The Roman and Byzantine Graves and Human Remains 9780876619391, 0876619391

Citation preview

ISTHMI A VOLU M E I X

THE ROM A N A N D BY Z A N TIN E GR AV ES A N D HUM A N R EM A INS

ISTHMI A R ESULTS OF EXC AVATIONS CON DUCT ED UN DER THE AUSPICES OF THE A MER IC A N SCHOOL OF CL A SSIC A L STUDIES AT ATHENS

I II III IV V VI VII VIII

Oscar Broneer, Temple of Poseidon (1971) Oscar Broneer, Topography and Architecture (1973) Oscar Broneer, Terracotta Lamps (1977) Mary C. Sturgeon, Sculpture I: 1952–1967 (1987) Timothy E. Gregory, The Hexamilion and the Fortress (1993) Steven Lattimore, Sculpture II: Marble Sculpture, 1967–1980 (1996) Isabelle K. Raubitschek, The Metal Objects (1952–1989) (1998) Catherine Morgan, The Late Bronze Age Settlement and Early Iron Age Sanctuary (1999)

ISTHMI A EXC AVATIONS BY THE UNI V ER SIT Y OF C A LIFOR NI A AT LOS A NGELES A ND THE OHIO STATE UNI V ER SIT Y UNDER THE AUSPICES OF THE A MER IC A N SCHOOL OF CL A SSIC A L STUDIES AT ATHENS

VOLUME IX

THE ROMAN AND BYZANTINE GR AVES AND HUMAN REMAINS BY

JOSEPH L . R IFE

T H E A M E R IC A N S C HO OL OF C L A S S IC A L S T U DI E S AT AT H E N S PR I NC ET ON, N E W J E R S E Y 2 012

© The American School of Classical Studies at Athens, 2012 ISBN: 978-0-87661-939-1 Library of Congress Cataloging-in-Publication Data Isthmia : excavations by the University of Chicago, under the auspices of the American School of Classical Studies at Athens v. cm. Includes bibliographical references and index. Contents: v. 1. Temple of Poseidon / by Oscar Broneer—v. 2. Topography and architecture / by Oscar Broneer—v. 3. Terracotta lamps / by Oscar Broneer— v. 4. Sculpture I : 1952–1967 / by Mary C. Sturgeon—v. 5. The Hexamilion and the fortress / by Timothy E. Gregory. 1. Isthmia (Greece)—Antiquities. 2. Greece—Antiquities. 3. Excavations (Archaeology—Greece—Isthmia. I. Title II. Broneer, Oscar, 1894–1992. III. Sturgeon, Mary C. (Mary Carol), 1943– IV. Gregory, Timothy E. V. University of Chicago. VI. University of California, Los Angeles. VII. Ohio State University. DF261.I85B76 938´7 76-362971 typography by ascsa publications 6–8 charlton street, princeton, new jersey printed in the united states of america by thomson-shore, incorporated, dexter, michigan

MATRI CARISSIMAE γειναμένωι χάριτας, γαμέται λέχος, ὑέι φάμαν ἄψογον ἐς πυμάταν ἀὼ ἐνεγκαμέναι

PREFACE

The study of the graves and human remains from the Isthmian Sanctuary and Fortress has had a long history. Since the 1950s, excavations at Isthmia by the American School of Classical Studies at Athens (ASCSA) have uncovered over 150 graves and over 100 skeletons. Apart from a few isolated interments and a scattering of disarticulated bones, these burials were investigated by the team sponsored by the University of California at Los Angeles (UCLA) under the direction of the late Paul Clement. The analysis and publication of this material has been undertaken by his successor, Timothy E. Gregory, who directs the team sponsored by the Ohio State University (OSU). The known burials were nearly all located in two separate areas representing disparate periods in the history of the site: the West Cemetery on the edge of the modern village of Kyras Vrysi, and the Fortress adjoining the Hexamilion just northeast of the Temple of Poseidon. The Late Archaic and Classical burials at the West Cemetery are the subject of a future volume that will describe the pottery, small finds, and bones from the graves and discuss the historical and social context of their interment.1 The present volume covers the graves and human remains belonging to the local community in Roman to Byzantine times, particularly the inhabitants of the fortifications and the ruins of the Sanctuary from the end of the 4th to the 7th or 8th centuries. The study of the graves and their contents has been an important component in the research program of the OSU project almost since its inception. At the invitation of Gregory, Myra Giesen conducted a preliminary examination of human bone from the site in 1989–1990. She was the first person to collate the pertinent records, to try to make sense of their depositional context, and to generate an inventory.2 I assumed responsibility for the publication of the human remains in 1995 as a part of my larger study of the historical setting of the graves. I completed a new comprehensive analysis of the mortuary and skeletal remains during two long study seasons at Kyras Vrysi in 1996 and 1997, and during shorter visits in 1998 and 2000 to 2004.3 John Robb, who at the time was teaching at Southampton but has since moved to Cambridge, joined me in 1998 to observe and to report on paleopathology. The introduction and many descriptions and identifications of lesions in Chapter 7 reflect his contribution. My lively dialogue with him has steered me away from many methodological and interpretive pitfalls, for which I am very grateful. Although I began the composing this book in 1997, various professional and personal distractions, beyond the vicissitudes of compiling and analyzing complex data that reside 1. For preliminary reports, see Clement 1969, pp. 142–143; 1970, p. 119; 1974, pp. 102–105; Clement and Thorne 1974. The West Cemetery previously was called Lambrou Field after the surname of the former landowner. Note that the graves discovered to the south across the modern road on the Filis property belong to the same necropolis. The author and Martha Risser

are preparing the volume on the West Cemetery. 2. Gregory and Kardulias 1989, pp. 16–17; O’Roark and Giesen 1991; Kardulias and Shutes 1992, p. 20; Rife and Giesen 1994. The observations and conclusions presented herein supersede any previously published. 3. Gregory, Kardulias, and Rife 1997, p. 4; Rife 1998.

viii

PREFACE

overseas, delayed its submission until 2005. At that point, recognizing that scholarship in related fields, particularly Late Antique archaeology, had increased dramatically in recent years and being aware of the duration of revision, editing, and production to come, I made the decision to stop incorporating new publications. Therefore, I have not cited works published after 2005, even though certain important studies have now appeared that complement the present volume in interesting ways.4 Graves and bones, particularly those of post-Classical date, are frequently encountered during excavation; almost all major sites, and many of secondary importance, have produced them. Yet one of the greatest challenges I faced in preparing this book was the absence of a clear precedent in published scholarship on Greek remains. This book brings a new perspective in various ways: it applies current theoretical approaches to the study of burial and society; it fully integrates the study of funerary ritual with the study of skeletal biology; it employs a contextual approach to understanding graves and human remains within the larger frameworks of a community and a region; it explores life and death in the Greek countryside; and it investigates social history during the transition from ancient to medieval times. Since this project contributes to an emerging scholarly discourse and the subjects it covers will not be familiar to many readers in the fields of classical archaeology and ancient history, I have considered it important to adopt a wide scope. Accordingly, I have explained my data and techniques in detail, have compiled a generous bibliography, and have tried to situate the Roman and Byzantine graves and human remains from the Isthmus in relation to published evidence from elsewhere in Greece as well as other regions of the world. This will, I hope, provide a more useful and meaningful study for a larger audience than a simple catalogue and commentary. I hope, too, that it will highlight the dire need for the standardized collection, description, and publication of the abundant mortuary and osteological records of Greece. I conducted the research for this book under the auspices of the American School of Classical Studies at Athens and with permission granted by the former Fourth Ephoreia of Classical and Prehistoric Antiquities at Nafplio and the Sixth Ephoreia of Byzantine Antiquities at Patras. For their interest and support I gratefully acknowledge Zoë Aslamatzidou and Panayiota Kasimis, archaeologists for the Ephoreia; William D. E. Coulson†, James B. Muhly, and Stephen Tracy, successive directors of the American School; and Scott Pike, former interim director of the Wiener Laboratory of the American School. The directors of the excavations at Isthmia, Timothy E. Gregory and Elizabeth R. Gebhard, have permitted my work and generously assisted in the preparation of this publication. In particular, I thank Gregory for recognizing the significance of the graves and human remains and for enabling my study over the long term. In undertaking this study and preparing its publication I have enjoyed the technical aid and intellectual counsel of several people. Nancy Buschini, Blanche Kim, and Steve Koob offered their expert advice concerning the conservation and storage of the bones in 1995 and 1996. I acknowledge the able assistance of several students in biological anthropology and Mediterranean archaeology, including Heather Heerssen (Kenyon College, 1995), Sarah Kielt (Bryn Mawr College, 1995), Victoria Semos (University of Texas-Austin, 1998), Nancy Tang (Cornell University and the University of London, 2000), and Dhruva Jaishankar (Macalester College, 2002). Tang and Jaishankar diligently tabulated the taphonomic and osteological data. I thank Nina Loney and especially Nora Dimitrova for translating publications in Bulgarian, Russian, and Serbo-Croatian during my research. All graphic material in this book appears by courtesy of the Isthmia Excavations. Many individuals have helped with the visual documentation of the graves and their contents. The line art is the work of several 4. One notable study is the thesis on the bioarchaeology of Byzantine Greece by P. Tritsaroli (2006).

PREFACE

ix

architects, illustrators, and excavators with the Chicago, UCLA, and OSU projects over the years, including myself. Many photographs of the site, the graves, and the human remains in situ, along with a few photographs of the artifacts, were taken by Emile Serafis and Paul Clement. Almost all photographs of the bones, several of the artifacts, and a few of the site and graves were taken by Daniel M. Curtis. He has also prepared all drawn and photographic images for digital reproduction and printing. I sincerely thank him for his skill and professionalism in completing this most important part of the project. I also owe my gratitude to many teachers, colleagues, and friends. The success of any interdisciplinary endeavor depends on the constructive collaboration of people with diverse viewpoints. I warmly acknowledge the following individuals for their help during the course of my research and writing, without burdening them with complicity in any errors or disparities that remain in this book. Gregory offered his unconditional encouragement and astute criticism at all stages. I have learned much about archaeology, history, religion, and the natural environment from discussion and correspondence with N. N. Ambraseys, William Caraher, Kevin Clinton, John Coleman, Florin Curta, Archibald Dunn, John V. A. Fine Jr., Ann Ellis Hanson, George Kacandes, P. Nick Kardulias, Paul Legutko, John Mansfield, Charles Pazdernik, David Potter, Raymond Rogers, Guy D. R. Sanders, John Shean, Alice-May Talbot, Thomas Tartaron, Lita Tzortzopoulou-Gregory, and Fikret Yegül. On topics of specialized scope and on classes of artifacts I have consulted individuals pursuing related research in the Corinthia: Professor Gregory, John Hayes, Jeanne Marty†, Melissa Moore Morison, Dr. Sanders, and Kathleen Warner Slane on Roman and Byzantine pottery; Birgitta Lindros Wohl on lamps; Liane Houghtalin and Orestes Zervos on coins; and Richard K. Dunn and Jay Stratton Noller on geology and geomorphology. I thank them for furnishing unpublished evidence and valuable observations. In osteological study I have profited from the advice of anthropologists Sandra Garvie-Lok, Kenneth A. R. Kennedy, Clark Spenser Larsen, and Douglas H. Ubelaker, who have discussed the project at various stages and assiduously reviewed drafts of Chapters 5 through 7. Finally, Mireille Lee has been my deepest source of intellectual guidance and personal support, for which I owe more than I can express. This book would not have been possible without the generous financial and administrative support of several institutions and offices. My study overseas in 1997 and 1998 was aided by grants from the Horace H. Rackham School of Graduate Studies of the University of Michigan. I completed the basic research and writing while I was James Rignall Wheeler Fellow at the American School of Classical Studies (1995–1996) and Townsend Assistant Professor of Classics at Cornell (1999–2002). The Department of Classics and Olin Library at Cornell provided the necessary resources and a conducive atmosphere for the completion of much of this book. The library staff of Macalester College, particularly Aaron Albertson, assisted greatly with interlibrary loan acquisitions. Lastly, I am grateful to the editorial staff of the ASCSA Publications Office, under the sure guidance of Charles Watkinson and Andrew Reinhard, for their hard work, high standards, and broad vision in exploring new avenues for scientific publication. I dedicate this book to my mother, Pat Rife (1943–2000). She made me a sandwich in Kyras Vrysi to nourish my work and bought me a computer in White Bear Lake to facilitate my writing. During the completion of this project, when I was far away, she developed leukemia. Like Prima at Corinth, the radiant woman survives in the memory of the living so long as the living preserve her glorious memory. She will be with me, and I will continue to learn from her, for a long time. Macalester College St. Paul, Minnesota

x

PREFACE

During the final preparation of this volume for production, I have not fundamentally altered the content or expanded the argument. This decision was motivated by a concern to preserve the study’s integrity and to ensure its rapid publication. I have, however, slightly enhanced the bibliography predating 2005 and have very selectively added citations of more recent publications pertaining to the northeastern Peloponnese and to human skeletal remains from Greece, Cyprus, and Asia Minor. Andrew Reinhard, Carol Stein, and Timothy Wardell at the ASCSA Publications Office deserve my sincere gratitude for the timely and careful attention they have given this manuscript. Publication of this book was aided by financial support from the offices of the Vice Provost for Research and the Executive Dean of the College of Arts and Sciences at Vanderbilt University, as well as from the von Bothmer Publication Fund of the Archaeological Institute of America, the Packard Humanities Institute, and the Ohio State University Excavations at Isthmia. Vanderbilt University Nashville, Tennessee

CONTENTS

LIST OF ILLUSTRATIONS LIST OF TABLES BIBLIOGRAPHY AND ABBREVIATIONS 1. THEMES, PROCEDURES, AND MATERIALS

xv xxi xxiii 1

THEMES AND APPROACHES

2

PROCEDURE OF STUDY

6

THE CREATION OF THE MATERIAL RECORD

11

PART I: THE ARCHAEOLOGICAL, HISTORICAL, AND SOCIAL CONTEXT OF THE GRAVES 2. THE GRAVES AND ASSOCIATED REMAINS

21

THE ARCHAEOLOGICAL DIMENSIONS OF MORTUARY BEHAVIOR

21

THE CHRONOLOGIES OF BURIAL

23

DESCRIPTION OF THE GRAVES BY AREA

25

NORTHEAST GATE

25

TOWER 2

56

TOWER 13

61

TOWER 14

62

LOUKOS FIELD

72

DECAUVILLE GRAVES

75

ROMAN BATH

76

HEXAMILION OUTWORKS

82

WEST FOUNDATION

88

PALAIMONION

91

THEATER CAVE

96

CATALOGUE OF BURIAL DEPOSITS CATALOGUE OF FUNERARY ARTIFACTS 3. BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTORY OF SETTLEMENT

99 103 113

PHASE I: THE ROMAN SANCTUARY (MID-1ST TO LATE 4TH CENTURIES)

114

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES)

120

PHASE IV: THE EARLY BYZANTINE SETTLEMENT (LATE 6TH TO 7TH/8TH CENTURIES)

135

PHASE V: THE MIDDLE TO LATE BYZANTINE FORTRESS (10TH TO 15TH CENTURIES)

144

EPILOGUE: KYRAS VRYSI, AYIOS IOANNIS PRODROMOS, AND ITS CEMETERY by Lita Tzortzopoulou-Gregory and Joseph L. Rife

146

xii

CONTENTS

4. FUNERARY RITUAL, MORTUARY VARIABILITY, AND SOCIETY

153

GREEK FUNERARY RITUAL DURING THE ROMAN AND BYZANTINE ERAS

159

MORTUARY FORM AND VARIABILITY

163

RITUALS BEFORE BURIAL

164

SELECTION, ACQUISITION, AND PREPARATION OF A SITE FOR BURIAL

167

MULTIPLE BURIAL

169

GRAVE DESIGN AND TYPOLOGY

172

BURIAL ORIENTATION

181

CORPOREAL POSITION

183

FUNERARY ASSEMBLAGE

187

SECONDARY BURIAL

199

HUMAN DISTURBANCE OF GRAVES

202

SOCIAL STRUCTURE AND IDEOLOGY

206

AGE

207

SEX AND GENDER

210

VERTICAL POSITION

213

HORIZONTAL POSITION

218

CHRISTIAN IDEOLOGY

222

SOCIAL AND IDEOLOGICAL DEVIANCY

230

PART II: THE OSTEOLOGICAL AND BIOARCHAEOLOGICAL CONTEXT OF THE HUMAN REMAINS 5. THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

235

THE CONDITION OF THE SKELETAL ASSEMBLAGE

238

OSTEOLOGICAL METHODOLOGY

247

DETERMINATION OF SEX

248

ESTIMATION OF AGE AT DEATH

249

MEASUREMENT OF BONES

253

OBSERVATION OF NONMETRIC TRAITS

254

THE COMPOSITION OF THE SKELETAL SAMPLE

258

PALEODEMOGRAPHY

259

PHYSICAL APPEARANCE AND DIVERSITY

265

GROWTH AND STATURE

282

NONMETRIC VARIABILITY

286

6. TEETH AND ORAL HEALTH

293

DIET AND ORAL HYGIENE

293

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

296

VARIATION IN DENTAL FORM, LOCATION, AND NUMBER

298

ENAMEL HYPOPLASIA

306

DENTAL WEAR

313

DENTAL TRAUMA

319

PLAQUE AND CALCULUS

324

DENTAL CARIES

326

ALVEOLAR DEFECTS AND AMTL

334

SUMMARY OF LOCAL DENTAL CONDITIONS

349

ORAL HEALTH IN THE GREEK WORLD DURING ROMAN AND BYZANTINE TIMES

351

CONTENTS

7. PALEOPATHOLOGY AND PALEOEPIDEMIOLOGY

xiii

367

CATALOGUE OF CONDITIONS

368

THE PALEOEPIDEMIOLOGY OF THE SKELETAL SAMPLE

405

CONGENITAL ANOMALIES

406

HEMATOLOGICAL DISORDERS

409

INFECTIOUS DISEASE

413

JOINT DISEASE

415

TRAUMA

429

NEOPLASIA

436

SUMMARY OF LOCAL PALEOPATHOLOGICAL CONDITIONS

437

DISEASE AND INJURY IN THE GREEK WORLD DURING ROMAN AND BYZANTINE TIMES

441

8. LIFE AND DEATH AT THE ISTHMUS BETWEEN ROME AND BYZANTIUM

457

APPENDIX: METRIC AND NONMETRIC TRAITS IN THE SKELETAL SAMPLE

469

INDEXES GENERAL INDEX

481

INDEX OF BURIAL SITES AND REGIONS

497

INDEX OF LOTS CITED

503

INDEX OF OBJECTS CITED

505

INDEX OF ANCIENT, BYZANTINE, AND EASTERN SOURCES

507

INDEX OF INSCRIPTIONS

511

ILLUSTRATIONS

FIGURES 1.1. 1.2. 1.3. 1.4. 2.1. 2.2. 2.3. 2.4. 2.5. 2.6. 2.7. 2.8. 2.9. 2.10. 2.11. 2.12. 2.13. 2.14. 2.15. 2.16. 2.17. 2.18. 2.19. 2.20. 2.21. 2.22. 2.23. 2.24. 2.25. 2.26. 2.27. 2.28. 2.29. 2.30. 2.31. 2.32. 2.33. 2.34.

The Isthmus of Corinth The Isthmian Sanctuary The Isthmian Fortress and adjacent graves Human remains from NEG 69-004, as stored by excavators in 1969 and retrieved for study in 1990 Northeast Gate, actual-state plan Northeast Gate, schematic plan Northeast Gate, restored plan NEG 69-103, 69-004, and 69-008 and surroundings Coverings of NEG 69-103 and 69-004 NEG 69-103A, B in grave NEG 69-004A, C–E, glass pitcher (7), and bottle (13) in grave NEG 69-004B and legs of A in grave Spiral and transverse fractures in long bones of NEG 69-004C–E from handling by Late Roman builders Covering of NEG 69-008 with lamp (5) Stones on floor of NEG 69-008 after excavation NEG 67-001A–K, coins (1–3), ring, and buckle (18) in grave Coins (1–3) and ring in NEG 67-001 West end of NEG 67-001 with cup (8) NEG 67-003B in grave after removal of other skeletons Covering of NEG 69-005 NEG 69-005A–C in grave Covering of NEG 69-001 Covering of NEG 69-001 NEG 69-001A–D in grave West end of NEG 69-001 with earring (19) East section of Trenches 69-5/95-1 northwest of the Northeast Gate Covering of NEG 69-007 Covering of NEG 69-007 NEG 69-007A, B in grave NEG 69-009, plan and sections Covering of NEG 69-009 and surroundings in 1969 Covering of NEG 69-009 in 1998 Inner layer of tiles sealing burial compartment of NEG 69-009 NEG 69-010, plan and sections Covering of NEG 69-010 and surroundings in 1969 Detail of cross on west end of covering of NEG 69-010 in 1969 Covering of NEG 69-010 in 1998 Inner layer of tiles sealing burial compartment of NEG 69-010

2 3 4 7 26 27 28 29 30 31 32 32 33 34 35 37 37 37 39 40 40 41 42 43 44 45 46 46 47 48 49 49 49 51 52 52 52 53

xvi 2.35. 2.36. 2.37. 2.38. 2.39. 2.40. 2.41. 2.42. 2.43. 2.44. 2.45. 2.46. 2.47. 2.48. 2.49. 2.50. 2.51. 2.52. 2.53. 2.54. 2.55. 2.56. 2.57. 2.58. 2.59. 2.60. 2.61. 2.62. 2.63. 2.64. 2.65. 2.66. 2.67. 2.68. 2.69. 2.70. 2.71. 2.72. 2.73. 2.74. 2.75. 2.76. 2.77. 2.78. 2.79. 2.80. 2.81. 2.82. 2.83. 2.84. 2.85. 2.86. 2.87.

ILLUSTRATIONS Tile fragments at west end of floor of NEG 69-010 Tower 2, actual-state plan Covering and marker of T2 68-006 T2 68-006A in grave T2 68-006 with stones lining walls after removal of skeleton Covering and marker of T2 68-002 T2 68-002A in grave Covering of T2 68-003 T2 68-003A, B in grave Reconstruction of tiles (IM 68-109) paving T2 68-003 after removal from cist Tower 14, actual-state plan T14 67-002A in grave T14 67-002 Architectural terracotta (IT 895) from under head of T14 67-002A T14 67-004A in grave Lower legs of T14 69-701A Covering over east end of T14 69-991 Upper body of T14 69-991A in west end of cist Lower body of T14 69-991A in east end of cist T14 69-002A in grave Covering of T14 69-003 T14 69-003A in grave Covering of possible grave northwest of Tower 14 Loukos Field, actual-state plan LOU 69-801A in grave LOU 69-801 DEC 69-901A, B in grave Tiles covering DEC 69-902 Pitched tiles covering DEC 69-902 after removal of outer tiles Roman Bath, actual-state plan Slab and debris enclosing channel with RB 76-002 RB 76-002 Differential preservation of femoral and tibial fragments of RB 76-002C and 76-002A Hexamilion Outworks, actual-state plan HO 70-901A–C Covering of HO 70-902 Top three skeletons in HO 70-902 Bottom seven skeletons in HO 70-902 West Foundation, actual-state plan and sections WF 62-001A in grave Reconstruction of tiles (IT 811–814) enclosing WF 62-001 after removal from cist Central area of the Sanctuary during its Antonine phase Palaimonion, actual-state plan and sections Temple of Palaimon during excavation in 1956 Covering of PAL 56-001 PAL 56-001A with earring (17) and beads (33) PAL 56-001 Theater Cave, actual-state plan TC 60-001A Coin of Leo I (1) Coin of Marcian (2) Illegible coin (3) Pentanummium (4)

53 56 57 57 58 59 59 60 60 61 63 64 64 64 65 66 67 67 67 68 69 69 71 73 74 74 76 77 77 79 80 81 81 83 84 86 87 87 89 90 90 92 93 94 94 94 95 97 98 103 103 103 103

ILLUSTRATIONS 2.88. 2.89. 2.90. 2.91. 2.92. 2.93. 2.94. 2.95. 2.96. 2.97. 2.98. 2.99. 2.100. 2.101. 2.102. 2.103. 2.104. 2.105. 2.106. 2.107. 2.108. 2.109. 2.110. 2.111. 2.112. 2.113. 2.114. 3.1. 3.2. 3.3. 3.4. 3.5. 4.1. 4.2. 5.1. 5.2. 5.3. 5.4. 5.5. 5.6. 5.7. 5.8. 5.9. 5.10. 5.11. 5.12. 5.13. 5.14. 5.15. 5.16. 5.17. 5.18. 5.19.

Lamp (5) Lamp (6) Glass pitcher (7) Coarse-ware cup (8) Fine-ware jar (9) Coarse-ware bowl (10) Unguentarium (11) Cooking pot (12) Pierced coin (pendant?; 14) Bronze finger ring (15) Pair(?) of bronze earrings (16) Bronze earring (17) Iron buckle (18) Pair of gold earrings (19) Beaded necklace with bone pendant cross (20, 21), suggested reconstruction Bone pendant cross (20) and various beads (21a–k) from necklace Iron loop (buckle or pendant?; 22) Bronze ring (23) Bronze buckle (24) Stone bead on glass rod (necklace?; 25) Iron disks (26) Iron rings (27) Iron nail (hobnail?; 28) Bone button (29) Iron buckle (30) Bronze loop (31) Shell bead (32) Ayios Ioannis Prodromos, Kyras Vrysi in 2005 Cemetery of Ayios Ioannis in 2000 Standard tomb type, Ayios Ioannis Tomb at ground level with concrete border, Ayios Ioannis Row of tombs, Ayios Ioannis Distribution of the orientations of 25 burials by period Survival of skeletal elements after secondary burial in HO 70-901 Surficial etching on right parietal and occipital of DEC 69-901A Surficial etching and structural degradation on left parietal and occipital of NEG 67-001C Cavitation, flaking, and fracturing on frontal of NEG 69-007A Differential rates of preservation of skeletal elements in 48 adults Differential rates of preservation of skeletal elements in 21 subadults Age structure among skeletons of Late Roman to Early Byzantine date Distribution of age at death by sex among skeletons of Late Roman to Early Byzantine date NEG 69-103B NEG 69-004B NEG 69-004C NEG 69-004D NEG 67-001B NEG 67-001C NEG 67-001D NEG 67-003B NEG 69-005C NEG 69-001A NEG 69-001D NEG 69-007A

xvii 104 104 105 106 106 107 107 107 109 109 109 109 109 109 110 110 111 111 111 111 111 111 111 112 112 112 112 148 149 150 150 151 182 200 240 240 240 242 243 270 270 271 271 271 271 272 272 272 273 273 273 273 274

xviii 5.20. 5.21. 5.22. 5.23. 5.24. 5.25. 5.26. 5.27. 5.28. 5.29. 5.30. 5.31. 5.32. 5.33. 5.34. 5.35. 5.36. 5.37. 5.38. 5.39. 5.40. 5.41. 5.42. 5.43. 6.1. 6.2. 6.3. 6.4. 6.5. 6.6. 6.7. 6.8. 6.9. 6.10. 6.11. 6.12. 6.13. 6.14. 6.15. 6.16. 6.17. 6.18. 6.19. 6.20. 6.21. 6.22. 6.23. 6.24. 6.25. 6.26. 6.27. 6.28. 6.29.

ILLUSTRATIONS NEG 69-007B NEG 69-999A T2 68-002A T2 68-003A T14 67-002A T14 67-004A T14 69-002A T14 69-003A LOU 69-801A RB 76-002A RB 76-002C HO 70-901A HO 70-901B HO 70-901C HO 70-902D HO 70-902F HO 70-902H HO 70-902I HO 70-902J WF 62-001A TC 60-001A Growth rates in five diaphyseal lengths in 17 individuals up to 12 years old Distribution of stature by sex among skeletons of Late Roman to Early Byzantine date Sites of Roman and Byzantine date with skeletons of known stature for comparative study NEG 69-103B NEG 69-004B NEG 69-004C NEG 69-004D NEG 69-004E NEG 67-001B NEG 67-001C NEG 67-001D NEG 67-003B NEG 69-005B NEG 69-005C NEG 69-001D NEG 69-007A T2 68-002A T2 68-003A T14 67-002A T14 67-004A T14 69-002A T14 69-003A LOU 69-801A DEC 69-901A DEC 69-901B RB 76-002B HO 70-901A HO 70-901B HO 70-901C HO 70-902D HO 70-902E HO 70-902F

274 274 274 275 275 275 276 276 276 277 277 277 278 278 278 278 279 279 279 280 280 284–285 287 290 300 300 300 300 300 300 300 300 300 301 301 301 301 301 301 301 301 301 302 302 302 302 302 302 302 302 302 302 303

ILLUSTRATIONS 6.30. 6.31. 6.32. 6.33. 6.34. 6.35. 6.36. 6.37. 6.38. 6.39. 6.40. 6.41. 6.42. 6.43. 6.44. 6.45. 6.46. 6.47. 6.48. 6.49. 6.50. 7.1. 7.2. 7.3. 7.4. 7.5. 7.6. 7.7. 7.8. 7.9. 7.10. 7.11. 7.12. 7.13. 7.14. 7.15. 7.16. 7.17. 7.18. 7.19. 7.20. 7.21. 7.22. 7.23. 7.24. 7.25. 7.26.

HO 70-902H HO 70-902I HO 70-902J WF 62-001A TC 60-001A HO 70-902?, linear enamel hypoplasia in C1L NEG 69-004B, fractured upper incisors and mandibular AMTL NEG 69-004C, calculus, periodontitis, and acute abscess in periapical granuloma in P3R–M3R NEG 67-001C, caries and periodontitis in P3R–M3R T14 67-002A, caries and periodontitis in M1R–M3R WF 62-001A, gross caries in M1R NEG 67-001C, caries, periodontitis, acute abscesses in periapical granulomata and periodontal cyst, and AMTL in upper and lower anterior teeth NEG 69-007A, periodontitis, acute abscess in periodontal cyst, and AMTL in P4R–M3R NEG 69-007A, periodontitis, acute abscess in periapical granuloma, and AMTL in C1L–M3L T14 67-002A, fracture, acute abscess in perapical granuloma, and AMTL in P4L–M3L HO 70-901C, caries, periodontitis, acute abscesses in periodontal cysts, and AMTL in P4R–M1R WF 62-001A, large periapical cavities around roots of M1L draining into maxillary sinus Distribution of ages of growth disruptions causing enamel hypoplasia Frequencies of defects and AMTL in alveoli of Late Roman to Early Byzantine date by age Frequencies of defects and AMTL in alveoli of Late Roman date by age Sites of Roman and Byzantine date with published teeth for comparative study NEG 69-004B, three right fibula fragments with periostitis; NEG 69-004?, unsided fibula fragment with periostitis NEG 69-004E, right shoulder with degenerative changes from rotator cuff disease: right proximal humerus; right glenoid cavity NEG 67-001A, sacrum with sacralized coccyx NEG 67-001A, right ulna with healed fracture NEG 67-001B, sternum with manubriomesosternal fusion and ossified first sternocostal cartilage NEG 67-001B, sacrum with sacralized coccyx NEG 67-001C, vertebral joint disease: C2–C3 with apophyseal ankylosis; T3–T4 with apophyseal ankylosis and spondylosis deformans NEG 67-001F, frontal with healing cribra orbitalia in both orbits NEG 67-003A, left radius with healed fracture NEG 69-001?, right metacarpal IV with enchondroma NEG 69-007A, ossified costal cartilage NEG 69-007A, right and left first carpometacarpal joints with severe osteoarthritis: right and left metacarpals I; left trapezium NEG 69-007B, left frontal with healed cribra orbitalia T2 68-003B, ossified left conoid ligament: coracoid process of scapula; clavicle T14 67-002A, sacrum with spina bifida occulta and sacralized coccyx T14 67-002A, right proximal humerus with degenerative changes from rotator cuff disease T14 67-002A, right and left first interphalangeal joints with severe osteoarthritis T14 67-002A, right and left radii(?) and ulnae with healed fractures; pseudarthrosis in right ulna T14 67-002A, left fifth finger with traumatic ankylosis from interphalangeal fractures T14 67-002A, left middle or lower rib with healed fracture T14 67-004A, sacrum with sacralized lumbar T14 67-004A, right and left shoulders with traumatic and degenerative changes from severe chronic anterior dislocation: right scapula and humerus and left scapula; right and left glenoid cavities T14 69-002A, sacrum with spina bifida occulta LOU 69-801A, left tibia with popliteal enthesophyte RB 76-002B, right ulna with healed fracture RB 76-002C, left tibia fragment with periostitis

xix 303 303 303 303 303 303 303 304 304 304 305 305 305 305 306 306 306 311 346 346 354 378 379 379 379 380 380 381 382 383 383 384 384 385 386 387 387 387 388 389 389 392 392 393 393 394 395

xx 7.27. 7.28. 7.29. 7.30. 7.31. 7.32. 7.33. 7.34. 7.35. 7.36. 7.37. 7.38. 7.39. 7.40. 7.41. 7.42. 7.43. 7.44. 7.45. 7.46. 7.47. 7.48. 7.49.

ILLUSTRATIONS RB 76-002C, frontal with healed depressed fracture RB 76-002C, L5 with bilateral spondylolysis HO 70-901A, left ulna and radius fragments with four healed fractures HO 70-901C, right and left tibiae with enthesophytes HO 70-901C, right parietal with perimortem perforation surrounded by radiating, concentric, and crushing fractures HO 70-902G, sacrum with sacralized coccyx HO 70-902H, right parietal with healed depressed fracture HO 70-902J, lumbarized sacrum with sacralized coccyx HO 70-902? (Z1), sternum with sternal aperture and ossified first sternocostal and xiphoid cartilage HO 70-902? (Z1), sacrum with sacralized coccyx HO 70-902? (Z5), sacrum with sacralized coccyx HO 70-902?, ossified costal cartilage, female type and male type HO 70-902?, left tibia fragment with periostitis HO 70-902? (Z3), right fibula with ossified anterior talofibular ligament and posttraumatic degenerative changes from severe ankle sprain HO 70-902? (Z5), right scapula with healed fracture T14 67-002A, lateral ends of left and right clavicles with degenerative porosity HO 70-902? (Z2), sternum with ossified first sternocostal and xiphoid cartilage HO 70-902? (Z4), L2–L5 with joint disease: L2–L5 with apophyseal osteoarthritis; L5 with intervertebral (osteo)chondritis and spondylosis deformans HO 70-902? (Z1), L3–L5 with joint disease: apophyseal osteoarthritis; spondylosis deformans NEG 67-001B, L1 with apophyseal osteoarthritis Vertebrae with intervertebral (osteo)chondritis and Schmorl’s nodes: NEG 67-001B, T6–T7; T14 67-004A, T8; T14 67-002A, middle Ts Skeletal distribution of 22 traumatic lesions Sites of Roman and Byzantine date with published bones for comparative paleoepidemiology

395 395 396 397 398 400 400 401 402 402 402 403 404 404 405 420 421 421 421 422 422 430 445

TABLES

1.1. 1.2. 1.3. 4.1. 4.2. 5.1. 5.2. 5.3. 5.4. 5.5. 5.6. 5.7. 5.8. 5.9. 5.10. 5.11. 5.12. 5.13. 5.14. 5.15. 5.16. 5.17. 5.18. 6.1. 6.2. 6.3. 6.4. 6.5. 6.6. 6.7. 6.8. 6.9. 6.10. 6.11. 6.12. 6.13. 6.14. 6.15. 6.16. 6.17.

Abbreviations for Areas of the Archaeological Site Periodization of the Early Roman to Early Modern Settlement Factors in the Preservation of the Graves and Human Remains Material Components of Mortuary Behavior in 30 Graves by Phase Distribution of the Position of the Arms in 22 Bodies by Period Periodization of 65 Skeletons Survival of Erupted Dentition in 61 Skeletons Sex and Age at Death of 65 Skeletons Postcranial Measurements in Subadult Skeletons of Late Roman to Early Byzantine Date Cranial and Mandibular Measurements in Adult Skeletons of Late Roman to Early Byzantine Date Cranial and Mandibular Measurements in Adult Skeletons of Late Roman Date Cranial and Mandibular Measurements in Seven Adult Skeletons of Disparate Dates Postcranial Measurements in Adult Skeletons of Late Roman to Early Byzantine Date Postcranial Measurements in Adult Skeletons of Late Roman Date Postcranial Measurements in Six Adult Skeletons of Disparate Dates Nonmetric Traits in Subadult Skeletons of Late Roman and Early Byzantine Date Nonmetric Traits in Adult Skeletons of Late Roman to Early Byzantine Date Nonmetric Traits in Adult Skeletons of Late Roman Date Nonmetric Traits in Six Adult Skeletons of Disparate Dates Cranial and Postcranial Indices among Skeletons of Late Roman to Early Byzantine Date Estimation of Living Stature in 26 Adults of Late Roman to Early Byzantine Date Comparison of Stature in Several Greek Skeletal Samples of Roman to Modern Date Coincidence of Nonmetric and Congenital Traits in Multiple Burials Variation in Dental Form and Location Instances of Linear Enamel Hypoplasia Frequency of Enamel Hypoplasia by Period and Sex Mean Wear Scores in Permanent Dentition of Late Roman to Early Byzantine Date by Age Mean Wear Scores in Permanent Dentition of Late Roman Date by Age Angle of Wear in Permanent First Mandibular Molars by Wear Stage Mean Wear Scores in Permanent Dentition by Period Mean Wear Scores in Permanent Dentition by Sex Instances of Dental Trauma Instances of Calculus Frequency of Calculus by Period Frequency of Calculus by Sex Instances of Dental Caries Frequency of Carious Teeth of Late Roman to Early Byzantine Date by Dental Class Frequency of Carious Lesions of Late Roman to Early Byzantine Date by Dental Site Frequency of Carious Teeth of Late Roman to Early Byzantine Date by Age Frequency of Carious Teeth of Late Roman Date by Age

8 9 12–13 154–158 185 237 246 250–251 255 256 257 258 260–261 262–263 264–265 266 267 268 269 270 287 288–289 291 299 308–309 310 315 315 316 317 317 320 325 326 326 328–329 329 330 330 330

xxii 6.18. 6.19. 6.20. 6.21. 6.22. 6.23. 6.24. 6.25. 7.1. 7.2. 7.3. 7.4. 7.5. 7.6. 7.7. 7.8. 7.9. 7.10. 7.11. 7.12. 7.13.

TABLES Frequency of Carious Teeth by Period and Sex Instances of Alveolar Defects Instances of AMTL Frequencies of Defects and AMTL in Alveoli of Late Roman to Early Byzantine Date by Age Frequencies of Defects and AMTL in Alveoli of Late Roman Date by Age Frequencies of Alveoli with Defects and AMTL by Period and Sex Dental Samples from Roman and Byzantine Sites Cariosity in Several Greek Dental Samples of Roman and Byzantine Date Paleopathology in 65 Skeletons Instances of Cribra Orbitalia Instances of Joint Disease by Site Frequency of Joint Disease by Joint in Adults of Late Roman to Early Byzantine Date Frequency of Joint Disease by Joint in Adults of Late Roman Date Frequency of Joint Disease in Adult Vertebrae of Late Roman to Early Byzantine Date Frequency of Joint Disease in Adult Vertebrae of Late Roman Date Distribution of Joint Disease during the Late Roman to Early Byzantine Periods by Age Distribution of Joint Disease during the Late Roman Period by Age Distribution of Joint Disease by Sex and Period Frequency of Trauma in Skeletons of Late Roman to Early Byzantine Date by Site Frequency of Trauma in Skeletons of Late Roman Date by Site Paleopathology in Several Greek Skeletal Samples of Roman and Byzantine Date

331 338–342 343 345 345 347 352–353 360–361 369–373 374 375–377 417 418 419 419 423 423 423 431 431 442–444

BIBLIOGRAPHY AND ABBREVIATIONS

Abrahamse, D. Z. de F. 1979. “Images of Childhood in Early Byzantine Hagiography,” Journal of Psychohistory 6, pp. 497–517. ————. 1984. “Rituals of Death in the Middle Byzantine Period,” GOTR 29, pp. 125–134. Adams, J. C., and D. L. Hamblen. 1999. Outline of Fractures Including Joint Injuries, 11th ed., Edinburgh. Adams, W. Y., and E. W. Adams. 1991. Archaeological Typology and Practical Reality: A Dialectical Approach to Artifact Classification and Sorting, Cambridge. Agelarakis, A. 1997. “Excavations at Polystylon (Abdera) Greece: Aspects of Mortuary Practices and Skeletal Biology,” ArchDelt 47–48, Α΄ (1992–1993), pp. 292–308. Agelarakis, A., and A. Agelarakis. 1989. “The Paleopathological Evidence at Polystylon, Abdera,” BF 14, pp. 9–26. Agora = The Athenian Agora: Results of Excavations Conducted by the American School of Classical Studies at Athens, Princeton V = H. S. Robinson, Pottery of the Roman Period: Chronology, 1959. VII = J. Perlzweig, Lamps of the Roman Period, 1961. XX = A. Frantz, The Church of the Holy Apostles, 1971. XXIV = A. Frantz, Late Antiquity: a.d. 267–700, 1988. Alexiou, M. 1978. “Modern Greek Folklore and Its Relation to the Past: The Evolution of Charos in Greek Tradition,” in The “Past” in Medieval and Modern Greek Culture, ed. S. Vryonis, Malibu, pp. 221–236. ————. 2002. The Ritual Lament in Greek Tradition, 2nd ed., rev. D. Yatromanolakis and P. Roilos, Lanham, Md. Amandry, P. 1989. “La ruine du temple d’Apollon à Delphes,” Bulletin de la Classe des lettres et des sciences morales et politiques de la Académie royale de Belgique ser. 5, 75, pp. 26–47. Ambraseys, N. N., and A. Jackson. 1990. “Seismicity and Associated Strain of Central Greece between 1890 and 1988,” Geophysical Journal International 101, pp. 663–708. ————. 1996. “Seismicity and Strain in the Gulf of Corinth (Greece) since 1694,” Journal of Earthquake Engineering 1, pp. 433–474. Anagnostakis, I. 1997. “Η χειροποίητη κεραμική ανάμεσα στην ιστορία και την αρχαιολογία ,” Βυζαντιακά 17, pp. 287–330. ————. 2000. “Οι πελοποννησιακοί σκοτεινοί χρόνοι: το σλαβικό πρόβλημα. Μεταμορφώσεις της Πελοποννήσου ή της έρευνας,” in Οι μεταμορφώσεις της Πελοποννήσου (4 ο ς –15 ος αι .), ed. E. Grammatikopoulou, Athens,

pp. 19–34. Anagnostakis, I., and N. Poulou-Papadimitriou. 1997. “Η πρωτοβυζαντινή Μεσσήνη (5ος–7ος αιώνας) και προβλήματα της χειροποίητης κεραμικής στην Πελοπόννησο,” Σύμμεικτα 11, pp. 229–322.

Anderson, J. K. 1967. “Corinth: Temple E Northwest, Preliminary Report, 1965,” Hesperia 36, pp. 1–12. Anderson-Stojanović, V. R. 1987. “The Chronology and Function of Ceramic Unguentaria,” AJA 91, pp. 105– 122. Anderson-Stojanović, V. R., and J. E. Jones. 2002. “Ancient Beehives from Isthmia,” Hesperia 71, pp. 345–376. Andreadis, M. 1980. Ἡ Κορινθία μετὰ τὸ 1821 1: 1823–1833, Athens. ————. 1996. Η Κορινθία την Οθωνική περίοδο 2: 1833–1862, Athens. ————. 2001. Η Κορινθία τον 19ο αιώνα 3: 1863–1872, Athens. Angel, J. L. 1941. “Physical Types in Ancient Corinth,” AJA 45, p. 88 (abstract). ————. 1944a. “A Racial Analysis of the Ancient Greeks: An Essay on the Use of Morphological Types,” AJPA 2, pp. 329–376. ————. 1944b. “Greek Teeth: Ancient and Modern,” HB 16, pp. 281–297. ————. 1945a. “Race, Type, and Ethnic Group in Ancient Greece,” HB 18, pp. 1–32. ————. 1945b. “Skeletal Material from Attica,” Hesperia 14, pp. 279–363. ————. 1945–1948. “Roman Tombs at Vasa, Appendix III: The Skulls,” RDAC 1945–1948, pp. 68–76. ————. 1951. Troy: The Human Remains, Princeton. ————. 1964. “Osteoporosis: Thalassemia?” AJPA 22, pp. 369–374. ————. 1966. “Porotic Hyperostosis, Anemias, Malarias, and Marshes in the Prehistoric Eastern Mediterranean,” Science n.s. 153, pp. 760–763. ————. 1967. “Porotic Hyperostosis or Osteoporosis Symmetrica,” in Diseases in Antiquity: A Survey of the Diseases, Injuries and Surgery of Early Populations, ed. D. R. Brothwell and A. T. Sandison, Springfield, Ill., pp. 378–389. ————. 1969. “Paleodemography and Evolution,” AJPA 31, pp. 343–353. ————. 1972. “Appendix B. Late Bronze Age Cypriots from Bamboula: The Skeletal Remains,” in J. L. Benson, Bamboula at Kourion: The Necropolis and the Finds Excavated by J. F. Daniel, Philadelphia, pp. 148–158. ————. 1973. “Early Neolithic People of Nea Nikomedeia,” in Die Anfänge des Neolithikums vom Orient bis Nordeuropa 8a.1: Anthropologie (Fundamenta B.3), ed. I. Schwidetzky, Cologne, pp. 103–112. ————. 1974. “Patterns of Fractures from Neolithic to Modern Times,” Anthropológiai Közlemények 18, pp. 9– 18.

xxiv

BIBLIOGRAPHY AND ABBREVIATIONS

————. 1975. “Paleoecology, Paleodemography and Health,” in Population, Ecology and Social Evolution, ed. S. Polgar, The Hague, pp. 167–190. ————. 1977. “Anemias in Antiquity,” in Porotic Hyperostosis: An Enquiry (Paleopathology Association Monograph 2), ed. A. Cockburn, Detroit, pp. 1–5. ————. 1978. “Porotic Hyperostosis in the Eastern Mediterranean,” Medical College of Virginia Quarterly 14.1, pp. 10–16. ————. 1979. “Osteoarthritis in Prehistoric Turkey and Medieval Byzantium,” in Arthritis: Modern Concepts and Ancient Evidence (Paleopathology Association Monograph 3 = Henry Ford Hospital Medical Journal 27.1), ed. E. Cockburn, H. Duncan, and J. M. Riddle, Detroit, pp. 38–43. ————. 1984. “Health as a Crucial Factor in the Changes from Hunting to Developed Farming in the Eastern Mediterranean,” in Cohen and Armelagos 1984, pp. 51–73. Angold, M. 1995. Church and Society in Byzantium under the Comneni, 1081–1261, Cambridge. Antoniadis-Bibicou, H. 1973. “Quelques notes sur l’enfant de la moyenne époque byzantine,” Annales de démographie historique 1973, pp. 77–84. Arnold, B., and N. L. Wicker. 2001. “Introduction,” in Gender and the Archaeology of Death, ed. B. Arnold and N. L. Wickers, Walnut Creek, Cal., pp. vii–xxi. Ashley, G. T. 1954. “The Morphological and Pathological Significance of Synostosis at the Manubrio-Sternal Joint,” Thorax 9, pp. 159–166. ————. 1956. “The Relationship between the Pattern of Ossification and the Definitive Shape of the Mesosternum in Man,” JAnat 60, pp. 87–105. Athanassopoulou-Pennas, V. 1981. “‘Θησαυρὸς’ νομισμάτων 6ου μ.Χ. αἰ. ἀπὸ τὴν περιοχὴ τῶν Θηβῶν,” ArchEph 1979, pp. 200–213. Aufderheide, A. C., and C. Rodríguez-Martín. 1998. The Cambridge Encyclopedia of Human Paleopathology, Cambridge. Aupert, P. 1980a. “Céramique slave à Argos (585 ap. J.-C.),” in Études argiennes (BCH Suppl. 6), Paris, pp. 373–394. ————. 1980b. “Objets de la vie quotidienne à Argos en 585 ap. J.-C.,” in Études argiennes (BCH Suppl. 6), Paris, pp. 395–457. ————. 1987. “Rapports sur les travaux de l’École française en Grèce en 1986: Argos. 3. Thermes A,” BCH 111, pp. 597–603. ————. 1988. “Rapports sur les travaux de l’École française en Grèce en 1987: Argos. 3. Temples et thermes A,” BCH 112, pp. 710–715. ————. 1989. “Les slaves à Argos,” BCH 113, pp. 417–419. Avramea, A. 1983. “Νομισματικοὶ ‘θησαυροὶ’ καὶ μεμονωμένα νομίσματα ἀπὸ τὴν Πελοπόννησο (ΣΤ΄–Ζ΄ αἰ.),” Σύμμεικτα 5, pp. 49–90. ————. 1997. Le Péloponnèse du IV e au VIIIe siècle: Changements et persistances (Byzantina Sorbonensia 15), Paris. ————. 2000. “Η παλαιοχριστιανική και πρωτοβυζαντινή Πελοπόννησος,” in Οι μεταμορφώσεις της Πελοποννήσου (4 ος –15 ος αι .), ed. E. Grammatikopoulou, Athens, pp. 9–18. ————. 2005. “Les villages de Thessalie, de Grèce centrale, et du Péloponnèse (ve–xive siècle),” in Lefort, Morrisson, and Sodini 2005, pp. 213–223. Avramea, A., and M. Kyrkou. 1988. “Inventaire topographique de Corinthe et sa région à l’époque chrétienne et byzantine,” in Géographie historique du monde méditerranéen (Byzantina Sorbonensia 7), Paris, pp. 31–45.

Bagnall, R. S. 1995. “Women, Law, and Social Realities in Late Antiquity: A Review Article,” BASP 32, pp. 65– 86. Bakirtzis, C. 1986. “ Ἀνασκαφὴ Πολυστύλου Ἀβδήρων,” Prakt 1983, A΄, pp. 13–19. ————. 1998. “Paul and Philippi: The Archaeological Evidence,” in Philippi at the Time of Paul and after His Death, ed. C. Bakirtzis and H. Koester, Harrisburg, Pa., pp. 37–48. Banaka-Dimaki, A. 1988a. “Δ´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Νομός Κορινθίας. Ανασκαφικές εργασίες. Αρχαία Κόρινθος,” ArchDelt 35, B΄1 (1980),

pp. 105–109. ————. 1988b. “Δ´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Νομός Κορινθίας. Ανασκαφικές εργασίες ,” ArchDelt 36, B΄1 (1981), pp. 86–94. ————. 1998. “Ρωμαϊκά ταφικά μνημεία του ´Αργους,” in Recherches franco-hélleniques 3: Argos et l’Argolide, topographie et urbanisme, ed. A. Pariente and G. Touchais, Paris, pp. 385–395. Barakari-Gleni, K. 1988. “ Δ´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Νομός Αργολίδας. Ανασκαφικές εργασίες. Άργος,” ArchDelt 35, B΄1 (1980), pp. 119–123.

Baratte, F. 1984. “Les témoignages archéologiques de la présence slave au sud du Danube,” in Villes et peuplement dans l’Illyricum protobyzantin (CÉFR 77), Rome, pp. 163–180. Barbouna IV = I. Hägg and J. M. Fossey, The Hellenistic Necropolis and Later Structures on the Middle Slopes, 1973–1977 (Excavations in the Barbouna Area at Asine IV), Uppsala 1980. Barford, P. M. 2001. The Early Slavs: Culture and Society in Early Medieval Eastern Europe, Ithaca. Barnes, E. 1994. Developmental Defects of the Axial Skeleton in Paleopathology, Niwot, Colo. ————. 2003. “The Dead Do Tell Tales,” in Corinth XX, pp. 435–443. Bass, W. 1995. Human Osteology: A Laboratory and Field Manual, 4th ed., Columbia, Mo. Beaton, A. E., and P. A. Clement. 1976. “The Date of the Destruction of the Sanctuary of Poseidon on the Isthmus of Corinth,” Hesperia 45, pp. 267–279. Beaucamp, J. 1977. “La situation juridique de la femme à Byzance,” Cahiers de civilisation médiévale, xe-xiie siècle 20, pp. 145–176. ————. 1990–1992. La statut de la femme à Byzance (4e–7e siècle), 2 vols., Paris. Beck, H.-G. 1979. Die Byzantiner und ihr Jenseits: Zur Entstehungsgeschichte einer Mentalität (SBMünch 6), Munich. Beck, L. A., ed. 1995a. Regional Approaches to Mortuary Analysis, New York. ————. 1995b. “Regional Cults and Ethnic Boundaries in ‘Southern Hopewell,’” in Beck 1995a, pp. 167– 187. Becker, M. J. 2005. “Appendix A: The Human Skeletal Remains,” in Sweetman 2005, pp. 370–383. Bedford, M. E., K. F. Russell, C. O. Lovejoy, R. S. Meindl, S. W. Simpson, and P. L. Stuart-Macadam. 1993. “Test of the Multifactorial Aging Method Using Skeletons with Known Ages-at-Death from the Grant Collection,” AJPA 91, pp. 287–297. Beeley, J. G., and D. A. Lunt. 1980. “The Nature of the Biochemical Changes in Softened Dentine from Archaeological Sites,” JAS 7, pp. 371–377. Bélios, G. 1933. L’histoire du paludisme en Grèce depuis l’antiquité jusqu’à la découverte de Laveran, Paris.

BIBLIOGRAPHY AND ABBREVIATIONS Berbati-Limnes = B. Wells and C. Runnels, eds., The BerbatiLimnes Archaeological Survey 1988–1990 (SkrAth 4º, 44), Stockholm 1996. Berry, A. C. 1978. “Anthropological and Family Studies on Minor Variants of the Dental Crowns,” in Development, Function, and Evolution of Teeth, ed. P. M. Butler and K. A. Joysey, London, pp. 81–98. Berry, R. J., and A. C. Berry. 1967. “Epigenetic Variation in the Human Cranium,” JAnat 101, pp. 361–379. Berryman, H. E., and S. Jones Haun. 1996. “Applying Forensic Techniques to Interpret Cranial Fracture Patterns in an Archaeological Specimen,” IJO 6, pp. 2–9. Berryman, H. E., and S. A. Symes. 1998. “Recognizing Gunshot and Blunt Cranial Trauma through Fracture Interpretation,” in Forensic Osteology: Advances in the Identification of Human Remains, 2nd ed., ed. K. J. Reichs, Springfield, Ill., pp. 333–352. Besser, H. 2003. Introduction to Imaging, rev. S. Hubbard and D. Lenert, Los Angeles. Biers, W. R. 1971. “Excavations at Phlius, 1970,” Hesperia 40, pp. 424–447. ————. 1973. “Excavations at Phlius, 1972,” Hesperia 41, pp. 103–120. Bigliani, L. U., P. M. Newton, S. P. Steinman, P. M. Connor, and S. J. McIlveen. 1998. “Glenoid Rim Lesions Associated with Recurrent Anterior Dislocation of the Shoulder,” The American Journal of Sports Medicine 26, pp. 41–45. Binford, L. R. 1971. “Mortuary Practices: Their Study and Their Potential,” in Approaches to the Social Dimensions of Mortuary Practices (Memoirs of the Society for American Archaeology 25), ed. J. A. Brown, Washington, D.C., pp. 6–29. Blackman, D. 1999. “Archaeology in Greece 1998–1999: Corinth,” AR 45, pp. 21–23. ———. 2000. “Archaeology in Greece 1999–2000: Corinth,” AR 46, pp. 24–25. ———. 2002. “Archaeology in Greece 2001–2002: Corinth,” AR 48, pp. 17–20. Bloch, M. 1982. “Death, Women, and Power,” in Death and the Regeneration of Life, ed. M. Bloch and J. Parry, Cambridge, pp. 211–230. Blouet, G.-A. 1838. Expédition scientifique de Morée, ordonnée par le gouvernement français: Architecture, sculptures, inscriptions et vues du Péloponèse, des Cyclades et de l’Attique 3, Paris. Boddington, A. 1987. “Raunds, Northamptonshire: Analysis of a Country Churchyard,” WorldArch 18, pp. 411– 425. Boddington, A., A. N. Garland, and R. C. Janaway, eds. 1987. Death, Decay and Reconstruction: Approaches to Archaeology and Forensic Science, Manchester. Boev, P. 1972. Die Rassentypen der Balkanhalbinsel und der Ostägäischen Inselwelt und deren Bedeutung für die Herkunft ihrer Bevölkerung, Sofia. Bon, A. 1951. Le Péloponnèse byzantin jusqu’en 1294, Paris. Bookidis, N., and J. E. Fisher. 1972. “The Sanctuary of Demeter and Kore on Acrocorinth. Preliminary Report IV: 1969–1970,” Hesperia 41, pp. 283–331. Bory de Saint-Vincent, J. B. G. M. 1834. Expédition scientifique de Morée, section des sciences physiques 2.1: Géographie, Paris. Boswell, J. 1988. The Kindness of Strangers: The Abandonment of Children in Western Europe from Late Antiquity to the Renaissance, Chicago. Bourbou, C. E. 2000. “Παλαιοπαθολογική μελέτη και ανάλυση του πληθυσμού,” in Πρωτοβυζαντινή Ελεύθερνα, τόμεας Ι 2, ed. P. G. Themelis, Rethymnon, pp. 291–319.

xxv

———. 2003. “Health Patterns of Proto-Byzantine Populations (6th–7th Centuries a.d.) in South Greece: The Cases of Eleutherna (Crete) and Messene (Peloponnese),” IJO 13, pp. 303–313. ———. 2004. The People of Early Byzantine Eleftherna and Messene (6th–7th Centuries a.d. ): A Bioarchaeological Approach, Athens. Bowden, W. 2001. “A New Urban Élite? Church Builders and Church Building in Late-Antique Epirus,” in Lavan 2001b, pp. 89–105. ———. 2003. Epirus vetus: The Archaeology of a Late Antique Province, London. Bowden, W., L. Lavan, and C. Machado, eds. 2004. Recent Research on the Late Antique Countryside (Late Antique Archaeology 2), Leiden. Bowersock, G. W. 1990. Hellenism in Late Antiquity, Ann Arbor. ———. 1996. “The Vanishing Paradigm of the Fall of Rome,” Bulletin of the American Academy of Arts and Sciences 149, pp. 29–43. Bowersock, G. W., P. Brown, and O. Grabar. 1999. “Introduction,” in Late Antiquity: A Guide to the Postclassical World, ed. G. W. Bowersock, P. Brown, and O. Grabar, Cambridge, Mass., pp. vii–xiii. Bozóky, E., and A.-M. Helvétius, eds. 1999. Les reliques: Objets, cultes, symboles. Actes du colloque international de l’Université du Littoral-Côte d’Opale (Boulogne-sur-Mer), 4–6 septembre 1997 (Hagiologia 1), Turnhout. Bräuer, G. 1980. “Bericht über die anthropologische Untersuchung eines Skelettes aus der Unterburg von Tiryns,” ArchKorrBl 10, pp. 291–293. Breccia, G. 2004. “L’arco e la spada: Procopio e il nuovo esercito bizantino,” Nea Rhome 1, pp. 72–99. Bridges, P. S. 1991. “Degenerative Joint Disease in Hunter-Gatherers and in Agriculturalists from the Southeast of the United States,” AJPA 85, pp. 379– 391. ———. 1992. “Prehistoric Arthritis in the Americas,” Annual Review of Anthropology 21, pp. 67–91. ———. 1993. “The Effect of Variation in Methodology on the Outcome of Osteoarthritic Studies,” IJO 3, pp. 289– 295. ———. 1994. “Vertebral Arthritis and Physical Activities in the Prehistoric Southeastern United States,” AJPA 93, pp. 83–93. Broneer, O. 1926. “Excavations at Corinth, 1925: Area North of Basilica,” AJA 30, pp. 49–57. ———. 1953. “Excavations at Isthmia, 1952,” Hesperia 22, pp. 182–195. ———. 1955. “Excavations at Isthmia, 1954,” Hesperia 24, pp. 110–141. ———. 1958. “Excavations at Isthmia, Third Campaign, 1955–1956,” Hesperia 27, pp. 1–37. ———. 1959. “Excavations at Isthmia, Fourth Campaign, 1957–1958,” Hesperia 28, pp. 298–343. ———. 1962a. “Excavations at Isthmia, 1959–1961,” Hesperia 31, pp. 1–25. ———. 1962b. “The Apostle Paul and the Isthmian Games,” BiblArch 25, pp. 2–31. ———. 1971. “Paul and the Pagan Cults at Isthmia,” HTR 64, pp. 169–187. Brook, A. H. 1984. “A Unifying Aetiological Explanation for Anomalies of Human Tooth Number and Size,” ArchOB 29, pp. 373–378. 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, pp. 227–238.

xxvi

BIBLIOGRAPHY AND ABBREVIATIONS

Brothwell, D. R. 1967a. “The Evidence for Neoplasms,” in Disease in Antiquity: A Survey of the Diseases, Injuries, and Surgery of Early Populations, ed. D. R. Brothwell and A. T. Sandison, Springfield, Ill., pp. 320–345. ———. 1967b. “Major Congenital Anomalies of the Skeleton: Evidence from Earlier Populations,” in Disease in Antiquity: A Survey of the Diseases, Injuries, and Surgery of Early Populations, ed. D. R. Brothwell and A. T. Sandison, Springfield, Ill., pp. 423–443. ———. 1981. Digging Up Bones, 3rd ed., Ithaca. ———. 1986. “The Human Bones,” in Excavations at Saraçhane in Istanbul I: The Excavations, Structures, Architectural Decoration, Small Finds, Coins, Bones, and Molluscs, ed. R. M. Harrison, Princeton, pp. 374–398. Brothwell, D. R., and P. Brothwell. 1998. Food in Antiquity: A Survey of the Diet of Early Peoples, exp. ed., Baltimore. Brown, E. A. R. 1985. “Burying and Unburying the Kings of France,” in Persons in Groups: Social Behavior as Identity Formation in Medieval and Renaissance Europe (Medieval and Renaissance Text and Studies 36), ed. R. C. Trexler, Binghamton, N.Y., pp. 241–266. Brown, J. A. 1981. “The Search for Rank in Prehistoric Burials,” in Chapman, Kinnes, and Randsborg 1981, pp. 25–37. ———. 1995. “On Mortuary Analysis—with Special Reference to the Saxe-Binford Research Program,” in Beck 1995a, pp. 3–26. Brown, M. A. 1983. “Grave Orientation: A Further View,” ArchJ 140, pp. 322–328. Brown, P. 1981. The Cult of the Saints: Its Rise and Function in Latin Christianity, Chicago. Bruce-Chwatt, L. J., and J. de Zulueta. 1980. The Rise and Fall of Malaria in Europe: A Historico-Epidemiological Study, Oxford. Bruneau, P. 1970. “Tombes d’Argos,” BCH 94, pp. 437– 531. Bryer, A. 2002. “The Means of Agricultural Production: Muscle and Tools,” in EHB 1, pp. 101–113. Buchet, J.-L., and J.-P. Sodini. 1984. “Les tombes,” in Aliki II: La basilique double (Études thasiennes 10), ed. J.-P. Sodini and K. Kolokotsas, Paris, pp. 211–242. Buikstra, J. E. 1997. “Paleodemography: Context and Promise,” in Integrating Archaeological Demography: Multidisciplinary Approaches to Prehistoric Population, ed. R. R. Paine, Carbondale, Ill., pp. 367–380. Buikstra, J. E., L. W. Konigsberg, and J. Bullington. 1986. “Fertility and the Development of Agriculture in the Prehistoric Midwest,” AmerAnt 31, pp. 528–546. Buikstra, J. E., and J. H. Mielke. 1985. “Demography, Diet and Health,” in Gilbert and Mielke 1985, pp. 359– 422. Buikstra, J. E., and D. H. Ubelaker. 1994. Standards for Data Collection from Human Skeletal Remains (Arkansas Archaeological Survey Research Series 44), Fayetteville, Ark. Burkert, W. 1983. Homo necans: The Anthropology of Ancient Greek Sacrificial Ritual and Myth, trans. P. Bing, Berkeley. Burns, P. E. 1979. “Log-Linear Analysis of Dental Caries Occurrence in Four Skeletal Series,” AJPA 51, pp. 637–648. ———. 1980. “A Multivariate Study of Dental Caries in Ancient Greece,” AJPA 52, p. 209 (abstract). ———. 1982. “A Study of Sexual Dimorphism in the Dental Pathology of Ancient Peoples” (diss. Arizona State Univ.). Bursian, C. 1872. Geographie von Griechenland 2: Peloponnesos und Inseln, Leipzig.

Bynum, C. W. 1995. The Resurrection of the Body in Western Christianity, 200–1336, New York. Caglar, E., O. O. Kuscu, N. Sandalli, and I. Arib. 2007. “Prevalence of Dental Caries and Tooth Wear in a Byzantine Population (13th c. a.d.) from Northwest Turkey,” ArchOB 52, pp. 1136–1145. Caillet, J.-P. 1996. “La transformation en église d’édifices publics et de temples à la fin de l’Antiquité,” in La fin de la cité antique et le début de la cité médiévale: De la fin du iiie siècle à l’avènement de Charlemagne, ed. C. Lepelley, Bari, pp. 191–211. Calcagno, J. M., and K. R. Gibson. 1988. “Human Dental Reduction: Natural Selection or the Probable Mutation Effect,” AJPA 77, pp. 505–517. Callegher, B. 2005. “La circulation monétaire à Patras et dans les sites ruraux environnants (vie–viie siècle),” in Lefort, Morrisson, and Sodini 2005, pp. 225–235. Cameron, A. 1993a. The Later Roman Empire a.d. 284–430, Cambridge, Mass. ———. 1993b. The Mediterranean World in Late Antiquity a.d. 395–600, London. ———. 1998. “The Perception of Crisis,” in Morfologie sociali e culturali in Europa fra tarda antichità e alto medioevo 1 (Settimane di studio del Centro italiano di studi sull’alto medioevo 45), Spoleto, pp. 9–31. ———. 2001. “A Response,” in Lavan 2001b, pp. 238–239. Cameron, A., B. Ward-Perkins, and M. Whitby, eds. 2000. Late Antiquity: Empire and Successors, a.d. 425–600 (CAH 2 14), Cambridge. Cannon, A. 2005. “Gender and Agency in Mortuary Fashion,” in Rakita et al. 2005, pp. 41–65. Cantarella, E. 1991. I supplizi capitali in Grecia e a Roma, Milan. Cantino Wataghin, G. 1999. “The Ideology of Urban Burials,” in The Idea and Ideal of the Town Between Late Antiquity and the Early Middle Ages, ed. G. P. Brogiolo and B. Ward-Perkins, Leiden, pp. 147–163. Capasso, L. 1985. L’origine delle malattie, Chieti. Capasso, L., K. A. R. Kennedy, and C. A. Wilczak. 1999. Atlas of Occupational Markers on Human Remains (JPaleopath Monographic Publications 3), Teramo. Caraher, W. R., L. J. Hall, and R. S. Moore, eds. 2008. Archaeology and History in Roman, Medieval and PostMedieval Greece: Studies on Method and Meaning in Honor of Timothy E. Gregory, Aldershot. Carington Smith, J. 1982. “A Roman Chamber Tomb on the South-East Slopes of Monasteriaki Kephala, Knossos,” BSA 77, pp. 255–293. Carlson, D. S., G. J. Armelagos, and D. P. Van Gerven. 1974. “Factors Influencing the Etiology of Cribra Orbitalia in Prehistoric Nubia,” Journal of Human Evolution 3, pp. 405–410. Carlson, S. J. 1990. “Vertebrate Dental Structures,” in Skeletal Biomineralization: Patterns, Processes, and Evolutionary Trends, ed. J. G. Carter, New York, pp. 531–556. Carpenter, R. 1929. “Researches in the Topography of Ancient Corinth–I,” AJA 33, pp. 345–360. Carr, C. 1995. “Mortuary Practices: Their Social, Philosophical-Religious, Circumstantial, and Physical Determinants,” Journal of Archaeological Method and Theory 2, pp. 105–200. Caseau, B. 2004. “The Fate of Rural Sanctuaries in Late Antiquity and the Christianization of the Countryside,” in Bowden, Lavan, and Machado 2004, pp. 105– 144. Caster, M. [1937] 1987. Lucien et la pensée religieuse de son temps, repr. New York.

BIBLIOGRAPHY AND ABBREVIATIONS Catling, H. W. 1988. “Archaeology in Greece 1987–88: Isthmia,” AR 34, pp. 21–22. Catling, H. W., and D. Smyth. 1976. “An Early Christian Osteotheke at Knossos,” BSA 71, pp. 25–47. Chandler, R. 1817. Travels in Asia Minor and Greece; or, An Account of a Tour Made at the Expense of the Society of the Dilettanti 2, 3rd ed., London. Chaniotis, A. 2005. “Ritual Dynamics in the Eastern Mediterranean: Case Studies in Ancient Greece and Asia Minor,” in Rethinking the Mediterranean, ed. W. V. Harris, Oxford, pp. 141–166. Chapman, R. 2005. “Mortuary Analysis: A Matter of Time?” in Rakita et al. 2005, pp. 25–40. Chapman, R., I. Kinnes, and K. Ransborg. 1981. The Archaeology of Death, London. Chapman, R., and K. Randsborg. 1981. “Approaches to the Archaeology of Death,” in Chapman, Kinnes, and Randsborg 1981, pp. 1–24. Charitonides, S. 1968. “ Ἀρχαιότητες καὶ μνημεῖα Ἀργολιδοκορινθίας ,” ArchDelt 21, B ΄ 1 (1966), pp. 121– 132. Charles, R. P. 1958a. “Les populations de la Grèce antique (état de la question en 1957),” Bulletins et mémoires de la Société d’anthropologie de Paris ser. 10, 9, pp. 169– 190. ———. 1958b. “Étude anthropologique des nécropoles d’Argos: Contribution à l’étude des populations de la Grèce antique,” BCH 82, pp. 268–313. ———. 1962. “Recherches anthropologiques en Grèce (mission 1961),” Cahiers ligures de préhistoire et d’archéologie 11, pp. 226–234. ———. 1963. Étude anthropologique des nécropoles d’Argos: Contribution à l’étude des populations de la Grèce antique, suite (Études péloponnésiennes 3), Paris. Chen, Y. M., and S. P. Bohrer. 1990. “Coracoclavicular and Coracoacromial Ligament Calcification and Ossification,” Skeletal Radiology 19, pp. 263–266. Cherf, W. J. 1991. “Appendix: Dhema,” in The Great Isthmus Corridor Route: Explorations of the Phokis-Doris Expedition I, ed. E. W. Kase, G. J. Szemler, N. C. Wilkie, and P. W. Wallace, Dubuque, Iowa, pp. 56–59. ———. 1992. “Carbon-14 Chronology for the Late-Roman Fortifications of the Thermopylai Frontier,” JRA 5, pp. 261–264. Child, A. M. 1995. “Towards an Understanding of the Microbial Decomposition of Archaeological Bone in the Burial Environment,” JAS 22, pp. 165–174. Cho, B. P., and H. S. Kang. 1998. “Articular Facets of the Coracoclavicular Joint in Koreans,” Acta Anatomica 163, pp. 56–62. Christides, V. 1981. “The Raids of the Moslems of Crete in the Aegean Sea: Piracy and Conquest,” Byzantion 51, pp. 76–111. Christophilopoulou, A. 1993. Βυζαντινὴ ἱστορία 1: 324–610, 2nd ed., Thessaloniki. Chuvin, P. 2004. “Christianisation et résistance des cultes traditionelles. Approches actuelles et enjeux historiographiques,” in Hellénisme et christianisme, ed. M. Narcy and É. Rebillard, Villeneuve d’Ascq, pp. 15–34. Clark, G. 1993. Women in Late Antiquity: Pagan and Christian Life-Styles, Oxford. Clark, W. G. 1858. Peloponnesus: Notes of Study and Travel, London. Clarke, N. G., S. E. Carey, W. Srikandi, R. S. Hirsch, and P. I. Leppard. 1986. “Periodontal Disease in Ancient Populations,” AJPA 71, pp. 173–183.

xxvii

Clarke, N. G., and R. S. Hirsh. 1991. “Physiological, Pulpal, and Periodontal Factors Influencing Alveolar Bone,” in Kelley and Larsen 1991, pp. 241–266. Clement, P. A. 1969. “Isthmia Excavations,” ArchDelt 23, B΄1 (1968), pp. 137–143. ———. 1970. “Isthmia,” ArchDelt 24, B΄1 (1969), pp. 116–119. ———. 1972. “Isthmia Excavations,” ArchDelt 25, B΄1 (1970), pp. 161–167. ———. 1974. “Isthmia Excavations,” ArchDelt 26, B΄1 (1971), pp. 100–111. ———. 1975. “The Date of the Hexamilion,” in Essays in Memory of Basil Laourdas, Thessaloniki, pp. 159–164. ———. 1976. “Isthmia Excavations,” ArchDelt 27, B΄1 (1972), pp. 224–231. ———. 1977a. “Isthmia Excavations,” ArchDelt 28, B΄1 (1973), pp. 143–149. ———. 1977b. “Alaric and the Fortifications of Greece,” in Ancient Macedonia 2, Thessaloniki, pp. 135–137. ———. 1984. “Isthmia,” ArchDelt 31, B΄1 (1976), pp. 65–70. ———. 1987. “Isthmian Notes,” in Φίλια ἔπη εἰς Γεώργιον Ε. Μυλωνὰν διὰ τὰ 60 ἔτη τοῦ ἀνασκαφικοῦ του ἔργου 2 (ΒΑΑΕ 103), Athens, pp. 381–383. Clement, P. A., and M. M. Thorne. 1974. “From the West Cemetery at Isthmia,” Hesperia 43, pp. 401–411. Clover, F. M., and R. S. Humphreys, eds. 1989. Tradition and Innovation in Late Antiquity, Madison. Cockshott, W. P. 1992. “The Geography of Coracoclavicular Joints,” Skeletal Radiology 21, pp. 225–227. Cohen, M. N. 1994. “The Osteological Paradox Re-considered,” CurrAnthr 35, pp. 629–631. ———. 1997. “Does Paleopathology Measure Community Health? A Rebuttal of ‘The Osteological Paradox’ and its Implications for World Prehistory,” in Integrating Archaeological Demography: Multidisciplinary Approaches to Prehistoric Population (Center for Archaeological Investigations Occasional Papers 24), ed. R. R. Paine, Carbondale, Ill., pp. 242–260. Cohen, M. N., and G. J. Armelagos, eds. 1984. Paleopathology at the Origins of Agriculture, Orlando. Connor, C. L. 2004. Women of Byzantium, New Haven. Cooper, P. D., J. H. Stewart, and M. S. McCormick. 1988. “Development and Morphology of the Sternal Foramen,” American Journal of Forensic Medicine and Pathology 9, pp. 342–347. Corinth = Corinth: Results of Excavations Conducted by the American School of Classical Studies at Athens I = H. N. Fowler and R. Stillwell, Introduction, Topography, Architecture, Cambridge, Mass., 1932. I.2 = R. Stillwell, R. L. Scranton, and S. E. Freeman, Architecture, Cambridge, Mass., 1941. I.3 = R. L. Scranton, Monuments in the Lower Agora and North of the Archaic Temple, Princeton 1951. I.4 = O. Broneer, The South Stoa and its Successors, Princeton 1954. III.1 = C. W. Blegen, O. Broneer, R. Stillwell, and A. R. Bellinger, Acrocorinth: Excavations in 1926, Cambridge, Mass., 1930. III.2 = R. Carpenter and A. Bon, The Defenses of Acrocorinth and the Lower Town, Cambridge, Mass., 1936. VIII.1 = B. D. Meritt, Greek Inscriptions, 1896–1926, Cambridge, Mass., 1931. VIII.3 = J. H. Kent, The Inscriptions, 1926–1950, Princeton 1966. XI = C. H. Morgan, The Byzantine Pottery, Cambridge, Mass., 1942. XII = G. R. Davidson, The Minor Objects, Princeton 1952.

xxviii

BIBLIOGRAPHY AND ABBREVIATIONS

XIII = C. Blegen, H. Palmer, and R. S. Young, The North Cemetery, Princeton 1964. XIV = C. Roebuck, The Asklepieion and Lerna, Princeton 1951. XVI = R. L. Scranton, Mediaeval Architecture in the Central Area of Corinth, Princeton 1957. XVII = J. C. Biers, The Great Bath on the Lechaion Road, Princeton 1985. XVIII.2 = K. W. Slane, The Sanctuary of Demeter and Kore: The Roman Pottery and Lamps, Princeton 1990. XVIII.3 = N. Bookidis and R. S. Stroud, The Sanctuary of Demeter and Kore: Topography and Architecture, Princeton 1997. XX = C. K. Williams II and N. Bookidis, eds., The Centenary 1896–1996, Princeton 2003. Corrucini, R. S. 1991. “Anthropological Aspects of Orofacial and Occlusal Variations and Anomalies,” in Kelley and Larsen 1991, pp. 295–323. Courville, C. B. 1962a. “Forensic Neuropathology IV: Significance of Traumatic Extracranial and Cranial Lesion,” JFS 7, pp. 303–322. ———. 1962b. “Forensic Neuropathology V: Complications of Cranial Fractures,” JFS 7, pp. 323–345. Crane, E. 1983. The Archaeology of Beekeeping, Ithaca. Crawford, S. 2000. “Children, Grave Goods, and Social Status in Early Anglo-Saxon England,” in Children and Material Culture, ed. J. S. Derevenski, London, pp. 169–179. Croke, B. 2005. “Procopius’ Secret History: Rethinking the Date,” GRBS 45, pp. 405–431. Crow, J. G. 1995. “The Long Walls of Thrace,” in Constantinople and its Hinterland, ed. C. Mango and G. Dagron, Aldershot, pp. 109–129. Csallány, D. 1956. “Pamiatniki vizantijskojo metalloobrabatyvajuščego iskusstva II,” ActaArchHung 4, pp. 261– 291. Curta, F. 1996. “Invasion or Inflation? Sixth- to SeventhCentury Byzantine Coin Hoards in Eastern and Southeastern Europe,” AIIN 43, pp. 65–224. ———. 2001a. “Peasants as ‘Makeshift Soldiers for the Occasion’: Sixth-Century Settlement Patterns in the Balkans,” in Urban Centers and Rural Contexts in Late Antiquity, ed. T. S. Burns and J. W. Eadie, East Lansing, Mich., pp. 199–217. ———. 2001b. The Making of the Slavs: The History and Archaeology of the Lower Danube Region, ca. 500–700, Cambridge. ———. 2005. “Female Dress and ‘Slavic’ Bow Fibulae in Greece,” Hesperia 74, pp. 101–146. Cutress, T. W., and G. W. Suckling. 1982. “The Assessment of Non-Carious Defects of Enamel,” International Dental Journal 32, pp. 117–122. Cybulski, J. S. 1977. “Cribra Orbitalia, a Possible Sign of Anemia in Early Historic Native Populations of the British Columbia Coast,” AJPA 47, pp. 31–40. Dagron, G. 1977. “Le christianisme dans la ville byzantine,” DOP 31, pp. 1–25. ———. 1984. “Les villes dans l’Illyricum protobyzantin,” in Villes et peuplement dans l’Illyricum protobyzantin (CÉFR 77), Rome, pp. 1–20. ———. 2002. “The Urban Economy, Seventh–Twelfth Centuries,” in EHB 2, pp. 393–461. Daim, F. 1998. “Archaeology, Ethnicity and the Structures of Identification: The Example of the Avars, Carantanians and Moravians in the Eighth Century,” in Strategies of Distinction: The Construction of Ethnic Communities, 300–800, ed. W. Pohl and H. Reimitz, Leiden, pp. 71– 93.

Danforth, L. M., and A. Tsiaras. 1982. The Death Rituals of Rural Greece, Princeton. Dantis, S. P. 1983. Ἀπειλητικαὶ ἐκφράσεις εἰς τὰς ἑλληνικὰς ἐπιτυμβίους παλαιοχριστιανικὰς ἐπιγραφάς, Athens. Daux, G. 1968. “Chronique des fouilles et découvertes archéologiques en Grèce en 1967: Isthmia, fouilles américaines,” BCH 92, pp. 773–784. Davidson, G. 1937. “The Avar Invasion of Corinth,” Hesperia 6, pp. 227–240. Davidson Weinberg, G. 1974. “A Wandering Soldier’s Grave at Corinth,” Hesperia 43, pp. 512–521. Deckers, J. G. 1993. “Das Hypogäum beim Silivri-Kapı in Istanbul,” JAC 36, pp. 140–163. de Grood, M. P. A. M. 1975. “Skull Fractures,” in Injuries of the Brain and Skull 1 (Handbook of Clinical Neurology 23), ed. P. J. Vinken and G. W. Bruyn, New York, pp. 387– 402. Deïlaki, E. 1977. “ Ἀρχαιότητες καὶ μνημεῖα ἈργολίδαςΚορινθίας 1971–1972: Ἄργος,” ArchDelt 28, B΄1 (1973), pp. 80–122. Dekoulakou, I. 1979. “ Ἀρχαιότητες καὶ μνημεῖα Ἀχαΐας 1974: Ἀνασκαφικαὶ ἐργασίαι. Πάτραι,” ArchDelt 29, B΄2 (1973–1974), pp. 382–397. Demakopoulou, K., N. Divari-Valakou, P. Åström, and G. Walberg. 1997–1998. “Excavations in Midea, 1995– 1996,” OpAth 22–23, pp. 57–90. DeMaris, R. E. 1995. “Corinthian Religion and Baptism for the Dead (1 Corinthians 15:29): Insights from Archaeology and Anthropology,” JBL 114, pp. 661– 682. Dengate, J. A. 1981. “Coin Hoards from the Gymnasium Area at Corinth,” Hesperia 50, pp. 147–188. Dennis, G. T. 1993. “Popular Religious Attitudes and Practices in Byzantium,” Proche-Orient chrétien 43, pp. 273–294. ———. 2001. “Death in Byzantium,” DOP 66, pp. 1–7. de Pina-Cabral, J. 1992. “The Gods of the Gentiles are Demons: The Problem of Pagan Survivals in European Culture,” in Other Histories, ed. K. Hastrup, London, pp. 45–61. Déroche, V. 1989. “Delphes: La christianisation d’un sanctuaire païen,” in Actes du XIe Congrès international d’archéologie chrètienne 3 (CÉFR 123), Rome, pp. 2713–2726. de Waele, F. J. 1933. “The Sanctuary of Asklepios and Hygieia at Corinth,” AJA 37, pp. 417–451. ———. 1935. “The Fountain of Lerna and the Early Christian Cemetery at Corinth,” AJA 39, pp. 352–359. Diacopoulos, L. 2004. “The Archaeology of Modern Greece,” in Mediterranean Archaeological Landscapes: Current Issues, ed. E. F. Athanassopoulos and L. Wandsnider, Philadelphia, pp. 183–198. Dias, G., and N. Tayles. 1997. “‘Abscess Cavity’—A Misnomer,” IJO 7, pp. 548–554. Díaz-Andreu, M. 2005. “Gender Identity,” in The Archaeology of Identity: Approaches to Gender, Age, Status, Ethnicity and Religion, ed. M. Díaz-Andreu, S. Lucy, S. Babić, and D. N. Edwards, London, pp. 13–42. Dieck, A. 1965. Die europäischen Moorleichenfunde (Hominidenmoorfunde 1), Neumünster. Dillon, B. D., and J. W. Verano. 1985. “Burial Illustration,” in The Student’s Guide to Archeological Illustrating (Archaeological Research Tools 1), 2nd rev. ed., ed. B. D. Dillon, Los Angeles, pp. 143–154. Dirkmaat, D. C., and J. M. Adovasio. 1997. “The Role of Archaeology in the Recovery and Interpretation of Human Remains from an Outdoor Forensic Setting,” in Haglund and Sorg 1997a, pp. 39–64.

BIBLIOGRAPHY AND ABBREVIATIONS Di Vita, A. 1988a. “Settore L,” in Gortina I, pp. 69–141. ———. 1988b. “Monete,” in Gortina I, pp. 153–163. Dixon, D. S. 1982. “Keyhole Lesions in Gunshot Wounds of the Skull and Direction of Fire,” JFS 27, pp. 555– 566. Dixon, S. 1992. The Roman Family, Baltimore. DOC I = A. R. Bellinger, Catalogue of the Byzantine Coins in the Dumbarton Oaks Collection and in the Whittemore Collection I: Anastasius I to Maurice, 491–602, Washington, D.C. 1966. Dodwell, E. 1819. A Classical and Topographical Tour through Greece, during the Years 1801, 1805 and 1806, 2 vols., London. Dorbarakis, P. C. 1986. “Τοπωνυμικὸν τῆς περιοχῆς Εὐρωστίνης τῆς Κορινθίας,” Platon 38, pp. 144–158. Dorrell, P. G. 1994. Photography in Archaeology and Conservation, 2nd ed., Cambridge. Douri, A. I. 1972. “ Ὀδοντολογικὰ γνωρίσματα κρανίων ἀπὸ τὸν Ἅγιο Ἀχίλλειο τῆς Πρέσπας,” ΕΕΠΘ 5 (1971–1972), pp. 455–461. Dreier, F. G. 1992. “The Paleopathology of a Finger Dislocation,” IJO 2, pp. 31–35. Drosinou, P. 1997. “ΚΕ´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Ανασκαφικές εργασίες. Νομός Χανίων. Κίσαμος,” ArchDelt 47, B΄2 (1992), pp. 579–581. Drosoyianni, P. A. 1972. “Βυζαντινὰ καὶ μεσαιωνικὰ μνημεῖα Πελοποννήσου. Α´. Ἀνασκαφαί. Ἀρχαῖον Λέχαιον,” ArchDelt 25, B΄1 (1970), pp. 206–208.

du Boulay, J. 1974. Portrait of a Greek Mountain Village, Oxford. Duncan, W. N. 2005. “Understanding Veneration and Violation in the Archaeological Record,” in Rakita et al. 2005, pp. 207–227. Dunn, A. 2001. Rev. of Avramea 1997, in Byzantine and Modern Greek Studies 25, pp. 232–238. ———. 2004. “Continuity and Change in the Macedonian Countryside from Gallienus to Justinian,” in Bowden, Lavan, and Machado 2004, pp. 535–586. Dupras, T. L., H. P. Schwarcz, and S. I. Fairgrieve. 2001. “Infant Feeding and Weaning Practices in Roman Egypt,” AJPA 115, pp. 204–212. Duyar, I., and Y. S. Erdal. 2003. “A New Approach for Calculating Dental Caries Frequency of Skeletal Remains,” Homo 54, pp. 57–70. Effros, B. 2003. “Monuments and Memory: Repossessing Ancient Remains in Early Medieval Gaul,” in Topographies of Power in the Early Middle Ages, ed. M. de Jong, F. Theuws, and C. van Rhijn, Leiden, pp. 93–118. EHB = A. E. Laiou, ed., The Economic History of Byzantium from the Seventh through the Fifteenth Century (Dumbarton Oaks Studies 39), 3 vols., Washington, D.C. 2002. Ekroth, G. 2002. The Sacrificial Rituals of Greek Hero-Cults in the Archaic to the Early Hellenistic Periods (Kernos Suppl. 12), Liège. Eliakis, C., E. Eliakis, and P. Iordanidis. 1966. “Détermination de la taille d’après la mensuration des os longs (recherches expérimentales),” Annales de médecine légale 46, pp. 403–421. El-Najjar, M. Y. 1977. “Maize, Malaria, and the Anemias in the Pre-Columbian New World,” YPA 20, pp. 329– 337. El-Najjar, M. Y., B. Lozoff, and D. J. Ryan. 1975. “The Paleoepidemiology of Porotic Hyperostosis in the American Southwest: Radiological and Ecological Considerations,” American Journal of Roentgenology, Radium Therapy and Nuclear Medicine 125, pp. 918–924. El-Najjar, M. Y. and A. Robertson. 1976. “Spongy Bones in Prehistoric America,” Science n.s. 193, pp. 141–143.

xxix

El-Najjar, M. Y., D. J. Ryan, C. G. Turner, II, and B. Lozoff. 1976. “The Etiology of Porotic Hyperostosis among the Prehistoric and Historic Anasazi Indians of the Southwestern U.S.,” AJPA 44, pp. 477–488. Emmanouelides, N. E. 1989. Το δίκαιο της ταφής στο Βυζάντιο (Forschungen zur byzantinischen Rechtsgeschichte 3), Athens. Engels, J. 1998. Funerum sepulcrorumque magnificentia: Begräbnis- und Grabluxusgesetze in der griechisch-römischen Welt mit einigen Ausblicken auf Einschränkungen des funeralen und sepulkralen Luxus im Mittelalter und in der Neuzeit (Hermes Einselschriften 78), Stuttgart. Erdal, Y. S. 2007. “Occlusal Grooves in Anterior Dentition among Kovuklukaya Inhabitants (Sinop, Northern Anatolia, 10th century AD),” IJO 18, pp. 152–166. Erdélyi, I. 1991. “Das slawische Urnengräberfeld von Olympia,” in Ex oriente lux: Mélanges offerts en hommage au professeur Jean Blankoff, à l’occasion de ses soixante ans 1: Histoire, archéologie, histoire de l’art, numismatique, Brussels, pp. 119–125. Eshed, V., B. Latimer, C. M. Greenwald, L. M. Jellema, B. M. Rothschild, S. Wish-Baratz, and I. Hershkovitz. 2002. “Button Osteoma: Its Etiology and Pathophysiology,” AJPA 118, pp. 217–230. Evangelatou-Notara, F. 1987–1988. “. . . καὶ τὰ πολλὰ τῆς Πελοποννήσου . . . σεισμοῦ γεγόνασιν παρανάλωμα . . . ,” in ΠΔΣΠΣ 3.2, pp. 427–448. Evershed, R. P., S. N. Dudd, V. R. Anderson-Stojanović, and E. R. Gebhard. 2000. “New Chemical Evidence for the Use of Combed Ware Pottery Vessels as Beehives in Ancient Greece,” JAS 30, pp. 1–12. Fahlström, G. 1983. The Glenohumeral Joint in Man: An Anatomic-Experimental and Archaeo-Osteological Study on Joint Function (Ossa 8 Suppl. 1), Solna. Fallat, L., D. J. Grimm, and J. A. Sarocco. 1998. “Sprained Ankle Syndrome: Prevalance and Analysis of 639 Acute Injuries,” Journal of Foot and Ankle Surgery 37, pp. 280– 285. Fazekas, I. G., and F. Kósa. 1978. Forensic Foetal Osteology, Budapest. Fedwick, P. J. 1976. “Death and Dying in Byzantine Liturgical Traditions,” Eastern Churches Review 8, pp. 152–161. Feissel, D. 1983. Recueil des inscriptions chrétiennes de Macédoine du IIIe au VIe siècle (BCH Suppl. 8), Paris. ———. 1988. “L’architecte Viktôrinos et les fortifications de Justinien dans les provinces balkaniques,” Bulletin de la Société nationale des antiquaires de France 1988, pp. 136– 146. ———. 2000. “Les édifices de Justinien au témoignage de Procope et de l’épigraphie,” Antiquité tardive 8, pp. 81–104. Feissel, D., and A. Philippidis-Braat. 1985. “Inventaires en vue d’un recueil des inscriptions historiques de Byzance. III. Inscriptions du Pélopponèse (à l’exception de Mistra),” TravMém 9, pp. 267–395. Ferrua, A. 1967. Rev. of Corinth VIII.3, in ByzZeit 60, pp. 368–370. Fildes, V. A. 1986. Breasts, Bottles, and Babies: A History of Infant Feeding, Edinburgh. Finley, J. H., Jr. 1932. “Corinth in the Middle Ages,” Speculum 7, pp. 477–499. Finnegan, M. 1978. “Non-Metric Variation of the Infracranial Skeleton,” JAnat 125, pp. 23–37. Finucane, R. C. 1981. “Sacred Corpse, Profane Carrion: Social Ideals and Death Rituals in the Later Middle Ages,” in Mirrors of Mortality: Studies in the Social History of Death, ed. J. Whaley, New York, pp. 40–60.

xxx

BIBLIOGRAPHY AND ABBREVIATIONS

Flatt, A. E., and V. E. Wood. 1975. “Rigid Digits or Symphalangism,” The Hand 7, pp. 197–214. Formicola, V. 1993. “Stature Reconstruction from Long Bones in Ancient Population Samples—An Approach to the Problem of Reliability,” AJPA 90, pp. 351–358. Fornaciari, G., F. Mallegni, D. Bertini, and V. Nuti. 1982. “Cribra Orbitalia and Elemental Bone Iron, in the Punics of Carthage,” Ossa 8, pp. 63–77. Fowden, G. 1988. “City and Mountain in Late Roman Attica,” JHS 108, pp. 48–59. ———. 1990. “The Athenian Agora and the Progress of Christianity,” JRA 3, pp. 494–501. ———. 1995. “Late Roman Achaea: Identity and Defense,” JRA 8, pp. 549–567. ———. 1998. “Polytheist Religion and Philosophy,” in The Late Empire, a.d. 337–425 (CAH 2 13), ed. A. Cameron and P. Garnsey, Cambridge, pp. 538–560. Fox, S. C. 1996. “Appendix: The Human Skeletal Remains from Alassa-Ayia Mavri, Cyprus,” in Excavations in the Kouris Valley II: The Basilica of Alassa, ed. P. Florentzos, Nicosia, pp. 39–64. ———. 2003. “Appendix 5: The Human Skeletal Remains, A Preliminary Report,” in Kalavasos-Kopetra, pp. 274– 277. ———. 2005. “Health in Hellenistic and Roman Times: The Case Studies of Paphos, Cyprus and Corinth, Greece,” in Health in Antiquity, ed. H. King, London, pp. 59–82. 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” (diss. Univ. of Arizona). Frantz, A. 1965. “From Paganism to Christianity in the Temples of Athens,” DOP 19, pp. 187–205. Frantz, A., H. A. Thompson, and J. Travlos. 1969. “The ‘Temple of Apollo Pythios’ on Sikinos,” AJA 73, pp. 397–422. Fraser, P. M. 1977. Rhodian Funerary Monuments, Oxford. Frend, W. H. C., and D. E. Johnston. 1962. “The Byzantine Basilica Church at Knossos (Knossos Survey 36),” BSA 57, pp. 186–238. Fully, G., and A. Dehouve. 1965. “Renseignements apportés dans la détermination de l’âge par l’étude anatomique et radiologique du sternum et des côtes,” Annales de médecine légale 45, pp. 469–474. Galloway, A. 1985. “Appendix: Report on the Human Skeletal Remains from the 1984 Excavations, Kourion City, Grid L 9,” RDAC 1985, pp. 302–306. ———. 1999. “Fracture Patterns and Skeletal Morphology: Introduction and the Skull,” in Broken Bones: Anthropological Analysis of Blunt Force Trauma, ed. A. Galloway, Springfield, Ill., pp. 63–80. Garland, A. N., and R. C. Janaway. 1989. “The Taphonomy of Inhumation Burial,” in Burial Archaeology: Current Research, Methods, and Developments (BAR-BS 211), ed. C. A. Roberts, F. Lee, and J. Bintliff, Oxford, pp. 15–37. Garland, L. 1988. “The Life and Ideology of Byzantine Women: A Further Note on Conventions of Behavior and Social Reality as Reflected in Eleventh and Twelfth Century Historical Sources,” Byzantion 58, pp. 361–393. Garland, R. 1995. The Eye of the Beholder: Deformity and Disability in the Graeco-Roman World, Ithaca. ———. 2001. The Greek Way of Death, 2nd ed., Ithaca. Garvie-Lok, S. J. 2001. “Loaves and Fishes: A Stable Isotope Reconstruction of Diet in Medieval Greece” (diss. Univ. of Calgary).

Geagan, D. J. 1989. “The Isthmian Dossier of P. Licinius Priscus Juventianus,” Hesperia 58, pp. 349–360. Geary, P. J. 1986. “Sacred Commodities: The Circulation of Medieval Relics,” in The Social Life of Things: Commodities in Cultural Perspective, ed. A. Appardurai, Cambridge, pp. 169–191. ———. 1994. “The Uses of Archaeological Sources for Religious and Cultural History,” in Living with the Dead in the Middle Ages, ed. P. J. Geary, Ithaca, pp. 30–45. Gebhard, E. R. 1973. The Theater at Isthmia, Chicago. ———. 1992. “The Early Stadium at Isthmia and the Founding of the Isthmian Games,” in Proceedings of an International Symposium on the Olympic Games, ed. W. D. E. Coulson and H. Kyrieleis, Athens, pp. 73–79. ———. 1993a. “The Isthmian Games and the Sanctuary of Poseidon in the Early Empire,” in Gregory 1993a, pp. 78–113. ———. 1993b. “The Evolution of a Pan-Hellenic Sanctuary: From Archaeology towards History at Isthmia,” in Greek Sanctuaries: New Approaches, ed. N. Marinatos and R. Hägg, London, pp. 154–177. ———. 2002. “Caves and Cults in the Isthmian Sanctuary,” in Peloponnesian Sanctuaries and Cults: Proceedings of the Ninth International Symposium at the Swedish Institute at Athens (SkrAth 4º, 44), ed. R. Hägg, Stockholm, pp. 59– 70. ———. 2005. “Rites for Melikertes-Palaimon in the Early Roman Corinthia,” in Schowalter and Friesen 2005, pp. 165–203. Gebhard, E. R., and M. W. Dickie. 1999. “MelikertesPalaimon, Hero of the Isthmian Games,” in Ancient Greek Hero Cult: Proceedings of the Fifth International Seminar on Ancient Greek Cult (SkrAth 8º, 16), ed. R. Hägg, Stockholm, pp. 159–165. Gebhard, E. R., F. P. Hemans, and J. W. Hayes. 1998. “University of Chicago Excavations at Isthmia, 1989: III,” Hesperia 67, pp. 405–456. Gejvall, N.-G., and F. Henschen. 1968. “Two Late Roman Skeletons with Malformation and Close Family Relationship from Ancient Corinth,” OpAth 8, pp. 179–201. Gerstel, S. E. J. 2005. “The Byzantine Village Church: Observations on its Location and on Agricultural Aspects of its Program,” in Lefort, Morrisson, and Sodini 2005, pp. 165–178. Gerstel, S. E. J., M. Munn, H. E. Grossman, E. Barnes, A. H. Rohn, and M. Kiel. 2003. “A Late Medieval Settlement at Panakton,” Hesperia 72, pp. 147–234. Giardina, A. 2000. “The Family in the Late Roman World,” in Cameron, Ward-Perkins, and Whitby 2000, pp. 392–415. Gilbert, R. I., and J. H. Mielke, eds. 1985. The Analysis of Prehistoric Diets, Orlando. Gilchrist, R. 2004. “Archaeology and the Life Course: A Time and Age for Gender,” in A Companion to Social Archaeology, ed. L. Meskell and R. W. Preucel, Oxford, pp. 142– 160. Glotz, G. 1906. Études sociales et juridiques sur l’antiquité grecque, Paris. Gojda, M. 1991. The Ancient Slavs: Settlement and Society, Edinburgh. Goldstein, L. G. 1981. “One-Dimensional Archaeology and Multi-Dimensional People: Spatial Organization and Mortuary Analysis,” in Chapman, Kinnes, and Randsborg 1981, pp. 53–69. Goodman, A. H. 1993. “On the Interpretation of Health from Skeletal Remains,” CurrAnthr 34, pp. 281–288.

BIBLIOGRAPHY AND ABBREVIATIONS Goodman, A. H., and G. J. Armelagos. 1985. “Factors Influencing the Distribution of Enamel Hypoplasia within the Human Permanent Dentition,” AJPA 68, pp. 479–493. Goodman, A. H., G. J. Armelagos, and J. C. Rose. 1980. “Enamel Hypoplasias as Indicators of Stress in Three Prehistoric Populations from Illinois,” HB 52, pp. 515–528. Goodman, A. H., and D. L. Martin. 2002. “Reconstructing Health Profiles from Skeletal Remains,” in Steckel and Rose 2002a, pp. 11–60. Goodman, A. H., and J. C. Rose. 1990. “Assessment of Systemic Physiological Perturbations from Dental Enamel Hypoplasias and Associated Histological Structures,” YPA 33, pp. 59–110. ———. 1991. “Dental Enamel Hypoplasias as Indicators of Nutritional Stress,” in Kelley and Larsen 1991, pp. 279–293. Gordon, C. C., and J. E. Buikstra. 1981. “Soil, pH, Bone Preservation, and Sampling Bias at Mortuary Sites,” AmerAnt 46, pp. 566–571. Gortina I = A. Di Vita, ed., Gortina I (Monografie della Scuola archeologica de Atene e delle missioni Italiane in Oriente 3), Rome 1988. Gounaris, G. G. 1989. “L’archéologie chrétienne en Grèce de 1974 à 1985,” in Actes du XIe Congrès international d’archéologie chrétienne 3 (CÉFR 123), Rome, pp. 2687–2711. ———. 1990. Τὸ βαλανεῖο καὶ τὰ βόρεια προσκτίσματα τοῦ Ὀκταγώνου τῶν Φιλίππων (ΒΑΑΕ 112), Athens. Grauer, A. L., and C. A. Roberts. 1996. “Paleoepidemiology, Healing, and Possible Treatment of Trauma in the Medieval Cemetery Population of St. Helen-on-the-Walls, York,” AJPA 100, pp. 531–544. Gregory, T. E. 1979. “The Late Roman Wall at Corinth,” Hesperia 48, pp. 264–280. ———. 1985. “An Early Byzantine Complex at Akra Sophia near Corinth,” Hesperia 54, pp. 411–428. ———. 1986. “The Survival of Paganism in Christian Greece: A Critical Essay,” AJP 107, pp. 229–242. ———. 1989. “Late Byzantine Pottery from Isthmia: New Evidence from the Korinthia,” in Recherches sur la céramique byzantine (BCH Suppl. 18), ed. V. Déroche and J.-M. Spieser, Paris, pp. 201–208. ———. 1992. “Kastro and Diateichisma as Responses to Early Byzantine Frontier Collapse,” Byzantion 62, pp. 235– 253. ———, ed. 1993a. The Corinthia in the Roman Period (JRA Suppl. 8), Ann Arbor. ———. 1993b. “An Early Byzantine (Dark Age) Settlement at Isthmia: Preliminary Report,” in Gregory 1993a, pp. 149–160. ———. 1993c. “Local and Imported Medieval Pottery from Isthmia,” in La ceramica nel mondo bizantino tra XI e XV secolo e i suoi rapporti con l’Italia, ed. S. Gelichi, Florence, pp. 283–306. ———. 1995. “The Roman Bath at Isthmia: Preliminary Report 1972–1992,” Hesperia 64, pp. 279–313. ———. 1996. “Archaeology and the Byzantine Dark Ages: Problems and Prospects,” in Acts of the XVIIIth International Congress of Byzantine Studies, Selected Papers 2: History, Archaeology, Religion, Theology, ed. I. Ševčenko and G. G. Litavrin, Shephardstown, W.Va., pp. 217– 224. ———. 2000. “Procopius on Greece,” Antiquité tardive 8, pp. 105–114.

xxxi

———. 2007. “Contrasting Impressions of Land Use in Early Modern Greece: The Eastern Corinthia and Kythera,” in Between Venice and Istanbul: Colonial Landscapes in Early Modern Greece (Hesperia Suppl. 40), ed. S. Davies and J. L. Davis, Princeton, pp. 173–198. Gregory, T. E., and P. N. Kardulias. 1989. “The 1989 Season at Isthmia,” OWAN 13.3, pp. 14–17. ———. 1990. “Geophysical and Surface Surveys in the Byzantine Fortress at Isthmia, 1985–1986,” Hesperia 59, pp. 467–511. Gregory, T. E., P. N. Kardulias, and J. L. Rife. 1997. “The Ohio State University Excavations at Isthmia: The 1995–1996 Field Seasons,” OWAN 20.2, pp 1–5. Gregory, T. E., and H. Mills. 1984. “The Roman Arch at Isthmia,” Hesperia 53, pp. 407–445. Grierson, P. 1986. “Iconografia, circolazione monetaria, e tesaurizzazione,” in La cultura bizantina, oggetti et messaggio: Moneta ed economia, Rome, pp. 29–57. Griffin, C. J., R. Powers, and R. Kruszynski. 1979. “The Incidence of Osteo-Arthritis of the Temporomandibular Joint in Various Cultures,” Australian Dental Journal 24, pp. 94–106. Grinsell, L. V. 1961. “The Breaking of Objects as a Funerary Rite,” Folklore 72, pp. 475–491. Gritsopoulos, T. A. 1973. Ἐκκλησιαστικὴ ἱστορία καὶ χριστιανικὰ μνημεῖα Κορινθίας 1: Ἱστορία, Athens. Grmek, M. 1989. Diseases in the Ancient Greek World, trans. M. Muellner and L. Muellner, Baltimore. Guy, H., C. Masset, and C. A. Baud. 1997. “Infant Taphonomy,” IJO 7, pp. 221–229. Hadjimarkos, D. M., and C. W. Bonhorst. 1962. “Flouride and Selenium Levels in Contemporary and Ancient Greek Teeth in Relation to Dental Caries,” Nature 193, pp. 177–178. Haglund, W. D., and M. H. Sorg, eds. 1997a. Forensic Taphonomy: The Postmortem Fate of Human Remains, New York. ———. 1997b. “Method and Theory of Forensic Taphonomy Research,” in Haglund and Sorg 1997a, pp. 13–26. Haldon, J. F. 1997. Byzantium in the Seventh Century: The Transformation of a Culture, rev. ed., Cambridge. ———. 2002. “Some Aspects of Early Byzantine Arms and Armour,” in A Companion to Medieval Arms and Armour, ed. D. Nicolle, Woodbridge (U.K.), pp. 65–79. Halffman, C. M., G. R. Scott, and P. O. Pedersen. 1992. “Palatine Torus in the Greenlandic Norse,” AJPA 88, pp. 145–161. “Halieis V” = W. W. Rudolph, “Excavations at Porto Cheli and Vicinity, Preliminary Report V: The Early Byzantine Remains,” Hesperia 48 (1979), pp. 294–320. Hamerman, D. 1989. “The Biology of Osteoarthritis,” New England Journal of Medicine 320, pp. 1322–1330. Hanson, R. P. C. 1977–1978. “The Transformation of Pagan Temples into Churches in the Early Christian Centuries,” JSS 23, pp. 257–267. Hapiot, L. 2003. “Malaria and Porotic Hyperostosis in the Aegean World from the Paleolithic to the Geometrical Period,” Paleopathology Newsletter 122, pp. 11–14. Härke, H. 1997. “The Nature of Burial Data,” in Burial and Society: The Chronological and Social Analysis of Archaeological Burial Data, ed. C. Kjeld Jensen and K. Høilund Nielsen, Aarhus, pp. 19–27. ———. 2001. “Cemeteries as Places of Power,” in Topographies of Power in the Early Middle Ages, ed. M. de Jong, F. Theuws, and C. van Rhijn, Leiden, pp. 9–30. Harl, K. W. 1990. “Sacrifice and Pagan Belief in Fifth- and Sixth-Century Byzantium,” PastPres 128, pp. 7–27.

xxxii

BIBLIOGRAPHY AND ABBREVIATIONS

———. 2001. “From Pagan to Christian in Cities of Roman Anatolia during the Fourth and Fifth Centuries,” in Urban Centers and Rural Contexts in Late Antiquity, ed. T. S. Burns and J. W. Eadie, East Lansing, Mich., pp. 301–322. Harper, N. K., and S. C. Fox. 2008. “Recent Research in Cypriot Bioarchaeology,” Bioarchaeology of the Near East 2, pp. 1–38. Harrington, K. D. 1979. “Degenerative Arthritis of the Ankle Secondary to Long-Standing Lateral Ligament Instability,” JBJS (American Volume) 61, pp. 354–361. Hather, J. G., L. Peña-Chocarro, and E. J. Sidell. 1992. “Turnip Remains from Byzantine Sparta,” Economic Botany 46, pp. 395–400. Hattersley-Smith, K. M. 1996. Byzantine Public Architecture between the Fourth and Early Eleventh Centuries a.d., with Special Reference to the Towns of Byzantine Macedonia, Thessaloniki. Haugh, A. J. 1997. “Pathology of Osteoarthritis,” in Arthritis and Allied Conditions: A Textbook of Rheumatology 2, 13th ed., ed. W. J. Koopman, Baltimore, pp. 1945–1968. Hauser, G., and G. F. De Stefano. 1989. Epigenetic Variants of the Human Skull, Stuttgart. Hayes, J. W. 1971. “A New Type of Early Christian Ampulla,” BSA 66, pp. 243–248. ———. 1975. Roman and Pre-Roman Glass in the Royal Ontario Museum, Toronto. ———. 1993. “Provisional Thoughts on Roman Pottery from the Sanctuaries at Isthmia,” in Gregory 1993a, pp. 113–114. Henderson, J. 1987. “Factors Determining the State of Preservation of Human Remains,” in Boddington, Garland, and Janaway 1987, pp. 43–54. Hengen, O. P. 1971. “Cribra Orbitalia: Pathogenesis and Probable Etiology,” Homo 22, pp. 57–75. Herrin, J. 1983. “In Search of Byzantine Women: Three Avenues of Approach,” in Images of Women in Antiquity, ed. A. Cameron and A. Kuhrt, Detroit, pp. 167–189. Herrmann, B. 1978. “Beschreibung eines Skelettfundes von Kastanas, Nordgriechenland,” Offa 35, pp. 138– 143. Hershkovitz, I., B. Ring, M. Speirs, E. Galili, M. Kislev, G. Edelson, and A. Hershkovitz. 1991. “Possible Congenital Hemolytic Anemia in Prehistoric Coastal Inhabitants of Israel,” AJPA 85, pp. 7–13. Hertzberg, H. T. E., E. Churchill, S. W. Dupertuis, R. M. White, and A. Damon. 1963. Anthropometric Survey of Turkey, Greece, and Italy, New York. Hesse, J. R., and M. Naeije. 1994. “The Biomechanics of the Temporomandibular Joint,” in Temporomandibular Disorders, 2nd ed., ed. S. L. Kraus, New York, pp. 41– 69. Hildebolt, C. F., and S. Molnar. 1991. “Measurement and Description of Periodontal Disease in Anthropological Studies,” in Kelley and Larsen 1991, pp. 225–240. Hill, B. H. 1966. The Temple of Zeus at Nemea, Princeton. Hill, M. C., and G. J. Armelagos. 1990. “Porotic Hyperostosis in Past and Present Perspective,” in A Life in Science: Papers in Honor of J. Lawrence Angel, ed. J. E. Buikstra, Kampsville, Ill., pp. 52–63. Hillson, S. 1996. Dental Anthropology, Cambridge. ———. 2000. “Dental Pathology,” in Katzenberg and Saunders 2000, pp. 249–286. Hjohlman, J. 2005. “Pyrgouthi in Late Antiquity,” in Pyrgouthi, pp. 127–266. Hodder, I. 1982. The Present Past: An Introduction to Anthropology for Archaeologists, New York.

Hodges, D. C. 1991. “Temporomandibular Joint Arthritis in a British Skeletal Population,” AJPA 85, pp. 367– 377. Hodges, R., and D. Whitehouse. 1983. Mohammed, Charlemagne and the Origins of Europe: Archaeology and the Pirenne Thesis, Ithaca. Hoffman-Axthelm, W. 1981. History of Dentistry, trans. H. M. Koehler, Chicago. Hohlfelder, R. L. 1970a. “A Small Deposit of Bronze Coins from Kenchreai,” Hesperia 39, pp. 68–72. ———. 1970b. “The ‘End’ of Classical Kenchreai: Some Numismatic Evidence,” AJA 74, p. 197 (abstract). ———. 1973. “A Sixth Century Hoard from Kenchreai,” Hesperia 42, pp. 89–101. ———. 1974. “A Conspectus of the Early Byzantine Coins in the Kenchreai Excavation Corpus,” Byzantine Studies 1, pp. 73–77. ———. 1975. “Barbarian Incursions into Central Greece in the Sixth Century of the Christian Era: More Evidence from Corinthia,” East European Quarterly 9, pp. 251– 258. ———. 1977. “Trans-Isthmian Walls in the Age of Justinian,” GRBS 18, pp. 173–179. Hopkins, K. 1983. Death and Renewal: Sociological Studies in Roman History 2, Cambridge. Horden, P. 2005. “Mediterranean Plague in the Age of Justinian,” in The Cambridge Companion to the Age of Justinian, ed. N. Maas, Cambridge, pp. 134–160. Horden, P., and N. Purcell. 2000. The Corrupting Sea: A Study of Mediterranean History, Oxford. Howell, C. L., and W. Blanc. 1995. A Practical Guide to Archaeological Photography (Archaeological Research Tools 6), 2nd ed., Los Angeles. Hubert, J., ed. 2000. Madness, Disability, and Social Exclusion. The Archaeology and Anthropology of “Difference,” London. Humphreys, S. C. 1993. “Death and Time,” in The Family, Women, and Death: Comparative Studies, 2nd ed., Ann Arbor, pp. 148–168. Hunt, E. D. [1981] 2001. “The Traffic in Relics: Some Late Roman Evidence,” in The Byzantine Saint, ed. S. Hackel, repr. Crestwood, N.Y., pp. 171–180. Hunter, J., and S. Dockrill. 1996. “Recovering Buried Remains,” in Hunter, Roberts, and Martin 1996, pp. 40– 57. Hunter, J., C. Roberts, and A. Martin, eds. 1996. Studies in Crime: An Introduction to Forensic Archaeology, London. Huxley, G. L. 1977. “The Second Dark Age of the Peloponnese,” Λακωνικαὶ Σπουδαί 3, pp. 84–110. Ibler, U. 1992. “Pannonische Gürtelschnallen des späten 6. und 7. Jahrhunderts,” ArhVest 43, pp. 135–148. IGCB I = N. A. Bees, ed., Inscriptiones graecae christianae veteres et byzantinae I: Peloponnesus, Isthmos, Korinthos, 1941, repr. Chicago 1978. ILGR = M. Šašel-Kos, ed., Inscriptiones Latinae in Graecia repertae: Additamenta ad CIL III, Faenza 1979. Imhoof-Blumer, F. W., and P. Gardner. 1964. Ancient Coins Illustrating Lost Masterpieces of Greek Art: A Numismatic Commentary on Pausanias, rev. A. N. Oikonomides, Chicago. Ingvarsson-Sundström, A. 2007. “Appendix B: The Human Remains,” in Midea II, pp. 469–478. İşcan, M. Y., and K. A. R. Kennedy, eds. 1989. Reconstruction of Life from the Skeleton, New York. İşcan, M. Y., and S. R. Loth. 1989. “Osteological Manifestations of Age in the Adult,” in İşcan and Kennedy 1989, pp. 23–40.

BIBLIOGRAPHY AND ABBREVIATIONS Isthmia = Isthmia: Results of Excavations Conducted under the Auspices of the American School of Classical Studies at Athens, Princeton I = O. Broneer, Temple of Poseidon, 1971. II = O. Broneer, Topography and Architecture, 1973. III = O. Broneer, Terracotta Lamps, 1977. IV = M. C. Sturgeon, Sculpture I: 1952–1967, 1987. V = T. E. Gregory, The Hexamilion and the Fortress, 1993. VI = S. Lattimore, Sculpture II: Marble Sculpture, 1967– 1980, 1996. VII = I. K. Raubitschek, The Metal Objects (1952–1989), 1998. VIII = C. Morgan, The Late Bronze Age Settlement and the Early Iron Age Sanctuary, 1999. Ivison, E. A. 1996. “Burial and Urbanism at Late Antique and Early Byzantine Corinth (c. a.d. 400–700),” in Towns in Transition: Urban Evolution in Late Antiquity and the Early Middle Ages, ed. N. Christie and S. T. Loseby, Brookfield, Vt., pp. 99–125. Jackson, R. 1988. Doctors and Diseases in the Roman Empire, Norman, Okla. Jacobsen, T. W., and T. Cullen. 1990. “The Work of J. L. Angel in the Eastern Mediterranean,” in A Life in Science: Papers in Honor of J. Lawrence Angel, ed. J. E. Buikstra, Kampsville, Ill., pp. 38–51. Jacoby, D. 1962. “Phénomènes de démographie rurale à Byzance aux XIIIe, XIVe et XVe siècles,” Études rurales 5–6, pp. 163–186. Jacques, F., and B. Bousquet. 1984. “Le raz de marée du 21 Juillet 365: Du cataclysme local à la catastrophe cosmique,” MÉFRA 94, pp. 423–461. Jaffe, H. L. 1958. Tumors and Tumorous Conditions of the Bones and Joints, Philadelphia. Janaway, R. C. 1996. “The Decay of Buried Human Remains and their Associated Materials,” in Hunter, Roberts, and Martin 1996, pp. 58–85. Jantz, R. L., D. R. Hunt, and L. Meadows. 1995. “The Measure and Mismeasure of the Tibia: Implications for Stature Estimation,” JFS 40, pp. 758–761. Jenkins, R. J. H., and A. H. S. Megaw. 1931–1932. “Researches at Isthmia,” BSA 32, pp. 68–89. Johansson, S. R., and S. Horowitz. 1986. “Estimating Mortality in Skeletal Populations: Influence of the Growth Rate on the Interpretation of Levels and Trends during the Transition to Agriculture,” AJPA 71, pp. 233–250. Johnson, J. S. 1994. “Consolidation of Archaeological Bone: A Conservation Perspective,” JFA 21, pp. 221–233. Johnson, M. J. 1997. “Pagan-Christian Burial Practices of the Fourth Century: Shared Tombs?” Journal of Early Christian Studies 5, pp. 37–59. Johnston, F. E., and L. O. Zimmer. 1989. “Assessment of Growth and Age in the Immature Skeleton,” in İşcan and Kennedy 1989, pp. 11–21. Jones, A. H. M. 1964. The Late Roman Empire 284–602: A Social, Economic, and Administrative Survey, 3 vols., Oxford. Jones, W. H. S. 1909. Malaria and Greek History, Manchester. Kajava, M. 2002. “When Did the Isthmian Games Return to the Isthmus? (Rereading Corinth 8.3.153),” CP 97, pp. 168–178. Kalavasos-Kopetra = M. Rautman, ed., A Cypriot Village of Late Antiquity: Kalavasos-Kopetra in the Vasilikos Valley (JRA Suppl. 52), Portsmouth, R.I. 2003. Kalavrezou, I., ed. 2003. Byzantine Women and their World, Cambridge.

xxxiii

Kalogeras, N. 2001. “What Do They Think about Children? Perceptions of Childhood in Early Byzantine Literature,” Byzantine and Modern Greek Studies 25, pp. 2–19. Kamp, K. A. 2001. “Where Have All the Children Gone?: The Archaeology of Childhood,” Journal of Archaeological Method and Theory 8, pp. 1–34. Kaplan, M. 1992. Les hommes et la terre à Byzance du VIe au XIe siècle: Propriété et exploitation du sol (Byzantina Sorbonensia 10), Paris. ———. 1999. “De la dépouille à la relique: Formation du culte des saints à Byzance de Ve au XIIe siècle,” in Bozóky and Helvétius 1999, pp. 19–38. Karagiorgou, O. 2001. “Demetrias and Thebes: The Fortunes and Misfortunes of Two Thessalian Port Cities in Late Antiquity,” in Lavan 2001b, pp. 183–215. Karayannopoulos, I. 1990. “Τὸ κιβώριο τῆς ἐκκλησίας τῆς Κορίνθου,” Λακωνικαὶ Σπουδαί 10, pp. 79–85. ———. 1996. “Zur Frage des Slavenansiedlung im griechischen Raum,” in Byzanz und seine Nachbarn (Südosteuropa-Jahrbuch 26), ed. A. Hohlweg, Munich, pp. 177–218. Kardulias, P. N. 1992. “Estimating Population at Ancient Military Sites: The Use of Historical and Contemporary Analogy,” AmerAnt 57, pp. 276–287. ———. 1993. “Anthropology and Population Estimates for the Byzantine Fortress at Isthmia,” in Gregory 1993a, pp. 139–148. ———. 1994. “Assessing Acculturation in the Byzantine Dark Age: An Archaeological Approach,” Abstracts of Papers—Byzantine Studies Conference 22, 1994, pp. 22–23. ———. 1995. “Architecture, Energy, and Social Evolution at Isthmia, Greece: Some Thoughts about Late Antiquity in the Korinthia,” JMA 8.2, pp. 33–59. ———. 2005. From Classical to Byzantine: Social Evolution in Late Antiquity and the Fortress at Isthmia, Greece (BAR-IS 1412), Oxford. Kardulias, P. N., and M. Shutes. 1992. “Regional Study in the Korinthia—The Korinthia Exploration Project 1991,” OWAN 15, pp. 20–26. Karivieri, A. 1995. “The Christianization of an Ancient Pilgrimage Site: A Case Study of the Athenian Asklepieion,” in Akten des XII. Internationalen Kongresses für christliche Archäologie 2 (JAC Suppl. 20.2), Münster, pp. 898–905. ———. 1996. The Athenian Lamp Industry in Late Antiquity (Papers and Monographs of the Finnish Institute at Athens 5), Helsinki. Karn, K. W., H. D. Shockett, W. C. Moffitt, and J. L. Gray. 1984. “Topographic Classification of Deformities of the Alveolar Process,” Journal of Periodontology 55, pp. 336–340. Karo, G. 1935. “Archäologische Funde vom Juli 1934 bis Juli 1935,” AA 50, pp. 159–244. Karpozelos, A. 1989. “Περὶ ἀποπάτων, βόθρων καὶ ὑπονόμων,” in Ἡ καθημερινὴ ζωὴ στὸ Βυζάντιο: Τομὲς καὶ συνέχειες στὴν ἑλληνιστικὴ καὶ ρωμαϊκὴ παράδοση. Πρακτικὰ τοῦ Α΄ Διεθνοῦς Συμποσίου, 15–17 Σεπτεμβρίου 1988, ed. C. G. Angelidi, Athens, pp. 335–352. Katz, D., and J. M. Suchey. 1986. “Age Determination of the Male Os pubis,” AJPA 69, pp. 427–435. Katzenberg, M. A., D. R. Herring, and S. R. Saunders. 1996. “Weaning and Infant Mortality: Evaluating the Skeletal Evidence,” YPA 39, pp. 177–199. Katzenberg, M. A., and S. R. Saunders, eds. 2000. Biological Anthropology of the Human Skeleton, New York. Kaza-Papayeoryiou, D. 1988. “Δ´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Νομός Αργολίδας. Ανασκαφικές εργασίες. Άργος,” ArchDelt 35, B΄1 (1980), pp. 116–119.

xxxiv

BIBLIOGRAPHY AND ABBREVIATIONS

Kazanski, M. 1999. Les slaves: Les origines (Ier–VIIe siècle après J.-C.), Paris. Kazhdan, A. 1982. “Small Social Groupings (Microstructures) in Byzantine Society,” in Akten des XVI. Internationalen Byzantinistenkongresses 2.2 (JÖB 32.2), Vienna, pp. 3– 11. Kazhdan, A., and A. Cutler. 1982. “Continuity and Discontinuity in Byzantine History,” Byzantion 52, pp. 429– 478. Kelley, M. A., and C. S. Larsen, eds. 1991. Advances in Dental Anthropology, New York. Kenchreai = Kenchreai, Eastern Port of Corinth: Results of Investigations by the University of Chicago and Indiana University for the American School of Classical Studies at Athens, Leiden I = R. L. Scranton, J. W. Shaw, and L. Ibrahim, Topography and Architecture, 1979. II = L. Ibrahim, R. L. Scranton, and R. Brill, The Panels of Opus Sectile in Glass, 1976. III = R. L. Hohlfelder, The Coins, 1978. IV = B. Adamscheck, The Pottery, 1979. V = H. Williams, The Lamps, 1981. Kendall, G. 1982. “A Study of Grave Orientation in Several Roman and Post-Roman Cemeteries from Southern Britain,” ArchJ 139, pp. 101–123. Kenna, M. E. 1976. “Houses, Fields, and Graves: Property and Ritual Obligation on a Greek Island,” Ethnology 15, pp. 21–34. ———. 1991. “The Power of the Dead: Changes in the Construction and Care of Graves and Family Vaults on a Small Greek Island,” Journal of Mediterranean Studies 1, pp. 101–119. Kent, S. 1986. “The Influence of Sedentism and Aggregation on Porotic Hyperostosis and Anaemia: A Case Study,” Man n.s. 21, pp. 605–636. Kerameikos XVI = B. Böttger, Die kaiserzeitlichen Lampen vom Kerameikos (Kerameikos XVI), Munich 2002. Keramopoullos, A. D. 1917. “Θηβαϊκά, Α´ μέρος—ἀνασκαφαὶ ἐν Θήβαις,” ArchDelt 3, pp. 2–252. ———. 1929. “Παλαιαὶ χριστιανικαὶ καὶ βυζαντιναὶ ταφαὶ ἐν Θήβαις,” ArchDelt 10 (1926), pp. 124–136. Kerr, N. W. 1988. “A Method of Assessing Periodontal Status in Archaeologically Derived Skeletal Material,” JPaleopath 2, pp. 67–78. ———. 1991. “Prevalence and Natural History of Periodontal Disease in Scotland—The Mediaeval Period (900–1600 a.d.),” Journal of Periodontal Research 26, pp. 346– 354. Kilian, K. 1980. “Zu einigen früh- und hochmittelalterlichen Funden aus der Burg von Tiryns,” ArchKorrBl 10, pp. 281–290. Kislinger, E. 1999. “Christians of the East: Rules and Realities of the Byzantine Diet,” in Food: A Culinary History from Antiquity to the Present, ed. J.-L. Flandrin, M. Montanari, and A. Sonnenfeld, New York, pp. 194–206. Koester, H. 1990. “Melikertes at Isthmia: A Roman Mystery Cult,” in Greeks, Romans, and Christians: Essays in Honor of Abraham J. Malherbe, ed. D. L. Balch, E. Ferguson, and W. A. Meeks, Minneapolis, pp. 355–366. Koilakou, C. 1997. “1η Εφορεία Βυζαντινών Αρχαιοτήτων. Ανασκαφικές εργασίες. Σπέτσες,” ArchDelt 47, B΄1 (1992), pp. 67–69. Kolias, T. G. 1988. Byzantinische Waffen: Ein Beitrag zur byzantinischen Waffenkunde von den Anfängen bis zur lateinischen Eroberung (Byzantina Vindobonensia 17), Vienna.

Konsolaki, E. 1987. “Β´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Ανασκαφικές εργασίες. Μέθανα,” ArchDelt 34, B΄1 (1979), pp. 71–72. ———. 1989. “ Β´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Ανασκαφικές εργασίες. Πόρος—Τροιζήνα— Μέθανα,” ArchDelt 37, B΄1 (1982), pp. 49–51. Konti, V. 1996–1997. “Βιοτεχνικὴ δραστηριότητα στὴν περιοχὴ τῶν Ἁλιέων Ἑρμιονίδας (6ος–7ος αἰ.),” in ΠΔΣΠΣ 5.2, pp. 335– 356. Koob, S. 1981. “Consolidation with Acrylic Colloidal Dispersions,” in Preprints of the Papers Presented at the Ninth Annual Meeting of the American Institute for the Conservation of Historic and Artistic Works, Washington, D.C., pp. 86– 94. ———. 1986. “The Use of Paraloid B-72 as an Adhesive: Its Application for Archaeological Ceramics and Other Materials,” Studies in Conservation 31, pp. 7–14. Köpstein, H. 1978. “Zu den Agrarverhältnissen,” in Byzanz im 7. Jahrhundert: Untersuchungen zur Herausbildung des Feudalismus (Berliner byzantinistische Arbeiten 48), ed. F. Winkelmann, H. Köpstein, H. Ditten, and I. Rochow, Berlin, pp. 1–72. Kordosis, M. S. 1981a. Συμβολὴ στὴν ἱστορία καὶ τοπογραφία τῆς περιοχῆς τῆς Κορίνθου στοὺς μέσους χρόνους , Athens. ———. 1981b. “ Ἡ σλαβικὴ ἐποίκηση στὴν Πελοπόννησο μὲ βάση τὰ σλαβικὰ τοπωνύμια,” Δωδώνη 10, pp. 381–444. ———. 1985–1986. “ Ἡ πρώτη μνεία τοῦ Ἑξαμιλίου καὶ ὁ δίολκος,” Ἱστορικογεωγραφικά 1, p. 199. ———. 1997. “Αρχαία και πρωτοβυζαντινή Τενέα,” Δωδώνη 26, pp. 465–580. Kosmetatou, E. 1996. “Midea in the Post-Bronze Age Period: A Preliminary Report,” OpAth 21, pp. 115–123. ———. 1998. “The Post-Bronze Age Period,” in G. Walberg, The Excavations on the Lower Terraces 1985–1991 (Midea I), Stockholm, pp. 168–174. ———. 2007. “The Post-Bronze Age Period,” in Midea II, pp. 189–194. Kosswig, K. 1986. “The Animal Bones and Molluscs,” in Excavations at Saraçhane in Istanbul I: The Excavations, Structures, Architectural Decoration, Small Finds, Coins, Bones, and Molluscs, ed. R. M. Harrison, Princeton, pp. 399– 401. Kötting, B. 1965. Der frühchristliche Reliquienkult und die Bestattung im Kirchengebäude, Cologne. Koukoules, P. I. 1940. “Βυζαντινῶν νεκρικὰ ἔθιμα,” EpetByz 16, pp. 3–80. ———. 1951. Βυζαντινῶν βίος καὶ πολιτισμός 4, Athens. Kounoupiotou, E. 1972. “Βυζαντινὰ καὶ μεσαιωνικὰ μνημεῖα Πελοποννήσου. 1. Ἀνασκαφαί,” ArchDelt 25, B΄1 (1970), pp. 208–210. Kountoura-Galaki, E., ed. 2001. Οι σκοτεινοί αιώνες του Βυζαντίου (Εθνικό Ίδρυμα Ερευνών, Ινστιτούτο Βυζαντινών Ερευνών, Διεθνή Συμπόσια 9), Athens. Kourinou-Pikoulas, E., Y. Pikoulas, and P. Faklaris. 1987– 1988. “Αγιονόρι, αρχαία χρόνια,” Ἱστορικογεωγραφικά 2, pp. 227–231. Kourkoutidou-Nikolaidou, E. 1997. “From the Elysian Fields to the Christian Paradise,” in The Transformation of the Roman World a.d. 400–900, ed. L. Webster and M. Crown, Berkeley, pp. 128–142. Krikos, A. 1935. “ Ἡ ἐξέλιξις τῆς τερηδόνος τῶν ὀδόντων ἐν Ἑλλάδι ἀπὸ τῶν ἀρχαιοτάτων μέχρι τῶν καθ᾿ ἡμᾶς χρόνων,” Πρακτικὰ τῆς Ἑλληνικῆς Ἀνθρωπολογικῆς Ἑταιρείας 12, pp. 3–18.

BIBLIOGRAPHY AND ABBREVIATIONS Kristalli-Votsi, K. 1984. “ Ἐφορεία Κλασικῶν Ἀρχαιοτήτων Ναυπλίου. Νομὸς Κορινθίας, Κεγχρεές,” ArchDelt 31, B΄1 (1976), pp. 64–65. ———. 1985. “Σικυών,” Ergon 1984 (1985), pp. 61–62. ———. 1988. “Ἀνασκαφὴ Σικυῶνος,” Prakt 1984, A΄, pp. 241– 242. Kritzas, C. B. 1976. “ Ἀρχαιότητες καὶ μνημεῖα Ἀργολιδοκορινθίας,” ArchDelt 27, B΄1 (1972), pp. 192–219. ———. 1979. “ Ἀνασκαφικαὶ ἐργασίαι. ῎Αργος,” ArchDelt 29, B΄2 (1973–1974), pp. 212–247. Krogman, W. M., and M. Y. İşcan. 1986. The Human Skeleton in Forensic Medicine, 2nd ed., Springfield, Ill. Künzl, E. 1983. Medizinische Instrumente aus Sepulkralfunden der römischen Kaiserzeit, Bonn. Kyriakakis, J. 1974. “Byzantine Burial Customs: Care of the Deceased from Death to the Prothesis,” GOTR 19, pp. 37–72. Laconia I = W. Cananagh, J. Crouwel, R. W. V. Catling, and G. Shipley, Methodology and Interpretation (BSA Suppl. 26), London 2002. Lai, P., and N. C. Lovell. 1992. “Skeletal Markers of Occupational Stress in the Fur Trade: A Case Study from Hudson’s Bay Company Fur Trade Post,” IJO 2, pp. 221–234. Laiou, A. E. 1985. “Observations on the Life and Ideology of Byzantine Women,” BF 9, pp. 59–102. ———. 2002. “Exchange and Trade, Seventh–Twelfth Centuries,” in EHB 2, pp. 697–770. ———. 2003. “Women in the History of Byzantium,” in Kalavrezou 2003, pp. 23–32. ———. 2005. “The Byzantine Village (5th–14th Century),” in Lefort, Morrisson, and Sodini 2005, pp. 31–54. Laiou-Thomadakis, A. E. 1977. Peasant Society in the Late Byzantine Empire: A Social and Demographic Study, Princeton. ———. 1981. “The Role of Women in Byzantine Society,” in Akten des XVI. Internationalen Byzantinisten-Kongresses 1.1 (JÖB 31.1), Vienna, pp. 233–260. ———. 1982. “Addendum to the Report on the Role of Women in Byzantine Society,” in Akten des XVI. Internationalen Byzantinisten-Kongresses 2.1 (JÖB 32.1), Vienna, pp. 198–203. Lallo, J. W., G. J. Armelagos, and R. P. Mensforth. 1977. “The Role of Diet, Disease and Physiology in the Origin of Porotic Hyperostosis,” HB 49, pp. 471–483. Lalonde, G. V. 2005. “Pagan Cult to Christian Ritual: The Case of Agia Marina Theseiou,” GRBS 45, pp. 91– 125. Lambert, P. M. 1997. “Patterns of Violence in Prehistoric Hunter-Gatherer Societies of Coastal Southern California,” in Troubled Times: Violence and Warfare in the Past, ed. D. L. Martin and D. W. Frayer, Amsterdam, pp. 77–109. Lambropoulou, A., E. Anagnostakis, V. Konti, and A. Panopoulou. 2001. “ Συμβολή στην ερμηνεία των αρχαιολογικών τεκμηρίων της Πελοποννήσου κατά τους ‘σκοτεινούς αιώνες,’” in Kountoura-Galaki 2001,

pp. 189–229. Lansjoen, O. 1998. “Diseases of the Dentition,” in Aufderheide and Rodríguez-Martín 1998, pp. 393– 412. Larsen, C. S. 1995. “Biological Changes in Human Populations with Agriculture,” Annual Review of Anthropology 24, pp. 185–213. ———. 1997. Bioarchaeology: Interpreting Behavior from the Human Skeleton, Cambridge.

xxxv

Larsen, C. S., and L. E. Sering. 2000. “Inferring IronDeficiency Anemia from Human Skeletal Remains: The Case of the Georgia Bight,” in Bioarchaeological Studies of Life in the Age of Agriculture: A View from the Southeast, ed. P. M. Lambert, Tuscaloosa, Ala., pp. 116–133. Laskaris, N. G. 2000. Monuments funéraires paléochrétiens (et byzantins) de Grèce, Athens. Lattimore, R. [1942] 1962. Themes in Greek and Latin Epitaphs, repr. Urbana. Lavan, L. 2001a. “A Brief Comment,” in Lavan 2001b, pp. 243–245. ———, ed. 2001b. Recent Research in Late Antique Urbanism (JRA Suppl. 42), Portsmouth, R.I. Lavigne, S. E., and J. E. Molto. 1995. “System of Measurement of the Severity of Periodontal Disease in Past Populations,” IJO 5, pp. 265–273. Lawson, J. C. [1910] 1964. Modern Greek Folklore and Ancient Greek Religion, repr. New York. Leake, W. M. [1830] 1968. Travels in the Morea, 3 vols., repr. Amsterdam. Lefort, J. 1993. “Rural Economy and Social Relations in the Countryside, ” DOP 47, pp. 101–113. ———. 2002. “The Rural Economy, Seventh–Twelfth Centuries,” in EHB 1, pp. 231–310. Lefort, J., C. Morrisson, and J.-P. Sodini, eds. 2005. Les villages dans l’Empire byzantin (ive-xve siècle), Paris Lerna II = J. L. Angel, The People (Lerna II), Princeton 1971. Leroy-Molinghen, A. 1967. “Un imbroglio suspect,” Byzantion 37, pp. 126–135. ———. 1968. “La mort d’Arius,” Byzantion 38, pp. 105– 111. Lesick, K. S. 1997. “Re-Engendering Gender: Some Theoretical and Methodological Concerns on a Burgeoning Archaeological Pursuit,” in Invisible Peoples and Processes: Writing Gender and Childhood into European Archaeology, ed. J. Moore and E. Scott, New York, pp. 31–41. Levers, R. G. H., and A. I. Darling. 1983. “Continuing Eruption of Some Adult Human Teeth of Ancient Populations,” ArchOB 28, pp. 401–408. Liebeschuetz, J. H. W. G. 2000. “Rubbish Disposal in Greek and Roman Cities,” in Sordes urbis: La eliminación de residuos en la cuidad romana. Actas de la reunión de Roma (Bibliotheca Italica 24), ed. X. Dupré Raventós and J.-A. Remolà, Rome, pp. 51–61. ———. 2001a. “The Uses and Abuses of the Concept of ‘Decline’ in Later Roman History, or, Was Gibbon Politically Correct?” in Lavan 2001b, pp. 233– 238. ———. 2001b. Decline and Fall of the Roman City, Oxford. Limberis, V. 2005. “Ecclesiastical Ambiguities: Corinth in the Fourth and Fifth Centuries,” in Schowalter and Friesen 2005, pp. 443–457. Lindenlauf, A. 2001. “Thrown Away Like Rubbish—Disposal of the Dead in Ancient Greece,” Papers from the Institute of Archaeology 12, pp. 86–99. ———. 2003. “The Sea as a Place of No Return in Ancient Greece,” WorldArch 35, pp. 416–433. Little, L. M., and J. K. Papadopoulos. 1998. “A Social Outcast in Early Iron Age Athens,” Hesperia 67, pp. 375– 404. Littleton, J., and B. Frohlich. 1993. “Fish-Eaters and Farmers: Dental Pathology in the Arabian Gulf,” AJPA 92, pp. 427–447.

xxxvi

BIBLIOGRAPHY AND ABBREVIATIONS

Liu, S. H., and W. J. Jason. 1994. “Lateral Ankle Sprains and Instability Problems,” Clinics in Sports Medicine 4, pp. 793–809. Lorantou-Papantoniou, R. 1999. Σολύγεια. Ἡ ἀνασκαφὴ τοῦ 1957–1958 (ΒΑΑΕ 194), Athens. Loukatos, D. S. 1940. “Λαογραφικαὶ περὶ τελευτῆς εἰδήσεις παρὰ Ἰωάννῃ Χρυσοστόμῳ,” Ἐπετηρὶς τοῦ Λαογραφικοῦ Ἀρχείου 2, pp. 30–117. ———. 1978. Εἰσαγωγὴ στὴν ἑλληνικὴ λαογραφία, 2nd ed., Athens. Loukopoulos, D., A. Hadji, M. Papadakis, P. Karababa, K. Sinopoulou, M. Boussiou, P. Kollia, M. Xenakis, A. Antsaklis, S. Mesoghitis, A. Loutradi, and P. Fessa. 1990. “Prenatal Diagnosis of Thalassemia and of the Sickle Cell Syndromes in Greece,” Annals of the New York Academy of Sciences 612, pp. 226–236. Lovejoy, C. O. 1985. “Dental Wear in the Libben Population: Its Functional Pattern and Role in the Determination of Adult Skeletal Age at Death,” AJPA 68, pp. 47–56. Lovejoy, C. O., and K. G. Heiple. 1981. “The Analysis of Fractures in Skeletal Populations with an Example from the Libben Site, Ottawa County, Ohio,” AJPA 55, pp. 529–541. Lovejoy, C. O., R. S. Meindl, R. P. Mensforth, and T. J. Barton. 1985b. “Multifactorial Determination of Skeletal Age at Death: A New Method with Blind Tests of its Accuracy,” AJPA 68, pp. 1–14. Lovejoy, C. O., R. S. Meindl, T. R. Pryzbeck, and R. P. Mensforth. 1985a. “Chronological Metamorphosis of the Auricular Surface of the Ilium: A New Method for the Determination of Age at Death,” AJPA 68, pp. 15– 28. Lovell, N. C. 1994. “Spinal Athritis and Physical Stress at Bronze Age Harappa,” AJPA 93, pp. 149–164. ———. 1997. “Trauma Analysis in Paleopathology,” YPA 40, pp. 139–170. ———. 2000. “Paleopathological Description and Diagnosis,” in Katzenberg and Saunders 2000, pp. 217–248. LRBC II = R. A. G. Carson and J. P. C. Kent, Bronze Roman Imperial Coinage of the Later Empire, a.d. 346–498 (Late Roman Bronze Coinage II), London 1978. LRP = J. W. Hayes, Late Roman Pottery, London 1972. Lucy, S. J. 1994. “Children in Early Medieval Cemeteries,” Archaeological Review from Cambridge 13.2, pp. 21–34. ———. 2005. “The Archaeology of Age,” in The Archaeology of Identity: Approaches to Gender, Age, Status, Ethnicity and Religion, ed. M. Díaz-Andreu, S. Lucy, S. Babić, and D. N. Edwards, London, pp. 43–66. Lukacs, J. R. 1989. “Dental Paleopathology: Methods for Reconstructing Dietary Patterns,” in İşcan and Kennedy 1989, pp. 261–286. ———. 1992. “Dental Paleopathology and Agricultural Intensification in South Asia: New Evidence from Bronze Age Harappa,” AJPA 87, pp. 133–150. Lyman, R. L. 1994. Vertebrate Taphonomy, Cambridge. MacDowall, D. 1965. “The Byzantine Coin Hoard Found at Isthmia,” Archaeology 18, pp. 264–267. MacIsaac, J. D. 1987. “Corinth: Coins, 1925–1926: The Theater District and the Roman Villa,” Hesperia 56, pp. 97–157. ———. 1995. “Coins and the Field Archaeologist: Numismatic Finds as Artifacts,” ArchNews 20, pp. 21–25. MacLaughlin, S. M., and B. L. Watts. 1992. “ManubrioSternal Fusion—Radiographic Investigation,” Annals of Human Biology 19, p. 216.

MacMullen, R. 1997. Christianity and Paganism in the Fourth to Eighth Centuries, New Haven. Mallegni, F. 1988. “Settore L: Analisi dei resti schletrici umani,” in Gortina I, pp. 339–401. Manchester, K. 1992. “The Palaeopathology of Urban Infections,” in Death in Towns: Urban Responses to the Dying and the Dead, 100–1600, ed. S. Bassett, London, pp. 8– 14. Mango, C. 1980. Byzantium: The Empire of New Rome, London. Mangwani, J., M. A. Hakmi, and T. W. D. Smith. 2001. “Chronic Lateral Ankle Instability: Review of Anatomy, Biomechanics, Pathology, Diagnosis and Treatment,” The Foot 11.2, pp. 76–84. Mann, R. W., W. M. Bass, and L. Meadows. 1990. “Time Since Death and the Decomposition of the Human Body: Variables and Observations in Case and Experimental Field Studies,” JFS 35, pp. 103–111. Manouvrier, L. P. 1892. “La détermination de la taille d’après les grands os des membres,” Mémoires de la Société d’anthropologie de Paris ser. 2, 4, pp. 347–402. Mant, A. K. 1987. “Knowledge Acquired from Post-War Exhumations,” in Boddington, Garland, and Janaway 1987, pp. 65–80. Manzi, G., L. Salvadei, A. Vienna, and P. Passarello. 1999. “Discontinuity of Life Conditions at the Transition from the Roman Imperial Age to the Early Middle Ages: An Example from Central Italy Evaluated by Pathological Dento-Alveolar Lesions,” AJHB 1, pp. 327–341. Maraval, P. 1985. Lieux saints et pèlerinages d’Orient: Histoire et géographie des origines à la conquête arabe, Paris. Marchetti, P., K. Kolokotsas, and C. Abadie-Reynal. 1995. Le nymphée de l’agora d’Argos: Fouille, étude architecturale et historique (Études péloponnésiennes 11), Athens. Marshall, T. K. 1976. “Changes after Death,” in Gradwohl’s Legal Medicine, 3rd ed., ed. F. E. Camps, Chicago, pp. 78–100. Martin, D. L., G. J. Armelagos, A. H. Goodman, and D. P. Van Gerven. 1984. “The Effects of Socioeconomic Change in Prehistoric Africa: Sudanese Nubia as a Case Study,” in Cohen and Armelagos 1984, pp. 193–214. Martin, R., and K. Saller. 1957. Lehrbuch der Anthropologie in systematische Darstellung, mit besonderer Berücksichtigung der anthropologischen Methoden, 2 vols., 4th ed., Stuttgart. Martin, R. E. 1999. Taphonomy: A Processual Approach, Cambridge. Marty, J. 1993. “Three Pottery Deposits and the History of Roman Isthmia,” in Gregory 1993a, pp. 115–129. Marty Peppers, J. 1978. “The Nymphs at Poseidon’s Sanctuary,” in Hommages à Maarten J. Vermaseren 2, ed. M. B. de Boer and T. A. Edridge, Leiden, pp. 931–937. ———. 1979. “Selected Roman Pottery, Isthmia Excavations, 1967–1972” (diss. Univ. of Pennsylvania). Mays, S. A. 1996. “Age-Dependent Cortical Bone Loss in a Medieval Population,” IJO 6, pp. 144–154. ———. 1998. The Archaeology of Human Bones, London. McCormick, W. F. 1980. “Mineralization of the Costal Cartilages as an Indicator of Age: Preliminary Observations,” JFS 25, pp. 736–741. McCormick, W. F., and J. H. Stewart. 1983. “Ossification Patterns of Costal Cartilage as an Indicator of Sex,” Archives of Pathology and Laborator y Medicine 107, pp. 206–210. ———. 1988. “Age Related Changes in the Human Plastron: A Roentgenographic and Morphologic Study,” JFS 33, pp. 100–120.

BIBLIOGRAPHY AND ABBREVIATIONS McCormick, W. F., J. H. Stewart, and L. A. Langford. 1985. “Sex Determination from Chest Plate Roentgenograms,” AJPA 68, pp. 173–195. McHugh, F. 1999. Theoretical and Quantitative Approaches to the Study of Mortuary Practice (BAR-BS 785), Oxford. Megas, G. A. 1940. “Ζητήματα ἑλληνικῆς λαογραφίας. 6. Τὰ κατὰ τὴν τελευτήν,” Ἐπετηρὶς τοῦ Λαογραφικοῦ Ἀρχείου 2, pp. 166–205. Megaw, A. H. S. 1968. “Archaeology in Greece, 1967–68: Isthmia,” AR 14, p. 7. Meinardus, O. 1970. “A Study of the Relics of Saints of the Greek Orthodox Church,” OC 54, pp. 130–278. Meindl, R. S., and C. O. Lovejoy. 1985. “Ectocranial Suture Closure: A Revised Method for the Determination of Skeletal Age at Death and Blind Tests of Its Accuracy,” AJPA 68, pp. 57–66. Meindl, R. S., C. O. Lovejoy, R. P. Mensforth, and L. D. Carlos. 1985a. “Accuracy and Direction of Error in Sexing of the Skeletons: Implications for Paleodemography,” AJPA 68, pp. 79–85. Meindl, R. S., C. O. Lovejoy, R. P. Mensforth, and R. A. Walker. 1985b. “A Revised Method of Age Determination Using the Os pubis, with a Review and Tests of Accuracy of Other Current Methods of Pubic Symphyseal Aging,” AJPA 68, pp. 29–45. Mensforth, R. P., and B. M. Latimer. 1989. “HamannTodd Collection Aging Studies: Osteoporosis Fracture Syndrome,” AJPA 80, pp. 461–479. Mensforth, R. P., and C. O. Lovejoy. 1985. “Anatomical, Physiological and Epidemiological Correlates of the Aging Process: A Confirmation of Multifactorial Age Determination in the Libben Skeletal Population,” AJPA 68, pp. 87–106. Mensforth, R. P., C. O. Lovejoy, J. W. Lallo, and G. J. Armelagos. 1978. “The Role of Constitutional Factors, Diet and Infectious Disease in the Etiology of Porotic Hyperostosis and Periosteal Reaction in Prehistoric Infants and Children,” Medical Anthropology 2, pp. 1– 58. Merbs, C. F. 1983. Patterns of Activity-Induced Pathology in a Canadian Inuit Population (Archaeological Survey of Canada Publications 119), Ottawa. ———. 1989a. “Trauma,” in İşcan and Kennedy 1989, pp. 161–189. ———. 1989b. “Spondylolysis: Its Nature and Anthropological Significance,” International Journal of Anthropology 4, pp. 163–169. Methana = C. Mee and H. Forbes, eds., A Rough and Rocky Place: The Landscape and Settlement History of the Methana Peninsula, Greece, Liverpool 1997. Metress, J. F., and T. Conway. 1975. “Standardized System for Recording Dental Caries in Prehistoric Skeletons,” Journal of Dental Research 54, p. 908. Meyer-Plath, B., and A. M. Schneider. 1943. Die Landmauer von Konstantinopel 2 (Denkmäler antiker Architektur 8), Berlin. Michaud, J.-P. 1970. “Chronique de fouilles et découvertes archéologiques en Grèce en 1968 et 1969: Isthme,” BCH 94, pp. 934–949. Midea II = G. Walberg, Midea: The Megaron Complex and Shrine Area, Excavations on the Lower Terraces 1994–1997 (Prehistory Monographs 20), Philadelphia 2007. Miles, A. E. W. 1962. “Assessment of the Age of a Population of Anglo-Saxons from their Dentitions,” Proceedings of the Royal Society of Medicine 55, pp. 881–886.

xxxvii

Miller, M. E., and J. R. Ada. 1992. “Injuries to the Shoulder Girdle,” in Skeletal Trauma: Fractures, Dislocations, Ligamentous Injuries, ed. B. D. Browner, J. B. Jupiter, A. M. Levine, and P. G. Trafton, Philadelphia, pp. 1291–1310. Miller, S. G. 1976. “Excavations at Nemea, 1975,” Hesperia 45, pp. 174–202. ———. 1977. “Excavations at Nemea, 1976,” Hesperia 46, pp. 1–26. ———. 1978. “Excavations at Nemea, 1977,” Hesperia 47, pp. 58–88. ———. 1979. “Excavations at Nemea, 1978,” Hesperia 48, pp. 73–103. ———. 1980. “Excavations at Nemea, 1979,” Hesperia 49, pp. 178–205. ———. 1981. “Excavations at Nemea, 1980,” Hesperia 50, pp. 45–67. ———. 1983. “Excavations at Nemea, 1982,” Hesperia 52, pp. 70–95. ———. 1986. “Poseidon at Nemea,” in Φίλια ἔπη εἰς Γεώργιον Ε. Μυλωνὰν διὰ τὰ 60 ἔτη του ἀνασκαφικοῦ του ἔργου 1 (BAAE 103), Athens, pp. 261–271. ———. 1988. “Excavations at Nemea, 1984–1986,” Hesperia 57, pp. 1–20. Milne, J. S. [1907] 1976. Surgical Instruments in Greek and Roman Times, repr. Chicago. Milner, G. R., and C. S. Larsen. 1991. “Teeth as Artifacts of Human Behavior: Intentional Mutilation and Accidental Modification,” in Kelley and Larsen 1991, pp. 357–378. Milner, G. R., J. W. Wood, and J. L. Bolsden. 2000. “Paleodemography,” in Katzenberg and Saunders 2000, pp. 467–497. Mitchell, S. 1993. Anatolia: Land, Men, and Gods in Asia Minor 2: The Rise of the Church, Oxford. Mittler, D. M., and D. P. Van Gerven. 1994. “Developmental, Diachronic, and Demographic Analysis of Cribra Orbitalia in Medieval Christian Populations of Kulubnarti,” AJPA 93, pp. 287–297. Mizoguchi, K. 1993. “Time in the Reproduction of Mortuary Practices,” WorldArch 25, pp. 223–235. Moffatt, A. 1986. “The Byzantine Child,” Social Research 53, pp. 705–723. Molleson, T. I. 1993. “The Human Remains,” in Poundbury 2: The Cemeteries, ed. D. E. Farwell and T. I. Molleson, Dorchester, pp. 142–214. ———. 1995. “The Importance of Porridge,” in Nature et culture: Colloque de Liège (Études et recherches archéologiques de l’Université de Liège 86), ed. M. Otte, Liège, pp. 479–486. Molleson, T. I., and D. Hodgson. 1993. “A Cart Driver from Ur,” Archaeozoologia 11, pp. 93–106. Monceaux, P. 1884. “Fouilles et recherches archéologiques aux sanctuaires des jeux isthmiques I–II,” GazArch 9, pp. 273–285, 354–363. ———. 1885. “Fouilles et recherches archéologiques aux sanctuaires des jeux isthmiques III,” GazArch 10, pp. 205–214, 402–412. Moore, W. J., and M. E. Corbett. 1971. “The Distribution of Dental Caries in Ancient British Populations. I: AngloSaxon Period,” CarRes 5, pp. 151–168. ———. 1973. “The Distribution of Dental Caries in Ancient British Populations. II: Iron Age, Romano-British, and Mediaeval Periods,” CarRes 7, pp. 139–153. ———. 1975. “The Distribution of Dental Caries in Ancient British Populations. III: The 17th Century,” CarRes 9, pp. 163–175.

xxxviii

BIBLIOGRAPHY AND ABBREVIATIONS

Moore-Jansen, P. M., S. D. Ousley, and R. L. Jantz. 1994. Data Collection Procedures for Forensic Skeletal Material, 3rd ed., Knoxville, Tenn. Moravcsik, G. 1931. “Il Caronte bizantino,” Studi bizantini e neoellenici 3, pp. 45–68. Moretti, L. 1953. Iscrizioni agonistiche greche, Rome. Morgan, C. 1936. “Excavations at Corinth, Autumn 1935– 1936,” AJA 40, pp. 466–484. ———. 1938. “Excavations at Corinth, Autumn 1938,” AJA 42, pp. 362–370. Morris, I. 1992. Death-Ritual and Social Structure in Classical Antiquity, Cambridge. Morrisson, C. 1986. “Byzance au VIIe siècle: Le témoignage de la numismatique,” in Byzantium: Tribute to Andreas N. Stratos 1, Athens, pp. 149–163. ———. 2001. “Survivance de l’économie monétaire à Byzance (viie–ixe siècle),” in Kountoura-Galaki 2001, pp. 377–397. Morrisson, C., and J.-P. Sodini. 2002. “The Sixth-Century Economy,” in EHB 1, pp. 171–220. Moutsopoulos, N. K. 1972. “ Ἀνασκαφὴ τῆς βασιλικῆς τοῦ Ἁγίου Ἀχιλλείου,” ΕΕΠΘ 5 (1971–1972), pp. 149–461. Moutzali, A. G. 1985. “Εφορεία Βυζαντινών Αρχαιοτήτων Σπάρτης. Νομός Αργολίδας. Ανασκαφικές εργασίες. Άργος,” ArchDelt 33, B΄1 (1978), pp. 105–107. ———. 1993. “ Από τον παγανισμό στον χριστιανισμό:

παλαιοχριστιανικές της Ελλάδος, που προήλθαν από μετατροπή αρχαίων ιερών ,” Αρχαιολογία 44, pp. 25–

34. ———. 2000. “Σλάβοι στη Βυζαντινή Πελοπόννησο,” Αρχαιολογία 74, pp. 64–72. Mundell Mango, M. 2000. “Building and Architecture,” in Cameron, Ward-Perkins, and Whitby 2000, pp. 918–971. Murphy-O’Connor, J. 1996. Paul: A Critical Life, Oxford. Musgrave, J. H. 1976. “Appendix A: The Human Remains,” in Catling and Smyth 1976, pp. 40–46. Mylona, D. 2005. “The Animal Bones from Pyrgouthi in the Berbati Valley,” in Pyrgouthi, pp. 301–308. Naji, S. 2005. “Death and Remembrance in Medieval France: A Case Study from the Augustinian Monastery of Saint-Jean-des-Vignes, Soissons,” in Rakita et al. 2005, pp. 173–189. Navani, S., J. R. Shah, and P. S. Levy. 1970. “Determination of Sex by Costal Cartilages,” American Journal of Roentgenology, Radium Therapy and Nuclear Medicine 108, pp. 771– 774. Nemea = Excavations at Nemea, Berkeley I = D. E. Birge, L. H. Kraynak, and S. G. Miller, Topographical and Architectural Studies: The Sacred Square, the Xenon, and the Bath, 1992. II = S. G. Miller, The Early Hellenistic Stadium, 2001. III = R. C. Knapp and J. D. MacIsaac, The Coins, 2005. Nemea Guide = S. G. Miller, Nemea: A Guide to the Site and Museum, 2nd ed., Athens 2004. Nestor, I. 1963. “La pénétration des Slaves dans la péninsule balkanique et la Grèce continentale: Considerations sur les recherches historiques et archéologiques I,” Revue des études sud-est européennes 1, pp. 41–67. Nettrour, L. F., E. L. Krufky, R. E. Mueller, and J. F. Raycroft. 1972. “Locked Scapula: Intrathoracic Dislocation of the Inferior Angle. A Case Report,” JBJS (British volume) 54, pp. 413–416. Nikolajević, I. 1984. “Nécropoles et tombes chrétiennes en Illyricum oriental,” in Actes du Xe Congrès international d’archèologie chrètienne à Thessalonique 1, Vatican City, pp. 519–535. Nilsson, M. 1908. “Das Ei im Totenkult der Alten,” ArchRW 11, pp. 530–546.

Novak, S. A. 2000. “Battle-Related Trauma,” in Blood Red Roses: The Archaeology of a Mass Grave from the Battle of Towton, a.d. 1461, ed. V. Fiorato, A. Boylston, and C. Knüsel, Oxford, pp. 90–102. Ogle, M. B. 1933. “The Sleep of Death,” MAAR 11, pp. 81– 117. Oikonomidou, M. 1992–1993. “ Ο μύθος του ΜελικέρτηΠαλαίμονα στα κορινθιακά νομίσματα,” in ΠΔΣΠΣ 4.2, pp. 135–140. Oikonomidou-Karamesini, M. 1991. “Πρώιμος βυζαντινός ‘θησαυρός’ χάλκινων νομισμάτων από την Κορινθία,” in ΑΡΜΟΣ. Τιμητικός τόμος στον καθηγητή Ν. Κ. Μουτσόπουλο 2, Thessaloniki, pp. 1289–1294. Oikonomou, A. 1988. “Lampes paléochrétiennes d’Argos,” BCH 112, pp. 481–502. ———. 1989. “ Συμβολὴ στὴν τοπογραφία τῆς περιοχῆς Ἄνω Ἐπιδαύρου στοὺς μέσους χρόνους ,” in Πρακτικὰ τοῦ Β´ Τοπικοῦ Συνεδρίου Ἀργολικῶν Σπουδῶν 1988 (Πελοποννησιακά Suppl. 14), Athens, pp. 303–312. Oikonomou-Laniado, A. 1998. “Les cimetières paléochrétiens d’Argos,” in Recherches franco-hélleniques 3: Argos et l’Argolide, topographie et urbanisme, ed. A. Pariente and G. Touchais, Paris, pp. 405–416. ———. 2003. Argos paléochrétienne: Contribution à l’étude du Péloponnèse byzantin (BAR-IS 1173), Oxford. Olajos, T. 1985. “Contribution à la chronologie des premières installations des Slaves dans l’Empire byzantin,” Byzantion 55, pp. 506–515. Olivier, G. 1960. Pratique anthropologique à l’usage des étudiants, Paris. Olynthus XI = D. M. Robinson, Necrolynthia: A Study in Greek Burial Customs and Anthropology, Baltimore 1942. Onasoglou, A. 1990. “Δ´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Ανασκαφικές εργασίες. Νομός Αργολίδας,” ArchDelt 40, B΄1 (1985), pp. 85–93. O’Roark, D., and M. J. Giesen. 1991. “Classical Archaeology and Biological Anthropology: A Cross-Disciplinary Study of Mortuary Data from the West Cemetery at Isthmia,” AJA 95, pp. 293–294 (abstract). Ortner, D. J., and W. G. Putschar. 1985. Identification of Pathological Conditions in Human Skeletal Remains (Smithsonian Contributions to Anthropology 28), rev. ed., Washington, D.C. O’Shea, J. 1981. “Social Configurations and the Archaeological Study of Mortuary Practices: A Case Study,” in Chapman, Kinnes, and Randsborg 1981, pp. 39–52. ———. 1984. Mortuary Variability: An Archaeological Investigation, New York. Packard, P. M. 1980. “A Monochrome Mosaic at Isthmia,” Hesperia 49, pp. 326–346. Pader, E.-J. 1980. “Material Symbolism and Social Relationships in Mortuary Studies,” in Anglo-Saxon Cemeteries 1979 (BAR-BS 82), ed. P. Rahtz, T. Dickinson, and L. Watts, Oxford, pp. 143–159. ———. 1982. Symbolism, Social Relations, and the Interpretation of Mortuary Remains (BAR-IS 130), Oxford. 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,” Journal of Comparative Human Biology 58, pp. 173– 190. Palkovich, A. M. 1987. “Endemic Disease Patterns in Paleopathology: Porotic Hyperostosis,” AJPA 74, pp. 527–537. Pallas, D. I. 1951. “Σαλαμινιακά (συνέχεια),” ArchEph 1950– 1951, pp. 163–181.

BIBLIOGRAPHY AND ABBREVIATIONS ———. 1955. “ Ἀρχαιολογικὰ τεκμήρια τῆς καθόδου τῶν βαρβάρων εἰς τὴν Ἑλλάδα,” Ἑλληνικά 14, pp. 87–105. ———. 1956. “Οἱ χριστιανικοὶ καμαρωτοὶ τάφοι,” ArchEph 1937, Γ΄, pp. 847–865. ———. 1957. “ Ἀνασκαφὴ ἐν Παλαιᾷ Κορίνθῳ,” Prakt 1954, pp. 210–218. ———. 1959a. “Δοκιμαστικὴ ἀνασκαφὴ ἐν Ἀρχαίᾳ Κορίνθῳ,” ArchEph 1956, Χρονικά, pp. 13–21. ———. 1959b. “Scoperte archeologische in Grecia negli anni 1956–1958,” RACrist 35, pp. 187–223. ———. 1961. “ Ἀνασκαφὴ βασιλικῆς ἐν Λεχαίῳ,” Prakt 1956, pp. 164–178. ———. 1964. “ Ἀνασκαφὴ τῆς βασιλικῆς τοῦ Λεχαίου,” Prakt 1961, pp. 137–154. ———. 1965. “ Ἀνασκαφὴ βασιλικῆς Λεχαίου,” Prakt 1959, pp. 126–140. ———. 1967. “ Ἀνασκαφικαὶ ἔρευναι ἐν Λεχαίῳ,” Prakt 1965, pp. 137–166. ———. 1972. “ Ἀνασκαφικὴ ἔρευνα εἰς τὴν βασιλικὴν τοῦ Κρανείου ἐν Κορίνθῳ,” Prakt 1970, pp. 98–117. ———. 1974. “Ἀνασκαφὴ τῆς βασιλικῆς τοῦ Κρανείου,” Prakt 1972, pp. 205–250. ———. 1975. “Investigations sur les monuments chrétiens de Grèce avant Constantin,” CahArch 24, pp. 1–19. ———. 1977. Les monuments paléochrétiens de Grèce découverts de 1959 à 1973 (Sussidi allo studio dell antichità cristiane 5), Rome. ———. 1978. “ Ἀνασκαφὴ τῆς βασιλικῆς τοῦ Κρανείου ἐν Κορίνθῳ,” Prakt 1976, A΄, pp. 163–195. ———. 1980. “ Ἀνασκαφὴ τῆς βασιλικῆς τοῦ Κρανείου ἐν Κορίνθῳ,” Prakt 1977, A΄, pp. 162–183. ———. 1981. “Données nouvelles sur quelques boucles et fibules considerés comme avares et slaves et sur Corinthe entre le VIe et le IXe s,” Byzantinobulgarica 7, pp. 295– 318. ———. 1989. “ Ἡ Ἀθήνα στὰ χρόνια τῆς μετάβασης ἀπὸ τὴν ἀρχαία λατρεία στὴ Χριστιανική,” Ἐπιστημονικὴ Ἐπετηρὶς τῆς Θεολογικῆς Σχολῆς ἐν Ἀθήναις 28, pp. 851–930. Pallas, D. I., S. Charitonides, and J. Venencie. 1959. “Inscriptions lyciennes trouvées à Solômos près de Corinthe,” BCH 83, pp. 496–508. Pallas, D. I., and S. P. Dantis. 1979. “ Ἐπιγραφὲς ἀπὸ τὴν Κόρινθο,” ArchEph 1977, pp. 61–83. Panayiotopoulos, V. 1985. Πληθυσμός και οικισμοί της Πελοποννήσου: 13ος–18ος αιώνας, Athens. Papacharalampous, G. C. 1965. Κυπριακὰ ἤθη καὶ ἔθιμα, Nicosia. ———. 1971. “ Ὁ Χάρωνος ὀβολός,” Κυπριακαὶ Σπουδαί 35, pp. 213–250. Papachristodoulou, I. 1968a. “ Ἀρχαιότητες καὶ μνημεῖα Ἀργολιδοκορινθίας. Ἀνασκαφαὶ Ἐθνικὴς Ὁδοῦ Κορίνθου-Πατρῶν,” ArchDelt 22, B΄1 (1967), pp. 166–

169. ———. 1968b. “ Ἀρχαιότητες καὶ μνημεῖα Ἀργολιδοκορινθίας. Νέαι Φυλακαὶ Τίρυνθος–Τίρυνς–Λέρνη (Μύλοι)–Κεφαλάρι Ἄργους–Τυχαῖα εὑρήματα ἐκ περιοχῆς ῎Αργους,” ArchDelt 22, B΄1 (1967), pp. 180–183. ———. 1970. “ Ἀρχαιότητες καὶ μνημεῖα Ἀργολιδοκορινθίας. Ἀνασκαφαί-εὑρήματα, Κορινθία 4–6,” ArchDelt 24, B΄1 (1969), pp. 103–104. Papadimitriou, I. K. 1948. “ Ὁ Ἰοβιανὸς τῆς βασιλικῆς τῆς Παλαιοπόλεως Κερκύρας,” ArchEph 1942–1944, Χρονικά, pp. 39–48. Papadopoulos, J. K. 2000. “Skeletons in Wells: Towards an Archaeology of Social Exclusion in the Ancient Greek World,” in Hubert 2000, pp. 96–118. Papalexandrou, A. 2003. “Memory Tattered and Torn: Spolia in the Heartland of Byzantine Hellenism,” in Archaeologies

xxxix

of Memory, ed. R. M. Van Dyke and S. E. Alcock, London, pp. 56–80. Papapostolou, I. A. 1979, “ Ἀρχαιότητες καὶ μνημεῖα Ἀχαΐας 1973. Ἀνασκαφικαὶ ἐργασίαι. Πάτραι,” ArchDelt 29, B΄2 (1973–1974), pp. 346–360. Parker, R. 1983. Miasma: Pollution and Purification in Early Greek Religion, Oxford. Parker Pearson, M. 1999. The Archaeology of Death and Burial, College Station, Tex. Patlagean, E. 1973. “L’enfant et son avenir dans la famille byzantine,” Annales de démographie historique 1973, pp. 85–93. ———. 1977. Pauvreté économique et pauvreté sociale à Byzance, 4 e–7 e siècles, Paris. ———. 1984. “Byzance et le blason pénal du corps,” in Sodalitas: Scritti in onore di Antonio Guarino 6, Naples, pp. 405–426. ———. 1987. “Byzantium in the Tenth and Eleventh Centuries,” in A History of Private Life 1: From Pagan Rome to Byzantium, ed. P. Veyne, trans. A. Goldhammer, Cambridge, Mass., pp. 551–641. Pearce, J. 2000. “Burial, Society, and Context in the Provincial Roman World,” in Burial, Society, and Context in the Roman World, ed. J. Pearce, M. Millett, and M. Struck, Oxford, pp. 1–12. ———. 2001. “Infants, Cemeteries, and Communities in the Roman Provinces,” in TRAC 2000: Proceedings of the Tenth Annual Theoretical Roman Archaeology Conference, ed. G. Davies, A. Gardner, and K. Lockyear, London, pp. 125–142. Peebles, C., and S. Kus. 1977. “Some Archaeological Correlates of Ranked Societies,” AmerAnt 42, pp. 421– 448. Pelekanides, S. 1977. “ Ἀνασκαφὴ Φιλίππων,” Prakt 1975, A΄, pp. 91–102. ———. 1978. “Kultprobleme im Apostel-Paulus-Octagon von Philippi in Zusammenhang mit einem älteren Heroenkult,” in Atti degli 9 o Congresso internazionale di archeologia cristiana 2, Vatican City, pp. 393–397. Pemberton, E. G. 1985. “Ten Hellenistic Graves in Ancient Corinth,” Hesperia 54, pp. 271–307. Pennas, V. 1996. “ Η ζωή στις βυζαντινές πόλεις της Πελοποννήσου: Η νομισματική μαρτυρία (8ος–12ος αι. μ.Χ.),” in Μνήμη Martin Jessop Price, ed. A. P. Tzamalis, Athens, pp. 195–264. Petrakakos, D. A. [1905] 1971. Die Toten im Recht nach der Lehre und den Normen des orthodoxen morgenländischen Kirchenrechts und der Gesetzgebung Griechenlands, repr. Aalen. Petrakos, P. 1977. “ Ἀρχαιότητες καὶ μνημεῖα Φθιώτιδος καὶ Φωκίδος. Κάλλιον,” ArchDelt 22, B΄2 (1972), pp. 375– 384. Petrakos, V. 1991. Rhamnous, Athens. Pettegrew, D. K. 2007. “The Busy Countryside of Late Roman Corinth: Interpreting Ceramic Data Produced by Regional Archaeological Surveys,” Hesperia 76, pp. 743–784. ———. 2008. “The End of Ancient Corinth? Views from the Landscape,” in Caraher, Hall, and Moore 2008, pp. 249–266. Pfister, F. [1909–1912] 1974. Der Reliquienkult im Altertum, 2 vols., repr. Berlin. Phenice, T. 1969. “A Newly Developed Method of Sexing in the Os pubis,” AJPA 30, pp. 297–301. Philippas, G. C. 1952. “Evidence of Function on Healthy Teeth: The Evidence of Ancient Athenian Remains,” Journal of the American Dental Association 45, pp. 443– 453.

xl

BIBLIOGRAPHY AND ABBREVIATIONS

Philpott, R. 1991. Burial Practices in Roman Britain: A Survey of Grave Treatment and Furnishing, a.d. 43–410 (BAR-BS 219), Oxford. Phougias, M. G. 1997. Ἱστορία τῆς ἀποστολικῆς ἐκκλησίας τῆς Κορίνθου ἀπ᾿ ἀρχῆς μέχρι σήμερον, 2nd ed., Athens. Piérart, M. 1998. “Panthéon et hellénisation dans la colonie romaine de Corinthe: La ‘redécouverte’ du culte de Palaimon à l’Isthme,” Kernos 11, pp. 85–109. Pinals, R. S. 1997. “Traumatic Arthritis and Allied Conditions,” in Arthritis and Allied Conditions: A Textbook of Rheumatology 2, 13th ed., ed. W. J. Koopman, Baltimore, pp. 1761–1777. Pindborg, J. J. 1982. “Aetiology of Developmental Enamel Defects Not Related to Fluorosis,” International Dental Journal 32, pp. 123–134. Pohl, W. 1988. Die Awaren: Ein Steppenvolk in Mitteleuropa 567–822 n. Chr., Munich. Polites, N. G. [1921] 1975. “Τὸ ἔθιμον τῆς θραύσεως ἀγγείων κατὰ τὴν κηδείαν,” in Λαογραφικὰ σύμμεικτα 2, 2nd ed., repr. Athens, pp. 321–340. ———. [1931] 1975. “Τὰ κατὰ τὴν τελευτήν,” in Λαογραφικὰ σύμμεικτα 3, 2nd ed., repr. Athens, pp. 323–362. Pollard, A. M. 1996. “Dating the Time of Death,” in Hunter, Roberts, and Martin 1996, pp. 139–155. Pomeroy, S. B. 1997. Families in Classical and Hellenistic Greece: Representations and Realities, Oxford. Popović, V. 1975. “Les témoins archéologiques des invasions avaro-slaves dans l’Illyricum byzantin,” MÉFRA 87, pp. 445–504. ———. 1978. “La descente des Kourtrigours, des Slaves et des Avars vers la mer Égée: Le témoignage de l’archéologie,” CRAI 1978, pp. 596–648. ———. 1980. “Aux origines de la slavisation des Balkans: La constitution des premières Sklavinies macédoniennes vers la fin du VIe siècle,” CRAI 1980, pp. 230–257. ———. 1981. “Note sur l’habitat paléoslave,” in P. Lemerle, Les plus anciens recueils des miracles de Saint Démétrius 2: Commentaire, Paris, pp. 235–241. Poulianos, A. N. 1972. “Βυζαντινὰ κρανία τῆς Πρέσπας,” ΕΕΠΘ 5 (1971–1972), pp. 439–450. ———. 1974. “Βυζαντινὰ κρανία τῆς Σάμου,” Ἄνθρωπος 1, pp. 61–68. Poush, J. R. 1991. “Distal Symphalangism: A Report of Two Families,” Journal of Heredity 82, pp. 233–238. Powell, M. L. 1985. “The Analysis of Dental Wear and Caries for Dietary Reconstruction,” in Gilbert and Mielke 1985, pp. 307–338. ———. 1991. “Ranked Status and Health in the Missisippian Chiefdom at Moundville,” in What Mean These Bones? Studies in Southeastern Bioarchaeology, ed. M. L. Powell, P. Bridges, and A. Mires, Tuscaloosa, Ala., pp. 22–51. Price, S. 1999. Religions of the Ancient Greeks, Cambridge. Proskynitopoulou, R. 1988. “Δ´ Εφορεία Προϊστορικών και Κλασικών Αρχαιοτήτων. Νομός Αργολίδας. Ανασκαφικές εργασίες,” ArchDelt 35, B΄1 (1980), p. 123. Protonotariou-Deïlaki, E. 1969. “ Ἀρχαιότητες καὶ μνημεῖα Ἀργολιδοκορινθίας. Κόρινθος,” ArchDelt 23, B΄1 (1968),

pp. 122–127. ———. 1970. “ Ἀρχαιότητες καὶ μνημεῖα Ἀργολιδοκορινθίας. Ἀνασκαφαί-εὑρήματα, Κορινθία 1–3,” ArchDelt 24, B΄1 (1969), pp. 102–103. ———. 1974. “ Ἀρχαιότητες καὶ μνημεῖα Ἀργολιδοκορινθίας. Κορινθία,” ArchDelt 26, B΄1 (1971), pp. 68–84. Pyrgouthi = J. Hjohlman, A. Penttinen, and B. Wells, eds., Pyrgouthi: A Rural Site in the Berbati Valley from the Early Iron Age to Late Antiquity. Excavations by the Swedish Institute at Athens, 1995 and 1997 (SkrAth 4º, 52), Stockholm 2005.

Rahtz, P. 1978. “Grave Orientation,” ArchJ 135, pp. 1–14. Rahtz, P., S. Hirst, and S. M. Wright. 2000. Cannington Cemetery (Britannia Monographs 17), London. Rakita, G. F. M., J. E. Buikstra, L. A. Beck, and S. R. Williams, eds. 2005. Interacting with the Dead: Perspectives on Mortuary Archaeology for the New Millenium, Gainsville, Fla. Ramoutsaki, I. A., H. Dimitriou, E. Galanakis, E. Stiakaki, and M. Kalmanti. 2001. “Aspects of Childhood Cancer during the Byzantine Period,” Pediatric Hematology and Oncology 18, pp. 161–166. Ramoutsaki, I. A., H. Dimitriou, and M. Kalmanti. 2002. “Management of Childhood Diseases in the Byzantine Period. I: Analgesia,” Pediatrics International 44, pp. 335–337. Rao, N. G., and L. M. Pai. 1988. “Costal Cartilage Calcification Pattern—A Clue for Establishing Sex Identity,” Forensic Science International 38, pp. 193–202. Rautman, M. 2004. “Valley and Village in Late Roman Cyprus,” in Bowden, Lavan, and Machado 2004, pp. 189–218. ———. 2005. “The Villages of Byzantine Cyprus,” in Lefort, Morrisson, and Sodini 2005, pp. 453–463. Rawson, B. 1986. “The Roman Family,” in The Family in Ancient Rome: New Perspectives, ed. B. Rawson, Ithaca, pp. 1–57. ———. 2003. Children and Childhood in Roman Italy, Oxford. Rebillard, É. 1993. “Κοιμητήριον et coemeterium: Tombe, tombe sainte, nécropole,” MÉFRA 105, pp. 975–1001. ———. 2003a. “Conversion and Burial in the Late Roman Empire,” in Conversion in Late Antiquity and the Middle Ages, ed. K. Mills and A. Grafton, Rochester, pp. 61–83. ———. 2003b. Religion et sépulture: L’église, les vivants, et les morts dans l’antiquité tardive (Civilisations et sociétés 115), Paris. Reese, D. S., K. Rielly, and M. J. Rose. 1987. “A Bone Assemblage at Corinth of the Second Century after Christ,” Hesperia 56, pp. 255–274. Resnick, D. 2002. Diagnosis of Bone and Joint Disorders, 5 vols., 4th ed., Philadelphia. RIC X = J. P. C. Kent, The Divided Empire and the Fall of the Western Parts, a.d. 346–498 (The Roman Imperial Coinage X), London 1994. Ricaut, F. X., and M. Waelkens. 2008. “Cranial Discrete Traits in a Byzantine Population and Eastern Mediterranean Population Movements,” HB 80, pp. 535– 564. Ricci, R., D. Mancinelli, R. Vargiu, A. Cucina, E. Santandrea, A. Capelli, and P. Catalano. 1997. “Patterns of Porotic Hyperostosis and Quality of Life in a II Century a.d. Farm near Rome,” Rivista di antropologia 75, pp. 117–128. Richardson, P. 2002. “Judaism and Christianity in Corinth after Paul: Texts and Material Evidence,” in Pauline Conversations in Context: Essays in Honor of Calvin J. Roetzel (Journal for the Study of the New Testament Suppl. 221) ed J. C. Anderson, P. Sellew and C. Setzer, London, pp. 42–66. Rife, J. L. 1998. “Roman and Byzantine Mortuary Practices in Rural Greece: The Case of Isthmia,” AJA 102, p. 382 (abstract). ———. 1999. “Death, Ritual, and Memory in Greek Society during the Early and Middle Roman Empire” (diss. Univ. of Michigan). ———. 2008. “Leo’s Peloponnesian Fire-Tower and the Byzantine Watch-Tower on Acrocorinth,” in Caraher, Hall, and Moore 2008, pp. 281–306. ———. In prep. “Food and Urban Society in Middle Byzantine Greece: A Deposit of Animal Bones from Corinth.”

BIBLIOGRAPHY AND ABBREVIATIONS Rife, J. L., and M. J. Giesen. 1994. “Regional Population Diversity and the Archaic to Early Byzantine Cranial Remains from Isthmia, Greece,” in Beyond the Site: Regional Studies in the Aegean Area, ed. P. N. Kardulias, Lanham, Md., pp. 223–263. Rife, J. L., M. M. Morison, A. Barbet, R. K. Dunn, D. H. Ubelaker, and F. Monier. 2007. “Life and Death at a Port in Roman Greece: The Kenchreai Cemetery Project 2002–2006,” Hesperia 76, pp. 143–181. Robb, J. 2002. “Time and Biography: Osteobiography of the Italian Neolithic Lifespan,” in Thinking through the Body: Archaeologies of Corporeality, ed. Y. Hamilakis, M. Pluciennik, and S. Tarlow, New York, pp. 153– 171. Robb, J., R. Bigazzi, L. Lazzarini, C. Scarsini, and F. Sonego. 2001. “Social ‘Status’ and Biological ‘Status’: A Comparison of Grave Goods and Skeletal Indicators from Pontecagnano,” AJPA 115, pp. 213– 222. Robert, L. 1960. “Épitaphes et acclamations byzantines à Corinthe,” in Hellenica 11–12, Paris, pp. 21–52. Roberts, C., C. Bourbou, A. Lagia, S. Triantaphyllou, and A. Tsaliki. 2005. “Health and Disease in Greece: Past, Present and Future,” in Health in Antiquity, ed. H. King, London, pp. 32–58. Roberts, C., and K. Manchester. 1995. The Archaeology of Disease, 2nd ed., Ithaca. Robinson, H. S. 1962. “Excavations at Corinth, 1960,” Hesperia 31, pp. 95–133. ———. 1964. “Corinth Excavations (1961),” ArchDelt 17, B΄1 (1961–1962), pp. 61–63. ———. 1965. “Excavations at Corinth,” ArchDelt 18, B ΄1 (1963), pp. 76–80. ———. 1966. “A Green-Glazed ‘Modiolus’ from Kenchreai,” in Χαριστήριον εἰς Ἀναστάσιον Κ. Ὀρλάνδον 3 (ΒΑΑΕ 54), Athens, pp. 179–185. ———. 1968. “ Βυζαντινὰ καὶ μεσαιωνικὰ μνημεῖα Πελοποννήσου. The Church of Haghia Paraskevi, Ancient Corinth,” ArchDelt 22, B΄1 (1967), pp. 218–219. ———. 1976a. “Excavations at Corinth: Temple Hill, 1968– 1972,” Hesperia 45, pp. 203–239. ———. 1976b. “Temple Hill, Corinth,” in Neue Forschungen in griechischen Heiligtümern, ed. U. Jantzen, Tübingen, pp. 239–260. Robledo, B., J. G. Trancho, and D. Brothwell. 1995. “Cribra Orbitalia: Health Indicator in the Late Roman Population at Cannington (Somerset, Great Britain),” JPaleopath 7, pp. 185–193. Rochow, I. 1991. Byzanz im 8. Jahrhundert in der Sicht des Theophanes. Quellenkritisch-historischer Kommentar zu den Jahren 715–813, Berlin. Rogers, J., and T. Waldron. 1995. A Field Guide to Joint Disease in Archaeology, Chichester. Rogers, J., T. Waldron, P. Dieppe, and I. Watt. 1987. “Arthropathies in Paleopathology: The Basis of Classification According to Most Probable Cause,” JAS 14, pp. 179–183. Rolley, C. 1965. “Chronique des fouilles et découvertes archéologiques en Grèce en 1964. II. Travaux de l’École française d’Athènes: Thasos. Sanctuaire d’Evraiocastro,” BCH 89, pp. 919–924. Rose, J. C., G. J. Armelagos, and L. S. Perry. 1993. “Dental Anthropology of the Nile Valley,” in Biological Anthropology and the Study of Ancient Egypt, ed. W. V. Davies and R. Walker, London, pp. 61–74. Rose, J. C., K. W. Condon, and A. H. Goodman. 1985. “Diet and Dentition: Developmental Disturbances,” in Gilbert and Mielke 1985, pp. 281–305.

xli

Roskams, S. 2001. Excavation, Cambridge. Rosser, J. 2005. “Dark Age Settlements in Grevena, Greece (Southwestern Macedonia),” in Lefort, Morrisson, and Sodini 2005, pp. 279–287. Rothaus, R. M. 1994. “Urban Space, Agricultural Space and Villas in Late Roman Corinth,” in Structures rurales et sociétés antiques: Actes du colloque de Corfou, ed. P. N. Doukellis and L. G. Mendoni, Paris, pp. 391–396. ———. 1996. “Earthquakes and Temples in Late Antique Corinth,” in Archaeoseismology (Fitch Laboratory Occasional Papers 7), ed. S. Stiros and R. E. Jones, Athens, pp. 105–112. ———. 2000. Corinth, First City of Greece: An Urban History of Late Antique Cult and Religion (Religions in the GraecoRoman World 138), Leiden. Rowell, G. 1977. The Liturgy of Christian Burial, London. Rush, A. C. 1941. Death and Burial in Christian Antiquity (The Catholic University of America Studies in Christian Antiquity 1), Washington, D.C. Russell, J. C. 1985. The Control of Late Ancient and Medieval Population (Memoirs of the American Philosophical Society 160), Philadelphia. Ryan, A. S. 1997. “Iron-Deficiency Anemia in Infant Development: Implications for Growth, Cognitive Develeopment, Resistance to Infection, and Iron Supplementation,” YPA 40, pp. 25–62. Saïd, S. 2005. “The Mirage of Greek Continuity: On the Uses and Abuses of Analogy in Some Travel Narratives from the Seventeenth to the Eighteenth Century,” in Rethinking the Mediterranean, ed. W. V. Harris, Oxford, pp. 268– 293. Saluja, G., K. Fitzpatrick, M. Bruce, and J. Cross. 1986. “Schmorl’s Nodes (Intravertebral Herniations of the Intervertebral Disc Tissue) in Two Historic British Populations,” JAnat 145, pp. 87–96. Salvadei, L., and G. Manzi. 1998. “Dental Anthropology, Paleobiology, and Environment: An Example from Archaeologically Controlled Contexts in Central Italy,” Rivista di antropologia 76, pp. 139–145. Salvadei, L., F. Ricci, and G. Manzi. 2001. “Porotic Hyperostosis as a Marker of Health and Nutritional Conditions During Childhood: Studies at the Transition Between Imperial Rome and the Early Middle Ages,” AJHB 13, pp. 709–717. ———. 2002. “Letter to the Editor: Response,” AJHB 14, pp. 418–419. Samellas, A. 2002. Death in the Eastern Mediterranean (50–600 a.d.). The Christianization of the East: An Interpretation (Studien und Texte zu Antike und Christentum 12), Tübingen. Samothrace XI.1 = E. B. Dusenbery, The Necropoleis and Catalogues of Burials, Princeton 1998. Sanders, G. D. R. 1999. “A Late Roman Bath at Corinth: Excavations in the Panayia Field, 1995–1996,” Hesperia 68, pp. 441–480. ———. 2002. “Corinth,” in EHB 2, pp. 647–654. ———. 2003. “Recent Developments in the Chronology of Byzantine Corinth,” in Corinth XX, pp. 385–399. ———. 2004. “Problems in Interpreting Urban and Rural Settlement in Southern Greece, a.d. 365–700,” in Landscapes of Change: The Evolution of the Countryside from Late Antiquity to the Early Middle Ages, ed. N. Christie, Aldershot, pp. 163–193. ———. 2005a. “Urban Corinth: An Introduction,” in Schowalter and Friesen 2005, pp. 11–24. ———. 2005b. “Archaeological Evidence for Early Christianity and the End of Hellenic Religion at Corinth,” in Schowalter and Friesen 2005, pp. 419–444.

xlii

BIBLIOGRAPHY AND ABBREVIATIONS

Saradi, H. 1996–1997. “The Use of Ancient Spolia in Byzantine Monuments: The Archaeological and Literary Evidence,” IJCT 3, pp. 395–423. Saradi-Mendelovici, H. 1990. “Christian Attitudes Toward Pagan Monuments in Late Antiquity and Their Legacy in Later Byzantine Centuries,” DOP 44, pp. 47–91. Sardis I = G. E. Bates, Byzantine Coins (Archaeological Exploration of Sardis I), Cambridge, Mass., 1971. Sarpaki, A. 2005. “The Archaeobotanical Material from the Site of Pyrgouthi in the Berbati Valley: The Seeds,” in Pyrgouthi, pp. 313–341. Šašel-Kos, M. 1978. “The Latin Inscriptions from Isthmia,” ArhVest 29, pp. 346–353. Sattenspiel, L., and H. Harpending. 1983. “Stable Populations and Skeletal Age,” AmerAnt 48, pp. 489–498. Sauer, N. J. 1998. “The Timing of Injuries and Manner of Death: Distinguishing among Antermortem, Perimortem and Postmortem Trauma,” in Forensic Osteology: Advances in the Identification of Human Remains, 2nd ed., ed. K. J. Reichs, Springfield, Ill., pp. 321–332. Saunders, S. R. 1978. The Development and Distribution of Discontinuous Morphological Variation of the Human Infracranial Skeleton (Archaeological Survey of Canada Papers 81), Ottawa. ———. 1989. “Nonmetric Skeletal Variation,” in İşcan and Kennedy 1989, pp. 95–108. ———. 2000. “Subadult Skeletons and Growth Related Studies,” in Katzenberg and Saunders 2000, pp. 135– 161. Savvides, A. G. C. 1990. “Morea and Islam, 8th–15th Centuries: A Survey,” Journal of Oriental and African Studies 2, pp. 47–71. ———. 1998. “Prosopographical Notes on the 9th-Century Byzantine Admiral Nicetas Ooryphas,” Mésogeios 1, pp. 84–96. Saxe, A. A. 1970. “Social Dimensions of Mortuary Practice” (diss. Univ. of Michigan). Schell, D. 1989. Griechische Totenbräuche: Ein quellenkritischer Beitrag zur Brauchforschung im modernen Griechenland (Mundus Reihe Volkskunde 4), Bonn. Scheuer, L., and S. Black. 2000. Developmental Juvenile Osteology, San Diego. Schiffer, M. B. 1987. Formation Processes of the Archaeological Record, Albuquerque. Schmidt, B. 1926. “Totengebräuche under Gräberkult im heutigen Griechenland,” ArchRW 24, pp. 281–318. Schmorl, G., and H. Junghanns. 1971. The Human Spine in Health and Disease, 2nd ed., New York. Schowalter, D. N., and S. J. Friesen, eds. 2005. Urban Religion in Roman Corinth: Interdisciplinary Approaches (Harvard Theological Studies 53), Cambridge, Mass. Schreiner, P. 1970. “Notes sur la fondation de Monemvasie,” TravMém 4, pp. 471–475. ———. 1976–1978. “Die byzantinische Lokalchronistik in der Peloponnes,” in ΠΔΣΠΣ 1.2, pp. 98–102. ———. 1977. Die byzantinischen Kleinchroniken 2: Historischer Kommentar (CFHB 12.2), Vienna. Schultz, M. 1997. “Microscopic Structure of Bone,” in Haglund and Sorg 1997a, pp. 187–199. Sciulli, P. W. 1978. “Developmental Abnormalities of the Permanent Dentition in Prehistoric Ohio Valley Amerindians,” AJPA 48, pp. 193–198. Scott, E. 1999. The Archaeology of Infancy and Infant Death (BAR-IS 819), Oxford. Sease, C. 1994. A Conservation Manual for the Field Archaeologist (Archaeological Research Tools 4), 3rd ed., Los Angeles. Selwitz, C. 1988. Cellulose Nitrate in Conservation, Marina del Rey, Cal.

Seremetakis, C. N. 1991. The Last Word: Women, Death, and Divination in Inner Mani, Chicago. Setton, K. M. 1950. “The Bulgars in the Balkans and the Occupation of Corinth in the Seventh Century,” Speculum 25, pp. 502–543. ———. 1952. “The Emperor Constans II and the Capture of Corinth by the Onogur Bulgars,” Speculum 27, pp. 351–362. Shay, T. 1985. “Differentiated Treatment of Deviancy in Death as Revealed in Anthropological and Archaeological Material,” JAnthArch 4, pp. 221–241. Shear, T. L. 1928. “Excavations in the Theater District and Tombs of Corinth in 1928,” AJA 32, pp. 474–495. ———. 1929. “Excavations in the Theater District and Tombs of Corinth in 1929,” AJA 33, pp. 515–546. ———. 1930. “Excavations in the North Cemetery at Corinth in 1930,” AJA 34, pp. 403–431. ———. 1931. “The Excavation of Roman Chamber Tombs at Corinth in 1931,” AJA 35, pp. 424–441. Shelley, J. M. 1943. “The Christian Basilica near the Cenchrean Gate at Corinth,” Hesperia 12, pp. 166– 189. Silverman, H., and D. B. Small, eds. 2002. The Space and Place of Death (Archaeological Papers of the American Anthropological Association 11), Arlington, Va. Singowitz, B. 1956. Studien zum Strafrecht der Ekloge, Athens. Sironen, E. 1997. The Late Roman and Early Byzantine Inscriptions of Athens and Attica: An Edition with Appendices on Scripts, Sepulchral Formulae, and Occupations, Helsinki. Sjøvold, T. 1984. “A Report on the Heritability of Some Cranial Measurements and Non-Metric Traits,” in Multivariate Statistical Methods in Physical Anthropology: A Review of Recent Advancements and Current Developments, ed. G. N. van Vark and W. W. Howells, Dordrecht, pp. 223–246. Skarmoutsou, K. 1997. “6η Εφορεία Βυζαντινών Αρχαιοτήτων. Ανασκαφικές εργασίες. Νομός Κορίνθου,“ ArchDelt 47, B΄1 (1992), pp. 164–168. Skarmoutsou-Dimitropoulou, K. 1995. “6η Εφορεία Βυζαντινών Αρχαιοτήτων. Ανασκαφικές εργασίες. Νομός Κορίνθου,” ArchDelt 45, B΄1 (1990), pp. 152–155. Skinner, M., and A. H. Goodman. 1992. “Anthropological Uses of Developmental Defects of Enamel,” in Skeletal Biology of Past Peoples: Research Methods, ed. S. R. Saunders and M. A. Katzenberg, New York, pp. 153–175. Slane, K. W., and G. D. R. Sanders. 2005. “Corinth: Late Roman Horizons,” Hesperia 74, pp. 243–297. Šlaus, M. 2002. The Bioarchaeology of Continental Croatia: An Analysis of Human Skeletal Remains from the Prehistoric to Post-Medieval Periods (BAR-IS 1021), Oxford. Smith, B. H. 1984. “Patterns of Molar Wear in HunterGatherers and Agriculturalists,” AJPA 63, pp. 39–56. ———. 1991. “Standards of Human Tooth Formation and Dental Age Assessment,” in Kelley and Larsen 1991, pp. 143–168. Smith, J. Z. 1987. To Take Place: Toward Theory in Ritual, Chicago. Smith, M. O. 1996. “‘Parry’ Fractures and Female-Directed Interpersonal Violence: Implications for the Late Archaic Period of West Tennessee,” IJO 6, pp. 84–91. Smith, P., O. Bar-Yosef, and A. Sillen. 1984. “Archaeological and Skeletal Evidence for Dietary Change during the Late Pleistocene/Early Holocene in the Levant,” in Cohen and Armelagos 1984, pp. 101–136. Snively, C. S. 1984. “Cemetery Churches of the Early Byzantine Period in Eastern Illyricum: Location and Martyrs,” GOTR 29, pp. 117–124.

BIBLIOGRAPHY AND ABBREVIATIONS ———. 1998. “Intramural Burial in the Cities of the Late Antique Diocese of Macedonia,” in Acta XIII Congressus internationalis archaeologiae christianae 2, ed. N. Cambi and E. Marin, Vatican City, pp. 491–498. Soames, J. V., and J. C. Southam. 1998. Oral Pathology, 3rd ed., Oxford. Sodini, J.-P. 1977. “Témoignages archéologiques sur la persistance à l’époque paléochrétienne et byzantine des rites funéraires païens,” in La mort au Moyen Âge: Colloque de l’Association des historiens médiévistes français réunis à Strasbourg, Strasbourg, pp. 11–21. ———. 1986. “Les ‘tombes privilégiées’ dans l’Orient chrétien (à la exception du diocèse d’Egypte),” in L’inhumation privilégiée du IVe au VIIIe siècle en Occident, ed. Y. Duval and J.-C. Picard, Paris, pp. 233–243. ———. 1993. “La contribution de l’archéologie a la connaissance du monde byzantin (IVe–VIIe siècles),” DOP 47, pp. 139–184. ———. 2003. “Archaeology and Late Antique Social Structures,” in Theory and Practice in Late Antique Archaeology (Late Antique Archaeology 1), ed. L. Lavan and W. Bowden, Leiden, pp. 25–56. Sørensen, M. L. S. 2000. Gender Archaeology, Cambridge. Soteriou, G. A. 1917. “Τὰ ἐρείπια τοῦ παρὰ τὸν Ἄρειον Πάγον βυζαντινοῦ ναοῦ,” ArchDelt 2 (1916), pp. 118–146. ———. [1942] 1962. Χριστιανικὴ καὶ βυζαντινὴ ἀρχαιολογία 1, repr. Athens. Southern, P., and K. R. Dixon. 1996. The Late Roman Army, New Haven. Southern Argolid = M. H. Jameson, C. Runnels, and T. H. van Andel, A Greek Countryside: The Southern Argolid from Prehistory to the Present Day, Stanford 1994. Spawforth, A. J. S. 1989. “Agonistic Festivals of Roman Greece,” in The Greek Renaissance in the Roman Empire (BICS Suppl. 55), ed. S. Walker and A. Cameron, London, pp. 193–197. Spieser, J.-M. 1976. “La christianization des sanctuaires païens en Grèce,” in Neue Forschungen in griechischen Heiligtümern, ed. U. Jantzen, Tübingen, pp. 309– 320. ———. 1984. “La ville en Grèce du IIIe au VIIe siècle,” in Villes et peuplement dans l’Illyricum protobyzantin (CÉFR 77), Rome, pp. 315–340. Spitz, W. U. 1993. “Injury by Gunfire,” in Spitz and Fisher’s Medicolegal Investigation of Death: Guidelines for the Application of Pathology to Crime Investigation, 3rd ed., ed. W. U. Spitz, Springfield, Ill., pp. 311–412. Spon, J. 1678. Voyage d’Italie, de Dalmatie, de Grèce, et du Levant fait en années 1675 et 1676 par Iacob Spon, docteur médécin aggregé à Lyon et George Wheler, gentilhomme anglois 2, Lyon. Sprague, R. 2005. Burial Terminology: A Guide for Researchers, Lanham, Md. Spyridakis, G. K. 1950. “Τὰ κατὰ τὴν τελευτὴν ἔθιμα τῶν Βυζαντινῶν ἐκ τῶν ἁγιολογικῶν πηγῶν ,” EpetByz 20, pp. 75–171. Staïs, E. 1906. “ Ἀνασκαφαὶ ἐν Ἰσθμίᾳ,” Prakt 1903, pp. 14–17. Steckel, R. H., and J. C. Rose, eds. 2002a. The Backbone of History: Health and Nutrition in the Western Hemisphere, Cambridge. ———. 2002b. “Conclusions,” in Steckel and Rose 2002a, pp. 583–589. Štefanovičová, T. 1997. “Slavic Settlements of Greece in the Light of the Archaeological Sources,” in Works of the VIth International Congress of Slavic Archaeology: Ethnogenesis and Ethnocultural Contacts of the Slavs 3, Moscow, pp. 352–361. Steinbock, R. T. 1976. Paleopathological Diagnosis and Interpretation: Bone Diseases in Ancient Populations, Springfield, Ill.

xliii

Steinhauer, G. 2001. “The Mesogaia in the Roman Period (1st–3rd century AD),” in Mesogaia: History and Culture of Mesogeia in Athens, ed. G. N. Aikaterinides and C. Doumas, Athens, pp. 143–144. Stewart, C. 1991. Demons and the Devil: Moral Imagination in Modern Greek Culture, Princeton. Stewart, T. D. 1979. Essentials of Forensic Anthropology, Springfield, Ill. Stikas, E. 1964. “Κοιμητηριακὴ βασιλικὴ Παλαιᾶς Κορίνθου,” Prakt 1961, pp. 129–136. ———. 1966. “ Ἀνασκαφὴ κοιμητηριακῆς βασιλικῆς Παλαιᾶς Κορίνθου,” Prakt 1962, pp. 51–56. Stiros, S. C. 1996. “Identification of Earthquakes from Archaological Data: Methodology, Criteria and Limitations,” in Archaeoseismology (Fitch Laboratory Occasional Papers 7), ed. S. Stiros and R. E. Jones, Athens, pp. 105–112. Stuart-Macadam, P. 1985. “Porotic Hyperostosis: Representative of a Childhood Condition,” AJPA 66, pp. 391–398. ———. 1989. “Porotic Hyperostosis: Relationship between Orbital and Vault Lesions,” AJPA 80, pp. 187– 193. ———. 1991. “Anemia in Roman Britain: Poundbury Camp,” in Health in Past Societies: Biocultural Interpretations of Human Skeletal Remains in Archaeological Contexts (BAR-IS 567), ed. H. Bush and M. Zvelebil, Oxford, pp. 101– 113. ———. 1992a. “Porotic Hyperostosis: A New Perspective,” AJPA 87, pp. 39–48. ———. 1992b. “Anemia in Past Human Populations,” in Diet, Demography, and Disease: Changing Perspectives on Anemia, ed. P. Stuart-Macadam and S. K. Kent, New York, pp. 151–170. Swanson, A. B., and G. de Groot Swanson. 1985. “Osteoarthritis in the Hand,” Clinics in Rheumatic Diseases 11, pp. 393–420. Swärdstedt. T. 1966. Odontological Aspects of a Medieval Population in the Province of Jamtland/Mid-Sweden, Stockholm. Sweetman, R. 2005. “Knossos Medical Faculty Site: Late Antique Graves,” BSA 100, pp. 331–386. Tainter, J. A. 1975. “Social Inference and Mortuary Practices: An Experiment in Numerical Classification,” WorldArch 7, pp. 1–15. ———. 1978. “Mortuary Practices and the Study of Prehistoric Social Systems,” in Advances in Archaeological Method and Theor y 1, ed. M. Schiffer, New York, pp. 105–141. Talbot, A.-M. 1984. “Old Age in Byzantium,” ByzZeit 77, pp. 267–278. Tartaron, T., T. E. Gregory, D. Pullen, J. S. Noller, R. M. Rothaus, J. L. Rife, L. Diacopoulos, R. Schon, W. Caraher, D. Pettegrew, and D. Nakassis. 2006. “The Eastern Korinthia Archaeological Survey: Integrated Methods for a Dynamic Landscape,” Hesperia 75, pp. 453– 523. Taylor, B. 1859. Travels to Greece and Russia, with an Excursion to Crete, London. Thillaud, P. L. 1992. “Retrospective Diagnosis in Paleopathology,” Paleopathology Newsletter 80 Suppl., pp. 1–4. Todd, T. W. 1920. “Age Changes in the Pubic Bone. I: The White Male Pubis,” AJPA 3, pp. 285–334. ———. 1921. “Age Changes in the Pubic Bone. II: The Pubis of the Male-Negro White Hybrid; III: The Pubis of the White Female; IV: The Pubis of the Female Negro-White Hybrid,” AJPA 4, pp. 1–70.

xliv

BIBLIOGRAPHY AND ABBREVIATIONS

Travlos, J., and A. Frantz. 1965. “The Church of St. Dionysius the Areopagite and the Palace of the Archbishop of Athens in the 16th Century,” Hesperia 34, pp. 157–202. Tritsaroli, P. 2006. “Pratiques funéraires en Grèce centrale à la période byzantine: Analyse à partir des données archéologiques et biologiques” (diss. Muséum national d’histoire naturelle, Paris). Trombley, F. 1978. “The Council in Trullo (691–2): A Study of the Canons Relating to Paganism, Heresy, and the Invasions,” Comitatus 9, pp. 1–18 ———. 1985. “Paganism in the Greek World at the End of Antiquity: The Case of Rural Anatolia and Greece,” HTR 78, pp. 327–352. ———. 2001a. Hellenic Religion and Christianization, c. 370– 529, 2 vols., Leiden. ———. 2001b. “Town and Territorium in Late Roman Anatolia (late 5th to early 7th c.),” in Lavan 2001b, pp. 217–232. Trotter, M. 1970. “Estimation of Stature from Intact Long Limb Bones,” in Personal Identification in Mass Disasters, ed. T. D. Stewart, Washington, D.C., pp. 71–83. Trotter, M., and G. C. Gleser. 1952. “Estimation of Stature from Long-Bones of American Whites and Negroes,” AJPA 10, pp. 463–514. ———. 1958. “A Re-evaluation of Estimation of Stature Based on Measurements of Stature Taken during Life and Long-Bones after Death,” AJPA 16, pp. 79–123. Tsigaridas, E. 1977. “Βυζαντινὰ καὶ μεσαιωνικὰ μνημεῖα Μακεδονίας-Θράκης. Α´. Κεντρικὴ καὶ δυτικὴ Μακεδονία,” ArchDelt 28, B΄2 (1973), pp. 477–501. Tsougarakis, D. 1988. Byzantine Crete from the 5th Century to the Venetian Conquest, Athens. Turkel, S. J. 1989. “Congenital Abnormalities in Skeletal Populations,” in İşcan and Kennedy 1989, pp. 109–127. Turner, C. G., II. 1979. “Dental Anthropological Indications of Agriculture among the Jomon People of Central Japan. X: People of the Pacific,” AJPA 51, pp. 619–636. Turner, C. G., II, and J. D. Cadien. 1969. “Dental Chipping in Aleuts, Eskimos and Indians,” AJPA 31, pp. 303–310. Turner, C. G., II, C. R. Nichol, and G. R. Scott. 1991. “Scoring Procedures for Key Morphological Traits of the Permanent Dentition: The Arizona State University Dental Anthropology System,” in Kelley and Larsen 1991, pp. 13–31. Turner, D. 1990. “The Politics of Despair: The Plague of 746–7 and Iconoclasm in the Byzantine Empire,” BSA 85, pp. 419–434. Tzortzopoulou-Gregory, L. 2008. “Cemeteries in the Countryside: An Archaeological Investigation of the Modern Mortuary Landscape in the Eastern Corinthia and Northern Kythera,” in Caraher, Hall, and Moore 2008, pp. 307–344. Ubelaker, D. H. 1987. “Estimating Age at Death from Immature Human Skeletons: An Overview,” JFS 32, pp. 1254–1263. ———. 1991. “Perimortem and Postmortem Modification of Human Bone: Lessons from Forensic Anthropology,” Anthropologie 29, pp. 171–174. ———. 1992. “Porotic Hyperostosis in Prehistoric Ecuador,” in Diet, Demography and Disease: Changing Perspectives on Anemia, ed. P. Stuart-Macadam and S. K. Kent, New York, pp. 201–217. ———. 1999. Human Skeletal Remains: Excavation, Analysis, Interpretation, 3rd ed., Washington, D.C. ———. 2000. “Temporal Trends in Old World Patterns of Morbidity,” Chungará 32, pp. 33–40.

Vaes, J. 1984–1986. “Christliche Wiederverwendung antiker Bauten: Ein Forschungsbericht,” Ancient Society 15–17, pp. 305–443. ———. 1989. “‘Nova construere sed amplius vetusta servare’: La réutilisation chrétienne d’édifices antiques (en Italie),” in Actes du XIe Congrès international d’archéologie chrètienne 1 (CÉFR 123), Paris, pp. 299–321. van Andel, T. H., and C. Runnels. 1987. Beyond the Acropolis: A Rural Greek Past, Stanford. van Nijf, O. 1997. The Civic World of Professional Associations in the Roman East (Dutch Monographs on Ancient History and Archaeology 17), Amsterdam. Varsik, V. 1992. “Byzantinische Gürtelschnallen im mittleren und unteren Donauraum im 6. und 7. Jahrhundert,” Slovenská archeológia 40, pp. 77–106. Vavřínek, V., ed. 1985. From Late Antiquity to Early Byzantium: Proceedings of the Byzantinological Symposium in the 16th International Eirene Conference, Prague. Velkovska, E. 2001. “Funeral Rites According to the Byzantine Liturgical Sources,” DOP 55, pp. 21–51. Verdelis, N. M. 1965. “ Ἀνασκαφὴ εἰς τὴν Ἀρχαίαν Σολύγειαν,” Prakt 1958, pp. 135–145. Vida, T., and T. Völling. 2000. Das slawische Brandgräberfeld von Olympia (Archäologie in Eurasien 9), Rahden. Vikatou, O. 2002. “Το χριστιανικό νεκροταφείο στην Αγία

Τριάδα Ηλείας. Συμβολή στη μελέτη της χειροποίητης κεραμικής,” in Πρωτοβυζαντινή Μεσσήνη και Ολυμπία: Αστικός και αγροτικός χώρος στη Δυτική Πελοπόννησο. Πρακτικά του Διεθνούς Συμποσίου, ed. P. G. Themelis and

V. Konti, Athens, pp. 238–270. Vinski, Z. 1974. “Kasnoantički starosjedioci u salonitanskoj regiji prema arheološkoj ostavštini predslavenskog supstrata,” Vjesnik za arheologiju i historiju Dalmatinsku 69, pp. 5–86. Vodanović, M, H. Brkić, M. Šlaus, and D. Željko. 2005. “The Frequency and Distribution of Caries in the Medieval Population of Bijelo Brdo in Croatia (10th–11th Century),” ArchOB 50, pp. 669–680. Vollgraff, W. 1956. Le sanctuaire d’Apollon Pythèen à Argos (Études péloponnésiennes 1), Paris. Völling, T. 1997. “Ein frühbyzantinischer Hortfund aus Olympia,” AM 110, pp. 425–459. ———. 2001. “The Last Christian Greeks and the First Pagan Slavs in Olympia,” in Kountoura-Galaki 2001, pp. 303–323. Volp, U. 2002. Tod und Ritual in den christlichen Gemeinden der Antike (VigChr Suppl. 65), Leiden. Von Endt, D. W., and D. J. Ortner. 1984. “Experimental Effects of Bone Size and Temperature on Bone Diagenesis,” JAS 11, pp. 247–253. Vroom, J. 2003. After Antiquity: Ceramics and Society in the Aegean from the 7th to the 20th Century A.C. (Archaeological Studies, Leiden University 10), Leiden. Vryonis, S., Jr. 1992. “The Slavic Pottery (Jars) from Olympia, Greece,” in Byzantine Studies: Essays on the Slavic World and the Eleventh Century, ed. S. Vryonis Jr., New Rochelle, N.Y., pp. 15–42. Wade, W., S. C. Bisel, J. Rosser, and N. C. Wilkie. 1983. “The Burials,” in Nichoria 3: Dark Age and Byzantine Occupation, ed. W. A. McDonald, W. D. E. Coulson, and J. Rosser, Minneapolis, pp. 398–404. Wadsworth, G. R. 1992. “Physiological, Pathological, and Dietary Influences on the Hemoglobin Level,” in Diet, Demography and Disease: Changing Perspectives on Anemia, ed. P. Stuart-Macadam and S. K. Kent, New York, pp. 63–104.

BIBLIOGRAPHY AND ABBREVIATIONS Wahlberg, G. 2007. “Summary: The Lower Terraces from the 13th, 12th, and 11th Centuries b.c. to the Roman and Byzantine Period,” in Midea II, pp. 193–199. Walbank, M. E. H. 2005. “Unquiet Graves: Burial Practices of the Roman Corinthians,” in Schowalter and Friesen 2005, pp. 249–280. Waldron, H. A. 1993. “The Health of the Adults,” in The Spitalfields Project 2: The Anthropology, The Middling Sort (CBA Research Reports 86), ed. T. I. Molleson and M. Cox, York, pp. 67–89. Waldron, T. 1987. “The Relative Survival of the Human Skeleton: Implications for Paleopathology,” in Boddington, Garland, and Janaway 1987, pp. 55–64. ———. 1993. “The Distribution of Osteoarthritis of the Hands in a Skeletal Population,” IJO 3, pp. 213–218. ———. 1994. Counting the Dead: The Epidemiology of Skeletal Populations, Chichester. Waldron, T., and J. Rogers. 1991. “Inter-Observer Variation in Coding Osteoarthritis in Human Skeletal Remains,” IJO 1, pp. 49–56. Walker, A. 2001. “A Reconsideration of Early Byzantine Marriage Rings,” in Between Magic and Religion: Interdisciplinary Studies in Ancient Mediterranean Religion and Society, ed. S. R. Asirvatham, C. O. Pache, and J. Watrous, Lanham, Md., pp. 149–164. ———. 2003a. “Home: A Space ‘Rich in Blessing,’” in Kalavrezou 2003, pp. 161–231. ———. 2003b. “Adornment: Enhancing the Body, Neglecting the Soul?” in Kalavrezou 2003, pp. 233–273. Walker, P. L. 1985. “Anemia among Prehistoric Indians of the American Southwest,” in Health and Disease in the Prehistoric Southwest (Arizona State University Anthropological Research Papers 34), ed. C. F. Merbs and R. J. Miller, Tempe, pp. 139–163. ———. 1986. “Porotic Hyperostosis in a Marine-Dependent California Indian Population,” AJPA 69, pp. 345–354. ———. 1989. “Cranial Injuries as Evidence of Violence in Prehistoric Southern California, Santa Barbara,” AJPA 80, pp. 313–323. Walker, P. L., G. Dean, and P. Shapiro. 1991. “Estimating Age from Tooth Wear in Archaeological Populations,” in Kelley and Larsen 1991, pp. 169–178. Walker, P. L., J. J. Johnson, and P. M. Lambert. 1988. “Age and Sex Biases in the Preservation of Human Skeletal Remains,” AJPA 76, pp. 183–188. Walter, C. 1976. “Death in Byzantine Iconography,” Eastern Churches Review 8, pp. 113–127. Waltzing, J.-P. [1895] 1979. Étude historique sur les corporations professionelles chez les Romains depuis les origines jusqu’à la chute de l’Empire d’Occident 1: Le droit d’association à Rome. Les collèges professionnels considérés comme associations privées, repr. New York. Wardle, K. A., and D. Wardle. 2004. “Glimpses of Private Life: Roman Rock Cut Tombs of the First and Second Centuries a.d. at Knossos,” in Knossos: Palace, City, State. Proceedings of the Conference in Herakleion (British School at Athens Studies 12), ed. G. Cadogan, E. Hatzaki, and A. Vasilakis, London, pp. 473–480. Ward-Perkins, B. 1997. “Continuists, Catastrophists, and the Towns of Post-Roman Northern Italy,” PBSR 45, pp. 157–176. ———. 1999. “Re-Using the Architectural Legacy of the Past, entre idéologie et pragmatisme,” in The Idea and Ideal of the Town Between Late Antiquity and the Early Middle Ages, ed. G. P. Brogiolo and B. Ward-Perkins, Leiden, pp. 25– 57.

xlv

———. 2005. The Fall of Rome and the End of Civilization, Oxford. Watson, P. J. C. 1986. “A Study of the Pattern of Alveolar Recession,” in Teeth and Anthropology (BAR-IS 291), ed. E. Cruwys and R. A. Foley, Oxford, pp. 123–132. Watts, D. 1989. “Infant Burials and Romano-British Christianity,” ArchJ 146, pp. 372–383. ———. 1991. Christians and Pagans in Roman Britain, London. Waywell, G. B., and J. J. Wilkes. 1994. “Excavations at Sparta: The Roman Stoa, 1988–91, part 2,” BSA 89, pp. 377– 432. ———. 1997. “Excavations at Sparta: The Roman Stoa, 1988–91, part 3,” BSA 92, pp. 401–434. Wedel, A., G. E. Carlsson, and S. Sagne. 1978. “Temporomandibular Joint Morphology in Medieval Skull Material,” Swedish Dental Journal 2, pp. 177–187. Weinberg, S. S. 1939. “Excavations at Corinth, 1938–1939,” AJA 43, pp. 592–600. Weiss-Krejci, E. 2001. “Restless Corpses: ‘Secondary Burial’ in the Babenberg and Habsburg Dynasties,” Antiquity 75, pp. 769–780. ———. 2005. “Excarnation, Evisceration, and Exhumation in Medieval and Post-Medieval Europe,” in Rakita et al. 2005, pp. 155–172. Weithmann, M. W. 1978. Die slawische Bevölkerung auf der griechischen Halbinsel: Ein Beitrag zur historische Ethnographie Südosteuropas (Beiträge zur Kenntnis Südosteuropas und des Nahen Orients 31), Munich. ———. 1979. “Strukturkontinuität und -diskontinuität auf der griechischen Halbinsel im Gefolge der slavischen Landnahme,” Münchner Zeitschrift für Balkankunde 2, pp. 141–176. ———. 1985. “Anthropologisches Fundgut zur Einwanderung der Slaven in Griechenland: Eine Materialzusammenstellung,” Homo 36, pp. 103–109. ———. 1994. “Interdisciplinäre Diskrepanzen in der ‘Slavenfrage’ Griechenlands,” Zeitschrift für Balkanologie 30, pp. 85–111. Wells, C. 1967. “Pseudopathology,” in Diseases in Antiquity: A Survey of the Diseases, Injuries and Surgery of Early Populations, ed. D. R. Brothwell and A. T. Sandison, Springfield, Ill., pp. 5–19. ———. 1982. “The Human Burials,” in Cirencester Excavations 2: Romano-British Cemeteries at Cirencester, ed. A. McWhirr, L. Viner, and C. Wells, Cirencester, pp. 135–201. Wesolowsky, A. B. 1973. “The Skeletons of Lerna Hollow,” Hesperia 42, pp. 340–351. ———. 1975. “Pathology of Human Remains from Stobi, 1970–1973,” in Studies in the Antiquities of Stobi II, ed. J. Wiseman, Belgrade, pp. 143–160. Wheler, G. 1682. A Journey into Greece, London. Whitby, M. 1988. The Emperor Maurice and His Historian: Theophylact Simocatta on Persian and Balkan Warfare, Oxford. ———. 2000a. “Armies and Society in the Later Roman World,” in Cameron, Ward-Perkins, and Whitby 2000, pp. 469–495. ———. 2000b. “The Balkans and Greece, 420–602,” in Cameron, Ward-Perkins, and Whitby 2000, pp. 701– 730. White, T. D. 2000. Human Osteology, 2nd ed., San Diego. Whittaker, D. K., G. Davies, and M. Brown. 1985. “Tooth Loss, Attrition, and Temporomandibular Joint Changes in a Romano-British Population,” Journal of Oral Rehabilitation 12, pp. 407–419.

xlvi

BIBLIOGRAPHY AND ABBREVIATIONS

Whittaker, D. K., T. I. Molleson, A. T. Daniel, J. T. Rose, P. Resteghini, and R. Resteghini. 1985. “Quantitative Assessment of Tooth Wear, Alveolar Crest Height and Continuing Eruption in a Romano-British Population,” ArchOB 30, pp. 493–501. Whittow, M. 1996. The Making of Byzantium, 600–1025, Berkeley. ———. 2001. “Was Gibbon Politically Incorrect, or Just Wrong?” in Lavan 2001b, pp. 241–243. Williams, C. K., II. 2003. “Frankish Corinth: An Overview,” in Corinth XX, pp. 423–434. Williams, C. K., II, E. Barnes, and L. M. Snyder. 1997. “Frankish Corinth: 1996,” Hesperia 66, pp. 7–47. Williams, C. K., II, and J. E. Fisher. 1975. “Corinth, 1974: Forum Southwest,” Hesperia 44, pp. 1–50. Williams, C. K., II, J. MacIntosh, and J. E. Fisher. 1974. “Excavations at Corinth, 1973,” Hesperia 43, pp. 1–76. Williams, C. K., II, L. M. Snyder, E. Barnes, and O. H. Zervos. 1998. “Frankish Corinth: 1997,” Hesperia 67, pp. 223– 264. Williams, C. K., II, and O. H. Zervos. 1982. “Corinth, 1981: East of Theater,” Hesperia 51, pp. 115–163. ———. 1983. “Corinth, 1982: East of Theater,” Hesperia 52, pp. 1–47. ———. 1984. “Corinth, 1983: The Route to Sikyon,” Hesperia 53, 83–122. ———. 1986. “Corinth, 1985: East of the Theater,” Hesperia 55, pp. 129–205. ———. 1987. “Corinth, 1986: Temple E and East of the Theater,” Hesperia 56, pp. 1–46. ———. 1988. “Corinth, 1987: South of Temple E and East of Theater,” Hesperia 57, pp. 95–146. ———. 1990. “Excavations at Corinth, 1989: The Temenos of Temple E,” Hesperia 59, pp. 325–369. ———. 1991. “Corinth, 1990: Southeast Corner of Temenos E,” Hesperia 60, pp. 1–58. ———. 1992. “Frankish Corinth: 1991,” Hesperia 61, pp. 133– 191. ———. 1995. “Frankish Corinth: 1994,” Hesperia 64, pp. 1– 60. ———. 1996. “Frankish Corinth: 1995,” Hesperia 65, pp. 1– 55. Williams, H. 2003. “Introduction: The Archaeology of Death, Memory and Material Culture,” in Archaeologies of Remembrance: Death and Memory in Past Societies, ed. H. Williams, New York, pp. 2–24. Wiltse, L. L., E. J. Widell Jr., and D. W. Jackson. 1975. “Fatigue Fracture: The Basic Lesion in Isthmic Spondylolisthesis,” JBJS (American Volume) 57, pp. 17–22. Winkelmann, F., and W. Brandes, eds. 1990. Quellen zur Geschichte des frühen Byzanz (4.–9. Jahrhundert): Bestand und Probleme (Berliner byzantinistische Arbeiten 55), Amsterdam. Wiseman, J. 1967a. “Excavations at Corinth, The Gymnasium Area, 1965,” Hesperia 36, pp. 13–41. ———. 1967b. “Excavations at Corinth, The Gymnasium Area, 1966,” Hesperia 36, pp. 402–428. ———. 1969. “Excavations in Corinth, The Gymnasium Area, 1967–1968,” Hesperia 38, pp. 64–106. ———. 1972. “The Gymnasium Area at Corinth, 1969–1970,” Hesperia 41, pp. 1–42. ———. 1978. The Land of the Ancient Corinthians (SIMA 50), Göteborg.

Wohl, B. L. 1981. “A Deposit of Lamps from the Roman Bath at Isthmia,” Hesperia 50, pp. 112–140. ———. 1993. “Lamps from the Excavations at Isthmia by UCLA,” in Gregory 1993a, pp. 130–138. Wolpoff, M. H. 1970. “Interstitial Wear,” AJPA 34, pp. 205–228. Wood, J. W., and G. R. Milner. 1994. “Reply,” CurrAnthr 35, pp. 631–637. Wood, J. W., G. R. Milner, H. C. Harpending, and K. M. Weiss. 1992. “The Osteological Paradox: Problems of Inferring Prehistoric Health from Skeletal Samples,” CurrAnthr 33, pp. 343–370. Wood, W. R., and D. L. Johnson. 1978. “A Survey of Disturbance Processes in Archaeological Site Formation,” in Advances in Archaeological Method and Theory 1, ed. M. B. Schiffer, New York, pp. 315–381. Wortley, J. 1982. “Iconoclasm and Leipsanoclasm: Leo III, Constantine V, and the Relics,” BF 8, pp. 253–279. Wozniak, F. E. 1982, “The Justinianic Fortification of Interior Illyricum,” in City, Town and Country in the Early Byzantine Era, ed. R. L. Hohlfelder, Boulder, pp. 199– 209. Wright, L. E., and F. Chew. 1998. “Porotic Hyperostosis and Paleoepidemiology: A Forensic Perspective on Anemia among the Ancient Maya,” American Anthropologist 100, pp. 924–939. Wyse, T. 1865. An Excursion in the Peloponnesus in the Year 1858 2, London. Xirotiris, N. 1979. “Rassengeschichte von Griechenland,” in Rassengeschichte der Menschheit 6: Europa 4. Ungarn, Rumänien, Bulgarien, Jugoslawien, Albanien, Griechenland, ed. I. Schwidetzky, Munich, pp. 157–192. Yalouris, N. P. 1962. “ Ἀρχαιότητες Ἠλείας-Ἀχαΐας,” ArchDelt 16, B΄ (1960), pp. 125–126. ———. 1963. “ Ἀρχαιότητες Ἀχαΐας-Ἠλείας,” ArchDelt 17, B΄1 (1961–1962), pp. 105–107. ———. 1966. “ Ἠλεία,” ArchDelt 19, B΄2 (1964), pp. 174–182. ———. 1967. “ Ἀρχαιότητες καὶ μνημεῖα Ἠλείας,” ArchDelt 20, B΄2 (1965), pp. 209–213. ———. 1968. “ Ἀρχαιότητες καὶ μνημεῖα Ἠλείας,” ArchDelt 21, B΄1 (1966), pp. 170–173. Yannopoulos, P. 1980. “La pénétration slave en Argos,” in Études argiennes (BCH Suppl. 6), Paris, pp. 323–371. ———. 1993. “Métropoles du Péloponnèse mésobyzantin: Un souvenir des invasions avaro-slaves,” Byzantion 63, pp. 388–400. Yerolymatou, M. 2001. “ Ἐμπορικὴ δραστηριότητα κατὰ τοὺς σκοτεινοὺς αἰῶνες ,” in Kountoura-Galaki 2001, pp. 347–364. Young, B., and D. Papadatou. 1997. “Childhood Death and Bereavement across Cultures,” in Death and Bereavement across Cultures, ed. C. Murray Parkes, P. Laugani, and B. Young, London, pp. 191–205. Zakythinos, D. A. 1966. “La grande brèche dans la tradition historique de l’hellénisme du septième au neuvième siècle,” in Χαριστήριον εἰς Ἀναστάσιον Κ. Ὀρλάνδον 3 (ΒΑΑΕ 54), Athens, pp. 300–327. Zwipp, H., S. Rammelt, and R. Grass. 2002. “Ligamentous Injuries about the Ankle and Subtalar Joints,” Clinics in Podiatric Medicine and Surgery 19, pp. 195– 229. Zygouries = C. W. Blegen, Zygouries: A Prehistoric Settlement in the Valley of Cleonae, Cambridge, Mass., 1928.

BIBLIOGRAPHY AND ABBREVIATIONS

xlvii

ABBREVIATIONS OF PERIODICALS, SERIES, AND LEXICA AA = Archäologischer Anzeiger AASS = Acta sanctorum, 71 vols., Antwerp 1643–1940; repr. Turnhout 1966–1971 ActaArchHung = Acta archaeologica Academiae scientiarum Hungaricae AIIN = Annali dell’Istituto italiano di numismatica AJA = American Journal of Archaeology AJHB = American Journal of Human Biology AJPA = American Journal of Physical Anthopology AM = Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung AmerAnt = American Antiquity Antiquity = Antiquity. A Quarterly Review of Archaeology AR = Archaeological Reports ArchDelt = Αρχαιολογικόν Δελτίον ArchEph = Ἀρχαιολογικὴ Ἐφημερίς ArchJ = Archaeological Journal ArchKorrBl = Archäologisches Korrespondenzblatt ArchNews = Archaeological News ArchOB = Archives of Oral Biology ArchRW = Archiv für Religionswissenschaft ArhVest = Arheološki Vestnik ΒΑΑΕ = Βιβλιοθήκη τῆς ἐν Ἀθήναις Ἀρχαιολογικῆς Ἑταιρείας BAR-BS = British Archaeological Reports, British Series BAR-IS = British Archaeological Reports, International Series BASP = Bulletin of the American Society of Papyrologists BCH = Bulletin de correspondance hellénique BF = Byzantinische Forschungen BiblArch = Biblical Archaeologist BICS = Bulletin of the Institute of Classical Studies BSA = Annual of the British School at Athens Byzantion = Byzantion. Revue internationale des études byzantines ByzZeit = Byzantinische Zeitschrift CAH 2 = Cambridge Ancient History, 2nd ed. CahArch = Cahiers archéologiques CarRes = Caries Research CÉFR = Collection de l’École française de Rome CFHB = Corpus fontium historiae byzantinae CIG = Corpus inscriptionum graecarum CP = Classical Philology CPG = E. L. Leutsch and F. G. Schneidewin, eds., Corpus paroemiographorum Graecorum, 2 vols., Göttingen 1839–1851 CRAI = Comptes rendus des séances de l’Académie des inscriptions et belles-lettres [Paris] CSHB = Corpus scriptorum historiae byzantinae CurrAnthr = Current Anthropology DHGE = A. Baudrillart, A. de Meyer, and R. Aubert, eds., Dictionnaire d’histoire et de géographie ecclésiastiques, Paris 1912– DOP = Dumbarton Oaks Papers ΕΕΠΘ = Ἐπιστημονικὴ Ἐπετηρὶς τῆς Πολυτεχνικῆς Σχολῆς τοῦ Ἀριστοτελείου Πανεπιστημίου τῆς Θεσσαλονίκης EpetByz = Ἐπετηρὶς Ἑταιρείας Βυζαντινῶν Σπουδῶν Ergon = Τὸ Ἔργον τῆς ἐν Ἀθήναις Ἀρχαιολογικῆς Ἑταιρείας GazArch = Gazette archéologique GGM = C. Müller, ed., Geographi Graeci minores, 2 vols., Paris 1855–1861 GOTR = Greek Orthodox Theological Review GRBS = Greek, Roman and Byzantine Studies HB = Human Biology

HTR = Harvard Theological Review IG = Inscriptiones graecae IJAnth = International Journal of Anthropology IJCT = International Journal of the Classical Tradition IJO = International Journal of Osteoarchaeology JAC = Jahrbuch für Antike und Christentum JAnat = Journal of Anatomy JAnthArch = Journal of Anthropological Archaeology JAS = Journal of Archaeological Science JBJS = Journal of Bone and Joint Surgery JBL = Journal of Biblical Literature JFA = Journal of Field Archaeology JFS = Journal of Forensic Sciences JGR = I. Zepos and P. Zepos, eds., Jus Graecoromanum, 8 vols., Athens 1931; repr. Aalen 1962 JHS = Journal of Hellenic Studies JMA = Journal of Mediterranean Archaeology JÖB = Jahrbuch der österreichischen Byzantinistik JPaleopath = Journal of Paleopathology JRA = Journal of Roman Archaeology JSS = Journal of Semitic Studies LSJ = H. G. Liddell, R. Scott, and H. S. Jones, A Greek-English Lexicon, 9th ed., Oxford 1940 MAAR = Memoirs of the American Academy in Rome Mansi = G. D. Mansi, ed., Sacrorum conciliorum nova et amplissima collectio, 53 vols., Florence 1759–1798; repr. Paris 1901–1927 MÉFRA = Mélanges de l’École française de Rome, Antiquité MGHAA = Monumenta Germaniae historica, auctores antiquissimi MGHSS = Monumenta Germaniae historica, scriptores MGHSSrM = Monumenta Germaniae, scriptores rerum Merovingicarum OC = Oriens christianus OCD3 = S. Hornblower and A. J. S. Spawforth, eds., The Oxford Classical Dictionary, 3rd ed., Oxford 1996 OpAth = Opuscula atheniensia OWAN = Old World Archaeology Newsletter PastPres = Past and Present ΠΔΣΠΣ = Πρακτικὰ τοῦ Διεθνοῦς Συνεδρίου Πελοποννησιακῶν Σπουδῶν (Πελοποννησιακά Suppl.) PG = J.-P. Migne, ed., Patrologiae cursus completus, series graeca, 161 vols., Paris 1857–1866 PL = J.-P. Migne, ed., Patrologiae cursus completus, series latina, 221 vols., Paris 1844–1880 Prakt = Πρακτικὰ τῆς ἐν Ἀθήναις Ἀρχαιολογικῆς Ἑταιρείας RACrist = Rivista di archeologia cristiana RBK = K. Wessel and M. Restle, eds., Reallexikon zur byzantinischen Kunst, Stuttgart 1963– RDAC = Report of the Department of Antiquities, Cyprus SBMünch = Sitzungsberichte, Bayerische Akademie der Wissenschaften (München), Philosophisch-historische Klasse SEG = Supplementum epigraphicum graecum SkrAth = Skrifter utgivna av Svenska Institutet i Athen SIMA = Studies in Mediterranean Archaeology TravMém = Travaux et mémoires. Centre de recherche d’histoire et civilisation byzantine, Paris VigChr = Vigiliae Christianae WorldArch = World Archaeology YPA = Yearbook of Physical Anthropology

xlviii

BIBLIOGRAPHY AND ABBREVIATIONS

ABBREVIATIONS FOR TERMS IN THE TEXT AC = alveolar crest AMTL = antemortem tooth loss ATFL = anterior talofibular ligament C = canine(s) (often followed by superscript or subscript number[s]); or cervical vertebra/-ae (often followed by number[s]) CEJ = cementum-enamel junction cm = centimeters d. = died D. = depth/deep Diam. = diameter H. = height/high km = kilometers I = incisor(s) (often followed by superscripted or subscripted number[s]) L = left, when referring to bones or teeth L = lumbar vertebra/-ae (often followed by number[s]) L. = length/long

M = molar(s) (often followed by superscript or subscript number[s]) m = meters masl = meters above sea level max. = maximum mm = millimeters mos. = months NB = notebook (i.e., primary record of fieldwork) P = premolar(s) (often followed by superscript or subscript number[s]) p. = preserved R = right, when referring to bones or teeth S = sacral vertebra/-ae (often followed by number[s]) sd = standard deviation T = thoracic vertebra/-ae (often followed by number[s]) Th. = thickness/thick TMJ = temporomandibular joint W. = width/wide

1

 THEMES, PROCEDURES, AND MATERIALS

T

his study describes and interprets the graves and human remains of Roman and Byzantine date in the area of the Panhellenic Sanctuary1 and the succeeding fortifications on the Isthmus of Corinth, which were excavated under the auspices of the American School of Classical Studies between 1954 and 1976 (Figs. 1.1–1.3). These burials belonged to local residents during several phases in the long history of their settlement. Most date to the Late Roman and Early Byzantine periods (end of the 4th to late 7th or 8th centuries), but a few date to the Early to Middle Roman (mid-1st to 4th centuries) and the Middle to Late Byzantine eras (12th to 15th centuries) or later. The 30 graves2 found in 11 separate locales in the central, northern, eastern, and far western3 areas of the archaeological site called Isthmia contained the bodies of at least 69 individuals4 and an assortment of artifacts. Bones from at least four other skeletons were found in secondary contexts to which they seem to have been moved during ancient building activities. Besides the much earlier burials in the West Cemetery, these remains comprise the largest body of mortuary and skeletal evidence so far recovered at the Isthmus.5 The graves are spare and mundane in form and content. They were built mostly from reused materials and were partially sealed, were seldom marked at the surface and apparently never displayed epitaphs, and infrequently contained objects, such as small vessels, lamps, coins, and articles of clothing and adornment. In contrast to the paucity and simplicity of the graves, the surviving hard tissues of the bodies were relatively well preserved, especially those from enclosed interments that were not extensively disturbed. Several large graves contained multiple bodies that had been added over a short period. Both men and women, adults and children were interred in the graves. For the most part the inhabitants led healthy lives, but suffered ailments that were not uncommon in ancient Mediterranean society, such as tooth loss, broken arms, and arthritic joints.

1. The word “Sanctuary” refers to the entire area associated with the Isthmian cults and festivals, whose exact limits remain unknown; see Isthmia II, p. 3, n. 10. Within the Sanctuary, the “central area” is the area occupied by the precincts of Poseidon and Palaimon and all contiguous structures (Figs. 1.2, 2.76). 2. This total includes one unexcavated feature near Tower 14 that was probably a grave (pp. 71–72). 3. One grave was found at the West Foundation ca. 2 km west of the Temple of Poseidon (pp. 88–91). 4. Significant portions of 65 skeletons were available for study, although two subadults were represented only by teeth (NEG 69-009A, 69-010A). The remains of at least two more

subadults had completely or almost completely disintegrated before the graves were discovered (NEG 69-008, DEC 69-902). Skeletons in two other graves (T13 54-001, PAL 56-001) were recorded in situ but not recovered. 5. On the West Cemetery, see p. vii, n. 1. In addition, excavations in trench GW2 over the Mycenaean Wall directly south of the central area (NB JGH, p. 356 [Oct. 3, 1959]) uncovered portions of one human right lower leg and foot (bone lots 83+89+97: metatarsals I, II, III, and V, first proximal phalanx, calcaneus, distal and shaft fragments of tibia and fibula). These remains, which were mixed with ovicaprid and canine bones, seem to represent undifferentiated refuse from an early period at the Sanctuary.

2

THEMES, PROCEDURES, AND MATERIALS

Figure 1.1. The Isthmus of Corinth

THEMES AND APPROACHES These graves and human remains furnish valuable evidence concerning society, religion, economy, and biology in the Corinthian countryside from Roman to Byzantine times, particularly for the Late Roman and Early Byzantine periods. When considered together with associated traces of habitation, the mortuary and skeletal remains bear vivid testimony not only to how these people died but also to how they lived. The graves, their contents, and their locations were used by the local community in an intentional and meaningful manner. When men, women, and children participated in the ritual process of mourning, interment, and commemoration, they identified themselves according to social relationships and expressed their beliefs concerning life and death. A close examination of the material remains of these funerary rituals can therefore reveal the social, economic, and religious character of the community and its composition by gender, age, descent, and status. Furthermore, study of the human bones and teeth can reveal the demographic structure, diet, occupational hazards, and health of local residents. The larger significance of the graves, the human remains, and the settlement emerges when they are considered within the frame of regional trends.6 The full variability of mortuary 6. For approaches to mortuary and biological variability on a regional scale, see, e.g., O’Shea 1984, Beck 1995a, and Silverman and Small 2002.

Figure 1.2. The Isthmian Sanctuary

4

THEMES, PROCEDURES, AND MATERIALS

Figure 1.3. The Isthmian Fortress and adjacent graves

behavior within a region can only be traced through intersite comparison. Since separate communities within a geographic area live in proximity, share resources and natural conditions, and often interact with one another, their burial practices usually exhibit basic similarities. But the material and spatial dimensions of funerary ritual also reflect the structural or ideological variation between contemporary, contiguous groups. Likewise, the examination of biological diversity within a region can reveal differences in life experiences and population boundaries stemming from locally distinct economic and subsistence strategies, settlement patterns, and familial relations. The evidence from the Isthmus has therefore been examined within the regional context of the Corinthia. During the Roman and Byzantine eras, the regional center was the city of Corinth, which controlled two nearby ports, Lechaion and Kenchreai, and maintained a dependence on the surrounding countryside (Fig. 1.1). The urban community at Corinth

THEMES AND APPROACHES

5

furnishes the most significant point of comparison to the community in its hinterland at the Isthmus. While the two places were variously linked by the movement of traffic, regional military strategy, and systems of production and trade, they display different scales and complexities of settlement. Thus, an understanding of the Isthmus is dependent on an understanding of Corinth, and vice versa.7 It has been worthwhile, as well, to consider evidence from locales elsewhere in the northeastern Peloponnese but still within the Corinthian orbit of communication, travel, and exchange. Although the publication of mortuary, skeletal, and habitational remains of Roman and Byzantine date from this large region has been inconsistent, particularly in peripheral areas, the evidence from several major sites in addition to Corinth, such as Nemea and Argos, has been documented extensively. When compiled for comparison, the published record provides an adequate basis for beginning to outline general patterns of variation across the northeastern Peloponnese. Such a contextual approach to the graves, human remains, and settlement at the Isthmus sheds light on several essential developments in the transformation of Greek society from classical antiquity to the Byzantine middle ages. These include the demise of pagan cults and the evolution of Christianity, the implementation of Imperial defensive schemes involving massive fortification, the diverse effects of plague, earthquake, and invasion on local communities, and the changing face of rural economy and settlement. The Greeks who lived and died at the Isthmus during these centuries saw their world fundamentally change. Although their settlement diminished, they survived during the Byzantine “dark age.” Throughout this transitional era, they were hardly confined by rustic isolation, cultural stagnation, or material destitution. They continued to encounter frequent travelers and to interact with neighboring settlements. Their shifting lifeways and funerary rituals speak to a capacity for adaptation and innovation. While the overarching goal of this study is to investigate these historical developments, the line of inquiry has been organized in order best to record and to interpret two rather different types of evidence, graves and human remains. To this end, the book is divided into four sections. The first section, the present chapter, introduces the main themes, the procedures, and the materials under study. The second section, which is comprised of three chapters (Part I: Chaps. 2–4), addresses the graves within their archaeological, historical, and social contexts. Chapter 2 describes the physical remains of the graves, including their form, contents, and associations. Chapter 3 discusses the historical and topographical setting of these physical remains by surveying long-term changes in settlement. Chapter 4 reconstructs the funerary rituals that led to the creation of the graves and examines the relationship between those rituals and the social structure and ideology of the local community. The third section, which also comprises three chapters (Part II: Chaps. 5–7), addresses the human remains in their osteological and bioarchaeological contexts. Chapter 5 discusses the morphological and demographic composition of the skeletal sample, including sex, age at death, living stature, and metric and nonmetric traits. Chapter 6 examines teeth as evidence for oral development and health. Chapter 7 addresses the paleopathology of the bones and the paleoepidemiology of the skeletal sample. The fourth and final section (Chap. 8) summarizes the mortuary and skeletal evidence from the settlement at the Isthmus and evaluates its significance for understanding the transformation of the Greek countryside. Although the description and analysis of the two primary data sets requires divided treatment, the mortuary and osteological sections of this study are integrated, so that discussion in one often depends on information presented in the other. Therefore, the descriptions 7. Recent scholarship on center-periphery interactions in Late Antique Greece and Asia Minor (e.g., Mitchell 1993, Lavan 2001b) has demonstrated the importance of viewing ur-

ban and rural communities as dual components in regional networks of settlement, production, and exchange.

6

THEMES, PROCEDURES, AND MATERIALS

of individual graves (Chap. 2) include summaries of skeletal traits (Chaps. 5–7), and the discussion of funerary rituals, social structure, and ideology (Chap. 4) takes into account the age, sex, and genetic character of the dead (Chap. 5). Likewise, the investigation of oral health and paleopathology (Chaps. 6–7) refers back to a reconstruction of the local settlement and its living community (Chap. 4). This repetition of information and conclusions in separate sections is intended to aid the reader by consolidating relevant evidence and assembling a composite picture of life and death in a changing world.

PROCEDURE OF STUDY It is important to outline the procedure of study at the outset. The basic goal was to document the graves and human remains as accurately and precisely as possible using standard techniques in funerary archaeology and physical anthropology. The procedure was also designed in such a way as to derive information that would be pertinent to the central themes outlined above. Finally, measures were taken both to facilitate the comparative analysis of the graves and bones at the Isthmus with those from other sites and to permit ready access to the remains for future investigation. The study began with the collection of all materials and primary records in the storerooms and workspaces at Kyras Vrysi, the modern village on the ancient site. These include the artifacts and bones recovered from the graves and the field notes, reports, drawings, and photographs recording their excavation. Inventoried and lotted artifacts and textual and visual records had been shelved and catalogued in the usual manner. The bones had remained since excavation in the storeroom of the Isthmia Museum (Fig. 1.4), in some cases for nearly three decades, in wooden crates and trays with handwritten labels. It was evident from references to the graves in notes, reports, and labels that the excavators had not systematically identified them. Sometimes they were numbered sequentially by area (e.g., “Tower 14 #2”), sometimes they were designated by an associated feature (“Roman Bath drain skeletons”), and sometimes they were given no identification at all. When numbers were assigned, those numbers were often used in field notes, on storage boxes for lotted pottery and bones, on plans and detail drawings, and in artifactual and photographic inventories. The old grave numbers have also appeared in previous publications, namely, the annual field reports in Αρχαιολογικόν Δελτίον and the description of the fortifications in Isthmia V. It was necessary to devise a new system for identifying burials from across the site that not only preserves the old numbers but also uniquely designates single graves and skeletons. These new burial numbers consist of five digits separated after the first two by a hyphen and followed by a letter. The first two digits indicate the year of excavation, the second three indicate the old number, and the letter indicates a single skeleton. For example, 69-004C is the third of at least three skeletons (A–C) from a grave called “Grave #4” that was excavated in 1969. If the excavators had not assigned a grave number, or had repeated a number already used at another locale in the same year, the second three digits of the new number were arbitrarily assigned. Thus, for example, the four graves that were the first ones discovered in four separate areas during the 1969 campaign, two of which had been designated “Grave #1” and two of which were nameless, were renamed 69-001, 69-801, 69-901, and 69-991. The assignment of letters was determined by the order in which the bones from one grave were examined, not the order in which they were either interred or excavated. Bones in multiple burials that could not be associated with individual cranial or pelvic remains were designated collectively by the grave number followed by a question mark (e.g.,

PROCEDURE OF STUDY

7

Figure 1.4. Human remains from NEG 69-004, as stored by excavators

in 1969 and retrieved for study in 1990

69-901?).8 One important class of information the new numbering scheme does not encode is the locale of excavation. Therefore, abbreviated names for the areas where human remains were found have been prefixed to the burial numbers (e.g., RB 76-002B). The abbreviations for areas are given in Table 1.1. In addition to the new numbering scheme, the study adopted a consistent terminology for referring to mortuary practices and to periods in the history of the local settlement. “Grave” and “interment” refer to the physical presence of a space containing one or more bodies and other objects. “Burial” refers to either the intentional, often ritual, act of interment or the product of that act, that is, one or more dead individuals and any funerary items placed in a grave. Archaeologists and historians sometimes use the term “cist” to refer to an unprotected hole in which a burial is made (like Modern Greek λάκκος). In this study, “cist” designates the cutting for the interment, without implying the form of either the cutting or its enclosure. In retracing the history of the settlement and its burials, it was necessary to identify remains according to periods of habitation. This periodization is defined by major historical events, artifactual developments, and architectural phases at the Isthmus, which will be 8. In one unusually complex grave containing ten bodies (70-902), no bones could be identified with a single skeleton (pp. 85–86). Therefore, separate skeletal elements (e.g.,

ulnae, femora) were given randomly assigned serial numbers after an altered form of the standard burial number (e.g., 70-902Z3).

8

THEMES, PROCEDURES, AND MATERIALS

TABLE 1.1. ABBREVIATIONS FOR AREAS OF THE ARCHAEOLOGICAL SITE NEG

Northeast Gate

T2

Tower 2

T13

Tower 13

T14

Tower 14

DEC

Decauville Graves

LOU

Loukos Field

RB

Roman Bath

HO

Hexamilion Outworks

WF

West Foundation

PAL

Palaimonion

TC

Theater Cave

discussed in Chapter 3. The periodization of local settlement from Roman to modern times is given in Table 1.2. Two major eras in the history of the site are commonly identified by a phrase linking sequential periods. “Late Roman to Early Byzantine” designates the era of most intensive, continuous occupation of the fortifications from the years just preceding their construction in ca. 410–420 through their renovation under Justinian, the Slavic incursions of the late 6th to early 7th centuries, and the “dark age” occupation that ended in the late 7th or 8th century. In this study, “late antiquity” will be synonymous with the Late Roman period,9 while the term “Early Christian” has been eschewed.10 “Middle to Late Byzantine” will designate the second era of dense habitation at the Fortress, beginning with the revival of the local settlement in the 10th century, reaching its apogee during the 12th to 14th centuries, and declining during successive military operations in the 15th century. The next stage in the study was the close examination of the artifacts, bones, teeth, and primary records in order to establish the original state of the graves. This required a reconstruction of how the graves had changed since the time of interment, as discussed below. The study of the mortuary remains involved the collection and recording of all relevant depositional, artifactual, and structural data from the early excavations. It proved helpful to return to several burial sites for further exploration to clarify their form and associations.11 The study of the human remains was a two-part process. The first was an inventory of all skeletal elements and an evaluation of their state of preservation. The second was an examination of the remains in order to establish sex and age at death, to collect metric and nonmetric data, and to document paleopathology. During the examination, special attention was devoted to the handling of the human remains and to their long-term storage.12 Before inspection, many bones and teeth had to be cleaned. Several retained soil matrix in cavities and fissures, and a few were coated with 9. The elusive term “late antiquity” has had a varied usage in past and current scholarship; see Bowersock, Brown and Grabar (1999, pp. ix–x) for a useful formulation. 10. Although the designation “Early Christian period” is still used (e.g., Nemea III, p. xxx), the author considers it a misnomer for general historical phasing. It is chronologically incorrect (the early phase of Christianity was the 1st century), politically inaccurate (Imperial, military, and local administration was in many respects neither distinctly nor uniquely Christian), and culturally biased (Christianity was only one factor in the development of Late Antique Mediterranean

religion and culture). Furthermore, as a historical designation, it identifies one period by a single religion, even though that religion existed both earlier and later, during periods identified by ruling regimes (“Roman” and “Byzantine”). 11. NEG 69-103, 69-004, 69-001, 69-007, 69-009, 69-010, T2 68-003, T14 69-002, LOU 69-801, RB 76-002, HO 70-902, PAL 56-001. 12. Sease (1994, pp. 48–52), Bass (1995, pp. 331–338), and White (2000, pp. 292–300) provide useful overviews of the proper procedures for conserving and storing human remains.

PROCEDURE OF STUDY

9

TABLE 1.2. PERIODIZATION OF THE EARLY ROMAN TO EARLY MODERN SETTLEMENT Early Roman

Mid-1st to early 2nd centuries

Middle Roman

Mid-2nd to late 4th centuries

Late Roman

End of the 4th to mid-6th centuries

Early Byzantine

Late 6th to 7th or 8th centuries

Middle Byzantine

10th to 12th centuries

Late Byzantine

13th to 15th centuries

Ottoman

16th to early 19th centuries

Early Modern

Mid-19th to early 20th centuries

a calcareous shell from environmental moisture.13 Once the bones and teeth were clean, they were conserved as needed. None was so friable that it required chemical consolidation. Many, however, were broken and marred by postmortem degradation in their depositional context, by damage during excavation, by crowded and unpadded packing in boxes and bags, and by occasional handling. Furthermore, during the 1970s and 1980s some fragmentary bones had been rejoined using masking tape, which caused staining and exfoliation, and several postmortem breaks had been mended with Elmer’s Glue, a poly(vinyl) acetate emulsion which discolored the bones and stripped contiguous surfaces after only a few years in storage.14 To rectify these earlier measures, the masking tape and white glue were carefully removed and selectively replaced with Acryloid B-72 in a 10% acetone solution.15 Since its initial application in 1995 and 1996, Acryloid has proven to be an optimal adhesive. The cleaned and restored remains were treated for clearer identification and better preservation. Burial numbers were written on all major bones, all skeletal elements that could be identified with individuals were segregated, and all elements from the same skeleton were bagged by type or anatomical region. Some bones had been wrapped by the excavators or their successors in newspaper or toilet tissue. These products had served well for cushioning in the short term, but in places the newsprint had bled onto the bones and the tissue had stuck to irregular surfaces, leaving tiny shreds in narrow crevices and cancellous tissue. These materials were removed and selectively replaced with Ethafoam, a polyethelene packing medium that is pliable, acid-free, and shock-absorbent.16 Finally, the remains were assigned by grave to numbered crates with the contents clearly marked. Rectangular wooden crates were chosen because they are sturdy, easy to lift and transport, can be repaired at minimal cost, and stack tightly but do not completely seal off the bones. They permit the circulation of air but prevent intrusion by the common denizens of the apothiki, lizards and rats. One of the most important components in the study of the graves and human remains was their visual documentation in drawings, photographs, and digital images. Drawing and photography are essential to archaeological research because they preserve visual data of a 13. These elements were washed in warm water without additives and dried in a shaded, protected location away from direct sunlight and blowing sediment. Dental instruments and soft-bristle brushes, both dry and moist, aided finer cleaning. Lime concretions were removed from bones using a brush dipped in a 5% acetone solution of hydrochloric acid and from teeth using wooden picks and a sparing wash in a 15% acetone solution of acetic acid. Special care was taken so as not to gouge the cortex or enamel, to erase pathological features, or otherwise to alter osseous or dental surfaces.

14. Koob (1981), Selwitz (1988), Johnson (1994, pp. 225– 226), and Sease (1994, p. 12) evaluate the utility of various adhesives in archaeological conservation, including PVAs (UHU) and cellulose nitrate resins (Duco, Durofix, UHU Hart). 15. Acryloid is a thermoplastic acrylic resin that is recommended for its long-term stability, transparency, penetration, durability, and reversibility (Brothwell 1981, p. 10; Koob 1986; Johnson 1994, p. 227). 16. Sease 1994, pp. 14–15.

10

THEMES, PROCEDURES, AND MATERIALS

contextual or material nature and allow for independent evaluation and comparative study. While traditional methods of photography have served these purposes well,17 digital technology offers a more versatile medium for recording and sharing visual data.18 In the study of the mortuary and skeletal remains at the Isthmus we have implemented new techniques for capturing, manipulating, and storing visual information bridge the transition from traditional photography to digital imaging. The visual documentation of the material was a three-stage process beginning with the initial drawing and photography of graves and funerary artifacts. The excavators in the 1950s to 1970s always plotted graves on actual-state plans but seldom prepared formal drawings of them.19 They always photographed the interments, which preserved crucial information, but procedures varied.20 The excavators stored the human remains without study or photography but photographed funerary artifacts on 35 mm or medium format black-and-white film using a copy stand indoors.21 Many of these inventory photographs were publishable with only slight modifications. The excavators stored all negatives and prints in filing drawers and albums on site. Over the years many negatives have been destroyed or lost, but in each case a high-quality print from the damaged or lost negative has survived. The second stage in the visual documentation of the mortuary and skeletal remains was conducted by Daniel M. Curtis over roughly six weeks in Kyras Vrysi during the summers of 1996, 1998, and 2000. His goal was to complete the photographic record for publication and archival storage. He visited several burial sites to capture features that had not been previously documented and shot new photographs of several funerary artifacts. Most of his efforts were devoted to photographing the human remains for the first time.22 He photographed the bones on Kodak TMAX ISO 100 and Ilford FP4 ISO 125 black-and-white film with a 35 mm SLR camera (Nikon 8008S or Canon A2) with a 50 mm lens for macro-focusing. The camera was mounted on a Hama 6229 copy stand with four rotatable 100-watt tungsten bulbs. The subject was placed on a plate of nonglare glass suspended 5.5 cm over a sheet of black cloth. The direction of lighting and orientation of the subject followed scientific convention.23 Due to financial and temporal constraints, Mr. Curtis did not compile a visual record of the total skeletal assemblage24 but did photograph many more elements than appear in this volume. Elements were chosen in order to create a primary visual archive of the bones and teeth and to illustrate published discussions of those remains.25 The third and final stage in the visual documentation of the mortuary and skeletal remains was completed by Curtis in the United States between 1998 and 2002.26 This involved the digitization of all photographs, the manipulation of those digital images, and provision for the long-term preservation of the visual record. All black-and-white negatives pictur17. Dorrell (1994), Howell and Blanc (1995), and Roskams (2001, pp. 119–132) offer useful introductions to archaeological photography. 18. Dorrell 1994, pp. 254–255; Besser 2003. 19. Dillon and Verano (1985) outlines the proper procedure for drawing graves. 20. Several photographs show the graves after the displacement or removal of their walls or bones: Figures 2.6, 2.46, 2.50, 2.54, 2.59. The correct procedure for photographing graves is discussed in several manuals: Dillon and Verano 1985, pp. 145–146; Dorrell 1994, pp. 132–133; Ubelaker 1999, p. 14; White 2000, pp. 284–286 (photography by Pieter Arend Folkens); Roskams 2001, pp. 130–131, pl. 23. 21. Howell and Blanc (1995, pp. 75–84) outline the proper procedure for studio photography. 22. Apart from the work of Curtis, Andrew Reinhard photographed the fragmentary long bones in Fig. 2.9, and John Robb photographed the pathological specimens in

Figs. 7.1, 7.10, and 7.26. 23. Buikstra and Ubelaker (1994, pp. 10–12), Hillson (1996, pp. 305–306), and White (2000, pp. 309–312, 517–519) outline these procedures for the photography of archaeological bones and teeth. 24. See the recommendations of Buikstra and Ubelaker 1994, pp. 10–11. 25. The following photographs were taken: all adult skulls, regardless of preservation, in the anterior, lateral, and posterior views; all upper and lower adult dentitions in the occlusal view; noteworthy details of the teeth, such as severe caries, attrition, or dental trauma; pubic symphyses and auricular surfaces, which are used to estimate age at death; all congenital defects, infectious lesions, trauma, and neoplasia; a representative sample of cases of cribra orbitalia and joint disease; and examples of postdepositional alteration to bones. 26. Besser 2003 is an introduction to digital image capture and digital asset management.

THE CREATION OF THE MATERIAL RECORD

11

ing the graves, artifacts, and human remains were scanned to Kodak Photo CD and Kodak Pro Photo CD master disks, which served as an excellent storage medium in terms of affordability and quality. All preserved line drawings of graves and artifacts, as well as the prints of negatives that had been destroyed or lost, were scanned at high resolution (800 dpi for drawings; 300 dpi for prints). Once all hard copies of visual data had been translated into 8-bit grayscale images, they were adjusted using Adobe Photoshop (versions 4.0 to 7.0) on a Macintosh computer. This involved the application of standard tools, the regularization of sizes, and the insertion of scales and arrows. These modifications have not distorted or erased original visual information essential to the subject; such adjustments are the same as those made during the traditional printing and retouching of photographs and the drafting and copying of drawings. After preparing the digital images, Curtis archived the visual record in a way that would promote long-term stability and easy accessibility. The original negatives, prints, and drawings were catalogued, cleaned, relabeled, and returned to the site in improved storage spaces.

THE CREATION OF THE MATERIAL RECORD Once these procedures had been established, it was possible to investigate the graves, artifacts, bones, and teeth as evidence for funerary rituals and skeletal biology. The gradual transformation of graves and bodies over time after death and burial can filter or even erase traces of both the osteological character of the deceased individual and the activities surrounding that individual’s death. The biological interpretation of human remains requires the assessment of the diachronic impact of natural and cultural agents.27 The osteologist must distinguish postmortem changes from ante- or perimortem disease and trauma.28 Moreover, understanding the state of preservation is necessary in order to examine the distribution and prevalence of pathological conditions in a skeletal sample.29 It is also important to determine the depositional and postdepositional pathways, or formation processes, by which the material components of mortuary behavior have moved from a cultural to an archaeological context.30 Reconstructions of burial chronology and settlement history, along with interpretations of funerary custom and social identity, rely on the evaluation of formation processes. The natural environment and human activity were primary factors in the creation of the osteological and mortuary records at the Isthmus. These factors are summarized in Table 1.3. The calculation of the relative survival and preservation of the bones (Table 1.3, right two columns) will be discussed in Chapter 5. The lithologic sequence and natural topography of the site of the Isthmian Sanctuary and Fortress constitute the depositional environment of the Roman and Byzantine graves.31 A bed of fine-grained, compact but soft marl is exposed on the north side of the Fortress and outside the circuit in the defensive ditch. A coarse limestone conglomerate bed overlying the marl crops out on the north edge of the enceinte. The soil overlying the conglomerate 27. Taphonomy is the study of the postmortem change of biomatter, especially bone and tooth, within a particular environment. White (2000, pp. 407–424) and Mays (1998, pp. 15–37) offer useful introductions; Lyman (1994) and Martin (1999) present definitive surveys of taphonomy; Haglund and Sorg (1997a) provide a collection of introductory essays on forensic taphonomy. The author thanks Thomas Tartaron and Raymond Rogers for informative discussions about taphonomy in general and depositional conditions at the Isthmus in particular. 28. Wells 1967 is a classic essay on “pseudopathology”; see

more recently Ubelaker 1991; Aufderheide and RodríguezMartín 1998, pp. 15–18; and Lovell 2000, p. 236. 29. Walker, Johnson, and Lambert 1988 (paleodemography); Waldron 1987; 1994, pp. 10–27 (paleoepidemiology). 30. O’Shea (1984, pp. 23–31) discusses factors affecting burial sites; Schiffer (1987) gives a basic survey of formation processes. 31. On local geology, see Isthmia VIII, pp. 3–4; on local topography, see Isthmia II, pp. 1–3; Isthmia V, p. 9; Isthmia VIII, p. 15. The author thanks Jay Noller for discussing the geomorphology of the Isthmus.

12

THEMES, PROCEDURES, AND MATERIALS

TABLE 1.3. FACTORS IN THE PRESERVATION OF THE GRAVES AND HUMAN REMAINS Burial Design

Environmental Factors

Area and Grave No.

No.

Covering

Lining

Depositional Matrix

Drainage Zone

Presence of Snails

Floriturbation

Sedimentary GraviturCompression1 bation2

NEG 69-103

2

Complete

None

Soil

Y

Y

N

N

Y

NEG 69-004

5

Complete

Partial?

Soil

Y

N

N

Y

N

NEG 69-008

1?

Partial?

Partial?

Soil

Y

Y

N

N

N

NEG 67-001

11

None

Complete

Soil

Y

N

Y

N

Y

NEG 67-003

5

None

Partial

Soil

Y

N

N

N

N

NEG 69-005

3

None

Partial

Soil

Y

Y

N

N

N

NEG 69-001

4

Complete

Complete

Soil

Y

N

N

N

Y

NEG 69-007

2

Complete

None

Marl

Y

Y

Y

N

N

NEG 69-009

1

Complete

None

Marl

Y

N

N

N

Y

NEG 69-010

1

Complete

None

Marl

Y

N

N

N

Y

T2 68-006

1

Complete

Complete

Soil

Y

N

Y

N

N

T2 68-002

1

Partial

Complete

Soil

Y

N

N

Y

Y

T2 68-003

2

Complete

Complete

Conglomerate

Y

N

Y

N

N

T13 54-001

1

Partial

Partial

Soil

N

?

?

?

?

T14 67-002

1

Partial

Partial

Soil

N

N

N

Y

Y

T14 67-004

1

None

None

Soil

N

N

N

N

N

T14 69-701

1

Partial?

Partial?

Soil

N

N

N

N

N

T14 69-991

1

Partial

Complete

Soil

N

N

N

N

Y

T14 69-002

1

Partial

None

Soil

N

N

N

N

Y

T14 69-003

1

Partial

None

Soil

N

N

Y

Y

Y

LOU 69-801

1

None

None

Soil

N

N

Y

N

N

DEC 69-901

2

Partial

Partial

Soil

Y

Y

Y

N

N

DEC 69-902

1?

Complete

None

Soil

Y

N

Y

Y

N

HO 70-901

3

None

None

Soil

Y

N

N

N

N

HO 70-902

10

Partial

Partial

Marl

Y

N

Y

Y

Y

RB 76-002

3

Complete

Complete

Stone-lined drain

Y

N

N

N

Y

WF 62-001

1

Partial

Partial

Soil

N

N

N

Y

N

N

N

N

N

Y

N

N

N

N

N

PAL 56-001

1

Complete

Complete

Concrete foundation

TC 60-001

1

None

None

Soil in marl cave

This includes only those graves in which damage to bones, artifacts, and graves could be confidently attributed to compression. Numerous teeth and ribs were also affected by sedimentary compression. 2 This includes only those graves in which bones or artifacts had clearly moved after deposition. Gravity affected all graves to a certain degree. 3 This only includes surficial damage that can be confidently identified as having occurred during excavation. 4 This includes only those graves from which bones were recovered after their initial excavation. 1

is a fine, clayey to sandy matrix containing a fraction of calcium carbonate and decomposed organic matter, with a neutral to slightly alkaline pH (7–8). The local terrain is dominated by the Rachi, a prominent ridge lying just south of the Temple of Poseidon, and by the Great Ravine, which borders the site on the north and defines the course of the Hexamilion. Three roughly parallel tributaries cut across the site from the high ground in the south toward the Great Ravine (Fig. 1.2). The drainage of the Sanctuary was facilitated not only by

THE CREATION OF THE MATERIAL RECORD

13

Anthropogenic Factors Disturbance Reuse for During Additional Construction Burial

Primary or Secondary Burial

Bones Removed after Reopening

Excavation Damage to Bones3

Intended Par- Unintended Survival Preservation tial Recovery Partial Recov- Rate of Bones Rate of Bones of Bones ery of Bones4 (%) (%)

Y

Y

Primary

N

N

N

N

65.02

40.63

Y

Y

Primary

Y

N

N

N

45.49

29.94

Y

Y

Primary

?

N

N

Y

0.40

0.21

N

Y

Primary

N

N

N

N

44.28

30.72

N

Y

Primary

Y

N

N

N

25.41

13.80

N

Y

Primary

N

Y

N

N

47.42

37.45

N

Y

Primary

N

Y

N

N

60.41

43.72

N

Y

Primary

N

N

N

N

38.63

34.90

N

N

Primary

N

N

N

N

0.00

0.00

N

N

Primary

N

N

N

N

0.00

0.00

N

N

Primary

N

Y

N

N

63.95

59.87

Y

Y

Primary

N

Y

N

N

41.85

32.94

N

N

Primary

N

Y

N

N

68.67

60.19

Y

N

Primary

?

?

Y

N

0.00

0.00

Y

N

Primary

N

N

N

N

88.84

85.62

Y

N

Primary

Y

N

N

N

90.13

89.70

?

N

Primary

?

N

Y

Y

12.02

10.52

Y?

N

Primary

N

N

N

N

32.19

20.17

Y

N

Primary

Y

N

N

N

69.10

67.06

Y?

N

Primary

N

N

N

N

78.97

74.03

N

N

Secondary?

N

N

N

Y

45.90

38.95

N

Y?

Primary

N

N

N

N

10.94

8.21

N

Y

Primary

N

N

N

N

0.00

0.00

?

Y?

Secondary

Y

N

N

N

24.89

19.81

N

Y

Primary

N

Y

N

Y

46.48

22.09

N

Y

Primary

N

Y

N

N

24.18

14.45

N

N

Primary

N

N

N

N

34.33

26.61

N

N

Primary

N

Y

N

N

0.00

0.00

N

N

Secondary

Y

Y

Y

N

7.73

7.73

deep gullies but also by artificial drains, several of which passed through the fortifications near the Northeast Gate and Tower 2.32 Once these channels were no longer maintained, run-off collected at the base of the ramparts and in the steep taphros (defensive ditch), eventually burying them under accumulating sediment. Microvariations in the local landscape generated small differences in the preservation of human remains. A few cists were cut into the marl and conglomerate, but most were cut into the soil above the conglomerate (Table 1.3). In general, this slightly alkaline, 32. Clement 1969, p. 142 (North Drain); Isthmia V, pp. 47, 48, 81, 116, figs. 8, 16, pls. 11:a–d, 14:a, 22:c, 42:b (drainage near the Northeast Gate and Tower 2).

14

THEMES, PROCEDURES, AND MATERIALS

well-drained soil and the semiarid, temperate climate favored the preservation of skeletal material.33 However, interment in calcareous strata promoted infiltration by soil water and the precipitation of salts; moisture then dissolved and leached out the organic and mineral constituents of hard tissues. As a result, surviving bones and teeth are sometimes fragile and display a thin encrustation or discoloration.34 A secure covering and lining in a grave could mitigate, but not prevent, the deleterious effects of these natural agents on the remains inside. It is, therefore, not surprising that the decomposition of hard tissues was most rapid in drainage zones, especially in partly or completely uncovered graves on the north side of the Fortress.35 Bones and teeth in cists that were poorly sealed and exposed to draining water have completely or almost completely disappeared.36 In contrast, the best-preserved skeletons were found in mostly sealed graves in the well-drained area on the west side of the Fortress. The flora and fauna of the Isthmus had a negligible impact on the preservation of human remains. In most areas of excavation, vegetation is grassy with shallow roots, but bushes and trees grow here and there, especially alongside buried or exposed architecture where water tends to collect. Roots of varying sizes had penetrated cists that were especially shallow and unshielded (Table 1.3). Invading roots sometimes marked the surface of bones, displaced them, or broke them down. There are scattered traces of burrowing rodents at the Isthmus, but earthworms and molluscs are frequent in the moister areas on the north edge of the site, in the gullies, and near the marl exposures. The most common terrestrial snails in cists exposed to moisture were typical Mediterranean species of turriculate and discoid gastropods (Rumina decollata and Eobania vermiculata; Table 1.3). Since these snails are not sarcophagous, they must have moved into the graves well after the decomposition of soft tissues, either burrowing to estivate or entering through worm action and erosional crevices. Although there is no sign that molluscs directly destroyed the human bodies, their presence implies that other factors, such as air, water, organic matter, and shallow depth, affected the burial environment.37 Natural processes also influenced the preservation of graves at the Isthmus. Gravity had a long-term effect on the position of the bodies after interment and skeletonization.38 It was not uncommon for bones to shift over time into unnatural positions once soft tissues had dissolved. Jaws that had been closed or tied shut for burial sometimes dropped open with the decay of the temporomandibular ligaments (see Fig. 2.56).39 The small bones of hands that had been placed over the chest, abdomen, or pelvis of the deceased often migrated to the floor of the cist with the decay of soft tissues (see Fig. 2.47). Chemical decay, sedimentary compression, and gravity determined the preservation and location of the funerary artifacts. The condition of these objects at the time of discovery depended on several factors, including material, soil chemistry and composition, and moisture.40 Nearly all iron objects are severely corroded (see Figs. 2.100, 2.104, 2.108–2.110, 2.112), and bronze objects often display green corrosion or have disintegrated (see Figs. 2.98, 2.113), while objects in gold, 33. See Henderson 1987, pp. 46–47; Lyman 1994, pp. 418– 419, 421–422; Janaway 1996, p. 69; Martin 1999, pp. 152–155. On soil as a burial environment in general, see Janaway 1996, pp. 58–63. 34. E.g., DEC 69-901B, HO 70-902 (adult teeth unidentified to individual). In both cases, the staining may have arisen as much from the decay of contiguous roots as from minerals in soil water. 35. Bones: NEG 67-003D-E, 69-007A, B, 69-009A, 69-010A, T2 68-002A, 68-003B, DEC 69-901A, B. Teeth: NEG 67-001K, 69-005C, 69-007A, T2 68-003A, B, DEC 69-901A, B.

36. NEG 69-008, 69-009, 69-010, DEC 69-902. 37. The author thanks Henk K. Mienis, curator of the National Mollusc Collections at the Hebrew University of Jerusalem and Tel Aviv University, for identifying snail specimens and discussing their ecology. 38. On the effects of gravity at archaeological sites, see Wood and Johnson 1978, pp. 346–352 and Schiffer 1987, p. 216. 39. NEG 69-103B, T2 68-002A, T14 67-002A, 69-003A. 40. Janaway (1996, pp. 78–81) surveys the decay of artifacts in burial environments.

THE CREATION OF THE MATERIAL RECORD

15

lead, stone, bone, and shell remain intact (see Figs. 2.101–2.103, 2.107, 2.111, 2.114). Terracotta lamps and vessels that had been buried intact were found in near perfect condition (see Figs. 2.88, 2.89, 2.91–2.94), but the condition of glass varies dramatically (see Figs. 2.7, 2.90). Textiles played an integral part in the funerary ritual, but did not survive to the time of the grave’s discovery. Not unlike bones and teeth, small, round artifacts such as buckles and beads sometimes moved from their original positions. Artifacts that had been worn on the torsos of the dead during the funeral and deposition, such as necklaces and pendants, remained over the chest until the skin and internal organs had dissolved, at which point they passed through the thoracic cavity and settled near the scapulae or vertebrae.41 In three graves, earrings were found where they had come to rest after the earlobes to which they were once attached had disappeared (see Figs. 2.21, 2.80, 2.81). Environmental agents did not significantly alter the form of the interments. Lined and covered cists typically incorporated durable materials in sturdy designs. One exception was the disintegration of the coffin or bier in the grave cut into the Palaimonion (56-001), which left behind only splinters and nails. Gravity had little effect on the form of the graves. Covering and lining elements shifted most when they were not supported by soil, stones, or mortar around the edges. The depth of sedimentation over the graves, especially those on the north side of the site, tended to enhance stability by encasing them. In some cases, however, the overburden displaced, cracked, or compressed coverings, particularly those using large flat tiles (see Fig. 2.62), which could crush bones.42 Downslope movement and soil creep did not affect the graves because most were interred in static deposits.43 Burrowing animals did not disrupt the graves, but tree roots dislodged the stones sealing the slabs over one grave (see Fig. 2.42), and smaller roots cracked and slightly displaced the tiles over another (see Figs. 2.62, 2.63). Successive phases of human activity in earlier burial areas significantly affected the condition of bones, teeth, and graves at the Isthmus (Table 1.3). During the early 5th century, the builders of the Hexamilion interrupted earlier graves south of the Northeast Gate and inside Tower 2. Workers attempted to preserve these interments by setting the foundation blocks around them, but during the excavation of the footing trench they could not avoid opening graves and often altering their form and contents (see Figs. 2.7–2.9). During the initial occupation of the Fortress in the early 5th to mid-6th centuries, local residents commonly added bodies to preexisting graves rather than digging fresh cists. This practice produced several multiple interments with as many as 10 or 11 bodies. Continued inhumation in single graves is evident in the purposeful modifications to the designs of the graves and the relative positions of the skeletons (see Figs. 2.12, 2.16, 2.18–2.20, 2.66, 2.70). These activities affected not only the situation but also the survival and preservation of the human remains. The displacement of bones led to breakage, the disruption of the original grave design often exposed bones to weathering, and occasionally bones were removed. Local residents also practiced secondary burial, the removal of human remains from one place and their redeposition in another, during late antiquity. The transplanting of remains led to a lack of protection over the new grave, an unconventional orientation of burial, the dense collection of bones, and the incomplete representation of skeletons. The process of secondary burial caused the loss of numerous small bones and damage to those that remained. 41. NEG 69-009, 69-010, T14 69-002, 69-991, RB 76-002, HO 70-902, PAL 56-001. Note, however, that the pocket of coins in NEG 67-001 was discovered somewhat higher in the cist, closer to where it had originally been placed (Figs. 2.12, 2.13). 42. E.g., T2 68-002A, T14 67-002A, 69-003A, WF 62-001A. Lyman (1994, pp. 423–424) and Aufderheide and Rodríguez-

Martín (1998, p. 17) discuss this process. 43. The surface markers above NEG 69-009, 69-010, T2 68002, and 68-006 were found in their original state, even though the first two had been erected on an incline (Figs. 2.27, 2.28, 2.31–2.33, 2.37, 2.40).

16

THEMES, PROCEDURES, AND MATERIALS

Construction and secondary burial also affected earlier burials during later phases in the occupation of the site (Table 1.3). The erection of small houses near Tower 14 during the Late Byzantine era cut into underlying Early Byzantine cists. As in earlier times, these building activities altered the design of graves and led to both the partial destruction and the selective extraction of skeletal remains (see Figs. 2.45, 2.46, 2.49, 2.54). The two latest graves from the site also reflect unusual burial conditions.44 In the first case, the skeleton might have been moved to the cist from another locale before the decomposition of the ligaments, and the extended legs were folded back over the chest (see Figs. 2.59, 2.60). In the second case, the body was already skeletonized when only the skull and a few other bones were redeposited in a cave above the ancient Theater, together with various animal bones (see Fig. 2.83). The modern residents of Kyras Vrysi have had no significant impact on the graves. Most interments are situated northeast of the village, buried well below modern ground level and protected within the archaeological site. However, olive cultivation south of the Fortress and bulldozing northeast of Tower 2 has disturbed the tops of three graves.45 Another important anthropogenic factor in the preservation of the mortuary and skeletal remains at the Isthmus was the method of excavation and recovery. The decisions and actions of archaeologists determined the data that were collected from burial contexts.46 The early excavators were unusually systematic and meticulous for the discipline of classical archaeology during the 1950s–1970s, but of course they did not attain the high standards set by archaeologists and anthropologists today.47 They had no expertise in the documentation and analysis of mortuary or osteological evidence and did not fully comprehend its historical and biological importance. One important class of evidence that did not receive adequate attention was stratigraphic context. The excavators recorded depositional units, but in many cases had difficulty defining graves because the walls of the cists were indistinct at the top. Depositional sequences in most of the burial areas can be retraced with confidence from a combination of elevations, written accounts, photographs, and artifacts. The lots from the early campaigns typically contained representative samples of finds from discrete excavated units, but excavators retained on average more material from graves than from other contexts. Occasionally they discarded small finds,48 but in most cases they saved well over half the sherds, and in some cases a total sample.49 Moreover, they almost always maintained the integrity of burial deposits, even when digging at the Northeast Gate and Tower 14, where habitational remains were dense and the risk of contamination high. The return to selected burial sites for inspection or cleaning in recent years has resulted in the recovery of crucial information concerning the stratigraphic contexts and designs of graves. In recording graves and saving their skeletal contents, the excavators were veritable pioneers during a time when their colleagues at other classical sites usually neglected burials of Roman and Byzantine date and discarded human bones. Their method of excavating graves involved opening the cist, clearing the fill to the level of the skeleton, and removing the bones. Frequent references in the notebooks show that sieving was a regular practice.50 But the excavators did not always recover and store the bones from individual skeletons sepa44. LOU 69-801A, TC 60-001A. 45. LOU 69-801, DEC 69-901, 69-902. 46. Schiffer (1987, pp. 339–364) discusses the practice of archaeology as a formation process. O’Shea (1984, p. 26), Haglund and Sorg (1997b, p. 20), and Mays (1998, pp. 14–15, 22–23) note the impact of investigative method on mortuary, osteological, and forensic evidence. 47. Most handbooks review techniques for excavating graves and recovering human remains: e.g., Parker Pearson 1999, pp. 198–204; Ubelaker 1999, pp. 3–38; White 2000, pp. 279–291. Hunter and Dockrill (1996, pp. 40–53) and

Dirkmaat and Adovasio (1997, pp. 39–50) provide useful introductions from the viewpoint of forensic archaeology. 48. NEG 69-004 (nail), 69-005 (nails), 67-001 (ring), T14 69-991 (beads?), 69-002 (pins), RB 76-002 (leather or textile), PAL 56-001 (nails, beads). 49. E.g., lot 67-T14-071 from T14 67-002 (see NB 1967 RP II, p. 171). 50. E.g., NB NEG I, p. 127 (69-004); NB NEG II, p. 45 (69-005); NB NEG III, p. 67 (69-001); NB NEG IV, p. 49 (69-007); NB JN-LO II, p. 103 (T2 68-003); NB T14 III, p. 141 (69-002).

THE CREATION OF THE MATERIAL RECORD

17

rately, particularly when they came from complex multiple interments (Fig. 1.4). Sometimes they did not recover all of the bones that survived in the graves, either out of disinterest or on practical grounds (Table 1.3). More frequently, they missed bones because they simply did not recognize them. Their main goal in excavating the total fill in each grave was to recover small artifacts, such as coins and jewelry, not diminutive or irregular bones, which are easily mistaken for rocks by the inexperienced. Some postmortem damage to the bones can be attributed to excavation and handling (Table 1.3). A comparison of the photographs of bones in situ to those of bones in the laboratory reveals that certain elements were drastically broken at some point after discovery. In one instance, a partially exhumed skeleton was left exposed over the Easter holiday and the skull was crushed, purportedly by vandals (see Figs. 2.46, 2.47).51 Certain fragile bones were inadvertantly destroyed during recovery. Numerous crania exhibit damage caused by excavation and storage. Some have perforations and gouges with clean, sharp edges and no soil adhering, which indicate forceful contact by digging implements.52 The relatively high frequency of such damage on the crania can be variously explained. First, this part of an extended and supine skeleton often rests at the highest elevation in the grave and is therefore the first element encountered when digging. Second, the excavators at the Isthmus transported and stored skulls in square or rectangular boxes, crates, and trays that did not always accommodate their form as tall, thin-shelled, spheroid structures. This led to the abrasion or compression of the bones when containers were stacked and shelved. Finally, skulls are handled more often than other bones during excavation because, unlike any other element in the human skeleton, they fascinate the imagination and signify the mortal condition. One cranium that could be reconstructed from around 60 pieces had perhaps at some point shattered when it was accidentally dropped (see Figs. 2.83, 5.40). The cumulative impact of environmental and anthropogenic factors on the creation of the material record at the Isthmus was not so profound that meaningful mortuary and osteological evidence was irretrievable. The varied remains that have been documented provide a broad basis for reconstructing funerary ritual and skeletal biology, as long as the effects of formation processes are taken into account. It is impossible to know how accurate such reconstructions are without excavating all graves and recovering all bones and teeth that exist at the site. Numerous graves belonging to the settlement around the Roman Sanctuary, the Late Roman fortifications, and the Byzantine community have yet to be discovered. The paucity of graves from the Middle Roman and Late Byzantine periods undoubtedly results from the selective nature of exploration. Archaeologists have not systematically investigated much of the area on the periphery of the Sanctuary where Roman graves are to be expected, or inside the Fortress where Byzantine graves are to be expected. Although the number of graves so far discovered at the Isthmus is small compared to other major Corinthian cemeteries, these 30 graves and their contents represent the mortuary interests and skeletal biology of over twice as many individuals, which is a significant total. The compiled mortuary and osteological evidence for the local settlement and its inhabitants contributes greatly to our understanding of the Greek countryside during a pivotal era in its history. 51. Another mishap at Easter two years later crushed the fragmentary cranium (69-999A) found in a secondary context at the Northeast Gate (NB NEG 1969 I, p. 123).

52. For a more specific discussion of such damage, see p. 239.

PART I THE ARCHAEOLOGICAL, HISTORICAL, AND SOCIAL CONTEXT OF THE GRAVES

2

 THE GRAVES AND ASSOCIATED REMAINS

T

his chapter surveys the graves and their associated remains. As has been previously discussed, various natural and human agents have shaped the condition of the graves since their creation. Although the mortuary remains at the Isthmus are not perfectly preserved, they are sufficiently intact to furnish substantial evidence for local topography and history, funerary rituals, and the social, economic, and religious character of the resident community over a period of several centuries. The first section of this chapter discusses the burial sites as the archaeological dimensions of mortuary behavior. The second discusses the techniques for dating the burials. The third describes the graves by area of excavation. The fourth catalogues the burial deposits, or the contents of the strata associated with each grave. The fifth catalogues the funerary artifacts, those objects purposefully deposited at the grave during or after the interment. This comprehensive account of the remains will establish the temporal and spatial contexts of burial. These contexts are crucial for understanding the development of the settlement within an evolving landscape (Chap. 3). Once the topographical and historical setting of the graves has been investigated, mortuary variability will be discussed in reference to the funerary rituals, social structure, and ideology of the local residents (Chap. 4).

THE ARCHAEOLOGICAL DIMENSIONS OF MORTUAR Y BEHAVIOR The graves and their associated remains constitute the archaeological dimensions of mortuary behavior. When people dispose of their dead, they conduct a series of ordered activities within a particular physical setting. In doing so, they often leave behind traces of these actions that can be recovered at a later date through careful investigation. Funerary ritual, though suffused with human emotion, is performed with intentional and decisive effort. The materials and spaces employed in ritual are imbued with complex meanings that express, amplify, or even contradict the concerns and interests of the living community. Archaeologists must document these physical remains completely and systematically to recover as much information as possible and to compile a record that is amenable to comparative study through the use of standard forms of description.1 The physical manifestations of funerary ritual can be classified broadly as formal, corporeal, artifactual, and locational. The description given for each of the graves will document the following features from among these four classes:2 the material, design, and 1. Sprague (2005) presents an excellent survey of the terminology of funerary archaeology. 2. The archaeological theory of mortuary behavior has recognized similar discriminatory categories: see Binford 1971,

p. 21; Goldstein 1981, p. 59; O’Shea 1984, pp. 39–41, table 3.2; Carr 1995, pp. 129–132, table III; Parker Pearson 1999, pp. 5–17.

22

THE GRAVES AND ASSOCIATED REMAINS

size3 of the interment (formal); the number of bodies buried, their orientation,4 and the position of their heads, torsos, and limbs5 (corporeal); the identification and location of artifacts associated with the grave (artifactual); and the spatial relationship between the grave and features in the vicinity, including other graves, architecture, and the natural terrain (locational). The archaeological dimensions of mortuary behavior are summarized in Table 4.1. In recording these physical conditions of burial, it will be necessary to account for how they have changed since deposition on account of environmental and anthropogenic processes. The data presented in this chapter will form the basis for reconstructing the history and topography of local settlement in Chapter 3 and for exploring ritual, society, and ideology in Chapter 4. Two other fundamental determinants of funerary activity must be considered: biology and time. The circumstances of these factors at death influenced the repertoire of activities chosen and enacted by the bereaved. The most important biological factors are sex and age; the structure and operation of all societies respond to the essential dichotomies of man-woman and adult-child. Abnormal conditions such as congenital anomaly or traumatic injury leading to physical deformity can also influence activities surrounding death and burial, inasmuch as they contribute to the social position of the deceased. At the same time, subsistence strategies, religious affiliation, dress, occupation, and status, all of which can be communicated through the treatment of the dead, may impact personal health. Therefore, the description of the graves will include a brief summary of the sex, age, and pathological conditions of each individual (see Tables 5.3 and 7.1). The osteological and bioarchaeological aspects of the skeletal sample will be investigated in Part II (Chaps. 5–7). Time is an abstract condition that informs mortuary behavior.6 When someone dies, the mourners perceive time on multiple levels. They decide whether the individual died prematurely, on time, or too late based on emotionally and culturally defined notions of when is the right time to die. The designation of the time of death can also vary, from a single moment, such as when the heart expires, to a series of moments, from the death of the body to the departure of the soul to the decay of the body. A certain amount of time elapses between the stages in the ritual process from death to the funeral, the deposition, and the graveside activities following burial. The death and funeral can also be defined temporally by their place on an extended scale, such as a season or a term of political or religious significance, and by their relation to important events in the life of the family and community, such as a wedding or a festival. If the individual is buried in a communal context, such as a cemetery, he or she enters into a succession of death that is mapped in space. Graves are sometimes opened and modifed after the primary burial either purposefully or inadvertantly. Moreover, the dead transcend ordinary natural time and enter a state that is preternaturally bounded. This may involve regeneration into another life cycle, passage into a timeless plane, or initiation into a separate history. Sometimes the dead are believed to move through stages of existence that can be marked on a calendar. Ritualized interactions between the living and the dead at the event of the burial and during later commemorative activities are often shaped by such prevailing notions of time after death. 3. The dimensions are the length, width, and depth of the grave and measurements of its main constituent parts, such as cover slabs or grave markers. Unless otherwise stated, the length and width of a grave measures the longitudinal and latitudinal distances between the interior walls of the cist, and the depth measures the distance below the enclosure to the floor of the cist. Measurements not given as ranges are maximum dimensions. Estimated measurements are marked with “ca.” 4. The orientation of the skeletons, nearly all of which were positioned supine with legs extended, is the cardinal direction in which the head lies as measured in degrees along the verte-

bral axis of the body. Since the vertebral columns were in varying states of preservation, the orientation of certain burials is estimated. Directional measurements should be accorded a range of error of +/- 5°. 5. Corporeal position is the specific relationship of the different parts of the body to each other. The description of position in this study adopts the procedure and terminology outlined by Ubelaker (1999, pp. 15–18). 6. Humphreys 1993; Mizoguchi 1993; Parker Pearson 1999, pp. 142–144; Härke 2001, pp. 18–19; Robb 2002, pp. 153–155.

THE CHRONOLOGIES OF BURIAL

23

Temporal perceptions would have framed the experience of death and mourning for residents at the Isthmus. Many cannot be easily recovered from the archaeological record because they are intellectual or emotional phenomena without material expression or historical testimony. Others can be determined from osteological, archaeological, or textual sources, namely, biological age at death, the duration of funerary rituals before and after burial, the historical date of interment, and the sequence of reuse and disturbance.

THE CHRONOLOGIES OF BURIAL The chronologies of burial define ritual events by their absolute, relative, and sequential place along a temporal continuum. The absolute chronology is the time of the burial in history. The relative chronology is the relationship in time between the burial and other events. The sequential chronology is the amount of time that elapsed between stages in the creation and use of the grave. In the case of most graves at the Isthmus, enough evidence exists to determine at least generally an absolute and relative date of interment. Often a sequential chronology can also be established, but separate stages of use can seldom be precisely dated. The primary burials are variously datable by their artifactual contents, their form, and their stratigraphic associations.7 In the study of the Isthmian remains, a combination of dating techniques has often succeeded when single classes of evidence were unavailable. Artifacts discovered in or near graves can only be used as markers of the date of interment if the circumstances of their deposition are understood. This is because all objects belonging to a burial deposit, the depositional context associated with a grave, did not enter that context through the same channels at one time. Some were intentionally placed in or near the grave as part of the funerary ritual, such as votive or commemorative offerings or items worn by the deceased. But many objects found in these deposits were unrelated to the event of burial, such as earlier sherds that were already in the fill poured over the body, or later sherds that slipped into the cist when it was reopened after initial use. Accordingly, the artifacts in any burial deposit should be identified as either coincidental, incidental, or accidental.8 Coincidental artifacts are those deposited in an area prior to its use for interment and without foreknowledge of that use. They can reveal the nature of local habitation and activity before the event of burial, and they can provide a terminus post quem for the date of interment. Most of the coincidental artifacts at the Isthmus are small sherds from earlier Roman times that survive in the burial fills. Incidental artifacts are those intentionally deposited in or near the grave during the funeral or on some later visit. They provide direct evidence for the appearance of the body when interred, the nature of the funeral, interment, and mourning, and the date of those activities. Most of the incidental artifacts at the Isthmus are vessels, jewelry, or personal articles inside the cist and lamps or vessels outside it. Accidental artifacts are artifacts deposited in an area after its use for burial and without recollection of that use. They can reveal the nature of local activity after the event of burial. Most often they represent the unintentional disturbance and contamination of burial contexts. The accidental artifacts at the Isthmus are a small number of sherds and other objects that entered cists during Byzantine construction or modern excavation. All recorded burial deposits from the graves at the Isthmus are catalogued below (pp. 99–102). Burials can be dated with a high degree of precision if they belong to periods, regions, and cultural groups in which the incidental artifacts are chronologically diagnostic objects, such as painted vases, lamps, coins, or epitaphs. The residents of the Greek countryside 7. Although classical archaeologists have traditionally relied on “grave goods” to establish a date of burial, they have many chronological indices at their disposal; see Laskaris

2000, pp. 267–268. 8. O’Shea 1984, pp. 24–25.

24

THE GRAVES AND ASSOCIATED REMAINS

during the Roman and Byzantine eras customarily interred their dead with few, if any, objects that can be precisely dated, and their graves seldom, if ever, had inscriptions.9 The few items that were most commonly interred at the Isthmus, simple jewelry and attachments for clothing, are only broadly datable by period or era. A few burials could be dated within a half century or less by bronze coins, lamps, and cooking- or coarse-ware forms. All recorded funerary artifacts from the graves at the Isthmus are recorded below (pp. 103–112).10 The form and composition of a grave can also indicate a broad date of interment. Some elements, such as distinct forms of rooftiles, were only produced during specific periods. However, tiles and stones can be deceptive chronological markers because mourners sometimes retrieved them from contexts of an earlier date for recycling in new burials. The design of a grave can also furnish a general date because distinct grave types were used most frequently during certain periods in Corinthian history. The precision and accuracy of typological dating depends on a knowledge of diachronic variation in mortuary forms across the region, which will be discussed in Chapter 4. In order to assess a date of interment from depositional associations, the grave must be placed in a continuous stratigraphic sequence by identifying the surface from which the cist was cut. The lotted finds from above and below the grave, as well as the coincidental artifacts in the burial deposit, can in the best case indicate chronological termini for the event of burial. The stratigraphic context of most graves at the Isthmus can be reconstructed in broad outline, but in a few instances it is uncertain because the ground level from which the cist was cut was very close to or even higher than modern ground level. Relative chronologies of interment can sometimes be determined from phases of building and collapse. This kind of stratigraphic evidence is especially applicable to dating the burials at the Isthmus because they are located in an area of high structural density, and many are directly associated with fortifications of known history. Artifacts, grave forms, and stratigraphy can also serve to establish a sequential chronology for activity at a burial site. In excavating and interpreting Roman and Byzantine graves, archaeologists have often overlooked or misread the full range of funerary rituals during and after the deposition of the body. Three distinct activities following primary interment are evident at the Isthmus: continued use for multiple burial, secondary burial, and human disturbance. The form of a grave and its contents must be examined closely in order to establish a date for these activities. Several factors must be considered when dating the frequency of use in multiple burials: the presence or absence of grave markers, the chronology and typology of the funerary artifacts, the number of burials, and the position of the bones. Graves that are plainly and permanently marked at the surface can be found and reopened for the addition of bodies over a period of many decades or even centuries. Unmarked graves can only be used for continued burial so long as a memory of their location survives; this would not exceed more than a few generations. Multiple burials containing incidental artifacts of the same type and date would not have been used for more than 50 to 100 years.11 In general, graves with numerous bodies probably had been used over a longer span of time than those containing only a few bodies. Finally, the rearrangement of the bones of previously buried bodies during the reopening of a multiple interment can only occur after hard tissues have become defleshed. 9. On the absence of epitaphs, see p. 172, n. 54. 10. O’Shea (1984, esp. pp. 13–14) and Chapman (2005) offer pertinent discussions of the impact of temporal scale both on the analysis of burials and on the interpretation of diachronic change in mortuary behavior and society at prehistoric sites in Europe and North America.

11. Almost all funerary artifacts in graves at the Isthmus were objects in contemporary use at the time of burial, not antiques or heirlooms. The only exception is a bronze coin (14) of the late 4th century worn as a personal ornament by a woman (T14 69-002A) interred during the Early Byzantine period.

DESCRIPTION OF THE GRAVES BY AREA

25

The rate of decomposition for the bodies at the Isthmus was relatively slow because of their intact state at death, their interment in dirt inside a lined and covered cist up to 1.50 m below ground level, and a dry, temperate burial environment.12 Under these conditions, the full decay of bodies to skeletons would have taken roughly five to 10 years.13 On the basis of these criteria, the use-life of a grave with multiple interments at the Isthmus was probably between roughly a decade and a half century, though a few might have been used over a shorter duration. The sequence of burial within that span is indicated by the superposition of the bodies, but the rate of interment cannot be ascertained. Secondary burials are often best dated by stratigraphic context because they sometimes lack an enclosure and funerary artifacts, which was the case at the Isthmus. The date of the primary burial, or the creation of the grave from which the bones were transplanted, can seldom be established. The only indication of the interval between the two burials is the condition of the skeletal remains. If many parts of the body are represented, the bones are partly articulated, and the remains are relatively intact, then the body was probably moved to the new grave before the soft tissues had completely disintegrated, within a few years of death. If, however, several skeletal elements are missing, the bones are disarticulated, and the remains display postmortem alteration from extended exposure to taphonomic agents, the body was probably moved to the new grave several years, decades, or even centuries after death. Establishing the date of activities that disturb a grave depends chiefly on the nature of the disturbance. Graves at the Isthmus were inadvertantly opened during the building of the Hexamilion and Byzantine houses, both of which are independently datable events. Such activities can introduce new material into a grave, such as fill containing sherds that indicate the date of the intrusion. Accidental artifacts can thus serve as chronological markers as long as an extended span of time separates burial and disturbance.

DESCRIPTION OF THE GRAVES BY AREA This section describes the graves and their contents by area of excavation. It begins with the numerous graves located in and around the enceinte and then moves outward to the dispersed graves in other parts of the site. This division by area allows a detailed investigation of the distinct physical settings, depositional contexts, and histories of activity in individual locales of settlement. Each subsection begins with a summary of the location and investigation of the area, then relates the formal, corporeal, artifactual, and locational dimensions of each grave in that area, and concludes with a chronological assessment.

Northeast Gate Early archaeologists at the Isthmus studied the Northeast Gate of the Fortress, but the first campaign to investigate closely the entire area was undertaken by UCLA between 1967 and 1969. These excavations uncovered 10 graves containing 33 individuals and numbered them in sequence of discovery. The interments can be placed in three groups (I–III) that are 12. Krogman and İşcan (1986, pp. 28–31), Mant (1987), and Mann, Bass and Meadows (1990) survey the factors in postmortem decomposition. 13. Marshall 1976, p. 91; Krogman and İşcan 1986, p. 29; Pollard 1996, p. 141. The soft tissues of the adults in the poorly sealed interments or cists without fill would have survived for a

shorter time, probably much less than ten years (NEG 69-007, 69-009, 69-010, T14 67-004, 69-002, 69-003, DEC 69-901, 69-902, T2 68-003, LOU 69-801, RB 76-002, WF 62-001). The two infants buried in unlined cists were probably defleshed within only a few years (NEG 69-009A, 69-010A).

Figure 2.1. Northeast Gate, actual-state plan

DESCRIPTION OF THE GRAVES BY AREA

27

Figure 2.2. Northeast Gate, schematic plan

spatially and chronologically distinct. The three earliest graves (3, 4, 8) were located south of the Gate along the Hexamilion, four later graves (1, 2, 5, 6) were located west of the Gate along the Hexamilion, and the three latest graves (7, 9, 10) were located north of the Gate in the northernmost wall of the taphros, the defensive ditch (Fig. 2.1). This subsection addresses the graves by group and concludes with contexts outside formal interments that also contained human remains. Since the architectural history and military function of this primary entrance into the Fortress have been published, a summary will suffice to provide a background for the present study (Figs. 2.1–2.3).14 In the second half of the 1st century, a tripylon was erected here over the main road from the northeast to serve both as a decorative archway and as an architectural landmark for those entering the Isthmian Sanctuary and the Peloponnese. With the construction of the Hexamilion and its enceinte around the second decade of the 5th century, the northern entrance into the Fortress incorporated the Roman Arch. The Hexamilion ran eastward up to the north pier of the Arch and continued southward from its south pier, 14. Clement (1969, pp. 139–143, pl. 82:b; 1972, pp. 164– 167) published the preliminary reports on the UCLA excavations of the Northeast Gate; Isthmia V (pp. 52–89) is the final publication. The results of exploration by P. Monceaux in 1883

and by R. J. H. Jenkins and A. H. S. Megaw in 1932–1933 are summarized in Isthmia V (pp. 2–3, 52). The reports on these early activities do not mention graves.

28

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.3. Northeast Gate, restored plan

while the two side passageways were blocked with towers and the road through the center was paved and enclosed by a secure gate. A fighting platform was created along the fortification wall with access by stairs on either side of the portal. In addition, the W-shaped taphros was cut into the hardpan north of the gate to a depth of 3.95–4.25 m below the base of the curtain.15 After the erection of the Hexamilion and the Fortress, floors and walls inside the Gate indicate that the area was occupied for domestic habitation during the mid-5th to early 6th centuries. Between 548 and ca. 555 Justinian saw to a rebuilding of the complex, which involved the construction of a square, buttressed bastion around the north tower. A blocking wall to shut the portal between the two towers, insubstantial walls inside the gate, and pottery, coins, and minor objects dating from the mid-6th to the late 7th or 8th centuries all show that the area was inhabited long after the initial Slavic incursions of the middle 580s. Thereafter, as elsewhere around the Fortress, traces of Middle to Late Byzantine occupation (12th to 15th centuries) in the form of walls, floors, and a lime kiln were found directly over Early Byzantine levels.16 15. The discovery of other sections of the taphros indicate that it ran in front of much, if not all, of the Hexamilion (Jenkins and Megaw 1931–1932, pp. 77–79, fig. 5; Isthmia V,

pp. 134–135, figs. 5, 7–9). 16. Isthmia V, pp. 12–14, 80–89, 144.

DESCRIPTION OF THE GRAVES BY AREA

29

Figure 2.4. NEG 69-103, 69-004, and 69-008 and surroundings, from northwest

Group I The first burial group consisted of three graves (3, 4, 8) located south of the portal. They were all unlined, slab-covered cists that show signs of disturbance by the footings for the Hexamilion. These burials date to the end of the 4th or early 5th century, or shortly before the construction of the fortifications in ca. 410–420. Grave 3 (69-103)17 was an unlined, slab-covered cist oriented southwest–northeast (226°) and measuring ca. 1.75 m long, 0.57 m wide, and 0.63 m deep (Figs. 2.4–2.6). It was located along the interior face of the Hexamilion 3.65 m southeast of the gateway (Fig. 2.1). Enclosing the cist were three rectangular blocks set horizontally across it.18 A circular depression near the edge of one slab and traces of dressing on another show that these had been recycled. The cutting for the grave extended ca. 0.60 m under the Hexamilion. The builders of the fortifications left a gap in the foundations to accommodate the grave, and, when they finished the inside face of the wall with a thick coating of mortar, it spilled over the top of the cover slabs. The slab underneath the Hexamilion was situated at a shallow angle to the horizontal plane of the cist, with its west end somewhat lower than its east and the foot of the grave uncovered. This probably resulted from disturbance during the construction of the wall. A fine-ware jar (9, Fig. 2.92) was discovered ca. 0.40 m below the cover at the level of the skeletons. The vessel, which stood upright, had been interred with the bodies. 17. NBs NEG I, pp. 17, 21, 85–91, 95, 106–107, 111, and NEG II, p. 136 (no trench number); Clement 1972, pp. 165– 166, pls. 133, 135:a; Isthmia V, p. 77, pl. 19:e. Allen Moore and Birgitta Wohl supervised the excavation of graves 3–5. Figure 2.6 (bottom center) shows 69-103A after the eye orbit had

slipped from its anatomical position during excavation and was put on top of the cranium. 18. The blocks, from west to east, measured: 0.58 x 0.98 x 0.22 m; 0.57 x 0.86 x 0.21 m; 0.40 x 0.76 x 0.14 m.

30

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.5. Coverings of NEG 69-103 and 69-004

The grave contained two skeletons, one (69-103B) directly above and slightly offset to the north of the other (69-103A). Both were in a supine and extended position, their heads were facing forward, and their arms were positioned over the pelves but not crossing, so that the hands rested on the thighs. The skeletal remains are only moderately well preserved (40.63% of expected bone elements) due to their lack of protection along the walls of the cist and the proximity to drainage, which attracted snails to the burial environment.19 The individual interred on the floor of the grave (A) was a female in her late 40s with periodontal disease, periapical lesions, and AMTL. The individual overlying her (B) was a male of 35 to 44 years with joint disease in the back and AMTL throughout his mouth (Figs. 5.8, 6.1). Grave 4 (69-004)20 was an unlined, slab-covered cist oriented southwest-northeast (217°) and measuring ca. 1.75 m long, 0.67–0.70 m wide, and ca. 0.50 m deep (Figs. 2.4, 2.5, 2.7– 2.9). It was located along the interior face of the Hexamilion 0.92 m southeast of 69-103 (Fig. 2.1). The cist, which extended ca. 0.75 m under the Hexamilion, was enclosed by four rectangular slabs.21 Various cuttings in these stones and three nails still intact on the bottom surface of the westernmost slab show that these were all reused blocks. As in 69-103, the easternmost slab under the Hexamilion was not aligned with the others but dipped to the west, and the foot of the grave was uncovered. The narrow gap between the top of the third slab from the west and the bottom of the first masonry course in the Hexamilion was sealed with mortar. Under the westernmost cover slab was a thin, rectangular slab leaning against the north wall of the cist. A similar slab ran under the Hexamilion and the third cover slab from 19. The fill in the middle of the cist was soft and contained snail shells (NB NEG I, p. 91). 20. NBs NEG I, pp. 10–13, 17, 21, 87, 96–99, 107–109, 111, 113–115, 117–121, 127, and NEG II, p. 136 (no trench

number); Clement 1972, p. 166, pls. 133, 135:b; Isthmia V, p. 78. 21. The slabs, from west to east, measured: 0.74 x 0.86 x 0.22 m; 0.40 x 0.88 x 0.21 m; 0.40 x 0.89 x ca. 0.20 m; 0.20 x ca. 0.89 x ca. 0.20 m.

DESCRIPTION OF THE GRAVES BY AREA

31

Figure 2.6. NEG 69-103A, B in grave, from southwest

the west, flat across the horizontal plane of the cist but slanting more deeply to the north.22 The unusual situation of these two slabs indicates that they had been moved from their original positions, where they functioned as, respectively, lining and covering members of a cist enclosure that predated the four uppermost covering slabs. An intact glass pitcher (7, Fig. 2.90) was found immediately under the cover slabs against the east edge of the lower, vertical slab ca. 0.20 m west of the Hexamilion, resting on top of a displaced cranium. Another glass vessel (13), an elongated bottle that might have been an unguentarium, was found along the north wall of the grave just east of the pitcher. This bottle was very fragile, and it disintegrated after examination. The excavators also found a piece of an iron nail, but this might have been residual in the fill.23 Immediately below the glass vessels were four skeletons in a jumbled mass against the Hexamilion, with the crania at the west end of the heap (69-004A, C–E). These bodies and the glass vessels had been interred over a short time in the same cist. Builders, however, disturbed the grave when digging the foundations of the wall, collected the uppermost skeletal remains, and redeposited them in a heap. The clean spiral and transverse breaks through the thinnest diaphysial segments (Fig. 2.9) show how they snapped when the builders dropped them back into the cist. Directly below this mass of bone was another skeleton (69-004B) that was mostly articulated up to the interior face of the Hexamilion, at which point the knees were severed. The builders had cut through the distal femoral epiphyses when digging the footing trench and had re-interred fragments of the tibiae, fibulae, tali, 22. The slab leaning against the north wall of the cist measured 0.18–0.44 x 0.31–0.56 x 0.06 m. The slab running under

the Hexamilion measured 0.55 x 0.66 x 0.06 m. 23. NB NEG I, p. 127.

32

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.7. NEG 69-004A, C–E, glass pitcher (7), and bottle (13) in grave, from southeast

Figure 2.8. NEG 69-004B and legs of A in grave, from northwest

DESCRIPTION OF THE GRAVES BY AREA

33

Figure 2.9. Spiral and transverse fractures in long bones of NEG 69-004C–E

from handling by Late Roman builders

and probably other foot bones with the overlying skeletons. They also poured earth over the intact skeleton, which caused a lateral fracture from the middle of the right parietal across the occipital, where large fragments broke off. This body (B) was supine and extended, with the head turned to its right side, the arms pulled slightly under the torso so that the hands were under the hips, and the upper legs pulled to the left. Over the chest and left shoulder of this individual were the extended leg bones of a child (A) whose head rested to the east approximately over the pelvis of the underlying body. The other elements of this skeleton had been redeposited in the overlying pile. These remains are poorly preserved (29.94% of expected bone elements) chiefly because of the disturbance by building and sedimentary compression. The skeletons collected in a heap represented a child of three or four years (A); a male over 35 years (C) with many oral conditions, including enamel hypoplasia, wear, calculus, caries, periodontal disease, periapical lesions, and AMTL (Figs. 5.10, 6.3); a female of 25 to 34 years (D) with similar dental pathology (Figs. 5.11, 6.4); and a female of 35 to 44 years (E) with similar dental pathology but also rotator cuff disease in one shoulder and slight articular degeneration in the other (Figs. 6.5, 7.2). The skeleton underlying these bones (B) was a female in her late 40s with a healed bone infection in one leg, joint disease in her back and one shoulder, broken upper front teeth, enamel hypoplasia, dental wear, calculus, and AMTL (Figs. 5.9, 6.2, 6.36, 7.1). More examples of periostitis and vertebral osteoarthritis were observed among the commingled adult bones that could not be associated with individual skeletons (Fig. 7.1). Grave 8 (69-008)24 was an unlined, slab-covered cist oriented southwest-northeast (243°) and measuring 1.40–1.60 m long, 0.54 m wide, and 0.75 m deep (Figs. 2.4, 2.10, 2.11). It was located along the interior face of the Hexamilion 0.70–1.00 m north of 69-103 (Fig. 2.1). At its top and up to the face of the Hexamilion the cutting was filled with tightly packed rubble, tile fragments, and eight substantial blocks, several of which were in secondary use. A lamp (5, Fig. 2.88) with its nozzle aligned with the widthwise axis of the grave was 24. NBs NEG IV, pp. 8, 41, 50–57, and NEG V, pp. 7, 43–49, 51 (no trench number); Clement 1972, p. 165, pls. 133, 134:a, b; Isthmia V, p. 78.

34

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.10. Covering of NEG 69-008 with lamp (5), from south

discovered intact on top of this rubble in its southwest corner, 1.07 m west of the Hexamilion and ca. 0.10 m north of the south wall of the cist. This appears to have been left at the graveside when the rubble was poured over the cutting. The covering rested over a shallow, loose fill containing large quantities of sherds and plaster fragments as well as a distinct layer of clay, which points to considerable drainage through the cist. The clay covered an irregular, roughly cut block situated transversely in the cist. Below the block was a more rectangular cover slab enclosing much of the west end of the grave up to the Hexamilion, under which the cutting extended ca. 0.20 m.25 Along the bottom of the masonry courses of the wall and the roof of the cist was a thin coating of mortar, as in 69-004. On the floor of the cist ca. 0.45 m below this lower slab was a cluster of stones including two rectangular pieces and seven smaller ones in no discernible arrangement.26 Among these stones was a bronze ring (23, Fig. 2.105) that had probably been worn on a finger of the deceased. The bones once contained in 69-008 had all but disintegrated before the grave was opened (0.21% of expected bone elements preserved). The excavators found only one human bone above the lowest stones, a brittle and deeply pitted diaphyseal fragment from a metacarpal, metatarsal, or clavicle. As has been discussed,27 the Northeast Gate, particularly the central passage once the portal was built and the road paved, was a primary route for drainage northward from the artificial plateau of the Fortress into the Ravine. The presence of water indicated by the clay sedimentation in the upper horizon of the cist would have accelerated the deterioration of bones. The comparatively unprotected nature of the interment would have assisted in this process. Furthermore, the short length of the grave in comparison to others at the Northeast Gate shows that the grave contained an infant or child. Immature skeletal remains would have been particularly vulnerable to rapid destruction by water.28 Both adult and subadult bones were better preserved in 69-103 and 69-004, but those interments were more consistently sealed and further from drainage in the roadway. An alternative explanation for the absence of bones from 69-008 is that they were completely removed from the grave when it was disturbed by the construction of the Hexamilion. This explanation is unlikely, when one considers that the same builders decided to leave numerous bones in 69-103 and 69-004. 25. The roughly cut block measured ca. 0.40 x ca. 0.40 m. The rectangular cover slab measured 0.60 x 0.42 x 0.33 m. 26. The rectangular stones measured 0.36 x 0.28 m and 0.34 x 0.33 m.

27. See pp. 12–14. 28. This occurred in NEG 69-009, NEG 69-010, and DEC 69-902 as well.

DESCRIPTION OF THE GRAVES BY AREA

35

Figure 2.11. Stones on floor of NEG 69-008 after excavation, north at top

The situation and contents of the three graves in Group I suggest a date of interment at the end of the 4th or early 5th century. The fact that the three graves were overrun by the Hexamilion indicates that they predate the second decade of the 5th century, but the artifacts suggest that they did not date much earlier. The jar (9) in 69-103 and the lamp (5) on top of 69-008 belong to the late 4th or early 5th century. The dating of the contextual artifacts from inside the cists of 69-103 and 69-008 corroborate this chronology (see p. 99). In addition to the artifacts, the similarities in design between the three graves, their proximity, and their common relationship with the Hexamilion point to a contemporary date. The amount of time between the original interments and their disturbance by construction must have been at least five to 10 years.29 This is an estimate of the amount of time it would have taken for the soft tissues of 69-004A, C–E to decay, allowing the builders to pile up the remaining bones. The sequence of primary burials cannot be retraced because of their disturbed state. The two individuals in 69-103 were probably buried in close succession, because both skeletons were found in a fully articulated and undisturbed state, while the disordered bones of the five individuals in 69-004 were more likely interred at separate times. Before the completion of the Fortress, the graves would have overlooked both the Roman Arch and the road into the Sanctuary from the northeast. When the builders of the Hexamilion incorporated the Arch into the Northeast Gate and erected the contiguous walls, they disturbed the graves in the area. The condition of 69-103 and 69-004 reveals that they attempted to preserve the interments. The foundation trench sheared off the legs of the lowest body in 69-004, substantially disturbed the overlying bodies, and probably dislodged the two thinner, lower slabs. The builders simply left the stones in the grave when they resealed it with more substantial slabs not unlike the plundered blocks they were building into the fortifications. The unusual situation of the east cover slabs of 69-103 and 69-004 may be explained by the disturbance of preexisting slabs during construction. It is difficult to determine how much of the design of these two graves belonged to the original interment and how much belonged to the rearrangement of the graves by builders. All skeletal remains, both cists, the jar (9) in 69-103, and the thinner, lower slabs in 69-004 probably predated the Hexamilion. When the foundation disturbed both graves, 69-103A and 69-103B were left in their original position, as were 69-004B, with the exception of the legs below the knees, and at least the legs 29. See pp. 24–25.

36

THE GRAVES AND ASSOCIATED REMAINS

of 69-004A. The jumble of bones from 69-004A, C–E was also created at this time, and the two glass vessels (7, 13) were placed on top. It is possible that the fragmentary lamp found in 69-103 had belonged to the original burial but was displaced and broken by the builders. Furthermore, it is unclear whether the cover slabs belonged to the initial interments or were added after disturbance. In this regard, it is noteworthy that only those slabs at the east end of the graves, underneath the Hexamilion, were positioned at an angle to the horizontal plane of the cists. The confused layering of stones throughout 69-008 also seems to have resulted from the sinking of foundations for the Hexamilion. The separate levels of stones in 69-008 appear to represent two phases of construction. The lowest stones formed the original floor, while either the large stone ca. 0.45 m above that or the more irregular stone at a slightly higher elevation belonged to the original covering. Presumably the mortar was applied to the masonry of the Hexamilion when it was erected in an attempt to seal the grave better. The uppermost deposit of rubble would have been poured over the original cutting at that time as well. It is unknown whether the lamp (5) was left on top of the grave at the time of the burial or was placed there when the grave was interrupted by construction.

Group II The second burial group consisted of four graves (1, 2, 5, 6) located west of the portal. Two of these graves were covered and three were at least partially lined. All four contained several individuals, and local residents apparently reused at least three and possibly all four for continued burial. They date to the mid- to late 5th century, which was apparently a period of relaxed military employment at the Fortress. Grave 1 (67-001)30 was a lined, uncovered cist oriented southwest–northeast (246°) and measuring ca. 2.10 m long, ca. 0.60 m wide, and ca. 0.70–0.80 m deep (Figs. 2.12–2.14).31 It was located immediately south of the stairway to the fighting platform and against the northern of the two interior walls flanking the road (Fig. 2.1). The flanking wall furnished the south wall of the cist, and the other three walls were lined with small, flat stones and fragments of tile and brick set in courses with mortar. The absence of covering is noteworthy. The cist was probably enclosed with slabs at first, but these were later discarded because the grave was opened so frequently. Among the eleven bodies interred were 10 bronze coins (e.g., 1–3, Figs. 2.84–2.86) ca. 0.30 m north of the south wall and ca. 0.75–0.95 m west of the east wall, concentrated in a pile near the middle ribs of one individual, and a bronze ring ca. 0.15 m south of them.32 Located ca. 0.40 m east, roughly over the pelvis of a child skeleton, was a severely corroded iron buckle (18, Fig. 2.100). A coarse-ware cup (8, Fig. 2.91) was also found along the base of the south wall 0.63 m east of the west wall. Finally, two eroded bronze earrings that possibly comprised a pair (16, Fig. 2.98) were found along the north wall amid a cluster of skulls 0.40–0.45 m east of the west wall. The coins were interred with the uppermost, or latest, bodies either in a small pile or within a purse that has since 30. NBs 1967 RMM I, pp. c, 75, 78, 86–89, 92, 101–102, 104–106, 144, 163, and 1967 RMM II, p. 198 (trench 67-2); Daux 1968, p. 779; Clement 1969, p. 140; 1972, p. 166; Isthmia V, p. 77, pl. 20:a, b (incorrect number of adults and children). Robert M. McClure supervised the excavation of graves 1 and 2. 31. The depth is a measurement of the grave from the top of the wall lining to the floor of the cist. 32. In addition to 1, 2, and 3, three coins were illegible (IC 945, 946, 947), three disintegrated in cleaning, and a

tenth coin was not inventoried (NB 1967 RMM I, pp. 89, 104, 105). Liane Houghtalin has observed (pers. comm.) that the flan of IC 946 is consistent with coins struck in the 5th century and that the flan of IC 947 is consistent with coins struck in the 4th to 5th centuries. Daux (1968, p. 779) and Clement (1969, p. 140; 1975, pp. 163–164, nos. 19–21) give an erroneous total (eight), omit IC 885, and incorrectly associate IC 887 with grave 1 (67-001). Clement (1987, pp. 382–383) corrected the previous publications, adding IC 885 to the burial context but removing IC 887. The bronze ring was not inventoried.

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.12. NEG 67-001A–K, coins (1–3), ring, and buckle (18) in grave, from northeast

Figure 2.13. Coins (1–3) and ring in NEG 67-001,

north at lower right

Figure 2.14. West end of NEG 67-001 with cup (8),

from southwest

37

38

THE GRAVES AND ASSOCIATED REMAINS

disintegrated.33 The earrings had been worn by one of the deceased for burial, as they were situated on either side of an adult skull. The ring and the iron buckle were also worn by two individuals. The cup, which was interred along with the lowermost, or earliest, bodies, was located just west of one skull. The grave contained eleven skeletons interred one over the other (67-001A–K). The three crowded furthest to the east represent the latest use of the grave, but it is difficult to discern separate burial events for the other eight. The bodies had been interred in the same basic supine position with slight variations. At least three skeletons had legs extended, and most seem to have had arms straight at their sides. At least six heads were turned on their left sides, and at least two heads were turned on their right sides. One of the last individuals interred, a child, along the south wall of the cist, had both legs tightly flexed to the left at a 70° angle to the vertebral axis. Another child lay directly below, oriented northwest–southeast and extending diagonally across the grave. The copious skeletal material in this grave was somewhat poorly preserved (30.72% of expected bone elements). The probable cause of degradation is the absence of covering and exposure to moisture, which is indicated by the presence of roots and plant matter (Fig. 5.2). The individuals interred in this grave were four adults and seven infants and children. One adult (A) was a male of 35 to 44 years with a sacralized coccyx, joint disease in the back, and a healed fracture in one forearm (Figs. 7.3, 7.4). Another adult (B) was a female of 35 to 44 years with a congenitally fused sternum and a sacralized coccyx, joint disease in the back, Schmorl’s nodes in the vertebrae, ossified rib cartilage, rotated teeth, dental wear, and calculus (Figs. 5.12, 6.6, 7.5, 7.6, 7.46, 7.47). Another adult (C) was a male in his late 40s with joint disease in the back, ossified rib cartilage, dental wear, calculus, periodontal disease, and numerous periapical lesions and AMTL (Figs. 5.13, 6.7, 6.38, 6.41, 7.7). The last adult (D) was a female of 20 or 21 years with joint disease in the back, Schmorl’s nodes in the vertebrae, slight dental wear, calculus, and AMTL (Figs. 5.14, 6.8). The seven subadults included one of seven to 11 years (E); two of four to eight years, including one with cribra orbitalia from a hematological disorder (F, G; Fig. 7.8); three of two to four years (H, I, K); and one of one or two years (J). Grave 2 (67-003)34 was a partially lined, uncovered cist oriented west–east (258°) and measuring ca. 2.20 m long, ca. 1.00 m wide, and ca. 0.75 m deep (Fig. 2.15).35 It was located along the south face of the Hexamilion at the base of the north stairway leading to the lower fighting platform, displacing the bottom steps (Fig. 2.1). This cist was bounded on the north by the fortification wall and on the south by the rubble wall on the north side of 67-001 extended 1.75 m to the west. Like the neighboring 67-001, this grave’s covering was missing, perhaps because it was discarded during reuse. No funerary artifacts were found in the interment. The grave contained five skeletons on top of one another (67-003A–E). The last body interred (B) had the head turned on its right side, the arms straight, and the legs extended. Below this was another extended body (A). The bones of three children (C–E) were concentrated below the heads of both top skeletons and to the south. This situation indicates that the bodies were buried on different occasions. The person(s) making the final burial removed the skull of 67-003A from the grave, perhaps while flesh still held the mandible to 33. Megaw’s suggestion (1968, p. 7) that these coins might have come from the strata into which the graves were cut is incorrect. Clement (1987, p. 383) refers to these coins as “Charon’s modest fees,” but they were not found in or near the mouth of a skeleton, pace Avramea (1997, p. 148; “l’obole dans la bouche du défunt à été repérée dans une tombe à Isthmia”).

34. NBs 1967 RMM I, pp. c, 85–87, 89, 101–102, 107, 162, and 1967 RMM II, pp. 198, 237–238, 244, 250 (trench 67-2); Daux 1968, p. 779 (incorrect number of skeletons); Clement 1969, p. 140; 1972, p. 166; Isthmia V, p. 77 (incorrect number of adults and children, repeated by Fowden [1995, p. 553]). 35. The depth is a measurement of the grave from the base of the stairway to the floor of the cist.

DESCRIPTION OF THE GRAVES BY AREA

39

Figure 2.15. NEG 67-003B in grave after removal of other skeletons, from south

the cranium. The skeletal remains in this grave were very poorly preserved (13.80% of expected bone elements). Their condition was largely the result of the incomplete enclosure of the cist and exposure to moisture, especially detrimental to the preservation of subadult bones. The skeletons were identified as two adults and three young children. The underlying adult skeleton (A) was a female of 35 to 44 years with joint disease in the back and a healed fracture in one forearm (Fig. 7.9). The overlying adult skeleton (B) was a male also of 35 to 44 years with enamel hypoplasia, dental wear, calculus, caries and AMTL (Figs. 5.15, 6.9). The skeletal remains clustered below their heads represented one child of three to five years (C), one within the first year (D), and one in the first six months (E). Grave 5 (69-005)36 was an unlined, slab-covered cist oriented west–east (258°) and measuring 1.97 m long, 0.62 m wide, and 0.60 m deep (Figs. 2.16, 2.17). It was located along the south face of the Hexamilion 0.90 m west of 67-003 (Fig. 2.1). This cist was covered with three slabs.37 The westernmost slab, which was split lengthwise into roughly two halves, was cleanly finished and smooth, while the other two were rough and somewhat thicker. All three had been set horizontally across the cist and rested on a shelf cut 0.13 m into the ashlar masonry facing the Hexamilion. The easternmost slab had been shifted slightly to the south of the lengthwise axis of the cist at some point after its initial setting. At the level of the bodies was an iron loop (22, Fig. 2.104), probably a buckle or pendant worn by one of the deceased, and a cluster of small nails around their feet, apparently hobnails.38 The grave contained three skeletons (69-005A–C). The offset position of the eastern cover slab reflects the opening of the cist for the addition of bodies after the primary interment. The uppermost individual (B) was positioned in an unusual manner, with the head turned on its right side, the torso and legs pulled slightly to the right, the right arm 36. NBs NEG I, pp. 15, 17, 79–81, 89, 120–122, 125, 142– 143, 145, 153, 165–166, 168, 171–172, and NEG II, pp. 2–3, 27, 45 (no trench number); Clement 1972, p. 166; Isthmia V, p. 78. 37. The slabs, from west to east, measured: 0.62 x 0.88 m;

0.57 x 1.03 m; 0.94 x 0.70 m. 38. The nails were not inventoried (NBs NEG I, p. 165, and NEG II, p. 45).

40

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.16. Covering of NEG 69-005,

from northeast

Figure 2.17. NEG 69-005A–C in grave, north at top right

straight, and the left arm bent back away from the body with the left hand sharply flexed. The body immediately below (A) was positioned on its left side, with the head on its left side against the Hexamilion and the legs flexed to the left at a ca. 85° angle to the vertebral axis. The body interred below this on the floor of the cist (C) was apparently supine and extended, with its head turned on its right side. The skeletal remains are moderately well preserved (37.45% of expected bone elements) due to the inadequate enclosure of the cist and its exposure to moisture. The individual on the bottom (C) was a male of 25 to 34 years with joint disease in the back, Schmorl’s nodes in the vertebrae, enamel hypoplasia, dental wear, calculus, caries, periodontal disease, and AMTL (Figs. 5.16, 6.11). The overlying individual (A) was a child of 10 to 12 years with cribra orbitalia from a hematological disorder, a congenital anomaly in the coronal formation of one molar called Carabelli’s trait, enamel hypoplasia, and slight dental wear. The uppermost individual (B) was a female

DESCRIPTION OF THE GRAVES BY AREA

41

Figure 2.18. Covering of NEG 69-001,

from northeast

of 20 to 22 years with cribra orbitalia from a hematological disorder, crowded incisors, extensive enamel hypoplasia, slight dental wear and calculus, and caries (Fig. 6.10). Grave 6 (69-001)39 was a brick-lined, slab-covered cist oriented west–east (258°) and measuring 1.60–1.88 m long, 0.48–0.52 m wide, and 0.54 m deep (Figs. 2.18–2.21). It was located along the south face of the Hexamilion just west of 69-005 but not contiguous with it (Fig. 2.1). The grave was mostly covered by a single rectangular slab with roughly cut edges and two smaller but thicker blocks on the east and west ends, all recycled from elsewhere.40 Unlike the covering of 69-005, these slabs did not rest on a ledge cut into the Hexamilion. Rather, three broken tiles had been pressed into the narrow gap between the north edge of the covering and the face of the wall. The slab at the east end was situated at a slight angle to the lengthwise axis of the grave. It rested over four large, irregular stones that were positioned 0.28 m beyond the east end of the cist proper. Under these four stones and the two other slabs to the west, the cutting was lined with tile and brick fragments set in mortar in two courses on the east wall, six courses on the south wall, and five courses on the west wall. The courses were evenly set and the individual bricks were arranged in an interlocking pattern by course. The uppermost two to three courses of the east wall had been removed after their construction. A pair of gold earrings (19, Fig. 2.101) was found on the floor of the cist, one 0.125 m east of the west wall and 0.17 m north of the south wall and the other 0.165 m east of the west wall and 0.285 m north of the south wall. The earrings were found near a few cervical vertebrae just south of and below an inverted cranium resting with its base against the west wall. Presumably the earrings were worn by that young woman (69-001C) during her funeral. 39. NBs NEG II, pp. 2–3, 7, 11, 147–148, and NEG III, pp. 7–8, 65–69, 74–75, 86–89, 100, 102–103, 119–125 (no trench number); Clement 1972, p. 166, pls. 135:c, 136; Isthmia V, p. 78, pls. 20:d, 21:a. Meri von Sternberg, Allen Moore, and

Birgitta Wohl supervised the excavation of grave 6. 40. The large slab measured 1.58 x 0.70 x 0.16–0.27 m. The block on the west end measured 0.25 x 0.70 m. The block on the east end measured 0.60 x 0.45 m.

42

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.19. Covering of NEG 69-001

The cist contained four skeletons in a confused state (69-001A–D). The one on top was almost completely articulated (D). The remains of two underlying individuals (A, B) had been collected and moved against the north wall of the cist. The cranium of a fourth individual (C), associated with the earrings, had been turned over so that its base was flat against the west wall of the cist, while the postcranial skeleton had been spread across the floor of the cist. The three lowest bodies (A–C) were the first interred in the grave, presumably in separate burial events. Their bones were later moved for the burial of the fourth body on top (D). This fourth interment required that the cist be enlarged slightly to the east. The uppermost brick courses of the east wall were removed, and the east end was enclosed with four additional stones, so that the legs of the deceased extended over and beyond the previous line of the wall. The head of the top skeleton (D) was turned to the left side, the arms were placed over the pelvis but not crossing, and the legs were extended. The original positions of the three lower skeletons (A–C) are unknown, but when they were moved the crania were placed in the west and the long bones were oriented along the west–east axis of the grave, as with an extended body. The skeletal remains in this grave were better preserved than in other graves in the vicinity (43.72% of expected bone elements) because it was more completely enclosed. The two individuals along the north wall of the cist were a female in her 50s or older (A) with joint disease in the back and shoulder and AMTL (Fig. 5.17) and a child of three or four years (B). The cranium against the west wall of the cist (C) represented a female of 20 to 24 years with cribra orbitalia from a hematological disorder and slight joint disease in the back. The individual interred across the top of the other three (D) was a female over 45 years with joint disease in the back, crowded incisors, and slight dental wear and calculus (Figs. 5.18, 6.12). One of the three adults also had a rare bone tumor (enchondroma) on a left hand bone (Fig. 7.10).

DESCRIPTION OF THE GRAVES BY AREA

43

Figure 2.20. NEG 69-001A–D in grave, from northeast

The burials in the four graves of Group II are dated by artifacts and architectural associations to the mid- to late 5th century. The rubble and mortar construction that delimits and retains both 67-001 and 67-003 indicates that they were contemporary. The coins of Marcian (2) and Leo I (1) in 67-001 furnish a secure date of interment for the latest bodies in the late third to fourth quarter of the 5th century, but the underlying bodies might have been interred a decade or more earlier, around the middle of the century. This dating on numismatic grounds for the two adjacent graves is supported by the residual artifacts in the burial fill of 67-003.41 Graves 69-005 and 69-001, which displayed structural features similar to 67-001 and 67-003, particularly the wall lining, in all likelihood also date to the mid- to late 5th century. Moreover, their similar association with the Hexamilion suggests that they were contemporary. Certain features show that these graves were repeatedly opened for burial. This disturbance represents a different, but no less conscientious, treatment of the graves and their contents than the disturbance of the graves of Group I, which builders rearranged not only for practical effect but also out of respect for the dead. In contrast, the graves of Group II were opened for the addition of bodies. The clearest instance is 69-001, which was enlarged eastward for the addition of 69-001D. The placement of the east slab over 69-005 also reflects disturbance, presumably for the interment of 69-005B on top. The large number of individuals in 67-001 and the unusual position of the two children also point to multiple burial events. Frequent reopening probably explains the absence of covering. Grave 67-003 also contained several individuals and lacked cover slabs. Here the burial of the children 41. See pp. 99–100.

44

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.21. West end of NEG 69-001 with earring (19),

from east

(C–E) preceded the burial of the adults (A, B). The mourners who reused the graves did not always leave earlier burials in the cist intact. For example, the absence of the head of 67-003A, though the skeleton is otherwise well preserved, indicates that those interring 67-003B, the final burial in the cist, removed it. Moreover, when 69-001 was expanded for the deposition of the uppermost body, the bones of the previously interred bodies were moved to accommodate the newest burial. While it is uncertain how long after the primary burial these graves were reused, there is no indication that extended periods of inactivity separated burial events. First, the artifacts in the burial fills are chronologically consistent. Second, since none of the graves seems to have been marked at the surface, their secondary use must have occurred within a generation or two of the primary interments, when some memory of their location survived. The two most populous graves (67-001, 67-003) would have been opened on several occasions. The interment of 69-001D over the displaced bones of 69-001A–C must have postdated the earlier interments by at least five to 10 years, or long enough for the soft tissues to decompose. The bodies in 67-003 and 69-005, however, appear to have been interred directly on top of older bodies in which the skeletons were still articulated by ligaments, which indicates reuse within roughly one or perhaps two decades. The duration of use of the graves with multiple burials would not have exceeded a half century, and it might well have been much shorter.

Group III The third burial group consisted of three graves (7, 9, 10) located north of the Gate in the northernmost wall of the taphros. The burial chambers were enclosed below the level of the marl slope, and two were marked at the surface with blocks of rubble and mortar finished in plaster. These graves date from the late 6th to very early 7th centuries, when the fortifications were falling into ruin after the Justinianic renovation. Grave 7 (69-007)42 was an unlined, slab-covered cist oriented west-east (260°) and measuring 2.11 m long, 0.65 m wide, and ca. 0.30–0.80 m deep (Figs. 2.22–2.25).43 It was located 42. NBs NEG III, pp. 10–11, 15, 117, 127–129, 132–135, 138–141, and NEG IV, pp. 2, 49 (trench 69-3); NB 14, pp. 97– 115 (trench 95-1); Clement 1972, pp. 164–165, pl. 130; Isthmia V, pp. 78, 80; Gregory, Kardulias, and Rife 1997, p. 4. The

trench supervisors were Allen Moore and Meri von Sternberg in 1969 and the author in 1995. 43. The depth is a measurement of the interior chamber below the slabs to the floor of the cist.

Figure 2.22. East section of Trenches 69-5/95-1 northwest of the Northeast Gate

46

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.23. Covering of NEG 69-007, from northwest

Figure 2.24. Covering of NEG 69-007

15.45 m north-northwest from the Hexamilion at the Northeast Gate (Fig. 2.1).44 The interment was made after the northern depression of the taphros had been filled to a maximum depth of 1.93 m with a coarse, reddish soil containing tile and rubble debris. From the ground level after the filling (ca. 27.800–28.175 masl) a rectangular cist was cut ca. 1.00– 1.60 m deep into the marl slope exposed by the ditch. The fine clay from the cutting of the cist was dumped on the ground directly in front of it until the backfilling of the interment, when it was spread out in a thin layer extending ca. 2.8 m southward (Fig. 2.22).45 The cutting went 44. Figure 2.22 is a composite of the east sections of trenches 69-3 and 95-1. The stratigraphy depicted in the northern depression of the taphros (95-1) is limited to the area bounded by the broken line and comprises the actual eastern escarpment of the trench. The remainder of the section (69-3) divides the trench along its lengthwise midline. The relationship between the stratigraphy in the 95-1 section and the stratigraphy directly over the slabs of grave 7 from 69-3 has been

extrapolated from a section drawn by the excavator (NB NEG IV, pp. 2–3). 45. A few pieces of obsidian and glass were found just south of the cutting on the surface of the fine sediment (NB NEG III, p. 127). These were most likely deposited there when the cist was cut and refilled and did not belong to the burial context.

DESCRIPTION OF THE GRAVES BY AREA

47

Figure 2.25. NEG 69-007A, B in grave, from southeast

straight down into the marl slope to an artificial ledge (W. ca. 0.30 m) at ca. 0.50–0.70 m below the surface. At this point the cist was sealed by four slabs resting on the ledge and leaning against the north wall. These included one small conglomerate slab over the west end, two large, roughly cut limestone slabs over the middle, and one marble fragment with faint traces of dressing over the east end.46 Eleven stones had been set in the gaps between the irregular slabs mostly along the north wall. Below these slabs, the cutting continued deeper but angled sharply to the north into the slope to create an interior chamber. This compartment contained two skeletons placed one on top of the other (69-007A, B). Both were situated supine and extended, with heads facing forward and arms crossed over the pelves. The uneven arrangement of the covering elements suggests that the two bodies were interred on separate occasions. Both skeletons are somewhat poorly preserved (34.90% of expected bone elements) because of exposure to moisture draining across the slope of fine marl (Fig. 5.3). The individual interred on the floor of the grave (A) was a female in her late 40s to middle 50s with lumbar ribs, joint disease in the back and shoulder, ossified rib cartilage, and osteoarthritis in both thumbs (Figs. 5.19, 6.13, 6.42, 6.43, 7.11, 7.12). The overlying individual (B) was probably a male of 16 to 18 years with cribra orbitalia from a hematological disorder, a congenital anomaly in the root formation of one tooth, and slight dental wear (Figs. 5.20, 7.13). Grave 9 (69-009)47 was an unlined, covered cist oriented west–east (273°) and measuring 0.82–1.25 m long, 0.54 m wide, and 0.17–0.29 m deep (Figs. 2.26–2.29).48 It was located directly north of the lime kiln and 20.25 m north-northwest of the Hexamilion at the Northeast Gate (Fig. 2.1). The grave was sealed tightly by a block-like covering built 46. The slabs, from west to east, measured: 0.34 x 0.52 m; 0.48 x 0.59 m; 0.83 x 0.70 m; 0.26 x 0.67 m. 47. NBs NEG I, pp. 2, 20, 83, 101–107, and NEG II, p. 1 (trench 69-1); NB 15, pp. 54–59, 71–81 (trench 98-1). The field supervisors for the exploration of graves 9 and 10 were Allen Moore, Birgitta Wohl, and Meri von Sternberg in 1969

and the author in 1998. After a sounding into grave 10 (lot 15-015), the excavators incorrectly concluded that these were not graves (NB NEG 1969 III, pp. 113–115; Isthmia V, p. 80; cf. NB 15, pp. 61, 63–65, 67). 48. The depth is a measurement of the interior chamber from below the tiles to the floor of the cist.

48

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.26. NEG 69-009, plan and sections

directly over the cist and against the marl slope, which had been cut back to receive it. This covering consisted of several small pieces of rubble and large tile fragments bonded with a coarse mortar into a rectangular block of varying thickness (1.61 x 0.72 x 0.25–0.32 m).49 Although the western portion of the top surface had been destroyed by the time of discovery, chiefly by water and vegetation, it appears that it had once displayed a slightly raised, square panel (L. ca. 0.45 m). All sides were coated with plaster (Th. 0.01–0.04 m) that had been applied to the flat surfaces but slopped over the edge onto the slope behind and the ground in front. This shows that the plaster was added after the deceased was buried and the grave enclosed, presumably right after interment. In front of this slab the ground level (ca. 28.900 masl) sloped gradually southward, with a strip of fine, white clay spread thinly over it. This shallow clay deposit extended from the south edge of the cover slab ca. 3 m south to where the foundations of the lime kiln later cut through it. As in the case of 69-007, this deposit is the chalky sediment that had been cut out of the marl slope and spread out in front of the fresh interment.50 49. The height of the grave marker is a measurement of the intact portion, not the fragmentary west end. 50. The pottery from the fill in the taphros immediately below the chalky sediment is lot 69-NEG-011 (6th century).

The thin band of chalky sediment here apparently produced no pottery. The pottery from the coarse fill directly over the marl slope down to the chalky sediment in this area is lot 69-NEG-007 (Late Roman).

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.27. Covering of NEG 69-009 and surroundings in 1969, from south

Figure 2.28. Covering of NEG 69-009 in 1998, north at top

Figure 2.29. Inner layer of tiles sealing burial compartment of NEG 69-009,

north at top

49

50

THE GRAVES AND ASSOCIATED REMAINS

The cist was cut straight down into the marl slope to a maximum depth of 0.48 m. At ca. 0.15–0.25 m below the surface, the cist was interrupted by a layer of tiles overlapping one another and leaning across the cist against the north wall. This layer consisted of large fragments of five different Laconian pantiles.51 Below this, the cist narrowed slightly and cut back northward into the slope ca. 0.05–0.08 m to create a compartment for the body. The grave contained a single individual represented only by four dental crowns near the west end of the cist, marking the location of the head of the body. In the west half of the grave were found an iron buckle (30, Fig. 2.112), a bronze loop, which was possibly another buckle (31; Fig. 2.113), and a shell bead (32, Fig. 2.114), perhaps from a necklace. These articles were worn by the deceased during the funeral. The location of the teeth and artifacts and the shape of the cist indicate that the body was positioned supine and extended. The stage of dental development shows that the individual was an infant in its first 10 months (69-009A). The remaining hard tissues had been destroyed over time by natural agents as the adjacent slope underwent erosion and drainage. Grave 10 (69-010)52 was an unlined, slab-covered cist oriented west–east (254°) and measuring 1.05 m long, 0.40 m wide, and 0.23–0.29 m deep (Figs. 2.30–2.35).53 It was located 1.43 m east of 69-009 and 0.78 m west of 69-007 (Fig. 2.1). Like 69-009, it was tightly sealed with a rectangular block that had been constructed over a vertical cut in the marl slope. The covering consisted of several stones of various sizes and irregular tile fragments bonded by a coarse mortar (1.37 x 0.58 x 0.20–0.31 m). The west end displayed a square panel that was slightly raised (0.38 x 0.35 m x 0.08–0.10 m). The top and sides of the covering had been coated with plaster, which extended onto the slope behind and the ground in front. In many places, the plaster displayed swirling lines that had been inscribed with two or three fingers. A cross with arms of equal length had been similarly drawn in the center of the raised panel, the horizontal bar before the vertical one. These designs were made after the plaster had been applied but was still wet, probably just after interment. The same hard, white clay deposit as was found in front of 69-007 and 69-009 was also discovered here, extending south from this plastered slab to where it was interrupted by the kiln.54 Like the neighboring graves, this cist was excavated down into the slope to a maximum depth of 0.98–1.06 m. The south wall of the cutting went straight down 0.50–0.63 m from the covering to an artificial ledge. However, the north wall immediately below the covering cut slightly northward and down into the slope ca. 0.12 m, from which point it angled back ca. 45° and continued down to the level of the ledge. At the ledge, three tile fragments were situated flat and lengthwise in a symmetrical pattern across the cist. At the west end was a triangular piece, in the middle was a roughly square piece, and at the east end was an intact Laconian pantile.55 At the level of the ledge, the cist had narrowed to an area measuring 1.27 m long and 0.41 m wide. The cist continued down below the tiles another 0.23–0.29 m to its floor, at which point it had narrowed even further to an area measuring 0.93 m long and 0.22–0.31 m wide. The only human remains in the interior chamber were 12 dental crowns in the west end, indicating the general location of the skull. The head rested on a single tile fragment on the floor of the cist, while another piece of tile was placed vertically 51. The measurements of the tile fragments, all maximum dimensions, are: 0.29 x 0.30 x 0.04 m; 0.43 x 0.38 x 0.02 m; 0.32 x 0.27 x 0.02 m; 0.33 x 0.28 x 0.02 m; 0.34 x 0.40 x 0.03 m. 52. NBs NEG I, p. 20, NEG II, pp. 1, 6, 130–132, and NEG III, pp. 113–115 (trench 69-2); NB 15, pp. 54–74, 81 (trench 98-1). 53. The depth is measurement of the interior chamber from below the tiles to the floor of the cist.

54. Digging in 1969 did not continue below the level of the fine, chalky sediment (28.720 masl), which the excavators identified with the deposit in front of grave 9 (NB NEG II, p. 130). The pottery from the coarse fill directly over the marl slope down to the chalky sediment in this area is lot 69-NEG018 (4th to 6th centuries). 55. The pantile measured 0.81 x 0.38 x 0.03 m.

DESCRIPTION OF THE GRAVES BY AREA

51

Figure 2.30. NEG 69-010, plan and sections

along the right side of the head. The location of the teeth and the bottom tiles and the shape of the cist indicate that the body was positioned supine and extended. The stage of dental development shows that the individual was a child of one to three years (69-010A). The destruction of the other skeletal remains in the cist, as in 69-008 and 69-009, was caused by environmental conditions. No intact objects accompanied the teeth but numerous sherds from a single vessel were found mixed with the fill at all levels of the cist. These sherds, several of which join, represent ca. 30% of a cooking pot (12, Fig. 2.95) that had been broken outside but near

52

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.31. Covering of NEG 69-010 and surroundings in 1969, from west

Figure 2.32. Detail of cross on west end of covering of NEG 69-010 in 1969,

north at top right

Figure 2.33. Covering of NEG 69-010 in 1998, north at top

DESCRIPTION OF THE GRAVES BY AREA

53

Figure 2.34. Inner layer of tiles sealing burial compartment of NEG 69-010,

north at top

Figure 2.35. Tile fragments at west end of floor of NEG 69-010,

north at top

the site of burial. The sherds entered the cist when the fine, chalky clay that had been cut from the marl to make the cist was shoveled back over the body after the funeral. It is uncertain whether the cooking pot was broken at this spot prior to or at the time of interment, and whether the inclusion of the sherds was accidental or intentional. The pot’s uneven fabric, with its coarse inclusions and voids, is highly friable, and it would have shattered when dropped on the ground. Natural processes of weathering tend to disperse and break down fragile objects left on the surface over time, particularly in areas such as the slope of the taphros, where local residents discarded refuse and water passed over the terrain seasonally. However, since the sherds found in the burial fill were large, their edges were relatively sharp, and several joined, the cooking pot could not have been broken long before the grave was cut. It seems likely that mourners at 69-010, just like the Greeks centuries later,56 threw the vessel on the ground beside the open grave during the funeral. 56. See p. 196 on the Byzantine and Modern Greek custom of smashing vessels at the graveside.

54

THE GRAVES AND ASSOCIATED REMAINS

An Early Byzantine date for the burials in the three graves of Group III can be established from their stratigraphy and contents. All three graves had a similar situation along the marl slope, and all three were angled cists with enclosed interior chambers. These correspondences point to contemporaneity. The depositional sequence of the northern portion of the taphros further indicates that the graves were roughly contemporary and date to the late 6th to 7th centuries. All three were associated with the same surface onto which the mourners dumped the fine, white, chalky clay when they cut the cists into the marl. This dump sealed deep deposits in the taphros. The initial filling of the taphros was either contemporary with or subsequent to the employment of the defenses in the mid-6th century for the Justinianic renovation of the Gate.57 Therefore, the fine clay from the cutting of the graves that overlies the deep deposits in the taphros postdates the mid-6th century. The latest residual pottery in the deep fill and the overlying clay dates to the 6th century.58 Sealing these strata, the graves, and the fine, chalky clay from the cutting of the cists were massive deposits containing debris from the deterioration of the Fortress. These overlying strata and related deposits to the south date to the late 6th to 7th centuries or somewhat later, with numerous sherds surviving from the Roman era.59 According to this stratigraphic sequence, the burials were interred after the filling of the taphros in the mid- or late 6th century and before the surface from which the graves were cut was sealed by the accumulation of sediments in the Early Byzantine period. The lime kiln south of the graves is a Middle to Late Byzantine construction (12th to 15th centuries).60 The ground level indicated by the elevation of the threshold on its east side (29.190 masl) is significantly higher than the surface associated with the graves and the succeeding deposits, which slope southward.61 The artifacts from the graves support this chronology. The residual pottery in 69-010 is mostly Early to Late Roman, and the large volume of fill in 69-009 and 69-010 produced no Middle to Late Byzantine artifacts. Although none can be precisely dated, the three articles worn by 69-009A (30–32) correspond in basic form to Corinthian objects of the Late Roman and Byzantine eras. The cooking pot (12), represented by numerous sherds throughout the fill, can be securely dated on the basis of parallels from Corinth, Nemea, and Argos to the late 6th to very early 7th centuries. The vessel serves as a reliable chronological marker for the interment, whether it was broken shortly before burial and unintentionally added to the cist or purposefully broken at the graveside during the funeral.

Summar y of Groups I–III and Bone from Secondar y Contexts In conclusion, the 10 graves containing 33 individuals at the Northeast Gate belong to three distinct burial phases. Group I is represented by 69-103, 69-004, and 69-008 southeast of the Gate. These three interments existed when the footings of the Hexamilion were set, an operation that significantly disturbed them. Their similarity in design and even spacing, which reflects planning by mourners, show that they were contemporary. The lamp (5) over 69-008, the glass vessel (7) inside 69-004, and residual pottery in the burial fills point to a 57. Gregory (Isthmia V, p. 83) speculates that the fill might have been intended to stabilize the Gate’s north tower. 58. The pottery from the deep fill in the northern depression of the taphros is lot 69-NEG-042 (4th to 5th centuries); the corresponding fill in the southern depression is lot 69-NEG034 (6th century). The pottery from the fill in the cutting directly over the slabs of 69-007, including the thin band of chalky sediment extending south from the cist (top elevation 28.280–28.335 masl), is lot 69-NEG-032 (Late Roman). The somewhat darker, coarser silt (top elevation 28.675 masl) directly over the chalky sediment to the north and interspersed with it to the south might have also been dumped when the

cist was backfilled, but that is uncertain (NB 14, p. 115). 59. Lots 69-NEG-021, -024, -025, -028, and -035. 60. Two other lime kilns of similar date have been found near the fortifications, one in the taphros just west of the Bath (Isthmia V, p. 37) and another just southwest of Tower 15 (Isthmia V, p. 102, pl. 33:b). 61. The Middle Byzantine pottery from lot 69-NEG-025, including a sgraffito pedestal bowl (IPB 69-14), is to be associated with the later occupation phase rather than with the strata sealing 69-007. Note, too, that the kiln was cut through the fine, chalky sediment in front of 69-009 and therefore postdates it.

DESCRIPTION OF THE GRAVES BY AREA

55

date of interment at the end of the 4th or early 5th century, shortly before the completion of the fortifications in ca. 410–420. Group II is represented by 67-001, 67-003, 69-005, and 69-006, all within or just west of the triangular space between the Hexamilion and the north flanking wall. These four graves were made after the construction of the Gate because the south face of the Hexamilion and the north face of the flanking wall in some cases delimit their cists, and in one instance the curtain wall supports cover slabs. Although Group II does not display the same consistency of design as Group I, 67-001, 67-003, and 69-001 were at least partially lined with rubble and tiles set with mortar in courses. Moreover, 67-001 and 67-003 shared a retaining wall, and 69-005 and 69-001 were almost contiguous. Because of the similar designs and regular placement of these graves, the burials seem to have been roughly contemporary. The coins (1–3) from 67-001 provide a date for 67-001 and 67-003 in the late third to fourth quarter of the 5th century, which is supported by residual pottery from 67-003. Moreover, many if not all of the graves of Group II were opened for additional interments, in some cases on many occasions but apparently only over short spans of time. Group III is represented by 69-007, 69-009, and 69-010 in the northernmost wall of the taphros. These interments, wholly unlike those inside the Gate in design, can be dated by stratigraphic associations and by the cooking pot in 69-010 (12) to the late 6th to very early 7th centuries. In addition to these 10 interments, excavations around the Northeast Gate recovered isolated human remains from three secondary contexts.62 First, the excavators found a human cranial bone resting on the southernmost slope of the taphros, ca. 5 m north of the curtain wall and ca. 11.75 m west of the northeast corner of the north tower (Fig. 2.1). It did not belong to a primary interment as no bones were found nearby at the same elevation (29.870 masl).63 It was probably redeposited here at some point not long after the cutting of the taphros in the early 5th century. The individual (69-999A), an adult, is represented only by an occipital fragment preserving the nuchal region and the lambdoid margin (Fig. 5.21). This bone could not be matched with any of the adult skeletons yet recovered in the area. Second, a fragment of a human maxilla, the anterior cheek bone, was found inside the Gate near the south flanking wall over a meter above the surface of the road pavement (32.490 masl; Fig. 2.1).64 It had been redeposited from another location long after the Theodosian construction of the Gate and probably after the burials inside it. Third, two human arm bones were found near 69-010, either in the fill of the cist along its south wall or immediately south of the grave in the surface fill from which it was cut (Fig. 2.1).65 These two bones, each of which belonged to a subadult but not necessarily to the same individual, were an unfused proximal radial epiphysis and a medial phalanx, probably from a hand. Since the primary burials in 69-007, 69-009, and 69-010 were undisturbed at the time of excavation, these two bones probably came from (an)other grave(s) in the area. These bones from secondary contexts, which came from skeletons and graves that have not yet been found, reveal that burial in the vicinity was more extensive than the uncovered graves indicate. Presumably the bones originated in graves that were disturbed by construction in Late Roman or Byzantine times, as was the case for the graves of Group I and those at Tower 14.

62. The bone discovered along with glass and coarse-ware fragments in the small pit just south of the south flanking wall was probably faunal and not human, as the excavators suggested (NB NEG II, pp. 139–140, 144). NB NEG IV, pp. 75, 93, 103, 159, records other remains from secondary deposits in trench 69-4 that were most likely also faunal. These bones were not saved. 63. NB NEG I, pp. 49, 123 (trench 69-1). The material im-

mediately above it was 69-NEG-002 (NB NEG I, pp. 41–49). The material immediately above it was lot 69-NEG-002 in trench 69-2 (NB NEG II, pp. 29–66, 134–136). 64. NB NEG IV, p. 103 (trench 69-4). The bone was not saved. 65. NB 15, p. 61 (trench 98-1). The bones were recovered from the fill in the sounding cut by the excavators in 1969, designated lot 15-015.

56

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.36. Tower 2, actual-state plan

Tower 2 The UCLA excavations in 1968 uncovered three graves in the northeast corner of the Fortress where the Hexamilion joins with the east wall of the Fortress at Tower 2 (Figs. 1.3, 2.36).66 The upper strata in the vicinity produced abundant evidence for permanent settlement during the Middle to Late Byzantine era (10th to 15th centuries).67 The three underlying graves predate this activity. One grave dug before the erection of the fortifications belonged to the settlement around the Roman Sanctuary, while the two other graves west of the tower belonged to Early Byzantine residents of the Fortress. Grave 1 (68-006)68 was a stone-lined, covered cist oriented southwest–northeast (245°) and measuring 2.13 m long, 1.12 m wide, and 0.43 m deep (Figs. 2.37–2.39). It was located 66. Clement 1970, pp. 116, 118, pl. 95; Isthmia V, pp. 110– 117. 67. Gregory 1989, pp. 206–207; Isthmia V, pp. 114–117. 68. NBs 1968 JN-LO I, pp. 137–144, 146, 149, 153, 176, 184,

194, and 1968 JN-LO II, pp. 134–135 (trench 68-3); Clement 1970, p. 118; Michaud 1970, pp. 934, 936, fig. 102; Isthmia V, p. 116. John Nicols and Louis Okin supervised the Tower 2 excavations.

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.37. Covering and marker of T2 68-006,

57

Figure 2.38. T2 68-006A in grave, from northeast

from northeast

2.62–4.75 m west of the Hexamilion and 0.80–1.92 m north of the Fortress wall (Fig. 2.36). It was cut from the surface (ca. 30.550 masl) of a fine, reddish silt stratum mixed with much rubble debris and decomposed mortar. The floor of the cist was hard-packed earth, but the elliptically shaped walls were lined with 19 stones of varying sizes, some rectangular slabs and some more amorphous fragments, as well as tile, all carefully set in courses. Three rectangular slabs (Th. ca. 0.15 m) and one triangular one at the east end were set horizontally over the stones lining the cist. A rectangular, unmarked block that served as a grave marker was set upright in the soil at the west end of the cist, near the edge of the slab but at a slightly higher elevation. Traces of plaster on its surface indicate that the stone had been previously used in the wall of a building. The grave contained a single skeleton, which was supine and extended (68-006A). The head was partly turned to its right side and the arms were over the chest but did not cross. The skeletal remains are well preserved (60.19% of expected bone elements) because of the close enclosure of the cist, which largely protected it from natural disturbance. The cranium, however, was pitted, etched, and split by roots that had grown through the west end of the cist. The individual interred here was a male of 35 to 44 years with joint disease in the back, dental wear, periodontitis, and AMTL. Grave 2 (68-002)69 was a stone-lined, covered cist oriented southwest–northeast (249°) and measuring 1.90 m long, 0.55 m wide, and 0.34 m deep (Figs. 2.40, 2.41). It was located 69. NBs 1968 JN-LO I, p. 160–163, 165–166, 168, 172, 174, 176–177, 182, 184, 194, and 1968 JN-LO II, p. 135

(trenches 68-1, 68-1a, 68-3); Clement 1970, p. 118; Michaud 1970, pp. 934, 936, fig. 102; Isthmia V, p. 116.

58

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.39. T2 68-006 with stones lining walls

after removal of skeleton, north at lower right

6.70–8.90 m west of the Hexamilion and 1.50–2.25 m north of the Fortress wall (Fig. 2.36). Although the covering was ca. 0.25–0.35 cm lower than that over 68-006, both interments were cut from the same surface. Like that grave, this cist also had a hard-packed earth floor and was roughly elliptical in shape. Its walls were lined with 11 flat stones set in courses, over which were placed two rectangular slabs and, at the east end, a rectangular stone. One unmarked stone with an irregular shape was set upright at the head of the cist at a slightly higher elevation than the covering to serve as a grave marker.70 The grave contained a single skeleton, which was supine and extended, with the torso pulled to the left (68-002A). The head was turned on its left side and the arms were crossed over the chest. The body was closer to the east end of the cist, leaving ca. 0.30 m between the cranium and the west wall. The skeletal remains, like those in 68-006, are well preserved (59.87% of expected bone elements) because the covering and lining elements had effectively protected the cist. The individual interred here was a male of 35 to 44 years with slight dental wear and calculus, caries, and AMTL (Figs. 5.22, 6.14). Grave 3 (68-003)71 was a partially tile-lined, stone-covered cist oriented north–south (333°) and measuring 1.97 m long, 0.98 m wide, and 1.00 m deep (Figs. 2.42, 2.43). It was located in the corner where the Hexamilion and the east wall of the Fortress join (Fig. 2.36). The cist was cut from the surface of the outcrop of conglomerate bedrock (29.000–29.320 masl) 70. The west cover slab measured 0.85 x 0.60 x 0.12 m. The east cover slab measured 1.26 x 0.40 x 0.17 m. The block at the foot of the cist measured 0.85 x 0.60 x 0.12 m. The grave marker measured 0.35 x 0.55 x 0.18 m.

71. NBs 1968 JN-LO I, pp. 10, 165, 167, 179, 187, 191, and 1968 JN-LO II, pp. 88–97, 99, 101, 103, 105, 107, 135–136 (trench 68-3); Clement 1970, pp. 116, 118, pl. 97; Michaud 1970, pp. 934, 936, fig. 102; Isthmia V, pp. 113–114.

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.40. Covering and marker of T2 68-002,

59

Figure 2.41. T2 68-002A in grave, from northeast

from north

on which Tower 2 rests. The southern part of the cist was enclosed by two slabs set in mortar and seven large stones arranged along the west and north edges of the cist.72 The two slabs did not entirely cover the cist, leaving the northernmost 0.65 m exposed. The cist also extended 0.37 m east under the Hexamilion and 0.82 m south under the Fortress wall. The two slabs, however, did not completely cover the cist under the walls. Both extended east only up to the face of the Hexamilion, while the south slab extended south under the wall 0.70 m. This unusual covering, which tightly sealed some areas but left others exposed, reflects disturbance and a subsequent attempt to enclose the grave at least partially. This happened when the builders of the Hexamilion were cutting the wall’s foundations and opened the earlier grave. Inside the cist, the interior walls were roughly cut from the conglomerate, and the floor (0.98 x 1.82 m) was neatly paved with seven or eight flat, square tiles, two of which were intact (Fig. 2.44).73 The grave contained two skeletons positioned side by side, one to the east (68-003A) and one to the west (68-003B). Since the original size of the cist could easily accommodate two bodies side by side and the bodies were symmetrically situated within the chamber, it seems that a double burial was planned from the outset. The eastern skeleton (A) was supine and extended, with the head turned on its left side, the arms crossed over the pelvis, and the legs straight but the heels together and the feet turned outward. The western skeleton (B) was 72. The north slab measured 0.65 x 0.89 x 0.35 m. The south slab measured 0.55 x 0.80 x 0.34 m. 73. Tiles (IM 68-109a–z, aa, bb): 0.54 x 0.54 x ca. 0.04–0.06 m. Two intact tiles and 27 fragments comprising five or six oth-

ers. They were flat, square tiles with slightly raised edges, all of the same form. Several have one surface incised shallowly by finger in broad stripes bordering each tile and crossing over the middle.

60

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.42. Covering of T2 68-003,

from northwest

Figure 2.43. T2 68-003A, B in grave,

from northwest

also supine and extended. The skeletal remains are somewhat poorly preserved (32.94%). When the Hexamilion was erected partly over this grave, builders rearranged its covering, leaving one skeleton (A) beneath the wall while the other (B) was covered only by loose slabs. Because of this discrepancy, A was less exposed to water, air, tree roots, and grassy vegetation than B. Therefore, the bones of A were largely intact, while the bones of B were brittle, fragmentary, and very difficult to recover. The individual interred to the east (A) was a male in his late 40s with joint disease in the back and shoulder, one rotated tooth, enamel hypoplasia, dental wear, caries, and AMTL (Figs. 5.23, 6.15). The individual interred to the west (B) was a female over 35 years with joint disease in the back and knee, an ossified shoulder ligament, dental wear, and calculus (Fig. 7.14). The three graves near Tower 2 represent two different phases of occupation.74 Grave 68-003 underlies the fortifications and therefore predated their construction in ca. 410–420. Its situation relative to the Hexamilion indicates that, when the builders of the wall came across the preexisting grave, they left a space in the foundations and fit the masonry tightly over the slab covering the south end of the grave (Fig. 2.42). The cist, which was cut directly into the bedrock, was sealed by a deposit of the late 4th to early 5th centuries that most likely represents the construction of the Hexamilion.75 While it is unclear by how long 68-003 predates the fortifications, its unusual floor lining and orientation, which are unparalleled among graves at the Fortress, suggest that it was made earlier in the Roman era. This grave on the northeast edge of the Sanctuary, which might have been contemporary with the Decauville graves (see below), would have overlooked the main road entering the Sanctuary from the northeast and the lower ground to the southeast. Graves 68-006 and 68-002 belong to the period after the rehabilitation of the fortifications. They both cut through a layer with large quantities of structural debris, domestic pottery, and coins that can be securely dated to the early–middle 580s.76 The upright stones 74. Cf. Isthmia V, pp. 114, 116, on chronology, which concurs only for 68-003. 75. E.g., lot 68-T20-022. Associated strata included a Phocaean ware bowl (IPR 68-36), a large spherical amphora with vertical handles (IPR 68-37), and a coin of the House of Valentinian (364–379, IC 68-36; Isthmia V, p. 114, nn. 3, 4). 76. E.g., lots 68-T20-016, -018, and -019, including a coin

of Justinian I (542/543, IC 68-25), seven coins of Justin II (ranging from 566 to 578, IC 68-19, 68-20, 68-23, 68-26, 68-28, 68-31, 68-34), a coin of Tiberius II (580/581, IC 68-27), an African red-slip form 99 bowl (IPR 68-38), a combed amphora (IPR 68-39), and a cooking pot with a plain vertical, triangular rim and a vertical, ovoid handle (IPR 68-42; Isthmia V, pp. 114–115, nn. 5, 6).

DESCRIPTION OF THE GRAVES BY AREA

61

Figure 2.44. Reconstruction of tiles (IM 68-109)

paving T2 68-003 after removal from cist

over the graves would have been visible at ground level, but the coverings over the cists would have been buried. Therefore, the relative elevations of the markers, the cover slabs, and the layer of debris all demonstrate that the markers were founded in, and the graves were dug into, the late-6th-century stratum either from the top surface of that layer or from another surface that shortly followed it. Deposits representing Middle Byzantine settlement in this area of the Fortress were found well above the graves.77 This depositional sequence shows that the two burials west of Tower 2 date after the early 580s but not long after, in the very late 6th or early 7th century.78

Tower 13 The University of Chicago excavations in 1954 discovered a grave (54-001) while exploring the west wall of the Fortress at Tower 13.79 The tile-covered cist, which was oriented roughly west–east, was located ca. 1.7 m west of two buildings against the west face of the Fortress wall and the north face of the tower (Fig. 1.3). The east end was enclosed by large tiles, but their arrangement is uncertain. The fragmentary state of these tiles and the discovery of two long bones and a mandible at a slightly higher elevation than the rest of the skeleton indicate that the grave had been disturbed after deposition. The remains represent a single skeleton, which was supine and extended, apparently with the head turned on its left side and the arms crossed over the pelvis. The bones, which appear to have been from an adolescent or adult, were not saved. A date of interment can be inferred from comparison with the remains at Tower 14. The two buildings immediately east of the grave seem identical in arrangement and construction to the Middle to Late Byzantine structure appended to the south face of Tower 14 and perhaps others between Towers 10 and 11.80 Moreover, the elevation of the top of the grave, which occurs in horizontal stratigraphy, was lower than the elevation of a floor 77. E.g., lots 68-T20-006 and -017, including a coin of Constantine VII (945–ca. 950, IC 68-29) and sgraffito vessels of the 12th century (IPB 68-7, 68-15; Isthmia V, p. 115, nn. 8–10). 78. Cf. Clement 1970, p. 118 (“They are doubtless early modern”).

79. NB 4, pp. 174–176. Oscar Broneer supervised the excavation of the grave. Broneer (1955, p. 124) reports on the investigations around Tower 13. 80. On Tower, 14, see p. 70, n. 103; on Towers 10 and 11, see Jenkins and Megaw 1931–1932, p. 75.

62

THE GRAVES AND ASSOCIATED REMAINS

associated with one of the buildings. The burial therefore predates these structures, both of which most likely belonged to the same settlement as the buildings clustering around the West Gate ca. 50 m to the north during the 13th to 15th centuries.81 It seems reasonable to propose that 54-001 was dug in the Early Byzantine period (late 6th to 7th centuries or later), or roughly contemporary with NEG Group III, T2 68-002 and 68-006, and the graves at Tower 14. The construction of later houses near Tower 13 might have disturbed the grave and displaced the bones.

Tower 14 The UCLA excavations in 1967 and 1969 in the area of Tower 14 and close to the west wall of the Fortress led to the discovery of six graves of Early Byzantine date (Figs. 1.3, 2.45).82 Although it is not possible to reconstruct exactly the stratigraphic context of each interment, the excavation records and the associated finds provide a clear picture of gross depositional history in the area of Tower 14. The excavators did not number the first three graves found in 1967, but they did number the three found in 1969.83 This section describes the graves in order of discovery and concludes with an unexcavated feature that is probably a seventh grave of comparable date. The first grave found in 1967 (67-002)84 was a stone-lined, covered cist oriented west–east (270°) and measuring 1.81 m long, 0.69 m wide, and ca. 0.75 m deep (Figs. 2.46, 2.47).85 It was located 3.4 m west of the Fortress wall and 3.1 m south of the south face of Tower 14 (Fig. 2.45). Lining the walls of the western portion of the cist were five rectangular blocks, which were probably reused from another context, and two thinner slabs set upright to either side of the head and shoulders. Another slab (0.66 x 0.84 m) was set horizontally across the upright members to enclose the east end of the cist, but the west end was found uncovered. The grave contained a single skeleton, which was supine and extended (67-002A). Three stones closely ringed the head, and a small fragment of a Greek architectural terracotta with a painted design had been reused to support the head as a pillow (Fig. 2.48).86 The cranium was turned to face north, but the detached mandible, found directly below the cranium, was facing east, its situation after the decomposition of the temporomandibular ligaments. Both the discrepancy in orientation between the face and the lower jaw and the fact that the eastern portion of the cist was securely sealed but the west end was uncovered indicate that the grave had been opened later in the Byzantine era. At that time, the head was lifted from the grave and redeposited on its base at a right angle to the lengthwise axis of the cist.87 The postcranial remains, however, were not disturbed. The arms were crossed over the abdomen, the legs were straight but pulled together at the knees, and feet were turned outward so that the heels almost touched. On the fifth finger of the right hand was an engraved bronze ring (15; Fig. 2.97). The skeletal remains are very well preserved (85.62% of expected bone elements) because of the grave’s design and topographic location away from drainage. The individual interred 81. Isthmia V, p. 107, further notes that Tower 13 had a door cut into its north side probably during this period, not unlike the portal cut into Tower 14 (p. 103). 82. Clement 1969, p. 142, pl. 84; 1972, pp. 163–164, pl. 127; Gregory 1989, pp. 204–206; 1993c, pp. 290–296, 302–305; Isthmia V, pp. 103–109. 83. Clement 1972, p. 164; Isthmia V, p. 109 (graves 1–3). 84. NB 1967 RP II, pp. 130, 157, 164–173, 181 (trench 671a). Ralph Platz supervised the excavations in 1967 at Tower 14. Figs. 2.46 and 2.47 show the skeleton after the cranium had been smashed during the Easter holiday by vandals

(NB 1967 RP II, p. 169). 85. The length and width measurements include the stone lining. 86. Architectural terracotta (IT 895): 0.088 x 0.092 x 0.047 m. Fragment preserving less than half. Terracotta slab, perhaps a paving tile. Monogram or structural design in black paint over thin buff slip. Probably of early date (Archaic or Classical?), painted much later (Roman or Early Byzantine?), and reused for burial in the late 6th to 7th centuries. Cf. IA 77-26 from the Roman Bath, an exact parallel for the painted figure. 87. NB 1967 RP II, p. 173.

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.45. Tower 14, actual-state plan

63

64

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.46. T14 67-002A in grave, from west

Figure 2.47. T14 67-002

Figure 2.48. Architectural terracotta (IT 895)

from under head of T14 67-002A. Scale 1:3

DESCRIPTION OF THE GRAVES BY AREA

65

Figure 2.49. T14 67-004A in grave, north at top

here was a male in his early to middle 50s who displayed several noteworthy conditions: two congenital anomalies of the lower vertebrae (spina bifida occulta and a sacralized coccyx); joint disease in the TMJ, back, the shoulder, forearm, wrist, and both thumbs; healed fractures in a rib, both forearms, and the last joint of the left fifth finger; and crowded teeth, enamel hypoplasia, heavy dental wear and fractures, calculus, caries, periodontal disease, periapical lesions, and AMTL (Figs. 5.24, 6.16, 6.39, 6.44, 7.15–7.20, 7.42, 7.47). The second grave found in 1967 (67-004)88 was an unlined, uncovered cist oriented west– east (273°) and measuring ca. 1.80 m long, ca. 0.70 m wide, and ca. 0.40 m deep (Fig. 2.49). It was located 6.6 m west of the Fortress wall and 5.6 m north of the north face of Tower 14 (Fig. 2.45). The grave contained a single skeleton (67-004A), which was supine and extended, with the head facing forward, the right arm over the pelvis so that the right hand rested on the right thigh, and the left arm across the abdomen. The interment was disturbed at some point before its discovery. This is indicated by the absence of both the left humerus and all phalanges, metatarsals, and three tarsals of the right foot, though the lower left arm and hand and the left foot were fully intact. This disturbance may also account for the absence of covering elements. The skeletal remains are almost perfectly preserved (89.70% of expected bone elements) because of the grave’s design and topographic location away from drainage. The individual interred here was a male of 25 to 34 years with a sacralized lumbar, Schmorl’s nodes in the vertebrae, severe chronic dislocations in both shoulders, rotated teeth, enamel hypoplasia, and slight dental wear and calculus (Figs. 5.25, 6.17, 7.21, 7.22, 7.47). The third grave found in 1967 (69-701)89 was a stone-lined, uncovered cist oriented west– east (264°) and measuring ca. 1.80 m long, ca. 0.45–0.55 m wide, and ca. 0.45 m deep (Fig. 2.50). It was located 3.7 m west of the Fortress wall and 10.5 m south of the south face of Tower 14 (Fig. 2.45). The cist, which contained a single skeleton (69-701A), had been cut directly over an earlier northwest–southeast wall. The walls of the cist were lined with stones but there were no cover slabs. Upon discovery, only the western two-thirds of the skeleton was exhumed, and no bones were saved. When the remainder of the grave to the east was discovered in 1969,90 the excavators recovered the right and left tibiae, fibulae, and foot 88. NB 1967 RP II, pp. 232–233 (trench 67-7). 89. NB 1967 RP II, p. 271 (trench 67-17). 90. NB T14 1969 II, pp. 291–297, 301 (trench 69-2). David

Wilson and Susan Kallemeyn supervised the excavation of this grave in 1969.

66

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.50. Lower legs of T14 69-701A, from west

bones, representing only 10.52% of expected bone elements. It is clear from the position of the legs and the size of the cist that the body was supine and extended. Little can be said about this individual except that he or she was an adult.91 Grave 1 (69-991)92 was a stone-lined, covered cist oriented west–east (269°) and measuring ca. 1.20 m long, ca. 0.75 m wide and ca. 0.45 m deep (Figs. 2.51–2.53).93 It was located 1.1 m west of the Fortress wall and 16.9 m south of the south face of Tower 14 (Fig. 2.45). The walls of the cist were lined with seven large slabs and four smaller stones set upright, but only the east end was covered, with three irregular slabs and a large tile fragment. The remainder of the covering might have been removed during construction later in the Byzantine era. The grave contained a single infant skeleton (69-991A), which was supine and extended, with the head facing forward and the arms crossed over the chest. A pendant cross (20) surrounded by at least 11 different beads (21a–k, Figs. 2.102, 2.103) was found near the upper right arm.94 These were part of a necklace worn by the deceased for burial, but its exact placement at interment cannot be ascertained from its position when found. Moreover, the string was not preserved, so the arrangement of the objects is unknown. Presumably the cross occupied the central position and hung over the middle of the chest; perhaps the beads alternated in shape, material, or color. The skeletal remains are poorly preserved (20.17% of expected bone elements) due to the incomplete enclosure of the cist and the fragile nature of immature bone. The individual interred here was aged one or two years. Grave 2 (69-002)95 was an unlined, covered cist oriented west–east (277°) and measuring ca. 1.65 m long, ca. 0.55 m wide, and ca. 0.45 m deep (Fig. 2.54). It was located 4.9 m west of the Fortress wall and 18.2 m south of the south face of Tower 14, directly under the foundations of a rubble wall (W. ca. 0.60 m) that formed the north side of a small building of later date (Fig. 2.45).96 The cist, which contained a single skeleton (69-002A), was covered with three irregular slabs, but its walls were not lined. The western slab was intact, and the later rubble wall was founded over the top of it.97 A rectangular stone was placed lengthwise 91. Rife and Giesen (1994, p. 232, table 11.3) incorrectly identify 67-004A as 69-701A. 92. NB T14 1969 III, pp. 7, 66–77 (trench 69-2); Isthmia V, p. 109. Judith Brunner and David Wilson supervised the excavation of graves 1–3. 93. The length and width measurements include the stone lining. 94. The excavators also noted other small metallic objects, some spherical, which might well have been additional beads (NB T14 1969 III, pp. 73–75).

95. NB T14 1969 III, pp. 123, 129–141 (trench 69-2). 96. Figure 2.54 shows the cist after excavators had removed a portion of its north wall that extended westward from a point adjacent to the left elbow. The original line of the north wall of the cist can be traced along the faint margin of dark soil that is visible ca. 0.20–0.30 m north of the left pelvic and leg bones. 97. NB T14 1969 III, pp. 123, 127, 129–131, 137 (trench 69-6), records the superposition of the wall over the grave; cf. Isthmia V, p. 109.

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.51. Covering over east end of T14 69-991, north at top

Figure 2.52. Upper body of T14 69-991A in west end of cist, from southwest

Figure 2.53. Lower body of T14 69-991A in east end of cist, north at top

67

68

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.54. T14 69-002A in grave, from east

across the neck and under the chin of the deceased, and large stones supported the head on both sides. In the grave were a very worn, pierced coin (14, Fig. 2.96) over the right shoulder near the neck (a necklace pendant or clothing attachment?), the shafts of a bone pin approximately over the left shoulder, and an iron pin (dress fasteners?). The body was supine and extended, with the head facing forward but tilted slightly to the right and the arms crossed over the abdomen. The right femur had been removed from its anatomical position and set over ca. 0.10 m of dirt along the vertical midline of the thoracic cavity from the chin to the hips. Moreover, although the skeleton was otherwise well preserved and fully articulated, the right innominate was missing, and the legs were severed just above the midshaft of the tibiae and fibulae, leaving no trace of the ankles or feet. This disturbance occurred when the grave was discovered during the pouring of the rubble foundation directly over it. The skeletal remains are well preserved (67.06% of expected bone elements) because of the grave’s design and topographic location away from drainage. The individual interred here was a female of 25 to 34 years with spina bifida occulta, cribra orbitalia from a hematological disorder, slight dental wear and calculus, and caries (Figs. 5.26, 6.18, 7.23). Grave 3 (69-003)98 was an unlined, covered cist oriented west–east (268°) and measuring ca. 2.00 m long, ca. 0.80 m wide, and ca. 0.40 m deep (Figs. 2.55, 2.56). It was located 1.9 m west of the Fortress wall and 7.5 m south of the south face of Tower 14 (Fig. 2.45). The cist, which contained a single skeleton (69-003A), was sealed tightly with four slabs over the west end and middle and three smaller blocks over the east end. The westernmost slab was roughly rounded to conform with the somewhat elliptical shape of the cist. A tile had been set across the chest of the deceased and up under the chin, and two blocks supported the head on either side. A worn pentanummium (4, Fig. 2.87) was found over the chest, where 98. NB T14 1969 III, pp. 169–179, 187, 194 (trench 69-5).

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.55. Covering of T14 69-003, from west

69

Figure 2.56. T14 69-003A in grave, north at right

it had been placed at the time of interment. The body was supine with the head facing forward, the left arm across the abdomen, the right arm across the chest, and the legs straight but the feet turned slightly outward. The skeletal remains are very well preserved (74.03%), mostly because of the grave’s location away from drainage. The individual interred here was a male of 18 to 24 years with Schmorl’s nodes in the vertebrae, a developmental defect called a foramen caecum in one molar, enamel hypoplasia, slight dental wear and calculus, caries, periodontal disease, and periapical lesions (Figs. 5.27, 6.19). The stratigraphic context of the six graves near Tower 14 shows that they were all roughly contemporary and date to the Early Byzantine period, the 7th century or later. The graves were cut from a surface that sloped northeast at a slight gradient from ca. 39.900– 39.800 masl over 69-002 and 69-991 to ca. 39.250 masl over 67-004. This surface postdated by several centuries two monumental features: the so-called Long Wall in ashlar masonry of the last third of the 1st century and the three-stepped structure with a northwest–southeast wall of the late 3rd to early 4th centuries (Fig. 2.45).99 Overlying these features were deposits of the 4th to 6th centuries containing rubble debris and a horizon of burned matter. Above this was the uppermost, or surface, stratum through which the graves were cut and in which the latest pottery belonged to the 6th century.100 Overlying the surface associated with the interments was an irregular stratum ca. 0.30–0.50 m deep containing only a few surviving sherds of much earlier date.101 Immediately above this was a broad horizon of darker, finer, 99. Long Wall: Clement 1969, p. 142, pls. 84:b–88:b; 1972, pp. 163–164, pl. 128:a; Marty Peppers 1979, pp. 142–212 (group A); Isthmia V, pp. 107–108; Marty 1993, pp. 117–121, figs. 2, 3 (group I). Three-stepped structure and wall: Clement 1972, p. 164, pls. 128:b, 129:a; Isthmia V, p. 108. 100. Lots representing Late Roman strata below graves:

67-T14-077 (67-002); 67-T14-135, 69-T14-050, -052 (69-701); 69-T14-099 (69-991); 69-T14-135 (69-003). 101. Lots representing thin stratum immediately above graves: 67-T14-134 (69-701); 69-T14-098 (69-991); 69-T14-117 (69-002); 69-T14-128 (69-003).

70

THE GRAVES AND ASSOCIATED REMAINS

looser silt dense with Middle to Late Byzantine material.102 Associated with the Byzantine levels were five rectangular buildings, two inside the Fortress, one appended to the south face of Tower 14, and one each in the northern and southern areas of excavation. This relatively dense occupation most likely began in the 13th century with the construction of the houses103 and the opening of the West Gate,104 and it continued into the 15th century. The buildings seem to have been abandoned shortly before or at the time of the renovation of the fortifications in 1415 under Manuel II. In this depositional sequence, the graves were cut during a single phase shortly after the Late Roman deposits, which continued well into the 6th century. The burial fills from the six graves (see pp. 100–101) contained mostly small, unidentifiable coarse-ware sherds (59%). Of the remaining pottery that was chronologically diagnostic, a majority was residual Late Roman (82%), with a small (6%) but consistent admixture of Early Byzantine, namely, sherds from 7th-century domestic wares and “Slavic ware” of the 7th or 8th century105 in 67-002, 69-991, and 69-003. A rare coin of Constantine IV (675–685) found in a secondary context just north of Tower 14 points to activity on the western side of the Fortress at least as late as the fourth quarter of the 7th century.106 The thin stratum directly above the graves, which produced relatively little cultural material, represents the subsequent period of inactivity before the dense settlement of the west side of the Fortress in the 13th to 15th centuries.107 The funerary artifacts corroborate the 7th-century date based on stratigraphic grounds. The necklace with pendant cross worn by the infant in 69-991 (20, 21) cannot be dated precisely, but the cross and beads are broadly comparable to Corinthian examples of the Late Roman to Early Byzantine periods. The bronze finger ring in 67-002 (15) has close parallels in jewelry from graves at Corinth and Nemea dated to the 6th century or somewhat earlier. The pentanummium in 69-003 (4), probably a funerary artifact, furnishes an imprecise terminus post quem of the beginning of the 6th century. The worn coin of the late 4th century from 69-002 (14) was recycled as a pendant after its monetary use and thus cannot be used to date the interment.108 The continued circulation of Late Roman bronze coins for several decades or even centuries was not unusual,109 and worn 4th-century coinage has often been found in sealed deposits of the 6th to 7th centuries in southwestern Asia Minor and southern Greece, including Corinth and the Isthmus.110 Moreover, coins minted in the 102. Lots representing Middle to Late Byzantine occupation above graves: 67-T14-065 (67-002); 67-T14-110 (67-004); 67-T14-131 (69-701); 69-T14-096 (69-991). 103. The buildings, which were contemporary (Isthmia V, p. 107), can be dated from the primary floor deposit in the building abutting Tower 14 (lot 67-T14-103). Noteworthy pottery included an Italian polychrome ware bowl (IP 3660), a cooking pot (IPB 67-4), two Spanish ware bowls (IPB 67-17, 67-20), four South Italian shield bowls (IPB 67-21, 67-22, 6724, 67-26), and two incised sgraffito bowls (IPB 69-40, 69-43: Gregory 1989, pp. 204–206, figs. 2–6; 1993c, pp. 290–293, 294–295, 302–304, nos. 6–9, 14, 24, 25, pls. 3–6, 11, 12; Isthmia V, pp. 103–106, n. 22). The same building produced several Venetian colonial coins (IC 846, 850, 852, 853, 856, 858, 862, 864, 935, 948; Isthmia V, p. 104, n. 21). 104. On the West Gate, see Isthmia V, pp. 103–104. The threshold pavement inside the gateway discovered by Monceaux (1884, p. 277) belonged to the design of the portal and was not associated with the Roman deposits, as Jenkins and Megaw proposed (1931–1932, p. 74). 105. On the chronology of the local assemblage of “Slavic ware,” see p. 138, n. 147. 106. IC 929 from lot 67-T14-122, trench 67-FW-9 (14th to 15th centuries). 107. Likewise, Jenkins and Megaw (1931–1932, p. 75) excavated two Byzantine houses directly overlying Late Roman

levels adjacent to the Fortress wall between Towers 10 and 11. The area around Tower 13 apparently had a similar history of occupation. 108. Another pierced coin of later date was found at Tower 14 (IC 897, anonymous follis class A1, 970–ca. 976). MacIsaac (1995, pp. 23–24) addresses the archaeological interpretation of altered coins. 109. See Kenchreai III, p. 1, n. 3; Dengate 1981, p. 157; MacIsaac 1987, p. 99. 110. E.g., Sardis I, pp. 1–3 (coins from early-7th-century destruction levels of the city); Kosmetatou 1996, pp. 118, 121, 123; 1998, pp. 170, 173–174; 2007, pp. 191, 193 (coins of the 4th century in habitation levels of the 5th to 6th centuries at Midea); Völling 1997, pp. 432, 440–441, n. 23 (coins of the 4th and 5th centuries in a 6th-century hoard at Olympia); Callegher 2005, p. 228 (coins of the 4th and 5th centuries continued to circulate with Early Byzantine denominations in Patras and nearby Achaian settlements). Dengate (1981, pp. 157–158, n. 21) lists numerous examples, such as the late-6th-century hoard from the Gymnasium area at Corinth; see also MacDowall 1965, p. 267, on minimi in the late-6th-century hoard from the Temple of Poseidon at the Isthmus. The unexpected discovery of hoards and stray finds by Byzantine Greeks sometimes led to the recirculation of very old, worn coins (Pennas 1996, pp. 199–200, nn. 17–19).

DESCRIPTION OF THE GRAVES BY AREA

71

Figure 2.57. Covering of possible grave northwest

of Tower 14, north at bottom

4th century that were either pierced for suspension or deposited as funerary offerings have been found in Corinthian graves.111 It is unknown whether the coin from 69-002 was a found object or valid currency when it was converted into jewelry, though the former seems more likely. The reuse of coins as personal accessories was common at Corinth not only during the Byzantine era but also during earlier periods.112 The condition of certain graves and their contents from Tower 14 indicates that they were disturbed after their deposition. The graves are all partially or completely lacking enclosing stones, which is unusual among the formal interments of all phases of occupation. In several cases (67-002, 69-701, 69-991), the absence of some covering elements but not others points to the selective removal of stones. The bones were treated in a similar manner. The cranium in one grave (67-002) and the right femur in another (69-002) were lifted out of the grave and replaced in a nonanatomical position, while the right humerus and right foot bones (67-004) and the right innominate (69-002) were removed in other cases. Such careful modifications to the skeletons must have occurred after the total decomposition of the soft tissues, or at least five to 10 years after interment. The relationship between the Early Byzantine graves and the Late Byzantine buildings suggests that the span of time was much greater. The builders of the rubble wall directly above 69-002 not only opened the old grave when they cut through the legs but also handled the bones, probably in the 13th century. The builders of the 13th-century house appended to Tower 14 must have hit the wall or covering of 67-002 when they dug the south foundations, at which point they removed the western slab(s) and handled the cranium.113 Builders might have also opened 67-004 and 69-991 and taken away stones and bones. Both graves are located in the general vicinity of later Byzantine walls, but the stratigraphic associations are not precisely documented. In addition to the six graves, excavations in 1967 found another probable interment in the vicinity. The excavators uncovered but did not open what appears to be the tile and stone covering for a grave 15.3 m north and 7.9 m west of the northwest corner of Tower 14 (Fig. 2.57). The covering was oriented west–east (ca. 260°) and measured ca. 1.50 m long and ca. 0.65 m wide.114 It was a rectangular arrangement that consisted of three large stones ringing the east end and at least 14 large tiles, some fragmentary, within the stones, and 111. A pierced coin of Constans I (346–350) was found in a Late Roman to Early Byzantine grave (no. 86) in the Gymnasium area (Wiseman 1969, p. 79, n. 28). Morgan (1936, pp. 473–474) reports the re-use of a coin of Arcadius “for purely symbolic reasons” in a Middle Byzantine grave in the Bema church in the Forum. 112. The author thanks Orestes Zervos for discussing the chronology and use of pierced coins at Corinth.

113. This wall is visible on the actual-state plan of the area (Fig. 2.45) but not the photograph of 67-002 (Fig. 2.46) because the excavators had removed it by the time they discovered the grave. 114. NB 1967 RP II, pp. 254–255, 257 (trench 67-11). A smaller arrangement of tiles, which is not described in the field notes, was discovered roughly 1 m to the north but was left unexcavated.

72

THE GRAVES AND ASSOCIATED REMAINS

extending to the west end of the cist. Thus, the walls seem to have been lined with stones and the top of the cutting (39.230 masl) was sealed with a layer of tile not unlike that found over HO 70-902 (see Fig. 2.70). If this was a grave, its dimensions are consistent with the burial of an infant or child. The feature postdates by several centuries the Long Wall over which it was constructed. The proximity of this putative grave to the other interments near Tower 14 suggests that it belonged to the same cemetery.115

Loukos Field The UCLA excavations in 1969 on the Loukos property outside the south wall of the Fortress, between Towers 9 and 10 and north of the modern road, revealed the remains of residential settlement in the Middle to Late Byzantine era (Figs. 1.3, 2.58).116 A single interment was discovered in the southeast corner of the northern sector of excavation, along the east balk. The grave (69-801)117 was a simple cist with no covering or funerary artifacts, oriented northwest–southeast (325°) and measuring 1.02 m long, 0.72 m wide, and 0.20 m deep (Figs. 2.59, 2.60).118 The cutting for the grave, which was elliptical and roughly defined, was first detectable ca. 0.35 m below the modern surface (43.131 masl). This area, however, has undergone erosion and cultivation over the centuries, and the surface from which the grave was cut might have been higher than this level. The cist’s only feature was on the floor nearest the east wall, directly under the pelvis, where three small tile fragments had been pressed flat into the earth. The grave contained a single skeleton (69-801A) with the head facing forward and the arms crossed over the pelvis. While both legs were still fully articulated from hip to foot, they were situated posterior side up to the left of and parallel with the vertebral column. To achieve this abnormal position, the muscles in the proximal femoral and gluteal regions and the hip ligaments must have first been cut or obliterated, and then the legs were folded back over the torso.119 The skeletal remains are somewhat poorly preserved (38.95% of expected bone elements) because of the unprotected nature of the cist. The individual interred here was a male over 35 years with joint disease in the back, ossified rib cartilage, an enthesophyte in the lower left leg, enamel hypoplasia, dental wear and chipping, calculus, and caries (Figs. 5.28, 6.20, 7.24). The burial can be dated by context to the Middle to Late Byzantine era. The interior of the cist, although constricted, produced several residual artifacts, the latest of which were three glazed sherds of the 12th to 14th centuries.120 The thin deposits of coarse sediment overlying 69-801 contained a few Late Roman sherds but mostly Middle to Late Byzantine pottery (late 12th to early 14th centuries).121 No architectural remains were found in the immediate vicinity,122 but a settlement was located along the nearby wall of the Fortress. This settlement is represented by a long, rectangular building with parallel rooms abutting the Fortress wall east of Tower 10 and, south of the walls and west of the grave, a scattering of bothroi for dumping refuse. The pottery and coins associated with these structures and the 115. The artifacts from the modern surface down to the level of the stone and tile covering in the south area of trench 67-11 were saved as lot 67-T14-126 (Late Byzantine). 116. Clement 1972, p. 163, pl. 125; 1977a, p. 149; Isthmia V, pp. 120–124. 117. NB Loukos III, pp. 101–106, 109 (trench 69-8-I); NB 14, pp. 123–148 (trench 95-2); Gregory, Kardulias, and Rife 1997, pp. 1, 4. The trench supervisors were Tom Jacoby and Ruth Scott in 1969 and the author in 1995. Figure 2.59 shows the skeleton after the excavators had dislodged several long bones and arrayed them alongside the cist. 118. The depth measures the cist from the point where it is first discernible to its floor. Although the uppermost portion

of the cist was not preserved, it could not have been much higher in its original form. 119. No cutmarks were observed on the bones of the pelvic girdle and upper legs, but this region is very fragmentary. 120. Although the uppermost horizon of fill in this cist was disturbed, the lower horizon, containing the skeleton and the latest sherds, was not. 121. Lots 69-LOU-107, including a slip-painted bowl (IPB 69-20; Gregory 1989, p. 207), 14-029, and 14-030. 122. The “cement floor” noted by the excavators to the south-southeast of the grave (NB Loukos III, p. 101) was a calcareous concretion that formed as a natural crust, not a constructed surface.

DESCRIPTION OF THE GRAVES BY AREA

73

Figure 2.58. Loukos Field, actual-state plan

bothroi date primarily to the 12th to 13th centuries but span the late 11th to 14th centuries.123 The grave must have belonged to this occupation of the Fortress. The unusual disposition of the body has no obvious interpretation. The irregular, unprotected character of the grave suggests that it was an informal or hasty interment. Moreover, the articulated state of the skeleton indicates that the body was interred when the ligaments, if not more soft tissues, were at least partly intact. The situation of the legs over the thorax indicates that the body had been intentionally folded over or severed at the hip. This perimortem mutilation of the body could have occurred before or shortly after death. Without any traces of cutting, chopping, or fracturing in the pelvic region it would be incautious to attribute the disfigurement to violence or massive injury. A more conservative explanation is that 69-801 is a secondary burial. Perhaps the body was exposed or buried elsewhere before it was transplanted to its final resting place, at which point the partly decomposed legs were folded back to fit a smaller burial space. 123. E.g., a glaze-painted bowl (IPB 69-5) and coins of Guillaume de Villehardouin (1245–1250, IC 69-29) and Guillaume de la Roche (1280–1287, IC 69-38; Isthmia V, pp. 123–124,

nn. 8, 10). On the pottery, see Gregory 1989, pp. 207–208, and Isthmia V, pp. 121–123, nn. 4, 9; on the excavations, see Clement 1972, p. 163.

74

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.59. LOU 69-801A in grave, north at top right

Figure 2.60. LOU 69-801

DESCRIPTION OF THE GRAVES BY AREA

75

Decauville Graves The UCLA excavations in 1969 discovered two graves of Roman date northeast of Tower 2 and just southwest of the modern highway to Epidauros (Figs. 1.2, 1.3).124 During the investigation of the Northeast Gate, a bulldozer turned up antiquities while clearing the area near the eastern terminus of the excavators’ “decauville,” a narrow-gauge railtrack. This prompted the study of a small rise above the southwest–northeast tributary to the Great Ravine overlooking the main road into the Sanctuary and the Peloponnese from the northeast. Since it was a salvage effort, exploration did not extend beyond the interments; it remains unclear whether the graves were isolated occurrences or belonged to a larger cemetery. In any event, burial in this area was sparse: the interments were ca. 14 m apart and two intervening trenches yielded nothing. Grave 1 (69-901)125 was a slab-covered cist oriented west–east (293°) and measuring 1.91 m long, 0.57 m wide, and 0.32 m deep (Fig. 2.61). It was located ca. 32 m northeast of Tower 2. The grave had been cut from a ground level ca. 0.20 m below the modern surface (24.295 masl). Set over the west end of the cist was a roughly cut slab (ca. 0.50 x 0.79 x 0.10 m), which extended from the edge of the cist to just below the shoulders of the skeletons. Two rectangular slabs (L. ca. 0.35 m) were set upright to the north and south of the knees, and three large stones surrounded the east end of the cist. The two skeletons inside had been interred one over the other on separate occasions (69-901A, B). They were supine and extended, with their heads facing forward but tilted slightly to the right, their arms straight but the hands turned slightly inward at the hips, and their legs straight but pulled to the right. The skeletal remains are very poorly preserved (8.21% of expected bone elements) because of the poor protection of the cist, exposure to moisture, and disturbance by vegetation (Fig. 5.1). The individual interred on top (A) was an adult over 45 years with extensive dental wear (Fig. 6.21). The individual interred on the floor of the cist (B) was an adult of unknown age, but probably younger than A, with less dental wear and a chipped tooth (Fig. 6.22). Grave 2 (69-902)126 was a tile-covered cist oriented west–east (282°) and measuring 0.93– 1.10 m long, 0.36–0.44 m wide, and ca. 0.35 m deep (Figs. 2.62, 2.63).127 It was located ca. 45 m northeast of Tower 2 and ca. 14 m north-northeast of 69-901. This grave had been cut from the same surface (24.265 masl) as 69-901. The bulldozer broke into the northern portion of the grave but left most of it intact. The covering consisted of two large, curved Laconian tiles (0.80 x 0.30 m) propped against each other at an acute angle in a pitched arrangement. This was capped at the west end with a fragmentary tile (0.44 x 0.40 m) that might have completely covered the grave before the bulldozing. This tile was pressed horizontally into a dense packing of fine clay around the walls of the tent structure. Smaller tile fragments had been set up against the east end of the upright members. No skeletal remains were found inside this well-sealed interment, only a small, circular depression or cutting in the unpaved soil to the west, perhaps where the head of the deceased had once rested. The size and shape of the cist indicate that the body of a child was interred here in an extended and probably supine position. The absence of any human remains is explained by the fact that the grave was exposed to moisture in a drainage zone and to the roots of shallow topsoil vegetation. This accelerated the disintegration of the immature bone, which by nature has low structural durability.128 A precise date of interment is difficult to establish. The proximity and similarity of these two graves suggest that they were contemporary. While neither contained significant 124. Clement 1972, p. 167. 125. NB NEG I, pp. 145, 147; NB NEG II, p. 47; NB ICBDEC 1969, pp. 126, 143–163 (trench 69-1). Judith Brunner supervised the excavation of the Decauville graves.

126. NB ICB-DEC 1969, pp. 128, 165–175 (trench 69-2). 127. These measurements include the tile covering. 128. Cf. NEG 69-008, 69-009, and 69-010, which contained few or no human remains when opened.

76

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.61. DEC 69-901A, B in grave, from east

artifacts, the subsurface zone immediately above 69-901 produced no pottery postdating the Late Roman period,129 and the shape of the tiles excludes a pre-Roman date. This particular pitched design was most characteristic of Roman and Late Antique graves in the Corinthia and Argolid.130 The arrangement of tiles over 69-902 differs from the use of tiles to cover WF 62-001, which is Roman, and both T13 54-001 and the putative grave north of Tower 14, which are most likely Early Byzantine. Thus, the two graves can be generally assigned to the Roman period, the late 1st to mid-6th centuries. The consistent preference for covered cists by the residents of the Fortress from the end of the 4th to the mid-6th centuries favors the earlier part of that range.

Roman Bath The UCLA excavations in 1976 discovered a grave (76-002)131 in the drain under the floor of room II in the Roman Bath (Figs. 2.64, 2.66). When the excavators cleared the floor of the massive deposit of stucco, revetment, window glass, brick, and mortar, they discovered 129. Lots 69-ICB-017 and -019. 130. For a full discussion of this type and its date, see p. 176. 131. NB MMT 1976, pp. 23, 30, 37, 40, 41, 45, 46, 100–104

(trench 76-2); Clement 1984, p. 67, pl. 60:b; Gregory 1995, pp. 289–290, n. 22. Margaret MacVeagh Thorne supervised the excavation of the skeletons in the Roman Bath.

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.62. Tiles covering DEC 69-902, from north

Figure 2.63. Pitched tiles covering DEC 69-902

after removal of outer tiles, from west

77

78

THE GRAVES AND ASSOCIATED REMAINS

a large break through the mosaic pavement in the room’s southwest corner. This exposed two perpendicular segments of channels lined with slabs and hydraulic cement. These had been constructed as drains for the Greek Bath and were incorporated into the Roman Bath. The two channels sloped down from the south and east into a deep, vertical shaft where the water drained. The breach in the floor exposing the juncture of the two channels and the adjacent shaft belonged to a larger system of makeshift drains cut unevenly through the floors of rooms III, IV, and VI. These unfinished channels were made at the time of other activities in the Bath during the 5th century, when the fortifications were erected and first employed.132 After the floor in room II was broken, a fine, gray silt washed into both channels and collected to a maximum depth of ca. 0.40 m just south of their juncture.133 Later, local residents converted the rectangular drain (H. 0.71 x W. 0.55 m) running east–west (282°) into a grave. The exposed channel was blocked by a large, reused slab with traces of mortar on the edges (0.85 x 0.65 x 0.33 m). The slab was set upright against the opening and directly on top of the gray silt. A sizable gap in the opening between the walls of the channel and the north margin of the slab was sealed with a stone and two tile fragments (Fig. 2.65). Beginning at ca. 0.50 m inside the slab and continuing 3.58 m east into the drain were the remains of three individuals lying roughly head-to-toe on top of the same silt as was found under the slab and in the adjacent north–south channel. Two objects were left at graveside outside the slab in the north–south channel either during or shortly after the funeral. A complete unguentarium (11, Fig. 2.94) was found in two pieces a short distance apart on top of the silt layer ca. 0.50 m south of the closing slab. The proximity of the vessel to the grave, its functional association with funerary rituals, and the fact that it has the same depositional relationship with the silt as the slab and the skeletons support its identification with the burial context. The chipped rim and the hairline fracture at the base of the neck indicate that the vessel snapped cleanly at its weakest point when it fell onto the hardened silt from above. An intact but worn lamp (6, Fig. 2.89) was also found near the grave and the unguentarium, ca. 1.80 m south of the closing slab, where it had been placed on top of a shallow, loose pile of mortar and plaster debris. This debris had previously accumulated at the juncture of the channels when the floor of room II was torn opened, but it was cleared back beneath the wall to the south when the east–west channel was converted into a grave. The lamp’s intact state differentiates it from the numerous lamps that were dumped as refuse in the northern rooms of the Bath during the early 5th century, all of which were highly fragmentary.134 Moreover, its proximity supports its association with the interment. The obliteration of surface decoration resulted from weathering as the lamp was exposed to the elements before the drain was completely filled. Behind the closing slab, the east–west channel contained the three skeletons (76-002A– C) and a few personal articles. The eastern (C) and western (A) skeletons were supine and extended. The head of A was turned on its left side, the arms were crossed over the pelvis, the right leg was bent, and the left leg was straight. The head of C had been displaced since burial, but the arms appear to have been crossed over the pelvis, and the legs were straight but pulled together at the heels. Between these two skeletons was a pile of bones including cranial fragments and a mandible in the west and two innominates lying together but disarticulated to the east. These remains represent a single skeleton (B) that was collected into a heap at some point after the initial interment of the body and the disintegration of its soft tissues, which would have occurred within roughly a decade. The rearrangement of 132. Gregory 1995, pp. 289–290; cf. Karivieri 1996, p. 50. 133. NB MMT 1976, pp. 32–35, 39–41. 134. Wohl 1981, pp. 119–121, 140; Corinth XVIII.2, p. 20, n. 82. Two intact lamps of earlier date were found in the fill in

the north–south channel under room VI: IPL 90-1, early 5th century (NB 3, p. 107); IPL 90-6, late 3rd to early 4th centuries (NB 5, p. 78).

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.64. Roman Bath, actual-state plan

79

80

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.65. Slab and debris enclosing channel

with RB 76-002, from northwest

the defleshed bones was not without method. The skull was intentionally placed along the south wall in the west, while the postcranial bones were piled together to the east. This was accomplished before the deposition of the two other articulated skeletons, because the one to the east (C) was placed slightly over the bones in the middle (B), and the right leg of the one to the west (A) was bent sharply inward so as not to disturb the head in the middle (B). The relative chronology of the two later bodies (A, C) is unclear, but they might have been deposited in close succession. It is hard to imagine that mourners would have bothered to place the eastern body (C) so far inside the tight channel if there was space just inside the opening, unless they were already planning to deposit a third body there (A). The unusual placement of the two individuals around the pile of collected bones and all three close to the opening into room II should be understood as the most practical solution to the channel’s narrow confinement. A bone button (29, Fig. 2.111) and a circular patch of textile or leather were found roughly in the middle of the grave.135 It is not, however, clear whether they should be associated with the middle body. The skeletal remains are very poorly preserved (14.45% of expected bone elements) because of their location in the drain. One skeleton was shielded from weathering, and one was partially exposed to it. The skeleton a few meters to the east of the opening (C) was well protected by the walls, floor, and ceiling of the channel. In contrast, the skeleton immediately inside the incomplete closure (A) was directly exposed to air, water, and sediments infiltrating the burial space. Due to the dry environment and thorough aeration of the channel, the bones of C are desiccated, porous, and powdery, and they are beginning to show thin, longitudinal fractures (Fig. 2.67 left). In contrast, the bones of A, which would have been moistened by seasonal rains until the opening to the channel was buried, are coated with a thin film of clay. The bones display not only slight lengthwise splitting but also marginal flaking, cortical crumbling, and ragged pockmarking, especially on the flat elements (Fig. 2.67 right).136 The individual interred to the west near the opening (A) was a female of 35 to 44 years with dental wear and a periodontal cyst (Fig. 5.29). The individual redeposited between the other skeletons (B) was probably a male of 35 to 44 years with a healed fracture of the right forearm, dental wear, and AMTL (Figs. 6.23, 7.25). The individual interred furthest east in the drain (C) was a female of 35 to 44 years with a healed bone infection in one leg, a small depressed fracture on the forehead, a fractured vertebra and joint disease in the lower back, dental wear, and AMTL (Figs. 5.30, 7.26–7.28). 135. The piece of textile or leather was not saved. 136. The middle skeleton (B) showed similar effects of weathering, but the process is more difficult to assess because

the bones had been moved when the other two individuals were interred.

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.66. RB 76-002

Figure 2.67. Differential preservation of femoral and tibial fragments

of RB 76-002C (left) and 76-002A (right), anterior view

81

82

THE GRAVES AND ASSOCIATED REMAINS

These burials predate the widespread destruction of the Bath in the late 6th century, when the channel was sealed under structural collapse, and they postdate the rupture of the channel, which should probably be assigned to the early 5th century. The skeletons, the closing slab, and the funerary artifacts were placed over the gray silt, which contained mostly pottery and architectural fragments from the Bath’s primary phase during the 2nd to 3rd centuries but also spirally grooved amphora sherds of the early to mid-5th century.137 The lamp outside the grave was manufactured in the late 5th to early 6th centuries; the unguentarium conforms to that date. The burial must therefore date to the late 5th to early 6th centuries, contemporary with or slightly later than Group II at the Northeast Gate. The presence of the grave inside the Bath further suggests that it succeeded the occupation of the area for the construction and initial employment of the Hexamilion, during which time the new drainage system seems to have been created.

Hexamilion Outworks The UCLA excavations in 1970 discovered two graves where the Hexamilion joins the north wall of the Roman Bath, one outside the building immediately opposite the northwest corner of room II and the other ca. 25 m northeast of room I in a corner of the curtain wall (Figs. 2.64, 2.68).138 The area north of the Bath was excavated in an exploratory fashion, and the present study cannot undertake a full evaluation of the complex sequence of building in the area. One monumental building of unknown function located northeast of the Bath was probably an earlier or contemporary construction. Other structures directly north of the Bath, which the excavators thought comprised a defensive outworks commanding the Ravine, a “gully bastion,” was either contemporary with or later than the construction of the Hexamilion.139 The Bath itself was largely destroyed at the end of the century by seismic catastrophe, when part of the roof and possibly the east wall collapsed. The building effort in the early 5th century incorporated the ruined Bath into the Hexamilion by buttressing the north walls of rooms I, II, and IV and adding an arcaded fighting platform. Traces of rubble walls, fires, numerous lamps, and a cache of storage amphoras in the ruined structure demonstrate that it was occupied periodically from the early 5th to mid-6th centuries. The effort to rehabilitate the fortifications under Justinian most likely involved the strengthening of the piers along the Bath’s north wall.140 Grave 1 (70-901)141 was an elliptical cist without covering or lining oriented south–north (186°) and measuring ca. 1.60 m long, ca. 1.10 m wide, and ca. 0.40 m deep (Fig. 2.69).142 It was located outside the Bath in the corner where room IV extends north beyond room II. The cist had been cut from a surface of light red silt (42.060 masl) immediately below the massive deposit of collapse debris containing tile, decomposed mortar, and rubble that largely constitutes the Byzantine to Modern subsurface strata north of the Bath. The grave contained the remains of three individuals (70-901A–C) in a disarticulated mass, with the skulls to the south near the wall and the leg bones lined up to the north. This was a secondary deposit of bones that did not include many skeletal elements. The situation of the heads to one side and the leg bones to the other reflects a concern to preserve a basic anatomical 137. The pottery from the silt is Lot 76-MMT-007. This sediment differs from the hard gray silt directly over the mosaic in room I (Wohl 1981, pp. 116–117; Gregory 1995, p. 287). 138. Clement 1974, pp. 109–111; 1977a, pp. 147–149, pl. 134; Isthmia V, pp. 41–44. 139. Isthmia V, pp. 41–43. 140. On the early-5th-century occupation of the Bath, see pp. 123–124. Isthmia V, pp. 38–41, and Gregory (1995, p. 286)

discuss the construction; Wohl (1981, pp. 116, 139, no. 47, n. 12) and Gregory (1995, p. 287) address the 6th-century pottery and lamps, which may be related to the activity of the renovations. 141. NB Gully Bastion 1970 II, pp. 41–45, 55 (trench 70-6). David Wilson supervised the excavations in the Hexamilion Outworks in 1970. 142. The dimensions measure the entire cist.

Figure 2.68. Hexamilion Outworks, actual-state plan

84

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.69. HO 70-901A–C, from north

arrangement. Their compact grouping and relative position indicate that they were deposited simultaneously. The skeletal remains are poorly preserved (19.81% of expected bone elements) because of the unprotected nature of the cist and the secondary nature of the burial. The crowns of numerous teeth, the innominates, the ends of the long bones, and the basicranial and posterior regions of the crania all display postmortem fractures as a result of transport from another locale. One individual (A) was a female of 35 to 44 years with healed fractures of the forearms, dental wear, and caries (Figs. 5.31, 6.24, 7.29). One individual (B) was a male of 35 to 44 years with dental wear, periodontal disease, and AMTL (Figs. 5.32, 6.25). One individual (C) was probably a male with enthesophytes in both lower legs, lethal fractures from a blow with a pointed object to the right side of the head, dental wear, caries, periodontal disease, and AMTL (Figs. 5.33, 6.26, 6.45, 7.30, 7.31).143 The best indication of the date of interment is stratigraphic context. The surface from which 70-901 was cut predates the massive accumulation of destruction debris north of the Bath, which began in the late 6th century. However, the surface itself contains a large amount of collapsed roofing material and window glass that came from the fenestration of rooms II and IV when the Bath was damaged at the end of the 4th century.144 Therefore, 70-901 was interred between these two periods of destruction, between the end of the 4th and the late 6th centuries. But it was a secondary burial, and the date and location of the primary burial and the reason for its displacement are unknown. One explanation is that the builders disturbed the original grave during the Theodosian project and then reinterred the bones where they would be safe from further disruption, outside but close to the completed fortification wall. Another explanation is that the original burial was made during a 143. Rife and Giesen (1994, pp. 233–234) incorrectly identify these three individuals with 70-902. 144. The pottery in the surface from which the grave was cut is lot 70-GB0-030 (4th to 6th centuries); the pottery in the lowest horizon of destruction debris immediately south and west of the grave is lot 70-GB0-031 (7th or 8th century), includ-

ing a “Slavic ware” cooking pot rim (IPB 70-6; Gregory 1993b, p. 153, no. 3). This surface is the same as the hardpan at a slightly higher elevation ca. 5.5 m to the east in trench 93-2 (NB 12, pp. 27–41, 55–148). There the pottery from the surface is lot 12-012 (Late Roman), and the pottery from the overlying destruction debris is lots 12-009 and 12-011 (7th century).

DESCRIPTION OF THE GRAVES BY AREA

85

period of relaxed military employment, such as the late 5th century, but was then disturbed and relocated during the mid-6th-century renovations in the vicinity. Grave 2 (70-902)145 was a tile-lined, tile-covered cist with two compartments oriented west–east (279°) and measuring ca. 2.15 m long, 1.27–1.35 m wide, and 1.30 m deep (Figs. 2.70–2.72). It had been cut from a surface (41.480 masl) into the marl bed north of the Bath inside the corner where the northward course of the Hexamilion returns east at a right angle. At ca. 0.40 m below the top of the cutting for the cist was a heap of about 12 fragmentary tiles and bricks, including a hypocaust element presumably from the Bath, which had originally enclosed the southern half of the grave. Below this, the ashlar facing of the Hexamilion constituted the west and north walls of the cist, while the south and east sides were partially lined with tiles and bricks set upright, some intact and some broken, with a few stones at the base. Here the cist was divided lengthwise by upright tiles into two rectangular compartments. The one to the south (ca. 1.80 x ca. 0.50–0.70 x D. ca. 0.85 m) was enclosed by the layer of tile fragments and stones, and the one to the north (ca. 1.80 x ca. 0.60–0.70 x D. ca. 0.40 m), which extended 0.12–0.13 m under the Hexamilion, was enclosed at a slightly lower elevation by at least three large tile fragments arranged in an east– west line.146 The middle portion of the north wall of the north compartment extended ca. 0.10–0.20 m beneath the lowest course of foundation blocks in the Hexamilion. This slight undercutting resulted from the collapse of a small portion of the marl under the massive masonry when the north compartment was excavated. The north compartment contained two skeletons, and the south compartment contained eight. The bodies in both compartments lay one on top of the other, and all were supine and extended. At least three had their heads turned partly to the left, one had the head turned on the left side, and at least three had their arms straight. The skeletons in the south compartment were crowded up against the tiles separating the two compartments, while the skeletons to the north were closer to the center of their burial space. Below the eight bodies on the floor of the south compartment were a coarse-ware bowl (10, Fig. 2.93) in the west end, a bronze buckle (24, Fig. 2.106), and in the middle a stone bead on a twisted glass rod (25, Fig. 2.107). The bowl had been placed in the grave before the deposition of the bodies, while the small objects had been worn by the deceased for burial but filtered to the bottom of the cist after the soft tissues had decomposed. The grave also contained iron disks and rings (26, 27, Figs. 2.108, 2.109), probably accessories worn by the deceased, and an iron nail (28, Fig. 2.110) of unknown function, perhaps a hobnail. The discrepancy in elevation between the two coverings and the different situations of the bodies in the two compartments suggest that they represent two burial phases, each involving multiple openings for continued interment. It is, however, unclear which of the compartments is earlier. In one scenario, the bodies to the south were interred first. As more bodies were added to the north, the older bones were crowded to the south and segregated from the more recent interments by the wall of tiles. In an alternative scenario, the southern compartment, which was higher in the cist, was a less spacious addition to the northern compartment, and therefore the only possible direction for expansion from that point was south into the marl. This seems to be the more likely of the two explanations. Since the skeletons were not recorded and collected individually, the exact situation of specific elements in the grave is unknown. The confusion of bones from 10 individuals after their removal called for a unique system to identify separate elements. Individuals were designated with a letter (A–J) only if bones could be matched with high certainty. Mandibles 145. NBs Gully Bastion 1970 I, pp. 177–179, and Gully Bastion II, pp. 6, 27, 47–53, 59–61, 64–79 (trench 70-4); NB 10, pp. 51–53, 56, 96–97 (trench 94-2); Marty Peppers 1979, pp. 295–298 (group L); Clement 1974, pp. 109–110; Isthmia V,

pp. 43–44. 146. Of the Roman tiles and bricks used to cover and line the grave, seven were inventoried (IM 70-49, 70-90, 70-91, 7092, 70-93, 70-94, 70-95).

86

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.70. Covering of HO 70-902,

north at left

were paired with crania, crania with vertebrae, and vertebrae with sacra and innominates on the basis of coincident wear patterns, tight articulation, relative size and shape, contiguous osteophytosis, and age- and sex-related features.147 In the end, only the subadult remains and the adult skulls, vertebral columns, and pelvic bones could be assigned letters as individuals. The numerous thoracic and appendicular bones that could not be positively associated with the adult skulls, vertebrae, and pelvic remains were identified generically as 70-902Z. Then they were grouped according to type (e.g., radii, calcanei, clavicles), and all surviving elements of a particular type were given a serial number (1–7). Consequently, bones of different types with the same serial number did not necessarily belong to the same individual; they merely occupied the same position in the counting sequence. For example, radius 70-902Z2 was not necessarily from the same skeleton as calcaneus 70-902Z2. They were the second radius and the second calcaneus examined of all radii and calcanei recovered from the grave. This idiosyncratic system allows for the unique identification of single skeletal elements when a preferable system based on individual skeletons is impossible because of irreparably commingled remains. Apart from their confusion during recovery, the skeletal remains from this multiple burial are poorly preserved (22.09% of expected bone elements) because of several factors. The grave was cut into the marl in an area of drainage in a corner of the Hexamilion. The cist 147. Simple correspondence in appearance was not a primary criterion because burial and storage can distort the color and texture of different bones from the same skeleton.

DESCRIPTION OF THE GRAVES BY AREA

87

Figure 2.71. Top three skeletons in HO 70-902,

Figure 2.72. Bottom seven skeletons in HO 70-902,

north at left

north at left

was incompletely protected and disturbed by plant roots. Finally, the excavators damaged bones during handling and discarded others in the spoilage.148 The remains of the three children could be segregated and identified as individual skeletons. One child was within the first year (A), another was two to four years old (B), and another with slight dental wear was five to six years old (C). Enough bones were preserved and could be identified with individual skeletons to determine age and sex for five adults, while the pelvic remains from the remaining two were too fragmentary to permit a confident evaluation of sex and age. One individual (D) was an adult of unknown sex over 25 years with dental wear and a chipped tooth, calculus, periodontal disease, and AMTL (Figs. 5.34, 6.27). One individual (E) was an adult of unknown sex over 25 years with dental wear, calculus, caries, periodontal disease, and AMTL (Fig. 6.28). One individual (F) was an adult of unknown sex over 25 years with dental wear and caries (Figs. 5.35, 6.29). One individual (G) was a female in her 50s or older with a sacralized coccyx and joint disease throughout the back (Fig. 7.32). One individual (H) was a male of 35 to 44 years with a Schmorl’s node in one vertebra, a small depressed fracture of the side of the head, advanced dental wear, calculus, periodontal disease, and AMTL (Figs. 5.36, 6.30, 7.33). One individual (I) was an adult of unknown sex over 15 years with enamel hypoplasia, dental wear, calculus, and caries (Figs. 5.37, 6.31). One individual (J) was a female of 15 to 24 years with a lumbarized sacrum and a sacralized coccyx, joint disease in her back, dental wear, and caries (Figs. 5.38, 6.32, 7.34). Either D or I was a male of 25 148. See pp. 239, 245.

88

THE GRAVES AND ASSOCIATED REMAINS

to 34 years, and either E or F was a male over 55 years. Among the commingled remains that could not be identified with individual skeletons were observed the following conditions: a sternal aperture; a sacralized coccyx; a healed bone infection in one leg; joint disease in the back, shoulder, elbow, ankle and ribs, ossified rib and xiphoid cartilage, and a Schmorl’s node in one vertebrae; a severe chronic ankle sprain and healed fractures of a rib and a scapula; and enamel hypoplasia, dental wear, calculus, and caries (Figs. 6.35, 7.35–7.41, 7.43–7.45). A date of interment in the early 5th century can be derived from architectural associations and funerary artifacts. The burial postdated the construction of the Hexamilion in ca. 410–420, while the bowl (10) and personal articles (24, 25) interred with the bodies point to the early part of the century. The coincidental artifacts in the burial fill support this date (see p. 102). The close dating of coincidental and incidental artifacts suggests that individual burials were not separated by long intervals. Furthermore, most or all of the skeletons were found articulated. This demonstrates that the bodies that were moved north or south when the cist was compartmentalized had not yet skeletonized and therefore had been buried for less than roughly a decade. The motivation for the separation of the bodies into separate spaces is unknown. The grave design and funerary artifacts do not suggest a social distinction, but the segregation may well reflect familial relations.149 The date of interment associates these 10 individuals with the initial employment of the fortifications. They were soon forgotten, because subsequent residents built a small oven on a floor roughly 0.10 m above the top of the grave.150

West Foundation The University of Chicago excavations in 1962 discovered a grave on the northeast side of the West Foundation. The structure is located ca. 2 km southwest of the Temple of Poseidon and ca. 70 m north of the modern road between Kyras Vrysi and Archaia Korinthos (Fig. 1.1). This monument of unknown function was comprised of a large rectangular foundation of ashlars, a sizable ash pit, and a deposit of burned offerings (Fig. 2.73). The grave (62-001)151 was a stone-lined, tile-covered cist oriented west–east (272°) and measuring ca. 1.65 m long, ca. 1.50 m wide, and ca. 0.40 m deep (Fig. 2.74).152 It had been cut through the dense stratum of irregular stones throughout the area enclosed by the walls on the south, east, and west and by a modern wall on the north. The grave was surrounded by the rubble on all sides except around the head of the deceased. The northern margin of the cist was covered by four complete Corinthian pantiles, which were broken after their placement (Fig. 2.75).153 The tiles to the west were positioned at a steep angle above the head and torso of the deceased, as though to keep the rubble from collapsing into the grave, while the easternmost tile was set almost flat over the ankles and feet. Apparently the west end of the cist and the entire middle and southern portions of the cutting were left exposed. The grave contained a single skeleton positioned on its right side (62-001A). The arms were bent back sharply in front of the chest, and the legs were semiflexed to the right at a 20° angle 149. For further discussion, see pp. 220–221. 150. NB Gully Bastion 1970 I, pp. 169–171, 177. The oven cannot predate the 6th century on the basis of an associated lamp fragment (IPL 70-90). It was not a funerary feature. 151. NB 43, pp. 114, 120–121, 124–125 (Section P); Isthmia II, p. 121, pls. 47:c, d, 78. Oscar Broneer supervised the excavation of the grave. For a full acount of the excavations and remains at the West Foundation, see Isthmia II, pp. 117–122. 152. The length and width measurements include the tile

covering; the depth is a measurement of the cist below the tile covering. 153. Tiles (IT 811, 812, 813, 814): L. 0.54, W. 0.35–0.44 m. Four complete Corinthian pantiles in several fragments. Flat, square terracotta tiles with slightly raised edges, all of the same form. The lower end narrows to 0.395 m at ca. 0.075 m from the lower end, where the edges terminate. Corners at the lower end are cut back to 0.353 m to fit inside the edges at the upper end of the overlapping tile. Isthmia II, p. 121, pl. 47:d.

DESCRIPTION OF THE GRAVES BY AREA

89

Figure 2.73. West Foundation, actual-state plan and sections

to the vertebral axis. The skeletal remains are poorly preserved (26.61% of expected bone elements) because of the incomplete enclosure of the cist. The individual interred here was a female over 35 years with crowded incisors, dental wear, caries, and periodontal and periapical lesions (Figs. 5.39, 6.33, 6.40, 6.46). This burial was interred long after the monument had stopped serving the purpose for which it was erected in the mid-4th century b.c.154 Most of the finds from the deposits above its tiles down to the bedrock in the immediate area of the grave were sherds of Classical date that were either residual in the fill or had been churned up from earlier deposits by later disturbance.155 The basic form of the tile-covered grave was used in the Corinthia and Argolid from the Classical to the Byzantine eras, but the distinct shape of the tiles enclosing the cist at the West Foundation—flat and square with raised edges tapering evenly at one end—belongs to the Early to Middle Roman periods.156 They were probably manufactured 154. O. Broneer (1962a, pp. 16–18; Isthmia II, pp. 121–122) hypothesized that the West Foundation supported a large exedra flanked with two wings that functioned as a “grandstand” facing south to the Isthmian hippodrome, which has yet to be discovered. See more recently OCD3, p. 772, s.v. Isthmia (C. A. Morgan; “hippodrome and hero shrine”) and Isthmia VIII, p. 435 (“It was probably a cenotaph, or a hero shrine, although many points of interpretation remain obscure”). 155. Pottery lots 1357, 1392, 1447. Excavation to the south-

east uncovered Roman pottery and a coin of Marcus Aurelius (Isthmia II, p. 121), but the depositional relationship between these remains and the grave is uncertain. 156. A tile of comparable shape and size covered a grave of the early 1st century in the ancient cemetery along the coastal road south from the harbor at Kenchreai toward Loutra Elenis (Corinth NB 249, pp. 31, 34–35; Robinson 1966, pp. 179–180, n. 2). For a full discussion of tile-covered graves in the northeastern Peloponnese, see p. 176.

90

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.74. WF 62-001A in grave, from south

Figure 2.75. Reconstruction of tiles (IT 811–814)

enclosing WF 62-001 after removal from cist

during the same era as the interment rather than recycled from an earlier context because they were placed in the grave intact. Furthermore, the unusual position of the body, lying on the right side with limbs flexed, demonstrates that the burial predated the numerous Late Roman interments on the Isthmus in which the arrangement of the deceased was typically supine and extended (see pp. 183–187). Therefore, the burial was not made after the 4th century and should probably be placed in the 1st to 2nd centuries. Although the excavators of 62-001 focused on monumental architecture, investigations over a larger area have elucidated the topography surrounding this grave. Excavation ca. 13 m south of the West Foundation revealed a long stretch of deep, well-worn wheelruts running east–west. This was most likely the main road between the Isthmian Sanctuary and Corinth.157 It is clear that the road operated for a long duration and undoubtedly was in use when 62-001A was buried. By that point, several centuries after the construction and operation of the West Foundation, the site would have been a low mound of fallen stones and earth. Recent surface survey in the olive grove east of the West Foundation has recorded a high density of potsherds of Archaic to Late Roman date, including several pieces of Roman fine ware and low to moderate concentrations of tile and cut stone. These remains indicate 157. Isthmia II, pp. 88, 122, pls. 48:b, 81.

DESCRIPTION OF THE GRAVES BY AREA

91

long-term, localized activity and perhaps settlement along the roadside during classical antiquity, including the period when the grave was made.158 It is reasonable to propose, though it cannot be proven, that the old woman who was interred among the ruins of the West Foundation had once lived nearby, perhaps in a farmstead or in a small community to the east of the monument.

Palaimonion The University of Chicago excavations in 1956 discovered a grave that had been cut into the foundations of the Temple of Palaimon after the building had fallen into disuse and despoliation. The areas and buildings associated with the cult of Melikertes-Palaimon developed throughout the history of the Roman Sanctuary. During the mid-2nd to early 3rd centuries the latest phase of the Palaimonion was located southeast of the temenos of Poseidon over the west end of the Early Stadium (Figs. 2.76, 2.77). The architectural complex consisted of three elements in an east–west sequence: a rectangular area to the east with a pit for holocaustic sacrifices in the center (Pit C), a passage with an entrance into the South Stoa at the north end and a small room at the south end, and a precinct to the west, which was approximately the same size as the eastern area. At the west end of the precinct was a long, underground reservoir that had originally supplied water for the Stadium but in the Roman era was interpreted as the sacred enclosure of Palaimon. The temple, a small monopteros (Diam. ca. 7 m), was built so as to give access to the enclosure. The roughly square podium, which consisted of a five-stepped core of opus incertum finished with ashlar masonry, had an entrance on the east side into a passageway that led through the foundations to the ancient reservoir.159 The grave (56-001)160 was situated in the northeastern quadrant of the monument’s foundation and extended out through the northern section of its façade (Figs. 2.77, 2.78). The cist was cut from above and to the east after the destruction of the temple’s superstructure and the removal of many blocks of the crepidoma. A triangular hole was cut into the exposed rubble and brick concrete foundation that widened into a more elliptical cutting ca. 0.80 m below the preserved top of the foundation. This irregular cutting formed a low concavity over the west end of the cist while the east end lay outside the concrete mass. The exact ground level from which the cist was cut cannot be ascertained. Recent study of the stratigraphy in this area based on the record of excavation in the field notebooks has shown that a distinct surface existed near the top of the dismembered structure, leaving only the uppermost ca. 0.20–0.50 m exposed.161 It is unknown whether the grave was cut while the surface was in use or earlier, before it formed. If the grave was cut from the ground level that existed during the removal of the temple’s facing blocks, the concrete core would have been visible to a height of ca. 0.80–1.00 m. But if the grave was cut from the distinct surface near the top of the foundation, only a small amount would have been above ground. Thus, while it is hard to know how much, some portion of the stripped monument was visible 158. The results of a surface survey in 2000 will appear in the final publication of the Eastern Korinthia Archaeological Survey, under the direction of T. E. Gregory, D. Pullen, and T. Tartaton; see in general Tartaron et al. 2006. The highest concentration of cultural debris immediately east of the West Foundation was designated LOCA 9064, though several survey units in the area (1254, 1258, 1260, 1834) produced moderate to high densities. Pettegrew (2008, p. 262, fig. 14.5) stresses the existence of concentrated settlement during the Early and Late Roman periods throughout the area west of Kyras Vrysi. 159. Broneer (Isthmia II, pp. 99–112) describes the remains

of the precinct of Palaimon and (pp. 27, 110, pls. 13:a, b, 22:a, b, 41:b, c, plans IV, IX) specifically addresses the Stadium Reservoir; Gebhard (2005, pp. 190–191, 197–200, fig. 6.7) gives the current understanding of the plan. 160. NB 9, pp. 75–78, 80–82 (trenches V, VI); Isthmia II, pl. 41:a, b, plans VII, IX. Broneer (1958, pp. 15–16, pl. 7:a) records the initial exploration of this area. Françoise Rosen supervised the excavation of the grave. 161. The author thanks Elizabeth Gebhard for her instructive discussions concerning the stratigraphy and chronology of the area.

92

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.76. Central area of the Sanctuary during its Antonine phase

when the grave was made. Moreover, the adjacent passageway seems to have been mostly or completely open at the time (see below). The grave was an unlined, covered cist oriented west–east (268°) and measuring 1.52 m long, 0.48–0.52 m wide, and 0.46 m deep. The cist was covered by three slabs that rested on a relatively level surface at the top of the monument’s bottom step (Figs. 2.78, 2.79, 2.81). The slabs were roughly dressed stones of varying sizes that had been cut back around the edges to fit the cist closely.162 A deep, rectangular clamp cutting in the middle slab shows that the stone was reused. Three large, square stones had been placed under the eastern edge of the eastern slab to support it and to create an even plane across the top of the covering. In order to close the gaps between the cover slabs and the walls of the cutting, 20–25 stones were placed along the northern and southern margins of the cist. A dense layer of hard-packed fill and small stones covered the cist’s far west end, adjacent to and at the same level as the western slab. Finally, strips of mortar were liberally applied (Th. ca. 0.01–0.03 m) across most of the top of the western slab and over the joint between the western and middle slabs (W. ca. 0.84 m), over the joint between the middle and eastern slabs (W. 0.06–0.09 m), and along the southern edge of the eastern slab (W. 0.02–0.04 m). Those who created the grave thus invested considerable effort in arranging the slabs, stones, and mortar to create a tightly sealed covering. The enclosed cist was not centered below the slabs but offset to the 162. The slabs, from west to east, measured: 0.66 x 0.51 x 0.13–0.18 m; 0.89 x 0.48 x 0.12–0.16 m; 0.86 x 0.49 x 0.08– 0.13 m.

DESCRIPTION OF THE GRAVES BY AREA

93

Figure 2.77. Palaimonion, actual-state plan and sections

north. The burial compartment was cleanly cut into the concrete so that the floor was level, the walls were mostly vertical, and the corners were slightly rounded. This undisturbed burial contained several funerary artifacts and a single skeleton (Figs. 2.80, 2.81). In the fill six large iron nails were found, one at each corner of the cist and two near the middle; others may have disintegrated before discovery of the grave.163 A piece of wood and fragments of plaster, one preserving the impression of wood grain, were also found near the middle of the cist at roughly the same level as the bones or slightly below. The nails and wood indicate that the cist contained a coffin or bier, while the plaster shows that either this wooden structure or the cist itself was lined with plaster that was still wet when the wood came into contact with it. The small quantity of plaster from the cist indicates that it more likely coated the wooden structure. The precise design of the coffin or bier is unclear because of the poor condition of the materials, which disintegrated after discovery. The grave also contained objects that had adorned the deceased during the funeral. A bronze earring (17, Fig. 2.99) was found ca. 0.20 m from the west wall and ca. 0.15 m from the south wall. Eight beads situated in a circle were located ca. 0.20 m further east and slightly to the north, and five others of similar size and appearance were recovered separately (33a–e). The location of these artifacts relative to the bones shows that the deceased had worn the ring in the right ear and a string of beads around the neck. Although the skeletal 163. One typical nail had a shank measuring 0.082 m long and a head measuring 0.015 m in diameter.

94

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.78. Temple of Palaimon during excavation

Figure 2.79. Covering of PAL 56-001, from east

in 1956, from east

Figure 2.80. PAL 56-001A with earring (17) and beads (33), north at bottom

remains (56-001A) were too poorly preserved to recover, sketches and photographs of the grave show that the body was supine with straight upper arms and legs. The skeletal remains (56-001A) could not be recovered intact and were only preserved in small pieces. Fragments of long bones and pulverized bone faintly tracing the outline of the skeleton disintegrated once the grave was opened. The enclosure of the skeleton within the solid concrete walls and under the sealed cover failed to prevent the destruction of the bone. Apparently water, air, and organisms entered the burial environment at the time of interment and perhaps later through small gaps in the covering. Atmospheric, residual, or seeping moisture within the cist then accelerated the decomposition of the wood and promoted bacterial and fungal activity. The rotting wood in turn would have broken down the contiguous bones.164 The size of the cutting and the presence of jewelry suggest that 164. On the effects of decaying wood on soft and hard tissues in archaeological contexts, see Henderson 1987, p. 51.

DESCRIPTION OF THE GRAVES BY AREA

Figure 2.81. PAL 56-001

95

96

THE GRAVES AND ASSOCIATED REMAINS

the deceased was an adolescent or adult female, but this identification has no osteological basis. A general date of interment can be determined from the grave’s structural and stratigraphic associations. The precinct of Palaimon developed in four distinct phases (I–III, V) after the Roman revival of the Isthmian Sanctuary in the mid-1st century.165 At that time, a pit for sacrifices (A) was opened inside a small precinct southeast of the Temple of Poseidon, and the easternmost manhole into the Stadium reservoir became a focus of cult activity (phase I). In the late 1st century, a second pit (B) was opened near Pit A within a larger precinct, and a Hadrianic Temple of Palaimon was erected to the east of the Temple of Poseidon (phase II). A third pit (C) was opened to the south inside the eastern of two conjoining precincts in ca. 130–150 (phase III). Finally, the erection of the second Temple of Palaimon in the western precinct, also a monopteros but slightly larger than its Hadrianic predecessor (Fig. 2.76), can be placed on numismatic grounds around the reign of Lucius Verus (161–169; phase V). The pottery and lamps from the uppermost fill in Pit C indicate that sacrifices and dedications continued well into the next century, until ca. 240.166 At that point the traditional sacrificial rite was interrupted, and it is unclear how long thereafter and in what capacity cult activity continued. The interment would not have been made in the facade of the temple when sacrifices were still observed, so it must postdate the early 3rd century. Furthermore, the cutting for the grave was made after the blocks facing the podium had been removed, probably for use in building the Hexamilion around the first decades of the 5th century. This is a terminus post quem for the burial. After the accumulation of as much as 1 m of sediment, up to the surface near the top of the concrete, the remnants of the podium, the grave, and the surrounding area were covered by a widespread, dense layer of rubble and marble fragments. This massive deposit filled the tunnel through the temple down to its floor (Fig. 2.78). The debris contained artifacts of the late 5th to late 6th or 7th centuries167 and may have originated from building activity at the fortifications in the mid-6th century. The unusual burial in the foundations of the dilapidated monopteros can therefore be generally placed in the early 5th to mid-6th centuries. It may well be contemporary with other graves associated with Late Roman fortifications (NEG Group II, RB 76-002, HO 70-901, 70-902).

Theater Cave The University of Chicago excavations in 1960 discovered human bones in chamber II of the Theater Cave, located along the southern edge of the cavea. This cave, which seems to have been a natural formation in the marl bed, was situated just east of the axis of the Theater, where the concrete foundations of the southeast corner of the Roman cavea would later be set (Figs. 2.76, 2.82). In Classical times it was expanded to accommodate two rectangular chambers with forecourts and rock-cut furniture. The chambers were reused in the Roman era, after which time they were apparently open for centuries and gradually filled with sediment. The exploration of the eastern chamber uncovered a jumbled heap of bones high in the fill. These were neither demarcated nor enclosed as a formal interment. 165. For a discussion of the archaeological history of the Roman hero-cult, see Gebhard 1993a, pp. 85, 89–93; Gebhard, Hemans, and Hayes 1998, pp. 416–417, 428–433, 436–444; Gebhard 2005, pp. 189–203; Isthmia VIII, pp. 17–18. Excavation in 1989 and subsequent study have permitted a revision of the original dating and construction sequence by O. Broneer (see n. 159, above). 166. Gebhard, Hemans, and Hayes 1998, pp. 443–444 (de-

posit II.4), 446, 453 (nos. 42–49); Hayes 1993, p. 114; Isthmia VIII, p. 18 (episode VII). 167. Pottery lots 645, 746, 749, 751, 754, 760, 762, 1184, 1262, 1297 from trenches II–VI and PIa–d. The massive deposit in the passageway through the Temple of Palaimon seen in Figure 2.78 (NB 9, p. 74) is pottery lot 751, dated early to mid6th century by Phocaean ware form 3H.

DESCRIPTION OF THE GRAVES BY AREA

97

Figure 2.82. Theater Cave, actual-state plan

The burial (60-001)168 was located in the southeast corner of chamber ca. 0.20 m below the top of the coarse fill of the cave. It was placed in a shallow, elliptical pit oriented south– north (152°) and measuring 1.55 m long, ca. 0.65 m wide, and 0.60 m deep (Fig. 2.83).169 The cranium was located against the south wall of the chamber face down; the detached mandible was found ca. 0.40 m to the north. In the area of the mandible and further north, particularly along the east edge of the pile, were found the long bones of an arm, one set of which was nearly articulated, and a disarticulated pelvic girdle. Almost no smaller bones were found. Mixed in with these human remains were assorted debris in no particular order: sizable, uncut chunks of limestone, large tile fragments, and numerous animal bones, including an equid metatarsus and a large mammal (probably equid) innominate. Of the human bones (60-001A), only the cranium (7.73% of expected bone elements), which was in very good condition at the time of discovery, was saved. On the basis of cranial morphology and the gross pelvic morphology visible in the photograph of the bones in situ, the individual is identified as an adult female of 15 to 34 years with benign osteomata on her cranial vault, enamel hypoplasia, dental wear, calculus, caries, and periodontal and periapical lesions (Figs. 5.40, 6.34). Although it is impossible to know for certain, all human bones in this location probably belonged to the same skeleton. The depositional context and composition of the interment suggest that it was a secondary burial of recent date. The woman was buried in the east chamber long after the Cave had 168. NB 33, pp. 210–212 (“Chamber III”). Broneer (1962a, pp. 5–7) describes the exploration of the Cave; Isthmia II (pp. 39–40, pls. 19:c, 57:a, b, 58:b [“Chamber II”]) describes in full the remains of the eastern chambers. Ione Mylonas Shear

supervised the excavation of the burial. 169. The dimensions are measurements of the area over which the bones spread since the cutting was indistinct.

98

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.83. TC 60-001A, from southwest

gone out of use. Ritual dining there ceased in the late 4th century b.c., and the chambers were apparently used for storage during the 2nd to 4th centuries a.d. At some point in the Roman era, a descending stairway (W. 0.95 m) was built in the entrance to chamber II and a hole (Diam. 0.63–1.05 m) was cut through the southwest corner of the ceiling and floor into a water channel below.170 From that time, debris accumulated throughout this chamber and adjacent ones until the time of excavation. Due to the situation of the bones in the uppermost horizon of the deep fill, it seems reasonable to propose, with due caution in the absence of conclusive chronological markers, a date of interment sometime between the Late Byzantine and Early Modern periods (15th to 19th centuries). Moreover, the absence of all the skeleton’s smaller bones and the disarticulated state of the remains show that this was a secondary burial. It is, however, noteworthy that the situation of the bones was not entirely random. The placement of the head at the south end, the arm bones together nearby, and the two innominates near each other to the north of the head reflects a rudimentary sense of skeletal anatomy. It is unknown how long after death and initial interment the secondary burial was made. At some point after the soft tissues of the body had decayed someone collected up the major bones of the head, pelvis, and limbs, together with a few bones from a horse or donkey and possibly other animals,171 and placed them in the back corner of the cave. No other graves were found in or around the Theater, which suggests that this was an isolated occurrence that did not belong to a larger burial zone.172 170. Isthmia II, pp. 40–46; Gebhard 1973, p. 3; Isthmia VIII, p. 18. Gebhard (2002) presents a general discussion of the chronology and function of the cult caves at the Isthmus. 171. Bones of cattle, sheep-goat, pig, chicken, dog, and horse (Equus caballus) were found in Late Roman deposits elsewhere in the cave chamber, both west of the burial and at the bottom of the hole (pottery lots 979, 997, and 1022, and bone lots 67, 70, and 74; NB 33, pp. 226, 320, 340). The author thanks David S. Reese for assistance and advice in examining

these remains. 172. Sparse human bones were found elsewhere in the Theater Cave complex in 1959: two thoracic vertebrae and one (human?) rib (bone lot 73; NB 25, p. 200); one thoracic vertebra and one mesosternum (bone lot 148; NB ERG, p. 165). The date and depositional context of these remains are uncertain. The author examined them in the Isthmia Museum in 2005.

CATALOGUE OF BURIAL DEPOSITS

99

CATALOGUE OF BURIAL DEPOSITS This catalogue describes the burial deposits by area and grave number.173 The graves are listed not numerically but rather by order of discovery. In most cases, the deposits are identified by one or more lot numbers because the excavators often segregated the material above and below the covering of the cist into distinct baskets. The artifacts from separate lots are classified as coincidental, incidental, or accidental after the definitions above.174 The artifacts in each category, both uninventoried and inventoried, are arranged by date and briefly described. The date and purpose of the incidental artifacts inside or above the graves are summarized here but fully described below in the catalogue of funerary artifacts.175 NORTHEAST GATE 69-103 Lot 69-NEG-009 above and around slabs, dated 4th to 6th centuries; lot 69-NEG-008 inside cist, dated late 4th to early 5th centuries (cf. Marty Peppers 1979, pp. 302–304, group N). Coincidental above and around slabs. Nine sherds total: one Early Roman (micaceous water jar body sherd); five Late Roman (one spirally grooved body sherd, two coarse-ware ring feet, one neck, one wheel-ridged body sherd); three Roman (coarse-ware body sherds). Coincidental inside cist. 10 sherds total: three Late Roman (two micaceous water jar body sherds, one wheel-ridged body sherd); seven undated (six coarseware body sherds, one cooking-ware body sherd); one red- and white-painted plaster fragment; IPL 69-23 (Athenian glazed lamp disk fragment, late 4th to early 5th centuries). Incidental inside cist. Fine-ware jar (9), late 4th to early 5th centuries, buried with bodies. Accidental. None. 69-004 Lot 69-NEG-012 above slabs, dated Late Roman; lot 69-NEG-013 inside cist, dated Late Roman. Coincidental above slabs. 23 sherds total: three Greek (two Geometric body sherds, one amphora rim); two Early Roman (micaceous water jar body sherds); four Late Roman (three cooking pot body sherds, one cooking pot rim); 14 undated (coarse-ware body sherds). Coincidental inside cist. Five sherds total: three Early Roman (one barbotine rim, one thin-walled cookingware pitcher rim, one thin-walled cooking-ware body sherd); two Late Roman (wheel-ridged body sherds); one tile fragment; one iron nail fragment (not saved), either coincidental or incidental. Incidental inside cist. Glass pitcher (7) and glass bottle (13, disintegrated) buried with bodies or deposited during later construction. Accidental. None. 173. The contents of deposits from T13 54-001 and the putative grave near Tower 14 are unknown. 174. See p. 23. 175. Although both skeletons and elements in the design

69-008 Lots 69-NEG-036, -037, -038, -056, and -057 inside cist, dated late 4th century with one intrusive Late Byzantine sherd. Coincidental inside cist. 107 sherds total: three Greek (black-glaze body sherds); one Middle Roman (Niederbieber 77 amphora handle); 18 Late Roman (two spirally grooved body sherds, 14 wheel-ridged body sherds, two African red-slip body sherds); 85 Roman (seven cooking pot body sherds, one ring foot, one handle, 76 coarse-ware body sherds); five tile fragments; one glass vessel body sherd; 24 red-painted plaster fragments; one iron nail fragment; IPL 69-131 (Athenian glazed globule lamp rim and disk fragment, 2nd to early 3rd centuries). Incidental outside cist. Lamp (5), late 4th century, placed over west end of covering during or after funeral. Incidental inside cist. Bronze ring (23), probably worn by deceased. Accidental inside cist. One Late Byzantine (glazed body sherd). 67-001 Lot 67-RMM-029 inside cist, dated late third to early fourth quarter of the 5th century (cf. Marty Peppers 1979, pp. 300–302, group M). Coincidental. 21 sherds total: one Early Roman (Eastern Sigillata B2 rim); seven Late Roman (five wheel-ridged body sherds, one red-slip base, one redslip body sherd); 13 undated (nine coarse-ware body sherds, one coarse-ware rim, one coarse-ware handle, two cooking-ware body sherds); one animal bone fragment; one tile fragment. Incidental. One bronze ring and 10 coins deposited with the uppermost bodies: Leo I (1), 457–474; Marcian (2), 450–457; illegible bronze (3), late 5th century; seven other Late Roman bronze (three illegible, three disintegrated, one not saved); coarse-ware cup (8) burof the interments, such as tiles, bricks, and spolia, in a strict sense belong to burial deposits, they have not been included here. They are covered in the foregoing description of the graves.

100

THE GRAVES AND ASSOCIATED REMAINS

ied with the lowermost bodies; two bronze earrings (16) and an iron buckle (18) worn by the deceased. Accidental. None. 67-003 Lot 67-RMM-027 inside cist, dated early 5th century; lot 67-RMM-079 in bottom of cist and below it, dated 3rd to 4th centuries. Coincidental inside cist. 24 sherds total: nine Late Roman (one spirally grooved body sherd, one micaceous water jar body sherd, three wheel-ridged body sherds, one wheel-ridged Palestinian amphora body sherd, one cooking pot rim, one red-slip body sherd, one coarse-ware basin rim); 13 undated (three cooking-ware body sherds, 10 coarse-ware body sherds); two glass vessel (one rim, one body sherd); IPL 67-28 (Athenian post-glaze lamp disk and rim fragment, early 5th century). Coincidental in bottom of cist and below it. 13 sherds total: one Greek (Protocorinthian body sherd); three Early to Middle Roman (one Eastern Sigillata B2 shoulder, one cooking pot rim, one Niederbieber 77 amphora body sherd); one glass vessel body sherd; IP 3739 (Broneer type XXVII lamp rim and disk fragments, signed ϹΕΚΟΥΝΔΟΥ, late 2nd to early 3rd centuries). Incidental. None. Accidental. None. 69-005 Lot 69-NEG-009A above slabs, mixed dates; lot 69NEG-015 inside cist, dated Late Roman with one intrusive Early Byzantine sherd (cf. Marty Peppers 1979, pp. 303–304, group P). Coincidental above slabs. 20 sherds total: four Early Roman (one thin-walled cooking-ware vase, two cooking pot rims, one red-slip body sherd); six Late Roman (two wheel-ridged body sherds, one Phocaean ware body sherd, two combed body sherds, one spirally grooved body sherd); three Roman (two coarse-ware handles, one coarse-ware base); seven Late Byzantine (four sgraffito body sherds, one cooking pot rim, one coarse-ware body sherd, one coarse-ware handle and rim). Coincidental inside cist. 16 sherds total: 12 Late Roman (two micaceous water jar body sherds, two micaceous water jar handles, three spirally grooved body sherds, four wheel-ridged body sherds, one wheel-ridged rim); four Roman (coarse-ware body sherds); one tile fragment. Incidental inside cist. Iron loop (22), probably a buckle or a pendant worn by the deceased; numerous small nails (not saved), probably hobnails worn by the deceased. Accidental inside cist. One Early Byzantine (“Slavic ware” cooking pot body sherd, 7th or 8th century).

69-001 Lot 69-NEG-022 above slabs, mixed dates; lot 69NEG-027 inside cist, dated Late Roman. Coincidental above slabs. 25 sherds total: two Greek (one pithos rim, one black-glaze skyphos handle); five Early Roman (one micaceous water jar body sherd, three coarse-ware rims, one Koan amphora toe); six Late Roman (wheel-ridged body sherd); 10 Roman (four coarse-ware handles, three flat coarse-ware bases, three cooking pot rims); one Early Byzantine (“Slavic ware” cooking pot rim, 7th or 8th century); one Late Byzantine (glazed body sherd). Coincidental inside cist. 16 sherds total: one Greek (black-glaze body sherd); three Early Roman (one redslip body sherd, two small coarse-ware rims); seven Late Roman (five wheel-ridged body sherds, one painted plate, one coarse-ware rim); five Roman (cooking pot body sherds); 11 tile fragments. Incidental inside cist. Pair of gold earrings (19) worn by the deceased. Accidental. None. 69-007 Lot 69-NEG-032 in cist above slabs, dated Late Roman. Coincidental. Four sherds total: three Late Roman (spirally grooved amphora body sherds); one Roman (coarse-ware body sherd); three tile fragments. Incidental. None. Accidental. None. 69-009 Lot 15-020 inside cist, undated. Coincidental. None. Incidental. Iron buckle (30), bronze loop (31), and shell bead (32) worn by the deceased. Accidental. None. 69-010 Lots 15-014, -015, -016, -017, and -018 inside cist, dated late 6th to very early 7th centuries. Coincidental. 68 sherds total: seven Early Roman (three micaceous water jar body sherds, one pitcher base with red-slipped interior, two red-slip body sherds, one Eastern Sigillata B body sherd); 61 undated, several probably Late Roman (11 cooking-ware body sherds, 47 coarse-ware body sherds, three coarse-ware body sherds); 68 tile fragments; one marble chip; one tessera; one green glass vessel body sherd; two human bones not associated with 69-010A (proximal epiphysis of radius, medial phalanx). Incidental. Cooking pot (12), late 6th to very early 7th centuries, probably broken during funeral. Accidental. None.

TOWER 2 68-006 No coincidental, incidental, or accidental finds. 68-002 No coincidental, incidental, or accidental finds.

68-003 Lot 68-T20-045 inside cist, dated Roman. Coincidental. Four Roman sherds (one slipped body, one coarse-ware body, two cooking-ware body). Incidental. None. Accidental. None.

CATALOGUE OF BURIAL DEPOSITS

101

TOWER 14 67-002 Lots 67-T14-071, dated 7th or 8th century, and 67T14-075, dated Late Roman, both inside cist. Coincidental. 44 sherds total: one Greek (amphora toe); one Early Roman (micaceous water jar body sherd); seven Late Roman (one spirally grooved body sherd, four wheel-ridged body sherds, one wheelridged shoulder and neck, one combed body sherd); three Early Byzantine (one “Slavic ware” cooking pot body sherd [7th or 8th century], one graffito body sherd with highly micaceous fabric [7th century], one amphora body sherd [7th century]); 30 undated (two semifine body sherds, 17 coarse-ware body sherds, two amphora body sherds, eight cooking-ware body sherds, one frying pan body sherd); 2 tile fragments; 2 iron fragments. Incidental. Bronze ring (15) worn by the deceased. Accidental. None. 67-004 Lot 67-T14-108 inside cist, dated 6th century. Coincidental. 13 sherds total: one Late Roman (brown-slip ring foot, 6th century); 12 undated, several probably Late Roman (10 coarse-ware body sherds, two cooking-ware body sherds, one overfired coarse-ware rim); one window glass fragment. Incidental. None. Accidental. None. 69-701 Lot 67-T14-134 above cist, mixed dates; lot 67-T14134A inside cist, dated Roman. Coincidental above cist. Six sherds total: five Early Roman (one black-slip rim, one black-slip body sherd, two Eastern Sigillata A body sherds, one Arretine body sherd); IPB 69-65 (proto-Majolica bowl rim fragment, 13th century); IM 3561 (unidentified architectural fragment); two glass vessel body sherds. Coincidental inside cist. One undated cooking-ware body sherd; six glass vessel body sherds, including one Early Roman ribbed bowl (cf. Kenchreai I, p. 136, pl. 56:e, 1st century). Incidental. None. Accidental. None. 69-991 Lot 69-T14-097 inside cist, dated 7th century with one intrusive Late Byzantine sherd. Coincidental. 26 sherds total: two Early Roman (one black-slip rim [Hellenistic or Early Roman], one plain Sigillata bowl rim [cf. Corinth XVIII.2, no. 76, Demeter and Kore, first half of the 2nd century]); 19 Late Roman (three coarse-ware bases, three coarse-ware handles, one micaceous water jar handle, one Phocaean

ware form 1A rim [late 4th to early 5th centuries], two spirally grooved body sherds, three wheel-ridged body sherds, two combed body sherds, one beehive sherd, two Late Roman 2 amphora handles, one coarse basin rim with folded, slightly concave rim in typical Late Roman 2 amphora fabric [5th to 7th centuries]); one Early Byzantine (African red-slip form 105 rim, cf. LRP, p. 167, no. 8, fig. 32, Porto Rafti, first half of the 7th century); four undated (coarse-ware body sherds). Incidental. Necklace with bone cross pendant and beads (20, 21) worn by the deceased; small metallic objects, some spherical (not saved), perhaps beads from the same necklace. Accidental. One Late Byzantine (glaze-painted plate rim, 14th to 15th centuries). 69-002 Lot 69-T14-119 inside cist, dated early 5th century. Coincidental. 48 sherds total: one Early Roman (cooking pot rim, cf. IPR 69-24 from Loukos Field, Early Roman); 22 Late Roman (six micaceous water jar body sherds, eight wheel-ridged body sherds, four spirally grooved body sherds, two Phocaean ware body sherds, one large basin with flaring rim [cf. IPR 67-1 from Northeast Gate, Isthmia V, pp. 70–74, n. 35, early 5th century], one coarse-ware ring foot in same fabric as basin); 25 undated (one cooking-ware body sherd, nine semifine body sherds, one coarse-ware rim, 14 coarseware body sherds); one glass vessel ring foot; seven tile fragments. Incidental. Coin of Theodosius I, Valentinian II, or Arcadius (14), 383–392, reused as pendant and worn by the deceased; bone and iron pins (not saved) worn by the deceased. Accidental. None. 69-003 Lot 69-T14-130 inside cist, dated 7th or 8th century. Coincidental. 62 sherds total: three Greek (one Corinthian body sherd, two black-glaze body sherds); one Early Roman (red-slip bowl rim with coarse, micaceous fabric); three Late Roman (one spirally grooved body sherd, one wheel-ridged body sherd, one coarse-ware rim); one Early Byzantine (combed “Slavic ware” body sherd, 7th or 8th century); 54 undated (20 semifine body sherds, 21 coarse-ware body sherds, 10 cookingware body sherds, one cooking-ware handle, one cooking-ware rim, one coarse-ware body sherd); six tile fragments; one Roman lamp nozzle fragment; one iron fragment. Incidental. Pentanummium (4), 6th century, deposited over the body. Accidental. None.

LOUKOS FIELD 69-801 Lots 14-031 and 14-034 inside cist, both dated 12th to 14th centuries.

Coincidental inside cist. 26 sherds total: eight Late Roman (six wheel-ridged body sherds, one micaceous water jar body sherd, one cooking-ware body sherd);

102

THE GRAVES AND ASSOCIATED REMAINS

three Late Byzantine (two plain-glaze body sherds, one plain-glaze rim); 15 undated (two cooking pot body sherds, two cooking-ware body sherds, two cookingware rims, one cooking-ware base, one cooking-ware handle, two coarse-ware handles, five coarse-ware body sherds); 15 tile fragments; three brick fragments; one

glass vessel body sherd; one iron fragment; two marble revetment fragments; one animal bone fragment (small mammal long bone shaft). Incidental. None. Accidental. None.

DECAUVILLE GRAVES 69-901 No coincidental, incidental, or accidental finds.

69-902 No coincidental, incidental, or accidental finds.

ROMAN BATH 76-002 Coincidental. None. Incidental outside grave. Lamp (6), late 5th to early 6th centuries, and unguentarium (11) deposited outside grave on floor of drain.

Incidental inside grave. Bone button (29) worn by the deceased; circular piece of textile or leather (not saved) worn by the deceased. Accidental. None.

HEXAMILION OUTWORKS 70-901 No coincidental, incidental, or accidental finds. 70-902 Lot 70-GB0-034 inside cist, dated late 4th to early 5th(?) centuries (cf. Marty Peppers 1979, pp. 295–298, group L). Coincidental. 18 total sherds: one Greek (black-glaze kylix body sherd); two Late Roman (micaceous wheelridged body sherd); 15 undated (seven cooking-ware body sherds, eight coarse-ware body sherds); two tile

fragments; IPL 70-100 (Athenian glazed lamp handle and rim fragment, late 4th century); IPL 70-105 (Athenian post-glaze lamp nozzle fragment, early 5th century?). Incidental. Coarse-ware bowl (10), late 4th century, interred with the bodies; bronze buckle (24) worn by one of the deceased; stone bead on a glass rod (25) worn by one of the deceased; iron disks (26) and iron rings (27), probably worn by one of the deceased; iron nail (28), perhaps a hobnail worn by one of the deceased. Accidental. None.

WEST FOUNDATION 62-001 Pottery lots 1357, 1392, and 1447 inside and around cist, dated Roman.

Coincidental inside and around cist. Numerous mixed Greek and Roman sherds. Incidental. None(?). Accidental. None(?).

PALAIMONION 56-001 Coincidental. None. Incidental inside cist. Six iron nails, plaster, and wood fragments (not saved) from coffin or bier; bronze ear-

ring (17) worn by the deceased; beaded necklace (33) worn by the deceased. Accidental. None.

THEATER CAVE 60-001 Coincidental inside grave. Several tile fragments; various animal bones, including equid metatarsus and large mammal (probably equid) innominate.

Incidental inside grave. None. Accidental inside grave. None.

CATALOGUE OF FUNERAR Y ARTIFACTS

103

CATALOGUE OF FUNERAR Y ARTIFACTS This catalogue gives descriptions of the funerary artifacts to accompany the discussion of the graves and the catalogue of burial deposits. These objects were intentionally placed inside or above the grave during or after the event of burial as offerings, articles of clothing or adornment, or ritual instruments.176 The entries are arranged first by functional type (coins, lamps, vessels, personal articles) and second by inventory number. COINS 1

(IC 885) Coin of Leo I Fig. 2.84 Obverse: head right, pearl-diademed. Reverse: monogram of Leo I in wreath. Mintmark [ ]C. Constantinople. 457–474. Cf. RIC X, p. 296, nos. 720, 721. Bibliography: Clement 1987, p. 382, pl. 60:d; Isthmia V, p. 77, n. 43. Buried in the middle of NEG 67-001 with the uppermost bodies in a small pile or purse (Figs. 2.12, 2.13). 2

(IC 888) Coin of Marcian Fig. 2.85 Obverse: [DN]M[ARCIANVS]FAVG, bust right, diademed, draped. Reverse: monogram of Marcian in wreath broken by exergual line. NI[CO] in exergue. Nicomedia. 450–457. Cf. RIC X, p. 283, no. 2468. Bibliography: Daux 1968, p. 779; Clement 1969, p. 140, 1975, p. 164, no. 20; Isthmia V, p. 77, n. 43. Buried in the middle of NEG 67-001 with the uppermost bodies in a small pile or purse (Figs. 2.12, 2.13)

3

(IC 889) Illegible coin Fig. 2.86 Obverse: bust right, diademed, draped. Reverse: illegible. In his unpublished catalogue and in Clement 1987, pp. 382–383, Paul Clement identified a monogram of Marcian (450–457) on the reverse, but Liane Houghtalin (pers. comm.) concurs with his earlier reading (Clement 1975, p. 164) of the reverse as illegible. Mint uncertain. Late 5th century. Bibliography: Clement 1969, p. 140; 1975, p. 164, no. 21; 1987, pp. 382–383, pl. 60:e; Isthmia V, p. 77, n. 43. Buried in the middle of NEG 67-001 with the uppermost bodies in a small pile or purse (Figs. 2.12, 2.13). 4

(IC 69-83) Pentanummium Fig. 2.87 Obverse: bust right, draped. Reverse: e. 6th century. Cf. DOC I, p. 44, no. 18. Placed on the chest of the young man buried in T14 69-003.

Figure 2.84. Coin of Leo I (1). Scale 1:1

Figure 2.85. Coin of Marcian (2). Scale 1:1

Figure 2.86. Illegible coin (3). Scale 1:1

Figure 2.87. Pentanummium (4). Scale 1:1

176. This catalogue does not include tiles and other building materials used in the design of the graves that were inventoried: seven or eight Roman tiles paving T2 68-003 (IM 68109a–z, aa, bb; Fig. 2.44); one reused Archaic or Classical(?) architectural terracotta under the head of T14 67-002A (IT

895; Fig. 2.48); seven Roman tiles and bricks covering and lining HO 70-902 (IM 70-49, 70-90, 70-91, 70-92, 70-93, 70-94, 70-95; Figs. 2.70–2.72); four Roman tiles covering WF 62-001 (IT 811, 812, 813, 814; Fig. 2.75).

104

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.88. Lamp (5). Scale 1:2

Figure 2.89. Lamp (6). Scale 1:2

LAMPS 5

(IPL 69-79) Lamp Fig. 2.88 L. 0.102 (restored), W. 0.081, H. 0.036 m Intact but small breaks around nozzle. Glaze is flaking. Fine clay with white inclusions, light reddish brown (5YR 6/4). Metallic glaze, red (2.5YR 5/6). Disk has central ring, rosette of 13 petals, one scalloped framing ring with small circles in interstices of scallops, and air hole. Plain rim has panels with branches and incised circles flanking handle, nozzle, and panels. Nozzle has single diagonal grooves, groove from air hole to wick hole. Solid handle with three grooves above, two uneven below. Base has two concentric grooves with plain interior. Made with plaster mold. Athenian manufacture. Late 4th century. Cf. Kerameikos XVI, no. 3548 (350–360); Agora VII, nos. 1695, 1694 = Karivieri 1996, no. 234 (late 4th

century); Wiseman 1967a, p. 37, pl. 15:a.7–8 (L4392, Gymnasium area, Corinth). Bibliography: Michaud 1970, pp. 945–946, fig. 124; Clement 1972, p. 165, pl. 134:a, b; Isthmia V, p. 78, n. 49. Deposited on top of the southwest corner of NEG 69008 during or after the funeral (Fig. 2.10). 6

(IPL 76-2) Lamp Fig. 2.89 L. 0.096, W. 0.064, H. 0.032 m Intact but surface weathered. Faint traces of plaster from mold near handle. Fine clay, reddish yellow (5YR 6/6). Unglazed. Disk somewhat rectangular near handle with channel to wick hole, plain Latin cross, central hole plus one under each horizontal arm and one in the channel. Uneven herringbone rim. Nozzle is flanked by two circles and displays burn marks. Solid handle is demarcated by transverse groove at handle base, cluster of

CATALOGUE OF FUNERAR Y ARTIFACTS

three faint circles on each side of handle, two grooves above and two below. Base has branch within double almond-shaped grooves, two circles flanking end of handle, three circles in triangle near nozzle. Made with plaster mold. Late 5th to early 6th centuries. Cf. Kerameikos XVI, no. 4698 (second quarter of the 5th century). For disk, Agora VII, no. 2488 = Karivieri 1996, no. 94 (first half of the 6th century); Agora VII,

105

no. 2507 = Karivieri 1996, no. 96 (second half of the 5th century). For base, Agora VII, no. 2496 = Karivieri 1996, no. 95 (first half of the 6th century); Karivieri 1996, no. 89 (Kerameikos, late 5th to early 6th centuries); Kenchreai V, no. 299. Bibliography: Wohl 1993, p. 135, fig. 8; Gregory 1995, p. 290, n. 22; Karavieri 1996, p. 50, n. 4. Deposited outside RB 76-002 during or after the funeral (Fig. 2.66).

VESSELS 7

(IM 69-42) Glass pitcher Fig. 2.90 H. 0.109, Diam. (base) 0.072, (rim) 0.054 m Intact but rim chipped. Yellowish green glass. Pitcher with concave bottom, globular body, and sloping shoulder. Short cylindrical neck flaring into slightly thickened, round rim. Strap handle with wide, vertical grooves in center. Cf. Hayes 1975, nos. 340, 437, 438 (late 3rd to early 4th centuries). Bibliography: Clement 1972, p. 166, pl. 135:b; Isthmia V, p. 78, n. 46, pl. 20:c. Buried above the bodies under the covering slab at the west end of NEG 69-004, alongside 13 (Fig. 2.7). 8

(IP 3688) Coarse-ware cup Fig. 2.91 H. 0.097, Diam. max. 0.090 m Intact but missing handle. Gritty fabric that flakes in layers, reddish yellow (5YR 6/6). Self slip. Cup with ovoid body on small disk base with wide mouth and flaring rim. Marks where vertical handle was attached. Bibliography: Marty Peppers 1979, p. 302, fig. 98:a (M344); Isthmia V, p. 77, n. 43. Buried in the west end of NEG 67-001 with the lowermost bodies (Fig. 2.14). 9

(IPR 69-76) Fine-ware jar H. 0.142, Diam. max. 0.108 m

Fig. 2.92

Intact. Fine fabric with fine grit and some mica, red (2.5YR 5/8). Fine slip, reddish brown (2.5YR 5/4). Small fusiform jar with grooved base, slightly flaring foot, and a single ridged handle. Body wheel grooved. Late 4th to early 5th centuries. Cf. Agora V, nos. F 42 (late 4th century), M 268 (late 4th century), M 295 (early 5th century). Bibliography: Michaud 1970, pp. 945, 946, fig. 125; Clement 1972, p. 165, pl. 135:a; Marty Peppers 1979, pp. 303–304, fig. 98:b (N345); Isthmia V, p. 77, n. 45. Buried in NEG 69-103 with the bodies. 10

(IPR 70-26) Coarse-ware bowl Fig. 2.93 H. 0.050–0.054, Diam. (rim) 0.162 m Intact. Heavy fabric with large orange and white inclusions, dark reddish gray (5YR 4/2). Shallow bowl on a flat base with flaring sides. Folded, flanged, concave rim. Faint wheel ridging inside. Late 4th century. Cf. Zygouries, pp. 178–179, no. 368, fig. 174; IP 3758 from Northeast Gate (second half of the 4th century); Corinth XVIII.2, p. 126, no. 275, fig. 33, pl. 16 (second half of the 4th century). Bibliography: Clement 1974, p. 110, pl. 95:a; Marty Peppers 1979, p. 298, fig. 92:c, d (L343); Isthmia V, p. 43, n. 32. Buried on the west end of the floor of the south compartment of HO 70-902 below the bodies.

Figure 2.90. Glass pitcher (7). Scale 1:2

106

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.91. Coarse-ware cup (8). Scale 1:2

Figure 2.92. Fine-ware jar (9). Scale 1:2

11

(IPR 76-2) Unguentarium Fig. 2.94 H. max. 0.139, Diam. max. 0.061 m Intact but rim chipped and thin crack at shoulder. Gritty fabric with sparkling inclusions, voids, and lime explosions, light yellowish brown (10YR 6/4). Fine slip unevenly applied, strong brown (7.5YR 5/8). Asymmetrical piriform unguentarium with flat base and high flaring rim. Light wheel ridging on outside of body and neck. String marks on base and excess clay adhering to body near base. Cf. IPR 70-20 from the East Field for a more slender form. Deposited outside RB 76-002 during or after funeral (Fig. 2.66). 12

(IPR 98-1) Cooking pot Fig. 2.95 Diam. (rim) 0.162 m 21 total sherds, nine joining, preserve over half of the rim and a quarter to third of the body at its maximum diameter. Coarse, red clay with numerous small stony inclusions, shiny flecks, and sparse voids, fired or burned uniformly dark gray (7.5R N3).

Globular body with outward-turned, thickened, pointed rim, slightly concave on interior. Vertical handles ovoid in section with a lateral groove, attaching at rim and point of maximum body diameter. Late 6th to very early 7th centuries. Cf. IPR 72-49 from the Roman Bath; Nemea II, p. 132, fig. 242 (P 1546 from the Stadium tunnel, Nemea, dated 580–585); Aupert 1980b, pp. 433–434, nos. 269, 285b, fig. 43 (Argos, dated 585 [p. 404], but should be late 6th to early 7th centuries [see Slane and Sanders 2005, p. 294, n. 108]); Williams and Zervos 1983, p. 29, no. 79, fig. 14 (from pit east of Theater, Corinth, dated late 6th century to ca. 600 [p. 28]); Sanders 1999, pp. 462, 470, no. 19, fig. 14 (Corinth, late 6th to early 7th centuries); Hjohlman 2005, pp. 148–149, 162, 223–225, 228, 230, nos. 40, 41, 266–268, 279, figs. 21, 87–89 (from the Tower and Southwestern Trenches, Pyrgouthi, Berbati Valley, “beginning of the 7th century”). On the development of this type between the mid-6th and late 7th centuries at Corinth, see Slane and Sanders 2005, p. 288. Broken outside NEG 69-010 quite possibly during funeral and sherds mixed into fill over the young child.

CATALOGUE OF FUNERAR Y ARTIFACTS

Figure 2.93. Coarse-ware bowl (10). Scale 1:2

Figure 2.94. Unguentarium (11).Scale 1:2

Figure 2.95. Cooking pot (12). Scale 1:2

107

108

THE GRAVES AND ASSOCIATED REMAINS

13

(Not inventoried) Glass bottle (unguentarium?) H. ca. 0.160–0.180, Diam. (body) 0.045–0.050, (rim) 0.028, Th. (rim) 0.0035 m Intact rim but fragmentary body, shoulder, and neck upon discovery. Disintegrated after recovery. Elongated glass vessel with bulb on bottom, cylindrical body, and sloping shoulder. Short neck flaring into

slightly thickened, round rim. Possibly an unguentarium. Bibliography: Clement 1972, p. 166 (“a disintegrated glass alabastron”). Buried above the bodies under the covering slab at the west end of NEG 69-004 alongside 7 (Fig. 2.7).

PERSONAL ARTICLES 14

(IC 69-74) Pierced coin (pendant?) Fig. 2.96 Obverse: bust left. Reverse: [GLORIAEREI-PV] BLICE, camp gate with two turrets. Thessalonica. Theodosius I, Valentinian II, or Arcadius (383–392). Cf. LRBC II, nos. 1856, 1857, 1861–1863. See pp. 70– 71, n. 111 on pierced coins from graves at Corinth. Reused in 7th century as a personal ornament, either a pendant or a clothing attachment. Worn near the neck of the adult woman buried in T14 69-002. 15

(IM 3533A) Bronze finger ring Fig. 2.97 Diam. 0.022, Th. (hoop) 0.003–0.005, (bezel) 0.008 m Intact but surface of bezel almost completely obliterated. Plain, delicate bronze hoop widening toward bezel. Flat to semicircular in section, with interior flat and exterior slightly convex. Oval bezel slightly raised with engraved design or lettering. Late Roman to Early Byzantine. Cf. Corinth XII, p. 229 (Type G), all with somewhat narrower diameter: e.g., nos. 1851 (hoop circular in section, 4th to 6th centuries), 1853 (rectangular bezel with incised cross, 7th century [cf. Ivison 1996, p. 114]), 1854 (more prominent band and slighter bezel, 4th to 6th centuries), 1862 (6th century); Ivison 1996, p. 115, fig. 5.6J (MF 7076 from grave 1937.15-19, South Stoa, Corinth, 7th century); Wiseman 1967b, p. 420, n. 44, pl. 88:e (MF 12703 from grave 29, Gymnasium area, Corinth, late 5th to early 6th centuries); Miller 1981, p. 48, pl. 12:d, e (GJ 65, 66 from grave 7, Nemea, 5th to 6th centuries). Worn on the right fifth finger of the old man buried in T14 67-002 (Fig. 2.47). 16

(IM 3553) Pair(?) of bronze earrings Fig. 2.98 Diam. max. (outer) 0.021 m Slightly corroded. One intact and one in three pieces. Perhaps a pair. Hoops of plain, thin bronze wire, circular in section. Simple hitch clasp preserved on one. Cf. 17; Corinth XII, no. 2012 (“Byzantine”); Miller 1980, p. 192, pl. 43:a (GJ 35a, b from grave 5, Nemea, 5th to 6th centuries). Bibliography: Marty Peppers 1979, pp. 301–302, fig. 97:a (MCon1). Worn by one or two of the adults buried in NEG 67001. 17

(IM 5019) Bronze earring Diam. max. (outer) 0.034 m

Fig. 2.99

Intact. Hoop of plain, thin bronze wire, circular in section. Ends twisted into hooks to form a simple hitch clasp. Cf. 16; same parallels. Bibliography: Isthmia VII, no. 259. Worn by the individual, probably an adolescent or adult woman, buried in PAL 56-001 (Figs. 2.80, 2.81). 18

(IM 67-6) Iron buckle Fig. 2.100 p.L. 0.072, p.Diam. (loop) 0.026–0.031, Diam. max. (hole) 0.008 m Severly corroded. Loop slightly broken on one side. Tongue snapped near base. Circular hoop, possibly circular in section. Long, thin tongue attached to loop, but shape and articulation marked by corrosion. Cf. Corinth XII, nos. 2175, 2176 (4th to 8th centuries) for buckles with same general form. Worn by a child buried near the top of NEG 67-001 (Fig. 2.12). 19

(IM 69-36, 69-37) Pair of gold earrings Fig. 2.101 Diam. 0.024 m Intact. Matching pair. Thick, heavy gold wire twisted into a cable, forming circular hoop ridged in section. Clasp with small globe on exterior of hoop for hitch. Bibliography: Michaud 1970, pp. 945, 946, fig. 126; Clement 1972, p. 166, pl. 136:a–c; Isthmia V, p. 78, n. 48, pl. 21:b. Worn by the young woman (C) buried in NEG 69001 (Fig. 2.21). 20

(IM 69-43) Bone pendant cross Figs. 2.102, 2.103 L. 0.024, W. 0.020 m Intact but surface scraped and pitted. Greek cross carved from bone with multifaceted knobs on ends of three lower arms. Top arm has circular suspension loop pierced lengthwise with ringed base. Lower arms somewhat circular in section but flattened. Square panel at intersection of arms with sharp rim and interior incised to create raised quadrants. Originally hung on a necklace with variegated beads (21a–k). Cf. Corinth XII, nos. 2080–2083 (“Most are from tenth–twelfth century contexts, but some might well be much earlier. . . . Concerning the chronology of the entire group of pendant crosses, few generalizations can be made,” p. 256); Miller 1988, p. 3, pl. 5:a (GJ 106, 109 from graves, Nemea, 5th to 6th centuries) for same basic design. Worn on a beaded necklace (21) by the infant buried in T14 69-991.

CATALOGUE OF FUNERAR Y ARTIFACTS

109

Figure 2.96. Pierced coin

Figure 2.97. Bronze finger

Figure 2.98. Pair(?) of bronze

(pendant?; 14). Scale 1:1

ring (15). Scale 1:1

earrings (16). Scale 1:1

Figure 2.99. Bronze

Figure 2.100. Iron buckle (18). Scale 1:1

earrings (19). Scale 1:1

earring (17). Scale 1:1 21 (IM 69-44–69-54) Beaded necklace Figs. 2.102, 2.103 The following 11 beads of various materials and forms, together with the bone pendant cross (20), comprised a necklace worn by the infant buried in T14 69-991. The string was not preserved, and the arrangement of the objects is unknown. Perhaps the beads were strung so that they alternated in shape, material, or color. Presumably the cross held the central position and hung over the middle of the chest. a (IM 69-44. L). 0.010 m. Intact. Smooth green stone. Roughly rectangular, trapezoidal in section, pierced lengthwise. b (IM 69-45). Diam. 0.005 m. Intact but surface weathered so color is not apparent. Flattened glass sphere.Cf. Corinth XII, nos. 2437–2440, 2442, 2444– 2447, 2463, 2464 (Roman to Byzantine). c (IM 69-46). L. 0.015 m. Intact but surface slightly worn. Rectangular lead bead, square in section, pierced lengthwise. d (IM 69-47). L. 0.013 m. One end broken and surface slightly weathered. Glass cylinder, pierced lengthwise. Five horizontal stripes painted in yellow. e (IM 69-48). L. 0.011, Diam. 0.013 m. One end slightly chipped. Terracotta cylinder with wide, globular body, pierced lengthwise. Chevrons painted in black on two sides but only partially preserved. Cf. Corinth XII, no. 2502 (probably 4th century). f (IM 69-49). L. 0.010 m. Intact. Bone cylinder with slightly flattened sides, irregular in section, pierced lengthwise. Cf. Corinth XII, nos. 2408 (Late Roman), 2476 (“Byzantine”), 2477 (3rd or 4th century). g (IM 69-50). L. 0.009 m. Intact but surface slightly corroded. Flattened lead sphere, pierced slightly offcenter.

Figure 2.101. Pair of gold

h (IM 69-51). L. 0.019 m. Intact. Thin, cylindrical shell with vertical grooves, finely pierced lengthwise with large hole. Shape and texture are perhaps its natural form. Cf. 32; Corinth XII, nos. 2425, 2472 (“Byzantine”). i (IM 69-52). L. 0.013 m. Intact. Rectangular green stone almost identical to 21a. j (IM 69-53). L. 0.007 m. Intact. Flattened glass sphere almost identical to 21b. k (IM 69-54). L. 0.008 m. Intact. Smooth purpleblack stone, rectangular in section, pierced lengthwise. 22

(IM 69-55) Iron loop (buckle Fig. 2.104 or pendant?) Diam. max. 0.050 m Intact, but severely corroded. Thick, circular iron loop flattened on one side. One end thicker and heavier, perhaps another iron piece fused to it. Possibly a buckle or pendant. Worn by one of the individuals buried in NEG 69005. 23

(IM 69-70) Bronze ring Fig. 2.105 Diam. 0.006, Th. 0.002 m Intact. Plain, thin bronze hoop, circular in section. No traces of a join, so probably cast. Possibly a finger ring, but very small. Cf. Corinth XII, nos. 1908–1913 (Type J, Roman to Byzantine); IM 2236 and 3181 from the temenos of Poseidon are larger versions of same design (Isthmia VII, pp. 61, 63, nos. 237, 238, pl. 38, “perhaps Roman period”). Probably worn by the deceased, mostly likely a child, in NEG 69-008.

110

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.102. Beaded necklace with bone pendant cross (20, 21),

suggested reconstruction. Scale 1:1

Bone pendant cross (20)

Stone bead (21a)

Glass bead (21b)

Lead bead (21c)

Glass bead (21d)

Terracotta bead (21e)

Bone bead (21f)

Lead bead (21g)

Shell bead (21h)

Stone bead (21i)

Glass bead (21j)

Stone bead (21k)

Figure 2.103. Bone pendant cross (20) and various beads (21a–k) from necklace. Scale 2:1

CATALOGUE OF FUNERAR Y ARTIFACTS

111

Figure 2.105. Bronze

ring (23). Scale 1:1

Figure 2.107. Stone bead on glass

rod (necklace?; 25). Scale 1:1

Figure 2.104. Iron loop (buckle

or pendant?; 22). Scale 1:1

Figure 2.106. Bronze

buckle (24). Scale 1:1

Figure 2.108. Iron disks

Figure 2.109. Iron rings

Figure 2.110. Iron nail

(26). Scale 1:1

(27). Scale 1:1

(hobnail?; 28). Scale 1:1

24

(IM 70-32) Bronze buckle Fig. 2.106 Diam. max. (ring) 0.024, L. (tongue) 0.024 m Intact. Small bronze ring, circular in section. Ends do not join. Straight bronze tongue, slightly rounded at end, circular in section but flattened where it folds over ring. Surface of tongue near fold is stepped, while surface of tongue at opposite end displays two shallow longitudinal grooves. Late Roman. Cf. IM 70-31 from East Field (end of the 4th century); Wiseman 1969, p. 79, pl. 25:c (MF 12929 from grave 86, Gymnasium area, Corinth, 5th century); Corinth XII, nos. 2175 (4th to 8th centuries), 2179 (4th century “or somewhat later”); Miller 1980, p. 192, pl. 43:a (BR 721 from grave 3, Nemea, 5th to 6th centuries) for same design but more elliptical loop. Bibliography: Clement 1974, p. 110, pl. 95:c; Marty Peppers 1979, p. 297, fig. 93:b (LCon2); Isthmia V, p. 43, n. 32. Found on floor of south compartment of HO 70-902. Worn by one of the deceased. 25

(IM 70-54) Stone bead on glass rod Fig. 2.107 (necklace?) p.Diam. (bead) 0.016, (hole) 0.005, p.L. (rod) 0.020 m Bead two-thirds preserved. Rod broken on both ends. Smooth black bead, flattened spherical and pierced. A segment of hollow, dark-colored (green or black) glass twisted like a screw. Glass rod found inserted through bead. Originally on a string, perhaps a necklace. For bead, cf. Wiseman 1967b, p. 428, no. 26, pl. 91:n

(MF 12712 from grave 47, Gymnasium area, Corinth, late 4th to early 6th centuries); Corinth XII, no. 2417; for rod, Buchet and Sodini 1984, p. 212, pl. 79:c (necklace elements from tombe 3, Basilique Nord, Aliki, Thasos, Early Byzantine); Vikatou 2002, pp. 247, 254, 270, pl. 22:β (necklace elements from τάφος 15, Ayia Triada, Elis, Early Byzantine). Bibliography: Clement 1974, p. 110, pl. 95:b; Marty Peppers 1979, p. 296, fig. 93:a (LCon1); Isthmia V, p. 43, n. 32. Found on floor of south compartment of HO 70-902. Worn by one of the deceased. 26

(IM 70-80) Iron disks Fig. 2.108 Diam. (ring) ca. 0.015–0.016, (hole) ca. 0.005 m Intact but severely corroded. Three iron disks, pierced. Fused into a mass so that they overlap but are slightly offset. Possibly another oblong piece runs through the rings. Unknown function. Cf. 27. Probably worn by one of the deceased buried in HO 70-902. 27

(IM 70-81) Iron rings Fig. 2.109 Diam. (ring) ca. 0.015–0.016, (hole) ca. 0.005 m Intact but severely corroded. Two iron disks, pierced. Fused into a mass so that they overlap. Another thin strip, apparently looped on one end, runs through them, perhaps to bindthem. Unknown function. Cf. 26. Probably worn by one of the deceased buried in HO 70-902.

112

THE GRAVES AND ASSOCIATED REMAINS

Figure 2.113. Bronze Figure 2.111. Bone button

Figure 2.112. Iron buckle (30). Scale 1:1

loop (31). Scale 1:1

(29). Scale 1:1

Figure 2.114. Shell bead

(32). Scale 1:1 28

(IM 70-82) Iron nail (hobnail?) Fig. 2.110 Diam. (head) 0.011, p.L. 0.017 m Intact head but shank broken. Severely corroded. Iron nail with flat, circular head and short square shank, slightly bent. Possibly a hobnail. Cf. Miller 1979, p. 85, pl. 28:d (IL 292 from grave, Nemea, 5th to 6th centuries). Interred with the bodies in HO 70-902. Perhaps worn by one of the deceased.

10 months) and the diameter of the loop count against its identification as a finger or hair ring. Cf. Corinth XII, nos. 1909 (iron finger ring, 7th century), 1961 (silver finger ring, 10th to 12th centuries), 2175 (bronze buckle, 4th to 8th centuries); Corinth XVIII.3, p. 383, no. 1, pl. 57:a (MF-72-37, -38 from grave 1972-6, Demeter and Kore, a child burial with “bronze hair rings,” 4th to 6th centuries). Worn by the infant buried in NEG 69-009.

29

32

(IM 76-2) Bone button Fig. 2.111 Diam. (hole) 0.008, (disk) 0.030, Th. 0.003 m Intact but surface pitted and edge chipped. Circular bone disk pierced through the center. Slightly convex on one side with small flange projecting from outer edge. Hole worn along inside edge of opposite side. Cf. Corinth XII, no. 2537 (“Byzantine”). Worn by one of the adults buried in RB 76-002. 30

(IM 98-2) Iron buckle Fig. 2.112 Diam. 0.022, Th. 0.003 m Broken into two pieces. Severely corroded. Iron buckle consisting of a round loop, circular in section, and a second piece (tongue?), hollow and circular in section, attached to the base of the loop like a hinge. The details of the attachment are unclear. For general design, cf. Corinth XII, nos. 2180–2183 (7th century); for shape, Corinth XII, no. 2179 (4th century “or perhaps somewhat later”). Worn by the infant buried in NEG 69-009. 31

(IM 98-3) Bronze loop Fig. 2.113 Diam. max. (outer) 0.019, (inner) 0.013, Th. 0.003 m Intact. Plain, oval bronze loop, circular in section but slightly flattened toward ends. Two ends join unevenly to close the ring. Perhaps a buckle; the age of the deceased (first

(IM 98-4) Shell bead Fig. 2.114 L. 0.013, Diam. 0.004 m Intact but small chips on edge. Smooth, cylindrical shell with thin walls, finely pierced lengthwise with large hole. Shape is perhaps natural. Originally on a string, perhaps a necklace. Cf. 21h; Corinth XII, nos. 2425, 2472 (“Byzantine”). Worn by the infant buried in NEG 69-009. 33

(Not inventoried) Beaded necklace The following five beads of blue glass, together with eight others that the excavator did not describe, comprised a necklace worn by the individual, probably an adolescent or adult woman, buried in PAL 56-001 (Figs. 2.80, 2.81). The string was not preserved and the arrangement of the objects is unknown. a. Diam. 0.014 m. Blue glass sphere with white swirls. Cf. Corinth XII, nos. 2430 (Roman), 2443, 2445 (“Late Roman [probably 4th century]”), 2448 (10th to 11th centuries). b. Diam. 0.0145 m. Flattened blue-green glass sphere. c. L. 0.013 m. Flattened blue glass cylinder. d. L. 0.014 m. Flattened blue glass cylinder. e. L. 0.0115 m. Flattened blue glass cylinder. Spherical and cylindrical beads in glass, often flattened, both plain and decorated, were common at Roman and Byzantine Corinth (Corinth XII, pp. 288– 290). For spherical examples, cf. Corinth XII, nos. 2429– 2450; for glass cylinders, Corinth XII, nos. 2472–2476.

3

 BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

T

he 30 graves containing at least 69 individuals uncovered by excavations at the Isthmus provide important evidence for the chronology, topography, and history of settlement. In order to trace diachronic patterns in the social and bioarchaeological character of the local residents as reflected by their graves (Chap. 4) and their bones and teeth (Chaps. 5–7), it is necessary to establish the long sequence of interment. The fact that graves of different dates have been found in every major area of excavation since 1954 proves that habitation was neither restricted nor brief. Since residents would not have buried their dead at great distances from where they lived, the location of graves should reflect the distribution of settlement. Archaeological and historical testimony reveals the nature of the settlement to which the graves belonged and thus elucidates the identity of the community, while the date and location of the graves shed light on the habitation of the Isthmus. The new synthesis presented here concentrates on those historical conditions that are most important for understanding the graves and human remains within the context of a living community and its changing landscape: domestic life and diet, social and economic intercourse, religion, political and military factors, foreign incursions, and natural disasters.1 This chapter will examine the chronological, topographical, and historical setting of the burials. They represent five major phases (I–V) from the Roman efflorescence of the Isthmian Games to the end of the Byzantine Empire and later. A majority of the interments date from the end of the 4th to the late 7th or 8th centuries (phases II–IV), with only a few earlier (phase I) and later (phase V) instances. The graves at the Northeast Gate can be dated to within a few decades, while those from most other locales can only be broadly dated. The burials in each phase were not necessarily synchronous events; in fact, it is doubtful whether any two of them were exactly contemporary. The periodization of the burials thus denotes a progression over time that was continuous on a larger scale, representing several successive generations of residents. These burial phases correspond with discernible eras in the history of occupation that can be summarized as follows. After the reinstitution of the Isthmian Games around the middle of the 1st century, the Panhellenic Sanctuary operated as a cult site well into the 3rd century, though the date of its final festival is unknown. During the acme of the Roman phase of the Sanctuary, a small community lived near the monumental core of the site, supporting its numerous visitors and religious functions and profiting from them. During the 3rd to 4th centuries, residents and visitors alike witnessed the decline of the Sanctuary and its cults and the deterioration of the sacred buildings from neglect and earthquakes. Despite these changes, the local settlement and memories of a cultic past survived. Residents during this 1. For the basic history and topography of the Isthmus during the Roman and Byzantine eras, see the following: Gebhard

1973; Gebhard, Hemans, and Hayes 1998; Gregory 1993a, pp. 75–160, 1995; Isthmia II; V; VIII, pp. 17–18; Kardulias 2005.

114

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

long era of prosperity and decline buried their dead near the road through the Roman Arch northeast of the Temple of Poseidon (phase I). The Hexamilion and the Fortress were erected in the early 5th century. The Bath and the central area were sparsely inhabited during the 5th and 6th centuries, while the primary settlement was concentrated within the enceinte. These residents survived the plague epidemic and large earthquakes that struck the Greek world during the 6th century. They were native Corinthians who occupied the fortifications probably as a rural militia, and many, if not all, were Christian. They depended on their gardens and fields for food but also benefited from exchange with other urban and rural settlements. They buried their dead as before along the northern edge of the site adjacent to the new curtain wall, with the exception of one grave in the foundations of the Temple of Palaimon (phases II–III). After the rehabilitation of the fortifications in the mid-6th century, the Fortress and Hexamilion began a period of dilapidation that would last for centuries. When the Slavic invaders reached the Peloponnese in the 580s, they quickly passed the Isthmus without resistance or any widespread destruction of the Fortress. The settlement that had occupied the site since the Roman Empire persisted until the late 7th or 8th century, but it fragmented into several areas, with foci in the Fortress, the Bath, and the central area. These residents were Christian Corinthians descended from the earlier inhabitants of the area, but they practiced a more modest, localized subsistence of agriculture and husbandry. They buried their dead mostly outside the north and west walls of the Fortress (phase IV). After an apparent hiatus of two to three centuries, when the community either moved, retracted into an even smaller area, or dissolved, the site was reoccupied during the 10th century. This settlement, which again was concentrated around the Fortress, grew dense during the 12th to 14th centuries. The fortifications were renovated and employed again in the 15th century, after which point the settlement returned to relative obscurity until the 19th century. Only two burials belong to this phase, one dating from the Middle to Late Byzantine era and another from the Late Byzantine to Early Modern era. Both are unusual interments outside the Fortress that do not represent conventional mortuary practices of the time. The regular cemetery has not been found; it must have been either inside the Fortress or, more likely, far outside its walls (phase V). During the mid-19th century, settlers founded Kyras Vrysi west of the Fortress. In 1883 Paul Monceaux conducted the first scientific investigation of the site. The village’s cemetery at Ayios Ioannis Prodromos, situated in the ruins of the Fortress reportedly since the 19th century, illustrates how Greek burial forms and rituals have persisted and transformed since ancient and medieval times.

PHASE I: THE ROMAN SANCTUAR Y (MID-1ST TO LATE 4TH CENTURIES) The first phase of burial includes T2 68-003, DEC 69-901 and 69-902, and WF 62-001, which together contained at least six individuals.2 These burials date to the Roman era, and most or all belong to the Early to Middle Roman periods. The dating of the burials in DEC 69-901 and 69-902 depends on the use of the pitched-tile covering in 69-902, which identifies them as Roman, more likely Early or Middle than Late Roman. The burial in WF 62-001, another tile-covered cist, can be placed more securely in the Early to Middle Roman periods by the form of the tiles. The burials in T2 68-003 predated the construction of the Hexamilion, and the grave’s unusual design points to a date before the late 4th to early 5th centuries. 2. This total is based on the assumption that 69-902, in which no skeletal remains were found, contained at least one individual.

PHASE I: THE ROMAN SANCTUAR Y (MID-1ST TO LATE 4TH CENTURIES)

115

The four graves do not constitute a coherent burial phase but rather a collection of single interments from separate periods and locales. In all probability they fall between the mid1st and the late 4th centuries, that is, during the Roman phase of the Sanctuary before the fortification of the Isthmus. The topography, settlement, and use of the site during the middle years of the Roman Empire, when the Isthmus was a thriving regional and international center, are amply represented in the archaeological record. After a period of abandonment in the Late Hellenistic period and the early decades of the Empire, the Games returned to the Isthmus in the mid1st century, and the Sanctuary was rebuilt.3 The road approaching the Isthmus from the northeast was adorned with the tripylon in the late 1st century. A large building represented by the Long Wall was located directly west of the area later occupied by the Fortress, while the Theater, rebuilt under Nero, and the exquisite Bath, erected in the mid-2nd century, were situated further west. The sacred complex consisted of the temenos of Poseidon and the precinct of the hero Melikertes-Palaimon, which underwent multiple stages of renovation during the Empire. The Stadium, which was remodeled perhaps in the late 1st century, was situated southeast of the temenos in the natural hollow east of the Rachi (Fig. 1.2).4 Other buildings occupied the area delimited by the Arch, the temenos, and the Bath. Exploration of the Fortress has revealed membra disjecta from elaborate monuments of Roman date, such as columns, entablature, statue bases, tesserae, and marble revetment not unlike those found in the Bath.5 The field south of the Theater has produced a marble statue of Heracles and the Lion from a fountain.6 These remains suggest that both the area later occupied by the Fortress and the zone extending west underwent considerable construction during the Empire. Some buildings presumably were used in the competitions but some were decorative and commemorative in nature, situated as they were along the primary route of traffic into and through the Sanctuary. Traces of residential settlement during this era have been found in situ within the Fortress, to its south, and between it and the temenos. Jenkins and Megaw frequently uncovered Roman pottery associated with simple buildings in the central and south parts of the enceinte.7 A surface survey in the area yielded numerous sherds dating from the 1st to 4th centuries and a vast amount of Roman tile.8 Excavation between Towers 9 and 10 and outside the South Gate has revealed rubble walls and debris from habitation in the Early and Middle Roman periods.9 The so-called East Field directly east of the temenos also contained a cluster of buildings that were occupied primarily in the 2nd to 3rd centuries.10 Nearby to 3. Gebhard (1993a, pp. 87–88) dates the return of the Games to 55 or 57, and Kajava (2002) dates it to 43; see also Gebhard 2005, p. 185, n. 76 (“The archaeological context cannot be dated more precisely than the middle of the 1st century C.E.”). For the ceramic evidence for the return of Games, see Hayes 1993, p. 113; Marty 1993, pp. 117–121; Gebhard, Hemans, and Hayes 1998, pp. 425–426, 445–446. 4. Tripylon: Gregory and Mills 1984. Long Wall: p. 69, n. 99. Theater: Gebhard 1973, pp. 63–134. Bath: Packard 1980; Gregory 1995, pp. 281–303; Isthmia VI, pp. 3–4. Temenos of Poseidon and the Palaimonion: Isthmia II, pp. 67–109; Gebhard 1993a, pp. 89–93; Gebhard, Hemans, and Hayes 1998, pp. 416–454; Gebhard 2005, pp. 189–203. Stadium: Isthmia II, pp. 55–63, 66; Gebhard, Hemans, and Hayes 1998, p. 416, nn. 27, 28; Isthmia VIII, p. 18 (episodes VI–VII). 5. Monceaux 1884, pp. 357–358, 361–363; Corinth I, pp. 66– 68; Clement 1969, pp. 140–142, plan 2; 1972, p. 167; Gregory and Kardulias 1990, pp. 498–500, fig. 23; Isthmia V, p. 130; Kardulias 2005, pp. 87–88, figs. 6.33–6.35. It is difficult to discern whether this material originated in the area later occupied

by the Fortress or was brought there when the fortitifications were erected. 6. Broneer 1962a, pp. 18–19; Isthmia IV, pp. 117–118, no. 30, pls. 52, 53:a (IS 405). 7. Jenkins and Megaw 1931–1932, pp. 79–81, fig. 7. The UCLA team excavated rubble walls, floors, and bothroi just southeast of the southeast corner of the modern churchyard, but the remains are not dated (Clement 1972, p. 167). Isthmia V, p. 98, notes deposits of Roman date in the area of Tower 15. 8. Gregory and Kardulias 1990, pp. 490–492, 496–498, figs. 17, 22; Kardulias 2005, pp. 76–77, fig. 6.20. 9. South of Towers 9 and 10: Clement 1972, p. 163, pls. 125:b, 126:b, c; Marty Peppers 1979, pp. 278–277 (Deposit I); Marty 1993, pp. 121–126; Isthmia V, p. 124. Outside South Gate: Broneer 1958, p. 21, pl. 8:c; 1959, p. 321, pl. 64:b. 10. Clement 1974, pp. 105–108; 1976, pp. 224–230; 1977a, p. 145; Beaton and Clement 1976; Marty Peppers 1979, pp. 215–277 (Deposits B–F); Marty 1993, pp. 121–126; Gregory, Kardulias, and Rife 1997, p. 2; Rothaus 2000, pp. 89–90, n. 23.

116

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

the northeast and due west of Tower 12 Jenkins and Megaw found three houses apparently of similar design and date.11 Thus, a substantial settlement extended east and northeast from the temenos, perhaps flanking the roads entering the central area.12 The general topography of the Roman phase of the Sanctuary can be reconstructed from these remains. One of the site’s faces was monumental and decorative. This consisted of impressive buildings that not only supported the cults and specific activities of the Games but also provided rich amenities for visitors and beautified the site. Apart from the Stadium, these buildings seem to have been concentrated to the west of the area later occupied by the Fortress. The site’s other face was comparatively mundane. This consisted of densely conglomerated buildings in rubble and mortar masonry with beaten earth floors. The buildings contained large quantities of coarse ware and refuse from a mixed diet of food prepared on site.13 The pottery, small finds, and faunal remains point to domestic occupation, while the large quantities of sherds and proximity to greater monuments suggest commercial use. The buildings were concentrated between the temenos and the area later occupied by the Fortress but probably extended further east and south into yet unexplored regions.14 The historical testimony portrays the site as bustling with throngs of visitors who came to watch the sacred competitions and to enjoy social and commercial exchange.15 The biennial Games were well attended due to their central location, which also benefited throughout history as a venue for political assembly and markets.16 During the Empire, the Sanctuary hosted numerous visitors, including a diverse group of itinerant intellectuals, performers, and sellers,17 while Corinthian aristocrats held banquets for both foreign and local visitors.18 Some spectators at the festival camped there,19 but athletes lodged in guestrooms near the Stadium that were used as shops in the off-season.20 The texts thus depict a sanctuary frequented by many visitors of diverse origins, professions, and standings both during the Festival and at other times. Apart from attending games, these people enjoyed various facilities, staying at the site if they wished and buying food, souvenirs, and other goods. This picture implies the site’s permanent occupation. While current understanding of the sacred officials is limited, it is conceivable that certain personnel with cultic or custodial responsibilities lived there. Moreover, heavy attendance generated not only a need for service and produce but also an opportunity for revenue to manufacturers, sellers, and renters. Various

11. Jenkins and Megaw 1931–1932, pp. 83–84, pl. 26 (Area B, Houses A–C). Monceaux (1885, p. 209) notes numerous foundations and walls (“maisons”) in this area; see Corinth I, p. 70. 12. On the Roman roads, see Gebhard, Hemans, and Hayes 1998, pp. 408–415, 420–422, and Isthmia VIII, pp. 18, 195–198, n. 4, which revise Isthmia II, pp. 87–89, plan IV on the east– west roads north of the Temple. The passage of traffic into the Roman temenos through the East Gateway (Isthmia II, pp. 74– 75) and later the Southeast Propylon (Isthmia II, pp. 77–78) is more problematic. Roads of various dates have been found in the East Field, but they are yet unpublished (e.g., Clement 1974, p. 106; 1976, pp. 227–228). 13. Preliminary study of the faunal remains from the East Field has identified the presence of cattle, horse, pig, sheep or goat, dog, chicken, and shellfish (D. S. Reese, pers. comm.). 14. Isthmia II, p. 2. 15. Spawforth (1989) generally discusses the atmosphere of agonistic festivals in Roman Greece. 16. Gebhard (1993b, pp. 165–169) discusses the role of the Isthmus as a meeting place for Greeks. In his description of the Isthmian Congress in 196 b.c., Livy wrote that people went to the Isthmus not only for the competitions but also because

it was a “marketplace” (mercatus) for every ware (33.32.1–2), echoing a commonplace in ancient discussions of the Corinthia (see, e.g., Thuc. 1.13.5; Strabo 8.6.20 [378]; Aristid. Or. 46.22–23; Philostr. sen. Nero p. 270.7–11; Lib. Decl. 25.46; John Chrys. Arg. ep. I ad Cor. 1 [PG 61.9–10]). 17. Dio Chrys. Or. 8.5–11 (cf. 9.1–2); Paus. 2.1.7–2.2.2. Philostr. VA 8.18 offers a similar portrait of the Olympic Games under the Empire. 18. Plut. Quaest. conv. 723A. Gritsopoulos (1973, pp. 45–48) emphasizes the cosmopolitan character of the community. 19. Dio Chrys. Or. 9.22; see also Gebhard 1992, pp. 78–79, n. 32, on camping visitors. 20. Inscriptions (IG IV 203; Corinth VIII.3, no. 306, with a join from Isthmia) recorded that P. Licinius Priscus Juventianus, a priest of Poseidon and agoranomos, made several improvements to the Sanctuary in the 2nd century, among which was a stoa near the stadium with 50 “vaulted houses” (κεκαμαρωμένοι οἴκοι). The agonothetai were to assign these chambers free of charge to athletes as “guest rooms” (ξενίαι) or “lodgings” (καταλύσεις) during the Games. When the competitions were not in session, Priscus controlled the allotment of rooms for commercial use as shops. Geagan 1989 is a complete edition and commentary.

PHASE I: THE ROMAN SANCTUAR Y (MID-1ST TO LATE 4TH CENTURIES)

117

finds point to the presence of a ceramic industry on the Isthmus, possibly within the settlement.21 Residents pursuing such activities would have lived in the area, because the site, which served both as a Panhellenic Sanctuary and a roadside shrine, was visited throughout the year.22 The twin motivations of convenience and profit suggest that this settlement was close to the central area, perhaps in the East Field and to the south of the Fortress.23 A residential settlement near the central area would have necessitated an adjacent burial ground. If 68-003, 69-901, and 69-902 date to the early or middle years of the Empire, they probably represent the inhabitants who supported and exploited the activities of the flourishing Sanctuary. These scattered graves are in the same general area but do not form a crowded cemetery, as no other interments were found in their vicinity. During the Roman era, this area was peripheral to the most dense zone of buildings, outside the entrance to the Sanctuary as demarcated by the Arch on the south side of the route from the northeast.24 It will be seen that three graves of phase II (69-103, 69-004, and 69-008), which slightly predated the conversion of the nearby Arch into the Northeast Gate of the Fortress, were also located on the south side of the road into the Sanctuary. Their placement excluded the dead from inhabited space by placing burials outside the formal limits of settlement and removed the pollution of dead flesh a safe distance from the temenos. The roadside position of the graves presumably reflects a need for accessibility during burial and perhaps later visitation; visibility does not seem to have been a dominant concern because the graves lacked markers at the surface. A similar motivation might have determined the placement of 62-001 to the north of the major east–west road passing the ruins of the West Foundation. The location of graves lining the roads into a settlement is a familiar feature of civic plans across the Roman world.25 If it were possible to investigate the ancient road to Corinth immediately west of the temenos, which today underlies the village, one would expect to find more graves of Roman date.26 The considerable distance of 62-001 west of the temenos and its proximity to surface remains of ancient habitation indicate that it probably belonged not to the residents around the Sanctuary but to a separate rural settlement.27 The occupation of the Isthmus continued throughout the Early and Late Roman periods while local residents and other Corinthians saw their world changing. Before turning to the settlement at the Isthmian fortifications in late antiquity (phases II–III), it is necessary to examine the historical transition that marked the end of the Sanctuary and its cults. Although the Middle to Late Roman remains of the central area are badly preserved and have not been fully studied, the available material remains, together with relevant textual sources, attest to the momentous but gradual shift from the Temple to the Fortress. After the prosperity of the 1st to 2nd centuries, cult practice and its physical setting slowly evolved during the 3rd to 4th centuries. Certain activities in the central area were waning 21. These include imitations in local wares of popular imported forms, stamped tiles, and the ritual apparatus of the hero-cult, particularly the wheelmade lamps and small mugs: Isthmia III, pp. 2–3, 26–54, 92 (Broneer type XVI); Hayes 1993; Marty 1993, p. 117, n. 6; Gebhard, Hemans, and Hayes 1998, pp. 444–454. 22. Gebhard 1992, p. 73; 1993b, pp. 165–166. 23. Isthmia II, pp. 89–96, surveys the numerous terracotta pipes and water channels that crossed the central area, all of Roman or later date, as well as a large cistern northeast of the Temple and just east of the Northeast Altar Terrace (p. 95, plans II, IV, pl. 55:a). These elaborate hydraulic facilities would have served both visitors and residents, perhaps in this general area (E. R. Gebhard, pers. comm.). 24. On this road, see Isthmia II, pp. 88–89, and Gregory and

Mills 1984, pp. 407–408. The movement of earth for the construction of the modern highway to Epidauros has drastically altered the landscape north and east of this area, so that it is now difficult to trace the route of the ancient road or the location of more graves in the vicinity of the Isthmian Sanctuary. 25. Paus. 2.2.4 similarly notes tombs lining the road from the Isthmus into Corinth from the east. 26. The western limit of the Sanctuary and its residential settlement is not known (Isthmia II, pp. 2–3). The modern road to Examilia and Archaia Korinthos approximates the ancient route. The West Cemetery gives a terminus intra quem for the Sanctuary proper and its settlement because excavation there revealed only graves and no buildings. 27. See pp. 90–91.

118

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

in the 3rd century. The last sacrificial deposit associated with the Palaimonion dates to the first half of the 3rd century, perhaps as late as ca. 240, around which time production of the distinctive lamps used in the cult ritual also ended.28 At the same time, strata across the central area show a marked decrease in pottery.29 Depositional contexts in the Stadium and the Theater reveal that both buildings fell into disuse and disrepair during the early to mid3rd century.30 Furthermore, the four latest epigraphic attestations of Isthmian victors date to the mid-3rd century.31 This collective testimony indicates that visitation to the site had decreased, and that traditional forms of worship and competition had greatly diminished or ceased, perhaps marking the end of the festival by the second half of the 3rd century. In the face of these changes, however, ritual and visitation did continue. The latest dedications found at the Sanctuary belong to the late 3rd and possibly 4th centuries. These include the famous portrait head of an intellectual or aristocrat made in the late 3rd century and a large inscribed monument in dark marble erected over the southeastern corner of the Temple of Poseidon at some point during the 4th to 6th centuries, quite possibly at the earlier end of that range.32 The buildings were dilapidated but still standing, and visitors and residents alike still beheld the images of old gods. Part of the great cult-statue of Amphitrite could be seen toppled in the Temple, while occupants of one simple house in the East Field at the end of the 4th century buried a hoard of coins and a cache of miniature sculptures depicting, among others, twin Cybeles, Asclepius, Hygieia, Telesphorus, and possibly nymphs.33 The latest reference in classical literature to contemporaneous cult-activity at the Isthmus occurs in a speech by Libanius to the emperor Julian dated 362. The orator discusses in vague terms the pious attachment of two Corinthian aristocrats, a father and son, to the cult of Poseidon (Or. 14.5, 14.7; cf. 14.8). The antipagan constitutions and Alaric’s invasion at the end of the 4th century have often been cited as decisive factors in the decline of the Isthmian Sanctuary.34 Theodosius I issued three edicts in 391 and 392 that together outlawed the essential activities and materials of public and private cult, namely, animal sacrifices, libations, incense, garlands, effigies, altars, and temples (Cod. Theod. 16.10.10–12).35 Although widely promulgated, the enforcement of Theodosius’s policy was not universal or effective: various forms of paganism survived into the 7th century or later across the eastern Mediterranean.36 Violent attacks on pagan temples occured in Syria and Egypt, particularly at centers such as Antioch, Apamea, Heliopolis, and Alexandria, and there is strong but dispersed evidence that Greeks forcefully appropri28. On Pit C at the Palaimonion, see p. 96, n. 166; on the Palaimonion lamps, see Isthmia III, pp. 35–52; Hayes 1993, p. 114; Gebhard, Hemans, and Hayes 1998, p. 446 (revised chronology). 29. Isthmia VIII, p. 18. 30. Catling (1988, p. 22) cites unpublished research by E. R. Gebhard (Stadium); Gebhard 1973, pp. 133–134 (Theater). 31. IG II/III2 3769 (after 244–245, Athens), 3169/3170 (ca. 253–257, Athens) = Moretti 1953, pp. 261–268, nos. 89, 90; IG VII 49 (mid-3rd century, Megara) = Moretti 1953, pp. 259–261, no. 88; Moretti 1953, pp. 253–257, no. 86 (after 229, Anazarbus, Cilicia). 32. Portrait sculpture (IS 446): Jenkins and Megaw 1931– 1932, p. 82; Isthmia IV, pp. 142–144, no. 85, pls. 70:d, 71 (dated ca. 270–280). Inscribed monument (IΣ 211a+c, b, d): NB GFS, pp. 99, 125 (trenches S2 and C6; May 1954). The author thanks Dan Geagan† for discussing this unpublished monument and its inscription. A full treatment will appear in his forthcoming study of the epigraphic corpus from the Sanctuary in this series. 33. Statue group in Temple of Poseidon (torso of Amphi-

trite is no. 17 A [IS 1]): Isthmia IV, pp. 76–113, nos. 17–20, pls. 34–51, esp. p. 84 on its end. Sculptures from East Field: Clement 1976, p. 229, pls. 164:a–165:c (IS 71-2–71-6); Beaton and Clement 1976, p. 267; Marty Peppers 1978; Isthmia VI, pp. 2, 10–14, 39–50, 55–56, 57, nos. 3, 4, 89–91, pls. 4, 31:b, 32; Rothaus 2000, pp. 123–124, figs. 28–30. This evidence may be seen to support Caseau (2004), who argues that the demise of rural temples as public religious institutions paralleled a shift of pagan cult to the private sphere. 34. Local impact of edicts: Isthmia I, p. 103; Gebhard 1973, p. 135; Beaton and Clement 1976, p. 277; Wohl 1981, p. 116. Local impact of Alaric: Beaton and Clement 1976, p. 277; Clement 1976, p. 229; 1977b; Wohl 1981, pp. 112–116; Marty 1993, p. 126, n. 46. 35. For general discussions of the Theodosian constitutions, see Jones 1964, pp. 167–169; Fowden 1998, pp. 548– 554; Price 1999, pp. 164–166. On Greece, see Agora XXIV, pp. 69–71. 36. See multa inter alia Bowersock 1990; Harl 1990 (pagan sacrifice); Trombley 2001a, vol. 1, esp. pp. 98–186 (Christianization in general), 283–332 (Athens and Attica).

PHASE I: THE ROMAN SANCTUAR Y (MID-1ST TO LATE 4TH CENTURIES)

119

ated old cult sites and images for Christian belief and practice, for instance, at Palaiopolis on Corfu in the early 5th century and at the Asklepieion on the Athenian Acropolis later that century.37 But these were rare events during a more advanced stage in the Christianization of Greek society.38 Many sanctuaries, like the Isthmian, had already begun to transform by the era of the Theodosian constitutions. The Visigothic invasion of the Isthmus early in 396 is recorded in historical and archaeological evidence, but the extent and impact of the event are difficult to ascertain. Several writers noted Alaric’s burning and enslavement of Peloponnesian cities, specifying the destruction of Corinth and neighboring communities apparently in the countryside.39 The best evidence for a barbarian presence at the Isthmus is a hoard of 97 coins buried in the East Field with the small sculptures of a cultic nature.40 This deposit does not prove a concerted program of destruction at the Sanctuary, but it does show that local inhabitants recognized an imminent danger. Earthquakes might have played a greater part than did legislation and invasion in the physical decay of those places where sacred rites had long been enacted. Buildings across the Corinthia suffered extensive damage in the second half of the 4th century. Archaeologists have often associated this damage with two massive earthquakes that, according to Late Antique and Byzantine sources, struck the Peloponnese in 365 and 375.41 But caution should be observed in assigning physical evidence for structural damage to a single seismic episode known from written sources. Not only is the dating of the material seldom precise enough for secure identification with the historical record, but the historical record itself is imprecise, and the extant texts do not record all contemporary earthquakes.42 The archaeological and epigraphic evidence for regional damage clusters around two dates, the middle 360s to the middle 370s and ca. 400.43 The buildings of the Isthmian Sanctuary also suffered from earthquakes in the late 4th century. The walls and roof of the Roman Bath partly collapsed around this time, as did the scene building of the Theater.44 Seismic force of such magnitude would have also damaged the buildings of the central area, but to what 37. Palaiopolis: Papadimitriou 1948; Spieser 1976, p. 312; Trombley 2001a, vol. 1, p. 330. Asklepieion in Athens: Gregory 1986, p. 238; Pallas 1989, pp. 876–879; Karivieri 1995; Trombley 2001a, vol. 1, pp. 308–310, 323, 342–344. Rothaus (2000, pp. 119–125, figs. 20–27) addresses the Christian mutilation of pagan statues at Corinth during the 5th to 6th centuries, including their consecration by the sign of the cross. See Caseau (2004, pp. 118–134) generally on the destruction of temples. 38. The attribution to Christians of the destruction of temples at Delphi (Amandry 1989, pp. 43–44, n. 20), Rhamnous (Petrakos 1991, pp. 29, 54; Moutzali 1993, p. 30), and Sicyon (Avramea 1997, p. 114, n. 26) remains largely hypothetical. 39. Eunap. VS 8.2.1–2; Zos. 5.6.3–4, depending on the history by Eunapius (e.g., fr. 64); Jer. Ep. 60.16; Claud. BGet 188–190, 611–614, 629–630; Cons. Hon.Con. 461–465, 471–473; Ruf. 2.186–191. 40. Beaton and Clement 1976, with a catalogue of IC 71-11 to 71-107 (none struck after 395); Clement 1976, pp. 228–229, pls. 160:b, 162. Isthmia V, p. 108, suggests that the burned layer overlying the Roman architecture near Tower 14 represents Visigothic destruction. Destruction horizons have been identified with the Visigoths at Corinth (e.g., Gregory 1979, p. 269; Williams and Zervos 1982, pp. 118, 144–145; 1983, pp. 23–24; 1984, p. 101, 1987, pp. 31–32; Corinth XVIII.3, p. 439), but some of these probably come from seismic events in the late 4th century. 41. Earthquake in 365: Amm. Marc. 26.10.15–19; Jerome Chron. a.d. 366 (p. 244.16–19); Oros. Hist. 7.32.5; Socrates

4.3.3–5; Fasti Hydatiani a.d. 365 (MGHAA 9.1, p. 240); Chron. pasch. a.d. 365 (p. 556); [Dionys.] Chron. a.g. 675–676 (I, p. 134); Sym. Met. V. s. Athan. Alex. 29 (PG 25.ccx); Theoph. Conf. a.m. 5859 (p. 56); George mon. II, p. 560.8–561.19. Earthquake in 375: Zos. 4.18.1–2, 5.6.3. Jacques and Bousquet (1984, p. 436 n. 52) and Evangelatou-Notara (1987–1988, pp. 429, 431, n. 17) suspect Zosimus’s chronology; Rothaus (1996, p. 106; 2000, pp. 17–18) impugns his veracity. 42. Schiffer (1987, pp. 231–233) addresses earthquakes as an archaeological formation process. Stiros (1996) discusses in detail the identification of earthquakes from archaeological data in Greece and Asia Minor. 43. Excavations at Corinth and Kenchreai have revealed extensive destruction during the last third of the 4th century: MacIsaac 1987, pp. 100–101; Williams and Zervos 1987, pp. 31–32; Rothaus 1996, pp. 106–107; 2000, p. 21, n. 52 (Corinth); Kenchreai I, pp. 75–76; II, pp. 1, 268–269, 410; Kristalli-Votsi 1984, p. 64 (Kenchreai). Some damage that has been associated with Alaric’s sack of Corinth, especially burning, was probably seismic (n. 40, above). Inscriptions from Corinth and Nauplia point to restorations under Imperial sanction in the 360s–370s, apparently in response to earthquakes: IG IV 674; Corinth VIII.3, nos. 504, 505; Feissel and Philippidis-Braat 1985, pp. 273–274, 274–275, nos. 6, 7, 9. 44. Gebhard 1973, p. 134, n. 65; Wohl 1981, pp. 116–118; Gregory 1995, p. 293, 303; Rothaus 1996, pp. 109–110. The concrete vault over the Northeast Reservoir might have fallen at this time as well (E. R. Gebhard, pers. comm.).

120

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

extent is a matter of speculation.45 Like the Temple of Zeus at Nemea,46 the substantial structures of the Temple of Poseidon and its stoas probably sustained damage but were not overturned. However, smaller buildings such as the domed Temple of Palaimon would have collapsed, and the area must have become littered with debris from fractured walls and severed pipes. During the 3rd to 4th centuries, as the competitions ended, as ritual activity and dedications slowly diminished, and as the sacred buildings deteriorated from a combination of growing disuse and sudden catastrophe, people still inhabited the site. They would have depended on local production and exchange with the numerous travelers who passed en route to Corinth.47 The settlement appears on the Tabula Peutingeriana, the illustrated itinerary of the mid-4th century, though it is not designated by a rectangular building with a gabled roof, the map’s usual icon for a sanctuary (seg. VII.5, col. 565). Buildings in the East Field have produced deposits of the 2nd to 3rd centuries as well as sherds and other debris from habitation in immediately overlying deposits of the 4th to 6th centuries.48 It is possible that the occupation of this area was punctuated by brief periods of reduced residence or even abandonment, such as during the Visigothic advance or the earthquakes, but the overall picture is one of stability. The graves on the north edge of the site (T2 68-003, DEC 69-901, and 69-902) may in fact belong to the settlement during the 3rd to 4th centuries rather than the mid-1st to 2nd centuries.

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES) The next two phases in the burial chronology will be examined together, because they represent the continuous occupation of the fortifications during their erection, early employment, and rehabilitation. Phase II includes NEG 69-103, 69-004, and 69-008 (Group I) and HO 70-902, which together contained at least 18 individuals.49 The burials in these graves can be securely dated from the end of the 4th to early 5th centuries, or around the time of the construction of the Hexamilion and Fortress in ca. 410–420. This date is based both on the situation of the graves relative to the footings for the curtain wall and on associated artifacts. At least two of these graves, 69-004 and 70-902, were reopened on several occasions for the addition of bodies. Phase III includes NEG 67-001, 67-003, 69-005, and 69-001 (Group II) and RB 76-002, which together contained 26 individuals. The burials at the Northeast Gate can be dated by architectural associations, burial deposits, and funerary artifacts to the mid- to late 5th century, while stratigraphy and artifacts place the burials in the Roman Bath in the late 5th to early 6th centuries. All these graves were repeatedly opened to add bodies. Two other graves, HO 70-901 and PAL 56-001, which together contained four individuals, belonged to either phase II or III. According to the local depositional sequence, the secondary burial of 70-901 occurred after the end of the 4th century but before the late 6th century. The burial of 56-001 is dated to the early 5th to mid-6th centuries by its stratigraphic and architectural context. 45. Rothaus (1996, pp. 108, 109; 2000, p. 89, n. 19) proposes drastic or complete seismic destruction to the Temple of Poseidon, but physical evidence is lacking. 46. Hill 1966, p. 1, n. 2; Miller 1977, p. 3; 1978, p. 59, n. 6; 1979, p. 74; 1980, p. 181; 1981, pp. 48–50; [Stella] Miller 1983, p. 72; Miller 1986, pp. 263–264; Nemea Guide, p. 154; Nemea I, pp. 44–46, n. 135. 47. The latest phase of the Roman road that passed through

the Arch and across the north side of the temenos of Poseidon was long, perhaps extending into Byzantine times (E. R. Gebhard, pers. comm.). 48. T. E. Gregory (pers. comm.); Rothaus 2000, pp. 89–90, n. 23. 49. This total is based on the assumption that 69-008, in which no significant skeletal remains were found, contained at least one individual.

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES)

121

The 11 graves comprising phases II–III, which contained at least 48 individuals, represent inhabitants in the area of the Fortress during roughly its first century and a half. It will be seen from the archaeological and historical sources that this occupation can be further divided into three subphases. These are the construction and initial employment of the fortifications, perhaps the time of densest occupation of the site; a period of relaxed defense and partial disrepair during the second half of the 5th century but probably extending into the 6th century; and the renovation of the fortifications in the mid-6th century.50 The era following the zenith and demise of the Roman Sanctuary (phase I) was dominated by the erection and operation of the fortifications. The disastrous intrusion of the Visigoths, perhaps coupled with the distant threat of the Huns, prompted intense fortification in the form of barrier walls and urban citadels throughout the Balkan peninsula during the early 5th century.51 The strength of the Isthmus of Corinth was crucial, because it was a geographic bottleneck and linchpin in the defense of the Peloponnese against northern invaders. A date of construction for the Hexamilion and its fortress early in the reign of Theodosius II, most likely during the second decade of the 5th century, is established by 18 coins from the interstices and clamp cuttings in the road pavement inside the Northeast Gate. Chronological evidence from throughout the fortifications corroborates this date and shows that the construction was a single, unified project.52 The planning and building would have been overseen by the praefectus praetorio of Illyricum and financed by the region through revenue and corvée.53 The duration of the building project is difficult to estimate, but it probably did not exceed a few years and might have taken much less time.54 The fortification of the Isthmus required the erection of a curtain wall, towers, coastal bastions, and an adjoining fort. The design of the Hexamilion and its stronghold accommodated existing topography by following natural contours, such as the Great Ravine, and incorporating earlier structures, such as the Bath and the Arch. The Northeast and South Gates marked the route of the major road entering the Sanctuary and the Peloponnese beyond through the enceinte. An artificial plateau was created for the Fortress by dumping soil and stone debris over the eastern portion of the building site, which would have obliterated or buried earlier structures. Once a level plain was created, the walls of the Fortress and its interior buildings were made from rubble, mortar, and recycled stone. Building material came from different sources, some of it freshly quarried but much of it reused, especially in the segments of wall closest to the Sanctuary. The construction of the fortifications involved the wholesale despoliation of the major architecture of the sacred precincts, including the Temples of Poseidon and Palaimon, the stoas and gates, the South Building, and the Theater. It is unknown when the removal of architectural members began, but it was certainly underway during the initial fortification in the first or second decade of the 5th century. These efforts not only stripped stone from the ancient buildings but also deposited structural debris and stone chips over the surface of the precincts and inside the waterworks.55 The next stage in the development of the Hexamilion and the Fortress was rehabilitation under Justinian roughly a century and a half later, the last known involvement of central 50. Gregory (Isthmia V, pp. 141–144) discusses the construction and early employment of the fortifications. 51. Clement 1975, p. 163; Isthmia V, pp. 11–12, 141–142. In general, see Gregory 1992, pp. 242–245; Crow 1995, pp. 118– 120. On local efforts, see Gregory 1979 (Corinth); Fowden 1988, pp. 58–59; 1995, pp. 553–556 (Athens); Cherf 1991, 1992 (Thermopylae). 52. Clement 1975, pp. 159–163; 1987; Isthmia V, pp. 74–77, n. 39. 53. Isthmia V, pp. 143–144, nn. 18, 19; Fowden 1995, pp. 551–

553; Avramea 1997, pp. 63–64. 54. Kardulias (2005, pp. 102–106, table 7.4) estimates the expenditure of 145,747 person-days on erecting the Fortress alone, but this must be considered a most general estimate based on several assumptions. 55. Isthmia I, pp. 1–2, 103; II, pp. 95, 96, plans II, IV, pl. 55:a; III, p. 3, n. 9. John Hayes has identified the pottery in destruction levels in the central area, which will appear in his forthcoming volume in this series.

122

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

government in their operation until the 15th century (phase V). Justinian’s activities on the Isthmus belonged to a global program to refurbish enfeebled defenses, not to build new ones.56 Procopius wrote (Aed. 4.2.27–28) that Justinian strengthened the ruinous wall over the Isthmus by building “forts and guardposts” (φρούριά τε καὶ φυλακτήρια), while the Peloponnesian cities remained without walls. A famous inscription discovered at the South Gate of the Fortress named one Victorinus as the supervisor of the project and invoked the protection of God (IG IV 204). These texts place the reconstruction most likely between 548 and ca. 555, after the death of Theodora, who is not mentioned epigraphically, but before Procopius completed his treatise on buildings.57 He did not completely rebuild the Isthmian fortifications but reinforced certain points, such as the section of wall along the north side of the Bath, and enlarged others, such as the bastions around Towers 15 and 19.58 This effort was not as massive as the initial construction, but it would have involved Imperial troops and a large labor force assembled from local residents. Apart from its ideological significance as a symbol of regional security and a statement of Imperial power, the operation was a practical response to both defensive and structural concerns. The Kutrigurs had invaded Greece as far south as the Isthmus in 539/40; they returned in 559 but did not pass Thermopylae.59 Furthermore, seismic damage would have necessitated repairs. Late Antique and Byzantine historians mention an earthquake that hit Dyrrachium and Corinth in 522.60 Historians of the 6th century also report an enormous earthquake striking a zone from Patras and the Corinthian Gulf to Thessaly and the Malian Gulf probably in the spring or early summer of 552.61 This event would have resulted from faulting and crustal extension in the rift valley of the Gulf of Corinth, tectonic processes that have caused frequent earthquakes in central Greece and the northern Peloponnese for thousands of years.62 In short, the Corinthia probably sustained seismic damage at different times during the 6th century, including both 522 and 552, if not on other occasions as well.63 Like major events in the 19th and 20th centuries, earthquakes in the 6th century could have debilitated the infrastructure of Corinthian communities. Coin hoards and collapsed structures at Corinth, Lechaion, and Kenchreai have been associated with these events,64 but the identification is not in every case conclusive. Procopius wrote that Justinian refortified

56. Gregory (1992, pp. 246–248; 2000) and Fowden (1995, p. 551) contrast the archaeological evidence for Justinian’s renovations with Procopius’s bland panegyric to Justinian’s innovations; Whitby (2000b, pp. 718–719) gives a more positive treatment. Dunn (2004) shows that the Justinianic program had a deep impact at least on the Macedonian countryside. 57. Isthmia V, pp. 12–14; see also IGCB I 1, pp. 1–5; Feissel and Philippidis-Braat 1985, pp. 279–281, no. 16; Avramea 1997, p. 65; Feissel 2000, pp. 92–93, nos. 21, 22. The date of the composition of Aed. has been much debated; see Croke 2005, pp. 427–429 for a recent treatment. Victorinus was perhaps an Imperial architect specializing in military construction or the praefectus praetorio. He was active throughout the Balkan provinces, according to four inscriptions from Byllis in Epirus Nova (SEG XXXVIII 530–533; Feissel 1988; 2000, p. 92, nos. 17–20). 58. Isthmia V, pp. 80–83, 101–102. 59. Procop. Bell. Pers. 2.4.10–11; Agath. 5.23.6; Kordosis (1981a, pp. 63–64) provides commentary. No traces of a Bulgarian invasion in the mid-6th century have been identified at either the Isthmus or Corinth. 60. Malalas 17.15.62–69; Evagrius Schol. 4.8; Theoph. Conf. a.m. 6014 (p. 168). 61. Procop. Bell. Goth. 4.25.16–23; Evagrius Schol. 4.23.

62. Ambraseys and Jackson (1990, 1996) document the impact of earthquakes in the Corinthian Gulf on settlements in contiguous regions during recent centuries. G. D. R. Sanders has denied (1999, pp. 474–475; 2004, pp. 170–171; 2005a, pp. 12–13; 2005b, p. 439, n. 33; see also 2002, p. 648) that the earthquake in 552 affected Corinth. Nonetheless, the cited studies of regional seismicity show that several disturbances since the mid-18th century (e.g, 1748, 1785, 1853, 1887, 1914) with epicenters in the areas of Achaea or Boeotia caused structural damage at Corinth or generated shock waves detectable from Corinth to the Spercheios delta and beyond. The author thanks N. N. Ambraseys for informative correspondence about the archaeoseismology of central and southern Greece. 63. Other passages identify either the event in 522 or (an) other one(s) in following years at Corinth (Procop. Anec. 18.42–45; Cosmas Ind. 1.22). 64. Coins: Corinth XVI, pp. 7–8; Gregory 1979, pp. 272–274; Dengate 1981, pp. 175–178; Avramea 1983, pp. 52–53, 56, nos. 2, 3, 8. Damage: Kordosis 1981a, p. 64; Evangelatou-Notara 1987–1988, p. 438; RBK IV, 1990, col. 749, s.v. Korinth (D. I. Pallas). One inscription records renovations at Corinth possibly after seismic damage in the 6th century (Corinth VIII.1, no. 245; IGCB I 2a, pp. 9–10; Robert 1960, p. 22; Feissel and Philippidis-Braat 1985, pp. 293–294, no. 33).

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES)

123

Corinth on account of seismic damage (Aed. 4.2.24); presumably this effort was concurrent with the Isthmian operations.65 Earthquakes would have contributed to the deterioration of the already dilapidated structures of the Isthmian Sanctuary. By the end of the 6th century, the roof and walls of the Roman Bath that had survived the 4th century toppled violently, most likely from severe groundshaking.66 This overview of the military and natural history of the Late Roman Corinthia provides a basis for investigating more closely the development of the Isthmian settlement. Geophysical and surface surveys within the enceinte have elucidated the nature of habitation. Rectilinear walls were detected in the western and southeastern areas, and large amounts of debris from the settlement, including tile and brick, marble revetment and detailing, tesserae, kiln elements, glass, and fine- and coarse-ware pottery were found on the surface. While it is difficult absolutely to distinguish Early to Middle from Late Roman or Byzantine remains inside the Fortress, much of the structural and domestic debris should probably be assigned to late antiquity, the era of densest settlement.67 A provisional plan has identified at least 10 buildings of various sizes, which might have included living quarters, storehouses, workshops, and stables. Spatial analysis and historical and modern analogy suggest that the capacity of the Fortress was about 2,000 occupants.68 Although this estimate is based on a hypothetical reconstruction of buildings and much of the actual plan remains undetected, the large area of the enclosure (27,000 m2) and comparison with other Late Roman forts suggest that the Isthmian Fortress could have held a few thousand permanent residents in close quarters. It is, however, unknown whether even in times of intensive use the Fortress was ever occupied to its full capacity. Although the occupation of the fortifications was concentrated in the enceinte, scattered traces of habitation during this era have also been found outside its walls. The temenos of Poseidon and the Theater have yielded evidence for activity after the demise of local cults and the despoiling of the sacred monuments. A substantial cistern was built east of the Temple to serve a large number of people settled nearby for a long time. This tank may belong to the time of the fortification project, when builders would have wrecked the eastern entrance into the temenos.69 The industrial or domestic use of the Theater is indicated by a small kiln or oven, a clay pit, scattered sherds, and a rubble wall perhaps forming a pen, found in different parts of the building.70 All these features apparently belonged to the 5th century, and it seems reasonable to associate them with the building and earliest occupation of the Fortress. The Bath, which was collapsing by the end of the 4th century from seismic damage, has also produced evidence for occupation during the early 5th century and probably later. Excavation uncovered a large amount of pottery, mostly amphoras, among the hypocausts of room IX, hundreds of fragments of lamps, mostly of the late 4th century, mixed with burned matter on the floor of rooms I and VI, and scorched patches from fires over the mosaic in room VI.71 The deep, jagged cuttings through the floor on the north side of room VI and from room IV to the southwest corner of room II were also made during the 5th century, apparently for drainage.72 The abundant pottery suggests that the local 65. Gregory (1979, pp. 272–274) addresses the archaeological evidence from Corinth; IG IV 204 might have adorned the new city wall (Isthmia V, p. 14). 66. Gregory 1993b, p. 149; 1995, pp. 301, 303. 67. Gregory and Kardulias 1990; Kardulias 2005. 68. Gregory and Kardulias 1990, pp. 504–505, table 3; Isthmia V, p. 131; Kardulias 1992; 1993; 2005, pp. 95–101. 69. Isthmia II, p. 96, plans IV, V, pl. 37:b. On waterpipes in the central area that might have served Late Roman residents east of the temenos and near to or inside the Fortress, see

n. 23, above. 70. Gebhard 1973, pp. 134–135. 71. Marty 1993, pp. 126–129 (amphoras); Wohl 1981, pp. 116–140, fig. 3; Corinth XVIII.2, p. 20, n. 82; Karivieri 1996, pp. 49–50 (lamps); Gregory 1995, pp. 286–287, 298, 301, 303. It is noteworthy that numerous lamp fragments of Late Roman date were also found in the Theater (Isthmia III, pp. 89–90). 72. Gregory 1995, p. 303, pls. 53:b, 56:c, 58:b.

124

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

population was large, while the stored amphoras and the dumped lamps point to habitation in the area. But the makeshift nature of the channels, the absence of permanent structures, and the presence of temporary hearths suggest transient use. In addition, thin deposits containing Late Roman domestic wares in the buildings of the East Field point to occupation after the decline of the Sanctuary. Since these buildings, which were damaged by earthquakes and plundering, were not renovated, this occupation too must have been short-lived. The local population was thus focused within the enceinte, and the occupation of the extramural zone of fallen buildings seems to have been sporadic. The main activities in peripheral areas, at least during the very early 5th century, seem to have been the demolition of earlier structures for building material and the disposal of refuse. Traces of habitation at the temenos and the Bath might have been associated with the initial fortification program, when numerous workmen labored and perhaps even camped amid the ruins of the Sanctuary. It is possible that areas outside the Fortress were occupied at a later date too, such as during the Justinianic operations. Local residents played a role in regional defense. Engineers and troops from the prefecture would have been involved in the construction and renovation efforts, and an expeditionary force might have occupied the Isthmus in times of peril. Otherwise, the permanent inhabitants of the fortifications would have been either regular soldiers or a militia assembled from rural smallholders.73 In the first case, stationary troops not unlike the limitanei who guarded the frontiers would have resided in the Fortress and manned the Hexamilion. As such, they would have been local recruits under military jurisdiction who could own land subject to civilian authority but who oversaw the security of a strategic line of defense. In the second case, a rural militia with little formal organization would have resided in and around the Fortress but served as defenders only when needed.74 The occupation of the Isthmian fortifications by farmers, like Thermopylae before its Justinianic garrison (Procop. Aed. 4.2.14–15; cf. Anec. 26.31), is most likely to have occurred during times of relaxed employment, namely, the late 5th to early 6th centuries. At other times, the Fortress might have housed a detachment of regular soldiers, such as during and after the renovations in the mid-6th century, also like Thermopylae in the 550s (Agath. 5.23.6). Such shifts in occupation are well documented in two Theodosian constitutions regarding the land-walls of Constantinople (Cod. Theod. 7.8.13, 15.1.51 = Cod. Just. 8.11.18).75 These laws show that local residents continued to occupy the area after the erection of the fortifications, that structural deterioration over time was anticipated, and that regular troops periodically moved in, at which times local residents accommodated them. The graves of those inhabiting the Isthmian Fortress were almost all placed along the fortification walls. It was not an uncommon practice in the Late Antique Greek world for those charged with defending fortifications or simply living near them to bury their dead along the walls or inside the towers.76 The individuals buried during phase II might have been affiliated with a detachment stationed at the ruined Sanctuary to build the fortifications, whether as soldiers or laborers.77 Yet the simple designs and contents of the graves from 73. See Isthmia V, p. 79; Gregory and Kardulias 1990, pp. 505–506; Kardulias 1993, p. 146; 2005, p. 121. 74. On the broad tendency toward local military autonomy in the Late Antique Balkans, see Wozniak 1982, pp. 202–204; Dagron 1984, pp. 17–18; Whitby 1988, pp. 181–182; 2000a, p. 482; 2000b, p. 729. 75. Fowden 1995, p. 553. 76. E.g., 7th-century graves of soldiers in a tower in the southwest circuit of Corinth (Corinth III.2, p. 68, fig. 49; Davidson 1937); late-6th-century mortuary chapel and tombs

in and around a tower in the northeast circuit of Thessalonica (Tsigaridas 1977, pp. 478–479, fig. 1, pls. 439:a–440:a); 5thto 6th-century graves of soldiers and their families lining the land-walls of Constantinople (Meyer-Plath and Schneider 1943, pp. 135, 139, 142–143, nos. 41a, 54, 72–78); hypogaeum with richly carved sarcophagus in necropolis along the same walls (Deckers 1993). 77. Isthmia V, pp. 44 (HO 70-902), 88 (NEG 69-103); Gregory 1995, p. 290 (RB 76-002).

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES)

125

both phases show no signs of military service. The easiest explanation is that those buried here were not regular soldiers, but the possibility that they were soldiers cannot be ruled out on the basis of the graves alone. They might have defended the Hexamilion, whether as regular troops or militia, but their station did not merit mortuary expression in the community. The fact that many of those buried were women and children demonstrates that families inhabited the Fortress. The age structure of the skeletal sample indicates that individuals of different generations were buried together, most likely representing parents, other adult relatives, and children. The proposal of familial relationships is supported by the relatively high frequency of certain genetically linked skeletal traits among individuals in the same grave.78 Multiple burials may even represent discrete households. Cohabitation within the settlement would not have undermined the site’s strategic importance either during episodes of relaxed employment or during temporary occupation by regular troops. Roman soldiers at post were legally permitted to marry and live with their families beginning in at least the early 3rd century.79 The burial of families at the Fortress on the Isthmus during phases II–III reveals that local residents enjoyed a fully domestic existence for several generations. The religious, social, and economic character of their community was complex. The earliest evidence for Christianity at the Isthmus belongs to the fortifications in the 5th century. Three crosses in raised relief framed the South Gate of the Fortress when it was built, and several Christian symbols, though difficult to date, were scrawled on the towers and the walls of the Fortress. Terracotta lamps bearing Christian imagery first appear during the mid- to late 5th century. A small Doric capital inscribed with a cross was deposited on the mosaic floor in room I of the Bath, perhaps during nearby military construction. A fragment of marble bearing an early chi-rho (monogram for Christ)was found in a Byzantine level inside the Northeast Gate, but its date and original location are unknown.80 To what extent these remains reflect personal belief and religious practice in the local community is a hard question. The use of apotropaic images such as the cross and pious invocations, which had a long pagan heritage, served in part to strengthen the Isthmian defenses and to shield their occupants from harm through divine protection. The lamps and graffiti point to the Christianity of the builders and residents. Even if the entire community had not converted, at least several inhabitants of the Isthmus already embraced monotheism by the 5th century. It is hard to know how long before the 5th century Christianity had reached the Isthmus and how rapidly it spread. The Church of Corinth was prominent from the time of Paul,81 who traveled through Kenchreai and probably visited the Isthmian Sanctuary.82 Considering the status of the Church and the international face of the Sanctuary, one might conclude that Christians occupied the Isthmus by at least the 4th century. During late antiquity the 78. For a full discussion of the osteological evidence, see pp. 291–292 and Table 5.18. 79. Jones 1964, pp. 630–631, citing Cod. Theod. 7.4.28 (a.d. 406) and 7.4.31 (a.d. 409) on rationing soldiers and their families in the Roman East; Southern and Dixon 1996, pp. 85–86. An anonymous Early Byzantine military treatise recorded that the cohabitation of soldiers and their families should be allowed in forts that were strong and well-supplied (De re strategica 9.29–33), but this explicitly refers to regular troops in permanent garrisons on the frontier. 80. Isthmia V, p. 93, pls. 28:d–29:b (crosses in relief); pp. 95, 96, 98, 103, 120, 125–126, 200, Ill. 23, pls. 29:d, 31:a, b (graffiti, some probably Byzantine); IA 72-75, NB MM II, p. 29 (Doric capital); IΣ 516+517 (trench 67-2, lot 67-RMM040; fragment with chi-rho); Isthmia III, pp. 78–79, 81, 89–90; Wohl 1981, pp. 139–140; 1993, p. 135 (Christian lamps; cf.

Karivieri 1996, p. 67, on the date of the first Christian symbols on Athenian lamps). 81. Paul at Corinth: Murphy-O’Connor 1996, pp. 26, 252– 273, 291–297. The first centuries of the Corinthian Church: Gritsopoulos 1973, pp. 48–74; Phougias 1997, pp. 37–131; Richardson 2002. 82. On Paul at Kenchreai, see Acts 18:18; Rom. 16:1; Const. apost. 7.46.10 (mentions the Cenchrean bishop but is questioned as a late invention in DHGE XII, 1953, col. 135, s.v. Cenchreae [R. Janin]). Paul would have passed through the Isthmus on his journey from Athens to Corinth. Broneer argued (1962b; 1971; Isthmia II, p. 55, n. 64) that Paul attended the Games in 51 on the basis of athletic exempla in his correspondence (e.g., I Cor. 9:24–27; II Tim. 2:5, 4:7–8); see however n. 3, above, on the date of the return of the Games to the Isthmus.

126

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

Corinthian Church exercised considerable influence, and the construction of basilicas in central locales at nearby towns in the 5th to 6th centuries points to sizable congregations on the Isthmus.83 While it is impossible to trace shifts in local belief and practice between the reigns of Theodosius II and Justinian, the number of Christian residents already at the Isthmian fortifications in the early 5th century would have only increased over time. No remains of an early church have yet been found, but one might well have existed in the Fortress by the 6th century.84 While some of the 48 individuals buried in graves of Late Roman date would have been Christians, in most cases neither the funerary artifacts nor the interments themselves overtly displayed the religious identity of the dead. One exception is the lamp with a cross that was deposited near 76-002 (6). It does not, however, follow that the other burials of phases II–III were pagan because they lacked lamps with crosses, including 69-008, which had a lamp without a cross (5). Traces of Christianity in the graves, such as artifacts with Christian imagery or epitaphs with Christian names, diction, and symbols, likely would not have been destroyed over time or missed during investigation. Indeed, the excavators recorded the presence of diverse perishable matter, including bone, wood, and textile or leather. If many or all of the deceased were Christians, then apparently they chose not to express their monotheism in death through material channels. Apart from the lamp, the first overt manifestations of Christian ideology in funerary ritual at the Isthmus belong to the Early Byzantine occupation (phase IV). Possible reasons for this slow transformation in mortuary behavior will be addressed in Chapter 4. The continuity of local settlement and material culture between the Early and Late Roman periods indicates that the inhabitants of the Fortress were in all likelihood indigenous Corinthians. They were probably descended from earlier inhabitants of the site, but families might have also moved in from neighboring areas. A degree of social differentiation is suggested by the variable quality of artifacts and architecture within the enceinte.85 The economic strategy of the residents depended on both local production and regional exchange. Primary subsistence activities would have included keeping diverse livestock, gardening and foraging, apiculture, and perhaps cultivating cereal grains either on a small scale inside the enceinte or more expansively outside the walls. The diverse faunal assemblage from Late Roman deposits shows that the inhabitants consumed various species that were raised or acquired locally.86 Chapters 6 and 7 will address the impact of these occupational activities and dietary habits on the teeth and bones of the residents. Apart from their local existence, to acquire additional foodstuffs, rare materials, and specialized services they would have communicated with other settlements in the region, including Corinth, the ports of Kenchreai, Schoinous, and Krommyon, and neighboring communities of tenants and laborers on the farmsteads and estates that filled the busy countryside.87 83. Gritsopoulos (1973, pp. 75–108) and Limberis (2005) survey the ecclesiastical history of Late Antique Corinth; RBK IV, 1990, cols. 759–787, s.v. Korinth (D. I. Pallas) and Sanders (2005b) cover the Christian monuments. 84. Gregory and Kardulias 1990, pp. 500–504; Isthmia V, p. 130; Kardulias 2005, p. 90. An early church may underlie Ayios Ioannis Prodromos; see pp. 147–148 on the long-term history of that building. 85. Gregory and Kardulias (1990, pp. 496, 500) and Kardulias (2005, p. 84) distinguish between officers and common soldiers. 86. Cf. Kardulias 2005, pp. 83–85, 120–121. Preliminary study of the faunal remains from the Fortress has identified cattle, pig, sheep-goat, chicken, and fish (D. S. Reese, pers. comm.). The evidence for beekeeping has often been discussed: Broneer 1959, p. 337, no. 17; Crane 1983, pp. 45–51;

Gregory and Kardulias 1990, pp. 489, 495–496, 509, n. 26, table 2, no. 3, pl. 77; Isthmia V, pp. 114–115, 127, nn. 5, 19, pl. 40:d; Evershed et al. 2000; Anderson-Stojanović and Jones 2002. Since the beehive sherds are hard to date precisely, they may belong to either the Late Roman or Early Byzantine periods. Vroom (2003, pp. 140, 144–145, 220, figs. 6.2, 6.40), discussing parallel examples from Boeotia, dates the sherds from the Isthmian Fortress to 550–600. 87. Gregory and Kardulias 1990, p. 506; Kardulias 1993, p. 146; 2005, pp. 112–113. Archaeological survey in the Corinthia and Argolid has revealed traces of flourishing settlement during late antiquity, including rural villas, though some sites operated at least into the 7th century: Gregory 1985; Rothaus 1994; Southern Argolid, pp. 400–404; Berbati-Limnes, pp. 338–340, 438–439; Avramea 1997, pp. 127–128; Tartaron et al. 2006, pp. 482–483, fig. 12; Pettegrew 2007; 2008.

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES)

127

In sum, the Late Roman community at the Isthmus consisted of Corinthians, many of whom were Christians. They represented the same settlement that had occupied the site during the efflorescence and decline of the Sanctuary. There is no historical testimony or material evidence such as weapons, military equipment, or soldiers’ epitaphs that a garrison of Imperial troops was permanently stationed and provisioned at the Fortress. Rather, for much of the 5th to 6th centuries the local community played a defensive role while engaging in small-scale, rural production and participating in a broader Corinthian network of trade and communication.88 Within this overarching pattern of relative stability and modest prosperity, minor changes in the occupation of the Fortress caused by evolving military use are likely to have occurred, especially fluctuations in the size of the resident population and intensified occupation, perhaps by Imperial troops, during building efforts. The local impact of the great plague that reached the Aegean Basin in early 542 is unknown, but the Corinthian countryside must have suffered, as did other rural parts of the eastern Mediterranean region.89 The topographical situation of the graves at the Isthmus attests to the concerns of the bereaved in planning for burial. Except for the unusual PAL 56-001, all graves from phases II–III were found on the northern margin of the site. Three explanations for this pattern are possible. The first is that the bereaved were concerned to separate corpses from the spaces of daily living. Dense remains of Late Roman settlement have been found in the central and southern areas of the enceinte. Because of the Great Ravine, residents could hardly have chosen a site for burial further removed to the north of the Fortress interior than the Northeast Gate.90 The second explanation is that local residents still considered the area to be the customary burial ground used by their predecessors (phase I). While the function and topographic significance of certain areas, such as the temenos of Poseidon and the Palaimonion, had evolved with the erection of the fortifications, the use and meaning of other areas apparently remained the same. The third explanation is that the residents were drawn to the main road for burial. If accessibility determined the situation of graves near the road in phase I, similar motivations might have influenced the placement of the three graves directly south of the Arch and alongside the road shortly before the construction of the Northeast Gate (Group I).91 All three of these explanations for the concentration of the graves of phases II–III on the northern edge of the site—segregation from inhabited space, recognition and use of an established burial zone, and proximity to the road—are plausible and, indeed, mutually consistent.92 Another pattern in the distribution of the graves is their location inside the curtain wall and in contact with or cutting into it. The only exception is 70-901, which was a secondary deposit of bones from another locale. This pattern is especially striking when compared to 88. Curta (2001a; 2001b, pp. 142–189) paints a different picture, based largely on evidence from the northern Balkans, of numerous frontier garrisons in a deserted or failing countryside dependent on centralized provisioning. It is unknown to what extent the inhabitants of the Isthmian Fortress benefitted from the annona militaris, if at all, but it did not sustain them. 89. Horden (2005) offers a recent survey of the “early medieval pandemic.” Several writers recorded its impact on Mediterranean countrysides: Procop. Anec. 23.20; V. s. Sym. styl. jun. 69; George heg. V. s. Theod. Syc. 8.1–2; V. s. Nich. Sion. 52; John Eph. HE 2, frag. F (pp. 231.20–232.17); Auctarium Marcellini, MGHAA 11.2, p. 107; [Zacharias Mytil.] HE 10.14. The graves at the Isthmus are too few in number and too general in date to be associated with plague victims. 90. Excavation in the immediate vicinity of the four midto late-5th-century graves at the Northeast Gate uncovered floors, pottery, and rubble walls west of Tower 19 dating gener-

ally to the mid-5th to mid-6th centuries (Isthmia V, pp. 83–84, Ill. 15, fig. 17; e.g., lots 67-RMM-026, -071, -072). This habitation does not seem to have been permanent or dense, and the dates of residence and burial are not precise enough to prove contemporaneity. 91. The Arch was probably in a damaged state when it was converted into the Northeast Gate (Isthmia V, pp. 53–55). In any event, the primary attraction for those planning burial would have been proximity to a functioning roadway. 92. An alternative hypothesis is that the dense concentration of burials inside the Northeast Gate reflects the proximity of a church. This is impossible to prove without further excavation, but the suggestion is intriguing: the Fortress probably had a church. The northern area of the enceinte is a strong candidate for the location of such a church, as no obvious ecclesiastical structures were located further south, and Ayios Ioannis Prodromos is due west of the Northeast Gate.

128

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

the distribution of graves during phase IV, a majority of which were outside the enceinte. The area north of the Hexamilion was, of course, external to the defenses. Even so, this would hardly have deterred residents from burying the dead there, unless it is to be assumed that Greeks during late antiquity always cowered behind their walls, too afraid of barbarian invasion to expose their burials to possible desecration.93 Such a scenario is particularly implausible for the Isthmus during the mid- to late 5th century (phase III).94 A better explanation is that inhabitants simply found interment within the Hexamilion convenient. The uneven and steeply eroded terrain north of the wall was less amenable to burial than the terrain south of it, which was level and easily reached, even if it was covered by ruins, such as the Bath. Until further excavation can illuminate this spatial relationship between the Hexamilion and Late Roman burial, the placement of known graves suggests that more are located inside the Fortress along the wall south and west of the Northeast Gate. The area of the Northeast Gate offers an opportunity to examine closely the spatial pattern of interment on a small scale (Figs. 2.1, 2.4). Burial advanced through space over time, from south of the Gate in the end of the 4th to early 5th centuries to west of the Gate by the late 5th century. Presumably those digging the graves were concerned to fill in vacant areas gradually and thus avoid overcrowding. The graves west of the Gate were, however, somewhat more closely arranged than those to the south. While the multiple burials undoubtedly represent families, there is no clear evidence for groupings of graves, unless the rubble wall shared by 67-001 and 67-003 was meant to associate those graves more closely with each other than with 69-005.95 All seven graves were arrayed along the walls at even intervals averaging ca. 0.80 m. This narrow separation may reflect a concern either to leave open ground for graveside rituals, which would have been very confined, or to ease movement for reopening cists. The careful arrangement of the graves at the Northeast Gate implies planning, so that none overlaps another. When the bereaved cut a fresh cist they remembered the location of earlier graves in the immediate vicinity and avoided placing the new grave there. The reuse of these graves for multiple interments further indicates that their location was remembered. Significantly, these graves do not seem to have been marked at the surface because the cists, some without cover slabs, were buried under shallow fill and no grave markers have been found.96 This raises the question of how those burying anew identified previous graves either to avoid or to reopen them. Perhaps they simply recalled the general placement of prior interments and then found the precise spot by quickly testing the topsoil with a spade. Recalling the location of the graves west of the Gate in this way would have been easy if they were interred over a short span of time in the late third to early fourth quarters of the 5th century. The situation of burials adjacent to the fortifications also facilitated the maintenance of an ordered burial ground and the discovery of earlier burials.97 In the absence of surface markers, preexisting walls or other structural features might have served as points of 93. Snively (1998, p. 496) uses such an argument to explain intramural burial in Late Antique Macedonia; see also Snively 1984, p. 121; Karagiorgou 2001, p. 195 (on Thessalian Thebes). Cantino Wataghin (1999) casts doubts on this interpretation of intramural burial in the Late Antique West, particularly Rome. 94. It is hard to imagine that rural Corinthians considered Hunnic activities in the northern Balkans during the mid- to late 5th century a real threat to their land and graves (Whitby 2000b, pp. 704–712). 95. The notion that burial areas were used by family mem-

bers is supported by the occurrence of a congenital anomaly in two skeletons buried at Tower 14 in the Early Byzantine period; see pp. 141, 408. 96. See p. 172, n. 54 on the absence of epitaphs at the Isthmus. Although grave markers might have been displaced between ancient burial and modern exploration, the surface markers for T2 68-002 and 68-006 were preserved in an area undergoing the same formation processes as the Northeast Gate. 97. Burial near the Northeast Gate does not seem to have extended much away from the walls, though it must be admitted that excavation did not reach very far into the enceinte.

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES)

129

reference for identifying nearby graves, which was a real need in a dense cemetery. The two most populous and therefore most frequently reopened graves, NEG 67-001 containing 11 individuals and HO 70-902 containing 10, are also the two graves with the most memorable sites. The former was right at the base of the stairway up to the platform over the Northeast Gate, while the latter was northeast of the Bath in the corner where the Hexamilion returns southward from its east–west course. The situation of the Northeast Gate graves relative to the fortifications also sheds light on the history of military occupation. Although the three graves south of the Gate predated the footings of the Hexamilion, they are evenly situated along the wall’s interior face and underlie its masonry by less than a meter. This consistency in placement along a wall not yet erected is inexplicable, unless those who cut the graves knew the approximate course of the Hexamilion prior to its completion. This strange circumstance makes sense if the bereaved did in fact participate in the wall’s construction and if its erection near the Gate occurred shortly after the burials. Decades later, the four graves west of the Gate were placed below the stairway up to the fighting platform of the north tower, the most strategic point in the defense of the north wall and the primary entrance into the Fortress. In particular, 67-003 displaced the bottom one or two steps of the staircase while 67-001 and 69-005 bordered it to the south and west. Local residents, whose choice of burial sites was intentional, were apparently unconcerned that anyone approaching the north tower would have to walk over at least two recent graves containing 14–16 corpses before climbing up a riser of nearly half a meter.98 Burial location at the Fortress thus suggests that the second half of the 5th century was a time of diminished employment and disrepair.99 Two graves of this period, PAL 56-001 and RB 76-002, exemplify the changing relationship between the local community and the Isthmian Sanctuary during the 5th to 6th centuries. The stone-covered cist 56-001 is remarkable for its placement within the temenos, its situation inside the monument, and its contents. It was the only grave cut in the central area during the entire Roman era (phases I–III), when the local population was sizable. All other graves were located a considerable distance from the precincts of Poseidon and Palaimon. Moreover, 56-001 violated the ancient prohibition against burial in sanctuaries, which Greeks had observed from early times.100 Although mourners made the interment in the Palaimonion after the sacred buildings had been dismantled and the sacrifices discontinued, they would have remembered the site’s significance. The dilapidated structures stood as graphic reminders of the cultic past until their plundering in the early 5th century and again in the mid-6th century. This process of gradual destruction left behind the scattered architecture and sculpture that excavators found centuries later. The central area was no longer operating when mourners buried the dead, but they would have been aware that the place was once a sanctuary preserved from the pollution of burial. The grave’s situation directly over the temple implies a deliberate choice to connect the burial with the monument. Mourners could have buried the corpse more easily in the unpaved ground nearby or inside the passageway through the temple’s foundations. Old water channels were not infrequently used for burial during late antiquity, as in the case of RB 76-002, because they afforded the same secure and lasting form as stone-lined 98. Pace Fowden 1995, p. 553 (“Only one step was removed, though, and there are several other staircases in the vicinity”); see the restored elevation of the stairway at Isthmia V, p. 66, Ill. 8. One similar stairway existed ca. 20 m south of the Gate (Fig. 2.3; Isthmia V, pp. 68–71, Ills. 10–12). 99. Isthmia V, p. 79, n. 51, compares the poor condition

of the city-circuit at Corinth, for which see Gregory 1979, pp. 272, 280. 100. Parker (1983, pp. 33–48) provides a classic discussion of death as a source of “pollution” (μίασμα) in early Greek religion.

130

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

cists. Instead, mourners made the impractical decision to cut the cist deep into the concrete foundation of the building’s facade and parallel to the tunnel through the podium (Figs. 2.77, 2.79). The entrance into the tunnel was a doorway 0.85 m above its floor, which might have been originally fitted with wooden stairs. The passageway continued to the center of the podium, where it connected with the earlier reservoir and extended underground over 17 m to the northwest. The prominent cutting for the interment alongside the tunnel would have disrupted the symmetrical arrangement of the podium’s front. That the opening of the tunnel was integral to the temple’s appearance is indicated by its inclusion as an arched portal in representations of the building on Corinthian coinage issued under Lucius Verus and continuing into the 3rd century.101 After the removal of the superstructure in the early 5th century, the entrance to the passageway and its course through the foundations would have been plainly visible until the entire area, including both tunnel and grave, was buried under a widespread deposit of debris by roughly the late 6th or 7th century.102 The form and contents of this grave also marked it as special among all known burials at the Isthmus. Its construction was unusually permanent and well sealed as a cist cut cleanly into solid concrete, furnished with a wooden coffin or bier, and tightly covered with slabs and mortar. Either the wooden structure or the walls of the cist were finished with plaster at the time of deposition. The deceased individual, probably an adolescent or adult woman, wore an earring (17) and a beaded necklace (33). Unlike other Late Roman burials at the site, 56-001 was a single burial isolated in an area without graves. In all these respects, those who buried this woman represented her as an outstanding individual who had achieved a measure of prestige in the community. She might have been a Christian, like others at the Isthmus during this era; the absence of Christian symbolism from the grave is not conclusive with regard to religious devotion.103 This exceptional burial calls for explanation. One possibility is that the mourners were trying to substantiate or augment the status of the deceased through their selection of an unusual burial locale. The situation of the cist would have appropriated the temple as a sort of funerary monument. The placement would have also displayed the grave for whomever passed by, including travelers on the east-west road across the ruined temenos of Poseidon. Such a conspicuous monumental site could have reinforced an image of the woman’s importance that mourners had already created by adorning the corpse, using a wooden structure and plaster, and carefully covering the grave. But it should be remembered that, since the ground level from which the cist was cut cannot be ascertained, it is unclear how much of the stripped podium was visible during and after the time of interment. It could have been as little as ca. 0.20–0.50 m or as much as ca. 0.80–1.00 m.104 In either case, while the bare concrete foundation would have been visible at the surface, the grave was hardly exalted by an elevated position or an impressive architectural setting. Two essential questions remain: why did mourners choose the Temple of Palaimon in particular, when other conspicuous locales for burial existed in the area, and why did they deliberately cut a cist into the facade but alongside the tunnel, where interment would have been easier? The purposeful nature of their actions reflects a consciousness about a peculiar meaning attached to the physical setting. During late antiquity, ancient ruins possessed a

101. Gebhard 1993a, p. 91; Gebhard, Hemans, and Hayes 1998, pp. 440–441. For examples of the coins, see ImhoofBlumer and Gardner 1964, p. 11, pl. B:XI, XII, XIII; Isthmia II, pp. 110–111, n. 14, pl. 42:a, b (IC 71-149 from the East Field). Oikonomidou (1992–1993) gives a general discussion of Co-

rinthian coinage depicting Melikertes-Palaimon. 102. See p. 96 for a stratigraphic summary. 103. On the identity of this individual, see further pp. 215– 216. 104. See pp. 91–92.

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES)

131

referential or symbolic value as repositories of memory, a quality made especially potent during ritual activities.105 The Late Roman community at the Isthmus necessarily constructed its identity in relation to the existing monumental landscape. If some memory of the site’s religious history survived during the 5th and 6th centuries, as it did throughout the 4th century, then the location of 56-001 within the precinct and over the temple at the tunnel’s entrance would have struck viewers as odd. The grave’s situation over the dilapidated temple drew a sharp and obvious contrast between the two, especially when the cist was cut, the graveside rite performed, and the body deposited. One possible explanation is that the association of the corpse with the temple bore ideological significance. The mourners might have been Christians promoting the degeneration of the Sanctuary by introducing the burial of a prominent believer. Indeed, if memories of the Palaimonion were strong, it would be difficult to reconcile polytheistic piety with the bold placement of the grave. Perhaps the structural violation of the hero’s shrine through Christian burial was a symbolic derogation of the hero’s cult. The use of burial as a dynamic medium for the visual and ritual negotiation of pagan and Christian space is well known in other contexts. There are numerous instances of martyria placed over pagan sites to desecrate earlier cults.106 For instance, in his biography of Syrian monks during the 4th to 5th centuries, Theodoret of Cyrrhus recounted how Thaleleius had consecrated a pagan sanctuary near Gabala by occupying it and later founding a martyrium there, after he “arranged the godly dead opposite those falsely called gods” (τοῖς ψευδωνύμοις θεοῖς τοὺς θείους ἀντιτάξας νεκρούς, Hist. rel. 28.1–2, 28.5). The same writer elsewhere praised the triumph of the “excellently victorious martyrs” (καλλίνικοι μαρτύροι) over the ancient deities through the construction of martyria at old places of worship and the replacement of pagan festivals by celebrations of martyrdoms (Affect. 8.68–69).107 The most famous case is Babylas, the bishop of Antioch who died in ca. 250 and whose martyrium was built over the Temple of Apollo at Daphne. Interring his coffin within the temenos silenced the oracle of Apollo. This prompted the emperor Julian on October 22, 362 to exhume the corpse and restore the temple, to which the Antiochenes then set fire.108 These examples obviously differ from the grave in the Palaimonion. The institution of martyrs’ relics in ancient precincts reflected an intense antipathy between eastern holy men and polytheistic cults. In contrast, there is no indication that the individual in 56-001 was a martyr, no church has been found in the central area, and the grave in the derelict shrine did not halt the worship of the hero. The Syrian martyria and the Isthmian grave did, however, share certain basic features: high status, conspicuous placement, and the superposition of a new burial over an old temple. If 56-001 was meant to deface the Palaimonion and thus countermand its ancient inviolability, the Isthmian Sanctuary could be added to only a few other known sites where Christian Greeks actively displaced traditional religion.109 According to one dominant and longstanding viewpoint, the demise of paganism was independent from the rise of Christian communities, represented chiefly by churches, at major cult 105. Cf. Smith 1987, pp. 83–95; Effros 2003, pp. 100–101. 106. Maraval (1985, pp. 53–54) summarizes the eastern evidence; MacMullen (1997, pp. 124–125, n. 66) collects additional examples from both East and West. 107. Trombley (2001a, vol. 2, pp. 143–173) generally addresses the activities of the Syrian monks at pagan sites, and (pp. 363–364) discusses the erection of the martyrium of St. George at Zorava in Trachonitis (Arabia) over a pagan temple. 108. The earliest sources (e.g., Euseb. Hist. eccl. 6.39.4) place Babylas’s death under Decius (249–251), but later hagiography places it under Numerianus (283–284). The tradi-

tion concerning Babylas’s burial at Daphne is extensive. The following comprise the major testimony: Julian. Mis. 361B–C; Lib. Or. 17.7, 18.282, 60.5, 60.9 et schol. ad loc.; John Chrys. Bab. 2–3 (PG 50.530–534), Bab. c. Jul. 14 (PG 50.554), Hom. IX in Eph. 4.2 (PG 62.71); Amm. Marc. 22.13.1–3; Philostorg. 7.8; Socrates 3.18; Theodoret HE 3.10.2–3; Sozom. 5.19.17–18; Evagrius Schol. 1.16; Theoph. Conf. a.m. 5854 (pp. 49–50); Passio s. Bab. (PG 114.967–982). Samellas (2002, pp. 163–165) discusses the historical circumstances. 109. See n. 37, above.

132

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

centers.110 Recent scholarship has shown that in many places pagans and Christians coexisted but in an uneven, sometimes adversarial counterpoint involving a slow process of absorption. The nature of interaction between polytheistic cult and Christian worship on a local scale in the Greek world was a complicated exchange of images, behaviors, and spaces as each group defined itself in relation to the other.111 The grave over the Temple of Palaimon may be considered a vivid illustration of this process. Regardless of whether the grave was intended to debase the hero cult, its intentional situation next to the underground passage established a connection between the two that recalled the tradition of Palaimon. Greek authors during the Empire wrote that the cult was observed at the boy’s burial112 and that the rituals were funerary and chthonic in nature.113 When Pausanias visited the Isthmus he saw the Hadrianic Temple of Palaimon in a precinct east of the temenos of Poseidon. He also saw in a separate place “the so-called sacred chamber and the descent to it underground” (ἄδυτον καλούμενον κάθοδος δὲ ἐς αὐτὸ ὑπόγεως), where it was said that Palaimon “had been hidden” or “buried” (κεκρύφθαι, 2.2.1). Philostratus the Younger seems to have imagined the same place when he wrote that Poseidon tore open a “sacred chamber” (ἄδυτον) in the earth to receive Palaimon (Imag. 2.16 [p. 362.27– 29]). Elizabeth Gebhard has identified this ἄδυτον with the Stadium Reservoir, which was rediscovered in the Early Roman period and associated with the hero.114 Here worshippers carried lamps for nocturnal rites involving mysteries; they also made holocaustic sacrifices over a pit located to the east. When the second temple of Palaimon was built immediately southeast of the Stadium Reservoir, the tunnel through its foundations linked the shrine directly to the ἄδυτον. It is uncertain what exactly occurred in this underground space, and various uses have been proposed.115 Surely it played an important part in the worship of the hero and, according to at least one well-known tradition, his body resided there. The Late Roman burial might thus have been viewed as intruding upon an ancient hero’s tomb. If mourners at the funeral and subsequent visitors remembered the hero cult,

110. E.g., Frantz 1965; Agora XXIV, pp. 70–71, 92 (Athens); Spieser 1976, pp. 314–315 (Olympia), 316–317 (Delphi); Spieser 1984, pp. 321, 328; Dagron 1977, p. 5; Hanson 1977–1978; Vaes 1984–1986, pp. 326–338; 1989, pp. 303–304; Déroche 1989; D. I. Pallas in Déroche 1989, pp. 2723–2725 (Delphi); Pallas 1989, pp. 887–888 (Attica); Caillet 1996, pp. 196–197; Avramea 1997, pp. 114–115. For a detailed refutation of cultic continuity at a site in Athens, see the critical discussion in Lalonde 2005. 111. E.g., Trombley 1985; Gregory 1986; Fowden 1990, pp. 500–501; Saradi-Mendelovici 1990, esp. p. 48, n. 8; Moutzali 1993; Saradi 1996–1997, pp. 404–405; Avramea 1997, pp. 148– 154, 162; 2000, pp. 17–18; Haldon 1997, pp. 327–329; MacMullen 1997, esp. pp. 103–159; Price 1999, pp. 164–171; Harl 2001, pp. 309–313; Trombley 2001a, vol. 1, pp. 116–118 (Athens); Chuvin 2004; Chaniotis 2005, pp. 147–148. Rothaus (2000) explores this interaction in the Late Antique Corinthia. 112. Aristid. Or. 46.40 (rituals performed “at his grave” [ἐπ᾿ αὐτῷ]); Clem. Al. Protr. 2.34.1 (games are “graveside festival” [ἐπιτύμβιος πανηγύρις]; cf. Iambl. VP. 52; Euseb. Praep. evang. 2.6.10); Philostr. Her. 53.4 (rituals performed “at the grave of Melikertes” [ἐπὶ Μελικέρτῃ]); cf. Schol. in Pind. Isth. hyp. c lines 9–12; Paus. 2.1.3 (Sisyphus buried boy on Isthmus); Zen. 4.38 (Corinthians buried boy at Corinth). The use of ἐπὶ Μελικέρτῃ or ἐπ᾿ αὐτῷ to qualify the institution of the Games and/or its rituals (e.g., [Apollod.] Bibl. 3.29; Plut. Thes. 25.4; Clem. Al. Strom. 1.21.137.1; Zen. 4.38; Hellad. Chrest. ap. Phot. Bibl.

279.533b.31; Tzetz. Schol. in Lyc. 107) illustrates the semantic proximity of ἐπί + dative meaning both “in honor of the deceased” and literally “over the body of” (LSJ s.v. B.I.1b). Philostr. VA 3.31 implies that Melikertes, Palaimon, and Pelops received “tombs” (τάφοι) and commemorative “contests” (ἀγωνίαι). Pfister ([1909–1912] 1974, pp. 405–412) lists sanctuaries containing the graves of heroes and (pp. 495–497) games established at the graves of heroes. 113. IG IV 203, line 9 (2nd century) called the sacrificial pit a “place for offerings to the dead” (ἐναγιστήριον; Geagan 1989, pp. 358–359); Philostr. jun. Imag. 2.16 (p. 363.4) called the sacrifices “offerings to the dead” (ἐναγίσματα); Philostr. Her. 52.3, 53.4 compares the rites of Melikertes with those enacted by the Thessalians at Achilles’ tomb in the Troad, which involved nocturnal singing, a mystery cult, and “offerings to the dead” (ἐναγίσματα). Ekroth (2002, pp. 74–128) discusses sacrifices to the dead in hero cults (pp. 80–81 specifically addresses the cult of Palaimon). For the funerary character of the cult of Melikertes-Palaimon from early times, see Burkert 1983, pp. 197–198; Gebhard 1993b, pp. 170–172; DeMaris 1995, pp. 665–666; Piérart 1998; Gebhard and Dickie 1999; Gebhard 2005. 114. Gebhard 2005, pp. 197–203. 115. E.g., Isthmia II, pp. 27, 110–111; Gebhard 1993a, pp. 93–94; Marchetti, Kolokotsas, and Abadie-Reynal 1995, pp. 203–205; Piérart 1998, pp. 100–106; Gebhard 2005, pp. 199–200, 202–203. Koester (1990, p. 364) considers it “difficult to explain.”

PHASES II–III: THE LATE ROMAN FORTIFICATIONS (END 4TH TO MID-6TH CENTURIES)

133

perhaps they imagined a relationship between the woman’s grave and Palaimon’s grave when they saw the cist and the tunnel in such close proximity. Again, it seems unlikely that pagans would have situated the grave in this way for positive effect because burial polluted precincts. The overt juxtaposition would carry special significance, however, if it was created by Christians. It was not uncommon for early Christians to locate places of worship and graves next to or directly over preexisting funerary monuments of non-Christians.116 This functional continuity over time was in part a practical choice because the use of specific locales for burial was already familiar to the community. Christians also modified or appropriated earlier tombs to control potentially threatening or contested sites. Pagan graves could be a source of danger. The vita of Theodore, the Galatian saint of the 6th to early 7th centuries, told how the holy man drove out numerous “spirits” (δαίμονες) that village diggers had accidentally liberated from ancient tumuli, cists, and sarcophagi (George heg. V. s. Theod. Syc. 115, 116–117, 118, 161).117 Several centuries earlier, according to the epitome of his acta by Gregory of Tours, St. Andrew had expelled aggressive demons from roadside tombs outside Nicaea (Mir. And. apost. 6). If such spirits were subjugated by holy intervention or turned away by divine signs, the place could be safely and effectively used. For instance, when Corinthians opened neglected chamber tombs of Early Roman date for continued burial or habitation in late antiquity, they sometimes inscribed or painted crosses or prayers on the walls.118 Although these were typical funerary symbols for Christian Greeks, in the context of pagan tombs they also served as apotropaic signs. There is no such evidence that exorcism or consecration accompanied the interment in the Temple of Palaimon. Nonetheless, the striking contrast between the underground crypt and the superimposed grave may point to the manipulation of space to convey a similar message. Any such message would have acquired potency through burial and commemorative rituals.119 In conclusion, 56-001 reflects a shift not only in the site’s topography but also in its sanctity. If mourners chose the Palaimonion as a location for burial to exploit its prominent appearance, they were recasting the stripped concrete foundation as a massive grave marker, though the benefit would have been slight. If they were motivated by Christian belief and memory of the cultic past, they were actively replacing the polytheism that had defined the site for over a millenium with a strange new ideology. The grave in the Temple of Palaimon would thus furnish rare evidence for a pivotal moment in the history of Greek cult-sites, where the relationship between Christianity and paganism seldom involved direct confrontation. Late Antique graves at Corinth, such as those at the Asklepieion and the Sanctuary of Demeter and Kore, were sometimes located near or over earlier sites of worship. But in neither of those cases does the sacred architecture determine burial placement as exactly or blatantly as it does at the Palaimonion.120 Graves were also located in the 116. E.g., Pelekanides 1978; Gounaris 1990, pp. 55–57; Bakirtzis 1998, pp. 41–48 (Octagon at site of Hellenistic chamber tomb used for hero cult, Philippi); Frantz, Thompson, and Travlos 1969 (Roman heroön transformed into Christian church, Sikinos); Vaes 1984–1986, pp. 338–340; 1989, p. 303 (western examples of mausolea, heroa and columbaria reused as Christian churches or crypts); Effros 2003, pp. 99–100, 110– 114 (various instances of reused ruins, including ancient burials and temples, Early Medieval Gaul). 117. Harl 2001, p. 307; Mitchell 1993, pp. 139–144 (on Theodore’s activities as exorcist). 118. E.g., Charitonides 1968, p. 122 (τάφος E II, north of Cheliotomylos, Corinth). The author has recorded prayers, invocations, and crosses inscribed on the walls of Early Roman chamber tombs reused during Late Roman and Byzantine times on the Koutsongila ridge north of the harbor at Ken-

chreai. On painted or inscribed crosses at sites of burial in the Corinthia, see pp. 178–179, n. 88; on crosses inscribed on pagan statues or architecture, see nn. 37, 80, above (IA 72-75). 119. Consider Horden and Purcell 2000, p. 422: “The ‘hardware’ of locality and physical form, including temple, church or tomb, is in practice infused with changing structures of meaning by ritual and observance.” 120. Asklepieion: de Waele 1933, p. 436; 1935, pp. 356–357; Corinth XIV, pp. 161–167; Spieser 1976, pp. 312–313; Rothaus 2000, pp. 51–54. Sanctuary of Demeter and Kore: Corinth XVIII.3, pp. 381–391. Corinth XVIII.3, pp. 389–391, 400, makes the intriguing suggestion that the preponderance of burials of women and children over the ruins of the Sanctuary of Demeter and Kore indicates a concern to associate the dead with the ancient deities. If so, this relationship between old cult and late burial would carry a positive connotation.

134

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

foundations of defunct monuments at Corinth. Cists were cut into Temples D, F, H, and the Babbius monument on the west terrace of the Corinthian Forum after those buildings had fallen out of use and collapsed. Funerary artifacts and grave designs place two of them generally in the Late Roman or Early Byzantine period, while the other two are probably contemporary or later.121 None of these graves, however, was as conscientiously sealed or as purposefully situated as 56-001, and none contained a wooden structure or multiple items of personal adornment. Furthermore, the Corinthian burials occurred in an area of the city that was rapidly changing into a burial zone in late antiquity, whereas the Isthmian burial was a unique event in the central area of the former Sanctuary. Any symbolic value that 56-001 may have had for participants in the funerary ritual and for later visitors should not be overestimated. The activities of the hero cult had ended many years before; the temple had been stripped down to its foundations and the other buildings of the central area were dilapidated or dismantled. If the grave did connote the Christian subversion of polytheism, it was a proclamation of victory over a ruinous heap, safely removed from the regular operation of the Sanctuary, before a largely sympathetic audience. The grave at the Palaimonion was a last look back to a waning experience. Now local residents were interring their dead in places that acquired meaning from new alignments of settlement and architecture. RB 76-002 was a Christian grave of three individuals inside a ruptured drain in the ruined Bath (Fig. 2.66). Like 56-001, 76-002 utilized a major monument in the celebration of the Games for the deposition of corpses. Unlike 56-001, which had been conspicuously cut into the facade of the Temple of Palaimon, 76-002 was hidden under the floor in a simple channel that had once emptied foul water from a public bath. The original purpose of the channel did not identify the dead as deviants or marginal figures. Theirs was a peaceful, private burial, and the bones and artifacts reveal that they were similar in health and social status to others buried at the Late Roman settlement, such as the individuals in HO 70-902 nearby. The deposition of three bodies in the drain was neither ostentatious nor motivated by ideological concerns; it was a pragmatic choice. Interment in derelict waterworks was a common practice in Late Roman and Byzantine Greece for the simple reason that, like lined cists, they were sturdy constructions enclosing a long, rectilinear space that could accommodate a corpse in a supine and extended position.122 By the early 5th century, the deteriorating sacred buildings no longer dominated the landscape. Many stones and entire structures had been stripped down and reconstituted into the defensive walls, contributing positively to the site’s new appearance and purpose. Through despoiling and recycling residents successfully managed available resources, transforming the ineffectual, fragmentary buildings of the Sanctuary into something meaningful for their community. The alteration of the drain for burial was a qualitatively similar reinvention of the architectural legacy. Structural features like the drain in the Bath still held a part of their old form but mediated a novel range of interests and applications.123 121. A square cist with covering slabs like PAL 56-001 was cut into the concrete podium in the northeast corner of the cella of Temple D (Corinth I.3, pl. 5, plan A). A tile-lined cist containing two skeletons and a Late Roman lamp was cut into the foundations of Temple F (Rothaus 2000, p. 55, n. 71, p. 150, n. 39). A vaulted tomb probably dating to the Late Roman or Early Byzantine period was built over the southeast corner of the stairway of Temple H alongside an earlier water channel, but it is unclear whether the juxtaposition was intentional (Corinth I.3, pl. 16, plan A; Williams, MacIntosh, and Fisher 1974, p. 10, n. 13). A “Byzantine” cist sealed with mortar was cut into the foundations at the southeast corner of the podium of the Babbius monument, which was a tholos like the Temple of Palaimon (Corinth I.3, p. 17, pl. 8.2, plan A). Early

excavations at Temple E, located to the west of these structures, revealed that 23 “Byzantine” cists had been cut deep into the concrete foundations (Corinth I.2, pp. 172, 182, pl. 18). 122. For examples, see p. 177, n. 79. 123. Kardulias (1995; 2005, pp. 102–105, table 7.3) interprets the dismemberment of the Isthmian Sanctuary as an adaptive behavior according to an evolutionary model of energy expenditure and resource management; Bowden (2001; 2003, pp. 105–159) applies a similar approach to the architectural history of early churches in Epirus Vetus. On the ideological and practical uses of spoliation in the Byzantine world, see Saradi 1996–1997, esp. pp. 396–405; Ward-Perkins 1999; Papalexandrou 2003. On recycled stones in burial design, see Chapter 4.

PHASE IV: THE EARLY BYZANTINE SETTLEMENT (LATE 6TH TO 7TH/8TH CENTURIES)

135

PHASE IV: THE EARLY BYZANTINE SETTLEMENT (LATE 6TH TO 7TH/8TH CENTURIES) The fourth phase in the burial chronology belongs to the agrarian settlement that succeeded the employment of the fortifications in the late 6th to 7th or 8th centuries. This phase includes NEG 69-007, 69-009, and 69-010 (Group III), T2 68-002, 68-006, T14 67-002, 67-004, 69-701, 69-991, 69-002, and 69-003, which together contained 12 individuals. The burials at the Northeast Gate and Tower 2 can be dated by stratigraphic associations and assemblages to the late 6th or 7th century. The burials at Tower 14 are dated on similar grounds to the 7th century but could extend into the 8th century. T13 54-001 is probably associated with these Early Byzantine graves. Finally, a feature resembling a tile-covered cist comparable in size to a child’s interment was discovered near Tower 14 but not excavated. If this was a grave, it probably belonged to the same burial ground and phase as the six neighboring interments. These burials made after the Justinianic renovation of the Isthmian fortifications belonged to a tumultuous period in Greek history. Eastern populations called Avars and Slavs that had settled north of the Danube during the 6th century launched a massive Balkan invasion by the end of the 570s and within roughly five years had crossed into the Peloponnese. They again overran the region in the first decades of the 7th century.124 These incursions resulted in the widespread, long-lasting settlement of foreign peoples in southern Greece. The sparsity and bias of the “dark age” evidence have obscured the group identity of the foreign settlers. Recent scholars have argued that the notion of early Slavic ethnicity was largely a Byzantine invention adopted by modern historians.125 In any event, the foreign settlers in Greece, whom this study will call “Slavs” for convenience, shared certain traits that distinguished them from the native population. Besides demographic and cultural changes in Greek society, the social, political, and economic structures that had defined urban and rural life during the Roman era were also transforming. A small body of textual and archaeological sources provides a basis for reconstructing society and settlement in Early Byzantine Greece, but interpretations of this poorly attested era have long been controversial.126 The agrarian settlement at the Isthmus provides evidence of singular importance concerning the response of local residents to foreign incursion, the vitality of the countryside during the Early Byzantine period, and the character of the rural population. The local community is represented by sparse traces of habitation and as many as 13 graves. The settlement existed more or less continuously from the 6th through the 7th centuries and possibly into the 8th, after which the site seems to have been abandoned until the 10th century. The fortifications did not halt the Slavic incursion in the mid-580s, which perhaps drove away some residents temporarily but did not end the occupation.127 Thereafter, a small group of Christian families of local descent who depended on a mixed subsistence of farming and husbandry inhabited the dilapidated fortifications, the collapsed shell of the Bath, and the ruins of the central area. The settlement was situated in the productive hinterland of Corinth, and it communicated with other settlements in the area. 124. For introductions to the Avars, Slavs, and their incursions into the Empire, see Pohl 1988, pp. 94–112; Whitby 1988, pp. 59–89, 138–191; Christophilopoulou 1993, pp. 310–317; Kazanski 1999, pp. 67–76. 125. Barford 2001, pp. 27–44; Curta 2001b. 126. The essays collected by Winkelmann and Brandes (1990) discuss the diverse sources for Early Byzantine history. Studies of the material record of Early Byzantine Greece, many addressing the Slavic question, have been numerous: e.g., Bon 1951, pp. 49–55; Pallas 1955; Nestor 1963; Popović 1975; 1978; 1980; Huxley 1977; Weithmann 1978, pp. 198–252; 1985; 1994, pp. 100–104; Yannopoulos 1980, pp. 353–360; Baratte 1984; Sodini 1993; Gregory 1996; Karayannopoulos 1996,

pp. 199–210; Avramea 1997, pp. 72–104; Štefanovičová 1997; Anagnostakis 2000; Moutzali 2000; Vida and Völling 2000, pp. 11–40; Dunn 2001; Kountoura-Galaki 2001; Curta 2005. For histories and gazetteers of the Early Byzantine Corinthia and Argolid, see Yannopoulos 1980; Kordosis 1981a, pp. 69– 87; Avramea and Kyrkou 1988; Avramea 1997, pp. 164–178. 127. The archaeological evidence disproves the unsubstantiated theory that Slavic invaders wiped out the settlement at the Isthmus: Hohlfelder 1977, p. 179 (“incontrovertible”); Popović 1980, p. 233 (“la fin brutale de l’habitat tardif”). See also Weithmann 1978, pp. 217, 227; 1979, p. 169; 1985, p. 103; 1994, p. 102.

136

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

Scattered traces of habitation reveal that the fortifications were occupied after the Justinianic rehabilitation, but there is no evidence for their structural maintenance or concerted employment. As during the previous era (phases II–III), the absence of artifacts of a military nature suggests that the Fortress at this time was not held by Imperial troops; local residents belonged to the same Corinthian population that had inhabited the site since Roman times. Several coins of Justin II (565–578) found around the enceinte and to its southwest show that occupation persisted after the renovations,128 and deposits inside the towers reveal occupation in the Fortress as late as the 580s.129 However, the fortifications fell into disuse soon after their renovation. The formidable taphros gradually filled with sediment and building debris during the second half of the 6th century as a result of the Justinianic project, natural erosion, or both. The walls of the enceinte and the Hexamilion slowly deteriorated as segments of ashlar masonry fell and the exposed rubble core disintegrated. On the eve of the initial Slavic advance, the site was still occupied, but the fortifications were in a neglected state. Historical sources describe foreign incursions during the late 6th to early 7th centuries followed by foreign settlement and Greek resistance until the early 9th century. The invaders cut a path of violent destruction southward in the 580s, seizing materials and animals, slaughtering residents, and desecrating churches. When they reached Corinth in the mid580s the chieftain stole the great ciborium.130 The most important but most debated source for the Slavic invasion and settlement of the Peloponnese is the Chronicon dictum Monembasiae, written by a Greek ecclesiastic between the 9th and early 11th centuries. The relevant portion of the narrative describes the southward advance of foreigners and the flight of urban residents and records the barbarian occupation of much of the Peloponnese for over two centuries, while the eastern parts remained under the control of a “governor” (στρατηγός) from Constantinople (pp. 12.86–18.144). Many historians accept the basic value of Chron. Mon. as a source for broad historical trends but agree that it offers a vague, incomplete account of the impact of invasion and that it paints too rigid a picture of dichotomous ethnic zones.131 After the initial incursions the foreigners again advanced into Greece during the early 7th century.132 These movements in ca. 610–626 led to the wide settlement of Slavic tribes in the Peloponnese,133 a gradual process that probably had begun already in the 580s. Archaeological evidence from Corinthian sites indicates disruption in the late 6th century that reasonably can be identified with the Slavic invasions attested in the textual sources. Coin hoards, burned horizons, and structural damage at Kenchreai and Corinth point to destruction and abandonment in the early to mid-580s.134 Excavations at Nemea have also 128. E.g., Northeast Gate (IC 934); Tower 2 (IC 68-19, 6820, 68-23, 68-26, 68-28, 68-34); Tower 15 (IC 68-13, 68-72, 6879); Tower 17 (IC 72-26); East Field (IC 70-24, 70-35, 71-109). 129. Isthmia V, pp. 96, 122. 130. John Eph. HE 3.6.25; cf. Barhebraeus (Gregory Abū ’l-Faraj) Chron. Syr. 89, 90 (I, pp. 83, 84). The account of desecration at Corinth, which was derived from a lost section of John’s report of the middle 580s, appears in the 12th-century chronicle by the Jacobite patriarch Michael I (10.21 [II, p. 362]). There is no reason prima facie to doubt the accuracy of this account or the integrity of the text, as, e.g., Karayannopoulos 1990; cf. Olajos 1985, p. 510. 131. The historicity of this document is corroborated by two notices that Monemvasia was founded by Laconian refugees in 582/3 (Chron. Byz. brev. 41.4a–b); see Schreiner 1970; 1976–1978; 1977, pp. 77–78. 132. Mir. s. Dem. 2.1.179; Isid. Sev. Chron. 144a (MGHAA 11.2, p. 479); Cont. Isid. Hisp. (MGHAA 11.2, p. 337.13–15); John Nik. Chron. 109.18; Thomas pres. Chron. [Libri calipharum]

a.g. 934 (I, p. 115); George Pis. Heraclias 2.75–76. 133. Medieval writers portrayed a land inundated by barbarians: Chrestomathia Straboniana 7.47 (GGM II, p. 574); Chron. Mon. p. 16.134–140; Hugeburc, V. s. Willibaldi, MGHSS 15.1, p. 93.12–13; Const. Porph. Them. 2.6.33–34; Nich. Gramm. Sent. 2.24 (PG 119.880). 134. Kenchreai: Hohlfelder 1970a; 1970b; 1973; 1974, p. 77; 1975, pp. 253–255; Kenchreai I, p. 67; Kenchreai III, pp. 4, 87, 106–107, n. 398; Avramea 1983, pp. 57–58, nos. 10, 11. Corinth: Broneer 1926, pp. 52–53, pl. II; Corinth XVI, pp. 8, 16, 27; Dengate 1981, pp. 153–175, pls. 37, 38; Avramea 1983, pp. 52, 55–56, nos. 1, 7. On Peloponnesian coin hoards during this period, see Athanassopoulou-Pennas 1981; Avramea 1983; 1997, pp. 72–80; Oikonomidou-Karamesini 1991; OikonomouLaniado 2003, p. 79. Such a wide pattern of strikingly close dates in coins often associated with destruction horizons (Whitby 1988, pp. 133–136, map 4) cannot be persuasively attributed to abandoned savings-deposits or intensive military presence, as Curta 1996; 2001b, pp. 169–181; 2005, pp. 120–121.

PHASE IV: THE EARLY BYZANTINE SETTLEMENT (LATE 6TH TO 7TH/8TH CENTURIES)

137

uncovered damaged and abandoned houses in the central area of the Sanctuary and a refuge site in the Stadium, all dated to the 580s.135 After the initial contact in the 580s, the second major phase of Slavic movements into southern Greece during the early 7th century did not leave as distinct a mark in the archaeological record.136 The remains at the Isthmian Fortress dating from the time of the Slavic invasions compare with evidence elsewhere in the region. A thick stratum of rubble debris mixed with pottery and nine coins was found just west of Tower 2.137 In addition, a hoard of 270 coins that had been buried in the mid-580s was discovered in the robbing trench for the northern foundations of the Temple of Poseidon.138 These discoveries suggest that sedentary habitation at the Isthmus was disturbed in the 580s, when residents fled before the coming invasion. The absence of evidence for burning suggests that the foreigners moved quickly past the Isthmus, probably even without a fight, heading to the wealthier, more populous communities of the Saronic coast and the Corinthian interior.139 No traces of disruption in the early 7th century have so far been identified at the site. The city of Corinth recovered from the invasions and continued to operate as a commercial, administrative, and ecclesiastical center from the 7th to 9th centuries, if in a reduced and less powerful capacity than before.140 Among the diverse sources for the history of Early Byzantine Corinth, none records that the Isthmus continued to play a role in the city’s political or religious affairs or in the region’s defense after the 6th century and the Slavic invasions. Nonetheless, the landbridge remained the prime route of passage for all civilian, military, and commercial travel between central Greece and the Peloponnese and from Gulf to Gulf. Expeditionary forces or even a provincial garrison might have employed the Isthmian fortifications, but such activities would have been brief, as there is no trace of sustained military occupation.141 It should be remembered that the exploration of the site over the past half century has not produced artifacts typically associated with garrisons (e.g., metal weaponry, fighting equipment, harness fittings, lead seals, large quantities of coins). There is no definite sign that inhabitants tried to maintain or to repair the deteriorating Hexamilion and the Fortress.142

135. Miller 1976, p. 202, pl. 44; 1979, pp. 99–103, pls. 36:c, 38:d, 39:a, 40:b; 1980, pp. 199–200, pl. 50:b; [Stella] Miller 1983, pp. 84–87, fig. 4, pls. 24:a–d, 25:a–g; Miller 1988, pp. 3–5, pls. 6:a–c, 7:a–c, 8:a, b; Nemea Guide, pp. 67–70, 72–73, 75, 104–106, 109, 110, 195, figs. 42–44; Nemea II, pp. 89, 132–134. Unlike the remains at Nemea, much of the reputed evidence for the Slavic invasion at Argos is chronologically and stratigraphically ambiguous. For the evidence, see Aupert 1980a; 1980b; 1987, pp. 599–603, fig. 20; 1988, pp. 710–712, fig. 17; 1989, p. 418; Yannopoulos 1980. For recent doubts concerning its interpretation, see Anagnostakis and Poulou-Papadimitriou 1997, pp. 271–273, 292–294; Vida and Völling 2000, pp. 20–22; Lambropoulou et al. 2001, p. 203; Slane and Sanders 2005, p. 294, n. 108. 136. The physical evidence for these invasions in the northeastern Peloponnese consists of coin finds: Avramea 1983, pp. 58–59, 77–80, nos. 12, 13; 1997, pp. 74–75. 137. Isthmia V, pp. 115–116; see p. 60, n. 76 on this stratum (lots 68-T20-016, -018, and -019). The coins seem to have formed a cluster that was later dispersed. 138. Broneer 1955, pp. 117, 136, pls. 46:a, 54:c; MacDowall 1965; Avramea 1983, pp. 56–57, no. 9. 139. The theories of Yannopoulos that a Byzantine garrison was wintering at the Isthmus when the invaders arrived

(1980, pp. 365–366) and that opposing forces engaged in a fight (1993, p. 392: “la bataille de l’Isthme”) have no archaeological or historical proof. 140. For surveys of published evidence concerning “dark age” Corinth, see RBK IV, 1990, cols. 787–793, s.v. Korinth (D. I. Pallas); Hattersley-Smith 1996, pp. 219–227; Lambropoulou et al. 2001, pp. 196–199; 200–202. Earlier scholars imagined that a disastrous Slavic invasion cast the city into severe decline or even total destruction and desertion: e.g., Finley 1932; Bon 1951, pp. 49–55; Corinth III.2, pp. 129–131, 189, 272; XI, pp. 3–4, n. 13; XII, p. 5; XVI, pp. 27–33. Sanders (2002, 2003) and Slane and Sanders (2005) reevaluate this historical model on the basis of recent discoveries and chronological refinements. 141. Constans II might have quartered at the Fortress during his visit to Corinth before embarking for Italy in 663 (Isthmia V, p. 87). A regional garrison would have been situated at Corinth, the seat of the governor whence the besieged Patrenses received military assistance at the beginning of the 9th century (Chron. Mon. p. 18.141–144; Const. Porph. Adm. Imp. 49.13–14). 142. Isthmia V, pp. 82–83, 93–94, proposes that the blocking walls in the Northeast and South Gates were erected during the Early Byzantine period.

138

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

The Isthmus was not, however, deserted after the Slavic incursions. A diminutive yet resilient community continued to inhabit the site. The occupation of the Fortress is represented by scattered pottery and walls within the enceinte, at the towers, and near the gates. Inside the Northeast Gate a square house with a pen, a dirt road, and numerous cooking pots were found. Presumably residents of this area were responsible for similar sherds scattered outside the Northeast Gate over the taphros, which was filled to a depth of nearly 2 m by the 7th century.143 They also buried their dead (Group III) in the north wall of the former defensive ditch. Discrete floor deposits, coins, and pottery of this period have also been found elsewhere around the Fortress.144 This evidence shows that the settlement at the Fortress was widespread. All locales of investigation have produced Early Byzantine material in small but consistent quantities. Some of this material belonged to a permanent, domestic settlement, such as the buildings and floors at the Northeast Gate, but some, particularly the pottery in the towers and on the surface, may belong to refuse dumps or construction fills from later phases. Much of the evidence for habitation is contained within the Fortress. The sparse artifacts outside the walls to the north and south were apparently dumped there by residents inside the enceinte. The deposits related to this occupation are reliably dated from the mid-6th until the late 7th or 8th centuries. The period from the mid-6th to mid-7th centuries is represented by amphoras and cooking pots from several locales, including the globular pot (12) that was broken near the grave of a small child (NEG 69–010) in the late 6th or very early 7th century. The second half of the 7th century is represented not only by a coin of Constans II (655/6) and possibly a bronze belt buckle145 in the same horizon at the Northeast Gate, but also by a coin of Constantine IV (675–685) at Tower 14, though it was found in a Late Byzantine deposit.146 The chronology of the local assemblage of “Slavic ware” is difficult to establish, because the total number of sherds is limited, and they are seldom associated with precisely datable material in sealed, primary contexts. The Isthmian variants can be confidently assigned to the 7th century, and at least some sherds, such as those in the same stratum as the coin of Constans II at the Northeast Gate, can be dated to the second half of the 7th century at the earliest. Comparanda for the incised decoration on certain sherds raises the possibility that residents continued to use this unique cooking ware during the 8th century.147 Thus, apart from a 143. Isthmia V, pp. 84–87, nn. 66, 67, 72, ill. 15, fig. 17, pls. 24:d–26:c. 144. Isthmia V, pp. 97–98, 102, 123–125; Gregory and Kardulias 1990, pp. 489, 493, 509–510, no. 4, table 2; Kardulias 2005, pp. 72, 79–80, 123, table 6.2, fig. 6.23; see also pp. 70, 100–101. 145. IM 3546 (Isthmia V, p. 85, n. 66, pl. 25:c). Florin Curta kindly informed the author that this buckle is related to the Boly-Želovce type dated chiefly to the second half of the 7th century, though a few instances have been dated earlier and the form might have persisted into the 8th century. Curta elsewhere (2005, pp. 122–123, n. 86) dates this example without discussion to the first half of the 7th century. Ibler (1992, pp. 138–140, 145–146, figs. 3, 4 [Boly-Želovce type]) and Varsik (1992, pp. 85–89, 93–94, 102–103, 105, tables 4, 5, 9 [Keszthely-Pécs type]) survey the type and cite several variants from Slovakia, Hungary, Romania, Croatia, Albania, and Greece (e.g., Corinth XII, nos. 2187–2189). None of these compares closely with the schematic, blocky example from the Isthmian Fortress; one analogy is an unperforated imitation from Constantinople (Csallány 1956, pp. 273–274, table VI:5, 5a). This form descended from Byzantine buckles of lyrate form in precious metals, like those found on Lesbos and Cyprus and in Italy (Vinski 1974, pp. 34–35, figs. 25.9, 26.1; Ibler

1992, p. 140, figs. 4.23. 4.24). Without closer comparanda, it is difficult to situate the Isthmian buckle precisely within the development of the type during the 7th century. 146. IC 929; see p. 70, n. 106. 147. Lambropoulou et al. 2001, p. 200. The most recent assessment of the cinerary urns in “Slavic ware” at Olympia (Vida and Völling 2000, pp. 52–59, 91–95, fig. 15) presents a typology in which decorated vessels begin in the second half of the 7th century and extend through the 8th century. One of the sherds found south of the Northeast Gate (IPR 67-51; Isthmia V, p. 86, n. 72, pl. 26:a) displays a decorative scheme of alternating waves and horizontal bands that conforms most closely with Gruppe IIIa at Olympia, dated to the 8th century (Vida and Völling 2000, pp. 19, 55–56, 120, 126, pls. 9:4, 18:1, 20, 23 [Gräber 7, 32]); Vroom (2003, p. 141) also refers to incised pottery from Isthmia as “characteristic for the late 7th and early 8th centuries.” However, the total Peloponnesian assemblage of “Slavic ware” displays great variability from site to site, which points to highly localized lines of development. Therefore, archaeologists and historians should hesitate to depend on simple formal similarities between artifacts from distant sites such as Isthmia and Olympia for the precise dating of specific pieces.

PHASE IV: THE EARLY BYZANTINE SETTLEMENT (LATE 6TH TO 7TH/8TH CENTURIES)

139

temporary disruption in the 580s, the occupation of the site seems to have been continuous for well over a century after the Justinianic renovations. Further evidence for permanent habitation at the Early Byzantine Isthmus has come to light at the Bath.148 Beaten earth floors, buildings of dry-set rubble and recycled stones, small structures for cooking and channeling water, and household deposits have been found overlying the dense stratum of collapsed brick, tile, and stone that buried the Bath in the late 6th century.149 In association with these remains were numerous coarse- and cookingware sherds, including both undecorated and decorated “Slavic ware,” animal bones, and rotary querns.150 The occupation is dated by the presence of amphoras and cooking pots of common 7th-century forms in the domestic deposits. The various examples of “Slavic ware,” which occurred in the same stratigraphic contexts as these diagnostic forms, can be dated to the 7th century or possibly later.151 The domestic architecture and pottery, the faunal remains, and the groundstone all point to sedentary occupation by a small number of individuals, perhaps an extended family or a group of families who processed grains and raised livestock in the vicinity. Other traces of settlement that can be assigned to this period have been found within the ancient temenos of Poseidon. Three rectangular houses made from rubble and spolia were erected on the south side of the precinct after the demolition of the monumental architecture and the massive accumulation of debris.152 They probably belonged to the same Early Byzantine settlement that occupied the Fortress and the Bath.153 Although remains in the East Field, the Bath, and the Theater indicate that these areas were used during the 5th century, the Late Roman habitation there seems to have been sparse and intermittent. In contrast, the three houses in the temenos represent permanent, self-sufficient, domestic occupation outside the fortifications. Excavation around the Temple of Poseidon has uncovered not only numerous fragments of “Slavic ware” dispersed over a wide area but also Early Byzantine buckles located within a few meters of the houses.154 These finds demonstrate that the area was inhabited during the 7th century. The archaeological record thus shows that, while the site remained occupied from the mid-6th to late 7th centuries or later, the composition and character of the local community was evolving. The relative paucity of the structural and ceramic evidence indicates that the number of residents during the late 6th to 7th centuries was much smaller than it had been during the early 5th to mid-6th centuries. Their homes and workplaces were no longer nucleated within the enceinte but covered the derelict fortifications and the ruins of the Sanctuary with little concern for their former use or significance. Within this broad pattern of dispersed occupation, settlement was focalized at the Fortress, the Bath, and the temenos. Furthermore, the spatial distribution of habitation seems to have shifted over time. Areas 148. Gregory 1993b; 1995, p. 286; Kardulias 1994. 149. Gregory 1995, pp. 286, 303. Another small building, perhaps a pen, that might have belonged to the Early Byzantine settlement at the Bath was found at the Theater (Gebhard 1973, p. 134). 150. The identifiable faunal remains were mostly ovicaprid limb and head elements (D. S. Reese, pers. comm.). 151. Vida and Völling (2000, p. 19, n. 123) note that the decoration on some of the sherds from the Bath (e.g., IPB 7231) compares with examples from Olympia dating to the 8th century (i.e., Gruppe IIIa). 152. Isthmia II, pp. 97–98, plan II, pls. 30:a, 32:b, 37:c. E. R. Gebhard has also noted (pers. comm.) the presence of structures postdating the demise of the Sanctuary at the northeast corner of the temenos of Poseidon near the courtyard of the

Northeast Cave (Isthmia II, pp. 33–37, pls. 15:a, 55). 153. Associated pottery indicates that the houses were constructed during the 6th century at the earliest (E. R. Gebhard, pers. comm.). The manner of construction, the installations for cooking, and the millstones parallel the remains of the 7thcentury occupation of the Bath. 154. Pottery: Gregory 1993b, p. 151, n. 17 (on the authority of J. W. Hayes); E. R. Gebhard (pers. comm.). Buckles: Isthmia VII, pp. 56, 60, no. 221 (IM 108), pl. 37; for comparanda dating to the early 7th century, see Corinth XII, nos. 2209, 2210; “Halieis V,” p. 320, no. 57, fig. 14; Avramea 1997, p. 91, pl. IV:c.8 (the Argolic islet of Plateia); Curta 2005, p. 123, n. 91. Another Early Byzantine buckle with a rectangular design was found over the central portion of the South Stoa (Isthmia VII, pp. 56, 60, no. 223 [IM 2150], pl. 37).

140

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

of the Fortress have produced traces of habitation in the late 6th century, but the ceramic assemblage from the Bath does not predate the 7th century. One plausible scenario is that the local settlement remained inside the enceinte around the time of the Slavic incursions but then expanded outside it to the west during the 7th century. The situation of the Early Byzantine graves also reflects changes in settlement and site use after the Late Roman period. Residents continued to bury their dead on the northern margin of the settlement in the general area of earlier graves, just outside the Northeast Gate in the northernmost slope of the taphros (69-007, 69-009, 69-010) and just inside the curtain near Tower 2 (68-002, 68-006). Both areas were unoccupied and littered when the cists were dug. Inhabitants were dumping their broken pottery in the defensive ditch, which had filled with structural collapse and downwashed sediment, and debris from the abandonment of the late 6th century remained at the surface in the northeast corner of the enceinte. Although these graves did not disturb previous interments at either place, mourners might not have known exactly where the graves of their forebears were situated because graves of the Roman era (phases I–III) were unmarked and already covered by soil. Indeed, structures, floors, and domestic deposits of the Early Byzantine period inside the Northeast Gate and in the northern rooms of the Bath were situated almost directly above graves of the 5th or early 6th century (NEG Groups I, II, and RB 76-002). Moreover, a new locale for interment was used for the first time during this period: the area directly west of the Fortress where the monumental architecture of the Sanctuary had once stood. Thus, except for T2 68-002 and 68-006, the graves of phase IV were all located outside the wall of the Fortress to its north or west, and in both areas the nearest contemporary habitation was inside the enceinte. These patterns suggest that the inhabitants of the Fortress separated mortuary space from residential space through extramural interment. In this regard, it is noteworthy that no graves were found in either the Bath or the central area, where graves had been placed in earlier centuries. Those who lived in the houses in both areas must have carried their dead some distance elsewhere for interment. Although no early church has yet been found at the Isthmus, the dispersal of known Early Byzantine graves around the Fortress suggests that the residents were not using a single churchyard cemetery during this period. Within the three burial areas of phase IV (NEG, T2, T14), the relative situation of the interments reflects more deliberate spatial patterning than in phases II–III (Figs. 2.1, 2.36, 2.45). Besides one grave (67-004) that was over 15 m north of the cluster of five graves near Tower 14, the graves in all three areas were spaced out evenly and generously with an average intervening distance of 2.7 m. They formed rows or lines; the graves near Tower 14 were arrayed side by side, though the southernmost graves (69-002, 69-701, 69-991) were not perfectly parallel, while the graves outside the Northeast Gate and those inside Tower 2 were arrayed head to foot. Such arrangements indicate that mourners were aware of the location of preexisting interments when, unlike their predecessors, they dug fresh cists in open areas without contiguous architecture. This was possible because several, if not all, graves were marked at the surface by means of either unadorned stones (T2 68-002, 68-006) or mortar and rubble structures (NEG 69-009, 69-010). Although no such features were discovered near the graves at Tower 14, stones that had originally marked these interments might have been removed during later leveling and building directly above the cists. The regular distribution, symmetrical alignment, and high visibility of the graves show that the community conscientiously planned and demarcated them for ordered burial and easy recognition. These developments speak to the growing importance of graveside rituals, especially mourning, which could be easily conducted without disturbing or stepping on adjacent interments. Since none of the Early Byzantine graves was repeatedly opened for the addition of corpses, as during the Late Roman period, the markers served to identify cists at

PHASE IV: THE EARLY BYZANTINE SETTLEMENT (LATE 6TH TO 7TH/8TH CENTURIES)

141

the time of the burial and during later visitations.155 The clustering of graves in these three areas may represent familial groupings, though there is no sign that collections of graves were formally delimited. This proposal gains strong support from the fact that two skeletons buried at Tower 14, an old male (67-002A) and a younger adult female (69-002A), were the only ones from the site to show a congenital anomaly of the sacrum (Figs. 7.15, 7.23). The subsistence and organization of the people who used these graves is attested by the remains of their daily existence. Residents tended livestock, processed grain, perhaps raised bees,156 and produced their own vessels in “Slavic ware” by hand or with a slow wheel, all on a domestic scale. The rural location and character of the settlement at the Isthmus suggest that the community depended on the local cultivation of gardens and fields, as well as hunting and gathering. There is no sign of surplus production or specialization; rather, residents seem to have practiced mixed farming chiefly for household consumption.157 While the remains in separate locales point to self-sufficiency, the proximity and essential similarity between these areas suggest that they formed a single community. The scattered inhabitants provided for their own families, but neighbors must have shared resources, labor, and produce. They were poor in material possessions and monetary wealth: the design of their houses is simple and practical, and associated strata have produced almost no coins, precious metals, or objects displaying complex decoration or skilled execution. This social and economic profile points to a small community of farmers at the Isthmus, not unlike others in the northeastern Peloponnese during the late 6th to 7th centuries, such as those at Pyrgouthi and Halieis.158 It will be seen (Chaps. 5–7) that they were short, sturdy people, not unlike Greeks today, whose occupation and diet had an impact on their skeletons. Their settlement seems to represent an intermediate entity in the evolution of the Greek countryside after antiquity: smaller and less coherent than a Byzantine “village” (χωρίον), inasmuch as that class of settlement is understood,159 but somewhat larger and more stable than a mere agglomeration of huts or an encampment of squatters or refugees. Notwithstanding its local subsistence, small scale, and rural setting, the diverse artifacts reveal that the Early Byzantine settlement at the Isthmus was not isolated from external contact. Residents acquired amphoras and cooking pots from workshops in the northeastern Peloponnese, millstones from the island quarries,160 and the lyrate buckle perhaps from the northern Balkan sphere. The centrally located settlement would have communicated not only with travelers across the Isthmus but also with other rural sites and coastal establishments, such as the port at Kenchreai and the villa at Akra Sophia, which in turn received maritime traffic from the Saronic islets and Aegina. The rural inhabitants of the Isthmus

155. The only multiple interment of phase IV (NEG 69007) contained two bodies, but it is uncertain whether they were buried simultaneously or on separate occasions. 156. The fragments of beehives found by surface survey inside the enceinte could belong to the Late Roman phase, the Early Byzantine phase, or both; see n. 86, above. The coastal villa of Early Byzantine date at Akra Sophia produced identical sherds (Gregory 1985, pp. 422, 428, nos. 37–39, pl. 111). 157. The settlement conforms broadly to the agricultural model for the Early Byzantine countryside outlined by Kaplan (1992, pp. 55–87, 127–133), Lefort (2002, pp. 248–267), and Laiou (2005, p. 45). See also pp. 294–295 for a discussion of the dietary regime of local residents. 158. Hjohlman 2005 (farmstead at Pyrgouthi in the Berbati valley); “Halieis V” (settlement of “serfs” associated with large farm or estate; see also Laiou 2005, p. 34, n. 21). Lefort (2002, pp. 243–248) outlines essential features of small Byzan-

tine farms. 159. Kaplan (1992, esp. pp. 111–134), Lefort (1993), and Lefort, Morrisson, and Sodini (2005) discuss the composition of the Byzantine village. Kopetra in south central Cyprus is an example of a 7th-century village larger than the contemporary settlement at the Isthmus (Kalavasos-Kopetra, esp. pp. 235–262; Rautman 2004, pp. 204–211; 2005, pp. 456–458). Steinhauer (2001, p. 144) reports the discovery of a “small settlement of farmers” postdating the mid-6th century located east of Spata in Attica, at the site of the Athens International Airport, but he provides few details. Better coverage is offered by the exhibit that opened in 2003 at the Airport. This documents a grape press and amalgamation of buildings (mid-6th century) at one site and a small farmstead (7th century) at another nearby to the west. 160. Kardulias 1994; 2005, pp. 122–123.

142

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

would have also communicated with Corinth. While the nature of this intercourse remains unclear, the infrequency of coins of this period at both the Isthmus and Corinth suggests that transactions involved trade or barter rather than monetary sale.161 The community at the Isthmus could offer its urban neighbors grain, meat, probably garden produce, honey, and labor, perhaps gaining in return pottery, implements, metalwork, fabric, or other domestic wares. The physical remains of the Early Byzantine agrarian settlement, including the graves, indicate that residents belonged to the same population that had inhabited the site in earlier centuries and that they were Christians. The continuous Late Roman to Early Byzantine ceramic series, the traditional design of the rotary querns, and the form of the houses, which are reminiscent of domestic architecture elsewhere in the region,162 reflect the long-term occupation of the site by Corinthians. A similar stability of local form and practice is evident in the Early Byzantine graves near the Fortress. The basic design of the covered cists is the same as the one used by the Late Roman occupants of the fortifications and resembles unadorned graves at other sites in the northeastern Peloponnese. Moreover, the classes of personal articles interred with the deceased during this period are all found in earlier periods as well.163 This artifactual, architectural, and mortuary continuity is echoed in biological continuity: one congenital anomaly that distinguishes the local population, sacralization of the coccyx, occurs in skeletons from both Late Roman and Early Byzantine graves (Table 7.1). The Christianity of the deceased was clearly expressed during the 7th century, when one infant was interred below a grave marker inscribed with a cross (NEG 69-010A, Figs. 2.30–2.33) and another was buried wearing a pendant cross (T14 69-991A, 20, Figs. 2.102, 2.103). The archaeological evidence thus supports the identification of the residents as Christian Corinthian farmers, certainly not professional soldiers or foreign settlers. The distribution and nature of the houses and associated domestic remains point to a modest, sedentary, but disparate and peaceful occupation, hardly an ordered garrison of provincial troops. Furthermore, the buildings are unlike Slavic houses in the northern Balkans, which are rectangular, semisubterranean structures made from stacked timbers.164 The local “Slavic ware” in a variable reddish, coarse fabric is exemplified by a flat-bottomed beaker with a rounded body, a slightly flaring rim, sometimes a vertical handle, and occasional incised decoration.165 These vessels belong to a larger class of vessels found at numerous sites in both northern and southern Greece often identified with barbarian invaders.166 But the unique assemblage at the Isthmus does not correspond closely with examples at other sites,

161. On Early Byzantine coinage at Corinth, see Sanders 2002, p. 649, table 1; 2003, p. 387, fig. 23.1, table 23.1. Callegher (2005) surveys numismatic circulation at Late Roman to Early Byzantine Patras and rural sites in Achaea, which reveals a pattern not unlike that in the Corinthia. On the monetary economy during this period generally, see, among others, Grierson 1986, pp. 44–53; Morrisson 1986, pp. 155–163; 2001; Yerolymatou 2001, pp. 347–349, 361–362. Laiou (2002, pp. 706–710) discusses the evidence for trading and “peasant” transactions of a small, independent nature in the Early Byzantine period. 162. Cf. Corinth XVII, pp. 12–13, 56, pls. 4:d, 41 (6th-century house in north half of colonnade, Great Bath on Lechaion Road); Pallas 1964, pp. 137–138, pl. 119:α (Early Byzantine complex near basilica, Lechaion); Miller 1983, p. 84, fig. 4, pl. 24:d, e; Nemea Guide, pp. 70, 104–106, figs. 43, 44 (5th- to

6th-century complex near basilica, Nemea); Kosmetatou 1996, pp. 117–121, figs. 4, 6; 1998, pp. 169–171, pls. 14, 15, 35, 55–57, 124a, 125d–f, 127a; 2007, pp. 190–191, figs. 11, 31–34, plate A (5th- to 6th-century buildings on lower terraces, Midea). 163. See pp. 172–175 on covered cists in the region and pp. 187–199 on objects buried with the dead. 164. Popović 1981; Gojda 1991, pp. 18–25; Kazanski 1999, pp. 83–135; Curta 2001b, pp. 277–285, figs. 60–62, tables 9, 10. 165. Gregory 1993b, pp. 151–153, 155; Isthmia V, pp. 85–86. 166. On “Slavic ware” generally, see Anagnostakis and Poulou-Papadimitriou 1997, pp. 250–291; Vida and Völling 2000, pp. 13–26; Vroom 2003, pp. 141–143. Examples of this broad ceramic class appear at sites during the Early Byzantine period, often without obvious Late Roman precedents but with demonstrable similarities to the indigenous wares of central and eastern Europe.

PHASE IV: THE EARLY BYZANTINE SETTLEMENT (LATE 6TH TO 7TH/8TH CENTURIES)

143

combining forms and decoration that appear in both Late Roman cooking ware and “Slavic ware” found elsewhere. The development of this unique local pottery can be attributed not only to the introduction of exogenous ceramic styles during the Slavic invasion and settlement but also to creative experimentation by Corinthian potters embracing self-sufficiency in a shifting rural economy.167 Finally, the graves at the Isthmus are not at all like four contemporary graves containing military equipment, weapons, and “Slavic” pots discovered on Acrocorinth and in the Corinthian Forum, which have been variously identified with Bulgarian, Slavic, or Avar soldiers and fashions.168 Needless to say, the interments at the Isthmus have nothing in common with the only graves convincingly associated with Slavs settled in southern Greece, a cemetery of cremation burials in “Slavic” beakers found at Olympia.169 The inhabitants of the agrarian settlement at the Isthmus during the Early Byzantine period essentially resembled their predecessors in the Fortress in terms of social, economic, and religious identity, even as the conditions of their daily life had changed with the larger world around them. It is uncertain exactly when and why this settlement at the Isthmus came to an end, but excavated deposits across the site show a lacuna between the Early and Middle Byzantine periods.170 While travelers must have continued to cross the Isthmus along the ancient road to Corinth, the community inhabiting the Fortress and its surroundings seems to have either dissolved or moved elsewhere. The disastrous epidemic of plague in 746–747 and repeated raids from the sea by Arabs had a significant impact on Peloponnesian communities during the 8th to 9th centuries, but whether these events led to the abandonment of the Corinthian countryside is uncertain.171 Other internal factors might have contributed to the demise of the Isthmian settlement. Perhaps the local population grew too small and secluded to replace itself, so that the community gradually disappeared as its families died. Residents might have left to find a site with more plentiful water, pasturage, and soil, or a place closer to the coast or the city where exchange was easier. With the emergence of the Byzantine village, perhaps diminutive clusters of farming families in rural areas like the Isthmus disbanded and merged into larger communities offering a greater range of social and economic benefits. Another possibility is that the community at the Isthmus did not end at all but contracted or shifted to another locale nearby that has so far eluded archaeological exploration. 167. See also Gregory 1993b, p. 159; Anagnostakis 1997, p. 316; Anagnostakis and Poulou-Papadimitriou 1997, pp. 280– 281, 283, 312; Avramea 1997, p. 86; Vida and Völling 2000, p. 19; Whitby 2000b, pp. 729–730. It is unknown whether Corinthians and foreign settlers were in direct contact. Slavic settlement in the northwest Corinthia, around modern Akrata and Zemeno, has been proposed on the basis of toponymy (Kordosis 1981a, pp. 133–134, 343–344; 1981b, p. 387, n. 2; Dorbarakis 1986). It is unclear why Rosser (2005, p. 279) assumes that the residents at the Isthmus were Slavs. 168. Graves: Corinth III.2, p. 68, fig. 49; Davidson 1937; Weinberg 1939, p. 592, fig. 1; Corinth XII, pp. 6, 271, 199, nos. 1455, 1468, 1533, 1537–1539, 1547–1549, 1555, 1561, 1568, 1864, 1909, 1934, 2174, 2177, 2181–2183, 2191, 2195; Davidson Weinberg 1974; Ivison 1996, pp. 114–115, 117, figs. 5.6, 5.7. Interpretations of the graves: Davidson 1937, pp. 232–238; T. Horváth in Davidson 1937, pp. 239–240; Setton 1950, pp. 520– 525; 1952, pp. 352–353; Corinth XII, pp. 5–6, n. 8, pp. 267– 269; Davidson Weinberg 1974, pp. 512–513, n. 3; Ivison 1996, pp. 118–119; Vida and Völling 2000, pp. 32–39; Morrisson and Sodini 2002, p. 190; Curta 2005, pp. 130–131.

169. Yalouris 1962; 1963, p. 107, pl. 117; 1966, pp. 174, 176, 1967, p. 209; 1968, p. 170, pl. 182:a, b; Erdélyi 1991; Vryonis 1992; Vida and Völling 2000, pp. 41–127; Völling 2001, pp. 310–315, figs. 5–8. 170. Occupation of the Northeast Gate continued until the late 7th or 8th century, after which point datable finds all but disappear for roughly two hundred years (Isthmia V, p. 87, n. 76, p. 145). L. Houghtalin has noted (pers. comm.) that no coin of the emperors between Constantine IV (668–685) and Leo VI (886–912) has been found by the University of Chicago or the UCLA/OSU Excavations at Isthmia. 171. Turner (1990, pp. 422–426) and Rochow (1991, pp. 160–161) discuss the plague. Savvides (1990, pp. 48–53) provides a useful short survey of Arab activities; Kordosis (1981a, pp. 88–89) and Rife (2008, pp. 289–290) address the Corinthia. The immediate threat to the Isthmus is demonstrated by attacks on Aegina in the early 9th century (Christides 1981, pp. 87–89) and by the naval expedition under Nicetas Oöryphas some time between 872/3 and 882/3 in the Corinthian Gulf (Savvides 1998, pp. 87–92).

144

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

PHASE V: THE MIDDLE TO LATE BYZANTINE FORTRESS (10TH TO 15TH CENTURIES) The fifth and last phase in the burial sequence is difficult to define chronologically, historically, and topographically because it is represented by only two graves, LOU 69-801 and TC 60-001. These interments most likely belong to two separate eras in the development of the Byzantine and post-Byzantine settlement. The burial in the Loukos Field can be dated generally to the 12th to 14th centuries on the basis of residual sherds and its association with habitation south of the Fortress. The burial in the Theater Cave apparently is the latest so far discovered at the site. Although it lacks any precise chronological index, the exceptional preservation of the bones and their situation high in the fill inside the cave support a general date of 15th to 19th centuries. The isolated location of both burials and the unusual treatment of the bodies indicate that the two interments were atypical. This raises the important questions of where and how local residents usually buried their dead during this era. After the demise or relocation of the Early Byzantine settlement in the late 7th or 8th century, activity at the site revived in the 10th century, when the Isthmian fortifications played a defensive role in response to Bulgarian invasions.172 Coin finds reveal a marked increase in local monetary circulation during the late 10th to 11th centuries. Coincidentally, the first known usage of the place-name “Hexamilion” appears in the 10th century.173 Beginning in the 11th or 12th century and continuing through the 15th century, a substantial settlement developed at the Fortress. For Corinth, its ports, and its countryside, the final Byzantine centuries and the Frankish and Venetian occupations were a time of economic recovery and prosperity, when local markets were booming and overseas commerce through its harbors was thriving.174 Movements into southern Greece by the Ottoman Turks between the late 14th and late 15th centuries led to strategic responses at the Isthmus by Byzantine and Venetian forces, including extensive renovations to the fortifications in 1415 under Manuel II. Local settlement in the Middle to Late Byzantine era was concentrated inside the Fortress but extended a short distance beyond its walls. Excavation and survey inside the enceinte and around the towers and gates have uncovered abundant remains of habitation, including floors, sherds, coins, and animal bones.175 Local residents built small houses in tight clusters against both faces of the south and west walls of the enceinte.176 Beyond these buildings south of the Fortress was an open area where inhabitants dumped refuse in bothroi and erected smaller accessory buildings (Figs. 1.3, 2.45, 2.58). The testimony of early travelers and the results of modern exploration reveal that the settlement inside the enceinte was also dense. Sir George Wheler and Jacob Spon in 1676, Richard Chandler in 1765, Paul Monceaux in 1883, Eustathios Staïs in 1903, and R. J. H. Jenkins and A. H. S. Megaw in 1932–1933 observed numerous walls and several churches, many of which most likely belonged to the final occupation of the Fortress.177 Monceaux misidentified two apsidal struc172. Isthmia V, p. 146. 173. Coins of of Leo VI (886–912) and Constantine VII (945–ca. 950) and Anonymous folles classes A1 and A2 have been found in the central area and the Fortress. Kordosis (1985–1986) notes that the Arab geographer ibn Ḥ awqal (Abū al-Qāsim ibn cAlī al-Nasīb) referred to the Isthmus as “Kasmīli” on the mappa mundi he compiled in the mid-10th century (I, p. 189). Rife (2008) discusses a possible fortification program involving fire beacons across the region initiated by Leo VI. 174. On the Corinthia in general, see Kordosis 1981a, pp. 91–129, 220–224; RBK IV, 1990, cols. 795–809, s.v. Korinth (D. I. Pallas); on Corinth, see Sanders 2002, pp. 650–654; 2003, pp. 396–397. Isthmia V, pp. 146–151, discusses this period in the history of the Isthmus. 175. Isthmia V, pp. 87–89, 94, 100, 102, 115, 125, 127, 120,

121–123, 127, 146; Kardulias 2005, pp. 79–81, figs. 6.22, 6.24. 176. Jenkins and Megaw 1931–1932, p. 73; Broneer 1955, p. 124, pl. 48:c; 1959, pp. 320–321; Isthmia V, pp. 94, 104–105, 107, 123, pls. 36:a, b, 37:b, 44:b. 177. Spon 1678, p. 294; Wheler 1682, p. 437; Chandler 1817, p. 274; Monceaux 1884, p. 356, pl. 38; Staïs 1906, pp. 15–16; Jenkins and Megaw 1931–1932, pp. 79–82, fig. 7; cf. Corinth I, p. 61, figs. 26, 66, 68. William George Clark (1858, p. 48) also noted “the remains of some houses, and a small church apparently of great antiquity” outside the west wall of the Fortress. Several walls recorded by geophysical and surface surveys in 1985–1986 may well belong to the Middle to Late Byzantine settlement rather than a Late Roman garrison, as Kardulias has argued (2005, pp. 88–91, table 6.3, figs. 6.36, 6.37).

PHASE V: THE MIDDLE TO LATE BYZANTINE FORTRESS (10TH TO 15TH CENTURIES)

145

tures in the southern area as “chapelles byzantines,” but he did record a church at the site of Ayios Ioannis Prodromos.178 It seems that the immediate predecessor of the modern church was erected no later than the mid-19th century, and it might well have been built during Byzantine times.179 No significant evidence for permanent habitation during this period was found in the temenos or the Bath.180 These remains reveal the character of the community at the Fortress during the Middle to Late Byzantine era. The distribution and quantity of artifacts and structures show that this settlement was more populous and dense than the Early Byzantine one but comparable to or smaller than the Late Roman one. The modest houses show that residents occupied a lower status in the regional social structure than contemporary urban elites. Nonetheless, the community enjoyed a degree of prosperity, engaging in both localized subsistence activities and exchange for imported ceramic wares.181 It is uncertain whether the inhabitants were soldiers who defended the Isthmus or rural residents who hosted occupying troops. Without a more systematic study of these remains in their depositional context it is difficult to trace how this settlement changed during the historical developments of the last Byzantine centuries. It is remarkable that excavation at the Isthmus over several decades has not uncovered more than a single grave of phase V. This absence proves that residents did not inter their dead in the customary burial areas of the Roman and Early Byzantine periods, that is, the northern margin of the site (phases I–IV) and the area directly west of the Fortress (phase IV). The discontinuity in the placement of burial presumably coincided with the decline of the ancient settlement. Indeed, local residents unknowingly founded their houses directly on top of earlier cists near Tower 14. The contrasting use of this area for burial in the Early Byzantine period and for domestic habitation in the Late Byzantine period resembles the situation in the Corinthian Forum between the 7th century and the late 9th to 13th centuries.182 The graves of the Isthmian community must have been located somewhere still unexcavated but not far from the Fortress. Perhaps there was an extramural cemetery west of the ancient temenos under the modern village, which seems unlikely because of its distance from the Fortress, or directly south or east of its walls. Another distinct possibility is that a churchyard cemetery existed inside the enceinte, perhaps at Ayios Ioannis, if the church Monceaux viewed was in fact “une chapelle byzantine.”183 The grave in the Loukos Field (69-801) did not belong to a larger cemetery. The deceased individual must have lived nearby, but it is uncertain whether he inhabited the nearby buildings along the south wall of the Fortress. Apart from its isolation, the unprotected and informal design of the grave and the atypical treatment of the corpse, with the legs folded back over the torso, marked both the grave and the man buried therein as anomalous. The circular, shallow cist mirrors the numerous refuse pits that littered the open area south of the houses abutting the Fortress. Several bothroi contained burned and chopped animal bones, and one contained three articulated bovine limbs.184 The proximity of this cist, in which the human body was arranged in an unnatural position, to the pits in which the butchered bones of livestock were dumped must have been striking for local residents.185 178. Monceaux 1884, p. 356, pl. 38; Corinth I, pp. 62, 66, fig. 26. Clark (1858, p. 49, pl. 2) had earlier noted a “little church” at the site of Ayios Ioannis, and Bursian (1872, p. 21) reported “eine kleine verfallene Kapelle” in the Fortress, presumably also Ayios Ioannis. On the misidentification, see Isthmia V, p. 2 n. 8, p. 130. 179. Cf. Gregory 2007, pp. 176, 191, table 9.6. 180. Small quantities of Middle to Late Byzantine pottery and Venetian colonial coins have been found along the Hexamilion and around the Roman Bath, presumably representing the latest military employments of the fortifications (Gregory 1993c, p. 305; 2007, p. 176).

181. Preliminary study of the faunal remains has identified sheep-goat, pig, chicken, and horse, some butchered (D. S. Reese, pers. comm.). Gregory (1989; 1993c) discusses pottery, including numerous Italian imports of the 14th to 15th centuries. 182. Ivison 1996, pp. 120–121. 183. Laiou (2005, p. 48) stresses the centrality of churches in the life of Byzantine villages, and Gerstel (2005) offers a probing study of the spatial and spiritual importance of churches in Late Byzantine villages. 184. Clement 1972, p. 163, pl. 62:a; Isthmia V, p. 123. 185. For a full discussion of the significance of this unusual burial, see pp. 230–232.

146

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

It is ironic that the most recent burial in the local chronology is also the most enigmatic. The last known grave before the modern cemetery is the secondary deposit of human remains found in the Theater Cave (60-001). The bones were most likely placed there during or after the Late Byzantine period, but the exact date of their interment and the nature of the community with which they are to be identified are uncertain. After the late 15th or 16th century, the settlement at the Isthmian Fortress again declined and either moved or disappeared. Presumably this burial belonged to a very small community or even a single household in the area during the Late Byzantine or Ottoman period or even somewhat later, but well before the 20th century, when villagers routinely buried at Ayios Ioannis Prodromos. No other human remains in a primary or secondary burial were found at the Theater or northeast of the ancient temenos of Poseidon. This demonstrates that, at least in its final form, 60-001 did not belong to a cemetery. The body had originally been interred elsewhere, probably outside the cave and some distance away. Sometime later it was exhumed, but whether purposefully or accidentally is unknown. Then several major elements of the skeleton were collected and deposited in a vaguely anatomical arrangement, together with scattered animal bones and rubble, along the cave’s wall. The person(s) responsible either tunneled in through the ancient entrance or descended through the hole in the ceiling. The choice of a cave as the site for interment is noteworthy. It not only hid the bones from view and shielded them from the weather, but it also resembled in its basic form a traditional rock-cut burial chamber.

EPILOGUE: KYRAS VR YSI, AYIOS IOANNIS PRODROMOS, AND ITS CEMETER Y by Lita Tzortzopoulou-Gregory and Joseph L. Rife The village of Kyras Vrysi now spreads up to and beyond the main archaeological zone of the Isthmian Sanctuary, the fortifications, and the West Cemetery (Fig. 1.2).186 This community is worth brief consideration even though it postdates the Roman and Byzantine eras that are the primary focus of this study. The history of Kyras Vrysi represents an important phase in the long-term development of settlement at the Isthmus. The village church of Ayios Ioannis Prodromos probably succeeds a Byzantine church inside the Fortress. The burial forms and mortuary practices of contemporary Corinthians as represented by graves at Ayios Ioannis show basic connections to those of Late Roman to Early Byzantine Corinthians as represented by nearby graves at the Northeast Gate. While the residents of Kyras Vrysi are relative newcomers to the Isthmus, they have moved into an area that was once densely settled, they use a church where their predecessors worshipped, and they bury their dead in a manner that has evolved since antiquity. The village, its church, and its cemetery therefore constitute a meaningful framework for examining life and death in the Corinthian countryside before modernity, apart from being important subjects in their own right. Travelers’ accounts and official documents imply that the Isthmus was largely deserted between the Byzantine and Modern eras. Europeans who journeyed from Athens to Corinth 186. This section is based on continuing research by Lita Tzortzopoulou-Gregory concerning modern cemeteries, rural landscape, and Greek cultural identity. She worked in Kyras Vrysi with permission from the Ministry of Culture and the former Nomarchia Korinthias and with approval from the Ethics Committee of LaTrobe University. Her evidence from the

Corinthia was collected in conjunction with the Eastern Korinthia Archaeological Survey; see Diacopoulos 2004, pp. 184, 193–194; Tartaron et al. 2006, p. 461; Tzortzopoulou-Gregory 2008, esp. pp. 320, 321–322, table 17.1. On Modern Greek mortuary studies in general, see pp. 159–160, n. 4.

EPILOGUE: KYRAS VR YSI, AYIOS IOANNIS PRODROMOS, AND ITS CEMETER Y

147

during the 17th and 18th centuries portrayed a sparsely inhabited and underutilized countryside. They observed that the ruins of the Sanctuary and Fortress were unoccupied and that no significant settlement besides Examilia was located between the Saronic Gulf and Corinth.187 This is supported by a recent regional survey that uncovered little or no evidence for concentrated settlement on the Isthmus during the Early Modern period.188 Whoever was responsible for the unusual secondary burial of TC 60-001A must have lived in isolation or with few neighbors. The Grimani Census of 1700, which listed all existing settlements in the Morea, does not appear to record habitation at the Isthmus.189 From the census of the Ottoman province of Corinth in 1822, the Expédition scientifique de Morée recorded 40,000 residents for “Isthmia,”190 though it is unclear whether this toponym designated the larger agricultural region of the Isthmus or a specific settlement. The modern settlement of Kyras Vrysi was not founded until roughly the mid-19th century. It probably came into existence after the War of Independence and the establishment of the Greek State in 1830. The new government offered incentives, including land allocation, to encourage people to move from the mountains down to plains and coasts that had grown inaccessible and insecure during the Ottoman dominion.191 Like many Corinthians today, most inhabitants of Kyras Vrysi are Αρβανίτες, Albanian-speakers, with ties of kinship to older villages of the Corinthian highlands such as Sophiko, Ayios Ioannis, Angelokastro, Ayionori, and Limnes. Their ancestors were nomadic peoples who settled in mountain villages during the 14th to 17th centuries.192 Kyras Vrysi has grown since its establishment, but it remains in many ways a small rural community. It has the appearance of a young settlement, lacking many of the features of old Greek villages: it has no plateia, no central church, no circuitous streets, and no traditional architecture. The core of the settlement is arrayed alongside the paved road that turns sharply off the highway to Epidauros and heads southwest to Examilia and Archaia Korinthos beyond. The few surviving early houses, which are modest, one-story structures typical of the 19th and early 20th centuries, front the main road.193 Since the archaeological plan showing modern buildings was drafted in 1978 (Fig. 1.2), grids have been instituted for rapid construction to the south up the slopes of the Rachi and to the north on the high ground flanking the Great Ravine. Many current residents, estimated at one to two thousand in number,194 work either in local agriculture, particularly vine and olive production, or in urban industry at Nea Korinthos, though more and more homes belong to seasonal occupants, many Athenians. Apart from the archaeological remains, the two features of Kyras Vrysi that predate its modern development are the road, which essentially follows the ancient and medieval route to Corinth, and the church. Although Ayios Ioannis now serves as the parish church, it is located in the northwestern corner of the enceinte, not in the center of the village (Figs. 1.2, 1.3). As has been discussed, several early travelers saw one or more churches inside the Fortress. In 1765 Chandler noted a particular “ruinous church” that contained spolia, including a column base reused for the altar. During the following century, Clark and 187. E.g., Spon 1678, p. 294; Wheler 1682, p. 437; Chandler 1817, pp. 274–275; Dodwell 1819, vol. 2, pp. 183–184; Leake [1830] 1968, pp. 286–294; Blouet 1838, p. 66; Clark 1858, pp. 49–53; Taylor 1859, pp. 29, 152, 153–154; Wyse 1865, pp. 332–333. 188. Gregory 2007, pp. 176–187. 189. Panayiotopoulos 1985, pp. 240–243, 291–292. 190. Bory de Saint-Vincent 1834. 191. Andreadis 1980, pp. 12–18; 1996, pp. 18–23; 2001, pp. 183–187. 192. Panayiotopoulos (1985, pp. 68–100) generally dis-

cusses Albanian immigration; Kordosis (1981a, pp. 137–138) specifically addresses the Corinthia. 193. Early archaeological plans and photographs show these houses: Jenkins and Megaw 1931–1932, pl. 26; Broneer 1953, pl. 55:a; Isthmia II, pls. 1, 2, 51. Site plans of the 19th century show no modern structures (e.g., Leake [1830] 1968, pl. 3; Clark 1858, pl. 2; Monceaux 1884, pl. 38; cf. Corinth I, p. 61 fig. 26), but they are very schematic and their purpose was to record antiquities. 194. The 1991 census recorded approximately 800 persons living in Kyras Vrysi.

148

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

Figure 3.1. Ayios Ioannis Prodromos, Kyras Vrysi in 2005, from southwest

Monceaux recorded the presence of a church at the site of Ayios Ioannis, which Monceaux called “une chapelle byzantine.”195 The church was probably founded in the Byzantine era, perhaps succeeding a still earlier church. The interior and exterior of the building have been altered over the past two decades, so that no traces of earlier construction are now visible (Fig. 3.1). However, its cross-inscribed format, which is unusual for a modern cemetery church, was a popular Byzantine design. Moreover, its situation in the Fortress suggests an association with an earlier phase of settlement. If so, settlers in the 19th century must have discovered this church within the Fortress on already consecrated ground and renovated it for their own purposes. The cemetery is located in a walled enclosure directly west and southwest of the church’s courtyard (Fig. 3.2). The burial precinct, which was greatly expanded over the winter of 1993–1994, now covers an area of ca. 2,000 m2, or approximately 7% of the interior of the Fortress (Fig. 1.3). Although the expansive cemetery directly overlies the Roman and Byzantine settlement, there is no policing mechanism to monitor or halt ongoing interment, and excavation for new burials has frequently disturbed the archaeological record. A committee including the village priest traditionally managed the cemetery, and parish members were granted the right to burial without charge. But since the introduction of the new administrative scheme under the Kapodistrias Plan in 2000, the cemetery has come under the jurisdiction of the Municipality of Loutraki-Perachora in the Prefecture of the Korinthia. Residents are now required to pay for space, unless they have access to an existing plot. This arrangement entitles non-Greeks and non-Orthodox Christians, who were previously excluded, to purchase land for burial within the precinct. As of 2002 the cemetery contained 240 graves marked at ground level. Although the earliest recorded epitaph dates to 1917, local residents claim that the cemetery has been active 195. See n. 178, above. Colonel Leake, who visited the Isthmus at the beginning of the 19th century, recorded a rectangular structure at the site of Ayios Ioannis as the “T. of Palaemon” ([1830] 1968, pp. 293–294, pl. 3). Although he called

the interior of the enceinte “a level pasture” (p. 287), he might have seen the ruins of a Byzantine church in its northeast corner and associated it with the Palaimonion.

EPILOGUE: KYRAS VR YSI, AYIOS IOANNIS PRODROMOS, AND ITS CEMETER Y

149

Figure 3.2. Cemetery of Ayios Ioannis in 2000, from northeast

since the 19th century, presumably since the first settlement of Kyras Vrysi. The absence of earlier grave markers is not surprising because permanent monuments and inscriptions are a relatively recent development in modern cemeteries. Before the early 20th century, many burials were not defined by monuments at the surface, and markers such as stones or wooden crosses were easily removed, displaced, or destroyed over time as graves were transplanted or bodies added. When graves with markers were made in old burial grounds, information about prior interments was often lost or ignored. Therefore, no mark of the earliest interments associated with Ayios Ioannis survives. Some may rest beneath more recent graves, while others may lie under the concrete pavement of the churchyard. Nearly 60% of all graves in the cemetery date to the last 30 years. This pattern can be attributed to the community’s growing population and wealth. The need to erect elaborate monuments may also reflect changing attitudes toward commemoration, genuine affection toward the deceased, or the unashamed exhibition of affluence. The form of the monuments, however costly, is nonetheless of poor quality, and many begin to collapse within a decade of installation, an ironic contradiction to the supposedly eternal preservation of ancestral memory. Recent study of the cemetery has documented its use, composition, spatial distribution, burial typology, and funerary assemblage.196 Almost all graves are familial plots designated epigraphically by the word “house” (οίκος) before the family name. Most of these graves are occupied by members of the nuclear family, usually husband and wife. When someone dies, the body is placed in a wooden casket fully dressed and wearing the wedding crown and ring if he or she was married. Sometimes coins are placed in the pockets or on the body. A new body can be added to the grave at least four years after the latest burial in a plot, so long as the remains of the previous occupant(s) are removed. The bones are washed in wine, dried, placed in a bag or box, and redeposited on top of the new casket containing the most recent 196. Tzortzopoulou-Gregory (2008, pp. 311–314) summarizes the methodology of study.

150

BURIAL CHRONOLOGY, TOPOGRAPHY, AND HISTOR Y OF SETTLEMENT

Figure 3.3. Standard tomb type, Ayios Ioannis

Figure 3.4. Tomb at ground level with concrete

border, Ayios Ioannis

dead. In some cemeteries, ossuaries or bone receptacles are employed to store the bones of exhumed individuals, because either the graves become overcrowded or neglected graves are taken over by different families who remove earlier remains to make room for new ones. Kyras Vrysi does not have an ossuary today, but it might have had one in the past. Many communities across Greece are closing and removing ossuaries because they regard them as unattractive and useless. The graves at Ayios Ioannis are arranged in orderly rows separated by narrow pathways aligned roughly north–south. The oldest graves, which are concentrated in groups bearing the same family name, are closest to the church, while those further away show less familial grouping. Most graves are oriented east–west and measure ca. 1 by 2 m, though a few are larger, especially double interments. The cemetery has four types of graves. The commonest has a rectangular monument in marble or concrete placed over the interment (H. ca. 0.50–1.00 m) and an upright stone or compartment erected at the head, or west end (Fig. 3.3). The second commonest type has a brick, stone, or concrete border encircling the interment and often a slab placed above it at ground level (Fig. 3.4). The other two types, which are represented by only four examples, are memorials for single persons and one mausoleum (Fig. 3.5). The graves are variously adorned with oil lamps, crosses, wreaths of flowers, decorative sculpture, and small gardens. The objects most frequently left on the grave or inside the compartment are photographs of the deceased, icons, lanterns, incense burners, bottles of oil, personal possessions or mementoes, and food and drink. The epitaphs not only identify the name and age of the deceased but also employ formulaic language, including religious and nonreligious themes, biblical or classical quotations, and verse or prose to express the anguish of the bereaved and their wishes for the departed. The community of Kyras Vrysi maintains a strong connection with its dead through commemoration. This involves daily visits to the family’s grave, usually just before sunset, to light a lantern; the washing of the monument and cleaning of its area before major holidays

EPILOGUE: KYRAS VR YSI, AYIOS IOANNIS PRODROMOS, AND ITS CEMETER Y

151

Figure 3.5. Row of tombs, Ayios Ioannis

(Easter, Saints’ Days, Christmas); and the performance of “memorial services” (μνημόσυνα) on special days, such as the anniversary of the death and “Souls’ Days” (ψυχοσάββατα), when κόλλυβα, a mixture of boiled wheat, sugar, nuts, and pomegranate seeds or raisins, is prepared and deposited at the graveside. Apart from these conventional rites, mourners regularly leave both fresh and plastic flowers and food, especially fruit, such as oranges and pomegranates; they also renew photographs, and burn incense. While these activites can continue for many years, the maintenance of most burial sites at Ayios Ioannis seems to end within a single generation, so that much of the cemetery is in a state of neglect. This survey of the village, its church, and its cemetery illuminates the long history of habitation and burial at the Isthmus. The settlers of Kyras Vrysi were not descended from the residents of the Late Roman and Byzantine Fortress. They did, however, discover and revive the church already existing in the enceinte as the site for the community’s cemetery. The area thus returned to its prior use for burial, even though the pattern of reuse was not intentional. Moreover, current graves and funerary activities at Ayios Ioannis can be traced back to the forms and rituals of local residents during the Late Roman to Early Byzantine era. The salient feature is the design of the monuments overlying many recent interments (Figs. 3.2, 3.3). This seems to be a formalized version of the rectangular blocks of rubble and mortar faced with plaster that covered NEG 69-009 and 69-010, only about 110 m to the east-northeast, in the late 6th or very early 7th century. As will be discussed (Chap. 4), certain objects left at the graveside today, such as flowers, lamps or candles, and personal possessions, were also deposited in earlier times, while the specific treatments of graves by villagers, such as the exhumation and redeposition of old bones and the commemorative visits, were also observed by their predecessors. This persistence of earlier practices is vividly revealed in the close juxtaposition of graves from different eras. But the settlers of the village, who arrived as outsiders in the 19th century, were not maintaining customs through a direct local lineage. The reminiscences of the past that survive in their mortuary behaviors are seen throughout Greece today, no matter how the traditions of burial have been preserved or revived from place to place.

4

 FUNERARY RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

T

his chapter surveys the historical evidence for funerary ritual in Greek society during the Roman and Byzantine eras, examines the graves at the Isthmus and their regional comparanda as evidence for the ritual process, and relates mortuary variability to social structure and ideology in the changing community. The mortuary and osteological data pertaining to this discussion are summarized in Table 4.1. Despite their small number, the 30 graves offer a reliable picture of local funerary ritual because they represent the interests of over twice as many individuals, a significant sample. The graves at the Northeast Gate, Hexamilion Outworks, and Tower 14 are particularly informative for the study of death and burial in the rural Corinthia during the Late Roman and Early Byzantine periods. They compare in general with interments at Corinth but display closer affinities to those at peripheral sites in the northeastern Peloponnese. Residents of the Isthmus used less elaborate mortuary forms and behaviors than urban residents, which reflects a discrepancy in social complexity and economic prosperity between the two communities. Moreover, changes in grave design and corporeal treatment at both the Isthmus and Corinth express the adoption of a Christian paradigm of death. A comparative study of the mortuary remains at different sites can thus reveal the range of social, economic, and religious diversity across the region. Funerary ritual, like other forms of ritual behavior, is a way of conceptualizing the world through words, images, actions, objects, and spaces.1 It is constructed in a certain time and place through a series of interrelated decisions on the part of those responsible for the burial, which in Greek society was usually the bereaved family. These decisions were determined by personal needs and communal expectations. For both participants and spectators, funerary ritual activated several modes of experience: sensual, emotional, intellectual, religious, and socioeconomic. It was a customary behavior that was not necessarily understood and could be variously interpreted; it was a channel for both prescribed and impromptu expressions of feeling; it was a religious event that recapitulated a theology (a conception of human and divine), a cosmology (a structure of the world), and an eschatology (a vision of death and the afterlife); it was an affirmation or contestation of the network of relations within a community. Of course, all the material, behavioral, and visual dimensions of past funerary rituals in any one place and time can seldom be recovered archaeologically. However, critical evaluation of the historical record and ethnographic analogy can help generate a more complete picture. The archaeological record of the Isthmus during the Roman and Byzantine eras, as recorded in Chapters 2 and 3, attests to the concrete activities and physical surroundings of mortuary behavior. As for aspects that have left no material residue, contemporary written sources and later practices provide a basis for the plausible recreation of local events surrounding death and burial. 1. The recovery and interpretation of funerary ritual in the material record has been much discussed: e.g., Pader 1982,

pp. 36–44; Morris 1992, pp. 8–15; Härke 1997; Parker Pearson 1999, pp. 32–34.

II

I

Phase

A (F, 35+ years)

A (F, late 40s); B (M, 35–44 years)

A (–, 3–4 years); B (F, late 40s); C (M, 35+ years); D (F, 25–34 years); E (F, 35–44 years)

(subadult)

WF 62-001

NEG 69-103

NEG 69-004

NEG 69-008

A (–, 45+ years); B (–, 15+ years)

DEC 69-901

(subadult)

A (M, late 40s); B (F, 35+ years)

T2 68-003

DEC 69-902

Sex and Age of Individual(s)

Area and Grave No.

Unknown

243

9 small stones in irregular cluster on floor (paving?)

Bones of A, C–E in a pile; leg bones of A extended straight; B supine and extended, head turned to R, arms slightly under back, legs pulled L

217

4 slabs and smaller Lining slab(?) dislodged at lower stones; cover level slab(?) dislodged at lower level

Rubble, tile, and 8 large stones; large stone and slab dislodged at lower level

Both supine and extended, facing forward, arms over pelvis but not crossing

Lying on R side, arms bent back sharply, legs semiflexed to R 20°

226

272

N, E, and S walls lined with rubble

Unknown

Both supine and extended, facing forward, arms straight, hands turned inward at hips, legs pulled R

293

282

Both supine and extended; A has head turned on L, arms crossed over pelvis

Corporeal Position

333

None

2 upright slabs and 3 stones lining middle and E end

Cutting into bedrock; floor paved with 7–8 tiles

Lining and Paving Orient. Elements (°)

3 slabs and smaller None stones

4 intact tiles at an angle covering N side

Pitched tiles, end tiles, and capping tile

1 slab over W end

2 mortared slabs and 7 large stones

Covering Elements and Grave Markers

0.67– 0.70

1.75*

0.54

0.57

1.75*

1.40– 1.60

1.50*

1.65*

0.57

1.91

0.36– 0.44

0.98

1.97

0.93– 1.10

Width (m)

Length (m)

0.75

0.50*

0.63

0.40*

0.35*

0.32

1.00

Depth (m)

0.57– 0.65 [0.61]

0.59– 0.61* [0.60*]

0.63*

0.99*

0.12– 0.17* [0.14*]

0.35

1.93

Volume (m3)

TABLE 4.1. MATERIAL COMPONENTS OF MORTUAR Y BEHAVIOR IN 30 GRAVES BY PHASE

End 4th c.– early 5th c.

End 4th c.– early 5th c. Glass pitcher (7); glass bottle (13); iron nail

Lamp (5); bronze ring (23)

End 4th c.– early 5th c.

Early–Middle Roman, probably 1st–2nd c.

Roman, but probably Early–Middle Roman

Roman, but probably Early–Middle Roman

Roman, before early 5th c.

Date

Fine-ware jar (9)

None

None

None

None

Funerary Artifacts

III

II–III

II

Phase

A (–, 0–1 years); B (–, 2–4 years); C (–, 5–6 years); D (–, 25+ years); E (–, 25+ years); F (–, 25+ years); G (F, 50s or older); H (M, 35–44 years); I (–, 15+ years); J (F, 15–24 years)1

A (F, 35–44 years); B (F, 35–44 years); C (M?, 25+ years)

(F, adolescent or adult)

HO 70-902

HO 70-901

PAL 56-001

All supine and at least 3 extended; at least 6 have heads turned on L, at least 2 have heads turned on R; most have arms straight; 1 child has legs flexed to L 70°

A, B are supine and extended; B has head turned on R; C, D, E are clustered at W end

246

NEG 67-003

258

A (F, 35–44 years); None B (M, 35–44 years); C (–, 3–5 years); D (–, 0–1 years); E (–, 0–6 months)

N wall is the Hexamilion; S wall is extension of the N wall of 67-001; E end abuts stairway

N, E, W walls lined with small, flat stones, tile, and brick in courses; S wall is the north flanking wall of the Gate

Supine and extended, upper arms and legs straight

268

Cutting into concrete foundation; wooden structure (bier? coffin?) around and/or under body

3 slabs sealed with mortar between them and 18–23 stones around them

Bones in a pile

186

None

None

All supine and extended; 1 has head turned on L; at least 3 have heads partly turned L; at least 3 have arms straight

Corporeal Position

N, W walls are Hexamilion; S, E walls lined with upright tiles and stones at base; 2 compartments divided by upright tiles; N wall below Hexamilion lined with 25 small stones and several tiles in uneven courses

279

Lining and Paving Orient. Elements (°)

12 tiles in a heap just below top of cist; tiles enclose 2 compartments at lower level

Covering Elements and Grave Markers

NEG 67-001 A (M, 35–44 years); None B (F, 35–44 years); C (M, late 40s); D (F, 20–21 years); E (–, 7–11 years); F (–, 4–8 years); G (–, 4–8 years); H (–, 2–4 years); I (–, 2–4 years); J (–, 1–2 years); K (–, 2–4 years)

Sex and Age of Individual(s)

Area and Grave No.

TABLE 4.1 (CONT.)

0.60*

2.10*

1.00*

0.48– 0.52

1.52

2.20*

1.10*

1.27– 1.35

2.15*

1.60*

Width (m)

Length (m)

0.75*

0.70– 0.80*

0.46

0.40*

1.30

Depth (m)

1.65*

0.88– 1.01* [0.95*]

0.34– 0.36 [0.35]

0.74*

3.55– 3.77* [3.66*]

Volume (m3)

Middle–late 5th c. 10 coins (1, 2, 3); coarse-ware cup (8); pair of bronze earrings (16); iron buckle (18); bronze ring

Middle–late 5th c.

Early 5th– middle 6th c. Bronze earring (17); beaded necklace (33)

None

Very late 4th c.–middle 6th c.

Early 5th c.

Date

None

Coarseware bowl (10); bronze buckle (24); stone bead on a glass rod (25); iron disks (26); iron rings (27); iron nail (28)

Funerary Artifacts

IV

III

Phase

None

None

4 slabs and 11 A (F, late 40s– small stones mid-50s); B (M?, 16–18 years)

NEG 69-007

A (–, 0–10 months) Plaster and rubble marker; large tiles at lower level

Stone-lined drain enclosed by upright slab, tiles, and rubble to W of bodies

A (F, 35–44 years); Bodies interred B (M?, 35–44 years); inside a defunct C (F, 35–44 years) drain intact on all sides

RB 76-002

NEG 69-009

N wall is Hexamilion; E, S, W walls lined with brick in even courses

A (F, 50s or older); 3 slabs, 3 tiles, and 4 large stones B (–, 3–4 years); C (F, 20–24 years); D (F?, 45+ years)

NEG 69-001

A, C supine and extended; A has head turned on L, arms crossed over pelvis, R leg flexed, L leg straight; C has arms crossed over pelvis, legs straight; bones of B in a pile

282

273

Supine and extended

Both supine and extended, heads facing forward, arms crossed over pelvis

Bones of A, B in a pile; bones of C scattered; D supine and extended, head partly turned L, arms over pelvis but not crossing

258

260

A has body and head turned on L, legs flexed to L 85°; B is supine and extended, has head turned on R, R arm straight, L arm bent back and hand flexed, legs pulled R; C is supine and extended(?), has head turned on L

Corporeal Position

258

N wall is Hexamilion

3 slabs

A (–, 10–12 years); B (F, 20–22 years); C (M, 25–34 years)

NEG 69-005

Lining and Paving Orient. Elements (°)

Covering Elements and Grave Markers

Sex and Age of Individual(s)

Area and Grave No.

TABLE 4.1 (CONT.)

0.48– 0.52

0.55

0.65

1.60– 1.88

3.58*

2.11

0.54

0.62

1.97

0.82– 1.25

Width (m)

Length (m)

0.17– 0.29

0.30– 0.80*

0.71

0.54

0.60*

Depth (m)

0.08– 0.20 [0.14]

0.41– 1.10* [0.75*]

1.40*

0.41– 0.53 [0.47]

0.73*

Volume (m3)

Late 6th– very early 7th c.

Late 6th– very early 7th c.

None

Iron buckle (30); bronze loop (31); shell bead (32)

Late 5th– early 6th c.

Middle–late 5th c.

Pair of gold earrings (19)

Lamp (6); unguentarium (11); bone button (29); piece of textile or leather

Middle–late 5th c.

Date

Iron loop (22); hobnails

Funerary Artifacts

IV

Phase

Sex and Age of Individual(s)

A (–, 1–3 years)

A (M, 35–44 years)

A (M, 35–44 years)

(Adolescent or adult)

A (M, early– mid-50s)

A (M, 25–34 years)

A (–, 15+ years)

A (–, 1–2 years)

Area and Grave No.

NEG 69-010

T2 68-006

T2 68-002

T13 54-001

T14 67-002

T14 67-004

T14 69-701

T14 69-991

3 slabs and tile fragment over E end

Slabs(?)

None

Slab over E end

Tiles(?)

2 slabs, 1 block, and stone marker

4 slabs and stone marker

Plaster and rubble marker; large tiles at lower level

Covering Elements and Grave Markers

264

269

11 slabs and small stones lining walls

273

270

–

249

245

254

Stones lining walls

None

5 large stones around W end

Tile lining(?)

Walls lined with 11 thin slabs

Walls lined with 19 large stones

None

Lining and Paving Orient. Elements (°)

Supine and extended, head facing forward, arms crossed over chest

Supine and extended

Supine and extended, head facing forward, R arm over pelvis, L arm across abdomen

Supine and extended, arms crossed over abdomen; 1 painted slab beneath head and 2 stones beside it

Supine and extended, head turned on L, arms crossed over pelvis

Supine and extended, torso and head turned on L, arms crossed over chest

Supine and extended, head turned to R, arms over chest but not crossing

Supine and extended; one tile on floor under head

Corporeal Position

TABLE 4.1 (CONT.)

1.12

0.55

–

0.69

2.13

1.90

–

1.81

0.45– 0.55* 0.75*

1.80*

1.20*

0.70*

0.40

1.05

1.80*

Width (m)

Length (m)

0.45*

0.45*

0.40*

0.75*

–

0.34

0.43

0.23– 0.29

Depth (m)

0.52*

0.36– 0.45* [0.41*]

0.50*

0.94*

–

0.36

1.03

0.10– 0.12 [0.11]

Volume (m3)

7th c. or later

7th c. or later

None

Bone pendant cross (20) on beaded necklace (21)

7th c. or later

7th c. or later

Early Byzantine

Very late 6th–early 7th c.

None

Bronze ring (15)

–

None

Very late 6th–early 7th c.

Late 6th– very early 7th c.

Fragments of cooking pot (12) None

Date

Funerary Artifacts

A (M, 18–24 years)

T14 69-003

TC 60-001

LOU 69-801

A (F, 15–34 years)

A (M, 35+ years)

–

A (F, 25–34 years)

T14 69-002

T14

Sex and Age of Individual(s)

Area and Grave No.

None

None

3 large stones ringing E end and at least 11 tiles extending W

4 slabs over W end and 3 smaller stones to E

3 slabs

Covering Elements and Grave Markers

None

None

152

325

~260

268

None

Stone lining (?)

277

None

Lining and Paving Orient. Elements (°)

Bones in a pile, mixed with animal bones and debris

Torso supine and extended, head facing forward, arms crossed over pelvis, legs folded back to L over chest

Unknown

Supine and extended, head facing forward, L arm across abdomen, R arm across chest; 2 stones beside head and one tile over chest

Supine and extended, head facing forward, arms crossed over abdomen; 2 stones beside head and 1 under chin

Corporeal Position

Notes: An asterisk (*) indicates that a measurement is an estimate. Values in square bracks ([ ]) are arithmetic means. 1 This grave contained two unidentified males, 70-902D or I and 70-902E or F.

V

IV

Phase

TABLE 4.1 (CONT.)

1.55

0.65*

0.72

1.02

0.80*

2.00*

0.65*

0.55*

1.65*

1.50*

Width (m)

Length (m)

0.60*

0.20

–

0.40*

0.45*

Depth (m)

0.60*

0.15

–

0.64*

0.41*

Volume (m3)

None

None

–

Coin (4)

Pierced coin (14); 1 bone and 1 iron pin

Funerary Artifacts

Late Byzantine–Early Modern (15–19th c.)?

12th–14th c.

Early Byzantine

7th c. or later

7th c. or later

Date

GREEK FUNERAR Y RITUAL DURING THE ROMAN AND BYZANTINE ERAS

159

GREEK FUNERAR Y RITUAL DURING THE ROMAN AND BYZANTINE ERAS Roman and Byzantine writings furnish abundant testimony concerning Greek funerary ritual. Since historians have examined these sources at length,2 a summary of ritual activities will suffice for the purposes of this study. The direct comparison of written evidence to an archaeological model of mortuary practices in the rural Corinthia can be problematic for a number of reasons.3 First, the homogeneity of literary accounts masks the actual regional variability of practices and forms, which, according to the material record, was extensive during all periods. Second, Greek authors of fiction, oratory, letters, chronicles, homilies, commentaries, and hagiography wrote for an educated audience that appreciated rhetorical and narrative artifice. Moreover, the funerary scene became a topos in Byzantine literature. As a consequence, the literary portrait of death and burial can be distorted. Third, in their depiction of current society these writers represent the perspective of the educated, urban ruling classes, often the clerical elite. In reality, the exact quality and scale of mortuary behaviors varied according to social and ideological factors. Despite these tendencies in the written accounts, funerary ritual was fundamentally conservative, and many basic activities and meanings were widely recognized across southeastern Europe and Asia Minor. It is important to bear in mind that, while literary accounts of burial were not always factually accurate, they were often described with a degree of verisimilitude that contemporary readers would have found believable. So long as the limitations of the evidence are observed, the texts can provide a useful framework for examining practices and beliefs at the Isthmus. It will also be worthwhile to consider funerary ritual and burial space in Modern Greek culture, a subject already explored in Chapter 3.4 Since certain beliefs and practices today resemble or refer to those in the Classical and Byzantine past, the traditional acts of contemporary Greeks can shed light on the activities of their ancestors. The applicability of ethnographic analogy for interpreting archaeological data has been debated,5 but the comparison of premodern and modern Greek mortuary behavior is a valid analytical approach. In this study, demonstrable correspondences between past and present practices do not imply either the intentional emulation by participants or the superiority of one model over another; the comparison serves largely as a heuristic device. Historians often compare their reconstructions of Late Antique and Byzantine villages with Modern Greek communities of 2. Rife (1999, pp. 54–142) surveys mortuary practices in the Greek world under Roman rule. Rush (1941) remains a useful resource for late antiquity; see now the opposing perspectives in Volp 2002; Rebillard 2003b; Samellas 2002 (on the Greek East). Loukatos (1940) offers a useful compilation of the voluminous testimony of St. John Chrysostom. Byzantine burial customs were the subject of several synthetic studies by Greek scholars during the early to mid-20th century: Petrakakos [1905] 1971; Koukoules 1940; 1951, pp. 148–248; Megas 1940; Spyridakis 1950. Kyriakakis (1974) and Alexiou (2002, pp. 24–35) give shorter, more recent treatments. See Emmanouelides (1989) for a full study of the legal dimensions of Byzantine death and burial. Dennis (2001) explores many aspects of the experience of death in Byzantine society. Fedwick (1976), Rowell (1977, pp. 31–37), and Velkovska (2001) discuss the Byzantine funeral liturgy. Walter (1976) discusses death in Byzantine iconography. Angold (1995, pp. 442–457) discusses beliefs concerning death and rituals of commemoration in Middle Byzantine thought and society. 3. Rife (1999, pp. 13–38) discusses these problems in the study of funerary ritual in Roman Greece; see also Morris 1992, pp. 10–11. Abrahamse (1984, pp. 125–126, 133–134) and Alexiou (2002) criticize the historiography of Byzantine

death on similar grounds. Dennis (2001) exemplifies the positivistic approach to written sources on Byzantine death. See also Volp (2002, pp. 96–101) on the limitations of the literary and material sources for early Christian death and Geary (1994, p. 35) on the use of sources for relations between the living and the dead and saint’s cults in the medieval West. 4. Polites ([1931] 1975) investigates the ancient and Byzantine roots of Modern Greek customs; Alexiou (2002, pp. 36–51) covers similar ground. Loukatos (1978, pp. 221– 225) and Papacharalampous (1965, pp. 139–162) cover Modern Cypriot practices. Schell (1989) provides an overview that explores the intellectual frameworks underlying the interpretation of burial custom since the founding of the Modern Greek state. Danforth and Tsiaras (1982, esp. pp. 38–45) colorfully document in word and image the ritual process of death and burial in the village of Potamia in northern Thessaly in 1979. Seremetakis (1991) offers a penetrating account of funerary ritual and gendered identity in villages of Inner Mani in the 1980s. Tzortzopoulou-Gregory (2008) presents an important study of cemeteries in the eastern Corinthia today. 5. Parker Pearson (1999, pp. 34–44) surveys the question of ethnographic analogy and the archaeology of death.

160

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

a traditional character to identify similarities and reveal plausible explanations.6 The two share a continuous cultural and religious history that can be documented and a common range of environmental conditions. To be sure, differences over time in the ethnic, social, and economic complexity of Greek society have generated differences in the materials and activities relating to death and burial, as well as differences in the perceived meaning of these things and behaviors. Such shifts can become more clear, however, when textual, archaeological, and ethnographic sources are viewed side by side. The diverse written testimony for the treatment of the dead reveals that a standard ritual process prevailed in the Greek world during the Roman and Byzantine eras. A review of this process will provide a useful basis for comparison with the Isthmian remains. There were four main stages: death and the activities before the funeral; the funeral procession; the graveside activities and deposition of the body; and the activities after deposition. When a person was on the verge of death, relatives and friends gathered at the bedside to bid farewell, hear the final utterances, and catch the last breath. In later periods, attendants said prayers, granted forgiveness, and administered the Eucharist as a viaticum (ἐφόδιον). Around this time the family contemplated the distribution of property and arrangements for the funeral in accordance with the last wishes of the dying. Considerations that might have required advanced planning included choosing a site for burial, acquiring the necessary materials, finding help to conduct the funeral, and preparing the grave. During both the Roman and Early Byzantine eras, the corpse remained in the home immediately after death, while the family, servants, or associates cleaned and dressed it and began the lamentation. The laying-out of the body (πρόθεσις) in the bereaved household provided an opportunity for relatives, neighbors, friends, and other members of the community to view and to mourn the deceased before the procession to the grave. The ancient tradition of a three-day moratorium between death and burial persisted into Roman times, but the duration was commonly shorter than three days, and in Late Antique and Byzantine times it was not unusual for the corpse to remain at home just over night. During this period, domestic activities were typically suspended, and the house took on a different appearance as its floors were swept clean and the walls, windows, and doors were draped in black. Wakes in aristocratic homes during the Roman era could be public spectacles attracting numerous visitors, while those of leading members in a Christian community could be overseen by the Church. But for most families, the laying-out was a private affair conceived and enacted on a small, personal scale. The handling of the corpse is vividly attested, especially in the Christian tradition. During both the Roman and Early Byzantine eras the body was cleaned and anointed, enshrouded or clothed, and placed on a bier or in a coffin for transport to burial. The body was usually washed in warm water or oil mixed with spices or flowers, but wine, milk, and honey alone or in combination were also used. Chrismation occurred either during or after the washing and was usually repeated during the funeral and deposition. The mourners applied aromatic substances to the corpse, such as myrrh, aloes, mastic, and unctuous perfumes. The practical intention was to cleanse the dead body of pathogens, to counteract the odors of natural decay, and to bestow on the corpse an attractive, lifelike appearance. The ritual was also symbolic: the anointing marked the glorious perfection of life, while the ablution cleansed mortal sins. The manner of dressing the washed corpse seems to have remained essentially the same from Roman to Byzantine times. Members of the family or their assistants wrapped the body in clean, white linens or clothed it either in everyday garb or in an elaborate outfit, sometimes adding jewelry or other articles that had belonged to the deceased in life. In some cases the head was covered and the feet were shod. In Greek and Roman 6. See Chap. 3, n. 159.

GREEK FUNERAR Y RITUAL DURING THE ROMAN AND BYZANTINE ERAS

161

practice a wreath could be placed on the corpse’s head to emblematize honor, victory, and sanctity, and though Christian theologians inveighed against it, the practice continued during the Byzantine era. Another common custom during the Roman era that persisted in Byzantine and modern times was the placement of a coin, “Charon’s obol” (δανάκη, βολός, περατίκι, πορθμήϊον), in the mouth or on the chest of the deceased to pay for the toll on the journey to the afterlife. The arrangement of the corpse was also dictated by custom. Immediately after death and perhaps even before the body was washed, anointed, and dressed, the mourners or their assistants moved it to a vehicle or container on which it would be transported to the grave. This was either a wooden bier resting on the ground, on attached legs, or on a “base” or “pedestal” (βάθρον, ἰκρίωμα), or a casket of wood, lead, or stone, variably either freestanding or resting on a bier or a movable support. Biers and caskets were most often wooden constructions that were either plain or decorated with metal fixtures or painting. The bier was most commonly called the “bed,” “cot,” or “bench” (κλινή, κλινίδιον, σκίμπους, σκιμπόδιον, χαμεύνιον, κράβαττος, κράββατος, κραββάτιον, κραββατοφόριον, λεκτίκιον, σκάμνος, σκάμνον, σκαμνίον), but it was also generically called a “carrier” (φέρετρον, φορεῖον). The container for the corpse was simply called a “casket” or “coffin” (σορός, λάρναξ, γλωσσόκομον, γλωσσοκομεῖον, κιβούριον). The family members or their assistants positioned the corpse while it was still pliable so that the deceased seemed asleep during the laying-out and funeral. They laid the body on its back, closed the eyes and mouth, straightened the torso and legs, and bound the jaws with a ribbon or cord under the chin and over the crown. The Christian ritual of arranging the corpse was particularly strict. The legs were pushed together and the arms drawn up over the chest or stomach in the attitude of prayer or in the form of a cross, with the limbs held in place by winding sheets or binding cords. Sometimes the head was supported by a pillow as if in slumber, and icons or other commemorative or symbolic objects could be placed over the hands. During the laying-out, the body was oriented on the bier or in the casket so that it faced the east, the direction not only of the rising sun, which signified divine advent, but also of the biblical paradise and coming resurrection. Ritual lamentation often began in the household after the preparation of the corpse, but it was confined to the procession and graveside rite in cases when the funeral shortly followed death. In classical antiquity, mourners wailed and wept, tore their hair, beat and scratched their skin, rubbed dirt or ashes over their limbs and face, and often performed a “dirge” (θρῆνος). The relatives and friends of the deceased and other visitors to the laying-out, especially women, lamented in view of the corpse. Intense, boisterous lamentation continued into the Byzantine era, but other activities accompanied outward signs of grief. Psalmody replaced threnody, and Christian mourners prayed, listened to readings from scripture, and sang hymns. These events often took place in an “all-night vigil” (παννυχίς) at the side of the deceased, when tapers or oil lamps were affixed to the ends and sides of the bier or coffin. In some Christian communities, the liturgy for the dead was performed at the church, to which the corpse was brought before the procession. The next stages in the ritual process were the funeral and deposition of the body. The procession was identified by several terms connoting the final departure of the deceased from home and community (ἐκφορά, ἐκκομιδή, ἐκκομισμός, ἔξοδος, ἐξόδιον, ἐξόδευσις). The duration and scale of the funeral varied considerably depending on the social status of the deceased. During the Roman era, aristocratic funerals could involve a public parade of colleagues and dignitaries and an overt display of wealth. In later centuries, a similar level of extravagance marked the funerals of ecclesiastical leaders. In such cases, the funeral followed a protracted wake, and the cortège moved over long distances through the central areas of the settlement. Most funerals, however, were attended only by a small group of relatives and

162

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

close friends. The procession, whether large or small, was typically led by the bier or casket born by friends, assistants, or official pallbearers, after which came the bereaved family. The behaviors of lamentation that began during the laying-out continued during the procession. Elaborate funerals in the Roman era sometimes employed professional threnodes and musicians. During late antiquity and the Byzantine era, the procession was accompanied by the singing of psalms. Participants in the procession carried any objects for graveside use or burial, as well as lamps or candles. These provided light during ceremonies at night or in the early morning, but even in daylight the sight and smell of their burning, the odor and the smoke served as powerful signs. Since early times Greeks employed lamps and torches in cultic activities, not only as apotropaic devices but also as a means of illuminating the figurative passage from shadow into light, which variously represented truth, life, salvation, and divinity. Moreover, lamps were commonly deposited at sacred sites as votive offerings in classical antiquity, and in Christian funerals the placement of a lamp or candle near the dead, on the bier or casket, or at the grave often accompanied a private prayer or liturgical utterance. Several different events might have occurred once the procession reached the site the family had selected and prepared for burial. Greek authors of the Roman era describe the conflagration of property from the bereaved household and the sacrifice of animals at elite funerals, but such shows of conspicuous consumption were unusual. During both the Roman and Byzantine eras, the participants in the funeral often listened to a speech before or after the deposition. If the deceased was an especially significant individual, a professional speaker, a leading member of the community, or a cleric delivered a “funerary oration” (ἐπιτάφιος λόγος) praising the life and service of the deceased and consoling the bereaved. The Christian ἐπιτάφιος, which was in general less encomiastic than the Greco-Roman form, typically discussed the meaning of death and invited prayer and psalmody. In the last moments, when the body still lay in plain view, lamentation reached a climax, and the mourners bid farewell to the deceased with a stroke, an embrace, or a kiss. Interment in a prepared grave (τάφος, θήκη, σῆμα, μνῆμα) involved the handling of the corpse and the deposition of objects both inside and outside the grave. The form of the interment varied considerably depending on the investment of time, effort, and resources. The body was placed either inside a container, such as a wood or lead coffin, a stone sarcophagus, or a compartment cut from rock, or inside a cutting in the ground, which was lined or unlined. If the body was carried on a bier without a casket, it was either moved from the transport into the grave or interred directly on the bier. The literary testimony of the Roman era does not mention any specific treatment of the interred dead, but Late Antique and Byzantine sources describe the strict arrangement of the body. Those handling the corpse typically situated it as it had been arranged during the laying-out and procession, with the head facing east, the eyes and mouth closed, the jaws tied shut, and the arms and legs bound. During the Byzantine era it was not uncommon to cover the entire body with a sheet and to pour ointment or wine over it or into the grave before inhumation. The mourners often placed items they had brought in the procession inside the grave, such as vessels containing unguents or other liquids, small personal objects such as souvenirs, jewelry, children’s toys, or money. Sometimes mourners left lamps or candles at the graveside, together with any phials, pitchers, or cups from the chrismation or libation and even fresh flowers. The body disappeared from view as earth was dumped over it, and the grave was usually shut off with a solid covering. The last phase in the ritual process was mourning and commemoration after the deposition. Those who attended the laying-out and funeral cleaned themselves and the bereaved house to expunge the pollution that had emanated from the corpse. Fervent lamentation

MORTUAR Y FORM AND VARIABILITY

163

continued after burial. In Byzantine times the normal duration was nine days, during which time the spouse and sometimes other relatives donned black, kept a disheveled appearance, and for the most part stayed indoors. One ancient custom that continued throughout the Roman and Byzantine eras was a “common meal” or “funerary meal” (περίδειπνον, σύνδειπνον, νεκρόδειπνον) shortly after interment. The meal was most often held in the home of the deceased or a close relative to console the bereaved, but in some cases it was held at the grave according to the belief that the spirit of the deceased could then partake. Another custom observed throughout antiquity and the Byzantine era, as today, were periodic “memorials” (μνῆμαι, μνεῖαι, μνημόσυνα). At these gatherings mourners not only lamented the deceased and consoled one another, but also performed graveside duties to accommodate the restless soul before its passage to the afterlife. They prayed that the soul may rest peacefully, ate, drank, and left food or poured wine over the grave to nourish the dead. During the Roman and Byzantine eras, as today, memorials were typically held at the grave on the third, ninth, and fortieth days after death and on the first anniversary. Authors propose several explanations for these spans of time, including the incremental decay of the body, the duration of the soul’s wanderings, and the intervals between Christ’s crucifixion and his resurrection, return, and ascension.

MORTUAR Y FORM AND VARIABILITY The burials associated with the Roman and Byzantine settlement at the Isthmus show that the inhabitants observed the same essential repertoire of customs as is attested in the written sources. However, because the rural Corinthians did not possess the high prestige, wealth, or power of the urban aristocracy, the clergy, and the crowd of saints and martyrs who populate the more vivid accounts, local practices were performed on a narrower, less elaborate scale. Moreover, certain parts of the ritual process are not evident in the physical evidence, either because they were expressed through behaviors such as lamentation and liturgy, rather than through material media, or because those media, such as shrouds and flowers, have not been preserved in the depositional environment. What have survived are the durable materials in stone, terracotta, metal, and glass that were employed in graveside rituals after the preparation of the corpse, the laying-out, and the procession. Some of these rituals continued or concluded activities that had begun at the house or in the procession, while others occurred only at the grave immediately before, during, or after the deposition of the body. Furthermore, the material record reveals certain important aspects of mortuary behavior, such as the diversity in grave design and funerary assemblages, that Roman and Byzantine authors were not concerned to record due to their literary program or topical focus. This section examines the physical traces of funerary ritual at the Isthmus and how those forms evolved as the material components of changing behaviors. The discussion will critically evaluate the literary evidence for separate stages or features of the ritual process. It will also address the relationship between graves at the Isthmus and those at other sites in the northeastern Peloponnese. The most significant comparanda come from published cemeteries at Corinth (the Asklepieion, Lerna Hollow, and the Gymnasium area; the Forum; Acrocorinth; the churches), Nemea, Argos, Halieis, and numerous other areas explored by excavation or survey in the Corinthian hinterland and the Argolid.7 These graves provide 7. The comparanda listed in the subsequent notes are arranged first by date (from Roman to Byzantine to post-Byzantine) and second by area within each era (beginning with the

city of Corinth, followed by its hinterland, and expanding into the greater Corinthia and Argolid).

164

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

relevant analogues for the remains at the Isthmus. Together they constituted a tradition of grave typology and funerary assemblage that was, in broad outline, regionally consistent. Many of these graves were made within comparable environmental settings and under the same social, economic, and religious conditions as the graves at the Isthmus. Perhaps most importantly, residents at the Isthmus would have known the surrounding mortuary landscape. When they buried their dead, they responded to the current mortuary forms and ritual actions at Kenchreai, for instance, or in the Corinthian suburbs. Outside of these regional comparanda, the present discussion will note a few parallels from elsewhere in the Greek world that offer especially apt illustration or explanation.8 The regional archaeological evidence for funerary ritual presents certain challenges that confine, but do not preclude, its application to comparative analysis.9 The publications vary dramatically in quality and extent, from passing references to short reports to detailed studies. Many reveal ignorance or disinterest with regard to the complexities of mortuary analysis, and therefore they lack crucial data such as depositional sequences and corporeal positions. Notwithstanding these shortcomings, the publications contain valuable information that historians and archaeologists cannot afford to disregard, with the provision that they discriminate between empirical evidence and subjective judgement. Another problem with the mortuary record of the Roman and Byzantine Corinthia and Argolid is chronology. Some publications assign no date; others give a date, but the rationale is questionable. Therefore, citations of comparanda in the present study will give a broad date by period or era based on the conclusions of the excavators or the evidence as published.10 This gross periodization of graves is adequate for tracing changes in mortuary practices over a long term of several centuries. It should, however, be emphasized that the chronological imprecision of the published data generates considerable blurring around the Late Roman/Early Byzantine transition. As a result, many graves originally identified as Late Roman may well belong to the late 6th to 7th centuries or even later. Indeed, recent refinements in the understanding of artifactual types and dates has in some cases led to the downdating of burial sites.11 Once the physical remains of funerary ritual at the Isthmus have been summarized and compared to the written sources, the regional mortuary record, and modern practices, their variability will be defined according to three essential factors: sex, age at death, and date of interment. This will determine whether men and women and adults and children received different treatment in death. There is no indication that other biological factors, such as pathological conditions or physical abnormalities, influenced funerary ritual at the Isthmus. Finally, any changes in the mortuary evidence will be traced over time to see whether funerary ritual evolved within the community. The final section in this chapter will interpret the dimensions of social and ideological experience that influenced mortuary variability.

Rituals before Burial No physical remains have been found at the Isthmus that can be positively identified with activities before the funeral. However, the nature of events during the first stages of preparation, mourning, and burial can be inferred from what is known about the contemporary 8. Laskaris (2000) synthesizes much archaeological evidence for Late Roman and Byzantine mortuary practices from sites throughout Greece. 9. The published record shows many of the same methodological and theoretical shortcomings as the early investigation of the graves at the Isthmus; see pp. 16–17.

10. Estimated dates will be signaled with a question mark (?). 11. A prime example is the revision of the burial chronology at the Asklepieion, Lerna Hollow, and the Gymnasium area north of Corinth from the late 4th to mid-6th centuries to the 5th to 7th centuries (Sanders 2004, pp. 180–184, 2005b, pp. 430–437; Slane and Sanders 2005, pp. 290–291).

MORTUAR Y FORM AND VARIABILITY

165

settlement. The activities immediately after death would have occurred in the bereaved house, unless they were held in an as yet undiscovered church. It is uncertain whether the few houses uncovered at the site belonged to particular families responsible for the known graves, but presumably they are typical residences. The Roman agglomeration in the East Field, which served in part as a residential quarter, and the Early Byzantine houses within the Bath and temenos were all modest structures that would have afforded confined living quarters for a few individuals. The size of these buildings would not have allowed an elaborate laying-out of the dead. Any activities involving corpses in these houses must have been limited, private affairs. The same tight spaces where occupants slept, dressed, and prepared their meals would have been dominated by the immediate presence of the deceased. The propinquity of the living and the dead must have generated an intense and emotionally charged atmosphere. Once mourners or their assistants had treated the body and prepared a site for its interment, the procession set out from the house. Local funerals probably never achieved the large scale of the processions for grandees and clerics recorded in the historical sources. They still would have affected the entire community. Due to their proximity, intercourse, and cooperation on various levels, the residents, like modern villagers,12 would have been aware of important events in the lives of their neighbors, such as births, marriages, and deaths. Furthermore, the spatial distribution of habitation and burial indicates that processions must have passed through or near the settlement, where the corpse and the train of mourners would have been open to view. Residents on the periphery of the Sanctuary during the Early to Middle Roman burial phase (I), such as those east of the central area, likely buried their dead along the northern limit of the site. Their funerals would have passed northward through an area of dense architecture where the Fortress would later stand. During the Late Roman burial phases (II–III), most residents lived within the enceinte but still buried their dead to the north. Their processions would have passed through the Fortress either northward to the Northeast Gate or westward to the Bath. The procession that terminated at the Palaimonion (56-001) would have been especially conspicuous, as the corpse was carried into the precinct past fallen buildings where any dead flesh was once forbidden. The route of local funerals during the Early Byzantine burial phase (IV) is more difficult to trace because habitation was dispersed between the Fortress, the Bath, and the temenos. Since several graves of this date were immediately north and west of the enceinte, at least some processions traveled a distance to reach their destination. The only materials primarily used before burial that have been excavated are the remnants of a bier or coffin in PAL 56-001. Splinters of wood, bits of plaster bearing the imprint of wood, and iron nails were found on the floor of the cist near its corners and middle. These remains constituted a wooden structure of planks fastened together with nails and probably lined with plaster that was arranged underneath and alongside the corpse. This was the bier or coffin on or in which the corpse was placed during the laying-out, the funeral, and the burial. Although its exact form and function are unclear,13 the structure was more likely a bier because the corpse was interred in a solid concrete chamber in which a coffin would have provided no real protective advantage or spatial confinement. The historical sources 12. See du Boulay 1974, p. 44 (Euboea, late 20th century); Kenna 1991, pp. 102–103 (a Cycladic island, 1960s–1980s); Alexiou 2002, p. 42. 13. One possible design is furnished by a well-documented example from Polystylon (Abdera) in Thrace. Bakirtzis (1986, p. 19) records the discovery of several nails around a skeleton in a Late Byzantine cist and reconstructs a “ξύλινον φερέτρον” in the form of a lidless casket. The photograph of the nails in

situ (pl. 19) shows that they were hammered into the structure from below, in which case they most likely attached cross supports to longitudinal planks. Gerstel et al. (2003, pp. 187, 199–200, fig. 63) record a stone-lined and covered cist (grave 1992-2) in the narthex of the Late Byzantine church at Panakton in Attica that had plastered walls and a wooden cover over the interior burial compartment.

166

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

and modern evidence for biers and coffins in Greek practice is extensive.14 Hagiographic sources even specify the use of wooden frames or boards nailed together.15 Several Corinthian graves have produced evidence for a bier or coffin similar to that in PAL 56-001. Excavations north of Corinth, south of Lechaion, and north of Kenchreai have revealed Roman tile-covered graves or loculi in chamber tombs containing iron nails of various sizes with occasional fragments or imprints of wood, all arrayed on the edges of cists or across them in lines.16 Further comparanda come from Late Roman or Byzantine graves near the urban center and on its periphery. Three graves from the Kraneion, which were lined with brick or marble and enclosed with slabs, contained wooden splinters and “coffin-nails,” while one stone-lined cist found under the new Panayia church contained fragments of square planks just above its floor.17 Coffins were also used at Early Roman and Late Antique Argos18 and in later centuries at Corinth.19 Many comparable wooden structures have undoubtedly been lost from the archaeological record through the natural decay of wood and iron. Within this admittedly small sample of comparanda, a plaster coating would have distinguished the structure at the Palaimonion, giving the planks a clean, finished appearance, perhaps bearing painted or engraved decoration. The simple design of this bier or coffin did not, however, approach the level of complexity or ornamentation of either the most lavish forms described by Roman and Byzantine writers or the elaborate metal fixtures for caskets found in Roman and Byzantine graves elsewhere in Greece.20 The wooden furniture in PAL 56-001 is unique among the graves at the Isthmus. No fragments of wood or nails in a similar pattern have been found in any other grave near the Sanctuary and fortifications, including three (NEG 69-001, T2 68-003, RB 76-002) that were as well protected against disturbance by environmental agents as the cist cut into the concrete foundations of the Temple of Palaimon. Other local residents must have used vehicles to transport corpses to burial but did not deposit them with the dead. Perhaps they used ad hoc constructions which were destroyed after the funeral, or one vehicle was shared and reused by members of the community. 14. Early to Middle Roman evidence: Rife 1999, pp. 68– 69. Late Roman and Byzantine sources: Loukatos 1940, pp. 65–66; Spyridakis 1950, pp. 115, 151–152; Koukoules 1951, pp. 173, 176–177; Abrahamse 1984, p. 131; Samellas 2002, pp. 230–231, n. 129. Modern practice: Polites [1931] 1975, p. 333, n. 8; Megas 1940, pp. 174, 179; Papacharalampous 1965, p. 150; Kenna 1991, pp. 102–103. 15. E.g., Agathang. A. s. Greg. Armen. 11.122 (AASS Sept. 8.375C; “coffins of cedar planks held together by sturdy nails” [λάρνακες . . . ἀπὸ σανίδων κεδρίνων ἐν στερεοῖς ἥλοις συγκλεισθεῖσαι], martyrs buried ca. 301, Valashabad, Armenia; acta written late 5th century or later); Method. V. s. Theoph. 56 (d. 817, Samothrace; vita written early 9th century), cf. Niceph. Sk. V. s. Theoph. Conf. p. 25.30; Nicet. Dav. Paph. V. s. Ign., PG 105.557 (d. 877, Constantinople; vita written end of the 9th to mid-10th centuries). 16. Corinth XIII, p. 76, n. 60, pp. 95, 297–298, 300, nos. 511, 512, 515, 528, 529, pl. 123 (North Cemetery, Early Roman); Walbank 2005, p. 274 (north of Corinth, Early–Middle Roman); Drosoyianni 1972, p. 206 (south of Lechaion, Roman); Rife et al. 2007, p. 160 (Koutsongila ridge, Kenchreai, Early– Middle Roman). 17. Carpenter 1929, p. 346 (Kraneion, “Byzantine”); Pallas 1959a, p. 19 (“διατηρούμενα λείψανα τῆς νεκρικῆς κλίνης ἐκ τετραγωνικῆς τομῆς [0.09 x 0.09 μ] ξύλων κέδρου ἢ κυπαρίσσου,” site of the new Panayia church, dated late 6th century).

18. Nails from coffins have been found in Early Roman (or earlier) graves in the Quartier Sud: Bruneau 1970, pp. 454, 490, 524 (tombes 30, 86). Single or multiple iron nails, most or all from coffins, have been found in Late Antique graves north and south of the ancient city: Oikonomou-Laniado 2003, pp. 27, 28, 32 (area nos. 180, 587, 649). Kritzas (1979, p. 242) records the discovery of iron nails and wood fragments in Byzantine (or later?) graves east of the amphitheater. 19. Corinth XVIII.3, p. 382, n. 6, cites “remnants of a wooden coffin” in grave 1928-18 in a paved court east of the Theater, dating “no earlier than the 11th or 12th century.” Williams and Zervos (1991, p. 39) and Williams, Barnes, and Snyder (1997, p. 21) note iron nails, a coffin hinge, and traces of wood in Byzantine, Frankish, or Ottoman graves east and south of the church southeast of Temple E. 20. Carington Smith (1982, pp. 282–284, figs. 3, 6, pls. 40:c, 41:a–d) describes a bronze lock and catch-plate, swinging handles, chains, rings, bosses, cotter pins, tacks, and iron nails found in a Roman chamber tomb southwest of Knossos; see also Wardle and Wardle 2004, pp. 477–479, figs 43.1–43.3 for similar remains from Early Roman tombs in the same area. Similar fixtures, sometimes associated with wood, are noted in Roman graves, sarcophagi, or chamber tombs at Athens, Patras, and Siphnos. Laskaris (2000, pp. 112, 176–177, 235, 272) records bronze and iron fixtures for coffins in Late Antique and Byzantine burials found at Thessalonica, Mystras, and Sparta.

MORTUAR Y FORM AND VARIABILITY

167

Selection, Acquisition, and Preparation of a Site for Burial Before the deposition of the body it was necessary to select, acquire, and prepare a site for burial.21 If this had not already been taken care of before death, it must have occurred shortly thereafter. It seems that in order to obtain and improve a place for interment, the residents of the Isthmus simply claimed an unoccupied spot and procured necessary materials. This diverges from the usual procedure in Corinth and its surrounding communities.22 There, according to Late Antique epitaphs, burial plots were usually “purchased” (ἠγοράσθη) for a specified “price” (τιμή), at which point they became personal property.23 There is no evidence for the commercial transaction of plots at the Isthmus. Moreover, while the use of the buildings of the Sanctuary and the fortifications was in part controlled by sacred law and Imperial edict, the burial zones were apparently unregulated. So long as the dead were kept away from homes and were not in immediate danger of disturbance by building or cultivation, the residents were free to choose sites for burial. The distribution of graves during different phases in the history of the settlement has been discussed at length in Chapter 3. Burials were always segregated from areas of active habitation, though all Late Roman and Early Byzantine graves were not outside the enceinte, where the local population was concentrated. Residents usually decided to place new graves near old ones (NEG, T14), but sometimes they placed graves in small groups (T2) or in isolation (RB, HO, PAL). All graves were accessible, but some were more prominently placed than others. Many of the graves from the era when traffic through the site was most frequent and habitation most populous (phases I–III) were situated alongside the road from the northeast, and, in the case of PAL 56-001, at a focal point of cult activity in the ancient Sanctuary. The presence of a local church might have also attracted burials. If a church associated with the initial occupation of the Fortress underlies Ayios Ioannis Prodromos, then the graves at the Northeast Gate could belong to a large churchyard cemetery. Over time, the desire for accessible or prominent locales for burial diminished or at least grew inconsistent. During the late 5th or early 6th century, mourners chose a secluded spot for RB 76-002 in a subfloor drain exposed in a narrow hole in the corner of a back room of the dilapidated Bath. Graves were still placed on the northern edge of the site during the Early Byzantine period, but a burial ground had also developed in the shadow of the Fortress’s western wall. This area would not have been visible to anyone passing through the enceinte or to its east, though it would have been seen by anyone using the postern gate next to Tower 15. The placement of graves within separate locales was influenced by preexisting graves and architecture. In some cases, the relationship between derelict buildings and graves reflects an evolving conception of the constructed landscape. The decision to place a grave conspicuously in the podium of the Temple of Palaimon (56-001) might have evoked the former significance of the monument, whereas the reuse of the drain in the Bath for burial (76-002) creatively adapted a derelict structure for a new purpose. The isolation of one of the latest graves and the unusual situation of the body (LOU 69-801) indicate that the dead man was marginalized or excluded from the community.24 In other cases, the location 21. Emmanouelides (1989, pp. 275–298) discusses this process in Roman and Byzantine law. 22. Danforth and Tsiaras (1982, pp. 57–58, n. 12) and Tzortzopoulou-Gregory (2008, pp. 310–311) note that graves in Modern Greek villages have often been free, whereas burial plots in cities are purchased or rented. This was the case for parish members at Kyras Vrysi until recently.

23. Corinth VIII.3, nos. 522, 530, 551, 552, 556, 558a, b, 561, 579, 584, 595; IGCB I 41; IG IV 437 = Pallas and Dantis 1979, pp. 63–64, no. 2, pp. 69–70, no. 8, p. 71, no. 10, pp. 82–83; Wiseman 1967b, pp. 422–424; 1972, p. 42 no. 35. Two epitaphs from Late Antique Argos record the purchase of graves for two νομίσματα (Oikonomou-Laniado 2003, p. 55, nos. 16, 20). 24. See pp. 230–232.

168

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

of graves reflects practical concerns. The location of five graves along the interior face of the Hexamilion25 might well have facilitated their rediscovery for the addition of corpses on later occasions. There are no walled enclosures or built connections proving that pairs or groups of graves were related to each other.26 It is, however, quite possible that clusters of Early Byzantine graves outside the Northeast Gate and near Towers 2 and 14 contained members of the same family.27 This proposal receives strong support from the fact that the only two skeletons from the site to display cleft neural arch defect of the sacrum, which is a congenital anomaly, were buried at Tower 14 (67-002A, 69-002A).28 One trend in the spatial distribution of graves within a burial area is their increasing separation over time. This development is most evident when the plan of the Late Roman graves inside the Northeast Gate is compared to the plans of the Early Byzantine graves outside the Northeast Gate and west of the Fortress (Figs. 2.1, 2.45). The differences between these locales suggest that residents were planning the situation of their graves more conscientiously over time, presumably to facilitate graveside rituals. During the Late Roman period, mourners at the large multiple interments inside the Northeast Gate must have huddled closely around the cist and almost up against the rampart in order not to step on, walk over, or disturb adjacent graves. Likewise, the narrow width of the drain in the Roman Bath would have permitted only one or two people to deposit and enclose the three corpses (76-002A–C), while other mourners could only watch from the floor above. In contrast, during the Early Byzantine period, people had ample space to gather next to the graves outside the Northeast Gate and near Towers 2 and 14, at least four of which were plainly marked. These residents arranged their graves to accommodate more spacious graveside activities than in earlier centuries but never built an impressive structural context for the observation of those rituals. Their graves had no permanent installations for funerary banquets or memorial services, such as the architectural features associated with several Roman and Early Byzantine tombs at Corinth, Lechaion, and Kenchreai.29 Once a site was selected and claimed, it was prepared for burial. Most cists were excavated from a cleared surface, but in some cases the creation of a grave required substantial alteration to the physical surroundings. The builders of NEG 69-005 used a coarse chisel to rabbet the cover slabs securely into the face of the Hexamilion (Fig. 2.16), just as was done in the construction of graves at Corinth in late antiquity.30 Graves disrupted the facade of the Temple of Palaimon (56-001), the stairway up to the fighting platform inside the Northeast Gate (67-003), and the northern wall of the taphros (69-007, 69-009, 69-010). Since the Early Byzantine graves near Towers 2 and 14 were located away from preexisting structures, their preparation would have simply required that the ground be cleared. 25. NEG 67-001, 67-003, 69-005, 69-001, HO 70-902. 26. The rubble wall shared by NEG 67-001 and 67-003 is merely part of an efficient design that would not have been visible at ground level. Wiseman (1969, p. 81) proposes that two Late Roman to Early Byzantine graves in the Gymnasium area at Corinth covered at the surface by earthen mounds and connected by a strip of stucco contained the bodies of related individuals. Nothing comparable has been found at the Isthmus. Few funerary precincts at Greek sites of Late Roman or Byzantine date have been documented; see Laskaris 2000, pp. 263–264. Robinson (1965, p. 79) records four Late Roman vaulted tombs within a peribolos near Anaploga; Fox Leonard (1997, pp. 363–369) discusses the skeletal remains from these burials. 27. See Corinth XVIII.3, pp. 381, 387, suggesting that separate groups of Late Roman or Byzantine graves over the Sanctuary of Demeter and Kore on the north slope of Acrocorinth (fig. 53, plan 6) may comprise “family burial plots.”

28. See p. 408. 29. Charitonides 1968, p. 122 (τάφος Z I with triclinium, Cheliotomylos, Corinth, Roman); Pallas 1959b, p. 214, fig. 29; 1975, pp. 7–9, figs. 10, 12; Kenchreai I, p. 11, pl. IX:A; Rife et al. 2007, pp. 155, 158, 161–162, fig. 7 (pedestals, benches, or offering tables in tombs, north and south of the harbor at Kenchreai, Roman); Williams and Fisher 1975, p. 16 (double-chambered tomb beneath possible burial chapel, southwest Forum, Corinth, 6th century or later); Pallas 1972, pp. 99–102 (triconch as triclinium, Kraneion basilica, Corinth, Late Roman); Pallas 1965, pp. 133–135, fig. 1, pl. 118:β (walled, raised area near tomb, south of Lechaion basilica, Early Byzantine?). Laskaris (2000, p. 268) cites examples of similar facilities at Eretria, Thessalian Thebes, and Kilkis. 30. Robinson 1962, p. 110, pl. 35:c (West Shops, Late Roman–Early Byzantine); Wiseman 1967b, p. 420, pl. 84:e (graves 41, 42, Gymnasium area, Late Roman–Early Byzantine).

MORTUAR Y FORM AND VARIABILITY

169

Multiple Burial An important decision facing the bereaved family was whether to inter the deceased alone in a fresh grave or with one or more bodies in an existing grave. If they chose an existing grave, they had to locate it. Since there is no evidence that multiple burials at the Isthmus were marked at the surface, mourners must have remembered where graves were situated relative to permanent features or landmarks. As has been discussed, cists containing multiple burials were used over a time no longer than roughly a half century, and probably in many cases the duration was much shorter.31 However, in no instance does the form or content of a grave indicate that many bodies were interred simultaneously or hastily out of panic or calamity, such as during war or epidemic. These individuals were buried conscientiously and peacefully over time; they were not the anonymous victims of plague or invasion.32 Mourners found the grave to be reused and dug out the overlying soil before the funeral, but the cist probably was not fully opened until it was time to deposit the dead. The removal of the slabs, tiles, or rubble covering the cist and the exposure of previously interred bodies must have made a strong impression on those present. They saw and smelled the corpses of persons they had once known alive in various stages of decay from putrefaction to skeletonization. Once the grave was open, those handling the corpse carefully placed it inside. Sometimes they pushed aside bodies already in the grave to create space for the new additions,33 and sometimes they enlarged the cist by moving one of its walls.34 Although they could have piled up the old bones, often they repositioned them so that the cranium pointed to the west, the pelvic remains were set nearby, and the appendicular diaphyses were placed to the east and aligned axially.35 In one case, bones were permanently removed from the grave.36 Once the mourners had deposited the recent corpse among the earlier ones, they usually closed the cist unevenly, leaving the slabs offset and imperfectly sealed or scattering the tiles and rubble.37 Two lined cists containing a total of 16 bodies were found without covering, presumably because they had been opened so frequently that the stones or tiles eventually had been lost or discarded.38 Those who chose multiple burial invested considerable time and effort in finding, opening, and altering a grave for continued use over a relatively short duration. Their activities were considered and purposeful, not impulsive or random. Individuals buried in the same cist had a significant relationship to one another. As will be discussed, the grave design and funerary assemblage of all multiple burials of the Late Roman period were essentially the same. This consistency gives the impression that individuals in one burial group were not distinguished from those in other burial groups by profession. Status differentiation in the community seems to have been slight. Moreover, inclusion in multiple burials was not determined by sex or age; the groupings comprised both males and females, adults and children. The most likely relationship between these individuals is familial.39 This would explain the distribution of individuals by age and sex 31. See pp. 24–25. 32. Four graves (NEG 69-103, 69-007, T2 68-003, DEC 69901) contained two bodies probably buried at different times. Assuming that the couples were related, their simultaneous burial would be inexplicable unless they coincidentally died at the same time or fell victim to the same illness or traumatic event. Such circumstances are conceivable but unlikely. On the other hand, all four graves were covered by stones that could have easily been removed for the addition of a corpse. 33. NEG 69-001, HO 70-902, RB 76-002. 34. NEG 69-001, HO 70-902. 35. NEG 69-004A, C–E, 69-001A–C, RB 76-002B. 36. Those who interred the last body (B) in NEG 67-003 removed an earlier skull (A) but left its postcranial bones. 37. NEG 69-005, 69-001, 69-007, HO 70-902, RB 76-002.

38. NEG 67-001, 67-003. Note that the covering on HO 70902, containing 10 bodies, was incomplete and fragmentary. 39. Gejvall and Henschen (1968, p. 189), Wiseman (1969, p. 85), Wesolowsky (1973, p. 351), and Ivison (1996, p. 108) arrive at the same conclusion regarding comparable burials of Late Roman to Early Byzantine date at Corinth; see also Pallas 1951, pp. 164–165 (Salamis); Oikonomou-Laniado 1998, p. 409; 2003, p. 33 (Argos). There is no positive evidence to support the hypotheses of J. Wiseman (1969, p. 81) that such burial groups were defined by racial affinity or membership in a “burial guild.” Emmanouelides (1989, pp. 224–225, 263–270) discusses family vaults in Roman and Byzantine law and epitaphs. Laskaris (2000, pp. 280–281) surveys numerous Late Roman and Byzantine examples of “tombes familiales” throughout Greece.

170

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

within multiple burials (Tables 4.1, 5.3). The ten graves associated with the Late Roman fortifications contained 17 female adults, 11 definite or possible male adults, two male or female adults, and 17 subadults, mostly infants.40 The distribution of adults and subadults in the same graves, presumably parents and children, points to the collective burial of families. The hypothesis is supported by the coincidence of heritable traits among the skeletons in multiple burials.41 The continued use of graves for the interment of several corpses was a common practice in the Greek world during late antiquity and the Byzantine era. The literary testimony for this custom is not as extensive as the archaeological evidence, but graphic accounts can be found. St. Gregory of Nyssa wrote that in 379 his sister, Macrina, was buried in “the tomb of her parents” (ἡ σορός τῶν γονέων), probably in a stone coffin on the family’s estate in Pontus. When the funeral procession arrived, Gregory was horrified by the thought of viewing the “ugliness” (ἀσχημοσύνη) of his mother and father, whose corpses already occupied the grave. Therefore, once the lid was removed, he covered their bodies with a sheet before moving the body of Macrina from the bier to her mother’s side (V. s. Macr., PG 46.994–996). Residents of the Isthmus who used multiple burials may well have had a similar reaction and may also have draped older bodies disfigured by decay when interring relatives. In his vita of John the Almsgiver written in 641–642, Leontius described how the saint was deposited in an earlier tomb in the oratory of St. Tychon at Amathous on Cyprus in 619/20. The two bishops previously interred there moved themselves aside to make room for John’s body in the middle of the cist (V. s. Ioan. Eleem. 405.16–27). This miraculous act of corporeal animation accomplished what mourners, such as those at the Isthmus, typically would have done by hand. The material record abundantly demonstrates that multiple burial was practiced for several centuries.42 Numerous Roman and Byzantine graves containing more than one body have been found across the northeastern Peloponnese. During the Roman era, relatively wealthy families buried their dead in chamber tombs either constructed above ground or cut from bedrock in prominent locations.43 In subsequent centuries, the practice of multiple burial took a variety of forms. Corinthians and Argives built tombs that were less impressive and smaller than their antecedents during late antiquity in urban, suburban, and rural settings. Some were chambers with stepped dromoi, entrance shafts or manholes, anterooms, and barrel-vaulted ceilings,44 while others were brick-lined and marble-faced chambers 40. This calculation assumes that NEG 69-008 contained a subadult. 41. See pp. 291–292, Table 5.18. 42. The roughly contemporary use of multiple burials in the basilica at Aliki on Thasos (Buchet and Sodini 1984, p. 235) and at the Knossos Medical Faculty site on Crete (Becker 2005, pp. 371–377, 380–381, table 3; Sweetman 2005, p. 366) closely resembles the treatment of the graves at the Isthmus. 43. On Early to Middle Roman chamber tombs in the Corinthia, see Rife 1999, pp. 235–250 (architecture), 319–321 (epitaphic commemoration of families). Walbank (2005, pp. 255–269, figs. 9.5–9.11) presents a detailed discussion of a group of chamber tombs north of Corinth and Rife et al. (2007, pp. 145, 147, 153–176, figs. 2, 3, 7–25) survey a major cemetery of Early to Middle Roman chamber tombs north of the harbor at Kenchreai. On similar tombs from the suburban cemeteries north and southwest of Argos, see Deïlaki 1977, pp. 109–110, fig. 17, pl. 105:a–d; Banaka-Dimaki 1998. 44. E.g., Pallas, Charitonides, and Venencie 1959, pp. 496– 497 (three skeletons; Longopotamos valley southwest of Solo-

mos, Late Roman); Papachristodoulou 1970, pl. 80:β (three skeletons; Palaio Scholeio north of Klenia, Late Roman); Kounoupiotou 1972, p. 208, pl. 196:γ (three skeletons; Kraneion, Late Roman or Early Byzantine); Banaka-Dimaki 1988b, p. 89, fig. 2 (τάφος XIV; Ayia Anna, Late Roman or Early Byzantine); Morgan 1938, p. 370, fig. 11; Corinth XVI, pp. 127– 128, pl. 17:1; Ivison 1996, p. 113 (15 skeletons; South Stoa, Early Byzantine); Anderson 1967, p. 6, figs. 3, 4, pls. 1:b, 2:a, 3 (northwest of Temple E, 6th century or later); Williams, MacIntosh, and Fisher 1974, pp. 9–10, pl. 2 (11 skeletons; west Forum, 6th century or later); Williams and Fisher 1975, pp. 15–17, pl. 6 (seven skeletons; southwest Forum, 6th century or later); Pallas 1965, p. 133 (south of Lechaion basilica, Early Byzantine?); Kritzas 1976, pp. 205–207, fig. 10; Kaza-Papayeoryiou 1988, p. 119; Oikonomou-Laniado 1998, pp. 405–406, 409, figs. 1, 2, 7, 2003, pp. 27–29, 30–31, 33, figs. 40–43, 47–50, 56, 57 (North and East Cemeteries [area nos. 180, 378, 587, 623], Argos, Late Roman–Early Byzantine); Konsolaki 1989, p. 50 (near Troezen, Late Roman–Byzantine). Pace Curta (2005, p. 126 n. 103), chambers with stepped dromoi are not an uncommon sepulchral form in Late Antique Greece.

MORTUAR Y FORM AND VARIABILITY

171

annexed to Christian churches.45 Other residents of the region, like those at the Isthmus, interred their dead with other bodies in less elaborate interments, some enclosed and demarcated using inexpensive materials and some not.46 Epitaphs from the Late Roman to Early Byzantine cemeteries of Corinth and its surrounding communities reveal that individuals in such burial groups belonged to nuclear families. Many inscriptions refer to the joint interment of spouses, siblings, and parents and children,47 while others refer to the interment of children, wives or husbands, and siblings in graves owned by their parents, husbands or wives, or siblings.48 These epitaphs imply that the owner of the site would also be interred there,49 but that probably did not happen in every case. The impulse for collectivity in death is also evident in one of the most striking and prevalent features of Corinthian chamber tombs: their continuous use for the deposition of numerous bodies. Many of the known family vaults display two broad phases of mortuary use: their construction and initial operation in the Early or Middle Roman period and the addition of several bodies, often together with lamps, pottery, and coins, in the Late Roman or Early Byzantine period.50 It is, however, uncertain whether the individuals placed in these tombs during late antiquity represented families or simply belonged to anonymous ossuaries. The frequency of multiple burials at the Isthmus changed dramatically over time. All 17 graves associated with the Roman and Early Byzantine settlement (phases I, IV) contained one or two individuals each, or on average 1.06 bodies per interment. In sharp contrast, the ten graves with well-preserved human remains associated with the Late Roman settlement (phases II–III) contained one to 11 individuals each, or on average 4.70 bodies per interment.51 45. Carpenter 1929, pp. 349–350, 352; Shelley 1943, pp. 176–179, 183; Pallas 1959b, pp. 204–205; 1972, p. 99; 1974, pp. 213–219 (Kraneion basilica); 1961, pp. 173–174, pl. 72:γ (Lechaion basilica). 46. Robinson 1962, pp. 116–117 (three graves with two, three, and three skeletons; near Anaploga, Late Roman); pp. 110–111, nn. 45, 46, fig. 2, pl. 35:c; Gejvall and Henschen 1968, p. 179, figs. 1, 2 (one grave with two skeletons, one grave with three skeletons; West Shops, Late Roman–Early Byzantine); Corinth XIV, pp. 164, 165, n. 17 (two to four skeletons); Wiseman 1967b, pp. 418–419 (three skeletons, grave 35 and eight skeletons, grave 53); 1969, p. 79 (six skeletons, grave 86), pp. 82–83 (three skeletons, grave 73 and three skeletons, grave 77), p. 85; 1972, p. 8 (two skeletons, grave 94); Wesolowsky 1973, pp. 341–342 (Asklepieion, Lerna Hollow, and Gymnasium area, Late Roman–Early Byzantine); Corinth III.1, pp. 36, 40–42, fig. 36, pl. IV (two skeletons; Upper Peirene, Acrocorinth, “Byzantine”); B. Robinson, pers. comm. (east and north apses, Peirene, Corinthian Forum, Early Byzantine); Pallas 1957, p. 215 (Skoutela basilica, Early Byzantine?); Corinth III.1, p. 25, fig. 27 (summit, Acrocorinth, Early or Middle Byzantine?); Davidson 1937, pp. 229–232 (seven, six, and two skeletons, graves I–III; tower in west circuit, Acrocorinth, Early Byzantine); Moutzali 1985, p. 107 (Liapi church, Argos, Early Byzantine?); “Halieis V,” pp. 300–301, fig. 2 (two skeletons each, graves 1, 2, 4, 11, Early Byzantine); Biers 1971, p. 432; 1973, p. 108, pl. 19:d (Phlius, Palati, Middle Byzantine?); Oikonomou 1989, p. 305 (three crania, τάφος 9; Laliotaiïka, Ano Epidauros, Middle Byzantine?); Miller 1988, p. 3 (eight graves with two or more skeletons; near basilica, Nemea, Late Byzantine). 47. Spouses: Corinth VIII.1, no. 161; Corinth VIII.3, possibly

no. 664. Siblings: Corinth VIII.3, nos. 531, possibly 536; Wiseman 1972, p. 41, no. 32. Parents and children: Corinth VIII.3, nos. 539, 547, possibly 585, 589; IGCB I 42; Wiseman 1972, pp. 41–42, no. 33. 48. Corinth VIII.3, nos. 530, 532, 560, 561, 564, 575, 646, 671; Pallas and Dantis 1979, pp. 64–65, no. 3; Wiseman 1967b, pp. 422–424; 1972, p. 42 no. 35. 49. E.g., Corinth VIII.3, nos. 552, 557, 582. 50. Several tombs have produced evidence for subsequent phases but this secondary usage is not precisely datable in most cases: e.g., Morgan 1936, p. 484, fig. 25 (Kokkinovrysi, Corinth); Shear 1931, pp. 434–436; Charitonides 1968, p. 122 (Cheliotomylos, Corinth); Robinson 1964, pp. 62–63; Fox Leonard 1997, pp. 295–308 (east of Baths of Aphrodite, Corinth); Robinson 1965, p. 77; Fox Leonard 1997, pp. 311–353; Walbank 2005, pp. 255–272 (north of Corinth); Papachristodoulou 1968a, p. 166, pl. 124:γ; ProtonotariouDeïlaki 1969, p. 126; 1970, p. 102, pls. 78:α–79:α; 1974, p. 72, fig. 4 (Kretika, Corinth); Skarmoutsou-Dimitropoulou 1995, pp. 154–155; Skarmoutsou 1997, p. 167 (northeast of Corinth); Morgan 1938, p. 370, fig. 12 (west of Examilia); Rife et al. 2007, pp. 152–153, 163, 172–173, 174–175 (Koutsongila ridge, Kenchreai); Karo 1935, pp. 201–202; Berbati-Limnes, pp. 295, 340, fig. 10 (Panayia area, west of Prosimna); Banaka-Dimaki 1998, p. 388 (cemetery southwest of Argos). The numerous skeletons on the floors of such chambers were not necessarily victims of political or natural catastrophe (e.g., Shear 1931, p. 435) and the Late Roman lamps probably did not belong to graverobbers (e.g.,Wiseman1978, p. 69). 51. These calculations do not include DEC 69-902 and NEG 69-008, in which no substantial human remains were preserved.

172

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

Grave Design and Typology When a site was chosen for a new burial, the design of the interment was determined, the materials were collected, and a grave was made. Residents of the Isthmus only disposed of their dead by inhumation. The interments can be grouped into four types that are well represented across the region: cists covered with slabs and/or rubble, tile, and brick, which are sometimes also lined and paved; cists covered with tiles in a pitched arrangement; burial in a drain; and burial in a small chamber with a shaft cut at an angle into the ground. This classification is based on the selection and arrangement of materials and on the ritual applications of the form.52 The four types differ in their design and use, but all applied the same common materials and required the same basic methods of manual construction. Since Greek authors of the Roman and Byzantine eras were not concerned to describe the exact form of simple “graves” (τάφοι), the written testimony contributes little to the interpretation of these four types.53 Each of these types and details of their designs, however, are represented in the archaeological record of the northeastern Peloponnese. The first type was the commonest at the Isthmus, with 20 examples. These covered, oblong cists were typically dug directly into the soil, but T2 68-003 was cut into conglomerate bedrock and PAL 56-001 was cut into a concrete foundation. In places where ancient ground level can be ascertained, the depth of burial was relatively shallow (ca. 0.50–1.50 m). Most graves at the Isthmus were unmarked at the surface, though the possibility cannot be ruled out that perishable materials, in particular wood, were used. No definite epitaphs have been found either associated with the graves or recycled in buildings at the site.54 This epigraphic poverty distinguishes the burials on the Isthmus from those at Corinth, where numerous Christian epitaphs have been found, mostly written on small plaques that were either attached to a base or erected upright. The only two exceptions among graves of the first type at the Isthmus, both of which are Early Byzantine, have unadorned, irregular stones standing like crude “colonettes” (κιονίσκοι) above and slightly west of the cuttings for the cists (Figs. 2.37, 2.40).55 This basic form of grave varied in shape and size, covering, internal structure, and materials. Most cists were roughly rectangular in form, but many were trapezoidal or rounded at the ends. The interior dimensions depended on the age of the deceased and how many bodies the cist was intended to hold. Covered cists containing single adults measured on average ca. 1.83 m long, 0.70 m wide, and 0.46 m deep, while those containing children were as deep but proportionately shorter and narrower. Thus, the enclosure was typically only big 52. Adams and Adams (1991, esp. pp. 182–193) discuss the formulation of archaeological typologies. O’Shea (1984, pp. 60–61) and Morris (1992, pp. 24–25) stress that, because typologies of graves and funerary artifacts cannot account for all attributes of a data set, they should account for those variables that are most relevant to the objectives of study. Many typologies of graves in Late Antique and Byzantine Greece have adopted formal analysis to explore architectural, settlement, and church history rather than construction technique or ritual function: e.g., Soteriou [1942] 1962, pp. 61–79; Pallas 1951, pp. 167–175; 1956; Corinth XVI, pp. 29–30; Nikolajević 1984, pp. 523–534; Ivison 1996, pp. 114–116; Laskaris 2000, pp. 76, 101–103, 142–143, 164, 262, 291–310. 53. Pallas (1951, pp. 167–175) attempts with little success to identify different forms of Middle to Late Byzantine graves at Ayios Dimitrios near Moulki on Salamis with the terms τάφος, θήκη, χαμοσόριον, καμάρα, μνῆμα, and κιβώριον in Late Antique and Byzantine literary and epigraphic sources. 54. Šašel-Kos (1978, p. 351, no. 9, pl. 4 = ILGR 119) identi-

fied as a possible tombstone one fragmentary Latin inscription found in the Roman Bath in 1972 (IΣ 72-6), but there is nothing decisive about its form or content. The author acknowledges with gratitude the assistance of Dan Geagan† and Matthew Trundle, who have discussed the absence of funerary texts in the epigraphic corpus from the Isthmus. 55. T2 68-002, 68-006. A parallel for these markers can be found in the stone or brick fragment jutting above an Early Roman cinerary urn in the North Cemetery at Corinth (Corinth XIII, p. 198, no. 517, pl. 17). Corinth VIII.3, p. 113, no. 278, pl. 23, records a fragment of a column bearing an epitaph probably of Early Roman date. Pallas and Dantis (1979, pp. 68–69, no. 7) record a “κιονίσκος” from Acrocorinth bearing a Late Roman inscription, but it was probably a dedication, not an epitaph. Small, roughly rectangular, unmarked and unfinished headstones have been found at the west ends of Frankish graves just west of the church southeast of Temple E at Corinth (Williams and Zervos 1992, p. 162; 1995, p. 14).

MORTUAR Y FORM AND VARIABILITY

173

enough to accommodate the corpse, whatever the age at death. Only DEC 69-901 and T2 68-002 were so large that the bodies were surrounded by open space. Covered cists containing more than one corpse were somewhat wider, longer, and deeper than those containing a single corpse, measuring on average ca. 1.95 m long, 0.75 m wide, and 0.70 m deep. The size of the graves for multiple burial depended on the number of bodies they were intended to hold. The two largest graves, HO 70-902 (10 skeletons) and NEG 67-001 (11 skeletons), measured on average ca. 2.13 m long, 0.96 m wide, and 1.03 m deep. Although they were relatively spacious, multiple burials would have become crowded over time, especially before the soft tissues of the corpses had decomposed. In some cases, the dense mound of bodies reached the level of the covering so that the mourners must have had difficulty setting the slabs tightly into place. These cists were always covered, often lined, and rarely paved with intact tiles, slabs, or fragmentary tiles, bricks, and rubble. The composition of the coverings and linings and the degree to which they sealed the cists varied. Several cists were covered with slabs,56 slabs in combination with small stones and/or tile and brick,57 or just tiles.58 The unusually large HO 70-902 had a secondary layer of tiles below the top of the cutting for the cist. While it is uncertain whether T14 69-701 was originally covered, its similarity with contemporary graves around the Fortress suggests that the stones lining its walls probably also supported slabs.59 Three other cists were found without covers, but the original covers had most likely been displaced. Most likely T14 67-004 was originally closed like its neighboring graves, but later builders who came across it removed any covering stones or tiles, along with a few bones.60 NEG 67-001 and 67-003, both of which contained several individuals, were lined but not covered. As has been noted, the coverings were probably discarded to facilitate the addition of bodies on several occasions. The lining and paving of these cists was less regular than the covering. Lining involved either stacking rubble, tile, and/or brick, setting those materials in courses, or placing tiles or stones upright. Several graves were completely or almost completely lined.61 The walls of NEG 67-001 and 69-001 were faced with even brick masonry. This construction technique compares with the walls of several Roman and Byzantine graves at Corinth and Argos.62 T2 68-003 and PAL 56-001 did not require any lining because they were cut into bedrock and concrete, and five other cists did not require lining on all sides because they abutted preexisting walls.63 HO 70-902 was the only interment at the Isthmus subdivided into compartments by a row of upright tiles. This format compares with more elaborate built tombs at Corinth and Argos containing separate chambers for multiple burial.64 The significance 56. NEG 69-005, T2 68-006, 68-002, T14 69-002, 67-002, DEC 69-901. 57. NEG 69-103, 69-004, 69-008, 69-001, 69-007, T2 68-003, T14 69-003, 69-991, PAL 56-001. The cover slabs of NEG 69004 and 69-008 were disturbed by the construction of the Hexamilion. 58. HO 70-902, T13 54-001. The putative grave near Tower 14 was covered by fragmentary tiles. 59. Cf. T2 68-002, 68-006, T14 67-002, 69-991. 60. Similar activity is evident at T14 67-002, but there the builders left one slab over the east end of the grave. 61. NEG 67-001, 69-001, T2 68-001, 68-006, T14 69-701, 69991, HO 70-902. T14 69-701 was partly excavated without documentation but presumably was at one time completely lined. 62. Wiseman 1969, p. 80, fig. 7 (grave 86; Gymnasium area, Late Roman); Shear 1931, pp. 439–440 (Cheliotomylos, Late Roman); Stikas 1964, pp. 133, 134, fig. 4, pls. 88, 91:α (graves in central aisle, τάφος H, and grave of bishop Eustathios; Kodratos basilica, Late Roman); Shelley 1943, pp. 176–177,

figs. 9, 10; Pallas 1974, pp. 213, pls. 201:β, 202:α, β (burial chambers, Kraneion basilica, Late Roman); Konsolaki 1987, p. 71, pl. 21:α, β; Methana, p. 88 (northwest of acropolis, Palaiokastro, Methana, Late Roman); Oikonomou-Laniado 1998, pp. 406, 407, figs. 2, 7; 2003, pp. 28–29, 30–31, figs. 40– 43, 48–50, 56, 57 (North and East Cemeteries [area nos. 587, 623], Argos, Late Roman–Early Byzantine); Vollgraff 1956, p. 95, plan XI; Oikonomou-Laniado 2003, p. 12 (Sanctuary of Athena on the Aspis, Argos, Late Roman–Early Byzantine); Pallas 1978, pp. 184–185, pl. 124:γ (cist in baptistery; Kraneion basilica, Corinth, Early or Middle Byzantine?). 63. NEG 67-001, 67-003, 69-001, 69-005, HO 70-902. 64. E.g., Wiseman 1969, p. 81, fig. 8 (grave 69, Gymnasium area); n. 45, above (burial chambers annexed to Kraneion and Lechaion basilicas); Kritzas 1976, pp. 205–207, fig. 10; Oikonomou-Laniado 1998, pp. 405–407, figs. 1–3, 5, 7; 2003, pp. 28–31, 33, figs. 40–43, 53–57 (North and East Cemeteries [area nos. 587, 620–623], Argos).

174

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

of this division of corpses into a group of two and a group of eight is uncertain, but it may reflect relationships between family members. Five cists were incompletely lined so that one or more wall(s) were left partly or entirely exposed,65 and five cists appear to have been unlined.66 The cists were only rarely paved. In one unusual case, the floor of T2 68-003, which was cut into bedrock, was completely lined with flat paving tiles. Similar pavements have been found in graves of Roman to Early Byzantine date in the Corinthia and Argolid.67 NEG 69-008 might have been partially paved with a layer of irregular stones, but this is difficult to ascertain because the grave had been disturbed by construction. These designs of enclosure indicate that the mourners were more concerned to cover the body from above than to enclose it from below or the side. They tried to place the coverings to seal the graves effectively. In certain cases, slabs were fit tightly together or reshaped to accommodate the cist, and gaps were blocked with irregular stones, tiles, and bricks or were sealed with mortar.68 When slabs were not used, rubble or tiles were spread out to cover the grave. Variation in the coverings and linings led to the differential decay of the bodies by environmental agents, particularly air, moisture, plants, and microbes. Those who made the cists would have understood these natural processes, and, if they had chosen, could have enclosed the cists more securely. The concentration of effort on the top of the graves and the disregard for the walls and floors suggest that the concern to close the interment outweighed the concern to protect and preserve the corpse. The covered cists were constructed from materials that mourners could have easily found on site. They often used slabs of varying thickness and shape to enclose or line cists. Chiselmarks, nail holes, and sockets on rectangular slabs, vestiges of prior use (Figs. 2.4, 2.5, 2.23, 2.24, 2.65), reveal that these stones were not freshly quarried but recycled from other contexts. Sometimes reused slabs were cut down to fit the shape of the cist (Figs. 2.37, 2.55, 2.79, 2.81).69 Besides slabs, irregular stones of all sizes from cobbles to blocks were extracted from fallen buildings or collected from nearby fields for secondary use. Those who built the graves also used roofing and paving tiles to cover and line several cists. HO 70-902 employed several fragmentary Laconian tiles to divide and enclose its interior compartments (Figs. 2.71, 2.72), while the pavement in T2 68-003 comprised square terracotta plaques not unlike those used for pavement in Roman buildings (Fig. 2.44). Bricks were set in level courses and bonded with mortar along the walls of NEG 67-001 and 69-001 (Figs. 2.12, 2.21). While the intact condition of bricks and paving tiles shows that they were not recycled, it is doubtful that they were manufactured expressly for use in burials. A more likely explanation is that they were acquired unused from stockpiles of building supplies or active construction sites. More commonly the cists were enclosed by fragments of tile and brick mixed with rubble, all of which were available among the ruins of the Sanctuary. The materials used to build the cists were economical because they were inexpensive, portable, and available on site in large quantity. Even the grave markers over T2 68-002 and 68-006 lacked ornament (Figs. 2.37, 2.40). But their application was not haphazard, and their form was not unsightly. In recycling materials, those who made the graves were efficient and opportunistic in their management of local resources. Moreover, the careful 65. NEG 67-003, 69-004, 69-005, T14 67-002, DEC 69-901. When the builders of the Hexamilion disturbed NEG 69-004 they dislodged a slab that had lined the cist. The putative grave near Tower 14 seems to have been at least partially lined with stones. 66. NEG 69-103, 69-008, T14 67-004, 69-002, 69-003. Lining elements in NEG 69-103, NEG 69-008, and T14 67-004 were removed during disturbance by subsequent construction. 67. E.g., Onasoglou 1990, p. 91, fig. 4 (Midea, Roman);

Konsolaki 1987, p. 71, pl. 21:β (northwest of acropolis, Palaiokastro, Methana, Late Roman); Banaka-Dimaki 1988b, p. 89, fig. 2 (τάφος XIV; Ayia Anna, Corinth, Late Roman or Early Byzantine); Kritzas 1976, p. 207; Oikonomou-Laniado 1998, p. 405, fig. 2; 2003, pp. 27, 28–29, 30, figs. 40, 42, 49, 55 (North Cemetery [area nos. 180, 587, 622], Argos, Late Roman–Early Byzantine). 68. E.g., T2 68-002, 68-006, PAL 56-001. 69. T2 68-002, 68-006, T14 69-003, PAL 56-001.

MORTUAR Y FORM AND VARIABILITY

175

situation of the slabs over the cists reveals a fundamental concern for visual symmetry (e.g., Figs. 2.37, 2.55). The value of the stone and terracotta that covered, lined, and paved the cists was not great, and its appearance was hardly impressive, but the selection and application of these materials was directed by a sense of purpose and order. This elementary form of interment, the covered cist for single or multiple burials, was virtually ubiquitous in Roman and Byzantine Greece.70 Graves closely resembling those at the Isthmus have been found at numerous urban and rural sites in the northeastern Peloponnese, including the cemeteries north of Corinth, in the Corinthian Forum, around the Corinthian churches,71 at Kenchreai,72 near Zygouries, Klenia, Dragatoura, and Prosimna, at Phlius, Nemea, Argos, Tiryns, Halieis and Ano Epidauros, and on Spetses.73 All instances share the same essential form of a quadrilateral enclosure that is at least partly lined, but they display minor variations in the manner of covering and lining, the dimensions of the cutting, and the choice and application of materials. The most elaborate examples are found in Lerna Hollow, the Asklepieion, and the Gymnasium area and inside or alongside the aisles of the Christian basilicas at Corinth and Lechaion. Many of their walls were constructed from intact bricks, spolia, or large stones and faced with marble revetment or stucco, creating a rectilinear chamber that resembles a sarcophagus. Most interments of this type at the Isthmus, however, do not display such decoration or solidity. The fragmentation and sparsity of their composition and the eccentricity of their design compare most closely with the modest graves of the small rural communities at Nemea, Zygouries, and Halieis during the Late Roman and Early Byzantine periods. The second type of grave represented at the Isthmus, which has two examples, is a cist covered by tiles in a pitched arrangement.74 In DEC 69-902, two long Laconian tiles were placed at an acute angle over the corpse and supported at the base and sides by packed 70. Exact parallels for the form of the Early Byzantine graves at the Isthmus can be found among the interments of the 6th to 7th centuries in Settore L at Gortyn (Di Vita 1988a, pp. 91–139, figs. 66, 86–158). 71. Banaka-Dimaki 1988b, p. 89, fig. 1 (Ayia Anna, Roman); Protonotariou-Deïlaki 1969, p. 125 (Kretika, Roman); Stikas 1964, p. 133, pl. 83; 1966, p. 52, pl. 47:β (Kodratos basilica, Late Roman); Corinth XVIII.3, p. 390 (north slope of Acrocorinth, Late Roman); Corinth XVI, pp. 29–31, 127 (South Basilica, west Forum, Central Shops, Late Roman–Early Byzantine); Robinson 1962, pp. 110–111, pl. 35:c (West Shops, Late Roman–Early Byzantine); Pallas 1959a, pp. 18–19, fig. 12 (site of new Panayia church, Late Roman–Early Byzantine); de Waele 1935, p. 356; Wiseman 1967b, pp. 418, 420, pl. 84:b, e; 1969, p. 79, fig. 7; Wesolowsky 1973, p. 341 (Asklepieion, Lerna Hollow, and Gymnasium area, Late Roman–Early Byzantine); Pallas 1965, pp. 133–134, pl. 118:β (Lechaion basilica, Late Roman–Early Byzantine); Davidson Weinberg 1974, pp. 513–515, pl. 110:b (South Stoa, Corinth, Early Byzantine); B. Robinson, pers. comm. (east and north apses, Peirene fountain, Early Byzantine); Robinson 1976a, pp. 221–222, pl. 48:a (Temple Hill basilica, Early Byzantine?); Blackman 1999, p. 23 (east of Late Roman bath, Panayia area, Early Byzantine); Pallas 1957, p. 215 (Skoutela basilica, Early Byzantine?); Pallas 1961, p. 173, pls. 72:γ–73:α (Lechaion basilica, Early Byzantine), p. 174 (Lechaion basilica, Middle Byzantine?); 1972, p. 112, pl. 156:α; 1974, pp. 219, 234–235, pls. 208:α, 214:β, 215:α; 1978, pp. 184–185, pl. 124:γ (Kraneion basilica, Early or Middle Byzantine?); Corinth III.1, pp. 10–11, 25–28, figs. 9, 27, 28, pl. 2 (Acrocorinth basilica and summit, Early or Middle Byzantine?), pp. 36–37, 40, 42, fig. 36, pl. IV (Upper Peirene, Acrocorinth, “Byzantine”).

72. Kristalli-Votsi 1984, p. 65; unpublished field records at the Ephoreia in Ancient Corinth (west of the harbor); Rife et al. 2007, pp. 147, 154, fig. 3 (Koutsongila ridge, north of the harbor). The author is currently studying the Late Roman cemetery west of the harbor on the former Athanasopoulou property. 73. Kordosis 1997, pp. 488, 492, 542, 553, fig. 6; cf. Wiseman 1978, p. 93 (near Vouno north of Klenia, Roman?); Onasoglou 1990, p. 91, fig. 5; Berbati-Limnes, pp. 293, 326– 328 (Psili Rachi and central valley near Prosimna, Roman); Koilakou 1997, p. 69 (Palaio Limani, Spetses, Late Roman?); Zygouries, pp. 70, 73, fig. 61 (Ampelakia west of Zygouries, Late Roman?); Miller 1977, p. 3, fig. 3, pl. 2:a; 1980, p. 192, fig. 6 (south of Temple, Nemea, Late Roman); Kritzas 1979, pp. 219–220, 245, fig. 16, pl. 167:α; Moutzali 1985, pp. 106– 107, fig. 1; Barakari-Gleni 1988, pp. 119–120; OikonomouLaniado 1998, pp. 406, 407, 408, figs. 3, 5, 7; 2003, pp. 13–14, 15, 16–17, 28–31, 33, figs. 11–13, 40–43, 53–57 (Lymberi, Gargassoula, Kephalari, and Doutsou basilicas, Liapi church, and North and East Cemeteries [area nos. 18, 37, 247, 405, 495, 587, 620, 621, 622, 623], Argos, Late Roman–Early Byzantine); Kourinou-Pikoulas, Pikoulas, and Faklaris 1987–1988, pp. 228, 231 (near Dragatoura, Late Roman–Early Byzantine?); Kilian 1980, p. 286, fig. 4, pl. 49:1 (Tiryns, Early Byzantine); “Halieis V,” pp. 298, 300–301, fig. 2, pl. 80:a (Early Byzantine); Biers 1971, p. 432, 1973, p. 108, fig. 2, pl. 19:d (Palati, Phlius, Middle Byzantine?); Oikonomou 1989, p. 305, pl. 13 (Laliotaiïka, Ano Epidauros, Middle Byzantine?); Miller 1988, p. 3 (near basilica, Nemea, Late Byzantine). 74. HO 70-902, T13 54-001, and the putative grave near Tower 14 were also covered with fragmentary tiles but did not display the distinctive pitched arrangement.

176

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

clay, while at least three more tiles were placed upright at the ends of the tent-like configuration and flat over the top (Figs. 2.62, 2.63). In WF 62-001, four large, square Roman tiles with raised edges were placed at an oblique angle over one side of the corpse and irregular stones closely surrounded the cutting (Figs. 2.74, 2.75). The tiles in both graves were manufactured for roofing but were probably never used for that purpose before they covered the graves. The cuttings for the tile-covered cists were roughly the same shape and size as those for the covered cists but they were shallower, so that the highest point on the tiles would have been very close to or at the surface. In functional terms, burial under pitched tiles is merely a variation on the basic covered cist and offers no better protection. The primary difference is the form of the tile covering, which would require less time and effort to erect than a covered, lined cist. This is a standard ancient form of grave in many regions during several periods. Numerous instances of graves covered by pitched tiles of Roman to Early Byzantine date have been found in the Corinthia and Argolid,75 but the same basic form also occurs in earlier76 and later contexts.77 For this reason, any one example is difficult to date without artifactual or stratigraphic assocations. In some cases, such as WF 62-001, the form and condition of the tiles themselves are general indications that the grave was Roman and not Late Antique or Byzantine. Moreover, a subtle but perceptible shift from Roman to Byzantine times is evident in the arrangement of the tiles covering cists in the cemeteries of Corinth. The pitched arrangement gradually became less regular, while the horizontal capping elements grew more uneven and smaller or were discarded altogether. Moreover, while the blocking elements at the ends are in all periods a basic feature, in late antiquity they were frequently replaced by irregular stones either at the extremities of the cist or extending around its sides.78 It should be stressed that these broad changes are imprecise chronological criteria. The variability of this type during all periods is so great that, for example, the least formal Hellenistic tile-covered graves are not so different than their Early Byzantine counterparts. The form of WF 62-001 is a degeneration of the standard type into a simple row of upright tiles. In contrast, the conventional form of DEC 69-902 compares most closely with other Corinthian examples of Roman date, though its regular design is most typical of the Early Roman period. 75. E.g., Corinth XIII, pp. 75, 297, nos. 511, 512, 513, fig. 6 (North Cemetery, Corinth, Early Roman); Charitonides 1968, p. 121, pl. 120:α (Cheliotomylos, Corinth, Early Roman); Robinson 1966, pp. 179–180, n. 2 (south of harbor at Kenchreai, Early Roman); Drosoyianni 1972, p. 206, pl. 190:α–γ (south of Lechaion, Roman); Protonotariou-Deïlaki 1969, pp. 125, 127; 1970, p. 102, pl. 79:α (Kretika, Corinth, Roman); Skarmoutsou-Dimitropoulou 1995, p. 154, fig. 4 (northeast of Corinth, Early or Late Roman?); Stikas 1964, p. 133, pl. 88 (Kodratos basilica, Corinth, Roman); Walbank 2005, p. 253, fig. 9.3 (north of Corinth, Middle Roman); Corinth XVIII.3, pp. 381–387, pls. 57:c, 58:a, c (Demeter and Kore, Late Roman), p. 390 (north slope of Acrocorinth, Late Roman); de Waele 1933, p. 436; 1935, pp. 356–357; Corinth XIV, pp. 162–163, fig. 32, pl. 67:4; Wiseman 1967a, pp. 32–33, pls. 13:a, 14:a, d; Wesolowsky 1973, p. 341; Sanders 2005b, pp. 430–432, figs. 16.7, 16.8 (Asklepieion, Lerna Hollow, and Gymnasium area, Late Roman–Early Byzantine); B. Robinson, pers. comm. (east apse, Peirene fountain, Early Byzantine); Blackman 1999, p. 23 (east of Late Roman bath, Panayia area, Early Byzantine); Miller 1981, p. 48, pl. 12:b; 1988, p. 3, pl. 4:c (near Temple and near basilica, Nemea, Late Roman); Kritzas 1979, pp. 219–220; Kaza-Papayeoryiou 1988, p. 119; Barakari-Gleni

1988, pp. 119–120; Oikonomou-Laniado 1998, pp. 406–407, 408, figs. 5, 8, 9; 2003, pp. 13–14, 28–30, 31–32, figs. 44, 45, 58 (Lymberi basilica, North Cemetery, and cemetery south of city [area nos. 180, 247, 378, 405, 587, 621, 649], Argos, Late Roman–Early Byzantine). 76. E.g., Corinth XIII, p. 74, nos. 354, 358, 365, pl. 16 (North Cemetery, Corinth, Classical); Pemberton 1985, pp. 272–273, 296–297 (various locales, Corinth, Hellenistic); Bruneau 1970, pp. 500–502, figs. 173, 174 (tombe 188, Quartier Sud, Argos, Hellenistic); Barbouna IV, pp. 28–127 (Barbouna area, Asine, Hellenistic). 77. E.g., Corinth XVI, pp. 29, 127–128 (Forum, Byzantine); Shelley 1943, p. 183, n. 19; Pallas 1972, p. 113, pl. 156:β; 1974, p. 234, pl. 214:α; 1978, p. 185; 1980, pp. 168– 170, pl. 106:β, γ (Kraneion basilica, Corinth, Early or Middle Byzantine?); Anderson 1967, p. 7 (northwest of Temple E, Corinth, Middle–Late Byzantine); Williams et al. 1998, p. 240 (west of church southeast of Temple E, Corinth, Frankish); Biers 1971, p. 432; 1973, p. 108 (Palati, Phlius, Middle Byzantine?). 78. The Early Byzantine graves recently found east of the Late Roman bath in the Panayia area at Corinth exemplify the development toward irregularity (Blackman 1999, p. 23).

MORTUAR Y FORM AND VARIABILITY

177

The third type of grave, which has only one example, is burial in a drain. Local residents transformed the ruptured water channel beneath the floor of the Bath into a burial chamber (76-002, Fig. 2.66). The width and height of the channel were large enough for bodies and the floor, walls, and ceiling were already lined with solid stone. The opening through which the bodies were inserted was easily sealed by means of a large slab and smaller stone and tile fragments, not unlike the coverings over common cists. The conversion of the abandoned drain into a grave reflects the adaptability of the local community to the changing conditions of its landscape. The reuse of defunct drains for burial was not uncommon during late antiquity and the Byzantine era. Skeletons have been found in drains and a balneum in the ruins of the Corinthian Forum. During the Roman era a cistern north of the city was transformed into a chamber tomb with the addition of arcosolia and niches in its walls, and in late antiquity the reservoirs on the south side of Lerna Hollow were converted into burial sites.79 The fourth type of grave, which has three examples, was a shaft cut at an angle into the ground and enclosed to create an inner compartment. These interments were found in the marl exposure that forms the north wall of the taphros outside the Northeast Gate (69-007, 69-009, 69-010, Figs. 2.22–2.35).80 The two graves containing infants (69-009, 69-010) were covered with solid blocks consisting of coarse rubble and a thin coat of white plaster. The western end of the top surfaces of these markers displayed raised panels, on one of which (69-010) a simple cross had been inscribed by finger (Figs. 2.31–2.33). These two markers were unique at the site and differ from the unadorned stones erected over the slab-covered graves near Tower 2 (68-002, 69-006). Below the surface, the three graves were excavated from the soft marl to a total depth of ca. 0.50–1.50 m, like other types of cists. At a point below the top of the cutting, the cist extended further back into the slope to a more confined interior chamber for the corpse. Large stones and tiles were placed on an artificial ledge and set obliquely across the cutting to shut off the shaft. These covering elements, both limestone and marble, had been recycled from elsewhere but were not used indiscriminately. The stones in 69-007 were carefully situated to cover the cist securely. The tiles in 69-010 were arranged in a symmetrical pattern by size and shape, so that the largest tile covered the extended torso of the infant and a triangular fragment was set directly over the head (Figs. 2.30, 2.34). In contrast, the tiles covering the young child in the smaller cist (69-009) were piled over the corpse in a less orderly manner. The interior compartment was left unlined. The shape and dimensions of the burial chambers were slightly different than those of the covered cists. The outline of the inner chamber was more rounded at the ends so that it fit more closely the form of a supine corpse. The length and width of these chambers were comparable to those of other graves containing adults and infants, but the depth was considerably shallower. The covering of the interior chambers would have been placed very closely over the corpses, perhaps even compressing them slightly. Graves of this type achieve the same basic goal of enclosing the dead as do more common cists of the first two types but are distinguished by their elaborate design. The fundamental 79. Papachristodoulou 1968a, p. 168, figs. 4, 5, pl. 125:ζ (Roman burial compartments in cistern in Kretika); Corinth I.2, pp. 55–57, n. 3, pl. II (five or six bodies placed in drain running under east foundations of Captives facade during Late Roman period); Ivison 1996, p. 114; cf. Corinth I.4, pp. 145– 151, plan V; Corinth XVI, pp. 23–24, 30–31 (tomb with eight bodies in drain behind shop 28, South Stoa; multiple burial in another drain; burial in the small Late Roman bath behind shops 25–26, South Stoa); de Waele 1935, p. 357; Corinth XIV, p. 164, plan A (Asklepieion and Lerna Hollow, Late Roman–

Early Byzantine); B. Robinson, pers. comm. (dense human bones found in 1911 in the West Supply Channel in the Peirene fountain, Early or Middle Byzantine deposits). 80. HO 70-902 does not represent this type of grave though it appears to in Figs. 2.71 and 2.72, which show the excavation of the cist in progress. The excavators did not remove all the burial fill systematically but tunneled downward in the west side of the cist until they found skeletons, at which point they dug horizontally to the east.

178

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

difference between the three types is the investment of effort and concern for appearance. The materials applied to all forms were equally inexpensive and mundane, but the excavation of a cist at an angle into a marl slope required more energy than digging a hole into level ground. The construction of separate layers of coverings also reflects a greater expenditure of thought, effort, and time than the creation of unmarked cists with simple slab, rubble, or tile coverings. The use of large gravemarkers with simple decorative features and a finish of white plaster points to a heightened interest in the visibility and appearance of the burial site. This design of grave was not uncommon in the northeastern Peloponnese during the Late Roman and Early Byzantine periods. The best comparanda for the examples at the Isthmus have been found in isolated locales or near farmsteads or small towns in the eastern Corinthia, namely, Perdikaria, Kenchreai, Solygeia, and Zygouries. In these instances, the shaft was cut at an angle into an escarpment, a steep slope, or a ledge of rock or soil down into a rectangular or elliptical burial chamber sealed with slabs, rubble, or tiles. In some cases, a rough shaft was cut straight down into the ground, and a compartment was excavated off to one side and sealed.81 The basic form of NEG 69-009 and 69-010, particularly the shaft leading down to a conscientiously arranged layer of tiles over the body, also resembles a child’s grave dug during the Early Byzantine period in the buildings immediately south of the atrium of the great basilica at Lechaion.82 Various graves similar to those at the Northeast Gate have also been found in the Late Antique cemetery at the Asklepieion, Lerna Hollow, and the Gymnasium area at Corinth. Several graves there consisted of a burial shaft, sometimes lined, extending down vertically or diagonally ca. 1.00 m from level or sloping ground to an interior chamber usually enclosed by tiles.83 Other graves had a related design involving a vertical shaft cut straight down into bedrock 0.50–1.00 m, at which point it opened into the east end (the foot) of a rectangular chamber, often displaying a slightly vaulted ceiling. In the most carefully executed versions of this form, the descent has one or more steps, and the shaft and chamber are evenly cut or constructed.84 Both of these variants of the same basic form were often marked at the surface by solid, rectangular or hemicylindrical mounds covering the area of the grave, as in NEG 69-009 and 69-010. These markers consisted of mortar, rubble, and sparse brick and tile coated with white plaster.85 Two particular features of the markers over NEG 69-009 and 69-010, the raised, square panel on the west end of their top surfaces and the inscribed cross, are also found in the Corinthian cemetery. One grave marker had a square pedestal at

81. Verdelis 1965, pp. 137–138; Lorantou-Papantoniou 1999, p. 53, pl. 7:β, γ (τάφοι Ε-Ζ, Κ-Λ, Solygeia, southwest of Galataki, Late Roman–Early Byzantine?); Zygouries, pp. 70, 72–74, figs. 57, 58, 60, 65 (tombs VIII, XVa, Ampelakia west of Zygouries, Late Roman?). Another unpublished example currently under study by the author has been found on the Koutsongila ridge north of the harbor at Kenchreai (G15; see Rife et al. 2007, p. 147, 154, n. 32, fig. 3). The author has discovered and recorded four graves of this type, all disturbed, on the central Isthmus during research in 2000–2001 with the Eastern Korinthia Archaeological Survey. One (ML 23) containing a fragmentary stone enclosure was cut into a bluff on the southeast side of the Ayios Dimitrios ridge near Philotheis. Three similar interments were found on hills flanking the natural basin known as Perdikaria ca. 0.5 km east of Examilia. One (ML 10.1) was cut into a low rise on the southeastern side of Rachi Simitra. Two others were found to the south on Rachi Boska, one with a side chamber sealed off with rubble (ML 11) cut directly into the ground and another with closure slabs (ML 15) cut into the northwest face of the ridge.

82. Pallas 1967, pp. 146, 148–149, fig. 1, pl. 196. 83. De Waele 1935, p. 356; Corinth XIV, p. 163; Wiseman 1967a, pp. 32–33, 34–35, pl. 14:c, d; 1969, p. 81. Other graves of comparable form were tunneled into vertical rock exposures in Lerna Hollow: de Waele 1935, pp. 356–357; Corinth XIV, p. 162, pl. 66:2. 84. De Waele 1933, p. 436, pl. L:1; Corinth XIV, p. 162, plan A, pl. 15:5; Wiseman 1967a, pp. 33–34, fig. 18, pl. 14:b; 1967b, pp. 419–420, pl. 84:d; 1969, pp. 81–85, figs. 8, 9, pls. 26:a, 27:a, b; 1972, pp. 8–9; Wesolowsky 1973, p. 341; Sanders 2005b, pp. 432, 434, 435, figs. 16.7, 16.8. Graves with the same basic form but rougher finishing have been found on the Koutsongila ridge north of the harbor at Kenchreai (Rife et al. 2007, pp. 147, 154, figs. 3, 7). 85. De Waele 1935, p. 356; Corinth XIV, p. 163; Wiseman 1967a, pp. 32–33, 35, fig. 17, pls. 13:c, 14:f; 1969, pp. 81–83, 85, figs. 9, 11, pls. 25:f, 26:a, 27:b, e; 1972, pp. 8–9; Wesolowsky 1973, p. 341. Laskaris (2000, pp. 307–308) discusses the form and its comparanda.

MORTUAR Y FORM AND VARIABILITY

179

the west end,86 other markers had inscribed plaques affixed to them, typically at one end,87 and three graves displayed crosses painted or inscribed on the plaster either on the interior walls of the chamber or on the exterior surface of the marker.88 The origin and purpose of this form of marker at Corinth and the Isthmus is unknown. The hemicylindrical mounds over the Corinthian graves may echo the vaults that typically covered the more monumental variety of Late Antique chamber tomb.89 This explanation cannot, however, account for the rectangular markers with square panels over the graves at the Isthmus. The panels may imitate plaques for epitaphs; they may also form a table for the placement of lamps or other vessels after burial.90 Another possibility is that the marker is a schematic representation of either a bed, whether a generic bed or the deathbed, or the bier itself with a low headrest. Markers of this form could have served both a practical and an iconographic purpose. Their effectiveness is confirmed by their survival until today: the form seems to be the model for the standard burial monument in Modern Greek cemeteries, including Ayios Ioannis at Kyras Vrysi (Figs. 3.2, 3.3, 3.5). More elaborate versions of this same essential form of grave were made during the main phase of use in the first Christian churches. The tomb of the bishop Eustathios in the Kodratos basilica, for example, consisted of slabs at floor level inscribed with the name of the deceased over the west end. These covered a shaft cut ca. 0.80 m down to an interior chamber for the corpse, which was sealed with more slabs. The burial annexes off the south aisle of the Kraneion basilica contained some graves with chambers extending back from an opening at the east end and others that were marked at the pavement with rectangular bases finished with marble revetment.91 In sum, this grave type, which was strictly a Late Antique development,92 displays several variants within the region. All forms, however, have a well-defined inner chamber which is entered from a vertical or diagonal shaft, and many have a marker or slab at their surface. The examples from the Early Byzantine settlement at the Isthmus most closely resemble examples from other rural sites in the Corinthia, while their markers compare best with those over graves in the large cemetery north of Corinth. In general, contemporary graves of this type in the periurban cemeteries and Christian basilicas of Corinth possess the same basic format as the rural examples, but they display greater regularity, decoration, and permanency. 86. Wiseman 1972, p. 8 (grave 98). 87. IGCB I 32; Corinth XIV, p. 165 (epitaphs on plaques set into markers); Wiseman 1969, pp. 81, 93, pls. 25:f, 30:e (epitaph on plaque set into marker, grave 80). Shear (1931, pp. 439–440, fig. 13) records an epitaph on a plaque cemented to one end of the cover of a Late Roman grave on Cheliotomylos. Late Roman graves in or near Christian basilicas also had epitaphs inscribed directly on covering slabs or on plaques at the west end of the interment over the head of the corpse: Stikas 1964, p. 133, fig. 4; 1966, pp. 52–53, pl. 48:α (τάφοι E, IB, Kodratos basilica); Pallas and Dantis 1979, pp. 64–65, no. 3 (north of Kraneion basilica). 88. Wiseman 1967b, p. 419, fig. 7, pl. 84:c (cross incised in plaster near west end of south wall of cist, grave 35); 1969, p. 85, fig. 11 (cross inscribed on west end of marker, grave 76); 1972, p. 8 (cross painted on south wall of shaft, grave 94). Crosses and other decorations were also painted on the interior walls of the Early Byzantine chambers annexed to the east end of the south aisle of the Lechaion basilica (Pallas 1961, p. 173, pls. 72:γ–73:α). One close parallel is to be found among the interments beneath Ayios Dionysios at the Areopagus in Athens. A cross was inscribed in the plaster facing

the interior west wall of a grave comparable in form to those at the Asklepieion, Lerna Hollow, and the Gymnasium area (n. 84, above) that is dated to the “Early Christian or transitional period” (Travlos and Frantz 1965, p. 183, pl. 49:b, c). 89. See n. 44, above. 90. Cf. Wiseman 1972, p. 8, on the square pedestal at the west end of the marker over grave 98 in the Gymnasium area. 91. Kodratos basilica: Stikas 1964, pp. 133–134, fig. 4 (τάφος Ε). Kraneion basilica: Shelley 1943, p. 183, n. 19, pl. XII (graves with entrance at east end, chambers C and D); Pallas 1959b, p. 205, fig. 19 (graves with markers, mausolea E, F, G). Photographs and plans from the excavation of the Bema church in 1936 show an otherwise unpublished Middle Byzantine(?) tomb in the north aisle that appears to have a design similar to the marked interments in the Gymnasium area and at the Kraneion basilica (Morgan 1936, n.p., figs. 8, 9; Corinth I.3, pl. 38:1; Corinth XVI, p. 44, fig. 3, pls. 5, 6:2). 92. The origins of the form are obscure, though certain examples resemble arcosolia or loculi from standard chamber tombs in the Roman Corinthia. The format of an interior compartment entered through a shaft or opening may be viewed as a miniature version of a tomb with a dromos.

180

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

Apart from these four types of graves at the Isthmus, four other types common in the region during the Roman and Byzantine eras are not represented at the site. The first is burial in a simple, “pit-like” (λακκοειδής) cutting in the ground without lining or covering.93 Three simple cists without lining or covering have been found at the Isthmus, but two or three of them contained secondary burials.94 The second type is burial in vessels (ἐγχυτρισμοί), most often amphoras, which were split either lengthwise or widthwise for the interment of children. This practice was common during the Roman era95 and persisted in later centuries.96 The third type is the sarcophagus, with or without sculpture, which was most commonly used during the Roman era.97 The fourth type is the chamber tomb, whose form, as has been noted, continued to evolve from Roman to Byzantine times. Among the four types of graves that were used at the Isthmus, the sex of the deceased did not influence the choice of type, because both males and females were interred in each type, but the age of the deceased did. The interments of young children are generally better sealed than the interments of adults. The three Early Byzantine graves containing infants were securely covered, as were the Roman and Late Antique graves that probably also belonged to juveniles.98 This pattern reflects a prevailing concern not simply to enclose but also to protect the graves of dead children. The preferred manner of primary burial changed over time. The basic form of the oblong cist persisted during the Roman and Byzantine eras, but the shape, size, and form changed significantly. The middle centuries of the Roman era are only represented by four graves (phase I), but this was the only period when graves covered by pitched tiles were used. These cists were only intended to hold one or two bodies. In contrast, except for one burial in a drain, all graves during the Late Roman period (phases II–III) were covered cists. They contained multiple bodies, so that the space for burial became exceedingly crowded (0.23 m3 average volume per individual).99 Moreover, apart from the unusual PAL 56-001, the Late Roman interments were less consistently sealed than earlier or later interments. The graves of the Early Byzantine period (phase IV) diverged in form from those in previous centuries. The covered cist remained the commonest type, but the Early Byzantine variety was better sealed and more often lined than its Late Roman predecessors. Shaft burials with inner

93. E.g., Skarmoutsou-Dimitropoulou 1995, p. 154, fig. 4 (northeast of Corinth, Early or Late Roman); Zygouries, pp. 62, 70–72, 73, figs. 62–64 (tombs XIII, XIV, XVII; Ampelakia west of Zygouries, Late Roman?); Broneer 1926, pp. 52–53, pl. II (behind Hemicycle, northeast Forum, Corinth, Late Roman– Early Byzantine); Corinth XIV, p. 163 (Lerna Hollow, Corinth, Late Roman–Early Byzantine); Corinth XVIII.3, pp. 382, 386– 388, fig. 53, pls. 58:e, 59:a–d (nos. 22, 26–29; Demeter and Kore, Late Roman–Late Byzantine?); Oikonomou-Laniado 1998, pp. 406–408, figs. 5, 8; 2003, pp. 16–17, 29, 31–32, fig. 13 (“tombe à fosse;” Doutsou basilica, North Cemetery, and cemetery south of city [area nos. 495, 621, 649], Argos, Late Roman–Early Byzantine); Corinth XVI, p. 127 (Forum, Byzantine); “Halieis V,” pp. 298, 300 (grave 1; Early Byzantine); Pallas 1972, p. 112 (Kraneion basilica, Corinth, Early or Middle Byzantine?). 94. HO 70-901, LOU 69-801, TC 60-001. 95. E.g., Williams and Zervos 1986, pp. 137–138, fig. 1, no. 8, pl. 30 (building 3, east of Theater); Papachristodoulou 1968a, p. 169; Protonotariou-Deïlaki 1969, pp. 125, 127; 1970, p. 102, pl. 78:γ, δ (Kretika, Corinth); Skarmoutsou-Dimitropoulou 1995, p. 154, fig. 4 (T13, northeast of Corinth); Drosoyianni 1972, p. 206, pl. 190:γ (Lechaion). 96. De Waele 1935, p. 357; Corinth XIV, pp. 163–164,

pl. 67:4; Wiseman 1967a, pp. 32, 34–35, pl. 14:c; 1967b, p. 418; Sanders 2005b, pp. 430–432, figs. 16.7, 16.8 (Asklepieion, Lerna Hollow, and Gymnasium area, Late Roman–Early Byzantine); Oikonomou-Laniado 1998, pp. 407–408; 2003, pp. 31, 47, nos. 4, 10, 11, figs. 62, 68, 69 (East Cemetery [area no. 623], Argos, Late Roman); Blackman 2000, p. 25; Sanders 2005b, pp. 427–428, fig. 16.5 (Panayia area, Corinth, Early Byzantine); “Halieis V,” p. 300 (grave 5; Early Byzantine). Pallas (1978, p. 185) records a “πιθοειδὴς τάφος” in the Kraneion basilica (Early or Middle Byzantine?). 97. Rife (1999, pp. 229–232) surveys the use of sarcophagi in the Roman Corinthia; see also Walbank 2005, pp. 252–253, fig. 9.2:a, b. A rare example of a Late Roman(?) sarcophagus on a small scale with sculpted decoration on the gabled lid (floral motifs and large crosses) was found in a room directly above the Temple of Artemis at Sicyon (Kristalli-Votsi 1985, p. 62, fig. 95; 1988, p. 242). 98. NEG 69-008, 69-009, 69-010, T14 69-991, DEC 69-902. The putative grave near Tower 14, which was covered with tiles and lined with stones, has the appropriate dimensions for a child’s grave. 99. This calculation does not include the three individuals in the secondary burial HO 70-901.

MORTUAR Y FORM AND VARIABILITY

181

chambers were only used during this period. In all but one case, a double interment (NEG 69-007), these graves contained one individual. Thus, compared to earlier interments, the Early Byzantine graves allowed on average more space per body within the burial enclosure (0.61 m3 average volume per adult; 0.26 m3 average volume per subadult). In addition, graves began to display simple decoration and monumentality, namely, the two block markers coated with plaster, one of which was inscribed, and the two upright stone markers.

Burial Orientation When mourners constructed a grave and deposited a body, they determined its orientation. The orientation of an oblong cist reflects the position of the body because the vertebral axis of an extended corpse always followed the lengthwise axis of the grave. The alignment of the corpse within the cist was highly consistent in single and double interments; even in multiple burials many bodies were interred in parallel (e.g., Figs. 2.71, 2.72). It was standard practice during the Roman and Byzantine eras at the Isthmus to inter the corpse with the head approximately to the west. The body was first oriented in this manner at home for the laying-out and was kept in that position during the graveside ritual and deposition. The heads of the bodies interred at the Isthmus were not pointed exactly in the cardinal direction of west (270°), but most were well within the western quadrant. Of the 28 graves with recorded orientations, the orientation of 22 fell within the range 243°–293° (Table 4.1).100 Of the six outliers, three were definite or probable secondary burials, two with the heads approximately to the south (HO 70-901, TC 60-001) and one with the head approximately to the north (LOU 69-801). The unusual orientation of the secondary burials will be discussed below. The orientation of primary burials did not vary significantly by the sex or age of the deceased, but it did change over time (Fig. 4.1). The 14 interments of the Roman era (phases I–III) show a distinctly wider deviation in their western orientation (217°–333°) than the 11 interments of the Early Byzantine period (phase IV), which cluster more tightly around true west (245°–277°). The orientations of the earlier graves are eccentric because their situation was determined either by the plan of preexisting structures or by an imprecise estimation of the direction of west–east. Of the six primary burials whose orientations were furthest from due west, three (NEG 69-004, 69-103, 69-008) were situated south of the Arch before it was converted into the Northeast Gate and three (T2 68-003, DEC 69-901, 69-902) were located in the open area that would later be occupied by the northeast corner of the Fortress. The interments near the Arch were aligned roughly parallel with the road (Fig. 2.1). Likewise, the grave at Tower 2 was cut parallel to the edge of the outcropping of bedrock, a geological feature that would later determine the line of the Hexamilion and the enceinte’s east wall. The Decauville graves were not close to artificial or natural landmarks. Those responsible for aligning them could have used several vague points of reference for an eastern orientation: the direction of the road that ran nearby to the north, the location of the Saronic Gulf, or even the position of the sun. The orientations of the other Late Roman graves were all dictated by their relationship to prior buildings: PAL 56-001 in the foundations of the Temple of Palaimon, NEG 67-001, 67-003, 69-001, 69-005, and HO 70-902 along the inner face of the Hexamilion, and RB 76-002 inside the drain. The interments of the Early Byzantine period more closely approximated true west than those in earlier centuries, probably because mourners were more concerned to follow a 100. This does not include T13 54-001 (oriented ca. 260°) or the putative grave near Tower 14.

182

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

Figure 4.1. Distribution of the orientations of 25 burials by period

consistent orientation than their predecessors. They had three possible points of reference for situating a grave. First, mourners could use nearby features as a general guide for alignment: the graves outside the Northeast Gate followed the northern wall of the taphros, the graves near Tower 2 were roughly parallel with the east wall of the Fortress, and the graves near Tower 14 were perpendicular to the west wall of the Fortress. Graves of this period, however, were mostly cut in open areas and not alongside standing architecture as in previous centuries. The interments at Tower 2 even deviated from the line of the Fortress wall to follow more closely a west–east axis. Thus, mourners might have considered landmarks as they oriented graves but did not exclusively depend on them. Second, mourners could rely on the orientation of preexisting graves that were marked at the surface. This would explain the comparable orientations of the four marked graves at the Northeast Gate and Tower 2. It would also support the hypothesis that the six graves at Tower 14, which exhibit strikingly regular alignments, were originally marked. As a result, the plans of the three Early Byzantine areas of burial have an orderly, deliberate appearance. Third, mourners could look to a geographic or celestial body in the east, such as the Aegean coast or the sun as it rose. To do so, those digging the cists at Towers 2 and 14 must have mounted the Fortress walls for an unobstructed view, and those responsible for the interments at the Northeast Gate must have climbed out of the partly filled taphros to higher ground. It should, however, be noted that there is no evidence that graves were dug or funerals held only at daybreak; by midmorning, the position of the sun far above the horizon would have been an imprecise point of reference for the horizontal alignment of a corpse. Whatever their particular point of reference, Early Byzantine residents were thinking of the direction east at the time of burial and attempted to align bodies consistently eastward. Their approximations of the cardinal direction were not perfectly accurate101 but were much more so than their predecessors’. This diachronic pattern in the orientation of graves at the Isthmus echoes a regional trend. During the first centuries of Roman Corinth many burials were oriented with the

101. Variation in corporeal alignment at the Isthmus cannot be attributed to dependence on solar orientation and death during different seasons, as the sun’s path across the sky changes with the seasons. This controversial explanation has been proposed for Late Roman and Early Medieval British cemeteries (e.g., Rahtz 1978; Russell 1985, pp. 112–114, table 31; Rahtz, Hirst, and Wright 2000, pp. 113–121) and applied to Late Byzantine or Ottoman graves south and east of the

church southeast of Temple E at Corinth (Williams and Zervos 1991, p. 39). Critics of the theory (e.g., Kendall 1982; Brown 1983; Boddington 1987, pp. 417–418) have shown that intrasite variability in burial orientation results from alignment with fixed local landmarks and from errors in estimating the direction of east. Both factors seem to have influenced burial orientation at the Isthmus.

MORTUAR Y FORM AND VARIABILITY

183

head to the west, but several were also oriented either to the south or the north, such as the Early Roman sarcophagi at the North Cemetery and in the Gymnasium area, whose orientation appears arbitrary.102 The frequent burials in chamber tombs were aligned according to the building’s plan, particularly the arrangement of the walls. Burial orientation, in other words, was inconsistent during the Early to Middle Roman periods because it was dictated chiefly by local features, not an overarching belief or pervasive custom.103 However, by the Late Roman period, the westward placement of the head of the corpse became so common that there is no need to cite parallels. Instances of primary interments during this era in which the heads were not to the west are very rare.104

Corporeal Position Once the grave was constructed, the funeral completed, and the corpse interred with the head at the west end, the mourners arranged the head and limbs of the corpse, sometimes to create a particular appearance and sometimes carelessly. The literary sources copiously attest to the importance of the appearance of the deceased and how this appearance was to be created, concerns that continue to influence modern practice.105 Residents of the Isthmus during the Roman and Byzantine eras always interred their dead in primary burials in a supine and extended position, as though asleep. The mourners would have arranged the corpse in this position at home after death and maintained it as they carried the body to the site of burial, lowering it into the cist by hand, in the coffin, or on the bier, as observed at the Palaimonion. Those interring the corpse usually aligned the vertical median axis of the torso and appendages with the lengthwise axis of the cist. According to Late Roman and Byzantine writers, when mourners laid out the corpse at home they stretched out the legs and tied the limbs in place using a winding sheet or cords.106 The bodies interred at the Isthmus, however, were not always perfectly straight. Sometimes the upper body was offset slightly to the right or left,107 the feet were pulled to the right or the left,108 or the legs were pushed closely together at the knees or ankles.109 These variations arose from different circumstances. During the laying-out, the muscles of the corpse began to rigidify, affixing the position of the limbs and neck in irregular poses if they had not been perfectly aligned by mourners. Those handling the body during its transportation to the grave and deposition often pulled the limbs to one side or the other as they lifted and lowered it. The flexed, everted position of the feet in a few skeletons (Figs. 2.41, 2.46, 2.47) was set when someone carrying the corpse wrapped their fingers and palms over the top of the foot and anchored their thumbs under 102. Corinth XIII, pp. 70–71, plan 1 (North Cemetery); Wiseman 1969, p. 87, pl. 27:c (Gymnasium area). A similar pattern of inconsistency is seen among the well-documented burials of earlier date at, for example, Argos (Bruneau 1970, pp. 524–525) and Olynthus (Olynthus XI, pp. 140–143). 103. Rife 1999, pp. 222–224. 104. Three graves at the Sanctuary of Demeter and Kore on Acrocorinth contained skeletons with heads to the south (nos. 9, 15, 20; Corinth XVIII.3, pp. 382, 284, 285, 286, plan 6). The two corpses in the unlined, uncovered grave behind the Hemicycle in the northeast Forum were arranged with heads to the south (Broneer 1926, pp. 52–53, pl. II). Sparse cranial remains were discovered in the eastern end of an unlined, uncovered cist at Ampelakia west of Zygouries, but it is uncertain whether this burial was disturbed (tomb XIII; Zygouries, p. 71, fig. 62). One grave in the North Cemetery at Argos (area no. 621) contained skeletons with heads to the south (Oikonomou-Laniado 1998, p. 406, fig. 5; 2003, p. 29). 105. Early to Middle Roman testimony: Rife 1999,

pp. 70–71. Late Roman and Byzantine testimony: Loukatos 1940, pp. 46–54; Spyridakis 1950, pp. 102–105, 111–112; Koukoules 1951, pp. 153–154, 156–158; Kyriakakis 1974, pp. 40–42, 55–58. Modern practice: Polites [1931] 1975, pp. 329–330; Megas 1940, pp. 171, 174; Papacharalampous 1965, pp. 146– 148, 152; Kenna 1991, pp. 102–103 (esp. evocative). 106. Legs stretched out: e.g., [John Chrys.] Hom. I in Iob 2 (PG 56.567); Geront. V. s. Melan. Iun. 68 (d. 439, Jerusalem; vita written mid-5th century); Theost. V. s. Nicet. heg. Med. 3.23, 7.49 (AASS Apr. 1.259B, 265B; d. 824 near Trigleia, Bithynia; vita written early 9th century); V. s. Nich. Stud., PG 105.921 (d. 868, Stoudios Monastery, Constantinople; vita written early to mid-10th century). Binding: e.g., John Chrys. Hom. XXVII in Mt. 4 (PG 57.348–349, 350); [John Chrys.] Hom. III in praed. jej., PG 60.715), Hom. suppl. V, PG 64.457. 107. NEG 69-005B, T2 68-002A, HO 70-902 (two bodies). 108. NEG 69-004B, 69-005B, DEC 69-901A, B. 109. T2 68-003A, T14 67-002A, RB 76-002C.

184

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

the arch. Sometimes the handlers needed to alter the position of the body at the last moment to accommodate the space available in the grave. If the cist was particularly narrow, as it often was, the legs may be pulled close together (e.g., Figs. 2.46, 2.47); if the grave already contained one or more bodies, a leg may be bent to avoid treading on another corpse (e.g., Fig. 2.66). Minor adjustments of this sort were practical decisions that were probably not particularly meaningful or symbolic. A peaceful and restful attitude was created by the situation of the head, which was the focal point for the viewing and lamentation of the dead. According to early Christian writers, mourners gave the head a natural, decorous appearance after death by straightening it, closing the eyes and mouth, and binding the jaws shut with a cord under the chin.110 The heads of many corpses interred on the Isthmus faced forward, as they had during the laying-out and funeral. But it was not unusual for the head to settle down against the chest or to tilt slightly to one side as the body was handled.111 In some cases the head turned partly or completely to one side, probably when the corpse came to rest in the grave.112 Some mourners were so concerned to support the head and to keep it pointed forward in perpetuity that they placed stones or tiles around it when they deposited the body. They knew that the skull would shift over time and the temporomandibular ligaments would eventually disintegrate, causing the mandible to drop open and even fall from the cranium, as happened in some cases (Figs. 2.20, 2.56). In four graves the heads of the bodies were propped up. A square tile and a terracotta slab were placed as pillows directly under the heads of NEG 69-010A and T14 67-002A (Figs. 2.30, 2.35, 2.46–2.48). A stone and a tile were placed over the chests and under the chins of T14 69-002A and 69-003A (Figs. 2.54, 2.56). The three bodies from the area of Tower 14 also had thin slabs or stones placed against either side of the head. The most common variation in the usual arrangement of the corpses was the placement of the arms (Table 4.2). The natural position for the upper limbs was alongside the body, and the arms of several bodies were either straight at the sides or slightly beneath the back.113 Another natural position for the arms was over the pelvis but not crossing, so that the hands rested on the thighs. A few skeletons at the Isthmus displayed this placement.114 In one unusual case (NEG 69-005B) the left arm was flexed at an angle with the hand bent back sharply. This clumsy misalignment probably resulted from careless treatment during burial because the legs and right arm were otherwise straight. Apart from these natural or unusual positions, the mourners often arranged the arms artificially so that they crossed over the pelvis,115 crossed over the abdomen,116 or crossed over the chest,117 though in some cases the left and right arms did not cover the same parts of the body and did not even touch.118 Early Christian and Byzantine writ-

110. E.g., Euseb. Hist. eccl. 7.22.9, citing Dionysius (Alexandria, late 3rd century); Greg. Nyss. V. s. Macr., PG 46.985 (west of Neocaesarea, Pontus, late 4th century); John Chrys. Hom. XXI ad pop. Antioch. 1 (PG 49.212), Stag. 2.10 (PG 47.465; late 4th century); [John Chrys.] Hom. I in Iob 2 (PG 56.567); Geront. V. s. Melan. Iun. 68 (Jerusalem, mid-5th century); V. s. Nich. Sion. 78.22–24 (Myra, Lycia, mid-6th century); see also Luc. Luct. 19 (mid-2nd century). 111. NEG 69-103A, B, 69-007A, B, T14 67-004A, 69-002A, 69-003A, 69-991A, DEC 69-901A, B. 112. Head partly turned to right: NEG 69-004B, 67-003B, T2 68-006A. Head partly turned to left: NEG 69-001D, HO 70902 (at least three). Head turned on right side: NEG 67-001 (at least two), 69-005A, B. Head turned on left side: NEG 67001 (at least six), 69-005C, T2 68-002A, 68-003A, T13 54-001,

RB 76-002A, HO 70-902 (at least one). The cranium of T14 67-002A faced left when the grave was excavated (Figs. 2.46, 2.47), but it had been removed from the cist and repositioned unnaturally by Byzantine builders. 113. NEG 67-001 (at least three), 69-004B, DEC 69-901A, B, HO 70-902 (at least three). The upper arm bones of PAL 56-001A were straight at the sides, but the lower arm bones were not preserved. 114. NEG 69-103A, B, 69-001D. 115. NEG 69-007A, B, T2 68-003A, T13 54-001, RB 76-002A. 116. T14 67-002A, 69-002A. 117. T2 68-002A, T14 69-991A. 118. T2 68-006A (arms over chest but not crossing), T14 67004A (left arm over abdomen, right arm over thigh), 69-003A (left arm over abdomen, right arm over chest).

MORTUAR Y FORM AND VARIABILITY

185

TABLE 4.2. DISTRIBUTION OF THE POSITION OF THE ARMS IN 22 BODIES BY PERIOD Roman Skeletons (no.)

Early Byzantine Skeletons (no.)

ARMS BESIDE BODY

Arms straight

8

0

Hands slightly under body

1

0

Arms flexed

1

0

One arm flexed, one arm straight

1

0

11

0

Arms crossed over pelvis

2

2

Arms crossed over abdomen

0

2

Total

1

ARMS OVER BODY

Arms crossed over chest

0

2

Arms over chest

0

1

Arms over abdomen and pelvis

0

1

Arms over abdomen and chest

0

1

Total

2

9

Several other skeletons of Roman date have their arms beside their bodies, but it is difficult to discern their precise position. 1

ers recorded that it was customary to place the arms of the dead over the chest or abdomen, sometimes situating them in the sign of the cross and binding them in place.119 One occasional deviation from the normal extended and supine position was the semiflexion of the legs. In three bodies, two of which were turned to one side, the legs were partly bent.120 The bending of the limbs was not necessitated by the narrowness of the cist or by the proximity of a preexisting body. This position might have resulted from the deposition of the corpse by hand, so that when placed the body rolled over onto its side and the arms and legs fell about the torso. This arrangement reflects a certain informality in the treatment of the body. Most mourners concerned with the display of the corpse during the laying-out and funeral would have handled it more carefully at the moment of deposition. The arrangement of dead bodies at the Isthmus resembles the treatment of corpses in contemporary interments throughout the region. It is difficult to trace patterns with precision because many skeletons in Corinthian graves are not well enough preserved to indicate the position of the body, and excavators have often failed to record this information. Nonetheless, the published remains furnish a general impression for comparative purposes. During the Roman era, the most consistent feature of corporeal position was the supine attitude, which was assumed when the body was carried to the grave. The skeletons in the Early Roman graves at the North Cemetery are all supine, their heads are often turned to one 119. Greg. Nyss. V. s. Macr., PG 46.985 (Pontus, late 4th century); John Chrys. Hom. XXVII in Mt. 4 (PG 57.348–349, 350, late 4th century; ); Geront. V. s. Melan. Iun. 68 (Jerusalem, mid-5th century); Greg. cler. Trans. s. Theod. Thess. 7 (p. 41.14– 15; Thessalonica, end of the 9th century); V. s. Theoph. imp. 23 (p. 16.18–19; Constantinople, end of the 9th to beginning of

the 10th centuries); V. s. Nich. Stud., PG 105.921 (Constantinople, early to mid-10th century); Mich. Psell. Encom. in matr. 23 (p. 37; Bithynia, mid-11th century); Sym. Thess. Ord. sepult., PG 155.676 (early 15th century). 120. NEG 67-001 (child with supine upper body but legs flexed left), 69-005A (on left side), WF 62-001 (on right side).

186

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

side, and most have their arms and legs extended.121 The custom of arranging the body with the head facing forward, the legs straight, and the arms placed over chest, abdomen, or pelvis, most often together but sometimes apart, was established by late antiquity and continued into Byzantine times. Skeletons in graves from the churches and the Forum at Corinth to the rural settlement at Halieis all display this treatment, though with varying degrees of adherence to the norm.122 As at the Isthmus, the flexion of the legs was one anomaly that persisted, but infrequently.123 The placement of objects under or beside the head to stabilize it is also found in Late Roman and Byzantine graves across the region. The most common feature is a “pillow” consisting of either an upward slope or raised area in the floor at one end of the grave124 or, as at the Isthmus, a slab, a tile, or stones placed under the head.125 Numerous graves of Late Roman, Byzantine, Frankish, or Ottoman date containing stones or tiles under the chin, over the chest or mouth, and/or alongside the head have also been found at Corinth and in the Argolid.126 121. Corinth XIII, nos. 503, 157/504, 214/505, 215/506, 145/507, 183/509, 511, 512, pls. 104, 122, 123; see also Walbank (2005, p. 274) on skeletons in Early to Middle Roman chamber tombs north of the city. 122. Stikas 1964, pls. 84:β, 85:α (at least one arm of skeletons in τάφοι A2 and A3 over abdomen or pelvis; Kodratos basilica, Corinth, Late Roman); Miller 1977, p. 5, fig. 3, pl. 2:a; 1979, pl. 28:e; 1988, pl. 4:d (arms crossed over abdomen or pelvis; Nemea, Late Roman); 1981, pp. 48–49 (left hand on chest, right hand on right thigh; near Temple, Nemea, Late Roman); Proskynitopoulou 1988 (arms over pelvis; Pasas or Inachos, Ayia Marina near Argos, Late Roman); Papachristodoulou 1968b, p. 181 (arms crossed over chest, τάφος VI; site of new penitentiary near Tiryns, Late Roman or Byzantine?); Robinson 1962, p. 110, n. 45; Gejvall and Henschen 1968, p. 180, fig. 2 (one body with arm over abdomen, other body with arms folded across the chest; West Shops, Corinth, Late Roman–Early Byzantine); Corinth XIV, p. 164 (hands crossed over torso; Lerna Hollow and Asklepieion, Corinth, Late Roman– Early Byzantine); Corinth XVIII.3, pp. 382, 384–388, 389, fig. 53, pls. 57:b, d, e, 58:b, d, e, 59:a–d (head forward, slightly tilted, or turned to side, and arms either over the chest, abdomen, or pelvis, often crossed, or, less commonly, at the side; Demeter and Kore, Acrocorinth, Late Roman–Late Byzantine?); Oikonomou-Laniado 2003, pp. 28–29, 31–32, figs. 45, 47–54, 58 (North Cemetery and cemetery south of city [area nos. 587, 620, 649], Argos, Late Roman–Early Byzantine); Ivison 1996, p. 113, fig. 5.5 (right arm over abdomen and left arm over pelvis; behind Hemicycle, northeast Forum, Corinth, Early Byzantine); Williams, MacIntosh, and Fisher 1974, pl. 2:a (arms over chest; west Forum, Corinth, 6th century or later); Moutzali 1985, p. 107 (arms crossed over chest; Liapi church, Argos, Early Byzantine?); “Halieis V,” pp. 298–301, fig. 2, pl. 80:d (heads tilted to one side and arms over chest, abdomen, or pelvis and sometimes crossed, but mostly irregular; Halieis, Early Byzantine); Corinth III.1, p. 36 (arms crossed; Upper Peirene, Acrocorinth, “Byzantine”); Zygouries, p. 39 (arms crossed; “Byzantine times or later”). Laskaris (2000, p. 274) notes that this pattern predominates in Late Antique and Byzantine burials across Greece. 123. E.g., Ingvarsson-Sundström 2007, p. 475, figs. 11, 53, pl. 9b (legs flexed to left, S7; lower terraces, Midea, Roman or Late Roman?); Stikas 1964, pl. 86:α (legs flexed to right, τάφος Δ; Kodratos basilica, Corinth, Late Roman); Broneer 1926, pp. 52–53, pl. II (one body on left side with legs slightly flexed; behind Hemicycle, northeast Forum, Late Roman–Early Byzantine); Oikonomou-Laniado 2003, pp. 28–29, figs. 48, 49 (legs flexed to right, tombes 9, 10; North Cemetery [area no. 587], Argos, Late Roman–Early Byzantine); Corinth III.1, pp. 36–38

(legs flexed to left; Upper Peirene, Acrocorinth, “Byzantine”). 124. E.g., Walbank 2005, pp. 263–264, 274, fig. 9.11 (slabs slope up at end in arcosolium, Painted Tomb, north of Corinth, Early–Middle Roman); Morgan 1936, p. 484, fig. 25 (two Late Roman[?] cists in floor of Roman chamber tomb have raised headrest; Kokkinovrysi); Stikas 1964, p. 134, fig. 4 (floor slopes up at west end to create headrest, τάφος E; Kodratos basilica, Late Roman); Wiseman 1967b, p. 418, pl. 84:c (raised strip in floor at west end to create headrest in rock-cut cist, grave 53; Gymnasium area, Late Roman– Early Byzantine); Pallas 1965, p. 133 (floor of brick-lined tomb rises to west end; Lechaion basilica, Late Roman–Early Byzantine?); Williams, MacIntosh, and Fisher 1974, p. 9, pl. 2:b (tile pavement slopes upward at west end of floor in chamber tomb; west Forum, 6th century or later); Kritzas 1976, p. 207; Oikonomou-Laniado 1998, p. 405; 2003, pp. 27–29, figs. 42, 43 (“ἐπικλινῆ προσκεφάλια”; North Cemetery [area nos. 180, 587], Argos, Late Roman–Early Byzantine). 125. E.g., Skarmoutsou-Dimitropoulou 1995, p. 154, fig. 4 (slabs under head; northeast of Corinth, Early or Late Roman?); Miller 1988, p. 3, pl. 4:d (Laconian tile under head in several graves; near basilica, Nemea, Late Roman); Robinson 1962, p. 111, n. 46 (stone headrest placed at angle under head; West Shops, Corinth, Late Roman–Early Byzantine); Pallas 1957, p. 216 (thin stone slab placed at angle under head; Skoutela basilica, Early Byzantine?); “Halieis V,” p. 300, fig. 2 (large pottery sherds under head, grave 7; Early Byzantine); Corinth III.1, pp. 10–11, fig. 9, pl. 2 (small stones placed under head; Acrocorinth basilica, “Byzantine”); Corinth XVIII.3, pp. 382, 387, fig. 53, pl. 59:b–d (stones placed under head, graves 27–29; Demeter and Kore, Acrocorinth, Late Roman–Late Byzantine?). 126. E.g., Demakopoulou et al. 1997–1998, p. 80, fig. 88; Kosmetatou 2007, p. 191; Ingvarsson-Sundström 2007, p. 472, figs. B-1b, 83, 86, pl. 15a (stone over upper chest and neck, S3; lower terraces, Midea, Roman or Late Roman?); Corinth XVIII.3, pp. 382, 387, pl. 58:e (stones around head, no. 26; Demeter and Kore, Acrocorinth, Late Roman–Late Byzantine?; several other unpublished examples from the Gymnasium, east of Theater, and the Temple Hill basilica are cited in nn. 5–7); Oikonomou-Laniado 2003, p. 32 (“une pierre sur la bouche,” tombe 2; cemetery south of city [area no. 649], Argos, Late Roman–Early Byzantine); Papachristodoulou 1968b, p. 181 (stone over chest, τάφος VI; site of the new penitentiary near Tiryns, Late Roman or Byzantine?); Williams and Zervos 1990, p. 350; 1991, p. 39; 1996, pp. 22, 25; Williams, Barnes, and Snyder 1997, pp. 23, 24; Barnes 2003, p. 437, figs. 26.2, 26.3 (stones or tiles around head, numerous graves; southeast of Temple E, Middle Byzantine, Frankish, Ottoman).

MORTUAR Y FORM AND VARIABILITY

187

The variations in corporeal position among the burials at the Isthmus display certain patterns. Bodies in multiple interments mostly held the same position.127 This suggests that mourners tried to arrange bodies coherently as they added them to preexisting graves to affect a consistency they deemed reverent, appropriate, or simply attractive. On the other hand, the bodies in a few multiple burials had varying positions,128 suggesting that the interment of the dead could be more disordered and even random. Corporeal position was not linked to sex or age because the bodies of men and women of all ages display the full range of positions. Even the head and arms of the well-preserved skeleton of an infant (T14 69-991A) had been carefully arranged (Fig. 2.52). The dominant attitude of the corpse changed over time (Table 4.2). During the Roman phases in the burial sequence (I–III), the range of corporeal positions is heterogeneous. The arms display various arrangements, the torsos and legs are often skewed from the vertebral axis, and three individuals were interred with flexed legs, two completely turned on one side. In contrast, the positions of the bodies interred during the Early Byzantine phase (IV) are strikingly regular. All corpses were supine and extended, and, as has been noted, they were oriented according to a more distinctly axial alignment than in earlier phases. The heads faced forward in all but one case where the skulls were preserved in situ. When the situation of the arms is discernible, they are never straight at their sides or flexed unevenly but are always over the pelvis, abdomen, or chest, often crossed. Finally, the use of tiles and stones underneath and around the head of the deceased was an exclusive practice of the Early Byzantine period.

Funerary Assemblage When the participants in the funeral interred the dead they deposited various objects at the grave. Some of these objects had been attached to the body of the deceased during its preparation at home, while others were added to the grave at the time of deposition. Mourners also placed objects outside the grave once the body was buried and covered, either directly after the funeral or during later visits. The set of all objects intentionally contributed to the burial context during or after the funeral is considered the funerary assemblage. Roman and Byzantine literary sources do not describe funerary artifacts in detail but do attest to the use of different objects as ritual instruments. Many of these objects have been found in contemporary graves throughout the northeastern Peloponnese. They attest to the distribution and variety of the typical assemblage in different periods, locales, and socioeconomic settings. At the Isthmus it has been possible to distinguish which artifacts were incidental to the burial context, rather than coincidental or accidental, on the basis of their nature, quality, and situation (Chap. 2). This local assemblage can be divided into three categories according to ritual use: objects worn by the deceased during the funeral and deposition; objects left with the dead at the time of deposition; and objects left outside the grave after deposition. While the rate of survival of funerary artifacts cannot be estimated, the objects that excavators recovered from the graves must represent only part of the total number at the time of burial. The survivability and degree of preservation of the artifacts depended on their composition, environmental conditions, and method of recovery. Of the objects that did survive, those in gold, lead, bronze, stone, bone, shell, or terracotta were typically well preserved and easily identified, while those in iron had suffered severe corrosion and were difficult to identify. Any garments worn by the deceased or any sheets or cords binding their 127. NEG 69-103A, B, 69-007A, B, DEC 69-901A, B, HO 70902A–J. 128. NEG 67-001A–K, 69-005A–C, 69-001A–D.

188

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

head or limbs had disintegrated long before the excavation of the graves. Overall, the graves did not contain a rich assemblage of great material value but rather one characterized by its functional diversity and aesthetic simplicity. Moreover, the location of different classes of artifacts within the burial environment reflects the manner and sequence in which certain items were worn or deposited.

Clothing and Adornment The first category of funerary artifacts consists of articles of clothing and adornment. Fasteners and metal fixtures provide the best evidence for the dress of the deceased. One type of fastener was buckles to bind leather or woven belts, typically worn at the waist by men. It is also possible that they attached straps over the shoulders or chest because they were sometimes found near the ribs. A bronze buckle was found in HO 70-902 (24), an iron buckle was found with the infant in NEG 69-009 (30), another iron buckle was found on the body of a child in 67-001 (18), and simple iron and bronze loops that might have been buckles were found in 69-005 (22) and 69-009 (31). These buckles were all simple in form and unadorned. Another type of fastener attaching layers of clothing were pins and buttons. An iron and a bone pin used either to affix garments or as hairpins were found near the left shoulder of the middle-aged woman in T14 69-002. A bone button was also found in RB 76002 (29), but it is unclear whether it belonged to a man or a woman and where on the body it was located. No traces of the garments with which these fasteners were associated were found. The only piece of clothing in any of the graves was a fragment of textile or leather in 76-002, which was not inventoried. Numerous iron studs (“hobnails”) around the feet of the three skeletons in NEG 69-005 indicate that one of the corpses, probably the middleaged male on the floor of the cist (C), was shod. Other small iron nails in 69-004 and 70-902 (28) might also have come from footwear but were found in isolation, not clustered. Alternatively, the nails might have been attached to wood, such as the six nails in PAL 56-001, or they might have been residual in the burial fill. Graves of Late Roman and Byzantine date in the region have produced similar evidence for fasteners and clothing worn by the dead. Bronze and iron buckles both plain and decorative are common in contemporary interments in the Corinthia and Argolid.129 Metal pins130 and buttons of various forms131 have also been found in graves, though they are often fragmentary. Surely pins were more common than the few recorded examples suggest. Likewise, clothing and footwear have seldom survived, though pieces of woven fabric and hobnails have been found in Late Antique or Byzantine graves at Corinth and Nemea.132 129. E.g., Wiseman 1969, pp. 79, 80, nn. 27, 30, pl. 25:b, c (bronze buckles; Lerna Hollow and Gymnasium area, Corinth, Late Roman–Early Byzantine); Miller 1980, p. 192, pl. 43:a (bronze belt buckle; near Temple, Nemea, Late Roman); 1988, p. 3, pl. 3:d (buckles; near basilica, Nemea, Late Roman and Late Byzantine); Konsolaki 1987, p. 72 (bronze buckles, τάφος 6; northwest of acropolis, Palaiokastro, Methana, Late Roman); Oikonomou-Laniado 1998, p. 406 (bronze belt buckles; North Cemetery, Argos, Late Roman–Early Byzantine); “Halieis V,” p. 301 (buckle found between middle ribs, grave 13; Halieis, Early Byzantine). 130. E.g., Wiseman 1967b, p. 428, nos. 23, 24, pl. 91:j, l (bronze pins; Gymnasium area, Corinth, Late Roman–Early Byzantine); Miller 1981, p. 48, pl. 12:d (bronze pins; near Temple, Nemea, Late Roman); Kritzas 1976, p. 207; Oikonomou-Laniado 2003, p. 30 (silver and bronze pins, tombes 1, 5; North Cemetery [area no. 622], Argos, Late Roman–Early Byzantine). Stikas (1966, p. 56) notes the discovery of ivory pins at the Kodratos basilica but does not explicitly associate

them with graves. 131. E.g., Corinth XII, no. 2582, pp. 297, 302, pl. 124 (bone button, child’s grave; Corinth, “Byzantine”); no. 2589, p. 302, pl. 124 (bone buttons; Acrocorinth, “Byzantine”); Williams et al. 1998, p. 242, no. 3, pl. 41:c (bronze and silver buttons, grave 1997-36; west of church southeast of Temple E, Corinth, Frankish); Williams and Zervos 1992, p. 170, no. 38, pl. 44 (10 bronze buttons; St. John the Theologian, Corinth, Ottoman). 132. Corinth XIV, p. 164 (“scraps of textiles” in graves; Asklepieion and Lerna Hollow, Corinth, Late Roman–Early Byzantine); Stikas 1964, p. 134, pl. 91:α (gold threads from episcopal vestments[?], τάφος Ε; Kodratos basilica, Corinth, Late Roman); Corinth XII, p. 292, no. 2434 (piece of cloth in grave; “Byzantine”); Miller 1979, p. 85, pl. 28:d (52 hobnails in one grave; near Temple, Nemea, Late Roman). Zygouries, p. 39 records the discovery of iron heelplates from boots in an unlined and uncovered cist at Zygouries but the date is uncertain (“Byzantine times or later”).

MORTUAR Y FORM AND VARIABILITY

189

The most frequent type of funerary artifact at the Isthmus is personal adornment. Dead men, women, and children wore earrings, finger rings, and necklaces. Two earrings in 67-001 (16) and a single earring in 56-001 (17) consisted of plain bronze wire with simple hitch clasps. Another remarkable pair from NEG 69-001 (19) is a more decorative version of the same basic design in gold, with a thick hoop of twisted cable and a delicate globe at the clasp. The gold earrings were perfectly preserved but the thinner bronze hoops had deteriorated into rough fragments; other earrings worn by bodies interred at the Northeast Gate might have disintegrated prior to excavation. The position of all these pieces near the west end of the cist indicates that they were originally on the ears of the deceased. The pair of gold earrings found next to each other near the cranium of a young woman (69-001C) had apparently been worn by her (Fig. 2.21). Only one definite finger ring was discovered, an intact bronze loop with an engraved bezel (15) worn on the right fifth finger of the old man buried in T14 67-002. Another simple, small bronze ring that might have been worn on the finger of an infant was found in NEG 69-008 (23). It is possible but unlikely that the large, thick bronze loop in NEG 69-009 (31) was not a buckle but a ring worn on a finger or in the hair of the dead infant (A). Numerous graves around the urban center of Corinth and elsewhere in the Corinthia and Argolid contained rings and ringlike objects of various designs and materials, most often bronze but occasionally iron, gold, and silver. The general impression is that it became more common to bury the deceased with rings in the Late Roman period, after which point it was almost a regular practice. On the basis of their form and location relative to the bones, these artifacts can often be identified as either earrings133 or finger rings.134 However, many graves have produced metal accessories with discoid, cylindrical, or ringlike features whose function is undetermined.135 Some of these might well have been either earrings or finger rings; 133. E.g., Walbank 2005, p. 274 (gold earrings, Painted Tomb; north of Corinth, Roman); Protonotariou-Deïlaki 1969, p. 127, pl. 68:β (gold earring; Kretika, Corinth, Roman); Rife et al. 2007, p. 160, fig. 11 (gold earrings; Koutsongila ridge, Kenchreai, Early–Middle Roman); Miller 1980, p. 192, pl. 43:a (three pairs bronze earrings, bronze wire loop; near Temple, Nemea, Late Roman); Papachristodoulou 1970, p. 104 (two bronze earrings; Palaio Scholeio north of Klenia, Late Roman); Konsolaki 1987, p. 72 (gold earring, τάφος 2; northwest of acropolis, Palaiokastro, Methana, Late Roman); Corinth XVI, p. 164 (earrings); Wiseman 1967b, pp. 418, 428, n. 39, no. 25, pls. 88:i, 91:m (bronze earrings, bronze hoop); 1969, p. 79, nn. 27, 28 (three bronze earrings; Asklepieion, Lerna Hollow, and Gymnasium area, Late Roman–Early Byzantine); Kritzas 1976, p. 207; Oikonomou-Laniado 2003, pp. 16, 27, 30 (bronze earrings; Doutsou basilica and North Cemetery [area nos. 180, 495, 622], Argos, Late Roman–Early Byzantine); Corinth XII, nos. 2030, 2036, 2037 (silver and bronze earrings from grave; Forum, Early Byzantine); Ivison 1996, pp. 113 (earring; South Stoa, Corinth, Early Byzantine), 114 (earrings; near Ayia Paraskevi, Corinth, Early Byzantine); Moutzali 1985, p. 107 (bronze earring; Liapi church, Argos, Early Byzantine?); Morgan 1936, p. 474, fig. 10; Corinth XII, nos. 2031, 2039 (bronze earrings; Bema church, Corinth, Middle Byzantine). 134. E.g., Rife et al. 2007, p. 160, fig. 12 (gold finger rings; Koutsongila ridge, Kenchreai, Early–Middle Roman); Miller 1981, pp. 48–49, pl. 12:d, e (two silver-plated finger rings; near Temple, Nemea, Late Roman); Papachristodoulou 1968b, p. 181 (iron finger ring, τάφος VI; site of new penitentiary near Tiryns, Late Roman or Byzantine?); 1970, p. 104 (iron finger ring; Palaio Scholeio north of Klenia, Late Roman); Konso-

laki 1987, p. 72 (bronze finger ring, τάφος 10; northwest of acropolis, Palaiokastro, Methana, Late Roman); Wiseman 1967b, p. 420, n. 44, pl. 88:e (bronze finger ring); 1969, p. 83, n. 35 (bronze finger ring); 1972, p. 8 (two iron finger rings; Gymnasium area, Late Roman–Early Byzantine); Kritzas 1976, p. 207; Oikonomou-Laniado 1998, p. 406; 2003, pp. 16, 28, 30 (bronze and gold finger rings; Doutsou basilica and North Cemetery [area nos. 495, 587, 622], Argos, Late Roman–Early Byzantine); Ivison 1996, p. 114 (finger rings; near Ayia Paraskevi, Corinth, Early Byzantine); Davidson 1937, p. 232; Ivison 1996, p. 114, fig. 5.6 (bronze finger rings, graves III, IV; tower in west circuit, Acrocorinth, Early Byzantine); Pallas 1974, pp. 234, 238, pl. 221:β (bronze finger ring; Kraneion basilica, Early or Middle Byzantine?); Corinth XII, nos. 1821, 1830, 1865, 1927, 1934, 1952 (bronze finger rings from graves; Forum, Byzantine); Miller 1988, p. 3, pl. 3:c (finger ring; near basilica, Nemea, Late Byzantine); Williams et al. 1998, p. 242, no. 4, pl. 41:d (bronze finger ring, grave 1997–15; west of church southeast of Temple E, Corinth, Frankish). 135. E.g., Shear 1931, pp. 426, 436 (bronze loop and rings; Cheliotomylos, Corinth, Roman); Corinth XVIII.3, pp. 382–383, no. 1, pl. 57:a (bronze cylinder, two bronze hair rings; Demeter and Kore, Acrocorinth, Late Roman); Kritzas 1976, p. 207 (“χαλκοῦν δισκόμορφον περίαπτον, μολύβδινον δισκάριον”; North Cemetery, Argos, Late Roman–Early Byzantine); Corinth XVIII.3, pp. 388–389, no. 27 (thin metal pierced disk from necklace or earring; Demeter and Kore, Acrocorinth, Late Roman–Late Byzantine?); Ivison 1996, p. 113, fig. 5.5 (bronze cylinder; Hemicycle, northeast Forum, Corinth, Early Byzantine); Corinth III.1, p. 36 (bronze rings; Upper Peirene, Acrocorinth, “Byzantine”).

190

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

large loops might have been rings for holding the hair in place, and smaller rings might have been pendants or attachments for clothing. A third subcategory of adornments is necklaces and other hanging objects. Two beaded necklaces with the same essential form but different elements were discovered in graves on the Isthmus. One in T14 69-991 consisted of a bone pendant in the form of a Greek cross with a finely incised geometric pattern (20) and at least 11 beads (21a–k) of various materials (stone, bone, shell, lead, terracotta, glass), shapes (cuboid, trapezoidal, cylindrical, globular, oblate), and dark colors (gray to black, brown, green, blue, purple). A second necklace in PAL 56-001 consisted of at least 13 beads, several of which were small spheres or cylinders in blue or blue-green glass (33a–e). This necklace, and probably the other one, originally had more beads than were found by the excavators. The described beads were intact but their strings had disintegrated, and so their arrangement is only conjectural (Fig. 2.102). The first necklace is noteworthy for the multifarious beads and the intricate cross, which would have hung in the middle of the string. Several beads in the second necklace are strikingly uniform in appearance. In the first example, the beads and pendant were found clustered near the upper right arm of the infant (69-991A) where they came to rest after slipping from the center of the chest either above or below the crossed forearms.136 In the second example, the beads were located in the area of the upper chest, but no thoracic bones were sufficiently preserved to confirm their exact location on the body. Other graves contained single small, pierced objects that were also worn by the deceased, probably on strings around the neck. A spherical black stone on a twisted glass rod was found in HO 70-902 (25), and an oblong shell was found in NEG 69-009 (32). A well-worn coin that had been pierced for reuse as an adornment was found in T14 69-002 (14). Its position over the right shoulder near the neck of the adult woman (A) suggests that it had hung on a necklace or may have been sewn to her garment. Finally, a series of corroded, round iron objects of unknown function, an iron loop in NEG 69-005 (22) and five disks or rings in two fused clumps in HO 70-902 (26, 27), might have been either pendants or decorative attachments for clothing, though the single loop was more likely a buckle. Similar necklaces and hanging objects have been found in graves across the region. Again, the general sense is that, like rings, they occur more frequently in Late Roman and Byzantine graves than in earlier interments, but this hypothesis cannot be proven without more complete publication of the evidence. Beads of various materials have been discovered in graves at Corinth and Nemea and on Methana. These beads were often found in large numbers and sometimes near the upper torsos, arms, or heads of skeletons, indicating that they had originally hung on necklaces.137 Small pierced objects such as beads, rings, or disks that were found in small numbers or in isolation within burial contexts might have been suspended from simpler necklaces or bracelets or sewn onto clothing.138 Individual pen136. It is tempting to imagine that the mourners centered the pendant over the crossed arms and above the chest of the infant. Symeon, archbishop of Thessalonica in the early 15th century, wrote (Ord. sepult., PG 155.676) that it was customary to place a psalterium, evangelium, or icon over the hands of the deceased, a practice that is still observed in the Orthodox liturgy (Kyriakakis 1974, pp. 56–57; cf. Danforth and Tsiaras 1982, pp. 39–40, pls. 1, 6–8; Kenna 1991, p. 103). 137. E.g., Corinth XII, nos. 2475, 2477–2480 (glass beads from a grave; Late Roman?); Konsolaki 1987, p. 72 (“ένα χρυσό αλυσιδωτό περιδέραιο με γαλάζιες χάντρες,” τάφος 6; northwest of acropolis, Palaiokastro, Methana, Late Roman); Wiseman 1967b, p. 418, n. 39 (glass beads, grave 53; Gymnasium area, Late Roman–Early Byzantine); Corinth XII, p. 289, pl. 148:c (ca. 300 glass beads in necklace; Forum, Early Byzantine); Ivison 1996, p. 114 (beads; near Ayia Paraskevi,

Corinth, Early Byzantine); Corinth XII, nos. 2434, 2467, 2469, 2476, 2490, 2492 (glass beads from one grave; “Byzantine”). 138. E.g., Corinth XIII, p. 296, no. 509-2 (glass bead, grave 509; North Cemetery, Early Roman); Shear 1930, p. 428 (gold bead; north of Cheliotomylos, Early Roman); Miller 1988, p. 3, pl. 5:a, (cross-incised bead; near basilica, Nemea, Late Roman); Wiseman 1967b, p. 418, n. 39, pl. 88:i (bronze hoop, possibly from necklace, grave 53), p. 428, no. 26, pl. 91:n (stone bead, grave 47; Gymnasium area, Late Roman–Early Byzantine); Corinth XII, nos. 2425 (faience bead; “Byzantine”), 2461 (glass bead; Forum, Early Byzantine); Morgan 1936, p. 474, fig. 10; Corinth XII, p. 307, no. 2657, pl. 126 (bronze trilobate ornament from chamber tomb; Bema church, Middle Byzantine); Anderson 1967, p. 7 (small rings, silver spangles around skeleton, possibly dress ornaments; northwest of Temple E, Middle–Late Byzantine).

MORTUAR Y FORM AND VARIABILITY

191

dants from graves on the Isthmus also have parallels. The basic form of the incised pendant cross in T14 69-991 compares with Late Roman examples from simple graves in the vicinity of the basilica at Nemea and Middle Byzantine examples from tombs at the Bema church on the south side of the Corinthian Forum.139 The closest parallel for the pierced coin in T14 69-002 is a pierced coin of Constans I (346–350) found in a multiple burial of Late Roman to Early Byzantine date in the Gymnasium area at Corinth. The coin had been out of circulation for at least several decades, perhaps a century or more, when it was perforated for suspension from the body.140 These objects indicate how the corpses interred at the Isthmus looked during the layingout, funeral, and deposition. The presence of buckles, a button and pins, and hobnails shows that dead men, women, and children were not shrouded anonymously but were dressed as they had been in life. Mourners might have partially covered the dressed corpse with a “linen,” “shroud,” or “winding-sheet” (ὀθόνη, πέπλος, σινδόνιον, σάββανον, λαζάρι), a common Byzantine practice that continues today.141 Although the standard wardrobe cannot be reconstructed in detail, the basic outfit of local residents evidently consisted of belts, shoes or sandals, and layered clothing, presumably tunics and over-garments. The simplicity of the fasteners and fixtures suggests that their dress was elementary and utilitarian, but woven fabrics or accessories in perishable materials might have displayed more intricate ornamentation. Late Roman and Byzantine art and literature attest to essential differences in men’s and women’s coiffure, cosmetics, jewelry, and footwear.142 A few individuals at the Isthmus wore jewelry for burial during the Roman, Late Antique, and Early Byzantine periods: one man wore a finger ring, women wore earrings and necklaces, and children wore rings and strung beads. Belts and hobnailed footwear were most likely items of masculine apparel.143 The hair of dead women was probably prepared for display, such as by pinning.144 In general, local residents dressed corpses attractively but not resplendently. The dead appeared not unlike many of their contemporaries buried in other Corinthian cemeteries. However, they were not as richly dressed and decorated as certain individuals who had been buried with a large and varied set of jewelry and fixtures for clothing. There is no evidence that the dead at the Isthmus bore the impressive finery of silk and gold vestments frequently cited by St. John Chrysostom,145 the sort of dress that typified urban aristocrats and ecclesiastical leaders. 139. Miller 1988, p. 3, pl. 5:a (bone and bronze cruciform pendants; near basilica, Nemea, Late Roman); Morgan 1936, p. 474, figs. 10, 11; Corinth XII, nos. 2072, 2078 (bronze cruciform pendants; Bema church, Corinth, Middle Byzantine). A cruciform pendant in silver on an earring was found in a Frankish grave west of the church southeast of Temple E at Corinth (Williams et al. 1998, p. 242, no. 2, pl. 41:b). Crosses have also been found in graves elsewhere in the Corinthia and Argolid: Kritzas 1976, p. 207; Oikonomou-Laniado 2003, p. 27 (small bronze cross, tombe 2; North Cemetery [area no. 180], Argos, Late Roman–Early Byzantine); Banaka-Dimaki 1988a, p. 105 (bronze cross; Kraneion, Corinth, Byzantine?); Konsolaki 1989, p. 50 (lead cross; near Troezen, Late Roman or Byzantine); Oikonomou 1989, pp. 305, 306 (τάφος 9; Laliotaiïka, Ano Epidauros, Middle Byzantine?). 140. Wiseman 1969, p. 79, n. 28 (grave 86). Oikonomou (1989, pp. 305, 307 [τάφος 9]) records the discovery of a pierced follis of Justinian (538–539) in a grave of Middle Byzantine(?) date at Laliotaiïka, Ano Epidauros. Laskaris (2000, p. 323) cites pierced coins of Middle Byzantine and Ottoman date in graves at Mystras, Glyki, and Platamon. See also pp. 70–71 on the continued circulation of Late Roman bronze coinage and their reuse for personal adornment. 141. Spyridakis 1950, pp. 110–111, 159; Abrahamse 1984,

pp. 129–130 (Byzantine custom); Megas 1940, pp. 171–172; Papacharalampous 1965, p. 147; Danforth and Tsiaras 1982, p. 39, pls. 1, 3, 4, 9, 10, 13, 14 (Modern Greece and Cyprus). 142. Koukoules 1951, pp. 342–418; see also RBK III, 1978, cols. 444–448, 485–487, s.v. Insignien B.XIII, XIV, C.IIId (K. Wessel) on Imperial dress. Clark (1993, pp. 105–118) discusses feminine dress in late antiquity. 143. Koukoules (1951, pp. 400–401, n. 3) examines the textual evidence for shoes with nailed soles. They were not only worn by male soldiers but also travelers and sometimes women. Byzantine authors record that men could be buried in shoes (e.g., V. s. Dan. Scet. p. 67 [d. late 6th century, Tambok, Egypt]; Sym. Thess. Ord. sepult., PG 155.676 [early 15th century]). Greeks today, including the residents of Kyras Vrysi, often place shoes in the casket with the corpse. 144. Cf. Koukoules 1951, p. 367. 145. Anna 5.5 (PG 54.675); Dros. mart. 6 (PG 50.692); Hom. XI in Heb. 3 (PG 63.94); Hom. XIII in Tim. 4 (PG 62.569); Hom. div. 31 (PG 63.811); Pan. Pel. 3 (PG 50.582); Pat., PG 60.725. Loukatos (1940, pp. 50–52) discusses this testimony. Other sources also refer to corpses wearing rich garments: Bas. Caes. Hom. div. 9 (PG 31.304); Greg. Nyss. Laud. Melet. epis., PG 46.857; V. s. Ephr. Syr., PG 46.837, 46.845; John Mosch. Prat. spir. 77 (PG 87.2932B–C).

192

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

Objects Left Inside the Grave: Coins and Vessels The second category of funerary artifacts consists of objects left by mourners inside the grave, namely, coins and terracotta or glass vessels. These objects would have been carried by participants in the procession and placed over and around the body once it was deposited in the grave. The first subcategory is coins, which were only recovered from two graves.146 A cluster of 10 bronze coins of the late 5th century was found in the thoracic area of the uppermost skeletons in 67-001 (1, 2, 3). The situation of these coins suggests that they were deposited as a pile or in a purse over the chest of one of the last corpses interred in this multiple burial (Figs. 2.12, 2.13). The only other coin was a single pentanummium in T14 69-003 (4). Like the coins in 67-001, it was found in the thoracic region of the skeleton. The mourners had left it over the chest of this young man when he was buried. The total number of coins is too small to represent any significant distribution by period. It is, nonetheless, notable that coins were placed in graves of both Late Roman and Early Byzantine date. These coins had no practical purpose but served as both symbolic offerings and personal possessions contributed to the burial. The placement of valued items like coins directly over the corpse during the laying-out or funeral signified to those present both a personal connection to the dead and a personal sacrifice. In the tradition of “Charon’s obol,” it was also an expression of farewell and a last chance for intimate contact as the dead left the world of the living. Whether mourners left one coin or several, the act carried the same message. It not only helped to satisfy the emotional needs of the mourners as a consolatory gesture but also exhibited the devotion of the bereaved in the public context of the graveside rite. Thus, it is perhaps not surprising that the simple but ancient custom of depositing coins on the corpse persisted in Late Roman Greece and throughout Byzantine times to the Modern era, though with variable meanings in separate contexts.147 Single or multiple coins were commonly deposited in graves across the northeastern Peloponnese. Many publications cite the discovery of one or more coins in graves of Roman, Byzantine, and later date in urban, suburban, and rural settings.148 In some cases it is 146. The pierced coin in T14 69-002 (4) was an article of adornment. 147. Lawson ([1910] 1964, pp. 108–114), Polites ([1931] 1975, pp. 330–332), Megas (1940, p. 175), Koukoules (1951, pp. 158–159, n. 7), and Papacharalampous (1965, pp. 148, 161–162; 1971, pp. 227–229, 231, 246–250) discuss various modern examples from mainland Greece, the Ionian and Aegean islands, Asia Minor, and Cyprus as evidence for the continuity of the practice since classical antiquity. Schell (1989, pp. 104–112) stresses the variable uses and meanings of the custom in Modern Greek society and discredits its interpretation in the context of ancient practice. Alexiou (2002, p. 39) notes the modern practice of placing coins on the forehead or in the mouth explicitly as an apotropaic device or fee for the ferry. Danforth and Tsiaras (1982, p. 40, pl. 6) record that mourners placed coins on the chest of corpses during wakes in a Thessalian village in 1979; cf. pp. 19, 85 on the placement of paper money on the skull during the exhumation. See p. 149 on this custom in the Corinthia today. Kyriakakis (1974, p. 58) incorrectly concludes from the “silence of the literary sources” that the practice must have been “sporadic.” Byzantine writers might not have been concerned to relate this kind of mortuary behavior because it was neither integral to the funeral liturgy nor evocative of Christian eschatology. Another possibility is that the members of the educated elite whose writings have survived were unaware of, or disinterested in, the practice because it was a popular or regional phenomenon.

148. E.g., Shear 1930, pp. 428, 430; 1931, pp. 426, 427– 428, 436 (north of Cheliotomylos, Corinth, Early Roman); Corinth XIII, pp. 296, 298, nos. 509–1, 515–1 (North Cemetery, Corinth, Early Roman); Rife et al. 2007, p. 160 (Koutsongila ridge, Kenchreai, Early–Middle Roman); Walbank 2005, p. 274 (north of Corinth, Roman); Protonotariou-Deïlaki 1969, p. 127; 1970, p. 102 (Kretika, Corinth, Roman); BanakaDimaki 1988b, p. 89 (τάφος V, Ayia Anna, Corinth, Roman); Drosoyianni 1972, p. 206 (south of Lechaion, Roman); Bookidis and Fisher 1972, p. 331, no. 100; Corinth XVIII.3, p. 386, no. 22 (Demeter and Kore, Acrocorinth, Late Roman); Corinth XIV, p. 164 (reservoir IV); Wiseman 1967a, p. 32; 1969, p. 82, n. 32 (grave 69), p. 83, n. 35 (grave 77), p. 84 (Lerna Hollow and Gymnasium area, Late Roman–Early Byzantine); Zygouries, p. 71 (tomb XIII; Ampelakia west of Zygouries, Late Roman?); Oikonomou-Laniado 1998, pp. 406, 407; 2003, pp. 28–29, 29– 30, 31, 32, 33 (North and East Cemeteries and cemetery south of the city [area nos. 587, 621, 622, 623, 649], Argos, Late Roman–Early Byzantine); Ivison 1996, p. 117 (South Stoa, Corinth, Early Byzantine); Pallas 1981, p. 298, n. 18 (Kraneion basilica, Corinth, Early Byzantine); Corinth XVIII.3, p. 390 (north slope of Acrocorinth, Early Byzantine); Shear 1928, p. 487 (east of Theater, Corinth, “Byzantine”); Oikonomou 1989, pp. 305, 307, 308 (τάφοι 9, 12; Laliotaiïka, Ano Epidauros, Middle Byzantine?); Corinth XVIII.3, p. 382 (Temple Hill basilica, Corinth, Middle Byzantine); Williams and Zervos 1988, p. 127, n. 40 (building 7, east of Theater, Middle

MORTUAR Y FORM AND VARIABILITY

193

difficult to determine whether these coins were intentionally or unintentionally deposited in the grave, but their high frequency suggests that many were funerary objects. If the coin in T14 69-003 was intentionally placed there, its deposition over the chest of the deceased during the laying-out or funeral may be viewed as an Early Byzantine echo of the pagan custom of “Charon’s obol.”149 Several graves dating to the Late Roman to Byzantine eras and later in the Corinthia and other parts of Greece contained coins that had been deliberately situated around the body, either in a hand, on the chest, on the head, or in the mouth.150 The deposition of several coins over the torso in NEG 67-001 also has parallels in Byzantine graves in the Corinthian Forum. A double interment of the late 6th century in the building behind the Hemicycle contained 56 coins, while a burial chamber of the late 10th century next to the Bema church contained at least 12 coins.151 Similarly, an isolated grave of the late 6th century containing 23 coins has been found near the cistern and basilica in the ruins of the Sanctuary of Pythian Apollo on the Aspis at Argos.152 Coin hoards have also been found in close proximity to Late Antique graves at Argos and Nemea,153 though it is uncertain whether the coins were formally deposited as part of a funerary ritual or were coincidental to the burial context. The second subcategory of objects placed inside the grave are vessels in terracotta or glass, five examples of which were found in four multiple interments. A coarse-ware cup with an ovoid body, a flaring rim, and a vertical handle was found along the base of the south wall of 67-001 (8). It was placed alongside the right shoulders and upper arms of the first bodies in the cist (Fig. 2.14). A fine-ware jar with a fusiform, wheel-grooved body and a vertical handle was found in 69-103 (9), where it was interred with the bodies of two adults. A shallow coarse-ware bowl with a flat base and a folded rim was found at the west end of the south compartment of 70-902 (10). It was set on the floor of the cist before the interment of the eight bodies but in the area of their heads. Finally, two glass vessels, an intact pitcher (7) and a bottle, possibly an unguentarium, which disintegrated upon recovery (13), were found directly under the covering slabs and over the displaced bones of four individuals Byzantine); Corinth XVI, p. 29 (South Basilica, Corinth, Middle Byzantine or Frankish); Williams and Zervos 1996, p. 23; Barnes 2003, p. 437 (church southeast of Temple E, Middle Byzantine or Frankish); Miller 1988, p. 3, pl. 4:a, b (near basilica, Nemea, Ottoman). Gold bracteates of Late Classical or Hellenistic coins from Sicyon have also been found in Roman burials north of the harbor at Kenchreai (unpublished evidence known to author) and north of Corinth (Robinson 1964, p. 63; Papachristodoulou 1968a, p. 166; Wiseman 1969, p. 87, n. 40; Walbank 2005, pp. 276–277, fig. 9.16). 149. The ancient custom at Corinth is represented in Roman graves in Kretika, where bronze coins were found in the mouths of skeletons (Protonotariou-Deïlaki 1969, p. 125). 150. Miller 1981, p. 49 (coin in right hand of skeleton; near Temple, Nemea, Late Roman); Oikonomou-Laniado 2003, p. 30 (“une monnaie de bronze près du crâne,” tombe 7; North Cemetery [area no. 622], Argos, Late Roman–Early Byzantine); Corinth XVIII.3, pp. 387–389, nos. 27, 29 (coins resting on skulls; Demeter and Kore, Corinth, Late Roman– Late Byzantine?); Morgan 1936, pp. 473–474 (coins on skulls in graves in nave and north aisle; Bema church, Corinth, Middle Byzantine). Examples from other regions: Papapostolou 1979, p. 355 (coins in mouth; Patras, Roman); Keramopoullos 1917, p. 104 (coins in mouths; Athens, Middle–Late Roman); Kourkoutidou-Nikolaidou 1997, p. 131 (coins in mouth and on chest; Thessalonica, Late Roman); Sweetman 2005, p. 366 (coins in burials; Knossos Medical Faculty site, Crete,

Late Roman); Di Vita 1988a, p. 109; 1988b, p. 161 no. 23 (coin in tomb 17 near west end of grave; Settore L, Gortyn, Early Byzantine); Pallas 1951, p. 166 (“νόμισμα . . . χρησιμοποιηθὲν ὡς δανάκη,” τάφος 2; Ayios Dimitrios near Moulki, Salamis, Middle Byzantine); Moutsopoulos 1972, pp. 196–201 (numerous coins among skeletons; Ayios Achilleios, Prespa, Macedonia, Middle–Late Byzantine); Gerstel et al. 2003, pp. 201, 218 (coin under or on chest of 1992-2a; Panakton, Attica, Late Byzantine); Agora XX, pp. 28–29 (coins in graves 6, 8, 9, some near vertebral column; Church of the Holy Apostles, Agora, Athens, early 19th century). Laskaris (2000, pp. 323) collects numerous examples from Roman, Byzantine, and later graves. 151. Broneer 1926, pp. 52–53; Corinth XVI, p. 8 (behind Hemicycle); Morgan 1936, p. 474, fig. 10; Corinth XVI, p. 42 (Bema church). 152. Vollgraff 1956, pp. 95, 97–98; Oikonomou-Laniado 2003, p. 12. 153. A hoard of 154 Late Roman bronze coins was found near the bottom of a trench also containing pottery and lamps situated close to several tile-covered and simple cist graves south of Argos (Oikonomou 1988, p. 481, and OikonomouLaniado 2003, pp. 31–32 [area no. 649]). Nemea II, p. 134, records that the scattered bones found near the west end of the tunnel into the Stadium were “essentially framed” by the coin hoard dated to the late 570s/early 580s, but it is unclear whether the coins were deposited with the bones; see also p. 192, below, on this burial.

194

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

in 69-004 (Fig. 2.7). It is uncertain whether these glass vessels were originally buried with the deceased or added when the skeletons were disturbed by construction. In all cases, the diminutive, undecorated vessels were common domestic forms deposited alone or in small numbers. The participants in the funeral placed them in the grave either beneath or beside the dead bodies, typically near the chest or head. The vessels were only found in the large multiple burials containing men, women, and children of all ages during the Late Roman occupation of the fortifications. These vessels might have had several funerary uses. Four of the vessels were designed to hold liquid or viscous substances and the fifth, a bowl, could have held liquids as well. One possibility is that they contained unguents, aromatics, or oil which mourners poured or sprinkled over the body during the funeral and deposition. Then they placed the container, either empty or partly filled, next to the head or torso.154 A second possibility is that the vessels were used to pour libations to the dead before the grave was closed. The narrowmouthed jar and glass vessels were better suited for ointments, while the cup and bowl were better suited for drinks. Both the chrismation during the funeral155 and the libation at the graveside156 are recorded in the Late Roman and Byzantine sources, though the libation was typically poured during a funerary banquet, a memorial service, or a visit after the body had been deposited and the grave closed. A third possibility is that these vessels contained consumables that were given to the deceased to symbolize a last meal. The excavators did not record organic remains, but foodstuffs seldom survive in the archaeological record.157 It should be remembered that the vessels themselves were objects acquired for the occasion or more likely brought from home. Their contribution to the burial was a minor but meaningful sacrifice on the part of the bereaved. Terracotta and glass vessels are commonly found in graves of Roman and Byzantine date across the region. During the first centuries of the Roman era, cups, jugs, bowls, plates, casseroles, and cooking pots of diverse forms were often deposited either inside or outside the interment, in both isolated graves and chamber tombs.158 Graves of Late Roman to Early or Middle Byzantine date also contained pottery from the domestic assemblage, including large or small amphoras, cooking pots, bowls or plates, cups or jars, pitchers or jugs, oinochoai, and lekythoi.159 The funerary lekythos is a distinct Late Antique type that seems to 154. A small bronze lidded vessel containing a black powder was found in a Late Roman grave near the acropolis on the Methana peninsula (τάφος 5; Konsolaki 1987, p. 72). The substance might have been an aromatic or ointment for application to the corpse. 155. E.g., Greg. Nyss. Laud. Melet. epis., PG 46.857 (Syrian Antioch, late 4th century); John Chrys. Hom. LXXXV in Io. 3 (PG 59.464–465), Pat., PG 60.725 (late 4th century); [Dion. Ar.] Ec. hier. 7.3.2 (PG 3.556), 7.3 (Cont.) 8 (PG 3.565; late 5th–early 6th centuries?); V. s. Iren. heg. Chrys. 9.89 (AASS Iul. 6.633C; d. ca. 940, Chrysovalantou Monastery, Constantinople; vita written late 10th–early 11th centuries). Rife (1999, pp. 84– 85) discusses testimony from the Roman era; Kyriakakis (1974, pp. 46–47) cites several hagiographic sources. 156. E.g., Greg. Naz. Or. 7.16 (PG 35.773–776), Or. 40.15 (PG 36.377), Poem. mor. 28.198–199 (late 4th century; PG 37.871); John Lyd. Mens. 4.26 (6th century); Mich. Attal. Hist. p. 279.15–17 (11th century). Rife (1999, pp. 85–86, 126–127) discusses the literary testimony for libations in the Greek world during the Roman era; Koukoules (1951, pp. 211–213) addresses the Late Roman and Byzantine sources for libations and the continuation of the practice in Modern Greece. 157. Sodini (1977, pp. 17, 20, n. 61, fig. 4) notes this possibility, citing the discovery of cereal grains in a pitcher in a Late Antique grave at Evraiokastro on Thasos (Rolley 1965,

p. 919, fig. 8). 158. Rife (1999, pp. 269–271) surveys the pottery found in Early to Middle Roman Corinthian burials. Melissa Moore Morison is conducting an important study of a major ceramic assemblage from the Early to Middle Roman cemetery on the Koutsongila ridge north of the harbor at Kenchreai; for an interim report, see Rife et al. 2007, pp. 166–175, figs. 17–25. 159. Corinth I.2, p. 55, n. 3 (“two small pots of coarse clay” from burial in drain; under the Captives facade, Forum, Late Roman); Corinth XVIII.2, p. 127, no. 276, pl. 16; Corinth XVIII.3, pp. 382–383, 384, no. 7, pl. 57:d (lekythos; Demeter and Kore, Acrocorinth, Late Roman); Robinson 1962, p. 116 (jug from grave; northeast of Anaploga, Late Roman?); Kristalli-Votsi 1984, p. 65; Rife et al. 2007, pp. 172–173, 174–175 (various vessels from cist graves; Kenchreai, Late Roman–Early Byzantine); Pallas 1961, p. 173, pl. 72:β (pitchers, lekythoi from graves annexed to south aisle; Lechaion basilica, Early Byzantine); Pallas, Charitonides, and Venencie 1959, pp. 496–497, figs. 1, 2 (oinochoe and lekane from chamber tomb; Longopotamos valley southwest of Solomos, Late Roman); Papachristodoulou 1970, p. 104 (19 vessels; Late Roman chamber tomb, Palaio Scholeio north of Klenia); Biers 1973, p. 108, n. 16, pl. 19:d, e (jug; Palati, Phlius, Middle Byzantine); Zygouries, pp. 63, 70–74, 176–179, nos. 322, 364–368, 567, 568, figs. 62, 63, 173–175 (cup, bowl, jugs, pots from

MORTUAR Y FORM AND VARIABILITY

195

have been manufactured expressly for use in burial at both Corinth160 and Argos.161 Glass vessels have been found in graves less frequently than pottery. It is uncertain whether this scarcity is an accurate indication of the historical distribution of funerary glass or a consequence of its low survivability in the archaeological record. Small, delicate glass vessels had a greater monetary value than table wares, which may explain their relative infrequency. But it can also be assumed that many more graves originally contained glass vessels than is revealed in the publications. The commonest glass forms, when they are discernible, are small bottles or pitchers, cups, and bowls.162 Vessels in other materials are rare, though a bronze vase has been found in a Late Roman grave in the North Cemetery at Argos.163 Overall, the funerary assemblage of vessels at the Isthmus resembles the range of forms and materials in contemporary graves throughout the region, but the primary difference is the absence of funerary lekythoi. As at the Isthmus, it is uncertain whether terracotta and glass vessels from Corinthian graves were used for chrismations or libations. The common cups and pitchers for pouring and drinking might well have been used for single libations during the funeral or deposition because they were typically found alongside the corpse. Lekythoi and glass bottles could have also been used for libations.164 However, like unguentaria, they might have contained

tombs XXXV, XIII, XIV, XVII, VIII, XI, XVa; Ampelakia west of Zygouries, Late Roman?); Proskynitopoulou 1988 (oinochoe; Pasas or Inachos, Ayia Marina near Argos, Late Roman); Konsolaki 1987, p. 72 (cups from τάφοι 4, 5, 11; northwest of acropolis, Palaiokastro, Methana, Late Roman); Corinth XIV, p. 164, pl. 67:5, 6 (jugs and lekythoi); Wiseman 1967a, pp. 34, 37–38, pl. 15:b (jar and lekythoi, graves 3 and 5); 1967b, pp. 418, 420, nn. 38, 43, pl. 88:a–c (jug and lekythoi, graves 31 and 53); 1969, p. 80, n. 30, pl. 25:d (lekythos and pitcher, grave 81), p. 82 (lekythos and jar, grave 69), p. 83 (lekythos, grave 77); 1972, p. 8 (“a few whole vessels”; Asklepieion, Lerna Hollow, and Gymnasium area, Late Roman–Early Byzantine); Kounoupiotou 1972, p. 208, pl. 196:α, β (oinochoai, cup; Kraneion, Early Byzantine?); Kritzas 1976, p. 207; Oikonomou-Laniado 1998, pp. 405–406, 407; 2003, pp. 27–32, 33–34, 35–50, figs. 59–102 (lagynoi, amphoras, amphoriskoi, oinochoai, lekythoi, pitchers, cooking pots, plates or bowls, cups; North, South and East Cemeteries and cemetery south of the city [area nos. 180, 587, 620–623, 649], Argos, Late Roman–Early Byzantine); Corinth XI, pp. 167–168, fig. 146:a, c (pitchers from cist graves; Forum, Early Byzantine); Williams, MacIntosh, and Fisher 1974, p. 9, no. 7, pl. 2 (oinochoe from chamber tomb; west Forum, Early Byzantine); Williams and Fisher 1975, pp. 16–17, nos. 15–18, pl. 6 (jugs and lekythos from chamber tomb; southwest Forum, Early Byzantine); Ivison 1996, pp. 113 (jug; South Stoa, Corinth, Early Byzantine), 114 (jug; near Ayia Paraskevi, Corinth, Early Byzantine); Robinson 1976a, p. 222 n. 54 (jug from burial chamber; basilica on Temple Hill basilica, Early Byzantine); Stikas 1964, p. 131 (“δύο μικρὰ ληκυθοειδῆ ἀγγεῖα” from burial chambers annexed to north aisle and apse; Kodratos basilica, Corinth, Early or Middle Byzantine?); Kenchreai IV, pp. 135–136, no. RC 73, pl. 35 (lekythos; basilica on south mole, Kenchreai, Early Byzantine); Oikonomou 1989, pp. 305, 307 (lekythos, τάφος 10; Laliotaiïka, Ano Epidauros, Middle Byzantine?). Sodini (1977, pp. 15–20) and Laskaris (2000, pp. 324–326) discuss evidence for the deposition of ceramic and glass vessels in Late Roman and Byzantine graves. 160. One common form has a cylindrical body and a flat bottom: e.g., Pallas 1961, p. 173, pl. 72:β (Lechaion basilica); Kounoupiotou 1972, p. 208, pl. 196:α (Kraneion); Corinth XIV, p. 164, pl. 67:5; Wiseman 1967a, pp. 37–38, pl. 15:b3–

5 (Asklepieion, Lerna Hollow, Gymnasium area); Corinth XVIII.2, no. 276, p. 127, pl. 16 (Demeter and Kore). Another common form has a piriform body and a flat bottom or a disk foot: e.g., Corinth XVI, p. 164, pl. 67:6; Wiseman 1967a, p. 37, pl. 15:b1, 2 (Asklepieion, Lerna Hollow, Gymnasium area); Williams and Fisher 1975, p. 17, no. 18, pl. 6 (southwest Forum); Kounoupiotou 1972, p. 208, pl. 196:α (Kraneion); cf. Kenchreai IV, pp. 135–136, no. RC 73, pl. 35. That urban residents buried lekythoi with their dead at least as late as the 7th century and perhaps in the 8th or 9th century is indicated by two late instances recently discovered east of the Late Roman bath in the Panayia area (Blackman 1999, p. 23). 161. As at Corinth, both the cylindrical and piriform shapes are present at Argos, though the former predominates: Oikonomou-Laniado 1998, pp. 406–407; 2003, pp. 27–30, 32, 33–34, 40–41, 45, 47–48, nos. 17–59, figs. 75–83 (North and South Cemeteries and cemetery south of the city [area nos. 180, 587, 620–622, 649], Argos, Late Roman–Early Byzantine). 162. E.g., Corinth XII, nos. 592, 675 (bowl and bottle; Early–Middle Roman grave), 629 (bowl; Early–Middle Roman tomb), 644 (cup; Early Roman grave, peribolos of Apollo); Shear 1930, p. 428; 1931, p. 436 (bottles; Early Roman graves, Cheliotomylos); Papachristodoulou 1968a, p. 169; Protonotariou-Deïlaki 1969, p. 127 (vessels; Roman graves, Kretika); Skarmoutsou-Dimitropoulou 1995, p. 154 (unguentaria; northeast of Corinth, Early or Late Roman?); Bookidis and Fisher 1972, p. 305, no. 18, pl. 59; Corinth XVIII.3, pp. 382–383, 385, no. 16 (bottle; Late Roman grave, Demeter and Kore, Acrocorinth); Pallas, Charitonides, and Venencie 1959, p. 496 (“des fragments d’un fin vase en verre”; Late Roman chamber tomb, Longopotamos valley southwest of Solomos); Kritzas 1976, p. 207 (“ὑάλινα ληκύθια”); Oikonomou 1988, p. 481; Oikonomou-Laniado 1998, pp. 405–407, 2003, pp. 14, 27, 31, 32 (Gargassoula basilica, North and East Cemeteries, and cemetery south of the city [area nos. 18, 180, 623, 649], Argos, Late Roman–Early Byzantine). 163. Oikonomou-Laniado 2003, p. 28 (“un vase en bronze sous le crâne,” tombe 9; North Cemetery [area no. 587]). 164. See Corinth XVIII.2, p. 127, no. 276, on the Corinthian assemblage.

196

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

small quantities of wine, honey, aromatic oil, or perfume for anointing the body.165 That Corinthians poured libations to the dead in late antiquity is proven by the famous case of the bishop Eustathios, whose interment in the nave of the Kodratos basilica was fitted with a pipe running diagonally from the pavement down into the burial chamber.166 Other Christian graves with inscribed and pierced cover slabs have been found elsewhere at Corinth167 and in Greece,168 apparently also for pouring liquids into the grave. A different ritual is represented by the numerous sherds comprising about one third of a cooking pot (12) that were mixed into the burial fill of NEG 69-010. The condition and quantity of the sherds show that the vessel must have been broken at the site of burial either shortly before the interment, in which case its presence inside the cist is coincidental, or during the funeral, in which case it was intentionally used as a ritual object. If the latter is the case, mourners presumably brought the pot from the kitchen of the bereaved household and smashed it either on the ground or on the pile of chalky clay next to the open grave. Once the corpse was interred, the cist was backfilled with loose clay and many of the dispersed sherds. The breaking of vessels during funeral processions and at the graveside is a widespread custom observed in many cultures both past and present, even today in Greece, Cyprus, and several European countries. The intact vessel could be used for pouring libations, anointing the corpse, or cleansing pollution from the mourners, but the significance of the act of breaking is obscure. Perhaps its breakage was intended to signify the dissolution of the body, to protect the living participants from demonic assault, or to render the dedicated vessel useless.169 In Greece the practice long predates the Modern era, but its history is difficult to trace because archaeologists traditionally have not paid close attention to fragmentary pottery in funerary contexts. The discovery of numerous joining sherds of Middle or Late Byzantine fine- or coarse-ware jugs or bowls in graves at Athens, Panakton, Moulki on Salamis, and Corinth has led excavators to conclude that these vessels were broken either during or shortly after the funeral.170 Although the analogous vessel at the Isthmus was not a drinking form, a globular pot could have contained and poured liquid. It was not as elegant as a glazed or matt-painted jug, but it would have been equally effective. The presence of cooking ware is not without parallels in Roman and Byzantine graves, but such utensils have usually been identified with a graveside banquet.171 The sherds of the cooking pot in 165. See Oikonomou-Laniado 2003, p. 41, on the Argive assemblage. 166. Stikas 1964, p. 134, fig. 4, pl. 90:β (τάφος E). Other graves and slabs inscribed with epitaphs belonging to the same phase of the basilica display similar holes: Stikas 1964, p. 133, pl. 86:β (τάφος ζ); 1966, pp. 53–54, fig. 2, pls. 49:β (τάφος ΛΒ), 51:γ. 167. Pallas and Dantis 1979, p. 67, no. 6 (dated 529, Kokkinovrysi). One tomb north of Corinth dated to the 1st to 2nd centuries contained a cist grave with a cover slab pierced for libations over the head; see Walbank 2005, p. 257, fig. 9.7, citing other Greek and Roman parallels (n. 6). 168. Sodini (1977, pp. 11–13) discusses the graves at the Kodratos basilica and notes analogous designs on Crete and at Larissa; see further Laskaris 2000, pp. 268–269. One parallel comes from the grave of the bishop Klematios in the basilica just southwest of Likavittos in Athens; see Pallas 1989, pp. 864, 866–867, fig. 2; Laskaris 2000, p. 269, n. 591, fig. A:2. 169. Polites [1921] 1975; Schmidt 1926, pp. 310–318; Megas 1940, pp. 180, 183–184; Papacharalampous 1965, pp. 152– 153; Alexiou 2002, p. 45. Grinsell (1961) provides a cross-cultural survey of the practice based on both ethnographic and archaeological evidence, including a discussion of its many possible meanings.

170. Soteriou 1917, pp. 136–138, n. 1, fig. 11, pl. 2; Travlos and Frantz 1965, p. 179 (Ayios Dionysios at the Areopagus, Athens); Gerstel et al. 2003, pp. 183, 200, 218, no. 49, fig. 36 (grave 1992-2; Panakton, Attica); Pallas 1951, pp. 164–166 (τάφος 1; Ayios Demetrios near Moulki, Salamis); Williams and Zervos 1992, pp. 162–163, no. 32, fig. 13, pl. 41; Williams et al. 1998, pp. 241–242, pl. 47:a (west of church southeast of Temple E, Corinth). The examples at Athens and Moulki are multiple burials or ossuaries that underwent frequent reuse, in which case vessels originally buried intact could have been damaged at a later date. Laskaris (2000, pp. 326–327) cites numerous other Late Roman and Byzantine graves that have produced fragmentary vessels. The most convincing evidence comes from the Corinthian graves of the late 13th century. 171. E.g., Corinth XIII, pp. 296–297, 309, pls. 77, 99 (grave 144/510; North Cemetery, Early Roman); Charitonides 1968, p. 122 (“χύτραι καὶ μαγειρικὰ γενικῶς σκεύη”; north of Cheliotomylos, Corinth, Roman); Oikonomou-Laniado 2003, pp. 29, 32, 34, 42, 43, 49, nos. 73, 78, figs. 96, 100 (“pot,” tombe 8 and “marmite,” fosse 2; North Cemetery and cemetery south of the city [area nos. 621, 649]; Argos, Late Roman–Early Byzantine); Keramopoullos 1929, pp. 126–127, 130–133 (sherds of χύτραι above graves and on surface in the cemetery; Ismeneion, Thebes, Late Roman–Middle Byzantine?). Laskaris (2000, p. 319)

MORTUAR Y FORM AND VARIABILITY

197

NEG 69-010 entered the cist when it was open to receive the corpse, and therefore the vessel could not have been used during a graveside meal, which normally followed burial.

Objects Left Outside the Grave: Lamps and Unguentaria The third category of funerary artifacts consists of objects such as terracotta lamps and an unguentarium left by mourners outside the grave. Two intact lamps have been found outside graves. One mourner carefully placed a lamp (5) on top of the southwest corner of the rubble covering NEG 69-008 (Fig. 2.10).172 Another mourner stood in the water channel before the slabs enclosing RB 76-002, turned to the south, and placed a lamp (6) on the debris that had been swept under the south wall of room II, perhaps seeking to shelter the light from the weather (Fig. 2.66). Both individuals deposited the lamps either immediately after the closure of the grave or during a later visit. In the case of 69-008, which was disturbed during the construction of the Hexamilion, one of the builders might have deposited the lamp. In the case of 76-002, the mourner could have left the lamp after the interment of any one of the three bodies. The use of artificial light during funerary rituals was a common practice in classical antiquity that continued in later times. Several Roman and Byzantine sources record the presence of lamps, candles, and torches at the bereaved home during the laying-out and funeral procession and at the graveside during and after the interment.173 Moreover, lamps have frequently been found associated with graves of Roman174 and Late Antique date175 in the Corinthia and Argolid. It is difficult to know whether these lamps, which most often occur outside the interment, were left at the graveside when the grave was made or at a later time. When they occur in large numbers, either they were deposited en masse at the conclusion of the funeral or, more likely, they accumulated over the course of several visitations. Apart from its practical purpose for illumination, the light from lamps signified passage or salvation and sanctified or even protected the burial. Moreover, the deposition of a lamp expressed the devotion of the mournful. This was probably the purpose of the two lamps

interprets cooking pots and plates from graves as evidence for funerary banquets. Nemea II, p. 132, records the discovery of two partly preserved cooking pots (P 368, P 1546, figs. 241, 242) that might have been deposited with an Early Byzantine grave in the Stadium tunnel. They came from the same context as two intact lamps (p. 132) that “framed” the disarticulated human bones (p. 134). 172. This conscientious placement compares with the situation of a lamp in a mortar and rubble grave marker over a Late Roman tomb (grave 77) in the Gymnasium area north of Corinth (Wiseman 1969, p. 83, pl. 27:e). It is hard to tell whether the lamp was built into the marker or left on top of it. 173. For discussions of the extensive testimony, see Rush 1941, pp. 221–228; Spyridakis 1950, pp. 161–162; Koukoules 1951, pp. 177, 179–180, 199–200; Abrahamse 1984, p. 130; Emmanouelides 1989, pp. 295–296, n. 93. Kenna (1991, pp. 102–103) discusses how burning candles in Modern Greek funerals (with specific reference to the Cyclades) can symbolize the soul’s freedom from bodily sin. 174. Rife (1999, pp. 276–279) surveys the published instances of lamps at Corinthian graves of Early to Middle Roman date, to which should be added Walbank 2005, pp. 275–276 (north of Corinth); Rife et al. 2007, pp. 162, 166, 168, 169, 171–172, 174, fig. 24 (Koutsongila ridge, Kenchreai); see Laskaris 2000, p. 320 (“Les lampes en terre cuite utilisées pour le culte funéraire datent exclusivement de l’époque paléochrétienne”). 175. E.g., Skarmoutsou-Dimitropoulou 1995, p. 155,

pl. 73:γ, δ (northeast of Corinth, Early or Late Roman?); Shear 1931, p. 441 (Cheliotomylos, Corinth, Late Roman); Shelley 1943, p. 178 (grave 3; Kraneion basilica, Late Roman); Skarmoutsou 1997, p. 167, pl. 54:γ (northeast of Corinth, Late Roman); Miller 1983, p. 87 (Nemea, Late Roman); Konsolaki 1987, p. 72 (τάφοι 6, 14; northwest of acropolis, Palaiokastro, Methana, Late Roman); de Waele 1933, p. 436; Corinth XIV, p. 164, pl. 67:1; Wiseman 1969, pp. 82–83, n. 33, pl. 26:c (grave 73); 1972, p. 8; Sanders 2005b, pp. 436–437 (Asklepieion, Lerna Hollow, and Gymnasium area, Corinth, Late Roman–Early Byzantine); Kristalli-Votsi 1984, p. 65; Rife et al. 2007, pp. 166, 173, fig. 25 (Kenchreai, Late Roman– Early Byzantine); Kritzas 1976, p. 207; 1979, p. 219; Oikonomou 1988, pp. 481, 500–501; Oikonomou-Laniado 1998, pp. 405–407; 2003, pp. 13–14, 27–29, 31, 34, 44–46, 49–50, nos. 86–112, figs. 106–113 (Lymberi basilica, North, South and East Cemeteries, and cemetery south of the city [area nos. 180, 247, 587, 621, 623], Argos, Late Roman–Early Byzantine). Nemea II, pp. 132–134, figs. 243–246, records that the disarticulated human bones found in the Stadium tunnel were “essentially framed” by two intact Early Byzantine lamps (L 238, L 239), but it is uncertain whether the lamps were deposited with the bones. The recorded location of the lamps indicates that they were well outside the area occupied by the bones. Oikonomou-Laniado (2003, pp. 28, 43–44, 49, no. 83, fig. 103) records an example of an incense container outside a Late Roman to Early Byzantine grave in the North Cemetery at Argos (“calice à ensens,” tombe 9 [area no. 587]).

198

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

deposited near graves at the Isthmus, both of which belonged to the Late Roman phases of burial (II–III). They were of common, inexpensive types that would have been readily available to local residents through the larger markets of Kenchreai and Corinth. The use of a lamp bearing the sign of the cross at RB 76-002 is particularly significant because it is the first instance of Christian iconography associated with a grave at the Isthmus. The other type of artifact found outside the graves at the Isthmus was an unguentarium (11). The vessel was found between the closure slabs of RB 76-002 and the south wall of room II. The clean break at the shoulder and the minute impact fracture on the lip indicate that it was dropped onto the floor of the channel by a mourner standing either in the channel or above it on the mosaic pavement. It is unknown whether the same person deposited the lamp (6) nearby to the south. This small-mouthed, flat-bottomed shape was designed to contain small amounts of liquids or viscous matter that would be poured slowly, such as water, wine, honey, oil, or ointment.176 The unguentarium from 76-002 should probably be identified with the chrismation of the dead, a ritual act that has already been discussed in connection with other vessels.177 But the possibility cannot be ruled out that the vessel was used to pour a libation at the graveside. The common series of bulbous unguentaria is well represented in Corinthian graves of the first generations after the Julian foundation, including several in the cemeteries north of Corinth and north and south of the harbor at Kenchreai.178 These ceramic unguentaria were gradually replaced by glass during the late 1st century a.d.179 Other common forms, such as the lekythos, might have assumed their role or shared it with glass equivalents in the funerary rituals of late antiquity.180 Despite these developments in the regional assemblage, the dumpy, uneven Late Roman unguentarium from the Isthmus is a clumsy throwback to the earlier series of globular, flat-bottomed shapes in terracotta. It was a local product that looked nothing like funerary lekythoi at Corinth or the Late Roman to Early Byzantine unguentaria that have been found at Corinth, Kenchreai, and Argos.181 Besides the deposition of lamps and the possible use of the unguentarium for libation, no evidence for postburial activities has been found associated with the graves at the Isthmus. Late Roman and Byzantine writers record not only periodic memorial services after the funeral and deposition but also the persistence of the ancient custom of eating and drinking at the graveside.182 These rituals might have been performed in part at home or at a church, 176. Anderson-Stojanović (1987) surveys the form, chronology, and function of terracotta unguentaria from graves throughout the Mediterranean; see esp. pp. 115–121 on their use. 177. Another class of artifacts found in Late Antique graves in the northeastern Peloponnese could have been used for chrismation. Bronze or bone utensils with long, thin shafts and a small bowl or spatula at the end have been found in Late Roman or Early Byzantine graves in the Gymnasium area (Wiseman 1967b, p. 420, n. 46, pl. 88:g [grave 42]) and the Kodratos basilica at Corinth (Stikas 1966, p. 56), near the Temple at Nemea (Miller 1981, pp. 48–49, pl. 12:d; Nemea Guide, pp. 73, 76, fig. 48 [grave 7]), and at the Liapi church at Argos (Moutzali 1985, p. 107). Corinth XII, pp. 181, 184, nos. 1318– 1333 discusses the series and their varied identifications as medical instruments, aids for personal hygiene, and cosmetic or unguent applicators. Anderson-Stojanović (1987, p. 115) suggests that a “small, long-handled spoon” might have been utilized to extract “solid perfume or unguent” from unguentaria. 178. Corinth XII, pp. 167, 294–299, 309, 327 (North Cemetery, Corinth); Walbank 2005, pp. 274–275, fig. 9.14 (north

of Corinth); Protonotariou-Deïlaki 1969, pp. 125, 127; 1970, p. 102 (“δακρυδόχοι”; Kretika, Corinth); Skarmoutsou-Dimitropoulou 1995, p. 154, pl. 73:α (“μυροδοχεία”; northeast of Corinth); Robinson 1966, p. 180, n. 3 (south of harbor at Kenchreai); Kenchreai I, p. 11; Rife et al. 2007, pp. 160, 168–169, fig. 20 (north of harbor at Kenchreai); Kenchreai IV, pp. 120– 121, nos. R 32, R 33, pl. 33 (area of harbor at Kenchreai). Rife (1999, pp. 273–275, 297–298) discusses the evidence. 179. Anderson-Stojanović 1987, pp. 113–114; see Corinth XVIII.2, p. 64, on the local chronology. 180. Sodini (1977, p. 14) notes that glass unguentaria in Late Antique graves might have been used for ointment. 181. Hayes 1971, pp. 245, 247; Slane and Sanders 2005, pp. 269, 272, 276–277, 279, nos. 3-34-39, 4-32, figs. 10, 12, 13 (Baths of Aphrodite and Panayia area, Corinth); Kenchreai IV, pp. 122–123, nos. RC 39, RC 40, pls. 33, 35, 38; cf. Agora V, p. 118, no. M 369 (area of harbor, Kenchreai); OikonomouLaniado 2003, pp. 30, 31, 44, 46, 49, nos. 84, 85, figs. 104, 105 (North and East Cemeteries [area nos. 621, 623], Argos). 182. Spyridakis 1950, pp. 166–171; Koukoules 1951, pp. 208–211; Maraval 1985, pp. 218–219; MacMullen 1997, p. 112, nn. 27, 28; Samellas 2002, pp. 279–286.

MORTUAR Y FORM AND VARIABILITY

199

if the settlement at the Isthmus had one, but they typically did involve a graveside visit. It is also possible that they were performed entirely at the site of burial but no traces have survived, particularly if the material components were perishable, such as foodstuffs.183 The presence of markers over the four Early Byzantine interments near the Northeast Gate and Tower 2 would have made it easier for mourners to find the grave for visitations after burial. As has been noted, the meaning and purpose of the distinctive shape of the constructions over NEG 69-009 and 69-010 is uncertain. One hypothesis is that the flat panel over the west end was used as a platform or table for the temporary placement of lamps, votive offerings of a more personal nature, containers for libations, or table ware for a meal. The west end of standard Modern Greek grave markers, which resemble in their basic form the Early Byzantine markers at the Northeast Gate (Figs. 3.2, 3.3), commonly serve as a sort of depository.184

Secondary Burial So far this chapter has only addressed primary burials, those containing bodies in their first and final resting place after death. Excavation at the Isthmus has uncovered two certain and possibly three secondary burials, those containing displaced and reburied human remains.185 One belonged to the Late Roman settlement (HO 70-901), and two belonged to the Middle to Late Byzantine settlement and later (LOU 69-801, TC 60-001; Figs. 2.68, 2.59, 2.60, 2.83). While the original sites of burial and the reasons for transplanting the bones are unknown, the primary burials were probably not far from the location of their discovery. HO 70-901 might have been made when the erection or renovation of the fortifications near the Bath disturbed three single interments, one single interment and one double interment, or one triple interment. The sparse bones in TC 60-001 were probably brought from some distance into the central area of the ancient Sanctuary. It is also possible that these remains suddenly turned up in a field, among ruined buildings, or even along a road, in which case it was the pious responsibility of the person(s) who found them to provide a proper burial.186 The amount of time that elapsed between the primary and secondary burials of HO 70-901A-C and TC 60-001A must have been many years because the bones were fully disarticulated upon reburial. If LOU 69-801 was a secondary burial, the corpse must have been deposited no longer than a year or two after death because much of the skeleton remained articulated. Those responsible for these burials chose sites that were isolated, though HO 70-901 and TC 60-001 would have been easy to find if the need arose. In contrast, LOU 69-801 was far from any conspicuous features in the landscape and, without a marker, would have been hard to locate after the bones were buried. Once a site for secondary burial was chosen, the remains were collected from the old grave and transported to the new one. The two definite secondary burials contained only a sample of the major skeletal elements from the primary burial, namely, crania, mandibles, innominates, and long bones. The preserved remains of HO 70-901A–C represent only 19.81% of the expected bone elements in the complete skeletons (Table 1.3, Fig. 4.2), while those of TC 60-001A probably represented a similar proportion at the time of discovery. Funerary artifacts and the small bones of the thorax, hands, and feet were lacking in both 183. For possible evidence for funerary banquets in the suburban cemeteries at Late Antique Argo, see Oikonomou 1988, pp. 500–501; Oikonomou-Laniado 1998, p. 409. 184. For a discussion of the use of such grave markers at Ayios Ioannis in Kyras Vrysi, see pp. 150–151. 185. The unique LOU 69-801 may be a secondary burial, but the unusual circumstances that caused the legs of the

corpses to be folded back over the chest remain unclear. 186. Spyridakis 1950, p. 131; Koukoules 1951, p. 196. The unexpected discovery of graves followed by the reburial of their contents is attested in hagiography: e.g., V. s. Agap. Syn. 9 (Phrygia, early 4th century); Ignat. V. s. Greg. Dec. 16 (Thessalonica, early 9th century).

200

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

Figure 4.2. Survival of skeletal elements after secondary burial in HO 70-901

interments. Since those who moved the remains did not want to spend the time to recover everything from the grave, they chose those bones that were most recognizable and portable to represent the body in its new resting place. They also might have lost bones by accident or took some for themselves during the transfer. The manner of transportation is unknown, but both the absence of vessels, lamps, and coins and the informality of the interments indicate that the event was unceremonious. Indeed, the human remains deposited in the Theater Cave were mixed with animal bones, rubble, and smashed tiles in the manner of a rubbish heap. The form of the secondary burials and the arrangement of the remains was informal but not utterly random. All three were irregular cists without lining, cover, or marker that would have required little time and effort to dig. Perhaps those who enacted the reinterment did not think that the bones required any protection because the flesh had already decomposed. The cists were roughly circular to elliptical in shape, not rectangular like other graves at the Isthmus. They were also much more compact, measuring on average ca. 1.39 m long, 0.82 m wide, and 0.40 m deep. Their shape and size are explained by the simple fact that they did not contain intact bodies. Nonetheless, just as in the multiple burials, the few bones that were redeposited were placed according to their relative positions in a supine, extended corpse. In HO 70-901 and TC 60-001 the remains of the heads were placed at the south end of the grave and the pelvic and appendicular remains were placed to the north, while the leg bones in HO 70-901 were lined up axially. This treatment shows that those who handled and interred the bones imagined the form of the human body, albeit incompletely.187 The

187. Compare the treatment of human remains during exhumation in the modern villages of Inner Mani, where mourners handled the skulls most carefully and often situated them

atop heaps of disarticulated bone (Seremetakis 1991, pp. 182, 189–190).

MORTUAR Y FORM AND VARIABILITY

201

heads in the secondary burials, including LOU 69-801, were placed in the south end of the cists, not the west as was typical in the primary burials of the Late Roman and Byzantine eras. Perhaps this blatant deviation reflects a notion that the exhumation of the primary interment revoked any need to preserve the normal orientation when redepositing the remains. There is no sign that, once the bones were reburied, the graves were visited or maintained. Textual and archaeological sources attest to the practice of secondary burial in a variety of forms. Numerous hagiographies of the 9th to 10th centuries recount “the translation of the relics” (ἡ μετάθεσις τῶν λειψάνων) of holy persons after their deposition for a period of time in temporary graves, often a year.188 The corpses were at first commemorated in the primary burial and then carefully handled and lavishly mourned upon transport to the secondary burial. This special treatment reflects the intrinsic sanctity of these individuals and the residual power of their remains. Secondary burial, however, was not restricted to holy men and women. Byzantine legal documents address the common ritual of moving bones from one grave to another or to an ossuary, and the practice of secondary burial might well have been more widespread than Byzantine literature indicates.189 In Modern Greece, families, particularly in rural communities, still observe the custom of “digging up” (ἀνόρυξις, ξέχωσμα, ξεχώνιασμα), “opening” (ἄνοιγμα), or “bringing back” (ἀνακομιδή) the mostly defleshed bones of relatives, cleaning and lamenting them, and redepositing them in boxes or larger depositories.190 While secondary burial obviously occurred at the Isthmus, the existing remains do not show that it was a common practice in the local community during the Roman and Byzantine eras. Its relative infrequency at the site supports its interpretation as a thoughtful reaction to the accidental discovery of earlier graves, rather than standard treatment for the bones in primary burials, as in recent Greek communities. In contrast to the customs attested in Byzantine textual sources, the human remains in secondary burials at the Isthmus were not treated with great care. Indeed, the manner of burial of LOU 69-801A and TC 60-001A was almost antithetical to that of saints and clergy. The bones were probably not even transplanted by bereaved relatives, who would have been more concerned to collect all the remains, to deposit them neatly and without corruption by refuse, and to protect them adequately in a new resting place.191 It is hard to evaluate the distribution and significance of secondary burial across Late Antique and Byzantine Greece, because the material evidence has not been consistently identified and fully recorded. Certainly it was practiced in the Corinthia and Argolid but apparently with infrequency. The published descriptions of certain Late Roman and Byzantine graves at Corinth, Nemea, and Halieis suggest that they contained secondary

188. E.g., Method. V. s. Theoph. 57–58 (d. 817, Samothrace; vita written early 9th century); Mich. mon. V. s. Theod. Stud., PG 99.229–231(d. 826, Principis or Chalcitis island, Marmara; vita written late 9th century); Sab. mon. V. s. Petri Atr. 95–97 (d. 837, Atroa; vita written mid-9th century); Nicet. Dav. Paph. V. s. Ign., PG 105.560 (d. 877, Constantinople; vita written end of the 9th to mid-10th centuries); Greg. cler. Trans. s. Theod. Thess. 3, 6–7 (pp. 38.29–39.12, 40.20–41.19; Thessalonica, end of the 9th century); V. s. Iren. heg. Chrys. 9.90 (AASS Iul. 6.634A; d. ca. 940, Chrysovalantou Monastery. Constantinople; vita written late 10th–early 11th centuries). Abrahamse (1984, pp. 132–133) discusses the Byzantine evidence. There was a long pagan tradition of translating human remains; Pfister ([1909– 1912] 1974, pp. 188–211, 433–444) still provides a fundamental discussion of the ancient practice. The relocation of bones in Europe during the Late Middle Ages and later is well known

among dynastic royalty and clergy in cases of political honor and canonization (Finucane 1981, pp. 53–54; Weiss-Krejci 2001, pp. 775–778; 2005, p. 168). 189. Emmanouelides 1989, pp. 379–383 (Byzantine law); Angold 1995, p. 454, n. 62 (suggesting that the ritual of exhumation is scarcely documented in Byzantine literature because it was “very largely a family matter”; cf. n. 147, above, on the absence of testimony concerning funerary coins). 190. Polites [1931] 1975, pp. 355–356, n. 6; Megas 1940, pp. 203–204; Kenna 1976, p. 32, 1991, pp. 106–110; Danforth and Tsiaras 1982, pp. 15–23, 85, 86, pls. 24–31; Seremetakis 1991, pp. 177–212; Alexiou 2002, pp. 47–48. 191. Consider the care with which modern Greeks exhume every bone, even those of the hands, from the grave of a relative or neighbor (e.g., Danforth and Tsiaras 1982, pp. 19–20, pls. 27–29).

202

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

burials. These interments share several salient features with HO 70-901, TC 60-001, and in some respects LOU 69-801: the skeletons are incomplete; the bones that are present are disordered; the grave is oriented north–south or south–north; the cists are small, shallow, elliptical to circular in shape; and they are completely or almost completely uncovered and unlined.192 If these are secondary burials, they represent a rather different practice than the large “ossuaries” (ὀστεοθήκαι, ὀστεοφυλάκια), massive depositories for exhumed bones, that sometimes developed in preexisting chamber tombs during late antiquity.193 Chambers serving as ossuaries were also erected near churches at Byzantine Corinth.194 The deposition of bones in ossuaries may be viewed as a continuation of the Roman practice of multiple burial in the monumental context of the chamber tomb. Mourners who used Byzantine ossuaries might have also participated in a funeral liturgy at the adjacent church. Unlike the Corinthian ossuaries, the secondary burials at the Isthmus, as well as the comparable graves at Corinth, Nemea, and Halieis, were informal, solitary interments. They did not introduce the bones into a community of the dead, and they probably did not involve a ritual of translation.

Human Disturbance of Graves One mortuary behavior at the Isthmus that did not occur during funerals was the human disturbance of graves after their creation.195 This occurred during two periods of intensive occupation, when the fortifications were erected in the early 5th century and when houses were built along the walls of the Fortress in the 13th century or later. Four graves were disturbed by the builders who dug the footings for the Hexamilion south of the Northeast Gate and nearby Tower 2 (Figs. 2.1, 2.4–2.8, 2.10, 2.36, 2.42).196 The builders replaced the coverings over the four graves and moved the funerary artifacts and bones in at least one of them, redepositing the crania to the west of a heap of postcranial elements (69-004). It is uncertain whether the builders had any foreknowledge of these graves, which were made shortly before the fortification effort. Likewise, men constructing walls just west of the Fortress during the Late Byzantine period inadvertently disturbed at least four and as many as seven Early Byzantine graves, which were not far below ground level at the 192. Corinth XVIII.3, p. 389, plan 1 (fragmentary leg bones in roughly oval pit, covered with slab; Demeter and Kore, Late Roman?); Miller 1983, p. 87, pl. 26:e (one skeleton in small, circular, unprotected cist oriented north–south, bones is disarray; south of Bath, Nemea, Late Roman); “Halieis V,” p. 300 (remains of two skeletons in cist, several bones missing, others damaged, many in disarray, grave 1; Early Byzantine); Anderson 1967, p. 7, fig. 3 (pit in corner above earlier tomb with only part of a skeleton, grave 4; northwest of Temple E, Middle–Late Byzantine). Nemea II, p. 134, fig. 247, records the discovery of disarticulated human bones from one skeleton scattered over a poorly defined area in the Stadium tunnel; see also Nemea Guide, pp. 75, 110, 195. These remains might well have been redeposited in the underground passageway from elsewhere during the Early Byzantine period. As in TC 60-001A, animal bones were present, though it is unclear whether they belonged to the same deposit as the human remains. Moreover, several skeletal elements were missing, but the remaining bones traced a roughly anatomical pattern, with the cranium facing west at the south end and, extending to the north, a rib, vertebrae, and upper and lower leg bones. The theory of S. G. Miller that “the corpse must have been deliberately and violently dismembered” (Nemea II, p. 134) or that the body was “pulled apart” (Nemea Guide,

pp. 75, 110) seems to have no basis. Moreover, the healed cranial lesion provides no osteological evidence for violent death, as he implies (Nemea II, p. 134, n. 320). The presence of possible funerary objects surrounding the bones (coins, cooking pots, lamps) would distinguish this secondary burial at Nemea from the more informal TC 60-001. 193. See n. 50, above. 194. E.g., Robinson 1968, p. 219 (Ayia Paraskevi, Byzantine); Robinson 1976a, pp. 221–222, pl. 48:b; 1976b, p. 256 (Temple Hill basilica, Early–Middle Byzantine); Corinth XVI, pp. 127– 128, fig. 12, pl. 17:2 (Peirene fountain, Byzantine); Pallas 1974, p. 213 (Late Roman burial vaults later converted into ossuaries, Kraneion basilica, Byzantine). Catling and Smyth (1976) publish a 7th- or 8th-century ossuary west of the Venizeleion, northwest of Knossos. Di Vita (1988a, pp. 107–109, figs. 106–109) and Mallegni (1988, pp. 341–342) recorded that tomb 17 in Settore L at Gortyn (early 7th century) contained the remains of 13 individuals, including nine crania that had been separated from skeletons and redeposited there. 195. On the disturbance of graves during building as an anthropogenic formation process of the material record at the Isthmus, see pp. 15–16. 196. NEG 69-103, 69-004, 69-008, T2 68-003.

MORTUAR Y FORM AND VARIABILITY

203

time.197 These builders could not have known of the graves that predated their activities by several centuries. The partial or total exposure of every interment near Towers 13 and 14 probably resulted from such disturbance. Those who opened the graves displaced two leg bones and the mandible of one skeleton (54-001), lifted the cranium of another and repositioned it over the cervical vertebrae (67-002A, Figs. 2.46, 2.47), removed the left humerus and several bones from the right foot of another (67-004A, Fig. 2.49), and removed the right innominate from another and placed the right femur alongside the vertebral column (69-002A, Fig. 2.54). This handling and removal of bones resembled the treatment of skeletal remains when bodies were added to multiple interments and bones were transplanted for secondary burial. The conscientious treatment of old graves at the Isthmus expresses a respectful attitude toward the dead among the living. The graves were not purposefully dug out and forcefully opened to plunder or vandalize their contents. The uncovering of at least six and as many as 11 interments could not be avoided during construction. Moreover, the bones were not dispersed during the looting of funerary objects. The calculated rearrangement of disarticulated crania and limbs reflects a concern to maintain the order of the buried remains. While it is impossible to determine whether those who opened the graves took away artifacts, in a few cases they obviously did not, such as the glass vessels (7, 13) that were dutifully placed with the heap of bones in NEG 69-004. Finally, bones were not thoughtlessly extracted in haste or in large quantity. The deliberate removal of either small, tractable bones (phalanges, metatarsals, tarsals) or large, identifiable bones (innominate, humerus) demonstrates that those who opened the graves carefully selected and lifted bones so as to preserve skeletons in a mostly articulated state. While the exact nature of the disturbance varied from grave to grave, the response to finding a grave was fundamentally the same during both the Late Roman and the Late Byzantine periods. The treatment of the interments interrupted by the Hexamilion was more invasive than the treatment of those underlying the buildings west of the Fortress, but the contents of all graves were replaced in an orderly fashion. The careful displacement or removal of bones was more common in the later graves at Tower 14 than in those at the Northeast Gate. However, in at least one case (NEG 67-003A) mourners reopening a multiple burial of Late Roman date removed the skull from a skeleton previously interred there. Moreover, it is quite possible that bones were retained during the transportation of human remains for secondary burial. Similar treatment of graves can be found in the archaeological record across the region. Skeletons with missing or moved bones, usually from the head and the limbs, have also been found in graves from Late Roman to Early Modern times at Corinth and Tiryns.198 Apart from these few examples, the displacement or removal of interred bones is poorly documented in Greece.199 This oversight probably results either from a lack of osteological expertise on most excavations or from a general disregard for the study of human remains in their funerary setting. In any case, the general pattern at these disturbed graves points 197. T13 54-001, T14 67-002, 67-004, 69-002. 198. Corinth XVIII.3, p. 386, no. 21 (mandible of child in anatomical position but cranium displaced over pelvis; Demeter and Kore, Acrocorinth, Late Roman); Robinson 1962, p. 110, n. 45; Gejvall and Henschen 1968, figs. 2, 3 (top skeleton in double burial missing both feet and distal right tibia and fibulae, while left tibia and fibula displaced over chest but under right arm; West Shops, Corinthian Forum, Late Roman–Early Byzantine); Kilian 1980, p. 286, pl. 49:3 (humeri of skeleton displaced over head, left ulna and radius missing;

Unterburg, Tiryns, Middle Byzantine); Barnes 2003, p. 437, fig. 26.5 (displaced skulls and leg bones interred over primary burials; church southeast of Temple E, Corinth, Frankish). This practice has also been observed in Ottoman or Early Modern graves excavated in the Panayia area at Corinth (G. D. R. Sanders, pers. comm.). 199. It is worth comparing the disturbance of interments in Medieval French abbeys, which involved both the piling up of bones unearthed in narrow spaces with dense burials and the selective extraction of skeletal elements (Naji 2005).

204

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

to controlled, focused, and mindful behavior that would not have left the area disordered, littered, or torn apart. This contrasts with the activities of disturbance in the cemeteries of rural Greece during the Modern era, including those in the Corinthia. It is not uncommon for new graves to deteriorate within a generation or two without active protection as the contents are scattered and the compartments reused. These conditions seem to result from financial and spatial pressures in the management of burial sites, as well as familial neglect and possibly a lack of respect for or identification with the community.200 Apparently this phenomenon did not pertain to the settlement at the Isthmus or to others in the region during earlier times. The disturbance of graves and the handling of human bones are well attested in Late Roman and Byzantine historical sources. The technical term for the disturbance of burials was τυμβωρυχία, literally “grave-digging.”201 The word was used for a range of activities encompassing the removal of stones, marble pieces, columns, or structural elements, the displacement or extraction of “remains” (λείψανα), and the “corpse-robbing” (νεκροσυλία) of items that may bring profit, such as money, jewelry, and clothing. The Church Fathers condemned these activities as impious and immoral,202 and Civil and Canon Law dictated harsh penalties for their practice from late antiquity until the last Byzantine centuries.203 The repeated promulgation of these laws suggests that the actions they prohibited were regular occurrences. The disturbance of graves and human remains was enacted on many levels in Late Antique and Byzantine society. There was an ancient tradition of defiling the burials of one’s military, political, or religious enemies by opening them and destroying their contents, especially the bones, which were scattered, burned, or crushed and cast to the wind.204 Among later perpetrators were the emperors Julian, who desecrated the remains of St. Babylas near Syrian Antioch and of St. John the Baptist and the prophet Elisha at Sebaste in Palestine,205 and Constantine V, who reputedly wrecked holy relics in his persecution of iconophiles.206 Late Roman and Middle Byzantine writers record other cases of graves having been opened and the bones of criminals or outcasts dispersed.207

200. Tzortzopoulou-Gregory 2008, pp. 332–333, 335–338, 340. Samothrace XI.1, p. 8, gives an ancient parallel for such neglect in the Archaic to Roman southern necropolis on the island, where mourners would indiscriminately churn up and discard the remains of earlier burials. 201. See in general Petrakakos [1905] 1971, pp. 127–129; Loukatos 1940, pp. 86–87; Koukoules 1951, pp. 202–203. Emmanouelides (1989, pp. 399–432) gives a close discussion of the definition of τυμβωρυχία. 202. E.g., Tert. Apol. 37.2; Bas. Caes. Ep. 217, Can. 66 (PG 32.800); Greg. Naz. Epigr. 38, 68 (PG 38.103, 38.117); Greg. Nyss. Ep. can., Can. 7 (PG 45.233–235); John Chrys. Bab. 2 (PG 50.531), Ec. hom. XI, div. et paup., PG 63.642, Exp. in Ps. XLVIII 48 (PG 55.239), Hom. XXXIV in ep. I ad Cor. 6 (PG 61.293), Hom. LXXXV in Io. 4 (PG 59.465). 203. E.g., Cod. Theod. 9.17; Cod. Just. 9.19; Dig. Just. 47.12; Nov. Val. III 23; Basil. 60.23 (late 9th century); Nov. Leon. 96 (I, p. 163; late 9th century); Epan. 52.50 (VI, p. 205; early 10th century); Proch. 39.57 (II, p. 223; early 10th century); Ec. ad Proch. mut. 37.18 (VI, pp. 299–300; mid-10th to mid-12th centuries); Mich. Attal. Jur. 35.59–62 (VII, p. 467; late 11th century); Syn. bas. min. T 15–18 (VI, p. 524; late 13th century); Matt. Blast. T 10 (PG 145.168–169; early 14th century). Rebillard (2003b, pp. 74– 83) addresses the Roman legal tradition and the Late Roman constitutions; Emmanouelides (1989, pp. 454–508) generally

discusses the punishment of τυμβωρυχία. 204. Petrakakos [1905] 1971, pp. 49–50 (e.g., Plut. Alex. 77, Paus. 1.9.8); Emmanouelides 1989, pp. 414–424; Lindenlauf 2001, p. 92. 205. John Chrys. Bab. c. Iul. 14 (PG 50.554); John mon. Pass. s. Artem. 57 (PG 96.1304–1305). 206. Const. T. Trans. s. Euph. Chal. 10; Theoph. Conf. a.m. 6258 (pp. 439–440); Niceph. Antirrh. 2.4–5 (PG 100.341–344); Theost. V. s. Nicet. heg. Med. 4.28 (AASS Apr. 1.260D). Emmanouelides (1989, pp. 414–416) addresses the emperor’s actions. Wortley (1982) evaluates the tradition but doubts that Constantine V actively pursued the “destruction of relics” (λειψανοκλασία). Regardless of whether Constantine actually carried out such hostilities, authors of the very late 8th or 9th century could readily imagine such behavior and considered it appropriate to his character. 207. [John Chrys.] Hom. suppl. III de eleem. 11 (PG 64.441); Theoph. Conf. a.m. 6235 (p. 420; the patrician Bactangius, ca. 772–773, Constantinople); cf. Zonaras 15.5.19–20 (III, pp. 267.14–268.2). Koukoules (1951, pp. 197–198) discusses such actions. Targeted desecrations are well documented in later settings of war or insurgency in western and central Europe, such as the French Revolution (Brown 1985; Weiss-Krejci 2001, p. 778; 2005, p. 169).

MORTUAR Y FORM AND VARIABILITY

205

Apart from these noteworthy instances, the most common setting for τυμβωρυχία would have been the rather less spectacular daily life of local communities. Some Byzantine graverobbers, like their modern counterparts, must have planned and coordinated their operations. Most, however, would have been petty thieves or enterprising farmers. They worked to find graves, to open them, and to take items for sale, especially the adornments of the corpse. In most cases their activities would have gone unnoticed and unpunished; even if the culprits were known, it is doubtful that local authorities always expended the effort to prosecute.208 Moreover, just as today, Late Roman and Byzantine workmen digging ditches, cutting foundations, and plowing fields inadvertently disrupted graves. On such occasions they would have inspected the remains and even taken a few things. It must have been a common occurrence in long-settled regions where grave markers and epitaphs were knocked over or recycled, interments forgotten, and buildings sometimes erected in earlier burial zones.209 Such unintended disturbance often would have gone without penalty. If the bereaved wanted to commemorate dead relatives for any duration but recognized the real risk of plundering and unauthorized reuse, perhaps their best hope was to display a prohibition or imprecation against violation. Greek epitaphs of this type were frequently used in late antiquity, and several of Roman or Early Byzantine date have been found at Corinth and Argos.210 It is unclear whether this measure was ever successful and, if so, for how long, especially when protection was enforced merely by a tone of moral censure or a threat of divine sanction.211 The disturbance of graves at the Isthmian fortifications should be viewed within this sphere of interactions between living communities and old graves. The opening of graves, removal of stones, and displacement of bones at the Northeast Gate and Towers 2 and 14 were not motivated by a desire either to desecrate peaceful graves or to plunder buried treasures. These activities followed the unintentional discovery of graves during building, a regular event at sites like the Isthmus with a long history of occupation. The situation of T14 67-002 and 69-002 immediately underneath later foundations can be compared with several instances of Middle to Late Byzantine houses built directly over Late Roman to Early Byzantine graves at Corinth and Nemea.212 This kind of disturbance was regulated by law as a form of τυμβωρυχία, but the builders probably were never punished. Presumably they had opened graves before and treated them in a similar manner, and one must wonder whether they even felt shame or feared retribution. It seems doubtful that builders with practical aims even considered their removal of covering slabs or walling stones destructive or disrespectful once the bodies they enclosed were skeletons.

208. George heg. (V. s. Theod. Syc. 116) relates how a farmer at Eukraa in western Galatia during the late 6th century was penalized for digging into a hillside grave either to prepare for building or to hunt for treasure. Mitchell (1993, pp. 148–149) discusses the historical circumstances. 209. The unexpected discovery of graves or bones beneath houses, at building sites, or in remote parts of the countryside was a topos of ancient Greek ghost stories and novels (Rife 1999, pp. 134–135). 210. Corinth: Corinth VIII.1, nos. 108, 135, 136; Corinth VIII.3, nos. 535, 539, 620, 636, 643, 644, 660; IGCB I 28; Pallas and Dantis 1979, pp. 70–71, no. 9. Argos: Oikonomou-Laniado 2003, pp. 51–57, nos. 1–4, 9, 10, 14, 15, 19–23, 26, figs. 114– 117, 120, 124–126. Sironen (1997, p. 120, n. 12) collects several examples from Late Roman and Early Byzantine Attica. 211. Dantis (1983) and Emmanouelides (1989, pp. 509– 531) survey the form and function of prohibitory epitaphs in

Greek tradition; Rebillard (2003b, pp. 86–95) discusses the epitaphic protection of tombs in late antiquity more generally. 212. E.g., Corinth XVI, p. 41 (stairs of Middle Byzantine house directly over Early Byzantine tile-covered grave; east of Peirene fountain, Corinth); Anderson 1967, pp. 6–7, figs. 2, 3 (13th-century house directly over 6th-century vaulted tomb and 11th-century cists [graves 2–3], disturbing contents; northwest of Temple E, Corinth); Williams, MacIntosh, and Fisher 1974, pp. 11–12, figs. 2, 3 (late 9th-century house with rubble walls directly over manhole into 6th-century vaulted tomb; west Forum, Corinth); Miller 1977, p. 3, fig. 3 (13th-century house with rubble walls directly over 5thor 6th-century cists, displacing coverings and removing bones; south of Temple, Nemea). Zygouries, pp. 39–40, pl. 34, records that walls “perhaps of mediaeval construction” were built over a Middle Helladic grave, disturbing the skull (tomb I).

206

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

Although those who disturbed graves were not seeking costly objects, they were not disinterested in what they found. Both their reverent disposal of the many bones they left behind and their conscientious extraction of single bones cannot be explained by reference to existing laws de sepulchro violato. Local residents considered these human remains inherently valuable. Their special treatment of the bones most closely resembles the manipulation of holy relics, which will be discussed below. In fact, those bones that received the most attention when graves were disturbed at the Isthmus are the same parts of the body that were typically collected as the remains of saints. An overwhelming majority of the venerated skeletal remains of eastern saints represent the head, jaws, hands, and feet, while shoulder blades and long bones are less common.213 Thus, the disturbance of earlier graves by builders at the Isthmus technically constituted τυμβωρυχία but qualitatively resembled the invention of relics. Since these graves were not loca sancta, the question of what specific value residents found in them remains. They might have handled and selected bones because they appreciated their antiquity, especially in the case of the Byzantine builders west of the Fortress. Another explanation is that they treated bones with reverence because they felt a personal attachment to them. In particular, builders might have considered graves they had opened to contain the burials of their own ancestors and therefore thought them worthy of reverence. This seems most likely at the Northeast Gate, where the laborers who dug into recent graves might have been enlisted from among the local inhabitants. Perhaps they even knew to which families the interments belonged.

SOCIAL STRUCTURE AND IDEOLOGY The physical remains of funerary ritual at the Isthmus provide important evidence for the social structure and ideology of the local community over time. The suite of behaviors and objects surrounding death and burial together symbolize a prevailing order and define the positions of both the deceased and the bereaved within that order by recreating their identities. This order is the social structure, the system of association and interaction between persons or groups that is recognized by the community. The social structure, however, does not inform all social activities on a daily basis; it represents, or even idealizes or distorts, the actual relationships that underpin the practical operation of the community. Furthermore, funerary ritual can serve as an arena for the expression or contestation of a particular ideology, such as a religious belief, a moral philosophy, a political agenda, or a hierarchy of power relations.214 In order to read social structure and ideology in the burials at the Isthmus, it will be useful to consider four fundamental dimensions of social identity: age; gender; vertical position, or status based on wealth, political authority, or religious superiority; and horizontal position, or identification with a kinship, professional, ethnic, or religious group. The manifestation of these dimensions in funerary rituals sometimes foregrounds single factors of identity but 213. This observation is based on the data in Meinardus 1970, which catalogues 3,602 relics from 476 eastern saints residing in churches, monasteries, and other institutions in Greece, Asia Minor, Cyprus, Palestine, and Egypt. The removal of the skull of NEG 67-003A after interment in the mid- to late 5th century can be compared with the absence of the cranium and much of the mandible from the Late Roman “bishop’s tomb” (tomb 8) in the Christian basilica at the Venizeleion, northwest of Knossos. The excavator proposed that Christians had taken the head from the grave as a relic (Frend and John-

ston 1962, pp. 195–197, pl. 49:b). Graves containing skeletons without crania have also been found at Kissamos on Crete (Late Roman; Drosinou 1997, p. 580) and Philippi (Middle Byzantine; Pelekanides 1977, p. 94, pl. 88:β). 214. The relationship between funerary ritual, social structure, and ideology has been widely discussed in the theoretical scholarship; see, among others, Hodder 1982, pp. 139–146; Pader 1982, pp. 54–56; Morris 1992, pp. 3–8; Beck 1995b, pp. 169–172; McHugh 1999, pp. 12–17; Parker Pearson 1999, pp. 23, 32–33.

SOCIAL STRUCTURE AND IDEOLOGY

207

more often combines them.215 The material expression of each dimension can be sought in the qualitative, quantitative, compositional, and spatial features of the mortuary remains addressed in the previous section. Variation in the remains reflects either synchronic complexity or diachronic change in the social structure. Situating the burials within the archaeological and historical context of the local settlement and Corinthian society will shed light on the nature of this social structure. The written sources, which record contemporary attitudes toward death and burial, will also contribute to understanding the ideological significance of ritual behaviors.

Age Funerary rituals in many cultures represent separate age categories through differential treatment of the dead. As an individual grows from infancy to adulthood, he or she acquires different social identities. These successive stages are sometimes reflected in the choice of funerary objects and dress, the location and form of the grave, and the arrangement of the body. Moreover, the relative cost and uniqueness of subadult burials can reveal the presence or absence of social stratification or horizontal complexity.216 The burials at the Isthmus reveal no fundamental distinctions by age during the Roman and Byzantine eras. The same designs, materials, and techniques were used in the graves of infants, children, adolescents, and adults, though the size of subadults’ cists were proportionately smaller than those of full-grown individuals.217 This homogeneity contrasts with the situation in urban cemeteries, where children were sometimes interred in a type of grave specific to their age: amphoras. The only perceptible age-linked differentiation among the burials at the Isthmus is that the graves of infants were generally better sealed than those of adults.218 This pattern should not be interpreted as a mark of differential status because the composition and complexity of the interments of adults and subadults were similar. Moreover, the construction of the relatively small grave T14 69-991 would have required less material, time, and effort than the construction of adjacent graves for adults. An alternative explanation is that the secure closure of cists containing infants was intended to shelter the child as precious or delicate in death. The protective treatment of subadult corpses was thus an emotional response by the parents who oversaw the interment. Such an expression of concern also validated the importance of the nuclear family by enacting the nurturing relationship between parents and their children. Subadults and adults occupied the same burial areas and displayed the same range of corporeal positions. No separate cemetery existed for children such as those found, for instance, at Late Roman sites in Macedonia and at Athens.219 It is noteworthy that many adult women (12) and subadults (16)220 and few men (seven) were interred in the Late Roman 215. See Binford 1971, esp. pp. 14, 17; Carr 1995; McHugh 1999, pp. 18–61. Pearce (2000, pp. 6–7) likewise discusses the variables of funerary ritual in Roman burial archaeology. Carr (1995, pp. 129, 133–134, table IV) records other practical, circumstantial, and physical determinants of mortuary practices across cultures. For the most part, these factors were either less influential than the four major dimensions or absent among the graves at the Isthmus. The practical concern to bury the corpse shortly after death in a protected cist that was accessible to mourners informed the standard funerary ritual. The nature of the secondary burial of HO 70-901 and TC 60-001 and the disturbance of the graves at the Northeast Gate and Tower 14 were determined by the circumstances of their unexpected discovery, not the age, sex, social position, or affiliations of the deceased.

216. Scott 1999; Gilchrist 2004; Lucy 2005. 217. Scott (1999, p. 124) stresses that neonates can receive very different mortuary treatment than older children. Apart from the relative tightness of the enclosure, no significant differences are apparent in the burials of subadults of different ages at the Isthmus, most of whom were interred in multiple burials. 218. This observation assumes that DEC 69-902 and NEG 69-008 originally contained subadults. 219. Laskaris (2000, p. 288) cites conglomerations of burials for children at Thessalian Thebes (modern Nea Anchialos), Vergina, Philippi, Rentina, and Athens. 220. This calculation assumes that NEG 69-008 contained at least one subadult.

208

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

graves inside the Northeast Gate, but the meaning of that pattern is unknown.221 The sample is not so large and the distribution by age and sex is not so skewed as to prove that this area was demarcated as a burial ground for women and children. Likewise, the presence of only one infant burial among at least five adult burials near Tower 14 cannot prove that infants were segregated in death or left without formal interment during the Early Byzantine period. The basic impression that men, women, and children were interred together, not separately, is supported by the distribution of skeletons by age and sex in the Late Roman multiple burials and in the three Early Byzantine graves outside the Northeast Gate. In addition, both subadult and adult bodies were situated in a supine and extended position for the laying-out, funeral, and deposition, and the orientations of their interments are similar. The shift to a more regular treatment of the head and arms and a more consistent alignment during the Early Byzantine period is evident among the graves of the young and the old alike. Both age groups were interred with the same classes of artifacts. A lamp (5) was probably placed at the graveside of a subadult inside the Northeast Gate (69-008), just like one (6) at the interment of three adults in the Roman Bath (76-002). Three multiple burials of men, women, and children contained vessels (7, 8, 10, 13) that were initially deposited with one corpse but became communal offerings for all corpses in the cists, regardless of age.222 If the sherds of the cooking pot (12) in NEG 69-010 represent a smashed funerary object, they demonstrate that pottery could be used in the funerals of infants. Children also wore the same kinds of personal articles as adults—rings, beaded ornaments, buckles. Certain objects, such as the tiny bronze ring (23) probably worn on the finger of a child in NEG 69008 and the diminutive iron buckle (30) worn by the infant in NEG 69-009, are miniaturized versions of articles worn by adults. Other objects, such as the heavy iron buckle (18) worn by the child at the top of NEG 67-001 and the necklace (20, 21) worn by the infant in T14 69-991, were full-size. Regardless of scale, all objects that parents chose for burial with their children were drawn from the world of adults, not the world of children, like the toys that have been found in Early Roman graves north of Corinth.223 Two infants in Early Byzantine graves, NEG 69-009A (30–32) and T14 69-991A (20, 21), wore as many articles as the welladorned PAL 56-001 (17, 33), but the objects were small and simple. These burials cannot be identified as the rich interments of infants who had been ascribed high social status because of a prominent lineage.224 The use of personal articles normally worn by older individuals would have instead marked the acceptance of these children into the adult realm. It could also be viewed as overcompensation on the part of adult mourners for an untimely death. The burials at the Isthmus therefore display no major differentiation between subadults and adults in grave type and location, corporeal position, orientation, or funerary assemblage. It is impossible to know whether the essential form of funerals for both age groups was also the same, though certain personal behaviors, such as the expression of emotion, probably varied in tone and intensity depending on the age of the deceased. The pattern contrasts sharply with the symbolic differentiation of subadults, especially infants, and adults 221. A similar preponderance of women and children has been noted in Late Roman or Byzantine graves around Argos (Kritzas 1979, p. 219; Oikonomou 1988, pp. 501–502; Oikonomou-Laniado 1998, pp. 406–407) and over the abandoned Sanctuary of Demeter and Kore on Acrocorinth (Corinth XVIII.3, pp. 391, 440). While it is possible that residents exterminated some children by exposure, particularly those with birth defects, the presence of children in graves rules out the possibility that they were exposed or sold into slavery in large numbers, as has been suggested for the Late Antique popu-

lation buried at Lerna Hollow in Corinth (Wesolowsky 1973, pp. 346–347). 222. NEG 69-004, 67-001, HO 70-902. Builders deposited the glass vessels (7, 13) above the bodies in 69-004 when they disturbed the grave. 223. Shear 1930, pp. 429–431, figs. 19, 20 (north of Cheliotomylos, 1st century a.d.). 224. See Binford 1971, pp. 21–22; J. A. Brown 1981, p. 29; 1995, p. 8.

SOCIAL STRUCTURE AND IDEOLOGY

209

that has frequently been observed in the burials of prehistoric and historic date in many regions, including Classical Greece and Republican and Imperial Italy.225 One explanation is that the inclusion of children among adult burials was a consequence of Christianity, which not only required the proper interment of all believers for bodily resurrection but also taught respect and care for the young. This theory has been proposed for Late Romano-British and Early Medieval cemeteries, where the frequency of graves containing infants is significantly higher than in pre-Christian burial grounds.226 Religious belief cannot, however, account for the high representation of subadult burials at the Isthmus. Children were interred in larger burial zones as early as the end of the 4th century (NEG 69-004, 69-008), and an earlier grave (DEC 69-902) probably also contained a subadult. While the exact course of the Christianization of the rural Corinthia is difficult to trace,227 these burials at the Isthmus would have belonged to either a pre-Christian or a transitional community. In addition, numerous infants and children were buried in tile-covered cists and amphoras in pagan cemeteries at Corinth during the Early Roman period.228 Adults and subadults were therefore already integrated in death by the time Christianity was widely adopted in the region. It is also significant that this pattern did not change during late antiquity and the Byzantine era.229 The continued inclusion of child burials would have accommodated the Christian ideologies of death and salvation, but eschatological needs were not the sole or prime factor in the uniformity of funerary ritual among individuals of all ages. The best explanation is that local residents recognized children of all ages as full members of the community.230 The burials of subadults do not communicate a unique experience of childhood. Instead, children were born into the sphere of juveniles and adults, donning similar articles of clothing and adornment, displaying the religious beliefs of their family, and even sharing their graves. This integration points to the centrality of the nuclear family in the social structure. The choice by the bereaved family to bury a child in a usual manner and the protective gesture of carefully enclosing the dead child indicate the close involvement of parents in their children’s lives, the cohesion of the family, and the intrinsic value of children for parents. These relationships fit the socioeconomic model for the Late Roman to Early Byzantine settlement based on habitational remains (Chap. 3). Parents depended on their children from a young age to participate in subsistence activities, such as feeding livestock, gardening, gathering, and herding sheep, goats, or pigs. Together they occupied the same small house and shared its limited resources. Those institutions that marked the maturation of children in the more prosperous families of large urban communities during classical antiquity and the Byzantine era, such as education, cohort associations, civic ritual,

225. E.g., McHugh 1999, pp. 20–21; Parker Pearson 1999, pp. 102–104 (general comments); Young and Papadatou 1997, pp. 192–195 (modern cultures); Scott 1999, pp. 90–123; Kamp 2001, pp. 6–8 (survey of evidence); Lucy 1994; Crawford 2000 (Romano-British and Anglo-Saxon cemeteries); Garland 2001, pp. 78–86 (Greece); Dixon 1992, pp. 99–100; Pearce 2001, pp. 126–127; Rawson 2003, pp. 336–363 (Italy); Samellas 2002, pp. 11–16 (Roman world, esp. the East). One anomalous burial chamber of the 6th to 7th century that was located just east of the basilica at the Knossos Medical Faculty site on Crete (tomb 223) contained 14 children under the age of 12 and one adult; its significance is unknown (Becker 2005, pp. 373, 383; Sweetman 2005, pp. 357, 367). 226. Watts 1989; 1991, pp. 40–51. However, Pearce (2001, pp. 132–137) stresses that the proportion of infants buried in

many Roman cemeteries in Gaul and Britain, regardless of their religious identity, varies considerably due to analytical and taphonomic biases; see also Lucy 1994, pp. 26–27; Scott 1999, pp. 114, 123, 124. 227. For a full discussion, see pp. 125–126, 226–228. 228. E.g., Corinth XIII, pp. 70, 294, 297 (graves 503, 513, 514); n. 95, above (amphora burials). 229. Children were commonly interred among adults in Late Roman and Byzantine times: Corinth XIV, pp. 163, 164; Wiseman 1967a, pp. 34–35; 1967b, p. 419, pl. 88:d; 1969, p. 69; Wesolowsky 1973, p. 340 (Asklepieion, Lerna Hollow, and Gymnasium area, Corinth); Robinson 1962, pp. 116–117 (near Anaploga); Williams, MacIntosh, and Fisher 1974, p. 9 (west Forum, Corinth); “Halieis V,” p. 300; see also n. 221, above, on Demeter and Kore at Acrocorinth and on Argos. 230. Cf. Pader 1980, p. 156, on Anglo-Saxon cemeteries.

210

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

and regular military service,231 were alien to rural residents at the Isthmus. Consequently, the burials of neonates, children, and adolescents show no differences indicative of a graded progression toward adulthood. To be sure, individuals at varying stages of biological and psychological growth had different capacities for social and economic involvement. This would have translated into separate functional roles for small children and adolescents. But the basic relationship between adults and subadults as constructed in funerary ritual was unequivocal: every resident contributed to the survival and success of a family, and the life of the community was foremost a collaboration of families.

Sex and Gender Like age, gender is an essential determinant of funerary ritual in most cultures. Roman and Byzantine historical sources relate that Greek women often prepared the wake and conducted a formal lament, while men were responsible for transporting the corpse in the procession. The distinction of funerary activities by gender remains a feature of traditional communities in Modern Greece and Cyprus.232 Moreover, many ritual behaviors were dictated by whether the deceased was a man or a woman. All people have a biological sex, which can be recovered from human remains by osteological analysis. They also possess a gender, or an identity defined in terms of socially generated concepts of masculinity and femininity that are related, but not restricted, to sexual categories. Gender can be expressed through a variety of material modes in funerary activity, especially the physical appearance of the body but also grave design, funerary artifacts, and burial location. The funerary representation of gender can also be connected to other qualities, such as social status.233 The burials at the Isthmus display little differentiation according to skeletal sex. Adult men and women234 during all periods received interments of the same composition, design, and orientation; they were interred in the same positions, and they occupied the same burial areas. As has been noted, more women than men were buried in the area of the Northeast Gate, but it is unclear whether this is a significant pattern. The ratio of men to women in all graves at the Isthmus is nearly equal.235 This shows that certain conditions that can generate an imbalance in the number of male and female burials, such as restricted status, female infanticide, male emigration for warfare, or polygamy, were absent in the community. It is difficult to trace the distribution of funerary artifacts by sex, because many objects were discovered in multiple interments where they could not be associated with individuals. Six vessels (7–11, 13) were found inside or outside cists with both males and females. Each of these vessels originally accompanied a single body to burial, and in several cases it must have been the first or second body in the grave. But even then the mourners would have known that the grave was intended for continued use, and in one case (NEG 69-004) builders added a glass pitcher and a glass bottle (7, 13) on top of previously interred corpses. In

231. Pomeroy (1997, esp. pp. 67–99) discusses age-related identities in the Classical and Hellenistic Greek family. Dixon (1992, pp. 133–138) addresses life stages in the Roman family. Jacoby (1962, pp. 167–168), Antoniadis-Bibicou (1973, pp. 77–78), and Patlagean (1973; 1977, pp. 145–149; 1987, p. 605) survey the progression of children through social institutions such as education and marriage in the Late Antique and Byzantine legal, historical, and hagiographical sources. See also Abrahamse 1979; Moffatt 1986, pp. 705–713; Kalogeras 2001 on the roles, experiences, and representations of Late Antique and Byzantine children. 232. See, e.g., Papacharalampous 1965 (Cyprus); Kenna

1991 (Cyclades); Seremetakis 1991 (Inner Mani); Stewart 1991, p. 53 (Naxos); Tzortzopoulou-Gregory 2008, p. 332 (Corinthia and Kythera). 233. Arnold and Wicker (2001, pp. xi–xvii) and Cannon (2005) provide general discussions. Lesick (1997), Sørensen (2000, esp. pp. 41–59), and Díaz-Andreu (2005) address the complex relationship between sex and gender in the archaeological record. 234. The osteological analysis did not determine the sex of subadult remains; see p. 248. 235. On the paleodemography of the skeletal sample, see pp. 259–265.

SOCIAL STRUCTURE AND IDEOLOGY

211

this way, as has been noted, the vessels were not only ritual implements in the burial of one man, woman, or child, but also communal offerings for individuals of all ages and both sexes. Furthermore, the utilitarian, generic forms of the vessels bore no overtly gendered connotations. All except perhaps the unguentarium were brought from the household, where they would have been commonly used by all members of the family. The ritual employment of these vessels effectively universalized the status of the deceased, neutralized any personal distinctions by age and sex, and signified the common condition of mortality. The articles of clothing and personal adornment worn by the deceased varied by gender, but the scanty artifacts permit only a hypothetical reconstruction of this differentiation. In the few cases in which artifacts could be identified with single bodies, males were buried with a coin (4) and a finger ring (15), while females were buried with a pierced coin (14), earrings (19), and pins. Several graves containing both males and females produced other accessories, such as buckles and various rings, loops, disks, and pendants of indeterminate function. All these items are frequently found in contemporary graves across the northeastern Peloponnese, but the sex of associated skeletons has seldom been assessed. The testimony of Late Antique and Byzantine iconography and literature, however, shows that in many ways dress constructed images of masculinity and femininity.236 Earrings, necklaces, and pendants were typically worn by women, not men. The presence of both an earring and a beaded necklace in PAL 56-001 in all probability identified the deceased as a woman. While it is uncertain that any of the personal articles in the graves was exclusively worn by men, belt buckles and hobnailed shoes probably belonged to the masculine outfit. Other features of personal appearance that have not survived in the material record might well have communicated gender, especially clothing and hair style. Dress has been a dynamic medium for the expression of gender distinctions in most cultures throughout history. The concentration of jewelry around the neck, ears, shoulders, and chest, perhaps in combination with facial cosmetics, would have created a visual focus on the head of the woman during funerals at the Isthmus. Hair was probably also fashioned for ornate display. Signs of femininity on the dead body would have referred to other gender-linked qualities, such as specific conceptions of physical beauty, moral virtue, or marital status. Hanging adornments beautified dead women in a manner that would not have been practical in daily routines at the Isthmus. Jewelry, together with clothing, hair style, and corporeal position, gave the deceased a special appearance that was considered attractive and perhaps signaled other abstract traits, such as orderliness, composure, or constancy.237 In the unusual case of the young woman (69-001C) who wore the gold earrings (19), the conspicuous display of costly and intricate pieces of jewelry framing the corpse’s face might have communicated not just high social status but a special class of woman, such as betrothed or married.238 Women acquired a succession of identities as fiancées, brides, mothers, and widows, all of which would have been accompanied by a socially defined set of expectations, responsibilities, and behaviors.239 While many small objects in the graves at the Isthmus signified the gender of the deceased, others were ambiguous. One old man buried near Tower 14 (67-002A) wore only a bronze 236. See n. 142, above. 237. Walker (2003b) gives a useful survey of feminine adornment and its meaning in the Byzantine world. 238. According to historical sources, women in Late Antique and Byzantine society typically wedded in their teens to early twenties, the age of this individual (Antoniadis-Bibicou 1973, p. 78; Laiou-Thomadakis 1977, pp. 272–273; Patlagean 1977, pp. 145–148; 1987, pp. 597–600; Clark 1993, pp. 13–15; Laiou 2003, p. 28). This young woman might have died from

complications of pregnancy or at childbirth. 239. It is worth comparing the different behaviors and dress of women at separate life stages in traditional Greek society during the 20th century (e.g., du Boulay 1974, pp. 121–123). Papacharalampous (1965, p. 147) and Alexiou (2002, p. 39, n. 14) note that young or newlywed Cypriot and Greek women sometimes wear nuptial attire to the grave. Residents of Kyras Vrysi are typically buried at Ayios Ioannis with symbols of their marital status.

212

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

finger ring, but rings were worn by both men and women in Late Antique and Byzantine society. It is possible that his ring, which once had a design inscribed on the bezel, showed that he had been married, because finger rings were often exchanged during nuptial ceremonies.240 Like women, men would have acquired new identities during their lives, including that of husband. In addition, coins were found in both a multiple burial with men, women, and children (1–3) and a grave with one man (4), though a recycled coin was also worn by one woman (14). As instruments of monetary exchange, coins symbolized commerce and belonged to the public sphere; when reused as pendants or garment attachments, they were personalized for dress. As has been noted, too, they were a traditional token of passage to the afterlife. Coins were thus not strictly gendered objects but rather multivalent symbols in the context of funerary ritual. Apart from the expression of gender through the choice of objects, the similar form and composition of the graves of males and females, the common treatment of their bodies, and their spatial integration communicated the social status and group membership of men and women over time at the Isthmus. During all periods, their funerals required the same investment of energy and resources, their graves looked the same, and they often occupied the same burial features. This uniformity reflects the equivalent position of men and women in the social structure.241 Both were active members of the community with no essential distinction by socioeconomic contribution. To a certain extent, they would have operated in separate spheres within the small agrarian settlement, men tending the fields and women keeping the home. But their spheres also overlapped. Women, along with children, would have engaged in small-scale gardening, raising livestock, and perhaps herding and gardening.242 It is uncertain who milled the grain brought into the home, but the discovery of rotary querns in association with animal bones, hearths, and other domestic remains shows that processing grain was an activity of the house or its yard. It is also uncertain who communicated with travelers across the Isthmus and who journeyed to neighboring communities either in the countryside or at Kenchreai or Corinth. It cannot, however, be assumed that subsistence, social and commercial activities outside the home were conducted only by men and not women. The survival, let alone success, of the small household in such a subsistence economy required the efficient collaboration and mobility of all members of the family. Women wielded no small measure of power in their daily management of limited vital resources such as clothing, shelter, food, and tools. Imperial legislation from Justinian to the end of the Byzantine Empire largely excluded women from public life, and contemporary literature portrayed women as subordinate, weak, and retiring.243 But the image of women in writings molded by literary and rhetorical convention and produced by and for the elite propertied class of the city cannot be directly applied to the largely invisible female inhabitants of the

240. Koukoules 1951, p. 108. Walker (2001) evaluates the figurative and functional aspects of Early Byzantine marriage rings, known examples of which display elaborate engravings and prominent bezels. 241. However, the gold earrings (19) worn by NEG 69-001C and the jewelry (17, 33) worn by the individual in PAL 56-001, probably an adolescent or adult female, might have marked differential status (pp. 215–216). The gold earrings, as noted, might have also signified a separate trait, such as marital status. 242. See Laiou-Thomadakis 1981, pp. 248–249, on the agricultural activities of Greek women in Byzantine society, and Clark 1993, p. 101, on Late Antique women.

243. Beaucamp (1977; 1990–1992) and Clark (1993, pp. 6–62) compile the extensive legal and documentary testimony for the status of women in Late Roman, Early Byzantine, and Middle Byzantine society; Herrin (1983) provides a brief account of Byzantine women. Laiou-Thomadakis (1981; 1982), Laiou (1985; 2003), Garland (1988), and Connor (2004) offer a more positive view of women’s social and political freedom in the Byzantine centuries, concentrating on exempla from the urban aristocracy and the Imperial court. Walker (2003a) surveys the textual and visual record for the domestic lives of Byzantine women, though again the evidence prefers the upper end of the social spectrum.

SOCIAL STRUCTURE AND IDEOLOGY

213

countryside.244 Among nuclear families in rural communities without complex public arenas, the socioeconomic positions of men and women approached an equilibrium. As will be discussed, the fundamental relationships between husband and wife and between parent and child that formed the internal structure of the family were sometimes even commemorated through joint burial at the Isthmus.

Vertical Position The material remains of funerary ritual can reveal the vertical position of the deceased, that is, status as defined by wealth and access to resources, political power, or religious supremacy. Status can be measured by the choice of material and expenditure of energy in the form of interment, the treatment of the corpse, the quantity and quality of the funerary assemblage, the prominence and spatial associations of the burial location, and the extent of the ritual of interment.245 The funeral is a theater for the performance of relations and divisions within society during which status is asserted or even challenged. The material symbolism of rank, however, can be difficult to read. The ritual display of wealth or authority may diminish or cancel the need to communicate these attributes through the material components of burial. Ideological factors can reduce or mask the investment of resources and energy in mortuary forms. It is important not to confuse vertical differentiation with horizontal differentiation or diachronic variation.246 Given these complexities, a contextual approach is required to interpret the expression of vertical position in the burials at the Isthmus. An examination of both historical sources and regional archaeological evidence elucidates the mortuary channels for variable status in Corinthian society during the Roman and Byzantine eras. The burials at the Isthmus exhibit little vertical differentiation. Although the location of the Roman burials near the road into the Sanctuary was dictated by a concern for accessibility, no grave of any period, except PAL 56-001, was prominently situated. Mourners did not purchase the site of burial but simply claimed it. The ensuing funerals would not have been lavish affairs. Processions were seen by local residents, but they did not attract numerous spectators, and the number of participants would have diminished over time with the shrinking number of residents. The materials used in the design of most interments were found on-site and recycled. Their construction for the most part did not call for specialized skill or a great amount of time and effort. Funerary objects were typically simple and inexpensive. Lamps and terracotta and glass vessels for chrismations or libations were either acquired in local markets, manufactured in the area, or brought from the household. The money buried with the dead, even the pocket of 10 small-denomination bronze coins (1, 2, 3), did not comprise large sums. The personal articles worn by the deceased were neither numerous nor extravagant. Overall, the homogeneity in the material cost, invested energy, prominence, and complexity of funerary rituals and burials does not point to a social structure with an extended hierarchy based on wealth or other factors of status. There are three possible explanations for the simplicity and paucity of the interments at the Isthmus: a religious belief, philosophy, or custom that deplored mortuary splendor; an unwillingness to relinquish valuable objects and expend time and energy on mortuary 244. Laiou-Thomadakis 1981, pp. 241–242, 248; Herrin 1983, pp. 168–169; Bagnall 1995, pp. 81–82 (critique of Beaucamp 1990–1992); contrast Garland 1988, pp. 363 (“it is at least likely that the ideology expressed through [Imperial women] is representative of their society as a whole”). 245. These mortuary dimensions of social status were the focus of several important early publications: e.g., Binford 1971;

Tainter 1975; 1978; Peebles and Kus 1977; J. A. Brown 1981. 246. Many important contributions, building upon and revising earlier theory (n. 245, above), have explored the funerary representation of social position: e.g., Chapman and Randsborg 1981, pp. 8–10, 13–14, 23–24; Pader 1982, pp. 56–62, 192–197; O’Shea 1984, pp. 3–22; Parker Pearson 1999, pp. 72–94.

214

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

activities; or the general poverty of the community. The first explanation might seem logical in light of Christian eschatology. The belief in the salvation of a heavenly afterlife that transcends the material trappings of an earthly existence could be considered a leveling mechanism in the expression of social position. Thus, it would have deleted ritual manifestations of status.247 Believers might have even applied social pressure to shun ostentatious burial in favor of more symbolic materials, spaces, and actions. The Church Fathers in their homilies and exegeses condemned funerary luxury, particularly in the form of expensive clothing for the deceased.248 However, apart from fanatical adherents or paradigmatic figures, such as the Syrian St. Ephraem,249 this intellectual discourse appears to have had little impact on the customs of most people in the eastern Mediterranean.250 In this respect, it was not unlike the Cynic diatribe concerning the futility of mortuary commemoration, which was a dominant theme in Lucian’s varied corpus of the 2nd century.251 The existence of such teachings proves the continuing relevance of the behaviors they rebuked. Indeed, according to several Late Antique and Byzantine accounts, even bishops, saints, and monks wore fine clerical or monastic vestments to the grave.252 On the other end of the social spectrum, a majority of burials in the eastern provinces were simple and poor because bereaved families were unable to invest large amounts in them, not because they were Cynics or Christians. Likewise, the simplicity and poverty of the Late Roman graves at the Isthmus cannot be attributed to an ideologically motivated rejection of a more splendid manner of interment. The written sources and the archaeological evidence from the region show that individuals with greater wealth and power in Roman and Byzantine society were willing to invest in elaborate burials. The direct proportionality of social status to the cost, scale, and visibility of funerary ritual is a recurrent theme in the Greek literature of the Empire.253 A comparison of this testimony with the physical remains of contemporary tombs reveals that literary discourse corresponded to actual practice. The funerals of the urban aristocracy would have been spectacular, probably advancing through the intramural districts of Corinth and Argos and attracting large crowds. They preferred interment mostly in the urban or suburban zones of Corinth, its ports, and Argos, where they erected conspicuous monuments either alongside roads or coasts or around churches. Even the more elaborate Late Antique graves in the vast cemetery near the Asklepieion and in the Gymnasium area would have required more time and effort to build than contemporary graves at the Isthmus. Several contemporary graves in the region contained numerous personal articles or pots that marked the deceased as wealthy members of their communities.254 In contrast, the typical graves at the Isthmus most closely resemble either the anonymous graves among the more impressive 247. Volp (2002, pp. 198–202, 267–268) observes the low material value versus high symbolic value of early Christian funerary objects. 248. See n. 145, above. 249. Greg. Nyss. V. s. Ephr. Syr., PG 46.837, 46.845 (d. 373, Edessa). 250. Rebillard (2003b, pp. 143–197) likewise discusses how early Christian writings on funerals, commemorative feasts, and prayers to the dead did not result in effective sanctions. Rather, mortuary behavior was chiefly determined by local customs and familial concerns through at least the 5th century. 251. Caster [1937] 1987, pp. 275–280. 252. E.g., Geront. V. s. Melan. Iun. 68 (Jerusalem, late 4th century); V. s. Eliae Spel. 35 (AASS Sept. 3.862; Calabria, late 10th century); Sym. Thess. Ord. sepult., PG 155.676 (early 15th century). Gold threads, probably from episcopal vestments, were found in the grave of bishop Eustathios in the so-called Kodratos basilica at Corinth (τάφος Ε; Stikas 1964, p. 134);

gold threads have also been found in a Late Roman to Early Byzantine funerary monument at Patras (τάφος X; Dekoulakou 1979, p. 385). Engels (1998, pp. 204–209) discusses the relative leniency with regard to funerary luxury in Byzantine practice, including the clerical elite. 253. Rife 1999, pp. 63–125. 254. E.g., Miller 1981, pp. 48–49, pl. 12:d; Nemea Guide, pp. 73, 76, fig. 48; cf. Avramea 1997, p. 128, 2005, p. 217 (silver ring, various accessories, grave 7; near Temple, Late Roman); Papachristodoulou 1970, p. 104, pl. 80:β (barrel-vaulted chamber tomb with three individuals, 19 vessels, one ring, two earrings; Palaio Scholeio north of Klenia, Late Roman); Corinth XII, pp. 6, 289, nos. 1821, 1830, 1927, 2030, 2036, 2037, 2461, pl. 148:c (several finger rings, earrings, necklace with almost 300 beads from grave; Forum, Late Roman–Early Byzantine). It is unclear why G. R. Davidson (Corinth XII, pp. 6, 289) called the numerous objects in the grave at the Forum “cheap jewelry.”

SOCIAL STRUCTURE AND IDEOLOGY

215

interments in the Corinthian and Argive cemeteries or isolated or clustered graves either in the countryside or at small settlements like Zygouries and Halieis. The regional mortuary record therefore indicates that the social structure of the community at the Isthmus was not as complex as in the cities. Corinth possessed an affluent and powerful stratum of provincial administrators, ecclesiastical officials, landowners, and entrepeneurs that flourished during Roman times and late antiquity and survived, albeit on a reduced level, through the Early Byzantine period. The rural residents were much poorer and exerted no supralocal influence in church or government. Their advantageous geographic location meant that the inhabitants of the Isthmus could communicate with frequent traffic and access coastal markets. They must have exploited these channels to trade or obtain materials and perhaps to offer services to travelers.255 But the settlement was able to sustain itself without depending solely on Corinth or Kenchreai. The composition of the local funerary assemblage illustrates the limits of the residents’ external interaction. Certain items must have been imported from Corinth, Kenchreai, Schoinous, or Krommyon, namely, the moldmade lamps (5, 6) and the glass pitcher (7) and bottle (13). The pottery (8, 10, 12) could have been manufactured in the countryside. Small objects like the pierced coin (14), the hand-carved bone cross (20), and the drilled stone, glass, and shell beads (21) might have been made on-site. The dumpy unguentarium (11) was a peculiar local product in imitation of a funerary form long since obsolete in the cities. One omission from the interments at the Isthmus was the funerary lekythos so common in Corinthian graves of Late Roman and Early Byzantine date. The distribution of the published examples by size of settlement reveals a clear pattern in the northeastern Peloponnese: the form has never been reported in graves outside urban or semiurban contexts, namely, Corinth, Lechaion, Kenchreai, and Argos.256 If this distribution accurately represents a historical trend, then it shows that the market for these specialized commodities was circumscribed. Lekythoi existed on the Isthmus (a piriform example was found at Corinth’s eastern port257), but apparently they never reached the Isthmian fortifications. Residents there used cups or bowls from the country household instead of lekythoi for anointing the corpse or pouring libations. Rural communities necessarily communicated with larger settlements, but the funerary assemblage at the Isthmus speaks to the localization of daily life. Within this general frame, the burials reflect a historical development in the vertical complexity of the community from Roman to Byzantine times. Although much less stratified than contemporary urban society, the social hierarchy at the Isthmus displayed a small degree of internal variation during the Roman era. Among the uniformly unremarkable graves of the Late Roman settlement (phases II–III), three stand out. First, many unique features distinguish PAL 56-001 from the more mundane interments in the years before and shortly after the building of the fortifications (phases I–II). The cutting of the even chamber into the solid concrete podium of the Temple of Palaimon would have involved considerable time and energy, and after the funeral the cist was conscientiously sealed (Figs. 2.79, 2.81). The burial is the only one at the site that used wooden furniture, namely, a bier or coffin probably finished with plaster. The deceased individual, probably an adolescent or adult woman, was deposited in an orderly manner wearing an earring (17) and a beaded necklace (33). Most significantly, the site of burial was isolated from all other graves in a highly visible spot in the facade of the dilapidated hero shrine at the center of the ruined Sanctuary. Moreover, the association of the new grave with the ancient temple might have been intended to evoke the funerary character of the hero cult and the crypt. 255. For a fuller discussion, see p. 126, 141–142. 256. See nn. 160–161, above. 257. Kenchreai IV, pp. 135–136, no. RC 73, pl. 35.

216

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

PAL 56-001 is an anomaly that may be viewed as a rare event in the social and ideological development of the community. The conspicuous grave established a direct and indelible connection with the hero shrine through its architectural situation. The procession of the decorated corpse on a vehicle and its deposition atop the stepped podium showcased the status of the deceased before everyone in attendance. The location of the burial was an enduring statement of superiority, because its permanent alteration of the monumental landscape was visible long after the jewelry and the finished wood had disappeared from view beneath the slabs. While this individual was exceptional for wealth or influence, the concurrence of burial and temple may also point to an ideological distinction. It is possible, though far from certain, that this burial was an assertion of Christian confidence within a community under conversion.258 Two Late Roman graves containing 14 individuals at the Northeast Gate, 67-001 and 69-001, are slightly more elaborate than contemporary and earlier interments. The first (67-001) was comparatively deep, with walls that were solidly lined with bricks and the stones were set in courses, and the bodies, though arranged inconsistently, were interred with several diverse objects (Figs. 2.12–2.14). The second grave (69-001) shows even more effort: it was carefully sealed even though it was used for multiple interments, the walls were carefully lined with bricks in even courses, and the bodies were relatively regular in their arrangement (Figs. 2.18–2.21). Moreover, this grave contained the young woman (69-001C) who wore the only object in precious metal found in the graves at the site, gold earrings (19). In comparison, two adjacent graves with eight individuals (67-003, 69-005) were poorly lined and the bodies, which have various poses, were accompanied by only a few objects of little value. The other grave of phase III (RB 76-002) required little effort to make because it occupied a preexisting drain (Fig. 2.66). Moreover, the construction of graves around the time of the erection of the fortifications or shortly thereafter (phase II) required a much smaller expenditure of energy and resources than NEG 67-001 and 69-001. The families responsible for these two unusual graves invested more than their neighbors, presumably because they had more to invest and wanted to display their resources. The dimensions and the skillful brick masonry of NEG 69-001 recall the construction of numerous cists built in rows among the cemeteries of Corinth and Argos and annexed to the Corinthian basilicas.259 But these two graves, particularly the less orderly 67-001, lacked the prominence and monumental context of display that urban or churchyard burials often derive from their architectural surroundings. The intricate gold earrings (19) were valuable objects that must have set the young woman apart from her peers. They did not, however, signify exceptional wealth or highly privileged access to resources. Gold ornaments are occasionally found in Roman and Byzantine graves at Corinth, but they have also been found in peripheral settlements, such as Palaiokastro on the Methana peninsula.260 The Late Roman graves at the Isthmus attest to a limited degree of status differentiation by wealth in the community.261 The mortuary variability, however, does not show sharp or deep distinctions in vertical position; no widely separated elite and nonelite strata are evident, as may be expected in a larger rural community. Furthermore, besides the unique PAL 56-001 and perhaps the gold earrings, which might, too, have signaled a gendered identity, the interments have produced no overt signs of individual prestige, political authority, or religious superiority. This minor variation disappears in succeeding generations when the graves of Early Byzantine inhabitants are thoroughly homogeneous (phase IV). It will be seen that certain 258. For fuller discussion of the significance of this grave and the identity of the deceased, see pp. 129–134. 259. See nn. 71, 73, above. 260. Konsolaki 1987, p. 72 (gold earring, τάφος 2; beaded necklace on gold chain, τάφος 6).

261. Likewise, Sodini (2003, pp. 45–46) proposes that social differentiation within a rural middle class is visible among the mortuary remains of the sizable Late Antique community at Olympia.

SOCIAL STRUCTURE AND IDEOLOGY

217

qualitative changes in grave form, burial orientation, and placement point to a shifting conception of the purpose and meaning of funerary ritual in the Christian community. But the expenditure of energy and resources on the graves of this period remains at the same low level as among the earlier graves. By this point in time, the diminished population was operating more strictly within a subsistence economy. People continued to interact with neighboring communities and travelers along the ancient road, obtaining their groundstone and certain coarse wares and amphoras from the outside. But the residents relied foremost on their families and each other for survival, building their small houses from local materials, raising their own animals, and even producing their own cooking ware that echoes a foreign influence. More so than before, the social structure was egalitarian, and the level of production and exchange was low. No council or representative elite class seems to have existed at the Isthmus, as was the case in more substantial villages of the age.262 The material poverty and localization of activities evident in the mortuary remains should not be equated with a deplorable or pathetic emotional or existential state. The simplicity and paucity of the graves of the Late Roman and Early Byzantine residents do not mean that they were sad people, their routine monotonous, or their habitation squalid. Certainly by modern standards their possessions were meager and their work hazardous, but no more so than in many pre-modern rural settlements. There has been a tendency to interpret Late Antique or Byzantine graves found in isolation or in small groups as mirrors to human suffering263 and products of physical violence264 or catastrophic circumstances.265 This image seems to stem in part from a scholarly disinterest in social differentiation through burial on a regional scale, which is a natural result of the urbanocentric and site-specific scope of traditional classical archaeology, and in part from an old presumption that life in a declining world or during a “dark age” was savage and fraught with calamity. Such interpretations are founded not upon positive evidence or a full consideration of mortuary variability but upon three misconceptions concerning normative practice. The first is that normal graves occurred in funerary monuments, in cemeteries, and/or near churches.266 The second is 262. Lefort 2002, pp. 279–280, citing George heg. V. s. Theod. Syc. 114, 115 (in Galatia, mid-6th century; vita written late 6th century) and Nicet. Am. V. s. Phil. El. p. 137 (in Paphlagonia, mid-8th century; vita written 821/2). 263. E.g., “Halieis V,” p. 304 (“[The graves show that] the livelihood of the serfs living at Halieis seems to have been rather miserable”); van Andel and Runnels 1987, p. 115 (“heartrendingly poor and miserable”); see also Konti 1996–1997, p. 339. 264. E.g., Corinth III.3, p. 38; Laskaris 2000, p. 80 (“hands [of the skeleton] were at the throat, and the whole attitude, and the absence of any apparent grave, indicated that he had met his end by violence or perhaps had been buried alive”; Acrocorinth); Nemea II, p. 134 (disarticulated bones apparently in secondary burial “suggests that the corpse must have been deliberately and violently dismembered”; Stadium tunnel; see also n. 192, above); Nemea Guide, p. 73 (“The final end to the pathetically small community of survivors was violent,” referring to the same grave in the Stadium tunnel); Kosmetatou 2007, p. 191 (a young child at Midea, S3 of Roman or Late Roman[?] date, “may have met a violent death . . . a large, thin sharp stone that may have served as the murder weapon was discovered at the height of the throat . . . may have been hurriedly placed in a shallow grave”; see also Demakopoulou et al. 1997–1998, p. 80. This overly dramatizes the archaeological evidence for an orderly burial and the osteological evidence for nonspecific perimortem trauma in poorly preserved cranial remains [Ingvarsson-Sundström 2007, p. 472, figs. B-1b, B-2b]). 265. E.g., Vollgraff 1956, p. 98; Laskaris 2000, p. 56 (Late Roman grave in the Sanctuary of Pythian Apollo on the Aspis,

Argos may be “un enterrement fait précipitamment sous le coup d’un désastre public: le mort a été enseveli avec le vêtement et la bourse qu’il portait à l’heure de son décès”); Corinth XII, p. 6 (improbable scenario of female and infant in Early Byzantine grave at Corinth cut down as she fled the barbarian invaders with child in arms); Miller 1988, p. 3 (evidence of Late Byzantine or Ottoman graves around the Christian basilica “suggests that some sort of calamity overtook the small settlement”); see also Kosmetatou 2007, p. 192 (“the possible murder of the child . . . , as well as the death of the male and the female buried nearby [at Midea], may be connected with the Slavic invasion,” but “the theory is still largely based on speculation”); Wahlberg 2007, p. 199 (“These skeletons may be associated with various catastrophes that took place during the 4th through the 7th century a.d., among them natural disasters, epidemics, economic problems, violence related to the introduction of Christianity, and the Slavic invasion”). See p. 232 for a more plausible example of interment during upheaval from Early Byzantine Cyprus. 266. E.g., Ivison 1996, p. 102 (“By the 9th century burial occurred nowhere else but in and around churches . . . , whether within or without cities”); Avramea 1997, pp. 154–155, n. 75; Trombley 2001b, p. 219. In a recent survey of Late Antique architecture, the only tombs treated were aristocratic monuments and monastic burials (Mundell Mango 2000, pp. 945, 954–955). Sodini (1986; 1993, pp. 140–141) identifies burials inside churches in the eastern provinces as the “tombes privilégiées” of the clergy or lay dignitaries, a subclass of the entire populace; see also Laskaris 2000, pp. 74–75, 142, 266.

218

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

that normal graves were completely lined and covered. The third is that human remains were normally interred as skeletons in a perfectly supine position with the limbs extended straight or neatly crossed. As has been seen, numerous graves at the Isthmus and other rural sites were not located in cemeteries or near churches, they often were only partly lined, the bodies were deposited in diverse positions, and bones were sometimes rearranged by the continued use, redeposition, or disturbance of interments after their creation. Yet these burials show no signs that the dead had succumbed to violence or disaster, or that they were interred thoughtlessly or under duress. Instead they exhibit a concerted effort to enclose the deceased from above, and a few, particularly during the Early Byzantine period, display an elementary level of symmetry, decoration, and monumentality. Although the design of mundane graves at the Isthmus would have been less costly than that of more elaborate graves at Corinth, it was no less intentional. The mourners in the countryside had developed intimate bonds with their families, and these personal relationships dictated that they observe a complete ritual process of funeral, interment, mourning, and remembrance. The creative and efficient reuse of both the material from ruined structures and the space of those structures for burial points to the innovative and adaptive capacities of local residents.267

Horizontal Position Funerary ritual and its material components communicate not only the vertical position but also the horizontal position of the deceased in society. Horizontal position refers to membership in a social group or a collective identity that does not necessarily depend on the differential investment of resources and does not essentially imply a hierarchical relationship to other groupings or identities. Many societies are divided horizontally into units such as lineages, clans, tribes, interest groups, sodalities, guilds, ethnicities, or religious sects.268 These divisions can be communicated in funerary ritual through behaviors before and during the funeral, the location and orientation of burial, the treatment of the corpse, and symbolic representation in the form of graves and their contents. It is sometimes difficult to distinguish expressions of rank from expressions of horizontal affiliation. Moreover, competition between separate groups can lead to status differentiation, so that horizontal and vertical identities overlap. Unlike contemporary Late Roman and Early Byzantine burials at Corinth, the burials at the Isthmus do not display horizontal differentiation by occupation. A subtype of graves found on Acrocorinth and in the Forum marked the deceased as soldiers through the inclusion of iron weapons.269 Moreover, a few Roman epitaphs from Corinth and Kenchreai advertise the identity of the deceased as a veteran or officer.270 The residents of the Fortress did play a role in the defense of the Isthmus but were primarily involved in the rural economy. Unlike the burials at Corinth, their interments did not express the military persona of the veteran, militiaman, regular conscript, or professional soldier. This corresponds with the results of excavation and survey around the Hexamilion and the Fortress, which have produced abundant traces of a sedentary, agrarian settlement but no remains that are 267. Pace Avramea 1997, p. 155 (“Le viie siècle, le siècle des transformations, trouvera les gens du pays vivant sans nouvelles créations, repliés dans les ruines du passé”); 2000, p. 18. 268. Horizontal differentiation has not received as much attention in anthropological theory as vertical differentiation. The most studied form of group membership has been kinship. For general discussions, see Binford 1971, pp. 18–23; O’Shea 1981, pp. 49–52; 1984, pp. 46–49, 252–254; McHugh

1999, pp. 40–50. 269. See Chap. 3, n. 168. 270. On the Early to Middle Roman evidence, see Rife 1999, pp. 316–319. Three or four Late Roman epitaphs from Corinth identify retired excubitores: Corinth VIII.3, nos. 541, 558a, b, possibly 553; Monceaux 1885, pp. 409–410 (stone found west of Corinth); IGCB I 3, pp. 11–12 (wrongly locating the findspot on the Isthmus).

SOCIAL STRUCTURE AND IDEOLOGY

219

usually found at military camps.271 Another professional class that is not represented among the burials at the Isthmus is the clergy.272 Monumental and often prominent graves of presbyters and bishops have been found at Corinth and Lechaion,273 indicating that the deceased held an office that conferred a high status. If the settlement at the Isthmus had a church that has not yet been discovered, leaders in the religious community probably would have been buried close to or inside it. Since the social structure exhibits little vertical complexity, it is hard to predict what the grave of a local cleric might have looked like. The burials do not identify the deceased with any other occupational activities. This does not mean that individuals or families did not associate with one another or collaborate in a variety of subsistence or commercial activities, such as trading, fishing, hunting, gardening, gathering, and raising livestock. The sizable population that occupied the Fortress during the 5th to 6th centuries probably included craft specialists but did not achieve the occupational diversity of the urban population, as is richly attested in the epitaphs of Late Antique Corinth.274 Neighbors presumably assisted each other in certain communal activities. The cooperation of families on this level probably intensified during the Early Byzantine period, when the settlement was smaller, more contained, and more self-sufficient than in earlier centuries. The interments, however, do not show that this kind of interaction led to the formation of dominant corporate entities. A long tradition of communal burial for members of social, religious, and familial associations existed in the Greek world until late antiquity. The funerary activities of private associations are well documented in the epigraphy of the Roman East, particularly Asia Minor, but these seem to have been strictly urban phenomena.275 Burial clubs like those in Rome are not attested in Greece during the Empire.276 Ethnicity is a second possible mode of horizontal differentiation. However, the uniformity of the design, location, and contents of the graves at the Isthmus does not reflect ethnic distinctions within the community. The signification of ethnic identity employs many media that do not leave physical traces, such as language and hair styles, but it also follows many that do, such as personal ornaments, domestic furnishings and structures, and grave forms.277 Cremation, a frequent manner of Slavic burial, is not represented at the Isthmus or elsewhere in the region. Furthermore, none of the interments at the Isthmus resembles the Corinthian graves containing Early Byzantine weapons, handmade cooking pots, and buckles that were once identified as “barbarian.”278 The local production of the so-called Slavic ware at the Isthmus reveals that residents and foreigners interacted, possibly as rural neighbors, but there is no sign that Avars or Slavs occupied the site during their invasion and settlement in southern Greece during the late 6th to early 7th centuries. In every respect, the 271. While it is conceivable that Christian residents of the Isthmus buried soldiers without signs of their military service (see Haldon 2002, p. 65), comparison with the burials at Corinth and consideration of the broader archaeological record at the Isthmus render such a conclusion unlikely. 272. Kaplan (1992, pp. 202–203, 227–231) generally discusses the social roles of the clergy in Early Byzantine villages. 273. E.g., Stikas 1964, pp. 133–134, fig. 4, pls. 89:α–91:α (τάφος E; Kodratos basilica, Late Roman); Wiseman 1969, p. 85, n. 37, fig. 10 (grave 84; Gymnasium area, Late Roman– Early Byzantine); Pallas 1961, pp. 173–174, pls. 72:β–73:α (Lechaion basilica, Early Byzantine). Laskaris (2000, pp. 266–267) collects numerous examples of graves in churches and cemeteries belonging to clerics and dignitaries. 274. E.g., Corinth XIV, p. 167 (Lerna Hollow). Sironen (1997, pp. 401–408), Avramea (2000, pp. 17–18), and Sodini (2003, p. 43) discuss Late Antique craftsmen who record their occupations on epitaphs as members of the urban

working class. 275. Van Nijf (1997, pp. 31–69) and Samellas (2002, pp. 239–242, nn. 150–157) address this phenomenon in Greece and the Roman East; Rebillard (2003b, pp. 51–71) critically examines its relationship to Christianity. Fraser (1977, pp. 58–70) discusses the relevant epigraphic and architectural evidence from Rhodes during the Late Hellenistic to Early Roman periods in its broader context. 276. On the Roman institution in general, see the still useful Waltzing [1895] 1979, pp. 141–153, 256–300; Hopkins 1983, pp. 211–217. The suggestion that burial clubs or “guilds” might have existed at Roman and Late Antique Corinth (Wiseman 1969, p. 81; Walbank 2005, p. 268) has no supporting evidence. 277. See Daim 1998, pp. 84–89, and Pohl 1988 on modes of expressing early medieval ethnicity. 278. See Chap. 3, n. 168.

220

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

graves at the Isthmus belonged to a regional tradition that was largely consistent throughout the northeastern Peloponnese. The rural residents lived in their homes and buried their dead like other indigenous Corinthians and would have been aware of their long history of occupation. With the advent of foreigners, they could have identified themselves according to several categories—Corinthian, Roman, Christian—all of which separated them from pagan Slavs. Such sharp distinctions lost meaning with the establishment and integration of the foreigners through agricultural cooperation, regular commerce, religious conversion, and even marriage. The process of cultural exchange would have influenced the perception of ethnic identity within both groups. One major dimension of horizontal position in the community at the Isthmus was kinship. As has been discussed, the almost egalitarian treatment of men, women, and children in funerary ritual reflects the centrality of families in the social structure. Children were fully integrated into society as members of the household, and men and women shared the demands of subsistence in the field and at home. The scattered remains of Early Byzantine houses show that, at least during that period, the basic dwelling for the family was compact and meager; similar living conditions might have existed in the Late Roman settlement. Since local families lived and worked together closely, the death of one person would have greatly affected not only the atmosphere of the household but also its practical efficiency and productive capacity. The most vivid expression of the importance of kin groups as the fundamental social and economic unit in the community is the practice of multiple burial in the Late Roman fortifications. The distribution of ages and the frequency of certain congenital traits demonstrate that these interments contained the remains of families.279 Once a grave was created for the burial of the first body, other bodies could be added by simply finding the site, removing the slabs, and depositing the recently deceased over the previously interred, as many as 11 deep. In one large grave (HO 70-902) the cist was divided into two compartments that separated either phases of burial or segments of a lineage. The multiple burials at the Isthmus served the same purpose as the chamber tombs and funerary monuments that lined the roads into Roman Corinth and later surrounded its churches. However, since the rural community was less prosperous and complex than the more affluent and diverse urban community, its graves were much less impressive. The form and contents of most multiple interments at the Isthmus reveal that these families occupied roughly the same social rank, except for NEG 69-001 and 67-001. The higher level of expenditure on those two burials demonstrates that vertical (social status) and horizontal position (kinship) could intersect in funerary ritual to identify the deceased not only as the member of a particular family but as the member of a leading family. Furthermore, formal variation among the collective burials implies a degree of competition between separate families.280 The composition of families can be traced in the sex and age of individuals interred together (Tables 4.1, 5.3). Several double and multiple burials containing men and women of roughly the same age presumably represent married couples,281 though it is conceivable that in some cases the relationship was one of siblings.282 Other graves contained pairs or small 279. See pp. 263–265, 291–292, Tables 5.3, 5.18. 280. Köpstein (1978, pp. 50–53) and Haldon (1997, p. 387) discuss “potential internal structural oppositions” in the Early Byzantine village, as attested particularly in the “Farmer’s Law” of the 7th or 8th century. 281. NEG 69-103A, and B; 69-004C and B, D, or E; 67-003A and B; 69-005B and C; T2 68-003A and B; HO 70-902,D or I, and J, E, or F and G. DEC 69-901 contained two adults of unknown age and sex.

282. The joint burial of siblings has been argued for two skeletons (a male, 25–30 years, and a female, 25–30 years) found in a Late Roman to Early Byzantine grave in the West Shops of the Corinthian Forum. The osteological study observed that both skeletons exhibited the same congenital deformity in the cervical vertebrae and concluded that they were most likely brother and sister (Gejvall and Henschen 1968; see also Robinson 1962, p. 110, n. 45).

SOCIAL STRUCTURE AND IDEOLOGY

221

groups of individuals who were otherwise related. One grave contained an old woman and a male adolescent (NEG 69-007A, B), perhaps a mother and son or a grandmother and grandson. Another containing old adults and infants (HO 70-902) might have also held grandparents and grandchildren. Two others, one containing a possible male and two women (RB 76-002)283 and another containing three women of different ages and a young child (NEG 69-001), likely included some combination of siblings. The remaining multiple burials contained a combination of adult men and women, some of whom were probably married, and subadults of different ages, presumably their children. The separation of one large cist into two compartments (HO 70-902) may indicate two generations of a family, but that is uncertain. The interments thus recreated in the space of death the intimate relationships that had existed in life at home between husband and wife, parent and child, and perhaps brother and sister. The possible inclusion of relatives outside the nuclear family, namely, brothers/uncles, sisters/aunts, and grandparents, shows that the household could include extended relatives but presumbly not whole second families. The organization of families and their importance at the Isthmus reflect broad developments in Late Roman and Early Byzantine society.284 Families with children were at the core of rural communities in Greece and Asia Minor. The survival of the nuclear family as the basic economic unit depended on internal cohesion, reciprocity in conjugal and parentalfilial relations, and equality between members. The traditional hierarchy of the Roman family founded upon paternal authority, the domination of husbands over wives, and agnatic lineage285 did not necessarily pertain to the Late Antique and Early Byzantine countryside. Furthermore, the elementary family, sometimes including close relatives,286 had gradually become an autonomous and indivisible social unit, while its network of connections with extended relatives and other lineages diminished. Rural settlements like the one at the Isthmus were aggregations of families who managed their own households, guarded their own possessions, and kept their own animals and crops independent of their neighbors. The solidarity of communities came from interpersonal links between friends, the sharing of pasturage, the management of water resources, cooperation during harvesting or haying, and mutual participation in group rituals, such as festivals, marriages, and funerals.287 It is noteworthy that multiple burials only occurred at the Isthmus during the Late Roman period. Graves and habitational remains predating the erection of the Fortress are too sparse to provide useful information on the composition of the local population and its settlement. But the physical remains of sedentary occupation, subsistence activities, and the domestic economy during the Early Byzantine period all point to the centrality of the family. The similar funerary treatment of men, women, and children during both the Roman era and the Early Byzantine period fits the model of the family as the basic social unit. The practice of multiple interment continued into the Early Byzantine period at Corinth, where large chambers for collective burial and ossuaries have been found in the ruins of the Forum. There are several possible reasons why families at the Isthmus apparently abandoned multiple burial after the late 5th or 6th century. One is that the composition of the community had changed in such a way that identification with a particular family was no longer 283. The secondary burial HO 70-901 also contained a possible male and two females, but it is uncertain whether these three individuals came from primary burials in one grave (p. 199). 284. Kaplan 1992, pp. 483–488; Haldon 1997, pp. 376–387; Lefort 2002, pp. 243–245; Laiou 2005, pp. 49–51. 285. Rawson 1986, pp. 15–31. 286. The husband-wife-child(ren) grouping was the essential socioeconomic cell, but the cohabitation of nuclear fami-

lies with adult siblings or grandparents, particularly when they were unmarried or widowed, is a pervasive pattern in Mediterranean societies past and present. See, e.g., Laiou-Thomadakis (1977, pp. 79–85) on extended households in the Macedonian countryside during the 14th century. 287. Kazhdan 1982, pp. 3–5; Kaplan 1992, pp. 195–198, 203–205; Haldon 1997, pp. 386–387; Lefort 2002, pp. 279– 281.

222

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

necessary. The number of residents in the Late Roman Fortress was considerably greater than the number in the Early Byzantine agrarian settlement. Members of such a large community would need to identify themselves with a particular family or else their neighbors may not recognize them immediately. In the considerably smaller community, neighbors would know to which family the deceased belonged without the placement of that individual in a familial interment. Another possibility is that the Early Byzantine residents did in fact bury their dead in familial groupings, but these consisted of a collection of graves in unmarked areas, not multiple interments. An individual’s kinship might thus have been expressed through interment near other family members even if those burial groups were not distinguished at the surface by a precinct or an inscription. The coincidence of skeletal markers that are genetically linked in certain burial areas offers some support for this hypothesis.288

Christian Ideology Apart from kinship, the dominant category of group identification in the community at the Isthmus was ideological, namely, the beliefs and practices of Christianity.289 As has been discussed, the Christianization of the community at the Isthmus is difficult to chart with precision. Christianity flourished in the urban setting of Corinth from the time of Paul’s missions, and by late antiquity Christian communities existed on the nearby coast at Kenchreai, Krommyon, and possibly Schoinous. The earliest definite traces of Christianity at the Isthmus appear in the construction of the Theodosian fortifications. Graffiti, lamps, and architecture bearing Christian signs occurred in the settlement during the 5th to 6th centuries.290 The unusual Late Roman grave cut into the Temple of Palaimon (56-001) might have belonged to a Christian because it overtly violated the sanctity of a venerable temple, an act that would not have been attempted by rural residents who were traditional polytheists. The first grave to display Christian symbolism is RB 76-002, near which was left the lamp bearing a cross (6). Over the following century or two, more graves displayed crosses, namely, the shaft burial outside the Northeast Gate with the inscribed marker (69-010) and the cist with the infant wearing the beaded necklace (20, 21) near Tower 14 (69-991). Thus, secure evidence for Christianity exists at the Isthmus from the 5th to 7th centuries. Specific stages in the spread of the new religion cannot be identified in the archaeological record, but in all likelihood the process was complex and gradual. The impact of Christianity on the funerary rituals of the local residents is likewise difficult to trace, apart from the use of explicit symbolism. Christians chose the same materials, the same forms and objects, and many of the same behaviors as their predecessors when they buried their dead. Several scholars have observed that Christians and pagans used the same burial grounds in Late Antique Greece, and their mortuary practices were so similar that little or no distinction is apparent at many sites.291 It can be reasonably assumed that the same homogeneity prevailed at the Isthmus. Christianity cannot therefore be identified sensu stricto as a dimension of horizontal differentiation. Exclusive indices of Christian burial are 288. See pp. 141, 407–408. 289. One important cross-cultural survey (Carr 1995) shows that philosophical and religious beliefs about the afterlife can be as important in determining mortuary practices as social organization. 290. See Chap. 3, n. 80. 291. E.g., Déroche 1989, p. 2722 (Delphi); Gounaris 1989, p. 2707 (Philippi); Pallas 1977, pp. 68–69; Kourkoutidou-Nicolaidou 1997, p. 129 (Thessalonica); Buchet and Sodini 1984,

p. 229 (Thasos); Oikonomou-Laniado 1998, pp. 407, 409; 2003, p. 30 (Argos); Samellas 2002, pp. 254–255, n. 197. Johnson (1997, pp. 50–56) discusses the pagan origin of Christian cemeteries and notes the sharing of cemeteries by pagans and Christians, particularly in the Roman West. MacMullen (1997, pp. 111–112, nn. 24–28) discusses the blurred distinction between Christian and non-Christian mortuary practices across the Roman world beginning in the 3rd century.

SOCIAL STRUCTURE AND IDEOLOGY

223

hard to find.292 Certain features of the graves, however, can be attributed to novel funerary rituals that expressed a new eschatology and theology in the rural community. These are the display of Christian symbolism, the particular treatment of the burial site and the corpse, the careful placement of the body, the regular eastward orientation, and the conscientious handling and removal of disturbed bones. It is also possible that Christianity motivated the ostentatious violation of the pagan precincts as it did at other Late Antique sites in Greece and the Near East. The growing frequency and prominence of the primary features of Christianization between the 5th and the 7th or 8th centuries (phases II–IV) parallels the slowly changing religious identity of the local residents. It should not, however, be concluded that Christians did not occupy the site before such changes were overtly expressed in the Early Byzantine burials, or that earlier burials lacking such features did not belong to Christians. During the long period of transition, many residents who had accepted the teachings of Christ and perhaps even participated in Christian worship still lived very much as their nonbelieving neighbors. The first burial that may be Christian, PAL 56-001, illustrates this ambiguity. Beyond its overt transgression of the sacred space of the hero shrine there is nothing about it that distinguishes it as Christian. The form, contents, and prominence of the interment marked the deceased as an individual of status in the community, not as a Christian. The mourners placed the head in the west and the feet in the east, a solar alignment that was consistently adopted in Christian funerals, but this orientation was dictated foremost by the plan of the Palaimonion. The grave was situated in the facade alongside the east–west tunnel through the foundations in order to juxtapose the late burial with the crypt. The unusual inclusion of wooden furniture in the cist probably also did not communicate the Christianity of the deceased. It has been proposed that Christians in Late Roman Britain buried the dead in coffins to preserve the corpse.293 However, coffins or biers were used for pagan burials during the Early Roman period in the North Cemetery at Corinth and the Koutsongila cemetery at Kenchreai. The containment of the corpse at the Temple of Palaimon inside the concrete chamber and on top of or within the wooden structure might have accommodated a belief that the body should be preserved, but it was not an exclusive sign of Christian identity. The first burial that can be identified as Christian is RB 76-002. The form of the grave as a multiple burial in a reused drain, the orientation and dress of the bodies, and the placement of the lamp (6) and unguentarium (11) outside are all typical of Late Roman and Byzantine graves in the region. The one feature that identifies the burial as Christian is the cross on the disk of the lamp. It might be argued that non-Christians sometimes used lamps with crosses simply because they were available on the market, or because they did not particularly care how the lamp was decorated so long as it worked. But accounts of death and burial by Roman and Byzantine writers plainly show that the selection of funerary objects was a deliberate and meaningful experience for the bereaved. Neither pagan nor Christian mourners would have disregarded the appearance and symbolic content of an object that was to be a visual and mental focus during the funeral procession or graveside visitation. 292. On the basis of his survey of Late Antique and Byzantine burial sites in Greece, Laskaris (2000, pp. 290–291) compiled a list of the distinct signs of a Christian grave: orientation, position of the hands, funerary stelai with Christian symbols, painted decoration with Christian motifs, datable funerary artifacts, location near Christian habitation or over a ruined building of Late Roman or “Early Christian” date, construction using spolia, certain forms of interment, especially the absence of cremation, and construction using rubble and brick. A few of these mortuary dimensions (orientation, position of

hands, presence of Christian symbolism, association with a church) are valid criteria because they can be directly linked to Christian belief and practice in most cases, if not exclusively. Others (construction techniques) cannot be positively linked to Christian belief and practice and were undoubtedly also applied by non-Christians during the Late Roman and Byzantine eras. Still others (various chronological markers) are independent of Christian belief and practice. 293. Philpott 1991, p. 238.

224

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

These events stimulated participants to contemplate the value of life and the meaning of death. The image of the cross was both a potent reminder of earthly sacrifice for heavenly salvation and an effective apotropaic device against demonic harassment.294 In many ways, this burial signals a transitional stage in the religious life of the local community during the late 5th to early 6th centuries. It belonged to the old Corinthian funerary tradition, but the inclusion of Christian symbolism points to the presence of new beliefs. The adoption of a Christian ideology of death can be further traced in several basic transformations in mortuary form and funerary ritual from the Late Roman to Early Byzantine periods. The later cists, which contained only one or two corpses, are on average more spacious than the earlier graves. The arrangement of covering members is slightly more even and, in some cases, more symmetrical than the typical Late Roman graves. It was during the Early Byzantine period that grave markers appeared, including permanent installations with a monumental quality. The arrangement of the bodies follows a much more regular pattern than before: the legs are more evenly extended, the arms are placed over the chest or abdomen, often crossed, and the head is often supported from below or the side. Burials are aligned more consistently toward the east than in previously centuries. The spacing between graves on the surface is greater and more regular than before. These developments are not attributable to changes in social status because they do not constitute a significant increase in the investment of time, effort, or resources by the bereaved. The energy expended to excavate a spacious cist for a single burial was no greater than the effort required to dig a chamber for multiple burial. The constructed markers over the two graves outside the Northeast Gate (69-009, 69-010) and the erection of amorphous stones over the two graves near Tower 2 (68-002, 68-006) would not have been significantly more time-consuming and labor-intensive than the lining or closure of many earlier interments.295 While several of the Early Byzantine graves were marked at the surface and none of the Roman and Late Roman graves seems to have been, the enhanced visibility was intended to improve access for mourners, not to display the wealth or prestige of the deceased. Furthermore, there is no shift in the inherent value and quantity of the funerary assemblage. Diachronic changes in burial do not therefore represent a dramatic growth in vertical complexity. Besides the three exceptional burials of Late Roman date (NEG 67-001, 69-001, PAL 56-001), mortuary variability by social status was basically the same during both periods. Instead of a divergent investment of energy and resources, the qualitative discrepancies in the grave designs of both periods reflect different degrees of attention to detail and preparation. The construction of the later graves exhibits a new intentionality and awareness of form that corresponds with innovations in funerary ritual. These changes reflect the adoption of a new eschatology in which the dead body is a sacred entity that would be resurrected in the afterlife.296 They also revolved around the growing importance of the graveside ritual and visitation after burial. The sign of the cross in burials asserted the religious affiliation of the deceased. It also recalled the foundationact of Christ’s end, evoked the message of victory through death, and served as a talisman of protection. The crossing of the arms might have been intended to turn the corpse into a perpetual cruciform figure. Local residents also handled dead bodies with care to create an attitude of peaceful or comfortable slumber. They placed stones or tiles beneath the head like a pillow and beside the head to prop it up, and they extended and straightened 294. On the use of crosses to consecrate and to protect Christian tombs, see also Chap. 3, n. 118 and n. 88, above. 295. NEG 69-004, 67-001, 69-001, T2 68-003, DEC 69-901, HO 70-902. 296. Bynum (1995, pp. 1–114) discusses various interpreta-

tions of the doctrine of the bodily resurrection in the Early Church. Outside of theological disputation, Greek and Latin epitaphs show that the simple idea of the corpse arisen and redeemed resonated with Christians across the Empire (Rush 1941, pp. 249–250).

SOCIAL STRUCTURE AND IDEOLOGY

225

the legs and torso to appear natural and restful. The two markers outside the Northeast Gate could have been used during activities after burial, but their form recalled either the deathbed or bier with a pillow or the figurative empty bed of the grave from which the deceased has arisen. These acts and forms gave the corpse a particular appearance of sleep that fit a Christian notion of death.297 While the association between sleep and death was an ancient concept, the metaphor of death as “sleep” (ὕπνος) and “dormition” (κοίμησις) is recurrent in New Testament scripture298 and appears more frequently than before in early Christian writings.299 In its new usage, death was not an existential end but a period of rest following life from which the pious would awake in resurrection. The metaphor was not restricted to the literary, liturgical, and theological discourses. Its popular currency is reflected in the common use of the term κοιμητήριον in Christian epitaphs, including many at Corinth, to identify the site of burial literally as a “dormitory” or “resting place.”300 The special treatment of corpses by the Early Byzantine residents of the Isthmus reflects the influence of this notion on local funerary ritual. The concept of resurrection following death, which was the key to Christian eschatology, dictated the arrangement of the body in the grave. The eastward orientation of burial was an ancient custom that was widely observed by early Christians, according to the literary sources, and continues to this day in Greece.301 The alignment of the corpse with the head in the west so that the body faced to the east was variously interpreted in antiquity. The Church Fathers wrote that, like the orant, the corpse was turned eastward to face divinity and the coming resurrection.302 In addition, the orderly arrangement of the limbs of the deceased not only gave the appearance of slumber during the funeral; it also prepared the deceased for salvation by facilitating his or her bodily departure from the grave as though arising from a bed. The growing importance of graveside rituals during the Early Byzantine period is suggested by the erection of grave markers over at least four graves and the widening space around each interment. These alterations to the earlier design allowed mourners to gather around graves at the time of the funeral and return to the grave with ease after the deposition of the body. The substantial markers built outside the Northeast Gate, apart from their visual reference to a kline, could have functioned as a table or pedestal for the placement of offerings after the enclosure of the cist. All these features facilitated the rituals of remembering the deceased. The cult of the dead was certainly not exclusive to Christianity, but it did become an important part of Christian funerary practice. Finally, the special treatment of human remains when graves were disturbed at the Northeast Gate and west of the Fortress points to a new concept of the significance of the body and the relationship between the living and the dead that emerged in late antiquity and 297. See Kyriakakis 1974, pp. 41–42. 298. E.g., John 11:11–13, I Cor. 15:18–20, Eph. 5:14, I Thess. 4:14–15. 299. E.g., Cyr. Jer. Cat. myst. 5.9 (Jerusalem, mid-4th century); Const. apost. 6.30.2 (Syria, late 4th century); Greg. Nyss. V. s. Macr., PG 46.984 (d. 379, west of Neocaesarea, Pontus; vita written late 4th century); John Chrys. Coem. 1 (PG 49.393– 394), Hom. XXXI in Mt. 2, 3 (PG 57.373, 57.374), Hom. LXVII in Gen. 4.13 (PG 54.577), Paralyt. 8 (PG 51.62; late 4th century); Epiphan. Expos. fid. 21.11 (III, 522; Cyprus, late 4th century). Ogle (1933), Rush (1941, pp. 13–22), and Lattimore ([1942] 1962, pp. 164–165, 306–307) compile numerous examples from Classical, ancient Hebrew, and early Christian literature and epigraphy, concluding that the metaphor’s frequency increases greatly in Christian contexts. 300. IGCB I, pp. 38–39, 68–72 (N. A. Bees); Feissel 1983,

p. 202; Emmanouelides 1989, pp. 255–256, 275; Sironen 1997, p. 120. Rebillard (1993) surveys the epigraphic usage of κοιμητήριον and coemeterium to designate a Christian tomb or burial area; see also Volp 2002, pp. 151–156, stressing its symbolic importance for group identification. 301. Loukatos 1940, p. 53; Spyridakis 1950, pp. 116–118; Kyriakakis 1974, pp. 55–56; Abrahamse 1984, p. 129. Recent burials at Ayios Ioannis in Kyras Vrysi adopt a similar orientation. 302. E.g., Greg. Nyss. V. s. Macr., PG 46.984 (d. 379, west of Neocaesarea, Pontus; vita written late 4th century); John Chrys. Pat., PG 60.725 (late 4th century; PG 46.984). Bas. Caes. Spir. S., PG 32.189–192, and Greg. Nyss. Orat. Dom., PG 44.1184 say that those praying face the biblical paradise in the east; see also Cyr. Jer. Myst. 1.9, on the eastward alignment of baptismal rites.

226

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

developed in subsequent centuries. The orderly rearrangement of the bones after handling shows a basic acknowledgment of the sanctity of the body in its postmortem state and a prevalent concern to maintain its integrity. Even when the corpse had dissolved into its constituent parts, those parts retained meaning as fragments of an anatomical whole. The bones could therefore be reassembled, however schematically, to approximate the corporeal form. Furthermore, the preferential removal from graves of bones, not utilitarian or decorative objects such as vessels or jewelry, shows that human remains as exiguous as a metatarsal and useless as an innominate were considered inherently valuable and even beneficial to their possessor. The closest analogy for these behaviors was the collection and conception of Christian relics.303 Since no standard liturgical process existed for the identification of relics from Late Roman to Middle Byzantine times, the recognition of sanctity in artifactual or biological matter was achieved through the composition of hagiography and celebration by monastic communities, or even just the force of popular interest.304 The material presence of saints in the form of their belongings or pieces of their bodies was a source of power for the faithful. Touching, viewing, or simply being near the remnants of the holy dead established direct contact with them. Several hagiographies recount how followers of saints and clergy stole cuttings of hair, shreds of clothing, and even body parts during funerary rites.305 The act of reclaiming bones from the mortuary setting was a means of overcoming death. Since relics harbored the innate and undying essence of the deceased, their acquisition and exchange as sacred commodities reintegrated the dead into living social and economic systems. The importance of bones as a material channel for communication across existential planes in Late Antique and Byzantine society is evident in the invention, veneration, and translation of relics.306 Of course, the men, women, and children buried at the Isthmus were not recognized as holy persons, and those who disturbed their graves years later probably did not think that they had unearthed clerics or saints. But the fascination with the physical remains of the deceased, the impulse to possess bones, the notion of the integrity of the fragmentary corpse, and the concept of its piecemeal value echoed a broader mentalité regarding a dynamic relationship between the living and the dead that had emerged in the early development of Christianity. The adoption of the new conception of death in Christian ideology thus led to fundamental changes in the funerary rituals of the Early Byzantine community. The question remains, however, as to why the first unequivocal evidence for the presence of Christianity at the Isthmian fortifications dates to the 5th century, while its influence on mortuary practices is not consistently evident for at least another century and a half. It seems highly doubtful that the numerous Late Roman graves excavated so far in separate areas at the Isthmus all belonged to non-Christians, or that all Christian graves at the site remain undiscovered.307 Throughout the Roman world, conversion to a particular religious identity did not involve burial in 303. P. Brown (1981, esp. pp. 78–79, 88–92) presents a classic treatment of Christian relics; see also Geary (1986) and Bynum (1995, pp. 104–108) on the cult of relics in the Late Antique and Medieval West, and various papers in Bozóky and Helvétius 1999. 304. Kaplan 1999. 305. E.g., Anton. V. s. Sym. styl. 29 (d. 459, near Beroea [Aleppo]; vita written late 5th century); Callin. V. s. Hypat. 51.10 (d. 466, near Chalcedon; vita written mid-5th or 6th century); Theod. Petr. V. s. Theod. p. 97.10–17 (d. 529, near Jerusalem; vita written mid-6th century); Eustrat. pres. V. s. Eutych. 10.97–98 (PG 86.2384; d. 582, Constantinople; vita written late 6th century); V. s. Eustr. heg. Abg. 39 (IV,

p. 393.23–394.6; d. 886, Constantinople; vita written end of the 9th–10th centuries); Nicet. Dav. Paph. V. s. Ign., PG 10.560 (d. 877, PG 105.560 Constantinople; vita written end of the 9th to mid10th centuries ); Greg. disc. V. s. Basil. jun. 55 (AASS Mart. 3.*38B; d. 944 or 952, Constantinople; vita written mid- to late 10th century); V. s. Nicon. 47.15–25 (John Malakenos d. early 11th century, Lacedaemon; vita written mid-11th century). Abrahamse (1984, p. 130) discusses this testimony. 306. Kötting 1965, pp. 15–24; Hunt [1981] 2001; Maraval 1985, pp. 222–224, 236–237; Emmanouelides 1989, pp. 417– 420. 307. Cf. p. 126.

SOCIAL STRUCTURE AND IDEOLOGY

227

a specific, exclusive area; adherents of separate groups, including pagans and Christians, were typically buried together.308 The religious history of the Corinthia and the varied evidence for Christianity dispersed across the site suggest that many of the known graves at the Isthmus did in fact belong to Christians. The delay in the funerary expression of Christianity is particularly curious because it contrasts with the cemeteries at Corinth, where graves comparable to those at the Isthmus in design and use contained Christian artifacts309 or displayed Christian epitaphs310 as early as the 5th century. Although the specific arrangement of bodies and burial orientations has not been consistently recorded in reports on graves in the northeastern Peloponnese, the mortuary innovations attested at the Isthmus in the Early Byzantine period had apparently developed earlier at Corinth. The funerals of the Late Roman residents of the Isthmian fortifications were not necessarily devoid of reference to their Christianity. Several behaviors or materials that have left no physical traces could have been used to show piety. The ancient sources relate that early Christians employed reading, singing, invocation, dress, and gesture to communicate their religious identity at funerals.311 These forms of self-identification during the transitional period in the religious development of the community did not require the adornment of graves and funerary objects with Christian symbols. In contrast, many Christian burials at Late Antique Corinth expressed the faith of the deceased through epitaphs and paintings. These decorative media required not only the services of craftsmen for their production but also the appreciation and literacy of viewers for their consumption. It is not, therefore, surprising that epitaphs and funerary paintings are absent from the settlement at the Isthmus. The rural residents were less affluent and educated than urban Corinthians and did not possess the means or the impulse to employ the regional epigraphical and painting industries. Christians at the Isthmus were apparently slower than contemporaries in the city to express their new conception of death and to accommodate their deeper concern for commemoration in corporeal treatment and burial design. This delay might well have arisen from the gradual nature of change in the local community. The manifestation of eschatological beliefs in the material components of funerary ritual by the 7th century was the final stage in a long transformation. Obviously the Corinthian Church existed and, indeed, thrived and expanded for centuries before the appearance of Christian iconography and ecclesiastical architecture in the material record. This chronological gap reveals the late development of such public media for the display of Christian identity rather than an absence of believers and worshippers in the city. A similar transition might have occurred, only at a somewhat slower pace, in the Corinthian countryside.312 Moreover, the shift in practice reflected in the Early Byzantine graves at the Isthmus might have been stimulated by external factors. The community in the Late Roman Fortress 308. Johnson 1997; Rebillard 2003a; 2003b, pp. 25–49. 309. On the Christian burials at the Asklepieion, Lerna Hollow, and the Gymnasium area, see, e.g., Corinth XIV, pp. 161–167; Wiseman 1967b, pp. 418–419; 1969, pp. 84– 85, n. 36; 1972, p. 8; on downdating this cemetery, see n. 11, above. Sanders (2005b) concludes that “[h]ard archaeological evidence for Christian worship . . . all appear [sic] no earlier and plausibly much later than ca. 475” (pp. 441–442). This survey does not give a detailed discussion of all evidence. The author admits that some material, such as lamps with Christian symbols associated with graves (pp. 436–437), belongs to the late 5th century. On the evidence for Argive Christians during the 5th century and later, see Oikonomou-Laniado 2003, esp. pp. 4, 23–25, 51–57, 77–80. 310. For a survey of the major Christian epitaphs from Corinth, see Feissel and Philippidis-Braat 1985, pp. 358–369.

Scholars have argued (N. A. Bees in IGCB I, pp. 30–37, 42, 43–48, 49–50, 54, 60–64, 77–79, 84–86, 90–94, 102–110, 114–119, 124–128; Ferrua 1967, pp. 369–370; D. Feissel in Feissel and Philippidis-Braat 1985, pp. 358–359) that several Christian epitaphs at Corinth belong to the 5th century. The earliest precisely datable text is the epitaph of Polychronios, erected in 431 (Pallas and Dantis 1979, pp. 63–64, no. 2; Feissel and Philippidis-Braat 1985, p. 368, no. 96). The epitaph of Selene contains a consular reference that dates it to 446? (Corinth VIII.1, no. 145; Feissel and Philippidis-Braat 1985, p. 368, no. 13). On Christianity in Attic epigraphy of the 4th to 6th centuries, see Sironen 1997, p. 376; Trombley 2001a, vol. 1, pp. 284–292. 311. Volp 2002, pp. 101–156. 312. Cf. Sanders 2004, p. 187.

228

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

showed only limited social stratification and no significant professional or ethnic differentiation. In such a coherent and homogeneous social context, neighbors would have known each other’s beliefs and practices without their overt expression. Instead, the primary distinction between residents was by familial grouping. The situation at Corinth, however, was different. The impulse to differentiate oneself horizontally as a Christian was greater in a complex society of many religious groups, including, among others, a prominent Jewish community and a staunchly pagan aristocratic element. The residents of the Isthmus did not face such religious diversity until the incursions of Avars and Slavs in the late 6th to early 7th centuries. The settlement of foreigners in southern Greece, possibly followed by their interaction with the farmers at the Isthmus, might have encouraged the natives to identify themselves not only as indigenous Corinthians but also as Christian Greeks. One salient feature of the slow and gradual tempo of conversion was the persistence of pre-Christian practices, sometimes referred to as “pagan survivals.”313 By the 7th century, theological orthodoxy was widely accepted in the Greek world. Although the Church considered these beliefs universal, the observation of traditional practices from before the advent of Christianity continued. Such practices should be distinguished from the less widespread phenomenon of independently operating pagan cults, which had been banned since the 4th-century edicts, and aristocratic or intellectual antiquarianism, which was fashionable only in certain circles, such as the Constantinopolitan elite. Fragmentary vestiges of polytheistic custom were ubiquitous in places characterized by an indirect or infrequent contact with central administration, by popular autonomy in belief and practice at the level of the village and the household, or by conservatism in social and religious life. One such region was the Mani, where “Hellenes” (Ἕλληνες) were Christianized only under Basil I in the late 9th century (Const. Porph. Adm. Imp. 50.71–76). But “Hellenic” religion was not restricted to the countryside: it could be found in many large communities in Asia Minor, Syria, Palestine, and Egypt with a long history as polytheistic cult centers,314 and it must have operated on some level in all urban centers. The 102 canons of the so-called Quinisext Council, convened by Justinian II in 691–692, dealt with the discipline of church leaders and the proper conduct of their subjects. Several address the concern that forms of “Hellenism,” such as pantomime and popular theater (51 [Mansi 11.968B]), prognostication (61 [Mansi 11.969E–972A]), and seasonal festivals (62 [Mansi 11.972A–C]), were too prevalent in Christian society and threatened the moral code of the Church.315 Such activities, however, were neither essentially unchristian nor deliberately antichristian. They preserved deeply rooted behaviors that countered the unpredictability and hazardousness of the human condition by defending individuals and their homes against harm and by defining their place in the natural order. Although originating in distant antiquity, they were closely relevant to the thoughts and actions of Christian Greeks, who adapted them to their needs. The religious life of local communities in the Late Roman and Early Byzantine Greek world thus accommodated the old and the new. It integrated a unitary system of orthodox teachings that authorized certain images and rituals among the devoted with a plurality of older attitudes and practices that served the daily needs of the people.316 313. See in general, de Pina-Cabral 1992; Haldon 1997, pp. 327–337; Trombley 2001a, vol. 1, pp. 147–151; on pagan and Christian burial customs, see Sodini 1977; MacMullen 1997, pp. 109–118, specifically pagan and Christian burial customs. For a discussion of parallel developments in the Medieval Slavic world, see Barford (2001, pp. 223–226) on the Church’s absorption of “pagan beliefs” and “rites.” 314. Jones 1964, pp. 941–943; Haldon 1997, pp. 329–331.

315. Trombley 1978, pp. 3–9; Dennis 1993, pp. 279–281; Angold 1995, pp. 457–460. 316. Stewart (1991) has demonstrated a similar coexistence of “great and little traditions,” or “doctrinal religion” and “local or practical religion,” in Modern Greek culture by exploring the continuum between supernatural forces in Orthodox cosmology and traditional demonic beings ([ε]ξωτικά) in the beliefs and actions of Naxiote villagers.

SOCIAL STRUCTURE AND IDEOLOGY

229

In this light, the terminology and underlying concept of “pagan survivals” are misleading. They are laden with the negative connotations of organized resistance or backwards peasantry and suggest a static continuity through time.317 An amalgam of traditions likewise characterized Corinthian mortuary practices. Iconography, language, the design of the grave, and the arrangement of the body communicated in different ways the fundamental eschatological doctrine of the Church.318 At the same time, both urban and rural Corinthians continued to observe ancient funerary customs and employ pre-Christian sepulchral forms. The fusion is illustrated by the funerary artifacts of RB 76-002: pieces of dress typically worn for burial (29), a late version of a bulbous unguentarium (11), and a Christian lamp (6). The same mourners who deliberately chose a lamp with a cross anointed the dead using an antiquated vessel shape. The absence of lamps from the Early Byzantine graves at the Isthmus is unusual because Christians normally carried lights during funerals and memorials. The residents probably used candles or torches that have not survived in the archaeological record. The only vessel that appears in the Early Byzantine graves is the cooking pot (12) that was probably smashed alongside NEG 69-010. Presumably it was used for libation, chrismation, or purification, all ancient practices, and its smashing might have been an apotropaic gesture. The placement of the coin (4) over the chest of T14 69-003A echoed the old practice of “Charon’s obol.” These traditions have all been recorded in Christian burials at Corinth and elsewhere. For instance, the interment of eggs with children, which probably symbolized growth, fertility, or even the resurrection in Christian contexts, was a durable practice attested not only in graves of the 5th to 4th centuries b.c. in the North Cemetery and near the amphitheater at Corinth, but also in a Hellenistic grave at Argos, a Roman tomb north of Corinth, a Middle Byzantine grave in the church in the Roman Stoa at Sparta, and Frankish graves near the church southeast of Temple E at Corinth.319 Such practices did not continue because they flourished in rustic isolation or escaped reform by zealots.320 They continued, often in an altered form, because they were still appropriate to the daily needs and routines of urban and rural society. The concepts of nourishment for the dead and fare for the Stygian ferryman originated in a pre-Christian view of the afterlife but echo in Greece today. Foods, such as fruit or κόλλυβα, are left beside graves during visitations in modern cemeteries like Ayios Ioannis in Kyras Vrysi. The terrible figure of Charon still inhabits the world of the folk songs, ritual laments, and verse-epitaphs alongside the Panayia, the saints, and the angels as “Charos” (Χάρος), alternately a sinister hunter and an angelic escort.321 These notions compensate the mourners’ sense of loss by imagining a soul with earthly appetites that will encounter dark personifications of death 317. Horden and Purcell (2000, pp. 408–411) warn against overestimating the continuity between ancient and modern religious practices, and also against underestimating the changing significance of traditional customs and capacity for reviving them. Saïd (2005) surveys “the mirage of continuity” and Romantic survivalism that color frequent analogies between ancient and Early Modern Greeks in the travelers’ accounts. 318. Beck (1979) argues that the teachings of the Middle to Late Byzantine Church concerning the afterlife had little or no impact on the popular imagination. Angold (1995, pp. 442–453) offers an important corrective discussion, stressing that the Eastern Church did not adopt a unified or fully articulated doctrine. That the essential message of life through death was widespread is proven by the ubiquity of the cross in funerary contexts. 319. Shear 1929, p. 537 (grave west of amphitheater,

Corinth); Corinth XIII, pp. 70, 84, no. 12 (North Cemetery, Corinth); Bruneau 1970, pp. 459, 525, 529 (tombe 58, Quartier Sud, Argos); Walbank 2005, p. 277 (grave in the Tomb of the Sarcophagi, north of Corinth); Waywell and Wilkes 1994, p. 388; 1997, pp. 423, 428 (Sparta); Williams et al. 1998, p. 241, n. 32 (southeast of Temple E, Corinth). Egg shells have been found in unpublished burial deposits of Early Roman and Byzantine date north and south of the harbor at Kenchreai. Nilsson (1908) collects early Greek evidence for this practice and discusses its symbolism; see also Olynthus XI, pp. 192–194. 320. See Avramea 1997, p. 154. 321. The evolution of this figure has been widely discussed: Lawson [1910] 1964, pp. 98–108; Moravcsik 1931; Papacharalampous 1965, pp. 142–144, 149; 1971, pp. 220–221; Loukatos 1978, pp. 221–222; Alexiou 1978; 2002, pp. 49–50; Stewart 1991, pp. 7, 15–16.

230

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

on an otherworldly journey. Moreover, the act of selecting possessions like kitchen wares or money to deposit at the grave draws a personal connection with the dead in a more immediate, concrete manner than the mere acceptance of an abstract notion of future salvation. Traditional beliefs and behaviors maintained an important role in early Christian funerary ritual because they were meaningful for mourners. The development of the new religion in the Corinthian countryside during late antiquity was not an abrupt caesura but a continuous process of integration and adaptation.

Social and Ideological Deviancy The graves discussed so far in this chapter belonged to individuals who were positive actors in the social structure of the Roman and Byzantine community at the Isthmus. The community would also have had its deviants or marginal members. Many cultures identify heterogeneous social figures through atypical funerary rituals or deny standard burial to those who perish under extraordinary circumstances.322 The denial of a proper funeral to enemies, outsiders, or the marginalized dead, sometimes accompanied by physical violation, can serve to punish and even to demolish an honorable or victorious image.323 The mortuary treatment of such individuals can delete them from the collective memory by subverting normal practices.324 Since these individuals are excluded from society, their burials often signify no vertical or horizontal position. For the residents at the Isthmus, an intentional act of social exclusion would entail interment in a noncanonical type of grave, the desultory or abusive handling of the corpse, and the absence of any familial or religious demarcation. One such burial is LOU 69-801, unique for its anomalous corporeal position, grave design, and location. The unnatural arrangement of this man’s legs around the time of death or shortly thereafter contradict the supine and extended position that prevailed among Christians and its preferred evocation of peaceful, beautiful repose. The sight of the decaying body and the folding back of the legs over the torso must have struck viewers as grotesque and offensive. The body might have been left unburied for some time before transport to the final site of interment, which is also without parallel at the Isthmus. Even the Byzantine builders who disturbed the earlier graves west of the Fortress showed greater reverence toward the human remains they found than those who created 69-801. The form of the cist was also abnormal. It was an unmarked, elliptical, shallow hole without covering or lining that was too small to accommodate a body in any usual manner. The treatment of the corpse, the form of the cist, and the absence of funerary artifacts all prove that this individual was not granted a typical funeral according to Christian belief and practice. Moreover, the interment was not placed in a formal burial area. It was located in an open area well outside the dense settlement in the south part of the enceinte. At the time, the vicinity was used for depositing refuse, and roughly contemporary bothroi of similar size to the pit-like cist have been found there, many containing cut animal bones and even the severed limbs of cattle.325 Such a stark juxtaposition must have led local residents to draw a qualitative comparison between the mutilation and interment of the corpse and the butchery of ani322. On the differentiated treatment of deviancy in death, see in general Shay (1985), revising Saxe (1970, pp. 11, 69, 71–74, 118–119) and Binford (1971, p. 17). Hubert (2000) explores many methods of social exclusion across human cultures, including burial. 323. Bloch 1982, pp. 228–229; Duncan 2005, pp. 211–213. 324. Williams (2003) explores the use of funerary rituals, materials, and spaces to remember and to forget.

325. The dumping of waste outside the limits of habitation was a common feature of Late Roman and Byzantine cities, such as 6th-century Emesa (Leont. V. s. Sym. Sal. p. 79.21–23) and 14th-century Chandax (Karpozelos 1989, pp. 345–346). Liebeschuetz (2000) explores how ancient communities handled rubbish disposal, stressing that the removal of refuse was largely a private obligation in Rome and the eastern provinces.

SOCIAL STRUCTURE AND IDEOLOGY

231

mals and dumping of garbage. This grave is the only one of Middle or Late Byzantine date yet found at the Isthmus, and the social structure of the local community during that period can only be inferred. Based on comparison to funerary rituals of Early Byzantine date at the site, these mortuary remains signaled that the deceased, at least during burial, had no social status, familial identity, or religious affiliation. In particular, the disposal of the corpse disregarded the Christian precept of salvation through bodily resurrection and the prevalent concern to keep the sanctity and incorruptibility of the physical self. The blatant deviation of this burial from the usual practices of interment in the Byzantine Corinthia marks the dead man, and perhaps the situation of his death, as aberrant. The skeletal remains bear no evidence of fatal trauma, revealing that he was not a casualty of war or invasion; rather, he was buried purposefully in an area peripheral to established settlement. The Church Fathers and Canon Law excluded certain persons from the celebration of funeral rites: non-Christians, unbaptized children, the excommunicated, heretics and schismatics, suicides, traitors, desecrators, and those condemned to death.326 The extraordinary nature of the burial of 69-801A, which was in many respects antithetical to the standard manner of Christian interment, could point to one or more of these marginalized personae, particularly religious outcast and criminal. The author knows of no historical or archaeological example of folding back the legs or amputation in this manner. It is basically analogous to the excision of facial, genital, or limb appendages, a common punishment for a gamut of crimes in Late Roman and Byzantine law.327 Capital punishment by public execution for murder, rape, heresy, and other crimes took many forms, from impaling to hanging to burning, all of which mutilated the body.328 Whatever place in society this man occupied, his burial could have signified many negative attributes. The exposure of the corpse before interment, if it occurred, would have left the flesh for scavengers, an expulsion from humanity to the realm of beasts.329 The decision to place the mutilated corpse in a cramped pit when it would have required little effort to cut a larger, rectangular cist expressed a deeper level of rejection from the social order. Throwing the bodies of criminals or lowlifes over cliffs and into chasms was synonymous with ignominious death in the imaginary world of Greek fiction during the Roman era.330 Apart from its manifest similarity with the adjacent bothroi, the basic form of 69-801 was a distant echo of the holes or caverns into which ancient Greeks had cast condemned persons for incarceration or execution. Athens had the notorious Βάραθρον and Sparta the Καιάδας or Κεάδας,331 while Corinth had the Κῶς, a gaping pit or cave where thieves and fugitives were imprisoned.332 One famous Corinthian who lived many centuries before 69-801A, Diogenes the Cynic, committed an ultimate antisocial act when he instructed his disciples not to inter his body properly but “to stuff [it] into a pit” (εἴς γε βόθρον συνῶσαι, Diog. L. 76.79). Late Roman and Byzantine legal texts discussed the proper use of toilets, cesspits, and

326. Petrakakos [1905] 1971, pp. 41–48, 112–115; Loukatos 1940, pp. 113–115; Koukoules 1951, p. 197; Emmanouelides 1989, pp. 90–105. Parker (1983, pp. 43–47) and Lindenlauf (2001) discuss the criteria for exclusion from regular burial in early Greece; Finucane (1981, pp. 54–58) addresses similar exclusions in western and northern Europe during the Middle Ages. 327. Singowitz 1956, pp. 18–22; Patlagean 1984. 328. Dennis 2001, pp. 6–7. 329. The author knows of no definite example of exposure from Greek archaeological contexts. G. D. R. Sanders has reported (Blackman 2002, p. 19) the discovery of an unusual burial of Ottoman date in the Panayia area at Corinth. The

grave contained the body of a young man who had been suspended at the left shoulder from an iron hook and then interred in a prone position. 330. Rife 1999, p. 104, n. 159; Lindenlauf (2001, pp. 88–89) gives earlier testimony. 331. Petrakakos [1905] 1971, p. 65, n. 5; Cantarella 1991, pp. 96–105; Lindenlauf 2001, pp. 88–89. 332. Ael. Herod. Pros. cath. 14 s.v. τὰ εἰς-ῶς (I, pp. 403.29– 404.3), cf. Steph. Byz. Eth. s.v. Κῶς (p. 402); for the name, see Strabo 8.5.7 (367) and Hesych. Alex. s.v. Κῶς (II, p. 560). Rife (1999, pp. 219–220) discusses the testimony and Corinthian topography. Contrary to Wiseman (1978, p. 14) the ancient sources do not record that it was the site of an “execution pit.”

232

FUNERAR Y RITUAL, MORTUAR Y VARIABILITY, AND SOCIETY

drainage systems, aiming to promote orderly waste management and to minimize stench in dense settlements. Structures that contained excrement were no place for an honorable death.333 One infamous story related how the arch-heretic Arius disgorged his organs during a fateful trip to the public latrine at the Forum of Constantine in 335. Likewise, the corpse of Thaleleius, the ungodly archbishop of Thessalonica, was found headfirst in the drainpipe of a toilet.334 Meeting one’s end in a cesspit, full of darkness and uncleanliness, was a stroke of divine punishment for blasphemy. Besides these enemies of the Church, the remains of persons who had suffered violence were sometimes thrown unceremoniously into wells or other dark, watery places. The bones of a man displaying cranial trauma have been found in an Early Iron Age well near the City Eleusinion at Athens. The remains of at least 11 individuals, several showing cut and slash marks indicative of a massacre, have been recovered from Frankish fill inside a manhole southeast of Temple E at Corinth.335 Corpses have been recovered from bogs in Thrace and on Crete, though the date and circumstances of their deposition are uncertain.336 A mass burial in a cistern has also been found near Kopetra in south central Cyprus. Excavation of the cistern in the court of the monastery at Sirmata revealed the remains of nine adults and children who had been buried under dense layers of rubble and tile debris and animal bones around the mid-7th century. While the human remains do not display traumatic injury, this grave was probably created during a time of social upheaval or natural disaster.337 In these cases, the unconventional placement of burial in facilities for water and the informal arrangement of deposition in unmarked space designated the deceased as outside the prevailing social order, whether because of their own status within the community or because of the community’s dissolution from without. A better understanding of the reason behind the strange treatment of the body in 69-801 would clarify the significance of this interment at the Isthmus. The only sure conclusion is that, at the moment of burial, this man was marked socially and ideologically as deviant within the Christian community. The strange case of 69-801 illustrates the depth of the social structure at the site during this period. It also shows how Byzantine Greeks could manipulate the material components of funerary ritual and mortuary space not just to preserve but also to destroy a person’s memory. An extreme form of burial reminded viewers to forget an individual whose place in society had been effectively obliterated. 333. Karpozelos 1989. 334. The tradition concerning Arius originated in the epistles of Athanasius during the late 350s (Ep. ad epis. Aeg. et Lib. 18–19 [PG 25.581A, B]; Ep. ad Serap. 3.3), where the heretic’s explosive end was compared to that of Judas (Acts 1:18). Epiphanus iterated the account in the mid-370s (Adv. haeres. III, pp. 146.52–147.2).The story grew more detailed and scatological as it flourished during the 5th century in both West and East: Faustinus and Marcellinus, Lib. prec. 7; Rufin. HE 1.13 (PL 21.485C–486B); Socrates 1.38.8–9; Sozom. 2.29.4–5; Theodoret, Haer. fab. com. 4.1 (PG 83.415B); HE 1.14.8; Hist. rel. 1.10 (probably an interpolation). Leroy-Molinghen (1967,1968) traces this legendary development. John Mosch. Prat. spir. 43 (PG 87.2897B, C) explictly compares the grotesque death of Thaleleius with that of Arius; see Leroy-Molinghen 1968, p. 109, n. 4. 335. Little and Papadopoulos 1998 (Athens); Williams, Barnes, and Snyder 1997, pp. 25–31; Barnes 2003, p. 436

(Corinth). Papadopoulos (2000) discusses the phenomenon of burial in wells in ancient Greece as a means of social exclusion, particularly during the Bronze Age. 336. Dieck 1965, pp. 17, 50–51, 82–83, nos. 15 (Nea Vyssa, Thrace [formerly Ahorköj]; burial dated 12th century), 309 (Iraklio, Crete; burial date unknown). The social identity of these individuals cannot be recovered from the scanty available information about their discovery. The better known bog burials in northern Europe, most of which date from the Iron Age to the Early Roman period, have been variously interpreted as criminals, social outcasts, and sacrificial victims. Perhaps the Greek bog burials recalled the better-known ancient practice of throwing the bodies of transgressors against state and deity into the sea (Cantarella 1991, pp. 91–105; Lindenlauf 2001, p. 89; 2003, pp. 421, 427). 337. Kalavasos-Kopetra, pp. 71, 90, fig. 3.18; Fox 2003, pp. 275, 277.

PART II THE OSTEOLOGICAL AND BIOARCHAEOLOGICAL CONTEXT OF THE HUMAN REMAINS

5

 THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

T

he human remains found at the Isthmus provide important evidence for the biological history of the Corinthian countryside from Roman to Byzantine times. In my discussion of the examination of the bones and teeth (Chaps. 5–7), I will consider both their osteological character as the hard tissues of a living skeletal structure and their bioarchaeological significance as dynamic factors in a particular natural and social environment. In the ensuing chapters I will describe the procedures and standards used, presenting the material through descriptions, tables, and figures, and I will use the data to reconstruct the biological identity of local residents, their diet, health and lifestyle, and their relationship to other groups. I will also consider, to the extent possible, how the local population changed through an era of historical transition, especially during the Late Roman and Early Byzantine periods. The validity of the reconstruction and interpretation of skeletal biology at the Isthmus will depend on the representativeness of the available data. The total assemblage of bones and teeth in variable states of preservation comes from 65 men, women, and children interred in graves,1 and at least four more individuals found outside graves. While this body of evidence is of considerable size and diversity, it must be stressed at the outset that the number of individuals under study comprises a relatively small sample for analysis, and therefore it is not known how accurately it represents the ancient living population. As a result, any conclusions drawn from the human remains about the constitution, life experience, and external relations of local residents must be considered strictly provisional until the excavation of more graves creates a larger skeletal sample. The approach adopted here is therefore one of cautious speculation. However, since so few skeletal remains at Greek sites from graves of Roman or Byzantine date, let alone those of other periods, have ever been analyzed and published, the full treatment of the material from the Isthmus makes a unique scientific contribution. Indeed, this study presents one of the largest Late Roman or Byzantine skeletal samples ever published (see Tables 6.24, 7.13).2 Both the historical importance of the Isthmian Sanctuary and Fortress and the richness of the associated funerary contexts justify a detailed study of the human remains. The archaeological, textual, and mortuary evidence covered in Part I furnishes the essential framework for examining skeletal biology at the Isthmus, especially during the Late Roman and Early Byzantine periods. The same population of native Corinthians occupied the site for several centuries. They lived in small, simple dwellings concentrated on the fortifications and dispersed among the ruins of the Sanctuary. They organized themselves as families 1. This total does not include the subadult buried in NEG 69-008, from whom only one bone survives. No human remains have survived for study from T13 54-001, DEC 69-902, or PAL 56-001, though photographs and drawings of fragmentary

bones in situ exist for the last one (Figs. 2.80, 2.81). 2. Laskaris (2000, pp. 284–288) briefly surveys skeletal studies of human remains recovered from Late Roman and Byzantine graves in Greece.

236

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

in which all members participated in the activities of a mixed agricultural subsistence. Planting, cultivating, and harvesting crops, tending animals, preparing foods, and constructing and maintaining buildings involved heavy physical labor and repetitive tasks that could be both stressful and hazardous.3 The local diet probably included cereal grains, fruits and vegetables, meat, fish, and dairy products.4 The composition of the skeletal sample, along with its oral conditions, and its health can be predicted from the influential factors of social organization, living environment, occupational activity, and diet. One would expect a skeletal sample with a homogeneous morphology and recurrent congenital traits. Infection and malnutrition, particularly among children, should have generated defects in developing dental enamel and anemia from iron deficiency. Teeth should show low to moderate levels of attrition and moderate levels of caries and caries-induced alveolar defects. Adults should exhibit infected leg bones from occupational trauma, joint disease in the back, shoulders, and chest from repetitive stress in the upper body, and injuries particularly in the forearms and shoulders from accidental injury during heavy labor. The distribution of activity-related conditions should be even or close to even between the sexes. It will be seen that the skeletal sample basically meets these expectations. Fine variation in certain conditions will clarify this general picture of skeletal biology and reveal both individual experiences and intrapopulational diversity. While the human remains from the Isthmus attest to the character, habitational setting, and life experience of local residents, it is hard to ascertain on osteological grounds their relationship with groups at other sites. Greek skeletal biology during the Roman and Byzantine eras has received little scholarly attention, and few publications exist.5 Only sparse evidence from the Corinthia and Argolid is available for direct comparison with the remains from the Isthmus. Therefore, it will not be possible to compile a regional survey of biological variability on the same scale as the survey of mortuary variability in Chapter 4. Similarities and differences between the remains at the Isthmus and those from other sites in the northeastern Peloponnese will be noted, but any apparent patterns, particularly the differentiation between rural and urban residents, must be considered hypothetical until more data are published. The most reliable results will emerge from the comparison of dental conditions at the Isthmus and at Corinth, as both sites have produced sizable dental samples. It will also be useful to compare the human remains from the Isthmus with published skeletons from Roman and Byzantine graves elsewhere in Greece, Asia Minor, and Cyprus. Comparative analysis on this level can reveal not only the degree of diversity across the Greek world but also the prevalence of conditions among separate groups living under similar circumstances. It will be seen that, compared to other pre-modern groups and to contemporary Greeks, including urban Corinthians, the rural inhabitants of the Isthmus suffered common afflictions but were relatively healthy. One of the greatest challenges in any osteological analysis is determining how to organize the skeletal data for meaningful interpretation. The most coherent and substantial subset of the Isthmian sample belongs to the Late Roman and Early Byzantine periods. The periodization of all 65 skeletons from the site is summarized in Table 5.1. This investigation will focus on the 58 individuals interred during the end of the 4th to the 7th or 8th centuries, phases II–IV in the burial chronology. The bones and teeth of these 17 men, 19 women, three adults of unknown sex, and 19 infants, children, and adolescents will be treated as a single, coherent data set for studying physical traits, illnesses, and injuries. Such a grouping 3. Bryer (2002) vividly evokes the strenuous, muscular activities of Byzantine farmers. 4. For a full discussion, see Chap. 6.

5. On the history and current state of physical anthropology and the study of human remains from archaeological contexts in Greece, see Roberts et al. 2005.

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

237

TABLE 5.1. PERIODIZATION OF 65 SKELETONS Total Individuals

Males or Probable Males

Females or Probable Females

Adults of Unknown Sex

5

T2 68-003A

T2 68-003B WF 62-001A

DEC 69-901A DEC 69-901B

46

NEG 69-103B NEG 69-004C HO 70-902D/I HO 70-902E/F HO 70-902H HO 70-901C NEG 67-001A NEG 67–001C NEG 67-003B NEG 69-005C RB 76-002B

NEG 69-103A NEG 69-004B NEG 69-004D NEG 69-004E HO 70-902G HO 70-902J HO 70-901A HO 70-901B NEG 67-001B NEG 67-001D NEG 67-003A NEG 69-005B NEG 69-001A NEG 69-001C NEG 69-001D RB 76-002A RB 76-002C

HO 70-902D/I HO 70-902E/F

NEG 69-004A HO 70-902A HO 70-902B HO 70-902C NEG 67-001E NEG 67-001F NEG 67-001G NEG 67-001H NEG 67-001I NEG 67-001J NEG 67-001K NEG 67-003C NEG 67-003D NEG 67-003E NEG 69-005A NEG 69-001B

12

NEG 69-007B T2 68-002A T2 68-006A T14 67-002A T14 67-004A T14 69-003A

NEG 69-007A T14 69-002A

T14 69-701A

NEG 69-009A NEG 69-010A T14 69-991A

Middle to Late Byzantine (12th–14th centuries)

1

LOU 69-801A

Late Byzantine to Early Modern (15th–19th centuries)

1

Period Early to Middle Roman (mid-1st to late 4th centuries)

Late Roman (end of the 4th to mid-6th centuries)

Early Byzantine (late 6th to 7th or 8th centuries)

Subadults

TC 60-001A

is valid because local residents over a span of three (or more) centuries experienced the same social, economic, environmental, and material conditions.6 While the Late Roman to Early Byzantine settlement did develop over time, particularly with the dispersal of habitation and the localization of production, whether these shifts had any perceptible impact on the skeletal structures of local residents cannot be determined with certainty due to the small skeletal sample of Early Byzantine date. In contrast, the 46 skeletons of Late Roman date represent a substantial number of individuals whose settlement at the Fortress can be more precisely characterized. Therefore, in order to aid both the interpretation of local patterns and intersite comparison, this study will arrange the data into two sets, one containing all Late Roman to Early Byzantine skeletons and one containing only Late Roman skeletons. Discrepancies or similarities between the Late Roman and the Early Byzantine skeletons will be noted, but the small data set for the later period cannot support robust conclusions about biological change in the Corinthian countryside. Evidence of the seven adult skeletons from graves predating and postdating the Late Roman to Early Byzantine periods will be 6. Note that other published skeletal samples from sites in Greece and Cyprus cover approximately the same time span (see Tables 6.24, 7.13). The grouping of the osteological data

from the Isthmus into a Late Roman to Early Byzantine data set therefore facilitates comparative study.

238

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

summarized but not included in discussions of the two primary data sets. These few chronological outliers come from disparate phases of settlement with distinct living conditions. The investigation of the osteological and bioarchaeological context of the human remains from the Isthmus is divided into three chapters. These address the demographic and morphological nature of the skeletal sample (Chap. 5), teeth and oral health (Chap. 6), and paleopathology and paleoepidemiology (Chap. 7). This chapter discusses the condition of the skeletal assemblage and its composition in terms of sex, age at death, and structural variation. The first section will describe and quantify the preservation and survival of bones and teeth as recovered, stored, and examined. The second section will outline the analytical techniques. The third section will use the osteological data to interpret the paleodemography, physical appearance and diversity, growth and stature, and nonmetric variability of local residents.

THE CONDITION OF THE SKELETAL ASSEMBLAGE Before addressing the human remains from the Isthmus as evidence for skeletal biology during Roman and Byzantine times, it is important to assess their condition. As discussed in Chapter 1, formation processes altered the graves and their contents between the time of deposition and the time of discovery and examination (Table 1.3). This study must evaluate the nature and distribution of postdepositional alterations in the skeletal assemblage so that postmortem effects are not mistaken for ante- or perimortem conditions. Moreover, calculating the preservation and survival of bones and teeth establishes a quantitative basis for determining the prevalence and significance of any condition observed in them. The hard tissues of the human body undergo chemical and structural changes after death and burial.7 The collagenous protein phase of bone slowly degenerates into peptides and amino acids by hydrolysis, and the crystalline matrix of the mineral phase (calcium hydroxyapatite) reconfigures. These internal processes are enhanced by water, foreign ions, and microbes in the burial environment, which further degrade the protein-mineral bond. The mineral phase tends to remain relatively stable over time, though ions dissolved in soil water can infiltrate and fill pore space with new minerals such as calcite or silica. While bones share a basic molecular design, their biomechanical and physiological functions require diverse sizes, shapes, textures, and densities, which lead to variable durability. Bones that are small, thin, flat, and porous tend to break down more rapidly than those that are large, thick, rugged, and compact. In general, the dense bone in shafts and external surfaces (cortical or lamellar tissue) exhibits greater survivability in buried contexts than the spongy bone in the ends of long bones, the vertebral centra, and the cores of many smaller and thinner elements (cancellous or trabecular tissue). Teeth are among the most durable elements in the skeleton. The interior of the tooth, which consists of dentine, is sealed by enamel over the crown and by alveolar bone and a thin layer of cementum over the root. Once an individual dies and the body is buried, the dense, rigid, prismatic structure of enamel deteriorates through demineralization, often by leaching. Once the dentine is exposed, its relatively elastic tissue degrades more rapidly than enamel through the loss of protein. As it decomposes, the crown does not split or flake 7. Pollard (1996, pp. 149–153) and Schultz (1997) discuss the composition and structure of human bone tissue. Garland and Janaway (1989, pp. 25–27), Lyman (1994, pp. 417– 433), and Child (1995) survey the diagenesis of bone. Carlson (1990) and Lyman (1994, pp. 79–80) describe the basic

structure of teeth in vertebrates. Beeley and Lunt (1980) and Hillson (1996, pp. 181, 196–197, 206) address the diagenesis of human teeth; Hillson (1996, pp. 21, 24–29, 35–37, 44–45, 49–51, 56–58) discusses the postmortem fragmentation of teeth by class.

THE CONDITION OF THE SKELETAL ASSEMBLAGE

239

like bone but powders, and the cementum-enamel junction (CEJ) eventually cracks. When the gingivae and periodontal ligaments have disintegrated, teeth can detach from the sockets and migrate from the jaws. Bones and teeth buried at the Isthmus underwent significant changes as a result of environmental and anthropogenic agents. Although these processes have been surveyed, it will be useful to describe their effects on the skeletal assemblage in greater detail as a preface to osteological analysis. Bones exposed to moisture by burial in a drainage zone or an unprotected cist became friable and porous, broke or crumbled when handled, and exhibit moderate to severe flaking, pitting, and splitting parallel to the fibrous structure (Figs. 5.19, 5.20, 5.39). Teeth buried in similar conditions display pitted, brittle, or even powdery enamel, flaky or absent cementum, soft to mushy dentine, and roots crumbling from the distal ends up (Figs. 6.22, 6.33, 6.40, 6.46). Cervical fractures of teeth were not uncommon, especially in the anterior dentition, so that crowns sometimes snapped from their roots (Figs. 6.13, 6.15, 6.26, 6.42, 6.43). Water in certain graves coated bones with thin sediment, either a fine clay or a granular silt, and several associated teeth displayed shallow encrustations or grayish to brown discoloration. Some bones are deformed or broken as a result of compression under sedimentary overburden. Ribs often cracked under the weight of overlying corpses, compacted burial fill, and sunken cover slabs or tiles, so that the ribs from large multiple interments are fragmentary and hard to identify accurately. Likewise, enamel in several teeth display vertical fractures or marginal chips caused by chemical degradation as well as sedimentary compression and movement within the cist after detachment from the jaws.8 Small roots entering graves sometimes adhered to bone surfaces, etching and staining them (Figs. 5.1, 5.2).9 In a few instances, sinuous roots grew through openings in bones and dislocated small, delicate elements; in other cases, roots degraded osseous regions already weakened by natural agents, so that small, irregular cavities developed, particularly in the cranial vault (Figs. 5.2, 5.3, 5.28), but also in the scapular and iliac fossae.10 Finally, the handling of bones during construction and secondary burial in Late Roman and Byzantine times led to the fracturing of bones, especially cancellous tissues and thin regions (Figs. 2.9, 2.69). Modern excavators inflicted damage on crania in particular. Several cracks, punctures, and scrapes visible on the crania were caused by picks, trowels, or shovels (Figs. 5.16–5.18, 5.22, 5.23, 5.30, 5.35, 5.38).11 Since formation processes at the Isthmus had a differential impact on the material record, all human remains have not survived in the same state. In order to assess the overall condition of the skeletal assemblage, the transformation of the bones and teeth after death and deposition was quantified using a procedure in several steps.12 The first step was to inventory those excavated remains that were found in storage.13 Next, all major bones and many subregions, such as the proximal, midshaft, and distal thirds of the long bones, were scored 8. E.g., NEG 69-004C (M1R, M2R, M1L, M2L), 69-005B (M1L), T2 68-003A (I1L), 68-003B (I1L), T14 67-002A (I1L), DEC 69-901B (I1R), HO 70-901B (I2R, C1R, M1R), 70-901C (I2R, M2L, M3L), 70-902D (M1L), 70-902? (P4R). 9. NEG 67-001C, T14 69-003A, DEC 69-901A, HO 70-902B, C, H. 10. NEG 67-001C, 67-001I, 69-007A, T2 68-006A, LOU 69801A. 11. E.g., NEG 69-001A (frontal eminence punctured), 69-001D (left sphenoid punctured, occipital scraped), 69005C (right sphenoid/frontal damaged), T2 68-002A (occipital and parietals fractured, punctured, scraped), 68-003A (left parietal fractured/scraped; occipital fractured), RB 76002C (posterior right parietal fractured), HO 70-902F (left

mastoid punctured), 70-902J (right lambdoid punctured). The ragged perforation in a mesosternum from HO 70-902 (Fig. 7.35) might have resulted from excavation or handling (p. 400). 12. This procedure was adapted from Buikstra and Ubelaker 1994, pp. 6–8, table 1, attachment 1. 13. Ribs were not consistently recorded because they were often too fragmentary to identify. The general impression is that they survive at roughly the same rate as the vertebrae, only in worse condition, and that the first and second ribs are better represented than the lower ones. Other minor bones that were not recorded separately were ethmoids, vomers, hyoids, nasals, lacrimals, inferior nasal conchae, mallei, incudes, stapes, coccyges, and sesamoids.

240

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Figure 5.1. Surficial etching on right parietal and

occipital of DEC 69-901A, superior-posterior view

Figure 5.2. Surficial etching and structural degradation on left

parietal and occipital of NEG 67-001C, superior-posterior view

Figure 5.3. Cavitation, flaking, and fracturing on frontal

of NEG 69-007A, superior-anterior view

THE CONDITION OF THE SKELETAL ASSEMBLAGE

241

on the following four-point scale of completeness: 0 = completely absent, 1 = 25% present, 2 = 50% present, 3 = 75% present, 4 = completely present. Scoring was done by rapid inspection, but the results were found to be accurate when double-checked. Certain surviving bones were not scored for completeness,14 certain bone types were combined into larger groups for scoring,15 and certain bones were scored by side of the body. The choice of bones and subregions to be scored was determined by the main interpretive concerns of osteological analysis: morphology, paleodemography, and paleopathology. In this scheme, the score of completeness for a single element indicates its relative analytical value. On the one hand, joint surfaces that scored 3 or 4 (75%–100% complete) can be observed for degenerative changes and are therefore useful for calculating the frequency of joint disease. On the other hand, joint surfaces scored 1 or 2 (25%–50% complete) are not useful because not enough of the surface exists to ascertain whether the joint was arthritic. On the basis of those numeric data, it was possible to assess the completeness of the entire skeletal sample after several centuries of alteration by natural and cultural agents. Two calculations were made for each skeletal element and subregion, one measuring the rate of survival and the other measuring the rate of preservation of those that survived. Before making these calculations, the data were classified by age at death (0–5 years, 5–15 years, all subadults, adults)16 in order to estimate the relative impact of age on bone survival and preservation. The rate of survival was the percentage of all bones or subregions expected that actually existed for examination.17 The number of bones and subregions expected depended on the premise that the number of individuals buried was 6918 and that those 69 individuals each had complete, normal skeletons19 at the time of death and primary burial. Because the total number of bones and subregions recorded per individual was 233, the total number expected for 69 individuals is 16,077. The rate of preservation was determined by calculating the arithmetic mean of all completeness scores, dividing the mean by four, and multiplying the quotient by the survival rate. For example, if 50% (68) of the total number of radial distal epiphyses expected (136) survived with an average score of 3.2, then the rate of preservation for that element would be 40%. The calculation of rates of survival and preservation revealed that the overall condition of the skeletal assemblage is relatively good, with several elements preserved in over half the 69 individuals and some in nearly two-thirds. Of the 16,077 total elements expected in the 69 individuals, 5,517 (34.32%) survive.20 This composite rate has little significance per se because preservation varied widely by age, element, and depositional context (Figs. 5.4, 5.5). Overall the bones of subadults are much more poorly preserved than the bones of adults. The rate of survival of the 21 subadult skeletons (25.69%) is 14. I.e., vertebrae, carpals, tarsals, metacarpals, metatarsals, and phalanges. Although sometimes vertebral centra have deteriorated or laminae have separated, adult vertebrae were sufficiently preserved to observe the presence or absence of abnormalities such as nodular impressions or osteophytes. If small hand and foot bones have survived, they are almost always intact, though occasionally the shafts of the metacarpals and metatarsals have snapped. Only tali and calcanei were scored for completeness because their degree of preservation varies considerably. 15. Phalanges of the hands and feet were not differentiated or sided. 16. The individuals once buried in NEG 69-008 and DEC 69-902, which preserved no significant human remains, were classed as subadults; HO 70-902C, aged 5–6 years, was classed as 5–15 years. 17. Waldron (1987, pp. 57–63, table 6.1) and Mays (1998,

pp. 22–23, fig. 2.6) provide good examples of survival calculations for large skeletal samples from Romano-British and Medieval British cemeteries. 18. This includes one individual from each grave in n. 1, above. It excludes the four individuals represented by isolated bones in secondary contexts at the Northeast Gate (p. 55). 19. A “complete, normal” adult skeleton has 206 bones, counting the sacrum and coccyx as units and assuming that there are 12 thoracic vertebrae and 24 ribs; the number of bones in the subadult skeleton is larger, depending on age. There were no cases of supernumerary vertebrae at the Isthmus. Rib counts were not regularly recorded, but at least one individual (NEG 69-007A) had lumbar ribs. 20. Note that this calculation underestimates by as much as 5% the proportion of the total assemblage that survived, because a number of fragments from each burial could not be identified and therefore were not scored.

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Bones and Subregions

242

Phalanges Humerus prox. epi. Carpals Tibia prox. epi. Patella Scapula body Femur dist. epi. Fibula prox. epi. Mesosternum Tibia dist. epi. Scapula acromion Ilium Auricular surface Fibula dist. epi. Calcaneus Ischium Manubrium Tarsals Radius prox. epi. Femur prox. epi. Pubis Humerus dist. epi. Ulna dist. epi. Sacrum Radius dist. epi. Lumbar vertebrae Tibia dist. third Tibia prox. third Humerus prox. third Ulna prox. epi. Zygomatic Metatarsals Clavicle Sphenoid Thoracic vertebrae Radius prox. third Tibia midshaft Fibula dist. third Talus Femur dist. third Radius midshaft Radius dist. third Fibula prox. third Palatine Metacarpals Femur prox. third Ulna dist. third Humerus dist. third Maxilla Cervical vertebrae Humerus midshaft Temporal Ulna prox. third Fibula midshaft Mandible Parietal Femur midshaft Ulna midshaft Frontal Occipital

0

10

20

30

40

50

60

% Preserved of Expected Total Figure 5.4. Differential rates of preservation of skeletal elements in 48 adults

70

Bones and Subregions

THE CONDITION OF THE SKELETAL ASSEMBLAGE

243

Mesosternum Talus Calcaneus Patella Carpals Manubrium Phalanges Tarsals Tibia prox. epi. Tibia dist. epi. Metatarsals Sacrum Femur prox. epi. Metacarpals Femur dist. epi. Auricular surface Thoracic vertebrae Lumbar vertebrae Occipital Zygomatic Ulna dist. epi. Palatine Radius dist. epi. Cervical vertebrae Radius prox. epi. Sphenoid Fibula prox. epi. Humerus prox. epi. Pubis Fibula dist. epi. Ulna prox. epi. Ilium Tibia prox. third Ischium Humerus dist. epi. Frontal Maxilla Tibia dist. third Femur prox. third Mandible Femur dist. third Radius dist. third Scapula body Scapula acromion Temporal Parietal Humerus prox. third Clavicle Tibia midshaft Femur midshaft Ulna dist. third Ulna prox. third Radius prox. third Radius midshaft Fibula prox. third Humerus dist. third Fibula dist. third Humerus midshaft Ulna midshaft Fibula midshaft

0

10

20

30

40

50

60

70

% Preserved of Expected Total Figure 5.5. Differential rates of preservation of skeletal elements in 21 subadults

244

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

considerably less than the rate of survival of the 48 adult skeletons (38.90%). When subadults are separated into age categories, the bones of 14 young children (0–5 years) survive at a rate of 21.31%, while the bones of five older children (5–15 years) survive at a rate of 32.96%.21 Furthermore, when the preservation of separate elements in subadults is compared to that in adults, in almost all cases, particularly the head, the vertebral column, and the hands and feet, preservation falls far below that in adults.22 This discrepancy in survival by age should be attributed largely to age-specific osteological conditions that promote natural factors of decay. The developing bones of children, especially neonates and infants, are less dense and smaller than those of older individuals and have less mineral and more collagen than mature bones.23 Therefore, the skeletons of young children are particularly susceptible to compression under sediment and dissolution by chemical decay. Many immature bones found at the Isthmus were crushed, while those in partly enclosed cists in drainage zones survived only in scattered traces, if at all, especially neonatal or infant remains.24 The bones of children in better sealed interments in welldrained areas were still more poorly preserved than adult remains in the same or adjacent graves.25 The methods of excavation did not facilitate the total recovery of subadult bones, which can be small and featureless, resembling pebbles or lumps of dirt. The loss of 93% of hand and foot bones and over 75% of vertebrae from subadults, when the same graves produced certain long bone shafts at a ca. 60% rate, must be attributed at least in part to procedural error. While adult bones were generally in much better condition than those of subadults, different parts of the skeletons also displayed separate rates of preservation. As may be expected, the cortical diaphyses were better preserved than the cancellous epiphyses, and their preservation decreases in direct proportion to distance from the midshaft. Bony structures of the pelvic and shoulder girdles, which are relatively thin, light, and irregularly formed, exhibited low preservation. Apart from these expected patterns, the preservation rates of certain bones call for special explanation. Among the best-preserved elements were parts of the skull. It is striking that delicate areas of the cranium, such as the maxilla and the palatine, survived more frequently than several durable appendicular shafts. One explanation is that the excavators treated the bones of the head preferentially. Although their incautious treatment of crania during excavation often caused superficial damage, many of them were saved in a scraped, fragmentary condition. This concern for the bones of the head may explain the unusually high frequency of cervical vertebrae, which are better represented than more robust thoracics and lumbars. The poor condition of bones in the anterior thoracic cavity probably resulted from crushing under sedimentation, collapsing cover slabs, or overlying bodies. The impact of natural and human disturbances on individual graves can also be calculated. The survival and preservation rates of bones by grave are summarized in Table 1.3. This tabulation shows that exposure to multiple detrimental factors had a cumulative effect on the condition of human remains in interments. For instance, bones in the graves at the Northeast Gate that were subjected to drainage, disturbed by builders, and incompletely enclosed were on average more poorly preserved than the relatively well-protected graves near Tower 14. While the influences of single factors are difficult to estimate for any given 21. The ages at death of the children buried in NEG 69-008 and DEC 69-902 are unknown. Since they have been excluded from these calculations of survival among younger versus older children, it can be assumed that the actual rates of survival for one or both subadult age classes are lower. 22. Note that the rate of preservation in skeletal remains of Late Roman to Early Byzantine date is roughly comparable to

that in the full skeletal sample from the site. 23. Gordon and Buikstra 1981; Von Endt and Ortner 1984; Guy, Masset, and Baud 1997; Pearce 2001, pp. 129–132. 24. NEG 67-003D, E, 69-009A, 69-010A. This was probably also the case for the subadults once buried in NEG 69-008 and DEC 69-902. 25. T14 69-991A, HO 70-902A–C.

THE CONDITION OF THE SKELETAL ASSEMBLAGE

245

burial environment, some general patterns are evident. Natural agents were responsible for consistent but limited destruction of the hard tissues, which was regulated by morphological differences between skeletal elements. The impact of anthropogenic factors can be evaluated on more secure grounds. The removal of bones from earlier graves, such as by Byzantine builders at Tower 14, accounted for the loss of less than 1% of the entire skeletal assemblage at the site. Secondary burial had a greater impact. The fragmentary remains that were reinterred in HO 70-901 comprise only ca. 20% of three adult skeletons, extracted almost entirely from the head, limbs, and innominates (Fig. 4.2). Secondary burial led to the loss of ca. 2%–3% of the entire skeletal assemblage, mostly representing the limbs and extremities. The most significant factor in the survival of bones within graves was the method of excavation. For example, the cleaning in 1994 of the spoilage from the excavation of HO 70-902 in 1970 led to the discovery of discarded bones and teeth that comprised ca. 4% of the ten skeletons in the cist.26 These bones were mostly small elements that are easily overlooked, such as vertebrae and hand and foot bones. In other cases, the excavators decided not to save bones from a grave, which led to the loss of large portions of the skeleton or of the entire body.27 Deficient techniques of recovery probably accounted for the loss of ca. 5%–10% of the entire assemblage, or up to 1,200 individual elements. A significant proportion of teeth were lost for similar reasons. The survival of erupted dentition in 61 skeletons is presented in Table 5.2. Since nearly all surviving teeth are intact,28 unlike bones, teeth were not scored for completeness and rates of dental preservation were not determined. The calculation of dental survival is a simple ratio of the number of teeth present to the number of teeth expected. Teeth in each of the four classes (incisors, canines, premolars, molars) were scored as present when they had erupted and were intact and as absent when they had erupted but were missing. The total number of teeth present is 725, most of which are molars and premolars. The number of teeth expected in the sample depended on the age distribution and on the prevalence of hypodontia (congenital absence) and antemortem loss (AMTL). The number of erupted teeth in individuals who have not reached biological maturity varies by age. Only subadults whose dental development could be accurately assessed on the basis of preserved teeth are included in the calculation.29 The total number of individuals scored for dental survival is 61, 47 adults and 14 subadults.30 Based on the calculation of expected teeth for each of the subadults and the assumption that all adults had 32 teeth, the total number of teeth expected is 1,784, of which 725 survive, or 40.64% (Table 5.2). All 61 individuals, however, do not have complete dental arcades because several display hypodontia and AMTL. The congenital absence of teeth, particularly the third molars, can only be ascertained in adults, whose permanent teeth are fully erupted. Moreover, AMTL was only observed in erupted teeth in adults at the Isthmus. Therefore, the rate of dental survival must be adjusted to account for teeth that were already missing from adults at death. It is uncertain how many adult teeth of the total 1,504 expected were either congenitally absent or lost before death because only 1,092 (905 adult and 187 subadult) alveoli survive.31 However, since a majority of adult alveoli were sufficiently intact for examination 26. NBs 10, pp. 47, 51–53 and 1990, pp. 124–129. In contrast, the excavators in other areas, such as the Northeast Gate, were able to recover the most exiguous bones of the human body: the sesamoids, the hyoid, and the auditory ossicles. 27. T13 54-001, T14 69-701, TC 60-001. 28. A very small number (less than 10) were too fragmentary to assign to a tooth class or observe for dental traits. 29. The infants NEG 67-003D, 67-003E, 69-009A, 69-010A, and HO 70-902A are excluded on these grounds, as are the

children buried in NEG 69-008 and DEC 69-902. 30. This total excludes the five isolated bones from at least four individuals in secondary contexts at the Northeast Gate, as well as T13 54-001, from which no bones were saved. NEG 69-007B is included in this calculation as an adult. 31. Technically speaking, scoring the congenital absence of teeth requires the preservation not of the sockets but of the alveolar regions where the teeth develop.

246

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.2. SURVIVAL OF ERUPTED DENTITION IN 61 SKELETONS Total Expected Teeth

Expected Teeth: 47 Adults

Incisors

488

376

112

151

113

Canines

236

188

48

102

Premolars

384

376

8

676

564

1,784

1,504

Tooth Class

Molars Total

Tooth Class

Total Preserved Sockets in Adults

Expected Teeth: Total Teeth Teeth Present: Teeth Present: 14 Subadults Present 47 Adults 14 Subadults

Total Survival Rate (%)

Survival Rate: 47 Adults (%)

Survival Rate: 14 Subadults (%)

38

30.94

30.05

33.92

89

13

43.22

47.34

27.08

177

173

4

46.09

46.01

50.00

112

295

220

75

43.64

39.01

66.96

280

725

595

130

40.64

39.56

46.43

AMTL (n)

AMTL Prevalence (%)

Sockets Absent in Adults

Predicted Hypodontia in Absent Sockets

Predicted AMTL in Absent Sockets

Predicted AMTL + Hypodontia in Absent Sockets

Hypodontia Hypodontia Prevalence (n) (%)

Incisors

228

0

0.00

21

9.21

148

0

14

14

Canines

125

0

0.00

6

4.80

63

0

3

3

Premolars

245

0

0.00

21

8.57

131

0

11

307 (109)

24 (24)

7.82 (22.02)

69

24.38

257

20

58

78

905

24

2.65

117

13.28

599

20

86

106

Molars (M3) Total

Tooth Class

Adjusted Total Expected Teeth

1 3

Adjusted Expected Teeth: 47 Adults

11 2

Adjusted Total Survival Rate (%)

Adjusted Survival Rate: 47 Adults (%)

Incisors

474

362

31.86

31.22

Canines

233

185

43.78

48.11

Premolars

373

365

47.45

47.40

Molars

598

486

49.33

45.27

1,678

1,398

43.21

42.56

Total

This percentage accounts for the fact that congenitally absent molars could not have been lost antemortem. Therefore, it is the ratio of 69 olars lost antemortem to 283 fully erupted molars. 2 This predicted number accounts for the fact that congenitally absent molars could not have been lost antemortem. 3 This percentage accounts for the fact that congenitally absent molars could not have been lost antemortem. Therefore, it is the ratio of 117 eeth lost antemortem to 881 fully erupted teeth. 1

(54%–66%, depending on dental class), the total number of teeth lost because of hypodontia and AMTL could be extrapolated from the known rate among the preserved sockets.32 If it is assumed that the prevalences of hypodontia and AMTL in the separate dental classes were consistent throughout the entire dental sample, then the 47 adults were missing an additional 106 teeth at death, mostly molars. This adjustment yields a more accurate estimate of 1,678 teeth expected in 61 individuals, in which case the 725 surviving teeth represent 43.21% of all those buried (Table 5.2).33 The chief reason for this low rate of survival was not natural destruction but the procedure of excavation. Only a few teeth display extensive postmortem deterioration. Moreover, except for the graves of young children, in which fragile dental crowns had mostly dissolved, the survival rates of deciduous (42.24%) and permanent teeth (40.34%) are comparable. The techniques of excavation were the main obstacle to the full recovery of teeth. Clearing 32. The calculation of the prevalence of AMTL among molars accounted for the fact that congenitally absent molars could not be lost before death.

33. Note that the rate of survival among teeth of Late Roman to Early Byzantine date was roughly comparable to that in the full dental sample from the site.

OSTEOLOGICAL METHODOLOGY

247

the spoil heap from HO 70-902 led to the discovery of six teeth, ca. 2% of those expected, which had been overlooked during sifting and were discarded. Many more teeth were probably lost when fill from the cist was dumped into the Great Ravine. That a large fraction of the teeth had already detached from their jaws by the time of excavation is indicated both by the fragmentary condition of the surviving mandibles and maxillae and by the number of teeth (23, or ca. 8%) recovered loose from the fill inside HO 70-902. Across the site, certain tooth classes survived more frequently than others. The adjusted total survival rates for premolars (47.5%) and molars (49.3%) are considerably higher than those for canines (43.8%) and incisors (31.9%). This variation reflects the recoverability of separate classes during excavation. Posterior and cheek teeth are easy to find because they are often anchored to the jaws by complex roots and are comparatively large. In contrast, anterior teeth have single roots and are small, particularly the deciduous incisors. The survival and preservation of bones and teeth have directly influenced the course of the osteological examination and the validity of its conclusions. The good preservation of skulls means that sex and age at death could be accurately determined for several individuals. The cranial and mandibular remains produced many metric and nonmetric data for the study of morphological diversity. The large number of surviving diaphyses allowed for a meaningful assessment of the prevalence of injury to limbs. While less than half of the teeth survive, many of these could be associated with particular skulls and are an important source of information on diet, development, and oral health. Sex and age are best determined from the bones of the pelvic girdle, which are only moderately well preserved. Similarly, the number of surviving vertebrae does not offer a complete picture of the severity or distribution of osteoarthritis in the spinal joints. The preservation of the ulna, radius, and femur is moderate to high, but these bones seldom survive intact, which limits the evidence for stature. Other skeletal regions that broke down after deposition were the thorax, most joints but especially the shoulder, elbow, and knee, and the hands and feet. The underrepresentation of the appendicular joints, hands, and feet hindered the classification of arthropathies.34 Subadult remains are more fragmentary than adult remains, which diminishes the record of infant and juvenile health. In sum, the composition of the skeletal assemblage as a result of diverse formation processes does not permit a perfect reconstruction of life in the Roman and Byzantine settlement at the Isthmus. It does, however, provide a solid basis upon which to build a provisional, approximate model.

OSTEOLOGICAL METHODOLOGY The osteological methods applied to the examination of the human remains from the Isthmus are commonly used by physical anthropologists. This section outlines the basic procedures and standards used for analysis.35 The accuracy and precision of the identifications of sex, age, and morphology in the skeletal sample depend on the degree of preservation and the choice of analytical strategies. The applicability of certain osteological methods to the sample from the Isthmus is an important consideration because many techniques for determining sex and estimating stature and age at death are derived from modern Americans.36 34. See Waldron 1987, p. 63. 35. The following manuals were consulted: Krogman and İşcan 1986; Buikstra and Ubelaker 1994; Moore-Jansen, Ousley, and Jantz 1994; Bass 1995; Ubelaker 1999; Scheuer and

Black 2000; White 2000. İşcan and Kennedy (1989) and Katzenberg and Saunders (2000) provide useful overviews of osteological analysis. 36. See White 2000, pp. 338–339.

248

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Determination of Sex Sexual dimorphism, the physiological and anatomical difference between males and females, is manifest not only in the soft tissues of the body but also in the skeleton. Since bones grow and change shape through infancy, childhood, and adolescence, with boys and girls maturing at different rates, sexual differences are most apparent in adult skeletons. Moreover, the bones of both sexes tend toward uniform gracility before biological maturity. For this reason, sex can be determined most accurately in adult remains. Several techniques have been proposed for sexing immature remains,37 but these are not universally accepted and have not been applied in the present study. When anthropologists determine the sex of adult bones from ancient contexts they use techniques from the study modern populations.38 The skeletal regions that give the most reliable indication of sex are the pelvis and the skull. The primary pelvic features observed in this study were the structures of the subpubic region (ventral arc, subpubic concavity, ischiopubic ramus),39 but other characteristics were also considered, namely, the relative width of the greater sciatic notch, the presence or absence of the preauricular sulcus, the relative rugosity and size of the acetabulum, and the elevation of the auricular surface of the ilium. The primary cranial and mandibular features observed in this study were the relative robusticity of the nuchal crest, the external occipital protuberance, and the mastoid process, the relative bluntness of the supraorbital margin, the relative prominence of the supraorbital ridge and glabellar region, the relative projection of the mental eminence, the degree of gonial eversion, and the overall rugosity of the bones. Certain areas of limb bones, particularly the heads of the humerus and femur, also exhibit sexual dimorphism in size. The measurements of these features in the human remains from the Isthmus were checked against standard size ranges for both sexes.40 The determinations of sex for 65 skeletons and the techniques applied for those determinations are recorded in Table 5.3. It was possible to determine sex in most skeletons because requisite elements were sufficiently preserved. The anterior and posterior innominate, the temporal and occipital, and the mandible often survived either intact or in substantial fragments in the depositional environment because these bones are dense and thick (Fig. 5.4). Therefore, in all but 11 cases,41 sex could be determined on the basis of both cranial and pelvic morphology. Overall the sample shows a high degree of sexual dimorphism, so that most adult skeletons can be confidently identified as either male or female. The sex of seven individuals is designated with a question mark because the remains were not sufficiently preserved for definite identification.42 The sex of an eighth individual (NEG 69-007B) is marked “M?” because the skeleton belonged to a young adult aged 16 to 18 years. The cranial and mandibular features in this individual are rather gracile, but the mature pelvic features indicate a male. In addition, the fragmentary crania of certain older females are sexually ambiguous.43 These specimens reflect the slight masculinization of cranial structures that often developes with age.44 37. Krogman and İşcan 1986, pp. 190–191, 200–208; Saunders 2000, pp. 138–141; Scheuer and Black 2000, pp. 15–17. 38. For summaries, see Buikstra and Mielke 1985, pp. 367– 385; Krogman and İşcan 1986, pp. 191–200, 208–247; Buikstra and Ubelaker 1994, pp. 16–21. 39. Phenice 1969. 40. Stewart 1979, p. 100 (vertical diameter of humeral head); Bass 1995, p. 230, table 3-29, after K. Pearson (vertical diameter of femoral head; bicondylar breadth). 41. NEG 67-001A, 67-003A, 69-001D, T2 68-006A, 68-003B, RB 76-002A, HO 70-902D–F, I, TC 60-001A.

42. NEG 69-001D (F?), RB 76-002B (M?), HO 70-901C (M?), 70-902D–F, I (M?). The pelvic morphology visible in the photograph of TC 60-001A in situ (Fig. 2.83) corroborates the determination of sex from the cranium as female. Although no bones from PAL 56-001A survive for examination, the funerary artifacts indicate that the individual was most probably female; see pp. 93–94. 43. RB 76-002A (35–44 years), HO 70-901A (35–44 years). WF 62-001A (mature adult). However, the cranial morphology of NEG 69-007A (45+ years) is relatively gracile. 44. See Meindl et al. 1985a, pp. 81, 85.

OSTEOLOGICAL METHODOLOGY

249

Estimation of Age at Death The human skeleton undergoes various changes throughout life, from formation and development in infancy and childhood, to growth and fusion in adolescence, and stability to degeneration in adulthood. Anthropologists can estimate the age at death of human remains from archaeological sites by reference to this progressive sequence in modern populations of known age. Age estimates based on immature remains are more accurate and more precise than those based on mature remains. Without the documentary evidence of, for instance, an epitaph or a certificate of decease, the age at death of an adult skeleton from an ancient grave cannot be confidently estimated within a year or two on osteological grounds. An adult skeleton can be assigned more securely to a broader age class of five years, ten years, or longer.45 This study employed three standard techniques to age subadult skeletons and three to age adult skeletons. Rather than preferring any one technique to age an individual, this study adopted a multifactorial approach in which several elements of each skeleton were evaluated in combination as segments of an integrated system. A strategy such as this improves the accuracy and reduces the bias of final age estimates.46 The main techniques used to age subadults were dental development,47 epiphyseal fusion, and the unification of the primary ossification centers in the vertebrae and pelvis.48 A mark of general age observed in the subadult skeletons is the maximum diaphyseal lengths of the long bones and the maximum width of the ilium.49 The seriation of bone sizes within the skeletal sample from the Isthmus also served as an indicator of relative age among subadults and as a discriminator between subadults and adults. The main techniques used to age adult skeletons were the metamorphosis of the pubic symphysis,50 the metamorphosis of the auricular surface of the ilium,51 and the closure of the cranial sutures.52 Moreover, advancing age was reflected in certain structural changes, such as dental wear, degenerative joint disease especially in the vertebrae, and the ossification of costal cartilage. In particular, once dental wear was correlated with ages derived from pubic symphyses and auricular surfaces, it served as a reliable method for identifying the age class of skeletons lacking other indices.53 The metamorphosis of the sternal rib end could not be used to age skeletons because too few ribs were sufficiently preserved at the Isthmus.54

45. Buikstra and Ubelaker 1994, p. 36; Waldron 1994, p. 20; White 2000, pp. 341–342. 46. For critical discussions of the multifactorial aging method, see Lovejoy et al. 1985b; Bedford et al. 1993. 47. Ubelaker 1999, pp. 63–65, fig. 71. Smith (1991) summarizes various techniques for aging by dental development. 48. Krogman and İşcan (1986, pp. 50–102) and Scheuer and Black (2000) offer full discussions; Buikstra and Ubelaker (1994, pp. 40–43, fig. 20) provide a useful summary. In addition, the fusion of the spheno-occipital synchondrosis was a useful age marker in young adult skeletons. 49. Ubelaker 1987, pp. 1258–1260; 1999, pp. 65–69. Since growth rates vary significantly both between and within populations, the application of these standards based on modern American or early Native American skeletons to archaeological samples from a different region and period can only provide a broad estimate of age. 50. This study applied both the Todd (1920, 1921) and Suchey-Brooks methods (Katz and Suchey 1986; Brooks and Suchey 1990). It was found that the two methods produced

comparable results. For a critical discussion of age determination from pubic remains, see Meindl et al. 1985b. 51. Lovejoy et al. 1985a. 52. Meindl and Lovejoy 1985. Buikstra and Ubelaker (1994, pp. 32–35, fig. 12) offer a more elaborate version of the same technique that includes endocranial and palatal sutural points; cf. White 2000, pp. 345–347, fig. 17.4. 53. See Walker, Dean, and Shapiro 1991, pp. 169–170. Anthropologists have devised several schemes for aging dental attrition in Neolithic to Medieval British (Miles 1962; Brothwell 1981, pp. 71–72, fig. 3.9) and Native American skeletal samples (Lovejoy 1985); for summaries, see Hillson 1996, pp. 239–242; Mays 1998, pp. 57–66. These standards cannot be directly applied to the inhabitants of the Isthmus because of dietary and cultural differences between peoples. It will be seen (Chap. 6) that the overall degree of the wear at the Isthmus is lower than that in these major reference populations. 54. İşcan and Loth (1989, pp. 27–29, fig. 2) discuss this technique.

250

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.3. SEX AND AGE AT DEATH OF 65 SKELETONS SEX

AGE AT DEATH

Pubic Symphysis

Auricular Surface

Cranial Suture Closure

Relative Size

-

-

+

+

-

-

-

-

+

+

+

-

-

+

-

-

-

-

-

-

-

+

-

+

-

35+

-

-

-

-

-

+

-

25–34

25–34

+

-

+

+

+

+

-

35–44

35–44

-

-

-

+

+

+

-

-

35–44

35–44

-

-

-

+

-

-

-

+

+

35–44

35–44

-

-

-

+

-

+

-

M

+

+

45+

Late 40s

-

-

-

+

-

+

-

F

+

+

15–24

20–21

+

-

+

+

+

+

-

NEG 67-001E

n/a

-

-

5–9

7–11

+

-

+

-

-

-

-

NEG 67-001F

n/a

-

-

5–9

4–8

+

-

+

-

-

-

-

NEG 67-001G

n/a

-

-

5–9

4–8

+

-

+

-

-

-

-

NEG 67-001H

n/a

-

-

0–4

2–4

+

-

+

-

-

-

-

NEG 67-001I

n/a

-

-

0–4

2–4

+

-

+

-

-

-

-

NEG 67-001J

n/a

-

-

0–4

1–2

+

-

+

-

-

-

-

NEG 67-001K

n/a

-

-

0–4

2–4

+

-

+

-

-

-

-

NEG 67-003A

F

+

-

35–44

35–44

-

-

-

+

-

-

-

NEG 67-003B

M

+

+

35–44

35–44

-

-

-

+

-

+

-

NEG 67-003C

n/a

-

-

0–4

3–5

+

-

+

-

-

-

-

NEG 67-003D

n/a

-

-

0–4

0–1

-

-

-

-

-

-

+

NEG 67-003E

n/a

-

-

0–4

0–6 mos.

-

-

-

-

-

-

+

NEG 69-005A

n/a

-

-

10–14

10–12

+

-

+

-

-

-

-

NEG 69-005B

F

+

+

15–24

20–22

+

-

+

+

-

-

-

NEG 69-005C

M

+

+

25–34

25–34

-

-

-

+

+

+

-

NEG 69-001A

F

+

+

45+

50s or older

-

-

-

+

+

+

-

NEG 69-001B

n/a

-

-

0–4

3–4

+

-

+

-

-

-

-

NEG 69-001C

F

+

+

15–24

20–24

+

-

+

+

-

+

-

NEG 69-001D

F?

-

+

Mature adult

45+

+

+

-

-

-

+

-

-

-

-

+

+

+

-

35–44

-

3–4

+

45+

Late 40s

+

Mature adult

+

+

+

+

M

+

NEG 67-001B

F

NEG 67-001C NEG 67-001D

Skull Morphology

-

Pelvic Morphology

Dental Development

Osseous Fusion

Techniques Applied Dental Wear

Techniques Applied

Age Class (years)

Age Estimate (years)

+

+

45+

Late 40s

+

+

35–44

-

-

0–4

F

+

+

NEG 69-004C

M

+

NEG 69-004D

F

NEG 69-004E

F

NEG 67-001A

Burial Number

Determination

NEG 69-103A

F

NEG 69-103B

M

NEG 69-004A

n/a

NEG 69-004B

NEG 69-007A

F

+

+

45+

Late 40s– mid-50s

NEG 69-007B

M?

+

+

15–24

16–18

+

-

+

-

-

-

-

NEG 69-009A

n/a

-

-

0–4

0–10 mos.

+

-

-

-

-

-

-

NEG 69-010A

n/a

-

-

0–4

1–3

+

-

-

-

-

-

-

OSTEOLOGICAL METHODOLOGY

251

TABLE 5.3 (CONT.) SEX

AGE AT DEATH

Osseous Fusion

Pubic Symphysis

Auricular Surface

Cranial Suture Closure

Relative Size

35–44

-

-

-

-

+

-

-

Skull Morphology

Age Estimate (years)

Pelvic Morphology

Age Class (years)

Dental Wear

Techniques Applied Dental Development

Techniques Applied

T2 68-006A

M

+

-

35–44

T2 68-002A

M

+

+

35–44

35–44

-

-

-

+

+

+

-

T2 68-003A

M

+

+

45+

Late 40s

-

-

-

-

+

+

-

T2 68-003B

F

+

-

Mature adult

35+

-

-

-

-

+

+

-

T14 67-002A

M

+

+

45+

Early–mid50s

-

-

-

+

+

+

-

T14 67-004A

M

+

+

25–34

25–34

-

-

-

+

+

+

-

T14 69-701A

–

-

-

Adult

15+

-

-

+

-

-

-

+

T14 69-991A

n/a

-

-

0–4

1–2

+

-

+

-

-

-

-

T14 69-002A

F

+

+

25–34

25–34

-

-

-

-

+

+

-

T14 69-003A

M

+

+

15–24

18–24

+

-

+

+

+

+

-

LOU 69-801A

M

+

+

Mature adult

35+

-

-

-

-

-

+

-

DEC 69-901A

–

-

-

Mature adult

45+

+

+

-

-

-

+

+

DEC 69-901B

–

-

-

Adult

15+

+

-

-

-

-

-

+

Burial Number

Determination

RB 76-002A

F

-

+

35–44

35–44

-

-

-

-

+

+

-

RB 76-002B

M?

+

+

35–44

35–44

-

-

-

+

+

+

-

RB 76-002C

F

+

+

35–44

35–44

-

-

-

-

+

+

-

HO 70-901A

F

+

+

35–44

35–44

-

-

-

-

+

+

-

HO 70-901B

F

+

+

35–44

35–44

-

-

-

-

+

+

-

HO 70-901C

M?

+

+

Adult

25+

-

+

-

-

-

+

-

HO 70-902A

n/a

-

-

0–4

0–1

-

-

+

-

-

-

-

HO 70-902B

n/a

-

-

0–4

2–4

+

-

+

-

-

-

-

HO 70-902C

n/a

-

-

5–9

5–6

+

-

+

-

-

-

-

HO 70-902D

M?

+

-

Adult1

25+

+

+

-

-

-

+

+

HO 70-902E

–

-

-

Adult2

25+

+

+

-

-

-

-

+

HO 70-902F

M?

+

-

Adult

25+

+

+

-

-

-

+

+

HO 70-902G

F

+

+

45+

50s or older

-

-

-

+

+

-

-

HO 70-902H

M

+

+

35–44

35–44

+

-

-

+

+

+

-

HO 70-902I

M?

+

-

Adult1

15+

+

-

-

-

-

+

-

HO 70-902J

F

+

+

15–24

15–24

+

-

-

+

+

+

-

WF 62-001A

F

+

+

Mature adult

35+

-

-

-

-

-

+

-

TC 60-001A

F

-

+

Young adult

15–34

+

-

+

-

-

+

-

2

Notes: The sex of subadults is marked “n/a” because it was not determined. A plus sign (+) indicates that a technique was applied to the skeleton. A minus sign (-) indicates that a technique was not applied, either because the skeleton was not sufficiently preserved or because it represented a subadult. 1 Either HO 70-902D or I should be assigned to age class 25–34 years. 2 Either HO 70-902E or F should be assigned to age class 45+ (estimated 55+ years).

252

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

The estimates of age at death for 65 skeletons and the techniques applied for those estimates are recorded in Table 5.3. The age of most subadult skeletons could be estimated because many teeth and axial and appendicular elements survived (Fig. 5.5). When dental development and osseous fusion were observable, the age of subadults could be estimated within one to four years and sometimes within less than a year. Individual skeletons were also assigned to one of three five-year age classes (0–4, 5–9, 10–14), which can be used to define composite age groups within the sample. In two cases (NEG 69-009A, 69-010A), the only remains found in the grave were dental crowns, which could be checked for their development. In another case (HO 70-902A) no teeth survived, but the fusion of the occipital could be checked.55 Relative size is the only index of age at death for two individuals represented by a few durable long bones (NEG 67-003D, 67-003E). The first skeleton (D) is aged 0 to 1 year because its humeral and fibular lengths are somewhat shorter than the contemporary NEG 67-001J, aged 1 to 2 years by dentition. However, the second skeleton (E) is aged 0 to 6 months, possibly perinatal, because its humeral and tibial lengths are considerably shorter than both D and another skeleton from a Late Roman grave, HO 70-902A, aged 0 to 1 year by occipital fusion. The very young age of D and E, which compares with NEG 69-009A (0 to 10 months), explains the poor preservation of these skeletons in an unprotected cist exposed to drainage.56 Subadult age estimates based on standardized tables for maximum diaphyseal lengths and iliac widths57 have a high correlation with age estimates based on dental development during early childhood, roughly the first five years of life. However, after that point, aging from bone measurements underestimates aging from dentition. This study adopted a conservative approach to aging adult skeletons by classes. If substantial portions of the pubic symphysis and/or the auricular surfaces of an individual were preserved, age could be estimated accurately within a decade (15–24, 25–34, 35–44) up to the middle 40s. Since skeletons of the youngest class furnished a wider array of indices than the older adult skeletons, their age could in most cases be estimated more precisely than 15–24 years. Furthermore, on account of the divergent ranges of estimates based on the pubic symphysis and auricular surface after the middle 40s and especially in the 50s, the age classes 45–54 and 55+ were not designated. Of the nine skeletons classed as 45+ years, roughly half should be aged in the late 40s and the remainder in the 50s or older.58 If only cranial suture closure or relative size were observable in a skeleton, age could be accurately identified only by the broader classes of young adult (15–34), mature adult (35+), and simply adult (15+). Although this scheme of age classes does not yield highly precise estimates like those often found in osteological reports, it does give an accurate picture of age distribution within the sample while mitigating the considerable error of exact estimates derived from common techniques.59 Furthermore, the broad classes of adulthood give an impression that might have been closer to the actual perception of age in Late Antique and Byzantine Greek society. Men and women would have marked the passing years of their lives but perhaps not with the same rigor or regularity as people today. In their daily existence, the residents of the Isthmus might have been content to categorize each other as young adult, middle-aged, or elderly according to general appearance and certain experiences or graded social and economic roles, such as marriage, parenthood, property management, and subsistence activities.60 55. The dimensions of the cists NEG 69-008 and DEC 69902 indicate that they were made for subadult bodies, but neither grave contained substantial skeletal remains when excavated. 56. See p. 39. 57. Ubelaker 1999, pp. 70–71, table 14, based on a large sample of protohistoric Arikara from South Dakota; Scheuer and Black (2000) include several other standards. 58. NEG 69-103A, 69-004B, 67-001C, T2 68-003A (late 40s),

NEG 69-007A (late 40s–middle 50s), T14 67-002A (early to mid-50s), NEG 69-001A, HO 70-902G (50s or older), 70-902E or F (over 55). 59. See Buikstra and Ubelaker 1994, p. 36; Waldron 1994, pp. 88, 90–91 n. 2; Ubelaker 1999, pp. 136–137. 60. On age-related constructions of identity, see pp. 209– 211. Talbot (1984, pp. 268–269) notes that the Byzantine conception of age was influenced by the ancient Greek division of the human life span into stages.

OSTEOLOGICAL METHODOLOGY

253

Although the main skeletal elements used for aging adults, namely, the pubic symphyses, auricular surfaces, and cranial vaults, were not consistently preserved (Fig. 5.4), enough survived to permit an estimation of age in nearly all adult skeletons. In the eight young adults, a combination of several markers for subadults and adults were consulted.61 When multiple indices were available but yielded different estimates, postcranial features were preferred to cranial suture closure. In a number of cases, the remains were so poorly preserved that two or more indices could not be considered.62 Apart from the three primary methods of age estimation in adult skeletons, other features were sometimes observed to assign a broad age class. One individual represented only by the lower legs (T14 69-701A) is identified as an adult from epiphyseal fusion in the tibia, fibula, tarsals, and metatarsals, and from the relative size of bones, which are similar to those in other full-grown adults in the sample.63 In another grave with poorly preserved bones (DEC 69-901), the relative size and dental development of both skeletons indicate that the deceased individuals were adults. One (A) can be aged as an adult by cranial sutures and as a mature adult (probably over 45 years) by relatively heavy dental wear. Reference to rates of occlusal attrition (Tables 6.4, 6.5) helped to narrow the age estimate for other individuals who were otherwise identified only as adults or mature adults. The relatively heavy attrition in mature adult NEG 69-001D indicates an age at death over 45 years. The degree of wear in adults HO 70-901C, 70-902D, 70-902E, and 70-902F indicates an age of death over 25 years. Finally, advancing age in most adult skeletons from the Isthmus is reflected in the relative severity of degenerative alterations to hard tissues, particularly osteoarthritis and the ossification of costal cartilage.64

Measurement of Bones The measurement of bones has long provided a basis not only for describing the physical appearance of deceased individuals but also for comparing skeletal samples. Morphological variation over time and space within and between groups arises from both genetic and environmental factors, though anthropologists generally agree that craniofacial shape is determined primarily by genes. The examination of morphology within a sample can also contribute to understanding sex and age, structural function, occupational stress, nutrition, and stature. Since immature bones are in a state of development, their measurement can provide important evidence for the process of growth.65 Subadult cranial and mandibular bones from the Isthmus were not sufficiently preserved for systematic measurement, but subadult postcranial bones were measured in 22 dimensions according to standard guidelines. A comprehensive approach was adopted for measuring adult cranial, mandibular, and postcranial remains. This study selected 89 standard measurements that anthropologists commonly use to identify genetic, sex, and age differences in adult skeletons. Full descriptions of all measurements are provided in the Appendix. Measurements of both subadult and adult skeletons were taken using osteometric instruments on the left antimere of bilateral elements, if it was available, or on the right antimere, if the left one was unavailable. Measurements were estimated when the bone was partly destroyed or reconstructed. Bones that were fragmentary or distorted by pathological or taphonomic alterations were not measured. 61. NEG 69-004D, 67-001D, 69-005B, 69-001C, 69-007B, T14 69-003B, HO 70-902J, TC 60-001A. 62. NEG 69-004C, 67-001A, 67-003A, 69-005B, 69-001D, T2 68-006A, LOU 69-801A, DEC 69-901A, HO 70-901C, WF 62001A, TC 60-001A. The aging of the seven adults in HO 70-902 (D–J) was complicated by the fact that the seven crania, six pubic symphyses, and five auricular surfaces cannot in every case be associated with each other. 63. The complete fusion and relative size of the isolated

occipital found in a secondary context near the Northeast Gate (69-999A) indicate that it came from an adult skeleton. Although the sparse, friable bones found in PAL 56-001 could not be recovered intact, the sizes of the cist and the bones as photographed (Figs. 2.80, 2.81) indicate that the deceased was an adolescent or adult. 64. See pp. 415–429 and Tables 7.3–7.7 65. E.g., Johnston and Zimmer 1989, pp. 18–20; Saunders 2000, pp. 146–147.

254

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

The postcranial measurements of subadults buried during both the Late Roman and Early Byzantine periods are presented in Table 5.4. Since the dimensions of skeletal elements in subadults are closely linked to their stage of growth, all metric data are presented and the individuals are arranged in order of estimated age. Many subadult skeletons could be extensively measured because the long bones of the arms and legs were the best-preserved elements (Fig. 5.5). The cranial, mandibular, and postcranial measurements of adults are summarized in Tables 5.5–5.10. These metric data are divided into three groups: the 58 individuals buried in graves during the Late Roman to Early Byzantine periods, the 46 individuals buried in graves during the Late Roman period, and the seven individuals buried before or after the Late Roman and Early Byzantine periods. In the first two groups, metric data from male and female skeletons have been separated and then averaged in order to show sexual dimorphism.66 The metric data for the remaining seven skeletons of disparate dates are given by individual. Although the preservation of the adult skeletons varies considerably (Fig. 5.4), numerous measurements could be collected, particularly from the skulls.

Observation of Nonmetric Traits Skeletal morphology is defined not only by its measured dimensions but also by the presence, absence, or quality of certain features variously called nonmetric, discrete, discontinuous, or epigenetic traits. These features can be classified as excessive ossification (bridge, spur), failed ossification (dehiscence, aperture), foramina, grooves and canals for blood vessels and nerves, supernumerary cranial sutures creating ossicles, craniobasal variations, spinal variations, prominent bony processes, and facet variants.67 Although the genetic basis of many nonmetric traits is uncertain, studies have demonstrated that variation is often linked to familial inheritance. Anthropologists often use nonmetric traits to identify group affinities, but in many cases it is difficult to know whether differences are due to genetic or environmental effects.68 In this study 30 cranial and 13 postcranial nonmetric traits were observed that are commonly recorded in osteological analysis. The observations of nonmetric traits in subadult and adult skeletons are presented in Tables 5.11–5.14. Full descriptions are provided in the Appendix.69 Most traits were scored dichotomously, often as present or absent, but certain ones were scored by appearance according to three grades. Bilateral traits were scored by anatomical side rather than by individual.70 While the collection of metric data required that the bone be mostly intact for measuring, the collection of nonmetric data was possible for many skeletons because traits were often observable even in fragmentary remains. Age and sex have been shown to affect the incidence of certain traits, particularly hyper-ostotic and hypostotic features.71 Therefore, the data from subadult and adult skeletons are segregated. Since only one of the subadults came from an Early Byzantine grave (T14 69-991A), the nonmetric data from this individual are included in a separate column alongside the combined nonmetric data from subadults of Late Roman date in Table 5.11. As in the presentation of metric data, the nonmetric data from adult skeletons were separated into 66. The entire skeleton NEG 69-007B (M?), 16–18 years, was measured according to the metric standards for adults except for the unfused diaphyses, which were not measured. These calculations do not include the measurements from adult skeletons with surviving remains that could not be sexed (NEG 69-999A, T14 69-701A, HO 70-902E, Z1–Z7). 67. Saunders 1989, p. 95, table 2. 68. For major discussions of the heritability of nonmetric traits, see Saunders 1978; Sjøvold 1984, pp. 236–243; Hauser and De Stefano 1989, pp. 1–15; Saunders 1989.

69. Other developmental traits that anthropologists sometimes class as nonmetric will be addressed later, namely, the congenital absence of the third molar (p. 299), cranial or caudal shift in the vertebral sequence, particularly at the lumbosacral border, and spina bifida occulta (pp. 406–408). 70. Saunders (1989, pp. 98–99) discusses the treatment of bilateral traits. 71. Hauser and De Stefano 1989, p. 9; Saunders 1989, pp. 99–101.

–

–

–

61.0*

6.5

–

–

FE LEN

FE D WID

FE DIA

TIB LEN

TIB DIA

FIB LEN

FIB DIA

5.5

–

10.0

104.0

10.5

33.0

130.0

6.5

75.0

–

22.0

–

NEG 67-003D 6.0

88.5

9.0

–

8.5

–

–

5.5

–

6.0*

–

8.0

22.0

83.0

–

–

–

–

–

–

–

–

0–1 yr.

NEG 67-001J 6.0

124.5

–

–

–

–

–

–

–

5.5

77.0

11.5*

28.0

125.5

31.0

18.0

–

–

–

–

5.0

58.0

1–2 yrs.

T14 69-991A 7.0

–

11.0

–

13.5

–

–

7.0

84.0

7.0*

–

11.0

20.0*

–

–

–

–

–

–

–

7.0

62.0*

1–2 yrs.

NEG 67-001K –

–

–

–

9.5

27.0*

–

7.0

86.5

8.0

99.5

–

–

–

–

–

–

51.5*

–

–

–

–

2–4 yrs.

HO 70-902B 6.5

–

13.5

138.0*

12.0

42.0

168.0

7.0

–

–

–

12.0

27.0

127.5

–

–

61.0

–

47.5

58.0

7.0

–

2–4 yrs.

NEG 67-001H 6.0

138.0

–

–

11.5

41.5

178.0

8.0

95.5

6.5

107.5

11.0

29.0

128.0

43.0

27.5

61.0

54.0

48.0

58.0*

8.5

70.0*

2–4 yrs.

NEG 67-001I 7.0

136.0*

–

–

12.0

41.5

160.5

7.5

98.5

7.0

109.0

11.0

26.5

133.0

44.0

–

58.5*

53.0*

47.0*

65.0

7.0

69.5*

2–4 yrs.

7.0

–

13.0*

125.0

13.5*

–

158.0*

8.0

93.0

8.0

103.0

12.5

30.0

130.0

–

29.5

64.0

–

50.0*

63.0

7.0

68.0

3–4 yrs.

NEG 69-004A

Notes: An asterisk (*) indicates that a measurement is an estimate. For trait abbreviations, see the Appendix.

–

–

UL LEN

RA DIA

8.0

6.5

HU DIA

–

–

–

84.0

75.0*

HU LEN

HU D WID

UL DIA

–

–

PU LEN

RA LEN

–

–

IS LEN

52.0*

–

42.0

–

–

SC WID

–

–

–

SC LEN

6.0

61.5

SC SPINE L

–

0–1 yr.

HO 70-902A

IL LEN

–

0–6 mos.

CL DIA

Age

NEG 67-003E

CL LEN

Trait

NEG 69-001B 6.5

141.0*

10.5

144.5

12.0

42.0

182.0

7.0

95.5

7.0

106.0

11.0

25.5

127.5

28.0*

–

–

–

–

–

6.0

72.0

3–4 yrs.

NEG 67-003C 6.0

–

12.0

–

11.0

–

–

7.0

–

6.5

–

9.5

–

–

–

–

–

–

–

–

5.0

65.5*

3–5 yrs.

HO 70-902C 7.0

–

13.5

145.0

16.0*

44.0*

–

8.0

102.0*

8.0

114.0

11.0

29.0

137.0

36.5

31.0

68.0

–

49.0

–

7.0

71.5*

5–6 yrs.

NEG 67-001G 8.0

172.0

–

–

14.0

–

–

8.0

120.0

9.0

132.0

13.5

33.0

158.0

35.0

29.0

66.0

–

–

–

7.0

81.0

4–8 yrs.

NEG 67-001F 8.5

175.5

–

–

16.5

48.5

–

9.0

119.5

9.5

134.0

11.0

35.0

158.5

43.0

32.0

64.0

70.0

58.5

84.0

8.5

81.0

4–8 yrs.

NEG 67-001E 8.0

205.0

–

–

16.0

–

–

9.0

143.0

8.5

132.0

13.0

36.0

192.0

50.0

34.0*

–

–

67.0*

91.5

7.0

88.0

7–11 yrs.

11.5

261.0

26.0

–

–

–

–

12.0*

–

13.0

–

16.5

44.0

220.0*

–

75.0*

–

–

–

–

9.0

–

10–12 yrs.

NEG 69-005A

TABLE 5.4. POSTCRANIAL MEASUREMENTS (MM) IN SUBADULT SKELETONS OF LATE ROMAN TO EARLY BYZANTINE DATE

11.5

–

22.0

–

22.0

57.0

352.0*

12.0

–

11.5*

–

17.0

47.0

–

61.0

53.0

118.0

–

72.0*

–

12.5

111.0

16–18 yrs.

NEG 69-007B

256

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.5. CRANIAL AND MANDIBULAR MEASUREMENTS (MM) IN ADULT SKELETONS OF LATE ROMAN TO EARLY BYZANTINE DATE Males

Females

Trait

n

Range

Mean

sd

n

Range

Mean

sd

MAX LEN

4

172.0–196.0

180.6

10.8

9

166.0–187.0

176.2

6.4

MAX BR

6

134.5–148.0

141.4

6.0

5

119.0–143.0

133.6

9.6

BAS-BREG

4

122.5–143.5

134.6

9.1

3

120.5–141.0

128.2

11.2

POR-BREG

6

119.0–133.0

128.0

5.4

10

112.0–128.0

118.3

4.8

BAS-NAS

5

88.0–109.0

101.6

8.1

3

93.0–96.0

94.7

1.5

BAS-PROS

5

86.0–101.0

94.1

7.6

3

90.0–99.0

93.8

4.6

MAX-ALV BR

5

58.0–67.0

62.8

4.2

2

59.5–65.0

62.3

3.9

MAX-ALV L

5

47.0–59.5

53.0

4.6

3

51.5–52.0

51.8

0.3

BIAUR BR

5

101.5–132.0

120.2

12.9

3

108.0–117.5

112.5

4.8

MIN FR BR

7

82.0–105.5

96.6

7.5

8

86.0–101.0

95.3

5.1

BIZYG BR

8

114.0–136.0

126.1

6.2

5

105.0–121.0

117.2

6.9

UFH

10

55.5–73.0

64.6

5.7

6

60.0–67.0

65.2

2.9

UFBR

8

87.5–113.5

102.3

8.7

5

91.0–112.0

99.6

8.0

ORBIT H

8

30.5–35.0

32.6

1.8

7

30.0–36.0

33.0

2.4

ORBIT BR

9

33.0–44.0

39.6

3.4

9

35.5–38.0

36.9

1.1

BIORB BR

8

81.0–102.5

92.5

7.0

5

85.0–90.0

87.9

2.3

INTERORB BR

8

16.5–22.5

19.8

2.1

7

15.0–23.5

19.4

3.1

NAS H

9

41.0–55.0

48.8

4.8

7

41.0–51.0

47.7

3.5

NAS BR

10

21.0–28.0

24.3

2.2

5

22.5–26.0

23.6

1.4

FRONT CHD

6

100.5–126.5

112.0

10.9

10

91.0–115.5

105.9

7.7

PAR CHD

8

100.5–139.5

115.0

12.6

7

105.0–119.0

112.4

4.8

OCC CHD

7

90.5–112.5

97.6

7.7

10

88.5–103.0

96.9

7.3

FOR MAG L

6

33.0–37.0

34.7

1.6

6

33.0–37.5

35.8

1.6

FOR MAG BR

8

26.0–33.0

29.9

2.5

6

26.0–30.0

28.2

1.6

MAST H

14

21.5–43.0

32.3

5.4

17

23.0–40.0

31.7

5.0

MAND L

6

101.0–107.0

104.6

2.5

5

86.5–121.0

99.9

12.8

BICON BR

4

119.0–129.0

122.5

5.6

3

102.0–123.0

114.7

11.2

MIN RAM BR

13

29.0–35.0

32.0

1.9

14

27.0–34.5

31.2

2.1

MAX RAM H

11

46.0–69.0

60.0

6.8

11

43.0–62.0

55.1

5.9

BIGON BR

6

98.0–111.0

105.2

4.9

5

85.0–91.0

89.1

2.4

12

49.0–69.5

62.6

5.6

12

46.0–64.0

57.5

4.9

SYMPH H

8

20.0–34.5

29.0

4.7

7

26.0–32.0

28.1

2.0

BIMEN BR

10

42.0–46.5

45.1

1.4

10

39.0–48.0

43.7

2.4

BODY H

10

21.5–33.5

28.3

3.9

10

20.5–31.0

25.8

3.3

COR H

Note: For trait abbreviations, see the Appendix.

three groups comprising individuals from Late Roman to Early Byzantine graves, individuals from Late Roman graves, and the remaining individuals from graves of disparate dates (Tables 5.12–5.14). The first two groups combine nonmetric traits from male and female adult skeletons, because segregating them would generate subsets that were too small for meaningful comparison. Nonetheless, the differential prevalence of traits in males and females was observed for noteworthy patterns. Moreover, since all bilateral traits did not con-

OSTEOLOGICAL METHODOLOGY

257

TABLE 5.6. CRANIAL AND MANDIBULAR MEASUREMENTS (MM) IN ADULT SKELETONS OF LATE ROMAN DATE Males Trait

n

Range

MAX LEN

2

172.0–182.0

MAX BR

2

BAS-BREG

1

POR-BREG

Females Mean

sd

n

Range

Mean

sd

176.0

5.7

7

166.0–183.0

175.2

5.5

134.5–145.0

139.8

7.4

3

130.0–141.0

135.3

5.5

–

139.0

–

2

120.5–123.0

121.8

1.8

2

119.0–132.0

125.5

9.2

9

112.0–123.0

117.2

3.6

BAS-NAS

2

–

105.0

–

1

–

95.0

–

BAS-PROS

2

96.5–101.0

98.8

3.2

1

–

92.5

–

MAX-ALV BR

2

62.5–67.0

64.8

3.2

2

59.5–65.0

62.3

3.9

MAX-ALV L

2

53.5–54.0

53.8

0.4

2

–

52.0

–

BIAUR BR

2

113.5–131.5

122.5

12.7

1

–

117.5

–

MIN FR BR

4

82.0–105.0

96.5

10.4

6

91.0–101.0

95.8

3.6

BIZYG BR

4

114.0–136.0

124.9

9.0

3

119.0–121.0

120.0

1.0

UFH

6

57.0–73.0

65.3

5.2

4

60.0–68.0

65.0

3.6

UFBR

5

93.5–113.5

104.6

8.1

3

91.0–102.0

96.7

5.5

ORBIT H

5

31.5–35.0

32.9

1.5

5

30.0–36.0

32.8

2.4

ORBIT BR

6

37.0–44.0

40.7

2.9

7

35.5–39.0

37.0

1.2

BIORB BR

5

85.0–102.5

94.2

6.7

3

86.0–90.0

88.2

2.0

INTERORB BR

5

16.5–22.5

20.2

2.3

5

15.0–23.5

19.2

3.0

NAS H

5

46.0–55.0

51.6

3.5

5

41.0–51.0

48.0

4.2

NAS BR

5

22.5–27.0

24.7

1.7

3

22.5–23.5

23.0

0.5

FRONT CHD

3

106.0–125.0

113.0

10.4

9

91.0–115.5

105.1

7.7

PAR CHD

4

100.5–112.0

107.1

5.1

6

105.0–117.0

111.3

4.2

OCC CHD

3

90.5–100.5

94.8

5.1

8

88.5–114.5

96.6

7.8

FOR MAG L

2

35.0–36.0

35.5

0.7

4

33.0–37.5

35.6

1.9

FOR MAG BR

3

28.0–32.0

30.3

2.1

4

26.0–29.5

27.9

1.7

MAST H

8

26.5–35.5

32.4

3.2

15

23.0–40.0

31.9

4.8

MAND L

2

103.0–106.0

104.5

2.1

4

86.5–121.0

101.1

14.4

BICON BR

1

–

119.5

–

2

102.0–123.0

112.5

14.8

MIN RAM BR

7

30.5–35.0

32.5

1.5

12

27.0–34.0

31.0

2.0

MAX RAM H

5

52.0–67.5

60.5

5.7

9

43.0–62.0

56.2

5.7

BIGON BR

1

–

105.0

–

3

85.0–90.0

88.2

2.8

COR H

6

57.5–66.0

62.8

3.4

10

52.0–64.0

58.8

3.5

SYMPH H

3

20.0–34.5

28.0

7.4

5

26.0–29.0

27.6

1.3

BIMEN BR

4

43.5–46.0

45.1

1.2

8

39.0–48.0

43.9

2.7

BODY H

5

24.0–33.5

28.9

3.7

9

20.5–31.0

25.6

3.4

Note: For trait abbreviations, see the Appendix.

sistently occur on both sides of the body througout the sample, the asymmetrical expression of those traits was observed. Finally, the frequency of traits among skeletons interred in multiple graves or within the same burial area was also checked. Any significant patterns in the spatial distribution of nonmetric traits may reflect the burial location of familial groups. The variation in the incidence of nonmetric traits by age, sex, anatomical side, and burial locale will be discussed below.

258

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.7. CRANIAL AND MANDIBULAR MEASUREMENTS (MM) IN SEVEN ADULT SKELETONS OF DISPARATE DATES T2 68-003A

T2 68-003B

TC 60-001A

WF 62-001A

M

F

M

–

–

F

F

187.0

–

181.5

–

–

178.5

171.0*

MAX BR

–

–

BAS-BREG

–

–

–

–

–

151.5

–

–

–

–

141.5

127.0

POR-BREG

132.0

–

129.0

–

–

136.0

122.0

BAS-NAS

–

–

–

–

–

104.5

–

BAS-PROS

–

–

–

–

–

99.5

–

MAX-ALV BR

–

–

–

–

–

62.0*

–

MAX-ALV L

–

–

–

–

–

49.5

–

BIAUR BR

–

–

–

–

–

132.5

–

MIN FR BR

Trait MAX LEN

LOU 69-801A DEC 69-901A DEC 69-901B

100.0

–

–

–

–

111.5

–

BIZYG BR

–

–

–

–

–

136.5

–

UFH

–

–

–

–

–

69.5

UFBR

–

–

–

–

–

108.0

ORBIT H

–

–

–

–

–

32.0

33.0*

ORBIT BR

–

–

–

–

–

40.5

39.0*

BIORB BR

–

–

–

–

–

98.5

–

INTERORB BR

–

–

–

–

–

24.0

–

NAS H

–

–

–

–

–

38.0*

NAS BR

–

–

–

–

–

27.0

FRONT CHD

111.0

–

107.0

–

–

107.5

–

PAR CHD

118.0

–

114.0

–

–

105.5

117.5

OCC CHD

–

–

87.0

–

–

111.0

91.5

FOR MAG L

–

–

–

–

–

29.5

34.5

FOR MAG BR

–

–

–

–

–

29.5

29.0

–

41.0

–

–

40.5

28.5*

MAST H

39.0

57.5* –

– 20.5

MAND L

–

–

106.0

–

–

–

93.0*

BICON BR

–

–

116.0

–

–

–

96.5* 30.0

MIN RAM

32.0

–

38.0

35.0

–

–

MAX RAM H

59.5

–

66.0

–

–

–

BIGON BR

–

106.0*

–

108.5

–

–

–

COR H

–

–

71.0

–

–

–

61.5*

SYMPH H

–

–

28.0*

–

–

27.5

BIMEN BR

–

43.0

39.0*

–

–

40.5*

29.5

28.0*

23.0*

–

25.0

BODY H

26.0

25.5* – 28.0

–

Notes: An asterisk (*) indicates that a measurement is an estimate. For trait abbreviations, see the Appendix.

THE COMPOSITION OF THE SKELETAL SAMPLE The composition of the skeletal sample from the Isthmus can be reconstructed from the analysis of the osteological data. The human remains provide evidence for demography, physical appearance, body size, and structural variation among local residents, particularly

THE COMPOSITION OF THE SKELETAL SAMPLE

259

during the Late Roman and Early Byzantine periods. On the basis of this reconstruction, it will be possible to compare the residents of the Isthmus with both contemporary Corinthians and Greeks of other periods and regions, even though the remains available for comparison are sparse.

Paleodemography The skeletal sample from the Isthmus is too small to provide an adequate basis for a detailed reconstruction of paleodemography, such as in the form of a life table.72 It is, however, noteworthy that the skeletons from graves of Late Roman to Early Byzantine date represent all ages and both sexes. The graves in areas so far excavated at the Isthmus were used by all members of society, not exclusively by men or women, adults or children. The distribution of age and sex is not atypical for a pre-modern group, with a usual frequency of subadults (29.23%) and a roughly equal number of males (20) and females (22). This is an important conclusion, because it suggests that the demographic distribution of the skeletal sample, despite its small size, is basically what would be expected in a living population.73 Although the exact demographic composition of all residents cannot be recovered, it is possible to trace the distribution of age within the Late Roman to Early Byzantine skeletal sample (Figs. 5.6, 5.7). As has been discussed, the skeletons were assigned to age classes of five years for subadults (0–4, 5–9, 10–14) and ten years or more for adults (15–24, 25–34, 35–44, 45+) so that the size of composite age groups could be compared (Table 5.3).74 The resulting patterns should be considered suggestive rather than definitive because of the data set’s limitations. A large proportion of all individuals (14/58,75 or 24.14%) died in the first four years, the period of life when many ancient people died due to primitive health care, poor hygiene and nutrition, and susceptibility to disease. Infection and malnutrition probably disrupted the dental enamel formation of several individuals in their second to fourth years. In addition, several skeletons display pitting in the superior orbital cavity, a condition called cribra orbitalia that is probably symptomatic of iron deficiency anemia in childhood.76 Many fewer individuals in the sample died in late childhood and adolescence (5/58, or 8.62%), when the human body undergoes substantial growth, increased structural durability, and improving physiological resilience. Then few died in early adulthood (11/58, or 18.97%), including several women aged 15–24, which probably reflects fatality from childbirth. Most adult males and females died at the age of 35–44 years (14/58, or 24.14%), after which point few individuals are represented. The causes of death for adults are unknown except for one individual (HO 70-901C), who died from a blow to the head with a pointed weapon (Fig. 7.31).77 In sum, the number of individuals in the skeletal sample peaks in infancy and in the late 30s–40s, with few individuals beyond the fifth decade present. There are slightly more females than males in young adulthood, but overall the age distribution of males and females is roughly equivalent. Among all preserved skeletons, the prevalence of subadults is 32.76%, and the prevalence of adults is 67.24%. The subadult frequency 72. Ubelaker (1999, pp. 135–141) presents a useful introductory discussion of paleodemography. 73. See Waldron 1994, pp. 23–24. 74. The age estimates for four subadults do not perfectly fit an age class (NEG 67-001E–G, 67-003C). These subadults were assigned to the most probable age class. 75. This total of 58 individuals includes seven not assignable to specific age classes. Three individuals, including two

males and one female (NEG 69-004C, 69-001D, T2 68-003B), were classed as “mature adult,” and four individuals, including two males (T14 69-701A, HO 70-901C, 70-902D or I, E or F), were classed as “adult.” 76. See pp. 306–312 on enamel hypoplasia and pp. 409– 413 on cribra orbitalia. 77. For a full discussion, see pp. 396–399.

260

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.8. POSTCRANIAL MEASUREMENTS (MM) IN ADULT SKELETONS OF LATE ROMAN TO EARLY BYZANTINE DATE Males Trait

Females

n

Range

Mean

sd

n

Range

Mean

sd

CL LEN

3

127.0–135.0

132.0

4.4

4

127.0–147.5

136.1

8.8

CL MID DIA AP

7

10.0–14.5

11.9

1.8

12

9.0–13.0

10.3

1.1

CL MID DIA IS

7

8.0–13.5

10.0

1.8

10

8.0–11.5

9.6

1.4

CL MID CIR

7

32.0–43.0

38.1

4.6

12

28.0–42.9

35.3

4.1

SC H

1

–

136.0

–

0

–

–

SC BR

2

110.0–122.0

116.0

8.5

1

–

103.0

–

SC GLEN H

6

29.0–39.0

34.8

3.5

8

25.0–36.5

33.4

3.8

SC GLEN BR

5

23.0–40.0

28.1

7.0

8

20.5–42.0

25.4

6.9

HU LEN

1

–

294.0

–

1

–

286.0

–

HU EPIC BR

4

32.5–62.0

53.4

14.1

7

49.0–59.0

53.6

3.5

HU HD DIA

2

45.0–47.0

46.0

1.4

4

38.0–41.0

39.4

1.3

HU MID ML

7

17.0–22.0

19.9

1.9

10

16.0–20.0

17.9

1.5

HU MID AP

7

17.0–24.0

21.2

2.2

11

15.5–20.5

18.4

1.4

HU LST CIR

6

55.0–73.0

63.3

6.6

13

52.0–67.0

57.7

4.9

RA LEN

5

215.5–237.0

228.1

7.8

7

201.0–230.0

216.6

12.1

RA HD DIA

3

19.0–21.0

20.0

1.0

9

17.0–22.5

19.8

1.7

RA MID DIA ML

8

12.5–16.5

14.9

1.1

8

11.0–14.5

12.4

1.1

RA MID DIA AP

8

9.5–12.5

10.9

1.0

8

8.5–10.5

9.4

0.8

UL LEN

2

–

241.0

–

4

229.0–254.0

243.5

10.5

UL PHYS LN

3

236.0–249.0

241.0

7.0

4

218.0–240.0

226.0

9.6

UL DIA ML

7

14.0–19.0

16.4

2.1

12

12.0–18.0

14.2

1.6

UL DIA AP

8

14.5–21.0

17.3

2.3

12

12.5–18.5

15.3

2.3

UL LST CIR

7

31.0–51.0

39.0

6.1

8

31.0–45.0

35.6

4.4

SAC ANT L

1

–

99.0

–

1

–

102.0

–

SAC ANT SUP BR

1

–

105.0

–

0

–

–

–

SAC M TR DIA B

3

46.0–52.0

48.7

3.1

1

–

INN H

0

–

–

–

0

–

–

–

INN IL LEN

0

–

–

–

0

–

–

–

INN PUB LEN

2

65.0–84.5

74.8

13.8

4

76.5–94.5

INN ISCH LEN

3

68.5–83.5

74.8

7.8

2

FE LEN

0

–

–

–

1

FE BIC LEN

0

–

–

–

FE BIC BR

0

–

– –

48.0

–

–

85.9

7.5

69.0–78.0

73.5

6.4

–

409.0

–

1

–

409.0

–

–

1

–

74.0

–

FE EPIC BR

0

–

–

1

–

70.0

–

FE HD DIA

3

43.5–47.0

45.2

1.8

7

38.0–42.5

40.6

1.5

FE SUB DIA ML

3

29.0–32.5

30.8

1.8

8

28.0–32.0

30.1

1.5

FE SUB DIA AP

3

24.5–31.0

27.3

3.3

8

21.0–27.5

24.5

2.3

FE MID ML DIA

6

25.5–30.0

28.4

1.6

9

25.0–29.0

26.8

1.5

FE MID AP DIA

6

25.0–34.0

29.7

3.4

9

23.0–29.0

25.8

2.0

FE MID CIR

6

81.0–106.0

93.9

8.7

8

81.0–94.0

85.3

4.9

FE LC DIA AP

0

–

–

–

1

–

54.0

–

FE MC DIA AP

0

–

–

–

1

–

57.0

–

THE COMPOSITION OF THE SKELETAL SAMPLE

261

TABLE 5.8 (CONT.) Males Trait

Females

n

Range

sd

n

Range

PAT LEN

4

31.5–41.0

Mean 36.6

4.0

6

33.0–40.0

Mean 36.3

2.6

PAT BR

5

37.5–48.0

42.9

4.4

6

36.0–43.0

39.9

2.6

TIB LEN

0

–

–

–

2

322.0–345.0

333.5

16.3

TIB M PROX BR

0

–

–

–

3

59.0–70.0

63.2

6.0

TIB M DIST BR

1

–

52.6

–

5

44.5–51.0

46.8

2.5

TIB NF DIA ML

2

22.5–26.0

24.3

2.5

6

21.0–24.0

21.8

1.2

TIB NF DIA AP

1

–

35.0

–

7

28.0–34.0

30.9

2.0

TIB NF CIR

2

–

92.0

–

3

83.0–94.0

88.3

5.5

FIB MAX LEN

4

339.0–397.0

362.5

25.6

8

304.0–355.0

321.6

15.5

FIB MID MAX DIA

3

10.5–17.0

14.2

3.3

11

8.5–15.0

13.0

1.9

TAL LEN

7

50.0–59.0

53.0

3.0

7

46.5–54.0

49.4

2.5

CAL LEN

4

69.0–79.0

74.8

4.2

3

69.0–75.0

71.7

3.1

CAL BR

1

–

38.0

–

0

–

–

sd

–

Note: For trait abbreviations, see the Appendix.

is slightly higher if the two small-sized cists found without substantial preserved bones (NEG 69-008, DEC 69-902) had once held subadults. The distribution of age at death in the Isthmian sample compares with that in many skeletal samples of Late Roman to Early Byzantine date from other sites. The published remains from Corinth offer evidence for comparison to the demographic profile at the Isthmus. Late Antique graves excavated at the Sanctuary of Demeter and Kore on Acrocorinth mostly contained infants and adult females. Of the few individuals whose age could be estimated, several died in young adulthood, though it is unclear whether the pattern is significant.78 A better picture of the paleodemography of Late Antique Corinthians is provided by the human remains from the Lerna Hollow cemetery. The frequencies of subadults (27%) and adults (73%) are similar to those at the Isthmus, and likewise few individuals in their fifth decade are represented. However, preservation of bone in this area was very poor. It is therefore quite possible that many remains of infants and young children had disintegrated before excavation, in which case the actual prevalence of subadults was higher.79 A sample of 44 individuals from burials at Corinth of mostly Roman date revealed a comparable frequency for individuals up to 12 years (30.9%), with few individuals in the teens and most adults in the third to fifth decades.80 Peter Burns conducted a demographic survey of 1,485 skeletons excavated at Corinth, many from the Roman chamber tombs on the north edge of the city and the Byzantine basilica on Temple Hill. He found that about 40% of skeletons of known age were younger than 18 years and that the average age at death for adult skeletons was roughly 35 years.81 The large skeletal sample recovered from the Frankish complex southeast of Temple E at Corinth points to similar frequencies of age groups, but with a higher prevalence of infants.82 Among other published remains from Late Antique or Byzantine graves across Greece, most adults are aged in their third to fifth decades, with few in their 78. Corinth XVIII.3, pp. 385–386, 387 (nos. 12, 15, 19, 20, 22, 23, 26). 79. Wesolowsky 1973, pp. 342–345, 351, table II. 80. Fox Leonard 1997, pp. 398–400, 453, table 8; Fox 2005,

pp. 65, 66, table 3.3. 81. Burns 1982, pp. 18–21, table 1. 82. Barnes 2003, p. 441.

262

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.9. POSTCRANIAL MEASUREMENTS (MM) IN ADULT SKELETONS OF LATE ROMAN DATE Males Trait

Females

n

Range

Mean

sd

n

Range

Mean

sd

CL LEN

3

127.0–135.0

132.0

4.4

4

127.0–147.5

136.1

8.8

CL MID DIA AP

7

10.0–14.5

CL MID DIA IS

7

8.0–13.5

11.9

1.8

12

9.0–13.0

10.3

1.1

10.0

1.8

10

8.0–11.5

9.6

1.4

CL MID CIR

7

32.0–43.0

38.1

4.6

12

28.0–42.9

35.3

4.1

SC H

1

–

136.0

–

0

–

–

SC BR

2

110.0–122.0

116.0

8.5

1

–

103.0

SC GLEN H

6

29.0–39.0

34.8

3.5

8

25.0–36.5

33.4

3.8

SC GLEN BR

5

23.0–40.0

28.1

7.0

8

20.5–42.0

25.4

6.9

HU LEN

1

–

294.0

–

1

–

286.0

HU EPIC BR

4

32.5–62.0

53.4

14.1

7

49.0–59.0

53.6

3.5

HU HD DIA

2

45.0–47.0

46.0

1.4

4

38.0–41.0

39.4

1.3

HU MID ML

7

17.0–22.0

19.9

1.9

10

16.0–20.0

17.9

1.5

HU MID AP

7

17.0–24.0

21.2

2.2

11

15.5–20.5

18.4

1.4

HU LST CIR

6

55.0–73.0

63.3

6.6

13

52.0–67.0

57.7

4.9

RA LEN

5

215.5–237.0

228.1

7.8

7

201.0–230.0

216.6

12.1

RA HD DIA

3

19.0–21.0

20.0

1.0

9

17.0–22.5

19.8

1.7

RA MID DIA ML

8

12.5–16.5

14.9

1.1

8

11.0–14.5

12.4

1.1

RA MID DIA AP

8

9.5–12.5

10.9

1.0

8

8.5–10.5

9.4

0.8

UL LEN

2

–

241.0

–

4

229.0–254.0

243.5

10.5

UL PHYS LN

3

236.0–249.0

241.0

7.0

4

218.0–240.0

226.0

9.6

UL DIA ML

7

14.0–19.0

16.4

2.1

12

12.0–18.0

14.2

1.6

UL DIA AP

8

14.5–21.0

17.3

2.3

12

12.5–18.5

15.3

2.3

UL LST CIR

7

31.0–51.0

39.0

6.1

8

31.0–45.0

35.6

4.4

SAC ANT L

1

–

99.0

–

1

–

102.0

–

SAC ANT SUP BR

1

–

105.0

–

0

–

–

–

SAC M TR DIA B

3

46.0–52.0

48.7

3.1

1

–

48.0

–

INN H

0

–

–

–

0

–

–

–

INN IL LEN

0

–

–

–

0

–

–

–

INN PUB LEN

2

65.0–84.5

74.8

13.8

4

76.5–94.5

85.9

7.5

INN ISCH LEN

3

68.5–83.5

74.8

7.8

2

69.0–78.0

73.5

6.4

FE LEN

0

–

–

–

1

–

409.0

–

FE BIC LEN

0

–

–

–

1

–

409.0

–

FE BIC BR

0

–

–

–

1

–

74.0

–

–

– –

–

FE EPIC BR

0

–

–

1

–

70.0

–

FE HD DIA

3

43.5–47.0

45.2

1.8

7

38.0–42.5

40.6

1.5

FE SUB DIA ML

3

29.0–32.5

30.8

1.8

8

28.0–32.0

30.1

1.5

FE SUB DIA AP

3

24.5–31.0

27.3

3.3

8

21.0–27.5

24.5

2.3

FE MID ML DIA

6

25.5–30.0

28.4

1.6

9

25.0–29.0

26.8

1.5

FE MID AP DIA

6

25.0–34.0

29.7

3.4

9

23.0–29.0

25.8

2.0

FE MID CIR

6

81.0–106.0

93.9

8.7

8

81.0–94.0

85.3

4.9

FE LC DIA AP

0

–

–

–

1

–

54.0

–

FE MC DIA AP

0

–

–

–

1

–

57.0

–

THE COMPOSITION OF THE SKELETAL SAMPLE

263

TABLE 5.9 (CONT.) Males Trait

Females

n

Range

sd

n

Range

PAT LEN

4

31.5–41.0

Mean 36.6

4.0

6

33.0–40.0

Mean 36.3

2.6

PAT BR

5

37.5–48.0

42.9

4.4

6

36.0–43.0

39.9

2.6

TIB LEN

0

–

–

–

2

322.0–345.0

333.5

16.3

TIB M PROX BR

0

–

–

–

3

59.0–70.0

63.2

6.0

TIB M DIST BR

1

–

52.6

–

5

44.5–51.0

46.8

2.5

TIB NF DIA ML

2

22.5–26.0

24.3

2.5

6

21.0–24.0

21.8

1.2

TIB NF DIA AP

1

–

35.0

–

7

28.0–34.0

30.9

2.0

TIB NF CIR

2

–

92.0

–

3

83.0–94.0

88.3

5.5

FIB MAX LEN

4

339.0–397.0

362.5

25.6

8

304.0–355.0

321.6

15.5

FIB MID MAX DIA

3

10.5–17.0

14.2

3.3

11

8.5–15.0

13.0

1.9

TAL LEN

7

50.0–59.0

53.0

3.0

7

46.5–54.0

49.4

2.5

CAL LEN

4

69.0–79.0

74.8

4.2

3

69.0–75.0

71.7

3.1

CAL BR

1

–

38.0

–

0

–

–

sd

–

Note: For trait abbreviations, see the Appendix.

50s.83 The remains from the basilica at Aliki on Thasos include numerous subadults whose ages show a distribution similar to the subadults at the Isthmus, as most died in the first five years and few after that point before adulthood.84 Paleodemographic research in recent decades has shown that the distribution of ages in a skeletal sample reveals more about fertility and birth rate than about mortality and death rate. In the traditional study of human remains from archaeological sites, those ages best represented in an assemblage were interpreted as the periods of life when most people died. It is now known that a young sample, one with a high frequency of subadults and a relatively low average age at death, reflects high fertility, while an old sample reflects stagnant or declining birth rate, assuming in both cases that the living populations were closed and stable.85 To be sure, historical sources indicate that many infants and children died in their first five years. Moreover, although certain royals, scholars, and monks lived into and well beyond their 60s, such longevity was probably unusual.86 But in terms of large-scale patterns in Greek paleodemography, skeletal samples from the Isthmus and Corinth, as well as other sites, in which ca. 30%–40% of individuals are subadults and the average age of death falls in the fourth decade, broadly attest to population growth during Roman and Byzantine times. If the skeletal samples in the Burns survey and from the Frankish complex are younger than the Isthmian sample, it seems that more individuals were entering the urban population through higher fertility in the Middle Ages than were entering the rural population in earlier centuries. Since the distribution of sex and age at the Isthmus by area and by burial has been discussed elsewhere,87 a summary will suffice here. There is no evidence for the segregation 83. E.g., Musgrave 1976, pp. 41, 45, table 3 (Knossos, 7th or 8th century); Mallegni 1988, p. 349, table 1 (Gortyn, 6th–7th centuries); Bourbou 2004, pp. 48–49, tables 13, 14, figs. 3, 4 (Messene, late 4th to mid-6th centuries, but see p. 351, n. 152 on the date; Eleutherna, 5th–7th centuries). 84. Buchet and Sodini 1984, pp. 234–235, pl. 91. 85. E.g., Sattenspiel and Harpending 1983; Buikstra, Konigsberg, and Bullington 1986; Johansson and Horowitz 1986.

Buikstra (1997) and Larsen (1997, pp. 337–340) give summaries. 86. Patlagean 1977, pp. 96–100, tables 4–6 (Late Antique epitaphs from Egypt, Syria, Palestine); Laiou-Thomadakis 1977, p. 296 (Late Byzantine Macedonian praktika); Talbot 1984, pp. 268–270; Dennis 2001, pp. 2–3 (on infant mortality citing Theod. Stud. Ep. 18, 29 and V. s. Mar. Jun. 4, 6 [AASS Nov. 4.693E–F, 694E–F]). 87. See pp. 207–208, 210.

264

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.10. POSTCRANIAL MEASUREMENTS (MM) IN SIX ADULT SKELETONS OF DISPARATE DATES

Trait CL LEN

T2 68-003A

T268-003B

LOU 69-801A

DEC 69-901A

DEC 69-901B

WF 62-001A

M

F

M

–

–

F

R

L

R

L

R

L

R

L

R

L

R

L

153.5

145.0*

139.0

–

–

157.0

–

–

–

–

–

–

CL MID DIA AP

14.0

13.5

11.5

–

14.0

15.0

–

13.0

–

–

11.5

11.0

CL MID DIA IS

14.0

13.5

9.5

–

12.0

12.0

–

–

–

–

8.5

8.5

CL MID CIR

48.0

46.0

35.0

–

46.0

45.0

–

43.0

–

–

38.0

36.0

SC H

–

–

–

–

–

–

–

–

–

–

–

–

SC BR

–

–

–

–

–

–

–

–

–

–

–

–

SC GLEN H

38.0

–

–

–

–

–

–

–

–

–

–

–

SC GLEN BR

25.0*

–

–

–

–

29.0

–

–

–

–

–

–

HU LEN

–

–

–

–

–

–

–

–

–

–

–

–

HU EPIC BR

–

–

49.0

49.0

–

–

–

–

–

–

–

–

HU HD DIA

–

–

–

–

–

47.5

–

–

–

–

–

–

–

–

23.0

22.5

17.0*

–

19.0*

–

15.0

18.5

HU MID ML

22.0*

23.0*

HU MID AP

24.0*

24.0*

–

–

22.0

22.0

21.0*

–

23.0*

–

19.0

18.0

HU LST CIR

71.0

72.0

–

–

72.0

69.0

67.0*

–

70.5*

–

60.5

58.5

RA LEN

–

229.0

–

220.0

–

–

–

–

–

–

–

–

RA HD DIA

–

22.5

–

–

–

–

–

–

–

–

–

–

RA MID DIA ML

17.5

15.5

–

–

16.5

16.0

–

–

–

–

14.5

13.5

RA MID DIA AP

13.0

12.5

–

–

12.5

12.0

–

–

–

–

10.0

9.5

UL LEN

–

–

–

–

–

–

–

–

–

–

–

–

UL PHYS LN

–

–

–

–

–

–

–

–

–

–

–

–

UL DIA ML

17.5

17.0

13.0

–

16.0

–

–

16.0

17.0

–

17.0

15.0

UL DIA AP

20.0

21.5

16.5

–

20.0

–

–

17.5

17.0

–

18.0

17.5

UL LST CIR

–

45.0

35.0

34.0

47.0

47.0

–

48.0

45.5

–

27.0

28.0

SAC ANT L

112.0

–

–

–

–

–

SAC ANT SUP BR

–

–

–

–

–

–

SAC M TR DIA B

48.0

–

–

–

–

–

INN H

–

–

–

–

–

–

–

–

–

–

–

–

INN IL LEN

–

–

–

–

–

–

–

–

–

–

–

–

INN PUB LEN

–

–

–

–

–

–

–

–

–

–

–

–

INN ISCH LEN

82.0

–

–

66.0

–

–

–

–

–

–

–

–

FE LEN

–

–

–

–

–

–

–

–

–

–

–

–

FE BIC LEN

–

–

–

–

–

–

–

–

–

–

–

–

FE BIC BR

83.0

81.5

–

–

–

–

–

–

–

–

–

–

FE EPIC BR

84.0

83.0

–

–

–

–

–

–

–

–

–

–

FE HD DIA

50.0

51.5

38.0

42.0

–

–

–

–

–

–

–

–

FE SUB DIA ML

33.0

–

–

31.0

–

–

33.0*

33.0

34.0*

32.0

–

26.5

FE SUB DIA AP

29.0

–

–

25.0

–

–

26.5

26.5

27.0*

28.0*

–

20.5

FE MID ML DIA

–

–

–

29.0*

29.0

29.5

26.0

28.0

28.0

28.0

–

19.0

FE MID AP DIA

–

–

–

26.0*

33.0

33.0

28.0

29.0

27.5

28.0

–

24.0

FE MID CIR

–

–

–

82.0*

99.0

98.5

87.0

92.0

85.5

89.5

–

70.5

FE LC DIA AP

62.5

60.0

–

–

–

–

–

–

–

–

–

–

FE MC DIA AP

62.0

55.0

–

–

–

–

–

–

–

–

–

–

THE COMPOSITION OF THE SKELETAL SAMPLE

265

TABLE 5.10 (CONT.) T2 68-003A

T268-003B

M Trait

LOU 69-801A

F

DEC 69-901A

M

DEC 69-901B

–

WF 62-001A

–

F

R

L

R

L

R

L

R

L

R

L

R

L

PAT LEN

–

–

41.5

42.0

–

–

–

–

–

–

–

–

PAT BR

–

–

45.0

46.5

–

–

–

–

–

–

–

–

–

345.0

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

TIB LEN TIB M PROX BR

76.5

TIB M DIST BR

54.0*

–

–

39.5

–

–

–

–

–

–

–

–

TIB NF DIA ML

–

24.5

–

–

25.0

26.0

–

–

–

–

23.5

24.5

TIB NF DIA AP

76.0*

–

32.5

–

–

38.0

39.0

–

–

–

–

18.5

19.0

97.0

–

–

–

102.0

102.0

–

–

–

–

75.5

75.5

–

–

–

–

–

–

–

–

–

–

–

–

FIB MID MAX DIA

15.5

–

–

14.5

18.0

18.0

–

–

–

–

–

–

TAL LEN

53.5

–

52.0

–

–

–

–

–

–

–

–

–

CAL LEN

–

–

–

–

–

–

–

–

–

–

–

–

45.0*

–

–

–

–

–

–

–

–

–

–

–

TIB NF CIR FIB MAX LEN

CAL BR

Notes: An asterisk (*) indicates that a measurement is an estimate. For trait abbreviations, see the Appendix.

of burial by sex or age. In general, men and women, adults and children were buried in the same areas, though more adult women and children than men are interred near the Northeast Gate. Finally, the distribution of age and sex within multiple burials (Tables 4.1, 5.3) supports the hypothesis that these interments contained relatives, such as adult siblings, spouses, or parents and children.

Physical Appearance and Diversity A general sense of the physical appearance and diversity of local residents during the Late Roman to Early Byzantine periods can be derived from measurements of the skeletal remains. Anthropologists often characterize the shape of heads and limbs in skeletons that have reached biological maturity by calculating standard indices from pairs of measurements. The indices used in this study are described in the Appendix. The values of five cranial indices and two postcranial indices calculated for the adult skeletons are summarized in Table 5.15. Since the shape of the cranial vault, face, and jaws, which are significantly controlled by genetic factors, are considered the most distinctive features of individual appearance, images of the existing frontal, lateral, and posterior views of all sufficiently preserved adult heads are provided in Figures 5.8–5.40 at a scale of 1:3. Many cranial indices were not calculated because paired measurements could not be taken on several fragmentary cranial vaults. The basic impression drawn from the few indices that were calculated is that the heads tended to be mesocranic (neither narrow and long nor broad and round) and metriocranic (neither high nor low in the perpendicular aspect), while the only two measurable examples were hypsicranic (high in the midsagittal aspect). Notwithstanding the paucity of the metric data, there does not appear to have been significant sexual dimorphism in the shape of the cranial vault, except possibly in the cranial breadth index. Among males, T14 67-004A is distinguished by a particularly short, low, broad cranium (Fig. 5.25), while T14 67-002A has a somewhat longer and higher cranium (Fig. 5.24). Among females, NEG 69-001A has a strikingly short, low, broad cranium (Fig. 5.17), while NEG 69-007A and T14 69-002A have longer, higher, narrower crania (Figs. 5.19, 5.26).

TABLE 5.11. NONMETRIC TRAITS IN SUBADULT SKELETONS OF LATE ROMAN AND EARLY BYZANTINE DATE Subadult Skeletons of Late Roman Date Trait METOPISM

Scored

Not Scored

Scoring of Traits (n)

Incidence of Traits (%)

T14 69-991A R

L

5

11

0 P, 5 A

0.0 P, 100.0 A

11

21

0 NF, 11 NP, 0 FP, 0 A

0.0 NF, 100.0 NP, 0.0 FP, 0.0 A

–

–

ANT ETH FOR

2

30

2 S, 0 XS

100.0 S, 0.0 XS

–

–

POST ETH FOR

1

31

0 P, 1 A

0.0 P, 100.0 A

–

–

INFORB SUT

9

23

8 P, 1 A

88.9 P, 11.1 A

–

–

INFORB FOR

9

23

3 M, 6 S

33.3 M, 66.7 S

–

–

ZYG-FAC FOR

8

25

4 M, 3 S, 0 A

57.1 M, 42.9 S, 0.0 A

S

–

FR-TEMP ART

2

30

0 P, 2 A

0.0 P, 100.0 A

–

–

EPIPT BONE

2

30

0 P, 2 A

0.0 P, 100.0 A

–

–

COR OSS

3

13

0 M, 1 S, 2 A

0.0 M, 33.3 S, 66.7 A

SUPORB STR

BREGMA BONE

–

–

5

11

0 P, 5 A

0.0 P, 100.0 A

11

21

3 P, 11 A

27.2 P, 72.8 A

OSS LAMBDA

5

11

0 P, 5 A

0.0 P, 100.0 A

–

LAMBDOID OSS

6

10

3 P, 3 A

50.0 P, 50.0 A

–

OSS ASTER

5

27

2 P, 3 A

40.0 P, 60.0 A

–

–

PAR NOT BONE

4

28

0 P, 4 A

0.0 P, 100.0 A

–

–

LS PAL FOR

4

28

0 M, 4 S

0.0 P, 100.0 A

–

–

PAL TORUS

9

7

0 P, 9 A

0.0 P, 100.0 A

PTER-SPIN BR

0

32

0 P, 0 A

0.0 P, 0.0 A

–

–

FOR OVALE

3

29

1 C, 2 IC

33.3 C, 66.7 IC

–

–

FOR SPINOSUM

3

29

1 C, 2 IC

33.3 C, 66.7 IC

–

–

AUD TOR

19

14

0 P, 18 A

0.0 P, 100.0 A

–

A

TYM DEHISC

18

14

1 P, 17 A

5.6 P, 94.4 A

–

–

OCC CON FAC

10

22

10 S, 0 D1

100.0 S, 0.0 D

–

–

HYPOGLOS CANAL

13

19

11 S, 0 D

100.0 S, 0.0 D

S

S

6

10

0 P, 5 A

0.0 P, 100.0 A

PHAR FOSSA

2

14

0 P, 2 A

0.0 P, 100.0 A

MENTAL FOR

24

8

0 M, 22 S

0.0 M, 100.0 S

S

S

MYLOHY ARCH

19

14

0 P, 18 A

0.0 P, 100.0 A

–

A

MAND TORUS

24

8

0 P, 22 A

0.0 P, 100.0 A

A

A

SSCAP F

19

13

0 P, 19 A

0.0 P, 100.0 A

–

–

CIRC SUL

16

16

0 P, 16 A

0.0 P, 100.0 A

–

–

ACR ART FAC

15

17

0 P, 15 A

0.0 P, 100.0 A

–

–

SEPT APERT

22

10

0 P, 22 A

0.0 P, 100.0 A

–

–

SUPCON PRO

PAR FOR

PRECON TUB

– –

–

–

A –

21

11

0 P, 21 A

0.0 P, 100.0 A

–

–

ALLEN

0

32

0 P, 0 A

0.0 P, 0.0 A

–

–

POIRIER

0

32

0 P, 0 A

0.0 P, 0.0 A

–

–

3RD TR

16

16

0 P, 16 A

0.0 P, 100.0 A

–

–

PAT EMARG

2

30

0 P, 2 A

0.0 P, 100.0 A

–

–

VAS NOTCH

2

30

0 P, 2 A

0.0 P, 100.0 A

–

–

SQUAT FACET

2

30

0 P, 2 A

0.0 P, 100.0 A

–

–

INF TAL ART

0

32

0 S, 0 D

0.0 S, 0.0 D

–

–

PER TUB

1

31

0 P, 1 A

0.0 P, 100.0 A

–

–

Note: For trait abbreviations, see the Appendix. 1 The occipital condyles in five young children from one large multiple grave (NEG 67-001E-I) had not completed fused, though they appeared to be developing with single facets (see Scheuer and Black 2000, pp. 58–59).

THE COMPOSITION OF THE SKELETAL SAMPLE

267

TABLE 5.12. NONMETRIC TRAITS IN ADULT SKELETONS OF LATE ROMAN TO EARLY BYZANTINE DATE Trait

Scored

Not Scored

Scoring of Traits (n)

Incidence of Traits (%)

METOPISM

22

14

1 PC, 0 PI, 21 A

3.8 PC, 0.0 PI, 96.2 A

SUPORB STR

42

30

2 NF, 29 NP, 6 FP, 5 A

4.8 NF, 69.0 NP, 14.3 FP, 11.9 A

ANT ETH FOR

13

59

10 S, 3 XS

76.9 S, 23.1 XS

POST ETH FOR

14

58

13 P, 1 A

92.9 P, 7.1 A

INFORB SUT

43

29

14 P, 29 A

32.6 P, 67.4 A

INFORB FOR

43

29

0 M, 43 S

0.0 M, 100.0 S

ZYG-FAC FOR

49

23

1 M, 43 S, 5 A

2.0 M, 87.8 S, 10.2 A

FR-TEMP ART

26

46

0 P, 26 A

0.0 P, 100.0 A

EPIPT BONE

21

51

0 P, 21 A

0.0 P, 100.0 A

COR OSS

15

21

0 M, 5 S, 10 A

0.0 M, 33.3 S, 66.7 A

BREGMA BONE

19

17

0 P, 19 A

0.0 P, 100.0 A

PAR FOR

38

34

26 P, 12 A

68.4 P, 31.6 A

OSS LAMBDA

23

13

2 P, 21 A

8.7 P, 91.3 A

LAMBDOID OSS

21

15

13 P, 8 A

61.9 P, 38.1 A

OSS ASTER

27

45

3 P, 24 A

11.1 P, 88.9 A

PAR NOT BONE

27

45

2 P, 25 A

7.4 P, 92.6 A

LS PAL FOR

27

45

9 M, 18 S

33.3 M, 66.7 S

PAL TORUS

22

14

1 P, 21 A

4.5 P, 95.5 A

PTER-SPIN BR

20

52

0 P, 20 A

0.0 P, 100.0 A

FOR OVALE

16

56

16 C, 0 IC

100.0 C, 0.0 IC

FOR SPINOSUM

17

55

11 C, 6 IC

64.7 C, 35.3 IC

AUD TOR

56

16

0 P, 56 A

0.0 P, 100.0 A

TYM DEHISC

54

18

1 P, 53 A

1.9 P, 98.1 A

OCC CON FAC

30

42

30 S, 0 D

100 S, 0.0 D

HYPOGLOS CAN

29

43

26 S, 3 D

89.7 S, 10.3 D

PRECON TUB

15

21

10 P, 5 A

66.7 P, 33.3 A

PHAR FOSSA

19

17

8 P, 11 A

42.1 P, 57.9 A

MENTAL FOR

53

19

2 M, 51 S

3.8 M, 96.2 S

MYLOHY ARCH

41

31

7 P, 34 A

17.1 P, 82.9 A

MAND TORUS

50

22

2 P, 48 A

4.0 P, 96.0 A

9

69

1 P, 8 A

11.1 P, 88.9 A

CIRC SUL

SSCAP F

17

61

11 P, 6 A

64.7 P, 35.3 A

ACR ART FAC

19

59

11 P, 8 A

57.9 P, 42.1 A

SEPT APERT

38

40

6 P, 32 A

15.8 P, 84.2 A

SUPCON PRO

42

36

2 P, 40 A

4.8 P, 95.2 A

ALLEN

26

52

1 P, 25 A

3.8 P, 96.2 A

POIRIER

20

58

5 P, 15 A

25.0 P, 75.0 A

3RD TR

27

51

2 P, 25 A

7.4 P, 91.6 A

PAT EMARG

20

58

0 P, 20 A

0.0 P, 100.0 A

VAS NOTCH

19

59

4 P, 15 A

21.1 P, 78.9 A

SQUAT FACET

16

62

2 P, 14 A

12.5 P, 87.5 A

INF TAL ART

31

47

24 S, 7 D

77.4 S, 22.6 D

PER TUB

21

57

13 P, 8 A

61.9 P, 38.1 A

Note: For trait abbreviations, see the Appendix.

268

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.13. NONMETRIC TRAITS IN ADULT SKELETONS OF LATE ROMAN DATE Trait

Scored

Not Scored

Scoring of Traits (n)

Incidence of Traits (%)

METOPISM

15

13

1 PC, 0 PI, 14 A

6.7 PC, 0.0 PI, 93.3 A

SUPORB STR

29

27

2 NF, 21 NP, 3 FP, 3 A

6.9 NF, 72.4 NP, 10.3 FP, 10.3 A

ANT ETH FOR

5

51

2 S, 3 XS

40.0 S, 60.0 XS

POST ETH FOR

6

50

5 P, 1 A

83.3 P, 16.7 A

INFORB SUT

27

29

6 P, 21 A

22.2 P, 77.8 A

INFORB FOR

29

27

0 M, 29 S

0.0 M, 100.0 S

ZYG-FAC FOR

35

21

1 M, 31 S, 3 A

2.9 M, 88.6 S, 8.5 A

FR-TEMP ART

15

41

0 P, 15 A

0.0 P, 100.0 A

EPIPT BONE

14

42

0 P, 14 A

0.0 P, 100.0 A

COR OSS

10

18

0 M, 4 S, 6 A

0.0 M, 40.0 S, 60.0 A

BREGMA BONE

13

15

0 P, 13 A

0.0 P, 100.0 A

PAR FOR

27

29

19 P, 8 A

70.4 P, 29.6 A

OSS LAMBDA

16

20

1 P, 15 A

6.3 P, 93.8 A

LAMBDOID OSS

16

20

11 P, 5 A

68.8 P, 31.3 A

OSS ASTER

16

40

3 P, 13 A

18.8 P, 81.3 A

PAR NOT BONE

16

40

1 P, 15 A

6.3 P, 93.8 A

LS PAL FOR

17

39

6 M, 11 S

35.3 M, 64.7 S

PAL TORUS

15

13

1 P, 14 A

6.7 P, 93.3 A

9

47

0 P, 9 A

0.0 P, 100.0 A

10

46

10 C, 0 IC

100.0 C, 0.0 IC

PTER-SPIN BR FOR OVALE FOR SPINOSUM

11

45

8 C, 3 IC

72.7 C, 27.3 IC

AUD TOR

41

15

0 P, 41 A

0.0 P, 100.0 A

TYM DEHISC

39

17

0 P, 39 A

0.0 P, 100.0 A

OCC CON FAC

17

39

17 S, 0 D

100 S, 0.0 D

HYPOGLOS CAN

14

42

12 S, 2 D

85.7 S, 14.3 D

PRECON TUB

8

20

5 P, 3 A

62.5 P, 37.5 A

PHAR FOSSA

13

15

7 P, 6 A

53.8 P, 46.2 A

MENTAL FOR

37

19

2 M, 35 S

5.4 M, 94.6 S

MYLOHY ARCH

29

27

7 P, 22 A

24.1 P, 75.9 A

MAND TORUS

34

22

0 P, 34 A

0.0 P, 100.0 A

6

54

0 P, 6 A

00 P, 100.0 A

12

48

8 P, 4 A

66.7 P, 33.3 A

SSCAP F CIRC SUL ACR ART FAC

12

48

8 P, 4 A

66.7 P, 33.3 A

SEPT APERT

27

33

3 P, 24 A

11.1 P, 88.9 A

SUPCON PRO

29

31

0 P, 100 A

0.0 P, 100.0 A

ALLEN

13

47

1 P, 12 A

7.7 P, 92.3 A

POIRIER

10

50

2 P, 8 A

20.0 P, 80.0 A

3RD TR

16

44

0 P, 16 A

0.0 P, 100.0 A

PAT EMARG

13

47

0 P, 13 A

0.0 P, 100.0 A

VAS NOTCH

12

48

4 P, 8 A

33.3 P, 66.7 A

9

51

0 P, 9 A

0.0 P, 100.0 A

INF TAL ART

SQUAT FACET

23

37

18 S, 5 D

78.3 S, 21.7 D

PER TUB

14

46

7 P, 7 A

50.0 P, 50.0 A

Note: For trait abbreviations, see the Appendix.

THE COMPOSITION OF THE SKELETAL SAMPLE

269

TABLE 5.14. NONMETRIC TRAITS IN SIX ADULT SKELETONS OF DISPARATE DATES T2 68-003A

T2 68-003B

F Trait

R

METOPISM SUPORB STR

LOU 69-801A

F L

R

A

DEC 69-901A

M L

R

–

– L

R

A –

–

NP

–

–

ANT ETH FOR

–

–

–

–

–

–

–

–

POST ETH FOR

–

–

–

–

–

–

–

–

INFORB SUT

A

–

–

–

–

–

–

–

INFORB FOR

S

–

–

–

–

–

–

ZYG-FAC FOR

–

–

–

–

A

–

FR-TEMP ART

–

A

–

–

–

–

EPIPT BONE

–

A

–

–

–

–

BREGMA BONE PAR FOR

A P

–

–

NP

A NP

S

S

–

–

P

P

–

–

A

A

–

–

–

S

S

S

–

–

–

A

M

A

–

–

–

A

A

A

–

–

–

P

P

–

–

–

A

– –

L

FP

A –

R

NP

–

– P

F L

A

–

–

R

–

FP

A

WF 62-001A

F L

NP

COR OSS

TC 60-001A

A

A

A A

P

A A

A

A

OSS LAMBDA

A

–

–

A

A

A

LAMBDOID OSS

–

–

–

P

P

P

OSS ASTER

–

A

–

–

–

–

–

–

P

A

A

–

PAR NOT BONE

–

A

–

–

–

–

–

–

A

A

A

–

LS PAL FOR

–

–

–

–

–

–

–

–

S

S

–

S

PAL TORUS

A

–

–

–

A

A

PTER-SPIN BR

–

–

–

–

–

–

–

–

A

A

–

–

FOR OVALE

–

–

–

–

–

–

–

IC

C

IC

–

–

FOR SPINOSUM

–

–

–

–

–

–

–

IC

C

–

–

–

AUD TOR

–

A

A

–

A

A

–

–

A

A

A

A

TYM DEHISC

–

A

A

–

A

A

–

–

A

A

A

A

OCC CON FAC

–

–

–

–

–

–

–

–

D

D

D

–

HYPOGLOS CAN

–

–

–

–

–

–

–

–

S

S

S

D

PRECON TUB

A

PHAR FOSSA

–

A

–

–

–

–

P

–

P

A

P

MENTAL FOR

–

S

–

S

S

S

–

S

–

–

–

M

MYLOHY ARCH

–

–

A

–

A

A

–

A

–

–

A

A

MAND TORUS

A

A

–

A

A

A

A

A

–

–

A

A

SSCAP F

–

–

–

–

–

–

–

–

–

–

–

–

CIRC SUL

–

–

–

–

–

A

–

A

–

–

A

–

ACR ART FAC

–

–

–

–

–

–

–

–

–

–

–

–

SEPT APERT

–

–

–

A

–

–

–

–

–

–

–

–

SUPCON PRO

A

A

A

A

–

–

–

–

–

–

A

A

ALLEN

P

P

–

A

–

–

–

–

–

–

–

–

POIRIER

P

P

–

A

–

–

–

–

–

–

–

–

3RD TR

–

–

–

P

–

–

–

–

–

–

A

A

PAT EMARG

–

–

A

–

–

–

–

–

–

–

–

–

VAS NOTCH

–

–

A

–

–

–

–

–

–

–

–

–

SQUAT FACET

A

A

A

–

A

–

A

–

–

–

–

–

INF TAL ART

D

–

S

–

–

–

–

–

–

–

–

–

PER TUB

–

–

–

–

–

–

–

–

–

–

–

–

Note: For trait abbreviations, see the Appendix.

270

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

TABLE 5.15. CRANIAL AND POSTCRANIAL INDICES AMONG SKELETONS OF LATE ROMAN TO EARLY BYZANTINE DATE Males Trait

n

Range

Females Mean

sd

n

Range

Mean

sd

CRANIAL IND

3

69.1–84.8

76.2

7.9

5

69.2–81.7

75.2

4.6

CRANIAL H IND

1

–

76.5

–

1

–

75.4

–

CRANIAL BR IND

3

90.2–97.6

94.6

3.9

1

–

98.6

–

ORBITAL IND

7

75.3–92.4

83.2

6.4

6

83.8–96.0

89.7

4.8

NASAL IND

9

42.9–59.8

50.2

5.7

5

45.0–54.2

48.9

3.9

PLATYM IND

7

75.0–87.5

82.8

4.4

7

66.7–89.7

80.2

8.8

PLATYC IND

7

58.5–68.1

64.7

3.4

8

62.1–72.1

68.5

3.1

Note: For trait abbreviations, see the Appendix.

16

No. of Individuals

14 12 10 8 6 4 2 0 0–4 years

5–9 years

10–14 years

15–24 years

25–34 years

35–44 years

45+ years

Age Class Figure 5.6. Age structure among skeletons of Late Roman to Early Byzantine date

8

Male Female

No. of Individuals

7 6 5 4 3 2 1 0 15–24 years

25–34 years

35–44 years

45+ years

Adult Age Class Figure 5.7. Distribution of age at death by sex among skeletons of Late Roman to Early Byzantine date

THE COMPOSITION OF THE SKELETAL SAMPLE

Figure 5.8. NEG 69-103B. Scale 1:3

Figure 5.9. NEG 69-004B. Scale 1:3

Figure 5.10. NEG 69-004C. Scale 1:3

Figure 5.11. NEG 69-004D. Scale 1:3

271

272

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Figure 5.12. NEG 67-001B. Scale 1:3

Figure 5.13. NEG 67-001C. Scale 1:3

Figure 5.14. NEG 67-001D. Scale 1:3

THE COMPOSITION OF THE SKELETAL SAMPLE

Figure 5.15. NEG 67-003B. Scale 1:3

Figure 5.16. NEG 69-005C. Scale 1:3

Figure 5.17. NEG 69-001A. Scale 1:3

Figure 5.18. NEG 69-001D. Scale 1:3

273

274

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Figure 5.19. NEG 69-007A. Scale 1:3

Figure 5.20. NEG 69-007B. Scale 1:3

Figure 5.21. NEG 69-999A

(occipital only). Scale 1:3

Figure 5.22. T2 68-002A. Scale 1:3

THE COMPOSITION OF THE SKELETAL SAMPLE

Figure 5.23. T2 68-003A. Scale 1:3

Figure 5.24. T14 67-002A. Scale 1:3

Figure 5.25. T14 67-004A. Scale 1:3

275

276

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Figure 5.26. T14 69-002A. Scale 1:3

Figure 5.27. T14 69-003A. Scale 1:3

Figure 5.28. LOU 69-801A. Scale 1:3

THE COMPOSITION OF THE SKELETAL SAMPLE

Figure 5.29. RB 76-002A. Scale 1:3

Figure 5.30. RB 76-002C. Scale 1:3

Figure 5.31. HO 70-901A. Scale 1:3

277

278

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Figure 5.32. HO 70-901B. Scale 1:3

Figure 5.33. HO 70-901C. Scale 1:3

Figure 5.34. HO 70-902D. Scale 1:3

Figure 5.35. HO 70-902F. Scale 1:3

THE COMPOSITION OF THE SKELETAL SAMPLE

Figure 5.36. HO 70-902H. Scale 1:3

Figure 5.37. HO 70-902I. Scale 1:3

Figure 5.38. HO 70-902J. Scale 1:3

279

280

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Figure 5.39. WF 62-001A. Scale 1:3

Figure 5.40. TC 60-001A. Scale 1:3

Facial structures provided more metric data than cranial vaults because they were better preserved throughout the sample. Indices calculated from the dimensions of the eye orbit and nasal aperture offer a more solid basis for characterizing the form of two dominant features of the human face, the nose and eyes. The shape of the eye orbits varies between individuals and between sexes. Both males and females are mesoconch (neither wide nor narrow orbits), but males have on average slightly wider and women on average slightly narrower orbits. Among males, T14 67-004A is distinguished by relatively round, narrow eye orbits (Fig. 5.25), while NEG 69-004C has proportionately wider and more angular eye orbits (Fig. 5.10). Among females, NEG 67-001D and 69-007A have round, narrow eye orbits (Figs. 5.14, 5.19), while HO 70-901B has wider, more angular eye orbits (Fig. 5.32). The shape of the nasal aperture is more variable than the shape of the eye orbit, though both males and females are on average mesorrhine to slightly platyrrhine (moderately to slightly broad nasal aperture). Among males, NEG 67-001C and T14 67-004A have unusually tall, narrow nasal apertures (Figs. 5.13, 5.25), while T14 67-002A and 69-003A have shorter, wider nasal apertures and very long nasal bones (Fig. 5.24). Among females, T14 69-002A is distinguished by a broad, wide nasal aperture (Fig. 5.26), while NEG 67-001B, 67-001D, and 69-007A all have relatively narrow nasal apertures (Figs. 5.12, 5.14, 5.19). Two measurements that are also indicative of basic facial appearance are bizygomatic breadth and bigonial breadth, which exhibited significant sexual dimorphism in this sample (Tables 5.5–5.7). The cheeks and jaws of males are on average noticeably wider in the anterior aspect than those of females. Compare, for example, the broad cheeks of the males NEG 69-004C and 67-003B (Figs. 5.10, 5.15) to the narrow cheeks of the female NEG 69-007A (Fig. 5.19), and the broad lower jaw of the male T14 67-004A and probable male HO 70-901C (Figs. 5.25, 5.33) to the narrow lower jaw of the same female (Fig. 5.19).

THE COMPOSITION OF THE SKELETAL SAMPLE

281

The relative flatness of long bone shafts in legs could be measured in several cases because diaphyses were well preserved. The femoral shafts in both males and females are platymeric (flat in the anterior-posterior aspect), though the variability in shape among females is quite high. The shape of tibial shafts also differ somewhat between the sexes, but both males and females can be characterized as mesocnemic to slightly eurycnemic (broad to slightly flat in the medial-lateral aspect). On average, the long bones of males are significantly larger and more robust than those of females. Several explanations have been proposed for the variable shape of the leg diaphyses by sex, including differential mechanical adaptation, stress during growth, and nutrition.88 Almost all cranial, mandibular, and postcranial measurements and indices in the Late Roman to Early Byzantine sample display normal distributions and small to moderate standard deviations (Tables 5.5, 5.6, 5.8, 5.9, 5.15). The sexual dimorphism visible throughout the skeleton but particularly in the face and limbs reflects essential differences in size and rugosity. It will be seen that the stature of adults was markedly dimorphic. Apart from basic physical variation between males and females, no individuals would have stood out because of their unusual physique; they belonged to a distinct group that was rather homogeneous in appearance. This uniformity suggests a level of genetic continuity among local residents from Roman to Byzantine times. Morphological homogeneity supports the historical model presented in Chapter 3 of a continuous occupation of the site by native Corinthians who did not permanently abandon their settlement because of natural disaster or foreign invasion in late antiquity. Certainly the skeletal morphology does not show that a biologically distinct group, such as Slavic settlers, lived and died at the site.89 It is interesting that the only skeleton that can be clearly differentiated from the others in terms of physical appearance, TC 60-001A, was also the latest individual interred at the site. This young woman’s cranium was strikingly short and wide, her eye orbits wide and angular, and her nasal aperture comparatively broad (Fig. 5.40). Without further evidence for the skeletal biology of the local residents of the Late Byzantine to Early Modern periods, her craniomorphological affinites are uncertain.90 Although it is impossible to distinguish morphological subgroups within such a small skeletal sample, two pairs of individuals sharing graves display similar facial features that may point to familial connections. One older and one younger woman in NEG 67-001 (B and D) both have unusually narrow nasal apertures and narrow, rounded eye orbits (Figs. 5.12, 5.14). If these two women died and were buried in the same grave over a short span of time, their ages (35–44 and 20–21 years) indicate that they might have been mother and daughter. If the older individual, however, outlived the younger one by several years, they might have been sisters. A familial relationship between the individuals in this grave is further supported by the coincidence of certain rare nonmetric traits, which will be discussed below. In addition, the old woman and the young adult, probably a male, in NEG 69-007 (A and B) both have exceptionally narrow, round eye orbits that were relatively close together (i.e., low interorbital breadth; Figs. 5.19, 5.20). The ages of these two individuals (late 40s– mid-50s and 16–18 years) indicate that they might have been mother and son. It seems less likely that they were siblings who died several decades apart. The relationship between this skeletal sample and others cannot be determined because no metric data set of sufficient size or coherent provenience from a contemporary site in 88. Larsen 1997, pp. 21–22, 213–218; Capasso, Kennedy, and Wilczak 1999, pp. 114–117, 128; Scheuer and Black 2000, p. 403. 89. Cf. Larsen 1997, pp. 330–331, discussing morphological distinctions between European settlers and indigenous peoples in the New World.

90. Orbital and tibial morphology in other measurable skeletons of Early to Middle Roman (T2 68-003A, B, WF 62-001A) and Middle to Late Byzantine date (LOU 69-801A) essentially conforms with the patterns observed in the Late Roman to Early Byzantine skeletal sample.

282

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

Greece has been published. Morphometric profiles have been reported for only a few Roman and Byzantine skeletons from Corinth, Argos, and Tiryns.91 The major craniomorphological surveys by J. Lawrence Angel include numerous skulls from Roman and Byzantine graves,92 but the geographic distribution of these sites is so wide that they hardly comprise a unified data set for comparison. Anthropologists in the past have often tried to identify skeletons from Greek sites with larger historical populations, or races, in Eurasian geography, a program of research pursued by Angel and his contemporary, Robert Charles.93 Scholars have since abandoned this approach for several reasons: the morphological types used by Angel and others were based on disparate samples; they were poor reflections of genetic reality; they were theoretical constructs without secure historical basis; they oversimplified biological diversity by applying a generalizing concept of races; and they underestimated the range of variability within single groups.94 Despite the shortcomings of this program, the basic goal of tracing genetic relationships between groups over space and time was an important one. The modern study of biological distance between skeletal samples at different sites through the statistical analysis of polygenic traits can reveal the relatedness or divergence between ancient peoples. Such analysis, however, requires the comprehensive collection and publication of both metric and nonmetric traits from human remains at different sites, the formation of samples that accurately represent the biological populations from which they are derived, and the selection of traits that are directly controlled by genes.95 As more human remains from Roman and Byzantine graves in the Corinthia are fully published, it will be possible to situate the skeletal sample from the Isthmus within the changing regional population by charting its distance from or proximity to other samples.96

Growth and Stature The size of postcranial bones in subadults and adults reveal patterns of growth and stature. As individuals age from birth to adulthood, they experience intensive phases of growth during the first years of life and adolescence, after which osseous fusion is completed and full stature is reached. Growth and stature are two basic aspects of an individual’s physical 91. Charles 1962, pp. 229, 231–234 (Corinth, Roman); Gejvall and Henschen 1968, pp. 179–185, 189–192 (West Shops, Corinth, Late Roman–Early Byzantine); A. Koumaris, in Davidson 1937, p. 230 n. 1 (grave 1; tower, Acrocorinth, Early Byzantine); Charles 1958b, pp. 290–292, 305, tables I–V, pls. XX, XXI (tombs 94, 96; Quartier Sud, Argos, Early and Late Roman); Charles 1963, pp. 57–60, 71, tables V–VIII, pl. XIX (tombs 200–202; cemetery southwest of city, Argos, 5th–18th centuries); Bräuer 1980, p. 292 (Unterburg, Tiryns, Middle Byzantine). 92. Angel 1944a, pp. 331, 366–367, tables 1, 8, 9 (“Roman and Late Roman: 150 b.c.–450 a.d.,” “Byzantine and Medieval: 450– 1300 a.d.,” from Boeotia, Attica, Corinthia, Argolid); 1945a, pp. 14–16, table 3 (same); 1945b, pp. 312–316, 324–325, 342–344, 352, 357, 360, tables 1–4 (“Roman period 1–450 a.d.,” “Byzantine ca. 450–1300 a.d.,” Attica); 1951, n.p., table 5 (“Greece, Imperial Roman: 120–600 a.d.”). 93. E.g., Angel 1941 (Corinth), 1944a, 1945a, 1945b (Attica), 1945–1948 (Cyprus); Charles 1958a (survey); 1958b, 1962 (Crete, Corinth), 1963 (Argos); cf. Boev 1972, pp. 166, 167, 169–170, tables 64, 65. Others continued to apply this strategy to the study of human remains from Late Antique and Byzantine graves: e.g., Poulianos 1972 (Ayios Achilleios, Prespa), 1974 (Samos); Boev 1972, pp. 170–171, table 66 (Ayios Achil-

leios, Prespa); Hermann 1978, pp. 142–143 (Toumba near Kastanas, Macedonia); Xirotiris 1979, pp. 168–170, pl. XXVII:3, 4 (survey); Mallegni 1988, pp. 373–380 (Gortyn, Crete). 94. Cf. Wesolowsky 1973, p. 344, criticizing Angel’s identification of two Late Roman or Early Byzantine skeletons buried in Lerna Hollow at Corinth as “negroid” (Wiseman 1969, p. 86, nn. 38, 39); see also the assessment in Roberts et al. 2005, pp. 41–42. 95. Larsen (1997, pp. 302–332) provides a general discussion of biological distance in skeletal studies. 96. An early study of the Isthmian sample (Rife and Giesen 1994) attempted to assess the biological distance between the Classical skeletons from the West Cemetery and the Late Roman to Early Byzantine skeletons from the fortifications based on select cranial and mandibular measurements. The study tentatively proposed biological continuity between the two samples but stressed the provisional nature of this conclusion. The case cannot be proven because of the small size of the sample and the simplicity of the statistical test. Moreover, without substantial skeletal remains of Hellenistic or Early Roman date for comparison, the apparent craniomorphological similarity between the earlier and later samples may be attributed to phenotypic stabilization or even reversion after a period of divergence rather than to genetic continuity.

THE COMPOSITION OF THE SKELETAL SAMPLE

283

appearance that arise from various biological and environmental factors. The reconstruction of subadult growth and adult stature in a skeletal sample is an important component of osteological analysis because it furnishes a criterion for evaluating somatic differences between groups. The inclusion of stature in anthropological reports serves another simple but important purpose in helping modern readers imagine what ancient people might have looked like. The human remains of Late Roman to Early Byzantine date from the Isthmus attest to the growth and stature of local residents. However, the significance of these local characteristics cannot be fully established because few metric data exist from other sites for comparison. Growth rates are usually calculated for postcranial bones prior to epiphyseal fusion using age at death as indicated by dental development. In the Late Roman to Early Byzantine skeletal sample from the Isthmus, growth was only measured in the maximum diaphyseal lengths of bones present in at least nine subadults, namely, the clavicle, humerus, ulna, radius, and fibula (Fig. 5.41). The ages used to determine the rate of growth are the means of the age ranges estimated for each individual. Since the oldest subadult was no older than twelve years, these rates reflect maturation only in infancy and childhood, not adolescence. Over this span of life, the limb bones grew gradually longer, with the greatest increase in length occurring in the fibula. Although many measurements could not be taken in several subadult skeletons, the general impression is that the diameters of all diaphyses and the dimensions of the scapula, ilium, ischium, and pubis grew proportionately throughout childhood. Growth is influenced by several factors, including genetic constitution, hormonal deficiencies, and especially nutrition and disease, which often operate in concert.97 The growth rate among subadults at the Isthmus followed a more or less regular progression. No metric data from contemporary Greek samples is available for comparison. The growth curves for limb diaphyses are essentially comparable to those of both early Native American skeletons and the Late Romano-British skeletons from the Poundbury cemetery at Dorchester, England.98 Adult stature is determined by numerous biological, environmental, and pathological factors that are specific to a genetic group and the conditions of its habitat.99 Living stature can be calculated from certain skeletal measurements because the total standing height of a human body correlates with the length of its limb bones during all developmental stages in life. However, this correlation is not constant either within or between populations.100 The most widely used method for calculating stature, the Trotter-Gleser formulae,101 is based on the study of modern Americans but is commonly applied to archaeological samples of much earlier dates in different regions. It must be stressed that estimates of stature in pre-modern, Old World skeletons made with these formulae are at best approximations, not perfectly accurate.102 The analysis of the human remains from the Isthmus used the Trotter-Gleser regression equations for modern “white” Americans of predominantly European ancestry.103 These equations required the maximum lengths of intact, nonpathological humeri, ulnae, radii, femora, and fibulae.104 Although the formula for each long bone generates a precise 97. Larsen 1997, pp. 8–13; Goodman and Martin 2002, pp. 19–21. 98. Ubelaker 1999, p. 69, figs. 78–80, 83 (Indian Knoll, Kentucky; Arikara skeletons, South Dakota); Molleson 1993, pp. 177–178, fig. 113 (Poundbury, Dorchester). 99. Larsen 1997, pp. 13–19; Goodman and Martin 2002, pp. 19–20, 21–22. 100. Krogman and İşcan (1986, pp. 302–351) survey and evaluate methods for measuring stature; see also Ubelaker 1999, pp. 60–63. 101. Trotter and Gleser 1952, 1958; Trotter 1970.

102. E.g., Formicola (1993) discusses the shortcomings of the Trotter-Gleser formulae for “whites” and “negroes” when applied to Neolithic skeletons from separate regions of Europe. 103. Trotter 1970, table 28, reprinted in Ubelaker 1999, p. 61, table 9, and White 2000, p. 372, table 17.2. 104. Tibiae were omitted because Jantz, Hunt, and Meadows (1995) have shown that the Trotter-Gleser regression equations for this bone were based on mismeasurements of the original specimens.

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

a. Clavicle (n = 11) 100 90

Length (mm)

80 70 60 50 40 30 20 10 0 0

2

4

6

8

10

Age (years)

b. Humerus (n = 13) 250

Length (mm)

200 150 100 50 0

0

2

4

6

8

10

12

Age (years)

c. Ulna (n = 11) 160 140 120

Length (mm)

284

100 80 60 40 20 0 0

2

4

6

Age (years)

8

10

THE COMPOSITION OF THE SKELETAL SAMPLE

285

d. Radius (n = 11) 160 140

Length (mm)

120 100 80 60 40 20 0 0

2

4

6

8

10

Age (years)

e. Fibula (n = 9) 300

Length (mm)

250 200 150 100 50 0 0

2

4

6

8

10

12

Age (years) Figure 5.41. Growth rates in five diaphyseal lengths in 17 individuals up to 12 years old

estimate, roughly two-thirds of all estimates from each formula will fall within a range of the actual stature. This is because the correlation between the lengths of specific long bones and actual stature varied within the skeletal sample from which the equations were derived. This range of variation is the standard error. Estimates based on certain bones have a greater standard error than those based on others. For this reason, it is preferable to calculate stature using those bones with the lowest standard errors in the reference sample, namely, the femur and fibula in both males and females. However, leg bones in the Isthmian sample are not as well preserved as arm bones (Fig. 5.4), which have a greater standard error in stature estimates. Therefore, in order to estimate the height of as many individuals of each sex as possible, all sufficiently preserved arm and leg bones from an individual have been used, and the final estimate of that individual’s stature is the arithmetic mean of the estimates

286

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

calculated for each bone.105 An adjustment was made to each estimate for individuals over the age of 30 to account for the gradual compression of the upper body with advancing age. In the end, it was possible to estimate the living stature of 12 of 17 adult males and 14 of 19 adult females representing the Late Roman to Early Byzantine periods.106 The calculation of living stature by sex is presented in Table 5.16 and Figure 5.42. Males are on average 166.8 cm (5'6") and females are on average 156.4 cm (5'1").107 Sexual dimorphism in body height is considerable, with only a little overlap in the distribution of stature by sex. The tallest man, whose body was distinguished by its overall robustness, is T14 69-003A at 172.7 cm (5'7"), while the shortest is NEG 67-001C at 160.6 cm (5'3"). The tallest woman is NEG 69-004B at 163.2 cm (5'4"), while the shortest, an unusually gracile individual, is NEG 69-005B at 150.2 cm (4'11"). If the skeletal sample from the Isthmus is representative, the regularity of stature among individuals of the same sex is noteworthy. It is difficult to ascertain the complex factors that contribute to stature in any ancient skeletal sample. Nonetheless, such a low statural variation among the rural residents would not be expected if environmental (e.g., diet, nutrition, living and occupational conditions) or genetic factors (e.g., influx of outsiders) were changing dramatically from the Roman to Byzantine eras. The height of these men and women can be compared to statures at eighteen Greek sites of Roman to Modern date (Table 5.17, Fig. 5.43).108 Since few stature estimates based on skeletal remains have been published, it is impossible to draw secure conclusions regarding interpopulational relations on the basis of stature. The initial impression given by a comparison of the scant data available is one of relative stability in Greek stature over the past two millenia, with perhaps a slight increase particularly in males since the Byzantine era. The ancient residents of the Isthmus were on average somewhat shorter than modern Greeks but roughly similar in stature to other Roman and Early Byzantine groups. The few skeletons from contemporary graves in the Sanctuary of Demeter and Kore, the Forum, and Lerna Hollow at Corinth are slightly taller than those from the Isthmus, while those from graves of earlier date around the city were slightly shorter. As at the Isthmus, most skeletons from the urban cemeteries display sexual dimorphism in stature. Among Late Roman to Early Byzantine skeletons from elsewhere in Greece, the Cretans (Eleutherna, Gortyn) were slightly taller, especially the men (Knossos), while the northern Greeks were roughly the same height (Toumba, Polystylon) or slightly shorter (Aliki on Thasos).

Nonmetric Variability In addition to demography, physical appearance, growth, and stature, the composition of a skeletal group is defined by the distribution of nonmetric traits. These skeletal features can vary by age, sex, and side of the body. Moreover, if it is assumed that they are genetically controlled, their incidence may indicate familial relation. Tracing the occurrence of certain traits in bones from Medieval and post-Medieval graves at Corinth has led to observations concerning genetic relatedness and familial burial patterns among urban residents.109 On similar grounds, nonmetric traits can be used in comparative analysis to examine biological 105. See Brothwell 1981, pp. 101–102. 106. The stature of the following males could not be calculated: NEG 69-007B, HO 70-901C, 70-902D or I, E or F, H. The stature of the following females could not be calculated: NEG 69-103A, RB 76-002A, C, HO 70-902G, J. Among skeletons predating or postdating the Late Roman to Early Byzantine sample, stature could not be calculated for the following: DEC 69-901A, B, LOU 68-801A, WF 62-001A, TC 60-001A. 107. Note that the two skeletons of Early to Middle Roman date whose stature could be calculated are close to the mean

measurement for the Late Roman to Early Byzantine skeletons: T2 68-003A (male) was 164.5 cm (5'5") and T2 68-003B (female) was 158.3 cm (5'2"). 108. Laskaris (2000, p. 287) lists several estimated statures for human remains from Late Roman and Byzantine graves across Greece. 109. The preliminary results of recent analyses by E. Barnes are noted in Williams et al. 1998, p. 242; Blackman 1999, pp. 22, 23; Barnes 2003, p. 440.

THE COMPOSITION OF THE SKELETAL SAMPLE

287

TABLE 5.16. ESTIMATION OF LIVING STATURE IN 26 ADULTS OF LATE ROMAN TO EARLY BYZANTINE DATE Sex

n

Male

12

Female

14

Mean in cm (ft)

sd

Range (cm)

166.8 (5'6")

3.8

160.6–172.7

156.4 (5'1")

4.0

150.2–163.2

7

Female Male

6

Count

5 4 3 2 1 0 140–145 (4'7"–4'9")

146–150 151–155 (4'9"–4'11") (4'11"–5'1")

156–160 161–165 (5'1"–5'3") (5'3"–5'5") Stature in cm (ft)

166–170 (5'5"–5'7")

171–175 (5'7"–5'9")

176–180 (5'9"–5'11")

Figure 5.42. Distribution of stature by sex among skeletons of Late Roman to Early Byzantine date

distance between groups. However, sufficient nonmetric data from other archaeological sites in Greece are not available for such a study.110 The incidence of nonmetric traits in the skeletal sample at the Isthmus shows little variation by age, though it must be admitted that the subadult skeletons do not comprise an adequate data set for comparison with older individuals. Certain hyperostotic traits that occur with variable frequency in adult skeletons are absent in subadult skeletons (Tables 5.11–5.14).111 This may reflect the progressive development of such traits that typically accompanies periosteal bone growth in life.112 Due to the small sample size, it is unclear whether other traits related to ossification, such as the completeness of the foramen ovale,113 show a differential prevalence by age. Little sexual dimorphism is visible in the prevalence of nonmetric traits, though it should be stressed that the size of the adult skeletal sample is too limited to permit confident conclusions. Among cranial and mandibular traits, only the ossicle at the asterion shows differential prevalence, occurring slightly more frequently in males (3/12 scorable sides, or 25.00%; 24 sides were unscorable) than in females (0/15 scorable sides, or 0.00%; 21 sides were unscorable).114 Among postcranial traits, the septal aperture occurs more frequently 110. Wesolowsky (1973, pp. 348–349) notes the presence of lambdoid ossicles, ossicle at the asterion, and metopism in the fragmentary bones from Late Roman to Early Byzantine graves in Lerna Hollow at Corinth. Fox Leonard (1997, pp. 416–421, 454–455, table 12) records the frequencies of several nonmetric traits in a sample of 94 skeletons from various graves at Corinth, mostly of Early to Late Roman date. These individuals display many of the same osseous anomalies as have been observed in the remains from the Isthmus: extrasutural bones, septal apertures, squatting facets, metopism, divided hypoglossal canal, and vastus notch. The significance of the

relative frequencies of these traits in the two skeletal samples cannot be determined without further comparanda. In an ambitious study of cranial nonmetric traits, Ricaut and Waelkens (2008) attempt to trace population movements in the eastern Mediterranean by comparing data from an 11th–13th-century churchyard at Sagalassos to several other skeletal samples. 111. Precondylar tubercle, mylohyoid arch, mandibular torus, third trochanter, peroneal tubercle. 112. See Saunders 1989, p. 101. 113. See Sjøvold 1984, p. 243. 114. See Hauser and De Stefano 1989, p. 9.

9 – 4

12 2 23 1 3 16 52 7 10 1 1

Hellenistic–Roman Roman (“post-1st c. a.d.”) Early–Late Roman, with one possible Classical– Hellenistic Late Roman–Early Byzantine (end 4th/early 5th/7th or 8th c.) Late Roman Late Roman (late 4th to mid-6th c.) Late Roman–Early Byzantine Late Roman–Early Byzantine Late Roman–Early Byzantine Late Roman–Early Byzantine (5th–7th c.) Late Roman–Early Byzantine (5th–7th c.) Late Roman–Early Byzantine (6th–7th c.) Late Roman–Early Byzantine (6th–7th c.) Late Roman–Early Modern (5th–18th c.)

Various areas, Paphos, Cyprus

Area DD, Lerna, Argolid

Various areas, Corinth, Corinthia

Various areas, Isthmia, Corinthia

Sanctuary of Demeter and Kore, Corinth, Corinthia

Various areas, Messene, Messenia

Forum, Corinth, Corinthia

West Shops, Corinth, Corinthia

Basilica, Aliki, Thasos

Basilica, Eleutherna, Crete

Basilica, Knossos Medical Faculty site, Crete

Settore L, Gortyn, Crete

Ayia Mauri, Alassa, Cyprus

Southwest of city, Argos, Argolid

3

1

Roman Early–Late Roman (150 b.c.–a.d. 450)

1

No. Males

Roman (1st–3rd c.)

Period (date)

Various sites, central and southern Greece

Lower terraces, Midea, Argolid

Quartier Sud, Argos, Argolid

Area, Site, Region

160 (5'3")

155.6 (5'1")

173.22 (5'8")

169.30 (5'7")

169 (5'7")

162 (5'4")

170.3 (5'7")

181 (5'11")

170.0 (5'7")

169.4 (5'7")

166.8 (5'6")

165.76 (5'4")

–

171.13 (5'7")

164.86 (5'4")

175 (5'9")

166–167 (5'6")

Average Male Stature, cm (ft)

–

4

3

6

21

1

1

1

12

2

14

5

2

7

–

–

–

No. Females

–

155.6 (5'1")

160.56 (5'3")

157.79 (5'2")

160 (5'3")

150 (4'11")

160 (5'3")

158.5 (5'2")

152.0 (5'0")

166.3 (5'5")

156.4 (5'1")

148.29 (4'10")

148.10 (4'10")

155.91 (5'1")

–

–

–

Average Female Stature, cm (ft)

Lerna II, p. 149, table 14 (149, 149a)

Trotter and Gleser formulae

Wesolowsky 1973, p. 349

Trotter and Gleser formulae

Bourbou 2000, p. 293; 2003, p. 305; 2004, pp. 30, 49 Becker 2005, p. 380, table 2 (tombs 151, 158, 109, 211, 236, 309) Mallegni 1988, pp. 349–353, table 2 (162, 171, 172, 191, 101, 11, 113, 51, 361, 362, 2, 161, 162, 363, 4) Fox 1996, pp. 52, 61 Charles 1963, p. 57, n. 3 (tombe 201)

Trotter formulae Trotter and Gleser formulae Manouvrier 1892; Trotter and Gleser formulae Trotter and Gleser formulae Manouvrier 1892; Olivier 1960, p. 261

Buchet and Sodini 1984, pp. 230–233 (tombes 5, 10, 12)

Gejvall and Henschen 1968, p. 180

Trotter and Gleser formulae

Not published

Bourbou 2003, p. 305; 2004, pp. 30, 491

Corinth XVIII.3, pp. 384–387, nos. 5, 7, 19, 26

See pp. 283–286, Table 5.16, Fig. 5.42

Trotter formulae

Not published

Trotter and Gleser formulae

Fox Leonard 1997, pp. 403–405, 453–454, 485; Fox 2005, pp. 65–66

Fox Leonard 1997, pp. 403–405, 453–454, 485; Fox 2005, pp. 65–66

Eliakis, Eliakis, and Iordanidis 1966

Eliakis, Eliakis, and Iordanidis 1966

Angel 1944a, p. 363, table VIII

Ingvarsson-Sundström 2007, p. 477 (S7)

Charles 1958b, p. 290 (tombe 94)

Reference

Not published

Trotter formulae

Not published

Technique

TABLE 5.17. COMPARISON OF STATURE IN SEVERAL GREEK SKELETAL SAMPLES OF ROMAN TO MODERN DATE

Early Byzantine (7th or 8th c.) Early Byzantine (6th–9th c.)

West of Venizeleion, northwest of Knossos, Crete

ΞA1 cemetery (basilica), Polystylon

6

Late Byzantine (mid-14th to early 15th centuries) Early Modern (19th c.)

Church, Panakton, Attica

Church of the Holy Apostles, Agora, Athens, Attica 113

Modern (1960–1961)

Various sites, Greece6

171.10 (5'7")

165.46 (5'5")5

170 (5'7")

170.64 (5'7")

165.74 (5'5")

168.1 (5'6")

–

169.8 (5'7")

168.74 (5'6")

171.07 (5'7")

169.69 (5'7")2

–

165.57 (5'4")

–

Average Male Stature, cm (ft)

–

–

87

1

52

34

1

16

52

Several

–

1

–

1

No. Females

–

–

158 (5'2")

159.83 (5'3")

156 (5'1")

159.9 (5'3")

156.4 (5'1")

154.9 (5'1")

156 (5'1")

154.97 (5'1")

–

156 (5'1")

–

151.6 (5'0")

Average Female Stature, cm (ft)

Agelarakis and Agelarakis 1989, p. 19; Agelarakis 1997, p. 303

Trotter and Gleser formulae

Agelarakis and Agelarakis 1989, p. 19; Agelarakis 1997, p. 303

Trotter and Gleser formulae

From living bodies

Not published

Not published

Hertzberg et al. 1963, p. 41, table 4.3

Angel 1944a, pp. 363, 367–369, table VIII

Agora XX, p. 30

Gerstel et al. 2003, pp. 198, 202–204, 209, 211, 214, 216, table 1 (graves 1991-3a, 1992-1, -2a, -2b, -2c, -3, -4a)

Brothwell 1986, pp. 388–389, table 6 (A)

Trotter and Gleser formulae

Trotter and Gleser formulae

Wade et al. 1983, p. 402 (burial 10-1)

Trotter and Gleser formulae

Angel 1979, p. 40

Musgrave 1976, p. 41

Trotter and Gleser formulae

Not published

Fox 2003, pp. 275–277, table 1 (tombs 1, 2, 3, sarcophagus, cistern)

Hermann 1978, p. 140

Trotter and Gleser formulae; Eliakis, Eliakis, and Iordanidis 1966 Eliakis, Eliakis, and Iordanidis 1966

Angel 1944a, p. 363, table VIII

Wade et al. 1983, pp. 399–400 (burial 9-1)

Trotter and Gleser formulae Not published

Reference

Technique

2

1

This table gives the stature measurements from Bourbou 2004; Bourbou 2003 gives slightly different averages (1.70 m for males and 1.60 m for females). This average does not include the shorter individual 21, whose sex was not determined. Otherwise the total average stature for eight individuals was 168.6 cm (5'6"; Fox 2003, p. 276). 3 This is the total number of adults of this sex in the skeletal sample; the number of adults yielding stature measurements was not published. 4 “The sample has the following geographic deviation: Doris and Phocis (1); Boeotia, C. Euboea, Megaris (5); Athens (5); Corinth and Argolis (5); rest of Peloponnese (11); Ionian islands (6)” (Angel 1944a, p. 369, n. 25). 5 However, “[t]he present sample probably underestimates stature” (Angel 1944a, p. 369, n. 25). 6 These military personnel originate from all major regions of Greece (Hertzberg et al. 1963, p. 91, table 7.3B).

37

Modern (1939)

Various sites, Greece4

143

163

Middle Byzantine (12th–13th c.)

Polystylon (Abdera), Thrace

ΞA2 cemetery (domed church),

13

Middle Byzantine (12th–13th c.)

–

Middle Byzantine (“13th century or not much later”)

Byzantine chapel, Nichoria, Messenia

Ayios Polyeuktos, Constantinople (Saraçhane, Istanbul, Turkey)

68

Early–Middle Byzantine (7th–12th c.)

163

Several

7

–

5

–

No. Males

Theotokos Kyriotissa, Constantinople (Kalenderhane Camii, Istanbul, Turkey)

(Abdera), Thrace

Early Byzantine (early–mid-7th c.)

Areas I and II, Kopetra, Kalavasos, Cyprus

Early Byzantine

Byzantine (450–1300)

Various sites, central and southern Greece

Toumba, Kastanas, Macedonia

“Probably Roman or later”

Period (date)

Tholos, Nichoria, Messenia

Area, Site, Region

290

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

1. Ayia Mauri, Alassa, Cyprus 2. Ayios Polyeuktos Constantinople (Saraçhane, Istanbul, Turkey) 3. Aliki, Thasos 4. Argos, Argolid 5. Athens, Attica 6. Corinth, Corinthia 7. Eleutherna, Crete 8. Gortyn, Crete 9. Isthmia, Corinthia 10. Knossos, Crete

11. 12. 13. 14. 15. 16. 17. 18. 19.

Kopetra, Kalavasos, Cyprus Lerna, Argolid Messene, Messenia Midea, Argolid Nichoria, Messenia Panakton, Attica Paphos, Cyprus Polystylon (Abdera), Thrace Theotokos Kyriotissa, Constantinople (Kalenderhane Camii, Istanbul, Turkey) 20. Toumba, Kastanas, Macedonia

Figure 5.43. Sites of Roman and Byzantine date with skeletons of known stature for comparative study

in females (5/18 scorable sides, or 27.78%; 16 sides were unscorable) than in males (1/15 scorable sides, or 6.67%; 13 sides were unscorable). This discrepancy is probably due to the fact that female humeri in this skeletal sample are more gracile than male humeri, which rendered them conducive to hypostosis.115 The peroneal turbercle occurs slightly more frequently in males (6/11 scorable sides, or 54.55%; 17 sides were unscorable) than in females (2/5 scorable sides, or 40.00%; 29 sides were unscorable). But it should be stressed that the number of observable calcanei from skeletons that could be sexed was quite low. If this is a 115. See Saunders 1978, pp. 113–118, 470, table 5.

THE COMPOSITION OF THE SKELETAL SAMPLE

291

TABLE 5.18. COINCIDENCE OF NONMETRIC AND CONGENITAL TRAITS IN MULTIPLE BURIALS

Burial No.

NEG 67-001

RB 76-002

Prevalence of Observable Individuals1 Showing Trait

Prevalence of Trait in Total Sample of Elements2

Prevalence Among Elements in Grave2

Trait

Individuals Showing Trait (sex, age)

Ossicle at asterion present

C (M, late 40s) G (–, 4–8 years)

40.00% (2/5)

21.1% (8/38)

50.0% (3/6)

Vastus notch present

A (M, 35–44 years) B (F, 35–44 years)

100.0% (2/2)

18.2% (4/22)

100.0% (2/2)

Multiple lesser palatine foramina

C (M, late 40s) D (F, 20–21 years) E (–, 7–11 years)

60.0% (3/5)

26.5% (9/34)

62.5% (5/8)

Sacralized coccyx

A (M, 35–44 years) B (F, 35–44 years)

100.0% (2/2)

35.0% (7/20)

100.0% (2/2)

Single coronal ossicle

A (F, 35–44 years) C (F, 35–44 years)

100.0% (2/2)

28.6% (6/21)

100.0% (2/2)

Pharyngeal fossa present

B (M?, 35–44 years) C (F, 35–44 years)

100.0% (2/2)

36.0% (9/25)

100.0% (2/2)

Mylohyoid arch present

D (M?, 25+ years) E (–, 25+ years) F (M?, 25+ years) H (M, 35–44 years) J (F, 15–24 years)

83.3% (5/6)

10.6% (7/66)

77.8% (7/9)

Sacralized coccyx

G (F, 50+ years) J (F, 15–24 years) Z1 = E or F (M, 55+ years) Z5 = E or F (M, 55+ years) or D or I (M, 25–34 years)

80.0% (4/5)

35.0% (7/20)

80.0% (4/5)

HO 70-902

1 “Observable individuals” are skeletons in the multiple burial possessing an element sufficiently preserved for the observation of a particular trait. 2 The prevalences in the total sample at the site and within single graves are the prevalences among both adults and subadults.

significant difference, it could be explained by the fact that the male calcanei in this skeletal sample were more robust than the female calcanei, which may generate a predilection for this hyperostotic trait. Bilateral traits were found to occur symmetrically, on both sides of the body, in most cases. This shows that environmental disruptions during development and cerebral dominance (right- and left-handedness) had little or no effect on their expression. The skeletal sample displays a marked correlation between paired traits in the cranium and mandible.116 Asymmetry, however, is apparent in certain postcranial traits. The septal aperture occurs more frequently in the left humerus (4/14 scorable sides, or 28.57%; 25 sides were unscorable) than in the right humerus (2/23 scorable sides, or 8.70%; 16 sides were unscorable), a pattern that accords with the frequency of this trait in many groups.117 The peroneal tubercle also occurs slightly more frequently in the left calcaneus (6/9 scorable sides, or 66.67%; 30 sides were unscorable) than in the right one (7/12 scorable sides, or 58.33%; 27 sides were unscorable). However, since the number of calcanei that could be observed for this trait was small, it is uncertain whether this is a meaningful distribution. Certain nonmetric traits, mostly cranial, that occur with relatively low frequency in the skeletal sample occur with higher incidence in some multiple burials (Table 5.18). The 116. See Hauser and De Stefano 1989, p. 10. 117. Saunders 1978, pp. 88–92, 470, table 2.

292

THE CONDITION AND COMPOSITION OF THE HUMAN REMAINS

disproportionate occurrence of these traits could indicate that individuals interred in multiple burials belonged to the same genetic lineage. Some of these traits have been shown to possess a significant heritability,118 but the genetic basis of others remains unknown. None of these traits, however, seems to demonstrate sex- or age-dependency within the skeletal sample. The three multiple burials that show the highest coincidence of nonmetric features are NEG 67-001 (11 individuals), RB 76-002 (3 individuals), and HO 70-902 (10 individuals). Most noteworthy are the high prevalence of one rare trait, vastus notch, in NEG 67-001 and the high prevalence of two other unusual traits, mylohyoid arch and sacralized coccyx, in HO 70-902.119 This coincidence of features with a possible genetic basis strongly supports the hypothesis that multiple burials contained members of the same family. While the precise assocations between these individuals cannot be ascertained, the two most likely relationships are parent-child and siblings. The coincidence of traits in adult skeletons of the opposite sex but approximately the same age in certain multiple interments points to the possibility that brothers and sisters could be buried together. The coincidence of nonmetric traits does not suggest any significant patterns in the spatial distribution of burial.

118. Sjøvold 1984, pp. 236–243, table 4 (ossicle at asterion and multiple lesser palatine foramina). 119. Agenesis of the third molar (p.299) is another trait

shared by individuals in HO 70-902, but it is not a rare condition.

6

 TEETH AND ORAL HEALTH

T

he mouth is the orifice through which oxygen, food, and drink pass from the natural environment into the human body. It is also the site of initial digestion and mastication, a tool for the manipulation of materials, a central feature of the facial anatomy, and a part of the body closely linked to the perception of personal identity, in no small part due to its role in articulating language. Injuries to the head can alter the shape of the mouth and impair its functions. The condition of the teeth and contiguous soft tissues contributes significantly to an individual’s physical appearance and comfort. For all these reasons, anthropologists study ancient jaws and teeth to understand the subsistence, activities, and experience of people in a particular place and time.1 The dental and alveolar remains of Roman and Byzantine date from the Isthmus provide important evidence for the oral health of residents. In this chapter I first discuss the major factors affecting the status of the teeth, diet and oral hygiene, and I then address several major conditions observed in the teeth and jaws. The remains of Late Roman to Early Byzantine date appear healthy relative to other pre-modern groups. There were few instances of congenital defects and the few enamel defects resulted from malnutrition or illness in childhood. Dental wear advanced at a normal pace but even in old age seldom reached a severe level. The frequencies of calculus, caries, alveolar lesions, and antemortem tooth loss (AMTL) were all low to moderate. The final section of this chapter compiles the dental data from Roman and Byzantine skeletons at other sites in Greece, most importantly Corinth, which have been sufficiently published for comparison with the evidence from the Isthmus.

DIET AND ORAL HYGIENE Before examining the various traits of teeth and jaws from the Isthmus, it is important to consider the evidence for dietary and behavioral factors that might have contributed to their prevalence among residents. The degree and location of dental wear, the presence of enamel defects, caries, and pulpitis, and the absence of teeth can be caused or directly influenced by several conditions that cannot be recovered from the historical or archaeological record, such as the habitual use of teeth as tools, bruxism (the grinding of teeth), orofacial injury, poor nutrition, and childhood disease. The next section will address the possible etiologies of these dental conditions within the socioeconomic model for the local community and the reconstruction of its settlement presented in Chapter 3. Contemporary historical and archaeological evidence from the Greek world for two other major components of oral health, diet and hygiene, is also discussed. 1. Kelley and Larsen (1991), Hillson (1996, 2000), Lansjoen (1998), and Soames and Southam (1998) provide useful introductions to dental anthropology.

294

TEETH AND ORAL HEALTH

The inhabitants of the site of the Sanctuary and fortifications pursued a diverse economy that depended on local produce but also drew upon other sources in the region, including producers and probably vendors or traders at Kenchreai and Corinth.2 The main subsistence activities close to the settlement would have been husbandry, the cultivation of garden crops and cereal grains, and apiculture. Residents probably also acquired foodstuffs off-site through exchange, hunting, gathering, or fishing. These activities are attested by faunal remains from Late Roman and Early Byzantine deposits across the site, numerous terracotta beehives of Late Roman or Early Byzantine date from the Fortress, several rotary querns from Early Byzantine strata in the Bath, and Late Roman or Early Byzantine structures in the areas of the temenos, the Theater, and the Northeast Gate that might have served as pens and/or bins.3 These remains provide direct evidence for several elements in the diet of local residents from roughly the 5th through the 7th or 8th centuries: meat from cattle, pigs, sheep or goats, chickens, and fish, as well as other products such as milk, eggs, honey, and grains, probably wheat and/or barley. Local residents surely consumed a wider array of foods. Historical and archaeological sources for the daily diet of Byzantine Greeks suggest the possible breadth of the dietary regimen at the Isthmus.4 Charred plant traces from deposits dated to the 11th to 12th centuries in the Roman Stoa at Sparta represent the kinds of foods consumed by the Middle Byzantine residents, many of them agrarian products: cereal grains (wheat and barley), legumes (chickpea, pea, lentil, broad bean, bitter vetch), fruits (grape, fig, olive), nuts (hazel), and a rare turnip. The documentation of land use in southeastern Macedonian villages during the 14th century reveals that residents cultivated vines, grazed sheep and goats, kept bees, and fished. The villagers divided their production of various foods between fields, gardens, and orchards: wheat, barley, rye, vetch, millet, legumes in fields; cabbage, carrots, garlic, onion, melon, cucumber, and zucchini in gardens; and pear, fig, walnut, cherry, almond, and mulberry in orchards. A similar subsistence prevailed in the Late Byzantine village at Panakton, where residents farmed and grazed the slopes of Panakton and the Skourta Plain below. Recent excavations at the site have uncovered the remains of sheep-goat, cattle, pig, horse, deer, fish, various pulses, grape, and barley. Residents of the Laconian countryside during the Byzantine and Ottoman eras exploited a comparable range of cultivated foods and livestock.5 The rural economy of the Corinthia during this time was characterized by the same diversity of activities and produce. Byzantine accounts, Venetian documents, and Early Modern itineraries record that the main crop was wheat, though grapes were also harvested, the primary livestock were cattle and sheep, and the coasts were fished.6 The closest analogy for the dietary regimen of the residents of the Isthmus comes from the recent excavation of an Early Byzantine farmstead at Pyrgouthi in the central Berbati Valley. The site produced a wealth of faunal and floral remains representing the abundant exploitation of local resources through cultivation, gathering, hunting, grazing, and even exchange: wheat, barley, oats, 2. See pp. 116–117, 126, 141–142. 3. Archaeobotanical remains have not been recovered from deposits at the site. Seeds of Ficus carica found in the sacrificial pits of the Palaimonion demonstrate that figs were consumed on-site at least during the rituals of the Roman Sanctuary (Gebhard, Hemans, and Hayes 1998, pp. 431–432 [Pit A, upper layer], 443–444 [Pit C, all deposits]). Presumably residents during Late Roman and Byzantine times also ate wild figs they collected from the trees growing throughout the countryside. 4. Brothwell and Brothwell (1998) survey the varieties and uses of food in antiquity. Garvie-Lok (2001, pp. 33–66) broadly discusses diet in Roman, Byzantine, Frankish, and Ottoman

times. Kaplan (1992, pp. 25–46), Kislinger (1999), Dagron (2002, pp. 445–449), and Lefort (2002, pp. 248–252), concisely investigate the historical evidence for the alimentary regime during the Early to Middle Byzantine periods. 5. Hather, Peña-Chocarro, and Sidell 1992; Waywell and Wilkes 1997, pp. 425–427 (Sparta); Laiou-Thomadakis 1977, pp. 28–32 (Macedonia); Gerstel et al. 2003, pp. 222–223, nn. 143–144 (Panakton); Laconia I, pp. 382–390. It has been suggested that the Early to Middle Byzantine residents at Polystylon in Thrace also consumed a mixed agricultural diet of cereal and seafood (Agelarakis and Agelarakis 1989, p. 22; Agelarakis 1997, p. 303). 6. Kordosis 1981a, pp. 380–383.

DIET AND ORAL HYGIENE

295

lentils, broad beans, peas, Spanish vetchling, bitter vetch, dwarf chickling, carob, walnut, pomegranate, figs, grapes, olives, and sheep, goats, pigs, cattle, hare or rabbit, small birds, and fish.7 Although this evidence derives from separate sites and different periods, residents of the Greek countryside from antiquity until the present day have utilized many of the same basic natural resources under similar conditions of climate and terrain. The evidence for the Byzantine Greek diet indicates that the range of foodstuffs consumed by the residents at the Isthmus during the Late Roman to Early Byzantine periods was undoubtedly wider than the surviving remains indicate. It is best characterized as a mixed agricultural diet combining meats, fruits, vegetables, cereals, and dairy. The residents would have eaten both soft, sticky, oily foods (fleshy fruits and vegetables, meat, honey, dairy), and hard, rough, fibrous foods (raw plants, roots and stalks, nuts and seeds, milled grain, occasional fruit or vegetable casings, husks, or even bone). They would have also ingested grit through airborn mineral particles and from the ground stone with which they processed their food. Soft and sticky, oily or liquid, and abrasive consumables would have had different effects on the progression of dental wear and caries. Furthermore, local residents ate several foods containing starches and sugars, in particular cereal grains, fruits and vegetables, and honey, which tend to promote plaque, tooth decay, and eventual periodontal disease. Apart from what they consumed, the way the residents of the Isthmus cared for their teeth affected their oral health. It is difficult to establish habits of dental hygiene and practices of oral surgery within the local community as no evidence for these activities has been found. Certain invasive actions, however, can be ruled out with certainty, and preventive behaviors can be extrapolated from ethnographic analogy and historical testimony. The hypothetical reconstruction of hygienic behaviors and primitive dentistry reveals not only how residents might have prevented oral pathology but how they might have responded to chronic pain. The evidence for the treatment of teeth in Roman and Byzantine times comes primarily from technical writings by ancient doctors and the remains of dental instruments.8 Extensive testimony for contemporary practices is provided by Greek and Roman scientific authors of the Empire such as A. Cornelius Celsus, Pliny the Elder, Scribonius Largus, Pedanius Dioscorides, Galen, and Oribasius, and Early Byzantine physicians such as Aëtius of Amida, Alexander of Tralles, and Paul of Aegina. These authors were educated, wealthy men who lived in cities and wrote in a long literary and intellectual tradition. Their writings were compilations of existing knowledge, and, inasmuch as they served as practical manuals, were intended for the urban elite. They describe several common dental conditions and ailments, in particular calculus, periodontitis, caries, abscesses, and trauma. They recommend a diversity of remedies, including the ingestion or application of narcotic herbs to ease toothaches, filling cavities with various mineral and organic matter, installing dental appliances to stabilize or replace teeth, using dentifrices and rasps to smooth teeth and remove accretions, and using elevators or forceps to remove diseased or loosened teeth. Numerous examples of dental bridges, false teeth, forceps, drills, and files have been found in the graves of aristocrats and medical practitioners throughout the Mediterranean. These historical and archaeological sources do not provide evidence for daily life at the Isthmus. They attest to high standards of pre-modern hygiene, expensive and exotic 7. Hjohlman 2005, pp. 250–253; Mylona 2005; Sarpaki 2005. A basically comparable range of foods was consumed at Eleutherna on Crete during the Late Roman to Early Byzantine periods (Bourbou 2000, pp. 303–305, n. 34). 8. Hoffman-Axthelm (1981, pp. 60–81) surveys the GrecoRoman and Early Byzantine traditions of dentistry, with an emphasis on the written sources; Milne ([1907] 1976, pp. 136–

139) reviews testimonia for oral surgery in the Greek, Roman, and Byzantine authors. Jackson (1988, pp. 118–121) presents a general discussion of textual and archaeological sources for Roman dental health and dentistry; Grmek (1989, pp. 113– 118) addresses dentition in the Greek world, especially attrition and caries.

296

TEETH AND ORAL HEALTH

cleansers and anodynes, and elaborate surgeries. Furthermore, they record model procedures that in reality might have been rarely performed and costly habits that pertained foremost to a particular social class. No dental appliances or instruments have been found in the graves, and none of the teeth or jaws shows obvious traces of filing, drilling, or removal.9 But presumably local residents did not ignore their teeth, which would gradually deteriorate, creating an ugly appearance and painful suffering. In the absence of public health measures and widespread hygienic conventions, dental care was a private concern that must have varied between communities and even families in the Greek countryside. People in most cultures possess basic strategies for cleaning debris from their mouths to assuage the irritation in the tongue, gums, and cheeks. Rural Corinthians might have used toothpicks, chewing sticks, or oral rinses. Moreover, they would have responded to the pain of toothache through popular remedies, such as the application of medicinal plants collected in the area.10 The residents would not have enjoyed the regular services of a dentist, but they might have had occasional access to medical expertise. A doctor’s grave containing iron dental forceps and other instruments dating to the late 3rd century was discovered at Kallion in Phocis, hardly a major urban center during the Roman era.11 Doctors traveled in the provinces and certainly many passed over the Isthmus of Corinth. Medical personnel would have accompanied any sizable expeditionary force that might have occupied the Isthmian fortifications in late antiquity but would have left the site with the troops. A large city such as Corinth would have been the home of doctors with knowledge of dentistry, though inhabitants of the hinterland probably did not benefit from their practice. The residents of the Isthmus must have treated their teeth to maintain function and comfort; perhaps they even cleaned their teeth to appear attractive. But in the end they possessed neither the resources nor the expertise of contemporary medical writers and their urban readers for advanced hygiene, not to mention surgery. As for most inhabitants of the ancient Mediterranean world, the sensations of loose or lost teeth, dull or flat cusps, sore gums, and frequent toothaches—sensations unfamiliar to many modern people with access to professional dental care—were not uncommon in the community at the Isthmus. The fact that many rural Corinthians pursued vigorous activities for long years as adults suggests that they accepted these oral conditions as a normal part of life. Whether they considered them deleterious to personal well-being and self-presentation, as would people in modern industrial societies, cannot be known.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES This section addresses the conditions observed in the teeth and jaws from the Isthmus in sequence from development through wear and eventual deterioration. These conditions include abnormal variation in form, location, and number, enamel hypoplasia, dental wear and fractures, the accumulation of plaque and calculus, caries, periodontitis, periapical lesions, and AMTL. All erupted deciduous and permanent dentition in near or full occlusion were examined, but pathological conditions were found only in the adult teeth. The study 9. It is difficult to distinguish purposeful extraction from AMTL. The forceful removal of teeth typically fractures the alveolar crest, but minute damage of this sort is often masked by healing before death and postmortem alterations. 10. See Mays (1998, pp. 152–153) on strategies for oral hygiene among recent tribal peoples. One of the most common remedies for toothaches in the Greek and Roman medical writings was henbane (Hyoscyamus), a member of the nightshade family that grows throughout the Aegean region. The

Late Roman physician Oribasius of Pergamum wrote that the drug ἐπιθύμον (Cuscuta epithymum), a parasitic plant on thyme, grew at Athens and on the Isthmus (Coll. med. 7.26.30). Ramoutsaki, Dimitriou, and Kalmanti (2002) summarize the historical sources for analgesic measures in Late Roman and Byzantine times, particularly as applied to children. 11. Petrakos 1977, pp. 375–378, fig. 319:α. Künzl (1983, pp. 40–44, fig. 11.1) lists other examples of medical instruments from archaeological sites in Greece.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

297

required that the teeth be cleaned, that fragmentary jaws be reconstructed, and that teeth recovered loose from graves be associated with surviving alveoli.12 Each dental arch was examined macroscopically using a 10x hand lens, all noteworthy features were described according to standard techniques, and dentitions were checked again to ensure the repeatability of identifications.13 The results of this examination are presented here in discussion, data tables, and images, including occlusal views of all surviving dental arcades at a scale of 1:2 (Figs. 6.1–6.34) and several details of dental pathology at varying scales (Figs. 6.35–6.46). In order to understand the significance of the observed conditions for the oral health of the Roman and Byzantine residents of the Isthmus, the survival of their teeth and jaws in the archaeological record must be considered. The oral remains are not perfectly representative of the ancient residents because all teeth and alveoli have not survived in all skeletons. While one or both jaws of 20 skeletons were completely lost due either to natural destruction or to selective recovery,14 the teeth and/or alveolar processes of 55 skeletons survive, either in a poor or intact state depending on the design of the grave and its exposure to drainage.15 The adjusted total survival rate for teeth is 43.21%, with molars and premolars surviving more frequently than canines and incisors. Nonetheless, the total number of erupted teeth that did survive for examination (725) is considerable (Table 5.2). A majority of the alveoli (1,092/1,784, or 61.21%) also survive. Moreover, the preservation rates of all mandibles (69.1%) and maxillae (70.6%) are high, and the surviving bones are relatively well preserved (Figs. 5.4, 5.5).16 Although the teeth and jaws do not fully represent the entire skeletal sample, they do comprise a substantial body of evidence for evaluating general trends in oral health among local residents. Due to differential survival by tooth class, the assessments of conditions in cheek teeth will be somewhat more reliable than those in front teeth. One important means of assessing the oral health of local residents is to determine the frequency of different conditions in the surviving remains. Anthropologists use several calculations to quantify the frequency of dental traits, each of which presents the data in a different manner.17 The first is the ratio of individuals affected to individuals examined. The second is the average number of traits per individual. The difficulty with both of these methods is that they assume that all teeth and alveoli in the sample have survived, which is not the case at the Isthmus nor most other archaeological sites. When these methods are applied to a skeletal sample with incomplete dentitions, the ratio and the average will be underestimates. In a sample with a total dental survival rate of less than 50%, such as the Isthmus sample, the degree of underestimation would be significant. The third calculation is the ratio of all teeth or alveoli affected to all teeth or alveoli examined. This calculation of true prevalence is unaffected by missing data. The present study uses all three to quantify the frequencies of traits in the dental sample. This permits the comparison of the skeletons at the Isthmus with other groups whose dental remains have been published in a variety of forms. 12. The association of loose teeth with alveoli was achieved only when a tooth of a certain class fit snugly into a socket in the proper region of the dental arcade and when its form, including its size, wear, and interproximal facets, aligned closely with its surroundings. 13. The teeth from the Isthmus were not studied radiographically or microscopically. Apart from measuring the location and size of certain defects and lesions, dental measurements were not collected. 14. NEG 67-001A (both absent), J (no maxilla), 67-003A (both absent), B (no mandible), D (both absent), E (both absent), 69-001A (no maxilla), C (no maxilla), 69-008A (both absent), T13 54-001 (both absent), T14 69-701A (both absent), DEC 69-901A (no maxilla), B (no maxilla), 69-902 (both absent), HO 70-902A (both absent), B (no mandible),

G (both absent), H (no maxilla), PAL 56-001 (both absent), TC 60-001A (no mandible). 15. One subadult skeleton (NEG 69-005A) preserved teeth but no alveolar processes. Two adult skeletons (NEG 69-103B, 69-001A) preserved intact alveolar processes but no teeth, many of which had been had lost before death. Therefore, the number of skeletons with surviving teeth is 54 (40 adults); the number of skeletons with surviving alveoli is 55 (42 adults). 16. The preservation of the alveolar regions of mandibles and maxillae was not calculated independently, but overall they seemed to be well preserved when the mandible or maxilla was present. 17. Lukacs 1989, pp. 271–273; Roberts and Manchester 1995, pp. 63–64; Mays 1998, pp. 147–148. Waldron (1994, pp. 42–60) discusses the measurement of disease frequency.

298

TEETH AND ORAL HEALTH

The demographic profile of the skeletal sample at the Isthmus must also be taken into account when interpreting the frequency of dental conditions. The distribution of traits by age and sex among residents can be studied because teeth and jaws from men, women, and children have survived for examination. Since conditions such as attrition and caries have the tendency to worsen with advancing age, old adults will typically exhibit more lesions than adolescents or young adults. Therefore, the overall age distribution of the skeletal sample has an influence on the prevalence of conditions. Of the total number of skeletons with preserved teeth from the site (54), a majority can be aged as adult (40). Few individuals represented at the Isthmus survived into their sixth decade. Within the Late Roman to Early Byzantine sample, the numbers of men and women who died at particular ages were generally similar (Fig. 5.7). This means that the frequencies of certain infectious and degenerative conditions, particularly severe wear, inflammatory lesions, and AMTL, can be reasonably compared by sex. Significant discrepancies in the prevalence of traits among men and women, however, may arise from differences in diet and behavior between the sexes. Dental conditions also respond to developments in subsistence, living conditions, and social customs over time. The majority of the skeletal sample from the Isthmus comprises a chronologically and historically coherent assemblage representing the Late Roman and Early Byzantine periods. It will be interesting to compare dental conditions in two subsets of the skeletal sample, Late Roman individuals (40 with observable remains) and Early Byzantine individuals (11 with observable remains).18 However, the later subset is so small that any patterns emerging from this comparison of dental conditions by period must be treated with great caution. The five earliest (T2 68-003A, B, DEC 69-901A, B, WF 62-001A) and the two latest skeletons (LOU 69-801A, TC 60-001A), representing distinct periods in the site’s history, will not be used in comparisons by period. Observations on these individuals’ oral remains, however, will be included in general assessments of the total skeletal sample.

Variation in Dental Form, Location, and Number As an individual grows from infancy to adulthood, the form and size of the dental arcade changes considerably. While human teeth tend to follow a similar pattern of formation, they also display variability in shape, location, and number. Since these morphological traits are to a degree controlled by genetic factors, they can provide an important basis for characterizing a skeletal sample and discriminating between samples.19 Although dental morphology was not systematically recorded in the skeletal sample from the Isthmus,20 salient anomalies in shape and situation were noted (Table 6.1). Three morphological variants were observed. First, M2R of T14 69-003A displays a foramen caecum, a distinct, conical pit at the apex of the buccal groove that occurs most commonly in the mandibular molars. Second, M1L of NEG 69-005A displays Carabelli’s trait, an accessory cusp on the base of the mesiolingual cusp that occurs most commonly in the upper molars. In this case, the trait is a large, free cusp.21 Third, C1R of NEG 69-007B is split into two rootlets, one of which is small and conical and extends from the lingual aspect. While most teeth in the sample occupied the normal 18. Late Roman subset (10 males and 14 females): NEG 69-103A, B, 69-004A–E, 67-001B–K, 67-003B, C, 69-005A–C, 69-001A–D, RB 76-002A–C, HO 70-901A–C, 70-902-F, H–J. Early Byzantine subset (six males and two females): NEG 69007A, B, 69-009A, 69-010A, T2 68-002A, 68-006A, T14 67-002A, 67-004A, 69-991A, 69-002A, 69-003A. 19. Hillson 1996, pp. 100–102; see, e.g., Corrucini (1991)

on malocclusion. 20. Turner, Nichol, and Scott (1991) outline a detailed procedure for recording dental variation; Buikstra and Ubelaker (1994, pp. 63–68) give an abbreviated version. 21. Scored 7 in the Arizona State University system (Turner, Nichol, and Scott 1991, p. 19; see also Hillson 1996, p. 92, fig. 3.7).

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

299

TABLE 6.1. VARIATION IN DENTAL FORM AND LOCATION Burial No.

Affected Tooth

Variant

NEG 67-001B

I1L, I2R

NEG 69-001D

I1R

Misalignment from crowding

NEG 69-004B

I2L

Distolingual rotation

NEG 69-005A

ML

Carabelli’s trait (large, free cusp)

NEG 69-005B

I2R

Misalignment from crowding

NEG 69-007B

C1R

Two rootlets

T2 68-003A

C1L

Distolingual rotation

T14 67-002A

IL

Misalignment from crowding

T14 67-004A

C1R P3R, P3L

T14 69-003A

M2R

Foramen caecum

WF 62-001A

I1R

Misalignment from crowding

1

1

Distolingual rotation

Mesiolingual rotation Mesiolingual rotation

alveolar site, a few locational variants were observed. Seven anterior and cheek teeth exhibit slight rotation from a straight alignment with the curvature of the dental arch (Figs. 6.2, 6.6, 6.15, 6.17), while four anterior teeth exhibit misalignment from crowding (Figs. 6.10, 6.12, 6.16, 6.33).22 Interarch occlusion was not recorded in the fragmentary skeletal sample; the best fit between the upper and lower teeth may not accurately represent occlusion in life.23 Variations in the number of teeth were systematically recorded. No instances of supernumerary teeth (hyperdontia) were observed but there were several instances of agenesis, the congenital absence of teeth (hypodontia), all third molars. The third molar is the most frequently absent tooth in ancient skeletons as in modern populations. Agenesis was identified when both a tooth socket and wear facets on the distal interproximal crown of the second molar were absent after the normal age of eruption of the third molar. In the full sample, 13 skeletons display agenesis of the third molar,24 comprising 35.14% (13/37) of all adult skeletons with observable third molar alveoli. In all, 24 third molars are congenitally absent, or 2.65% (24/905) of all preserved tooth sockets in adults and 22.02% (24/109) of all preserved third molar sockets in adults (Table 5.2). Agenesis of third molars is more common in mandibular dentition (15) than in maxillary (9). Within the Late Roman to Early Byzantine sample, 37.50% (12/32) of observable adult skeletons display agenesis of the third molar, while 3.01% (23/763) of all preserved tooth sockets and 23.96% (23/96) of all preserved third molar sockets in adults are congenitally absent. Dental agenesis has long been identified as an inherited trait, and its frequency may provide an index for genetic relationships within and between groups.25 Therefore, the occurrence of agenesis in individuals occupying the same graves at the Isthmus is noteworthy.26 This distribution supports the theory that multiple burials contained individuals from the same lineage, such as parents and children or adult siblings.27 22. Figure 6.4 gives the distorted impression that I2L was misaligned in life. Rather, the tooth was incorrectly positioned for the photograph. 23. See Hillson 1996, pp. 107–108. However, possible malocclusion was noted in two individuals (NEG 69-004B, T14 67002A) during the close examination of dental fractures; see pp. 321, 323. 24. NEG 69-103A, 69-004B, 69-004E, 67-001D, 69-005B, C,

69-001D, T14 69-002A, 69-003A, LOU 69-801A, HO 70-901A, 70-901C, 70-902H. 25. E.g., Berry 1978, p. 94; Brook 1984; Hillson 1996, pp. 100, 114. 26. NEG 69-004B (F, late 40s) and E (F, 35–44 years); 69-005B (F, 20–22 years) and C (M, 25–34 years); HO 70-901A (F, 35– 44 years) and C (M?, adult). 27. See pp. 220–221, 291–292, Table 5.18.

300

TEETH AND ORAL HEALTH

Figure 6.2. NEG 69-004B

(note AMTL). Scale 1:2

Figure 6.3. NEG 69-004C (note wear and caries). Scale 1:2

Figure 6.4. NEG 69-004D.

Figure 6.5. NEG 69-004E.

Scale 1:2

Scale 1:2

Figure 6.1. NEG 69-103B

(note AMTL). Scale 1:2

Figure 6.8. NEG 67-001D

(note AMTL). Scale 1:2

Figure 6.6. NEG 67-001B. Scale 1:2

Figure 6.7. NEG 67-001C (note wear, alveolar defects, and AMTL). Scale 1:2 Figure 6.9. NEG 67-003B (note wear and AMTL). Scale 1:2

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

Figure 6.11. NEG 69-005C (note wear, alveolar defects, and AMTL). Scale 1:2

Figure 6.12. NEG 69-001D.

Figure 6.14. T2 68-002A (note caries and AMTL). Scale 1:2

Figure 6.15. T2 68-003A (note caries and AMTL). Scale 1:2

Figure 6.16. T14 67-002A (note

Figure 6.17. T14 67-004A.

Figure 6.18. T14 69-002A.

wear, fractures, caries, alveolar defects, and AMTL). Scale 1:2

Scale 1:2

Scale 1:2

Figure 6.10. NEG 69-005B

(note caries). Scale 1:2

Figure 6.13. NEG 69-007A

(note alveolar defects and AMTL). Scale 1:2

Scale 1:2

301

302

TEETH AND ORAL HEALTH

Figure 6.21. DEC 69-901A. Scale 1:2

Figure 6.19. T14 69-003A

(note wear and caries). Scale 1:2

Figure 6.20. LOU 69-801A. Scale 1:2

Figure 6.22. DEC 69-901B. Scale 1:2

Figure 6.23. RB 76-002B

Figure 6.24. HO 70-901A.

(note AMTL). Scale 1:2

Scale 1:2

Figure 6.25. HO 70-901B (note alveolar defect and AMTL). Scale 1:2

Figure 6.26. HO 70-901C (note wear, alveolar defects, and AMTL). Scale 1:2

Figure 6.27. HO 70-902D

Figure 6.28. HO 70-902E

(note alveolar defects and AMTL). Scale 1:2

(note alveolar defects and AMTL). Scale 1:2

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

Figure 6.29. HO 70-902F. Scale 1:2

Figure 6.30. HO 70-902H.

Figure 6.31. HO 70-902I

Figure 6.32. HO 70-902J

Scale 1:2

(note caries). Scale 1:2

(note caries). Scale 1:2

Figure 6.33. WF 62-001A (note caries and alveolar defects). Scale 1:2

Figure 6.34. TC 60-001A

Figure 6.35. HO 70-902?,

(note alveolar defects and AMTL). Scale 1:2

linear enamel hypoplasia in C1L, labial view

Figure 6.36. NEG 69-004B, fractured upper incisors

and mandibular AMTL, labial view

303

304

TEETH AND ORAL HEALTH

Figure 6.37. NEG 69-004C, calculus, periodontitis, and acute

abscess in periapical granuloma in P3R–M3R, buccal view

Figure 6.38. NEG 67-001C, caries and periodontitis

in P3R–M3R, buccal view

Figure 6.39. T14 67-002A, caries and periodontitis

in M1R–M3R, buccal view

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

305

Figure 6.40. WF 62-001A, gross

caries in M1R, occlusal view

Figure 6.41. NEG 67-001C, caries, periodontitis, acute abscesses

in periapical granulomata and periodontal cyst, and AMTL in upper and lower anterior teeth, labial view

Figure 6.42. NEG 69-007A, periodontitis, acute abscess in

periodontal cyst, and AMTL in P4R–M3R, buccal view

Figure 6.43. NEG 69-007A, periodontitis, acute abscess in periapical granuloma, and AMTL in C1L–M3L, buccal view

306

TEETH AND ORAL HEALTH

Figure 6.44. T14 67-002A, fracture, acute abscess in perapical

granuloma, and AMTL in P4L–M3L, buccal view

Figure 6.45. HO 70-901C, caries, periodontitis, acute abscesses in

periodontal cysts, and AMTL in P4R–M1R, buccal view

Figure 6.46. WF 62-001A, large periapical cavities around

roots of M1L draining into maxillary sinus, superior view

Enamel Hypoplasia The nature and prevalence of enamel hypoplasia in a skeletal sample is an index of a group’s health during childhood. In this condition, a flaw develops in the otherwise smooth surface of the dental crown when the ameloblasts fail to secrete the organic matrix that miner-

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

307

alizes incrementally to form enamel. The hypoplastic lesion is a transverse narrowing in enamel thickness that occurs during the maturation of permanent tooth germs from birth to 7 years, though defects take many forms.28 One of the commonest varieties is distinct, horizontal bands on the labial aspect of the anterior dentition, but similar grooves can also occur in the molars, where they are often closer to the CEJ.29 Systemic metabolic stress, congenital defects, or localized trauma can interrupt dental development and cause enamel defects. Conditions resulting in amelogenetic arrest include, among others, severe viral and bacterial infection, intestinal parasites, rickets, scurvy, sicknesses with high fever, measles, allergic reactions, whooping cough, pneumonia, vitamin deficits (especially A and D), general malnutrition, anemia, and chronic diarrhea.30 Moreover, a high incidence of enamel hypoplasia in a sample correlates with agricultural subsistence and sedentary settlement, which are often accompanied by seasonal deficiencies in nutrients and the transmission of illness among children.31 Thus, enamel defects in permanent dentition preserve a record of developmental insult under particular environmental conditions, even if the precise cause of every defect cannot be determined. Since dental maturation follows a predictable timetable, the ages of growth disruption in children can be traced by measuring the location of hypoplasias in adult teeth from a skeletal sample. The widely used Goodman method, which has been tested on prehistoric skeletons from Illinois, correlates the CEJ-lesion distance with half-year periods in enamel formation over the first seven years of life, not including the third molar.32 This approach has weaknesses, including its assumption of a constant growth rate for the formation of the crown surface and its use of a single developmental standard, whereas development varies between individuals and populations.33 In the absence of better standards, it remains worthwhile to record CEJ-lesion distances and to apply the Goodman method, provided that the researcher bears in mind that the resulting estimates of age of insult carry an error of unknown magnitude.34 A relative chronology of age-specific illnesses or episodes of stress in an individual’s childhood can thus be reconstructed. The examination of enamel hypoplasia in all teeth recovered from the Isthmus followed standard procedures.35 The only form of enamel hypoplasia observed was linear banding, which occurred in 79 permanent teeth in 16 individuals (Tables 6.2, 6.3; Fig. 6.35). A higher frequency of anterior teeth (20.59%) than posterior teeth are hypoplastic (10.92%). This is a typical distribution by tooth class that reflects not only different times of development but also variable susceptibility among the dental crowns.36 Within the Late Roman to Early Byzantine sample, 14.46% of the observable teeth (73/505) display hypoplasia.37 But this is 28. Goodman and Rose 1990, pp. 65–70; Hillson 1996, pp. 165–172, table 6.2; Soames and Southam 1998, pp. 7–13. 29. Goodman and Armelagos (1985) discuss the distribution of enamel hypoplasia within the dentition. 30. Cutress and Suckling 1982; Pindborg 1982; Rose, Condon, and Goodman 1985, pp. 284–285. Certain enamel defects, known as amelogenesis imperfecta, are heritable, but these are very rare and none was observed in skeletons from the Isthmus. 31. Roberts and Manchester 1995, pp. 59–60; Hillson 1996, p. 176. 32. Goodman, Armelagos, and Rose 1980, p. 519, fig. 1, modified from Swärdstedt 1966, which was based on the developmental standards of M. Massler, I. Schour, and H. G. Poncher. Goodman and Rose (1990, pp. 97–99) summarize alternative methods for converting defect locations to developmental ages. 33. Goodman and Rose 1991, pp. 287–289; Hillson 1996,

pp. 172–174. 34. Skinner and Goodman 1992, p. 165; Buikstra and Ubelaker 1994, pp. 56–57; Larsen 1997, p. 48; Goodman and Martin 2002, pp. 24–25. 35. Goodman and Rose 1990, pp. 91–96; Buikstra and Ubelaker 1994, pp. 56–58. The study of the dental remains from the Isthmus employed a wood pick to test the texture of teeth and thus distinguish true lesions from pronounced perikymata. Only lesions visible under good lighting without magnification were recorded. Their location was measured using dial calipers as the distance from the CEJ to the midpoint of the hypoplasia. 36. Goodman and Armelagos 1985. 37. An unknown number of enamel defects were erased from the dental record through attrition or fractures in 11 teeth (see Skinner and Goodman 1992, p. 164; Katzenberg, Herring, and Saunders 1996, p. 185).

308

TEETH AND ORAL HEALTH

TABLE 6.2. INSTANCES OF LINEAR ENAMEL HYPOPLASIA Burial No.

NEG 67-003B

NEG 69-004B

Hypoplastic Tooth

CEJ-Lesion Distance (mm)

Correlated Ages1

I1R

3.4

2.5

I2R

3.8

3.0

2

IL

4.3

3.0

1

CR

4.5

3.0

C1L

4.6

3.0

I1R

1.3, 3.2

4.0, 3.0

IL

4.5

2.5

2

Age of Disruptions: Positive

Age of Disruptions: Negative

3.0–3.5

0–3.0, 3.5–6.0

3.0–3.5

0–3.0, 3.5–6.0

C1R

3.9

3.0

NEG 69-004C

P4R

1.0

6.0

–

0–6.0

NEG 69-004D

M2R

1.8

5.5

–

0–7.0

4

PL

1.5

5.0

M2R

1.6

6.0

–

0–6.0

P3R

2.7

4.5

M2L

2.8

5.5

I1L

3.9

1.5 1.5–2.0

0–1.5, 2.0–7.0

1.5–2.0, 2.0–2.5, 3.0–3.5, 3.5–4.0, 4.0–4.5, 4.5–5.0

0–1.5, 2.5–3.0, 5.0–7.0

NEG 69-004E

NEG 69-005A

NEG 69-005B

I2R

4.0

2.0

I2L

4.9

1.5

P3R

3.8

3.0

P3L

3.3

3.5

M1L

1.8

2.5

I1R

0.8, 5.0

4.0, 2.0

I2R

1.9

3.5

1

CR

3.2, 6.5

4.0, 2.0

1

CL

3.2, 6.7

4.0, 2.0

P3R

2.0

4.5

P3L

3.4

4.0

4

PR

3.5

4.0

4

PL

3.4

4.5

M1R

2.8

2.0

M2R

3.3

5.0

I1R

3.1

2.0

I1L

3.0

2.0

I2R

3.4

2.5

I2L

3.0

2.0

C1R

3.8, 7.3

3.5, 1.5

C1L

4.0, 7.7

3.5, 1.5

P3R

3.8

3.0

P4R

3.8

3.0

P4L

3.7

4.0

M1R

4.2

1.5

M1L

4.2

1.5

M2R

4.1

4.5

M2L

4.1

4.5

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

309

TABLE 6.2 (CONT.) Burial No. NEG 69-005C T2 68-003A

T14 67-002A

T14 67-004A

Correlated Ages1

C1R

4.8

3.0

C1L

5.6

2.4

C1R

2.0, 3.6, 6.5

4.5, 3.5, 2.0

1

CR

2.2

4.5

1

CL

2.7

4.0

I1R

3.0

2.0

I1L

3.9

2.0

I2L

3.6

2.0

C1R

2.9

4.0

C1L

3.8

3.5

P3R

2.5

4.0

M3R

2.1

–

1

CR

2.4

4.5

C1L

1.5

5.0

P3R

1.5

5.0

3

PL

1.7

5.0

4

PR

1.3

5.5

P4L

1.7

5.0

M2L

1.1

6.5

P3R

1.0, 3.3

5.0, 3.5

P3L

1.1

5.0

P4R

0.9, 2.1

6.0, 5.0

P4L

1.3

6.0

M1R

1.3

3.0

Age of Disruptions: Positive

Age of Disruptions: Negative

–

0–6.0

–

0–6.0

2.0–2.5, 4.0–4.5

0–2.0, 2.5–4.0, 4.5–6.0

5.0–5.5, 6.0–6.5

0–5.0, 5.5–6.0, 6.5–7.0

–

0–7.0

3.5–4.0

0–3.5, 4.0–7.0

M1L

1.4

2.5

1.0

6.0

M2L

1.3

6.0

1

IR

5.0

2.0

1

IL

7.0

1.5

C1R

3.8

3.5

C1L

3.5

3.5

HO 70-902I

P4R

1.0, 2.1

6.0, 5.0

–

0–7.0

HO 70-902?

1

CL

3.0, 5.1, 6.3

4.0, 3.0, 2.0

–

–

P4L

2.5

5.0

C1R

6.4

2.0

–

0–7.0

P3L

4.9

2.5

LOU 69-801A

2

WF 62-001A

2

CEJ-Lesion Distance (mm)

M2R

T14 69-003A

1

Hypoplastic Tooth

The correlated ages are determined using the graph at Goodman, Armelagos, and Rose 1980, p. 520, fig. 1. This tooth represents some individual in this multiple burial other than 70-902I.

a deceptive calculation of the prevalence of defects because all dentitions in which hypoplastic teeth were observed are incomplete, and several had lost many teeth either antemortem or postmortem.38 Therefore, the frequency of hypoplasia was calculated only in C1L, or C1R 38. See Lansjoen 1998, p. 407.

310

TEETH AND ORAL HEALTH

TABLE 6.3. FREQUENCY OF ENAMEL HYPOPLASIA BY PERIOD AND SEX Individuals with Scorable Maxillary Canines (n)

Individuals with Hypoplastic Maxillary Canines (n)

Prevalence of Individuals with Hypoplastic Maxillary Canines (%)

Late Roman Males

4

2

50.00

Late Roman Females

9

1

11.11

15

4

26.67

Early Byzantine Males

5

3

60.00

Early Byzantine Females

2

0

0.00

7

3

42.86

9

5

55.56

11

1

9.09

22

7

31.82

Period and Sex

Total Late Roman

Total Early Byzantine Late Roman–Early Byzantine Males Late Roman–Early Byzantine Females Total Late Roman–Early Byzantine

if the left antimere was absent. Among Late Roman to Early Byzantine skeletons with surviving maxillary canines, roughly one third display enamel defects (7/22, or 31.82%). Within the Late Roman sample, which is quite small, 14.11% of the observable teeth (47/333) display hypoplasia, or 33.33% (5/15) of individuals with surviving maxillary canines. The low survival rate of teeth severely limited analysis of the distribution of enamel hypoplasia by period and sex (Table 6.3). Enamel defects were observed in both males and females of the Late Roman period, but only in males of the Early Byzantine period. The sample sizes for each period when divided by sex are very small, but the frequencies of defects by sex for both periods combined still show that a much greater proportion of males (55.56%) than females (9.09%) possess hypoplastic teeth. Notwithstanding the unequal size of the sample subsets by period, the prevalence of hypoplastic maxillary canines is higher among the Early Byzantine skeletons (42.86%) than among the Late Roman skeletons (26.67%). Correlation of the location of enamel hypoplasias with dental age can provide a better sense of developmental disturbance in life than the raw scores of frequency. The Goodman method was applied to the teeth from the Isthmus to determine the sequence of enamel defects. The first step was to correlate CEJ-lesion distance with developmental age in halfyears. Then each half-year period was identified as positive evidence for growth disruption, negative evidence for growth disruption, or insufficient evidence for growth disruption. A half-year period was classed positive if two or more teeth displayed a defect that developed at that time. A half-year period was classed negative if at least four sections of enamel could be checked but no more than one of these was hypoplastic. A half-year period was classed insufficient evidence if four sections of enamel could not be checked and less than two were hypoplastic.39 The application of this method identified six individuals of Late Roman to Early Byzantine date with positive evidence for 13 discrete disruptions in their growth (Table 6.2). Isolated enamel hypoplasias were observed in the other ten individuals, but these did not show a pattern of occurrence that could be positively identified with disturbance at a particular age. A majority of permanent teeth from the sample provided negative evidence for developmental disturbances. Among dentitions that provide positive evidence for the age of growth disruption, most episodes (10) occurred between the ages of 1.5 and 4.5, centering 39. Goodman, Armelagos, and Rose 1980, p. 519, fig. 1; see also Rose, Condon, and Goodman 1985, p. 293, fig. 9.4. Since the dentitions had not survived in a complete state, all teeth

were examined rather than a selection (see Goodman, Armelagos, and Rose 1980, p. 526; Goodman and Rose 1990, p. 91).

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

311

Number of Growth Disruptions

4 3 2 1 0 0–0.5 0.5–1 1–1.5 1.5–2 2–2.5 2.5–3 3–3.5 3.5–4 4–4.5 4.5–5 5–5.5 5.5–6 6–6.5 6.5–7 Half-year Periods from Birth to 7 Years Figure 6.47. Distribution of ages of growth disruptions causing enamel hypoplasia

around the middle to late third year, but a few (three) occurred between the ages of 4.5 and 6.5 (Fig. 6.47). No disturbances are indicated before 1.5 years. It should, however, be noted that severely worn and deeply chipped teeth would not preserve defects located far from the CEJ and close to the occlusal margin, those developing earliest in life. Several disturbances could have caused enamel hypoplasia among local residents during the Late Roman to Early Byzantine periods. There is no positive correlation between hypoplasia and other pathological conditions, particularly cribra orbitalia, which can result from the same conditions that cause enamel defects.40 While dental enamel formation is highly sensitive to stress, hypoplasia is a nonspecific index of metabolic status.41 Among the commonest causes of childhood illness and death in pre-modern or developing societies are infectious disease and malnutrition. Dietary deficiencies, especially in protein, reduce the body’s resistance to infection, while systemic infection and diarrhea raise the physiological need for nutrients, exacerbating a deficit that was already serious at the onset of the disease. Similar stresses might well have afflicted children living in the Corinthian countryside during Roman and Byzantine times. Presumably many children died from these ailments at the Isthmus within their first four years, the age at death for 14 individuals recovered from the site (Fig. 5.6). Some children, however, survived multiple episodes of metabolic arrest that generated an unusually high number of dental defects. The nature of subsistence and settlement at the Isthmus from the Late Roman to Early Byzantine periods was conducive to cycles of malnutrition and infection. The diet would have involved protein-poor foods such as cereals, and habitation was close and sedentary. Several studies have explained diachronic increases in prevalences of enamel hypoplasia among prehistoric peoples from both Old and New World sites as a consequence of either the shift from hunting-gathering to agriculture or the advancement of agricultural exploitation.42 The proportion of individuals with enamel hypoplasia ranges from 40% to over 90% in many ancient and modern groups subjected to poor nutrition and high infection.43 The proportion of afflicted individuals in the skeletal sample from the Isthmus (31.82%) falls somewhat below this range, but the actual proportion might well have been higher than is 40. See pp. 412–413. 41. Goodman and Rose 1990, pp. 101–102; 1991, pp. 284– 285, 290; Skinner and Goodman 1992, p. 164; Hillson 1996, p. 177. 42. Sciulli (1978) and Goodman, Armelagos, and Rose (1980) present important studies of prehistoric groups in

Ohio and Illinois. Angel (1984) and Smith, Bar-Yosef, and Sillen (1984) observe similar trends in the eastern Mediterranean, including the Aegean and the Near East. 43. Goodman and Rose 1991, pp. 282–283; Skinner and Goodman 1992, p. 161, table 2; Goodman and Martin 2002, pp. 25–26.

312

TEETH AND ORAL HEALTH

indicated by the surviving teeth. Other skeletal samples from sedentary communities that depended at least in part on an agricultural economy have demonstrated a prevalence of individuals with hypoplastic teeth in the 35%–40% range.44 If the frequency of enamel hypoplasia in the limited dental sample from the Isthmus does reflect juvenile health, it is noteworthy that the prevalence of afflicted individuals and enamel defects increases from the Late Roman to Early Byzantine periods. Such a shift may indicate agricultural intensification accompanied by reduced alimentary diversity over time. This theory gains archaeological support from the discovery of rotary querns in habitation contexts of Early Byzantine date. The appearance of ground stone, including imported querns, points to an increase in cereal processing during the late 6th to 7th centuries. The meaning of the apparent disparity in hypoplasia frequencies between men and women is unknown. If it does not arise from preservational bias, it may reflect a differential sensitivity to stress or dietary differences in boys and girls. The ages of disruptions calculated by the Goodman method provide a general chronology of illness among these children. The clustering of ages for enamel defects in the range 1.5–4.5 years corresponds with the range 2–4 years observed in other prehistoric and historic groups.45 Recent research has emphasized that the apparent concentration in this age period arises at least in part from the fact that the third year is the time of peak enamel deposition, and therefore most observed hypoplasias will occur around that time.46 However, the range of ages of disruption at the Isthmus is wide (1.5–6.5), and the apparent peak in the middle to late third year is minimal. This observation points to a relatively uniform level of postnatal stress over the first six years of life. Indeed, it is over this span of life, but especially in the first four years, that most subadults at the Isthmus died. One age-specific stress that has been widely discussed in conjunction with the timing of enamel defects is weaning.47 During the transition from breast-feeding to eating food, a child becomes more susceptible to environmental pathogens, particularly infection, because the immunological components in breast milk are no longer available. In many cultures, the age of weaning falls somewhere between the first and third years, though in antiquity it was generally later than in modern times. The major Greek medical writers of the Roman and Early Byzantine periods, including Oribasius, Aëtius, and Paul of Aegina, recommended the cessation of breast-feeding around the age of two.48 These technical works, however, do not attest to the actual variation in weaning practices from place to place and household to household. For instance, analysis of stable nitrogen and carbon isotopes in the subadult bones of the Late Roman inhabitants of the Dakleh Oasis in Egypt, together with study of archaeological and documentary evidence, has shown that infants were fed animal milk at around six months but weaning was not done until the third year.49 Many studies have associated the modal age of growth disruptions derived from enamel hypoplasias with the age of weaning. However, a direct relationship between the two cannot necessarily be assumed, in part because the peak age of occurrence may be a distortion.50 As in all peoples, the physiological impact of weaning on children at the Isthmus would have been significant. Weaning might well have contributed to the malformation of dental enamel in these children, but when in their lives it occurred, and which hypoplasias resulted from it, are hard to ascertain. 44. E.g., Smith, Bar-Yosef, and Sillen 1984, pp. 123–124, table 5.7 (40%; Neolithic Jericho); Stuart-Macadam 1985, pp. 392–393, table 3 (38.5%; Late Roman Poundbury, Dorchester, England). 45. Goodman and Rose 1990, pp. 76–83; Goodman and Martin 2002, p. 26. 46. Skinner and Goodman 1982, pp. 167–168, fig. 8; Larsen 1997, p. 49.

47. Hillson 1996, pp. 176–177; Katzenberg, Herring, and Saunders 1996, pp. 182–185, table 1. 48. Fildes 1986, pp. 26–39, 60–61, table 1.1. 49. Dupras, Schwarcz, and Fairgrieve 2001. 50. Skinner and Goodman 1982, pp. 167–168; Katzenberg, Herring, and Saunders 1996, pp. 185–186, 193–194; Larsen 1997, pp. 48–49.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

313

Dental Wear All teeth gradually lose coronal height due to wear from contact with other teeth or foreign substances. Dental attrition involves the progressive destruction of incisal and occlusal topography, especially the ridges of the anterior teeth and the cusps of the cheek teeth. Coronal surfaces usually are not involved in occlusal attrition, but elliptical facets typically develop at the contact points between teeth from their minute mobility on the cushion of the periodontal ligaments during mastication.51 Normal wear begins with the removal of the enamel and continues with the degradation of the dentine. The dental-alveolar system compensates for attrition by generating secondary dentine in the pulp chamber and by hypereruption, the continued emergence of teeth from their sockets, so that the CEJ grows away from the alveolar crest.52 Massive deterioration of the enamel and exposure of the pulp promotes the onset of dental caries and eventually periapical or periodontal infection, which can result in AMTL.53 Dental wear varies with tooth size, morphology, and angle. The rate of attrition is usually consistent in antimeres, with the regression of the lower molars preceding that of the upper molars. A helicoidal plane of occlusion typically develops in worn molars so that the first molars are steeper buccally than lingually, the second molars, are flattened, and the third molars retain a more lingual slope. Dental wear has a variety of dietary, behavioral, and mechanical causes. The fundamental factor is the use of teeth for slicing, grinding, and chewing food. High frequencies of coarse, abrasive substances in the diet can substantially change the shape of occlusion. Other activities that contribute to attrition are bruxism, defective jaw structure leading to unusual masticatory stress, painful lesions that promote eating on one side of the mouth, and technological and cultural activities such as holding hard objects between the teeth or altering dental shape to generate a certain appearance.54 The recording of dental wear in the skeletal sample from the Isthmus followed standard methods. Wear was scored in all permanent dentition rather than a representative side of the dental arcade, as few arcades were complete and many teeth were found in isolation. Almost every surviving tooth could be scored for occlusal wear. Teeth were identified with one of eight stages of occlusal wear for all dental classes established by Holly Smith in her study of molar attrition in prehistoric and historic groups from several global regions.55 The angle of occlusal wear in fully erupted permanent first mandibular molars was also measured according to Smith’s procedure.56 Other metric data, such as crown heights, coronal diameters, and areas of wear facets, were not collected from the teeth and no examination of microwear was conducted.57 Mean wear scores for teeth of Late Roman to Early Byzantine date are summarized in Tables 6.4, 6.5, 6.7 and 6.8. Descriptions of dental wear by individual are recorded in the summary of paleopathology in Table 7.1. Deciduous teeth in infants and young children are typically unworn or polished (score 1), while those in a child aged 7–11 years (NEG 67-001E) show minor facets, slight blunting of the cusps, and pinpoint or hairline 51. Wolpoff 1970. 52. Levers and Darling 1983; Whittaker et al. 1985. 53. Ortner and Putschar 1985, pp. 454–455; Lansjoen 1998, p. 399. 54. Smith 1984; Powell 1985, pp. 308–312; Roberts and Manchester 1995, pp. 52–54; Lansjoen 1998, pp. 398–399. 55. Smith 1984, pp. 45–46, table 2, fig. 3. Interproximal wear at contact points between teeth (Hillson 1996, p. 242) was not systematically recorded in teeth from the Isthmus. 56. Smith 1984, pp. 43–45, fig. 2. This measurement was only done on molars that were neither dislocated nor carious

but were attached to a relatively intact alveolus and jaw. The right side of the mouth was used, unless only the left side survived. Angle of wear was determined by placing a protractor across the talonid basin and measuring the angle of deviation in the lingual or buccal direction relative to a straight edge positioned along the horizontal plane. Gradients were recorded by wear stage. 57. See Buikstra and Ubelaker 1994, pp. 61–62; Hillson 1996, pp. 233–236 (measurements), 243–251; Larsen 1997, pp. 262–267 (microwear).

314

TEETH AND ORAL HEALTH

exposure of dentine (score 2). It seems doubtful that any deciduous tooth reached the level of moderate wear (score 3–4) before shedding around the tenth year. Since wear progressed throughout life and the dental sample represents all ages, permanent teeth exhibit the full range of attrition, as is apparent in the images of the dental arcades (Figs. 6.1–6.34). Although interproximal wear was not scored, the general impression is that it was slight and that no dentine was exposed between teeth. The mean wear score for all permanent teeth in the skeletons of Late Roman to Early Byzantine date is 2.99 (Table 6.7). That score represents a distinct line of exposed dentine in the incisors and canines and full cusp removal and/or some exposed dentine in the premolars and molars. Occlusal attrition was slightly greater in the canines and molars than in the incisors and premolars, and slightly greater in the lower molars than the upper ones. As is typical, occlusal attrition increased with advancing age among the Late Roman to Early Byzantine inhabitants of the Isthmus (Tables 6.4, 6.5). All classes of permanent teeth gradually wore down from childhood through the late 40s and beyond, though molars underwent the severest attrition. From adolescence into the late 20s, the rate of attrition was relatively slow, and most teeth showed only small areas of exposed dentine and blunted cusps, particularly in the molars. Typical for this age group are the upper teeth of NEG 67001D (female, 20–21 years; Fig. 6.8), the lower teeth of NEG 69-005B (female, 20–22 years; Fig. 6.10), and the upper teeth of T14 69-003A apart from M1R and M1L, which have severe coronal caries (male, 18–24 years; Fig. 6.19). The rate of wear increased slightly between the late third and the early fifth decades, during which time the hard enamel on the grinding surfaces was slowly obliterated and increasingly large areas of the softer underlying dentine were exposed. These developments are evident in the teeth of T14 69-002A (female, 25–34 years; Fig. 6.18), NEG 69-004C (male, over 35 years; Fig. 6.3), NEG 67-001B (female, 35–44 years; Fig. 6.6), and T2 68-002A (male, 35–44 years; Fig. 6.14). If individuals survived into their late 40s or older and retained their teeth, they suffered a more severe degree of occlusal attrition in which the cusps were mostly obliterated and large areas of dentine were exposed and sometimes cupped. Good examples of this high level of wear among old residents are the teeth of NEG 67-001C (male, late 40s; Fig. 6.7) and T14 67-002A (male, early–middle 50s; Fig. 6.16). The rate of wear, however, varied among adults: certain younger individuals, particularly Late Roman men, suffered unusually advanced attrition for their age, such as NEG 69-005C (male, 25–34 years; Fig. 6.11) and NEG 67-003B (male, 35–44 years; Fig. 6.9).58 The angle of the occlusal plane in mandibular molars also changed with increasing attrition throughout life (Table 6.6). Although only 20 skeletons possessed permanent lower molars and mandibles in adequate condition to study, the distribution of measurements reveals a distinct pattern. The grinding surface gradually tilted in a buccal direction with advancing wear.59 The occlusal plane of mandibular molars in a normal, unworn dental arch creates a curve in which the surface inclines lingually. In one child at the Isthmus with permanent first mandibular molars that were fully erupted but completely or mostly unworn (score 2), the teeth tilted in toward the tongue at an angle of approximately 5°. The slow progression of molar attrition into the mid-20s (score 3) degraded the buccal cusps, so that the occlusal plane grew flat. As wear in the first molars accelerated during the third and fourth decades of life (score 4–8), the occlusal plane sloped outward toward the cheeks up to 12°–15°. One 58. In comparison to the Late Roman to Early Byzantine residents, LOU 69-801A displays a typical degree of wear, while TC 60-001A displays unusually worn teeth for a young adult female (Fig. 6.34). She lived during a later period in the history of the site, when different dietary or behavioral factors might have led to higher attrition at a younger age than in earlier

times. The individuals buried during the Middle Roman era displayed dental wear comparable to that of later inhabitants of the site, particularly T2 68-003A, DEC 69-901A, and WF 62-001A with numerous surviving teeth. 59. Smith (1984, pp. 40–41) and Hillson (1996, pp. 56, 237– 239, fig. 2.28) explain the structural mechanics of molar wear.

6

6

41

12

62

0–14

15–24

25–34

35–44

45+ 86

110

98

119

36

Total Permanent Teeth (n)

4.15

3.67

2.38

2.18

1.56

Mean Wear Score: All Teeth

Mean Wear Score: Canines 2.00 (n = 2) 2.24 (n = 17) 2.36 (n = 11) 3.84 (n = 19) 4.18 (n = 11)

Mean Wear Score: Incisors 1.54 (n = 13) 2.00 (n = 22) 2.47 (n = 19) 3.84 (n = 19) 4.14 (n = 14)

4.26 (n = 27)

3.38 (n = 34)

2.48 (n = 21)

2.03 (n = 36)

2.00 (n = 4)

Mean Wear Score: Premolars

5.50 (n = 10)

4.35 (n = 17)

3.53 (n = 15)

2.84 (n = 19)

1.33 (n = 15)

Mean Wear Score: First Molars

4.19 (n = 13)

3.80 (n = 15)

3.00 (n = 12)

2.05 (n = 20)

2.00 (n = 2)

Mean Wear Score: Second Molars

4.67 (n = 6)

2.00 (n = 6)

1.88 (n = 8)

1.67 (n = 6)

–

Mean Wear Score: Third Molars

6

4

2

10

4

0–14

15–24

25–34

35–44

45+

52

65

34

70

36

Total Permanent Teeth (n)

4.34

4.15

3.50

2.11

1.56

Mean Wear Score: All Teeth

Mean Wear Score: Canines 2.00 (n = 2) 2.00 (n = 10) 3.00 (n = 6) 4.45 (n = 11) 3.83 (n = 6)

Mean Wear Score: Incisors 1.54 (n = 13) 1.75 (n = 8) 3.22 (n = 9) 4.67 (n = 6) 4.25 (n = 8)

Note: Dental wear is scored according to the standards at Smith 1984, pp. 45–46, table 2, fig. 3.

Total Individuals (n)

Age Class (years)

4.00 (n = 18)

3.90 (n = 20)

3.60 (n = 5)

1.95 (n = 22)

2.00 (n = 4)

Mean Wear Score: Premolars

5.83 (n = 6)

4.57 (n = 14)

3.71 (n = 7)

2.77 (n = 13)

1.33 (n = 15)

Mean Wear Score: First Molars

4.23 (n = 11)

4.00 (n = 11)

3.60 (n = 5)

2.17 (n = 12)

2.00 (n = 2)

Mean Wear Score: Second Molars

TABLE 6.5. MEAN WEAR SCORES IN PERMANENT DENTITION OF LATE ROMAN DATE BY AGE

5.00 (n = 2)

2.33 (n = 3)

2.00 (n = 2)

1.67 (n = 6)

–

Mean Wear Score: Third Molars

Note: Dental wear is scored according to the standards at Smith 1984, pp. 45–46, table 2, fig. 3. 1 Since it is uncertain whether HO 70-902D or I is 25–34 years old, the dental data from neither were included in this calculation. 2 Since it is uncertain whether HO 70-902E or F is 45+ years old, the dental data from neither were included in this calculation. This calculation does include NEG 69–001D, aged mature adult but probably over 45 years old.

Total Individuals (n)

Age Class (years)

TABLE 6.4. MEAN WEAR SCORES IN PERMANENT DENTITION OF LATE ROMAN TO EARLY BYZANTINE DATE BY AGE

316

TEETH AND ORAL HEALTH

TABLE 6.6. ANGLE OF WEAR IN PERMANENT FIRST MANDIBULAR MOLARS BY WEAR STAGE Wear Stage

Individuals (side used)

Age (years)

Angle/Direction of Wear (°)

Average Angle of Wear (°)

2

NEG 67-001E (L)

7–11

-5.0*

3

NEG 67-001D (R) NEG 69-005B (R) NEG 69-007B (R) NEG 69-001C (R) T2 68-002A (R) T14 67-004A (R) T14 69-003A (R)

20–21 20–22 16–18 20–24 35–44 25–34 18–24

-6.5* -4.0 -3.5* -1.0 -1.0 -1.0 +2.0

-2.29

41

NEG 67-001B (R) T14 69-002A (R)

35–44 25–34

+0.0* +2.0

1.00

52

HO 70-901A (R) NEG 67-001C (R) HO 70-902F (L) HO 70-902D (L)

35–44 Late 40s 15–45+ 15–45+

+1.5 +2.5 +3.0 +5.0

3.00

6

HO 70-902H (L) RB 76-002B (R) NEG 69-001D (R) HO 70-901C (L)

35–44 15–45+ 35+ 35–44

+6.0* +7.0 +9.0 +15.0

9.25

8

NEG 69-005C (R)

25–34

+13.0*

13.00

-5.00

Notes: Angles marked with an asterisk (*) were measured on individuals with poorly preserved mandibular molars. The minus (-) before the angle measurement indicates a lingual inclination, and the plus (+) indicates a buccal inclination. 1 The angle of wear (+2°) in WF 62-001A (wear score 4) is similar to that in later skeletons. 2 The angle of wear (+9°) in LOU 69-801A (wear score 5) is greater than that in earlier skeletons.

young adult man (NEG 69-005C) had a severely worn M1R with concave dentine (score 8) that was inclined bucally approximately 13°, while M1L had been destroyed down to the mesiobuccal root before death (Fig. 6.11). Thus, for individuals who retained their first molars into old age, progressive wear reversed the initial lingual inclination bucally by as much as 15°–20°. The degree of occlusal wear in permanent dentition varied by period and sex, though it should be remembered that the significance of any observed patterns is limited by the small sample size (Tables 6.7, 6.8). Wear was overall slightly higher among the Late Roman residents (mean score 3.09) than among the Early Byzantine residents (mean score 2.80). Moreover, wear was higher overall among males than females, a discrepancy that is especially marked among the Late Roman residents. Seldom could pathological conditions be directly linked to dental wear. Functional alterations to occlusion due to heavy attrition can exert abnormal mechanical stress on the temporomandibular joint (TMJ) and cause its degeneration. Such alterations were observed in one skeleton (T14 67-002A) in the sample from the Isthmus. A high level of occlusal wear predisposed local residents to caries, but it was only a mild predisposition, and most carious lesions occurred in unworn or slightly worn areas on the cervical and interproximal areas (Table 6.15). Severe wear could lead to osseous inflammation. For one man in his 50s (T14 67-002A), the heavy attrition of P4L (scored 5) and M3L (scored 8) induced fracturing, caries, massive cavitation of the dentine, exposure of the pulp chamber, and periodontitis and periapical granuloma (Figs. 6.16, 6.44). For one woman in her late 40s–mid-50s (NEG 69-007A), the heavy attrition of C1L and P4R (scored 7) exposed the pulp chambers, which led to pulpitis and abscessed granulomata in both alveoli (Figs. 6.13, 6.42, 6.43). A similar

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

317

TABLE 6.7. MEAN WEAR SCORES IN PERMANENT DENTITION BY PERIOD Total Permanent Teeth (n)

Mean Wear Score: All Teeth

Mean Wear Score: Incisors

Mean Wear Score: Canines

Mean Wear Score: Premolars

Mean Wear Score: First Molars

Late Roman

340

3.09

2.31 (n = 59)

3.24 (n = 45)

3.21 (n = 91)

3.53 (n = 70)

3.39 (n = 55)

2.70 (n = 20)

Early Byzantine

176

2.80

2.79 (n = 42)

2.16 (n = 25)

2.67 (n = 55)

3.57 (n = 21)

2.57 (n = 21)

2.42 (n = 12)

Total

516

2.99

2.50 (n = 101)

3.13 (n = 70)

2.92 (n = 146)

3.54 (n = 91)

3.16 (n = 76)

2.59 (n = 32)

Period

Mean Wear Mean Wear Score: Score: Second Molars Third Molars

TABLE 6.8. MEAN WEAR SCORES IN PERMANENT DENTITION BY SEX Sexable Individuals (n)

Total Teeth (n)

Mean Wear Score: All Teeth

Late Roman Females

14

171

3.06

Late Roman Males

10

120

4.03

Early Byzantine Females

2

34

3.35

Early Byzantine Males

6

142

2.80

Late Roman–Early Byzantine Females

16

205

3.11

Late Roman–Early Byzantine Males

16

262

3.36

Period and Sex

process in these individuals and many others probably led to the loss of numerous teeth before death, though how frequently cannot be ascertained.60 The main cause of dental attrition in the residents of the Isthmus was eating. If the predominant mode of subsistence at the Isthmus was mixed agriculture, the local diet would have consisted of both hard and soft foods, many finely processed and cooked, so that residents consumed low quantities of tough, fibrous matter such as stalks, roots, or shells. One exception might have been dried fish or shellfish, which can contain small bones, scales, and sand. On the basis of the faunal and archaeological remains, however, it seems unlikely that marine products were a predominant part of the Isthmian diet. Dental wear should most likely be attributed to the abrasive matter in the food, not the food itself. Coarse particles might have entered foodstuffs during processing through several channels. Minute pieces of stone could have detached from querns and become mixed with cereal grains during milling. If grain was not intensively sieved and ground, seed husks bearing sharp phytoliths could have persisted in bread flour. Windblown grit could have blended with grain during winnowing, and loose dirt could have adhered to poorly washed foods during domestic storage and preparation. Such contaminants probably had a greater impact on dental wear than dietary content. The development of the wear plane in lower molars also reflects subsistence strategies at the Isthmus. Smith has shown that the occlusal inclination of molars in agriculturalists, who mill grain and cook food in water, shifts considerably from lingual to buccal throughout life, while the grinding surfaces of molars in hunter-gatherers, who consume tougher, more fibrous meats and vegetables, tend to flatten over time.61 The progression of the wear plane in residents of the Isthmus follows the pattern among the agricultural peoples she surveyed, but the degree of buccal inclination falls between her hunter-gatherers and 60. See further pp. 346–348. 61. Smith 1984, pp. 45–55, fig. 5; Larsen 1997, pp. 254–256.

318

TEETH AND ORAL HEALTH

agriculturalists. This difference between the residents at the Isthmus and full agriculturalists can be explained by the mixed nature of local subsistence. In comparison to other ancient groups, the degree of dental wear at the Isthmus was low to moderate. The slight wear in deciduous teeth may reflect the mushy texture of the child’s diet. Absent or negligible wear in the teeth of children around or just older than 2–3 years62 could have resulted from a weaning diet, as Theya Molleson has argued for groups in Neolithic Syria and Roman England.63 Mothers at the Isthmus might have supplemented their breast-milk with soft, porridge-like foods made from grain or fed their children such food after weaning, in which case deciduous teeth would develop little or no wear. As children grew older and reached adulthood, dental attrition increased, but wear scores seldom exceed 5 even in old people, and wear was overall heavier in the posterior than in the anterior teeth. Moreover, caries or TMJ disease can be positively linked to attrition in only a few instances. This contrasts with dental conditions among well-studied prehistoric to medieval samples in Britain and North America.64 In comparison to the residents of the Isthmus, these peoples must have consumed softer, less abrasive foods, such as finely milled grain, fleshy fruits and vegetables, and well-cooked meat. A low to moderate level of occlusal and interproximal attrition is not unusual for a population dependent on a mixed agricultural subsistence.65 Moreover, consumption of soft foods required more chewing and grinding with the back teeth than tearing and cutting with the front teeth, which explains the differential wear rates across the dental arch in the sample from the Isthmus. The distribution of dental wear by age and sex also illuminates dietary habits. Residents of the Isthmus show a relatively constant rate of dental wear over life, notwithstanding the apparently slight increase in adulthood. This suggests that the young and the old consumed foods of a similar texture. The discrepancy in overall wear between Late Roman males and females is not easily explained.66 One possibility is that men during this period ate coarse, unprocessed foods that they either gathered or brought with them during daily activities away from the home, whereas women ate finer, processed foods that they prepared at home. If such an occupational and dietary division existed, the evidence of dental wear seems to indicate that it did not persist into the Early Byzantine centuries. Agricultural intensification and a shift to a less diverse diet of more highly processed foods may explain the general decrease in attrition over time suggested by the surviving dental evidence. Apart from food preparation and diet, other factors apparently had little influence on dental wear at the Isthmus. It is impossible to know how psychological stress and the clenching of teeth influenced dental attrition; there is no sign that acidic agents altered grinding surfaces. Differential wear in one dentition indicates assymetrical mastication. An older man (NEG 69-004C) preferred to chew his food on the back left side of his mouth because of painful lesions afflicting M3R, which eventually fell out. Over time, the left upper and lower molars degraded somewhat more quickly than their right counterparts, which acquired a substantial shelf of calculus (Figs. 6.3, 6.37).67 Another old man (T14 67-002A) might have 62. NEG 69-004D, 67-001H, 67-001I, 67-001K, 67-003C, 69-001B, HO 70-902B. 63. Molleson 1993, pp. 179, 183 (Poundbury, Dorchester); 1995 (Abu Hureyra, Syria); Katzenberg, Herring, and Saunders 1996, p. 186. 64. E.g., Miles 1962; Brothwell 1981, pp. 71–72, fig. 3.9; Whittaker et al. 1985; Hodges 1991 (Neolithic–Medieval British attrition, TMJ arthritis); Lovejoy 1985 (prehistoric Native American attrition). 65. Many anthropologists have observed the fundamental difference in overall dental wear between hunting-gathering

and agricultural groups (Larsen 1997, pp. 250–252). A marked reduction in attrition accompanies the shift from nondomesticated to domesticated plants and technological improvements in food production. 66. Larsen (1997, pp. 257–258) surveys intrapopulational variability in dental wear by sex. In most documented groups, gender-specific activities lead to heavier attrition in women. 67. The imbalanced masticatory stress that arose from this long-term behavior does not seem to have affected either TMJ. While the mandibular rami and condyles are not preserved, the fossae show no signs of arthritic remodeling.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

319

suffered a similar condition on the opposite side of his mouth before the evulsion of his lower left molars (Fig. 6.16). One cause of dental wear in many cultures is the application of teeth as tools. Using the anterior or posterior teeth as scrapers, piercers, or clamps to work materials or to hold them in place can modify dental form in anomalous ways, particularly in the coronal region.68 One old woman (NEG 69-004B) had an I1R with an unusually sharp incisal edge and lingual flattening that may be attributed to the regular use of the front teeth for scraping and/or holding (Fig. 6.36), as discussed below. Finally, none of the teeth from the Isthmus exhibits intentional alterations for ornamentation, orthodonture, or oral surgery. One rare morphological feature that is sometimes associated with the mechanics of chewing is an elongated bony protrusion in the oral cavity called a torus. Anthropologists have debated whether the occurrence of mandibular, maxillary, and palatine tori is influenced more by genetic or environmental factors.69 High frequencies of tori have been observed in far northern and Arctic peoples who consume very tough, lightly cooked, or raw foods. Such a diet exerts heavy masticatory stress on oral structures and thus favors the development of osseous butresses along the roof and the floor of the mouth.70 None of the skeletons in the sample from the Isthmus displays mandibular or palatine tori,71 even those individuals with heavy dental wear. Therefore, tori cannot be associated with abnormally high mechanical demands on the jaws among local residents and do not seem to have been a genetic marker at the Isthmus.

Dental Trauma Apart from the gradual degradation of teeth from attrition, dental structures can be altered more quickly and dramatically by localized trauma. Cracks, chips, or snaps of various shapes and sizes in isolated teeth or segments of the dental arch can result from biting down on hard substances, accidental injury, or using teeth as tools.72 Anthropologists must distinguish genuine antemortem trauma from postmortem damage. Several teeth from the Isthmus displayed vertical cracks, chips, or brittle enamel that had been caused by weathering in the burial environment.73 Antemortem damage was identified when the edges of broken enamel were blunt and not crisp, stained and not light and untinted, and sometimes infilled with soil particles or encased in a silty crust.74 Eight individuals buried at the Isthmus displayed definite antemortem fractures in a total of 15 teeth, all from the permanent dentition (Table 6.9). In the total dental sample from the site, the frequency of antemortem trauma is 2.07% (15/725). The prevalence of this condition in teeth of Late Roman to Early Byzantine date is 1.58% (10/631), while the prevalence during just the Late Roman period is 1.35% (6/446). During these main periods in the history of the local settlement, teeth with physical damage occurred in both men and women but only in mature adults. Four individuals (NEG 69-004E, 69-005C, HO 70-901A, 70-902D) display similar damage on a small scale, while two (NEG 69-004B, T14 67-002A) display extensive destruction. The first six individuals display minor traumatic alterations to teeth that were most likely sustained during eating. This common form of dental fracturing was a small triangular or rectangular chip broken from the enamel along the front or back edge of the occlusal 68. Milner and Larsen 1991, pp. 364–367, 371–373; Larsen 1997, pp. 258–262. 69. Hauser and De Stefano 1989, pp. 174–185; Scheuer and Black 2000, pp. 129, 142. 70. Halffman, Scott, and Pedersen 1992; Larsen 1997, pp. 237–239, fig. 7.3. 71. The examination of nonmetric traits in the skeletal

sample did not include the maxillary torus. 72. Milner and Larsen 1991, pp. 367–371; Larsen 1997, pp. 267–268; Capasso, Kennedy, and Wilczak 1999, pp. 147–161. 73. See p. 239, n. 8. 74. See Ortner and Putschar 1985, p. 453; Milner and Larsen 1991, p. 367.

320

TEETH AND ORAL HEALTH

TABLE 6.9. INSTANCES OF DENTAL TRAUMA Burial No.

Affected Tooth

Description

Probable Cause

I1L

Large enamel-dentine chip, labial face from incisal edge down to 2.5 mm below CEJ

Injurious fall

I1R

Large enamel-dentine chip, mesiolabial face from incisal edge down to 1.5 mm below CEJ; lingual attrition

Injurious fall; possible extramasticatory use

NEG 69-004E

I2R

Small–medium enamel chip, labial occlusal margin, shallow extension down coronal face

Microtrauma

NEG 69-005C

M1L

Small enamel chip, mesiolingual face from occlusal surface down to 2 mm below CEJ

Microtrauma

P4L

Large enamel-dentine-pulp chip, labial face from occlusal surface down to 3 mm below CEJ

Major fracture

M2R

Large enamel-dentine chip, mesiolingual face from occlusal surface down to 3 mm below CEJ

Major fracture

M3R

Large enamel-dentine chip, mesiolingual face from occlusal surface down to 4 mm below CEJ

Major fracture

M1L

Large enamel-dentine chip, distobuccal face from occlusal surface down to 4 mm below CEJ

Major fracture

NEG 69-004B

T14 67-002A

M1R

Small enamel chip, mesial interproximal

Attrition or abrasion

M1R

Small enamel chip, mesiobuccal occlusal margin

Microtrauma

P3R

Small enamel chip, distobuccal occlusal margin

Microtrauma

P4R

Medium enamel chip, lingual coronal face from occlusal margin down to CEJ

Microtrauma

M1R

Small enamel chip, distobuccal occlusal margin

Attrition or abrasion

HO 70-901A

M1R

Small enamel chip, distobuccal occlusal margin

Attrition or abrasion

HO 70-902D

ML

Small enamel chip, mesial occlusal margin

Attrition or abrasion

DEC 69-901B

LOU 69-801A

1

surface either near its contact with the neighboring tooth or at the corner. Such minute flakes exposed only a small area of dentine, but in two cases deeper chips were removed from the crown. The probable causes of minor chipping of the enamel are attrition or abrasion and microtrauma. When teeth are moderately or heavily worn so that the dentine is exposed and cupped, the surviving rim of enamel around the occlusal surface grows unstable and fragile. Under such conditions, the enamel is prone to chipping from either the usual pressures of attrition or the abrasion of the grinding surface by coarse particles. This seems to have been the case in DEC 69-901B (M1R; Fig. 6.22), LOU 69-801A (M1R; Fig. 6.20), HO 70-901A (M1R), and 70-902D (M1L; Fig. 6.27). Caries can lead to similar destabilization of the occlusal margin and eventual fracturing of the enamel. This seems to have been the case in NEG 69-005C (M1L; Fig. 6.11). Sometimes biting down forcefully on dense, hard matter can also cause a small, local fracture in a tooth that was not severely worn. This seems to have been the case in NEG 69-004E (I2R) and LOU 69-801A (M1R, P3R; Fig. 6.20). There is no evidence that these six individuals damaged their teeth by using them as tools, such as for textile, leather, or food preparation. Similar dental trauma, sometimes called pressurechipping, has been studied in skeletons from other regions.75 While the overall status of dental wear among residents at the Isthmus was low to moderate, they did consume abrasive matter that occasionally fractured their teeth, and heavy attrition predisposed certain adults 75. Turner and Cadien 1969; Milner and Larsen 1991, pp. 367–370; Larsen 1997, pp. 267–268.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

321

to enamel chipping, particularly in premolars and molars. Even so, the frequency of dental trauma in this skeletal sample is considerably lower than in many prehistoric groups with a coarser diet, a more demanding masticatory regime, and various extramasticatory uses for teeth.76 In addition to these typical examples of enamel chipping, two individuals suffered major dental trauma that requires futher comment. One woman in her late 40s (NEG 69-004B) suffered fractures and differential attrition in her upper front teeth and AMTL in her lower front teeth (Figs. 6.2, 6.36). I1L displays a broad transverse snap cutting across the regular attrition plane that removed the entire labial half of the crown at a ca. 75° angle to the incisal edge, down to 2.5 mm below the CEJ. I1R displays a large longitudinal chip cutting across the regular attrition plane that removed the mesiolabial quadrant down to 1.5 mm below the CEJ on the labial surface. Both of these fractures have destroyed the enamel and deeply exposed the dentine. The wear planes on the fractured surfaces of the two teeth are contiguous. There are, however, no signs of caries, pulpitis, or periapical infection in either tooth, and the margins of the fractures exhibit only slight wear. Moreover, I2L shows no wear opposite the fracture on I1L. These observations suggest that the breaks occurred not long before death. The most probable cause of these fractures is accidental injury. The injury snapped off the upper front teeth probably when the woman was in her 40s. Fractures in the central maxillary incisors most commonly result from striking a hard object or surface face first, such as when falling on the ground.77 It is clear that trauma was localized, because the surrounding mandibular and maxillary bone was not damaged. The sharpness of the fractures, the force of the blow, and the absence of lower incisors suggest that this woman also suffered soft tissue damage, such as lacerations to the lips, nose, chin, cheeks, and forehead. While the exact context for this injury cannot be recovered, it is possible to imagine several scenarios involving falling on hard pavement, an earthen floor, or rocky terrain in the daily life of a physically active woman residing at the Fortress. In addition, the surviving distal crown of I1R has an abnormally sharp incisal edge, and the lingual concavity and marginal ridge are distinctly flattened. This heavy wear occurred over a long period of use prior to the fractures in the upper central teeth and would have promoted their breaking through structural weakening. Alteration mostly to the lingual aspect of the crown can be variously interpreted. It might have arisen from a slight overbite, but the mandibular incisors were lost before death and cannot be checked for compensatory remodeling. It might have been caused by dietary habit, in particular holding fibrous, abrasive substances between the tongue and upper teeth, such as during peeling or sucking. One food available at the Isthmus that might have been eaten in this manner was artichoke. But this explanation seems unlikely because it would mean that this one woman ate artichoke habitually over a long period, unlike her neighbors. Another possible cause is processing food or other materials, such as stripping off pieces or scouring a surface by drawing a coarse substance in a downward motion through the clenched teeth.78 Differential wear to incisor crowns has been noted in skeletons from Late Byzantine Panakton and identified with extramasticatory use to soften some material, perhaps leather.79 Moreover, lingual surface degeneration of the maxillary teeth, sometimes coupled with deterioration 76. See Milner and Larsen 1991, pp. 368–369, table 1. 77. Ortner and Putschar 1985, pp. 452–453; Milner and Larsen 1991, pp. 370–371. 78. Similar explanations have been proposed for lingual wear in the anterior maxillary dentition of prehistoric peoples who processed manioc and worked animal hides with the front teeth (Larsen 1997, pp. 260–262, fig. 7.9). Lingual wear in the

incisors, especially maxillary, of prehistoric peoples of the eastern Arabian coast and the Arabian Gulf has been attributed to food preparation or occupational activities (rope or basket production), but there it was coupled with occlusal flattening in the premolars and incisal wear in the canines (Littleton and Frohlich 1993, pp. 432–433, 441). 79. See pp. 357–358.

322

TEETH AND ORAL HEALTH

of the mandibular condyles and other modifications of the hands, pelvis, and legs, has been observed in skeletons of Byzantine or later date from sites on Cyprus and on the northern coast of Turkey and interpreted as evidence for processing fiber, or specifically spinning.80 However, edge modification from preparing foods or other materials usually generates a more pronounced and regular pattern of grooving than is visible in NEG 69-004B. Without the complete anterior dentition, it cannot be determined with confidence whether the lingual wear in I1R was caused by an overbite or some extramasticatory use. Finally, I1R and I1L were lost before death, and the sockets were completely resorbed. AMTL has several causes, including infection and trauma. This woman suffered dental disease in her posterior dentition, where several upper and lower molars (M1R, M1L, M2R, M3L; M1R, M1L, M2L) were lost before death, and the alveolar bone shows moderate or complete resorption (see Table 6.20 below). It is unclear whether periodontal or pulpal infection afflicted I1R and I1L, though the closest mandibular teeth (C1R, P3R) are relatively healthy for a woman of this age (I2R and I2L were lost postmortem). The intact state of the mandibular bone around the lower central incisors shows that they were not lost through massive injury to the lower jaw. They might have been affected by trauma on a smaller scale, in which case minor breakage in the walls of the sockets has been obliterated by resorption. A less likely explanation is that the utilization of the front teeth during some habitual behavior led to their loss, perhaps in relation to the lingual wear in I1R.81 Whatever its cause, evulsion must have occurred well before death because the alveolar tissue was fully remodeled. Therefore, it was probably not related to the same traumatic episode that fractured the upper central incisors. The second person with unusual dental trauma, a man in his early to mid-50s (T14 67002A), suffered severe fractures and AMTL in his lower back teeth (Figs. 5.24, 6.16, 6.39, 6.44). On the left side of the mouth, P4L has a large rectangular chip cut from the occlusal surface through the enamel and dentine to 3 mm below the CEJ, exposing the pulp chamber. The chip begins at a ca. 30° angle and then extends down in a roughly vertical slice, removing the buccal half of the tooth. The preserved lingual portion of the occlusal surface is severely worn, more so than P4R. M1L, M2L, and M3L were lost before death. The alveolar tissue was partially resorbed around M1L, completely around M2L, and almost completely around M3L. The cause of the evulsion of these three teeth is unknown. Above this, the distobuccal cusp of M1L has a large rectangular chip cutting through the enamel and dentine vertically from the occlusal surface to 4 mm below the CEJ. The neighboring M2L was lost before death and the alveolar tissue was almost completely resorbed. M3L is severely worn and displays a massive carious lesion on the distobuccal side of the crown, which led to periodontitis and alveolar destruction. On the right side of the mouth, M2R and M3R both have large triangular chips on the mesiolingual side that cut through the enamel to the dentine but not the pulp. The chip in M2R extends from the occlusal surface to 3 mm below the CEJ at a ca. 50° angle, while the chip in M3R extends from the occlusal surface to 4 mm below the CEJ at a ca. 70° angle. The cavities on the buccal aspects of these teeth do not communicate with these fractures. All margins of the fractures are blunt and worn, which proves that they occurred long before death. The fractures in the right lower posterior teeth do not constitute a continuous wear plane, and their shape and angulation varies. This demonstrates that they occurred in isolated episodes. Other old individuals in this sample display a high degree of advanced wear, but such massive fracturing is unique. 80. Harper and Fox 2008, p. 19 (Athienou-Malloura, Cyprus, “Medieval or Venetian”; Polis Chrysochous, Cyprus, 7th– 11th centuries); Erdal 2007 (Sinop, Turkey, 10th century).

81. Merbs (1989a, pp. 172–173) summarizes his research on dental loss from the use of teeth as tools and related trauma, with special reference to the Eskimo.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

323

In addition to these fractures, the right TMJ shows nascent arthritic degeneration. The medial half of the articular surface of the mandibular condyle displays diffuse pitting and a narrow osteophytic ridge along the anterior margin, while the lateral half displays slight sclerosis. The anterior plate of the mandibular fossa shows very slight pitting but no largescale remodeling. The condition of the left TMJ is uncertain because the fossa is fragmentary and the condyle was destroyed after deposition, perhaps when the grave was disturbed by builders in the Byzantine era. This old man suffered several forms of dental disease that were mutually related. While the exact order of events is difficult to reconstruct, the status of his teeth at death was evidently the culmination of a long degenerative process. The fracturing of large blocks from four teeth seems to have resulted from a combination of considerable wear and unnatural masticatory stress. Heavy attrition of the enamel and dentine typical of an individual in his sixth decade substantially weakened the occlusal surface of these teeth. Uneven stress leading to differential wear and severe fracturing can be caused by mandibular ramus hypoplasia, TMJ deformation, or malocclusion. This man’s left mandibular condyle and fossa have not survived intact, but the left ascending ramus mirrors its right counterpart, indicating that it is not defective.82 The most likely explanation for the fractures is that asymmetrical demand was placed on the posterior teeth when this man preferred to chew on one side of his mouth. He suffered massive infections on the left side of his mouth, presumably at first in the lower molars and the upper second molar, all of which were eventually lost, and then in the upper first and third molars. It is quite possible that the pain arising from these chronic ailments compelled him to favor the right side of his mouth. Once the right molar crowns were significantly degraded by attrition, abnormally forceful biting on one side of the dentition, perhaps together with the chance presence of a hard particle in the mouth, could readily snap off a large fragment from a tooth. The cause of the fractures in P4L and M1L is unknown, but the latter must have occurred before the loss of the lower molars. Again, it is possible that the weakening of the entire masticatory apparatus with the gradual loss of molars in an unknown sequence exerted an unnatural pressure on these teeth that promoted their coronal disintegration. The arthritis in the right TMJ was probably related to these developments.83 Mild arthropathy resulted from the prolonged mechanical stress of imbalanced chewing and the reduced integrity of the dental arch with the eventual loss of five molars. Chewing food on one side of the mouth distributes the bite force unevenly across the occlusal plane and creates a disproportionate load on the joints, stressing one joint capsule more than the other.84 The loss of numerous teeth causes occlusal malfunction as the mandible shifts away from its centric position to achieve new but abnormal biting surfaces. The repeated displacement, or internal derangement, of the joint leads to microdamage in the cartilaginous disc and eventually degenerative remodeling in both condyle and fossa.85 Numerous studies of skeletons from different regions have shown a close correlation between AMTL and osteoarthritis at the TMJ.86 The ailment would have been painful but probably not as bothersome as the infectious lesions in the back teeth. 82. The reconstructed skull shown in Fig. 5.24 gives the distorted impression of a marked overbite. This resulted from less than tight mending at the temporalosphenoid sutures, which placed the mandibular fossae behind their anatomical position. An example of a defective ramus “apparently as a result of masseter paralysis” has been recorded in a skull of Roman date from Kambi near Vasa on southern Cyprus (Angel 1945–1948, p. 68 [CKV I]). 83. Although the left TMJ does not survive, it, too, might

have displayed arthritic deformation. 84. Hesse and Naeije 1994, pp. 63–67, figs. 2–11, 2–12, 2–13. 85. Lansjoen 1998, p. 400; Soames and Southam 1998, p. 327; Resnick 2002, vol. 2, p. 1738. 86. E.g., Wedel, Carlsson, and Sagne 1978, pp. 184–185, fig. 7 (medieval Swedish); Griffin, Powers, and Kruszynski 1979, pp. 105–106 (northern European and Australian); Whittaker, Davies, and Brown 1985; Hodges 1991, p. 374 (ancient British).

324

TEETH AND ORAL HEALTH

Plaque and Calculus Dental plaque is a membranous matrix of salivary glycoprotein, food particles, and oral bacteria that adheres to the tooth surface. Calculus is mineralized plaque chiefly made up of crystallites of inorganic salts from the saliva. Areas near salivary ducts, namely, the buccal upper molars and the lingual lower incisors, often display the densest concentrations. The accumulation of plaque and the development of calculus are linked to poor dental hygiene and the consumption of carbohydrates, on which plaque bacteria feed. Plaque is the major prerequisite for the development of dental caries and periodontal disease.87 Calculus is not always well preserved in skeletons from archaeological sites because it can be easily broken from the tooth surface by postdepositional disturbance, hasty recovery, or vigorous cleaning. Thus, not all instances of calculus in skeletons from the Isthmus have survived for study.88 All observable instances of calculus were recorded by location and severity.89 Supragingival calculus was observed in 157 teeth in 25 skeletons at the site (Tables 6.10– 6.12).90 The frequency of individuals with calculus among those with preserved teeth buried during the Late Roman to Early Byzantine periods is 46.81% (22/47) and 66.67% among adults (22/33). The frequency of calculus in the Late Roman to Early Byzantine dental sample is 23.77% (150/631) and 29.94% among teeth in adults (150/501). Among Late Roman skeletons with preserved teeth, 42.11% (16/38) of all individuals, 64.00% (16/25) of all adults, and 20.85% (93/446) of all teeth display calculus. Nearly half of the instances were observed in mandibular incisors, many on the lingual aspect, and in maxillary molars, many on the buccal aspect; the remaining instances are distributed roughly evenly over the other tooth classes. The predominance of calculus at these two sites is determined by the salivary ducts, which channel nutrients for bacterial plaque. The most common form of calculus is as a low, well-defined ridge just above the CEJ; only occasionally does it expand slightly toward the occlusal surface and out from the crown (e.g., Fig. 6.36). Calculus seems to have a low correlation with age in the sample: several young adults display teeth with slight calculus91 but most cases occur in mature adults.92 The frequency of calculus is slightly higher among Early Byzantine adults (32.39%) than among Late Roman adults (28.61%). It is noteworthy that six of eight Early Byzantine adults with surviving teeth show calculus in multiple locations. The overall distribution of calculus by sex seems even. Due to the very small sample size, it is uncertain whether the discrepancy between men and women in the Early Byzantine sample is meaningful. The primary causes of dental plaque and calculus among local residents were poor hygiene and the consumption of sugars and starches. While they probably did clean their teeth, they most likely did not usually employ an abrasive implement or dentifrix, which would have prevented plaque formation and removed nascent deposits. As a result, mineral buildups were a usual part of the oral environment of most adults in the community. Another factor contributing to the growth of plaque was diet. While most plaque microbes acquire nutrients from saliva and gingival crevice fluid, they also metabolize carbohydrates eaten by their hosts. The primary sources of carbohydrates for the residents at the Isthmus would have been cereal grains, fruits, and vegetables. Rural residents consuming a similar 87. Hillson 1996, pp. 259–260; 2000, p. 259. 88. Calculus seemed absent in DEC 69-901A, 69-901B, RB 76-002A–C, and WF 62-001A, but the poor condition of the enamel made observation difficult. It is striking that no calculus was observed in HO 70-901A–C and 70-902D–J, some of whom were mature adults. There are several possible causes for the destruction of calculus in these cases: secondary burial (HO 70-901), the detachment and movement of teeth within the burial environment (HO 70-902), excavation, and storage. 89. Calculus was scored according to the standards in

Brothwell 1981, p. 155, fig. 6.14b. 90. It is probable, if one considers the frequency of periodontitis in the sample (pp. 336–345, Tables 6.19, 6.21, 6.22), that sub-gingival calculus once existed but is now unobservable due to surficial destruction of dental roots. 91. NEG 69-004D, 67-001D, 69-005B, 69-005C, T14 67004A, 69-002A, 69-003A. 92. NEG 69-004B, 69-004C, 69-004E, 67-001B, 67-001C, 69-001D, T2 68-002A, T14 67-002A.

TABLE 6.10. INSTANCES OF CALCULUS Burial No. NEG 67-001B NEG 67-001C NEG 67-001D NEG 67-003B NEG 69-001D NEG 69-103A NEG 69-004B

Affected Tooth

NEG 69-004D

Size1

C1R, C1L, P4L, M1R, M1L

Buccal

S

C1R, P4L, M1R, M1L, M2R, M2L, M3R, M3L

Buccal

S

P3R, P3L, P4R, P4L

Lingual

S–M

C1R, P3R, P4R, M1R, M1L, M2R, M2L

Buccal

S–M

C1R, C1L

Labial+lingual

M1L

Mesiolingual

S

I1R, I2R, I2L, C1R, C1L, P3R, P3L, P4R

Labial/buccal

S

M1R, M1L, M2R, M2L

Buccal+lingual

S S

S

M1R, M1L, M2R, M2L

Buccal+lingual

M2R, M2L

Buccal

C1R

Buccal+lingual

P4R

Buccal

M

CL

Buccal

M

P3L, P4L

Buccal

S

P4R, M1R, M2R, M3R

Buccal

C

I1R, I2R, I1L, I2L

Mesial+distal+labial

S

C1L, P3R

Buccal

S

P4R

Buccal+lingual

S

M1R, M2R

Buccal+lingual

M–C

M1R, M2R

Buccal

1

NEG 69-004C

Location

S M–C

S

M1L

Buccal+lingual

S

NEG 69-004E

I1R, I1L, I2R, I2L

Mesiolingual

S

NEG 69-005B

I1R, I1L, I2R, I2L

Labial

S

NEG 69-005C

I2R, I2L, P3L

Labial

S

NEG 69-007A

M R, M R

Buccal+lingual

C

I1R, I1L, I2R, I2L, C1R, C1L, P3R, P3L, P4R, P4L, M1R, M1L

Labial/buccal

S

I1R, I1L, I2R, I2L

Labial+lingual

S

C1R, C1L

Mesial+distal

S

T2 68-002A T2 68-003B

1

3

CR

Mesiobuccal

S

I1R, I1L, I2R, I2L

Mesial+distal

S

1

T14 67-002A

I2L

Distolingual

S

Mesiolingual

S

T14 67-004A

C1R, C1L I1L, I2L, C1R, C1L, P3R, P3L, P4R, P4L, M1R, M1L, M2R, M2L, M3R, M3L I1R, I1L, I2R, I2L, C1R, C1L

Labial/buccal

S

Labial/buccal+lingual

S

I1R, I1L, I2R, I2L, C1R, C1L

Labial/buccal+lingual

S

I1R, I1L, I2R, I2L

Mesial+distal+lingual

S

M1R

Mesial+distal

S

I1R, I1L, I2R, I2L

Labial

S

C1L

Buccal

M

T14 69-002A T14 69-003A LOU 69-801A

PL

Buccal

S

M1L

Distobuccal

S

3

HO 70-902D

I2L

Distolingual

S

P4R, P4L, M1R

Lingual

M

M1R

Mesiobuccal

S

P4R, P4L

Lingual

M

HO 70-902H

M2R

Buccal

S

HO 70-902I

C1L

Buccal

S

IR

Mesial+distal+lingual

S

I2L

Mesiolingual

C

HO 70-902E

2

HO 70-902?

TC 60-001A 1

M1R

Distolingual

S

M2R

Distolingual

S

MR

Mesial+distal

M

1

Size abbreviations: S = slight; M = medium; C = considerable (Brothwell 1981, p. 155, fig. 6.14b).

326

TEETH AND ORAL HEALTH

TABLE 6.11. FREQUENCY OF CALCULUS BY PERIOD Total Erupted Teeth (n)

Total Teeth in Adults (n)

Total Teeth with Calculus (n)

Prevalence of Calculus in All Teeth (%)

Prevalence of Calculus in Teeth in Adults (%)

Late Roman

446

325

93

20.85

28.61

Early Byzantine

185

176

57

30.81

32.39

Total

631

501

150

23.77

29.94

Period

TABLE 6.12. FREQUENCY OF CALCULUS BY SEX Sexable Individuals (n)

Total Teeth (n)

Teeth with Calculus (n)

Prevalence of Calculus (%)

Late Roman Females

14

171

47

27.49

Late Roman Males

10

120

32

26.67

2

34

14

41.18

Period and Sex

Early Byzantine Females Early Byzantine Males

6

142

43

30.28

Late Roman–Early Byzantine Females

16

205

61

29.76

Late Roman–Early Byzantine Males

16

262

75

28.63

diet in central Italy during the Roman era and the early Middle Ages exhibit prevalences of calculus per individual and per tooth roughly comparable to those at the Isthmus.93 If the frequency of plaque did increase over time, one possible explanation is a shift to a more carbohydrate-rich diet, perhaps related to an intensification of agriculture in the Early Byzantine settlement. A similar explanation was proposed for the increase in the prevalence of enamel hypoplasia and the decrease in the severity of occlusal wear. The manner of eating influenced the distribution and severity of calculus in one individual. A mature adult male (NEG 69-004C) displayed massive shelves of calculus covering the entire buccal aspect of his right upper fourth premolar and molars from the occlusal margin to 1–3 mm below the CEJ (Figs. 6.3, 6.37). Several other teeth had less prominent deposits. This man suffered severe caries and abscessing in his upper right molars, which must have been very painful. As a result, he chewed his food on the left side of his mouth. This masticatory preference led to differential molar wear. The normal action of grinding food, which moves the teeth against the cheeks, could not dislodge matter adhering to the buccal coronal surface, and thus large calculus deposits formed.

Dental Caries Dental caries is an infectious disease involving the decalcification of the inorganic component of teeth and the destruction of the organic component. Structural disintegration is caused by acid from the fermentation of carbohydrates by oral bacteria. Carious lesions are characterized by their form and location.94 Once the enamel or cementum has been opened by cavitation, the dentine quickly deteriorates in a lateral progression. If caries reaches the pulp chamber, it can lead to pulpitis, alveolar infection, periapical abscesses, and AMTL. Once its path reaches the jaws, caries can further promote lethal cranial infections, such as meningitis or sinus thrombosis, and systemic infection, causing, for instance, endocarditis, pneumonia, or hematogenous osteomyelitis.95 93. Manzi et al. 1999, pp. 332, 334, 335–337, tables 2, 3 (66.7% and 26.9%, Lucus Feroniae, 1st–3rd centuries; 63.6% and 27.1%, La Selvicciola, 7th century). 94. Hillson 1996, pp. 269–275; 2000, pp. 260–266; Soames

and Southam 1998, pp. 26–34. 95. Ortner and Putschar 1985, pp. 123, 434, 442; Calcagno and Gibson 1988, pp. 509–510; Dias and Tayles 1997, p. 552.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

327

Several factors influence the frequency of dental caries within a group.96 The essential conditions are the presence of susceptible dentition, the consumption of fermentable carbohydrates, and the growth of dental plaque. A direct correlation therefore exists between a group’s diet of sugars and starches and the incidence of caries.97 Another dietary variable in the prevalence of dental caries is food texture. A soft, sticky diet generally promotes caries because those substances are easily trapped on occlusal surfaces and within interproximal spaces. A coarse diet generally reduces caries by cleaning teeth of food debris, deterring the accumulation of plaque, destroying calculus, and flattening the occlusal surfaces to which softer substances adhere. Trace elements ingested in food and drinking water, especially fluoride, can also influence the local occurrence of caries. Other factors contribute to the onset of dental caries.98 The morphology of teeth and the mechanics of chewing generate a variable caries rate across the dental arch. Foodstuffs typically spend less time in contact with the sharp-edged crowns of the anterior teeth, while the occlusal surfaces of the premolars and molars tend to trap food, especially when it is soft and sticky. Such differences promote caries in the posterior dentition. Two other conditions can influence the frequency of decay, though the precise nature of the causal relationships is unclear. First, enamel hypoplasia can generate zones conducive to coronal decay by weakening the surface of teeth. Second, dental attrition aids the rapid progression of decay if it exposes the pulp chamber. Finally, behavioral factors such as the duration and frequency of meals and preventive cleaning contribute to the occurrence of caries in a group. The examination of teeth from the Isthmus revealed 59 lesions of chronic caries in at least 21 adults99 of both sexes from all periods under study (Table 6.13); there is no evidence that subadults suffered from acute juvenile caries.100 Definite carious lesions were described using the Moore and Corbett system originally based on Ancient and Medieval British skeletons and later refined.101 Instances were recorded by individual, affected tooth, site on tooth, and size according to a standard four-point system. Large lesions were scored as “gross coronal” or “gross cervical” if the general region of initiation was evident, or simply “gross caries” if most of the crown and cervix was destroyed. The teeth from the Isthmus provide ample data for investigating the occurrence of caries among local residents (Tables 6.14–6.18). The frequency of Late Roman to Early Byzantine skeletons with caries among those with preserved teeth is 36.17% (17/47) and 51.51% among adults (17/33). But this calculation is deceptive, because several skeletons have very few surviving teeth. Since some number of teeth lost antemortem and postmortem were carious, the actual frequency of individuals with caries and the actual average number of lesions per mouth could well be higher. Both these numbers were probably much higher because the total survival rate of teeth at the site is 40.64% and the prevalence of AMTL is 13.28% (Table 5.2). Caries was probably the main cause of AMTL among local residents.102 A more accurate impression of the prevalence of caries in the Late Roman to Early Byzantine sample is provided by the ratio of carious teeth to all erupted and surviving teeth, which is 7.45% 96. Powell 1985, pp. 313–317, fig. 10.1. 97. Hillson 1996, pp. 276–279; Soames and Southam 1998, pp. 25, 26. 98. Powell 1985, pp. 315–316; Hillson 1996, pp. 283–284; Larsen 1997, pp. 65, 67. 99. Ten teeth with 13 carious lesions recovered from HO 70-902 could not be associated with individual skeletons in that grave. 100. Open cavities with sharp edges were scored, but teeth were not viewed microscopically for nascent caries, which forms as a discolored opacity. Special attention was given to distinguish caries both from noncarious cavities, such as those caused

by morphological variation, enamel hypoplasia or postmortem erosion, and from dark-colored dentine exposed by dental attrition or fracture (see Lukacs 1989, p. 267; Buikstra and Ubelaker 1994, p. 55; Lansjoen 1998, p. 405; Mays 1998, pp. 146–147). 101. Moore and Corbett 1971, p. 157; 1973, p. 143; see later modifications in Metress and Conway 1975; Lukacs 1989, p. 267; Buikstra and Ubelaker 1994, pp. 54–55; Hillson 1996, pp. 280–281, table 12.3; 2000, pp. 264–265. 102. See p. 347. Skeletons that are edentulous (NEG 69103B) or missing many teeth (NEG 67-001C, 69-001A, 69007A) might well have had multiple carious teeth.

328

TEETH AND ORAL HEALTH

TABLE 6.13. INSTANCES OF DENTAL CARIES Burial No.

NEG 67-001C

Affected Tooth

NEG 69-004C

Mesial gross cervical

2

P4R

Mesial CEJ

2

M1L

Distal CEJ

1

M2R

Buccal gross cervical

2

Buccal CEJ

2

Distobuccal root

1

M3R

Buccal CEJ

1

PL

Mesial interproximal

1

I 1R

Distal CEJ

1

I2L

Distal CEJ

1

MR

Distal CEJ

2

3

2

MR

Mesial gross cervical

2

Mesial CEJ

1

Occlusal (buccal groove)

1

M2R

Occlusal (central fossa)

2

3

NEG 69-004E NEG 69-005B

Size of Lesion1

C1L

M2L

NEG 67-003B

Site of Lesion

M2L

M2L

Occlusal (central fossa)

2

NEG 69-005C

M1L

Mesial CEJ

1

NEG 69-007A

C1R

Mesial CEJ

2

4

PL

Distal interproximal

2

ML

Mesial interproximal

1

M1L

Distolingual gross coronal

2

I2L

Lingual CEJ

1

MR

Distal interproximal

1

T2 68-002A T2 68-003A

1

1

ML

Mesiobuccal root

2

M2R

Mesial gross cervical

2

M3L

Distobuccal gross coronal

2

M2R

Buccal gross cervical

2

M3R

Buccal CEJ

1

M1R

Occlusal (central fossa)

1

M1L

Occlusal (central fossa)

1

1

MR

Gross caries

3

1

T14 67-002A

T14 69-002A T14 69-003A

ML

Gross caries

3

M2R

Distal CEJ

2

C1R

Distal CEJ

2

PR

Distal CEJ

2

HO 70-901C

M3R

Buccal CEJ

2

HO 70-902E

M2R

Buccal CEJ

1

HO 70-902F

P3R

Mesial CEJ

1

HO 70-902I

MR

Mesial interproximal

1

3

ML

Occlusal (distal groove)

1

M3L

Occlusal (distal groove)

1

LOU 69-801A HO 70-901A

HO 70-902J

1

4

1

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

329

TABLE 6.13 (CONT.) Affected Tooth

Burial No.

Site of Lesion

C1R

Distal root

1

P3R

Mesial CEJ

2

4

PR

Distal CEJ

2

PR

Mesial CEJ

2

P4L

Mesial interproximal

1

M2R

Gross cervical

3

P4R

Distal CEJ

2

Occlusal (mesial fossa)

1

Occlusal (distal fossa)

1

Occlusal (buccal groove)

1

M1R

Occlusal (lingual groove)

1

Occlusal (lingual groove)

1

M2L

Buccal CEJ

1

P4L

Gross caries

3

ML

Mesial gross cervical

2

4

HO 70-902?

P3L

1

WF 62-001A

TC 60-001A

Size of Lesion1

P4R

Gross caries

3

M1R

Mesial gross cervical

3

M1R

Mesial interproximal

1

1 Size abbreviations: 1 = small pit or fissure; 2 = medium–large cavity destroying less than half the crown (or root); 3 = large cavity destroying over half the crown (or root); 4 = complete destruction of crown, leaving only the roots (Metress and Conway 1975, no. 5; Lukacs 1989, p. 267).

TABLE 6.14. FREQUENCY OF CARIOUS TEETH OF LATE ROMAN TO EARLY BYZANTINE DATE BY DENTAL CLASS Maxillary Dentition Total Teeth Tooth Class Erupted (n)

Total Carious Teeth (n)

Mandibular Dentition

Caries Prevalence Total Teeth (%) Erupted (n)

Total Carious Teeth (n)

All Dentition

Caries Prevalence Total Teeth (%) Erupted (n)

Total Carious Teeth (n)

Caries Prevalence (%)

Incisor

74

3

4.05

55

0

0.00

129

3

2.33

Canine

48

3

6.25

39

2

5.13

87

5

5.75

Premolar

78

8

10.26

72

3

4.17

150

11

7.33

Molar

131

14

10.69

134

14

10.45

265

28

10.57

Total

331

28

8.46

300

19

6.33

631

47

7.45

(47/631). The frequency is even higher among adults (47/501, or 9.38%).103 Among Late Roman skeletons with preserved teeth, 34.21% (13/38) of all individuals, 52.00% (13/25) of all adults, and 7.40% (33/446) of all teeth display caries. The distributions of caries by dental class and lesions by site were also recorded (Tables 6.14, 6.15).104 Caries is more prevalent in maxillary teeth than in mandibular teeth. The most frequently carious tooth class overall is molars (10.57%), though upper premolars 103. These calculations are only slightly different if the two carious teeth in LOU 69-801A and TC 60-001A are included. 104. The distributions of caries by dental class and dental

site were generally similar during both the Late Roman and Early Byzantine periods.

330

TEETH AND ORAL HEALTH

TABLE 6.15. FREQUENCY OF CARIOUS LESIONS OF LATE ROMAN TO EARLY BYZANTINE DATE BY DENTAL SITE Caries in Incisors (n)

Caries in Canines (n)

Caries in Premolars (n)

Caries in Molars (n)

Total Lesion (n)

Prevalence of Carious Lesions (% of 52 total)

Occlusal sites, including grooves, pits, and cusps

0

0

2

10

12

23.08

Buccal/lingual coronal sites, minus grooves

0

0

0

0

0

0.00

Interproximal sites, below the contact point minus CEJ

0

0

3

3

6

11.54

Gross coronal caries

0

0

0

1

1

1.92

Total Coronal Sites

0

0

5

14

19

36.54

Tooth Site

22 15 7

4 6

7 0

2 0

2 1

Mesial/distal Buccal/lingual

10

7

2

3

All CEJ sites

41.31 28.85 13.46

Gross cervical caries

0

1

0

5

6

11.54

Total Cervical Sites

3

3

7

15

28

53.85

Root surface

0

1

0

2

3

5.77

Gross caries

0

0

0

2

2

3.85

TABLE 6.16. FREQUENCY OF CARIOUS TEETH OF LATE ROMAN TO EARLY BYZANTINE DATE BY AGE Age Class (years)

Total Individuals (n)

Total Teeth (n)

Carious Teeth (n)

Prevalence of Carious Teeth (%)

0–14

14

128

0

0.00

15–24

6

119

6

5.04

25–34

4

98

3

3.06

35–44

12

110

6

5.45

45+

6

86

13

15.17

1

2

Since it is uncertain whether HO 70-902D or I is 25–34 years old, the dental data from neither were included in this calculation. 2 Since it is uncertain whether HO 70-902E or F is 45+ years old, the dental data from neither were included in this calculation. The calculation does include NEG 69–001D, aged mature adult but probably over 45 years old. 1

TABLE 6.17. FREQUENCY OF CARIOUS TEETH OF LATE ROMAN DATE BY AGE Age Class (years)

Total Individuals (n)

Total Teeth (n)

Carious Teeth (n)

Prevalence of Carious Teeth (%)

0–14

13

119

0

0.00

15–24

4

70

2

2.86

25–34

2

34

3

8.82

35–44

10

65

4

6.15

45+

4

52

5

9.62

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

331

TABLE 6.18. FREQUENCY OF CARIOUS TEETH BY PERIOD AND SEX Period and Sex

Individuals (n)

Total Teeth (n)

Total Prevalence of Carious Teeth (n) Carious Teeth (%)

Late Roman Males

10

120

16

13.33

Late Roman Females

14

171

6

3.59

38

446

33

7.40

Early Byzantine Males

6

142

11

7.75

Early Byzantine Females

2

34

3

8.82

9

185

14

7.57

Total Late Roman

Total Early Byzantine Late Roman–Early Byzantine Males

16

262

27

10.31

Late Roman–Early Byzantine Females

16

205

9

4.39

47

631

47

7.45

Total Late Roman–Early Byzantine

exhibit cavities with comparatively high frequency (10.26%). The accuracy of this distribution is difficult to assess because the adjusted total survival rates for all tooth classes is below 50%, the lowest being for incisors (31.86%; Table 5.2). Nevertheless, the progression in the prevalence of caries from front to back among surviving teeth probably arose from the increasing susceptibility of dentition due to the distinctive morphology of each dental class. Most often carious lesions with an identifiable point of origin occur in the cervical (53.85%) rather than in the coronal region (36.54%). Among cervical lesions, a large fraction developed near the CEJ (41.31%), most often between teeth; among coronal lesions, they occur most often on the occlusal surface (23.08%) and considerably less frequently on the interproximal enamel (11.54%). In a few cases, caries afflicted larger areas of the dentition, creating adjacent or coalescent lesions on separate teeth in the same jaw (e.g., M2R–M3R in NEG 69-004C; Figs. 6.3, 6.37).105 Dental caries progressed with age among the Late Roman to Early Byzantine residents of the Isthmus (Tables 6.16, 6.17). It never occurs in subadult skeletons and seldom in young adults. Exceptions are a young man (T14 69-003A) whose maxillary first molars displayed gross caries and two young women (NEG 69-005B, HO 70-902J) with small to medium cavities on the occlusal surfaces of their molars. In the Late Roman to Early Byzantine dental sample, the frequency of caries increases significantly between the late 30s (5.45%) and late 40s (15.17%).106 As will be seen, this pattern broadly coincides with an increasing prevalence of alveolar defects and AMTL, which in many cases probably resulted from dental caries. The advancement of caries by age in the Late Roman sample appears to be more gradual, but it is important to stress the small size of this subset of the total sample. Many men and women who reached mature adulthood acquired multiple lesions, such as NEG 67-001C (7; Fig. 6.38), NEG 69-004C (4; Fig. 6.3), and T14 67-002A (7; Figs. 6.16, 6.39). In some cases these lesions were so severe that they must have gravely altered the quality of life. Adjacent lesions in the upper right back molars in one mature adult male (NEG 69-004C) leading to alveolar resorption and abscessing caused so much pain that he preferred to eat on the left side of his mouth. This in turn caused the accumulation of massive buccal calculus in his right teeth and differential attrition in his left teeth (Figs. 6.3, 6.37). Among skeletons of earlier date, one mature woman (WF 62-001A) displayed severe cavities directly associated with alveolar bone loss and periapical fistulae that together would have caused great pain and perhaps even systemic illness (Figs. 6.40, 6.46). 105. Another example of earlier date is P4L–M1L and P4R–M1R in WF 62-001A (Figs. 6.33, 6.40).

106. The increase is even sharper if NEG 69-004C (four carious teeth, 35+ years) is considered.

332

TEETH AND ORAL HEALTH

The frequency of caries also varied by sex and period (Table 6.18). During the Late Roman to Early Byzantine periods, carious teeth were more common in men (10.31%) than in women (4.39%). This discrepancy by sex was particularly large during the Late Roman period, which is represented by the most teeth. The overall prevalence of carious teeth does not appear to have changed over time from the Late Roman to the Early Byzantine periods, but the dental sample from the later period is too small to serve as a basis for secure conclusions about diachronic change. Several factors contributed to dental caries among the local residents. The impact of eating behavior and hygiene on the prevalence of caries is unknown. Whatever habits of cleaning teeth were probably too inconsistent or insubstantial to prevent tooth decay effectively. Also unknown is the level of natural fluoride in the ancient water supply, which would have inhibited the development of caries among local residents. Were the level high they would have consumed large amounts of fluoride if they regularly ate marine foods.107 There are no instances of caries at the site of hypoplastic lesions, and only in a few cases did occlusal cavities develop in large areas of dentine exposed by attrition or abrasion (Figs. 6.16, 6.18, 6.44).108 One major cause of dental caries was dietary carbohydrates. The most cariogenic carbohydrates are sugars, which oral bacteria can quickly metabolize. While the ancient residents did not consume high quantities of sucrose and refined sugars like modern peoples, they did ingest glucose and fructose. These sugars occur in small amounts in fruits available on the Isthmus, such as fig, grape, and pomegranate, and are highly concentrated in honey, a common sweetener that was harvested in the settlement at the Fortress.109 Complex carbohydrates, namely starches, are not directly metabolizable, but salivary and bacterial enzymes eventually break them down if they remain in the mouth. Starches were probably the main cariogenic element in the local diet at the Isthmus. The discovery of rotary querns shows that residents consumed cereal grains, and comparison with other Byzantine Greek sites and historical sources suggests that they probably also ate roots and tubers (onion, garlic, radish, turnip, carrot), leafy vegetables (cabbage, chard, cardoon), and beans. Residents also consumed fats, oils, and proteins in meats, olives, and dairy products, which are all cariostatic, but they would not have consumed them exclusively or even predominantly. The distribution of carious lesions by location on teeth reflects the texture of the local diet (Table 6.15). The low to moderate overall level of dental attrition shows that foodstuffs were not highly coarse but were more abrasive than the soft, sticky foods that characterize the diet of most modern industrial societies. The relatively low prevalence of occlusal caries at the Isthmus means that small amounts of gummy foods, such as honey or fig, adhered to the uneven biting surfaces of the posterior teeth. The more common pattern of decay began when starchy particles from, for instance, wheat bread or carrots collected along the gingival sulcus or settled between the teeth, where they remained until they were broken down by enzymes and metabolized by plaque bacteria. The carious process was facilitated by the continuous eruption of teeth with advancing attrition, which exposed the cervix and increased the interproximal space where food debris could become impacted. Consequently, a majority of the lesions in the teeth from the Isthmus occur at cervical sites, most commonly on the mesial and distal sides of the neck, while many at coronal sites were also interproximal above the CEJ. 107. Although Athenian drinking water is poor in fluoride, Athenians since antiquity have gained a healthy dietary supplement from sea salt and fish (Hadjimarkos and Bonhorst 1962; Grmek 1989, p. 118).

108. T14 67-002A (M3L, P4L, M2R), 69-002A (M1R, M1L). 109. Brothwell and Brothwell (1998, pp. 73–84) generally discuss sources of sugar in the ancient diet, stressing the prominence of honey.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

333

The prevalence and distribution of caries among the local residents is indicative of a mixed agricultural subsistence. Surveys of numerous skeletal samples from across the world have shown that the nature of decay varies according to subsistence strategies and dietary regimes.110 Prehistoric or historic peoples who depend on hunting and gathering but consume a meat-rich, carbohydrate-poor diet typically have a low caries rate. Agriculturalists who depend on cereal grains and other starchy foods typically have a higher caries rate, with most lesions occurring at cervical sites. Modern peoples who consume large quantities of carbohydrates, including refined sugars, have the highest caries rate, with most lesions occurring on the occlusal surfaces. The overall caries rate for the Late Roman to Early Byzantine dental sample at the Isthmus (7.45%) falls between the rates calculated for peoples practicing a mixed subsistence (4.4%) and those practicing agriculture (8.6%).111 The overall frequencies of teeth with calculus (23.77%, though probably an underestimate) and carious teeth at the Isthmus compare with the moderate calculus and low caries rate among prehistoric islanders on the Arabian Gulf who depended on a mixed subsistence of pastoralism or fishing and agriculture.112 Likewise, the caries rate at the Isthmus falls near the threshold (7%) between archaeological groups of eastern Native Americans who were foragers and those who were agriculturalists.113 Caries rates approximating those at the Isthmus are also exhibited in skeletal samples both from Roman, Anglo-Saxon, and Medieval sites in Britain and from a Roman site in central Italy, where residents predominantly practiced agriculture but consumed a wide range of foodstuffs.114 While all these groups lived in different environments and developed their own economies, their basic strategies for food production and their consumption of carbohydrates were fundamentally similar. However, the residents of the Isthmus did not display as high a caries rate as many groups with a cereal-rich diet, such as the Late Roman inhabitants of Poundbury Camp at Dorchester, England (15.8%) or the Early Medieval inhabitants of La Selvicciola in northern Latium, Italy (12.6%).115 This indicates either that sugary, sticky foods like honey and figs were relatively infrequent in the mostly starchy diet at the Isthmus, or that cariostatic foods, such as mutton and pork, olive oil, dairy, and fish, were relatively frequent. The different frequencies of caries in men and women may reflect dietary and behavioral variation among the inhabitants of the Isthmus. The higher caries rate among men than among women, particularly during the earlier era of occupation, contradicts the usual pattern of higher cariosity among women than men in archaeological groups. This difference has been explained by the fact that men are chiefly occupied with hunting and therefore eat more meat than women, whereas women are chiefly occupied with planting, collecting, and processing field or garden products and therefore eat more starchy plants than men.116 One possible explanation for the higher caries rate among men at the Isthmus is that men consumed more cariogenic foods than women, such as honey, which they might have harvested. Another possible explanation is that men ate more frequently than women 110. E.g., Powell 1985, pp. 316–321; Hillson 1996, pp. 280–283; Larsen 1997, pp. 67–72. Moore and Corbett (1971, 1973, 1975) construct a useful epidemiological survey of caries among Ancient to Modern British groups. 111. Turner 1979, p. 622, table 3. 112. Littleton and Frohlich 1993, pp. 433–436, 441–442, 444–445, tables 4, 5, 8 (Group 2, esp. Failaka Island, Kuwait, ca. 300–100 b.c., and Umm an-Nar Island near Abu Dhabi, ca. 2,500 b.c.). 113. Larsen 1997, pp. 68–70, fig. 3.2. 114. Roberts and Manchester 1995, pp. 48–50, table 4.1 (survey of sites in Britain); Manzi et al. 1999, pp. 331–332, 337, tables 2, 3 (Lucus Feroniae, Latium, Italy, 1st–3rd centuries).

The data from other sites are comparable. Adult teeth from rural sites in Early Medieval (6th–8th centuries) and Late Medieval (10th–11th centuries) Croatia exhibit similar cariosity (Šlaus 2002, p. 50, table 103; Vodanović et al. 2005, pp. 671– 677, tables 3–7). The skeletons of 13th-century residents at Nicaea (İznik), who likewise depended on a mixed diet, display a somewhat higher caries frequency (~11%) than the skeletons from the Isthmus (Duyar and Erdal 2003, pp. 60–63, table 1; Caglar et al. 2007). 115. Molleson 1993, pp. 183–184 (Poundbury, mostly 4th century); Manzi et al. 1999, pp. 331, 332, 335–336, table 3 (La Selvicciola, 7th century). 116. Larsen 1997, pp. 72–76.

334

TEETH AND ORAL HEALTH

but cleaned their teeth less regularly, which led to the dense impaction of food debris. Perhaps men ate more meals than women because they worked away from the home in fields, where they could gather wild edibles or snack on foods they brought from home during breaks from physical labor. Furthermore, Late Roman men displayed slightly heavier dental attrition than contemporary women, which partly reflects the slightly higher average age of the male sample of this period. Heavier wear in men than in women would lead to greater continuous eruption of teeth, which would expose cervical and interproximal areas in male teeth to caries more extensively than in female teeth. It is difficult to trace changes in local subsistence over time by reference to the frequency of caries in the Late Roman and Early Byzantine periods because the later dental sample is very small. Nonetheless, the surviving evidence does not point to an increase in the prevalence of caries. The increase in oral plaque suggested by an increasing frequency of calculus (Table 6.11) does not therefore seem to have led to more prevalent tooth decay among Early Byzantine residents than among Late Roman residents.

Alveolar Defects and AMTL Among all peoples, advancing age leads to the development of defects in the alveolar processes that involve the remodeling of crestal bone and the underlying cancellous tissue. Sometimes alveolar bone is degraded to such a degree that the root loses all structural support and the tooth falls out (AMTL). If the tooth exfoliates around the time of death, the socket remains open. If it exfoliates well before death, the bone surrounding the socket resorbs and eventually disappears, leaving a smooth, rounded surface. Alveolar defects and AMTL are detectable in archaeological bone. Alterations to the alveolar process, however, have a complex etiology. Anthropologists need to discern whether alveolar defects and AMTL observed in skeletal specimens originated in periodontal or pulpal pathology, though confident diagnosis is seldom possible. Periodontal disease is an inflammatory reaction to one or more irritants in the tissues surrounding the teeth. Its first stage is gingivitis, the inflammation of the thin gingival membrane cupping the inferior margin of the dental crown.117 Eventually a gap develops in the periodontal membrane around the CEJ, exposing the cementum and promoting the accumulation of sub-gingival plaque. In the advanced stage of the disease, called chronic periodontitis, destructive inflammation spreads into the connective and bony tissues and the root is gradually denuded.118 Advanced periodontitis can induce abscess formation at the lateral alveolar margin or between roots and even the evulsion of the infected tooth. Over time the alveolar bone displays macroporosity and periostitis and begins to resorb. Early in the process of resorption, the dense cortical plate thins at the interproximal crest and becomes porotic and either sharp and ragged, if inflammation is active, or smooth, if it is quiescent. As bone loss progresses, the buccal, lingual, and interproximal alveolar septa deteriorate either in a horizontal pattern involving several adjacent teeth or in a vertical pattern concentrating around single or neighboring teeth, independent of any horizontal or periapical activity. Irregular defects in the bone lining the socket result in a crater, trench, or moat around the tooth, and those involving the outer cortical plates result in a ramp or plane in the margin. 117. Hillson 1996, pp. 262–263; 2000, pp. 268–269; Soames and Southam 1998, pp. 99–104. 118. Hildebolt and Molnar 1991, pp. 226–227; Hillson

1996, pp. 263–267; Soames and Southam 1998, pp. 104–107, 113–114.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

335

The progression from gingivitis to periodontitis in mature adults, sometimes culminating in AMTL, is the most common form of periodontal disease in modern populations. There are several causes. Dental plaque is an essential factor in gingivitis, and when plaque mineralizes into calculus it can irritate contiguous soft tissues. As has been discussed, the prevalence of plaque is influenced by oral hygiene and carbohydrate consumption. The progression of periodontal lesions can also be affected by other local and systemic factors that promote plaque and affect immune response, such as dental anatomy and occlusal relationships, heredity, vitamin and protein deficiencies, hormonal imbalances, age, trace elements, and blood diseases.119 Pulpal pathology also leads to the destruction of dental tissues, but its origin and formation differ from those of periodontal disease.120 It begins when dental wear, caries, or trauma expose the pulp chamber to bacterial infiltration, infection, and inflammation. Persistent pulpitis can induce inflammation in the adjacent periodontal ligament and gingivae, which in turn promotes osseous lesions not unlike those caused by periodontitis. Once pulpitis spreads to the root canal, bacteria can pass through the apical foramen, inflame the surrounding tissues, and alter alveolar bone in various ways.121 Chronic periapical inflammation can lead to the production of a periapical granuloma. The granulation tissue destroys the thin lamellar bone lining the alveolus and forms a small, smooth chamber in the resorbed bone measuring up to ca. 2–3 mm in diameter beyond the wall of the root. Sometimes this remodeling creates a small fenestration in the buccal plate. The periapical granuloma can remain stable or develop in two directions. One is the replacement of the granulation tissue by fluid and the expansion of the chamber to form a radicular cyst, or a smooth-walled cavity larger than ca. 3 mm in diameter beyond the wall of the root. The other is the accumulation of pus in the granulomatous cavity, creating a periapical abscess. In acute abscesses, no bone cavity forms but the pus tracks out through the trabeculi and vascular channels in the lingual or buccal plates and erupts through the oral mucosa or skin. The development of acute abscesses in granulomata or cysts makes the walls of the cavity slightly rough. In chronic abscesses, so much pus is produced that it must drain either out through the crown into the oral cavity or, more often, out through a distinct, rounded fistula in the buccal/ labial or lingual plate of the alveolar process, through the angle of the palatine process, into the maxillary sinus, or onto the buccal or lingual aspect of the mandibular body. In dry bone specimens, it can be difficult to distinguish between a fenestrated periapical granuloma and a periapical abscess with a fistula, though in the latter the walls of the cavity are slightly roughened. The deterioration of periodontal tissues and alveolar bone from such inflammatory lesions can destablilize the tooth and ultimately release it from the socket. Inflammatory edema and suppuration resulting from pulpitis can spread into the soft tissues of the face and neck. Acute periapical abscesses induced by especially virulent microorganisms can expand into neighboring bone marrow spaces and generate osteomyelitis, which in its acute form can lead to septicemia and ultimately death. It has been argued that pulpal, not periodontal, disease was the main cause of alveolar resorption and AMTL in ancient peoples.122 Alveolar resorption, periapical lesions, and AMTL are readily visible in most skeletal remains. Care should be taken not to mistake postmortem alterations for antemortem 119. Hildebolt and Molnar 1991, pp. 228, 236–237; Hillson 1996, pp. 266–269; Soames and Southam 1998, pp. 95–99. 120. Clarke and Hirsch 1991, pp. 248–251; Soames and Southam 1998, pp. 54–57, 59, 63–69; Hillson 2000, pp. 269– 271. Dias and Tayles (1997, p. 550, fig. 2) give a useful diagram

showing the developmental relationships between pulpitis and periapical infections. 121. Dias and Tayles 1997. 122. Clarke et al. 1986, pp. 182–183; Clarke and Hirsch 1991, pp. 248–251; Hillson 2000, pp. 270–271.

336

TEETH AND ORAL HEALTH

defects. The destruction of alveolar margins by erosive agents in the burial environment or by vigorous excavation and careless handling can resemble lesions and lead to the loss of teeth.123 Once antemortem defects have been accurately identified in skeletal specimens, the pathogenesis of alveolar destruction can be hard to identify. The condition of alveolar bone is influenced by physiological, periodontal, and pulpal factors. Large intervals (> 2 mm) between the CEJ and the alveolar crest (AC) normally arise from continuous eruption to compensate for dental wear and craniofacial growth throughout life. Failure to recognize the physiology of the CEJ-AC relationship has led researchers to overestimate the frequency and severity of horizontal bone loss from periodontitis in ancient skeletal samples.124 In several mature adults at the Isthmus, it was possible to distinguish hypereruption with an enlarged CEJ-AC distance caused by moderate to severe dental wear from genuine horizontal bone loss (Figs. 6.39, 6.45).125 Periodontitis can be accurately identified in skeletal remains using proper criteria. The main diagnostic features are the loss of the cortical plate to expose the porous cancellous bone and the alteration of the alveolar crest. A CEJ-AC distance greater than 2.5 mm in a relatively young individual, coupled with slight to moderate dental wear, points to pathological bone loss if the alveolar region displays characteristic remodeling. The absence of pulpal lesions further supports a gingival etiology for resorption.126 A pulpal rather than a periodontal origin for alveolar defects can be identified in skeletal remains when localized lesions occur but crestal bone is intact. Moreover, the presence of alveolar defects together with a pulp chamber exposed by trauma, wear, or caries and/or a periapical lesion point to pulpal pathology.127 Different processes of periapical bone loss can be identified from the form and size of cavities, fenestrations, and fistulae.128 All alveolar processes from the Isthmus were examined, and the form and severity of antemortem defects were recorded. Apart from a few exceptions,129 most alveolar processes were sufficiently intact for examination. First, localized defects involving alveolar remodeling and bone loss were described and diagnosed according to the Karn typology,130 and resorptive alterations to interproximal crests were classified according to the five Kerr classes.131 These defects were recorded only in cases where the affected tooth was preserved or was lost shortly before death, so that only slight to moderate remodeling had occurred. Second, the location, size, and form of periapical lesions were documented and diagnosed according to a standardized procedure.132 Third, instances of AMTL and the degree of resorption were recorded on a four-point scale from slight to complete. In the end, an attempt was made to determine whether alveolar defects and AMTL had a periodontal or a pulpal origin based on the nature of deformation and the presence or absence of calculus, caries, severe attrition, and pulp exposure. The total skeletal sample from the Isthmus displays several different alveolar defects and numerous cases of AMTL in 27 adults but no subadults. The full range of periodontal and periapical lesions were observed in 78 teeth in 18 skeletons from the site (Table 6.19). Such 123. In one mandible from the Isthmus (NEG 69-004E; Fig. 6.5) it is hard to distinguish extensive postmortem abrasion and fracturing to the left alveoli from genuine antemortem defects. In examining this individual, therefore, no defects were recorded. 124. Clarke et al. 1986; Watson 1986; Clarke and Hirsch 1991. Clarke and Hirsch (1991, pp. 247–248) note other developmental factors that can reduce the alveolar margin, including dehiscence, fenestration, root proximity, and enamel pearls. 125. NEG 69-004B, 67-003B, 69-001D, T14 67-002A, LOU 69-801A, HO 70-901A–C, 70-902H, TC 60-001A. 126. Clarke and Hirsch 1991, pp. 256–259; Hildebolt and

Molnar 1991, p. 232; Lavigne and Molto 1995. 127. Clarke and Hirsch 1991, pp. 253–254. 128. Dias and Tayles 1997, p. 553; Hillson 2000, pp. 269– 271, 278. 129. T2 68-003B, 68-006A, DEC 69-901A, B, RB 76-002A–C, HO 70-901A, C, 70-902D–J, WF 62-001A. Only one tooth and no maxillary or mandibular bone survived from NEG 67-003A. 130. Karn et al. 1984, summarized in Hillson 1996, p. 267, table 12.1. 131. Kerr 1988; 1991, pp. 347–348, adapted in Hillson 1996, p. 268, table 12.2. 132. Hillson 2000, pp. 269–271, 278, following Dias and Tayles 1997.

CONDITIONS OF THE TEETH AND ALVEOLAR STRUCTURES

337

lesions occur in 70 teeth in 16 Late Roman to Early Byzantine skeletons, or in 53 teeth in 12 Late Roman skeletons. A large majority (82.86%) of these 70 affected teeth are posterior (19 premolars and 39 molars), while few anterior teeth (6 incisors and 6 canines) show defects. Most defects and lesions occur in maxillary teeth (39/70, or 55.71%).133 The commonest periodontal defects among the Late Roman to Early Byzantine skeletons are localized irregularities in socket walls, cortical plates, and underlying cancellous bone, while horizontal bone loss involving a row of cheek teeth occurs at only three locations in two individuals (NEG 67-001C, 69-004C). Various defects in individual or neighboring alveoli were diagnosed as gingivitis (9), acute, chronic,134 or quiescent periodontitis (21), or aggressive periodontitis in an acute or quiescent state (11). The commonest apical cavities were created by small granulomata at or beside the tip of the root, but often these developed into larger cysts. Various periapical lesions were diagnosed as periapical granulomata sometimes with abscesses, opened by fenestrations or fistulae (12), and periodontal cysts sometimes with abscesses, opened by fistulae or larger eruptions (8). Since apical cavities were observed without X-rays, only those instances visible through open sockets or exposed by fenestrations or fistulae were recorded.135 Therefore, more periapical lesions might have gone undetected in the dental sample. There are no instances of infection spreading from abscesses to local bone marrow to produce osteomyelitis. The extensive destruction of alveolar bone led to the antemortem loss of 117 teeth in 24 adults from the site (Tables 5.2, 6.20). Within the Late Roman to Early Byzantine dental sample, 22 adults suffered AMTL in 112 teeth; within the Late Roman dental sample, 18 adults suffered AMTL in 88 teeth. A majority of the teeth lost before death in skeletons of Late Roman to Early Byzantine date were molars (64/112, or 57.14%), followed by premolars and incisors in equal proportion (21/112, or 18.75%), and a very small fraction of canines (6/112, or 5.36%). Half of the lost teeth were maxillary (56/112, or 50.00%).136 In 14 of these lost teeth, alveolar defects or periapical cavities were noted in association with the slightly or moderately resorbed sockets. In the remaining 98 cases, no defects or cavities are visible, either because they never existed or because they were obliterated by resorptive remodeling. All stages of resorption following AMTL were observed, from nascent or active remodeling (e.g., Figs. 6.9, 6.27) to advanced or complete ossification (e.g., Figs. 6.1, 6.13). Alveolar defects and AMTL were relatively common among adult residents of the Late Roman and Early Byzantine periods. The frequency of skeletons with either alveolar defects or AMTL among those with preserved alveolar processes is 50.00% (24/48), or 68.57% of all adults with preserved alveolar processes (24/35). Among Late Roman skeletons, the frequencies are 48.72% (19/39) and 70.37% (19/27). Within the Late Roman to Early Byzantine sample, the frequencies of skeletons with defects of periodontal origin, defects of pulpal origin, and AMTL are, respectively, 29.17% (14/48), 25.00% (12/48), and 45.83% (22/48). These 48 individuals display in total 168 defective alveoli, for an average of 3.50 per individual or 4.80 per adult. Within the Late Roman sample, the frequencies of skeletons with defects of periodontal origin, defects of pulpal origin, and AMTL are, respectively, 133. Similar distributions of lesions by dental region are apparent in the Late Roman sample. 134. The term “chronic periodontitis” generally describes an established disease that progressively destroys periodontal tissues and alveolar structures, though in fact destruction occurs in separate phases of acute bursts and quiescence evolving from a nascent to an aggressive state. “Chronic periodontitis” is here applied to the diagnosis of alveolar defects such as craters, trenches, and ramps that formed over a longer time

than a single intensive episode (see Dias and Tayles 1997, pp. 552–553). The texture and form of defective septa provide a more precise sense of the acute or quiescent nature of pathogenesis. 135. E.g., a peridontal cyst without fenestration or fistula is apparently located at the apex of the lingual root of M1L in T14 69-003A. 136. Similar distributions of lesions by dental region are apparent in the Late Roman sample.

NEG 69-004C

NEG 67-001C

Burial No.

Description of Defects

Tooth lost postmortem; apparently a labial cratered ramp to D. ca. 6.0 mm; both interproximal walls are rough, porous and sloping inward sharply (class 5) C1R Tooth lost postmortem; medium-sized periapical cavity with buccal fistula showing slightly roughened walls (Diam. ca. 2.3 mm allowing for estimated root W.); mesial interproximal wall shows many small foramina and small grooves (class 2) C1L Tooth lost postmortem; large, elliptical periapical cavity with buccal sinus showing ragged but distinct margins, rough, porous walls, small perforation into left nasal aperture apparently antemortem (W. max. 8.1 mm x D. max. 11.0 mm from pres. AC); mesial interproximal wall is porous and ragged, sloping sharply to D. 10.1 mm (class 5) I1R, I1L, I2R, AMTL and slight-moderate resorption (1–2); all labial and lingual margins are sharp and I2L ragged, showing only slight postmortem damage, possible lingual and labial ramp to D. ca. 9.5 mm; interproximal walls are sharp, ragged, and sloping to D. 3–5 mm (class 5) C1R AMTL and slight resorption (1); medium-sized periapical cavity with buccal fistula showing slightly roughened walls (Diam. ca. 2.5 mm allowing for estimated root W.); buccal plate probably displayed a sinus but unclear because of postmortem damage; mesial interproximal wall is gone (cf. I2R) but distal interproximal wall shows many small foramina and small grooves (class 2) C1L Medium-sized periapical cavity with buccal fistula showing slightly roughened walls (Diam. ca. 2.2 mm allowing for root W.), though note that alveolus distorted by long diagonal crack through mental eminence; buccal plate probably displayed a sinus but unclear because of postmortem damage; mesial interproximal wall is gone (cf. I2R) but distal interproximal wall shows many small foramina and small grooves (class 2) P3R, P4R, M1R, P3R lost postmortem but others survive; large CEJ-AC interval in surviving teeth (P4R M2R, M3R = 3.9 mm, M1R = 8.0 mm, M2R = 7.7 mm, M3R = 5.9 mm); somewhat irregular and porous buccal margin with porosity, esp. around P4R, M1R; mesiobuccal trench opening furcation of M1R to D. 7.8 mm; all interproximal walls show numerous foramina and grooved or sharp, ragged edges though slightly rounded between molars (class 3–4) P3L, P4L, M1L, M3L lost postmortem but others survive; large CEJ-AC interval in surviving teeth (P3L M2L, M3L = ca. 5.6 mm, P4L = 5.9 mm, M1L = 6.3 mm, M2L = 6.5 mm); somewhat irregular and porous buccal margin with porosity, esp. around M1L; mesiobuccal trench opening furcation of M1L to D. 6.6 mm; all interproximal walls show numerous foramina and grooved or sharp, ragged edges though slightly rounded between molars (class 3–4); interproximal wall between M2L and M3L destroyed to D. 7.9 mm and shows numerous foramina P3R, P4R, M1R Somewhat large CEJ-AC interval in teeth (P3R = 3.7 mm, P4R = 2.5 mm, M1R = 3.4 mm); irregular, porous buccal margin with ragged edges from mesial P3R to distal P4R but slightly smoother around M1R; all interproximal walls show numerous foramina and grooves (class 2+), though walls of P3R show more sharp, ragged texture (class 2–3) 2 MR Shallow mesial ramp into narrow one-surface crater extending 1.9 mm down mesiobuccal root; mesial interproximal wall shows breakdown of contour, porosity and grooving, somewhat sharp and ragged texture around margin of crater (class 3); buccal margin is sharp and ragged, ramps distally to 5.4 mm below CEJ; distal interproximal wall shows complete, deep destruction of contour with sharp, ragged texture and frequent porosity (class 5) M3R Mesial interproximal wall shows complete, deep destruction of contour with sharp, ragged texture and frequent porosity (class 5); complete thinning of buccal plate to expose buccal roots; small–medium periapical cavity communicating with the lingual root with buccal fistula showing showing slightly porous, roughened walls (Diam. 2.6 mm allowing for root W.)

I1R

Affected Tooth

Diagnosis Plaque?

Pathogenesis

6.37

Chronic periodontitis in acute Plaque forming heavy buccal(mesial) and acute aggressive mesial calculus shelf; mesial gross cervical caries on M3R (distal) phases

Plaque forming heavy buccalmesial calculus shelf; mesial gross cervical caries exposing pulp

6.37

Plaque forming heavy buccal calculus shelf Horizontal bone loss from nascent acute periodontitis (esp. P3R)

Acute aggressive periodontitis in mesial region; fistulated acute abscess in a periapical granuloma

5.13

Cervical caries in M1L, M2L

Horizontal bone loss from chronic periodontitis

6.37

6.38

Cervical caries in P4R, M2R, M3R

Horizontal bone loss from chronic periodontitis

5.13 6.41

5.13 6.38 6.41

Cervical caries exposing pulp

Cervical caries? (cf. C1L)

5.13 6.41

5.13 6.7

5.13 6.7 5.13 6.7

Figure

Fistulated acute abscess in a periapical granuloma; gingivitis

Fistulated acute abscess in a periapical granuloma; gingivitis

Caries and/or wear? Fistulated acute abscess in a periapical granuloma; gingivitis Erupted acute abscess in peri- Caries and/or wear? Plaque? odontal cyst draining into upper left lip and nostril; acute aggressive periodontitis Acute aggressive periodontitis Plaque?

Acute aggressive periodontitis

OLAR DEFECTS

NEG 69-007A

NEG 69-005C

NEG 69-004E

NEG 69-004D

Burial No.

Description of Defects

Mesial interproximal wall shows breakdown of contour and formation of a depression with a porous, somewhat sharp texture (class 3); also a distal ramp into a crater extending to D. 5.2 mm; remainder of distal interproximal wall displays scattered foramina, ridges, and slight cupping (class 2–3) Mesial interproximal wall shows scattered foramina and ridges (class 2); also a distal ramp into broad crater extending to D. 4.7 mm; remainder of distal interproximal wall displays scattered foramina, ridges, and slight cupping (class 2) AM or postmortem tooth loss; small periapical cavity with buccal fenestration showing well-demarcated and smooth walls (Diam. ca. 1.7 mm allowing for estimated root W.) Medium-sized periapical cavity with buccal fistula showing slightly roughened, porous walls (Diam. 2.7 mm allowing for root W.), surrounding distal root Tooth lost postmortem; small periapical cavity with buccal fenestration showing welldemarcated and smooth walls (Diam. 1.8 mm allowing for estimated root W.) Large moat around all roots to D. ca. 11.9 mm on buccal side and D. ca. 7.3 mm on lingual side; mesial interproximal walls destroyed to D. ca. 6.2 mm and distal interproximal wall destroyed to D. ca. 8.9 mm, both showing porous, jagged, somewhat sharp texture (class 5) Distal ramp and moat defect to D. 8.3 mm; all walls show numerous foramina, but mesial is more smooth and distal is more rough (class 2–3)

ML

Fenestrated periapical granuloma Fistulated acute abscess in periapical granuloma Fenestrated periapical granuloma Acute aggressive periodontitis

Gingivitis progressing into periodontitis

Chronic periodontitis in acute (mesial) and nascent acute (distal) phases

Diagnosis

M2R

P4R

C1R, P3R

M1L

Only the mesial root of M1L survives from a point ca. 1.5 mm below the CEJ, entire crown and distal root are worn away; slight resorption in distal portion of alveolus; broad buccal ramp across both roots to D. ca. 13.0 mm, showing rounded margin; interproximal walls show ragged, sharp texture and considerable breakdown of contour, esp. distal (class 3) AMTL and slight resorption of C1R (1); large area of irregular destruction with porosity, ridges, and a somewhat smooth texture ramping down at a sharp angle from distal I2R and mesial P4R to D. ca. 11.0 mm (class 5); before slight postmortem abrasive damage of alveolar bone, AMTL teeth probably showed a ramp and/or crater defect; a partly resorbed periapical cavity apparently located at P3R root; impossible to know original form, but size is smaller than the adjacent cavity at P4R root (Diam.