity was also able to draw upon a new demographic group created by the exclusion of the chronically ill from other healing services. The Christian emphasis on charity also appears to have been responsible for the development of the hospital as a widely available health care institution. The origins of hospitals have been traced by Crislip (2005) to the surprisingly comprehensive and efficient health care th systems that appeared in early Christian monasteries by the 4 century AD. Monasteries contained multiple components recognizable in modern health care such as infirmaries or th inpatient facilities and professional staffs of doctors and nurses. In the late 4 century, prominent individuals such as Basil of Caesarea and John Chrysostom began the practice of extending the monastic health care system to the general population in what are viewed by some as the first true hospitals (Crislip 2005; Miller 1984). Despite these positive developments, the possible effects of the collapse of earlier Roman infrastructure on health deserve some mention. Roman cities contained a number of institutions and services carefully designed to promote public health including aqueducts, baths, public latrines, and sewers (Rosen 1958; Scarborough 1981). As discussed previously, those institutions 62 were often neglected at the expense of ecclesiastical architecture in late antique urban areas, especially in the Balkans where cities also contracted into smaller, more fortified sites. The decline or absence of public health infrastructure, when combined with population aggregation in urban communities, was likely a frequent cause of problems related to sanitation, access to clean water, and disease transmission (Manchester 1992). Based on paleopathological data, there is good evidence that this scenario occurred at places like Butrint and the suburban Vrina Plain, th th Albania around the 6 -7 centuries AD (Beatrice et al. 2009, forthcoming; Fenton et al. forthcoming). The extent to which rural areas would have been affected by this process is unclear. The Early Christian Community at Nemea The conditions that allowed for continued prosperity in the Greek countryside during Late Antiquity led to the occupation of the site of ancient Nemea by a farming community during th th the 5 -6 centuries AD (Wright et al. 1990). Archaeological evidence for sustained activity during this time comes mainly from modifications made to the Sanctuary of Zeus. While modest in scale, the Early Christian community seems to have thrived until the site was abandoned th abruptly in the last quarter of the 6 century. Consistent with common alterations to classical cities and towns associated with Christianization, two main basilicas were constructed at Nemea. Only limited excavation has been carried out in the area of the basilica located atop the Evangelistria Hill southeast of the Sanctuary of Zeus. Landon (1990) points out that this structure is similar in both design and scale to the valley basilica, which has received much greater attention. The valley basilica was erected th on top of the remains of the Xenon, a 4 century BC hotel that probably housed the athletes who 63 competed in the Nemean games (Kraynak 1990). The plan of this basilica is typical for Early Christian churches in the region and features two aisles, a central nave with an apse at the east end, and a narthex at the west (Landon 1990). It was constructed primarily with building materials reused from earlier sanctuary buildings including the Temple of Zeus. A baptistery and a structure identified as an Early Christian housing complex are associated with the basilica, the latter perhaps serving at least in part as the residence for the clergy (Miller 1983; Miller 1988). The valley basilica was likely the religious focal point for the community, as it would have dominated the modified landscape of the former sanctuary (Landon 1990). Other evidence for activity at Nemea sheds additional light on the nature of the Early Christian community. Numerous farming trenches have been discovered across the sanctuary as well as in surrounding areas of the Nemea Valley (Landon 1990; Miller 1975, 1977, 1978, 1979, 1980, 1981, 1982; Miller 1983, 1984). These trenches typically appear in parallel rows and in many cases their construction caused considerable damage to underlying Classical and Hellenistic structures. The distribution of these farming trenches suggests that agricultural th th activity during the 5 -6 century AD must have been extensive. In fact, most of the areas within the sanctuary that have been found to be devoid of agricultural activity were later discovered to contain Early Christian burials (Landon 1990; Miller 1977). Among the crops typically cultivated during the Late Roman and Byzantine periods, grain, beans, grapes, and olives were the most important (Bourbou et al. 2011; Kazhdan 1997; Rautman 2006; Stathakopoulos 2007a). The excavation reports do not speculate on exactly what was grown at Early Christian Nemea, but it was likely some combination of those foods. A dependence on grain and terrestrial animal products can be inferred from the results of GarvieLok (2001), who analyzed carbon and nitrogen isotope ratios in bone samples from 29 64 individuals from Early Christian graves. Archaeological data from regional sites corroborates the historical and isotopic evidence. The analysis of architectural and botanical remains from a late antique farmstead at Pyrgouthi in the Berbati Valley to the southeast of Nemea indicates the cultivation of a few varieties of grain and legumes, along with the production of wine and perhaps olive oil. Additionally, there is evidence of meat consumption. Faunal remains, some exhibiting cutmarks, show that sheep and goats were the most common domesticated animals at the site, followed by pigs and cattle (Hjohlman 2002). The results of the Nemea Valley Archaeological Project, which synthesize data from previous excavations, geological studies, and more recent archaeological surveys, suggest that the Nemea region was characterized by population growth and stability in Late Antiquity. As discussed in Chapter One, Wright et al. (1990) have demonstrated that draining the valley properly is crucial to the success of agriculture and probably also for habitation generally. Any project involving the clearing of natural channels and the creation of a suitable irrigation system would have required a substantial and sustainable workforce. While it is difficult to estimate the size of the Early Christian community with precision, the construction of substantial ecclesiastical structures and the scale of the agricultural activity at Nemea are consistent with a relatively large group of permanent residents. Given the modest size of Nemea and its location in an upland valley, it could be argued that the Early Christian community was marginal and isolated. On the other hand, Sanders and Whitbread (1990) demonstrate that Nemea was located along a network of reasonably well connected towns in antiquity. The Abandonment of Early Christian Nemea Bearing in mind the results of regional archaeological surface surveys and the evidence for agricultural and religious activity at Nemea, there is good reason to believe that the instability 65 which characterized much of the Balkans and the West during Late Antiquity was slow to reach rural communities in southern Greece. The evidence for the abandonment of Nemea during the th 6 century AD adds to that assertion. According to the numismatic evidence, the Early Christian th community was most prosperous during the second half of the 6 century (Landon 1990). In addition, construction projects at the site were ongoing and included the multi-roomed domestic structure located southeast of the Hellenistic Bath described above (Miller 1983). Activity at Nemea ends abruptly, however. The life of the Early Christian housing complex seems to have been quite short. Miller (1983) dates the construction, use, and demise of th the structure to within the second half of the 6 century, based on coins of Justin II. Its domestic contents along with the coins were left in place, suggesting that abandonment occurred rapidly. Archaeological evidence of additional activity in the Sanctuary of Zeus is lacking until around th the 12 century AD. While there is ceramic evidence for continued activity in the valleys around th Nemea in the early to mid-7 century (Christian Cloke, personal communication 2012), the surface surveys carried out by the Nemea Valley Archaeological Project generally produced little th th material that could be reliably dated to between the 7 and 9 centuries AD (Athanassopoulos 1993). These results are suggestive of a lengthy period of relative inactivity in the region. The traditional explanation for the abandonment of Early Christian Nemea is the Slavic invasion of the Peloponnese in the 580s (Landon 1990; Miller 1983). Evidence of violent destruction at a small settlement south of the sanctuary at Nemea as well as at other sites in the region such as Corinth and Argos has been attributed to Slavic invaders (Landon 1990; Miller 1976). Furthermore, possible Slavic settlements have been identified in the region, for example 66 at Isthmia (Gregory 1993) and Olympia (Völling 2001). Additional evidence of a violent end to the Early Christian settlement at Nemea comes from the tunnel of the Hellenistic Stadium. The skeletal remains of a male individual discovered within the stadium tunnel exhibit a defect on the left parietal bone consistent with an injury caused by a sharp-edged weapon. The fact that this individual did not receive a proper Christian burial, in addition to the discovery nearby of animal bones, cooking vessels, and a coin hoard, suggested to the excavators that the tunnel may have served as a place of refuge during a period of crisis (Garvie-Lok 2010; Miller 1979, 1980). Interestingly, it seems likely that there was no perceived threat of invasion among the Early Christian community at Nemea as there is no evidence of the fortification of the settlement. In addition, as discussed previously, linking evidence of destruction and abandonment to processes described in historical narratives is problematic. Especially in light of the debate over the Slavic presence in southern Greece, additional explanations should be sought for the rapid abandonment of Early Christian Nemea. One such explanation is the so called “Justinianic Plague” that swept through much of the Byzantine world beginning in 541 AD. Outbreaks of plague were numerous and widespread in the Mediterranean and, according to the written sources, continued in waves probably into the th mid-8 century (Stathakopoulos 2007b). Nearby Corinth was supposedly hard hit and, according th 9 to Procopius, lost around half of its population in the mid-6 century. The impact of plague was felt in the countryside as well as in more densely populated areas. Population losses were apparently great enough that it was necessary for the state to resettle groups of peasants in highly affected areas such as Thrace and Constantinople (Little 2007). 9 Procopius, Secret History 18.148-160. 67 th The presence of plague at Nemea during the middle or latter half of the 6 century is difficult to prove, however. If the causative agent of the Justinianic Plague was Yersinia pestis as a number of recent scholars hypothesize, then no specific indicators would be present on the skeletal remains of those affected. 10 Furthermore, the archaeological evidence at Nemea is inconsistent with the death of a large portion of the Early Christian community occurring around the same time. In that case, one would expect to find “plague pits,” containing groups of individuals buried together, perhaps haphazardly, in a short period of time (Antoine 2008; McCormick 2007). Instead, the Early Christian burials at Nemea are in large part single inhumations that display a high degree of consistency and organization. These burials are introduced in detail in Chapter Six. Greece and Byzantium in the Middle to Late Byzantine Period Prosperity under the Komnenoi th A period of revitalization in Byzantium following the so-called “Dark Ages” of the 7 th 9 centuries resulted in a degree of growth and prosperity in the Greek countryside and the repopulation of the Nemea Valley after centuries of minimal activity (Bintliff 1996). The th th renewed strength of the Byzantine Empire in the late 11 and 12 centuries was due in large part to the policies of emperor Alexios Komnenos (AD 1081-1118) and his immediate successors, who were part of a new trend in which powerful administrative positions were held by aristocratic families (Cameron 2006). In addition to reviving a weakened military, Alexios successfully navigated increased interaction with the West and restructured the economic base of 10 th See for example the molecular identification of Y. pestis in a 6 century double burial from southern Germany by Wiechmann and Grupe (2005) and the synthesis of the written, DNA, and epidemiological evidence by Sallares (2007). 68 th the Empire (Gregory 2010). The occupation of Nemea again in the 12 century was almost certainly an effect of the population growth and reuse of previously abandoned land that characterized the Greek provinces during this period (Herrin 1985; Kazhdan and Epstein 1985). A more active countryside was accompanied by expansion in urban areas. Gregory (2010) points out that the sites of classical cities were frequently rebuilt or at least reconfigured to feature churches and relocated city centers. This can be seen at Corinth, where new construction in the th area of the forum around the end of the 11 century coincides with increased local production and greater distribution of higher quality pottery (Sanders 2002, 2003). Factors Precipitating Fragmentation th th The 11 -12 centuries saw a growing interest in the east by western powers. On the one hand, this was a positive development, as increased trade with Italian merchants at urban centers such as Corinth was one outcome. The development of opportunities in adjacent rural areas like Nemea was probably an indirect result of this renewed economic activity (Athanassopoulos 1993, 2010). On the other hand, western ambitions in some cases extended beyond trade. Greece was periodically invaded by the Normans, who sacked Corinth and Thebes in 1147 AD and Thessalonica in 1185 (Herrin 1985). A weakening of the Byzantine Empire’s defenses, precipitated by a gradual breakdown in the relationship between the central government and th provincial officials near the end of the 12 century, was partly to blame (Herrin 1975, 1985). As the provincial government became increasingly disconnected from Constantinople, various local officials competed with one another by collecting their own taxes from the peasantry (Herrin 1975). At the same time, the wealth of Constantinople grew at the expense of the provinces as the central administration was intent on maximizing the productivity of the 69 countryside and acquiring its products (Herrin 1985; Lock 1995). Also, as early as the 10 th century, there was an increase in the number of powerful landowners who sought to obtain available smaller holdings, forcing the owners of those properties to become paroikoi or dependent peasants (Lefort 1993; Oikonomides 1996). Peasants were further exploited for their land, which could be taken as soon as farmers were sent away on forced military duty. Some local landowners, increasingly frustrated by the deteriorating conditions in provincial areas, led resistance movements and attempted to establish independent territories (Herrin 1975, 1985). Leon Sgouros, for example, infamously recruited his own army and conquered territory in th central Greece and the Peloponnese in the early 13 century (Gregory 2010; Herrin 1975). The invasions by western powers and the changes in patterns of land ownership described above seem to have been precursors to the profound alteration of Byzantine society that occurred th at the beginning of the 13 century. By the time of the Crusades, the relationship between Constantinople and the West had already grown tenuous as a result of fundamental differences in outlook between eastern and western Christianities (Gregory 2010). Byzantine policy toward the Crusaders did little to diffuse those tensions. While Constantinople supported western troops on their journey to the Holy Land, it was sometimes suspicious of their intentions. Emperors also used crusading armies to achieve their own military goals (Herrin 1985). During the Fourth Crusade mutual mistrust, a weakened Byzantine state, and a desire for wealth by the West were ignited by broken promises and internal strife in Constantinople upon the arrival of the Crusaders (Cameron 2006; Gregory 2010). 70 The Fourth Crusade and the Conquest of the Peloponnese In April of 1204 AD the Crusaders sacked Constantinople and instituted a period of Latin rule in the east. The territory of the Byzantine Empire was divided into Frankish states, minor lordships, and Italian colonies (Reinert 2002). Byzantine resistance to Latin occupation was confined to separate successor states such as Epirus in northwest Greece and Nicaea in Anatolia (Cameron 2006; Reinert 2002). Conquest of the Peloponnese began a short time later as an army of Frankish knights passed through central Greece and laid siege to Corinth and Nauplion. Greek resistance, led by Leon Sgouros, held out at the fortress of Acrocorinth until 1210 AD and slightly longer at other strongholds in the northeastern Peloponnese such as Nauplion and Argos (Gregory 2010; Lock 11 1995). From this position, the Franks proceeded gradually to control territory in southern Greece. While they encountered few additional episodes of significant local resistance, conquest of the entire Peloponnese was not complete until 1249 AD. Lock (1995) points out that the Frankish holdings remained somewhat insecure and that the Frankish period in Greece was characterized by violence and instability. Conflicts with Byzantine successor states who aimed to re-conquer the Peloponnese continued into the later th 13 century (Jacoby 2008; Lock 1995; Runciman 1980). The Byzantines under Michael VIII Palaiologos were finally successful in reestablishing themselves in southern Greece in 1262, after the Frankish knights ceded Monemvasia, Maina, and Mistra back to Byzantine rulers (Lock 1995). However, raids and sporadic fighting between the Franks and local Greeks intensified and became more widespread, resulting in what Lock (1995:84) describes as “daily hostilities, the destruction of towns, the depopulation of the countryside, the death of countless men cut down in 11 Sgouros himself died in either 1208 or 1209 (Gregory 2010). 71 their prime and the compulsive search for plunder.” The inability of the Frankish rulers to maintain control over the Peloponnese necessitated an appeal for western aid and, by the 1270s, the Angevins under Charles of Anjou were in possession of the Frankish Morea (Lock 1995). Thus began a centuries-long period during which the French, Aragonese, and Italians competed for power and influence in southern and central Greece. Frankish Greece The establishment of Frankish Greece was part of the larger pattern of fragmentation of Byzantine territory that altered the political, economic, and demographic trends of the Aegean in th the 13 century (Jacoby 2008). Under the Latin Empire, territory in Greece was divided into fiefdoms and bishoprics (Gregory 2010; Ilieva 1991; Williams 2003). The Peloponnese became the principality of Achaia and a western-style feudal system was established in the province by Frankish lords accustomed to a rigid social hierarchy (Jacoby 1973, 1986, 1989a, 2008). Powerful Greek landowners were gradually incorporated into the class of feudatories (Jacoby 1989a). Life for the average rural peasant is usually assumed not to have undergone substantial changes, aside from the destination of their taxes (Kardulias 1997). However, there are reasons to believe that the western system imposed on Greek peasants did in fact have a negative impact on their well being. In contrast to Late Antiquity when peasants were largely independent, the stratified society of the Frankish lords made peasants’ position at the bottom of the social hierarchy both legal and hereditary (Jacoby 1973, 1989a, 2008; Lock 1995; Rautman 2006). In addition to losing their free status, the burden of taxes on peasants, which consisted of the products of their agricultural labor, was severe. Runciman (1980:26) suggests that peasants were required to give everything beyond what was necessary for “bare subsistence” to their lords. Peasants also lost their legal power and could no longer 72 direct a complaint against a lord. When abuses did take place, the courts apparently looked the other way. Jacoby (1989b) describes a case in which a landholder avoided any formal prosecution after he accidentally killed a paroikos for refusing to obey him. In a number of ways, then, the relationship between peasant and landlord deteriorated under Frankish rule when compared to the pronoia landholding system that developed during the Middle Byzantine period (Ilieva 1991; Jacoby 1989a). Archaeological evidence from well-excavated urban centers like Corinth suggests that the long period of conflict and social transformation outlined above negatively affected living conditions in Frankish Greece. As previously discussed, Corinth appears to have enjoyed th economic prosperity during the 12 century, perhaps even following the attack by the Normans in 1147 AD (Sanders 2002). However, Williams (2003) reports that the first half of the 13 th century witnessed contraction, a decline in living standards, and a reduction in the size of the city’s population. Examination of human skeletal material from Frankish Corinth reveals evidence of violence and poor health, indicating that urban life was precarious (Barnes 2003; th Snyder and Williams 1997). It is only in the later 13 century, when the Franks had been long established in the northeastern Peloponnese, that conditions in the city appear to have improved (Williams 2003). Less archaeological evidence is available that might add to the historical picture of conditions in the countryside of Frankish Greece. Evidence from archaeological field surveys in central Greece points to the proliferation of both nucleated communities and small rural sites (Bintliff 1996). However, the opposite pattern is apparent in the northeastern Peloponnese. Athanassopoulos (2010), for example, has recently demonstrated that the conditions around 73 th Nemea deteriorated in the late 13 century to the extent that the inhabitants abandoned the valley in favor of the fortified hilltop of Polyphengi. Social and Religious Change in the Middle to Late Byzantine Period Religion The events of the Fourth Crusade were, in part, the culmination of religious tension between the East and West that characterized the Middle Byzantine period. Disagreements between the Orthodox Byzantines and the Latin Catholics over a number of aspects of Christian th doctrine had become vocal by the 9 century AD (Krueger 2006). In the East, religious practice took many forms including services that involved community participation and acts that could be more personal such as visiting healing shrines possessing saints’ relics and the use of icons or pendants (Gerstel and Talbot 2006). Monasticism continued to be prominent and its influence on lay observance may have actually increased from Late Antiquity. Krueger (2006) points out that the distinction in the Early Byzantine period between monastic and parish liturgies broke down during the Middle Byzantine period. As a result, Christians were encouraged to participate in ascetic forms of religious behavior such as personal devotion and fasting. The shift from public forms of worship to those that were more private was an important distinguishing feature of Byzantine Christianity in the later centuries of the Empire. In terms of the liturgy, large outdoor processions gave way to indoor services characterized by less participation by the laity (Krueger 2006). Greater emphasis was placed on the personal connection between an individual and religious imagery (Ousterhout 2008). This social transition is also reflected in the architecture of the period. Large basilicas were replaced by smaller 74 churches, many of which were used as private family chapels (Herrin 1985; Ousterhout 2008). 12 Additionally, internal design changes such as the development of the iconostasis created a greater physical separation between the clergy and the laity (Gerstel and Talbot 2006; Krueger 2006). The establishment of the Latin Church in the lands conquered during the Fourth Crusade did not result in the disappearance of Byzantine Christianity. Rather the two traditions seem to have managed an uneasy coexistence. In the Morea, Greeks supposedly accepted Frankish rule on certain conditions, one of which was that they were able to keep their faith (Angold 1989). The presence of the Catholic Church there was established quickly as representatives were present in the armies of the Frankish knights that conquered the Peloponnese. Later, Catholic bishops were present in a number of towns including Patras, Corinth, and Argos. However, at least some Greek priests remained in the villages of the Morea, although they may have been treated poorly and, at least initially, subjected to high taxes (Ilieva 1991). Richard (1989) demonstrates that even the control of Greek monasteries by the Latin Church was not always based on the idea of the suppression of those institutions. Greek monks were not necessarily forced to leave and an Orthodox lifestyle might continue in those instances. In the less accessible mountainous or rural regions, Orthodox practices probably continued uninterrupted (Ilieva 1991). In her study of the interaction between the Franks and local Greeks in the Peloponnese, Ilieva (1991:246) is more comfortable characterizing the religious climate as one of “contact” and “opposition” rather than “symbiosis.” 12 Small family chapels were also used for the burial of extended family during the Middle and Late Byzantine periods (Gerstel and Talbot 2006; Kazhdan and Epstein 1985). 75 Gender and the Family The increasing popularity of private forms of religious devotion in the Middle Byzantine period had corresponding developments in the conception of the Byzantine family. By the 9 th century AD, participation in social life had become narrower and greater emphasis was being placed on the nuclear family. Byzantine society had always been patriarchal, but the activity of women in public life that was characteristic of Late Antiquity declined sharply during this period and in its place were greater expectations within the family (Kazhdan and Epstein 1985). A high priority was placed on child-care and Talbot (1997) points out that bearing children was viewed as the most important role of women. The relative seclusion of women was suggested also to avoid attracting the attention of unrelated men (Kazhdan and Epstein 1985; Talbot 1985). Leaving the home even for the purpose of public religious worship may have been frowned upon by some (Talbot 2006). Again, it should be noted that conformity to these ideas was probably greatest among the elite women of Byzantium. The extent to which they were observed by men and women of modest means in the countryside is unclear. It is unlikely, however, that women in rural areas and agricultural communities were confined to the household (Cameron 2006). Peasant women probably played significant roles in activities such as maintaining gardens and vineyards and feeding animals. They might also have assisted with grain harvests during periods of warfare or other calamities (Talbot 1997). Jacoby (1989a:207) suggests that the stratified society established in the Frankish Morea, “ensured social segregation” between the Greeks and the Latin elite. While this might be overstating the case, mixed marriages were certainly discouraged and the illegitimate children of Latin fathers and Greek mothers could not become members of the feudal class. If western social 76 customs did influence gender relations in the Morea, the effect was not likely to have been positive for women given that traditional Frankish society was male-dominated. Laiouth Thomadakis’ (1977) study of peasant society in the 14 -century Macedonian countryside shows that the social unit normally consisted of a patriarchal nuclear family. Men were generally regarded as the heads of households and were also typically the property owners and tax payers. Perhaps most telling, registers of households indicate that the sex ratio of males to females was high. While an overrepresentation of males in these data could be the result of bias in the registration process, it might also be attributed to greater female mortality resulting from the preferential treatment of males. Other researchers (e.g., Talbot 1997) have noted that childhood was more hazardous for Byzantine girls as a result of factors such as unequal attention and an earlier weaning age. Medicine and Public Health Byzantine medicine in general remained highly influenced by the work of Galen of Pergamon, but contemporary medical authors have been criticized for their emphasis on summarizing his work and for their lack of fresh ideas and original research (Nutton 1984). Galen’s brand of preventative medicine is apparent in the large number of guidelines related to health and hygiene produced in both the East and West during the Middle Ages (Rosen 1958). While texts such as the Regimina Sanitatis contained a number of ideas that remain accepted (e.g., moderation in food and drink, regular exercise), they were aimed mostly at elites (Sotres 1998). For most people in the Byzantine Empire, the need for medical attention was probably met first at the household level. Simple remedies that involved the use of medicinal plants or amulets were widely known even in the countryside and did not require trained specialists (Rautman 2006). 77 The attitude of the public toward doctors during the Middle Byzantine period varies depending on the source and also changes through time. As in Late Antiquity, hagiographic texts are highly critical of doctors and often contrast their greed and lack of skill to the miraculous th cures of saints. On the other hand, medical practice in the 12 century became the subject of satires, many of which describe doctors in a more positive light, remarking, for example, on their surgical abilities (Kazhdan and Epstein 1985). Hospitals, which had their origins in monasteries and which opened their doors to the general public during Late Antiquity, as was discussed above, developed into prominent th institutions during the Middle Byzantine period. As early as the 7 century AD, hospitals were no longer exclusively charitable establishments. Miller (1984) demonstrates that patients included wealthier members of society in addition to the poor and the homeless. The range of medical specialists employed in hospitals also increased during this period. The role of nursing was professionalized and portions of hospitals were dedicated to certain types of ailments and forms of treatment such as surgery. Hospitals were typically associated with cities, and large th population centers such as Constantinople and Antioch had more than one. By the 12 century, Byzantine hospitals contained an experience-based hierarchy of physicians and an administrative staff that was chosen from those among that group (Miller 1984, 1997). Despite these developments, the impact of hospitals on the health care of the average member of Middle to Late Byzantine society is not exactly clear. Some researchers (e.g., Horden 2008) have questioned whether the quality of hospital care was actually much different from what was available outside of hospitals. If hospitals did provide a superior level of medical treatment, one might still wonder about their accessibility. As an urban phenomenon, how often 78 were they visited by people living in the countryside? Furthermore, it seems likely that the events of the Fourth Crusade brought about a decline in the presence of hospitals. Comparable th institutions did not exist in the West until the 13 century and it may be telling that it is difficult th to find records of hospitals outside of Constantinople in the 14 century. Even Mistra, a major center of Byzantine culture in the post-Frankish Peloponnese, apparently did not have one (Miller 1984). It is important to bear in mind that the health of any community is dependent first on a reliable subsistence base and second on institutions such as hospitals or doctors. A number of factors, both natural (e.g., climate shifts, insects and other pests) and anthropogenic (e.g., political instability, economic crises), affected the production and distribution of food in Byzantium (Rautman 2006). Stathakopoulos’ (2004) extensive compilation of references to outbreaks of famine and plague in the Byzantine world demonstrates the consequences of such problems and attests to the importance of proper food storage during times of plenty. People living in the countryside might keep an entire year’s harvest in large storage jars or pithoi as a buffer against misfortune (Grünbart 2007). Prolonged storage is also problematic, however, as it may deplete food of certain nutrients (Mays 2008). Food production and storage have important implications for the Middle to Late Byzantine community at Nemea. Like its Early Christian counterpart, it was a rural community whose livelihood was based on intensive agriculture. The Middle to Late Byzantine Community at Nemea 79 Based on the evidence in and around the Sanctuary of Zeus at Nemea, there were few differences 13 in the activities of the Early Christian and Middle to Late Byzantine communities. th The remains th of farming trenches and irrigation ditches dating to the 12 -13 centuries AD closely resemble those of the Early Christian community (Landon 1990; Miller 1975). The similarities in the impressions left on the landscape of Nemea by these two communities should be of no surprise as agricultural practices probably changed little during the course of the Byzantine Empire (Athanassopoulos 1993; Rautman 2006). Once again, there is little data available in the excavation reports to provide a clear indication of the types of produce cultivated at Nemea during this period. However, Athanassopoulos (2010) suggests that gardens and probably vineyards were in use. Richard (1985) adds that wine grapes prospered in the Morea, and that wine and olives were commonly produced. Isotopic data from bone samples that may be used to reconstruct dietary composition is available for 25 Middle to Late Byzantine individuals from Nemea studied by Garvie-Lok (2001). Her data show that the diet of the later community at Nemea was nearly identical to that of the Early Christian community, which focused on grain along with some animal products. In addition to the evidence of agriculture in the sanctuary, domestic activity is suggested th by a two-room structure with associated cisterns dated to around the 13 century (Miller 1977). All other evidence of activity within the sanctuary during the Middle to Late Byzantine period is associated with the Early Christian basilica, the use of which differed markedly from Late Antiquity. At least portions of that structure had probably fallen out of use by this time, evidenced by a number of graves cut into its floor and walls (Frey 1998; Miller 1988). However, 13 The distinct differences that exist are related to the patterns of burial during this later occupation. These will be treated separately in Chapters Five, Six, and Eight. 80 the presence of graves and the construction of a small chapel adjacent to or within one of the rooms of the basilica demonstrate that it retained some religious significance (Cousin and Dürrbach 1885; Landon 1990; Miller 1988). Burials continued to be placed within or adjacent to the remains of the basilica through the Late Byzantine period (Miller 1988). Archaeological surveys in the Nemea Valley have recently produced a clearer picture of regional settlement patterns during the Middle to Late Byzantine periods. Based on concentrations of Byzantine pottery, Athanassopoulos (1997, 2010) has identified two relatively th th large settlements dating to the 12 -13 centuries AD in the area to the east of Nemea. Site 600, just to the east of the Sanctuary of Zeus, is the likely location of the hamlet whose inhabitants farmed the land and buried their dead at Nemea in this period (Figure 3.1). The identification in the region of a number of smaller sites that probably represent th th farmsteads suggests that agricultural activity was expanding during the 12 -13 centuries. Athanassopoulos (2010) also points out that the presence of high-quality glazed pottery attests to a growth in the disposable income of peasants at this time. Once again, however, the settlement th pattern in the Nemea Valley changes. By the late 13 century, almost certainly as a result of the need for greater security during a tumultuous period in the Peloponnese, dispersed activity in the valley ceased and the fortified village on the hilltop of Polyphengi became the center of habitation (Athanassopoulos 1997, 2010). 81 Figure 3.1: Map Showing Medieval Sites from the Nemea Valley Survey (Athanassopoulos [2010: Figure 1, Rosemary Robertson]). Reprinted with permission without modification or alteration (see Appendix C, p. 372). 82 Summary This chapter outlines in detail the historical and archaeological contexts of the Early Christian and Middle to Late Byzantine settlements at Nemea. For each period, I have discussed developments in religious practice, conceptions of gender, and health care that may have shaped daily life in each community. I have also outlined some of the major sociopolitical transformations and events that may have had an impact on each community. Finally, I have provided in-depth discussions of the archaeology of each community at Nemea, with an emphasis on demonstrating what the observed patterns tell us about how those communities fit into the broader transformative processes of their respective time periods. I argue that the Early Christian settlement at Nemea may have enjoyed a greater degree of prosperity than is usually assumed based on a revised understanding of historical circumstances traditionally viewed as th devastating, such as the Slavic “invasion” of the 6 century AD. The events associated with the Fourth Crusade and the Frankish conquest of southern Greece are more likely to have had negative consequences for the health of the Middle to Late Byzantine community. 83 CHAPTER 4: MORTUARY PRACTICE IN BYZANTIUM As discussed in Chapter Two, the study of mortuary practice as reflected in the archaeological analysis of burials may be informative about aspects of sociocultural practice such as religious beliefs, status differentiation, family structure, and gender relationships. While the study of burial practices in the Roman Empire is well developed, comparatively little attention has been paid to burial in the Byzantine world. Furthermore, much of the scholarship on the subject of mortuary practice in Byzantium has focused on funerary rituals. Fewer studies have closely examined the context of burial and the organization of Byzantine cemeteries. Ignoring those components of mortuary practice limits our ability to answer a number of related questions about burial in Byzantium. For example, who decided where people were buried? Were family plots typical? How common were single inhumations and multiple burials, and what was the significance of each type? Did a systematic practice of exhumation and grave reuse exist in Byzantium? What determined whether or not someone received secondary burial treatment? What did the location of burials symbolize? Did there exist a “hierarchy of sacred space” as some have argued (e.g., Ivison 1993)? Did the rules governing cemetery organization change over time? Particularly after 1204 AD, can western influence be detected in burial practices? Acknowledging that Byzantine burial practices were likely to have been context specific, this study is uniquely positioned to shed light on those questions as they pertain to the Greek countryside. In this chapter, I provide an overview of funerary ritual in Late Antiquity and the Byzantine period and discuss what is known about common aspects of burial practice in Byzantium. I then introduce the hypotheses about the association between burial location and 84 social status that are tested in this research using the archaeological evidence and skeletal samples from Nemea. Early Christian and Byzantine Funerary Ritual The evidence for Early Christian and Byzantine funerary ritual comes mainly from sources such as sermons and hagiographic texts. Rebillard (2009) has pointed out that it is difficult to construct a late antique Christian death ritual based the limitations of those sources, which offer only snapshots of practices that were locally variable. It is nevertheless useful to briefly examine the aspects of Christian funerals that appear in the written sources to better understand the outlook and worldview of early Christians and the ways in which those views changed during the Byzantine period. Furthermore, it is the archaeological evidence of funerary rituals in general that provides another avenue for the investigation of sociopolitical change in past societies (Morris 1992). Certain aspects of Christian funerary rites in Late Antiquity were not dissimilar from those of Roman funerals. While Rebillard (2009) has shown that the Church was not as responsible for control over funerary rituals as previously assumed, it should probably be taken as significant that the early Church broke with tradition gradually. For example, when certain pagan practices were rejected, Christian substitutions for those practices were often made (Rush 1941). Additionally, Alexiou (2002) reports that aspects of ancient ritual that survived in Late Antiquity were sometimes simply adopted by the Church, even in the case of rites that had been previously denounced. Based on the tension apparent in Christian and pagan dialogue with regard to funerary practice, however, Rebillard (2009) notes that Christian funerals must have had a distinctive character. Rush (1941) provides an overview of common features of Early Christian funerary ritual and his work is the basis of the following discussion. 85 Among the first acts performed after death were closing of the eyes and mouth and stretching out of the hands and feet, which were intended to give the deceased the appearance of sleeping in addition to facilitating the soul’s departure from the body (Kyriakakis 1974; Rautman 2006; Rush 1941). Throughout the Byzantine period, these initial rites were normally carried out by close family members (Kyriakakis 1974). A final kiss was given, although this served to demonstrate the connection between the living and the dead among Early Christians and was not necessarily connected to the catching of the soul with the last breath in the pagan tradition (Toynbee 1971). Interestingly, the “kiss of peace” demonstrates that Christians did not have the same anxiety about contact with the corpse that pagans did. Unlike Roman funerals, which required purification rites of those who came into close proximity to the body (Jones 1987; Lindsay 2000), Christian practices dissociated a ritual or legal state of death-pollution from contact with the dead (Samellas 2002). Christianity retained a sense of the deceased person’s need for protection on the journey to the next life but, as Rush (1941) demonstrated, this was an instance where a pagan tradition was modified or replaced with something new. The Christian version of the viaticum, in which the Eucharist was received at the time of death, was substituted for the act of placing of a coin in the mouth for Charon’s fee. Conclamatio, the custom of calling out to and lamenting the deceased, was not formally adopted by Christianity. However, it is clear that some form of deliberate excess in the manifestation of grief similar to the pagan tradition continued to be practiced despite the vehement opposition of Church fathers such as John Chrysostom and Basil of Caesarea (Alexiou 2002; Rebillard 2009; Rush 1941). The care of the body after death was the responsibility of relatives or, in some cases, the local Church community (Rush 1941). As in the Roman tradition, the body was washed and 86 anointed. However, it must have been common among Christians to restrict the use of perfumes to funerals as texts produced by non-Christians deride that practice (Rebillard 2009). The body was usually dressed in a white winding-sheet or sometimes in the finest clothes owned by the deceased (Alexiou 2002; Constas 2006; Rush 1941). The hands and feet were bound and the former were tied to either the stomach or chest (Kyriakakis 1974). The use of simple clothing or a shroud of white linen was promoted by Church Fathers, who spoke out against competitive display among wealthier Christians that developed from the practice of clothing the dead in special garments (Kyriakakis 1974; Rush 1941). The early Church also rejected the pagan practice of placing a crown upon the head of the deceased because this act likened the dead to the gods and was viewed as a form of idolatry (Rebillard 2009; Rush 1941). Despite the pagan association, it was popular in Byzantium to crown the dead with a wreath of flowers (Kyriakakis 1974). Following the preparation of the body, Byzantine Christians held a wake or prothesis similar to that which was typical in Roman funerals (Alexiou 2002; Jones 1987; Kyriakakis 1974; Rush 1941; Toynbee 1971). While this might have involved the deceased lying-in-state at home, the wake could also be held at the grave in cases when burial happened quickly after th death. Rush (1941) points out that it was not until the 4 century AD, when Christianity was more established, that church services began to be associated with the death of its members and that wakes were held at the church. In the Byzantine period, the body might be taken to the church and displayed in the narthex or nave following the prothesis at the house of the deceased (Constas 2006; Gerstel and Talbot 2006). Christian funeral processions were ideally simpler and more subdued than those carried out in the Roman tradition. The singing of psalms was substituted for musicians and though 87 professional mourners and excessive displays of grief are attested, they were, in the preference of the Church, not part of this rite (Alexiou 2002; Constas 2006). The protests by Church Fathers of extreme mourning practices were based on the idea that they were not useful to the dead and on the revised understanding of death in Christianity, which was supposed to elicit feelings of hope rather than sorrow (Rebillard 2009). The former notion was apparently also behind criticism of Early Christians who continued to offer items such as food, clothing, and candles at the grave (Alexiou 2002). In the Byzantine East, funerary services or prayers to commemorate the dead were held on the third, ninth, and fortieth days after death. Those dates were adopted and modified from the Greco-Roman tradition (the thirtieth day was replaced by the fortieth), with justification coming from Christian theology (Alexiou 2002; Gerstel and Talbot 2006; Paxton 2008; Rebillard 2009; Velkovska 2001). Burial in Byzantium The significance of a Byzantine cemetery to its community has been underscored by Rautman (2006:164), who states that “as a ceremonial focus and literal repository of its past, the cemetery was an enduring part of any village’s identity.” The Byzantine period was characterized by a closer physical association between the living and the dead, as cemeteries developed alongside churches that were increasingly located within towns and cities. Despite the strength of the relationship between cemetery and community, the creation and organization of cemeteries in Byzantium is not well understood. This section outlines general features of burial practice in the Byzantine world and explores how evidence from Byzantine cemeteries might be used to infer aspects of community social structure. 88 Grave Construction, Arrangement of the Body, and Grave Contents Especially in the countryside, most Byzantine graves were simple pits that might be tilelined or stone-lined and were often covered with tiles or stone slabs (Constas 2006; Rautman 2006). As in the later Roman period, inhumation rather than cremation was the norm for Byzantine burials. It was standard practice for Christian graves to be oriented east-west with the head to the west, so that the deceased would face east where Christ would appear at the resurrection. Bodies were usually arranged extended in a supine position, with the arms crossed over the chest or abdomen. Ivison (1993) has argued that coffins were probably used infrequently in Byzantine burials before Latin rule in 1204 AD, except perhaps among the wealthy. The practice of propping up or protecting the head using a stone pillow and placing tiles or fieldstones along the sides of the skull discussed by Gerstel and Talbot (2006) as well as Ivison (1992, 1993) may also be an indication of western influence. Ivison (1993) has demonstrated that this type of attention to the head of the deceased does not appear in Byzantium th until the 13 century. In contrast to Roman burials, which typically contained a variety of grave goods (Toynbee 1971; Jones 1987), Christian burials in Byzantium normally contain few mortuary artifacts. Constas (2006) and Rautman (2006) suggest that objects such as cross-inscribed ceramics, flasks, and occasionally coins might be placed with the deceased at burial. Standard articles of clothing such as belt buckles and items of personal adornment such as earrings and finger rings are also frequently recorded (Ivison 1993, 1996). It is possible that children and adolescents were buried with items of personal adornment more frequently than adults. The concern of Byzantine parents for the well-being of their children extended to the funerary context and it has been argued that jewelry and toys appearing in the graves of children indicate affection 89 and perhaps attempts to offer them an additional degree of protection in death (Ivison 1993; Pitarakis 2009; Talbot 2009; Tritsaroli and Valentin 2008). Collective Burial and Secondary Burial Gerstel and Talbot (2006) point out that it is not uncommon for Byzantine graves to contain multiple individuals. In most cases, such graves probably represent multiple primary interments (Ivison 1993). Large burial vaults specifically constructed to accommodate multiple extended burials have been described by Ivison (1993, 1996) at a number of Byzantine sites including Corinth. Smaller graves were used in a similar way, for example at Polystylon-Abdera (Agelarakis 1997) and Panakton (Gerstel et al. 2003), although earlier interments needed to be moved aside to make room for each subsequent burial. While Constas (2006) notes that families were normally buried in the same cemetery, the extent to which multiple burials can be understood to contain family members is unclear. Talbot (2009:300) suggests that this was a common practice, stating that children might be buried in “family graves that held multiple burials.” The presence of children in graves containing multiple individuals may in fact be a good indication of family use as Gerstel and Talbot (2006) suggest that mothers and children were often buried together. In the case of multiple individuals included in a single grave or clusters of individuals buried near one another, one must also consider the possibility of a disruptive event such as an epidemic disease (Gerstel and Talbot 2006). For example, Barnes (1996, 2003) has related clusters of children’s burials at Frankish Corinth to a series of epidemics, possibly of malaria. A similar situation may have produced clusters of predominantly children’s burials at the site of Butrint, Albania (Beatrice et al. forthcoming). Secondary burial in ossuaries was also carried out during the Byzantine period and may have been particularly common among monastic communities and saints (Abrahamse 1984; 90 Constas 2006). A number of researchers have suggested that this practice extended to laypeople as well (Constas 2006; Ivison 1993; Rautman 2006; Talbot 2009). Where this occurred, it involved exhuming the remains of the deceased after a few years and reburying the bones in community or family ossuaries or burial chapels (Constas 2006; Rautman 2006). Talbot (2009) points out that Byzantine children were included in secondary burial practices as their remains have also been found in ossuaries. Ivison (1993) has argued that the popularity of secondary burial in Byzantium may be overestimated based on the tendency of archaeologists to misidentify sites of multiple primary burials as ossuaries. He demonstrates that true ossuaries can be identified based on their layout, which does not have to account for extended bodies, and on the pattern of disarticulation and sorting of the bones present. Empty or incomplete primary burials may also serve as indirect evidence of the practice of secondary burial. The Spatial Organization of Byzantine Cemeteries Burial Location and Religious Beliefs The Roman practice of burying the dead outside the boundaries of towns began to break down in Late Antiquity, facilitated by the growth in popularity of martyr cults and the movement th of relics from extramural to intramural churches (Harries 1992). By the end of the 4 century AD, graves were being placed within city walls (Davies 1999). Only a century later the public spaces of classical towns were being used for burial (Constas 2006; Ivison 1996). Early Christian burial sites also began to be closely associated with the basilicas and mortuary chapels that appeared in both East and West around this time (Davies 1999). By the Middle Byzantine period, burials were almost exclusively located either within or immediately outside of churches (Ivison 1996). The religious significance of basilicas was great enough that they retained their function 91 as places of burial in instances where they had been damaged or had otherwise fallen out of use (Caraher 2008, 2010). The extent to which Christian burial locations were kept spatially distinct from those of pagans has been the subject of debate. It seems likely that practices varied from region to region and perhaps even between local communities. For example, Davies (1999) notes that a clear th separation between pagan and Christian burials was maintained at the 4 century AD cemetery at Poundbury. Other reviews of mortuary behavior in Late Antiquity suggest that, in general, little effort was expended to separate burial space based on religious belief (Harries 1992; Johnson 1997). Rebillard (2003, 2009) has argued that a number of statements by early religious leaders that appear to confirm a Christian preference for spatial segregation have been misinterpreted. He suggests that the Church held no official position on the sharing of burial space and that decisions about this aspect of burial location were left largely to the families. This conclusion is supported by Johnson (1997), who adds that Christianization was a lengthy process. While it might have divided families along religious lines, it probably would not have broken the tradition of maintaining family tombs in those cases. From an archaeological perspective, classifying burials into categories of pagan or Christian is not always straightforward because religious identity is one of many identities that might be expressed through the act of burial (Jenny 2011). Furthermore, such a classification presupposes the existence of strict religious affiliations that in reality may have been much more flexible. Burial Location and the Social Hierarchy One of the most fascinating ideas about the organization of Byzantine burials is the suggestion that their arrangement reproduced or was at least related to the social hierarchy. At a 92 basic level, this goes to the question of who received burial within a church. Most of the historical and epigraphic evidence suggests that church burial was reserved for the clergy and perhaps wealthy donors, while the burials of most laypeople were located in cemeteries around the church (Davies 1999; Ivison 1993; Samellas 2002; Yasin 2005). Individuals given church burial might be provided with tombs sunk into the floor, most often in the narthex or in an associated burial chapel (Constas 2006; Gerstel and Talbot 2006; Teteriatnikov 1984). In other cases, individuals were placed in burial chambers or arcosolia, which were arched niches carved into the walls of the church or chapel (Constas 2006; Teteriatnikov 1984). A driving force behind the burial of privileged Christians within churches was the desire to receive burial in close proximity to the bodies or relics of saints (Harries 1992; Paxton 2005). It was believed that, through ad sanctos burial, the saint would intercede for the deceased and thus provide them with divine protection. Samellas (2002) has demonstrated that it was in fact beneficial to the Church to maintain control over relics because it allowed religious leaders to determine who had access to more sacred burial locations. Effros (1997) adds that the Church actively promoted the idea that ad sanctos burial offered not only protection for the soul, but also social prestige. Although the degree of control exerted by the Church over the placement of burials, especially those of the laity in Late Antiquity, has been questioned by Rebillard (2009), it seems likely that competition for burial space developed over time and that social distinctions were reflected in the spatial organization of Christian tombs and cemeteries. Paxton (2008:394) captures this argument in the following remarks: In both East and West, burials were placed according to a moral scale, with the holiest of the dead at the center and others around them in proportion to their own claims to holiness: saints, bishops, abbots, monks, and pious (and wealthy) lay men and women…. For the most part, the medieval economy of death reproduced the social hierarchies of the living. 93 Ivison (1993) has expanded upon and refined the idea that burial space within Byzantine churches was organized in a hierarchical fashion. Using archaeological evidence along with written sources, he showed that tendencies in the location of elite burials at both Constantinople and Mistra were also visible in the churches of a number of provincial cities. In the pattern he identified, burials of important individuals were located in the naos or nave. Those closer to the east end of the church carried greater distinction because of their proximity to the altar and holy relics. Priests, for example, might be buried in the bema, a space they typically occupied during church services. The place of greatest honor was on the south side of the naos, toward the east end. This designation drew upon a symbolic association between the right side of the basilica and the idea of the right side as “that of The Saved” (Ivison 1993:74). It is important to emphasize here that access to sacred burial space, which was highly limited, was not just important to the deceased. The prestige associated with a prominent burial location probably extended to the deceased’s living relatives and may have been a factor in the legitimization of power relationships in Byzantine communities (Ivison 1993). Although Ivison (1993) demonstrated that his model has wide applicability in Byzantium, the church burials he examined are from relatively large cities and towns. One of the goals of the present research is to test whether or not the associations between burial location and social status proposed by Ivison and others can be detected in the context of the Byzantine Greek countryside. Gerstel and Talbot’s (2006:97) assertion that “burial patterns in villages mirror those from urban contexts, though on a more modest scale” supports the application of Ivison’s model to places like Nemea. Furthermore, while Ivison (1993) focused on evidence from the Middle to Late Byzantine period, it is probable that social hierarchies were reflected in the spatial organization of Early Christian cemeteries as well. For example, Al-Shorman (2004) has 94 th shown that three distinct cemeteries surrounding a 6 -century AD church at Yasieleh in Jordan separated the congregation based on social rank. In this way, entire funerary landscapes created during the Byzantine period might be imbued with information about social distinctions. Summary This chapter has examined fundamental aspects of mortuary practice during Late Antiquity and the Byzantine period. I have shown that Early Christian and Byzantine funerary rituals borrowed elements from Roman practice, but often modified them to maintain consistency with Christian tradition. Despite the disapproval of religious leaders, funerals may have contained a mix of Christian and pagan themes and this has implications for what might be observable archaeologically in burials, especially from the Early Christian period. I have also reinforced the idea that Byzantine communities had powerful physical and symbolic connections with their cemeteries. This was expressed in the close association between cemeteries and community churches as well as through the organization of burial space both within and outside of churches. Archaeological and historical evidence suggests that status distinctions which existed during life were often maintained and delineated spatially in death. In Chapter Five, I outline the research questions and hypotheses used to test that assertion, as well as those used to investigate the impact of sociopolitical change on health in the Byzantine Greek countryside. 95 CHAPTER 5: RESEARCH QUESTIONS, HYPOTHESES, AND EXPECTATIONS In chapters three and four I examined the sociopolitical context of Early Christian and Byzantine Nemea, as well as more detailed aspects of cultural practices associated with those periods. I highlighted instances where our understanding of the impact of certain events and cultural transformations is incomplete and also posed a number of questions that might be explored using an approach that augments the available historical and archaeological evidence with data from human skeletal analysis. In this chapter, I lay out the primary research questions that are addressed in this dissertation. Investigating the Biological Impact of Sociopolitical Change As discussed in Chapter Two, anthropological models of stress (e.g., Goodman et al. 1984, 1988) demonstrate that changing cultural and environmental circumstances can exceed the limits of an individual’s biological resistance and result in physiological disruption. Especially in the case of chronic, unmitigated stress, physiological disruption is often reflected in the human skeleton as inhibited growth, pathological lesions, and elevated mortality. The need to adapt to changing conditions was common to all past populations. However, groups such as Greek th th peasants living in the Byzantine Peloponnese during 12 -13 centuries AD experienced changes in their physical and social environments that were particularly direct and dramatic. While the archaeological evidence discussed in Chapter Three indicates that the Early Christian community at Nemea was relatively prosperous just before the abandonment of the site in the late 6 th century, the Middle to Late Byzantine community existed during a period characterized by increasing central and local administrative instability, a violent invasion, and the imposition of a feudal system. The effects of those processes on aspects of living conditions such as workload 96 and access to dietary resources may be assessed through the paleopathological analysis of the extant human skeletal remains. This study is also concerned with the statuses and roles of men and women in the Byzantine Greek countryside and their implications for health. Finally, this study applies data on skeletal health to questions about Byzantine cemetery organization. The research questions investigated in this dissertation are as follows: th 1. Did the sociopolitical changes of the 12 -13 th centuries AD diminish quality of life among the Middle to Late Byzantine farming community at Nemea? 2. Are differential treatment and/or different spheres of activity for men and women in Late Antiquity and the Byzantine period evidenced by disparities in the skeletal health of males and females at Nemea? 3. Was the burial space at Nemea organized according to the social hierarchy? The following section provides testable hypotheses that are used to address each of these questions in turn. Research Hypotheses and Expectations Hypothesis 1: The Middle to Late Byzantine skeletal sample from Nemea will exhibit greater prevalence rates of physiological stress indicators than the Early Christian skeletal sample. The Middle to Late Byzantine community at Nemea faced three major problems unknown to the late antique community: (1) administrative instability that strained the provinces th in the later 12 century; (2) the Frankish conquest of the Peloponnese; and (3) a decline in the socio-legal status of the Greek peasantry following the introduction of a western feudal system by the Franks. 97 As Herrin (1975, 1985) has pointed out, the breakdown in the relationship between the th central and provincial administration of the Byzantine Empire in the late 12 century led to impoverishment and a degree of social upheaval in the countryside. Peasants would have been pressured to increase production in order to meet the demands of provincial officials collecting taxes. It is anticipated that workload associated with agricultural activities increased for both men and women. Greater prevalence rates of osteoarthritis and occupational trauma among the Middle to Late Byzantine skeletal sample would be consistent with that expectation. It is also anticipated that the quantity and quality of the Byzantine community’s diet declined as a greater proportion of its agricultural products needed to be sold, given to a landlord, or handed over in the payment of taxes. If this was the case, I expect to find greater prevalence rates of linear enamel hypoplasias, porotic hyperostosis, and cribra orbitalia in both children and adults in the Middle to Late Byzantine sample. Furthermore, the deterioration of social conditions evidenced by separatist movements in the provinces is expected to have contributed to the appearance of those indicators by reducing individual community members’ resistance to nutritional and disease-related stressors. Decreased resistance to infection as a result of living in an unstable social environment is also likely to have increased the prevalence of periosteal reactions among both adults and children at Middle to Late Byzantine Nemea. It is hypothesized that living conditions for the peasantry at Nemea deteriorated further in th the 13 century when Frankish conquerors installed a western-style feudal system along with a rigid social hierarchy onto a Byzantine provincial social structure that was already becoming more stratified. Feudalism in the Peloponnese preserved Byzantine taxes while legalizing peasants’ unfree status (Ilieva 1991; Jacoby 1973, 1989a, 2008). It also reduced the proportion of peasants’ agricultural products that were used to meet their own subsistence needs (Runciman 98 1980). Furthermore, the potential for violent conflict almost certainly increased during this period. The mistreatment of peasants by landholders and provincial officials even prior to Frankish conquest was common. For example, Kazhdan (1997) reports that an inability or refusal to pay one’s taxes frequently resulted in physical abuse. Conquest, however, brought with it periodic warfare throughout southern Greece as well as the potential for peasants to become caught between feuding lords attacking one another’s villages. It is expected, then, that trauma reflecting interpersonal violence will be more prevalent at Middle to Late Byzantine Nemea. Warfare also contributes to problems such as food crises resulting from the disruption of agricultural cycles, the destruction of crops, and the increased strain of providing soldiers with food (Stathakopoulos 2004). Hypothesis 2: Prevalence rates of physiological stress indicators associated with nutritional stress and infection will be greater among females than males in both phases at Nemea. Prevalence rates of activity related stress indicators will be greater among males in both phases. As discussed in Chapter Three, historical sources indicate that women were a marginalized group in both Late Antiquity and the Byzantine period. While there is some debate as to the extent to which contemporary descriptions reflect the reality of women’s experiences (Cameron 2006; Laiou 1981), their activities are usually suggested to have revolved around the household. Frequently cited examples include grinding grain, preparing meals, washing and making clothes, and gardening, in addition to child care (Talbot 1997). By most accounts, child bearing, which was considered to be the most important role of women, took a serious toll on women’s health (Connor 2004; Hill 1997; Talbot 1997). The emphasis placed on child bearing in Byzantine society is perhaps understandable given certain demographic realities. For example, 99 based on average birth rates, life expectancy, and rates of infant mortality during the Late Byzantine period, Laiou-Thomadakis (1977) has estimated that women would have needed to have six female children before it was likely that any of them would survive well into adulthood. However, having children was a dangerous process and health problems or even death resulting from complications of pregnancy and childbirth were not uncommon (Clark 1993). In Late Antiquity and the Byzantine period women were assumed to be inferior to men both physically and intellectually. Boys were given preferential treatment from an early age, which led to health disparities during development (Talbot 1997). For example, Talbot (1997) has suggested that weaning occurred earlier for girls and this may have contributed to decreased disease resistance and elevated mortality relative to male children. This discrepancy may have carried over into adult life and, along with the hazards of childbirth, may have been a factor in the average life span, which was greater for males than females (Connor 2004; Rautman 2006; Talbot 1997). Previous skeletal studies in Greece have reported lower life expectancies for th th women at Hellenistic and Roman Corinth (Fox-Leonard 1997; Fox 2005) as well as at 6 -7 century AD Crete and the southern Peloponnese (Bourbou 2003). While Christianity did offer women the option of a life devoted to the Church rather than to marriage and child care, it also reinforced the idea that young women should stay at home in order to help preserve their virginity (Talbot 1997). The Early Christian emphasis on maintaining one’s virginity also affected dietary recommendations. Girls and young women were encouraged to have meager diets consisting of vegetables, bread, water, and occasionally fish. Never satisfying their hunger was thought to play a critical part in controlling the body and reducing sexual desire (Alberici and Harlow 2007; Grimm 1995). As Bonvillain (2001) points out, chronic 100 malnutrition resulting from such practices may lead to women’s comparatively poor health status and, in turn, perpetuate ideas about their weakness and inferior social status. If dietary restrictions and differential treatment of men and women were a reality at Nemea, then females in the Early Christian and Middle to Late Byzantine skeletal samples will exhibit greater prevalence rates of linear enamel hypoplasias, periosteal reactions, porotic hyperostosis, and cribra orbitalia when compared to males. While women’s activities in the countryside likely extended beyond the household (Cameron 2006; Laiou 1981), it is expected that the most physically demanding tasks such as plowing and harvesting were more often carried out by men. Thus, it is anticipated that males will exhibit greater prevalence rates of osteoarthritis and occupational trauma than females in both phases. It should be noted, however, that women probably assisted in harvesting and other intensive activities more regularly during stressful times such as military conflicts (Talbot 1997). If this occurred at Nemea, then the prevalence of degenerative joint disease and occupational trauma will be greater among Middle to Late Byzantine females than among Early Christian females. Hypothesis 3: There will be significant differences in the prevalence rates of physiological stress indicators between groups of individuals based on grave location. As outlined in Chapter Two, the spatial analysis of mortuary sites, in conjunction with evidence from historical sources and skeletal paleopathology, has great potential to contribute to our understanding of changes in social structure and their impact on the daily lives of past societies (Ashmore and Geller 2005; Buikstra and Charles 1999; Goldstein 1981; Silverman 101 2002; Tzortzopoulou-Gregory 2008). An interdisciplinary bioarchaeological approach to the burials at Nemea is thus a crucial component of this research. The Early Christian and Byzantine cemeteries at Nemea will be examined in detail in Chapters Six and Eight. However, it is useful here to briefly describe the patterns of burial at the site, as it is otherwise difficult to discuss the expectations associated with Hypothesis Three. Frey (1998) has demonstrated that the burial practices of the Early Christian and Middle to Late Byzantine communities at Nemea were alike in that the dead were generally buried in simple graves with very few mortuary artifacts. However, marked differences between the two communities existed with respect to the choice of grave location and the number of individuals per grave. With few exceptions, graves in the Early Christian period consisted of single inhumations dispersed throughout the former sanctuary. Middle and Late Byzantine graves, by contrast, were concentrated in and around the remains of the Early Christian basilica and the associated chapel which was in use at that time. It is anticipated that those patterns are tied to the social and religious changes that occurred in Byzantium from Late Antiquity to the Middle to Late Byzantine period. The first is the fundamental shift in Christian practice from public worship in the Early Christian period to individual devotion during the Middle Byzantine period (Kazhdan and Epstein 1985; Krueger 2006). The second is the heightened emphasis on the nuclear family and concern for one’s lineage that developed in the Middle Byzantine period. Private dynastic chapels became popular as burial places among the elite, a process Kazhdan and Epstein (1985:104) referred to as the “architectural expression of the importance of self and family.” Because burial patterns in rural settings emulated those of urban contexts (Gerstel and Talbot 2006), it is anticipated that this aspect of burial practice was not confined to urban elites and that burials containing multiple 102 individuals located around the remains of the basilica and the small Byzantine chapel at Nemea 14 represent family tombs. In the near absence of mortuary artifacts, it is expected that any status distinctions at Nemea will be indicated by variability in other aspects of the burial program such as the spatial organization of the graves. As discussed in the previous chapter, Ivison (1993) has argued that there was a strong association between burial location and status in the Middle and Late Byzantine periods. While tombs could be located in different areas within and around churches, a spatial hierarchy existed that was meant to reflect the social hierarchy. Exclusive rights to the most sacred tomb locations were granted to prominent individuals such as church founders and high-ranking clergy. In the case of burials that took place within churches, the location of graves was associated with the positions that people occupied during the church service. Priests were buried in or close to the bema, while laypeople, if allowed church burial at all, were to be buried in the narthex or naos (Constas 2006; Ivison 1993). The place of greatest honor was “an eastern position on the south side of the naos” (Ivison 1993:70). With Ivison’s (1993) model in mind, Frey’s (1998) observation that Middle to Late Byzantine burials at Nemea were concentrated near the southeast corner of the basilica is particularly noteworthy. Especially given the concern over lineage that arose during this period, it is possible that this pattern resulted from competition or at least effort among Byzantine family groups at Nemea to be buried in privileged space. Consistent with Ivison’s (1993:74) assertion that the location of Byzantine tombs could be used by living relatives to affirm their own status, 14 Because this study is primarily concerned with the assessment of skeletal health, metric and non-metric data useful in investigating the possibility of genetic relationships among the individuals in those graves were not collected. Such data constitute an important area of future research at Nemea. For the purposes of this dissertation, the possibility of family tombs is explored using the demographic profiles established for each multiple burial. 103 the concentration of multiple burials near the place of honor at the Nemea basilica could be interpreted within the framework of Goldstein’s (1976, 1981) argument that corporate groups might legitimize their rights to resources by emphasizing descent through specialized disposal areas. Based on the historical developments and archaeological evidence discussed above, status differences are expected to have been more strongly expressed in the Middle to Late Byzantine community at Nemea. However, spatially distinct cemeteries representing differences in social rank have also been reported in Early Byzantine mortuary landscapes (e.g., Al-Shorman 2004). Thus, the correlation between grave location and social status can be tested for both periods at Nemea. If status distinctions provided certain individuals with privileged burial locations in addition to advantages such as improved access to adequate nutrition, decreased risk of infections, and lower levels of psychosocial stress, then it is possible that skeletal health disparities will exhibit spatial patterns. It is anticipated that individuals buried closer to the basilica during the Early Christian period will exhibit lower prevalence rates of physiological stress indicators than those buried farther away. During the Middle to Late Byzantine period, when graves are mostly concentrated near or inside of the basilica, lower prevalence rates of physiological stress indicators are expected be associated with greater proximity to the place of honor in the southeast end. Summary One of the goals of this research is to elucidate the effects of sociopolitical change on health in the Early Christian and Byzantine Greek countryside. It is hypothesized that more immediate and intense changes occurred during the Middle to Late Byzantine period, when the relationship between the central and provincial administrations became strained and the 104 Byzantine Empire was divided up during the violent transition to Latin rule. If the countryside was impoverished and subjugated as the historical record suggests, then living conditions for Greek peasants should have deteriorated and this should be reflected in poor skeletal health at Middle to Late Byzantine Nemea relative to Early Christian Nemea. It is also hypothesized that females will show greater prevalence rates of stress indicators associated with nutritional problems and infections than males during each period. However, greater prevalence rates of activity related stress indicators such as osteoarthritis are expected to be found among males. Another goal of this research is to test archaeological models that link burial location with social status in the Byzantine world. If the mortuary space at Nemea was organized in a hierarchical fashion, then it is possible that prevalence rates of physiological stress indicators will decrease with greater proximity to the basilica. The lowest prevalence rates of physiological stress indicators should be found among those individuals buried within the basilica, especially toward the east end. The cemeteries and samples of human skeletons that are used to test these hypotheses are the subject of the next chapter. 105 CHAPTER 6: MATERIALS AND METHODS This chapter introduces the burials and the samples of human skeletons recovered from Nemea. Here I also provide an overview of the methods used to collect and analyze demographic, paleopathological, and archaeological data in this research. The primary scoring criteria and recording system for demographic and paleopathological variables followed the protocols suggested by Buikstra and Ubelaker (1994). However, in some instances, modifications were made in order to facilitate comparison of the collected data with those from other studies of human skeletal remains from Byzantine Greece. Specific alterations are outlined below and also appear in Appendix A, which presents the data recording scheme utilized in this research. In most cases, significant deviations from the criteria suggested by Buikstra and Ubelaker (1994) were made for the recording of certain paleopathological variables. The alternative recording schemes for those variables followed the methods of the Global History of Health Project (Steckel et al. 2002a, 2006). Research Materials History of the Excavation of Burials at Nemea Human burials have been excavated from within the Sanctuary of Zeus at Nemea since at least the 1920s, when excavation rights passed to the American School of Classical Studies at Athens (Frey 1998; Miller 1990). Earlier investigations carried out by French archaeologists in th the late 19 century focused, among other areas, on the remains of a small chapel that was constructed on top of the Early Christian Basilica (Cousin and Dürrbach 1885). It is possible that graves were discovered by the French team given their great concentration in the area of the basilica. However, none were recorded. 106 Since American excavations began in 1924 under the direction of Bert Hodge Hill and Carl Blegen, nearly 300 burials from the Sanctuary of Zeus at Nemea have been recorded. However, Frey (1998) points out that a precise figure for the total number of graves is difficult to calculate. This problem is due mainly to inconsistencies in the recording of burials during nearly a century of excavations carried out by researchers from various institutions. References to burials in the Nemea excavation notebooks suggest that over the years graves were partially excavated and later re-recorded. In other instances skeletal remains were reburied in locations other than those from which they were excavated. Coupled with the fact that a number of graves were empty upon excavation, the total number of human skeletons is smaller than the number of graves which have been systematically excavated and recorded. The division of the graves at Nemea into two broad chronological periods (Early Christian and Middle to Late Byzantine) follows the burial typology developed by Frey (1998) and also the dates provided by Steven Miller and Stella Miller in the published excavation reports. The Early Christian Burials During the Early Christian period, the Sanctuary of Zeus had an important function as a disposal area for the community’s dead. The large number of graves placed within the sanctuary is overshadowed only by the ecclesiastical buildings and farming trenches when examining the th th most conspicuous pieces of evidence for activity at the site during the 5 -6 century. In general, the chronology of the burials at Nemea is difficult to establish because the graves are remarkably uniform in certain respects. With only a single exception, all burials are oriented east-west. Furthermore, they contain very few mortuary artifacts such as grave goods or items of personal adornment. The construction of the graves is also somewhat repetitive across 107 the site. With some overlap, the burials fall into a small number of basic types: simple pits, graves lined and covered with roof tiles, graves lined and covered with field stones, and tombs constructed of large blocks previously used in earlier monumental architecture. In some cases, graves could be dated confidently based on their association with architectural features. For example, some graves were cut into the floor and walls of the basilica and thus likely date later than its use as a place of worship during the Early Christian period. Others are located in the immediate vicinity of Byzantine farming trenches, which strongly suggests an earlier date. Aspects of the location, construction, and contents of the datable graves have been used to sort the less clear examples into either the Early Christian or Middle to Late Byzantine period (Frey 1998). Early Christian burials at Nemea are located in three distinct areas of the sanctuary (Figure 6.1): 1) Northwest and south of the Temple of Zeus 2) Northeast of the Hellenistic Bath 3) Within the Early Christian basilica and outside of its east, west, and south walls Frey (1998) suggests that within the Early Christian period, the twenty-eight graves associated with the temple are the earliest, followed by the ninety-six graves associated with the bath. The precise position of the twenty-one Early Christian graves associated with the basilica in the chronology is less clear. 108 Figure 6.1: Spatial Distribution of the Nemea Burials by Period. For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation. One of the sharpest distinctions between the Early Christian burials at Nemea and their Middle to Late Byzantine counterparts is a clear emphasis on single inhumations. This subject 109 will be treated more extensively in Chapter Eight. However, it is important at this point to note that there are very few examples of multiple and commingled burials dating to this period. As discussed briefly above, a second interesting feature of the Early Christian burials is the frequency with which graves were completely empty upon excavation. While the degree of preservation of human skeletal remains does vary across the site, the fact that many empty graves were found adjacent to and at the same elevation as graves containing relatively complete skeletons begs a different explanation. It is possible that these graves represent cenotaphs or that it was common practice among the Early Christian community to exhume bodies, perhaps to reuse graves. The treatment of the body is very consistent among Early Christian graves that contained preserved skeletons. Individuals were placed extended in a supine position and were oriented east-west with their heads to the west. The Middle to Late Byzantine Burials The Middle to Late Byzantine community at Nemea also utilized the Sanctuary of Zeus for the burial of the dead. However, instead of being widely dispersed throughout the sanctuary, burials dating to this period were concentrated in and around the collapsed ruins of the Early Christian basilica and the chapel constructed within or adjacent to it (see Figure 6.1). These burials were more frequently stone lined, although some were constructed using roof tiles or field stones. The well-built tombs that made use of portions of classical buildings date exclusively to this period. In addition to an emphasis on proximity to the basilica, Middle to Late Byzantine burials are often distinguished by the presence of multiple individuals. Out of a total of 83 graves, only 21 graves having bones available for analysis contained single inhumations. Also unlike the Early Christian burials, graves dating to this period always contained skeletal remains. 110 Middle to Late Byzantine burials show the same consistency in the arrangement and orientation of the body as Early Christian burials. Among those for which body position was recorded, individuals lay extended and supine with their arms crossed either over the chest or abdomen (Frey 1998). Frey (1998) points out that one of the only unique features of some of the burials associated with the basilica is greater attention paid to the head of the deceased. In such cases, the head rested upon a stone pillow or was enclosed and supported by stones or tiles. As discussed in Chapter Four, Ivison (1992, 1993:86) has argued that these “cephalic burials” indicate western influence based on their presence in the Latin West and on their distribution in areas of the Byzantine Empire that were under Frankish control following 1204 AD. As discussed in Chapter Three, the Middle to Late Byzantine graves at Nemea are believed to be those of the community that occupied the settlement just east of the sanctuary th th during the 12 -13 centuries AD. It should be noted that a subset of graves located outside of th th the Early Christian basilica near its southeast corner may in fact date to the 14 or 15 century (Miller 1988). 15 For the purposes of this research, those graves and the skeletal remains recovered from them have been grouped together with the rest of the archaeological and skeletal material dating to the Middle to Late Byzantine period. The Human Skeletal Samples from Nemea As outlined above, the graves at Nemea can be categorized generally into either the th th th th Early Christian (5 -6 century AD) or Middle to Late Byzantine (12 -13 century AD) periods using the typology developed by Frey (1998). Thus, the collection of human skeletons recovered from Nemea represents two communities that occupied the site during distinct periods of time. 15 The numbers for these graves assigned by Miller (1988) are 1-16 and 18-26 from grid square M-19. 111 As a result of factors such as poor preservation, multiple interments, and extensive commingling in certain graves, the figures for the number of individuals in each period are minimum counts. The minimum number of individuals (MNI) for each commingled burial was determined after sorting skeletal elements by age and anatomical side. The most frequently duplicated element provided the basis for the MNI estimate. Table 6.1 shows the MNI recovered and analyzed from each period of occupation. One hundred thirteen individuals date to the Early Christian period. The MNI for the Middle to Late Byzantine period (146) is greater even though the contents of fewer graves containing human remains were available for analysis. This is due, again, to the marked tendency of graves from that period to contain multiple individuals. Table 6.1: Minimum Number of Individuals in Each Period. Period Total Early Christian 113 Middle to Late Byzantine 146 Total Sample 259 All individuals available for analysis were examined in this dissertation research. The number of adults and subadults in each period are shown in Table 6.2 below. In the Early Christian sample, adults make up 69.9% (79/113) of the total sample and subadults make up 30.1% (34/113) of the total sample. Similarly, adults in the Middle to Late Byzantine sample make up 72.6% (106/146) of the total, while subadults form 27.4% (40/146) of the total. The number of subadults is distributed relatively evenly between phases at Nemea, while more adults 16 are represented in the Middle to Late Byzantine period than in the Early Christian period. 16 Many of the adult individuals in the Middle to Late Byzantine phase are from commingled burials and their presence is based on minimum counts. The number of discrete human skeletons in this phase is actually smaller than that in the Early Christian phase. 112 Table 6.2: Number of Adults and Subadults in Each Period. Period Subadult Adult Total Early Christian 34 79 113 Middle to Late Byzantine 40 106 146 Total Sample 74 185 259 As discussed above, the individuals dating to the Early Christian period come from graves distributed across different areas of the site, and it is worth briefly examining the age demographics of each burial area to determine in a preliminary manner whether or not they represent specialized disposal areas. Table 6.3 shows the number of adults and subadults in each of the major Early Christian burial areas in the Sanctuary of Zeus. While no burial area was exclusive to adults or subadults, differences in their proportions in each area are apparent. Those differences will be explored in detail in Chapter Eight. All 146 individuals dated to the Middle to Late Byzantine period were recovered from burials located either within or adjacent to the basilica. Table 6.3: Number of Early Christian Individuals by Burial Location. Area Subadult Adult Total Temple 3 19 22 North of Bath 24 44 68 Basilica 7 16 23 Total 34 79 113 Research Methods Skeletal Analysis The osteological analysis of the human skeletal material recovered from Nemea was carried out using criteria suggested by Buikstra and Ubelaker (1994) and (Steckel et al. 2002a, 2006). Skeletal inventories that included notations on the presence and completeness of each bone and tooth were completed for each individual (or grave in the case of multiple, commingled burials). Additional osteological data collected in this research included morphological 113 observations appropriate for the estimation of age and sex in adults, and developmental indicators used to estimate age in subadults. Osteometric measurements of the postcranial skeleton were taken whenever possible. The bulk of the data recorded and analyzed in this research pertains to skeletal and dental paleopathologies. Indicators of physiological stress were recorded by presence and expression, while descriptive observations were made for specific pathological conditions and extensive trauma. As discussed in Chapter Three and in the sections above, the graves from the Sanctuary of Zeus often contained multiple, commingled individuals. The frequent practice of including more than one individual in a grave resulted in a challenging osteological analysis that complicated efforts to determine basic demographic parameters such as the MNI for individual graves and age and sex estimates for their occupants. The MNI determined for graves containing multiple individuals at Nemea ranged from as few as two to as many as sixteen. Because individuals from a particular grave could not always be separated out, observations on paleopathological conditions were recorded by skeletal element. This approach, as opposed to recording the presence of such conditions by individual, enables data comparison between subgroups within the total skeletal sample (e.g., males and females, age categories, temporal phases) at Nemea. Demographic Data Collection Estimation of Age The methods used in the estimation of age were dependent upon the specific skeletal elements available for each burial. In both subadults and adults, age ranges generated by multiple techniques were considered when possible. In most cases, at least one of the aging techniques outlined in Buikstra and Ubelaker (1994) was able to be applied. Age estimation from well- 114 preserved pelvic bones utilized the pubic symphysis (Brooks and Suchey 1990; Suchey and Katz 1986, 1998). In many cases, however, only the auricular surface (Lovejoy et al. 1985; Meindl and Lovejoy 1989) was available for use. Cranial suture closure (Buikstra and Ubelaker 1994) was used less frequently, but especially in cases of unassociated crania from commingled burials. In some instances, preservation of sternal rib ends permitted the application of the aging technique developed by İşcan et al. (1984, 1985). Whenever possible, individuals were placed into one of the following adult age categories recommended by Buikstra and Ubelaker (1994): young adult (20-35 years); middle adult (35-50 years); and old adult (50+). Subdault ages were estimated using dental formation and eruption (Ubelaker 1989), diaphyseal length measurements, and degree of epiphyseal union (Schaefer et al. 2009; Scheuer and Black 2000). Hillson (1996) has pointed out that aging techniques based on dental development provide the most accurate estimates of age at death in subadults. This is because the development of the dentition—especially its formation—is under stronger genetic control than skeletal development (Saunders 2008). Thus, when teeth were available, estimates derived from their development were preferred. Each subadult was placed into one of the following age categories recommended by Buikstra and Ubelaker (1994): fetal ( 83.842 > Females Total 73 During the data collection process, each individual was placed into one of the following categories suggested by Buikstra and Ubelaker (1994): undetermined, probable female, female, ambiguous, probable male, and male. However, in order to increase cell sizes, individuals 136 originally placed into the “probable male” and “probable female” categories were reclassified into the categories of “male” and “female.” Table 7.3 shows the number of individuals in each category of sex. Sex was unable to be assigned in two cases due to the absence or poor preservation of bones from the pelvis, cranium, and limbs. Table 7.3: Distribution of Sex among Adults in the Early Christian Sample. Sex Category Number % Early Christian (n) Adults Male 41 51.9 Female 36 45.6 Ambiguous 0 0 Undetermined 2 2.5 Total 79 100.0 Demographic Patterns: Subadults The estimation of age at death among subadults from the Early Christian period relied primarily on the degree of formation and eruption of the dentition. This is because dental development is less influenced by extrinsic factors than skeletal development (Saunders 2008). In the absence of the dentition, subadult ages were based on degree of epiphyseal union and diaphyseal length measurements (Schaefer et al. 2009; Scheuer and Black 2000). Each subadult individual was placed into one of the following categories recommended by Buikstra and Ubelaker (1994): fetal (1/2 Diaphysis) (<1/4 Surface) (<1/2 Surface) L R L R L R L R Adults 31 30 1 1 1 2 4 3 Subadults 1 1 1 2 1 1 0 0 Most periosteal reactions in this sample were scored as “sclerotic,” which means that they were healed at the time of death (Table 7.15). Active lesions appearing as woven bone were observed in one adult and one subadult. Table 7.15: Activity of Periosteal Reactions in Early Christian Adults and Subadults. Tibia Woven Sclerotic Mixed Reaction L R L R L R Adults 1 0 32 30 3 6 Subadults 1 0 1 1 1 3 High prevalence rates of periosteal reactions are exhibited by both males and females in this phase. Again, more consistent figures would likely be generated with larger subsample sizes. The prevalence rate for the left tibia is greater among females, while that for the right tibia is greater in males. The difference between males and females for neither the right (p = .307) nor the left tibia (p = .153) is significant. Table 7.16: Prevalence Rates of Periosteal Reactions in Early Christian Males and Females. Tibia Present Absent Total % Affected L R L R L R L R Males 22 22 6 2 28 24 78.6 91.7 Females 12 9 0 3 12 12 100.0 75.0 Males and females in this sample show no clear differences in the expression of periosteal reactions (Table 7.17). Extensive reactions are more common in males, but the differences in subsample sizes make this pattern difficult to evaluate. 146 Table 7.17: Expression of Periosteal Reactions in Early Christian Males and Females. Tibia Accentuated Slight Discrete Moderate Extensive Reaction Striations Patches Involvement (>1/2 Diaphysis) (<1/4 Surface) (<1/2 Surface) L R L R L R L R Males 18 18 0 1 1 0 3 3 Females 10 7 1 0 0 2 1 0 For both males and females, most periosteal reactions were scored as “sclerotic” (Table 7.18). The only woven (active) lesion observed among sexed adults was present on the left tibia of a male. Table 7.18: Activity of Periosteal Reactions in Early Christian Males and Females. Tibia Woven Sclerotic Mixed Reaction L R L R L R Males 1 0 18 18 2 4 Females 0 0 11 7 1 2 Osteoarthritis The prevalence of osteoarthritis in all Early Christian adults is reported for each major appendicular joint in Table 7.19 below. All joint categories examined in the Early Christian sample showed some evidence of osteoarthritis with the exception of the right ankle. The most frequently affected joints were the shoulder and hip, with the right shoulder exhibiting the highest prevalence rate at 36%. In general the elbow, wrist/hand, knee, and ankle/foot joints show much lower prevalence rates. The ankle/foot joint was the least affected by osteoarthritis. None of the differences between right and left sides are statistically significant. 147 Table 7.19: Osteoarthritis Prevalence by Joint in the Early Christian Period. Joint Present Absent Total Joints % Affected L. Shoulder 6 15 21 28.6 R. Shoulder 9 16 25 36.0 L. Elbow 2 23 25 8.0 R. Elbow 3 22 25 12.0 L. Wrist/Hand 2 18 20 10.0 R. Wrist/Hand 4 30 34 11.8 L. Hip 13 27 40 32.5 R. Hip 11 32 43 25.6 L. Knee 4 20 24 16.7 R. Knee 3 24 27 11.1 L. Ankle/Foot 1 19 20 5.0 R. Ankle/Foot 0 15 15 0.0 When the available joints in the Early Christian sample are compared by sex, greater prevalence rates are seen in males for most joint categories (Table 7.20). The exceptions are the left knee and the right ankle/foot, which have problematic subsample sizes. While sample sizes are small for all joints, especially among females, the difference between males and females for the left hip was found to be statistically significant. A similar pattern is apparent for the right hip, although the difference only approaches the level of significance using Fisher’s exact test. Table 7.20: Osteoarthritis Prevalence by Joint in Early Christian Males and Females. Males Females Joint Pr Ab % Affected Pr Ab % Affected χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No L. Shoulder 6 10 37.5 0 5 0.0 1 0.262 No R. Shoulder 7 10 41.2 2 4 33.3 1 1.000 No L. Elbow 1 12 7.7 0 5 0.0 1 1.000 No R. Elbow 2 14 12.5 0 5 0.0 1 1.000 No L. Wrist/Hand 1 11 8.3 0 3 0.0 1 1.000 No R. Wrist/Hand 2 19 9.5 0 4 0.0 1 1.000 No L. Hip 11 11 50.0 2 13 13.3 5.261 1 0.022 YES R. Hip 10 17 37.0 1 12 7.7 1 0.068 No L. Knee 3 14 17.6 1 3 25.0 1 1.000 No R. Knee 3 15 16.7 0 6 0.0 1 0.546 No L. Ankle/Foot 1 11 8.3 0 5 0.0 1 1.000 No R. Ankle/Foot 0 9 0.0 0 1 0.0 *Empty cells indicate that Fisher’s exact test was used instead of Pearson’s chi-square test. 148 Antemortem Trauma The analysis of antemortem trauma focused on fractures. The presence or absence of fractures in the Early Christian period was recorded by bone so that the data are comparable to the figures for the Middle to Late Byzantine period and to other regional studies. As discussed in Chapter Six, cranial bones were included in the bone count if they were at least 75% complete, and long bones were included if at least three out of five segments (proximal epiphysis, proximal diaphysis, middle diaphysis, distal diaphysis, and distal epiphysis) were 75% or more complete. Because no examples of antemortem trauma were observed in Early Christian subadults, the following discussion includes adults only. The prevalence rates of fractures in the major cranial bones are reported for the Early Christian period in Table 7.21. Fractures were observed only on bones of the cranial vault, including the left frontal and the left and right parietals. The left parietal was the most frequently affected element, with three bones out of 40 (7.5%) exhibiting fractures. One left frontal bone and one left parietal bone were also affected. Four of the cranial fractures observed in this period were classified as depression fractures and were well healed at the time of death. A defect on the left parietal of a young adult male was classified as a penetrating fracture and was also well healed at the time of death (Table 7.22). 149 Table 7.21: Fracture Prevalence by Cranial Bone in the Early Christian Period (Adults Only). Bone Present Absent Total Bones % Affected L. Frontal* 1 25 26 3.8 R. Frontal* 0 24 24 0.0 L. Parietal 3 37 40 7.5 R. Parietal 1 36 37 2.7 Occipital 0 34 34 0.0 L. Temporal 0 19 19 0.0 R. Temporal 0 15 15 0.0 L. Zygomatic 0 28 28 0.0 R. Zygomatic 0 23 23 0.0 L. Maxilla 0 15 15 0.0 R. Maxilla 0 12 12 0.0 L. Mandible* 0 35 35 0.0 R. Mandible* 0 27 27 0.0 *Observations were made separately on the left and right portions of the frontal bone and the mandible following the inventory form in Buikstra and Ubelaker (1994). Table 7.22: Type and Healing Status of Early Christian Cranial Fractures. Skeleton Bone Fracture Type Healing Sex Age SK 066 L. Frontal Depression Well Healed Male 40-55 SK 076 L. Parietal Depression Well Healed Male 35-50 SK 155 L. Parietal Depression Well Healed Male 35-50 SK 160 L. Parietal Penetrating Well Healed Male 24-34 SK 009 R. Parietal Depression Well Healed Female 20-29 Among postcranial bones, the right clavicle was the most frequently fractured element. The prevalence rate for the right clavicle (10.0%) was also the highest of any bone in the Early Christian sample. Fractures were also observed on a right ulna and a right femur (Table 7.23). All postcranial fractures were classified as oblique fractures and were well healed with the exception of one midshaft clavicle fracture that was partially healed (Table 7.24). 150 Table 7.23: Fracture Prevalence by Postcranial Bone in the Early Christian Period (Adults Only). Bone Present Absent Total Bones % Affected L. Clavicle 0 25 25 0.0 R. Clavicle 3 27 30 10.0 L. Humerus 0 26 26 0.0 R. Humerus 0 31 31 0.0 L. Radius 0 20 20 0.0 R. Radius 0 30 30 0.0 L. Ulna 0 17 17 0.0 R. Ulna 1 27 28 3.6 L. Femur 0 30 30 0.0 R. Femur 1 32 33 3.0 L. Tibia 0 16 16 0.0 R. Tibia 0 14 14 0.0 L. Fibula 0 8 8 0.0 R. Fibula 0 5 5 0.0 Table 7.24: Type, Location, and Healing Status of Early Christian Postcranial Fractures. Skeleton Bone Fracture Type Location Healing Sex Age SK 012 R. Clavicle Oblique Lateral Well Healed Male 30-40 SK 116 R. Clavicle Oblique Lateral Well Healed Indeterminate Adult SK 023 R. Clavicle Oblique Middle Partially Healed Male 35-50 SK 116 R. Ulna Oblique Middle Well Healed Indeterminate Adult SK 014 R. Femur Oblique Proximal Well Healed Male 40-50 For bones exhibiting fractures, prevalence rates are compared between Early Christian males and females in Table 7.25. While none of the differences between males and females for individual bones are statistically significant, there is a clear tendency for males to be affected by fractures more frequently than females. When all ten fractures in the Early Christian sample are considered, seven occurred in males, one female was affected, and the remaining two are exhibited on unassociated elements of indeterminate sex. The only fracture exhibited by a female was observed on a right parietal. The fractured right ulna and a third fractured right clavicle, which were not assigned a sex and do not appear in Table 7.25, are unassociated elements from the same commingled burial. 151 Table 7.25: Fracture Prevalence by Bone in Early Christian Males and Females. Males Females Bone Pr Ab % Affected Pr Ab % Affected χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No L. Frontal 1 15 6.7 0 10 0.0 1 1.000 No L. Parietal 3 20 13.0 0 17 0.0 1 0.248 No R. Parietal 0 20 0.0 1 16 5.9 1 0.459 No R. Clavicle 2 18 10.0 0 8 0.0 1 1.000 No R. Ulna 0 15 0.0 0 6 0.0 R. Femur 1 22 4.3 0 9 0.0 1 1.000 No *Empty cells indicate that Fisher’s exact test was used instead of Pearson’s chi-square test. The Middle to Late Byzantine Skeletal Sample The minimum number of individuals represented in the Middle to Late Byzantine skeletal sample from Nemea is 146. As was the case for the Early Christian sample, all of those individuals were analyzed in this research. While the preservation of the Middle to Late Byzantine sample is much better, with very few examples of erosion of the cortical surfaces, it presents its own unique limitations. The skeletal material in this sample was recovered from 61 graves containing human remains. The marked disparity between the number of graves and the minimum number of individuals is due to the fact that Middle to Late Byzantine burials often contained multiple individuals that were extensively commingled. This situation complicates the analysis of both demographic and paleopathological data. The latter is frequently reported here by bone or tooth rather than by individual in order to include isolated elements while controlling for the possibility of over-reporting prevalence rates. As the presence and preservation of bones is not uniform across graves, the number of individuals in a given set of results may vary greatly depending on the variable being analyzed. The Middle to Late Byzantine sample includes 106 adults (72.6% of the total) and 40 subadults (27.4% of the total). The data on age and sex among the adults is reported first. 152 Demographic Patterns: Adults Age at Death The number of individuals placed into each adult age category recommended by Buikstra and Ubelaker (1994) appear in Table 7.26 below. Individuals in the generic “adult” category were missing the skeletal elements (innominates, crania, ribs) used to estimate age at death for adults in this research. Individuals placed into that category were usually those represented by duplicated elements such as long bones that were developmentally complete. Table 7.26: Number and Proportion of Middle to Late Byzantine Individuals in Each Adult Age Category. Age Category Number % Middle to Late (n) Byzantine Adults Young Adult (20-35) 29 27.4 Middle Adult (35-50) 36 34.0 Old Adult (50+) 12 11.3 Adult (20+) 29 27.4 Total 106 100.0 Sex In a sample composed of a large number of multiple and commingled burials, the approach to the determination of sex was somewhat different. Assessments of sex were carried out on a grave by grave basis. This first involved identifying and analyzing the reliable indicators of sex that were duplicated in each grave. Other skeletal elements that might indicate sex through metric measurements (e.g., femur) were relied upon secondarily. However, due to the number of incomplete skeletons and isolated skeletal elements, sex determinations relied to a greater degree on metric assessments of robusticity. Demarking points were developed for a number of bones for the Middle to Late Byzantine sample and are presented in Appendix B. Each was derived from measurements of adult individuals for which sex was established using pelvic or cranial 153 remains. In the analysis of paleopathologies, sex was assigned to isolated skeletal elements using those points. The number of adult individuals in each category of sex appears in Table 7.27. There is a marked difference between the number of males and females in this phase. The larger number of individuals of undetermined sex in this sample is due to the greater frequency with which individuals were represented by partial skeletons or isolated skeletal elements not useful in sex determination. Table 7.27: Distribution of Sex among Adults in the Middle to Late Byzantine Sample. Sex Category Number % Middle to Late (n) Byzantine Adults Male 67 63.2 Female 24 22.6 Ambiguous 0 0 Undetermined 15 14.2 Total 106 100.0 Demographic Patterns: Subadults As was the case for the Early Christian sample, the analysis of subadult demography focused on the estimation of age. Sex was assigned only to two older adolescents, one male and one female. Table 7.28 shows the number of individuals placed into each subadult age category recommended by Buikstra and Ubelaker (1994). Again, children are the best represented category, making up more than half of the subadults buried at Nemea during the Middle to Late Byzantine period. 154 Table 7.28: Number and Proportion of Middle to Late Byzantine Individuals in Each Subadult Age Category. Age Category Number % Middle to Late (n) Byzantine Subadults Fetal (1/2 Diaphysis) (<1/4 Surface) (<1/2 Surface) L R L R L R L R Adults 43 50 1 3 1 1 0 0 Subadults 7 4 0 0 0 0 0 0 Only one lesion, observed on the left tibia of a subadult, was scored as “woven” or active at the time of death. Adults exhibited almost exclusively “sclerotic” or healed periosteal reactions. 162 Table 7.40: Activity of Periosteal Reactions in Middle to Late Byzantine Adults and Subadults. Tibia Woven Sclerotic Mixed Reaction L R L R L R Adults 0 0 45 53 0 1 Subadults 1 0 4 2 2 2 A comparison of prevalence rates by sex shows an inconsistent pattern, probably due to the smaller number of females than males (Table 7.41). The prevalence of periosteal reactions is greater in males for the left tibia, although the difference is not significant (p = 0.730). Females are more frequently affected when the right tibia is considered. In that case as well the difference does not reach the p <0.05 level of significance (p = 0.309). Table 7.41: Prevalence Rates of Periosteal Reactions in Middle to Late Byzantine Males and Females. Tibia Present Absent Total % Affected L R L R L R L R Males 27 29 11 10 38 39 71.1 74.4 Females 8 15 5 2 13 17 61.5 88.2 When the degree of expression is considered, females exhibited only reactions that were scored as “accentuated striations.” While the periosteal reactions observable in most males were also placed into that category, a small number of individuals exhibited more severely expressed lesions (Table 7.42). All periosteal reactions exhibited by males and females in this sample were healed at the time of death (Table 7.43). Table 7.42: Expression of Periosteal Reactions in Middle to Late Byzantine Males and Females. Tibia Accentuated Slight Discrete Moderate Extensive Reaction Striations Patches Involvement (>1/2 Diaphysis) (<1/4 Surface) (<1/2 Surface) L R L R L R L R Males 25 27 1 2 1 0 0 0 Females 8 15 0 0 0 0 0 0 163 Table 7.43: Activity of Periosteal Reactions in Middle to Late Byzantine Males and Females. Tibia Woven Sclerotic Mixed Reaction L R L R L R Males 0 0 27 29 0 0 Females 0 0 8 15 0 0 Osteoarthritis The presence or absence of osteoarthritis was recorded on all major non-vertebral joints available in the Middle to Late Byzantine sample. Subsample sizes for each joint category in this phase are generally larger than those in the Early Christian sample due to better preservation and to the greater number of individuals represented. The prevalence of osteoarthritis in all Middle to Late Byzantine adults is reported by joint and anatomical side in Table 7.44 below. As discussed in Chapter Six, one bone was chosen to represent each joint in the case of commingled burials because of the problematic nature of re-associating their constituent skeletal elements. 19 The most frequently affected joint in this sample was the knee, with 40% or more of the observable joints showing evidence of osteoarthritis. High prevalence rates of osteoarthritis are also found for the shoulder and hip joints. The elbow and wrist/hand show more moderate prevalence rates. The ankle/foot joint was least affected by osteoarthritis. None of the differences between the right and left sides are statistically significant. 19 See Table 6.4 for a list of the bones selected to represent each joint. 164 Table 7.44: Osteoarthritis Prevalence by Joint in the Middle to Late Byzantine Period. Joint Present Absent Total Joints % Affected L. Shoulder 9 29 38 23.7 R. Shoulder 17 25 42 40.5 L. Elbow 8 45 53 15.1 R. Elbow 14 45 59 23.7 L. Wrist/Hand 8 32 40 20.0 R. Wrist/Hand 8 32 40 20.0 L. Hip 16 35 51 31.4 R. Hip 13 32 45 28.9 L. Knee 21 30 51 41.2 R. Knee 20 30 50 40.0 L. Ankle/Foot 2 34 36 5.6 R. Ankle/Foot 3 36 39 7.7 Prevalence rates of osteoarthritis were compared between males and females, utilizing metric assessments of robusticity to estimate sex in cases of unassociated long bones. 20 Males show greater prevalence rates of osteoarthritis than females for most joints and the smaller number females in this sample makes the patterns that deviate from that trend difficult to interpret (Table 7.45). For example, males show a greater tendency toward arthritic changes in the shoulder, while for the elbow and hip the greater sex-based prevalence rate depends on the anatomical side considered. It is likely that the inconsistency in the data for the latter joints is an artifact of the relatively small number of female observations rather than a reflection of differences in patterns of behavior. Similarly, while females show no evidence of osteoarthritis in the wrist/hand joint, this is almost certainly due to the size of the subsamples. Fisher’s exact tests indicate that none of the apparent differences between males and females in the prevalence of osteoarthritis are statistically significant. 20 See Appendix B for a list of demarking points used in the estimation of sex from isolated long bones. 165 Table 7.45: Osteoarthritis Prevalence by Joint in Middle to Late Byzantine Males and Females. Males Females Joint Pr Ab % Affected Pr Ab % Affected χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No L. Shoulder 8 20 28.6 0 9 0.0 1 0.159 No R. Shoulder 14 17 45.2 3 8 27.3 1 0.477 No L. Elbow 4 29 12.1 3 10 23.1 1 0.385 No R. Elbow 9 27 25.0 3 16 15.8 1 0.511 No L. Wrist/Hand 4 18 18.2 0 4 0.0 1 1.000 No R. Wrist/Hand 6 20 23.1 0 3 0.0 1 1.000 No L. Hip 14 26 35.0 2 8 20.0 1 0.468 No R. Hip 8 25 24.2 5 6 45.5 1 0.256 No L. Knee 18 22 45.0 2 5 28.6 1 0.682 No R. Knee 17 20 45.9 3 7 30.0 1 0.481 No L. Ankle/Foot 0 23 0.0 1 4 20.0 1 0.179 No R. Ankle/Foot 1 22 4.3 0 6 0.0 1 1.000 No *Empty cells indicate that Fisher’s exact test was used instead of Pearson’s chi-square test. Antemortem Trauma The presence or absence of fractures was recorded by bone in the Middle to Late Byzantine sample. As was the case for the Early Christian sample, cranial bones were included in the bone count if they were at least 75% complete, and long bones were included if at least three out of five segments (proximal epiphysis, proximal diaphysis, middle diaphysis, distal diaphysis, and distal epiphysis) were 75% or more complete. No examples of antemortem trauma were observed in Middle to Late Byzantine subadults and the results reported here include adults only. Table 7.46 shows the distribution and prevalence of cranial fractures in the Middle to Late Byzantine sample. Only two bones, each a left parietal belonging to an adult male, were affected. Both defects were categorized as depression fractures and were well healed at the time of death. 166 Table 7.46: Fracture Prevalence by Cranial Bone in the Middle to Late Byzantine Period (Adults Only). Bone Present Absent Total Bones % Affected L. Frontal* 0 22 22 0.0 R. Frontal* 0 20 20 0.0 L. Parietal 2 25 27 7.4 R. Parietal 0 26 26 0.0 Occipital 0 30 30 0.0 L. Temporal 0 20 20 0.0 R. Temporal 0 17 17 0.0 L. Zygomatic 0 16 16 0.0 R. Zygomatic 0 21 21 0.0 L. Maxilla 0 14 14 0.0 R. Maxilla 0 19 19 0.0 L. Mandible* 0 35 35 0.0 R. Mandible* 0 29 29 0.0 *Observations were made separately on the left and right portions of the frontal bone and the mandible following the inventory form in Buikstra and Ubelaker (1994). Table 7.47: Type and Healing Status of Middle to Late Byzantine Cranial Fractures. Skeleton Bone Fracture Type Healing Sex Age SK 107a L. Parietal Depression Well Healed Male 21-34 SK 144b L. Parietal Depression Well Healed Male 40-50 Fractures were present on a number of postcranial bones, although prevalence rates for individual bone types are low. The most frequently fractured elements were the left clavicle and the left radius, with two bones out of 28 (7.1%) affected in each case. It should be noted that many of the fractures reported in Tables 7.48 and 7.49 below appear on skeletal elements belonging to the same adult male individual (SK 091). This individual exhibits bilateral impacted fractures of the distal humeri and proximal radii and ulnae. An additional transverse fracture is present on the distal third of the left ulna. It is also possible that the fractured left tibia and left fibula from the commingled burial designated “SK 117” belong to the same individual. All fractures recorded in this sample were well healed at the time of death. 167 Table 7.48: Fracture Prevalence by Postcranial Bone in the Middle to Late Byzantine Period (Adults Only). Bone Present Absent Total Bones % Affected L. Clavicle 2 26 28 7.1 R. Clavicle 0 20 20 0.0 L. Humerus 1 29 30 3.3 R. Humerus 1 38 39 2.6 L. Radius 2 26 28 7.1 R. Radius 1 21 22 4.5 L. Ulna 1 22 23 4.3 R. Ulna 1 25 26 3.8 L. Femur 0 39 39 0.0 R. Femur 0 38 38 0.0 L. Tibia 1 27 28 3.6 R. Tibia 0 28 28 0.0 L. Fibula 1 19 20 5.0 R. Fibula 1 16 17 5.9 Table 7.49: Type, Location, and Healing Status of Middle to Late Byzantine Postcranial Fractures. Skeleton Bone Fracture Type Location Healing Sex SK 127 L. Clavicle Transverse Middle Well Healed Male SK 141b L. Clavicle Comminuted Lateral Well Healed Female SK 091 L. Humerus Impacted Distal Well Healed Male SK 091 R. Humerus Impacted Distal Well Healed Male SK 126 L. Radius Oblique Distal Well Healed Female SK 091 L. Radius Impacted Proximal Well Healed Male SK 091 R. Radius Impacted Proximal Well Healed Male SK 091 L. Ulna* Impacted Proximal Well Healed Male SK 091 L. Ulna* Transverse Distal Well Healed Male SK 091 R. Ulna Impacted Proximal Well Healed Male SK 117 L. Tibia Oblique Proximal Well Healed Indeterminate SK 117 L. Fibula Oblique Middle Well Healed Indeterminate SK 100a R. Fibula Compression Proximal Well Healed Male *Bone exhibits multiple fractures. Age 34-44 35-60 34-44 34-44 21-35 34-44 34-44 34-44 34-44 34-44 Adult Adult 45-60 For bones exhibiting fractures, prevalence rates are compared between Middle to Late Byzantine males and females in Table 7.50. No significant differences are found between the sexes and small subsample sizes especially in females make definitive statements about the sex distribution of fractures difficult. At first glance, fractures are exhibited far more frequently in males. However, the distribution of fractures between the sexes is relatively even when the 168 multiple fractures exhibited by individual SK 091 are taken into consideration. As stated above, the fractured left tibia and left fibula, which were not assigned a sex and do not appear in Table 7.50, are unassociated elements from the same commingled burial. Table 7.50: Fracture Prevalence by Bone in Middle to Late Byzantine Males and Females. Males Females Bone Pr Ab % Affected Pr Ab % Affected χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No L. Parietal 2 16 11.1 0 8 0.0 1.000 No L. Clavicle 1 15 6.3 1 5 16.7 0.481 No L. Humerus 1** 23 4.2 0 6 0.0 1.000 No R. Humerus 1** 26 3.7 0 12 0.0 1.000 No L. Radius 1** 20 4.8 1 5 16.7 0.402 No R. Radius 1** 19 5.0 0 2 0.0 1.000 No L. Ulna 1** 15 6.3 0 4 0.0 1.000 No R. Ulna 1** 15 6.3 0 4 0.0 1.000 No L. Tibia 0 23 0.0 0 3 0.0 L. Fibula 0 15 0.0 0 2 0.0 R. Fibula 1 10 9.1 0 3 0.0 1.000 No *Empty cells indicate that Fisher’s exact test was used instead of Pearson’s chi-square test. **Fractured bone from SK 091. Summary of Significant Results Pertinent to Hypothesis Two One of the goals of this research is the identification of sex-based differences in the prevalence of physiological stress indicators within each skeletal sample that potentially reflect health disparities between men and women in each community at Nemea. If women were a marginalized group in Byzantine society, then it is anticipated that the prevalence rates of most skeletal indicators of stress will be greater among females. The results from the first two sections of this chapter are used to test the second hypothesis of this dissertation. Hypothesis 2: Prevalence rates of physiological stress indicators associated with nutritional stress and infection will be greater among females than males in both phases at Nemea. Prevalence rates of activity related stress indicators will be greater among males in both phases. 169 Intraphase comparisons of prevalence rates of physiological stress indicators between males and females that produced statistically significant results are presented in Table 7.51. While this analysis reveals very few significant differences between males and females, those that do exist are in accordance with Hypothesis Two. In the Early Christian sample, the prevalence rate of cribra orbitalia, a non-specific stress indicator usually linked with nutritional stress, is greater in females than in males. The prevalence rate of osteoarthritis affecting the left hip joint, an activity-related condition, is greater in males than in females. In the case of the latter, it is worth noting that although the p <0.05 level of significance is reached only for the left side, the differences between males and females when the corresponding hip joint is considered does approach significance (p = 0.068). Although significant differences expressed bilaterally would make a stronger case for sex-based differences in activity patterns, these figures at least suggest that males experienced greater levels of activity-related stress that affected the hip joint. The significant result for linear enamel hypoplasias in the Middle to Late Byzantine sample is more problematic. While the frequencies of the presence and absence of hypoplasias on the left mandibular canine differ from those expected by chance and the prevalence is greater in males, the subsample sizes especially in the case of females are quite small. Furthermore, the results for the corresponding right mandibular canine show no evidence of a similar trend. The absence of a systemic pattern and the small sample sizes together suggest that there is no broader significance to this particular result. 170 Table 7.51: Summary of Statistically Significant Results Pertinent to Hypothesis Two. Period Paleopathological Significant Result (p <0.05) Condition (frequencies differ from those expected by chance) Early Christian Cribra Orbitalia Cribra orbitalia is present more frequently in females than in males. Early Christian Osteoarthritis Osteoarthritis is present on the left hip joints of males more frequently than on those of females. Middle to Late Linear Enamel Linear enamel hypoplasias are present on the left Byzantine Hypoplasias mandibular canines (tooth #22) of males more frequently than on those of females. It should be noted that although no statistically significant differences between males and females were found when the frequencies of fractures were compared by bone, there is a tendency for fractures to appear in males. In fact, the inability to detect significant differences is probably due only to the fact that there are a relatively small number of fractures affecting many different bones on which prevalence rates were calculated independently. Among fractured skeletal elements that could be assigned a sex in the Early Christian sample, 80% (4/5) of cranial fractures occurred in males and all three postcranial fractures affected males. The pattern is less clear for the Middle to Late Byzantine sample because many of the observed fractures affected a single individual. While it is not verified statistically then, the pattern for antemortem trauma, at least for the Early Christian period, is consistent with Hypothesis Two. Interphase Comparison This section presents the results of the skeletal analysis of the Early Christian and Middle to Late Byzantine skeletal samples from Nemea using a comparative temporal approach. Comparisons of prevalence rates of linear enamel hypoplasias are limited to the maxillary and mandibular canines because those teeth were the most frequently affected in each sample. As in the previous sections, associations between demographic subgroups and the physiological stress indicators used in this research were tested using Pearson’s chi-square test or Fisher’s exact test 171 when any expected cell frequency was less than 5. The level of significance is set at p <0.05. The demographic profiles of each sample are compared first. Demography The results of the demographic analysis reveal both similarities and distinct differences between the Early Christian and Middle to Late Byzantine skeletal samples. The age and sex distributions for each sample are reproduced in Table 7.52 below. Earlier it was mentioned that one of the points of similarity between the two samples was their age at death distributions. As can been seen in Table 7.52, the numbers of individuals are similar or even match in every age category except for the generic “adult” category. The differences apparent in that category can be explained by the difficulties of establishing narrower age ranges for adults represented by incomplete skeletons or isolated bones from commingled graves. Given the fact that the individuals represented by these samples lived roughly half a millennium apart, the degree of concordance in the number of individuals in each age at death category is extraordinary. Table 7.52: Age and Sex Demographics of the Early Christian (EC) and Middle to Late Byzantine (B) Samples. Age Category Male Female Ambiguous Undetermined EC B EC B EC B EC B Fetal (1 91 85.8 33 84.6 TOTAL 106 100.0 39 100.0 A comparison of MNI categories by sex shows that the proportions of Early Christian males and females in each category are quite similar (Table 8.24). Regardless of sex, most individuals were buried in single graves. Comparable proportions of males and females were buried in multiple burials. Table 8.24: Frequency of Early Christian Males and Females by Grave MNI. MNI Males Females (n) % (n) % 1 34 79.1 29 78.4 2 5 11.6 3 8.1 3 4 9.3 5 13.5 TOTAL 43 100.0 37 100.0 In general, males and females buried during the Middle to Late Byzantine period exhibit the same pattern with regard to grave MNI. Especially in the case of burials containing one and two individuals, the proportions of males and females in categories of MNI do not show large differences (Table 8.25). 230 Table 8.25: Frequency of Middle to Late Byzantine Males and Females by Grave MNI. MNI Males Females (n) % (n) % 1 11 16.2 4 16.0 2 22 32.4 9 36.0 3 12 17.6 2 8.0 4 4 5.9 2 8.0 5 9 13.2 2 8.0 6 1 1.5 0 0.0 8 5 7.4 2 8.0 16 4 5.9 4 16.0 TOTAL 68 100.0 25 100.0 When the categories of grave MNI are collapsed into single and multiple burials, the proportions of males and females are very nearly identical (Table 8.26). This clearly demonstrates that there is no association between sex and grave MNI in the Middle to Late Byzantine period. Table 8.26: Males and Females in Single versus Multiple Burials in the Middle to Late Byzantine Period. MNI Males Females 2 (n) % (n) % df Two-tail p <0.05 χ p-value Yes/ No 1 11 16.2 4 16.0 0.000 1 0.984 No >1 57 83.8 21 84.0 TOTAL 68 100.0 25 100.0 Grave Orientation The analyses of spatial orientation in this dissertation focus on the graves themselves rather than on the bodies within them. This is because information on the orientation of the bodies was often either unavailable or problematic, such as in the case of commingled burials. As discussed in Chapter Six, there is almost no variation whatsoever in the orientation of the graves at Nemea. Aside from a single clear example of a burial aligned north-south, all graves were oriented east-west. 231 Figure 8.9 shows the orientation of the Early Christian graves. The single exception to the pattern of east-west oriented graves is dated to this period and is located southwest of the Hellenistic bath. Interestingly, this is one of the few commingled graves dating to the Early Christian period. It was found to contain a minimum of three individuals: two young adult females and one adolescent of indeterminate sex. Among all other graves, which are oriented east-west, the only observable variation is that graves associated with the temple and the basilica show a tendency toward less precision in their orientation. For example, some of the graves northwest of the temple follow the orientation of that structure, which is not precisely east-west. By contrast, the graves associated with the bath are generally much more regular and accurate in their east-west orientation. There is one Early Christian grave for which the orientation is unknown. It is, in fact, not truly a grave but another example of a unique occurrence in the mortuary space at Nemea. An isolated, fragmentary cranium was discovered northwest of the temple and is represented in Figure 8.9 by the small hollow shape near the temple’s northwest corner. In the excavation notebook, the cranium is described as wedged between fallen temple blocks. The orientation of the graves dated to the Middle to Late Byzantine period is illustrated in Figure 8.10. Once again, all graves dated to this period were oriented east-west. Some variation in the precision of that orientation is apparent, most notably in the graves within the nave of the basilica. 232 Figure 8.9: Grave Orientation of Early Christian Burials. 233 Figure 8.10: Grave Orientation of Middle to Late Byzantine Burials. 234 Head Treatment Some burials at Nemea showed evidence of additional attention given to the head of the deceased. In some cases, a stone or tile “pillow” was placed underneath the head. Other variations on this practice included enclosing the head with stones or tiles. Each grave containing an individual with head treatment was placed into one of the following categories, which are based on variation in the manner and type of material used in this practice: stone pillow, enclosed with stone, tile pillow, enclosed with tile, and mixed stone and tile enclosure. Graves were also categorized more generally based on the presence or absence of head treatment, although this was somewhat problematic because observations on head treatment were typically recorded in the excavation notebooks only when it was present—the absence of head treatment was not necessarily recorded. If no statement was made about the presence or absence of head treatment, then the grave was placed into the category “unknown.” The spatial patterns in the maps below were evaluated bearing in mind that, while it is possible that some of the “unknown” burials did contain examples of head treatment, it is likely that most of them did not. The presence of head treatment seems to have been an unusual finding and it is probably the case that the graves for which there is no description of elaboration associated with the head truly lacked evidence of that practice. Clear spatial patterns in the presence of graves in which increased attention was given to the head have been demonstrated by Frey (1998) and are reconfirmed in this analysis. Based on the available data, the emphasis on propping up or protecting the head of the deceased was apparent only among burials associated with the basilica. A small number of burials dated to the Early Christian period by Miller (1988) contained individuals exhibiting some form of head treatment. Those four burials are illustrated in Figure 8.11 below. Each of the four graves is 235 located within the boundaries of the Early Christian houses immediately south of the basilica. From west to east, two of the graves are located in house one, one of them is in house two, and the remaining grave is in house four. A high/low cluster analysis carried out using ArcGIS 10.0 confirms that Early Christian burials exhibiting head treatment are more highly clustered than would be expected by chance (p = 0.027; z-score = 2.22). 236 Figure 8.11: Presence or Absence of Head Treatment in Early Christian Burials. 237 The type of head treatment used in these four burials is illustrated in Figure 8.12. It is interesting to note that, although two different styles of head treatment were employed, tiles were the only material used in the Early Christian burials. Three graves in which the occupant’s head rested on a tile pillow are located close to one another in houses one and two. The grave in which the occupant’s head was enclosed with tiles is located in another small cluster of graves in house four. Using the available data, it can be determined that at least three of the four Early Christian individuals exhibiting head treatment were adults. Among the three tile pillow graves, all of which contained a single individual, the two individuals buried closest together in house one were middle adult males and the individual buried to the east in house two was an older adult female. The picture for the larger grave located in house four is more complicated. It was found to contain the remains of three individuals: a child around nine years of age, a middle adult male, and a middle adult female. While it is clear from the notebook description that the head of one of those individuals was enclosed by tiles, there is not enough information to determine exactly which individual is being referred to. Based on the small number of Early Christian graves exhibiting head treatment then, it is apparent only that this practice was not exclusive to either males or females. It is possible that only adults received such treatment or at least that it was more common in the burials of adults. 238 Figure 8.12: Spatial Distribution of Head Treatment Types in Early Christian Burials. Increased attention given to the head of the deceased seems to have been more common during the Middle to Late Byzantine period, as eleven graves dating to that period were recorded as exhibiting some form of head treatment. The spatial distribution of head treatment presence also demonstrates that individuals with head treatment were concentrated in a different area during the Middle to Late Byzantine period (Figure 8.13). Nearly all of the Middle to Late 239 Byzantine graves containing individuals with head treatment are found outside of the basilica near its east end. The single exception is located centrally in the nave. A high/low cluster analysis indicates an absence of significant clustering of Middle to Late Byzantine graves exhibiting head treatment (p = 0.149; z-score = -1.44). Figure 8.13: Presence or Absence of Head Treatment in Middle to Late Byzantine Burials. 240 The results of recorded examples of head treatment are organized by type in Table 8.27 and illustrated spatially in Figure 8.14. In contrast to the Early Christian burials, which utilized only tiles, stone pillows or enclosures were used frequently during the Middle to Late Byzantine period. Enclosing the head with stone was the most frequent type of elaboration, followed by tile and stone pillows. In one case an individual’s head was protected using a combination of tiles and stones. Table 8.27: Frequency of Head Treatment Types in Middle to Late Byzantine Burials. Type of Head Treatment Middle to Late Byzantine (n) % Stone Pillow 2 18.2 Enclosed with Stone 5 45.5 Tile Pillow 3 27.3 Mixed Tile and Stone 1 9.1 TOTAL 11 100.0 While again the small number of burials for which a type of head treatment could be identified precludes any broad statements about spatial patterning, it can be said that graves containing individuals with similar types of head treatment do appear in close proximity to one another. As may have been the case for the Early Christian burials, all Middle to Late Byzantine individuals exhibiting head treatment were adults. Assessing a relationship between sex and head treatment is again problematic. Two of the burials in this period contained multiple individuals and the excavation notebooks refer to only one of the skeletons as having some form of elaboration focused on the head. As the analysis of sex is only now completed, it is unclear which individual received that treatment. Among graves containing a single individual whose remains were available for analysis, two males and three females were given head treatment. 241 Figure 8.14: Spatial Distribution of Head Treatment Types in Middle to Late Byzantine Burials. 242 Grave Goods Grave goods were uncommon among the burials at Nemea. However, some graves did contain items, such as coins and ceramic vessels, which fall under the category of grave goods or mortuary artifacts. Graves at Nemea also occasionally contained objects more appropriately classified as items of personal adornment, such as finger rings, earrings, crosses, and belt buckles. In this analysis, each grave was assessed for the presence of grave goods—in this case a general category that included either mortuary artifacts or items of personal adornment. Graves were also classified more specifically into one of the following categories, based on the type of object or objects they contained: items of personal adornment, mortuary artifacts, and both items of personal adornment and mortuary artifacts. As was the case for the analysis of head treatment and grave orientation, the reasons for which the analysis of grave goods was carried out primarily by grave rather than by individual should be made clear. First, some of the data on grave goods was gleaned from burials containing human remains that were not saved. Thus, estimates of age and sex are not available in all cases. Second, the phenomenon of large and in some cases extensively commingled burials makes it very difficult, especially decades after excavation, to determine whether or not a particular item in a grave was associated with a particular individual. As a result of these issues, it is practical to use the data on grave goods in conjunction with data on age and sex only in the case of single inhumations. While this approach provides results that are more limited in scope, it is nevertheless useful as an avenue of exploration. When compared by period, the proportions of graves containing grave goods are very similar (Table 8.28). Around 20% of graves in each period contained either mortuary artifacts or 243 items of personal adornment. The presence of grave goods in Middle to Late Byzantine graves (23.7%; 18/76) is slightly greater than that in Early Christian graves (20.1%; 28/139). Table 8.28: Presence or Absence of Graves Goods in Each Period. Grave Goods Early Christian Middle to Late Byzantine Total (n) % (n) % (n) % Present 28 20.1 18 23.7 46 21.4 Absent 111 79.9 58 76.3 169 78.6 TOTAL 139* 100.0 76* 100.0 215 100.0 *The presence or absence of grave goods was unknown for six Early Christian and seven Middle to Late Byzantine graves. The spatial distribution of grave goods among the Early Christian burials is outlined by burial area in Table 8.29 and illustrated in Figure 8.15. Among burials associated with the temple, grave goods are concentrated in graves closest to the northwest corner. The burials to the north of the temple but located near the middle of the structure contained no evidence of grave goods. Similarly, only one of the nine graves clustered to the south of the temple contained grave goods. The burials associated with the bath show few clear patterns when the presence of grave goods is considered, although all clusters of graves in that area contain examples of burials with grave goods. One such burial that is of additional interest is the relatively isolated grave located southwest of the bath. It is a commingled burial containing a minimum of three individuals (two young adult females and one adolescent of indeterminate sex) and, as described previously, it is also the only grave at Nemea with a north-south orientation. The grave was covered with stones and a large pot was placed directly at its north end at the level of the covering. Fragments of the base of a second large pot were found within the grave, associated with the remains of one of the occupants. The characteristics of this grave—its location, orientation, and contents—are unique among the Early Christian graves at Nemea and perhaps suggest a different religious or ethnic group affiliation for the occupants. 244 Many examples of burials containing grave goods were found near the Early Christian basilica. All burials located to the west of the basilica and most of those located to the south within the Early Christian houses contained either mortuary artifacts or items of personal adornment. If the presence of grave goods is compared between the general burial areas (temple, bath, basilica), the largest proportion of graves containing those items is found among the burials associated with the basilica (Table 8.29). A chi-square test shows that the differences between 2 the areas are statistically significant (χ = 12.665, df = 2, p = 0.002), although one cell has an expected count of less than five. As a result, Fisher’s exact test was used when necessary to compare the differences between two areas at a time. The differences between the graves associated with the temple and the bath (p = 0.359) and between those associated with the temple 2 and the basilica (χ = 3.429, df = 1, p = 0.064) are not significant. However, a significant association was found when the presence or absence of grave goods was compared between the graves associated with the bath and those associated with the basilica (p = 0.001). Table 8.29: Frequency and Distribution of Grave Goods in Early Christian Burials. Grave Goods Temple Bath Basilica (n) % (n) % (n) % Present 6 22.2 12 13.2 10 47.6 Absent 21 77.8 79 86.8 11 52.4 TOTAL 27* 100.0 91* 100.0 21 100.0 *The presence or absence of grave goods was unknown for six Early Christian burials. 245 Figure 8.15: Spatial Distribution of the Presence of Grave Goods in Early Christian Burials. 246 The spatial distribution of Middle to Late Byzantine burials containing grave goods is not dissimilar to that of burials exhibiting some form of head treatment (Figure 8.16). Grave goods were frequently found among the burials outside of the basilica that lie immediately to the east of that structure. A few burials within the nave of the basilica as well as some that lie either along or within its southern walls also contained grave goods. A high/low cluster analysis indicates that Middle to Late Byzantine burials with grave goods are not clustered in space to a greater degree than would be expected by chance (p = 0.939; z-score = -0.077). 247 Figure 8.16: Spatial Distribution of the Presence of Grave Goods in Middle to Late Byzantine Burials. The spatial distribution of graves containing grave goods was also examined with regard to the type of items found within the graves. Although the number of Early Christian burials 248 containing grave goods is small, certain patterns are apparent in their distribution—especially if burial areas are compared against each other (Table 8.30). The area around the temple exhibits the lowest proportion of graves containing items of personal adornment and the greatest proportion of graves containing mortuary artifacts. Moving to the graves associated with the bath, there is an increase in the proportion of items of personal adornment, while the proportion of graves containing mortuary artifacts decreases. Finally, grave goods among the burials associated with the basilica are exclusively items of personal adornment. Fisher’s exact test demonstrates that the differences between temple and basilica burials in the proportions of graves with items of personal adornment and those containing mortuary artifacts are statistically significant (p = 0.022). Table 8.30: Frequency and Distribution of Grave Good Types in Early Christian Burials. Grave Good Type Temple Bath Basilica (n) % (n) % (n) % Items of Personal Adornment 2 33.3 9 75.0 10 100.0 Mortuary Artifacts 3 50.0 3 25.0 0 0.0 Items of Personal Adornment 1 16.7 0 0.0 0 0.0 and Mortuary Artifacts TOTAL 6 100.0 12 100.0 10 100.0 It is possible that the differences in the types of grave goods between burial areas are a reflection of the development of the mortuary space of Nemea through time. Based on criteria such as datable finds, relationships to datable architecture, and variation in grave construction, Frey (1998) has suggested a progression through time for the burials in these three main areas. In this model, the burials associated with the temple are the earliest, those concentrated around the bath occupy a transitional position, and those associated with the basilica are the latest in time. The objects discovered within the graves also offer potential clues in support of this chronology. For example, one of the burials located near the northwest corner of the temple (colored orange in Figure 8.17) is particularly unique because of its apparent wealth. In stark contrast to the rest 249 of the burials at Nemea, the young adult female occupying this grave was buried with a number of mortuary artifacts including bronze items, an incised stone plaque, and a coin, in addition to many pieces of jewelry (Frey 1998; Miller 1981). Frey (1998) has posited that this grave could be an example of a burial carried out with Christian and pagan syncretism. The presence of mortuary artifacts such as pots and fragments of bronze in two adjacent graves perhaps lends support to that argument. In terms of broader patterns, the suggestion that the graves associated with the temple are the earliest at Nemea seems even more likely when viewed in the context of the distribution of grave good types across the site. 250 Figure 8.17: Spatial Distribution of Grave Good Types in Early Christian Burials. 251 Among Middle to Late Byzantine burials, items included within the graves were most often items of personal adornment (Table 8.31). Four graves contained mortuary artifacts that included coins, a cooking pot, and a sgraffito bowl. As in the case of the Early Christian burials, one grave contained both items of personal adornment and mortuary artifacts, although the contents of this grave, a bronze finger ring and a bronze needle, were nowhere near as copious as those of the Early Christian grave described above. Coincidentally, this Middle to Late Byzantine grave also contained the skeletal remains of a young adult female. The spatial distribution of grave good types during this period is illustrated in Figure 8.18. No obvious clustering of any particular type is apparent upon visual inspection. Graves containing items of personal adornment as well as those containing mortuary artifacts were located both inside and outside of the basilica. The greatest concentrations of the former appear in the southwest portion of the nave and along the exterior walls of the southeast corner. The latter are dispersed widely in the area of the basilica. The only grave containing both classes of grave goods is located outside of the apse at the basilica’s east end. Table 8.31: Frequency of Grave Good Types in Middle to Late Byzantine Burials. Grave Good Type Middle to Late Byzantine (n) % Items of Personal Adornment 13 72.2 Mortuary Artifacts 4 22.2 Items of Personal Adornment 1 5.6 and Mortuary Artifacts TOTAL 18 100.0 252 Figure 8.18: Spatial Distribution of Grave Good Types in Middle to Late Byzantine Burials. As discussed at the beginning of this section, the possibility of an association between the presence of grave goods and age at death was investigated only for those burials with a MNI of one. While this conservative approach excludes a number of graves, mostly dating to the Middle to Late Byzantine period, it ensures the ability to discern a relationship between the individual and the items contained within each grave considered. 253 Among Early Christian single burials, the proportions of adults and subadults buried with either items of personal adornment or mortuary artifacts are almost identical (Table 8.32). There is no association between age at death and the presence or absence of grave goods in this period. Table 8.32: Presence of Grave Goods in Early Christian Adult and Subadult Burials. Grave Goods Adults Subadults 2 (n) % (n) % df Two-tail p <0.05 χ p-value Yes/ No Present 14 24.6 7 25.0 0.002 1 0.965 No Absent 43 75.4 21 75.0 TOTAL 57 100.0 28 100.0 The proportion of subadults buried with grave goods is greater than that of adults among single inhumations dated to the Middle to Late Byzantine period, although the subsample sizes are quite small (Table 8.33). Table 8.33: Presence of Grave Goods in Middle to Late Byzantine Adult and Subadult Burials. Grave Goods Adults Subadults 2 (n) % (n) % χ * df Two-tail p <0.05 p-value Yes/ No Present 2 14.3 2 33.3 1 0.549 No Absent 12 85.7 4 66.7 TOTAL 14 100.0 6 100.0 *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. If all burials from the Middle to Late Byzantine period are considered regardless of the number of individuals they contained, then the elevated proportion of subadults buried with grave goods is still apparent (Table 8.34). However, the differences between those age categories remain non-significant. Table 8.34: Presence of Grave Goods in All Middle to Late Byzantine Burials by Age. Grave Goods Adults Subadults 2 (n) % (n) % df Two-tail p <0.05 χ p-value Yes/ No Present 16 15.7 10 25.6 1.859 1 0.173 No Absent 86 84.3 29 74.4 TOTAL 102 100.0 39 100.0 254 Age based patterns in the types of grave goods recovered from the Nemea burials were also investigated. In the Early Christian period, adults were buried with all categories of grave goods, while objects included in subadult graves were exclusively items of personal adornment such as earrings (Table 8.35). It is possible that this pattern is suggestive of a preference for subadults to be buried with such items, but more examples of graves containing grave goods would be necessary to confirm that interpretation. Using the available figures, the differences between adults and subadults are not significant even when the adult buried with both categories of grave goods is collapsed into the category of mortuary artifacts (p = 0.521). Table 8.35: Frequency of Early Christian Adults and Subadults by Grave Good Type. Grave Good Type Adults Subadults (n) % (n) % Items of Personal Adornment 11 78.6 7 100.0 Mortuary Artifacts 2 14.3 0 0.0 Items of Personal Adornment 1 7.1 0 0.0 and Mortuary Artifacts TOTAL 14 100.0 7 100.0 The problem of small subsample sizes is even more serious for Middle to Late Byzantine burials. Among single inhumations with grave goods, one adult was buried with an item of personal adornment and another was buried with an item of personal adornment as well as a mortuary artifact. Each of the Middle to Late Byzantine subadults with grave goods was buried with items of personal adornment. Together with the data from the Early Christian graves, these results may at least hint at an association between subadult status and the inclusion of items of personal adornment in burials throughout the Byzantine period at Nemea. The distribution of grave goods among males and females was also investigated, again with a focus on single inhumations. In the Early Christian period, the proportions of males and females buried with grave goods are very similar (Table 8.36). There is no association between sex and the presence or absence of grave goods. 255 Table 8.36: Presence of Grave Goods in Early Christian Male and Female Burials. Grave Goods Males Females 2 (n) % (n) % df Two-tail p <0.05 χ p-value Yes/ No Present 8 24.2 6 23.1 0.011 1 0.917 No Absent 25 75.8 20 76.9 TOTAL 33 100.0 26 100.0 The pattern is quite different among single inhumations dated to the Middle to Late Byzantine period. In this case, only females were associated with the presence of either mortuary artifacts or items of personal adornment (Table 8.37). Fisher’s exact test demonstrates that the difference between males and females in the presence or absence of grave goods is statistically significant. Table 8.37: Presence of Grave Goods in Middle to Late Byzantine Male and Female Burials. Grave Goods Males Females 2 (n) % (n) % χ * df Two-tail p <0.05 p-value Yes/ No Present 0 0.0 2 66.7 1 0.033 YES Absent 11 100.0 1 33.3 TOTAL 11 100.0 3 100.0 *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. The broader significance of this finding is difficult to evaluate because of the small sizes of the subsamples involved. If all Middle to Late Byzantine burials are considered, then the proportions of males and females associated with grave goods are similar. However, this approach introduces a different type of bias—the greater number of males in the sample together with many multiple burials means that the proportion of males associated with grave goods is exaggerated. For that reason, it is perhaps best to tentatively suggest that an association existed between females and the presence of grave goods, based on the figures for single inhumations. Finally, single burials containing grave goods were examined for associations between grave good type and the sex of the occupants. In the Early Christian period, items of personal 256 adornment were the only grave good type found in male burials. In contrast, females were associated with all categories of grave goods (Table 8.38). The differences between males and females, however, are not statistically significant even if the female buried with both grave good types is placed into either individual category. Table 8.38: Frequency of Early Christian Males and Females by Grave Good Type. Grave Good Type Males Females (n) % (n) % Items of Personal Adornment 8 100.0 3 50.0 Mortuary Artifacts 0 0.0 2 33.3 Items of Personal Adornment 0 0.0 1 16.7 and Mortuary Artifacts TOTAL 8 100.0 6 100.0 Among Middle to Late Byzantine single inhumations containing grave goods, one female was associated with items of personal adornment and the other was associated with both items of personal adornment and mortuary artifacts. Spatial Analysis of Demography and Physiological Stress Indicators In this section, the analysis of spatial and temporal patterns in the mortuary space of Nemea sharpens its focus on the occupants of the graves. Here I incorporate the results of the skeletal analysis in order to examine the spatial distribution of age classes and sex categories, as well as spatial patterns in the presence and absence of variables associated with health status. This portion of the mortuary analysis is concerned with identifying spatial variability in patterns of health that may provide insight into the organization of the mortuary space. Based on the possibility that individuals of higher status experienced lower levels of physiological stress and were granted access to privileged burial locations, the results presented in this section will test the hypothesis that the burial space at Nemea was organized according to the social hierarchy. 257 Demography and Burial Location Age A visual inspection of the spatial distribution of age at death among the Early Christian burials across the site as a whole reveals no obvious clustering of adults or subadults. Adults and subadults were found in graves in all areas of the site, which strongly suggests that no particular burial area was reserved for children or adults. An absence of significant spatial clustering of adults or subadults is confirmed using a high/low cluster analysis (p = 0.390; z-score = -0.859). When burial areas are compared against each other, the burials associated with the temple were found to contain an elevated proportion of adults and a lower proportion of subadults (Table 8.39). However, a chi-square test demonstrates that there are no statistically significant 2 differences in the frequencies of adults and subadults between burial areas (χ = 3.708, df = 2, p = 0.157). This holds true when temple burials are compared with bath and basilica burials independently (Tables 8.40 and 8.41). Table 8.39: Frequency and Distribution of Adults and Subadults in Early Christian Burials. Age Category Temple Bath Basilica (n) % (n) % (n) % Adult 19 86.4 44 64.7 16 69.6 Subadult 3 13.6 24 35.3 7 30.4 TOTAL 22 100.0 68 100.0 23 100.0 Table 8.40: Frequency of Adults and Subadults in Temple versus Bath Burials. 2 Age Category Temple Bath df Two-tail p <0.05 χ p-value Yes/ No (n) % (n) % Adult 19 86.4 44 64.7 3.713 1 0.054 No Subadult 3 13.6 24 35.3 TOTAL 22 100.0 68 100.0 258 Table 8.41: Frequency of Adults and Subadults in Temple versus Basilica Burials. Age Category Temple Basilica χ2* df Two-tail p <0.05 p-value Yes/ No (n) % (n) % Adult 19 86.4 16 69.6 1 0.284 No Subadult 3 13.6 7 30.4 TOTAL 22 100.0 23 100.0 *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. When particular burial areas are viewed at a micro level, a small number of clusters of adult burials are apparent. For example, based on the skeletal remains available for analysis, the cluster of burials located immediately south of temple contained only adults (Figure 8.19). Figure 8.19: Spatial Distribution of Age at Death among Temple Burials. Similarly, certain rows of graves within the burial clusters located to the northeast of the bath included only adults (Figure 8.20). 259 Figure 8.20: Spatial Distribution of Age at Death among Bath Burials. A general absence of clustering is also observed upon visual inspection when Middle to Late Byzantine burials are examined (Figure 8.21). This is confirmed by a high/low cluster analysis, which demonstrates that neither adults nor subadults are clustered in space to a greater degree than would be expected by chance (p = 0.448; z-score = 0.759). Adults and subadults 260 were found both within and outside of the basilica and neither age group shows an overwhelming concentration in those areas. A chi-square test confirms that there is no association between adult and subadult status and burial location within or outside of the basilica (Table 8.42). Figure 8.21: Spatial Distribution of Age at Death among Middle to Late Byzantine Burials. 261 Table 8.42: Frequency of Adults and Subadults Inside versus Outside of the Basilica. 2 Age Category Inside Outside df Two-tail p <0.05 χ p-value Yes/ No (n) % (n) % Adult 31 67.4 75 75.0 0.917 1 0.425 No Subadult 15 32.6 25 25.0 TOTAL 46 100.0 100 100.0 As was demonstrated previously but is apparent again in Figures 8.21 and 8.22, burials containing multiple individuals frequently included both adults and subadults. This pattern may suggest that graves were used as family tombs during this period. Figure 8.22: Detail of Age Distribution among Middle to Late Byzantine Burials in the Nave. Sex The spatial distribution of sex was also compared across burial areas using both visual inspection and statistical analysis. Among Early Christian graves containing human remains for 262 which sex could be estimated, the distribution of males and females appears unremarkable (Figures 8.23 and 8.24). As was the case for adults and subadults, males and females do not show a tendency to concentrate in any particular area of the site. A high/low cluster analysis indicates that neither sex is more spatially clustered than would be expected by chance (p = 0.916; z-score = 0.359). Furthermore, the proportions of males and females in each burial area are remarkably consistent (Table 8.43). A chi-square test demonstrates that there are no 2 significant differences in the frequencies of males and females between those areas (χ = 0.110, df = 2, p = 0.947). Table 8.43: Spatial Distribution of Sex in Early Christian Burials. Age Category Temple Bath Basilica (n) % (n) % (n) % Male 10 55.6 24 52.2 9 56.2 Female 8 44.4 22 47.8 7 43.8 TOTAL 18 100.0 46 100.0 16 100.0 A few examples of potentially interesting patterns are visible within burial clusters. One of those concerns the two neat rows of burials immediately south of the temple, discussed above because the skeletons from those graves are all adults (Figure 8.23). This somewhat isolated group of graves contains the remains of five males, two females, and one individual of indeterminate sex represented only by an additional set of permanent dentition. The northernmost grave of each row is occupied by a female and it is tempting to hypothesize that each row or perhaps the entire group of graves is representative of a family burial plot. 263 Figure 8.23: Spatial Distribution of Sex among Temple Burials. The other interesting sex-based pattern in the spatial arrangement of the Early Christian graves appears in the cluster of graves immediately north of the bath (Figure 8.24). Within this group, all individuals available for analysis from the easternmost row of graves were determined to be female. The osteotheke, which was discussed earlier, lies immediately to the west of this row of burials. That structure was found to contain the fragmentary remains of a minimum of three individuals—all consistent with males based on metric assessments of robusticity. 264 Figure 8.24: Spatial Distribution of Sex among Bath Burials. Among Middle to Late Byzantine burials, there appears to be a relative scarcity of females to the immediate south and southwest of the basilica (Figure 8.25). However, a high/low cluster analysis indicates that there is an absence of significant spatial clustering of either males or females during this period (p = 0.722; z-score = -0.355). Males and females were found within 265 the basilica (Figure 8.26) as well as in all areas along its external walls, if in lower concentrations in the case of females to the south and southwest of that structure. A chi-square test demonstrates that there is no association between sex and burial location with respect to the basilica in this period (Table 8.44). Figure 8.25: Spatial Distribution of Sex among Middle to Late Byzantine Burials. 266 Figure 8.26: Detail of Sex Distribution among Middle to Late Byzantine Burials in the Nave. Table 8.44: Frequency of Males and Females Inside versus Outside of the Basilica. 2 Sex Inside Outside df Two-tail p <0.05 χ p-value Yes/ No (n) % (n) % Male 18 72.0 50 73.5 0.022 1 0.883 No Female 7 28.0 18 26.5 TOTAL 25 100.0 68 100.0 Physiological Stress and Burial Location Linear Enamel Hypoplasias As discussed in Chapters Six and Seven, the prevalence rates of linear enamel hypoplasias were calculated and analyzed by tooth rather than by individual in this research. As the analysis of the spatial distribution of linear enamel hypoplasias for all anterior teeth independently would be cumbersome and unlikely to yield different results, the results for only 267 the mandibular canines are used. The mandibular canines were chosen because they have been demonstrated to be among the most susceptible to hypoplasia formation (Goodman and Rose 1990), a conclusion supported by the results from Nemea. The spatial distribution of the presence of linear enamel hypoplasias affecting the left mandibular canines (tooth #22) among Early Christian burials is illustrated in Figures 8.27, 8.28, and 8.29. As is clear upon visual inspection as well as from the figures in Table 8.45 below, individuals from all three burial areas show moderate to high prevalence rates. The presence of linear enamel hypoplasias does not appear to be concentrated in any one of those areas. An absence of significant spatial clustering of either the presence or the absence of the condition is confirmed through a high/low cluster analysis (p = 0.413; z-score = 0.819). 268 Figure 8.27: Spatial Distribution of Linear Enamel Hypoplasias on Left Mandibular Canines among Temple Burials. 269 Figure 8.28: Spatial Distribution of Linear Enamel Hypoplasias on Left Mandibular Canines among Bath Burials. 270 Figure 8.29: Spatial Distribution of Linear Enamel Hypoplasias on Left Mandibular Canines among Early Christian Basilica Burials. Table 8.45 presents the frequency data related to the spatial distribution of the presence of linear enamel hypoplasias on observable left mandibular canines. While small subsample sizes make the evaluation of spatial patterns difficult, individuals buried in association with the bath appear to exhibit fewer hypoplastic left mandibular canines than those buried in association with 271 the temple or the basilica. However, Fisher’s exact test demonstrates that the differences between those burial areas are not statistically significant (Table 8.46). Table 8.45: Spatial Distribution of Linear Enamel Hypoplasias (LEH) on Left Mandibular Canines (#22) among Early Christian Burials. LEH Temple Bath Basilica 2 #22 Pr Ab % Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) (n) (n) p-value Yes/ No 9 2 81.8 19 14 57.6 9 2 81.8 3.524* 2 0.172 No *At least one cell has an expected count less than five. Table 8.46: Comparison of Linear Enamel Hypoplasias (LEH) on Left Mandibular Canines (#22) between Bath and Temple/Basilica Burials. LEH Bath Temple/Basilica** 2 #22 Pr Ab % Pr Ab % χ * df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 19 14 57.6 9 2 81.8 1 0.278 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. **The figures for the temple and basilica burials are identical. A similar pattern is apparent when the right mandibular canines (tooth #27) are considered (Table 8.47). High prevalence rates of linear hypoplasias are found in each burial area and a high/low cluster analysis confirms that neither the presence nor the absence of the condition is clustered in space to a greater degree than would be expected by chance (p = 0.137; z-score = 1.488). Again, individuals from burials located near the bath show a lower prevalence rate of linear enamel hypoplasias when compared with those from burials located near the temple and the basilica. However, Fisher’s exact tests indicate that there is no statistically significant association between burial location and the presence or absence of enamel hypoplasias (Tables 8.48 and 8.49). 272 Table 8.47: Spatial Distribution of Linear Enamel Hypoplasias (LEH) on Right Mandibular Canines (#27) among Early Christian Burials. LEH Temple Bath Basilica 2 #27 Pr Ab % Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) (n) (n) p-value Yes/ No 5 1 83.3 24 11 68.6 12 2 85.7 1.823* 2 0.402 No *At least one cell has an expected count less than five. Table 8.48: Comparison of Linear Enamel Hypoplasias (LEH) on Right Mandibular Canines (#27) between Bath and Temple Burials. LEH Bath Temple #27 Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 24 11 68.6 5 1 83.3 1 0.651 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. Table 8.49: Comparison of Linear Enamel Hypoplasias (LEH) on Right Mandibular Canines (#27) between Bath and Basilica Burials. LEH Bath Basilica #27 Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 24 11 68.6 12 2 85.7 1 0.297 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. Among Middle to Late Byzantine burials containing individuals with observable left mandibular canines (tooth #22), the presence of linear enamel hypoplasias does not appear to cluster in any particular area inside or outside of the basilica (Figure 8.30). As was the case for the Early Christian burials, most left mandibular canines exhibited at least one hypoplasia. A high/low cluster analysis confirms that there is no statistically significant spatial clustering of the presence or absence of linear enamel hypoplasias on this tooth (p = 0.671; z-score = -0.425). The same pattern is observed when the corresponding right mandibular canine (tooth #27) is considered. The results for this tooth also indicate that neither the presence nor the absence of linear enamel hypoplasias is spatially clustered to a greater degree than would be expected by chance (p = 0.387; z-score = 0.865). 273 Figure 8.30: Spatial Distribution of Linear Enamel Hypoplasias on Left Mandibular Canines among Middle to Late Byzantine Burials. A comparison of the prevalence of linear enamel hypoplasias on the left and right mandibular canines between Middle to Late Byzantine burials located inside and outside of the basilica reveals a potentially interesting pattern, although small subsample sizes once again 274 prove problematic. For each of those teeth, burials located within the basilica show greater prevalence rates than those located outside of the basilica. All left mandibular canines belonging to individuals buried inside the basilica were hypoplastic, while 92.9% (13/14) of right mandibular canines from such burials exhibited linear enamel hypoplasias (Tables 8.50 and 8.51). Figure 8.31 shows a detailed view of the distribution of the presence and absence of hypoplasias on the latter tooth among burials in the nave of the basilica. Despite this apparent pattern, Fisher’s exact test demonstrates that there is no association between burial inside or outside of the basilica and the presence or absence of linear enamel hypoplasias for both the left and right mandibular canines (Tables 8.50 and 8.51). Table 8.50: Comparison of Linear Enamel Hypoplasias (LEH) on Left Mandibular Canines (#22) Inside versus Outside of the Basilica. LEH Inside Outside #22 Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 4 0 100.0 18 8 69.2 1 0.550 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. Table 8.51: Comparison of Linear Enamel Hypoplasias (LEH) on Right Mandibular Canines (#27) Inside versus Outside of the Basilica. LEH Inside Outside #27 Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 13 1 92.9 15 6 71.4 1 0.203 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. 275 Figure 8.31: Detail of the Spatial Distribution of Linear Enamel Hypoplasias on Right Mandibular Canines among Middle to Late Byzantine Burials in the Nave. 276 Porotic Hyperostosis As demonstrated in Chapter Seven, porotic hyperostosis was not common in the Early Christian skeletal sample and this is represented visually in Figures 8.32, 8.33, and 8.34. Upon visual inspection, individuals exhibiting porotic hyperostosis appear to be concentrated around the temple. Only a few individuals associated with the bath exhibit porotic hyperostosis and among basilica burials only a single case was recorded. A high/low cluster analysis demonstrates that the spatial distribution of the presence and absence of porotic hyperostosis among Early Christian burials across the site does not differ significantly from what would be expected by chance (p = 0.989; z-score = 0.014). 277 Figure 8.32: Spatial Distribution of Porotic Hyperostosis among Temple Burials. 278 Figure 8.33: Spatial Distribution of Porotic Hyperostosis among Bath Burials. 279 Figure 8.34: Spatial Distribution of Porotic Hyperostosis among Early Christian Basilica Burials. The figures for the presence and absence of porotic hyperostosis in Early Christian burial areas are compared in Table 8.52. Again, the proportion of temple burials containing crania exhibiting porotic hyperostosis is elevated when viewed against the burials associated with the bath and the basilica. To avoid violating the assumptions of the chi-square test, those burial areas were compared against each other using Fisher’s exact test. Despite the apparent pattern, the differences between the temple burials and those associated with the bath and the basilica only approach the p <0.05 level (Tables 8.53 and 8.54). 280 Table 8.52: Spatial Distribution of Porotic Hyperostosis (PH) among Early Christian Burials. PH Temple Bath Basilica 2 Pr Ab % Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) (n) (n) p-value Yes/ No 5 14 26.3 3 40 7.0 1 19 5.0 6.011* 2 0.050 No *At least one cell has an expected count less than five. Table 8.53: Comparison of Porotic Hyperostosis (PH) between Temple and Bath Burials. PH Temple Bath Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 5 14 26.3 3 40 7.0 1 0.050 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. Table 8.54: Comparison of Porotic Hyperostosis (PH) between Temple and Basilica Burials. PH Temple Basilica Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 5 14 26.3 1 19 5.0 1 0.091 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. Among Middle to Late Byzantine burials, most individuals exhibiting porotic hyperostosis are located outside of the basilica along its south and southeast walls (Figure 8.35). Only three individuals with porotic hyperostosis were recovered from graves within the basilica. There were no examples of porotic hyperostosis in individuals with observable parietal bones from graves located at the east end of the basilica outside of the apse. Despite this apparent pattern in the distribution of the condition, a high/low cluster analysis shows that neither the presence nor the absence of porotic hyperostosis is clustered in space to a greater degree than would be expected by chance (p = 0.274; z-score = 1.093). Similarly, the differences in the presence of porotic hyperostosis between individuals whose graves were inside the basilica and those whose graves were outside of that structure are not statistically significant (Table 8.55). 281 Figure 8.35: Spatial Distribution of Porotic Hyperostosis among Middle to Late Byzantine Burials. Table 8.55: Presence of Porotic Hyperostosis (PH) Inside versus Outside of the Basilica. PH Inside Outside Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 3 23 11.5 12 53 18.5 1 0.541 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. 282 Cribra Orbitalia The spatial distribution of cribra orbitalia among Early Christian burials is illustrated in Figures 8.36, 8.37, and 8.38. The condition was a relatively common observation in skeletons recovered from graves associated with the temple and the bath. Fewer examples of cribra orbitalia were observed among the individuals buried closest to the basilica. No obvious clustering of cribra orbitalia in any particular burial area is apparent upon visual inspection and this is confirmed by a high/low cluster analysis, which indicates an absence of spatial clustering of the presence or absence of the condition (p = 0.655; z-score = -0.447). Figure 8.36: Spatial Distribution of Cribra Orbitalia among Temple Burials. 283 Figure 8.37: Spatial Distribution of Cribra Orbitalia among Bath Burials. 284 Figure 8.38: Spatial Distribution of Cribra Orbitalia among Early Christian Basilica Burials. The figures detailing the presence and absence of cribra orbitalia among Early Christian burials are compared between burial areas in Table 8.56. While the prevalence of cribra orbitalia is lower among individuals recovered from graves associated with the basilica, the differences between those individuals and the subsamples from the temple and the bath are not statistically significant (Tables 8.57 and 8.58). 285 Table 8.56: Spatial Distribution of Cribra Orbitalia (CO) among Early Christian Burials. CO Temple Bath Basilica 2 Pr Ab % Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) (n) (n) p-value Yes/ No 5 9 35.7 11 22 33.3 2 12 14.3 2.051* 2 0.359 No *At least one cell has an expected count less than five. Table 8.57: Comparison of Cribra Orbitalia (CO) between Basilica and Temple Burials. CO Basilica Temple Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 2 12 14.3 5 9 35.7 1 0.385 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. Table 8.58: Comparison of Cribra Orbitalia (CO) between Basilica and Bath Burials. CO Basilica Bath Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 2 12 14.3 11 22 33.3 1 0.288 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. The distribution of cribra orbitalia among individuals buried in Middle to Late Byzantine graves is similar to that of porotic hyperostosis (Figure 8.39). Cribra orbitalia was present in just three individuals from two graves within the basilica—one grave is located in the western portion of the nave and the other in the southern room of the narthex. No examples of cribra orbitalia were observed in individuals buried outside of the basilica adjacent to the apse at its east end. Most of the individuals exhibiting the condition were recovered from graves located outside of the basilica along its south wall and adjacent to the southeast corner. A high/low cluster analysis demonstrates that this pattern is statistically significant—the presence of cribra orbitalia is clustered in space to a greater degree than would be expected by chance (p = 0.032; z-score = 2.139). There is, however, one caveat. When compared to porotic hyperostosis, which showed a very similar but non-significant pattern, the results for cribra orbitalia include more individuals scored as “unobservable” because of missing or unobservable eye orbits. It must be considered 286 then, that the significant result for the spatial clustering of cribra orbitalia might be due to essentially removing some of the outliers in terms of distance from the primary concentration of burials from the analysis. Still, the results for both porotic hyperostosis and cribra orbitalia at least hint at the possibility that there is significant spatial patterning of the presence of those two conditions and it is noteworthy that they both appear most frequently in graves located immediately southeast of the basilica. When the presence of cribra orbitalia is compared between individuals buried inside and outside of the basilica, those buried outside exhibit a slightly elevated prevalence rate. However, Fisher’s exact test demonstrates that the differences between those two subgroups of burials are not statistically significant (Table 8.59). 287 Figure 8.39: Spatial Distribution of Cribra Orbitalia among Middle to Late Byzantine Burials. Table 8.59: Presence of Cribra Orbitalia (CO) Inside versus Outside of the Basilica. CO Inside Outside Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 3 9 25.0 14 30 31.8 1 0.738 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. 288 Periosteal Reactions As discussed in Chapter Seven, periosteal reactions having a minimum degree of expression were extremely common in individuals from both periods at Nemea. Tables 8.60 and 8.61 present the figures for the spatial distribution of periosteal reactions on the left and right tibiae of individuals recovered from Early Christian burials. For both anatomical sides, the prevalence rates of periosteal reactions are quite consistent between each burial area. No statistically significant differences were found when burial areas were compared against each other two at a time using Fisher’s exact test. The widespread distribution of periosteal reactions is confirmed with a high/low cluster analysis, which indicates that neither the presence nor the absence of periosteal reactions is clustered to a greater degree than would be expected by chance (left tibia: p = 0.197; z-score = 1.290; right tibia: p = 0.100; z-score = 1.642). Table 8.60: Spatial Distribution of Periosteal Reactions (PR) on Left Tibiae among Early Christian Burials. PR Temple Bath Basilica 2 L. Tib Pr Ab % Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) (n) (n) p-value Yes/ No 10 3 76.9 13 4 76.5 17 4 81.0 0.135* 2 0.935 No *At least one cell has an expected count less than five. Table 8.61: Spatial Distribution of Periosteal Reactions (PR) on Right Tibiae among Early Christian Burials. PR Temple Bath Basilica 2 R. Tib Pr Ab % Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) (n) (n) p-value Yes/ No 11 2 84.6 13 3 81.2 16 3 84.2 0.076* 2 0.963 No *At least one cell has an expected count less than five. Similarly, the presence of periosteal reactions shows no tendency to cluster in any particular area among graves dating to the Middle to Late Byzantine period. For both anatomical sides, individuals exhibiting tibial periosteal reactions are found within the narthex and nave of the basilica, as well as outside of the basilica along the southeast wall and outside of the apse at 289 the east end. The absence of significant spatial clustering of periosteal reactions is confirmed using a high/low cluster analysis (left tibia: p = 0.805; z-score = -0.247; right tibia: (p = 0.643; zscore = -0.463). When the presence of periosteal reactions on the tibiae is compared by burial location inside versus outside of the basilica, the right and left anatomical sides show opposite patterns. When the left tibia is considered, a greater prevalence rate is found among individuals buried outside of the basilica. For the right tibia, by contrast, periosteal reactions were observed more frequently among individuals buried within the basilica. Chi-square tests confirm that there is no statistically significant relationship between burial location with respect to the basilica and the presence or absence of tibial periostitis for either anatomical side (Tables 8.62 and 8.63). Table 8.62: Presence of Periosteal Reactions (PR) on Left Tibiae Inside versus Outside of the Basilica. PR Inside Outside 2 L. Tib Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) p-value Yes/ No 12 11 52.2 40 21 65.6 1.272 1 0.259 No Table 8.63: Presence of Periosteal Reactions (PR) on Right Tibiae Inside versus Outside of the Basilica. PR Inside Outside 2 R. Tib Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) p-value Yes/ No 18 5 78.3 40 17 70.2 0.537 1 0.464 No Osteoarthritis The spatial distribution of osteoarthritis among Early Christian burials is presented in Table 8.64 below. Statistical comparisons between burial areas are problematic due to the small sizes of the subsamples. The differences between burial areas in the presence and absence of osteoarthritis were found to be significant for the left knee joint. However, the differences for the right knee joint are non-significant and, furthermore, the cell sizes violate the assumptions of the 290 chi-square test. To produce statistically valid figures, comparisons were also made between two burial areas at a time using Fisher’s exact test, for all combinations of burial areas and for every joint. None of those comparisons yielded statistically significant results, which demonstrates that there is no association between burial location and the presence or absence of osteoarthritis in the Early Christian period. Table 8.64: Spatial Distribution of Osteoarthritis (OA) among Early Christian Burials. OA Temple Bath Basilica 2 Joint Pr Ab % Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) (n) (n) p-value Yes/ No L. Shoulder 0 4 0.0 2 7 22.2 4 4 50.0 3.578* 2 0.167 No R. Shoulder 1 6 14.3 5 5 50.0 3 5 37.5 2.291* 2 0.318 No L. Elbow 0 2 0.0 0 8 0.0 2 13 13.3 1.449* 2 0.485 No R. Elbow 0 7 0.0 1 6 14.3 2 9 18.2 1.387* 2 0.500 No L. Wrist 1 4 20.0 0 6 0.0 1 8 11.1 1.235* 2 0.539 No R. Wrist 0 8 0.0 1 11 8.3 3 11 21.4 2.462* 2 0.292 No L. Hip 4 6 40.0 7 12 36.8 2 9 18.2 1.448* 2 0.485 No R. Hip 3 8 27.3 5 11 31.2 3 13 18.8 0.679* 2 0.712 No L. Knee 0 8 0.0 0 6 0.0 4 6 40.0 6.720* 2 0.035 YES R. Knee 0 8 0.0 0 6 0.0 3 10 23.1 3.635* 2 0.162 No L. Ankle 0 5 0.0 0 4 0.0 1 10 9.1 0.861* 2 0.650 No R. Ankle 0 3 0.0 0 3 0.0 0 9 0.0 *At least one cell has an expected count less than five. The presence or absence of osteoarthritis in individuals buried during the Middle to Late Byzantine period was also examined for evidence of spatial patterns. The results of high/low cluster analyses for each joint appear in Table 8.65 below. The only joint for which the presence of osteoarthritis was found to be spatially clustered at a level of statistical significance was the left ankle. However, the number of individuals with affected left ankle joints (2) is extremely small and this result is due to the fact that those individuals happen to come from burials relatively close together south of the basilica. Especially given that no other joint (including the corresponding ankle joint) shows significant results, there is no broader significance to this pattern. 291 Table 8.65: Results of High/low Cluster Analyses of Osteoarthritis (OA) Presence by Joint among Middle to Late Byzantine Burials. Joint Two-tail p <0.05 z-score p-value Yes/ No L. Shoulder 0.115 No 1.575 R. Shoulder 0.670 No -0.426 L. Elbow 0.803 No -0.250 R. Elbow 0.972 No 0.035 L. Wrist/Hand 0.475 No 0.714 R. Wrist/Hand 0.939 No 0.076 L. Hip 0.958 No -0.052 R. Hip 0.772 No -0.290 L. Knee 0.082 No 1.738 R. Knee 0.477 No -0.711 L. Ankle/Foot 0.038 2.076 YES R. Ankle/Foot 0.771 No 0.291 The figures for osteoarthritis were also compared by joint between individuals buried inside and outside of the basilica (Table 8.66). The differences for the presence and absence of that condition approach the p <0.05 level of significance only for the left knee, which is affected more frequently in individuals buried outside of the basilica. However, in addition to this result being non-significant, the figures trend in the opposite direction when the corresponding knee is considered. These results provide no evidence for a relationship between the presence or absence of osteoarthritis and burial either within or outside of the basilica. 292 Table 8.66: Presence of Osteoarthritis (OA) Inside versus Outside of the Basilica. OA Inside Outside 2 Joint Pr Ab % Pr Ab % χ * df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No L. Shoulder 3 7 30.0 6 22 21.4 1 0.673 No R. Shoulder 7 5 58.3 10 20 33.3 1 0.174 No L. Elbow 1 13 7.1 7 32 17.9 1 0.665 No R. Elbow 4 16 20.0 10 29 25.6 1 0.753 No L. Wrist/Hand 3 8 27.3 5 24 17.2 1 0.660 No R. Wrist/Hand 2 8 20.0 6 24 20.0 1 1.000 No L. Hip 4 7 36.4 12 28 30.0 1 0.723 No R. Hip 3 9 25.0 10 23 30.3 1 1.000 No L. Knee 3 11 21.4 18 19 48.6 3.107 1 0.078 No R. Knee 6 3 66.7 14 27 34.1 1 0.130 No L. Ankle/Foot 0 9 0.0 2 25 7.4 1 1.000 No R. Ankle/Foot 2 9 18.2 1 27 3.6 1 0.187 No *Empty cells indicate that Fisher’s exact test was used instead of Pearson’s chi-square test. Antemortem Trauma Due to the relatively small number of fractures affecting any given bone type and anatomical side, this analysis focuses on the spatial location of graves containing individuals with fractures rather than on the distribution of fractures by bone. The figures in this section also include the presence of fractures on less substantial skeletal elements (e.g., vertebrae, metacarpals) that were not considered in the trauma analysis presented in the skeletal analysis results chapter. The spatial distribution of fractures among Early Christian graves that contained human remains is illustrated in Figures 8.40, 8.41, and 8.42. Graves containing individuals with fractures are not clustered in any particular area upon visual inspection and an absence of statistically significant clustering is confirmed through a high/low cluster analysis (p = 0.411, zscore = -0.822). 293 Figure 8.40: Spatial Distribution of Early Christian Temple Graves Containing Individuals with Fractures. 294 Figure 8.41: Spatial Distribution of Early Christian Bath Graves Containing Individuals with Fractures.* *Two graves associated with the basilica containing individuals exhibiting fractures are visible. 295 Figure 8.42: Spatial Distribution of Early Christian Basilica Graves Containing Individuals with Fractures. When burial areas are compared against each other, the graves associated with the temple exhibit the greatest proportion of individuals affected by fractures (Table 8.67). Fewer graves associated with the basilica contained individuals with fractures and only a small proportion of individuals buried in graves adjacent to the bath exhibited fractures. While a chi-square test shows that the differences in the presence and absence of fractures between burial areas is statistically significant, the expected cell sizes require that those areas are compared using Fisher’s exact test. 296 Table 8.67: Spatial Distribution of Early Christian Graves Containing Individuals with Fractures. Temple Bath Basilica 2 Pr Ab % Pr Ab % Pr Ab % df Two-tail p <0.05 χ (n) (n) (n) (n) (n) (n) p-value Yes/ No 7 14 33.3 5 54 8.5 4 15 21.1 7.478* 2 0.024 YES *At least one cell has an expected count less than five. The figures for the temple and basilica burials, which are fewer in number, are relatively similar and Fisher’s exact test shows that the differences between those two areas are not statistically significant (Table 8.68). Table 8.68: Comparison of Fractures between Temple and Basilica Burials. Temple Basilica Pr Ab % Pr Ab % χ2 df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 7 14 33.3 4 15 21.1 1 0.488 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. Similarly, no significant differences are found when the frequencies of graves containing individuals with fractures are compared between the areas of the bath and the basilica (Table 8.69). Table 8.69: Comparison of Fractures between Bath and Basilica Burials. Bath Basilica Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 5 54 8.5 4 15 21.1 1 0.210 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. The only comparison for which significant differences were revealed included the areas of the temple and the bath (Table 8.70). In this case, the proportion of temple burials containing individuals with fractures is much greater than that of bath burials. Table 8.70: Comparison of Fractures between Temple and Bath Burials. Temple Bath Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 7 14 33.3 5 54 8.5 1 0.011 YES *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. 297 Among Middle to Late Byzantine burials, graves containing individuals with fractures show a tendency to be located around the margins of the basilica as opposed to inside of the structure (Figure 8.43). Those graves are located to the north, south, and east of the basilica and a high/low cluster analysis demonstrates that they are not clustered in space to a greater degree than would be expected by chance (p = 0.263, z-score = -1.120). Figure 8.43: Spatial Distribution of Middle to Late Byzantine Graves Containing Individuals with Fractures. The frequencies of Middle to Late Byzantine graves containing individuals with and without fractures are presented by location in Table 8.71. Although a larger proportion of graves 298 outside of the basilica contained individuals exhibiting fractures, Fisher’s exact test demonstrates that there is no statistically significant relationship between grave location with respect to the basilica and the presence or absence of fractures. Table 8.71: Presence of Fractures in Graves Inside versus Outside of the Basilica. Inside Outside Pr Ab % Pr Ab % χ2* df Two-tail p <0.05 (n) (n) (n) (n) p-value Yes/ No 1 12 7.7 11 36 23.4 1 0.433 No *Empty cell indicates that Fisher’s exact test was used instead of Pearson’s chi-square test. Summary of Significant Results Pertinent to Hypothesis Three One of the goals of this research is to examine patterns in the mortuary space of Nemea for indications that burial location may have been determined by or was at least related to social differentiation. If differences in social status contributed to health disparities and were a factor in decisions about where a deceased individual was buried, then this might be reflected in patterns of skeletal stress when groups of individuals from different burial areas are compared against each other. With this in mind, the results presented above in the second section of this chapter are used to test the third hypothesis launched in this dissertation: Hypothesis 3: There will be significant differences in the prevalence rates of physiological stress indicators between groups of individuals based on grave location. In some cases, significant results were found when the frequencies of the presence and absence of a particular physiological stress indicator were compared between subgroups of burials for which sample sizes were small enough to violate the assumptions of the chi-square test (expected cell counts less than five). Such results are not included here. The summary that 299 follows contains only significant results produced using a statistically valid test, such as Fisher’s exact test in the cases of small subsamples. The results of this analysis, which integrates the osteological and archaeological datasets, revealed very few significant differences between burial subgroups. Furthermore, there is variation in the extent to which those differences can be taken as meaningful and indicative of broader patterns of physiological stress and, by extension, spatial organization of the mortuary space based on social differentiation. The significant results are summarized in Table 8.72. Table 8.72: Summary of Statistically Significant Results Pertinent to Hypothesis Three. Period Paleopathological Significant Result (p <0.05) Condition (frequencies differ from those expected by chance) Early Christian Antemortem Trauma Fractures are more common in burials associated with the temple when compared to those associated with the bath. Middle to Late Cribra Orbitalia Spatial clustering of the presence of cribra orbitalia. Byzantine Middle to Late Osteoarthritis Spatial clustering of the presence of osteoarthritis on Byzantine the left ankle joint. The differences between Early Christian burials associated with the temple and those associated with the bath in the frequency with which they contained individuals exhibiting fractures are of interest because temple burials do occupy a noteworthy spatial location—lying adjacent to pagan monumental architecture and simultaneously at the greatest distance from the Early Christian basilica of all burial areas. Despite the relatively small size of the temple burial subsample, it is possible that the results for antemortem trauma have significance for cemetery organization and this will be discussed in greater detail in Chapter Nine. The implications of the significant result for the spatial clustering of the presence of cribra orbitalia among Middle to Late Byzantine burials are less clear. If the presence of cribra orbitalia is truly clustered in space, it is concentrated primarily in burials located outside of the basilica along its south wall and adjacent to the southeast corner. As discussed previously, 300 however, an alternative interpretation that must be considered is that this pattern is the result of the large number of individuals not included in the high/low cluster analysis due to the absence of observable eye orbits. Finally, and also as discussed in the previous section, the significant result for the spatial clustering of the presence of osteoarthritis on the left ankle joint among Middle to Late Byzantine burials can be dismissed as an artifact of a very small subsample of observable bones recovered from graves in close proximity to one another. Summary This chapter had two main foci—each one using osteological and archaeological data in conjunction: (1) an examination of the relationships between attributes of the graves at Nemea and the individuals buried within them and (2) an exploration of the spatial organization of the graves and the prevalence of physiological stress indicators in the human skeletal remains they contained. While Hypothesis Three, which anticipates differences between burial areas in the prevalence rates of physiological stress indicators, cannot be completely rejected, it is certainly not strongly supported. Instead, the results of the bioarchaeological analysis demonstrate that the presence and absence of skeletal stress markers are in large part randomly distributed across the mortuary space. At the same time, the results presented in this chapter have identified important examples of variability in the mortuary space at Nemea and the interpretation of those patterns is one of the principal foci of Chapter Nine. In Chapter Nine I also review the results of the skeletal analysis and discuss their implications for health and living conditions in the Early Christian and Middle to Late Byzantine communities at Nemea. 301 CHAPTER 9: DISCUSSION At the outset of this dissertation I introduced four principal goals. The first three were carried out using a comparative analysis of prevalence rates of skeletal indicators of physiological stress and activity. They are as follows: (1) to investigate the possibility that living conditions deteriorated at Nemea during the Middle to Late Byzantine period; (2) to investigate the possibility of differences in life experiences between men and women through the identification of health disparities between males and females; and (3) to provide a broader perspective on the health status of the Nemea communities using data on physiological stress from additional skeletal samples in the region. The fourth goal, an exploration of the possibility that the organization of the burials at Nemea was based on status distinctions, was met using an approach that combined a spatial analysis of the graves with comparative assessments of prevalence rates and spatial distributions of physiological stress indicators. With these goals in mind, I now reexamine and interpret the results of the skeletal and mortuary analyses presented in Chapters Seven and Eight. This chapter is divided into two major sections. The implications of the results of the skeletal analysis are treated first. Within this section, I draw upon the paleodemographic profiles and the results of the paleopathological analysis of the Early Christian and Middle to Late Byzantine skeletal samples to reconstruct aspects of health and living conditions in each community at Nemea. The results of the tests of the first two hypotheses related to skeletal health posed in the dissertation are subsequently addressed in turn. Finally, the implications for health in a regional context are considered. The second section of the chapter is devoted to the results of the mortuary analysis. Here I use patterns of grave attributes identified in Chapter Eight together with the data on demography and skeletal stress to reconstruct aspects of burial practice in each 302 period at Nemea. Particular emphasis is placed on the results of the test of the third dissertation hypothesis and alternative explanations for the organization of burials at Early Christian and Middle to Late Byzantine Nemea. Reconstructing Health and Living Conditions at Byzantine Nemea Before addressing the first two hypotheses posed in this dissertation, it is useful to discuss in a general way the implications of the results of the skeletal analysis for each community at Nemea and whether or not those results are consistent with the historical narratives and expectations presented in earlier chapters. To review briefly, Greece was characterized by changing political and social landscapes as the Byzantine East took shape during Late Antiquity. But while early Byzantine communities experienced transformation on a large scale, there is also substantial archaeological evidence for the continuity of classical cultural traditions in Greece during this period (Gregory 1984, 1986; Kardulias 1995; Rothaus 2002). This was perhaps especially true in rural areas. Furthermore, although traditional narratives paint the picture of a late antique Greek countryside devastated by events such as the Slavic invasion, recent examinations of written sources and archaeological survey data (e.g., Curta 2001b; Pettegrew 2010) suggest that other interpretations are possible. Adding to this evidence, the state of the Early Christian community at Nemea suggested by the archaeology of the site is one of relative prosperity up until the abandonment of the site in the th late 6 century AD. th th Rural Byzantine communities of the 12 -13 centuries AD also existed during a period characterized by substantial sociopolitical transformation. While settlement patterns identified by archaeological surveys suggest that the southern Greek countryside experienced economic and 303 th demographic expansion during the 12 century (Athanassopoulos 1993, 1997, 2010), this growth occurred against a background of increasingly strained relationships between the central and provincial administrations (Herrin 1975, 1985; Ilieva 1991). It also took place on the eve of an invasion by western powers that resulted in the rapid dismantling of the territory of the empire. At the very least, the Frankish invasion of the Peloponnese had the potential to increase levels of psychosocial stress through a real or perceived decrease in security. It is also possible that the establishment of a feudal system in the newly formed Principality of Achaia made life more physically demanding as workload increased to meet obligations to local landowners. The results of the skeletal analysis carried out in this research provide more support for the revised narrative of late antique Greece in which conditions in the countryside were more or less favorable. What was unanticipated is the extent to which the results suggest that conditions were similar during the Middle to Late Byzantine period. The evidence from paleodemography is instructive in this regard. While the composition of a skeletal sample cannot be understood as representative of a living population because of factors such as differential preservation and variability in burial practices, the demographic profile nevertheless serves as an important means of evaluating living conditions in the past (Roberts and Manchester 2005). Figure 9.1 shows the percentage of the total number of individuals in each phase at Nemea that could be placed into each age category suggested by Buikstra and Ubelaker (1994). As presented in Chapter Seven, the mortality curves of the Early Christian and Middle to Late Byzantine skeletal samples are nearly identical. 304 Figure 9.1: Mortality Curves of the Nemea Skeletal Samples. 35 30 25 20 % Early Christian 15 10 Middle to Late Byzantine 5 0 83.842 > Females Total 73 The Middle to Late Byzantine Sample Table B.2: Number of Middle to Late Byzantine Femora and Demarking Point for Maximum Head Diameter. Sex Number Mean Femur Maximum Demarking Point (mm) (n) Head Diameter (mm) Male 49 48.941 Female 12 41.183 Males > 45.062 > Females Total 61 Table B.3: Number of Middle to Late Byzantine Femora and Demarking Point for Midshaft Circumference. Sex Number Mean Femur Midshaft Demarking Point (mm) (n) Circumference (mm) Male 31 95.032 Female 11 80.273 Males > 87.653 > Females Total 42 Table B.4: Number of Middle to Late Byzantine Tibiae and Demarking Point for Circumference at the Nutrient Foramen. Sex Number Mean Tibia Nutrient Foramen Demarking Point (mm) (n) Circumference (mm) Male 22 100.727 Female 7 87.143 Males > 93.935 > Females Total 29 367 Table B.5: Number of Middle to Late Byzantine Humeri and Demarking Point for Epicondylar Breadth. Sex Number Mean Humerus Demarking Point (mm) (n) Epicondylar Breadth (mm) Male 19 64.895 Female 7 57.586 Males > 61.241 > Females Total 26 Table B.6: Number of Middle to Late Byzantine Humeri and Demarking Point for Vertical Head Diameter. Sex Number Mean Humerus Vertical Demarking Point (mm) (n) Head Diameter (mm) Male 29 47.428 Female 10 39.850 Males > 43.639 > Females Total 39 Table B.7: Number of Middle to Late Byzantine Humeri and Demarking Point for Minimum Circumference. Sex Number Mean Humerus Minimum Demarking Point (mm) (n) Circumference (mm) Male 23 66.130 Female 10 58.000 Males > 62.065 > Females Total 33 Table B.8: Number of Middle to Late Byzantine Radii and Demarking Point for Tuberosity Circumference. Sex Number Mean Tuberosity Demarking Point (mm) (n) Circumference (mm) Male 20 55.350 Female 8 50.125 Males > 52.738 > Females Total 28 368 Appendix C: Permission to Reprint Figures 369 370 371 372 373 BIBLIOGRAPHY 374 BIBLIOGRAPHY Primary Sources in Translation Pausanias. Description of Greece. Translated by W.H.S. Jones and H.A. Omerod. Loeb Classical Library Volumes. Cambridge, Harvard University Press. 1918. Procopius. Secret History, Translated by R. Atwater. University of Michigan Press, Ann Arbor. 1961. Secondary Sources Abrahamse, D., 1984. “Rituals of death in the Middle Byzantine period.” Greek Orthodox Theological Review 25: 125-134. Adler, C.P., 2000. Bone Diseases: Macroscopic, Histological, and Radiological Diagnosis of Structural Changes in the Skeleton. Berlin, Springer. Agelarakis, A., 1997. “Excavations at Polystylon (Abdera) Greece: aspects of mortuary practices and skeletal biology.” Archaiologikon Deltion 47-48 (1992-1993 A’1): 293308. Al-Shorman, A., 2004. “Three cemeteries and a Byzantine church: a ritual landscape at Yasieleh, Jordan.” Antiquity 78: 306-313. Alberici, L.A. and M. Harlow, 2007. “Age and innocence: female transitions to adulthood in Late Antiquity,” in: A. Cohen and J.B. Rutter (eds.), Constructions of Childhood in Ancient Greece and Italy (Hesperia Supplements Vol. 41): 193-203. New Jersey, The American School of Classical Studies at Athens. Alcock, S., 1993. Graecia Capta: The Landscapes of Roman Greece. Cambridge, Cambridge University Press. Alexandre, M., 1992 “Early Christian women,” in: P.S. Pantel (ed.), A History of Women in the West: I. From Ancient Goddesses to Christian Saints, trans. A. Goldhammer: 409-444. Cambridge, The Belknap Press of Harvard University Press. nd Alexiou, M., 2002. The Ritual Lament in Greek Tradition. 2 Edition. Revised by D. Yatromanolakis and P. Roilos. Lanham, Rowman and Littlefield. Allen, L. and J. Casterline-Sabel, 2001. “Prevalence and causes of nutritional anemias,” in: U. Ramakrishnan (ed.), Nutritional Anemias: 7-21. Boca Raton, CRC Press. 375 Angel, J.L., 1966. “Porotic hyperostosis, anemias, malarias, and marshes in the prehistoric eastern Mediterranean.” Science 153(3737): 760-763. Angel, J.L., 1967. “Porotic hyperostosis or osteoporosis symmetrica,” in: D.R. Brothwell and A.T. Sandison (eds.), Diseases in Antiquity: A Survey of the Diseases, Injuries, and Surgery of Earlier Human Populations: 378-389. Springfield, Charles C. Thomas. Angel, J.L., 1978. “Porotic hyperostosis in the eastern Mediterranean.” Medical College of Virginia Quarterly 14(1): 10-16. Angold, M., 1989. “Greeks and Latins after 1204: the perspective of exile,” in: B. Arbel, B. Hamilton, and D. Jacoby (eds.), Latins and Greeks in the Eastern Mediterranean after 1204: 63-86. Totowa, Frank Cass. Antoine, D. 2008. “The archaeology of ‘plague’,” in: V. Nutton (ed.), Pestilential Complexities: Understanding Medieval Plague, Medical History, Supplement 27: 101-114. ArcGIS 10 Resource Center. Retrieved April 17, 2012. “How high/low clustering (Getis-Ord General G) works.” http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//005p0000000q000000 Armelagos, G.J., 2003. “Bioarchaeology as anthropology,” in: S.D. Gillespie and D. Nichols (eds.), Archaeology as Anthropology: 27-40. Arlington, VA, Archaeological Papers of the American Anthropological Association. Armelagos, G.J. and D.P. Van Gerven, 2003. “A century of skeletal biology and paleopathology: contrasts, contradictions, and conflicts.” American Anthropologist 105(1): 53-64. Armstrong, P., 1993. “Byzantine Thebes: excavations on the Kadmeia, 1980. The Annual of the British School at Athens 88: 295-335. Ashmore, W. and P.L. Geller, 2005. “Social dimensions of mortuary space,” in: G.F.M. Rakita, J.E. Buikstra, L.A. Beck, and S.R. Williams (eds.), Interacting with the Dead: Perspectives on Mortuary Archaeology for the New Millennium: 81-92. Gainesville, University Press of Florida. Athanassopoulos, E.F., 1993. “Intensive Survey and Medieval Rural Settlement: The Case of Nemea,” unpublished Ph.D. dissertation, Department of Anthropology, University of Pennsylvania, Philadelphia, Pennsylvania. Athanassopoulos, E.F., 1997. “Landscape archaeology of medieval and pre-modern Greece: the case of Nemea,” in: P.N. Kardulias and M.T. Shutes (eds.), Aegean Strategies: Studies of Culture and Environment on the European Fringe: 79-105. Lanham, Rowman and Littlefield. 376 Athanassopoulos, E.F., 2008. “Medieval archaeology in Greece: a historical overview,” in: W.R. Caraher, L.J. Hall, and R.S. Moore (eds.), Archaeology and History in Roman, Medieval, and Post-Medieval Greece: Studies on Method and Meaning in Honor of Timothy E. Gregory: 15-35. Aldershot, Ashgate. Athanassopoulos, E.F., 2010. “Landscape archaeology and the medieval countryside: settlement and abandonment in the Nemea region,” International Journal of Historical Archaeology 14: 255-270. Avalos, H., 1999. Health Care and the Rise of Christianity. Peabody, Hendrickson. th Barnes, E., 1996. “Diseases and disorders in a 13 century Frankish cemetery from Corinth, Greece. American Journal of Physical Anthropology Supplement 22: 65. Barnes, E., 2003. “The dead do tell tales,” in: C.K. Williams II and N. Bookidis (eds.), Corinth, The Centenary: 1896-1996: 435-443. Athens, American School of Classical Studies. Bass, W.M., 2005. Human Osteology: A Laboratory and Field Manual. Columbia, Missouri Archaeological Society Special Publication No. 2. Beatrice, J.S., T.W. Fenton, C.M. Rauzi, J.C. Wankmiller, L.L. Jenny, D.R. Foran, 2009. “Life and death at Butrint, Albania: paleopathology and mortuary behavior in Late Antiquity.” th Poster presented at the 78 annual meeting of the American Association of Physical Anthropologists, Chicago, Illinois. Beatrice, J.S., T.W. Fenton, C.V. Hurst, L.L. Jenny, J.C. Wankmiller, M. Mutolo, C. Rauzi, and D.R. Foran, forthcoming. “The human skeletons from the Triconch Palace and Merchant’s House,” in: W. Bowden and R. Hodges (eds.) Butrint 3: Excavations at the Triconch Palace (Vol. 2). Oxford, Oxbow. Binford, L.R., 1971. “Mortuary practices: their study and potential,” in: J.A. Brown (ed.), Approaches to the Social Dimensions of Mortuary Practices: 6-29. Washington D.C., Society for American Archaeology. Bintliff, J., 1996. “The Frankish countryside in central Greece: the evidence from archaeological field survey,” in: P. Lock and G.D.R. Sanders (eds.), The Archaeology of Medieval Greece (Oxbow Monograph 59): 1-18. Oxford, Oxbow. Birge, D.E., L.H. Kraynak, and S.G. Miller, 1992. Excavations at Nemea I: Topographical and Architectural Studies: The Sacred Square, the Xenon, and the Bath. Berkeley and Los Angeles, University of California Press. Blegen, C.W., 1927. “Excavations at Nemea, 1926,” American Journal of Archaeology 31(4): 421-440. 377 rd Bonvillain, N., 2001. Women and Men: Cultural Constructs of Gender. 3 Edition. New Jersey, Prentice Hall. th th Bourbou, C., 2003. “Health patterns of proto-Byzantine populations (6 -7 centuries AD) in south Greece: the cases of Eleutherna (Crete) and Messene (Peloponnese).” International Journal of Osteoarchaeology 13: 303-313. th th Bourbou, C., 2004. The People of Early Byzantine Eleutherna and Messene (6 -7 Centuries A.D.): A Bioarchaeological Approach. Athens, University of Crete. th th Bourbou, C., 2010. Health and Disease in Byzantine Crete (7 -12 Centuries AD). Farnham, Surrey, Ashgate. Bourbou, C. and S.J. Garvie-Lok, 2009. “Breastfeeding and weaning patterns in Byzantine times: evidence from human remains and written sources,” in: A. Papaconstantinou and A.M. Talbot (eds.), Becoming Byzantine: Children and Childhood in Byzantium: 65-83. Washington, D.C., Dumbarton Oaks. Bourbou, C., and M.P. Richards, 2007. “The Middle Byzantine menu: paleodietary information from isotopic analysis of humans and fauna from Kastella, Crete.” International Journal of Osteoarchaeology 17: 63-72. th th Bourbou, C. and A. Tsilipakou, 2009. “Investigating the human past of Greece during the 6 -7 Centuries A.D.,” in: L.A. Schepartz, S.C. Fox, and C. Bourbou (eds.), New Directions in the Skeletal Biology of Greece: 121-136. Princeton, The American School of Classical Studies at Athens. Bourbou, C., B.T. Fuller, S.J. Garvie-Lok, and M.P. Richards, 2011. “Reconstructing the diets th th of Greek Byzantine populations (6 -15 centuries A.D.) using carbon and nitrogen stable isotope ratios.” American Journal of Physical Anthropology 146: 569-581. Bowden, W., 1997. “Urban transformation in early-Byzantine Epirus: the example of Butrint,” in: G. De Boe and F. Verhaeghe (eds.), Medieval Europe (Brugge 1997) Vol. 3: 155-169. Zellik, I.A.P. Bowden, W., 1999. “The city in late-antique Epirus: the example of Butrint,” in: P. Cabanes (ed.), L’Illyrie méridionale et l’Épire dans l’Antiquité III. Actes du IIIe colloque international de Chantilly (16-19 Octobre 1996): 335-340. Paris, Boccard. Bowden, W., 2009. “Christian archaeology and the archaeology of medieval Greece,” in: J. Bintliff and H. Stoger (eds.), Medieval and Post-Medieval Greece: The Corfu Papers: 93-100. Oxford, BAR-IS 2023. Bowersock, G.W., 1990. Hellenism in Late Antiquity. Ann Arbor, University of Michigan Press. 378 Brickley, M. and R. Ives, 2006. “Skeletal manifestations of infantile scurvy.” American Journal of Physical Anthropology 129: 163-172. Brickley, M. and R. Ives, 2008. The Bioarchaeology of Metabolic Bone Disease. Boston, Academic Press. Bridges, P.S., 1994. “Vertebral arthritis and physical activities in the prehistoric southeastern United States.” American Journal of Physical Anthropology 93: 83-93. Brooks, S.T. and J.M. Suchey, 1990. “Skeletal age determination based on the os pubis: a comparison of the Acsádi-Nemeskéri and Suchey-Brooks methods.” Human Evolution 5: 227-238. Brown, J.A., 1971. “The dimensions of status in the burials at Spiro,” in: J.A. Brown (ed.), Approaches to the Social Dimensions of Mortuary Practices: 92-111. Washington D.C., Society for American Archaeology. Brown, J.A., 1981. “The search for rank in prehistoric burials,” in: R. Chapman, I. Kinnes, and K. Randsborg (eds.), The Archaeology of Death: 25-37. Cambridge, Cambridge University Press. Brown, J.A., 1995. “On mortuary analysis – with special reference to the Saxe-Binford research program,” in: L.A. Beck (ed.), Regional Approaches to Mortuary Analysis: 3-26. New York, Plenum Press. Brown, J.A., 2003. “The Cahokia Mound 72-Sub 1 burials as collective representation,” in: J.D. Richards and M.L. Fowler (eds.), The Wisconsin Archaeologist 84(1&2), A Deep Time Perspective: Studies in Symbols, Meaning, and the Archaeological Record – Papers in Honor of Robert L. Hall: 81-97. Brown, P., 1971. “The rise and function of the Holy Man in Late Antiquity.” The Journal of Roman Studies 61: 80-101. Brown, P., 1989. The World of Late Antiquity. New York, W.W. Norton & Company. Brown, M. and D.J. Ortner, 2011. Childhood scurvy in a medieval burial from Mačvanska Mitrovica, Serbia. International Journal of Osteoarchaeology 21: 197-207. Buikstra, J.E., 1977. “Biocultural dimensions of archaeological study: a regional perspective,” in: R.L. Blakely (ed.), Biocultural Adaptation in Prehistoric America: 67-84, Proceedings of the Southern Anthropological Society, No. 11. Athens, Georgia, University of Georgia Press. 379 Buikstra, J.E. and D.K. Charles, 1999. “Centering the ancestors: cemeteries, mounds, and sacred landscapes of the ancient North American midcontinent,” in: W. Ashmore and A.B. Knapp (eds.), Archaeologies of Landscape: Contemporary Perspectives: 201-228. Oxford, Blackwell. Buikstra, J.E. and A. Lagia, 2009. “Bioarchaeological approaches to Aegean archaeology,” in: L.A. Schepartz, S.C. Fox, and C. Bourbou (eds.), New Directions in the Skeletal Biology of Greece: 7-29. Princeton, The American School of Classical Studies at Athens. Buikstra, J.E. and D.H. Ubelaker, 1994. Standards for Data Collection from Human Skeletal Remains (Research Series, no. 44). Fayetteville, Arkansas Archaeological Survey. Bush, H., 1991. “Concepts of health and stress,” in: H. Bush and M. Zvelebil (eds.), Health in Past Societies: Biocultural Interpretations of Human Skeletal Remains in Archaeological Contexts: 11-20. BAR International Series 567. Oxford, Tempvs Reparatvm. Cameron, A., 1985. “Rome and the Greek East,” in: R. Browning (ed.), The Greek World: Classical, Byzantine, and Modern: 201-214. London, Thames and Hudson. Cameron, A., 2006. The Byzantines. Malden, Blackwell. Cannon, A., 1989. “The historical dimension in mortuary expressions of status and sentiment,” Current Anthropology 30(4): 437-458. Caraher, W.R., 2008. “Constructing memories: hagiography, church architecture, and the religious landscape of Middle Byzantine Greece: the case of St. Theodore of Kythera,” in: W.R. Caraher, L.J. Hall, and R.S. Moore (eds.), Archaeology and History in Roman, Medieval, and Post-Medieval Greece: Studies on Method and Meaning in Honor of Timothy E. Gregory: 67-84. Aldershot, Ashgate. Caraher, W.R., 2010. “Abandonment, authority, and religious continuity in post-classical Greece.” International Journal of Historical Archaeology 14: 241-254. Carr, C., 1995. “Mortuary practices: their social, philosophical-religious, circumstantial, and physical determinants.” Journal of Archaeological Method and Theory 2(2): 105-200. Caseau, B., 2004. “The fate of rural temples in Late Antiquity and the Christianization of the countryside,” in: W. Bowden, L. Lavan, and C. Machado (eds.), Recent Research on the Late Antique Countryside: 105-144. Leiden, Brill. Charles, D.K., 1995. “Diachronic regional social dynamics: mortuary sites in the Illinois Valley/American Bottom region,” in: L.A. Beck (ed.), Regional Approaches to Mortuary Analysis: 77-99. New York, Plenum Press. 380 Charles, D.K. and J.E. Buikstra, 2002. “Siting, sighting, and citing the dead,” in: H. Silverman and D. Small (eds.), The Space and Place of Death. Archaeological Papers of the American Anthropological Association, Vol. 11: 13-25. Arlington, VA. Cherry, J.F., J.L. Davis, and E. Mantzourani, 2000. “Mycenaean Nemea: the results of the Nemea Valley Archaeological Project archaeological survey,” from The Nemea Valley Archaeological Project Archaeological Survey: Internet Edition. Accessed 11/08/10. http://classics.uc.edu/NVAP/MycNVAP.html. Chesson, M.S., 1999. “Libraries of the dead: early Bronze Age charnel houses and social identity at urban Bab edh-Dhra’, Jordan,” Anthropological Archaeology 18: 137-164. Clark, G., 1993. Women in Late Antiquity: Pagan and Christian Life-styles. New York, Oxford University Press. Cohen, M.N., 1994. “The osteological paradox reconsidered,” Current Anthropology 35(5): 629-637. Cohen, M.N., 1997. “Does paleopathology measure community health? A rebuttal of ‘The Osteological Paradox’ and its implications for world prehistory,” in: R.R. Paine (ed.), Integrating Archaeological Demography: Multidisciplinary Approaches to Prehistoric Population (Center for Archaeological Investigations, Occasional Paper 24): 242-260. Carbondale, Ill. Conkey, M.W. and J.D. Spector, 1984. “Archaeology and the study of gender,” in: M.B. Schiffer (ed.), Advances in Archaeological Method and Theory. Vol. 7: 10-38. Orlando, Academic Press. Connor, C.L., 2004. Women of Byzantium. New Haven, Yale University Press. Constas, N., 2006. “Death and dying in Byzantium,” in D. Krueger (ed.), Byzantine Christianity: A People’s History of Christianity, Vol. 3: 124-145. Minneapolis, Fortress Press. Cousin, G. and F. Dürrbach, 1885. “Inscriptions de Némée.” Bulletin de Correspondence Hellénique 9: 349-356. Crass, B.A., 2000. “Gender in Inuit burial practices,” in: A.E. Rautman (ed.), Reading the Body: Representations and Remains in the Archaeological Record: 68-76. Philadelphia, University of Pennsylvania Press. Crass, B.A., 2001. “Gender and mortuary analysis: what can grave goods really tell us?,” in: B. Arnold and N.L. Wicker (eds.), Gender and the Archaeology of Death: 105-118. Walnut Creek, AltaMira. 381 Crislip, A.T., 2005. From Monastery to Hospital: Christian Monasticism and the Transformation of Health Care in Late Antiquity. Ann Arbor, University of Michigan Press. Crown, P.L. and S.K. Fish, 1996. “Gender and status in the Hohokam pre-classic to classic transition.” American Anthropologist 98(4): 803-817. Cruse, A., 2004. Roman Medicine. Stroud, Tempus. Curta, F., 2001a. “Peasants as ‘makeshift soldiers for the occasion’: sixth-century settlement patterns in the Balkans,” in: T.S. Burns and J.W. Eadie (eds.), Urban Centers and Rural Contexts in Late Antiquity: 199-217. East Lansing, Michigan State University Press. Curta, F., 2001b. The Making of the Slavs: History and Archaeology of the Lower Danube Region, c. 500-700. Cambridge, Cambridge University Press. Curta, F., 2004. “Barbarians in Dark-Age Greece: Slavs or Avars?,” in: T Stepanov and V. Vachkova (eds.), Civitas Divino-Humana. In Honorem Annorum Lx Georgii Bakalov: 513-550. Sofia, TANGRA TanNakRa. Curta, F., 2006. Southeastern Europe in the Middle Ages 500-1250. Cambridge, Cambridge University Press. Curta, F. 2008., “The making of the Slavs: between ethnogenesis, invention, and migration.” Studia Slavica et Balcanica Petropolitana 2: 155-172. Danforth, L.M. and A. Tsiaras, 1982. The Death Rituals of Rural Greece. Princeton, Princeton University Press. Davies, J., 1999. Death, Burial and Rebirth in the Religions of Antiquity. London, Routledge. Doucette, D.L., 2001. “Decoding the gender bias: inferences of atlatls in female mortuary contexts,” in: B. Arnold and N.L. Wicker (eds.), Gender and the Archaeology of Death: 159-177. Walnut Creek, AltaMira. Edelstein, E.J. and L. Edelstein, 1945a. Asclepius: A Collection and Interpretation of the Testimonies. Vol. II Part I. Baltimore, Johns Hopkins Press. Edelstein, E.J. and L. Edelstein, 1945b. Asclepius: A Collection and Interpretation of the Testimonies. Vol. II Part II. Baltimore, Johns Hopkins Press. Effros, B., 1997. “Beyond cemetery walls: early medieval funerary topography and Christian salvation.” Early Medieval Europe 6(1): 1-23. 382 El-Najjar, M.Y., J. Andrews, J.G. Moore, and D.G. Bragg, 1982. “Iron deficiency anemia in two prehistoric American Indian skeletons: a dietary hypothesis.” Plains Anthropologist 27(97): 205-209. Elm, S., 2000 [1994]. Virgins of God: The Making of Asceticism in Late Antiquity. Oxford, Oxford University Press. Fain, O., 2005. “Musculoskeletal manifestations of scurvy.” Joint Bone Spine 72: 124-128. Fenton, T.W., A. Soler, C.V. Hurst, and J.S. Beatrice, forthcoming. “The human skeletons from the Vrina Plain,” in: S. Greenslade and R. Hodges (eds.) The Evolution of an Urban Landscape: Excavations of the Vrina Plain, Butrint. Oxford, Oxbow. Fontaine, J., 1989. “Discussion,” in: Acts of the 11th International Congress of Christian Archaeology: 152-213. Rome, Pontifical Institute of Christian Archaeology. Fox, S.C., 2005. “Health in Hellenistic and Roman times: the case studies of Paphos, Cyprus and Corinth, Greece,” in: H. King (ed.), Health in Antiquity: 59-82. London, Routledge. Fox-Leonard, S.C., 1997. “Comparative Health from Paleopathological Analysis of the Human Skeletal Remains Dating to the Hellenistic and Roman Periods, from Paphos, Cyprus, and Corinth, Greece,” unpublished Ph.D. dissertation, Department of Anthropology, University of Arizona, Tucson, Arizona. Frey, J.M., 1998. “Burial Sites at Nemea: Early Christian, Byzantine, Other? An Attempt at Developing Categories of Burial Identification,” unpublished M.A. thesis, Graduate Group in Ancient History and Mediterranean Archaeology, University of California, Berkeley, California. Frey, J.M., 2006. “Speaking Through Spolia: The Language of Architectural Reuse in the Fortifications of Late Roman Greece,” unpublished Ph.D. dissertation, Graduate Division in Ancient History and Mediterranean Archaeology, University of California, Berkeley, California. Garnsey, P. and C. Humfress, 2001. The Evolution of the Late Antique World. Cambridge, Orchard Academic. Garvie-Lok, S.J., 2001. “Loaves and Fishes: A Stable Isotope Reconstruction of Diet in Medieval Greece,” unpublished Ph.D. dissertation, Department of Archaeology, University of Calgary, Calgary, Alberta. Garvie-Lok, S.J., 2009. “Population mobility at Frankish Corinth: evidence from stable oxygen isotope ratios of tooth enamel,” in: L.A. Schepartz, S.C. Fox, and C. Bourbou (eds.), New Directions in the Skeletal Biology of Greece: 245-256. Princeton, The American School of Classical Studies at Athens. 383 Garvie-Lok, S., 2010. “A possible witness to the sixth century Slavic invasion of Greece from the stadium tunnel at ancient Nemea.” International Journal of Historical Archaeology 14: 271-284. Gerstel, S.E.J. and M.E. Talbot, 2006. “The culture of lay piety in medieval Byzantium 10541453,” in: M. Angold (ed.), The Cambridge History of Christianity, Vol. 5: Eastern Christianity: 79-100. Cambridge, Cambridge University Press. Gerstel, S.E.J., M. Munn, H.E. Grossman, E. Barnes, A.H. Rohn, M. Kiel, 2003. “A late medieval settlement at Panakton.” Hesperia 72(2): 147-234. Giannicola, G., F.M. Sacchetti, A. Greco, G. Gregori, F. Postacchini, 2010. “Open reduction and internal fixation combined with hinged elbow fixator in capitellum and trochlea fractures: a retrospective study of 15 patients followed for 29 months.” Acta Orthopaedica 81(2): 228-233. Gillespie, S.D., 2001. “Personhood, agency, and mortuary ritual: a case study from the ancient Maya.” Journal of Anthropological Archaeology 20: 73-112. Gladykowska-Rzeczycka, J.J., 1998. “Periostitis: form and frequency in paleopathology.” Mankind Quarterly 38(3): 217-236. Glencross, B. and P. Stuart-Macadam, 2000. “Childhood trauma in the archaeological record.” International Journal of Osteoarchaeology 10: 198-209. Goldstein, L., 1976. “Spatial Structure and Social Organization: Regional Manifestations of Mississippian Society’, unpublished Ph.D. dissertation, Department of Anthropology, Northwestern University, Evanston, Illinois. Goldstein, L., 1981. “One-dimensional archaeology and multi-dimensional people: spatial organization and mortuary analysis,” in: R. Chapman, I. Kinnes, and K. Randsborg (eds.), The Archaeology of Death: 53-69. Cambridge, Cambridge University Press. Goldstein, L., 1995. “Landscapes and mortuary practices: a case for regional perspectives,” in: L.A. Beck (ed.), Regional Approaches to Mortuary Analysis: 101-121. New York, Plenum Press. Goldstein, L., 2000. “Mississippian ritual as viewed through the practice of secondary disposal of the dead,” in: S.R. Ahler (ed.), Mounds, Modoc, and Mesoamerica: Papers in Honor of Melvin L. Fowler. Scientific Papers Series, Vol. 28: 193-205. Springfield, Illinois State Museum. Goldstein, L., 2006. “Mortuary analysis and bioarchaeology,” in: J.E. Buikstra and L.A. Beck (eds.), Bioarchaeology: The Contextual Analysis of Human Remains: 375-387. Amsterdam, Elsevier Academic Press. 384 Goodman, A.H., 1993. “On the interpretation of health from skeletal remains.” Current Anthropology 34(3): 281-288. Goodman, A.H. and G.J. Armelagos, 1985. “Factors affecting the distribution of enamel hypoplasias within the human permanent dentition.” American Journal of Physical Anthropology 68: 479-493. Goodman, A.H. and J.C. Rose, 1990. “Assessment of systemic physiological perturbations from dental enamel hypoplasias and associated histological structures.” Yearbook of Physical Anthropology 33: 59-110. Goodman, A.H., D.L. Martin, G.J. Armelagos, and G. Clark, 1984. “Indications of stress from bone and teeth,” in: M.N. Cohen and G.J. Armelagos (eds.), Paleopathology at the Origins of Agriculture: 13-49. New York, Academic Press. Goodman, A.H., R.B. Thomas, A.C. Swedlund, and G.J. Armelagos, 1988. “Biocultural perspectives on stress in prehistoric, historic, and contemporary population research.” Yearbook of Physical Anthropology 31: 169-202. Gregory, T.E., 1984. “Cities and social evolution in Roman and Byzantine south east Europe,” in: J. Bintliff (ed.), European Social Evolution: Archaeological Perspectives: 267-276. Bradford, University of Bradford. Gregory, T.E., 1986. “The survival of paganism in Christian Greece: a critical essay.” The American Journal of Philology 107 2: 229-242. Gregory, T.E., 1993. “An Early Byzantine (dark-age) settlement at Isthmia: preliminary report," in T.E. Gregory (ed.), The Corinthia in the Roman Period, Supplement 8. Journal of Roman Archaeology: 149-60. Gregory, T.E., 2010. A History of Byzantium. 2 nd Edition. Malden, Wiley-Blackwell. Grimm, V., 1995. “Fasting women in Judaism and Christianity in Late Antiquity,” in: J. Wilkins, D. Harvey, and M. Dobson (eds.), Food in Antiquity: 225-240. Exeter, Exeter University Press. Grünbart, M., 2007. “Store in a cool and dry place: perishable goods and their preservation in Byzantium,” in: L. Brubaker and K. Linardou (eds.), Eat, Drink, and Be Merry (Luke 12:19): Food and Wine in Byzantium: 39-49. Aldershot, Ashgate. Hamlin, C., 2001. “Sharing the load: gender and task division at the Windover site,” in: B. Arnold and N.L. Wicker (eds.), Gender and the Archaeology of Death: 119-135. Walnut Creek, AltaMira. 385 Harries, J., 1992. “Death and the dead in the Late Roman West,” in: S. Bassett (ed.), Death in Towns: Urban Responses to the Dying and the Dead, 100-1600: 56-67. Leicester, Leicester University Press. Heather, P., 1996. The Goths. Oxford, Blackwell. Herrin, J., 1975. “Realities of Byzantine provincial government: Hellas and Peloponnesos, 1180-1205.” Dumbarton Oaks Papers 29: 253-284. Herrin, J., 1985. “A Christian millenium,” in: R. Browning (ed.), The Greek World: Classical, Byzantine, and Modern: 233-250. London, Thames and Hudson. Herrmann, N., 2002. “GIS applied to bioarchaeology: an example from the Río Talgua Caves in northeast Honduras.” Journal of Cave and Karst Studies 64(1): 17-22. Hershkovitz I., B. Ring, M. Speirs, E. Galili, M Kislev, E. Edelson, and A. Hershkovitz, 1991. “Possible congenital hemolytic anemia in prehistoric coastal inhabitants of Israel.” American Journal of Physical Anthropology 85: 7-13. Hershkovitz I., B.M. Rothschild, B. Latimer, O. Dutour, G. Leonetti, C.M. Greenwald, C. Rothschild, and L.M. Jellema, 1997. “Recognition of sickle cell anemia in skeletal remains of children.” American Journal of Physical Anthropology 104: 213-226. Hertz, R., 1960 [1907]. “A contribution to the study of the collective representation of death,” in: R. Needham and C. Needham (trans.), Death and the Right Hand: 27-86. New York, Free Press. Hill, B., 1997. “Imperial women and the ideology of womanhood in the eleventh and twelfth centuries,” in: L. James (ed.), Women, Men, and Eunuchs: Gender in Byzantium: 76-99. London, Routledge. Hillson, S., 1992. “Dental enamel growth, perikymata, and hypoplasias in ancient tooth crowns.” Journal of the Royal Society of Medicine 85: 460-466. Hillson, S., 1996. Dental Anthropology. Cambridge, Cambridge University Press. Hillson, S. and S. Bond, 1997. “Relationship of enamel hypoplasia to the pattern of tooth crown growth: a discussion.” American Journal of Physical Anthropology 104: 89-103. Hjohlman, J., 2002. “Farming the Land in Late Antiquity: The Case of Berbati in the Northeastern Peloponnese,” unpublished Ph.D. dissertation, Department of Classical Archaeology and Ancient History, Stockholm University, Stockholm. 386 Hodder, I., 1982. “The identification and interpretation of ranking in prehistory: a contextual perspective,” in: C. Renfrew and S.J. Shennan (eds.), Ranking, Resource and Exchange: Aspects of the Archaeology of Early European Society: 150-154. Cambridge, Cambridge University Press. Hollimon, S.E., 2000. “Sex, health, and gender roles among the Arikara of the Northern Plains,” in: A.E. Rautman (ed.), Reading the Body: Representations and Remains in the Archaeological Record: 25-37. Philadelphia, University of Pennsylvania Press. Hollimon, S.E., 2011. “Sex and gender in bioarchaeological research,” in: S.C. Agarwal and B.A. Glencross (eds.), Social Bioarchaeology: 149-182. Malden, Blackwell Hood, S., 1970. “Isles of refuge in the Early Byzantine period.” The Annual of the British School at Athens 65: 37-45. Horden, P., 2008. “Sickness and healing,” in: T.F.X. Noble and J.M.H. Smith (eds.), Early Medieval Christianities, c. 600-c. 1100 (The Cambridge History of Christianity, Vol. 3): 416-432. Cambridge, Cambridge University Press. Ilieva, A., 1991. Frankish Morea (1205-1262): Socio-cultural Interaction between the Franks and the Local Population. Athens, S.D. Basilopoulos. İşcan, M.Y., S.R. Loth, and R.K. Wright, 1984. “Age estimation from the ribs by phase analysis: white males.” Journal of Forensic Sciences 29: 1094-1104. İşcan, M.Y., S.R. Loth, and R.K. Wright, 1985. “Age estimation from the rib by phase analysis: white females.” Journal of Forensic Sciences 30: 853-863. Ivison, E.A., 1993. “Mortuary Practices in Byzantium (c950-1453): An Archaeological Contribution,” unpublished Ph.D. dissertation, Centre for Byzantine and Modern Greek Studies, School of Antiquity, University of Birmingham, Birmingham. Ivison, E.A., 1996. “Burial and urbanism at Late Antique and Early Byzantine Corinth (c. AD 400-700),” in: N. Christie and S.T. Loseby (eds.), Towns in Transition: Urban Evolution in Late Antiquity and the Early Middle Ages: 99-125. Aldershot, Scholar Press. Jackson, R. 2000. Doctors and Diseases in the Roman Empire. London, British Museum Press. Jacoby, D., 1973. “The encounter of two societies: Western conquerors and Byzantines in the Peloponnese after the Fourth Crusade.” The American Historical Review 78(4): 873-906. Jacoby, D., 1986. “Knightly values and class consciousness in the crusader states of the eastern Mediterranean.” Mediterranean Historical Review 1(2): 158-186. 387 Jacoby, D., 1989a. “Social evolution in Latin Greece,” in: K.M. Setton, H.W. Hazard, and N.P. Zacour (eds.), A History of the Crusades, Vol. VI: Impact of the Crusades on Europe: 175-221. Madison, University of Wisconsin Press. Jacoby, D., 1989b. “From Byzantium to Latin Romania: continuity and change,” in: B. Arbel, B. Hamilton, and D. Jacoby (eds.), Latins and Greeks in the Eastern Mediterranean after 1204: 1-44. Totowa, Frank Cass. Jacoby, D., 2008. “After the Fourth Crusade: the Latin Empire of Constantinople and the Frankish States,” in: J. Shepard (ed.), The Cambridge History of the Byzantine Empire c. 500-1492: 759-778. Cambridge, Cambridge University Press. Jenny, L.L., 2011. “A Bioarchaeological Study of Local Roman Identity: Skeletal Stress and Mortuary Treatment in the Butt Road Cemetery,” unpublished Ph.D. dissertation, Department of Anthropology, Michigan State University, East Lansing, Michigan. Johnson, M., 1999. Archaeological Theory: An Introduction. Malden, Blackwell. Johnson, M.J., 1997. “Pagan-Christian burial practices of the fourth century: shared tombs?” Journal of Early Christian Studies 5.1: 37-59. Jones, R., 1987. “Burial customs of Rome and the provinces,” in: J. Wacher (ed.), The Roman World (Vol. 2): 812-837. London, Routledge. Joyce, R.A., 2001. “Burying the dead at Tlatilco: social memory and social identities,” in: M.S. Chesson (ed.), Social Memory, Identity, and Death: Anthropological Perspectives on Mortuary Rituals. Archaeological Papers of the American Anthropological Association, Vol. 10: 12-26. Arlington, VA. Judd, M.A., 2002. “Comparison of long bone trauma recording methods.” Journal of Archaeological Science 29: 1255-1265. Judd, M.A., 2004. “Trauma in the city of Kerma: ancient versus modern injury patterns.” International Journal of Osteoarchaeology 14: 34-51. Judd, M.A., 2008. “The parry problem.” Journal of Archaeological Science 35: 1658-1666. Judd, M.A. and C.A. Roberts, 1999. “Fracture trauma in a medieval British farming village.” American Journal of Physical Anthropology 109: 229-243. Jurmain, R.D., 1991. “Degenerative changes in peripheral joints as indicators of mechanical stress: opportunities and limitations.” International Journal of Osteoarchaeology 1: 247252. 388 Jurmain, R.D. and L. Kilgore, 1995. “Skeletal evidence of osteoarthritis: a paleopathological perspective.” Annals of the Rheumatic Diseases 54: 443-450. Kardulias, P.N., 1995. “Architecture, energy, and social evolution at Isthmia, Greece: some thoughts about Late Antiquity in the Korinthia.” Journal of Mediterranean Archaeology 8.2: 33-59. Kardulias, P.N., 1997. “Reconstructing medieval site locations in the Korinthia, Greece,” in: P.N. Kardulias and M.T. Shutes (eds.), Aegean Strategies: Studies of Culture and Environment on the European Fringe: 107-122. Lanham, Rowman and Littlefield. Kardulias, P.N., T.E. Gregory, and J. Sawmiller, 1995. “Bronze Age and Late Antique exploitation of an islet in the Saronic Gulf, Greece.” Journal of Field Archaeology 22: 321. Kazhdan, A.P., 1997. “The peasantry,” in: G. Cavallo (ed.), The Byzantines: 43-73. Chicago, University of Chicago Press. Kazhdan, A.P. and A.W. Epstein, 1985. Change in Byzantine Culture in the Eleventh and Twelfth Centuries. Berkeley, University of California Press. King, H., 2005. “Introduction: what is health?,” in: H. King (ed.), Health in Antiquity: 1-11. London, Routledge. King, T., L.T. Humphrey and S. Hillson, 2005. “Linear enamel hypoplasias as indicators of systemic physiological stress: evidence from two known age-at-death and sex populations from postmedieval London.” American Journal of Physical Anthropology 128: 547-559. Klaus, H.D., C.S. Larsen, and M.E. Tam, 2009. “Economic intensification and degenerative joint disease: life and labor on the postcontact north coast of Peru.” American Journal of Physical Anthropology 139: 204-221. Knüsel, C.J., S. Göggel, and D. Lucy, 1997. “Comparative degenerative joint disease of the vertebral column in the medieval monastic cemetery of the Gilbertine Priory of St. Andrew, Fishergate, York, England.” American Journal of Physical Anthropology 103: 481-495. Kraynak, L., 1990. “The Xenon I,” in: S.G. Miller (ed.), Nemea: A Guide to the Site and Museum: 96-103. Berkeley, University of California Press. Kron, G., 2005. “Anthropometry, physical anthropology, and the reconstruction of ancient health, nutrition, and living standards.” Historia: Zeitschrift für Alte Geschichte 54(1): 68-83. 389 Krueger, D., 2006. “Introduction: the practice of Christianity in Byzantium,” in: D. Krueger (ed.), Byzantine Christianity: A People’s History of Christianity, Vol. 3: 1-15. Minneapolis, Fortress Press. Kyriakakis, J., 1974. “Byzantine burial customs: care of the deceased from death to the prothesis.” Greek Orthodox Theological Review 19: 37-72. Kyriakou, A., S.C. Sawa, I. Sawides, E. Pangalou, Y.S. Ioannou, S. Christou, N. Skordis, 2008. “Gender differences in the prevalence and severity of bone disease in thalassemia.” Pediatric Endocrinology Reviews 6(Suppl. 1): 116-122. Lagia A., C. Eliopoulos, and S. Manolis, 2007. “Thalassemia: macroscopic and radiological study of a case.” International Journal of Osteoarchaeology 17: 269-285. Laiou, A.E., 1981. “The role of women in Byzantine society.” Jahrbuch der Österreichischen Byzantinistik 31(1): 233-260. Laiou, A.E., 2009. “Priests and bishops in the Byzantine countryside, thirteenth to fourteenth centuries,” in: D.G. Angelov (ed.), Church and Society in Late Byzantium (Studies in Medieval Culture XLIX): 43-57. Kalamazoo, Medieval Institute Publications, Western Michigan University. Laiou-Thomadakis, A.E., 1977. Peasant Society in the Late Byzantine Empire: A Social and Demographic Study. Princeton, Princeton University Press. Landon, M., 1990. “The basilica and the Early Christian community,” in: S.G. Miller (ed.), Nemea: A Guide to the Site and Museum: 78-96. Berkeley, University of California Press. Larsen, C.S., 1997. Bioarchaeology: Interpreting Behavior from the Human Skeleton. Cambridge, Cambridge University Press. Larsen, C.S., 2006. “The changing face of bioarchaeology: an interdisciplinary science,” in: J.E. Buikstra and L.A. Beck (eds.), Bioarchaeology: The Contextual Analysis of Human Remains: 359-374. Boston, Elsevier. Lefort, J., 1993. “Rural economy and social relations in the countryside.” Dumbarton Oaks Papers 47: 101-113. Leonard, M. and D. Reidy, 2008. “Simultaneous bilateral elbow dislocation with associated bilateral radial head fracture.” European Journal of Orthopaedic Surgery and Traumatology 18: 43-45. Leyerle, B., 1994. “John Chrysostom on almsgiving and the use of money.” Harvard Theological Review 87(1): 29-47. 390 LiDonnici, L.R., 1995. The Epidaurian Miracle Inscriptions: Text, Translation, and Commentary. Atlanta, Scholars Press. Lindsay, H., 2000. “Death pollution and funerals in the city of Rome,” in: V.M. Hope and E. Marshall (eds.), Death and Disease in the Ancient City: 152-173. London, Routledge. Little, L.K., 2007. “Life and afterlife of the first plague pandemic,” in Plague and the End of Antiquity: The Pandemic of 541-750: 3-32. New York, Cambridge University Press. Lock, P., 1995. The Franks in the Aegean, 1204-1500. London, Longman. Lovejoy, C.O., R.S. Meindl, T.R. Pryzbeck, R.P. Mensforth, 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: 15-28. Lovell, N.C., 1997. “Trauma analysis in paleopathology.” Yearbook of Physical Anthropology 40: 139-170. Maat, G.J.R., 2004. “Scurvy in adults and youngsters: the Dutch experience. A review of the history and pathology of a disregarded disease.” International Journal of Osteoarchaeology 14: 77-81. McCormick, M., 2007. “Toward a molecular history of the Justinianic pandemic,” in: L.K. Little (ed.), Plague and the End of Antiquity: The Pandemic of 541-750: 290-312. New York, Cambridge University Press. MacKinnon, M., 2007. “Osteological research in classical archaeology.” American Journal of Archaeology 111: 473-504. Manchester, K., 1992. “The paleopathology of urban infections,” in: S. Bassett (ed.), Death in Towns: Urban Responses to the Dying and the Dead, 100-1600: 8-14. Leicester, Leicester University Press. Marchand, J.C., 2009. “Kleonai, the Corinth-Argos road, and the ‘Axis of History’.” Hesperia 78: 107-163. Martin, R.B., D.B. Burr, and N.A. Sharkey, 1998. Skeletal Tissue Mechanics. New York, Springer. Mays, S., 2006. “A paleopathological study of Colles’ fracture.” International Journal of Osteoarchaeology 16: 415-428. Mays, S., 2008. “A likely case of scurvy from Early Bronze Age Britain.” International Journal of Osteoarchaeology 18: 178-187. 391 Meindl, R.S. and C.O. Lovejoy, 1989. “Age changes in the pelvis: implications for Paleodemography,” in: M.Y. İşcan (ed.), Age Markers in the Human Skeleton: 137-168. Springfield, Charles C. Thomas. Metcalf, P. and R. Huntington, 1991. Celebrations of Death: The Anthropology of Mortuary nd Ritual. 2 Edition. Cambridge, Cambridge University Press. Miller, Stella G., 1983. “Excavations at Nemea, 1982.” Hesperia 52(1): 70-95. Miller, Stella G., 1984. “Excavations at Nemea, 1983.” Hesperia 53(2): 171-192. Miller, Steven G., 1975. “Excavations at Nemea, 1973-1974.” Hesperia 44(2): 143-172. Miller, Steven G., 1976. “Excavations at Nemea, 1975.” Hesperia 45(2): 174-202. Miller, Steven G., 1977. “Excavations at Nemea, 1976.” Hesperia 46(1): 1-26. Miller, Steven G., 1978. “Excavations at Nemea, 1977.” Hesperia 47(1): 58-88. Miller, Steven G., 1979. “Excavations at Nemea, 1978.” Hesperia 48(1): 73-103. Miller, Steven G., 1980. “Excavations at Nemea, 1979.” Hesperia 49(2): 178-205. Miller, Steven G., 1981. “Excavations at Nemea, 1980.” Hesperia 50(1): 45-67. Miller, Steven G., 1982. “Kleonai, the Nemean games, and the Lamian War.” Hesperia Supplements 20: 100-108. Miller, Steven G., 1988. “Excavations at Nemea, 1984-1986.” Hesperia 57(1): 1-20. Miller, Steven G. (ed.), 1990. Nemea: A Guide to the Site and Museum. Berkeley, University of California Press. Miller, Steven G., 2006. Ancient Greek Athletics. New Haven, Yale University Press. Miller, T.S., 1984. “Byzantine hospitals.” Dumbarton Oaks Papers 38 (Symposium on Byzantine Medicine): 53-63. Miller, T.S., 1997. The Birth of the Hospital in the Byzantine Empire. Baltimore, Johns Hopkins University Press. Milner, G.R., 1995. “An osteological perspective on prehistoric warfare,” in: L.A. Beck (ed.), Regional Approaches to Mortuary Analysis: 221-244. New York, Plenum. Morris, I., 1991. “The archaeology of ancestors: the Saxe/Goldstein hypothesis revisited.” Cambridge Archaeological Journal 1(2): 147-169. 392 Morris, I., 1992. Death-Ritual and Social Structure in Classical Antiquity. Cambridge, Cambridge University Press. Nauth, A., M.D. McKee, B. Ristevski, J. Hall, E.H. Schemitsch, 2011. “Distal humeral fractures in adults.” Journal of Bone and Joint Surgery 93-A(7): 686-700. Nithyananth, J.M., V.M. Cherian, K. Venkatesh, and R. Amritanand, 2008. “Bilateral HahnSteinthal fracture: a case report and review of literature.” European Journal of Orthopaedic Surgery and Traumatology 18: 395-397. Nutton, V., 1984. “From Galen to Alexander: aspects of medicine and medical practice in Late Antiquity.” Dumbarton Oaks Papers, Vol. 38: Symposium on Byzantine Medicine: 1-14. O’Gorman, J.A., 2001. “Life, death, and the longhouse: a gendered view of Oneota social organization,” in: B. Arnold and N.L. Wicker (eds.), Gender and the Archaeology of Death: 23-49. Walnut Creek, AltaMira. Oikonomides, N., 1996. “The social structure of the Byzantine countryside in the first half of the Xth century.” Byzantina Symmeikta 10: 105-125. nd Ortner, D.J., 2003. Identification of Pathological Conditions in Human Skeletal Remains. 2 Edition. Boston, Academic Press. Ortner, D.J. and M.F. Ericksen, 1997. “Bone changes in the human skull probably resulting from scurvy in infancy and childhood.” International Journal of Osteoarchaeology 7: 212-220. Ortner, D.J., E.H. Kimmerle, and M. Diez, 1999. “Probable evidence of scurvy in subadults from archaeological sites in Peru.” American Journal of Physical Anthropology 108: 321-331. O’Shea, J., 1981. “Social configuration and the archaeological study of mortuary practices: a case study,” in: R. Chapman, I. Kinnes, and K. Randsborg (eds.), The Archaeology of Death: 39-52. Cambridge, Cambridge University Press. Ousterhout, R., 2008. Master Builders of Byzantium. Philadelphia, University of Pennsylvania Museum of Archaeology and Anthropology. Paine, R.R., R. Vargiu, A. Coppa, C. Morselli, and E.E. Schneider, 2007. “A health assessment of high status Christian burials recovered from the Roman-Byzantine archaeological site of Elaiussa Sebaste, Turkey.” HOMO – Journal of Comparative Human Biology 58: 173190. Papageorgopoulou, C., and N.I. Xirotiris, 2009. “Anthropological research on a Byzantine population from Korytiani, west Greece,” in: L.A. Schepartz, S.C. Fox, and C. Bourbou (eds.), New Directions in the Skeletal Biology of Greece: 193-221. Princeton, The American School of Classical Studies at Athens. 393 Papathanasiou, A., C.S. Larsen, and L. Norr, 2000. “Bioarchaeological inferences from a Neolithic ossuary from Alepotrypa Cave, Diros, Greece.” International Journal of Osteoarchaeology 10: 210-228. Papathanasiou, A., E. Zachou, and M.P. Richards, 2009. “Bioarchaeological analysis of the human osteological material from Proskynas, Lokris,” in: L.A. Schepartz, S.C. Fox, and C. Bourbou (eds.), New Directions in the Skeletal Biology of Greece: 223-235. Princeton, The American School of Classical Studies at Athens. Parker Pearson, M., 1999. The Archaeology of Death and Burial. College Station, Texas A&M University Press. Paxton, F.S., 2005. “Communities of the living and the dead in Late Antiquity and the early medieval West,” in: M. Williams (ed.), The Making of Christian Communities in Late Antiquity and the Middle Ages: 49-62. London, Anthem. Paxton, F.S., 2008. “Birth and death,” in T.F.X. Noble and J.M.H. Smith (eds.), The Cambridge History of Christianity, Vol. 3: Early Medieval Christianities, c. 600-c. 1100: 383-398. Cambridge, Cambridge University Press. Pechenkina, E.A., and M. Delgado, 2006. “Dimensions of health and social structure in the early intermediate period cemetery at Villa El Salvador, Peru.” American Journal of Physical Anthropology 131: 218-235. Peebles, C.S. and S.M. Kus, 1977. “Some archaeological correlates of ranked societies.” American Antiquity 42: 421-448. Perkins, J., 1995. The Suffering Self: Pain and Narrative Representation in the Early Christian Era. London and New York, Routledge. Perry, M.A., 2002. “Health, Labor, and Political Economy: A Bioarchaeological Analysis of Three Communities in Provincia Arabia,” unpublished Ph.D. dissertation, Department of Anthropology, University of New Mexico, Albuquerque, New Mexico. Perry, M.A., 2007. “Is bioarchaeology a handmaiden to history? Developing a historical bioarchaeology.” Journal of Anthropological Archaeology 26: 486-515. Peterson, J.D., 2000. “Labor patterns in the southern Levant in the Early Bronze Age,” in: A.E. Rautman (ed.), Reading the Body: Representations and Remains in the Archaeological Record: 38-54. Philadelphia, University of Pennsylvania Press. Petroutsa, E.I., M.P. Richards, L. Kolonas, and S.K. Manolis, 2009. “Isotope paleodietary analysis of humans and fauna from the Late Bronze Age site of Voudeni,” in: L.A. Schepartz, S.C. Fox, and C. Bourbou (eds.), New Directions in the Skeletal Biology of Greece: 237-243. Princeton, The American School of Classical Studies at Athens. 394 Pettegrew, D.K., 2007. “The busy countryside of Late Roman Corinth: interpreting ceramic data produced by regional archaeological surveys.” Hesperia 76: 743-784. Pettegrew, D.K., 2010. “Regional survey and the boom-and-bust countryside: re-reading the archaeological evidence for episodic abandonment in the Late Roman Corinthia.” International Journal of Historical Archaeology 14: 215-229. Phenice T., 1969. “A newly developed visual method of sexing the os pubis.” American Journal of Physical Anthropology 30: 297-302. Pitarakis, B., 2009. “The material culture of childhood in Byzantium,” in: A. Papaconstantinou and A.M. Talbot (eds.), Becoming Byzantine: Children and Childhood in Byzantium: 167-251. Washington, D.C., Dumbarton Oaks. Porubcan, P.J., 2000. “Human and nonhuman surplus display at Mound 72, Cahokia,” in: S.R. Ahler (ed.), Mounds, Modoc, and Mesoamerica: Papers in Honor of Melvin L. Fowler. Scientific Papers Series, Vol. 28: 207-219. Springfield, Illinois State Museum. Pullen, D.J., 1986. “The Early Bronze Age settlement on Tsoungiza Hill, ancient Nemea,” in R. Hägg and D. Konsola (eds.), Early Helladic Architecture and Urbanization (SIMA 76): 73-78. Göteborg, Goterna, Kungalv. Rakita, G.F.M., and J.E. Buikstra, 2005. “Introduction,” in: G.F.M. Rakita, J.E. Buikstra, L.A. Beck, and S.R. Williams (eds.), Interacting with the Dead: Perspectives on Mortuary Archaeology for the New Millennium: 1-11. Gainesville, University Press of Florida. Rautman, M., 2006. Daily Life in the Byzantine Empire. Westport, Greenwood. Rebillard, E., 2003. “Conversion and burial in the Late Roman Empire,” in: K. Mills and A. Grafton (eds.), Conversion in Late Antiquity and the Early Middle Ages: Seeing and Believing: 61-83. Rochester, University of Rochester Press. Rebillard, E., 2009. The Care of the Dead in Late Antiquity. Ithaca, Cornell University Press. Reinert, S.W., 2002. “Fragmentation (1204-1453),” in: C. Mango (ed.), The Oxford History of Byzantium: 248-283. Oxford, Oxford University Press. Richard, J., 1985. “Agricultural conditions in the crusader states,” in: N.P. Zacour and H.W. Hazard (eds.), The Impact of the Crusades on the Near East (A History of the Crusades, Vol. V): 251-294. Madison, The University of Wisconsin Press. Richard, J., 1989. “The establishment of the Latin Church in the Empire of Constantinople (1204-1227),” in: B. Arbel, B. Hamilton, and D. Jacoby (eds.), Latins and Greeks in the Eastern Mediterranean after 1204: 45-62. Totowa, Frank Cass. 395 Roberts, C., 1991. “Trauma and treatment in the British Isles in the historic period: a design for multidisciplinary research,” in D. Ortner and A. Aufderheide (eds.), Human Paleopathology: Current Syntheses and Future Options: 225-240. Washington, D.C., Smithsonian. Roberts, C. and K. Manchester, 2005. The Archaeology of Disease. Ithaca, Cornell University Press. Roberts, C., C. Bourbou, A. Lagia, S. Triantaphyllou, and A. Tsaliki, 2005. “Health and disease in Greece: past, present and future,” in: H. King (ed.), Health in Antiquity: 32-58. London, Routledge. Rose, J.C., 1977. “Defective enamel histology of prehistoric teeth from Illinois.” American Journal of Physical Anthropology 46: 439-446. Rosen, G. 1958. A History of Public Health. New York, MD Publications. Rothaus, R.M., 1996. “Christianization and depaganization: the late antique creation of a conceptual frontier,” in: R.W. Mathisen and H.S. Sivan (eds.), Shifting Frontiers in Late Antiquity (Papers from the First Interdisciplinary Conference on Late Antiquity, The University of Kansas 1995): 299-308. Hampshire, Aldershot. Rothaus, R.M., 2000. Corinth: The First City of Greece: An Urban History of Late Antique Cult and Religion. Leiden, Brill. Rothschild, B.M., 2002a. “Porotic hyperostosis as a marker of health and nutritional conditions.” American Journal of Human Biology 14: 417-420. Rothschild, B.M., 2002b. “Hair standing on end as a manifestation of iron deficiency?” Radiology 224(2): 609-610. Rothschild, B.M. and Martin, L.R., 2006. Skeletal Impact of Disease. Albuquerque, New Mexico Museum of Natural History and Science, Bulletin 33. Rothschild, B.M. and Rothschild C., 1995. “Treponemal disease revisited: skeletal discriminators for yaws, bejel, and venereal syphilis.” Clinical Infectious Diseases 20(5): 1402-1408. Ruchelsman, D.E., N.C. Tejwani, Y.W. Kwon, K.A. Egol, 2008. “Coronal plane partial articular fractures of the distal humerus: current concepts in management.” Journal of the American Academy of Orthopaedic Surgeons 16(12): 716-728. Runciman, S., 1980. Mistra: Byzantine Capital of the Peloponnese. London, Thames and Hudson. Runnels, C.N. and T.H. van Andel, 1987. “The evolution of settlement in the Southern Argolid, Greece: an economic explanation.” Hesperia 56(3): 303-334. 396 Rush, A.C., 1941. Death and Burial in Christian Antiquity. Washington, D.C., The Catholic University of America Press. Safont, S., A. Malgosa, and M.E. Subirà, 2000. “Sex assessment on the basis of long bone circumference.” American Journal of Physical Anthropology 113: 317-328. Sallares, R., 2007. “Ecology, evolution, and epidemiology of plague,” in: L.K. Little (ed.), Plague and the End of Antiquity: The Pandemic of 541-750: 231-289. New York, Cambridge University Press. Samellas, A., 2002. Death in the Eastern Mediterranean (60-600 A.D.): The Christianization of the East: An Interpretation. Tübingen, Mohr Siebeck. Sanders, G.D.R., 2002. “Corinth,” in: A. Laiou (ed.), The Economic History of Byzantium From the Seventh through the Fifteenth Century, Vol. 2: 647-654. Washington, D.C., Dumbarton Oaks. Sanders, G.D.R., 2003. “Recent developments in the chronology of Byzantine Corinth,” in: C.K. Williams II and N. Bookidis (eds.), Corinth, The Centenary: 1896-1996: 385-399. New Jersey, The American School of Classical Studies at Athens. Sanders, G.D.R., 2005. “Archaeological evidence for early Christianity and the end of Hellenic religion in Corinth,” in: D.N. Showalter and S.J. Friesen (eds.), Urban Religion in Roman Corinth: Interdisciplinary Approaches: 419-442. Cambridge, Harvard University Press. Sanders, G.D.R., and I.K. Whitbread, 1990. “Central places and major roads in the Peloponnese.” The Annual of the British School at Athens 85: 333-361. Sandmark, H., C. Hogstedt, and E. Vingård, 2000. “Primary osteoarthrosis of the knee in men and women as a result of lifelong physical load from work.” Scandinavian Journal of Work, Environment and Health 26(1): 20-25. Saradi-Mendelovici, H., 1988. “The demise of the ancient city and the emergence of the mediaeval city in the eastern Roman Empire.” Echos du Monde Classique: Classical Views (Special Issue: Problems and Method in Greek History) 32 n.s. 7 No. 3: 365-401. Sauer, N.J., 1998. “The timing of injuries and manner of death: distinguishing among antemortem, perimortem, and postmortem trauma,” in: KJ Reichs (ed.), Forensic Osteology: Advances in the Identification of Human Remains (Second Edition): 321-332. Springfield: Charles C. Thomas. Saunders, S.R., 2008. “Juvenile skeletons and growth-related studies,” in: M.A. Katzenberg and nd S.R. Saunders (eds.), Biological Anthropology of the Human Skeleton. 2 Edition: 117147. Hoboken, NJ, John Wiley & Sons. 397 Saunders, S.R., and A. Keenleyside, 1999. “Enamel hypoplasia in a Canadian historic sample.” American Journal of Human Biology 11: 513-524. Saxe, A.A., 1970. “Social Dimensions of Mortuary Practices,” unpublished Ph.D. dissertation, University of Michigan, Ann Arbor. Scarborough, J., 1981. “Roman medicine and public health,” in: T. Ogawa (ed.), Public th Health: Proceedings of the 5 International Symposium on the Comparative History of Medicine-East and West, 1980: 33-74. Tokyo, Saikon. Schaefer M., S. Black, and L. Scheuer, 2009. Juvenile Osteology: A Laboratory and Field Manual. Burlington, Academic Press. Schepartz, L.A., S.C. Fox, and C. Bourbou (eds.), 2009. New Directions in the Skeletal Biology of Greece. Princeton, The American School of Classical Studies at Athens. Schepartz, L.A., S. Miller-Antonio, and J.M.A. Murphy, 2009. “Differential health among the Mycenaeans of Messenia: status, sex, and dental health at Pylos,” in: L.A. Schepartz, S.C. Fox, and C. Bourbou (eds.), New Directions in the Skeletal Biology of Greece: 155174. Princeton, The American School of Classical Studies at Athens. Scheuer, L. and S. Black, 2000. Developmental Juvenile Osteology. Amsterdam, Elsevier Academic Press. Schindler, O.S., 2003. “Bilateral capitellum humeri fracture: a case report and review of the literature.” Journal of Orthopaedic Surgery 11(2): 207-212. Schreiner, P., 1997. “Soldiers,” in: G. Cavallo (ed.), The Byzantines: 74-94. Chicago, University of Chicago Press. Schultz, M., 2001. “Paleohistopathology of bone: a new approach to the study of ancient diseases.” Yearbook of Physical Anthropology 44: 106-147. Shanks, M. and C. Tilley, 1982. “Ideology, symbolic power and ritual communication: a reinterpretation of neolithic mortuary practices,” in: I. Hodder (ed.), Symbolic and Structural Archaeology: 129-154. Cambridge, Cambridge University Press. Siegel, S., 1956. Nonparametric Statistics for the Behavioral Sciences. New York, McGraw Hill. Silverman, H., 2002. “Narratives of identity and history in modern cemeteries of Lima, Peru,” in: H. Silverman and D. Small (eds.), The Space and Place of Death. Archaeological Papers of the American Anthropological Association, Vol. 11: 167-190. Arlington, VA. Šlaus, M. and Z. Tomičić, 2005. “Discriminant function sexing of fragmentary and complete tibiae from medieval Croatian sites.” Forensic Science International 147: 147-152. 398 Snyder, G.F., 1999. Inculturation of the Jesus Tradition: The Impact of Jesus on Jewish and Roman Cultures. Harrisburg, Trinity Press International. Snyder, L.M. and C.K. Williams II, 1997. “Frankish Corinth: 1996,” Hesperia 66(1): 7-47. Soler, A., 2011. “Life and Death in a Medieval Nubian Farming Community: The Experience at Mis Island,” unpublished Ph.D. dissertation, Department of Anthropology, Michigan State University, East Lansing, Michigan. Sotres, P.G., 1998. “The regimens of health,” in: M. Grmek (ed.), Western Medical Thought from Antiquity to the Middle Ages: 291-318. Cambridge, Harvard University Press. Spieser, J.M., 2001a. “The city in Late Antiquity: a reevaluation,” in: J.M. Spieser (Variorum Collected Studies Series) Urban and Religious Spaces in Late Antiquity and Early Byzantium: I 1-14. Aldershot, Ashgate. Spieser, J.M., 2001b. “The Christianization of the city in Late Antiquity,” in: J.M. Spieser (Variorum Collected Studies Series) Urban and Religious Spaces in Late Antiquity and Early Byzantium: III 1-10. Aldershot, Ashgate. Spieser, J.M., 2001c. “The Christianization of pagan sanctuaries in Greece,” in: J.M. Spieser (Variorum Collected Studies Series) Urban and Religious Spaces in Late Antiquity and Early Byzantium: VI 1-13. Aldershot, Ashgate. Stark, R. and S. Garvie-Lok, 2012. “What can radiology contribute to paleopathological examinations of juvenile scurvy?” American Journal of Physical Anthropology 147: 277. Stathakopoulos, D.C., 2004. Famine and Pestilence in the Late Roman and Early Byzantine Empire: A Systematic Survey of Subsistence Crises and Epidemics. Birmingham Byzantine and Ottoman Monographs, Vol. 9. Aldershot, Ashgate. Stathakopoulos, D.C., 2007a. “Between the field and the plate: how agricultural products were processed into food,” in: L. Brubaker and K. Linardou (eds.), Eat, Drink, and Be Merry (Luke 12:19): Food and Wine in Byzantium: 27-38. Aldershot, Ashgate. Stathakopoulos, D.C., 2007b. “Crime and punishment: the plague in the Byzantine Empire, 541749,” in: L.K. Little (ed.), Plague and the End of Antiquity: The Pandemic of 541-750: 99-188. New York, Cambridge University Press. Steckel, R.H., P.W. Sciulli, and J.C. Rose, 2002a. “A health index from skeletal remains,” in: R.H. Steckel and J.C. Rose (eds.), The Backbone of History: Health and Nutrition in the Western Hemisphere: 61-93. Cambridge, Cambridge University Press. Steckel, R.H., J.C. Rose, C.S. Larsen, and P.L. Walker, 2002b. “Skeletal health in the western hemisphere from 4000 B.C. to the present.” Evolutionary Anthropology 11: 142-155. 399 Steckel, R.H., C.S. Larsen, P.W. Sciulli, and P.L. Walker, 2006. “The global history of health project data collection codebook.” . Stuart-Macadam, P., 1985. “Porotic hyperostosis: representative of a childhood condition.” American Journal of Physical Anthropology 66: 391-398. Stuart-Macadam, P., 1989. “Porotic hyperostosis: relationship between orbital and vault Lesions.” American Journal of Physical Anthropology 80: 187-193. Stuart-Macadam, P., 1992. “Porotic hyperostosis: a new perspective.” American Journal of Physical Anthropology 87: 39-47. Suchey, J.M. and D. Katz, 1986. “Skeletal age standards derived from an extensive multiracial sample of modern Americans.” Abstract. American Journal of Physical Anthropology 69: 269. Suchey, J.M. and D. Katz, 1998. “Applications of pubic age determination in a forensic setting.” In: K.J. Reichs (ed.), Forensic Osteology: Advances in the Identification of Human Remains: 204-236. Springfield: Charles C. Thomas. Sweetman, R. and M.J. Becker, 2005. “Knossos medical faculty site: late antique graves and other remains.” The Annual of the British School at Athens 100: 331-386. Tainter, J.A., 1977. “Woodland social change in west-central Illinois.” Midcontinental Journal of Archaeology 2: 67-98. Tainter, J.A., 1978. “Mortuary practices and the study of prehistoric social systems,” in: M.B. Schiffer (ed.), Advances in Archaeological Method and Theory (Vol. 1): 105-141. New York, Academic Press. Talbot, A.M., 1997. “Women,” in G. Cavallo (ed.), The Byzantines: 117-143. Chicago, University of Chicago Press. Talbot, A.M., 2006. “The devotional life of laywomen,” in: D. Krueger (ed.), Byzantine Christianity: A People’s History of Christianity, Vol. 3: 201-220. Minneapolis, Fortress Press. Talbot, A.M., 2009. “The death and commemoration of Byzantine children,” in: A. Papaconstantinou and A.M. Talbot (eds.), Becoming Byzantine: Children and Childhood in Byzantium: 283-308. Washington, D.C., Dumbarton Oaks. Tayles, N., 1996. “Anemia, genetic diseases, and malaria in prehistoric mainland Southeast Asia.” American Journal of Physical Anthropology 101: 11-27. 400 Teteriatnikov, N.B., 1984. “Burial places in Cappadocian churches.” Greek Orthodox Theological Review 29(2): 141-174. Triantaphyllou, S., 2001. A Bioarchaeological Approach to Prehistoric Cemetery Populations from Central and Western Greek Macedonia. Oxford, BAR-IS 976. Trigger, B.G., 1989. A History of Archaeological Thought. Cambridge, Cambridge University Press. Tritsaroli, P., 2006. “Pratiques Funéraires en Grèce Centrale à la Période Byzantine: Analyse à Partir des Données Archéologiques et Biologiques,” unpublished Ph.D. dissertation, Departément de Préhistoire, Institut de Paléontologie Humaine, Muséum National d’Histoire Naturelle, Paris. Tritsaroli, P. and F. Valentin, 2008. “Byzantine burials practices for children: case studies based on a bioarchaeological approach to cemeteries from Greece,” in F. Gusi Jener, S. Muriel, and C. Olària (coords.), Nasciturus, Infans, Puerulus Vobis Mater Terra: La Muerte en la Infancia: 93-116. Servei d'Investigacions Arqueològiques i Prehistòriques (SIAP): Diputació de Castelló. Trombley, F.R., 1985. “Paganism in the Greek world at the end of antiquity: the case of rural Anatolia and Greece.” The Harvard Theological Review 78(3/4): 327-352. Tzortzopoulou-Gregory, L., 2008. “Cemeteries in the countryside: an archaeological investigation of the modern mortuary landscape in the eastern Corinthia and northern Kythera,” in: W.R. Caraher, L.J. Hall, and R.S. Moore (eds.), Archaeology and History in Roman, Medieval, and Post-Medieval Greece: Studies on Method and Meaning in Honor of Timothy E. Gregory: 307-344. Aldershot, Ashgate. Ubelaker, D.H., 1989. Human Skeletal Remains. 2 Press. nd Edition. Washington, D.C., Taraxacum Ubelaker, D.H., 1992. “Porotic hyperostosis in prehistoric Ecuador,” in: P. Stuart-Macadam and S. Kent (eds.), Diet, Demography, and Disease: Changing Perspectives on Anemia: 201217. New York, Aldine de Gruyter. Van der Merwe, A.E., M. Steyn, and G.J.R. Maat, 2010. “Adult scurvy in skeletal remains of late th 19 century mine workers in Kimberley, South Africa.” International Journal of Osteoarchaeology 20(3): 307-316. van Gennep, A., 1960 [1908]. The Rites of Passage. Chicago, University of Chicago Press. VanPool, T.L. and R.D. Leonard, 2011. Quantitative Analysis in Archaeology. Malden, WileyBlackwell. 401 Velkovska, E., 2001. “Funeral rites according to the Byzantine liturgical sources.” Dumbarton Oaks Papers 55: 21-51. Völling, T., 2001. “The last Christian Greeks and the first pagan Slavs in Olympia,” in: E. Kountoura-Galake (ed.), The Dark Centuries of Byzantium (7th-9th c.) (National Hellenic Research Foundation, Institute for Byzantine Research, International Symposium 9): 303323. Athens, National Hellenic Research Foundation. Waldron, T., 2007. Palaeoepidemiology: The Measure of Disease in the Human Past. Walnut Creek, Left Coast. Waldron, T., 2009. Paleopathology. Cambridge, Cambridge University Press. Walker P.L., 2005. “Greater sciatic notch morphology: sex, age, and population differences.” American Journal of Physical Anthropology 127: 385-391. Walker, P.L., D.C. Cook, and P.M. Lambert, 1997. “Skeletal evidence for child abuse: a physical anthropological perspective.” Journal of Forensic Sciences 42(2): 196-207. Walker, P.L., R.R. Bathurst, R. Richman, T. Gjerdrum, and V.A. Andrushko, 2009. “The causes of porotic hyperostosis and cribra orbitalia: a reappraisal of the iron-deficiency anemia hypothesis.” American Journal of Physical Anthropology 139: 109-125. Wapler, U., E. Crubézy, and M. Schultz, 2004. “Is cribra orbitalia synonymous with anemia? Analysis and interpretation of cranial pathology in Sudan.” American Journal of Physical Anthropology 123: 333-339. Weatherall, D. and D. Kwiatkowski, 2003. “Hematologic manifestations of systemic diseases in children of the developing world,” in: D.G. Nathan, S.H. Orkin, D. Ginsburg, and A.T. Look (eds.), Nathan and Oski’s Hematology of Infancy and Childhood: 1810-1833. Sixth Edition. Philadelphia, W.B. Saunders. Weglian, E., 2001. “Grave goods do not a gender make: a case study from Singen am Hohentwiel, Germany,” in: B. Arnold and N.L. Wicker (eds.), Gender and the Archaeology of Death: 137-155. Walnut Creek, AltaMira. Weiss, E. and R. Jurmain, 2007. “Osteoarthritis revisited: a contemporary review of aetiology.” International Journal of Osteoarchaeology 17: 437-450. Weston, D.A., 2008. “Investigating the specificity of periosteal reactions in pathology museum specimens.” American Journal of Physical Anthropology 137: 48-59. Whitby, M., 2000. “The Balkans and Greece,” in: A. Cameron, B. Ward-Perkins, and M. Whitby (eds.), The Cambridge Ancient History, Vol. XIV: Late Antiquity: Empire and Successors, A.D. 425-600: 701-730. Cambridge, Cambridge University Press. 402 Wickkiser, B., 2008. Asklepios, Medicine, and the Politics of Healing in Fifth-Century Greece: Between Craft and Cult. Baltimore, Johns Hopkins University Press. Wiechmann, I. and G. Grupe, 2005. “Detection of Yersinia pestis in two early medieval skeletal th finds from Aschheim, (Upper Bavaria, 6 century A.D.).” American Journal of Physical Anthropology 126: 48-55. Williams, C.K. II, 2003. “Frankish Corinth: an overview,” in: C.K. Williams II and N. Bookidis (eds.), Corinth, The Centenary: 1896-1996: 423-434. New Jersey, The American School of Classical Studies at Athens. Williams, C.K. II and O.H. Zervos, 1996. “Frankish Corinth: 1995.” Hesperia 65(1): 1-55. Williams, C.K. II, L.M. Snyder, E. Barnes, and O.H. Zervos, 1998. “Frankish Corinth: 1997.” Hesperia 67(3): 223-281. Wood, J.W., G.R. Milner, H.C. Harpending and K.M. Weiss, 1992. “The osteological paradox.” Current Anthropology 33(4): 343-370. Wright, J.C., 1982. “Excavations at Tsoungiza (Archaia Nemea).” Hesperia 51(4): 375-397. Wright, J.C., J.F. Cherry, J.L. Davis, E. Mantzourani, S.B. Sutton, R.F. Sutton Jr., 1990. “The Nemea Valley Archaeological Project: a preliminary report.” Hesperia 59(4): 579-659. Wright, L.E. and C.J. Yoder, 2003. “Recent progress in bioarchaeology: approaches to the osteological paradox.” Journal of Archaeological Research 11(1): 43-70. Yasin, A.M., 2005. “Funerary monuments and collective identity: from Roman family to Christian community.” The Art Bulletin 87(3): 433-457. Zuckerman, M.K. and Armelagos, G.J., 2011. “The origins of biocultural dimensions in bioarchaeology,” in: S.C. Agarwal and B.A. Glencross (eds.), Social Bioarchaeology: 1543. Malden, Blackwell. 403