Aphrodite’s Kephali: An Early Minoan I Defensive Site in Eastern Crete (Prehistory Monographs) [Illustrated] 9781931534710, 1931534713

Citation preview

Aphrodite’s Kephali An Early Minoan I Defensive Site in Eastern Crete

Aphrodite’s Kephali An Early Minoan I Defensive Site in Eastern Crete

Frontispiece. Pithos 77 (AK 9) from Aphrodite’s Kephali. Photo P. Betancourt.

PREHISTORY MONOGRAPHS 41

Aphrodite’s Kephali An Early Minoan I Defensive Site in Eastern Crete

by Philip P. Betancourt

with contributions by Kostas Chalikias, Heidi M.C. Dierckx, Andrew J. Koh, Evi Margaritis, Floyd W. McCoy, Eleni Nodarou, and David S. Reese

Published by INSTAP Academic Press Philadelphia, Pennsylvania 2013

Design and Production INSTAP Academic Press, Philadelphia, PA Printing Hoster Bindery, Inc., Ivyland, PA

Library of Congress Cataloging-in-Publication Data Betancourt, Philip P., 1936Aphrodite’s Kephali : an early Minoan I defensive site in eastern Crete / by Philip P. Betancourt ; with contributions by Kostas Chalikias, Heidi M.C. Dierckx, Andrew J. Koh, Evi Margaritis, Floyd W. McCoy, Eleni Nodarou, and David S. Reese. pages cm. — (Prehistory monographs ; 41) Includes bibliographical references and index. ISBN 978-1-931534-71-0 (alk. paper) 1. Minoans—Greece—Crete—Antiquities. 2. Minoans—Material culture. 3. Excavations (Archaeology)—Greece—Crete. 4. Fortification—Greece—Crete. 5. Bronze age—Greece—Crete. 6. Crete (Greece)—Antiquities. I. Title. DF221.C8B538 2013 939’.18—dc23 2012040345

Copyright © 2013 INSTAP Academic Press Philadelphia, Pennsylvania All rights reserved Printed in the United States of America

Contents

List of Tables in the Text. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vii List of Figures in the Text. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xi Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xvii Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xix List of Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xxi INTRODUCTION 1. Introduction, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 2. The Isthmus of Ierapetra, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 3. Geology and Geologic History, Floyd W. McCoy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 4. The FN/EM I Settlement Patterns in the Northern Part of the Isthmus of Ierapetra, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 5. The FN/EM I Settlement Patterns in the Southern Part of the Isthmus of Ierapetra, Kostas Chalikias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 THE EVIDENCE 6. The Excavation of the Site, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53 7. The Architecture, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57 8. The Pottery, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75

vi

APHRODITE’S KEPHALI

9. The Ground and Chipped Stone Tools, Heidi M.C. Dierckx. . . . . . . . . . . . . . . . . . . . .101 10. The Faunal Remains, David S. Reese. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .109 11. Arboriculture at Aphrodite’s Kephali, Evi Margaritis. . . . . . . . . . . . . . . . . . . . . . . . . . .111 DISCUSSION AND INTERPRETATION 12. Hillforts and Watchtowers, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . . . . . . . . . . .117 13. The Place of Aphrodite’s Kephali in the Early Development of Fortifications, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123 14. The Significance of Aphrodite’s Kephali, Philip P. Betancourt. . . . . . . . . . . . . . . . . . . .131 Appendix A. Petrographic Analysis of the Pottery, Eleni Nodarou. . . . . . . . . . . . . . . . . . . . . . . .151 Appendix B. Gas Chromatography Analysis of the Pottery, Andrew Koh and Philip P. Betancourt. . .171 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .225 Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .243

List of Tables in the Text

Table 3.1.

The geologic time scale for the Cenozoic Era, with subdivisions into periods, epochs, and stages, and identification of notable geologic events in the tectonic and geologic history of the area surrounding the Aphrodite’s Kephali archaeological site. . . .16

Table 3.2.

Sedimentary rock classification scheme. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29

Table 3.3.

Summary of data on rocks exposed and sampled at Aphrodite’s Kephali. . . . . . . . . . . . .31

Table 10.1. Faunal remains from Aphrodite’s Kephali. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .110 Table 11.1

Archaeobotanical remains by context: complete/fragmented. . . . . . . . . . . . . . . . . . . .112

Table A.1.

Concordance for the vessels sampled for petrographic analysis. . . . . . . . . . . . . . . . . .152

Table B.1.

GC-MS chromatogram for 2 (AK 4; ARCHEM sample 139, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177

Table B.2.

GC-MS peak report for 2 (AK 4; ARCHEM sample 140, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .179

Table B.3.

GC-MS peak report for 10 (AK 12; ARCHEM sample 157, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .181

viii

APHRODITE’S KEPHALI

Table B.4.

GC-MS peak report for 10 (AK 12; ARCHEM sample 158, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .183

Table B.5.

GC-MS peak report for 17 (AK 21; ARCHEM sample 163, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .185

Table B.6.

GC-MS peak report for 17 (AK 21; ARCHEM sample 164, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .187

Table B.7.

GC-MS peak report for 35 (AK 7; ARCHEM sample 145, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .189

Table B.8.

GC-MS peak report for 35 (AK 7; ARCHEM sample 146, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .191

Table B.9.

GC-MS peak report for 38 (AK 11; ARCHEM sample 155, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .193

Table B.10. GC-MS peak report for 38 (AK 11; ARCHEM sample 156, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .195 Table B.11. GC-MS peak report for 51 (AK 20; ARCHEM sample 161, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .197 Table B.12. GC-MS peak report for 51 (AK 20; ARCHEM sample 162, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .199 Table B.13. GC-MS peak report for 52 (AK 8; ARCHEM sample 147, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .201 Table B.14. GC-MS peak report for 52 (AK 8; ARCHEM sample 148, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .203 Table B.15. GC-MS peak report for 60 (AK 5; ARCHEM sample 141, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .205 Table B.16. GC-MS peak report for 60 (AK 5; ARCHEM sample 142, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .207 Table B.17. GC-MS peak report for 61 (AK 1; ARCHEM sample 135, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .209 Table B.18. GC-MS peak report for 61 (AK 1; ARCHEM sample 136, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .211 Table B.19. GC-MS peak report for 77 (AK 9; ARCHEM sample 149, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .213 Table B.20. GC-MS peak report for 77 (AK 9; ARCHEM sample 150, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .215 Table B.21. GC-MS peak report for 87 (AK 19; ARCHEM sample 159, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .217 Table B.22. GC-MS peak report for 87 (AK 19; ARCHEM sample 160, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .219

LIST OF TABLES IN THE TEXT

ix

Table B.23. GC-MS peak report for 90 (AK 6; ARCHEM sample 143, DCM) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .221 Table B.24. GC-MS peak report for 90 (AK 6; ARCHEM sample 144, E) showing total ion current (TIC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .223

List of Figures in the Text

Figure 1.1. Map of Crete. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 Figure 1.2. Map of eastern Crete. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 Figure 1.3. Topographical map of the isthmus of Ierapetra. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 Figure 1.4. The site of Aphrodite’s Kephali as seen from the north–south road near modern Episkopi. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6 Figure 1.5. The road to Aphrodite’s Kephali from the northwest. . . . . . . . . . . . . . . . . . . . . . . . . . . .6 Figure 1.6. The site, looking south, with olive groves around it. . . . . . . . . . . . . . . . . . . . . . . . . .6 Figure 1.7. Consolidating the architecture in 2006. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 Figure 2.1. The Ottoman castle at Episkopi, on the crest of a hill overlooking the north–south highway. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 Figure 2.2. Pseira and Chrysokamino as seen from Alatsomouri, looking north. . . . . . . . . . . . . . . .11 Figure 2.3. The village of Pacheia Ammos as seen from Alatsomouri, looking east. . . . . . . . . . .11 Figure 2.4. The Cha Gorge at the east of the isthmus of Ierapetra. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 Figure 2.5. The hill of Alatsomouri, as seen from across the Pacheia Ammos harbor. . . . . . . . . . . .12 Figure 3.1. Overall tectonic scheme in the eastern Mediterranean region. . . . . . . . . . . . . . . . . . . . . .17

xii

APHRODITE’S KEPHALI

Figure 3.2. Sketch illustrating the progressive closure between the African and Aegean-Anatolian tectonic plates, from about 30 million years ago. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 Figure 3.3. Progressive southwesterly migration of the Aegean arc in response to the active closure between the African and Eurasian plates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Figure 3.4. Generalized structural geology map of Crete and surrounding seafloor. . . . . . . . . . . . .20 Figure 3.5. Map of the structural geology in the area surrounding the Aphrodite’s Kephali archaeological site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22 Figure 3.6. Hazard map for Greece outlining areas prone to significant seismic damage. . . . . . . . . .23 Figure 3.7. Map of earthquake epicenters in the Gulf of Mirabello and Ierapetra Isthmus area for 1998–2008. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23 Figure 3.8. Map portraying the general rock units exposed on Crete. . . . . . . . . . . . . . . . . . . . . . . . . . .24 Figure 3.9. Stratigraphic chart of the Neogene sedimentary rocks in the Ierapetra area generalized from stratigraphic sections in four areas of the Ierapetra basin. . . . . . . . . . . . . . . . . . . .25 Figure 3.10. Locations of field station sites and identification numbers. . . . . . . . . . . . . . . . . . . . .26 Figure 3.11. Positions and alignments of cross-sections A–A', B–B', and C'–C'''. . . . . . . . . . . . . . .26 Figure 3.12. Cross-sections through the ridge and hill at the Aphrodite’s Kephali archaeological site. . .27 Figure 3.13. Cross-section down the north slope of the ridge and hill at Aphrodite’s Kephali. . . . . . .27 Figure 5.1. Map of the isthmus of Ierapetra with sites mentioned in the text. . . . . . . . . . . . . . . . . .42 Figure 5.2. Map of Crete with sites mentioned in the text. . . . . . . . . . . . . . . . . . . . . . . . . . . . .43 Figure 6.1. Plan of the 1996 excavation at Aphrodite’s Kephali. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 Figure 6.2. Plan of the 2003 excavations at Aphrodite’s Kephali. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 Figure 7.1. The hill of Aphrodite’s Kephali in 2007 with the plan of the architecture. . . . . . . . . . .58 Figure 7.2. The view from Aphrodite’s Kephali looking toward the southern coast in 2006. . . . . . . .59 Figure 7.3. The single-faced west wall of the small building at the south of the site, with bedrock incorporated into the wall, looking west. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60 Figure 7.4. The mouth of the cave in 2007, looking south. . . . . . . . . . . . . . . . . . . . . . . . . . . . .60 Figure 7.5. Plan of the architecture at Aphrodite’s Kephali after cleaning in 2006 and 2007, showing the numbers of the surviving walls. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61 Figure 7.6. Plan of the north end of Aphrodite’s Kephali, showing the elevations above sea level. . .63 Figure 7.7. Plan of the small building at the south end of the hill. . . . . . . . . . . . . . . . . . . . . . . . . . . . .64 Figure 7.8. The bench at the north of the small building, looking north. . . . . . . . . . . . . . . . . . . . .65 Figure 7.9. The support for the base of the post in the small building, looking north. . . . . . . . . . . . .65

LIST OF FIGURES IN THE TEXT

xiii

Figure 7.10. Hearth 1, near the east end of the bench, and Hearth 2, about a meter south of it, looking north. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .66 Figure 7.11. Semicircular feature at the south of the small building, looking southeast. . . . . . . . . . . . .66 Figure 7.12. The southeastern corner of the small building, looking north. . . . . . . . . . . . . . . . . . . . . . .68 Figure 7.13. The semicircular buttress at the south end of the east wall of the small building. . . . . . . .68 Figure 7.14. Plan of the fire area north of the small building. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69 Figure 7.15. The north wall of the small building with the fire area beyond it, looking north. . . . . . . .70 Figure 7.16. The fire area from above, with north at the top of the photograph. . . . . . . . . . . . . . . . .70 Figure 7.17. Part of the collapsed interior circuit wall (Wall 15) at the north of the courtyard after cleaning in 2007. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70 Figure 7.18. Pithos sherds lying on the EM I surface under the fall of stones shown in Figure 7.17. . . .71 Figure 7.19. Partial restoration of the plan of the fort. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72 Figure 7.20. Aerial view of the southern part of the site showing the cave and its relation to the small building, looking south. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72 Figure 8.1. AK Pottery Class 1 (1–12). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78 Figure 8.2.

AK Pottery Class 2 (13–26). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .81

Figure 8.3.

AK Pottery Classes 2 (27–31), 3 (32–35), 4 (36, 37), and 5 (38–41). . . . . . . . . . . . . . .83

Figure 8.4.

AK Pottery Classes 6 (42, 43) and 7 (44–64). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .87

Figure 8.5.

AK Pottery Classes 8 (65–75) and 9 (76). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .91

Figure 8.6.

AK Pottery Class 10 (77). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .94

Figure 8.7.

AK Pottery Class 10 (78–80). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95

Figure 8.8.

AK Pottery Class 10 (81–86). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97

Figure 8.9.

AK Pottery Classes 11 (87), 12 (88–90), and 13 (91). . . . . . . . . . . . . . . . . . . . . . . . . . .98

Figure 8.10. Comparisons between pottery from Aphrodite’s Kephali and EM I ceramics from other Minoan sites. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .99 Figure 9.1. The mortar (92) in situ, looking south. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102 Figure 9.2.

Ground stone implements from Aphrodite’s Kephali. . . . . . . . . . . . . . . . . . . . . . . . . . .103

Figure 9.3.

Chipped stone from Aphrodite’s Kephali. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .105

Figure 11.1. Olive stone: (a) Two views of one olive stone; (b) charred grape with arrow pointing to its pip; (c) two views of part of a complete fig with arrows pointing to voids and fig seeds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .113

xiv

APHRODITE’S KEPHALI

Figure 14.1. The chronological development of closed vessels decorated in the Hagios Onouphrios Style between EM IA and EM IB. . . . . . . . . . . . . . . . . . . . . . . . . .134 Figure 14.2. Walls built using two rows of stones with the largest blocks on the exterior side of the space and the interior of the wall filled with mud mortar and small stones: (a) EM IA, Aphrodite’s Kephali; (b) LM I, Gournia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .135 Figure 14.3. Walls that incorporate bedrock (marked BR) into the lowest course in order to accommodate the uneven ground level: (a) EM IA, Aphrodite’s Kephali; (b) LM I, Pseira. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .136 Figure 14.4. Semicircular buttresses placed below ground level to help support foundation walls against lateral thrust: (a) EM IA, Aphrodite’s Kephali; (b) LM I–II, Knossos. . . . . . . . . . .137 Figure 14.5. Large squared blocks placed at the ends of walls to provide extra support at this location: (a) EM IA, Aphrodite’s Kephali; (b) LM I, Gournia. . . . . . . . . . . . . . . . . . . .137 Figure 14.6. Comparison of fortifications with curved perimeter walls: (a) EM IA, Aphrodite’s Kephali; (b) MM I, Chamaizi. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .138 Figure A.1. Geological map of the isthmus of Ierapetra and the south coast. . . . . . . . . . . . . . . . . .154 Figure A.2. Petrography sections: (a) Fabric Group 1a, tempering with calcite; (b) Fabric Group 1a, tempering with micritic limestone; (c) Fabric Group 1b, tempering with calcite and grog; (d) Fabric Group 2, tempering with grog, high fired; (e) Fabric Group 2, tempering with grog, low fired; (f) Fabric Group 2, grog fragment preserving surface treatment. . . . . . . .156 Figure A.3. Petrography sections: (a) Fabric Group 3a, south coast, high fired; (b) Fabric Group 3b, south coast, low fired; (c) Fabric Group 3c, south coast with fine grained phyllite; (d) Fabric Group 3d, south coast, with argillaceous inclusions; (e) Fabric Group 4, gray firing with quartzite; (f) Fabric Group 4, 33. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .159 Figure A.4. Petrography sections: (a) Fabric Group 5 with granodiorite; (b) Fabric Group 6 with quartz, micritic limestone, and metamorphics; (c) Fabric Group 7 with quartz, micritic limestone, and chert; (d) Fabric Group 8 with quartz, chert, and micritic limestone; (e) Sample AKE 08/22 (91) fine fabric with TCFs; (f) Sample AKE 08/59 (87), fabric with grog. . . . . . . .161 Figure B.1. GC-MS chromatogram for 5 (AK 4; ARCHEM 139, DCM). . . . . . . . . . . . . . . . . . . . .176 Figure B.2. GC-MS chromatogram for 5 (AK 4; ARCHEM 140, E). . . . . . . . . . . . . . . . . . . . . . . .178 Figure B.3. GC-MS chromatogram for 12 (AK 12; ARCHEM 157, DCM). . . . . . . . . . . . . . . . . . .180 Figure B.4. GC-MS chromatogram for 12 (AK 12; ARCHEM 158, E). . . . . . . . . . . . . . . . . . . . . .182 Figure B.5. GC-MS chromatogram for 17 (AK 21; ARCHEM 163, E). . . . . . . . . . . . . . . . . . . . . .184 Figure B.6. GC-MS chromatogram for 17 (AK 21; ARCHEM 164, DCM). . . . . . . . . . . . . . . . . .186 Figure B.7. GC-MS chromatogram for 35 (AK 7; ARCHEM 145, DCM). . . . . . . . . . . . . . . . . . . .188 Figure B.8. GC-MS chromatogram for 35 (AK 7; ARCHEM 146, E). . . . . . . . . . . . . . . . . . . . . . . . . .190 Figure B.9. GC-MS chromatogram for 38 (AK 11; ARCHEM 155, E). . . . . . . . . . . . . . . . . . . . . . . . .192 Figure B.10. GC-MS chromatogram for 38 (AK 11; ARCHEM 156, DCM). . . . . . . . . . . . . . . . .194

LIST OF FIGURES IN THE TEXT

xv

Figure B.11. GC-MS chromatogram for 51 (AK 20; ARCHEM 161, E). . . . . . . . . . . . . . . . . . . . . . . . .196 Figure B.12. GC-MS chromatogram for 51 (AK 20; ARCHEM 162, DCM). . . . . . . . . . . . . . . . .198 Figure B.13. GC-MS chromatogram for 52 (AK 8; ARCHEM 147, E). . . . . . . . . . . . . . . . . . . . . . . . . .200 Figure B.14. GC-MS chromatogram for 52 (AK 8; ARCHEM 148, DCM). . . . . . . . . . . . . . . . . . . . . .202 Figure B.15. GC-MS chromatogram for 60 (AK 5; ARCHEM 141, DCM). . . . . . . . . . . . . . . . . .204 Figure B.16. GC-MS chromatogram for 60 (AK 5; ARCHEM 142, E). . . . . . . . . . . . . . . . . . . . . . . . .206 Figure B.17. GC-MS chromatogram for 61 (AK 1; ARCHEM 135, E). . . . . . . . . . . . . . . . . . . . . . . . .208 Figure B.18. GC-MS chromatogram for 61 (AK 1; ARCHEM 136, DCM). . . . . . . . . . . . . . . . . .210 Figure B.19. GC-MS chromatogram for 77 (AK 9; ARCHEM 149, DCM). . . . . . . . . . . . . . . . . .212 Figure B.20. GC-MS chromatogram for 77 (AK 9; ARCHEM 150, E). . . . . . . . . . . . . . . . . . . . . . . . . .214 Figure B.21. GC-MS chromatogram for 87 (AK 19; ARCHEM 159, DCM). . . . . . . . . . . . . . . . .216 Figure B.22. GC-MS chromatogram for 87 (AK 19; ARCHEM 160, E). . . . . . . . . . . . . . . . . . . . . . . . .218 Figure B.23. GC-MS chromatogram for 90 (AK 6; ARCHEM 143, DCM). . . . . . . . . . . . . . . . . .220 Figure B.24. GC-MS chromatogram for 90 (AK 6; ARCHEM 144, E). . . . . . . . . . . . . . . . . . . . . . . . . .222

Preface

The conclusions based on the evidence published in this volume challenge some of the commonly held views about Crete in the third millennium B.C. This period is often called the Prepalatial period, a nomenclature based on assumptions about its character and its role in a future history of Minoan Bronze Age politics. A number of writers have suggested that Crete was populated at this time by villages with an egalitarian social structure with little social ranking and that they mainly existed at a subsistence level economically, with social stratification developing gradually during the course of the millennium (Cherry 1983, 40; Whitelaw 1983, 333–334; Branigan 1995, 39). In response to these opinions, recent research has shown that, in fact, considerable social differentiation already existed at the beginning of the third millennium B.C., and craft specialization and trade were already well advanced (Wilson and Day 1994; Day, Wilson, and Kiriatzi 1997; Day and Wilson 2002). This volume goes further, suggesting that rather than being a precursor to a socially complex state that would arise later, early polities involving several communities probably already existed in the isthmus of Ierapetra during Early Minoan I. Advances in technology had already led to craft specialization in the production of metals, ceramics, and stone tools, and in some cases entire sites specialized, which can only be an indication of decisions that were based on regional goals. Recent excavations and studies have identified long-distance trade involving the Cyclades during the Neolithic (Zachos 2007). By Early Minoan I trade was routine, and in Crete products traveled for substantial distances and included both pottery up to the size of pithoi and also the commodities transported in ceramic

xviii

APHRODITE’S KEPHALI

vessels. Social and economic differentiation existed on a regional, not just a local level, and decisions for mutual defense could involve collaboration by groups of workers, including the building of the watchtower that is the focus of this volume. No evidence suggests that this situation was necessarily replicated throughout Crete. On the contrary, evidence for differences throughout the island can be found in many characteristics of the material record, including aspects of culture like burial customs that are often important social markers. Like the isthmus of Ierapetra, other regions in Crete must be considered on their own evidence.

Acknowledgments

Many different people and institutions provided assistance for this project. First, thanks are due to Vili Apostolakou, Director of the 24th Ephorate of Prehistoric and Classical Antiquities, for inviting the author to be a member of the team that studied and consolidated the site in 2006 and 2007 and for granting permission to publish this site. Without her help and encouragement, this important archaeological location could not have been studied so thoroughly. Thanks are also extended to Theodore Eliopoulos for permission to publish the excavations of 1996. The two archaeologists who supervised the work in 1996 and 2003, Nikos Panagiotakis and Maria Kyriakaki, provided invaluable assistance and advice. Alekos Nikakis provided his crucial knowledge and expertise, and he supervised the site’s consolidation. The staff of the Institute for Aegean Prehistory (INSTAP) Study Center for East Crete in Pacheia Ammos made many contributions to this project. Director Thomas M. Brogan and Assistant to the Director Eleanor J. Huffman offered help in numerous ways. Stephania N. Chlouveraki provided important work with conservation, ably assisted by Matina Tzari. The long and patient work of Chlouveraki and Tzari in restoring the earliest Minoan pithos makes a lasting contribution to the history of this class of storage container. Doug Faulmann drew pottery profiles, and he helped sort out the minimum number of pithoi represented by the sherds from scattered locations. Eleni Nodarou conducted the petrographic analysis at the William A. McDonald Petrographic Laboratory at the INSTAP Study Center, and she offered many valuable insights into the pottery production. Chronis Papanikolopoulos provided photographic support. Beginning

xx

APHRODITE’S KEPHALI

in 2006, Andrew Koh, Chlouveraki, Kathy Hall, and Michel Roggenbucke established a program for sampling pottery for organic residue analysis in the William D.E. Coulson Conservation Laboratory at the Study Center in collaboration with: Archaeochemistry Research in the Eastern Mediterranean (ARCHEM); the Museum of Cretan Ethnology Research Centre in Vori, Crete; the Department of Art History, Temple University, Philadelphia; the Foundation for Research and Technology-Hellas (FORTH), Heraklion, Crete; and the University of California, Los Angeles (UCLA). This program successfully analyzed samples from Aphrodite’s Kephali, with results from 100% of the samples from this site that were tested. Study and mapping in the field was accomplished with the contributions of: Floyd W. McCoy, Professor of Geology at the University of Hawaii; Susan C. Ferrence, Director of Publications for INSTAP Academic Press; and Stephania N. Chlouveraki, Chief Conservator for the INSTAP Study Center. Graduate students from Temple University (Judith Papit, Ariel Pearce, and Heather Hicks) and the University of Pennsylvania (Andrew Insua and Miriam Clinton) provided important assistance with documentation of the cleaning by the 24th Ephorate personnel and assisted with drawing and measuring the architecture. Andrew Insua measured the topography and the architecture for mapping purposes using a Topcon Total Station. As always, substantial work was necessary in cataloging and studying the finds and in writing reports. The work was accomplished between 2006 and 2009. Mary A. Betancourt developed the database system used for the organization of the cataloged finds, and she and her assitants cataloged the objects with advice from Heidi M.C. Dierckx on stone implements. Susan C. Ferrence and Floyd W. McCoy helped supervised some of this work in preparation for publication. David S. Reese studied the faunal remains. Heidi M.C. Dierckx studied the stone implements and drew them. Graduate students from Temple University included Lily A. Bonga, Jeannine A. Beckman, Rachael Fowler, Whitney Krukenberg, Allyson McCreary, Judith Papit, Ariel Pearce, and Sarah Peterson. Graduate students from the University of Pennsylvania included Miriam Clinton, Nurith Goshen, and Andrew Insua. Rebecca Mullin was a graduate student from University College Dublin. Financial support for this project was provided by Temple University in Philadelphia, the Institute for Aegean Prehistory, the University of Pennsylvania, and private donors. Other acknowledgments are listed in the individual chapters.

List of Abbreviations

AK

excavation accession number

kg

kilogram(s)

AKE

petrography sample number

km

kilometer(s)

cc

counterclockwise

kPa

kilopascal

cm

centimeters

kV

kilovolt

d.

diameter

LM

Late Minoan period

DCM

dichloromethane

m

meter(s)

dim.

dimension

M

E

ethanol

maximum earthquake magnitutde (Richter)

EB

Early Bronze

Ma

million years before present

EC

Early Cycladic period

m asl

meters above sea level

EH

Early Helladic period

max.

maximum

EM

Early Minoan period

min

minute(s)

FN

Final Neolithic period

ml

mililiter(s)

GC-MS

gas chromatography and mass spectrometry

mm

millimeter

MM

Middle Minoan period

gm

gram(s)

Mw

movement magnitude

h.

height

µl

microliter(s)

IGME

Institute of Geology and Mineral Exploration

µm

micrometer(s)

xxii

APHRODITE’S KEPHALI

NAFZ

North Anatolian Fault Zone

sp.

species

pers. comm.

personal communication

TCF

textural concentration feature

pers. obs.

personal observation

th.

thickness

pres.

preserved

wt.

weight

sec

second

yr(s)

year(s)

Part I

Introduction

1

Introduction Philip P. Betancourt

Aphrodite’s Kephali is a translation of the Greek phrase tis Aphroditis to Kephali (της Αφροδίτης το κεφάλι). It is the local name for a small hilltop in eastern Crete where a Minoan archaeological site was excavated in two seasons of work, in 1996 and 2003. The location is named after the property owner. The word kephali (κεφάλι) is a local Cretan noun in the neuter case. It means a small rounded hillock that stands out from the surrounding landscape. The word is closely related to the feminine word for “head” (κεφάλη). As explained by Eliopoulos (1998, 301) in regard to the contrast between the two nearby sites of Kephala Vasilikis and Kephali Vasilikis, both of which are located near one another just north of Aphrodite’s Kephali, the word kephala (κεφάλα) signifies a larger hillock than a kephali. The situation of this small hill and its role in early Cretan history is a direct result of its position in the topography of the largest Aegean island. Because most of Crete is mountainous and land travel was difficult, a high percentage of the most important Bronze Age sites were situated near the northern or

southern coasts where travel by sea offered a more convenient means of movement (Fig. 1.1). The largest Minoan palaces of the Middle and Late Bronze Age—Knossos, Malia, and Phaistos—were all situated where they could take advantage of travel and trade by sea. Only in the eastern part of the island where the Gulf of Mirabello cuts into the land from the north was the island easily traversed from north to south (Fig. 1.2). Here, at the isthmus of Ierapetra, a low-lying valley extended between the northern and southern coasts (Fig. 1.3). This topography created a natural route across the island from north to south. Hills and mountains were situated both to the east and to the west of this low and relatively level region. A number of Minoan and later sites took advantage of the valley and established themselves on the nearby hills with access to this natural trade route between the Aegean and the Libyan Sea. One of those sites was Aphrodite’s Kephali. Aphrodite’s Kephali is located on the western side of this isthmus, 7 km north of the Libyan Sea and 5 km south of the Gulf of Mirabello (Fig. 1.4).

4

Figure 1.1. Map of Crete.

Figure 1.2. Map of eastern Crete.

PHILIP P. BETANCOURT

INTRODUCTION

5

0

1

Figure 1.3. Topographical map of the isthmus of Ierapetra. Contour interval is 100 m.

5

6

PHILIP P. BETANCOURT

Aphrodite’s Kephali

Figure 1.4. The site of Aphrodite’s Kephali as seen from the north–south road near modern Episkopi. Photo P. Betancourt.

Aphrodite’s Kephali

Figure 1.5. The road to Aphrodite’s Kephali from the northwest. Photo P. Betancourt.

Figure 1.6. The site, looking south, with olive groves around it. Photo P. Betancourt.

INTRODUCTION

It is the easternmost peak of the hill of Smaïlongosi (Σμαϊλογγόσι), which consists of a series of several promontories west of the modern village of Episkopi. The site is easily visible as a peak whose upper part is pale colored and devoid of vegetation because of modern bulldozing. It has an elevation of 211.10 meters above sea level. Access to the peak is easier from the northwest than from the low isthmus, and an unpaved farm road leads to the site from this direction (Fig. 1.5). Modern Aphrodite’s Kephali is surrounded by olive groves (Fig. 1.6), and it has had no habitation of any kind since the opening phases of the Early Bronze Age. The Kephali has a small, almost level spot at the top, and this elevation overlooks the modern north–south road. Anyone positioned at the summit has an especially good view toward the south, with visibility all the way to the sea. Only part of the archaeological site on Aphrodite’s Kephali survives because the hill was partly destroyed by bulldozing to create new terraces for olive cultivation. The eastern and western sides of the hill were removed by perhaps as much as 3 or 4 m in some places, with less removed at the north and south before the work was halted by the 24th Ephorate in order to preserve this unique and important early evidence of the human occupation of this part of Crete.

Figure 1.7. Consolidating the architecture in 2006. Photo P. Betancourt.

7

Two seasons of archaeological excavation were conducted on the hill. The 24th Ephorate first undertook rescue excavations at Aphrodite’s Kephali in 1996 under the direction of Theodore Eliopoulos. These excavations, which were supervised by Nikos Panagiotakis, were a part of the official investigation of this part of Crete in anticipation of the possibility of building a new airport north of the city of Ierapetra (Eliopoulos 1998, 312). The airport was never constructed. Work at the site was continued in 2003 under the direction of Vili Apostolakou, with the work supervised by Maria Kyriakaki. Cleaning and conservation work was conducted on the hilltop in 2006 and in 2007 under the direction of Vili Apostolakou on behalf of the 24th Ephorate of Antiquities in order to consolidate and preserve the architecture (Fig. 1.7). Alekos Nikakis, the Chief Conservator for the Ephorate, supervised the cleaning and consolidation. The author was invited to be a member of this team, and with the assistance of graduate students from Temple University and the University of Pennsylvania, he studied the site and prepared the site plan in 2006 and 2007. The geological study by Floyd McCoy and the study of stone tools and other remains took place at the same time as well as in subsequent years.

8

PHILIP P. BETANCOURT

The work revealed a small, fortified site from the Early Minoan (EM) I period. With its good view of the entire valley stretching all the way to the Libyan Sea in the south, the site would have been a dramatic and easily visible landmark at the beginning of the Minoan era. All those who traveled up the road from its terminus at the coast could have looked up and seen its lofty setting. Scholars do not agree on the absolute chronology for EM I. Its beginning has been dated to ca. 3600/3000 B.C. by Peter Warren and Vronwy Hankey (1989, 169) and to ca. 3100–3000 B.C. by Sturt Manning (1995, 168–170). A later date of 2600 B.C. once suggested by Nicolas Platon (1974, 118) is no longer accepted. Although it is one of the least known periods of the entire Cretan Bronze Age, it is extremely important because it contains some of the earliest evidence for the history of the cultural group we call the Minoans. This is the first EM I site of its kind to be excavated and published. A comparison with the famous walled hilltop establishment at Chamaizi is inevitable (Xanthoudides 1906; Davaras 1972, 1992). Like Aphrodite’s Kephali, Chamaizi is situated on a defensible hill in East Crete. Both places

were surrounded by fortifications that were constructed with curved outer faces that followed the contours of their hills, but the date of Chamaizi is Middle Minoan. Though both sites were well situated for observing the countryside, the EM I date of the site on Aphrodite’s Kephali sets it apart from Chamaizi, and the later settlement also has no courtyard to allow for an increase in population in times of danger. The large amount of storage at Aphrodite’s Kephali in contrast with the sparse evidence for the number of regular residents also makes the site in the Isthmus of Ierapetra very different from its much later counterpart. A tiny watchtower whose size and whose artifact assemblage suggest it was not a simple farm implies a degree of specialization along with a set of regional objectives that have not been previously documented for this early in Minoan Crete. The site represents a new chapter in early Cretan history, and it must have played at least a small role in the dynamic events that marked the beginning phases of the Early Minoan period. Several preliminary reports have been published: Betancourt 2008a, 2008b, 2010; Koh and Betancourt 2010; Betancourt et al. 2012.

2

The Isthmus of Ierapetra Philip P. Betancourt

The isthmus of Ierapetra is the narrowest part of Crete by a substantial margin. It consists of a long valley extending 12 km north and south across the eastern part of the island between the Gulf of Mirabello at the north and the Libyan Sea at the south (Fig. 1.3). The valley is much flatter than the land that flanks it on both the east and the west. Because the rest of Crete is mountainous, this valley provides the most easily traversed land bridge between the northern and southern coasts, and a well-traveled modern road lies along it. The modern paved highway follows the easiest passage across the isthmus, and it must be close to the path used by travelers in all previous periods. The road has always been an important artery for trade and communication across the southern Aegean island, and in some periods of history it has required a physical presence by the political power that controlled it. The ruins of a Venetian tower overlook the road from the hill of Vaïnia Stavromenos on the western side of the Isthmus, a testament to the need to observe the southern part of the north–south artery when Venice ruled the island of

Crete (Spanakis 1991, I, 173). An Ottoman period watchtower overlooking the same highway still sits on the crest of a hill near the modern village of Episkopi, across the modern road from Aphrodite’s Kephali (Fig. 2.1). With this later history, it is not surprising to see a similar response in other periods when land communication across Crete was an important aspect of the society. Geologically, the isthmus is a graben, a low part of the landscape bounded by faults separating it from uplifted blocks called horsts (see Ch. 3). The relative uplift and subsidence of grabens and horsts occurs as a result of tectonic movements of the earth’s crust. Crete is near the center of an active tectonic area called the Hellenic Arc where the European tectonic plate is moving south over the African one. The movements of the plates in the Hellenic region have been very complex, and a variety of rock types have been displaced and broken into segments during the millions of years that these plates have been active (for general reviews, see Baumann et al. 1976; Baumann, Best, and Wachendorf 1977; Fortuin 1977, 1978; Seidel et al. 1981; Seidel, Kreuzer, and

10

PHILIP P. BETANCOURT

Harre 1982; Gifford and Myer 1984; Fassoulas, Kilias, and Mountrakis 1994; Higgins and Higgins 1996; Fassoulas 1999). At the isthmus of Ierapetra, the low and relatively flat graben extends out to sea at both the north and the south. The eastern fault line can be traced easily at the north of the depressed region. Sheer cliffs on the western side of the island of Pseira are part of this massive fault (Farrand and Stearns 2004, 17), which extends out to sea beyond the small island (Fig. 2.2). The eastern fault can also be seen at Chrysokamino (Farrand 2006, 29), and it can be

traced south from there along the eastern side of the Gulf of Mirabello (Fig. 2.3). Farther south, the same fault line forms lofty cliffs near the Cha Gorge (Fig. 2.4). Its counterpart is visible at the northwest where the hill of Alatsomouri forms the western side of the bay of Pacheia Ammos (Fig. 2.5). These cliffs at the north of the isthmus are formed of hard rocks. The hills along both the west and the east are less conspicuous at the south of the valley because the rocks there are softer, and they erode to form more gentle slopes, but they still rise more than 200 meters above sea level at the sides of the low valley.

The Early Settlements of Eastern Crete The history of the isthmus of Ierapetra must be understood in relation to what was happening elsewhere in East Crete at the same time. The isthmus region’s population was always in contact with nearby settlements, and its history was interwoven with that of nearby areas. New research and several recent excavations and surveys have added substantial information to our knowledge of the opening phases of the Early Bronze Age in the broad region of East Crete that includes the Gulf of Mirabello and the isthmus of Ierapetra. Although some of the problems discussed in previous studies of the period still remain (for discussion of problems, see Karantzali 1996, 17), we are now in a much better position to understand some of the complex details of this period. Intensive surface surveys have made substantial contributions in the past 20 years, under the direction of many individuals: David Blackman and Keith Branigan (1977); Keith Branigan (1998, 1999a, 2000), Stella Chryssoulaki and Leonidas Vokotopoulos (Vokotopoulos 1998, 2000), Theodoros Eliopolos (1998, 302, 312), Donald Haggis (2005), Barbara Hayden (2003a, 2003b, 2004), Krzystof Nowicki (2000, 2002, 2006, 2008a), Norbert Schlager (2001a, 2001b), Metaxia Tsipopoulou (1989a), Yannis Tzedakis and colleagues (Tzedakis et al. 1989, 1990, 1991–1993, 1999), Vance Watrous and Harriet Blitzer (Watrous and Blitzer 1995; Watrous et al. 2000, 2012), and Philip Betancourt, Costis Davaras, and Richard Hope Simpson (Betancourt, Davaras, and Hope Simpson, eds., 2004, 2005).

In addition to these surveys, new publications and new excavations in recent years have been conducted for a long series of Final Neolithic (FN)–EM sites in East Crete in addition to Aphrodite’s Kephali (Figs. 1.1, 1.2, 1.3): Alatzomouri (Apostolakou, Betancourt, and Brogan 2007–2008) Azoria, near Kavousi (Haggis and Mook 2004; Haggis et al. 2004, 2007) Chalepa (unpublished) Chrysokamino (Betancourt 2006; Catapotis, Pryce, and Bassiakos 2008) Hagia Photia Cemetery (Betancourt 2003, 2008c; Davaras and Betancourt 2004, 2012; Betancourt and Muhly 2007; Karantzali 2008; Muhly 2007, 2008) Hagia Photia Kouphota (Tsipopoulou 2007) Hagios Antonios (Haggis 1993) Kato Kastelas (Vokotopoulos 1998; 2000, 138–139) Kalo Chorio (Haggis 1996b) Kokkino Phroudi (Chryssoulaki 1996; Vokotopoulos 2000, 139–141) Mochlos (Soles and Davaras 1992, 1996) Monastiraki Katalimata (Nowicki 2001, 2008b) Petras Kephala (Papadatos 2007, 2008)

THE ISTHMUS OF IERAPETRA

11

Ottoman Castle

Figure 2.1. The Ottoman castle at Episkopi, on the crest of a hill overlooking the north– south highway. Photo P. Betancourt.

Pseira

Chrysokamino

Figure 2.2. Pseira and Chrysokamino as seen from Alatsomouri, looking north. The cliffs visible at the west of the island are part of the fault line at the east of the isthmus of Ierapetra. Photo P. Betancourt.

Figure 2.3. The village of Pacheia Ammos as seen from the hill of Alatsomouri, looking east with the fault line beyond the village. Photo P. Betancourt.

12

PHILIP P. BETANCOURT

Priniatikos Pyrgos (Hayden et al. 2007) Pseira (Betancourt and Davaras, eds., 1995, 1998a, 1998b, 1999, 2003; Betancourt, Davaras, and Hope Simpson, eds., 2004, 2005) Traostalos (Vokotopoulos 2000, 138–139; Chryssoulaki 2001, 63)

In spite of all this new research, however, problems still remain. The pottery sequences and the correlations between sites are still poorly understood (Wilson and Day 2000, 54). Most of the sources for the pottery are still unknown, although good progress has been made on characterizing several of the fabrics (Myer, McIntosh, and Betancourt 1995; Day 1991, 1997; Day, Wilson, and Kiriatzi

Figure 2.4. The Cha Gorge at the east of the isthmus of Ierapetra, flanked by steep cliffs, looking east from the floor of the valley. Photo P. Betancourt.

Figure 2.5. The hill of Alatsomouri, as seen from across the Pacheia Ammos harbor, looking west. Photo P. Betancourt.

THE ISTHMUS OF IERAPETRA

1997, 1998; Haggis 2000; Vaughan 2002; Day, Joyner, and Relaki 2003; Day et al. 2005; Day, Relaki, and Faber 2006; Day and Wilson 2002, 2004). At Knossos, the Neolithic sequence (Evans 1964) is now much better understood (Tomkins 2007, 2008). In the Gulf of Mirabello region, the boundary between FN and EM I is still uncertain, especially in regard to the dark-surfaced, coarse, burnished ceramics with thick walls. Both absolute and relative chronologies are hotly debated. The role of technological expansion (especially metallurgy) in the development of these periods is imperfectly understood. Most sites have been recognized only from surface survey. Excavations have taken place at just a few places, and we will need much more evidence to clarify many of the new questions that have emerged in regard to the era. The Final Neolithic and Early Minoan I periods were a time of dynamic change in the Aegean when many new settlements were founded in Crete. Increases in population occurred at the beginning of Early Helladic as well (Rutter 1993, 772). The period probably spanned the end of the fourth millennium and the beginning of the third millennium B.C. (for an earlier view that the period began ca. 2500 B.C., see Hutchinson 1962, 138; for a transition from FN to EM I ca. 3600 to 3000 B.C., see Warren and Hankey 1989, 169; for a date ca. 3000 B.C., see Manning 1995, 168–170). A movement of new people to Crete during the Final Neolithic and EM I periods, a view that was suggested by a number of scholars beginning many years ago (Hutchinson 1962, 140; Branigan 1970a, 11; 1988, 66; Renfrew 1972, 229; Warren 1973; Hood 1990a; 1990b), is now known to have affected the entire island. The crucial evidence for this conclusion is the large number of new foundations during this period. The following studies record so many expansions in population that an influx of people from elsewhere is virtually certain: Western Crete (Warren and Tzedhakis 1974, 338; Tzedakis 1984, 5; Moody 1987, 292–294; Nixon, Moody, and Rackham 1988, 171) Central Crete (Vagnetti 1972–1973; 1973; 1975; 1996; Vagnetti and Belli 1978; Vasilakis 1987; Tomlinson 1995, 64;

13

Watrous and Hadzi-Vallianou 2004, 221–226; Panagiotakis 2004; 2006, 169) Far Eastern Crete (Branigan 1999a, 2000; Muhly 2008; Nowicki 2000, 2002, 2006, 2008a; Schlager 2001a, 2001b) Gulf of Mirabello and the isthmus of Ierapetra (Betancourt 1999; Hayden 2003a; 2003b; Watrous et al. 2000, 474; 2012; Nowicki 2001; 2002; 2004; 2006) In the local area discussed in this study, more new sites have been recorded at the north of the isthmus rather than at the south. Watrous and Blitzer have established relative dates for this immigration into the region of the isthmus of Ierapetra. Their surface survey near Gournia discovered 3 FN sites and 25 EM I sites (Watrous et al. 2000, 474), clearly showing that the large increase of population in this region occurred after the end of the Neolithic. The new settlements along the Gulf of Mirabello and just south of it included both coastal and inland places, with some of them on defensible peaks and others on lower ground. They tended to be on the low hills, not on the flatter land of the plain at the north of the isthmus (called the Kampos). Many were near fresh water sources, suggesting that water management was often an important concern. It is not surprising that people would settle there at an early period. The soils on the hills that adjoin the isthmus are fertile, the land is not as mountainous as most other parts of Crete, and the many harbors offer good opportunities for seafaring. The climate was probably somewhat wetter during EM I than in modern times (Allen and Katsikis 1990, 14; Rackham and Moody 1996, 39). In addition, the opportunities for land trade are excellent because communication between the northern and southern coasts is easier than elsewhere on the island. Many of the most important sites were on the hills around the valley where the residents had access to the region’s pale colored soils (marls and phyllitic sediments). The residents of this region built houses of stone or mudbrick on stone foundations, with earth floors and flat roofs. The archaeological evidence suggests that they practiced both farming and animal husbandry, using

14

PHILIP P. BETANCOURT

the fertile low hills for their agriculture and the more marginal land for pastures (Betancourt 2006, 233–259). Sheep and goats were the main animals. Both coastal communities and inland

sites engaged in trade, and obsidian from Melos was a common material, but exchange networks were primarily with other places on the island of Crete.

3

Geology and Geologic History Floyd W. McCoy

Tectonic Development SUMMARY Late in the geological history of the eastern Mediterranean region at the beginning of the Neogene Period of the Cenozoic Era (23 million years ago; Table 3.1), newly established tectonic forces resulted in significant deformation of the Aegean crust in a new direction—east and west. This new force was added to dominant north-northeast–south-southwest forces responding to the continuing closure between the African and Eurasian tectonic plates that had been ongoing for over 200 million years since the Mesozoic Era (Table 3.1; Fig. 3.1). Accordingly, the southern edge of the Aegean-Anatolian microplate where these two sets of orthogonal forces were active was now undergoing both east–west stretching and north-northeast–south-southwest compression. Increased stretching resulted in the collapse of the central Aegean plate to create a series of large basins (tectonic grabens) that became fresh-water lakes. Along the southern boundary of the plate, a series of

basins (grabens) opened onto the predecessor of the modern Mediterranean Sea, the Tethys seaway, to form marine embayments. Finally, the entire Aegean plate foundered below sea level, and the interior lakes were replaced by marine waters: the Aegean Sea and its archipelago of islands was born. On Crete, local tectonism during the late Neogene uplifted and deformed some of the flooded embayments into hills and mountains: the knoll of Aphrodite’s Kephali was created at this time.

CRETE AND THE AEGEAN SEA Crustal extension, thus, was added to an older, and longer (throughout the Mesozoic, continuing today), tectonic episode dominated by compressional forces. The African tectonic plate had moved north to collide with the Aegean tectonic plate during the Mesozoic Era

16

FLOYD W. MCCOY

(Table 3.1), and that collision was destroying the ancient Tethys seaway to form the roots of today’s Hellenide and Anatolide mountain systems (Fig. 3.2). It is estimated that approximately 2,100–2,500 km of closure—of crustal shortening—has occurred here since the Jurassic period (van Hinsbergen et al. 2005). Contemporary closure rates are on the order of 4 cm/yr (McKenzie 1978; Le Pichon and Angelier 1979; Jolivet and Patriat 1999; ten Veen and Kleinspehn 2003; Nyst and Thatcher 2004). Also during this tectonic episode, continental fragments (slivers of land removed from the African plate) within the ancient Tethys seaway were heading for the collision zone (Fig. 3.2). With collision, these slivers encountered the edge of the Aegean-Anatolian plate and subsequently were folded and stacked one on top of the other. Other slivers, however,

Million Years

Era

Period Quaternary

0.01

Epoch

remained attached to the ancient sea-floor crust and went through a more complicated history: 1. Subduction beneath the AegeanAnatolian plate 2. Metamorphism of rocks into quartzites, phyllites, and marbles 3. Exhumation 4. Uplift and merger with the stack of slivers already present Continued compression, combined with the subsequent tensional forces in a very complex tectonic setting, compressed this terrain into a sandwich of stacked and imbricated crustal fragments, thus creating the series of nappes characteristic of contemporary Cretan geology.

Stage

Holocene Continued uplift of the AK area

Pleistocene

1.8 Pliocene

Cenozoic

5 Neogene Miocene

23 35

Oliogene Paleogene

57

Eocene Paleocene

65

Events

Messinian Tortonian Serravallian Langian Lower

Uplift of Crete and the AK area Formation of the Aegean Sea Refilling of the Mediterranean Sea Dessication of the Mediterranean Sea Deposition of sediments that form today’s AK Subsidence of the AK area beneath sea level to shallow depths

Uplift creating Cretan landmass Assemblage of tectonic nappes to form Crete; progressive destruction of the Tethys Seaway with closure between the African and Aegean plates

Mesozoic

Cretaceous

245

Jurassic

Broad Tethys Seaway between Africa and Europe with slivers of continental land as large islands (which will later form the tectonic nappes of Crete)

Triassic

Table 3.1. The geologic time scale for the Cenozoic Era, with subdivisions into periods, epochs, and stages, and identification of notable geologic events in the tectonic and geologic history of the area surrounding the Aphrodite’s Kephali archaeological site (AK). Note that time units are not to scale. Table utilizes information from: Potsma, Fortuin, and van Warmel 1993; ten Veen and Potsma 1999; Fassoulas 2001, 2004; Reuter, Brachert, and Kroeger 2006; van Hinsbergen and Meulenkamp 2006.

GEOLOGY AND GEOLOGIC HISTORY

17

Figure 3.1. Overall tectonic scheme in the eastern Mediterranean region. Dashed arrows along curved line in southwest depict directions of the Aegean and Anatolian microplates’ motion relative to Africa; numbers indicate measured rates of plate motion in mm/yr. For the African tectonic plate, plate motion is directed toward the northeast at rates on the order of 20–35 mm/yr (not shown here for simplification). Curved dashed line with arrow indicates the motion of the Aegean-Anatolian plate along the North Anatolian Fault Zone (NAFZ) relative to a fixed Eurasian plate. The dashed line connecting the NAFZ and the Kephalonia Fault represents the complex and largely unmapped tectonic boundary between the Eurasian and Aegean and Anatolian microplates through mainland Greece. Small triangles indicate sites of active volcanism along the volcanic arc of the southern Aegean Sea. “Trench” is in quotation marks because the referenced sea-floor features are not trenches in the accepted tectonic definition of the word. Modified from ten Veen and Kleinspehn 2003.

18

FLOYD W. MCCOY

Figure 3.2. Sketch illustrating the progressive closure between the tectonic African plate and the Aegean and Anatolian microplates, from about 30 million years ago (Oligocene; top cross-section) to modern time (lowermost cross-section). By the Oligocene, other landmasses had already moved north and been incorporated as tectonic nappes into Crete. Sometime shortly after the Oligocene, likely in the Miocene, the Hellenic arc started its southward movement, as depicted in this sketch, increasing the rate of closure between the two plates. Note that continental crust apparently was also subducted, which may be unique to the Hellenic arc (van Hinsbergen et al. 2005). Illustration modified from Sakellariou, pers. comm. Solid triangle indicates location of volcanoes.

Accompanying the process of subduction, an island arc system—the Hellenic Forearc— developed over the convergence zone (Fig. 3.1). This first formed somewhere near modern Lesbos some 11 million years ago (in the Miocene) and then migrated south, with an evolving subduction zone, to its present position as the Hellenic Forearc (Fig. 3.3; ten Veen and Potsma 1999; ten Veen and Kleinspehn 2003). Rates today of southerly motion are on the order of 3.0–4.5 mm/yr (Caputo, Monaco, and Tortorici 2006). With progressive forearc migration to the south, lateral expansion of the frontal edge of the Aegean-Anatolian plate occurred. Consequent tensional forces led to stretching; the brittle crust cracked, and a series of large grabens formed during the earliest Miocene to form lake basins. As stretching continued, the floors of these grabens dropped below sea level, and the

sea intruded to create marine sedimentary basins both as the Aegean Sea and as smaller basins on Crete. Uplift on Crete came with a new compressional pulse during the Pliocene epoch that persisted into the early Pleistocene, lifting these sedimentary basins well above sea level (Table 3.1). Rates of uplift were on the order of 10–30 cm/1,000 yrs (Angelier et al. 1982; Peters, Troelstra, and van Harten 1985). Later foundering of some of these basins due to tensional forces, accompanied by the large oscillations in sea levels during the PlioPleistocene epochs in response to global glaciation, lead to the geography of islands and the complex structural patterns seen today. In the Ierapetra area, sediments deposited in the Mio-Pliocene marine basins were lithified into limestones, sandstones, marls, chalks, breccias, and conglomerates, then

GEOLOGY AND GEOLOGIC HISTORY

19

Figure 3.3. Progressive southwesterly migration of the Aegean arc in response to the active closure between the African and Eurasian plates. Note that with each step in the migration, significant stretching of the advancing front occurs along the arc, resulting in the break-up and collapse of the landscape into basins, such as those in the Ierapetra area. Arc migration toward the advancing tectonic front is seen at other tectonic settings throughout the world and is known as “roll back.” Black lines with triangles outline the front of the advancing plate through time; time is indicated as Ma (million years before present). After ten Veen and Kleinspehen 2003.

uplifted to form the Siteia and Psiloritis/Dikte Mountains—a foothill to the latter is the location of the Aphrodite’s Kephali archaeological site—and the Ierapetra Isthmus between the mountains. These tectonic movements continue. Modern marine sedimentary basins on the seafloor north and south of Crete, as well as non-marine basins on Crete (e.g., the Lasithi Plateau), are defined by major fault systems (Fig. 3.4). These locations are characterized by high sedimentation rates as Crete is denuded through tectonism and weathering. The future is

suggested from new geophysical data and seafloor mapping—it appears that a portion of the African continent may now be in contact with western Crete (Mascle et al. 1999; ten Veen and Kleinspehn 2003). This situation will result in significant uplift of that part of Crete (perhaps already underway and reflected in historic neotectonic movements and palaeoseismicity there, such as uplift in West Crete). Meanwhile, the Aegean island arc will continue its southerly movement, the African plate will continue its northerly march, and further continental collision

20

FLOYD W. MCCOY

Figure 3.4. Generalized structural geology map of Crete and surrounding sea floor. Tectonic stresses define two regional trends of faulting north-northeast–south-southwest and east-northeast–west-southwest directions (azimuth averaging 070˚ and 020˚). Faults in the Cretan area are dominantly normal faults that reflect crustal extension, whereas those to the east are dominantly strike-slip faults due to crustal slippage between this tectonic terrain and the terrain located to the east near Anatolia and Cyprus. Crete has been broken into numerous small blocks, some of which respond today to these tectonic forces through significant and measurable counterclockwise (cc) rotations during the past 5 million years (since the Messenian stage of the last Miocene; see Table 3.1 for the latest scale). Abbreviations are: cc = counterclockwise, ESCT = East South Cretan Trough, M = Mesara graben (valley), and WSCT = West South Central Cretan Trough. Shaded patterns denote areas where crustal motion has formed depressions (grabens) that create topographic basins between areas of uplifted crust (horsts). AK and arrow indicate the position of the Aphrodite’s Kephali archaeological site. Modified from ten Veen and Kleinspehn 2003.

over the next few million years should see the Mediterranean Sea gone, replaced by a vast mountain range, incorporated within which will be the

modern Cretan landscape with its scattered prehistoric and historic ruins.

Geology STRUCTURAL SETTING The hill with the archaeological site of Aphrodite’s Kephali at its summit is part of a ridge extending off an eastern spur of the low hills east of the Dikte Mountains. These foothills are a complex tectonic terrain of small, normal faults forming the western boundary of the Ierapetra Isthmus. The eastern boundary of the isthmus is defined by a

series of major faults along the Ierapetra Isthmus (Fig. 3.4) that create spectacular cliffs in the modern topography (Figs. 2.2–2.4). Regional alignment of fault trends is north-northeast–south-southwest, the consequence of continued crustal extension resulting in a downthrown fault-block (graben) that is today’s isthmus (Fig. 3.4). A second set of regional fault lines is oriented east-northeast–west-southwest

GEOLOGY AND GEOLOGIC HISTORY

and forms the southern boundary of the Dikte Mountains as well as being reflected in the geographic trend of the southern coastline of Crete in the vicinity of Ierapetra. This fault zone is part of a major structural feature in the region, known as the East South Cretan Trough (labeled ESCT in Fig. 3.4), formed in response to transverse crustal slippage (transpression) along the southeastern side of the Aegean arc (Fig. 3.4). This structural fabric is a regional pattern seen both onshore and offshore of eastern Crete. It is here that compressional forces due to tectonic closure are translated into transpressional forces along the eastern boundary of the closure zone. Breakage of the crust forms smaller tectonic blocks seen in the topography offshore and onshore as grabens and mountains, troughs, and the isthmus, with each block often going through differential rotation and tilting (Fig. 3.4). Crustal deformation is seen in the numerous smaller faults mapped in the area of Aphrodite’s Kephali (Fig. 3.5). Local fault-line trends follow those mapped in regional trends. Faults in the vicinity of the site are normal faults, reflecting the tensional stress regime, and they subdivide the knoll and its environs into a series of small fault blocks. Each of these blocks has been rotated and shifted relative to the others, as is apparent from abrupt changes in local bedding orientations near their contact with the faults, much like the larger blocks in the regional setting (Fig. 3.5). Fault planes are vertical or nearly vertical. Displacements along faults at Aphrodite’s Kephali are on the order of no more than one meter where a displacement can be discerned. Faults are particularly evident by gouge (ground rock resulting from movement along the fault zone) that is easily eroded. The cave at the top of the hill at the archaeological site, for example, has been dug into an easily excavated fault gouge that marks a pair of closely spaced faults. All faults may be considered active, given the current crustal dynamic regime of the region. The youthful morphology of the Ierapetra Fault Zone, as well as sheared alluvial and colluvial deposits at the base of the fault scarp, indicate recent movement along this fault (Caputo, Monaco, and Tortorici 2006). Estimates of seismotectonic parameters for this fault zone based upon fault morphology and characteristics by Caputo, Monaco, and Tortorici (2006)

21

suggest that maximum earthquake magnitudes on the order of M=6.3–6.6 may be characteristic, with a mean recurrence interval of 260 years for earthquakes of this magnitude over the past 13,000 years (with an uncertainty range of 200–371 years; in other words, an earthquake of M=6.3–6.6 might be expected to occur every 260 years, or perhaps every 200–371 years, given the uncertainty of the calculations and field observations). The area east of the Ierapetra Isthmus in the Siteia Mountains is mapped as a high-risk zone for larger earthquakes of magnitudes up to M=9–10, with the adjacent area to the west considered a zone where magnitudes of M=7–8 may be expected (Fig. 3.6). Although no seismic activity of such magnitudes has been documented from 1997 to 2008 (Fig. 3.7), the hazard map (Fig. 3.6) clearly notes that the area has been subject to strong seismic activity in the past. A search of seismic databases for this portion of eastern Crete lists 82 earthquakes for the 11-year period, 1998–2008, all with magnitudes of 3.0 (minimum magnitude for search) to 4.7 (maximum listed), most at subcrustal depths of 0–35 km, in a spatial pattern outlining a variety of fault traces in the area (Fig. 3.7). Contrary to the hazard mapping depiction, low levels of seismicity such as seen in recent years may be more characteristic of the Ierapetra Isthmus, according to Caputo, Monaco, and Tortorici (2006). Tectonic movements thus may be accommodated through a number of faults, resulting in numerous low-magnitude earthquakes that disallow significant stress accumulations and consequent large seismic disturbances focused on a few major faults. It is interesting to note that seismic activity beneath Crete seems constrained to the upper 20–40 km, and that an aseismic zone occurs at depths greater than 20–40 km (30 km beneath central Crete), apparently representing the interface between the subducting African plate and the overriding Aegean-Anatolian plate (Meier et al. 2004; Snopek et al. 2007). In today’s morphology, the small hill with the archaeological site and the ridge connecting it to the hills to the west are the result of active erosional processes controlled, first, by the structural fabric of the area whereby fractured rock along faults (rock gouge) is readily eroded, and, second, by erosion of the less-indurated stratigraphic units of limestones and marls. The consequence is a landscape primarily shaped by this

22

FLOYD W. MCCOY

Figure 3.5. Map of the structural geology in the area surrounding the Aphrodite’s Kephali archaeological site. Strike/dip symbols on the knoll where the site is located and some areas on the north slope of the ridge are averages from more than one reading of bedding attitude. For bedding attitudes: number adjacent to the strike/dip symbol is the dip angle in degrees; symbols with no numbers are from IGME (1959); circle with cross at the archaeological site indicates vertically dipping beds. Solid heavy lines depict faults mapped during this study; dashed where inferred. Symbols describing fault-plane attitude and relative motion along faults are noted by: tick mark with adjacent number indicates dip direction and angle where it could be determined; no mark indicates a vertical fault plane; U indicates upthrown block; D indicates the downthrown block where displacements could be determined. Long dashes represent either suggested continuation of faults or faults not seen in exposure but inferred from abrupt changes in bedding orientations. Short dashes trace the approximate positions of faults mapped by IGME (1959); fault plane attitude and relative motion along these faults is inferred, based upon the assumption that the Aphrodite’s Kephali ridge is a horst and graben structure like similar nearby topographic features. Data are from field mapping during 2006, 2007, and 2008. For positions of stations and sampling localities, see Fig. 3.10. Contours are in meters; contour interval is 4 m.

tectonic fabric into a geologically young morphology, which is modified by additional natural processes such as sheet-wash down slopes, small landslides especially where terrace walls have collapsed, and stream erosion along the northern side of the hill. Anthropogenically related erosion via grazing animals, mechanical equipment, and such has, today, overwhelmed geologic processes.

GEOLOGICAL SETTING The rocks exposed at Aphrodite’s Kephali and its environs are part of the extensive Neogene sequence of sedimentary rocks on Crete (Fig. 3.8). These represent the extensive fills of sediments deposited in the numerous structural grabens that formed during the east–west expansion of the arc

GEOLOGY AND GEOLOGIC HISTORY

23

Figure 3.6. Hazard map for Greece outlining areas prone to significant seismic damage. Note that the Ierapetra Isthmus–Gulf of Mirabello region is a high risk zone for destructive seismic activity, based on historic activity. Map is after Higgins and Higgins 1996.

Figure 3.7. Map of earthquake epicenters in the area of the Gulf of Mirabello and Ierapetra Isthmus for 1998–2008. Plotted are 82 epicenters. All have magnitudes (Mw) of less than 5, with most occurring at crustal depths of less than 33 km. Open triangles depict final epicenter positions from the Bulletin of the International Seismological Centre (Berkshire, UK). Filled triangles depict epicenter positions that remain tentative pending additional study of seismological data as of this writing, from the Monthly Hypocenter Data File distributed by the National Earthquake Information Service of the U.S. Geological Survey. Map compiled with data from the Institute of Geodynamics (National Observatory of Athens), the U.S. Geological Survey, and the Incorporated Research Institutions for Seismology, Washington, DC. For geographic reference, selected archaeological sites (black dots) and selected towns (large open circles) are shown.

24

FLOYD W. MCCOY

Figure 3.8. Map portraying the general rock units exposed on Crete (after Higgins and Higgins 1996). The archaeological site at Aphrodite’s Kephali is indicated by a large star.

front accompanying crustal extension during the Miocene (Figs. 3.1, 3.3, 3.4). These grabens were shallow-water marine basins (with depths of less than perhaps a few hundred meters within the oceanic photic zone) with high rates of sedimentation from surrounding land areas where uplift kept pace with rates of basin subsidence (Meulenkamp 1971; Potsma, Fortuin, and van Wamel 1993; Fassoulas 2001; Duermeijer et al. 1998, 2000). Local underwater ridges, hills, and scarps provided sites for coral and algal reefs, these likely representing tectonic features such as fault scarps (Fassou-las 2001; Reuter, Brachert, and Kroeger 2006). Shallow-water conditions resulted in high biological productivity in both the pelagic and benthic environments, as indicated by high carbonate contents and fossil assemblages (ichnofossil, microfossil, and megafossil remains). Fossil assemblages and the lack of minerals derived from anoxia within sediments indicate well-oxygenated conditions throughout the water column and within sediments, with an oceanographic regime of vigorous water motion and mixing. An input of organic debris from terrestrial and oceanic sources is certainly evident from twig impressions and ichnofossil traces such as burrows, tracks, and trails

from organisms feeding on particulate organic debris mixed into sediments. High rates of sedimentation resulting in quick burial of organic material may locally have caused some anoxia in sediments, as indicated by pyrite-filled small burrows. Evidence for these high sedimentation rates also comes from sedimentary structures (e.g., turbidites, slump deposits, scoured channels, sediment fills, conglomerates, and breccias). The geology of the Ierapetra Isthmus and Gulf of Mirabello area, incorporating the archaeological site at Aphrodite’s Kephali, is shown in Figure 3.8. The general stratigraphic section in the Ierapetra area is depicted in Figure 3.9.

STRATIGRAPHY AND LITHOLOGY A detailed stratigraphic section over 120 m thick was mapped below the archaeological site along the north, east, and south flanks of the ridge (Figs. 3.5, 3.10). Additional mapping was done on the hill adjacent and west of the site, as well as into the foothills above, and up-section of, the area shown in Figure 3.5. Because extensive terracing severely limits exposures, particularly along ridge slopes, it

GEOLOGY AND GEOLOGIC HISTORY

25

Figure 3.9. Stratigraphic chart of the Neogene sedimentary rocks in the Ierapetra area generalized from stratigraphic sections in four areas of the Ierapetra Basin: Phaneromeni (F), Asari/Vasiliki (A/V), Kalamafka/Makrylia (K/M), and Parathiri (P). The stratigraphic sequence forming the hill and ridge at Aphrodite’s Kephali is outlined with a solid black line; the dashed line indicates the stratigraphic section exposed west of the archaeological site in the hills above the site. The presence of numerous unconformities (depicted by a thick wavy line)—each representing a significant and regional hiatus in the depositional history—speaks to active tectonism shaping, and ultimately destroying, the sedimentary basins where these sediments accumulated. Two unconformities extend across the entire section, one forming the upper boundary of the section (representing final uplift and destruction of the Neogene basins; this is essentially today’s ground surface), and the second in the latest Miocene during the Messinian stage (representing the dessication and disappearance of the Mediterranean Sea for approximately 400,000–500,000 years by the closure of the Beltic Seaway connection across southern Spain to the Atlantic Ocean). See Table 3.1 for a geologic time scale. Symbols depicting lithologic types are those defined in Table 3.2, with the exception of those used in this chart for breccias (open and solid triangles in a row) and limestones (vertical lines within narrow bands). Additional symbols here: inverted open triangles = gypsum; curved parallel lines pointing to the right = slumped beds; curved feature with an inscribed V in the Kalamafka Formation depicts a coral/algal bioherm. Abbreviations are: A/V = Asari/Vasiliki; F = Phaneromeni; Fm = formation; K/M = Kalamafka/Makrilia; Mb = member; P = Parathiri; SP = Stratified Prina Series of Potsma, Fortuin, and van Wamel 1993. Modified from Fassoulas 2004.

26

FLOYD W. MCCOY

Figure 3.10. Locations of field station sites and identification numbers. The contour interval is 4 m.

Figure 3.11. Positions and alignments of cross-sections A–A', B–B', and C–C'''. The offset from C' to C'' approximately follows a field road. The star identifies the position of the archaeological site.

was not possible to map or infer a complete composite stratigraphic section. Exposures are best on the northern and eastern slope of the ridge. Topographic and stratigraphic data from these slopes, with structural geology data, are shown in two cross-sections (for positions of the cross-sections, see Fig. 3.11; for cross-sections, see Fig. 3.12). A composite stratigraphic section extrapolated onto an additional cross-section down the north flank of the ridge below the archaeological site is shown in Figure 3.13, sketched from field observations (for the alignment of this cross-section, see Fig. 3.11).

The regional stratigraphic sequence (Fig. 3.9) is a series of sedimentary rocks that are, from the lower/ older portion of the section (thalweg of the ravine along the northern side of the Aphrodite’s Kephali ridge) to the upper/younger portion of the section (in the low hills east of the Psiloritis Mountains), part of the Makrilia Formation (Tepheli Group) overlain by the Ammoudares Formation (Vrysses Group) and the lower portion of the Myrtos Formation (Phinikia Group). Rocks forming the ridge along which the archaeological site and the hill immediately to the east are sited as part of the

GEOLOGY AND GEOLOGIC HISTORY

27

Figure 3.12. Cross-sections A–A' and B–B' through the ridge and hill at the Aphrodite’s Kephali archaeological site, depicting faults, relative fault motion, and stratigraphic dips. For locations of the cross-sections, see Fig. 3.11. Note: faults likely extend deeper than shown; question marks indicate presumed fault motion inferred from geological mapping or nearby fault motion; dips are apparent dips rotated onto the plane of the cross-sections, unless the dip is parallel to the cross-section plane; patterns portraying rock units have no lithologic inference, rather they are diagrammatic to illustrate dip or apparent dip orientations. Vertical exaggeration is ca. 1.5x.

Figure 3.13. Cross-section C'–C''' (in two parts) down the north slope of the ridge and hill at Aphrodite’s Kephali (see Fig. 3.11 for location of this cross-section). This sketch (not to scale) depicts terraces and locations of outcrops relative to terraces. While this sketch is accurate for relative spacing and heights of terraces along the line of the cross-section, it is not accurate elsewhere because terraces often merge, split, terminate, or collapse. Note the relative paucity of outcrops. Bedding contacts are masked by terracing and dense vegatation, and the depiction of these contacts in this sketch are diagrammatic. Areas of unknown geological characteristics are left blank. Lithologic symbols follow those in Table 3.2.

28

FLOYD W. MCCOY

Makrilia Formation. The other formations are present in outcrops in the foothills above the ridge. No stratigraphic contacts were seen or mapped between these formations because of ground cover and terraces, and the succession is inferred from lithologic criteria. Geologic age is Late Miocene, from the late Tortonian stage (Makrilia Formation) through the Messinian stage (Table 3.1; Fig. 3.9). It is not clear if this section extends into the Pliocene, as shown on the Ierapetra Sheet of the Geological Map of Greece (IGME 1959). Fassoulas (2004) places the Myrtos Formation in the lowermost Pliocene, whereas the Geologic Map of Greece (IGME 1977) and Potsma, Fortuin, and van Wamel (1993) place this formation in the uppermost Miocene (Messinian). Critical to this question, and not seen in the field here, is the profound unconformity between the uppermost Miocene and Pliocene representing the approximately 400,000–500,000 year duration when the Mediterranean underwent dessication and evaporation of its seawater to precipitate enormous thicknesses of gypsum (Ryan et al. 1973; Cita 1982; Hsu 1983). A limited exposure of gypsum, however, was found in the hills above the archaeological site, and it may represent the latest Miocene dessication event. Rock types exposed at the Aphrodite’s Kephali archaeological site were determined by: (1) hand specimen observations using both a hand lens and a binocular microscope, (2) thin sections using a petrographic microscope, and (3) smear slides made from both sedimentary rock and residues after carbonate removal also using a petrographic microscope (Table 3.2; Fig. 3.13). Total content of calcium carbonate (CaCO3) was determined using the acid-weight loss method by dissolving small rock fragments (40–50 g) in dilute 1–2N hydrochloric acid. The residue remaining after the chemical reaction had stopped was washed three times with distilled water, dried, and weighed. This residue represented the noncarbonate fraction of the rock (Table 3.3 lists data from CaCO3 analyses). The total CaCO3 content was a critical component in identifying lithologic types (Table 3.2). Dominant rock types (Tables 3.2, 3.3) include the following: clastic and bioclastic marls and limestones (including calcarenites), limestones, marls, and minor thin intercalations of claystones. Soft poorly indurated claystones and marls are

interbedded as thin laminae within some units. In a thick sequence of soft calcareous marls and limestones/chalks near the base of the stratigraphic section near the streambed in the ravine north of the ridge, a cave has been excavated (Fig. 3.13). The cave, whose diameter is about 1.5 m, extends approximately 15 m into the hill where it terminates in a small domed room. Characteristic ichnofossils are horizontal networks of filled worm burrows along bedding planes. Burrows are preserved as molds with diameters up to 2 cm, with lengths on the order of decimeters, often in elaborate anastomosing networks, and with slight swelling at junctions. These fossils appear to be Thalassionoides sp. burrows left by small anthropods feeding within sediments and along the sediment surface. Oblique (to the bedding planes) worm burrows of the same size, or smaller, are present, but rare; they may be Planolites sp., the burrows made by wormlike animals feeding within sediments. Smaller burrows, only a few millimeters in diameter, are less common and noticeable only by black/brown iron stains. The stains are likely from pyrite fills in the burrows, representing chemically reducing conditions either within the burrows post-mortem or within the sediments due to high rates of sedimentation; these fossils may be Chrondites sp. The assemblage and abundance of ichnofossils is indicative of a well oxygenated benthic environment, with some supply of organic debris derived from productivity within the water column and perhaps washed in from land (impressions of twigs were seen in some rocks), and they have been found in sedimentary rocks of similar age elsewhere in East Crete (Potsma, Fortuin, and van Warmel 1993). Megafossils are less common and include bivalves (e.g., Graphyea sp., Pecten sp., Ostrea sp., and other undetermined types), algal fragments, bryozoan frgments, corals (Porites?), and rare broken echinoid spines. They were likely detrital components washed into, and mixed within, basin sediments, and are indicative of nearby rocky substrates. This assemblage of fossils is characteristic of similar sedimentary rocks of equivalent age elsewhere in Central and East Crete (Potsma, Fortuin, and van Warmel 1993; Fassoulas 2001; Reuter, Brachert, and Kroeger 2006). Carbonate content varied between 38%–95% with an average of 73% (Table 3.3). Carbonate

GEOLOGY AND GEOLOGIC HISTORY

29

Table 3.2. Sedimentary rock classification scheme. Nomenclature and graphic patterns are those commonly used in sedimentary geology, with some simplification for this study. Additional symbols for special types of sediment not shown on this chart: gypsum = open inverted triangles (aligned in a row); chert = solid triangles with base down (aligned in a row; note exception to this in Fig. 3.9). 1Mean diameter of the dominant size of particles forming the sediment, following the standard Wentworth grade scale: gravel >2 mm; sand 2–1/16 mm (62 μm); silt 62–64 μm; clay 50% clastic sand particles composed of calcium carbonate, thus could be fragments of eroded carbonate rocks or biogenic material (e.g., shells). 5Chalk refers to a less-indurated, softer, friable limestone, usually white, that easily rubs onto hard surfaces or into one’s hand, as differentiated from limestone, which can be very hard and well-indurated with colors ranging from white to black. 6Claystone is usually massive without fissility (splitting easily into thin slabs along closely spaced roughly parallel surfaces such as bedding planes), whereas shale has marked fissility.

components—determined from thin sections—are dominantly subangular and sand- to coarse siltsized limestone fragments (“ind calc frags” in Table 3.3) at varying amounts (trace amounts to 40%), in addition to bioclastic particles (foraminifera, shell fragments, algal plates, coral fragments, ostracodes[?]: see Table 3.3), in a micritic cement. Microfossils often are poorly preserved due to diagenetic alteration. Noncarbonate detrital components, as determined both from thin sections and smear slides of residues from acid-wash carbonate-removal processing, are mainly quartz grains, with clays,

micas, chert(?), and some heavy minerals (usually an amphibole). Sedimentary structures mapped during field surveys included: normal size-graded bedding (coarse to finer-sized grains, from base to top of layer), cross bedding, scour features, flute marks (grooves carved into underlying beds by down slope movement of gravels), and load casts (depressions in underlying soft-sediment beds due to rapid accumulation of sediment). These structures are characteristic of turbidite sequences, each representing a single sediment cascade onto the ancient seafloor. Complete turbidite sequences (Bouma sequences)

30

FLOYD W. MCCOY

were not common, but present. A few rock units were massive sequences (up to 3 m) of sandstone with no apparent sedimentary structures. Thin calcarenites composed of broken fragments of bivalves were often present as thin interbeds. Some imbrication structures (alignment of disk-shaped particles into oblique stacks) were noticed in gravels (granules and pebble sizes), representing vigorous near-bottom currents. Most spectacular was a large, approximately 1.5 m in diameter, olistolith (displaced boulder of biogenic material) of an algal/coral bioherm (reef). Corals were only slightly out of growth position, and the boulder had bowed the underlying sediments downward indicating it was a displaced block deposited on firm sediment. Biogenic components in the olistolith included algal mats much like stromatolites, branching corals (Porites?), clam shells, and calcareous worm tubes, tightly cemented in a fine-grained carbonate matrix. Pockets of calcarenites were also present. Thin sections showed bacterial filaments, broken shell fragments, and algal plates in a micritic groundmass. Similar olistoliths are common and form stratigraphic markers in sedimentary rocks of similar age on East Crete such as the Kalamafka Formation in the lower part of the Tepheli Group (Fig. 3.9; Potsma, Fortuin, and van Warmel 1993; Fassoulas 2001).

STONE/FOSSIL FINDS AND SOURCES As used here, “stone” refers to both rock and fossil remains found during excavations at the archaeological site. Many of the rocks found in exposure are well-indurated and hard, thus good for building stones such as those found at the archaeological site, or for shaping into tools (e.g., the mortar found at the site); Table 3.3 provides a qualitative summary of these stones. In the mapped area surrounding the site, no locations where quarrying activities might have occurred were clear in the harder, more indurated rock units except for a series of three to four flat ledges, 20 to 40 cm wide, of indefinite length (extending beneath thick thorn bushes and brush), which may represent remnants of a quarry (noted on Fig. 3.13). In the vicinity of

field site 28 (Fig. 3.10) on the northwestern side of the saddle between the hill with the archaeological site and the higher hill to the northwest, a lowgradient slope appears unusual in this topography and may represent slope alteration due to ancient quarrying activity. Modern terracing, however, masks evidence to confirm this speculation. Also in this saddle, many well-rounded, diskshaped pebbles of black Tripolitza limestone were found scattered on the surface. While it is not clear if they represent a modern or ancient input, these pebbles were also found during archaeological excavations, suggesting that the input may be ancient. No source nearby was obvious. The nearest exposure of a conglomerate with these pebbles was found in a road cut a few kilometers away in the foothills northwest of the site, interbedded in a complex stratigraphy of breccias, petroliferous limestones, sandstones, and gypsum (lower portion of the Myrtos formation of the Phinikia Group; see Fig. 3.9). No quarry was obvious at this exposure; these pebbles occur in a poorly cemented conglomerate and are easily eroded and removed. During archaeological excavations, small fragments of broken dolomitic limestone, with conchoidal fractures resembling chert, were recovered during flotation, one of which is a broken fragment of a cast of a Pecten sp. shell. They are probably derived from dolomitic interbeds within the Tripolitza formation (found elsewhere in the vicinity; in Fig. 3.8 it is noted as “Permian Limestone”). A broken fragment of an echinoid (sea urchin) spine was recovered during flotation. Similarly large spines were not found in sedimentary rocks near the site, and it must be assumed that this has been brought from elsewhere (such large spines are known from Messinian marly chalks elsewhere in the region [e.g., Potsma, Fortuin, and van Wamel 1993], including exposures to the west of the site). Fragments of bivalve shells were recovered during excavations of the site from a number of locations. These fossils are common in rocks exposed at Aphrodite’s Kephali. All stone and fossil fragments found during the excavations, while found in an archaeological context, are displaced from their geological stratigraphic context, so it is difficult to assign them either a source or geological age.

GEOLOGY AND GEOLOGIC HISTORY Sample CaCO3 (%) No.1

Color2

Ind.3

Dominant Minerals and Fossils4

31 Rock Type6

1 [1]

58

pale yellow (2.5Y 8/3)

M

qtz, HM, bryozoa, algal plates, shell frags, foraminifera, burrows, bivalves (Gryphea? Ostrea?)

bioclastic marl

2 [5]

74

white (10YR 8/1)

M

burrows, bivalves (Gryphea? Ostrea?)

clastic limestone

3 [6]

75

pale yellow (2.5Y 7/4)

M

qtz, burrows

clastic limestone

4 [7]

82

white (10YR 8/1)

W

plag, fdpr, hblde, HM, calc nodules

clastic limestone

5 [7]

93

white (10YR 8/1)

W

burrows (pyrite-filled)

limestone

6 [8]

90

very pale brown (10YR 7/4)

W

burrows, shell frags, honeycomb structures(?)

limestone

7 [9]

82

very pale brown (10YR 7/3)

W

qtz, HM, burrows(?)

clastic limestone

8 [10]

69

very pale brown (10YR 7/3)

W

qtz, HM, burrows(?)

clastic limestone

9 [11]

78

very pale brown (10YR 7/3)

P

qtz, HM

clastic limestone

10 [11]

67

white (10YR 8/1)

P

qtz, HM, shell frags, algal plates, foraminifera

bioclastic limestone

11 [11]

72

white (10YR 8/1)

M

qtz, HM, shell frags, algal plates, foraminifera

bioclastic limestone

12 [11]

64

white (10YR 8/1)

P

qtz, HM, shell frags, algal plates, foraminifera, burrows

bioclastic limestone

13 [12]

81

white (10YR 8/1)

M

qtz, HM, shell frags, algal plates, foraminifera

bioclastic limestone

14 [15]

75

white (10YR 8/1)

M

qtz, HM, shell frags, algal plates, foraminifera

bioclastic sand7 and claystone

Table 3.3. Summary of data on rocks exposed and sampled at Aphrodite’s Kephali. 1Sample number here is a laboratory designator. Those not italicized indicate that observations employing smear slides and hand specimens using a hand lens were the basis for sediment/fossil descriptions and rock indentification. Italicized numbers indicate thin sections were also employed for descriptions and identifications. Numbers enclosed within brackets refer to the field station number (for locations of field stations, see Fig. 3.10); samples 25 and 26 are from exposures in the hills above Aphrodite’s Kephali (off map to left in Fig. 3.10). 2Colors are identifiers from the Munsell Color Chart. 3Ind. = induration, refers to the relative degree of cementation, thus a qualitative measure of rock hardness in terms of: P = poor = poorly cemented, powdery, either crumbles with light pressure in one’s hand or breaks easily when hit with a rock hammer, poor for use in buildings or tools; M = mod. = moderately well cemented, breaks rapidly when hit with a rock hammer, marginal for use in buildings or tools; W = well = well cemented, difficult to break when hit with a rock hammer, good for use in buildings and tools. 4Components listed in relative abundance (first is most abundant, etc.); most are present in minor (trace to