Archaeological Survey of the Gournia Landscape: A Regional History of the Mirabello Bay, Crete, in Antiquity (Prehistory Monographs) [Illustrated] 9781931534673, 1931534675

Citation preview

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE A Regional History of the Mirabello Bay, Crete, in Antiquity

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE A Regional History of the Mirabello Bay, Crete, in Antiquity

PREHISTORY MONOGRAPHS 37

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE A Regional History of the Mirabello Bay, Crete, in Antiquity by L. Vance Watrous, Donald Haggis, Krzysztof Nowicki, Natalia Vogeikoff-Brogan, and Maryanne Schultz contributions by John Hayes, Angeliki Kossyva, and Eberhard Zangger

Published by INSTAP Academic Press Philadelphia, Pennsylvania 2012

Design and Production INSTAP Academic Press, Philadelphia, PA Printing and Binding Thomson-Shore, Inc., Dexter, MI

Library of Congress Cataloging-in-Publication Data An archaeological survey of the Gournia landscape : a regional history of the Mirabello Bay, Crete, in antiquity / by L. Vance Watrous . . . [et al.]. p. cm. — (Prehistory monographs ; 37) Includes bibliographical references and index. ISBN 978-1-931534-67-3 (hardcover : alk. paper) 1. Minoans. 2. Mirabello Bay Region (Greece)—Antiquities. 3. Excavations (Archaeology)—Greece—Mirabello Bay Region. I. Watrous, Livingston Vance, 1943DF221.C8A738 2012 939′.18—dc23 2012015212

Copyright © 2012 INSTAP Academic Press Philadelphia, Pennsylvania All rights reserved Printed in the United States of America

Στο Μάρκο Γεωργίου Περρονικολή, Ψυχή των Γουρνιών

Table of Contents

List of Tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ix List of Maps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xi List of Figures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xiii List of Plates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xvii Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xxi Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xxiii Chronology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xxv List of Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xxvii PART I. THE GOURNIA PROJECT AND THE REGION 1. The Field Project, L. Vance Watrous. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 2. The Region, L. Vance Watrous and Eberhard Zangger. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 PART II. HISTORY OF SETTLEMENT DURING ANTIQUITY 3. Final Neolithic Period: Egalitarian and Heterogeneous Communities, L. Vance Watrous and Maryanne Schultz. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 4. Early Minoan I–II Periods: Emergence of a Stratified Society, L. Vance Watrous and Maryanne Schultz. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

viii

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

5. Early Minoan III–Middle Minoan IA Periods: Disruption and Social Reorganization, L. Vance Watrous and Maryanne Schultz. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 6. Middle Minoan IB–II Periods: Growth of Regional Factions and Conflict, L. Vance Watrous and Maryanne Schultz. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 7. Middle Minoan III–Late Minoan I Periods: The Rise of a Regional State, L. Vance Watrous and Maryanne Schultz. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 8. Late Minoan IIIA–IIIB Periods: Depopulation and Mycenaean Occupation, L. Vance Watrous. ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 9. Late Minoan IIIC–Geometric Periods: Retreat into Refuge Settlements, Krzysztof Nowicki............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69 10. Orientalizing–Classical Periods: Population Nucleation and Development of the Polis, L. Vance Watrous. ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .77 11. Hellenistic and Roman Periods: Expansion of the Isthmus in an International Era, Natalia Vogeikoff-Brogan...........................................................................81 PART III. CONCLUSIONS 12. Conclusions, L. Vance Watrous. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 PART IV. APPENDICES A. Catalog of Sites, L. Vance Watrous.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 B. Neolithic and Bronze Age Pottery, Donald Haggis. ............................................135 C. Late Minoan IIIC–Orientalizing Pottery, Donald Haggis. ......................................155 D. Archaic–Hellenistic Pottery, L. Vance Watrous. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 E. Roman Pottery, John Hayes and Angeliki Kossyva. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .175 Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .189 Tables Maps Figures Plates

List of Tables

Table 1.

List and settlement hierarchy of Final Neolithic sites.

Table 2.

List and settlement hierarchy of EM I–II sites.

Table 3.

List and settlement hierarchy of EM III–MM IA sites.

Table 4.

List and settlement hierarchy of MM IB–II sites.

Table 5.

List and settlement hierarchy of MM III–LM I sites.

Table 6.

Neopalatial settlement changes by percentage.

Table 7.

List and settlement hierarchy of LM IIIA–IIIB sites.

Table 8.

List and settlement hierarchy of LM IIIC sites.

Table 9.

List and settlement hierarchy of Protogeometric sites.

Table 10.

List and settlement hierarchy of Geometric sites.

Table 11.

List and settlement hierarchy of Orientalizing–Archaic sites.

Table 12.

List and settlement hierarchy of Classical sites.

Table 13.

List and settlement hierarchy of Hellenistic sites.

Table 14.

List and settlement hierarchy of Roman sites.

Table 15.

Early and Late Roman imports.

List of Maps

Map 1.

Gournia, Kavousi, and Vrokastro Survey areas.

Map 2.

Gournia Survey, transects, 1992.

Map 3.

Gournia Survey, transects, 1993.

Map 4.

Gournia Survey, transects, 1994.

Map 5.

Map of East Crete.

Map 6.

Map of the Mediterranean.

Map 7.

Map of the Aegean.

Map 8.

FN sites in the survey area.

Map 9.

EM I–II sites in the survey area.

Map 10.

Plan of Gournia in EM II.

Map 11.

Plan of Sphoungaras cemetery.

Map 12.

Plan of Vasiliki in EM II.

Map 13.

Prepalatial off-site pottery density, EM I–MM IA sites.

Map 14.

Prepalatial off-site pottery types, EM I–MM IA sites.

Map 15.

Neolithic–EM II off-site chipped stone density.

Map 16.

Neolithic–EM II off-site chipped stone types.

xii

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

Map 17.

EM III–MM IA sites in the survey area.

Map 18.

Plan of Gournia in EM III–MM IA.

Map 19.

Plan of Vasiliki.

Map 20.

MM IB–II sites in the survey area.

Map 21.

MM IB–II site clusters.

Map 22.

Plan of Gournia in MM IB–II.

Map 23.

MM IB–II off-site pottery density.

Map 24.

MM IB–II off-site pottery types.

Map 25.

MM III–LM I sites in the survey area.

Map 26.

MM III–LM I off-site pottery density.

Map 27.

MM III–LM I off-site pottery types.

Map 28.

Plan of Gournia in MM III–LM I.

Map 29.

Plan of Gournia town.

Map 30.

Plan of Gournia palace.

Map 31.

Economic organization of Late Minoan I Gournia.

Map 32.

LM IIIA–IIIB sites in the survey area.

Map 33.

LM IIIC sites in the survey area.

Map 34.

LM IIIC off-site pottery density.

Map 35.

LM IIIC off-site pottery types.

Map 36.

PG sites in the survey area.

Map 37.

Geometric sites in the survey area.

Map 38.

Orientalizing–Archaic sites in the survey area.

Map 39.

Classical sites in the survey area.

Map 40.

Hellenistic sites in the survey area.

Map 41.

Hellenistic off-site pottery types.

Map 42.

Early Roman sites in the survey area.

Map 43.

Roman off-site pottery density.

Map 44.

Roman off-site pottery types.

Map 45.

Early Roman imported pottery types.

Map 46.

Early Roman imported pottery origins.

Map 47.

Late Roman sites in the survey area.

Map 48.

Late Roman imported pottery types.

Map 49.

Late Roman imported pottery origins.

List of Figures

Sherds without specific site or transect attributions in the captions lack specific site or transect provenance.

Figure 1.

Neolithic pottery (B1–B13).

Figure 2.

Neolithic pottery (B14–B21).

Figure 3.

Neolithic pottery (B22–B33).

Figure 4.

EM I–II pottery (B34–B44).

Figure 5.

EM I–II pottery (B45–B59).

Figure 6.

EM I–II pottery (B60–B74).

Figure 7.

EM I–II pottery (B75–B91).

Figure 8.

EM I–II pottery (B92–B106).

Figure 9.

EM I–II pottery (B107–B123).

Figure 10.

EM I–II pottery (B124–B130).

Figure 11.

EM III–MM IA pottery (B131–B145).

Figure 12.

EM III–MM IA pottery (B146–B160).

Figure 13.

EM III–MM IA pottery (B161–B176).

Figure 14.

EM III–MM IA pottery (B177–B190).

xiv

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

Figure 15.

EM III–MM IA pottery (B191–B203).

Figure 16.

EM III–MM IA pottery (B204–B217).

Figure 17.

EM III–MM IA pottery (B218–B236).

Figure 18.

EM III–MM IA pottery (B237–B255).

Figure 19.

MM IB–II pottery (B256–B270).

Figure 20.

MM IB–II pottery (B271–B292).

Figure 21.

MM IB–II pottery (B293–B311).

Figure 22.

MM IB–II pottery (B312–B326).

Figure 23.

MM IB–II pottery (B327–B342).

Figure 24.

MM IB–II pottery (B343–B355).

Figure 25.

MM IB–II pottery (B356–B362).

Figure 26.

MM II–III pottery (B363–B370).

Figure 27.

MM II–III pottery (B371–B382).

Figure 28.

MM II–III pottery (B383–B399).

Figure 29.

MM II–III pottery (B400–B408).

Figure 30.

MM III–LM I pottery (B409–B428).

Figure 31.

MM III–LM I pottery (B429–B463).

Figure 32.

MM III–LM I pottery (B464–B482).

Figure 33.

MM III–LM I pottery (B483–B502).

Figure 34.

MM III–LM I pottery (B503–B520).

Figure 35.

LM IIIC pottery (C1–C12).

Figure 36.

LM IIIC pottery (C13–C22).

Figure 37.

LM IIIC pottery (C23–C35).

Figure 38.

PG pottery (C36–C48).

Figure 39.

PG pottery (C49–C62).

Figure 40.

PG–Geometric pottery (C63–C70).

Figure 41.

Geometric pottery (C71–C85).

Figure 42.

Geometric pottery (C86–C99).

Figure 43.

Geometric pottery (C100–C111).

Figure 44.

Geometric–Orientalizing pottery (C112–C123).

Figure 45.

Orientalizing pottery (C124–C138).

Figure 46.

Orientalizing pottery (C139–C143).

Figure 47.

Archaic–Classical pottery (D2, D4, D6–D9, D27–D28, D30, D32–D34, D38).

LIST OF FIGURES

Figure 48.

Classical pottery (D39, D44, D46–D47, D49, D51–D52, D56–D57, D67, D79).

Figure 49.

Hellenistic pottery (D82, D84, D86–D88, D90).

Figure 50.

Hellenistic pottery (D92, D94).

Figure 51.

Roman pottery (E1–E5).

Figure 52.

Roman pottery (E6–E9).

Figure 53.

Roman pottery (E10–E16).

Figure 54.

Roman pottery (E17–E25).

Figure 55.

Roman pottery (E26–E31).

Figure 56.

Roman pottery (E32–E35).

Figure 57.

Roman pottery (E36–E43).

Figure 58.

Roman pottery (E44–E51).

Figure 59.

Roman pottery (E52–E61).

Figure 60.

Roman pottery (E62–E69).

Figure 61.

Roman pottery (E70–E78).

Figure 62.

Roman pottery (E79–E86).

Figure 63.

Roman pottery (E87–E96).

Figure 64.

Roman pottery (E97–E101).

xv

List of Plates

Plate 1A.

Survey team members, 1992. Front row, left to right: John Zielinski, Maria Mastropavlou, Ann Reynolds, Vance Watrous. Second row, right to left: Donald Haggis, Christina Papadaki, Eleana Vardaki, Yvonne Widner, Angeliki Kossyva. Rear: Matthew Franklin.

Plate 1B.

Survey team members, 1993. Front row, left to right: Ann Reynolds, Susan Petrakis, Yvonne Widner. Second row, right to left: Vance Watrous, Mihkriban Oxbasaran, Angeliki Kossyva, John Zielinski, Dimitra Mylona, Donald Haggis hugging Matthew Watrous. Rear: Matthew Franklin.

Plate 1C.

View of Pacheia Ammos and the northern Isthmus.

Plate 2A.

Isthmus of Ierapetra, looking west-southwest toward Episkopi and the Plain of Ierapetra.

Plate 2B.

Aerial photo of Gournia valley.

Plate 2C.

View north to Pera Alatzomouri ridge from south of Gournia.

Plate 3A.

View from north of Episkopi basin, Prophetes Elias (150) hill in background.

Plate 3B.

View from west of cave, site 130.

Plate 3C.

View of Cha Gorge and site of Katalimata, looking east.

Plate 4A.

View westward of Vasiliki village and lepidochoma clay bed.

Plate 4B.

View of EM site 132 from north.

Plate 4C.

View of site 131 from north.

Plate 5A.

Cyclopean wall at site 131 from west.

xviii

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

Plate 5B.

Cyclopean wall at site 131 from west.

Plate 5C.

View of site 144 from west.

Plate 6A.

View from west of Cyclopean terrace at site 25.

Plate 6B.

View north of Cyclopean terrace at site 46.

Plate 6C.

View east of Cyclopean wall at Selimas, site 120.

Plate 7A.

View eastward from Gournia central court to peak sanctuary on Mount Ephendis Stavromenos.

Plate 7B.

Minoan sherds and pebbles from Mount Ephendis Stavromenos.

Plate 7C.

Minoan sherds and pebbles from Mount Ephendis Stavromenos.

Plate 8A.

View from Ephendis Stavromenos north to Mochlos.

Plate 8B.

LM III chamber tomb at Gournia from west (Pera Alatzomouri).

Plate 8C.

Aerial view of shipshed at Gournia.

Plate 9A.

Palace at Gournia, view from north toward northwest corner.

Plate 9B.

Roman aqueduct near Hagia Pelagia (Gournia).

Plate 9C.

Roman trapetum from site 66.

Plate 10A.

Roman olive press near site 56.

Plate 10B.

Roman millstone from site 117.

Sherds without specific site or transect attributions in the captions lack specific site or transect provenance.

Plate 11.

Neolithic pottery (B1–B8).

Plate 12.

Neolithic–EM pottery (B9–B18).

Plate 13.

EM pottery (B19–B33).

Plate 14.

EM pottery (B34–B40).

Plate 15.

EM pottery (B41–B56).

Plate 16.

EM–MM IA pottery (B57–B65).

Plate 17.

EM III–MM IA pottery (B66–B71).

Plate 18.

EM III–MM IA pottery (B72–B81).

Plate 19.

EM III–MM IA pottery (B82–B101).

Plate 20.

EM III–MM IA pottery (B102–B115).

Plate 21.

EM III–MM IA pottery (B116–B121).

Plate 22.

EM III–MM IA pottery (B122–B132).

Plate 23.

MM IA–MM II pottery (B133–B148).

Plate 24.

MM IB–MM II pottery (B149–B155).

LIST OF PLATES

Plate 25.

MM IB–MM II pottery (B156–B165).

Plate 26.

MM III–LM I pottery (B166–B178).

Plate 27.

MM III–LM I pottery (B179–B186).

Plate 28.

LM IIIC pottery (C1–C10).

Plate 29.

LM IIIC pottery (C11–C15).

Plate 30.

LM IIIC pottery (C16–C19).

Plate 31.

LM IIIC pottery (C20–C27).

Plate 32.

LM IIIC–PG (C28–C36).

Plate 33.

PG pottery (C37–C51).

Plate 34.

PG–Geometric pottery (C52–C67).

Plate 35.

Geometric–Orientalizing pottery (C68–C82).

Plate 36.

Orientalizing pottery (C83–C86).

Plate 37.

Orientalizing pottery (C87–C99).

Plate 38.

Archaic pottery (D1, D3–D5, D9–D15).

Plate 39.

Archaic pottery (D16–D26, D29).

Plate 40.

Classical pottery (D31, D33–D38, D40–D42).

Plate 41.

Classical pottery (D43–D46, D48, D50–D51, D53–D56, D58).

Plate 42.

Classical pottery (D59–D73).

Plate 43.

Classical–Hellenistic pottery (D74–D81, D83, D85, D89–D91, D93).

Plate 44.

Roman basins (E1–E12).

Plate 45.

Late Hellenistic–Early Roman imported amphorae (E13–E20).

Plate 46.

Local Roman amphorae (E21–E29).

Plate 47.

Roman beehives (E30–E34).

Plate 48.

Local Roman cooking pottery (E35–E47).

Plate 49.

Local Roman cooking pots (E48–E60).

Plate 50.

Roman imported wares (E61–E67).

Plate 51.

Roman plain ware lamps (E68–E72).

Plate 52.

Roman roof tiles (E73–E78).

Plate 53.

Terra Sigillata pottery, Group I (E79–E89).

Plate 54.

Terra Sigillata pottery, Group II (E90–E99).

Plate 55.

Phocean Red Slip stamped vase from Transect 393 (E100).

Plate 56.

Imported African Red Slip Wares (E101–E111).

xix

xx

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

Plate 57.

Red Color ware and Cypriote Red Slip Wares (E112–E117).

Plate 58.

Local Roman wasters (E118–E123).

Plate 59.

Local Roman wasters (E124–E126).

Preface

This project was conceived in 1991 by L.V. Watrous. At that time, the Mirabello Bay area had been the focus of excavation at various sites—e.g., Vasiliki, Mochlos, Kavousi, and Pseira—for a number of years. Moreover, two surveys had just been completed in the Istron and Kavousi areas. All of this work left the central part of the Mirabello Bay area, the Isthmus of Ierapetra, and the palatial site of Gournia untouched. For this reason, Watrous approached Costis Davaras, Ephor of East Crete, with the idea of undertaking a regional survey around Gournia to be followed by excavation at Gournia itself. Davaras agreed, and the three-year project began in 1992 with the general goal of providing a regional context for the site of Gournia. The area included the northern coast of Crete in the Gournia Valley and the northern half of the Isthmus of Ierapetra, ending in the valley of Episkopi (Map 1, middle area). Archaeological survey work has been carried out in Crete for over fifty years (Gkiasta 2008). Surveys record archaeological data horizontally, that is, across the present land surface. For this reason, regional survey evidence differs fundamentally from excavation data. Excavations proceed vertically and produce immense amounts of material from one location over time. Excavation data, therefore, is particularly suited to answer questions about the subsistence, technology, exchange, social organization, and ideology of households within a single settlement, both synchronically and diachronically. Survey evidence, on the other hand, has a broader focus, and is particularly informative on the environment, population(s), and social organization of an entire region. In most cases, previous surveys have been content to document their pattern of settlement and limit their interpretation to questions answerable by survey data alone. As a consequence, the interpretations of settlement

xxii

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

patterns and their development remained, in these studies, rather abstract and attributed on a tenuous basis. This study will combine both sets of evidence. Hence, this volume differs from other survey publications in two main ways. First, as this survey was the last and most central of three surveys (Hayden 2004a; Betancourt, Davaras, and Hope Simpson 2005; Haggis 2005) conducted in the Mirabello region, it will attempt to tie together the data from all three of these surveys. Second, it goes beyond the survey data to consider, at some length, the evidence from local excavations, so as to provide an in-depth and integrated picture of the regional socio-economic development. This is particularly important because previous publications of excavations and surveys have concentrated on their limited areas, and also because settlement patterns by themselves, based as they are on the broad sweep of a ceramic sequence, inevitably can mask specific events and developments within a region. This volume, therefore, is meant as a regional archaeological study of the Mirabello Bay area during antiquity. Harriet Blitzer will publish the post-antique finds from this project.

Acknowledgments

This project was carried out with a permit from the Greek Ministry of Culture under the auspices of the American School of Classical Studies at Athens in cooperation with the 24th Ephorate of Prehistoric and Classical Antiquities, Hagios Nikolaos, Crete. Especially important was the support of Costis Davaras, the Ephor of East Crete, who co-directed the project during 1992. Throughout our work Metaxia Tsipopoulou and Vili Apostolakou of the 24th Ephorate gave us their official support. We would also like to thank Philip Betancourt, Barbara Hayden, and Tom Brogan for their advice and collegial help. Watrous first learned about the Minoan pottery from the region from Gerald Cadogan, the excavator at Myrtos Pyrgos near Ierapetra. Haggis also worked with the Myrtos Pyrgos material as well as with the excavation finds from Kavousi and Vronda. Kellee Barnard and Eva Sikla discussed their deposits from Mochlos with us. John Hayes helped us identify and date the Roman and later pottery from the survey. Taylor Dabney took the object photographs and Roxanne Doxan drew many of the ceramic profiles and excellent maps. We would like to thank Georgos Koinakis for the many trips to look at archaeological monuments around Pacheia Ammos and for information about the archaeological history of the region. We would also like to thank Emmanuel Koinakis for his unfailing philoxenia and support during our stay in Pacheia Ammos. Finally, we would like to acknowledge the copy-editors, Kapua Iao, Carol Leyba, and Barbara Hayden, who carefully helped in putting this manuscript together, and the anonymous reviewer for the constructive comments. Funding for the project was provided by the Institute for Aegean Prehistory and the University at Buffalo.

xxiv

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

We thank the following people who worked on the field project: Harriet Blitzer, Yvonne Widner, Justin Van Ness, John Zielinski, Pauline Gleeson, Kristen Hannold, Dena Gilby, Katerina Skourtopoulou, Amy Schwartzott, Vasiliki Zografaki, Jo Ann Polley, Cheryl Floyd, Dimitra Mylona, Eleni Hatzidopavlaki, Maria Mastropavlou, Eleana Vardaki, Christina Papadaki, Francesca Tronchin, Ann Reynolds, Fani Stavroulaki, Angeliki Kossava, Mihkriban Ozbasaran, Susan Petrakis, and Georgos Haroulis, our ipoptis (ephoreia supervisor) in 1994 (Pls. 1A, 1B). Maryanne Schultz wrote the sections describing the sites in the Gournia, Kavousi, and Vrokastro surveys as part of her Ph.D. thesis (Schultz 2007). As the general editor, Watrous shepherded the manuscript through to completion.

Chronology

All prehistoric dates (calibrated 14C) are based on Coleman 1992, Manning 1999, and Tomkins 2007.

Final Neolithic Early Minoan I Early Minoan II Early Minoan III Middle Minoan IA Middle Minoan IB Middle Minoan II Middle Minoan III Late Minoan I Late Minoan II Late Minoan IIIA Late Minoan IIIB

4000–3500 B.C. 3500–2900 B.C. 2900–2200 B.C. 2200–2100 B.C. 2100–1900 B.C. 1900–1850 B.C. 1850–1800/1775 B.C. 1800/1775–1725/1700 B.C. 1725/1700–1570/1540 B.C. 1570/1540–1490/1450 B.C. 1490/1450–1375/1360 B.C. 1375/1360–1220/1200 B.C.

Late Minoan IIIC Subminoan Protogeometric Geometric Orientalizing Archaic Classical Hellenistic Early Roman Late Roman Early Byzantine

1220/1200–1100 B.C. 1100–970 B.C. 970–800 B.C. 800–700 B.C. 700–600 B.C. 600–500 B.C. 500–323 B.C. 323–66 B.C. 66 B.C.–A.D. 400 A.D. 400–700 A.D. 700–828

List of Abbreviations

BG c. ca. EC EG EMC EMMC EH EM est. FN ha km LG LM

Black Glaze century approximately Early Cycladic Early Geometric Early Minoan cooking ware Early Minoan medium coarse Early Helladic Early Minoan estimated Final Neolithic hectares kilometer(s) Late Geometric Late Minoan

LN LPMC LPP m m asl MG MM NPP pers. comm. PG PGB PPC PPMC pop. SM

Late Neolithic Late Prepalatial medium coarse Late Prepalatial meter(s) meters above sea level Middle Geometric Middle Minoan Neopalatial phyllite personal communication Protogeometric Protogeometric B Protopalatial cooking ware Protopalatial medium coarse population Subminoan

Part I

The Gournia Project and the Region

1

The Field Project L. Vance Watrous

This chapter begins with a summary of the early archaeological research in the Mirabello region that led up this survey, followed by a description of the

goals, methods, and results of the field project. The final section lays out the theoretical background for the sociopolitical emphasis of the study.

Previous Archaeological Research in the Mirabello Region Archaeological research in the Mirabello area began during the first two decades of the 20th century. Detailed discussions of archaeological work in the Vrokastro and Kavousi survey areas can be found in Shaw 1990; Hayden 2000; and Haggis 2005, 2–4. As the section below reveals, the extensive archaeological work carried out in the Mirabello region provides a wealth of detail that can be of help in interpreting the settlement data from the Gournia Survey. A summary of this work follows. In 1900, Harriet Boyd (later Harriet B. Hawes), a student at the American School of Classical Studies at Athens, excavated at the Early Iron Age sites of Kavousi and Vronda just east of our survey zone (Map 1). The following year Boyd was led to the coastal site of Gournia by George Perakis of

Vasiliki. Boyd excavated at Gournia during 1901, 1903, and 1904, using on average 100 workmen, 10 women, and several assistants, including Richard Seager and Edith Hall (Hawes et al. 1908; Fotou 1993; Watrous 2000). At Gournia, Boyd uncovered a portion of the town, including several Middle Minoan (MM) houses, a Late Minoan (LM) I palace and central court, cobbled streets, houses (Hawes et al. 1908), and Early Minoan (EM) II–MM III house tombs (Soles 1992, 1–40). Along the coast of Gournia, Boyd excavated a MM I–II house and a LM I shipshed (“Shore House”). On the ridge top (Pera Alatzomouri) immediately east of Gournia, Boyd dug a MM I–II house and LM III chamber tombs. In 1904 and 1906, Richard Seager excavated at Vasiliki, uncovering a settlement spanning the EM

4

L. VANCE WATROUS

II–LM I and Roman periods (Seager 1905, 1907, 1908). In 1906 and 1907, Seager dug a portion of the Final Neolithic (FN)–LM III town and cemetery on the island of Pseira in the Bay of Mirabello (Seager 1910). Also in 1906, he uncovered some 60 LM III vases from a cemetery at Episkopi (Seager 1907) and a LM IIIC tholos tomb at Vasiliki Kephala (Seager 1907, 129–132, pl. 30). In 1908, Seager visited the offshore island of Mochlos and uncovered a wealthy EM II–LM III settlement and cemetery (Seager 1909, 1912). Later, in 1914 and 1915, he discovered a MM I–LM I cemetery of pithos burials on the shore at Pacheia Ammos (Seager 1916). Edith Hall began work at Gournia, uncovering a large assemblage of EM III–MM IA pottery deposited in a pit outside the wall of the upper town (Hall 1904), and part of the coastal cemetery at Sphoungaras (Hall 1912). In 1910 and 1912, she excavated at the LM IIIC–Orientalizing hilltop site of Vrokastro, east of our survey area (Hall 1914). In 1912, Hall turned her attention to the nearby coastal EM II–LM I and Roman–Byzantine settlement of Priniatikos Pyrgos (Hall 1914). As a result of these two early decades of work, by 1918, five of the largest Bronze Age settlements— Gournia, Mochlos, Vasiliki, Pseira, and Priniatikos Pyrgos—and three Early Iron Age sites—Kavousi, Vrokastro, and Vronda—in the Mirabello Bay area had been partly excavated and published. For the next 50 years (1919–1970), no major archaeological work took place in our survey zone. Members of the Greek Archaeological Service did, however, undertake rescue excavations. In 1919, S. Xanthoudides discovered a LM IIIB chamber tomb just north of Episkopi (Xanthoudides 1920–1921, 157–162). In 1940, two more LM IIIA–IIIB chamber tombs at Episkopi were excavated by N. Platon (Kanta 1980, 140–160). In 1951, S. Alexiou dug a LM IIIA–IIIB chamber tomb on Alatzomouri just east of Gournia (Alexiou 1954; Kanta 1980, 143, 144). In 1963, J. Sakellarakis excavated a LM IIIA–IIIB chamber tomb at the southwest edge of Pacheia Ammos, near Seager’s villa (Alexiou 1967; Kanta 1980, 144). Archaeological interest in the area was rekindled in the 1970s (see Muhly and Sikla, eds., 2000 for a history of American work in the Mirabello region). In 1971, J. Soles and C. Davaras began a long-lasting collaboration focusing on Gournia

and Mochlos (Soles 2003). In 1971 and 1972, they cleaned the tombs at the north end of the upper town (Soles 1992, 1–40), in the process of which they discovered a cache of EM II–MM I vases, a seal, and gold and silver objects (Soles 1992, 13–17). In 1970, A. Zois started new excavations at the site of Vasiliki. His investigations (1970–1982, 1990–1993) have been quite important because he has clarified the architectural phases and their plans at Vasiliki. Late in the 1970s, the Ministry of Culture in Athens requested that the foreign archaeological schools return to sites that their compatriots had previously excavated earlier in the century. Given the American work in East Crete during the first two decades of the 1900s, American archaeologists turned to the Mirabello region in East Crete to reopen old excavations. During 1979 to 1992, G. Gesell, L. Day, and W.D.E. Coulson carried out an initial study and excavation at Kavousi Kastro and Vronda. Excavation at Kavousi Kastro revealed a small hilltop settlement dating to the LM IIIC–Orientalizing period (Coulson et al. 1997). At Vronda they uncovered 12–15 houses of the LM IIIC period and Early Iron Age tombs (Gesell, Day, and Coulson 2000; Day, Klein, and Turner 2009). In the 1970s, Soles and Davaras started investigations at Mochlos; excavations began in 1989 and lasted until 1994, followed by a second phase during 2004–2005. Excavation resumed in 2010. Their research has uncovered more of the EM/LM I–III settlement, as well as a late Hellenistic fortress (Soles and Davaras 2000; Vogeikoff-Brogan 2004). In 1986, N. Papadakis dug nine LM III tombs on the mainland opposite Mochlos. In the Istron area, west of our survey zone, B. Hayden began in 1979 to study the Early Iron Age settlement at Vrokastro; this work expanded by 1986 into the Vrokastro Project, a regional survey project (Map 1, left area) directed by Hayden and J. Moody (Hayden 2003b, 2004a). This project intensively surveyed a southwestern portion of the Mirabello region, including the Priniatikos Pyrgos–Istron coast and the large inland Meseleri Valley (Hayden 2004a, fig. 2). Sites from the Vrokastro survey mentioned in the text are derived from Hayden 2003b and 2004a. In 2005, Hayden began to assist M. Tsipopoulou in excavating in and around the coastal EM–LM and Roman–Byzantine

THE FIELD PROJECT

site of Priniatikos Pyrgos. This project now continues as a cooperative venture with the Irish Institute of Hellenic Studies at Athens. During 1988–1990, D. Haggis conducted an intensive survey of the Kavousi area (Map 1, right area), immediately north and east of our survey zone (Haggis 2005). Sites from the Kavousi survey mentioned in the text are drawn from Haggis 2005. In 2003–2006, Haggis directed an excavation at the LM IIIC–Archaic site of Azoria near Kavousi (Haggis et al. 2004). In 1984, P. Betancourt and C. Davaras began to map the architecture exposed by Seager on the island of Pseira, and subsequently to reopen excavations there, resulting in a much clearer picture of this EM–LM III town. In 1988, a survey of the island was carried out (Betancourt, Davaras, and Hope Simpson, eds., 2005). In 1996 and 1997, Betancourt dug an EM–MM IA metallurgical establishment at Chrysokamino on the coast near Kavousi (Betancourt 2006). Within the last 15 years, M. Tsipopoulou, W.D.E. Coulson, and K. Nowicki have excavated at two nearby Early Iron Age sites, Chalasmenos and Katalimata, located next to the Cha Gorge at the eastern edge of our survey zone (Tsipopoulou and Coulson 1994; Nowicki 2000, 2008; Tsipopoulou 2001, 2004a). T. Iliopoulos exposed a LM IIIC shrine complex and settlement on the Kephala hill south of Vasiliki (Iliopoulos 1998). Tsipopoulou also excavated burial caves near Kavousi and a LM IIIC tholos tomb near Vasiliki (Tsipopoulou, Vagnetti, and Liston 2003). Since 2005, the 24th

5

Ephorate of Prehistoric and Classical Antiquities has excavated a rock shelter (Apostolakou, Betancourt, and Brogan 2007–2008) and a small industrial area near Pacheia Ammos. In 1991, it was proposed that an airport should be built in the Isthmus of Ierapetra, directly south of the site of Vasiliki (Schultke and Zois 1993). Accordingly, the Greek Archaeological Service undertook the responsibility to carry out an archaeological survey of the area south of the village of Vasiliki, west of the asphalt Ierapetra road, as far south as Episkopi. While unpublished, the survey revealed several important sites, including a massive Minoan tower at Skourdeli tou Metochi and an EM I hilltop site at Aphrodite Kephala, south and west of Episkopi. South of the Mirabello region, three other recent investigations deserve mention. In 1967 and 1968, P. Warren conducted an excavation at Phournou Koriphi, on the south coast about 10 km west of Ierapetra, exposing an EM IIA–IIB hamlet (Warren 1972). Between 1970 and 1982, G. Cadogan excavated portions of the EM II–LM I site of Myrtos/Pyrgos on the south coast, about 2 km west of Phournou Koriphi, revealing an ashlar hilltop villa at the center of a small settlement (Cadogan 1977–1978, 1992). Finally, rescue excavations by the 24th Ephorate of the Greek Archaeological Service in and near the town of Ierapetra have revealed many Roman structures of this important polis (summary reports have appeared in Archaiologikon Deltion from 1960 to the present).

Project Goals and Methods Goals At the beginning of our fieldwork we laid out the four basic questions our survey project was designed to address (Watrous et al. 2000, 472): 1. How did the people of Gournia exploit the surrounding environment? 2. How large was the population of Gournia and the settlements around it? 3. What kind of economic relations did Gournia have with its region?

4. What was the role of Gournia in the political organization of the region? These questions, focused on the site of Gournia, remain basic to our study. In preparing our finds for publication, however, we have broadened our geographic scope to include the entire Mirabello region, and deepened our investigation into the regional causes of the cultural developments within the whole Mirabello area, which includes not just Gournia but other major sites such as Priniatikos Pyrgos, Mochlos, Pseira, and Vasiliki.

6

L. VANCE WATROUS

Hence, this present volume has three basic goals. The first, which includes question 2 above, is to describe the results of the archaeological survey carried out around the site of Gournia and the northern Isthmus of Ierapetra in East Crete during 1992–1994. This survey parallels the recently published surveys in the immediate vicinity to the east (Kavousi) and west (Vrokastro) of the Mirabello region (Hayden 2004a, 2005; Haggis 2005). The second goal is to combine all data-sets from the three surveys in order to synthesize the material into a diachronic picture of settlement within the

Bay of Mirabello region from the Neolithic period through the Late Roman period. The third goal, which includes questions 1, 3, and 4, above, is to use the survey data and the information from all local excavations in the area to trace how the population, economy, and sociopolitical organization of the region evolved during antiquity, and to interpret these developments in light of current theories of cultural change. Given that our survey area reached its highest level of social complexity at Gournia in the Neopalatial period, this will be a main focus of our conclusions (Ch. 12).

Fieldwork Methods During fieldwork (1992–1994), we surveyed five days a week and studied our finds for a half-day on Saturdays. Our field methods were based on Watrous’s experience during the Mesara Survey (Watrous, Hadzi-Vallianou, and Blitzer 2004, 12– 14). Our methods differed from this earlier survey in three important ways: we conducted our survey with a single team, we picked up all artifacts during survey, and we tried to walk transects of a consistent 100 x 100 m size. Field walking started at 8:00 a.m. and ended by about 3:00 p.m. After a rest, we washed and studied pottery from 5:00 to 7:00 p.m. Our field team consisted of 10 persons who walked 100-m-long transects on line at 10-m intervals (Pls. 1A, B). An eleventh person (D. Haggis) carried the map and paced out the length of each transect. We used 1:500 scale topographic Greek Army maps. Watrous, one of the field walkers, carried a camera and a logbook that described the daily activities of the team. Field walkers picked up all artifacts, placing them in a plastic bag. At the end of each transect, walkers filled out a form for each transect, which was placed in the bag before it was closed and put in a backpack. Transect forms asked the walker to identify their place in line; the type of soil (red, light marly, or brown); the rough degree of visibility (poor = less than 20%; medium = 30%–70%; good = more than 70%); the type of land use (e.g., uncultivated, plowed, garden, terraced, irrigated, hothouses); the type of vegetation (e.g., olives, beehives, wild brush/thistles, grass/fodder, almond trees, carob trees); and any noteworthy features or structures

(e.g., field house, spring, well, cistern, walls). Haggis plotted the location of each transect on the map, numbered it, and noted its direction. As field-walking was in progress, a walker, noting a distinct rise in the number of sherds, or the appearance of tiles or walls, relayed this information to Watrous and to Haggis, who marked this information on the map. If the number of sherds increased dramatically, one or more walkers would shout out this information to Haggis for notation on the map. In this way, the rough size and location of a site could be recorded during field-walking. If the walker was not certain that the artifacts constituted a site, the transect-walking was stopped while Watrous and Haggis went to the spot to make a decision. We identified a site on the basis of the quantity of artifacts (usually sherds or chipped stone) relative to the areas immediately surrounding it. No absolute totals were used. The most common difficulty in identifying a site involved the date(s) of the artifacts being found. That is, if an area had substantial land use during more than one period, then the rise in the number of artifacts could be mistaken for a site. As the field walkers learned to date the pottery, such mistakes became less and less frequent. Fields near a site often possessed substantial numbers of sherds from that site (the halo effect). Defining the extent of a site (as opposed to its adjacent sherd-filled fields) was determined by counting the number of sherds in adjacent transects: when the numbers climbed 200%–300%, the area was considered part of the settlement. Consequently, defining site size often

THE FIELD PROJECT

did not take place during transect-walking. In some cases, sites were identified days later, when dating and counting sherds revealed a notable rise in the number of sherds of a certain date (which would be subsequently reinvestigated in the field). Sites were not investigated until the entire area around them had been surveyed. Once we had determined a site existed, the practice was to revisit the site on another day to answer certain questions. When the rough center of the cluster was found, we collected sherds within a 5-m-radius circle, in lines radiating from the center along the four cardinal points of the compass. This was usually a fairly accurate and objective method of determining the site boundaries. That finished, diagnostic sherds within the four quadrants of the site were gathered separately. In many cases, the 5-m-radius circles were too small to give us the information we desired. This was especially true for multi-period sites. We were interested in trying to determine the size of a site during its several separate periods of use because such information can be very revealing about population trends. In these cases, we often enlarged the collection areas substantially to get the needed data. At times, it was necessary to revisit and recollect on a site in order to determine its size at different periods. While on site, the team members filled out a site form that asked the following questions: name of site (toponym); nature of finds (e.g., pottery, tiles, chipped or ground stone, walls); possible date; possible function; transect location; topographic situation (e.g., hill, ridge, slope, valley bottom); land use; location of possible water source; disturbance (e.g., erosion, plowing, colluvium, construction, planting); visibility on site; vegetation; collection interval and length of each transect line; and site size. Finally, a rough map of the site was drawn, placing it in relation to recognizable landmarks (e.g., road, ravine, church, field wall). In the afternoons, Watrous and Haggis studied the pottery washed from previous days. Students washed pottery that been collected that day, or took turns learning how to identify the shapes and dates of the pottery with Watrous and Haggis. Pottery to be studied was laid out on a table and sorted by clay fabric. The distinctive local fabric types, defined by Haggis and Mook (1993), were crucial in helping us to date coarse-ware sherds (which was what we usually picked up on survey).

7

Each field-walker’s collection within a single transect was counted, identified by shape (e.g., pithos, carinated cup, roof tile, tsoukali or cooking pot), and dated in a corroborative fashion by Haggis and Watrous. The combined experience of these two ceramic experts meant that virtually every sherd collected in the survey could be dated. Differences of opinion were extremely rare: the only ongoing issue concerned the identification of EM III–MM IA pottery. Thanks to Haggis’s restudy of the site finds in 1994, Watrous came to accept much of Haggis’s dating for this period. This method of recording the survey pottery made it possible to create a precise mosaic of finds, located accurately to within 100 m across the Cretan landscape. During our first field season in 1992, we covered approximately 3 km2 in 227 transects (Map 2). Starting from our western boundary, we surveyed the valley of Gournia, as well as the promontories of Pera Alatzomouri, Phylakeion, and the immediate area of Pacheia Ammos. We spent a good deal of time defining the settlement of Gournia and its immediate surroundings. The following season (1993), we continued in an eastward direction, along the coast between Pacheia Ammos and Halepa, south of the asphalt road to Kavousi, and across the Isthmus floor in the areas of Evraika and Kamina, as far south as Vasiliki village and the Cha Gorge (Map 3). That year we walked some 13 km2 in 577 transects. Our aerial coverage increased because the floor of the Isthmus was largely flat and devoid of archaeological sites. In 1994, we continued southward, east of the asphalt road to Ierapetra, through the Episkopi–Papadiana–Kato Chorio basin, stopping just south of the village of Kato Chorio (Map 4). We walked about 18 km2 in 459 transects. Defining the giant Early Iron Age site of Prophetes Elias (150) required a week. By the end of the survey, we had surveyed 24 km2, walked 1,263 transects, and discovered a total of 156 new archaeological sites in our area. One way in which the project differed from other field projects is that we believed that a successful research project should have a strong educational aspect to it. Our students were urged to learn some modern Greek; they attended seminars on historical and archaeological topics that we arranged, and they traveled together to museums,

8

L. VANCE WATROUS

churches, and archaeological sites on free days. The best background for comprehending the regional data collected on a survey is to understand traditional life in the region. The ultimate goal of our studies is to understand the ancient peoples and

the systems—social, economic, and political—by which they lived. The collected artifacts are only a means to these goals: they are not ends in themselves.

Methods of Study and Analysis As the study progressed, we began to publish our preliminary findings. In 1995, we read a paper outlining our survey results at the annual meeting of the Archaeological Institute of America (Watrous and Blitzer 1995, 313). In 1996, Haggis delivered a multi-authored preliminary report on the Gournia Survey at the Seventh Cretological Congress in Herakleion (Watrous et al. 2000). In 2000, an article on the Gournia region during the Neopalatial Period appeared in the Malcolm Weiner Festschrift (Watrous and Blitzer 1999). In 2001, Watrous examined the relationship between the LM IIIA–IIIB population in the Isthmus and the LM IIIC sites there (Watrous 2001a). In 2006–2007, Schultz used the Neolithic–Late Minoan I survey data as the basis for her Ph.D. thesis (Schultz 2007). In 2008, papers based on our survey findings (Watrous 2007, 2008) were delivered at the annual meetings of the Archaeological Institute of America in Chicago and at the INSTAP Study Center of East Crete. In the years following fieldwork, we made three major changes in our approach to the material. First, we expanded our perspective to take into account the entire Mirabello region, which included the Vrokastro and Kavousi survey areas and all local excavations. Hence, we focused less exclusively on Gournia and tried to understand regional developments from a wider perspective. Second, we adopted a systemic approach to our study of the data, exploring the interconnected roles of the environment, population growth and/or decline, trade, industry, and social organization. Third, we used this information as a way of focusing on the question of social organization within specific settlements and the wider region through time. The reasons for this are laid out in some detail below, in the conclusions (Ch. 12).

We categorized settlements by site type, according to size (Watrous, Hadzi-Vallianou, and Blitzer 2004, 24–26). The smallest habitation is a field site that measures no more than 0.1 hectare in area. Field sites are typically agricultural field houses, temporary work areas, or seasonal camps. Next, a farm site may range in size from 0.1 to 0.2 ha. Hamlets have the largest variability in site size, with a span of 0.2 to 1.0 ha. In some cases, large hamlets clearly approach the size of a village, which is 1.0 ha and over. No sites from the Late Neolithic through the Neopalatial reach the size of a center (10+ ha). Although these categories are somewhat arbitrary, they are useful in comparing settlement data. One of the primary lessons to come out of our study of survey-based settlement patterns, from a systemic perspective, is that the coarse-grain chronological phases set up by survey pottery (often coarse-ware fabrics) do not necessarily match the contemporary sequence of developments as revealed from other perspectives, such as excavation data. In periods of overall continuity (e.g., Neolithic, EM I–II, and MM IB–II), survey data and the resultant settlement patterns seem harmonious with what is known from excavated evidence for trade, technology, and social organization during the period. But, for periods when a series of more complex changes took place, as in EM III–MM IA and MM III–LM I, the coarse settlement patterns may mask important changes that need to be pulled out of the chronologically finer-grained data obtained from excavation. For this reason, one cannot interpret the settlement pattern of these latter periods merely on the basis of survey data. In this volume, we have tried to integrate the various sources in our interpretation of regional developments.

2

The Region L. Vance Watrous and Eberhard Zangger

Mirabello Region The Gournia survey area is located at the center of the Mirabello region (Map 1, middle area) in eastern Crete. At the heart of this area is the semicircular Bay of Mirabello, 17 km wide east–west and 10 km north–south. Five rivers run down from the mountains and extend through river valleys to the coast, emptying into the Mirabello Bay—at Hagios Nikolaos, Ammoudara, Istron, Pacheia Ammos, and Tholos (Kavousi). Each of these valleys has had a major settlement, either inland (as Neapolis, Kritsa, Kalo Chorio, Episkopi, and Kavousi) or coastal (as Mochlos), although tourism has swollen the coastal settlements at Hagios Nikolaos, Istron, and Pacheia Ammos. Finally, at the eastern edge of Mirabello Bay are located the island of Pseira as well as the village of Mochlos, which is based at the head of its own narrow coastal valley. Physically, each of these valleys is a separate area enclosed by mountains (Pl. 1C), but they are linked together via their only access to the outside world, Mirabello Bay. This volume takes the geographical boundaries of the Mirabello region as its area of focus. The survey

areas—around Vrokastro (Hayden 2003a, 2003b, 2004a, 2004b, 2005), Gournia, and Kavousi (Haggis 2005)—are naturally considered together. Geographically tangent, they each consist of a central river valley framed on both sides by extensive mountains. Geologically part of a karstic landscape, the mountains are limestone and the valley floors are a mixture of terra rossa and marl soils. Each valley has an extensive drainage system originating in the surrounding mountains, consisting of deeply incised valleys that empty into at least one river. The bedded limestone mountain slopes produce seepage and springs sought out by ancient and modern farmers and shepherds. Environmentally, these valleys are part of a single coastal Mediterranean ecozone, marked by similar soils and common vegetation: wild shrubs and grasses, planted trees (tamarisks, olive trees, carob), crops (principally grain and vines), and indigenous trees (mainly inland), such as pines, oak, and juniper. Hayden, Moody, and Rackham (2004b) and Haggis (2005, 9–22) provide a detailed description of the environments of the

10

L. VANCE WATROUS AND EBERHARD ZANGGER

Vrokastro and Kavousi areas. Some 50 years ago, before the advent of deep wells and mechanized irrigation, the basic local agricultural economy of each valley and its terraced slopes was similar, consisting of grains, olives, vines, almonds, and pears, complemented by shepherding (sheep and goat) and fishing. The Mirabello region is encircled by mountains: the foothills (500–700 m) of the Dictaean Range on the west and southwest and the steep Thriphti Range (1,476 m) on the east (Map 1). To the west, the upland plain of Lasithi nestles between the Dictaean Range (2,148 m) and the valley of Neapolis, which connects the Lasithi region with the north coast of Central Crete and the towns of Malia, Chersonissos, and Herakleion (Map 5). To the south, the Isthmus opens up into a large coastal plain dominated by the seaside town (pop. 8,575 in 1981) of Ierapetra (ancient Hierapytna). West of Ierapetra, the coast is lined by villages, including Myrtos, which extend as far as the Mesara. To the east, the mountainous end of the island is settled by small coastal and inland villages, such Makrygialos, Zaros, Zakros, and Palaikastro, and their regional center, the north coast town of Siteia. The largest valley within the Mirabello region is the Isthmus of Ierapetra (Pl. 2A), the main focus of the Gournia Survey. Throughout most of (pre)history, the Isthmus has been the heartland of the Mirabello region: the Minoan palace at Gournia, the large Early Iron Age town of Prophetes Elias, the powerful polis of Hierapytna, and the Christian bishopric at Episkopi were situated there. The importance of the Isthmus derives from its strategic location and areas of fruitful agricultural land. As Harriet Boyd recognized (Hawes et al. 1908, 20), the Isthmus, the narrowest north–south section of Crete, has always been a potential trade corridor linked to the north and south coasts of the island and to the Aegean and North Africa, beyond. All east–west traffic along the north coast of Crete, whether by sea or land, had to pass along the

coast of the Mirabello area; hence the maritime sites at the good harbors of Mochlos and on the island of Pseira. The Isthmus offered a short route between the Aegean world and points south, to Egypt, North Africa, and the southern Levant (Maps 6, 7). This is especially apparent during the Roman period. Sea routes passing through the area included those along Kasos, Karpathos, Rhodes, and along the Anatolian coast to Cyprus and the Levant, northward to Thera and to the other Cyclades, and westward along the western string to the metal-rich islands of Siphnos, Kythnos, and Lavrion on the Attic coast. The importance of the sea traffic to this area explains the existence of a settlement on the rocky, dry isle of Pseira. The relative lack of water in the Mirabello region has, until recently, restricted the size of most local settlements—at Kalo Chorio (pop. 612 in 1981), Gournia (est. LM I pop. about 400), Vasiliki (pop. 245 in 1981), Kavousi (pop. 715 in 1981), and Episkopi (pop. 654 in 1951). Before the introduction of deep wells for irrigation and the construction of agriculture-producing greenhouses in the 1960s, the few exportable commodities of the area were, according to villagers, carobs, tomatoes, and olive oil. The northern Isthmus and its adjacent valleys of Vrokastro and Kavousi do not match the size of the coastal plain around the town of Ierapetra (pop. 8,575 in 1981) or the larger valleys elsewhere on Crete, such as the Mesara and the plain of Malia, where the first Minoan palaces arose. In this respect, the Mirabello region was secondary to the major urban centers of the island. But the region is interesting in its own right, partly because it represents the majority of the Cretan (and Aegean) landscape, and partly because this small area has been more intensively investigated by archaeologists than any other area of the island, which allows us to trace its socioeconomic development in great detail.

Gournia Survey Area The Gournia survey zone was partially defined by other, tangent survey projects (Map 1), and by the need to include an area of sufficient size to

generalize about the entire Mirabello and Isthmus region. As a result, the survey covered the coastal plain on the Bay of Mirabello, the area between

THE REGION

Gournia and Kavousi, and the northern half of the Isthmus of Ierapetra, as far south as Episkopi, an area of about 24 km2. Four kilometers to the west, the Kalo Chorio Valley, part of the Vrokastro survey region (Hayden 2003b, 2004a, 2005), is somewhat topographically similar to the Gournia area in that it runs down to the Bay of Mirabello. This area is separated from the Isthmus by a range of mountains that extends down to the coast. On the other hand, the area surveyed by Haggis (2005), the valley running from Kavousi to the north coast and surrounding mountains, represents a small northward branch or extension of the Isthmus. Specifically, the western boundary of the survey was defined by the mountain ridge and the road to the Monastery of Hagia Phaneromeni at the western edge of the Gournia Valley. This was the eastern border of the Vrokastro survey area. Along the Isthmus, the survey ventured up the lower slopes of the Dictaean foothills west and north of the village of Vasiliki. The road from Vasiliki village to the main Pacheia Ammos–Ierapetra asphalt road was the southwestern border. Farther south, the Pacheia Ammos–Ierapetra asphalt road was our western border. The hilly area south of Vasiliki village, as far south as Episkopi, was surveyed by the 24th Ephorate in preparation for the construction of a proposed new airport (still unrealized). To the north, the survey extended from the sea at the small bay near the Gournia camping complex (about 1 km west of Gournia) to the promontory of Halepa (some 600 m east of Pacheia Ammos). The rocky isle of Konida, about 1 km offshore of Pacheia Ammos, was investigated, but no sign of ancient settlement was found. Farther east, the asphalt road leading to Kavousi, which was the southern edge of the Kavousi Survey (Haggis 2005), formed our northern boundary. The eastern border of the survey was along the base of the scree-covered slope (up to 200 m from the valley floor) of Thriphti. South of the Cha Gorge, the entire slope of Thriphti was walked, up to the very top of the ridge of Xylogournes, 824 m above the floor of the Isthmus, and as far south as the peak of Prophetes Elias, located some 2 km southeast of Episkopi. Between 2008 and 2011 Kostas Chalikias surveyed the island of Chryssi off-shore of Ierapetra as part of his Ph.D. thesis (Chalikias 2011). Colluvium within our survey zone is largely concentrated in a 200-m mantle of light brown soil

11

along the coast, at the base of the marly hill slope running between Pacheia Ammos and Vasiliki, and in the valley basin between the villages of Episkopi, Epano Chorio, and Kato Chorio. The remaining land surfaces are red paleosols visible south and west of Gournia, on the Isthmus floor between Pacheia Ammos, Kavousi, and Episkopi, and on the lower slopes of Thriphti. Hence, we would estimate that roughly 85% of the survey zone consists of ancient land surfaces, compared to roughly 2% for the recent Pylos Messenia Survey (E. Zangger, pers. comm.). Until the drilling of deep wells in the 1960s, the life and economy in the northern Isthmus was severely restricted by lack of water. Available water in the area came from the perennial river at Gournia, seasonal streams that run from Vasiliki to Pacheia Ammos and from the Cha Gorge to Halepa, numerous small springs along the slope of Thriphti, and traditional shallow wells. Limited to December through March, rainfall is relatively low. Aside from kitchen gardens located on the outskirts of Vasiliki, Monastiraki, Papadiana, and Episkopi, the rest of the bottomland is dry most of the year and suitable only for growing grains, primarily barley, and for grazing. Field-walking for the survey began in the coastal valley of Gournia (Pl. 2B). This small area is transected by the Gournia River, which flows from the uplands of Asari (near the Vrokastro survey area) some 5 km northward toward the coast. The Gournia Valley is defined by a steep mountain ridge to the west and south. A tributary streambed from this western slope flows past the Venetian-period Church of the Panagia down to the river. The gentle marly slope of the valley west of the river is said by locals to be especially good farmland for grain and is heavily terraced with walls that probably date to the Venetian period. The valley slope runs down to the coast, opening between the ridge of Palaiobarda on the west, and the slope of Sphoungaras and marl ridge of Pera Alatzomouri on the east (Pl. 2C). Portions of the coast are formed by 10- to 30-m high cliffs that drop into the sea. Recent research (Watrous, forthcoming) has shown that the present coastline of this valley has subsided several meters since antiquity. East of the river, the Minoan settlement of Gournia, excavated by Harriet Boyd during 1901–1904 (Boyd 1904, 1904–1905), occupies a low ridge that runs parallel to the river down

12

L. VANCE WATROUS AND EBERHARD ZANGGER

to the coast. South and east of Gournia, the twin Cretaceous limestone hills of Selli and the ridge of Pera Alatzomouri close off the valley. The Church of Hagia Pelagia is located at the southern edge of the Minoan site of Gournia. A settlement dating to the Roman period was found near the church. It was fed by a small aqueduct built of cement and stone, which runs up the slope of the valley toward Asari. To the east, a small ravine separates Alatzomouri from the next promontory of Phylakeion (with its chapel of Hagios Dimitrios) and the village of Pacheia Ammos. The village is at the western end of a kilometer-long sandy coastline that forms the northern mouth of the Isthmus of Ierapetra. Pacheia Ammos, only recently grown to a village, is situated on the slope and base of Phylakeion, west of a small valley and stream that run down from the village of Vasiliki, approximately 3 km to the south. Fifty years ago, Pacheia Ammos was mainly seasonally inhabited. Fishermen and farmers used shallow wells and cisterns to water their fields. Roman and Minoan settlements lie under Pacheia Ammos. The Minoan community buried its dead along the coast in a cemetery excavated by Richard Seager in 1914–1915 (Seager 1916). The eastern end of the coast, called Halepa, is marked by the new Church of Hagia Eirene and the mouth of a stream that runs from the Cha Gorge. A small boat harbor was recently dug on the coast at the base of Mount Halepa. Freshwater springs surface around Halepa and the EM I–IIA site on the coast, currently under excavation by the 24th Ephorate. Rising immediately east of Halepa, a ridge formed by Mount Halepa and Mount Schinias (Haggis 2005, fig. 2) separates the Bay of Pacheia Ammos from the smaller north–south valley of Kavousi, which was the focus of Haggis’s survey (Haggis 2005). The Gournia survey area continues southward from the coast into the Isthmus of Ierapetra. At its northern end, the Isthmus is a 2-km-wide corridor (Pl. 1C) flanked by the mountain ranges of Thriphti on the east and Mount Dicte on the west. Moving south from the coast the ground rises gently, leading into a flat, open plain consisting of dry red Pleistocene soils mixed with limestone, conglomerate, and sandstone. Today this area, called Evraika and Kamina, is covered with new, miniature Italian olive trees (morelia) and is divided in half by the north–south asphalt road that connects Pacheia Ammos to Episkopi and Ierapetra. The dry, steep

limestone slopes of Thriphti define the eastern edge of the plain. On the west, the border is marked by the stream that runs past Vasiliki and adjacent marl foothills of Mount Dicte. The old village of Vasiliki, with its locally spring-fed gardens and natural clay beds, is located on the western edge of this plain. Moving farther south, the valley floor begins to rise and narrow, merging with the surrounding gentle Pliocene marl hills, called Kazarma and Papadiana on the east and Tourloto Kephali on the west. At the base of Thriphti, the slope from Monastiraki to Prophetes Elias consists of a series of natural flysch shelves from which small springs leak. In antiquity, these locations were favored for settlement. Across the valley, south of Vasiliki (surveyed by the 24th Ephorate), marly foothills alternate with small valleys. Nearing Episkopi, the valley floor reaches its narrowest and highest point (elevation 100 m). Around Episkopi, several hills, notably Kazarma and Prophetes Elias, jut out to create a small, semienclosed valley basin (Pl. 3A), ringed by the villages of Episkopi, Papadiana, Epano Chorio, and Kato Chorio. These villages sit on slopes overlooking the confluence of mountain streams in the valley below. With springs, natural runoff, and heavily colluviated marl soil, this basin is rich farmland, hence the local density of settlement. South of Episkopi, the valley opens into a large coastal plain that gradually slopes down to the coast and Ierapetra. Our survey stopped at the northern edge of this coastal plain. The Gournia survey area contains several distinctive micro-environments that have influenced the ancient pattern of settlement. The sea is the first such zone. It links the villages of Mirabello by providing a common transportation route and sustenance. All of the Mirabello villages—Hagios Nikolaos, Kalo Chorio (Istron), Vasiliki (Pacheia Ammos), Kavousi (Tholos), and Mochlos—had and continue to have coastal harbors used by local fishermen. Despite the influx of cash-generating tourism, villagers still find it worthwhile to fish in the bay. According to Zangger, the coastline at Ierapetra has also changed little. Beach rock, along the Ierapetra coastline, dates to 7000 B.P., again suggesting little change. The second environmental zone includes the mountain slopes, which are situated at the edges of the Gournia survey area. These slopes would

THE REGION

have been forested in the Bronze Age. Both the Thriphti and Dictaean mountain ranges possess upland slopes and plateaus used by shepherds from local villages who graze their flocks of sheep and goats during the warmer months. In the early 20th century, summer settlements existed at Asari, Thriphti, and Melises (Hayden 2004a, 3, 18, 36, 88; Haggis 2005, 21). During the months of December through March, these flocks, a valuable source of wool and meat, were pastured in the lowland valleys. The mountains and upper slopes could also have produced wood (e.g., oak, pine), acorns, grapes, almonds, and pears. Stream-fed valleys are the third environmental zone, which includes the small agricultural streambeds near Gournia, Vasiliki, and Episkopi.

13

Soils there are a mixture of red Pleistocene paleosols and colluvial marl eroded from the slopes. Streams are lined with lush green vegetation, including reeds, rush, platanos, and fruit trees. Not surprisingly, the greatest density of ancient settlement was tangent to this zone. The fourth zone is the plain on the floor of the Isthmus. Although large in area, this zone was, until the introduction of deep wells and mechanized irrigation in the 1960s, limited in its potential because of its dryness and the fact that the red soil does not retain water well. Older villagers who can remember the area in the 1920s recall that the red soils in the middle of the valley floor were not farmed, but were used primarily for grazing.

Part II

History of Settlement during Antiquity

3

Final Neolithic Period: Egalitarian and Heterogeneous Communities L. Vance Watrous and Maryanne Schultz

This chapter begins by describing the Final Neolithic settlement data in the Gournia survey area in terms of site numbers, size, hierarchy, and population density. Contemporary site data from the survey projects at Kavousi and Vrokastro are

summarized in order to provide maximum coverage of the Mirabello region. Finally, the chapter explores wider questions of subsistence, exchange, and social organization in the region during the FN period.

Gournia Survey The earliest settlements known in the Gournia survey area date to the FN period. The Gournia Survey Project identified four FN habitation sites— one hamlet, one farm, and two field sites (Table 1; Map 8). Final Neolithic pottery was identified at Stavromenos (130), a hamlet-sized cave site on a steep slope about 1 km east of the modern village of Epano Chorio; at the farmstead Hagios Georgios (98), which is approximately 2 km southeast of Vasiliki in the southwest corner of the Gournia Valley; and at field sites at Panagia (3), a location 700 m west and upslope from the site of Gournia, and site 137, located east of the modern village of Kato Chorio. In addition, Boyd (Hawes et al. 1908,

37) reported one FN burial at Sphoungaras (10), near the north coast of Mirabello Bay. The presence of a burial there would suggest that the EM I–II site on the steep hill of Alatzomouri above Sphoungaras (10) may have started in the FN period. Recent archaeological work has produced two additional sites near the survey area. Nowicki (2008) has excavated a FN settlement at Katalimata, located high (elevation 340 m) up the extremely steep cliffs of Thriphti, directly overlooking the Cha Gorge at the eastern border of the Gournia survey boundary. Perched on a small shelf (invisible from below) on a cliff, Katalimata served as a FN refuge settlement consisting of one or perhaps two small

18

L. VANCE WATROUS AND MARYANNE SCHULTZ

houses. T. Iliopoulos (1998) excavated a FN, LM IIIC settlement on the hilltop of Kephala about 600 m south of Vasiliki village. Situated in the foothills at the southwest corner of the Gournia Valley, site 3 yielded FN pottery and chipped stone around the Church of the Panagia. At an elevation of 100 m and about a kilometer from the coast, the site overlooks a gentle slope that provides some of the best arable land in the region today. The few Neolithic sherds located on the site suggest that the settlement was small, perhaps occupied by a single family. The remaining three FN sites in the Gournia survey area are located in the Isthmus of Ierapetra. The site of Hagios Georgios (98) is situated on a slope of the Thriphti mountain range, approximately 2 km southeast of Vasiliki village, on the east side of the Isthmus Plain. Hagios Georgios, a church 200 m northwest of the FN site, marks the location of a perennial spring. Shelves of arable

land immediately to the east and west of the site offer ample environs for growing crops. The pottery and chipped stone spanned an area of 30 x 30 m, or 0.09 ha, and suggests that the site was a farmstead of one or two families. Approximately 1 km south of site 98, the hamletsized site 130 is on a steep slope of Thriphti, 200 m southeast of the modern Church of Hagios Stavromenos. Site 130 (Pl. 3B) is situated at a cave (elevation of 340 m) near Kephalovrysi, the largest spring in the Isthmus. A spacious ridge of arable land lies 200 m northwest of the sizable spring. The site consists of FN sherds and a scatter of obsidian over 50 x 50 m or 0.25 ha. East of the modern village of Kato Chorio, one Neolithic sherd was discovered at site 137. It is possible that this site is associated with the knapping site 138, located some 350 m to the west. Site 137 has immediate access to cultivable land on the plain of Ierapetra.

Kavousi Survey To the east of Gournia, the Kavousi Survey (Haggis 2005) investigated at least six FN sites (one hamlet, three farms, one field site, two burials, a metalworking site, and two possible field sites). Excavation below the plateia of Kavousi village (K24) produced much prehistoric (including FN) pottery. Situated at a prime, well-watered location at the edge of the valley, site K24 was probably an important FN settlement. Excavations at the site of Azoria (K71) indicate its size as 0.36 ha, or as large as 1.0 ha (Haggis 2005, 38, 60), making it the largest known FN settlement in the area. Azoria seems to have a fortified settlement plan that covers a significant portion of the hilltop (Haggis et al. 2004, 2007) and a permanent water source at the Kambos and Vronda springs. The small sites at Chondrovolakes (K67) and Vronda (K77) are located in the mountains south of the modern village of

Kavousi. In the foothills elevated above the Kambos Plain, the farm site at Alykomouri (K8) is situated near a perennial spring and good agricultural land, similar to the location of other FN farmstead sites in the Gournia Survey. The field site K16 sits on the Kambos Plain (Haggis 2005, 38, 59–62). Excavation has revealed a FN habitation site or farm (K34), which was probably associated with the burial caves at Theriospilio (K31), dug by Hall (Haggis 2005, 113), and the metallurgical site (K32) at Chrysokamino (Betancourt 2006). The latter site juts out over the Bay of Mirabello on the west side of Mount Chomatas, exposing it to the north and south winds needed by the smelting furnaces. A sheltered cove (Agriomandra) southwest of the site would have provided a safe landing for ships and a relatively short 2-km trek to the workshop area (Betancourt 2006).

Vrokastro Survey The publication (Hayden 2004a, 35–48) of the Vrokastro Survey combined the FN, EM I, and

EM IIA time periods in their discussion. This procedure masks at least two separate and distinct

FINAL NEOLITHIC PERIOD: EGALITARIAN AND HETEROGENEOUS COMMUNITIES

phases of settlement (see below). It is possible, however, to distinguish the fabrics (and hence the sites) of the two periods. Haggis (2005, 47) has provided a clear exposition, based on his excavation of the EM I site at Kalo Chorio, of the EM I (not FN) date of quartz and calcite-tempered ware. Since “Hearth ware” is not restricted to the FN period, but continues into EM I, we have identified a site as FN only when the preponderance of the ceramic assemblage consists of thick, darkburnished wares without quartz and calcite/marble inclusions. Hence, we have decided for the sake of

19

clarity to regard the quartz-and-marble-tempered fabric as a primary hallmark of the EM I period, although it may begin slightly earlier. Following these qualifications, the Vrokastro area produced one certain FN site (KP6). This farm, which yielded worked chert and a stone ax (similar to the ax described in Hall 1914, 121), sits on the high defensive hill of Vouno (elevation 150–180 m), near a spring situated about a half-kilometer from the sea. Its location is strikingly similar to the FN site of Panagia (3) near Gournia, which is removed from the coast and near a water source.

Final Neolithic Settlement in the Mirabello Region During the Final Neolithic period, except for its last phase, the first settlers in the Mirabello region chose inland sites located on low slopes along the edges of arable valleys near a perennial water source. Haggis (2005, 61) came to the same conclusion for the Kavousi area. All of the Gournia survey sites conform to this pattern. In terms of settlement hierarchy, FN sites were usually small: hamlets, farms, and seasonal field houses or camps. Most of the sites were located one or more kilometers apart from one another. Valley bottomlands apparently were avoided, probably for a variety of reasons— that is, they were wooded, were relatively dry with poor soils, and were indefensible. At least four local sites—Katalimata, Vronda (K77), Vasiliki Kephala, and Azoria (K71; Map 1)— are dramatic exceptions to this pattern: they were located inland, on high defensive hilltops or slopes. Unlike the other Neolithic sites, at least two of these are larger settlements. Katalimata, excavated by Nowicki (2008), is a refuge settlement perched on the cliff face (at an elevation of 300 m) above the Cha Gorge (Pl. 3C). Access was via an extremely steep scree slope and narrow path. The settlement apparently consisted of two houses, dated by Nowicki to the end of the FN I period (Nowicki 1999, 575). Although the Neolithic site on the mountain shoulder at Vronda was probably small (Haggis 2005, fig. 8), Vasiliki Kephala (Iliopoulis 2004), located on the steep ridge top south of Vasiliki village, was a substantial settlement that dates to FN–EM I. The recently excavated site of Azoria (Haggis 2005, 132) is located on the first hill

750 m south of Kavousi; the Neolithic settlement, on the southern acropolis, was village-sized. These defensive sites suggest a chronological phase within the FN period (extending into EM I) when safety was a major concern. Having studied FN sites throughout Crete, Nowicki (2008) has recently suggested that these defensible sites were founded by both Cretan refugees and new settlers during the FN period (and continuing into EM I), at a time when new immigrants began to arrive in Crete (Nowicki 1999, 579). Research by Nowicki (2008) has shown that settlers moved into the Dodecanese and the offshore islands of Crete in the late FN period for a short time before moving to another location, probably Crete. The pottery of the last FN phase has features that are foreign to Crete, including quartz temper (Nowicki 2008). Day and Wilson (2006) have linked this ceramic tradition with the Cyclades. Sites of the last FN phase are usually defensively located, often on marginal coastal areas. All of this evidence suggests that the immigration during the FN II phase was the cause of the flight to refuge sites such as Katalimata (Nowicki 2008, 72–76). What remains uncertain is the date of other FN sites, such as Kavousi (K24), Kambos near Priniatikos Pyrgos (Kalpaxis et al. 2006, 174–180), and Panagia (3), and their role in the foundation of refuge sites, such as Azoria and Katalimata. Possibly, they predate the arrival of the late immigrants who caused the flight to the hills beginning in FN II. Late FN coastal sites, such as Sphoungaras (10), may therefore represent the last Neolithic arrivals on the island.

20

L. VANCE WATROUS AND MARYANNE SCHULTZ

Final Neolithic Society The earlier FN settlement pattern in the Mirabello region consists primarily of a sparse scatter of hamlets, small farms, and camps. Such a pattern suggests a mobile subsistence strategy composed of a combination of hunting/gathering, small-scale mixed farming, and the keeping of livestock. The size and spacing of these farm sites (e.g., 3, 98, K8, and K16) suggest that they were agriculturally self-sufficient. This economic picture accords with what is known of the Neolithic economy elsewhere on the island. Based on the osteological data (caprines, cattle, and pig) at Neolithic Knossos, Halstead (1980) has concluded that the Neolithic economy on Crete was a combination of cereal cultivation and animal husbandry. Neolithic Nerokourou in West Crete produced spindle whorls and possible loomweights, which point to flocks of sheep and goats (Vagnetti, Christopoulou, and Tzedakis 1989). Late Neolithic levels at Phaistos yielded wild goat, hare, boar, badger, large birds, cattle, caprines, shellfish, and fish. These findings indicate a subsistence based on hunting, gathering, and seasonal pastoralism, supplemented by animal husbandry and spring-fed agriculture (Watrous, Hadzi-Vallianou, and Blitzer 2004, 221–231). To what extent the local economy changed in FN II remains unknown. Our knowledge of FN industries and exchange in the Mirabello region is hindered by lack of information from excavated sites. Obsidian at the FN site 98 demonstrates connections with the island of Melos in the Cyclades. The metalworking (crushing and smelting of imported ores) site at Chrysokamino (K32; Betancourt 2006), may have been established toward the end of the FN period when connections with the Cyclades began to become strong on Crete (see Ch. 4). Pottery production, however, seems to be mostly local; inclusions are either granodiorite from the Gournia–Vrokastro area (Whitelaw et al. 1997), or phyllite from the Kavousi area (Haggis 2005, 47). The “sandy” FN fabric from Sphoungaras noted by K. Nowicki (pers. comm.), however, may have been influenced by the Cycladic ceramic tradition (Day and Wilson 2006). Local exchange in ceramics seems likely, since the Kavousi Survey recorded

finding pottery with granodiorite inclusions typical of the Vrokastro area (Haggis 2005, 47–48), although the type of regular long-distance ceramic trade demonstrated between FN Phaistos and Knossos seems lacking (Wilson and Day 1994). Evidence for local social organization is largely indirect. For the most part, FN I settlements seem heterogeneous and egalitarian rather than hierarchical or heterarchical. Field sites are almost twice as numerous as hamlets; farms equal the number of hamlets, and hamlets themselves probably represent no more than two to six families of perhaps four to six persons each (Hayden 2004a, 47–78; Watrous, Hadzi-Vallianou, and Blitzer 2004, 25). The basic social unit of such settlements is presumed to be the nuclear family. It is possible that the foundation of larger defensive settlements in FN II may have increased social hierarchy, although this remains conjecture. Absence of evidence for social hierarchy within the regional settlement pattern does not completely preclude a developed social heterarchy (McGuire 1983; Ehrenreich, Crumley, and Levy 1995). At the FN village-sized Phaistos, for example, data for vertical inequality is also absent, but social heterarchy seems inferable from the artifactual record (Watrous, Hadzi-Vallianou, and Blitzer 2004, 226). Subsistence there included a wide range of activities: hoe/dibble stick agriculture, gathering, fishing, hunting, animal husbandry, and shepherding. If hunting, fishing, and shepherding can be identified as male activities, then gathering, textile production, shellfish and plant food preparation, and perhaps agriculture, may have been female activities. More similar to our sites, the small FN settlement at Miamou (Taramelli 1897), in the Asterousia Mountains south of Phaistos, kept cattle and sheep/goat, hunted deer and hare, gathered shellfish and marine crab, and practiced a grain-based agriculture. As Brian Hayden (1995) has shown, groups with a similar economic basis survive by means of sharing, interfamily exchange, and cooperation. In addition, local FN domestic pottery used for the preparation and serving of food, likely female roles, is elegantly and carefully produced, an indication of the high status of such roles.

4

Early Minoan I–II Periods: Emergence of a Stratified Society L. Vance Watrous and Maryanne Schultz

The first section of this chapter presents the EM I–II settlement data from the Gournia Survey, followed by a summary of the contemporary information from the Kavousi and Vrokastro projects and, finally, a discussion of the regional pattern of settlement in the Mirabello area. Building on this

framework, the second section of the chapter draws on information from local excavated sites to discuss various aspects of EM II society in the Mirabello region, including subsistence, industry, trade, and social organization.

Gournia Survey During EM I–II, there was a virtual explosion of new settlements established in the area. The number of sites increased from five definite FN habitation sites to a total of 38 in EM I–II. Early Minoan I–II sites consist of 3 villages, 14 hamlets, 9 farms, and 12 field sites, as well as 4 knapping sites and 3 cemeteries (Table 2; Map 9). Within the Gournia survey area, all FN sites continued to be occupied into EM II, except for site 98, which was deserted in EM I. Final Neolithic site 3 expanded in size. The FN hamlet 130 decreased in size, as new settlements (121, 131, 133) were established nearby.

In the southwest corner of the Gournia Valley, the small FN settlement 3 at the Panagia Church expanded to the size of a hamlet. In the EM I–II period, the site consisted of two concentrations of pottery, one (40 x 50 m or 0.2 ha) around the Church of the Panagia (where FN sherds were also found) and a second (80 x 80 m or 0.64 ha), some 30 m to the north, next to a small streambed. About 200 m to the east of 3, the knapping site 6 produced a dense (200 x 200 m or 4 ha) scatter of chipped stone, including red chert and obsidian blades.

22

L. VANCE WATROUS AND MARYANNE SCHULTZ

Almost a kilometer north of site 3, the hamlet 7 is situated along the edge of a high cliff on the coast. During EM II, the site measured 60 x 40 m or 0.24 ha, the size of a small hamlet. Ancient walls are visible, including one discrete rectangular structure that measures 16 x 12 m. A large quantity of chipped obsidian and chert found at the site suggests that it was involved in craft production and trade rather than being exclusively agricultural. Located on a steep slope at the western edge of the Palaiobarda Bay, the knapping site 4 yielded 2 pieces of obsidian and 21 pieces of chert. Both sites 4 and 7 are a new type of EM II site, characterized by specialized activities linked to industry and trade. The ridge top site (Boyd’s excavated “Acropolis”) of Gournia (8), about 500 m from the coast, was inhabited in this period (Map 10). The Gournia River provided a perennial source of water, and the inhabitants probably took advantage of the excellent arable land in the valley located to the west. Pera Alatzomouri, a high ridge to the northeast, shielded the settlement from harsh winter winds (Pl. 2C). Boyd excavated the “acropolis” ridge of Gournia in 1901, 1903, and 1904 (Hawes et al. 1908). Although the exact dimensions of the EM I–II site remain unknown, the presence of Early Minoan sherds found below the Late Minoan houses and the cemeteries on the east and west sides of the northern edge of the ridge indicate that the settlement was larger than three or four houses, and probably should be considered a village in the EM II period. Tomb III at Gournia was built in EM II, accompanied by rock shelters V and VI. Early Minoan II artifacts were recovered from Tomb I (in a pit) and under room Dh 59 in the town (Soles 1992, 1–34). A settlement (10) was located on the ridge of Pera Alatzomouri (Map 10; Pl. 2C). Early Minoan II sherds were found on the hilltop and on its southwest and east slopes as well as on the southern hilltop of Pera Alatzomouri. Immediately below site 10, the cemetery at Sphoungaras was used for EM IIA inhumations in at least two rock shelters and in Hall’s “Deposits A and B” (Hall 1912). Hall’s map (1912, fig. 4; Map 11) of Sphoungaras shows the walls of what are probably two partly preserved house tombs. She also recorded EM I vases (e.g., a jug with dots punctated around its neck), as well as Vasiliki-style (EM IIB) egg cups, plates, jugs, bowl, mugs, goblets (Hall 1912, figs. 20, 22; Cooper

2002, figs. 8–11, 17:b, 18:b), clay rellas (polishers) used by potters (Hall 1912, fig. 27), gold jewelry, bronze tweezers, and stone “rings” similar to examples described by Warren (1972, pl. 79:d). About 700 m to the southeast of 10, an EM IIB hamlet (14) was founded on the shoulder of the ridge (Phylakeion) west of Pacheia Ammos. At the eastern edge of the Pacheia Ammos Bay, a site (23) was established at Halepa, on a small promontory near the newly constructed Church of Hagia Eirene. Current excavations by the Greek Archaeological Service have revealed EM walls in the north-facing scarp of the site. The site extends 200 m from the coastline, giving it an area of 150 x 200 m (or 3 ha), making it one of the largest EM I communities in the Isthmus. Large amounts of obsidian found on the site point to Cycladic contacts. The presence on the site of calcite/marbletempered pottery, a Cycladic ceramic tradition, suggests that the site may have been a Cycladic foundation (Day and Wilson 2006). Approximately 40 m directly south of Halepa and across the asphalt Kavousi road, the hamlet-sized site of Kale (22) was probably an extension of site 23. This settlement at Halepa was abandoned by EM IIA. South of Pacheia Ammos, a string of five sites (two hamlets [26, 28] and three field sites [29, 30, 31]) was established along the river between Vasiliki and the coast. Site 26 consists of three dense concentrations of pottery uncovered by plowing on the eastern slope of the river. One of the sherd clusters (30 x 70 m or 0.21 ha) dates predominantly to EM I–II. A smaller cluster of EM II sherds was also found in a higher eastern sector of the site, which could indicate that the settlement had grown during the EM II period. About 100 m south of 26, a hamlet (28) sits on top of and along the slopes of a ridge on the east side of the river. Its catchment would have consisted of the river valley below for cultivation and the red soils to the east for grazing. About 250 m south of 28, a small field site (29) could have used the river valley below and the marl slope to the west for cultivation. Some 300 m farther south, site 30 consists of two discrete sherd concentrations on the north and south slopes of a ridge overlooking the river. Another 400 m to the south, field site 31 is positioned on a gentle marl slope above the west bank of the river. Obsidian and pieces of pumice, the size of lemons, found on the site suggest some kind of specialized activity.

EARLY MINOAN I–II PERIODS: EMERGENCE OF A STRATIFIED SOCIETY

Immediately north of the modern village of Vasiliki, a hamlet-sized site (33) is based on a ridge above the river. Early Minoan I–II pottery is scattered over an area of 90 x 40 m (0.36 ha) to the north and east of the Church of Hagios Ioannis. Immediately west of the site (Pl. 4B), a large outcrop of pure clay (lepidochoma), wasters (undated), and a Venetian–Ottoman-period kiln on the site associate it with pottery production (see below). A half-kilometer south of Vasiliki village is the archaeological site of Vasiliki (135), excavated by Seager (1905, 1907, 1908) and reexamined by Zois (1992 for summary). Founded in EM IIA, Vasiliki consisted of a minimum of four houses, three of which (Theta, Xsi, and P) yielded remnants of tool production, including knapping debris, ground stone tools, obsidian, and a bronze tweezer. Following a destruction at the end of EM IIA, Vasiliki was reorganized, and a new two-story structure (Zois’s “Red House”) was built (Map 12). The Red House was an impressive structure possessing massive exterior walls, a new paved courtyard to the north and west, and an 8-m-deep well hewn into the bedrock. The presence of two courts and two basement-level southern magazines, and the space between rooms 13 and 39, suggest that the Red House complex was actually two houses (Watrous 2001b, 171 and n. 77). Seager found two broken bronze axes, a knife, much obsidian, amphorae and loomweights, and many pithoi (south of room 43) in the Red House. Zois’s excavations in the Red House produced a steatite seal (Pini, ed., 1975, 24 [CMS I, no. 27]) and a large vase stamped with an EM II seal (Zois 1979, 326). Later in EM IIB, Vasiliki was reorganized once again. Two new houses (Zois’s West and Southwest Houses) were built on top of the paved west court as annexes to the Red House. The West House consisted of a large central room, around which were storage annexes and other rooms. The Southwest House produced fragments of pithoi and 70 loomweights. Across the floor of the Isthmus to the east, a site (67) was settled on a hill occupied today by the Church of the Holy Spirit near the mouth of Cha Gorge (Pl. 3C). The location provided ample yearround water and access to arable land for cultivation. Approaching the size of a farm, the site consists of an EM II sherd scatter measuring 30 x 30 m (0.09 ha) around the church. A field site (65), possibly an offshoot of 67, may have been established on the floor of the Isthmus valley as early as EM II. South

23

of the Cha Gorge, another small site (70) was located on the slope of Thriphti, near a streambed. Worked chert debris from 70 points to the site being used seasonally for knapping. South of Monastiraki village, a similar field site (84) sits on the slope of Thriphti near a spring. Lack of nearby arable land suggests the site had a pastoralist function. To the south, two small sites (90 and 98) were clustered next to the hamlet 97. Located on a shoulder (elevation 221 m) of Thriphti below a spring, site 90 consists of two clusters of EM I–II sherds above the Church and spring of Hagios Georgios. About 200 m south of site 90, near a stretch of arable land, field site 98 produced over 100 pieces of chipped stone, obsidian, and chert. The largest site in the area, hamlet 97 has access to a 400 x 200 m (8 ha) stretch of flat, arable land on the slope of Thriphti and a spring some 300 m to the north. The EM I–II occupation of the settlement is represented by a concentration of sherds over 60 x 80 m or 0.48 ha. About 600 m south of site 90, the farm-size site 109, located near a stream and limited arable land on the slope of Thriphti, consists of three small concentrations of EM I–II pottery, which cover an area of 15 x 40 m or 0.16 ha. Proceeding southward, the next large settlement is site 114. Located on a conical hill that overlooks the valley and stream south of Kazarma hill, site 114 measures 80 x 90 m or 0.72 ha. In addition to pottery, this agricultural hamlet produced an obsidian blade, a stone chopper, and a quern. To the west of site 114, on the floor of the Isthmus, two smaller sites—105 and 106—were founded near arable land and a stream. Site 105 yielded EM I pottery, obsidian, and a stone celt. Nearby site 106 had much EM IIA pottery and chipped stone. During this period, several new sites were scattered on the upper slopes of Thriphti above site 114. Near the Church of Ephendis Christos, the farm 117 (elevation 240 m) is on a flat spur of arable land next to a spring. Another farm, 121, had access to two areas of arable land, 150 m to the west and 70 m below, and water provided by a stream to the north. Higher still, at Selimas, the farm 118 (elevation 380 m) is situated on a small shoulder overlooking six springs. Nearby, the field site 119 produced over 40 pieces of chipped (chert) stone. Farther south, a cluster of four new sites (131, 132, 133, 136) was settled near the FN–EM I cave site 130, on the slopes of Thriphti near the

24

L. VANCE WATROUS AND MARYANNE SCHULTZ

Kephalovrysi spring (the largest spring in the Isthmus). The hamlet 131 is naturally protected on three sides by cliffs and has one point of access along a gorge that flanks the south edge of the site. Site 132 is also defensively positioned in a small saddle behind a steep hill (Pl. 4A), hidden from view from below. The farm site 136 is located on top of a steep ridge that projects outward at the base of Thriphti. Hamlet 133 is placed at the base of Thriphti on a gentle colluvial marl slope, 120 m from a riverbed that would have provided water. In the middle of the Episkopi Valley, site 123 sits on a low hill, near arable land and a stream. The site was measured as 0.26 ha, but it may have been larger, since the modern road to Ierapetra has destroyed part of it. Four of the sites in this area (131, 132, 136, 144), at the entrance to the large southern Ierapetra Plain, were situated on defensive locations. Four more sites, 142, 137, 138, and 144, remain at the southernmost limits of the Gournia survey area. East of the village of Kato Chorio, the farm 142 is located on a level slope of Thriphti above good arable land near a ravine. Possibly already settled in the FN period, the field site 137 is located by a stream and good land. Knapping site 138, possibly associated with 137, yielded a large (90 x 300 m) concentration of chipped stone (286 pieces collected). On the slope southeast of Kato Chorio, the hamlet 144 sits on top of a prominent ridge that overlooks the Episkopi Valley. Surrounded on all sides by steep slopes, the site is naturally protected, except for the narrow ridge top that forms a natural entrance to the site on the southwest. A stream immediately below the site would have supplied water. Settlement in the survey zone exhibited six major changes in the EM I–II periods. First, there is the rise in the number of settlements, from a total of 5 definite FN habitation sites to 38 EM I–II sites. It is likely that this phenomenon began at the

end of the FN period, as the excavated sites of Priniatikos Pyrgos and Sphoungaras produced FN pottery, but this expansion actually accelerates in the EM I–IIA period. In the Mirabello region, by EM II there are at least 14 more settlements than in EM I (an additional 4 hamlets, 3 farms, and 7 field sites). There is a tremendous expansion of sites, especially the smaller farm- and hamlet-sized settlements, near good agricultural land and water. Second, EM I–II settlers sought out new types of locations: on the coast (7, 8, 10, 22, 23), on the valley bottom of the Isthmus (26, 28, 29, 30, 31), and in higher mountainous areas (118, 119, 121, 131). Some sites are situated more than 200 m from water (70, 105, 106, 131, 142, 144); others are some distance from arable land (23, 70). Third, the overall size of local settlements increased. While the largest FN site (130) had been of a hamlet size, in EM I–II, 3 villages and 14 hamlet-sized settlements existed. An overall increase in all site categories implies population growth. Fourth, settlement hierarchy changed. In the FN period, sites were commonly quite small, but in EM I–II, hamlets outnumbered farms and field sites. Growth seems concentrated at the top of the settlement hierarchy, that is, the largest increases took place in hamlets and in villages. Fifth, some EM I–II sites occurred in discrete clusters, that is, around a larger site (e.g., 7, 10, 14 around Gournia [8]; 84, 90, 98, 99 around hamlet 97). The fact that Halepa (23), which had an excellent location with fresh water and a small harbor, was abandoned in EM IIA, is significant. The result was that Gournia consolidated its position as the sole harbor settlement on the north coast of the Isthmus. Sixth, some EM sites (4, 7, 23) may have had specialized functions, as they produced large quantities of imported obsidian and are some distance from good land, a water source, or both.

Kavousi Survey The Kavousi Survey Project (Haggis 2005) identified 10 EM I–II sites in the survey area: 2 villagesized (one a harbor); a hamlet, 2 farms, 3 field sites, 1 metalworking site, and burials in a rock shelter. There were three FN sites (K16, K67, K71), two of

which (K67 and K71) were abandoned in this period. Newly founded sites included a large coastal site (K1) at Tholos and two field sites (K7, K62). Haggis (2005, 62) stated that during EM I–II, the total number of sites in the survey area did not appreciably

EARLY MINOAN I–II PERIODS: EMERGENCE OF A STRATIFIED SOCIETY

increase, although the size of some settlements did become larger. Nevertheless, if the Kavousi FN sites are regarded (see Ch. 3) as belonging to one of two separate chronological phases, dating to early FN (K24, K34) or to later FN (K31, K32, K67, K71, K77), the number of EM I–II sites in the Kavousi area did increase significantly—a general pattern observed in the Gournia and Vrokastro areas and throughout Crete. New areas were exploited in EM I–II, particularly locations suitable for rain-fed agriculture and near a spring (Haggis 2005, 63–64). The coastal site of Tholos (K1; see Map 1), established during EM I, was an active port by EM II; the site itself produced an obsidian core and blade, and the imported Cycladic jug, stone vases, and obsidian found in the nearby Hagios Antonios cave (K6) probably came to Crete via this site. The

25

Chrysokamino settlement (K34; see Map 1) and an associated burial cave site (K31), and the metalworking site (K32), continued through EM I–II. To the south, the hamlet-sized settlement of Alykomouri (K8) continued to use the Hagios Antonios cave (K6) for burials. A field site (K7) adjacent to the settlement appeared in this period. Located beneath the present-day Kavousi village, site K24 was a major settlement, probably the size of a village. In the mountains south of Kavousi, the large FN site of Azoria (K71) was not occupied during the EM I–II period. Nevertheless, there was an expansion of settlement elsewhere in this mountainous area—namely, K62. These sites and Vronda (K77) may have been used seasonally by the settlers at Kavousi (K24).

Vrokastro Survey In the reports of the Vrokastro Survey (Hayden 2004a), FN, EM I, and EM IIA sites are considered together in Chapter 2, and EM II and EM III sites together in Chapter 3. As a result, we have turned to the Vrokastro site catalog to draw up a list of EM II sites that can be compared with the Gournia and Kavousi surveys. Since the way in which the dates of EM sites were designated varies within the Vrokastro survey catalog, we consider only sites that have been explicitly dated to the EM I–II period, which equals only 31 sites. This means that we have almost certainly underestimated the number of EM I–II sites in the Vrokastro area. Nevertheless, for our purposes, which are to establish a comparable database for the EM I–II Mirabello region for examining general trends in the settlement pattern, and to produce comparable lists of Vrokastro EM I–II and EM III–MM IA sites, this list should be sufficient. Beginning in the EM I period and continuing into EM II, a string of new settlements was established along the coast of the Vrokastro area. Many of the EM I sites established on the coast and inland in the survey area were situated on defensive locations (Hayden 2005, 35–53). Early Minoan I sites such as GN4, KP10, PhR1, and PN1 were founded in EM I on inhospitable locations at the top of steep hills, peaks, or summits.

Other sites (IM1, IM2, IM6, IM7, IM8, IM9, IM11) were settled (perhaps later?) on lower land on or near the coast. To the east, the large promontory of Vrionisi had two coastal EM I settlements, VN1 and VN2, which, like Halepa, was discontinued by EM IIB. Five more sites (APh1, APh7, APh9, APh11, PhR1) were located on or near the coast to the east. Site PhR1, an EM I–II hamlet, is based on a hill near a small beach. Just 200 m east of PhR1, in the Gournia survey area, is the knapping site 4, and a little farther east, site 7. Priniatikos Pyrgos (PP1; see Map 1) continued to be the largest harbor site along the coast in the Vrokastro region during EM I–EM II. Several sites (PP2, IS2, GN2A1 and 2, IM1) were founded nearby, around the edge of the coastal plain dominated by PP1. This pattern of settlement, a center surrounded by small agricultural sites, may reflect some form of economic and social hierarchy. Barbara Hayden (2004a, 72) speaks of Priniatikos Pyrgos as a “primary center” for her region in this period. The Vrokastro area experienced a tremendous rise in the number of settlements during the EM I–II period (Hayden 2004a, 61–80). Settlement in EM I–II also expanded into the valleys and mountains of the hinterland (Hayden 2004a, 70, 75). Settlements also apparently became larger, which

26

L. VANCE WATROUS AND MARYANNE SCHULTZ

Hayden (2004a, 73) attributes to population growth (Hayden 2004a, 73–76) made possible by an intensification of local land cultivation for grains, olives, legumes, and grapes, as well as herding on the floor of the Istron Valley and the mountain slopes. Several coastal sites (PP1, NP1,

VN1) probably functioned as harbors in this period. Hayden (2004a, 70) also points out that some of the new Early Minoan settlements were situated in marginal environments, on poorer soils distant from a water source, a trend also noted in the Gournia survey area.

Early Minoan I–II Settlement in the Mirabello Region Turning to the Mirabello–Isthmus region as a whole, we see that the number of settlements rose from a total of 31 settlements (8 hamlets, 8 farms, and 15 field sites), 3 site-specific work areas (1 metalworking site and 2 knapping sites), and 4 burials in the FN period, to 92 settlements (10 villages, 27 hamlets, 20 farms, and 35 field sites), 6 work areas (1 metalworking site and 5 knapping sites), and 6 burials by late EM II. Pseira and Mochlos were two important settlements along the northern edge of the survey areas that were settled in EM I (Betancourt et al. 1999; Betancourt 2003). Sites founded in EM I often favor defensive locations, on steep hills or on high upland areas. South of Episkopi, the prominent hilltop site of Aphroditi’s Kephala (Betancourt 2008) is a good example: excavations there have revealed what has been interpreted as an EM I fort. By EM IIA, some of these sites are abandoned for more hospitable locations, for example, the ridge top site of Vasiliki Kephala is deserted, probably for Vasiliki below. The sharp rise in EM I settlements throughout the Mirabello region has been explained, at least in part, by immigration, coming from the direction of the Cyclades (Betancourt 1999; Nowicki 1999; Hayden 2003a). Excavation at the cemetery established at Hagia Photia in EM I near Siteia has produced strong evidence—in its tomb types, pottery, lithics, and metallurgy—that the settlers were from the Cyclades. Most recently, Day and Wilson (2006) have pointed out that the new settlements on the north coast of Crete in EM I use Cycladic ceramic technology, and have suggested that these new sites were established by Cycladic immigrants. By the EM II period, settlement in all three survey zones had begun to expand into new areas,

especially along the coast, inland, and on the valley floors. A number of settlements almost reach the size of villages, including Priniatikos Pyrgos (PP1), Kato Arniko (KA1), Gournia (8), Alatzomouri (10), Halepa (23), Vasiliki (135), Kavousi (K24), and Mochlos. Many sites were situated to take advantage of nearby water and arable land for agricultural purposes. Upland areas near a perennial water source (SK9, K77) were also settled. In the Vrokastro area, the site hierarchy changed in ways similar to the Gournia zone. By EM II, field sites more than doubled in number (9:20) while farms tripled (3:10); hamlets more than doubled (6:13) and new village sites (three) appeared. As in the Gournia area, the increase takes place at the top of the settlement hierarchy. Some large coastal settlements—such as Priniatikos Pyrgos, GourniaAlatzomouri, and perhaps Mochlos—may have been supported by small outlying agricultural sites. Overall, the changes in the Early Minoan settlement pattern strongly suggest population growth in the Mirabello region. Taking into consideration the length of the EM I–II period and the sites sizes, Hayden (2004a, 73) estimates a doubling of the population since the FN period for the Vrokastro zone. The same seems true for the Gournia survey area. By the EM II period, the earlier FN–EM I immigrants seem to have been assimilated into the regional landscape; the dispersed and evenly spaced EM II pattern of settlement is largely keyed to locations of prime agriculture potential, that is, near water and arable land. Such a pattern points to a relatively independent agricultural population, without centralized political control.

EARLY MINOAN I–II PERIODS: EMERGENCE OF A STRATIFIED SOCIETY

27

Early Minoan II Society in the Mirabello Region Subsistence Studies of data from Greek excavations (Hansen 1988; Pullen 1992) have shown that Early Bronze Age agriculture was marked by the introduction of new crops, such as legumes, and the increased importance of plowed cereals, and probably olives and vines. The plow made it possible to dry-farm crops in marginal areas without irrigation. Livestock were no longer just sources of meat and wool, but their milk was converted to long-lived cheeses that could be stored and traded (Sherratt 1981). One may be able to see the initial results of this new, intensified agriculture in the surprisingly extensive halos of off-site Prepalatial sherds distributed around EM I–II sites (Map 13). Fine-ware sherds distributed around Gournia (Map 14) might be explained by manuring, but other pottery (e.g., coarse ware and cookwares) around Gournia and in the Isthmus probably represent agricultural use of the land. The intensity of land use around Gournia is especially striking. Off-site chipped stone (Map 15) follows off-site pottery distribution to an extent; chipped stone also forms halos around Prepalatial sites but is a bit more dispersed, especially on the floor of the Isthmus, perhaps associated with grazing. Off-site stone types (Map 16) show three chert clusters: around Gournia, west of site 97, and around Kato Chorio. The latter two clusters probably represent local chert sources. Obsidian is, with a few exceptions, concentrated along the coast. Several aspects of the EM I–II pattern of settlement in the Mirabello region can be explained in terms of this new subsistence strategy. The most detailed source of information comes from the excavated EM II hamlet of Phournou Koriphi on the south coast near Ierapetra (Whitelaw et al. 1997). Like some of the Mirabello EM I–II sites (e.g., 26, 28, 29, PP2, IS2), the site is located on a coastal hill of marly slopes next to a small river valley. The community subsisted on barley, wheat, olives, vines, and livestock, primarily sheep and goats, but also pigs and oxen. Some 90% of the

bones from the site are sheep and goat (Manning 1994, 232, table 8.3). Wood charcoal from the site consisted almost entirely of olive, pointing to the importance of olive cultivation. In this respect, EM IIB Phournou Koriphi may have looked like the traditional village of Vasiliki today, which is surrounded by a halo of ancient olive trees. Recent analysis of pottery from the EM II site indicates that many of the jars were used to store wine (Hayden 2004a, 79 n. 90). Phournou Koriphi seems to have been agriculturally self-sufficient and, more significantly, was able to produce surpluses that made trade in a new variety of goods possible. Early Minoan sites in the Mirabello region seem to be an outgrowth of this new polycrop agriculture, which made the settlement of small permanent farms and hamlets feasible and economically worthwhile. Permanent farms and hamlets were settled across the landscape in this period, reflecting a more intensive use of the land. Most of the Gournia survey EM I–II sites are located near a water source and marl land that is good for the dry-farming of grains. The large numbers of ground stone tools, used for the processing of cereals and other agricultural products, in addition to large storage vases at these EM I–II sites, point in the same direction. Settlement in upland areas, some probably seasonal, is a sign of new interest in animal husbandry and shepherding. The production of an agricultural surplus in EM I–II made possible a greater diversification and specialization in craft and trade activities, hence the appearance of sites (e.g., 4, 7, 23, 33, and the metallurgical installation at Chrysokamino) that seem to specialize in specific types of crafts such as pottery, stone vase manufacture, and metallurgy.

Trade and Industry During the EM II period, the Mirabello region had close connections with the Cyclades, Attica, and indirectly with the Near East (Davis 1992). At Mochlos, a Syrian cylinder seal (Aruz 1984), two

28

L. VANCE WATROUS AND MARYANNE SCHULTZ

Sixth Dynasty Egyptian stone vases (A. Bevan, pers. comm.), and tin alloyed in three daggers (Branigan 1991, 100) attest to some form of international connections with the Near East (Watrous, HadziVallianou, and Blitzer 2004, 243). The well-known stone pyxis from Mochlos with a dog depicted on its lid probably indicates that similar pyxides found at Zakros and Hagia Triada (Warren 1969, 82) were trade items from Mochlos. An unfinished stone bowl from Tomb I at Mochlos is evidence of local production (Seager 1912, 20). Stone vases found at Phournou Koriphi, Pseira, Sphoungaras, and Vrokastro have also been identified as probable trade goods from Mochlos (Branigan 1991). Several of the silver/lead items from Mochlos tombs have been shown to be imported from the Cycladic island of Siphnos (Stos-Gale 1985; Stos-Gale and Macdonald 1991), although copper objects are from Attic ore (Branigan 1991, 100). Early Minoan I sites in the Mirabello region, including the excavated site of Kalo Chorio, produce large amounts of obsidian (Haggis 1996a, 662; Hayden 2003b, 386). The Hagios Antonios burial cave (K6) produced imported metal, stone vases (perhaps made in Mochlos or Pseira), and a Cycladic juglet (Haggis 2005, 62). The Kavousi village site K24 yielded obsidian imported from Melos. Haggis (2005, 63) mentions the presence of Hagios Onouphrios ware from the Mesara. In the Vrokastro area, there was a production center, perhaps centered at Priniatikos Pyrgos (PP1), for large vases (Mirabello Ware), which have been found in the Gournia and Kavousi areas, and at Phournou Koriphi on the south coast (Hayden 2004a, 74 and n. 96). Melian obsidian was imported and widely available in the Gournia survey area: 14 Early Minoan sites (3, 4, 6, 7, 23, 31, 98, 105, 114, 117, 121, 135, 136, 144) produced obsidian. Mirabello wares were widely traded, including along the south coast, indicating a regional-scale economic network (Whitelaw et al. 1997, 267). The Mirabello area exhibited a sharp rise in craft production and trade in the EM I–EM II periods. Stone vases were certainly manufactured at Mochlos and Pseira. Copper-bearing ore was imported from the Cycladic island of Kythnos, and crushed and smelted on-site at Chrysokamino from where copper was traded locally (Betancourt 2006). The three sites associated with this operation—the workshop, habitation site, and burial cave—were located together but isolated from other local sites,

suggesting that the operation was an independent one, run perhaps by Cycladic islanders. In EM IIB, the major production source of high-quality, redpolished Vasiliki Ware was near Vasiliki, perhaps at or near site 33 with its extensive clay beds (Pl. 4B; Day 1997). Loomweights found at Vasiliki are evidence of weaving of textiles. A bronze ax mould from Vasiliki (Mavriyannaki 1984) indicates tools were also being made there. The tombs from Mochlos have provided a wealth of imported material: a bronze dagger, an electrum necklace, amethyst (from the Peloponnese?), beads, gold jewelry (Davaras 1975), lead, and a Syrian cylinder seal. Lead objects from Mochlos have a composition consistent with origins from Siphnos and Lavrion (Stos-Gale and Macdonald 1991). Gold may also have come from Siphnian mines. On the south coast, the well-studied finds from the hamlet of Phournou Koriphi show how extensive EM II trade was in our area. Over half the pottery— that is, Vasiliki and Mirabello Ware—from this small site was imported from the Mirabello region. Vasiliki Ware was also exported to the Mesara and North-Central Crete (Knossos). Cycladic trade was important for the Mirabello region. The workers who crushed and smelted the ores at Chrysokamino and probably lived at site K34 may have been from the Cyclades. The many metal artifacts—a Cycladic jug and stone vases from the Hagios Antonios cave (K6)—may have belonged to these workers. Moreover, the large amounts of obsidian at village-sized Halepa (23) indicate that those Cycladic boats stopped there as well as at Mochlos, Pseira, Gournia (8), Tholos (K1), and Priniatikos Pyrgos (PP1). The EM I–II artifacts from the settlements at Mochlos, Halepa (23), Gournia/Pera Alatzomouri, and Priniatikos Pyrgos show signs of craft specialization. The Mochlos tombs are quite wealthy relative to other known cemeteries, such as Theriospilio (K31), Sphoungaras (10), and Gournia (8). At Mochlos, the concentration of large amounts of imported materials suggests that by EM II some of the inhabitants had become seafaring traders rather than relying on Cycladic mariners. As EM II pottery is virtually unknown in the Aegean islands and Early Helladic (EH) II mainland, the settlers from other sites may have bartered local agricultural produce (e.g., olives/oil, grains, wine, legumes, and wool) for goods from Cycladic traders.

EARLY MINOAN I–II PERIODS: EMERGENCE OF A STRATIFIED SOCIETY

All in all, local EM II artifacts suggest that economic power in the region was concentrated in the hands of groups resident at Mochlos, Vasiliki, and Priniatikos Pyrgos. In this respect, the sites seem to play complementary roles: Mochlos had maritime access to overseas metals and Priniatikos Pyrgos made coarse-ware vases, while inland Vasiliki seems to have been the production center for the export of fine pottery.

Ideology Evidence for ideology is limited to cult activities in domestic and funerary contexts. Kernoi,

29

stone slabs carved with small cupule-like depressions, have been found outside houses at Vasiliki and at Phournou Koriphi, and at Trypetos (inside a house) on the south coast near Lebena. Kernoi have been connected with cult, as offering platforms, and also as game boards (Whittaker 2002). A house at Phournou Koriphi produced a small domestic shrine consisting of a bench on which a statuette (“the Goddess of Myrtos”) of a female divinity(?), holding a jug, had been placed. At Hagia Photia, the graves in the cemetery had small platforms outside their doors, used for offerings placed in vases, suggestive of ancestor worship (Davaras and Betancourt 2004).

Society Social Organization Identifying the type of social organization in the Mirabello region during the EM II period is complicated, partly because social complexity can take different forms. Our approach here is based on the discussion by Watrous, Hadzi-Vallianou, and Blitzer (2004, 20–22, 244–250). Social complexity can be heterogeneous or hierarchical (McGuire 1983). Social hierarchy consists of the vertical inequality in relation to differential access to material and social resources in a society. On the other hand, social heterogeneity is a matter of horizontal diversity, that is, the degree to which groups, based on kin, gender, occupation, and institution, are distributed within a society. Ethnographically based studies have focused on one particular form, the chiefdom, which is defined as an autonomous and hierarchical regional polity comprising a number of communities organized along kinship lines under the two-tiered control of a paramount leader (Flannery 1972; Carneiro 1981; Earle 1991; see Tainter 1988, 26–31 for earlier definitions of social complexity). A society becomes ranked at the onset of inherited wealth. A stratified society is one in which members have privileged access to limited natural resources through organizational, geographical, or physical means, or by virtue of their specialized status or occupation. Once stratified, the elite members of a society develop an

institution to oversee the unequal distribution of wealth. This is the state, defined here as a stratified society with at least three elements of administrative hierarchy and a permanent, centralized, and institutionalized political authority that has coercive power over its subjects. For our purposes here, more general categories of social hierarchy—egalitarian, ranked, stratified, and state—will be used (Fried 1967, summarized in Watrous, Hadzi-Vallianou, and Blitzer 2004, 20–22, 244–250). A number of scholars (Flannery 1972; Peebles and Kus 1977; Wright 1977, 1984; Earle 1991; Stein 1999) have compiled a list of archaeological correlates as evidence for a level of hierarchical complexity. These correlates include: 1. Agricultural intensification 2. Multilevel settlement hierarchy 3. Differentiation in grave goods 4. Monumental public architecture 5. Concentrations of precious materials at regional centers 6. Long-distance trade in precious materials 7. Attached craft specialization 8. Centralized storage and redistribution 9. Complex centralized administration

30

L. VANCE WATROUS AND MARYANNE SCHULTZ

Let us consider each of these correlates within the Mirabello region.

Intensified Agriculture Signs of intensified agriculture are, by and large, missing in the EM I–II period. There does not seem to be the density or hierarchy of settlement that would suggest the raising of a sizable agricultural surplus for the purpose of paying taxes or tribute to a regional head. In EM IIB, however, the “Red House” at Vasiliki did contain “many” pithoi found south of room 43 by Seager (1905, 1907). The Vasiliki storage facility may, therefore, have been used for some form of redistribution, possibly linked to the production of Vasiliki Ware.

Hierarchy The regional settlement pattern exhibits hierarchy, in the form of field sites, farms, hamlets, and villages. Small villages, at Mochlos, Halepa (23), Vasiliki, Gournia, site 97, and Priniatikos Pyrgos, range in size between 1 and 3 ha. Using the population estimate of 30–50 persons per ha of settlement space (Watrous, Hadzi-Vallianou, and Blitzer 2004, 24–26), no EM Mirabello settlement would appear to have a population of more than 90–150 people. On the basis of ethnographic and historical cases, a population of about 500 has been suggested as the threshold for the necessity of authoritative officials (Naroll 1962; Kosse 1990). In a random sample of 45 non-industrial societies, Lekson (1984) found a boundary between simple and politically complex societies at a settlement size of 2,000–2,500 persons. Our largest settlements, therefore, do not necessarily suggest complex social hierarchy. While our village-sized settlements are distributed at even distances of over 5 km from each other, only site 97, Gournia (8), and Priniatikos Pyrgos (PP1) have a cluster of lower-level, second-order hamlets and farms surrounding them. In discussing the role of Priniatikos Pyrgos, Hayden (2004a, 76) suggests that the site, like Mochlos, may have had social classes based on wealth (social ranking). Early Minoan II village centers, such as Priniatikos Pyrgos, Halepa, Vasiliki, and Mochlos, procured foreign materials, such as obsidian, copper, and gold, and manufactured items such as tools,

weapons, jewelry, and stone vases, which would have been purchased by smaller regional settlements in exchange for agricultural produce.

Differentiation of Grave Goods The EM II cemetery at Mochlos exhibits a differentiation in the grave goods deposited there. As Soles (1988) has pointed out, two tombs (I/II and IV/VI) on the west terrace of the cemetery have unusually large amounts of precious/exotic goods, such as gold and silver jewelry, as compared with other graves on the southern slope of the cemetery. The west terrace tombs (I/II and IV/V/VI) contained many and diverse types of gold jewelry, as well as large-sized stone vases, some metal objects (e.g., a copper dagger), imported lead and silver, and a Near Eastern seal. The smaller tombs on the south slope were limited, for the most part, to clay and stone vases (some small, some large), although a few examples of gold jewelry (Seager 1912, nos. XVI:13, broad armlet; XIX:13, gold bands) did occur. Interpreted literally, the grave goods suggest that EM II Mochlos consisted of 4 elite families (Tombs I/II/III, IV/V/VI, XIX, XXI), and at least 15 non-elite families (Tombs VII–XI, XII, XV–XXIII). If the rich personal possessions in these tombs came by virtue of their occupation (as maritime traders), which seems probable, they would be evidence of social stratification (though not chieftain status). Finds, such as gold and silver jewelry, bronze tweezers, and stone vases from the EM II cemeteries at Gournia and Sphoungaras indicate that wealthy families were not unique to Mochlos in this period.

Monumental Architecture, Precious Materials, and Trade The monumental scale of the Mesara tholos tombs is unknown in the EM II Mirabello region. A possible exception is the “Red House” at Vasiliki. Probably two houses rather than one (Watrous 2001b, 171), the large “Red House” possesses thick walls, two stories, red painted interior walls, and an exterior courtyard, which stand out in comparison with other EM II houses, such as those known at Phournou Koriphi. Little concentration of precious materials at regional centers is known beyond the grave goods mentioned above, largely because

EARLY MINOAN I–II PERIODS: EMERGENCE OF A STRATIFIED SOCIETY

Vasiliki is the only local EM II village-sized settlement that has been excavated. Bronze tools were found in the “Red House,” however. Long-distance trade in precious materials is evident at the Mochlos cemetery, in the jewelry of gold and silver imported from the Cyclades or farther north. Tomb I at Mochlos produced an Early Bronze Age Syrian cylinder seal (Aruz 1984).

Attached Craft Specialization A case for this phenomenon could be made if we knew something about the craftsmen who made the stone vases, the silver cup from Mochlos, and the high-quality Vasiliki Ware. One might argue that the many pithoi from the Red House were used to provision attached craftsmen, although this remains a hypothesis.

Discussion Overall, the archaeological evidence points to a stratified EM IIB society in the Mirabello region. A multileveled settlement hierarchy, differentiation in grave goods, and long-distance trade in precious materials certainly existed by EM II. The evidence for centralized storage/redistribution and administration is ambiguous. Both at Vasiliki and at Mochlos, the community seems to have been headed by two separate family groups who

31

are distinguishable by their residences, privileged access to precious materials, weapons (bronze daggers), international connections, and craft production. The gold and silver jewelry found in Tomb I at Gournia and at Sphoungaras might be a sign of similar families there. Other such families may have existed at Kavousi (K24) and Priniatikos Pyrgos (PP1). The degree to which, and the means whereby these family groups were able to exert economic and political power over their communities, remain unknown. In EM IIB, economic and political power seems to have been consolidated at three centers in the region: Mochlos, Vasiliki, and Priniatikos Pyrgos. This balance of power did not last long, however, for Vasiliki, Mochlos (Soles 2009), Phournou Koriphi, and Myrtos/Pyrgos suffered destructions by fire in EM IIB. Who was responsible for these destructions? Two scenarios seem possible. The first would involve intraregional conflict. If we imagine the region as organized around centers at Mochlos and Vasiliki (and perhaps Priniatikos Pyrgos), then conflict may have arisen among rival settlements. The second scenario would involve outside intruders, perhaps from other parts of the Aegean. Malia also suffered a destruction by fire in EM IIB and Palaikastro may have been deserted for nearby Kastri at this time (Watrous 2001b, 179 n. 170). The coastal sites of Phournou Koriphi and Myrtos Pyrgos would fit such a scenario.

5

Early Minoan III–Middle Minoan IA Periods: Disruption and Social Reorganization L. Vance Watrous and Maryanne Schultz

This chapter begins by presenting the EM III–MM IA settlement data from the Gournia Survey. It then summarizes contemporary data from the Kavousi and Vrokastro Surveys to form a basis for discussing the regional pattern of settlement in the Mirabello area. As the EM III–MM IA ceramic phasing remains controversial, the sequence of developments at local excavated sites (e.g., Vasiliki, Myrtos/Pyrgos,

Gournia, Mochlos, and Pseira) during this period are described in order to provide an additional framework. The following section treats the more general question of EM III–MM IA chronology in East Crete. The chapter ends with a discussion of Late Prepalatial industry, trade, and social organization in the Mirabello region.

Gournia Survey The Gournia Survey Project reported a total of 28 EM III–MM IA habitation sites: 4 villages, 14 hamlets, 6 farms, 4 field sites, and 2 burials (Table 3; Map 17). These numbers represent a sharp drop from the total of 38 EM I–II habitation sites (3 villages, 14 hamlets, 9 farms, and 12 field sites, as well as 4 knapping sites and 3 cemeteries). Along the coastal sector of the Gournia Survey, seven sites—3, 4, and 6 (both knapping sites), 7, 22,

23, and 24 (a cemetery)—were abandoned after EM II. Gournia (8), Pera Alatzomouri (10), and a nearby hamlet (14) continued to be occupied, while two new sites (1 and 11) were settled nearby. Just south of Gournia, site 11 might have had two to four houses. Its catchment may have been the hill slopes to the south, but the land was poorly suited for cultivation. The advantage to this site may have been its protective height and distance from the coastline. It is also

34

L. VANCE WATROUS AND MARYANNE SCHULTZ

possible that this site was established by inhabitants who abandoned the coastal sites in EM II. Site 1 was settled in MM IA at the western end of the Gournia Valley, about 700 m away from Gournia, in an area removed from any immediate water source. It is just possible that the settlers from the nearby EM II sites 3, 6, and 7, which were deserted in EM II, moved to site 1, although this seems unlikely since sites 3 and 6 possessed superior catchments. The size of site 1 in the MM IA period was small, but it grew into a large settlement in the MM IB–II period. Gournia (8) grew appreciably in this period (Map 18). Boyd (Hawes et al. 1908, 20, 57; Fotou 1993, plan B) discovered Late Prepalatial pottery in “an early house” northwest of House Ea and in the North Trench. The survey found contemporary sherds immediately north of the acropolis ridge, along the river west of the acropolis, and also extending north of the asphalt road as far as the coast. Pera Alatzomouri (10) grew somewhat from its EM II size; it produced Late Prepalatial sherds on the north hilltop and west slope, and possible traces on three spots on the south hill. Two ridges to the east, a large new settlement (17) was established on the ridge of Phylakeion overlooking the Bay of Pacheia Ammos. Located on a saddle hidden from the sea, the site consists of a concentration of sherds over an area of 125 x 250 m. Such a large settlement cannot be accounted for by nucleation and may have been established by newcomers. It is a striking development that two new settlements appear east and west of Gournia in this period. Farther inland in the Vasiliki area, six of the seven EM II sites along the stream continued into this period; only one EM II site (29) was deserted. Our most detailed evidence for the developments in this area comes from the excavations at Vasiliki (discussed below). Across the Isthmus valley to the east, in the area of the Cha Gorge and Monastiraki, occupation of the EM II site 67, located next to the mouth of the Cha Gorge, probably ceased during this period, while site 65 doubled in size to include perhaps two houses. Site 65 produced fine ware as well as a jar and amphora. At the same time, four new sites (44, 54, 60, 63) were established on the dry valley floor in a previously unoccupied area of the Isthmus. Site 44 was a small farm using cookware, pithoi, jars, amphorae, and basins. Its cemetery was located in

site 45, two caves several hundred meters to the north. Site 54 was another small farmstead; pottery consisted of cookware and basins, including a scored basin (beehive?). Village-sized in the subsequent Protopalatial period, site 60 produced sherds (cookware and jars) in two spots, 100 m apart; hence, it may have already been this size at this time. Site 63, a cemetery, produced pithoi (one containing human bones) and larnakes. In the southern portion of the survey zone, south of the villages of Vasiliki and Monastiraki, the area had 23 sites in EM I–II, but by EM III–MM IA, the number was reduced to 11. Most of the abandoned sites were small, either farms or field sites. The cluster of dispersed small EM I–II sites was replaced in the Late Prepalatial period by a new cluster of hamlets (84, 105, 114, 118, 131, 144, 151). Several groups of small EM II sites (117, 121, 130, 132, 136) clustered around a hamlet (131) disappeared, while the hamlet remained. Several of these hamlets were either newly enlarged (84, 105, 118, 131) or newly founded (151). The new site 151 was established on a small spur in the middle of an upland valley (elevation 400 m). Site 84 consists of several houses situated high on the slope of Thriphti next to a spring. At the base of Kazarma hill, a farm or small hamlet (105) produced sherds, chipped stone, and a celt. On a steep hill, perhaps two farmhouses (114) used jars, cookware, and cups. A hamlet (118) was established at Selima, high on the steep mountain slope next to several springs. Site 131 is based on a natural ledge within a ravine (Pl. 4C), with cliffs on three sides. In this period (or MM IB–II?), its only natural access has a cyclopean tower (Pl. 5A) blocking the entrance to the site. Cyclopean fortification walls (Pl. 5B) run along the north and south edges of the cliffs. Southeast of Episkopi, site 144 sits on a hill surrounded on all sides by steep slopes and massive defensive walls protecting its entrance (Pl. 5C). Early Minoan III–MM IA was a period of important change in the Gournia survey area. First, there was a sharp reduction in the overall number of sites in this period, from 38 EM I–II habitation sites to 28 Late Prepalatial habitation sites. Sites along the coast (4, 7, 22) were abandoned. The largest EM II settlements, Gournia (8), Vasiliki (135), and sites 97, 114, and 131, however, remained the same size or grew. Second, new sites (11, 118, 131, 144) were established in defensive locations.

EARLY MINOAN III–MIDDLE MINOAN IA PERIODS: DISRUPTION AND SOCIAL REORGANIZATION

Third, large settlements appeared at the same time that many small sites were deserted. In some cases, nucleation may have been a cause; Gournia (8), for instance, seems to have grown appreciably in this period (Map 18). In other cases, new sites were established in previously unoccupied areas. Site 1, for instance, was settled on the slope at the west end of the small Gournia Valley. At the same time, two ridges to the east of Gournia, a large settlement (17) was established on the ridge overlooking the Bay of Pacheia Ammos. (This settlement is probably the source of the tombs excavated by the 24th Ephorate in 2007 [Apostolakou 2008].) Site 17 is new in this period, and neighboring sites either continued into EM III–MM IA or were deserted in EM IIA, long before its foundation. Neither of the two large new settlements (1, 17) can be accounted for by nucleation. Hence, it seems likely that they were founded by newcomers (see below). Fourth, new settlements were often established in marginal environments. For example, three new sites, 44, 54, and 60, were established on the dry valley floor in an unoccupied area of the Isthmus. Fifth, the EM III–MM IA settlement pattern consisted of a series of compact site clusters separated from one another by a distance of a kilometer or more of unoccupied land. North of the Cha Gorge, sites 44, 54, 60, 63, and 65 formed such a cluster. Between Monastiraki and Episkopi, the dense scatter of EM I–II sites was transformed into two separate clusters (84, 90, 97, 105, 114, 118, and 131, 137, 144, 151). These clusters represent a fundamental change from the evenly dispersed pattern of settlement in the EM II period.

35

Sites in the Gournia survey area fall into two categories: those on defensible, hilltop, or mountainous locations (8, 10, 14, 17, 84, 97, 114, 118, 131, 135, 144, 151), or those in lowland areas near a water source (26, 28, 30, 31, 33, 90, 99, 105, 137). Rather than view these two different types of sites as completely contemporary, it seems more likely, given what we know about the EM III–MM IA period (see the discussion below), that the sites in the first group were continuously occupied during this period, and that those in the second group represent MM IA settlement into good agricultural areas, including the Isthmus Plain southwest of Kavousi (K44, 54, and perhaps 60), an expansion that continued into MM IB–MM II. All in all, the EM III–MM IA pattern in our area suggests the following sequence: initial abandonment of smaller sites, nucleation at larger sites, and settlement of new sites in EM III, followed by general settlement expansion later in MM IA. The off-site distribution of Prepalatial sherds (Maps 13, 14) is particularly interesting because in some areas, that is, around sites 11, 17, 60, and 99, the sherds must represent a newly intensified use of marginal land during the Late Prepalatial period, since few or no earlier sites existed in the area. The predominant off-site vessels found, cooking pots and jars or pithoi, correlate closely with agricultural use of the land—namely, for farmers’ meals and the storage of water for animals. Only immediately around Gournia were fine-ware vases recognized, which might be taken as a sign of manuring, the products of which were initially stored in or near the household.

Kavousi Survey Haggis (2005, 65–69) reports that sites in the Kavousi survey area increased from a total of 11 to 16 in the transition from EM I–II to EM III–MM IA (EM I–II: two villages, one hamlet, two farms, three field sites, one workshop, two burials; EM III–MM IA: three villages, four hamlets, four farms, five field sites, one workshop, four burials). This overall increase in site numbers is in contrast to the pattern observed in the Gournia and Vrokastro areas, where there was a substantial reduction (33%–50%) in the number of sites. As Haggis (2005, 65) suggests the

increase recorded may be due to his familiarity with the ceramic sequence, based on local excavation. On the other hand, Haggis displays a willingness to assume continuity at sites with little or no supportive evidence. His Late Prepalatial sites K5, K8, K45, and K92 each produced a single EM III–MM IA sherd, and his sites K1, K6, K24, and K41 yielded no EM III–MM IA pottery. Haggis (2005, 65) identifies site K5 as village-sized, although he says “the exact size of the site (K5) in EM III is unknown.” Additionally, that a lowland (K5) and

36

L. VANCE WATROUS AND MARYANNE SCHULTZ

defensive hilltop (K71) site should grow at the same time that the major settlement center (K24) in the area was deserted seems inherently unlikely. Skepticism about Haggis’s interpretation is further fueled by the high degree of discontinuity in the settlements of the Kavousi area during the Late Prepalatial period. Four of the 11 EM I–II sites— K1, K7, K24, K62—were deserted, including the central village site at Kavousi (K24) and the harbor at Tholos (K1). The major site under Kavousi village (K24) was abandoned, probably for the hilltop site of Azoria (K71) immediately above it. Additionally, 11 of the 15 habitation sites in EM III–MM IA are new foundations, including villages K5, K41, and

K71(?); hamlets K45 and K92; farms K28, K36, K44, and K68; and K31 and K35. As at Gournia, settlements were founded in new clusters: around K5, in the area (Haggis 2005, fig. 10) of Chrysokamino (K32) and Azoria (K71). The distinct cluster of eight sites around Chrysokamino is especially striking. Six of these sites were newly established in this period, probably by newcomers (see below). Excavation (Betancourt 2006) has shown that the Chrysokamino habitation site (K34) continued in EM III–MM IA, and the Chrysokamino metallurgical work site (K32) continued into EM III–MM IA. During this period, new sites (K71, K92) were also founded in marginal upland areas.

Vrokastro Survey The publication of the Vrokastro Survey Project unfortunately considered local site data from EM II and EM III together as one period (Hayden 2004a, figs. 16, 17). An additional obstacle to interpreting the Vrokastro sites is their inconsistent designations in the catalog as “EM,” “EM II–III,” “EM II–MM,” “EM–MM,” or “EM, MM I–II.” Following our method in Chapter 4, we have only counted sites that were specifically dated “EM III” or “EM III– MM I” or that were recorded in the catalog as producing EM III–MM IA pottery. Hence, our list consists of 13 secure EM III–MM IA sites, whereas the Vrokastro catalog lists 19 EM II/EM III–MM I sites. Our list, therefore, is a minimal one, but this method of selection has the advantage of allowing us to compare the EM I–II and EM III–MM IA settlement patterns in the Vrokastro region. In the Vrokastro area, at least 13 settlements were occupied in the EM III–MM IA period, down from 30 EM II sites (46 EM I–II sites). A large number of EM II sites were abandoned, especially along the

coast, including VN1, VN2, and IM8. Hayden specifically identifies 12 EM II sites that have a gap in occupation between EM II and MM IB—namely, AC4, APh11, GN2A:1, KM2, IM2, IM5, IM6, IM8, KA1, PT1, IS2, and PI1. Four additional EM II sites (APh2, GN2A:2, GN4, KP6) ended in the Early Minoan period and do not continue in the Middle Minoan period. Small sites were especially vulnerable: only 9 of the 20 EM II farm sites continued, and only 4 of the 10 EM II hamlets continued in this period. Only two of the EM III–MM IA settlements, KM1 and SK9, were new in this period, and both are at high locations in the mountains. Upland, inland sites were also favored: KM1, SK9, PT4, VK7, SK8, and SK9. Five sites—KM1, PT4, VK7, SK9, and SK9—are inland at high elevations. The only coastal site that maintained its EM IIB size was the large settlement at Priniatikos Pyrgos (PP1; B. Hayden, pers. comm.). Several large sites (PP1, SK9, VK7, APh1, KA2) may indicate a nucleation of population.

Regional Settlement The settlements in the three survey areas exhibit similar developments. First, many EM II sites across the Mirabello region, especially small ones, were abandoned. Small coastal sites (7, 22, 23, K1) in particular were abandoned. Mochlos suffered a

destruction in EM IIB (Soles 2009, 10). The major settlement under Kavousi village (K24) was deserted, probably for Azoria (K71) immediately above it. Two large settlements, however, existed on the coast, at Priniatikos Pyrgos and Gournia. Second,

EARLY MINOAN III–MIDDLE MINOAN IA PERIODS: DISRUPTION AND SOCIAL REORGANIZATION

sites were established or grew in defensive locations. Several of these sites in the Gournia area— 11, 118, 131, and 144—had massive perimeter walls. Third, new sites were established in separate clusters, as if for defensive reasons. All of these developments point to an influx of newcomers who settled along the coast and in unoccupied inland areas (see below).

Excavated Settlements Local excavated sites provide details that support the more general changes exhibited by the settlement patterns outlined above. As we have seen in Chapter 4, EM IIB was characterized by increased international connections, the appearance of social ranking (mainly visible in grave goods), and specialized craft production. The regional balance of economic power in EM IIB did not last long, however, as several local sites were destroyed by fire: Phournou Koriphi, Mochlos, Myrtos/Pyrgos, Vasiliki, and Malia. Some sites, namely, Phournou Koriphi and Myrtos/Pyrgos, remained deserted during EM III–MM IA. Vasiliki is particularly important because it was an EM IIB center, and Seager’s excavations there give us some specific details of what happened to the site in the EM III–MM IA period. Seager (1905, 218) reported that after the EM IIB destruction of the site, the succeeding (his period IV) inhabitants “built their hovels only over the southeast corner, as there are no signs of their ware on any other part of the hill.” In a later report, Seager (1907, 118) wrote, “The area in which the sherds of period IV were found was small, measuring about eight by ten metres. The settlement is even less extensive than that of the preceding period (period III, EM II) and seems to have consisted of

37

poorly built hovels huddled against the massive outer walls of the big house of period III, already in ruins.” By MM IA, however, Vasiliki had recovered and grown considerably (Zois 1992). During MM IA, the settlement possessed at least four houses, Seager’s Houses A and B, and Zois’s Houses Gamma, Phi and X, as well as other remains of well-built structures on the central hilltop (E-50) and on the south slope (PB1). These houses were protected by a massive fortification wall and tower (Map 19; Watrous 2001b, 170 n. 218, fig. 9). The town, however, was destroyed by fire in MM IA. At Gournia the excavators found a large deposit (“North Trench”) of pottery at the north edge of the settlement that had been dumped over a wall (Hall 1904). At Mochlos several sizable deposits of MM IA pottery have been found in blocks A and C, and under House D (Seager 1909). More recent excavations at Mochlos by Soles and Davaras have revealed an EM III–MM IA building (Soles and Davaras 1994, 1996). The hilltop of Myrtos/Pyrgos was also reoccupied in this period. In MM I, the settlement around the hilltop grew substantially, and the top of the settlement was crowned by a massive tower and an elite structure described as a large and rather rich building that was the predecessor to the Neopalatial villa (Cadogan 1977–1978, 74). At Pseira, extensive traces of settlement first appear on the entire hill northeast of the harbor, which the excavators suggested belonged to newly arrived settlers (Davaras, Betancourt, and Farrand 1992). As the chronological phasing of this period, which is basic to any historic reconstruction, has been controversial, it seems best to review the relevant evidence before reconstructing the sequence of events during the Late Prepalatial period.

Early Minoan III–Middle Minoan IA Chronology and Settlement History in East Crete The history of the Late Prepalatial period (EM III–MM IA) in the Mirabello Bay region begins in 1904 with the publication of Edith Hall’s article on the North Trench at Gournia. Digging north of the

LM I town, Hall (1904) discovered a massive pottery deposit characterized by geometric whiteon-dark decoration, which Hall (1904, 203–204) correctly placed between the light-on-dark EM II

38

L. VANCE WATROUS AND MARYANNE SCHULTZ

ware found by Seager at Vasiliki and the Middle Minoan “Kamares Ware” (MM I–II) ware. Since “Kamares Ware” was thought to date to MM I–II, Hall called her deposit EM III. Given the knowledge of Minoan ceramics at the time, her label made good sense—it was only some 20 years later that Evans (1921–1935, I, 172) realized that Middle Minoan I consisted of two ceramic phases, Prepalatial MM IA and Protopalatial MM IB. Nevertheless, the EM III date of the North Trench deposit was solidified in 1984 with Betancourt’s book East Cretan White-on-Dark Ware that analyzed a number of East Cretan light-on-dark ware deposits stylistically. In his book, Betancourt dated the North Trench as following the Myrtos EM IIB pottery (Betancourt 1984, 9–10). This assertion is easily refuted. Betancourt’s comparison (1984, figs. 2–4, 2–5) of the main shapes of the two deposits is not valid as he did not represent them to scale (e.g., the EM II bowls from fig. 2–4 are much larger than the North Trench cups in fig. 2–5). In fact, the main EM IIB shapes, the shallow bowls, curved bowls, jugs, and spouted jugs from Myrtos (Warren 1972, figs. 42–59), have no counterparts in the North Trench. This fact should indicate that there is a chronological gap between the two deposits. More recently, archaeologists at Knossos came to the same conclusion. Warren and Hankey’s fundamental book, Aegean Bronze Age Chronology, pointed out that cross links between North-Central and East Cretan pottery deposits indicate that the eastern white-on-dark style continued to be produced during the Knossian MM IA period (Warren and Hankey 1989, 20). A burial deposit at Pacheia Ammos (Apostolakou 2008) with geometric whiteon-dark decoration, but no curvilinear designs popular in the North Trench, has been identified as EM III (P. Betancourt and T. Brogan, pers. comm.). As a result, most Cretan pottery specialists now agree that the North Trench deposit dates to MM IA— excavations in 2010 have revealed that the level above the deposit dates to MM IB—and that, with the exception of the small Pacheia Ammos deposit, there exists a gap in the ceramic sequence between EM IIB and MM IA at published sites in the Mirabello Bay area. This ceramic gap corresponds to a disruption in the sequence of settlement in East Crete. At

Myrtos/Pyrgos, for example, the earliest post–EM II architecture at the site is defined by deposits with polychrome pottery, i.e., no earlier than MM IA (Cadogan 1977–1978, 71). In the Kavousi area, both the coastal site at Tholos (K1) and the important settlement in the village (K24) were deserted, probably for the resettled site at Azoria (K71) on the hilltop behind and above the village. At Palaikastro, the coastal settlement was abandoned for the hilltop of Kastri in EM III (Watrous 2001b, 179 n. 170; Driessen, MacGillivray, and Sackett, eds., 2003, pl. 6). In the Gournia coastal area (see above), seven sites were deserted after EM IIB. What would cause such a pattern in the ceramic sequence and pattern of settlement? The answer is Cycladic immigration. Many sites in the Cyclades, including Akrotiri on Thera, were deserted at this time (Davis 2001, 74; Nikolakopoulou 2006). At the same time, signs of Cycladic influence and/or presence appear along the north coast of Crete. Cist graves, a Cycladic type, and pithos burials, a western Anatolian type, appear locally at Sphoungaras, Pseira, and Pacheia Ammos, as well as at many other north coast sites. The hallmark fabric for this period, Haggis and Mook’s XX/XXI (Haggis and Mook 1993; Haggis 2005, 50–51), is characterized by crushed calcite or quartz inclusions, which is a feature of Cycladic pottery production in the Early Cycladic and Middle Cycladic periods (Day and Wilson 2007). At Chrysokamino, where ore was brought from Kythnos, the fabric of the local ware, including the furnace chimneys, includes crushed angular quartz (Haggis 2005, 114; Betancourt 2006, 71–97 n. 28). The above evidence suggests that Cycladic immigrants arrived on Crete at this time, which was probably the cause of local site abandonment, particularly along the coast. New sites in deserted areas, such as sites 1, 17, and IM9, as well as the site clusters around site 60, and in the Chordika and Tholos areas of the Kavousi valley, were probably founded by Cycladic immigrants. The largest sites, i.e., Gournia and Priniatikos Pyrgos, however, either grew at this time, probably through nucleation, or maintained their EM IIB size. Excavation and survey work in East Crete suggest the following sequence of events during the period: conflicts on Crete during EM IIB, arrival of new settlers, abandonment and/or reduction of

EARLY MINOAN III–MIDDLE MINOAN IA PERIODS: DISRUPTION AND SOCIAL REORGANIZATION

some sites in EM III, and a nucleation of population at other sites as well as settlement on new, defensively located, upland sites. In the MM IA

39

period, there was an expansion of settlement at sites, some of which were fortified.

Society Trade and Industry Toward the end of the third millennium B.C., Egypt, the Levant, Anatolia, and the Aegean were swept by destructions, internal chaos, the abandonment of sites, and the arrival of new peoples. Early in the second millennium B.C., the cities of the Levant and Egypt reestablished relations with one another, and Crete reentered the larger commercial orbit of the eastern Mediterranean. Crete exported MM IA vases overseas to Aegina, Lerna, Samos, and Cyprus (Watrous 2001b, 197–198). Sites in the Mirabello region participated in this trade. At Mochlos, a MM IA clay kantharos (Seager 1912, pl. XIV:37), based on a ceramic Anatolian shape, points to contacts with this silver-rich area in MM IA, a connection corroborated later by the wellknown MM IB–MM II silver kantharos from Gournia (Hawes et al. 1908, 56; Davis 1979). A dagger (Seager 1912, fig. 45:XI:22) from Mochlos Tomb XI, found with MM IA pottery, has been considered a Near Eastern type, similar to examples from Byblos (Dietz 1971, 17). Beginning in MM I, Mochlos seals were made from ivory, a material imported from Syria and/or Egypt (Krzyszkowska 1988). Imported materials at sites in the Mirabello region also point to a revival of foreign trade connections. Ores from the Cyclades were being crushed and smelted at Chrysokamino for use within the region (Betancourt 2006). The cemetery at Sphoungaras produced a heart-shaped piece of jewelry (Hall 1912) made of imported gold that may date to this period (Higgins 1980, 64). Middle Minoan IA Mirabello workshops produced goods for export and local use. Some of the MM IA–II gold jewelry from the Mochlos tombs dates to the Late Prepalatial period (Watrous 2005a). A number of Egyptianizing stone vases, jewelry, and seals of this period (Watrous 2005a,

stone vases: nos. II:k, II:o, VI:7, VI:9, XXI:2; jewelry: XXI:19; seals: II:41, II:42) found in the Mochlos tombs indicate that local craftsmen specialized in making copies of exotic foreign items, much like the contemporary “White Piece” workshop of Egyptian-like seals on the south coast (Pini 2000). Mochlos workshops also produced clay and stone vases (Seager 1912, pl. XII:1, right and left; Watrous 2005a, 112, fig. 18), metal weapons (Seager 1912, fig. 44 XIX:27 and probably XXI:22), and perhaps seals. Priniatikos Pyrgos (PP1) and its immediate area probably continued as a major pottery production center for the export of Mirabello Wares that have been found at Knossos (Warren 1965, 23–28), Malia, and Petras near Siteia (Hayden 2004a, 74, 99).

Ideology Funerary customs at major sites in the Mirabello region change dramatically. At Mochlos, a platform was constructed outside of Tomb IV/V/VI in MM IA (Soles 1992, 220). Found with MM I stone vases on it (Soles 1992, 62, fig. 24), this platform served as an altar for offerings (Soles 1992, 56–57). One of these vases (Soles 1992, fig. 24:1) is a miniature amphora common in Middle Kingdom contexts (Warren 1969, 361, 363, fig. 28), and hence, is probably MM I in date. The second vase is a MM I tumbler (Warren 1969, 91–92). At Gournia, Tomb II was built in MM IA (Soles 1992, 23). It was unique among the Gournia tombs in that a platform with a kernos on its upper surface was affixed to the tomb for offerings. At Myrtos/Pyrgos, near the top of the hill, a monumental ossuary and processional way were constructed in MM IA (Soles 1992, 176–179). This ossuary contained exclusively male burials and elite offerings. A kernos for offerings was set in the pavement next to this tomb.

40

L. VANCE WATROUS AND MARYANNE SCHULTZ

This type of public offering-place at a cemetery is a new phenomenon in this period. The Gournia, Mochlos, and Myrtos/Pyrgos constructions point to the appearance of social groups that distinguished themselves by a public display honoring their ancestral ties, and, by extension, their preeminent social status. These innovations should probably be regarded as the local Mirabello version of developments in Central Crete—that is, the construction of the Chrysolakkos tomb at Malia and the paved terrace outside Building 6 in the Archanes cemetery (Watrous 2001b, 188–189), both elite structures used for public ceremony. It should be added that already in 1945, when Pierre Demargne published the Chrysolakkos tomb, he recognized its many Egyptian features, which suggested it was designed as a Cretan version of an Egyptian mastaba (Demargne 1945). Not only were the elite Mirabello tombs larger than other local examples and designed for public display, but they also incorporated foreign objects and forms. One of the stone vases dedicated outside Tomb IV/V/VI, for example, was an imitation of an Egyptian shape. The Mochlos tombs contained Egyptianizing funerary gifts (Watrous 2005a, 107–114). The clay larnakes in the Sphoungaras and Pacheia Ammos cemeteries are also ultimately an Egyptian shape. At Myrtos/Pyrgos, the processional funerary ramp also has monumental parallels in Egyptian royal funerary architecture. At the time that these elite tombs were established, new burial types also appeared at the bottom of the social scale. In this period, poor individual burials, in jars and clay coffins, became common. Such burials are known from Sphoungaras (Hall 1912), Pacheia Ammos (Seager 1916, pl. III), and Mochlos (Soles 1992, fig. 34).

Social Organization The settlement pattern for this period indicates that the local population consisted of different ethnic groups living in clusters. Fortified settlements point to a lack of security and any central authority. Literacy, identified at palatial sites in Central Crete, remains unknown in the Mirabello region during this period.

On the basis of tomb contents, we can detect the presence of prosperous families at MM IA Vasiliki (well-built houses A, B, Gamma, Phi, and X), at Mochlos (rich tombs II, VI, XIX, and probably III and XXII), and perhaps at Gournia. Beginning in the MM IA period (and continuing into MM IB–MM II), several changes are apparent in the internal social hierarchy within the Mochlos community. West-terrace tombs continued to produce much gold jewelry, numerous stone vases (some based on local ceramic shapes and others on Eastern shapes), and two seals of imported ivory (Seager 1912, nos. II:41, II:42) with Egyptian-like designs. The smaller south-slope tombs both imitated the west-terrace tombs with a few gold items (Seager 1912, nos. XIX:11, XIX:13, XIX:23) and jewelry (Seager 1912, no. XXI:19, a necklace of gold tubular beads; probably no. XIX:14, a necklace of carnelian, amethyst, crystal, and gold beads), but primarily contained clay vases and many small stone vases (Seager 1912, nos. XIX:3, carinated bowl; XIX:7, alabaster cup; XXI:2, globular alabastron, XI:12, 15, 17, 19–21; XV:e, f). Judging solely from the tombs, the Late Prepalatial Mochlos community seems to have consisted of one preeminent social group (Tomb IV/V/VI), three other elite groups (Tombs I/II/III, XIX, and XXI), and at least fifteen commoner families (Tombs VII–XI, XIII, XV– XXIII). Certain Late Prepalatial innovations, such as the appearance of simple individual burials and the public use of exotic foreign objects, combined with the introduction of new people in the area and signs of regional conflict, may indicate the beginnings of intensified social hierarchy and greater social diversity. The elite tombs at Mochlos, Gournia, and Myrtos/Pyrgos point to groups that distinguished themselves by a public display honoring their ancestral ties, and by extension, their right to landownership and preeminent social status. It is possible, therefore, that elites and commoners played new social roles, created in response to changing ethnic, religious, and demographic pressures. These trends in social reorganization became more obvious in the succeeding MM IB–II period, when evidence is stronger.

6

Middle Minoan IB–II Periods: Growth of Regional Factions and Conflict L. Vance Watrous and Maryanne Schultz

This chapter begins with an account of the MM IB–II sites located in the Gournia Survey Project area. Following a summary of contemporary survey data from the Kavousi and Vrokastro areas, the chapter analyzes the Protopalatial settlement pattern

of the Mirabello region. The final sections discuss excavated settlements, land use, craft production, exchange, and social organization of Protopalatial society in the Mirabello Bay region.

Gournia Survey The Gournia Survey recorded a total of 76 Protopalatial sites (Table 4; Map 20): 13 villages, 29 hamlets, 17 farms, and 17 field sites, as well as 7 burials (and 3 possible burials). This is a significant rise from the 28 EM III–MM IA sites (4 villages, 14 hamlets, 6 farms, 4 field sites, 1 work area, and 2 burials). A total of 48 sites are fully “new” in the Protopalatial period: 2 villages, 18 hamlets, 13 farms, and 15 field sites. Twenty-eight of the 76 habitation sites had been occupied previously (11 villages, 11 hamlets, 3 farms, and 3 field sites). That leaves 48 sites that are new in MM IB–II (2 villages,

19 hamlets, 13 farms, and 15 field sites). This represents an impressive rise in number from the Late Prepalatial period, where there were 28 sites (4 villages, 14 hamlets, 6 farms, 4 field sites, 1 work area, and 2 burials). Even though all site categories grew substantially, expansion is noticeably sharp at larger settlements. For example, Gournia (8), Sphoungaras (10), and Vasiliki (135) grow to become substantial settlements in this period. Seven other Late Prepalatial sites become village-sized (1, 14, 60, 84, 90, 97, 114). Significantly, the sites that increase the

42

L. VANCE WATROUS AND MARYANNE SCHULTZ

most in size are new in the Late Prepalatial period. The appearance of numerous individual burials across the landscape is especially noticeable. All Late Prepalatial sites except one (30) continue into the Protopalatial period. The settlement pattern is striking in that almost all site growth in this period took place exclusively in and around clusters (Map 21) already established in the Late Prepalatial period. Haggis (2005, 70–73 and fig. 20) noticed the same pattern in the Kavousi area, and he suggested that these clusters functioned similarly to villages, possessing a leading household and occupying separate agricultural areas, and controlling miniature territories and local water supplies. All four of the site clusters in our area had at their center one or more large settlements (8–10; 17; 60; and 83–84–90–97–100), surrounded by an expanding halo of sites. Each site cluster was separated from the others by areas of unoccupied land. Spatially, the clusters resemble a traditional village surrounded by field houses (used to keep tools and to store the harvest) and farm- to hamlet-sized metochia or field houses lived in seasonally. But two of the Protopalatial clusters, around Gournia and south of Monastiraki, have more than one large central settlement, suggesting that they represent a social group larger and probably more complex than a village. Immediately north and east of Episkopi, an east–west strip of unoccupied land seems to form a boundary between our area and the southern Isthmus. Settlements can be grouped into two types: ones whose foundation can be traced back to the FN/ EM I–EM IIA periods (Map 21, black circles), and newer ones whose foundation dates to EM III–MM I (Map 21, gray circles). In some clusters, there seems to have been an ethnic mixture of populations, one predominating. We can see this in the Gournia Valley, where Gournia (8), Alatzomouri (10), 11, and 14 were older settlements, whereas sites 1, 2, 5, 6, and 7 were more recently settled (EM III–MM I). As we have seen in Chapter 5, the later sites and clusters probably represent Cycladic immigrant groups that arrived in Crete in EM III. The first cluster, along the coast around Gournia, consisted of at least three village-sized settlements (1, 8, 10) on adjacent hills no more than 300 m away from each other. By this period, sites 8 and 10 had probably become one site, centered at Gournia. Site 1 produced many ancient walls

(recently destroyed in part), a full range of coarse and fine wares, loomweights, chipped stone, hand axes, querns, and a whetstone. Several small new sites (2, 4, 5, 6, 7, 12, 13) expanded along the coast and up into the hills to the south. Sites 2, 4, 5, 12, and 13 were single farmsteads, site 11 was a hamlet, and site 14 a small village. On the coast, a rectangular structure (7; a house?) may have been a potter’s establishment in this period. The second cluster, around site 17 near Pacheia Ammos, included sites 16, 20, 22, 24, 25, 26, 28, and 43. Site 20, founded underneath the modern village of Pacheia Ammos, is of uncertain size due to modern construction. Visible sherds cover an area 100 x 150 m or 0.5 ha, but it could have been larger. In addition to domestic vessels, site 20 produced larnax fragments and other possible grave goods, which means the site also partly overlapped Seager’s cemetery (18) under the village. The hamlet 22 may have included farms and a potter’s workshop. Next to 22, site 24 produced evidence of settlement (cookware, amphorae, and fine ware), pottery production (a waster), and burials (larnakes and jars). A hamlet, 25, was founded at the east bank of the stream, along the eastern edge of Pacheia Ammos. A rectangular cyclopean terrace (Pl. 6A) formed a retaining wall to support a house that was positioned to take advantage of the rich alluvial land alongside the stream. To the south, sites 26, 28, and 43 were agricultural establishments. Sites 22 and 43 had clay vats, probably for pressing grapes. The third cluster, around Plakoures (60), expanded from 5 Late Prepalatial sites to 15 sites, covering an area of 3 ha. A string of new sites on the dry plain in this cluster (61–67) may have been connected to the Cha Gorge along an artificial water channel (a long channel cut into bedrock runs out of the northern side of the mouth of Cha Gorge, but its date is unknown). Similar water channels (irrigation ditches) were maintained between Cha and Pacheia Ammos for watering the fields before the introduction of deep wells some 40 years ago. The main site 60 consists of a central cyclopean (recently destroyed) building, surrounded by rubble houses and much pottery, including large jars, pithoi, cookware, and basins (some scored for beehives?). The farm 44 and new hamlet 46 had their own cemetery. Site 46 consisted of a central structure set on a cyclopean retaining wall (Pl. 6B) surrounded

MIDDLE MINOAN IB–II PERIODS: GROWTH OF REGIONAL FACTIONS AND CONFLICT

by outlying rubble buildings. Site 44 yielded evidence for agriculture (storage jars, a beehive, and spouted basins) and amphorae for wine production. Sites 61–66 seem to have been farming settlements with their own cemeteries. Site 67 produced a scored basin (beehive?). The slopes north of site 60 were also newly settled with field sites, farms, and a hamlet (55, 34, 49, 51, and 52). Most of these sites produced both agricultural vessels and funerary vases, suggesting that they were both farms and burial sites. The fourth site cluster south of Monastiraki, in the area of the large spring at Kephalovrysi, also became larger. Rich in springs, this area supported 7 sites in the Late Prepalatial period; in the Protopalatial period the number rose to 25. Sites 68, 69, and 70 were settled in a new area, on the slopes around the village of Monastiraki and its spring. Both sites 67 and 68 show some sign of associated burials. Several Late Prepalatial sites (84, 90, 97, 114) became village-sized. An additional 15 new sites were settled (villages 83 and 100; hamlets 76, 78, 95, 107, and 112; farms 82, 96, 108, and 120; field sites 75, 99, 104, and 109). The growth in this inland area was so extensive that one wonders if it was not based on settlers who had lived along the coast in the Late Prepalatial period. New settlement grew up to take advantage of arable parts of the plain (82, 100, 104, 105) and the slopes of Thriphti (95, 96, 109, 112, 120), often near a water source.

43

All the sites in this cluster seem to have been agricultural in nature, yielding a common assemblage of cookware, storage jars, basins (100 and 114 had a beehive), and amphorae, although a few (83, 97, 98) held larnakes. Site 99 is unique in that the settlement overlies a large chert source and produced two crucible fragments. Four sites were found along the stream between Pacheia Ammos and Vasiliki village. Two of the sites, 32 and 39, are new. These sites may have been part of a cluster centered at Vasiliki (135), but this remains uncertain because the area south and west of Vasiliki lay outside our survey zone. Farther south, the number of sites around and south of Episkopi grew from three to seven in the Protopalatial period. Separated from the Monastiraki cluster by a kilometer of unoccupied land, these seven sites (123, 131, 137, 140, 143, 144, 151) look as if they belonged to a separate cluster located south and/or west of Episkopi, outside our survey zone. This collection of sites also produced an agricultural assemblage: jars, cookware, basins (123 had a spouted jar for pressing wine), and amphorae; site 143 included evidence of burials. At the edge of Kato Chorio, site 140 may have had one or two cyclopean buildings. Much destroyed, the site consists of cyclopean foundations, a water channel, querns, and a large chunk of granodiorite (imported for pottery production?).

Kavousi Survey In the MM IB–II period, the total number of sites in the Kavousi survey area increased from 21 to 71 (Haggis 2005, fig. 11). Protopalatial sites consisted of 3 villages, 10 hamlets, 14 farms, 40 field sites, and 4 burials. Hence, Kavousi had a total of 45 new settlements in the Protopalatial period (6 hamlets, 8 farms, and 31 field sites). In the Protopalatial period, villages remained consistent in number (3), while hamlets more than doubled (4 to 10); farms grew by 3.5 times (4 to 14); and field sites rose by a factor of 8 (5 to 40). Almost every site from EM III–MM IA continued to be occupied in MM IB–II. Most sites remained the same size. Each of the four EM III–MM IA site clusters expanded. Most new sites were small and located to exploit nearby land.

The harbor site Tholos (K1) was reoccupied and grew to be village-sized. New sites K2, K3, and K4 clustered around Tholos. Both Late Prepalatial site clusters, around Chrysokamino and around Azoria, expanded dramatically. The small Chrysokamino cluster grew from 9 Late Prepalatial sites to 17 Protopalatial sites. Haggis (2005, 72) suggests that the cyclopean structure on site K44 in this cluster may represent the community leader. Azoria (K71), the hilltop refuge site, was abandoned for site K24, located on the plain immediately below. The Late Prepalatial cluster of 3 sites around Azoria grew to 24 in the Protopalatial period. On the Xerambela Ridge, excavations at Vronda (K77) have revealed earlier material, including a sealing and a large

44

L. VANCE WATROUS AND MARYANNE SCHULTZ

Protopalatial wall. The cyclopean structure on site K68 may mark it as the center of this cluster (Haggis 2005, 73). On the Kambos Plain, the 2 Late Prepalatial sites clustered around Hagios Antonios (K5) grew to at least 11 sites and probably included Tholos (K1),

bringing the total to 16. New sites in this cluster consisted of farms (K15, K16, K17) and field sites (K12, K13, K18, K19) founded at the edge of the plain. Alykomouri (K8) was deserted in MM IB, probably for the hamlet K11 on the plain below.

Vrokastro Survey Settlement in the Vrokastro area also expanded vigorously during the MM IB–II period. From the Late Prepalatial total of 13 sites (or, following Hayden’s count [2004a, 71], 47), the Protopalatial total grew to 88 sites (Hayden 2004a, fig. 17). In either case, the increase was quite large. Protopalatial sites in the Vrokastro area consisted of 13 villages, 33 hamlets, 13 farms, 26 field sites, and 3 burials. The number of villages grew almost sevenfold from two to 13. Hamlets more than doubled from 16 to 33; farms increased by a factor of over three (4 to 13); and field sites increased from 9 to 26. Hayden (2004a, 94) reports settlement expansion both along the coast and inland. The Istron River valley was densely occupied by a string of four hamlets and four villages, three of which were new. The distribution of new sites near good arable land suggests that the exploitation of the land was a factor in new settlement location. Inland settlement consisted largely of small dispersed sites located near water sources (Hayden 2004a, 98). Hayden also noted a trend toward difficult-of-access or fortified sites (e.g., the peak of Vrokastro [VK1]) in this period.

Several of these sites (KPh5, VK7, TM7) possessed a cyclopean structure at their center. Site APh3 produced cyclopean walls, high-quality pottery, and a “halo” of surrounding hamlets (APh1, APh11) and farms (APh2, APh4/6). Near the coast, site APh1 has well-built, hammer-dressed walls. On a ridge overlooking the Gournia Valley to the east, site KM2 was a large settlement with cyclopean walls in the Protopalatial period. Hayden interpreted these as residences of a rural elite. Protopalatial settlement formed clusters, as in the Gournia and Kavousi areas. In the mouth of the Istron River, the large site of Priniatikos Pyrgos (PP1) was surrounded by large (IS2, KA1) and small (IS1, KP1, KA2, GN2A1–2, GN2B) sites. To the northwest, a village (IM2), five hamlets (IM1, IM5, IM9, IM11, IM12), three farms (IM3, IM8, IM10), and three field sites (IM6, IM13, IM14) form another cluster. Inland, the dense pattern of settlement and rugged mountainous terrain make it difficult to distinguish other clusters, although the sites around APh3 and the three adjacent villagesized sites, TM1–TM3, might qualify.

Regional Settlement Across the Mirabello region, the number of settlements rose sharply in the Protopalatial period. Virtually all Late Prepalatial sites continued into this period. Smaller sites especially multiplied. Nevertheless, if the estimated population of each site size category is taken into consideration, the greatest expansion took place at the largest settlements. This pattern certainly suggests a regional demographic increase. Hayden, for example, calls the Protopalatial growth in the Vrokastro area a “population explosion” (Hayden 2004a, 93, 97–98, 115).

More diverse environments were settled and exploited. Sites ranged from those high up in the hills to sites on the valley floor. In the Vrokastro and Kavousi areas, there was a tremendous expansion of settlements into the hills. Site location includes a diversity of soil types. In the Kavousi area, for example, sites K15 and K16 were on marl slopes; sites K45, K46, K57, K61, K63, and K65 were on red soils (in the Kavousi Plain); and sites K95, K96, K109, K112, and K120 were on limestone slopes (in the south). In the Gournia area, site

MIDDLE MINOAN IB–II PERIODS: GROWTH OF REGIONAL FACTIONS AND CONFLICT

62 sits on sheer scree from Mount Thriphti. This site distribution points to an intensification and diversification of agriculture. In a break with earlier practice, sites were often located a considerable distance from a water source, which implies a greater reliance on dry farming. In the Gournia area, sites also produced a larger combination of agricultural artifacts: pithoi and large jars (storage), basins (food storage and preparation, milking?), cooking ware (cooking), scored basins (beehives from 44, 60, and 67), and a clay vat/slab (wine press from 100). Settlement expansion in all three survey areas continued to be concentrated in site clusters. In several cases, sites were clustered around a single large settlement (e.g., Gournia, 60, K44, PP1). In the Kavousi area, Haggis (2005, 73) sees the Protopalatial organization of five clusters of sites, each controlled by one social group that is responsible for local cooperative activities. Haggis suggests that these central sites were the residences of a community leader of some sort. Defensively located or fortified sites were also a feature of the Protopalatial settlement pattern—for example, Vrokastro (VK1), Phanourios (APh3), KPh5, PN1, PV1, Gournia (8), Vasiliki (135), sites 120 and 131, and Vronda (K77). Site 120, for example, is located high on the slope of Thriphti below several springs. The main building is based on a terrace formed by a large cyclopean wall (Pl. 6C). Settlement in all three survey areas was concentrated in tightly formed site clusters, some of which were old (pre–EM II) foundations, and others that were probably settled by Cycladic (and perhaps other) immigrants in EM III. At Gournia, there may have been an ethnic mixture, with the older population predominating. The construction of forts (such as Skourdeli tou Metochi, located about 1 km south of Vasiliki village) and fortified or defensively situated sites point to uneasy relations between these groups.

Excavated Settlements Local excavations provide details about the development of major sites in the region. Gournia achieved its maximum size in the Protopalatial period, reaching at least 4 ha (Map 22). Settlement

45

seems to have extended from the excavated acropolis, with at least one house near the coast, along most of the Pera Alatzomouri Ridge to the east, to the adjoining hill on the west bank of the Gournia River. Two new houses (Aa and Ek) and house tombs (I, II, and IV) were built at the northern end of the acropolis. Middle Minoan I–III pottery and human bones found by the survey near the cement guardhouse on the western slope of the acropolis indicate the existence of other tombs. Protopalatial remains on the acropolis ridge include Houses Aa and Ek, as well as deposits under Houses Ff and El on the eastern slope, and under the central court (Soles 1979). Boyd dug, but did not publish, Protopalatial house remains that are still visible today, near the coast and on the north hill of Pera Alatzomouri. At the top of the acropolis, a cyclopean tower located at, but not bonded with, the northeast corner of the Neopalatial palace (Soles 1991, 68–70) may be a Protopalatial construction. This date is suggested by the fact that such towers are typical of the Protopalatial Cretan landscape (e.g., at Skourdeli tou Metochi, a kilometer south of Vasiliki, and on MM I–II sites like Myrtos/Pyrgos). These towers do not often occur later, or as a part of the Neopalatial palaces. The acropolis may not have been fortified, since the wall Boyd (Hawes et al. 1908, 20) identified as “a town wall of the Middle Minoan Period” is a terrace. At the end of the Protopalatial period the town was extensively rebuilt, the result of a destruction in MM IIB, as happened at many Cretan sites. On the island of Pseira, a town known primarily through scattered house remains has been estimated to have consisted of 30 to 40 houses (Betancourt, Davaras, and Hope Simpson, eds., 2005, 286). At Myrtos/Pyrgos, the MM IB–II settlement grew dramatically. At the top of the settlement were a tower, a cistern, and a large elite building (or buildings) that was the predecessor of the Neopalatial villa (Cadogan 1977–1978, 74). The tower and cistern point to defensive concerns. Vasiliki continued to be a prosperous and large settlement: at least four MM II houses are known from the site (Zois 1992, 279). At Mochlos, Seager (1909, 277, 283, 288, 290, 302), mentions “MM I” (i.e., Protopalatial) remains west (Blocks A and C) and north of the

46

L. VANCE WATROUS AND MARYANNE SCHULTZ

Church of Hagios Nikolaos and on the southeastern slope. Recent excavations have also revealed Protopalatial houses and scattered deposits beneath a LM I plateia and below Houses D and C.3 (Soles and Davaras 1992, 422–423, 426–428; 1996, 180–184; Soles pers. comm.). Excavators (Seager 1912; Soles and Davaras 2000, 28) at Mochlos have reported recovering relatively few finds from the Protopalatial period. Apparently the settlement shrank in size, in contrast to Gournia. Many of the tombs in the cemetery are said to have gone out of use in the Protopalatial period (Soles 1992), although six of the Mochlos tombs produced MM IB–II material, including

pottery, stone vases, daggers, ivory seals, and a Syrian cylinder seal (Davaras and Soles 1995; Watrous 2005a). Seager, however, noted later burials at Mochlos around the edge of the Early Minoan cemetery (Seager 1912, 87–88); he published only a few MM III–LM I vases from these burials, but it is possible that MM IB–II pithos and jar burials were also made at this location. Excavation at the cliff-side site of Katalimata (Nowicki 2008) has shown that this extraordinary refuge site was briefly reoccupied at the end (MM IIB) of the Protopalatial period, evidence that the last part of the period witnessed a displacement of population.

Subsistence, Industry, and Exchange Protopalatial sites in the Mirabello region were situated near good-quality arable land for reasons of subsistence (Haggis 2005, 71). Intensification of land use in marginal hillside areas probably signals a wider variety of crop use (sheep/goat forage, honey, grapes, olives, grains), given the different types of environments involved. Newly settled sites along the valley floor probably imply an emphasis on dry farming. What is new in this period is the intensity and diversity of land use—mainly the use of the dry valley floor and elevated slopes. As Hayden (2004a, 98) suggests for the Vrokastro area in this period, more of the carrying capacity of the region has been brought into play. This seems especially true in the light of new settlement (e.g., 12, 13, 27, 61–66) on marginal (waterless or thin-soiled) upland areas. A number of these new sites were small—field houses and farms, also a sign of a more intense use of rural land. Their recovered assemblage in this period invariably consists of cookware, one or more heavy basins and jars, pithoi, amphorae, and cups. Larger vases indicate the storage of produce (e.g., wine, honey, milk/cheese, and grain). This evidence is consistent with the Pseira Survey, which showed that the land there was used intensively in the Protopalatial period (Betancourt, Davaras, and Hope Simpson, eds., 2005, 286–290). Off-site data (Map 23) demonstrate the striking intensity of land use in this period. The halos of sherds around sites clearly show that the land was

intensively used by each settlement. Coastal settlements around Gournia relied heavily on the lower slopes to the south within a distance of 1 km. Settlers on the Thriphti slope north and south of Monastiraki farmed large expanses of the valley floor. Given the precision of our identifications (Map 24), we can discern at least three sub-regional patterns of land usage. In the area around Monastiraki, by far the commonest vessel found (by a ratio of 1:10) in the off-site fields was the cooking pot, used, one suspects, by farmers to prepare meals when they spent the day in the field. Next are pithoi and jars, used to catch and store water. On the slopes of the northeast sector of our survey zone, the off-site distribution pattern differs somewhat. Cookpots are less numerous and only outnumber the next most popular vessel, the basin, by 2:1. Basins were used more frequently in this area (and on the slopes south of Gournia) than around Monastiraki, possibly for the collection and pressing of grapes. Amphorae and jars may be part of this pattern too. Yet another pattern of off-site vessels occurs around Gournia. Fineware sherds are the most numerous, followed by cookware; both might be the result of manuring. The Mochlos tombs produced evidence for local industry, including clay and stone vases (clay: Seager 1912, figs. 13, 28, XI:11, XI:13, 50, 84; stone: Seager 1912, fig. 28, XI:12; Soles 1992, 62, MM I–II), bronze items (probably the daggers from Seager 1912, figs. XX:9, XXI:22), and ivory seals (e.g., Seager 1912, II:41, II:42). The tombs at

MIDDLE MINOAN IB–II PERIODS: GROWTH OF REGIONAL FACTIONS AND CONFLICT

Mochlos contained some 15 (probably) local seals of the MM I–II period, of ivory, marble, and steatite (Platon, ed., 1969, 471–478; Platon, Pini, and Salies, eds., 1977, 249–255). The Gournia tombs of this period also yielded local goods, such as bronze daggers and tweezers, stone vases, stone seals, Kamares Ware (including kantharoi imitating the well-known silver example), and perhaps ivory inlays (Soles 1992, 8–17, 40). No evidence of a dominant or centralized source of pottery production existed, as in EM IIB (Vasiliki Ware). The fine ware from the excavated site of Myrtos/Pyrgos on the southern coast has been shown to be locally made (Knappett 1999). In contrast to the situation at Malia, pottery production of kitchen wares at Myrtos/Pyrgos was non-centralized—that is, characterized by diversity in fabric, potting, and finishing technique, suggesting small and dispersed workshops. Fine wares found at Myrtos/Pyrgos were produced with high levels of skill, standardization, and labor investment (Knappett 1999, 631), suggesting potters were attached craftsmen working for an elite patron. These communities were literate and had a local accounting system, since at Myrtos/Pyrgos a number of amphora handles and two four-sided seals bore hieroglyphic signs (Weingarten 1990, 105; Knappett 1999, 633). With the abandonment of the Early Minoan metallurgical center and emporium at Chrysokamino in this period, the local communities within our

47

region must have obtained copper through some other means. The sudden appearance of MM IB–II pottery in the Cyclades (Papaiannopoulou 1991) suggests that this Aegean exchange had changed hands, and that the Cretans had begun to play a more assertive role in this trade. Coastal communities in our region would have become competitors in this venture. A MM IIB shipwreck has been recently discovered off the south coast of Pseira (Hadjidaki and Betancourt 2006; Betancourt and Hadjidaki 2009). The ship carried many amphorae and jugs (for wine?) of local manufacture, suggesting it was part of the regional maritime network. The people of the region exported a considerable amount of Mirabello Ware vessels—oval-mouth amphorae, pithoi, and pithoid jars—to Quartier Mu and Myrtos/Pyrgos (Knappett 1999, 630). Amphorae from the Mesara were traded to Myrtos/Pyrgos (Knappett 1999, 632 n. 98). Stylistically, the fine wares at Malia and Myrtos/Pyrgos are very similar; Knappett argues that the elite family at Myrtos/ Pyrgos was imitating palatial elites at Malia. Mirabello Ware vessels have also been found at Mochlos, Pseira, and Petras, and in Lasithi (Hayden 2004a, 99; personal observation). Overseas connections with Anatolia and Syria can be seen in the silver kantharos from Gournia (Davis 1979) and the cylinder seal dated to 1850–1720 B.C. from Tomb L at Mochlos (Davaras 1973a).

Ideology Changing burial practices reveal new social and political ideologies. At Gournia, two house tombs (I and II) were built in the MM I period. Tomb I contained what was apparently a reburial of skulls and EM II pottery in a pit with MM I stone vases and the well-known silver kantharos. Tomb II contained stone vases, a seal, and bronze tweezers, and had a kernos slab for offerings set on an altar-like base attached to its exterior (Soles 1992, 3–28). Soles (1992, 23) has pointed out that the earlier skulls found in Tomb I were purposely saved, suggesting a reverence for ancestors. This development is paralleled at Myrtos/Pyrgos in MM I–II, where a large rectangular house tomb with two attached ossuaries built on the top of Pyrgos hill housed elite

male burials. At Myrtos/Pyrgos, the elite ossuary (Cadogan 1977–1978, 73, fig. 5) contained about 65 males. Spread over the MM I–LM IA period, perhaps 500 years, this implies that three to four males were buried per generation, a number that probably includes more than a single nuclear family. The tomb was reached by a long paved processional road, ending in a kernos by the tomb, which implies that one social group had reached special status manifested in public cultic ceremony (Cadogan 2006). At Mochlos, the same phenomenon is evident in the ramp and altar in front of Tomb IV/V/VI. Seager (1912, 37–38) reported finding MM I and MM III pottery in Tombs I/II/III and IV/V/VI: “It is not easy

48

L. VANCE WATROUS AND MARYANNE SCHULTZ

to understand the presence of M.M. I and M.M. III objects in and near these big chamber tombs, as they are seldom associated with human remains. They might be taken as evidence of a species of ancestor worship practiced by the later inhabitants of Mochlos at the graves of their forebears.” Unlike the royal Chrysolakkos tomb, the funerary monuments in the Mirabello region are smaller and found at several settlements, implying that power in the region had not yet been centralized. Nevertheless, they seem to have served the same

function, that is, to justify and express elite status of a social group. In this respect, these Mirabello groups were only following the expansionist behavior of elites in Central Crete. By the Protopalatial period, the sanctuary and cult offerings in the Dictaean Cave indicate that Malia had extended its territory to include the upland plain of Lasithi (Watrous 1996, 49–52). In MM IIIA, Knossos had made inroads into the Pediada region by establishing a palace at Galatas (Rethemiotakis 1999), and a royal residence at Hagia Triada (La Rosa 1997).

Social Organization Burials at Gournia, Myrtos/Pyrgos, and Mochlos indicate these communities were organized around one, or perhaps two, paramount social groups. Contemporary with these elite tombs, simple individual jar or pithos burials appear across the region, at sites 45 and 63, Sphoungaras (10), Pacheia Ammos (18), and at Mochlos. In the Protopalatial period, 12 sites in the Gournia area are reported as having burials (10, 17, and 18 in association with 20, 44, 45, 49, 57, 63, 83–84, 97, and 98), while 7 others (16, 24, 39, 54, 60, 65, 143) have evidence of possible interments. These burials from rural farms and hamlets suggest the formation and growth of a lower class closely linked to the rural land. Soles (1992, 115) came to the same conclusion about the new types of Protopalatial burials, remarking that “the choice between one form of burial and the other (house tombs vs. pithos and larnax burials) reflects a division in the social make up of the population.” Protopalatial society in the Mirabello region seems to have been divided into two groups: elite families and commoners. At each center, the elite apparently consisted of one or more families. This group of one or more families, then, would have wielded political power within their communities as a corporate body (Blanton et al. 1996). A new lower class is also visible archaeologically in the proliferation of small farms and field sites in marginal environments and in the masses of simple jar burials that begin to appear in rural areas.

The material culture of this period shows the same division. Elite objects, mainly from tombs, are more clearly distinguishable than in earlier periods. At palatial sites, such as Knossos, Phaistos, and Malia, unique and ostentatious goods, such as the well-known leopard-shaped ax and the gold-hilted swords from Malia, were elite, probably royal, objects. Similar examples occur in the Mirabello region, but at a reduced, second-level scale—for instance, the silver kantharos from Gournia, the Mochlos silver cup, and Kamares Ware vases (Cadogan 1977–1978, figs. 12, 13; Knappett 1999, figs. 11–14). In our region, the highest elite items were not of palatial quality and, significantly, were not concentrated at a single site, but were found at a number of major settlements such as Mochlos, Gournia, and Myrtos/Pyrgos. Craftsmen, perhaps attached to elite patrons from these centers, also produced gold-foil jewelry, necklaces of semiprecious imported and local stones (carnelian, amethyst, rock crystal, marble, and steatite), ivory seals, and many stone vases. No single site, or combination of sites, seems to dominate the Mirabello region during the Protopalatial period (as Mochlos and Vasiliki did in EM II). Instead, settlement took the form of major site constellations. Some central settlements possessed a single great house built in cyclopean masonry that seems to represent a leading family. Though unexcavated, site 60 is a good example: a villagesized community of 3.4 ha with at least 10 rubble

MIDDLE MINOAN IB–II PERIODS: GROWTH OF REGIONAL FACTIONS AND CONFLICT

houses clustered around a single cyclopean structure. Impressive cyclopean structures were found at several smaller sites in this period. A few of these secondary-type sites with heavy cyclopean walling are located in strategic locations, for example, sites 137, K44, APh1, APh3, and KM2. These site clusters fall into two types: old communities that can be traced back to FN/EM I–EM II, and newer ones whose foundation dates to the EM III–MM IA period (Map 21). In one case, there seems to be a mixture of populations. In the Gournia Valley, where there are two large settlements— Gournia and Pera Alatzomouri, communities going back to EM I–II were joined in EM III–MM IA by site 1 on the opposite side of the valley. It seems unlikely that Gournia would have allowed a community the size of site 1 to begin using its small valley on terms of equality. What this suggests is that these two sites may have formed a client-patron relationship, which, if true, would have increased the power of the Gournia elites. These newly arrived populations, visible in the dense settlement pattern, lived literally (on the poorest land) as well as economically at the margins of society, forming a large lower class under conditions of economic circumscription (Carneiro 1981). A close historical parallel for such a situation existed in Greece after the Asia Minor Disaster in 1922. With little more than the clothes on their backs, refugees from Asia Minor arrived in Greece where they formed large settlements located in marginal urban and rural areas. On Crete, one can see them in lower-class suburbs, such as Nea Halicarnassis on the edge of Herakleion, in poor villages such as Skinias (Tents) in the undesirable eastern Mesara, or Filissia located in the Pediada, and especially in the slum of the “temporary” Gypsy camp near the Herakleion airport. Such conditions of economic circumscription served to create a large lower class, described by Stratis Myrivilis in The Mermaid Madonna (1959), which destabilized Greek society and resulted in the dictatorship of General Ioannis Metaxas in 1936. Within their circumscribed areas, different Protopalatial social groups lived uneasily, side by side with one another. The groups that expanded the most during this period were the newer

49

communities (Map 21, gray circles). This process is a natural one, visible today, where immigrant groups encourage high birth rates for economic and social support. Mirabello society must have been marked by a certain amount of competition and antagonism among these social groups, similar to the factional competition between social groups described by Brumfiel and Fox (1994). Under similar economic conditions, Greek villages today and in the last century compete with one another, often in an atmosphere of mistrust and hostility. Each village pursues its own economic interest within the existing local socioeconomic hierarchy. Two local examples illustrate this process. During the 1950s, the villages of Pacheia Ammos and Episkopi (Map 1) came into conflict over who would control the water from the Cha Gorge that they used to irrigate their fields. And, beginning in the 1960s, deep wells and thermokipia (hothouses) transformed Ierapetra into a tremendously wealthy town—so much so that in the 1990s it assumed control of the regional political structure, hitherto composed of independent village koinotita, and attempted to construct a giant airport virtually on top of the village of Vasiliki. The Mirabello region is a small area severely circumscribed by mountains, and in such a situation, social and political transformations often come about in the form of a competitive zero-sum game. Some version of this process seems to have taken place in the Mirabello region toward the end of the Protopalatial period. Pseira, Myrtos/Pyrgos, Vasiliki (House Gamma), and perhaps Priniatikos Pyrgos were destroyed by fire in MM IIB. At Mochlos, Seager reported masses of MM III/LM IA pottery “found in all parts of the site” under the LM I floors, which came from the initial rebuilding of the Neopalatial town or a destruction at that time (Seager 1909, 284, 290). Pseira was briefly abandoned after its destruction at the end of MM IIB and was resettled along different lines at the very same time that site 1 was abandoned. It may be no coincidence that burials continue to be made on site 1 in the Neopalatial period, at the time that the old Pseira cemetery went out of use. The sequence of abandonments and destructions at the end of the Protopalatial period led to a retreat to

50

L. VANCE WATROUS AND MARYANNE SCHULTZ

defensive sites such as the fortified refuge site of Katalimata (Nowicki 2001, 2008). In MM IIIA, Mirabello Ware vessels, the chief export of the Vrokastro area, were abruptly replaced within the Isthmus by ceramic production from the vicinity (eastern isthmus) of Gournia. The town produced a large number (14) of weapons—daggers, swords,

and a spearhead—spanning the MM II–LM I period (Hawes et al. 1908, pl. IV). These conflicts are probably one major cause for the drop in population visible on sites in our region in the following period. As a result of the events described above, Gournia seems to have gained control of the Mirabello region early in the Neopalatial period.

7

Middle Minoan III–Late Minoan I Periods: The Rise of a Regional State L. Vance Watrous and Maryanne Schultz

The first section of this chapter begins by presenting the Neopalatial data from the Gournia Survey, which is followed by a summary of the Kavousi and Vrokastro survey results. In the next section, the chapter gives an overview of the Neopalatial settlement pattern in the Mirabello region. The third

section includes information from excavations at Gournia, Vasiliki, Pseira, Mochlos, and Myrtos/ Pyrgos. The final section examines certain aspects of Mirabello society, including craft production, trade, and social organization.

Gournia Survey Neopalatial settlement (Table 5; Map 25) in our survey zone consisted of 72 settlements (6 villages, 22 hamlets, 21 farms, and 23 field sites), and 4 burials (including 4 possible burials). Compared with the previous Protopalatial period, when there were 76 settlement sites (13 villages, 29 hamlets, 16 farms, and 18 field sites), and 7 burials (including 3 possible burials), Neopalatial settlement dropped by about 1%. Any impression of continuity, or even simple decline, between the two periods based on site totals is misleading: 18 Protopalatial sites, 2 of which were villages, did not continue after MM II. Moreover, 15 new sites were settled in the Neopalatial period. The

number of hamlets dropped by two, farms remained the same, and the number of field sites increased by five. Most of the Protopalatial sites that did not carry over were larger-category sites (two villages: 1 and 17; eight hamlets: 7, 22, 28, 33, 95, 105, 144, and 151; six farms: 2, 26, 44, 57, 96, and 131; and two field sites: 63 and 98), whereas newly settled sites were small (four hamlets: 15, 47, 81, and 94; seven farms: 48, 50, 58, 59, 85, 121, and 129; and four field sites: 35, 53, 56, and 92). Protopalatial cemeteries (10, 20, 24, and 83) continued to receive burials in this period. Four other potential burials (16, 39, 49, and 97) were recorded.

52

L. VANCE WATROUS AND MARYANNE SCHULTZ

Along the Mirabello coast, Gournia (8) clearly dominated the area, as neighboring sites (1, 2, 7) disappeared in the Neopalatial period. The hamlet PhR1 on the coast, at the western edge of the survey zone, was also abandoned in this period. The large settlement 1 was deserted in MM IIIA, but later larnax fragments found on the site indicate that it was reused as a cemetery, suggesting their inhabitants had moved somewhere in the region— perhaps to Pseira, which was resettled at this time—but returned to the site to bury their dead. To the east of Gournia, the coastal hamlet 22 at Halepa was no longer inhabited. Farther south, along the river toward Vasiliki, occupation of the farm 26 and hamlet 28 ceased, as did occupation of the hamlet 33 north of Vasiliki. Only one new site was founded in the Gournia area: hamlet 15 on the Phylakeion ridge southwest of Pacheia Ammos. It is noticeable that Neopalatial sites produced less pottery and a reduced range of shapes (cookware, “coarse” ware, and cups), as compared with the Protopalatial period. Site 6, however, yielded a spouted tub for wine pressing and many cups. The cluster of sites around Plakoures (60) expanded from 15 Protopalatial to 20 Neopalatial sites. Thirteen sites of the Protopalatial cluster continued to be occupied. Two sites directly adjacent to Plakoures, 44 and 57, were abandoned. Significantly, Plakoures itself seems to have shrunk in size, as the eastern and southern edges of the site produced little or no Neopalatial sherds. The cluster expanded to the north with the establishment of eight new sites: hamlet 47; farms 48, 50, 58, and 59; and field sites 35, 53, and 56. Only one site, 52, decreased in size from a farm to a field site. Sites 35, 58, and 59 were located at the edge of the plain on low rising slopes, although sites 56 and 55 were on the floor of the plain. These changes suggest an intensification of land use. The string of sites located on the dry valley floor (54, 58, 59, 61, 62, 64, 65) may have been using water channeled from the Cha Gorge to irrigate crops. One might explain this type of site expansion into marginal areas as a natural ecological response to increased population, but since the Neopalatial population in our area seems to have dropped, another cause must be sought. These sites yielded artifactual evidence for their common agricultural function, for example, storage jars, cookware, basins, and amphorae. Wasters from

the hamlet 47 point to pottery production. Sites 35, 47, and 58 yielded evidence (spouted jars or vats) of wine production. Additionally, sites 47 and 65 produced beehives; site 60 had loomweights. Larnakes were found at sites 47 and 48. The Plakoures site cluster was surrounded by a halo of unsettled land to the north, west, and east. To the north, a kilometer of land empty of sites separates this cluster from another equally distinct cluster (Haggis 2005, fig. 12) of contemporary sites around Kavousi (K24). To the south, despite decreases in site size and the abandonment of a few sites, the site cluster south of Monastiraki may have expanded in size. Two villages sites (83, 114) shrank in size, suggesting a nucleation of population at the nearby four villages. Smaller sites 95, 96, and 98 were abandoned, replaced perhaps by the new nearby hamlet 94. Six new sites were founded: two were hamlets (81, 94) and the others were small farms (85, 121, 129) or field sites (92). On the south edge of the cluster, Protopalatial sites 96 and 105 disappeared, and sites 114, 118, and 120 shrank in size, but farms 121 and 129 were added. All sites produced a common basic ceramic assemblage—cookware, pithoi or jars; fewer had basins and amphorae. Sites 85 and 100 contain wine-pressing vats and site 84 had a beehive. Sites 90 and 97 yielded probable larnakes. Farther south, three of seven Protopalatial sites (hamlets 144 and 151, and farm 131) were deserted, widening the gap between the Monastiraki site cluster and the sites on the Ierapetra Plain to the south. This area around Episkopi, then, may have been a boundary zone between the territories of Gournia and Ierapetra. Sites produced the basic pottery assemblage. Site 121 had a possible larnax burial. Off-site data (Map 26) collected by the survey provides economic information. The heaviest concentrations of off-site materials surround the largest settlements. These halos of sherds around settlements imply that the inhabitants were visiting and using the land around their communities. Material around Gournia, the largest settlement, is also the most diverse. These off-site collections resemble domestic pottery assemblages in that they include multiple examples of fine ware, basins, jars, pithoi, and cookware. For this reason, this off-site pottery has been convincingly interpreted as the simple residue of domestic manuring practices (Bintliff and

MIDDLE MINOAN III–LATE MINOAN I PERIOD: THE RISE OF A REGIONAL STATE

Snodgrass 1988; Betancourt and Hope Simpson 1992; Ault 1994; Watrous and Blitzer 1995). On the other hand, the types of Neopalatial vases found in more rural off-site areas (Map 27) do not always correspond closely to the contemporary domestic assemblages. Fine-ware vases are found clustered around Gournia and sites 90 and 97, but the predominant off-site shapes elsewhere are actually different. They are limited to cooking pots and large storage jars (pithoi and pithoid jars), indicating manuring may not be the correct explanation. Other ethnographically documented agricultural activities besides manuring may be the answer. For example, traditional farmers are known to have stored water in pithoi in their fields and used water jars to irrigate plants. Large basins have been used as beehives, as water and food troughs for animals, as milking pans, and as containers for grapes. Farmers ate their

53

mid-morning and afternoon meals in the field rather than trekking back into the village. Food was carried out to the field in a container called the seferta; food may have been actually heated or cooked in the field. It seems likely that the rural off-site distribution in our area bears witness to some of these agricultural uses of the countryside. Neopalatial off-site data differ from the Protopalatial pattern. The drop in numbers likely reflects a less intense land use in this period, which might be expected with a population loss. The shift in recorded vessel types is more difficult to interpret. Cooking pots are still the predominant type, followed by pithoi and jars, but basins and fine ware are rare, although this pattern may be influenced to some extent by problems of identification following the disappearance of Mirabello Ware.

Kavousi Survey In the Kavousi area, Haggis (2005, 74–79) records changes similar to those in the Gournia zone: a decrease in the number of sites, population nucleation at some sites, and an expansion of new settlements on the plain (Haggis 2005, fig. 12). The 53 Protopalatial sites in the area were reduced to 40 in the Neopalatial period (Haggis 2005, 75). New “megalithic farmsteads” or rural houses (K28, K34, K69, K87) were founded on hilltops, and Haggis (2005, 75–77) suggests that they may be “country houses” with specific agricultural or economic functions. Small sites—field sites and farms—were deserted as new hamlets were settled. Site clusters lost some of their sites. A number of small Protopalatial sites on the plain (K15, K16, K17, K23, K27, K52) grew into evenly spaced, hamlet-sized settlements in this period. Haggis interprets this change as a new sign of systemic agricultural intensification of the landscape (Haggis 2005, 75). Productive upland areas, previously well settled, were largely abandoned, presumably for the coastal centers, a move that seems inexplicable in subsistence terms, or as a process that self-sufficient agriculturalists would voluntarily undertake. Sites are also located along north–south routes as if they were part of a larger economic

network. This new density of settlement in the plain cannot be related to population pressure, since the Protopalatial population was higher than in the Neopalatial period. Haggis remarks perceptively (2005, 79), “The regionally varied relationship between human settlement and the environment, local patterns of land use, and local-level social and economic interaction appear, in LM I, to have been superseded by a regional hierarchical structure.” The site cluster at Chordakia in the vicinity of Chrysokamino (K34) was reduced from 18 Protopalatial to 12 Neopalatial sites. It looks as if the population of this cluster was reduced and nucleated at the two hamlets. The cluster of sites around Kavousi went from 27 Protopalatial sites (1 village center [K24], 3 hamlets, 2 farms, and 21 field sites) to 23 Neopalatial sites (Kavousi [K24], the center, 6 hamlets, and 16 field sites) in this period. Farms were deserted for larger hamlets on the valley floor. Especially striking is the line of four hamlets (three of which were new in this period) on the plain north of Kavousi. Near the coast, Tholos (K1) and Hagios Antonios (K5) swell in size, and a number of new sites were established nearby on the poor soils of the plain. Haggis (2005, 77–78) believes that this new coastal cluster may be a

54

L. VANCE WATROUS AND MARYANNE SCHULTZ

response to the increased importance of maritime trade. According to Haggis (2005, 79), this settlement pattern was organized to intensively exploit

the landscape in a systematic way, and the impetus for this new regional hierarchical structure was the establishment of the LM I palaces.

Vrokastro Survey Overall settlement in the Vrokastro survey region decreased 40% during the MM III–LM I period (Hayden 2004a, fig. 18). The 88 Protopalatial sites (13 villages, 33 hamlets, 13 farms, 26 field sites, and 3 burials) were reduced to a total of 54 sites during the Neopalatial period: 52 habitation sites (9 villages, 16 hamlets, 6 farms, 21 field sites) and 3 burials. Hamlets were reduced (33 to 16), farms from 13 to 6, villages from 13 to 9, and field sites from 26 to 20. Based on site sizes, Hayden (2004a, 115) estimates that the Protopalatial population of 2,165 in the Vrokastro area dropped by about 800 to 1,267 in the Neopalatial period. Most of the Protopalatial settlements in the Vrokastro region continued to be inhabited during the Neopalatial period. Priniatikos Pyrgos (PP1) remained the dominant site in the region. The abundance and quality of the fine wares at Priniatikos Pyrgos (PP1) suggests its prominence in LM I (Hayden 2004a, 114). Of the 52 Neopalatial habitation sites, 9 were newly founded (1 hamlet, 1 farm, and 7 field sites). Although most MM I–LM I sites remained the same size, there were significant reductions in site size. Protopalatial villages PT1, VK6, and KA1 shrank drastically in size to farms or field sites. Four Protopalatial hamlets (APh1, IM1, IM5, OL12) became field sites. Site growth was smaller: the field site AG2 became a village, the field site PI5 expanded in size to a hamlet, and the field site SP2 increased to farm size. In the Ioannimiti promontory northwest of Priniatikos Pyrgos (PP1), the 12 Protopalatial sites (1 village, 4 hamlets, 3 farms, and 4 field sites) decreased to 5 sites (1 village, 2 hamlets, and 2 field sites) by the Neopalatial period. This represents a 58% drop in settlement on Ioannimiti, which may be partly accounted for by nucleation at site IM2 or at Priniatikos Pyrgos (PP1). According to Barbara Hayden (pers. comm.), the overall dimensions of Priniatikos Pyrgos are unknown, largely because of the heavy alluvium overlying it, especially to the south and east. Nevertheless, if the

Neopalatial foundation of the large Kambos site some 200 m east of PP1—which is less than the distance between the Gournia “acropolis” and the coastal portion of the town—is taken into consideration, then there may be considerable evidence for Neopalatial expansion at Priniatikos Pyrgos in the MM III–LM IA period. While Priniatikos Pyrgos remained the main site in the Istron River valley, settlement in the rest of the valley became sparse. In the Protopalatial period, 19 sites—3 villages, 7 hamlets, 2 farms, and 7 field sites—were fairly evenly spaced along the river valley. In the Neopalatial period, only two villages remained: Priniatikos Pyrgos at the coast and KK1 at the southernmost end of the Istron River valley. The other two Protopalatial villages (KA1, IS2) were almost completely deserted, and only a few hamlets, farms, and field sites were occupied, a 40% reduction in sites. The Vrokastro hinterland also suffered a loss of settlement. The 20 Protopalatial sites (5 villages, 8 hamlets, 1 farm, and 6 field sites) were reduced to 14 settlements in the Neopalatial period (2 villages, 3 hamlets, and 9 field sites). Larger settlements were abandoned, while the number of farms and field sites increased. Five (VK5, PS3, GT2, GT3, S03) of the 14 Neopalatial settlements were new. Several sites (GT2, GT3, and S03) were isolated structures. In the southernmost area of the survey zone, at Tzamachi and Schinavria Koriphi, only two Neopalatial villages remained of the three Protopalatial villages, along with five hamlets, one farm, and one field site, an 80% decrease in settlement. Several specific aspects of the Neopalatial settlement pattern recorded in the Vrokastro area seem significant. First, the overall size of the site reduction is quite large, much more so than in the other two survey zones. Second, with the exception of Priniatikos Pyrgos (PP1), the coast is virtually swept clean of sites. This change mirrors the same phenomenon around Gournia in this period. Third, site clusters are replaced in some instances by a

MIDDLE MINOAN III–LATE MINOAN I PERIOD: THE RISE OF A REGIONAL STATE

single small site (Hayden 2004a, 113). The majority of small sites—single farms or field houses—are new foundations in this period, which may indicate some form of reorganization or redistribution of land. As Hayden (2004a, 113–114) notes, “There is an increase in the number of small sites comprising one structure (probable farm or field house) in the landscape coupled with loss of some larger settlements and site clusters.” Hayden (2004a, 114) suggests that the foundation of these small sites may imply a reorganization of settlement for exploiting

55

the landscape. Fourth, Hayden (2004a, 116) discusses the possibility that one massive Neopalatial structure (SP2, and perhaps GN1:1 and KM2) may have been a fort or watch station located along a natural route, near water and good agricultural land in the Istron Valley. Examples of these have been described elsewhere in East Crete (Tzedakis et al. 1989), although a number of similar local structures (KPh5, TM7, VK7) were built earlier, in the Protopalatial period.

Regional Settlement All three surveys agree that the number of settlements decreased in the Mirabello Bay region during the Neopalatial period. The settlement pattern immediately around Gournia hints at one cause. Sites in the immediate vicinity of Gournia (1, 2, 17) were abandoned, leaving Gournia the sole settlement on the coast of the Isthmus. This demographic relocation gave the LM I settlement at Gournia direct access to three times more arable land than it had earlier. Archaeological studies of the settlement patterns elsewhere, as in the Valley of Mexico, have shown that as the local society became more complex, certain sites within the settlement hierarchy developed larger catchment areas, presumably in order to be able to raise the agricultural surplus necessary to support emergent elite nonproducers (e.g., craftsmen, bureaucrats, and warriors; see Brumfiel 1976; Steponaitis 1981). Perhaps the newly established palace at Gournia took similar steps to consolidate its power. Had the coastal reduction been simply nucleation at Gournia, we would expect Gournia to have grown at this time, but this does not happen (Map 28). The dramatic drop, which is limited to sites on the exposed north-facing Vrokastro–Gournia–Isthmus coast, may therefore also be related to the destructive effects of the Theran eruption. A more likely explanation, therefore, is that there was nucleation at Gournia in MM III, the results of which were largely wiped out by local conflict and the Theran eruption. Nevertheless, the overall reduction in regional site numbers and size suggests a drop in the local population. Hayden (2004a, 119) postulates local emigration, seismic activity, or the effects of

the Theran eruption as possible causes. Local warfare may also have reduced the population. As we have seen, the changes in the Neopalatial settlement pattern are complex, and this implies that there were several causes rather than one primary cause. For example, the reduction in sites had already started in MM IIB (at Pseira, Vasiliki, and Mochlos) and continued in MM IIIA (at Vrokastro, VK7, and 1). The discontinuity between earlier Protopalatial sites and new Neopalatial sites may suggest significant changes in landownership, with certain new groups replacing older groups. In the early Neopalatial period, a local palace was constructed at Gournia, and in LM IB, that palace may have come under Knossian control (see below). Hence, the local political sequence for the Neopalatial period may include as many as three different phases: late local Prepalatial (MM IIIA), local palace rule (MM IIIB–LM IA), and Knossian clientage (LM IB). Changes we observe in the Neopalatial settlement pattern may date predominantly to one of these phases, but the chronological coarseness of survey pottery prevents us from definitively sorting out each of these changes in their proper sequence. During MM III, the new settlement pattern “signals a significant sociopolitical restructuring of the landscape” (Haggis 2005, 75). In the Kavousi area, Protopalatial farmhouses were abandoned for large nucleated sites (K1, K5) in the plain near the coast. Other site clusters were abandoned for a large single building, a “megalithic farmstead,” strategically located on an inland route. These new farms, which Haggis calls the residences of “emergent rural elites” (Haggis 2005, 77), were also located

56

L. VANCE WATROUS AND MARYANNE SCHULTZ

so as to take advantage of well-defined (and recently abandoned) good land. The changes in the Gournia and Kavousi areas were complementary (see Table 6), indicating that the same phenomenon took place in both areas. Gournia and Kavousi were similar in that smaller sites increased and rural losses were relatively minor—the end result being an expanded use of the countryside around Gournia. The Vrokastro area, however, developed independently. The same overall development centered at Gournia took place there around Priniatikos Pyrgos. Recent excavations have uncovered a large paved area at the site dating to LM IA (B. Hayden, pers. comm.). Had the Vrokastro area been part of the new Gournia polity, we could expect that small sites would have increased so as to raise a surplus for Gournia, but this did not happen. Instead, a consolidation of sites around Priniatikos Pyrgos, parallel to what happened in the northern Isthmus, occurred. The pattern of the Vrokastro data suggests that the area was independent of the Isthmus in the Neopalatial period. A second piece of evidence tends to corroborate this suggestion. In the Neopalatial period, a peak sanctuary was established on the chapel hill of Ephendis Stavromenos about 10 km south of and overlooking the entire Meseleri Valley. Minoan peak sanctuaries are informative because their purview is believed to mark the political territory of the worshipping communities (Cherry 1986; Peatfield 1990). The Vrokastro area seems to have had its own peak sanctuary by the LM I period, which would indicate that it was an independent polity at that time. If the peak sanctuary located on the peak of Mount Thylakkas dates to the Bronze Age, as well as later, as claimed by Sakellarakis (1970), then another Minoan polity existed in the area of Kritsa and Lato. The discovery of a third peak sanctuary by K. Nowicki at Anatoli Stavromenos above Myrtos/ Pyrgos indicates that that site was the center of a polity in the Myrtos Valley. Inevitably, the question arises as to the southern border of the Gournia state. Did it extend to the south coast, or was it limited to a portion of the Isthmus? The southern half of the Isthmus is a selfcontained valley, with mountains along its north side that faces south to the coast of the Libyan Sea. Today the entire Isthmus is dominated by the city of Ierapetra. In the Classical–Hellenistic period,

Ierapetra (ancient Hierapytna) was the urban center of a powerful polis that controlled most of East Crete. During the Roman Empire, Hierapytna was one of the most prosperous cities in Crete. Hence, the temptation is to assume that a major Bronze Age settlement lies underneath the Roman–modern city. Surprisingly, there is virtually no evidence to support such a hypothesis. Extensive rescue excavations in Ierapetra by the Ephorate have been carried out over the years and have been reported in the Archaiologikon Deltion since the 1950s. No Minoan remains have ever been found, although this could be explained by the fact that Ephorate excavations stopped when they struck Roman levels. Hierapytna only enters the archaeological record in the 4th century B.C., when, according to Strabo (Geo. 10.4.475), the polis forced the residents of Larisa, probably Prophetes Elias (150), to join them on the coast. Despite this lack of Bronze Age evidence, there are two bits of indirect evidence that the Gournia state extended no farther south than the hill of Kazarma and Episkopi. The first is the massive tower at Skourdeli tou Metochi north of Episkopi. Looking northward, this tower is situated in a way that suggests that it is the northern border of the territory to the south. The second is that during the survey, it was noticed that the Minoan pottery found on sites south of Episkopi was distinct from north-coast pottery, that is, this pottery was not the product of north-coast production centers. In the Mirabello region, settlement of the rural land was reorganized. The farther the distance from Gournia, the greater these changes were. Single families were selected to care for the rural land, but the fact that the new cyclopean farms away from Gournia were situated in different locations from the Protopalatial sites suggests a certain discontinuity, namely, that the new LM IA farm owners around Kavousi and Vrokastro were not the same families that had controlled the Protopalatial clusters. If viewed through kinship lines, the changes in the Mirabello LM IA settlement pattern suggest the establishment of a new dendritic relationship headed by Gournia and Priniatikos Pyrgos. Land was intensively farmed to produce an agriculture surplus to feed the new elite, freeing them to man the military and navy, produce goods for exchange and export, and carry out administrative and ceremonial activities. This

MIDDLE MINOAN III–LATE MINOAN I PERIOD: THE RISE OF A REGIONAL STATE

was necessary because the harvest time for grain, pulses, and grapes (Blitzer 2004) falls at the same time as the sailing season (April to August). In the Neopalatial period, a peak sanctuary was established about 8 km east of Gournia on the mountaintop of Ephendis Stavromenos (Map 1), the highest peak (elevation 1,476 m) in East Crete and the highest visible from Gournia (Pl. 7A). The LM IA date of this small sanctuary is suggested by the fact that many of the LM fine-ware sherds (Pls. 7B, C) from the site are burnished. The actual site of the peak sanctuary has been almost completed destroyed by the construction of the Church of

57

Ephendis Stavromenos and a nearby cistern. Nevertheless, the presence of Minoan pottery, including tiny cup fragments and a possible leg from a figurine on this isolated peak, make the identification reasonably secure. From the top of this peak, the Minoan sites of Gournia, Pseira, and even Mochlos (Pl. 8A), which is hidden from the Isthmus by the mountain of Kapsas, are all clearly visible. In this case, the location of the Ephendis Stavromenos sanctuary implies the existence of a Neopalatial polity that included the northern Isthmus. The palace at Gournia, then, would have been its administrative center (Soles 1991).

Excavated Settlements Gournia was probably the main Neopalatial settlement in the northern Isthmus. The unexcavated Neopalatial settlement extended from the acropolis westward to the edge of the river, and eastward probably to the base of Pera Alatzomouri Ridge, where the Hill House was excavated (Map 28). Colluvium prevents us from determining the exact dimensions of the town on the east, but the cemetery of Sphoungaras-Alatzomouri provides indirect evidence, since it extends southward along the base of the Alatzomouri Ridge (Cooper 2002, fig. 29), almost as far as its southern end. This would imply that the east–west dimension of the settlement reached from the river eastward to the edge of the present asphalt road at the base of the Alatzomouri Ridge, a total of some 350 m. On the west side of the river, a patch of Neopalatial pottery next to (south of) the asphalt road may be the remains of a house. The town did not extend farther south than the Hill House, as the sherds around the Church of Hagia Pelagia on the hill to the south are entirely Roman. The upper town was densely settled to about 40 m north of the old cement guardhouse. Small clusters of Neopalatial sherds along both sides of the river may indicate that the settlement consisted of widely spaced single houses along its edges. Tombs placed against the steep slopes of the north edge of the acropolis were in use until the Neopalatial period, as indicated by the MM IIIA cup decorated with tortoise-shell ripple from Tomb I (Soles 1992, 11, fig. 4; cf. Seager 1912, fig. 31, XII:f, from a MM III context). These elite tombs go out of use at the same

time that the palace was first constructed. Apparently, traditional elite funerary customs came to an end with the establishment of the palace. North of the acropolis there was a short stretch of unoccupied land about 100 m in north–south extent (Boyd [Hawes et al. 1908, 20] likewise reports finding no traces of settlement in this small area). Farther north, on the eastern side of the river, stretching from roughly 50 m north of the asphalt road to the coast (approximately 250 m north–south and 150–200 m east–west), another area seems to have been given over largely to agricultural terraces. Several stretches of fortification wall lined the coast. A shipshed (“Shore House”) and attached workrooms were located on the coast. The cemetery of Sphoungaras at the western base of the Alatzomouri Ridge ran along its eastern edge. The survey found a pithos burial, a snake tube, and cups there. On the northern hill of the ridge top of Alatzomouri, where there had been a Protopalatial (and earlier) settlement, the light scatter of Neopalatial finds included coarse-ware sherds, fine-ware cups, and cooking ware. A loomweight, potter’s wheel, and waster from this area may be Protopalatial or Neopalatial in date. Thus, the north hill may have included a few residences and workshops. At the base of the north hill (where the dirt road crosses the ridge today), there is a small area (about 50 m north–south) called Skala, with almost no signs of ancient activity. Farther south along the ridge (recently planted in morelia), Neopalatial finds consist of coarse-ware sherds, a

58

L. VANCE WATROUS AND MARYANNE SCHULTZ

pithos (burial?), conical cup, and a snake tube. On the southern hill, called Amygdalokephalo, the survey found coarse, fine, and cooking ware. Two LM IIIA–IIIB chamber tombs (Pl. 8B), now destroyed, were visible there in 1992. Hence, the southern ridge top seems to have had pithos burials while residences may have been located on the southern hill. All in all, the inhabited area of Neopalatial Gournia was roughly 3 ha in size, so perhaps occupied by 300–900 people, probably some 500 individuals. There does not appear to be any area of Neopalatial Gournia that was not already settled in the Protopalatial period, whereas the reverse is not true. Two areas of Protopalatial settlement in particular, along the northwestern slope and northern edge of the acropolis ridge (south of the asphalt road), and on the hill immediately west and across the river from Gournia, were not occupied in the Neopalatial period. Additionally, the Protopalatial sherd density was much higher than that of Neopalatial pottery in all areas of the town. Why should the density of Neopalatial sherds be low relative to Protopalatial sherds? Certainly, at Gournia and throughout our survey zone, the quantities of Protopalatial coarse ware found (Mirabello Ware with granodiorite temper: Haggis 2005, 53, 169–170, Types II and VI) always exceeded recognized Neopalatial coarse wares by a ratio ranging from 1:2 to 1:4 (and occasionally even more). The relative lengths of the two chronological periods do not explain the difference, since the Neopalatial period is the longer one. A contributing factor may well be that, because Mirabello Ware is easily recognizable, and because in excavations at Mochlos this ware stops at MM IIIA, during the survey we took it as a hallmark of the Protopalatial period (K. Barnard, pers. comm.). But it seems likely that it continued in the Vrokastro area (J. Moody, pers. comm.) and probably at Gournia, with the result that we may have underestimated the Neopalatial presence somewhat. Additionally, the quality of LM IB pottery in East Crete is generally poorer than in LM IA—less burnished and decorated— which makes it harder to identify. Another possible explanation for the low density of Neopalatial finds has a parallel on the excavated acropolis of Gournia. Boyd found that certain parts of the town (Houses Ce, Cf, Cl, Ff, Fh, and Quarter D) lacked finds datable to the last

period of occupation, which convinced her that they had been abandoned before LM IB. With our present understanding of LM IA–IB pottery, we can now recognize a few pieces of LM IB that allow us to date some of these areas more accurately to LM IB (Soles 2002). This does not, however, explain the relative scarcity of finds in these houses noted by Boyd, and one wonders if the low density of LM I finds may not be connected with the socioeconomic status of the households in these areas, meaning that the residents were of a lower class whose material culture left less recognizable traces in the archaeological record. If this is true of the ceramic finds, one suspects it might also be true of the architecture. Boyd remarks on the poor quality of the house remains below (north of) the acropolis (Hawes et al. 1908, 20), and that is probably why she concentrated her efforts on the “acropolis” ridge where the houses were constructed of massive cyclopean and ashlar masonry. Parts of the outer sectors of Gournia, on the coast and on the Pera Alatzomouri Ridge, were built in less impressive rubble masonry. These parts of the town would have been inhabited by residents of lower status. This same pattern, a central cyclopean structure surrounded by a large outlying area of rubble remains, has been noted on several sites (e.g., 46 and 60). At the northern end of the settlement, on the coast, stood a large cyclopean building (Pls. 2B, 8C) with two galleries about 5 m in width and at least 16 m in length, which identify it as a shipshed, similar to examples at Kommos (Shaw 1986) and at Katsambas in Herakleion (A. Vasilakis, pers. comm.). The structure was designed to store seagoing ships belonging to Gournia. Subsequent fieldwork along the coast carried out by Watrous in 2008 and 2009 has revealed a system of fortification walls with towers crossing the entire promontory, behind which were agricultural terraces (Watrous 2010) In 1993, the geomorphologist Eberhard Zangger worked with the project to answer several fundamental questions about the area. To address the issue of seacoast change, Zangger examined the coast north of Gournia. He found that the pattern of water erosion visible on the sea cliffs indicated that the local tectonic uplift has been perhaps ca. 1 meter. As a result, he concluded that the coastline within the Gournia Valley has changed only slightly since antiquity. Zangger’s measurement of the

MIDDLE MINOAN III–LATE MINOAN I PERIOD: THE RISE OF A REGIONAL STATE

two small bays incised into the sea cliff, flanking the Gournia promontory on which the Minoan shipshed (Boyd’s “Shore House”) is located (Pl. 8C), showed that the two bays were identical in size, suggesting to him that they were man-made— that is, cut during the Minoan period to provide anchorage for ships. Based on the size of the drainage basin of the Gournia River, together with the assumption that the wooded mountains south of Gournia held more water in the Bronze Age than now, and the fact of the size of the boulders carried downstream, Zangger (pers. comm.) suggested that the river was perennial in the Bronze Age. If true, Gournia was situated next to a year-round water source. The survey project also found that the coastal cemetery of Sphoungaras, excavated by Hall (1912), actually continues southward for several hundred meters. In the Neopalatial period, simple pithos, larnax, and jar burials occur along the eastern edge of the Gournia settlement at the base and top of the Alatzomouri Ridge. In MM III (Hawes’s Town Period), the town (Map 29) was remodeled, perhaps following earthquake damage. Hawes’s excavation on the acropolis ridge of Gournia uncovered at least 47 Neopalatial houses built in the MM III period. Neopalatial houses were built using two masonry styles—either small boulders (Hawes et al. 1908, 21, fig. 5) or, for the first time at the site, in a cyclopean style (Hawes et al. 1908, 21, fig. 6). Hawes comments perceptively, “These (Cyclopean houses) belong to the latter half of the Town Period (i.e., LM I) when there was increased mechanical skill among the people, or some social change, that placed the labor of many at the bidding of a few” (Hawes et al. 1908, 21; emphasis my own). At the same time, several Gournia houses yielded evidence that its owners possessed weapons. Hawes (1908, 26 and pl. IV:49, 56, 57, 58) illustrates a bronze horde, including a long sword and daggers, from the so-called Hill House as well as a spearhead (from A8), and a short sword (from F14) from elsewhere in the town. At some point in MM III–early LM IA (Hawes et al. 1908, 24; Soles 1991, 26), a palace (Map 30) was built on top of the “acropolis” ridge. The appearance of the impressively built cyclopean houses and the construction of a palace were probably part of the same development, one that transformed Gournia. The palace, constructed of rubble and cyclopean

59

masonry, consisted of central and west courts, storerooms, and a north section that probably included a reception hall and residences (Soles 1991, 21–22; 2002), signaling that by LM IA, Gournia had become the political and administrative center for a state in the Mirabello region (Soles 1991, 73–76). Trade at this time reached the Cyclades, e.g., Akrotiri on Thera (Marinatos 1974, pls. 10 [in color], 78:b, 79:b–79:c; Kriga 2010). Given the early architectural plan (Soles 1991), it seems that this palace already had the main functions—storage, redistribution, and communal ceremony—of the later LM IB palace. At a later date, almost certainly after the Theran eruption in LM IB (Soles 1991, 21–31), the palace was remodeled: a facade in ashlar masonry (Pl. 9A), quarried at Mochlos, was added onto the west side of the building (and probably to the portico on the central court), the ashlar south wing was added, and the interior reception hall was given an ashlar face. The remodeling increased the size of the palace by one-third and added greatly to its monumentality. The ashlar masonry and masons’ marks, and architectural parallels with the palace at Knossos, leave little doubt that the remodeling was undertaken by Knossian architects and masons (Driessen 1989–1990). Soles’s careful 1991 article demonstrates how the palace at Gournia was modified to enhance its monumentality and to add the west and north porticoes for cult purposes. Surmounted by stone horns of consecration, the north portico with its bench-like stairs and altar-like slab, fit for bull sacrifice, faced out onto the central court. Given its width of 40 Minoan feet after these additions, the central court, at least in its present form, probably also dates to this period, but an earlier version of the court probably already existed; similar LM IA court areas at Myrtos/Pyrgos and Pseira would suggest as much. In form and location, the portico on the west side of the central court imitates other palaces (the “tripartite shrine” at Knossos, room V11 at Malia, and room 23 at Phaistos). It faces out onto the central court and eastward toward the mountain range of Thriphti. In 1991, Soles wrote prophetically that the portico “probably linked the palace in some way to a real peak sanctuary on a nearby mountain.” In 1995, we discovered a peak sanctuary on the top of Mount Ephendis Stavromenos, the highest point in East Crete. As at

60

L. VANCE WATROUS AND MARYANNE SCHULTZ

Knossos, the eastern side of the central court at Gournia was kept at ground level so that the mountain ridge and the appearance of the sun and moon were fully visible. As Goodison (2004) has shown for the Knossos Throne Room, the rising of the sun at certain times of the year was incorporated into the orientation of the palace, probably to play a part in some form of royal ceremony. We can imagine the same thing happened at Gournia. The cosmological aspect of the new palace seems to have been a Knossian ideological innovation. If this remodeling of the palace at Gournia took place at the beginning of LM IB, what form did the architectural predecessor of the LM I palace take? If we subtract the later remodeling, we are left with the north facade, the interior portion of the west facade wall, the storage wing (rooms 1–12), perhaps a Minoan suite in the area of rooms 20–21, and probably the north tower. The west and central courts also existed. It is noticeable that the large storage wing, with its three setbacks, faces out onto the west court, emphasizing the wealth of the structure. This storage wing can be directly linked to the agricultural intensification of the landscape that is seen in the local Neopalatial settlement pattern. Both these developments suggest that the architectural predecessor of the LM IB palace already had a redistributive function. That would imply the existence of literate officials. The storage capacity of the palace also implies that personnel with specialized functions, such as craftsmen and craftswomen, received rations. Pottery was produced and exported, as we know from the Gournia vases found at pre-eruption Thera. The same may have been true for textiles and metal goods. The LM IB palace at Gournia (Soles 1991) served several interrelated functions (Map 30). The many pithoi in rooms 3–9 and 23 identify them as storerooms for agricultural produce; smaller storerooms (8, 9, 11, 12, and 12a) are to the south. Room 2 contained stone and duplicate clay lamps, possibly made there; a bronze saw and ax came from the area of room 1. Soles (1991, 71) has suggested that room 29, with its bin and built drain, was used for the production of textiles. A stone kernos in room 18, the massive altar-like slab next to the main entrance, and the stone baetyl and kernos at the southwest corner of the palace look as if they were used in ritual ceremonies. Surmounted by stone horns of consecration (Soles 1991, 50, fig. 46), the

north portico, with its bench-like stairs, faced out onto the large central court, where ceremonies of some sort would have taken place. Inside the palace, the large central suite and ashlar-faced open court (rooms 20 and 21) look like a reception hall. Upper-floor rooms (restored by Soles 1991, 65, fig. 67) may have served a variety of purposes, including residences, a banquet hall, workrooms, and archives (the last indicated by clay sealings, two nodules, and a seal fallen from above into storerooms and the west court). Recent excavations at Mochlos have helped define the Neopalatial occupation sequence there. According to the excavators (Soles and Davaras 2000; Soles 2009), much of the Neopalatial town was rebuilt, beginning in MM IIIA, after a period of apparent decline. The MM IIIA architecture has been characterized as “modest” (Soles 2009, 10). Blocks of at least 17 houses, some three stories high, along the south shore of the island, have been uncovered at Mochlos. Three or four major streets ran north–south up the slope, intersected by a few east–west streets; one street may have crossed the Isthmus to the mainland (Soles and Davaras 2000, 29). Houses, including the large House D, were built in LM IA, damaged in the Theran destruction, and repaired in LM IB. In LM IB, several of the houses (C2 and C3) contained signs of Giali obsidian/stone vase manufacture and bronze caches (foundry deposits?). According to Seager (1909, 285), in LM IB the settlement had “passed its prime and a period of decadence had begun.” At that time, Building B.2, with its unusually large size, ashlar facade, elaborate storage spaces, and votive deposits, was built as the administrative and public center of the settlement. The building seems to have been to Mochlos what the LM IB villa at Myrtos/Pyrgos was to that settlement. After the Theran eruption, an industrial quarter consisting of at least two workshops, settled perhaps by Cycladic refugees, was built on the mainland (Soles 2003, 93–95). Although a few of the earlier tombs contained Neopalatial objects, the cemetery seems to have consisted largely of jar and pithos burials. Three of the tombs used in the Neopalatial period contained weapons: a sword (Tomb II), an arrowhead (Tomb XIX), and three spearheads (Tomb XX). After a destruction in MM IIB, and apparently a short gap in MM III (the area of House BNW

MIDDLE MINOAN III–LATE MINOAN I PERIOD: THE RISE OF A REGIONAL STATE

produced a MM IIIA bridge-spouted jar), the settlement at Pseira was rebuilt, but along different lines from the earlier site. The site grew and reached its maximum size in LM IB, consisting of perhaps as many as 60 houses (Betancourt and Davaras 2000). Signs of extensive trade contacts are evident during this period. The old cemetery also ceased to be used, and it is unknown where the dead were buried in this period (although burials of this period are found in abandoned site 1). In LM IB, a main house (Plateia House) faced onto a central open space (Floyd 1998). The fine ware for the settlement was made at Gournia, and it appears as if the site became a seaward extension and deepsea harbor for Gournia in this period (the coast at Gournia lacks a good harbor). Betancourt (2004) has shown how Knossian influences in wall painting and pottery became important in LM IB. Vasiliki is said to have shrunk in size but continued to be a prosperous settlement in LM I; the remains of five houses (M, N, K, S, 50) are known (Zois 1992). Houses M and N may have been deserted in LM IA. House K may have been the principal house in the LM IB settlement. At Myrtos/Pyrgos, following a MM IIIA destruction (Driessen and Macdonald 1997, 216), an ashlar villa (“country house”) was built, probably in LM IA. Following earthquake damage (from the

61

Theran eruption?), the building was repaired. In LM IB, the villa contained storerooms, living quarters, a shrine, two clay sealings, and two Linear A tablets, and looked out onto a paved court. During LM IA, the tomb for the local elite continued to receive burials but went out of use in LM IB, presumably as the result of Knossian control. All of the excavated settlements in the region produced evidence of rebuilding early in the Neopalatial period. Gournia, however, differs from the other sites during MM IB–II in that it was not destroyed (like Pseira and Myrtos/Pyrgos) or abandoned (like Pseira); nor did it shrink (like Mochlos and Vasiliki). Instead, Gournia grew during this period. According to Barbara Hayden (pers. comm.), in the MM II–LM IA period, Priniatikos Pyrgos may have been the industrial sector for the main Kambos settlement, located immediately to the east. Few signs of LM IB, however, have been uncovered at Priniatikos Pyrgos. The original excavator proposed that the site was a “shipping station” throughout this period (Hall 1914). More recently, this site has been called a “gateway community” (Rehak and Younger 1998) that functioned as a primary place for exchanged goods to enter and leave the area.

Exchange and Industry Regional exchange continued actively in the Neopalatial period. Coastal settlements conducted trade with one another and with the Cyclades, Cyprus, and the Near East. The Mirabello region received vases imported from the Levant, Cyprus, and Rhodes in LM I (Betancourt and Banou 1991). A Syrian spindle bottle (Hawes et al. 1908, pl. VIII:25), Cypriot copper, and Near Eastern tin came to Gournia in this period. In LM IA, Gournia exported pottery to Pseira, Mochlos, and Thera. Late Minoan IB foundry hordes, consisting of bronze tools, scrap, and copper ingots, have been found at Mochlos and at Gournia. Harriet Boyd’s excavation at Gournia provided much evidence that the town (Map 31) was the main industrial center for the north Isthmus. Household industries at Gournia included the production of

textiles, ceramics, and metal objects, such as cauldrons, axes, saws, chisels, daggers, knives, and sickles (Watrous and Heimroth 2011, fig. 18.4). Trade was especially important, as seen in the wide distribution of balance pans, chains, and weights in the settlement (Watrous and Heimroth 2011, fig. 18.6). In LM IB Gournia, 23 households produced signs of industrial activity (Watrous and Heimroth 2011). Four houses appear to be economically independent—that is, they possessed adequate foodstorage facilities to support their inhabitants. Hence, the economic pattern at LM IB Gournia seems to be that a substantial portion of the households were economically active, but the majority of these households did not appear to have been economically self-sufficient. These households may have manufactured items that were sold to locals in

62

L. VANCE WATROUS AND MARYANNE SCHULTZ

the Mirabello region, who paid for their purchases with food, or may have been supplied with rations from the palace. The Knossian remodeling of the Gournia palace in LM IB is part of an island-wide phenomenon that takes place at Galatas (MM IIIA), Hagia Triada (LM IA), and Zakros (LM IB), and surely suggests some form of Knossian control over much of Crete (Wiener 2007). At Zakros, Platon (1999) has demonstrated that during LM IA, industrial activities were distributed throughout the entire town, but by LM IB, elite industrial activity had been moved into the palace. The only high-quality goods found outside the palace during LM IB were found in three houses—House Z, the East House, and House A (the last identified as a customs house). Unlike at Zakros, industry at Gournia does not appear to have been so completely centralized, at least in a spatial sense. Elite goods were distributed throughout Gournia in both elite (Houses Ah, De, Ec, and G) and non-elite households (Houses Ad, Ba, Cf, Cg, Co, Da, Dd, Eg, Eh, Fg/Fh, and Fd), but it appears that they were being produced by, rather than used by, non-elite households (Schwartzott 1997; Watrous and Heimroth 2011). For example, elite metal items in the town, with the exception of those in Houses Dd and Ad, were found in areas of metalworking, and hence may not have been personal possessions. Platon suggests that House Z at Zakros, which contained elite goods, produced wine for the palace storerooms. Houses Ab, Ac, and Ad, located close to the Gournia palace, contained wine presses and a total of eight rhyta, and so they may have produced wine for the palace storerooms.

The seizure of regional power by Gournia is visible in economic terms. By LM I, Gournia had obtained control of the production and export of certain goods within the region. Thus, by LM IA, Gournia supplied most of Pseira’s pottery, and the same may also have been true for Mochlos (K. Bernard, pers. comm.). The ceramic producers of Gournia, at least 10 workshops in LM I, held a privileged position. The same would have been true for the craftsmen producing metal weapons, tools, and jewelry, and probably for the manufacture of stone vases. Mochlos shows the same hierarchical organization in LM IB. Like Gournia, it was connected to overseas ports (Barnard and Brogan 2003; Soles 2005). Built in ashlar masonry and larger than other houses, House B.2 seems to have been the cultic and political center of the community. House C.7 held workshops preparing perfumed oils, wine, and textiles (Koh 2006). In House C.2, there was a seal carver (Soles 2003, 96). After the Theran eruption, a quarter consisting of four workshops (Houses A, B, C, and D, the latter two unexcavated) was established, probably by Cycladic refugees, at the southern edge of the settlement. Houses A and B contained work areas for the production of stone vases and textiles. Two potters worked in and outside of House B. A bronze worker in House A produced bronze tools and other utilitarian objects (Soles 2003, 93–95). The organization of the houses indicates the occupants were full-time craft workers. Lack of storage facilities in the houses suggests that they were dependent on households on the island for raw materials (bronze) and food.

Social Organization The socioeconomic situation at Gournia and its territory in MM IIIB–LM IA is more difficult to understand than in LM IB. Floor deposits from the earlier period have rarely been found or are unpublished. Nevertheless, certain facts are obvious. The early Neopalatial urban architecture at Gournia does not seem radically different from the LM IB phase, with the exception of the remodeled palace and the addition of the unexcavated ashlar house

south of House Dd and the Hill House. Mochlos seems to have had the same architectural sequence: the basic plan and the houses of the settlement were established in the MM III–LM IA period. Following reconstruction after the Theran eruption, an elite civic structure, Building B.2, was erected in the settlement in LM IB. Pseira also follows suit: in LM IB, portions of the site were rebuilt, and the Plateia House apparently was the

MIDDLE MINOAN III–LATE MINOAN I PERIOD: THE RISE OF A REGIONAL STATE

“residence for the local administrative authority on the island” (Floyd 1998, 233). The above evidence indicates that Gournia and its polity seem to have come under Knossian control in LM IB. Two of the sealings from Gournia were stamped with two nearly identical gold rings depicting bulls leaping. Sealings from these same two rings have also been found at Hagia Triada, in the Mesara, and Sklavokampos, west of Knossos and Zakros. Scholars have recognized that these rings are Knossian and that they imply some form of administrative connection, if not control, by Knossos in LM IB (Betts 1967; Weingarten 1986; Hallager and Hallager 1995). On the basis of their large size, cyclopean masonry, and contents, six LM IB house structures at Gournia stand out as being elite. All six of these structures also have evidence of bronze or stone vase production, so these industries may have enjoyed elite status (Watrous and Heimroth 2011, fig. 18.8). The Hill House contained a horde of bronze vessels, fine vases, and an elegant long sword. Boyd (Hawes et al. 1908, 26) remarks, “Assuredly the Hill House was the home of a leader.” The head of the Hill House seems to have been a military officer, similar probably to the elite warriors depicted on the ships of the Theran Fleet Fresco, which are being paddled by lower-class enlisted men. Signs of trade at Gournia occurred

63

only in elite structures, indicating that trade and commerce also seem to have been associated with elite status. Structures whose storage assured them economic independence were all elite (with the exception of Houses Fe and Ad, which look as if they were used as storage areas). In contrast, nonelite households typically were not involved in industrial activities (except, occasionally, for ceramics and textiles), had houses constructed of rubble masonry, had inadequate storage for selfsufficiency (or none at all), and were not engaged in trade. In LM IB, Gournia seems to have included four rough socioeconomic tiers. These tiers include (1) the palace; (2) the industrial and self-sufficient houses grouped around the open north market; (3) the industrial houses that were not economically self-sufficient; and finally (4) the houses without any industry or storage. Tier 2 appears to have been involved in regional trade activities. Tier 3 structures were involved in stone vase production and metalworking, and were associated with smaller houses also involved in industries. It is possible that these tier 3 houses were also producing goods for the market area and the palace in exchange for sustenance. Individuals from tier 4 houses may therefore have formed a lower class, probably agricultural workers, portions of whose produce went to the palace.

8

Late Minoan IIIA–IIIB Periods: Depopulation and Mycenaean Occupation L. Vance Watrous

Following the widespread destruction of sites (Gournia, Mochlos, Pseira, and Vasiliki) at the end of the LM IB period, the number of sites in the Gournia survey area dropped precipitously. During the Neopalatial period, there had been 72 settlements (7 villages, 27 hamlets, 16 farms, and 22 field sites). In the LM IIIA–IIIB period, these numbers dropped to a total of 10 sites, with only 6 settlements: 1 village (8), 2 hamlets (123, 144), 1 farm (81), 2 field sites (84, 104), which represent an 86% decline in the number of settlements (Table 7; Map 32 ). Nonhabitation sites fell in number by 67%, with a 60% decrease in burials and the cessation of the Ephendis Stavromenos peak sanctuary. The coastal area from Kavousi to Pacheia Ammos was swept clean of settlement—only Gournia (8) continued to be inhabited. No new sites were settled in this period, except for site 144. No LM IIIA site exhibits enough growth to suggest a nucleation of population; the drop in site numbers must reflect severe depopulation. Aside from Gournia (8), all LM IIIA–IIIB sites were located inland, well away from the coast. Settlements during this period were smaller than their Neopalatial predecessors. Only the inland site

near Episkopi (123) grew, from the size of a farm to that of a hamlet. A few settlements in the southern portion of our survey area (Map 32) seem to have survived the LM IB catastrophe. Site 81 diminished from a hamlet in the Neopalatial period to a large farm in LM III. Previously a village (1.2 ha), site 84 contracted sharply in size (by 95%) to that of a field site (0.06 ha). Its continuation may be due to its location on a high defensible slope, next to a spring. The field site 104 was reduced in size (by 0.02 ha) in this period. The only new LM IIIA–IIIB settlement in the survey area was the hamlet-sized site 144 (Pl. 5C). Some 300 m southeast of Kato Chorio village, and near a spring, the site is situated on a steep, high hill whose top is protected by fortification walls. Off-site finds were limited to a cookpot found 500 m south of Gournia (8), a lone pithos at the extreme northeast corner of our survey area, a cookpot midway between sites 104 and 107, and a fine-ware sherd west of site 84. Site 123 was the only local settlement that grew in this period. A small farm in LM I, the site expanded into a large hamlet (125 x 65 m or 0.81 ha). Located in the middle of the Episkopi Basin, the site is not

66

L. VANCE WATROUS

visible from either the north or south coast. Late Minoan IIIA:1–IIIB/LM IIIC tombs associated with this settlement were excavated by Seager, Xanthoudides, and Platon in Episkopi. Kanta (1980, 146–160) describes the saved pottery, mostly stirrup jars, amphoroid kraters, and larnakes. Three stirrup jars and a hydria are Mycenaean imports (Kanta 1980, 154); six stirrup jars and a kylix are products of the Chania workshop. A well-known larnax (Kanta 1980, fig. 63) from Episkopi from this cemetery was painted by the same artist who decorated the larnax from Pacheia Ammos (Kanta 1980, fig. 56). A few vases indicate that the cemetery continued into LM IIIC. Settlement in the Kavousi zone follows the same depressing pattern (Haggis 2005, 79–81, fig. 13): a large drop in site numbers, reduced site size, settlement away from and hidden from the coast, and settlement near a water source. For the Vrokastro area, Hayden (2004a, 129–133, fig. 19) reports a 67% drop in LM IIIA–IIIB settlement. As in the Gournia and Kavousi areas, the coast was largely deserted; the sites that remained in LM IIIA–IIIB are found inland and in elevated and defensible locations (Hayden 2004a, fig. 19). Late Minoan I sites were reduced in size in the LM III period. No new settlements were founded in this period. Priniatikos Pyrgos, the Neopalatial center, produced only a few scattered LM IIIA–IIIB sherds. Gournia (8) was reinhabited after its destruction by fire (Hawes et al. 1908, 21) in LM IB. Reoccupation had begun at least as early as LM IIIA, as A. Evans (1921–1935, II, 139, fig. 70) illustrates a LM IIIA:1 stand from the site. Settlement at Gournia was located on the old acropolis and within newly constructed Houses He, Eh, Hf, and Ei around the edge of the settlement (Hawes et al. 1908, 26, plan). Of the LM III reoccupation, Boyd (Hawes et al. 1908, 23) remarks, “The greater part of our site was never reoccupied after the Town’s sudden demolition at the close of the first late Minoan period . . . In the Reoccupation period . . . only a few scattered dwellings on the old acropolis were in use.” A small bench shrine (Hawes et al. 1908, 47–48, pl. XI) containing statuettes of goddesses, figurines, and snake tubes (lamps or incense burners) was erected over an earlier street, on top of the acropolis north of the palace. In contrast to the town dwellings, two of the Mycenaean-period houses (He and Hf) were large ashlar structures.

One, House He, is a Mycenaean type “Corridor house” with a central megaron. This house contained a bull figurine, horns of consecration, transport as well as painted amphorae, and stirrup jars (Hawes et al. 1908, pl. X:8; Fotou 1993, 96; P. Pantou, pers. comm.). It looks as if LM IIIA–IIIB Gournia was occupied by a scatter of families living on the acropolis in reused houses under the control of Mycenaean officials resident in the outlying elite mansions He and Hf. Around Gournia, LM IIIA–IIIB burials were made in chamber tombs (Pl. 8B) at Pera Alatzomouri (10), in the Neopalatial House Ej, and near Pacheia Ammos (19). Boyd (Hawes et al. 1908, 20) claimed that two trenches dug west of the river produced evidence of LM IIIB occupation and a burial; the survey found no LM III pottery west of the river. One of the two chamber tombs (19) near Seager’s House dates to LM IIIA–IIIB, and the other dates to LM IIIB. Each tomb housed three larnakes. Grave goods in both chamber tombs hint at the occupants’ professions (the northeast tomb produced a bronze scale, beads, and gold rings, suggesting the burial of a merchant), and the southern tomb possessed a stirrup jar and jug, cups, and a seal stone (used for administrative purposes). It is possible that (but unclear if) the northeast tomb was the “Neolithic Cave” mentioned by Boyd (Hawes et al. 1908, 46). The Gournia area suffered a tremendous loss of population as a result of the Mycenaean destructions in LM IB. Population loss seems to have been the real cause, since rich agricultural areas, such as around Vasiliki and east of Episkopi, were not settled in this period. The settlements at Gournia, Pseira, and Mochlos were burnt down. Seager (1909, 301) records finding skeletons in many of the destroyed LM IB houses at Mochlos. At Mochlos there appears to have been a gap in occupation after the LM IB destruction of the town but before its reoccupation (Brogan, Smith, and Soles 2002; Soles and Davaras, eds., 2009). Boyd found no human remains at Gournia, so it appears that the inhabitants escaped. Vasiliki was deserted during LM IIIA–IIIB. The population drop evident in the LM IIIA–IIIB settlement seems to have been the result of warfare and probably emigration. By LM IIIA, Gournia was occupied by Mycenaean invaders. Other excavated sites, namely, Mochlos and Pseira, experienced the same fate. The

LATE MINOAN IIIA–IIIB PERIODS: DEPOPULATION AND MYCENAEAN OCCUPATION

evidence from Mochlos is especially valuable because it comes from recent excavations. After its destruction in LM IB, Mochlos (Davaras and Soles 2000) was reoccupied by at least six haphazard structures (Smith 2005, 187). By LM IIIB, the number of small houses had doubled. One house, however, as at Gournia, was distinguished by its large size (and the possession of an imported Canaanite amphora). In LM II, Knossian imports and pottery in Knossian LM II style appear at Mochlos (Smith 2005, 188–189; 2009). In Late Minoan IIIA:1, Mochlos used imported and locally made wares in the Knossian style as well as local shapes. The LM II–IIIA:1 ceramic practice at Mochlos shows strong Knossian influence, and Smith (2005, 190) remarks that “This fits well into the general picture of Late Minoan II and IIIA:1 Crete, when the surviving palace at Knossos seems to have extended its influence over much of Crete.” If the construction of an elite mansion (House He) at Gournia can be dated to LM IIIA:1, it may have been the residence of one of the Mycenaean officials mentioned in the Linear B tablets from Knossos. The Knossian ceramic connections visible at LM II–IIIA:1 Mochlos may also suggest that the Mirabello region was part of the Knossian territory at this time, perhaps controlled by a Mycenaean official at Gournia (contra Bennet 1985, 243). During LM IIIA:2–IIIB, the new Mycenaean settlers in the Gournia area seem to have been prosperous and connected to an active trade network in LM IIIA–IIIB. Two vases from Gournia (Hawes et al. 1908, pls. X:1, X:2; Kanta 1980, 140) were imported from Chania and the Mycenaean mainland, respectively. Likewise, the Episkopi cemetery

67

produced West Cretan and Mycenaean imports. As noted above, the well-known larnax from Episkopi was painted by the same artist who decorated a larnax from Pacheia Ammos (Kanta 1980, 143, 156). The many amphorae and stirrup jars at Gournia and its tombs point to the local production and export of wine, and perhaps oil. During LM IIIA:2–IIIB, vases from Chania, Palaikastro, Knossos, and the Mycenaean mainland were traded to Mochlos (Smith 2005). With the destruction of the palace at Knossos early in LM IIIA:2, Mochlos pottery became predominantly East Cretan in style. By LM IIIA:2, a new cemetery of at least 30 chamber tombs was established on the mainland coast. Pseira was violently destroyed in LM IB, sparsely resettled in LM IIIA, destroyed again in LM IIIB, with only a few settlers in LM IIIC (Davaras and Betancourt 2000, 42). Tombs in Kritsa at the western edge of the Mirabello region also contained Mycenaean and West Cretan imports (Kanta 1980, 134–139). Mochlos too received Chaniot and Mycenaean pottery (Smith 2005). Mycenaean (LH IIIA–IIIB) and Chaniot vases have also been reported at Petras (Tsipopoulou 2005). The LM IIIA–IIIB population in the Mirabello region seems to have consisted of two groups. The first comprises the settlements at Gournia and Mochlos, where the occupants lived in “small, haphazard dwellings” within earlier houses under the supervision of a Mycenaean official. Second, some of the occupants of the inland sites, such as 84, 104, 107, 123, and 144, may have been part of the native population. This is especially likely in the case of defensive site 144, which was newly settled in LM IIIA.

9

Late Minoan IIIC–Geometric Periods: Retreat into Refuge Settlements Krzysztof Nowicki

Late Minoan IIIC Period In the LM IIIC period, the number of local sites rises from eight in LM IIIA–IIIB to nine (Table 8; Map 33 ). Gournia (8) was abandoned, and a cluster of new sites (81, 84, 90, 97) was settled in the Isthmus valley. Off-site pottery (Map 34) found by the survey was minimal and was mostly dispersed around the cluster of sites in the Cha Gorge– Monastiraki area. Types of off-site pottery included cookware, pithoi, basins, and fine ware (Map 35). Higher up, site 150, Katalimata, Kephala, and Chalasmenos also were established. The area covered by the Gournia Project included mostly the lowland and foothills of the northern and central Ierapetra Isthmus, which were densely inhabited from the end of FN or early EM I through the LM I periods. Despite occasional and rather short-lasting settlement crises during these periods, caused probably by non-local historical factors, a quick recovery was stimulated by the richness of natural resources and the role this region played in the communication network between central and very eastern Crete. These early (pre–LM IIIC) short “intervals,” charac-

terized by a partial shift of population up to neighboring hills and mountains, may be very helpful for understanding the phenomenon of an unprecedented dramatic settlement crisis that took place in most of Crete at the turn of the 13th century B.C., and which changed the situation for many centuries to come. This event (or series of events) must have been related to the geographically broader political and social upheavals observed at that time in the eastern Mediterranean (Karageorghis and Morris, eds., 2001) and recorded in the Egyptian and Near Eastern texts. It seems that the well-established Late Bronze Age political systems ended shortly before 1200 B.C., and life became too risky in coastal and lowland Crete, as was the case in many other Aegean regions. What happened then to the population of those abandoned areas can be deduced from careful analysis of the settlement patterns (not only within the surveyed area, but also around it) following this collapse. The interpretation of the data concerning settlement history, yielded by the Gournia Survey Project, must therefore also take into consideration

70

KRZYSZTOF NOWICKI

neighboring mountains where the local inhabitants may have withdrawn whenever the political security system failed to protect them on the Isthmus Plain. The Ierapetra Isthmus was particularly exposed to external seaborne threat from both directions (south and north), but natural refuge areas could be easily found in the mountains immediately to the east, and to a lesser degree, in the hilly country to the west. The huge barrier of the West Siteia Mountains, which rises abruptly on the eastern side of the Isthmus, must always have been seen as the most important natural stronghold in case of insecurity; this phenomenon has been traced back to the Final Neolithic period and influenced life during historical times, too. The Gournia Survey extended to two such defensible strongholds, well rooted in the local settlement history: (1) the rocky slopes and cliffs above Monastiraki, and (2) the ridges above Kato and Epano Chorio (Map 1). The survey area, however, did not cover all the elements of these zones. The ridges above the Cha Gorge (in the vicinity of Kavousi) also played an important role in the LM IIIC security system of the Ierapetra Isthmus (Haggis 2005, 81–85), as did the entire massif of Katalimata (which stretches between Hagios Ioannis to the east and Vainia), east of the surveyed area (Nowicki 2004, 267–275; 2008). The analysis of the settlement history in LM IIIC through the Geometric period should also include the sites in the Asari Valley (Map 1) and near Vasiliki (Nowicki 2000, 106–107), west and south of the surveyed area. One of the most difficult problems in reconstructing the processes that led to the foundation of defensible settlements at the beginning of LM IIIC is the situation during the period immediately preceding this phenomenon; in other words, what was the settlement pattern during the LM IIIA–IIIB periods, and how stable (or unstable) was it? The general picture suggests a dramatic depopulation following the LM IB destructions, and slow revival during the LM II–IIIA:1 periods (Watrous et al. 2000, 477; Haggis 2005, 79). By the LM IIIA:2–IIIB period, the main excavated settlements in the region, such as Gournia and Vasiliki, did not recover, but Gournia was definitely substantially repopulated (Watrous 2001a, 89). A similar situation was suggested for the site at Hagios Theodoros, north of Kavousi (Haggis 2005, 80). A

settlement east of Episkopi (Map 32; Watrous et al. 2000, 477), together with a rich cemetery unearthed in the same region (Kanta 1980, 146–160), may suggest that there was some inland relocation of the coastal population in the southern part of the Isthmus. This conclusion is supported further by a cluster of LM III sites in the southern Isthmus, near Vainia, with the foundation of a defensive settlement on Stavromenos either in late LM IIIB or very early LM IIIC (Nowicki 2000, 88). Late Minoan IIIA−IIIB reoccupation (such as at Gournia and Mochlos [Brogan, Smith, and Soles 2002, 98–100]) and possible relocation (in particular on the southern coast) were complemented by the dispersion of a LM III population through the region. This latter phenomenon is represented by a scatter of small sites (81, 84, 100, 104, 107), which must have represented individual and isolated houses or small hamlets (Map 32; Haggis 2001, 45; 2005, 80–81), and by the pattern of individual tombs or small groups of tombs belonging to as yet unidentified habitation sites. All in all, the population in the Isthmus in the second half of the 13th century B.C. did not reach the level of the Neopalatial period, but a few sites may have been of the size of large villages (perhaps 123), with elements of higher (than a village) rank in the settlement “stratigraphy” (comparable to LM I towns?). The dispersion of settlement and the unclear character and scale of LM III reoccupation at LM I sites may be partly responsible for the underestimation of the LM IIIA:2–IIIB population, as proposed on the basis of surface evidence and excavations. These uncertain gazetteers of LM IIIB settlements are sometimes put against the lists of more easily identifiable LM IIIC sites. Such unreliable statistics might suggest that there was a growth of population between these two periods. That would be a wrong conclusion, however, since the archaeological visibility of LM IIIB and LM IIIC sites is not comparable. There is another problem of the LM IIIC gazetteers: as a rule, they group together the sites occupied during the LM IIIC period, though they were inhabited in different phases. Despite some criticism of the reliability of dating precision within one period (Dickinson 2006, 65), our present knowledge of LM IIIC pottery nevertheless allows differentiation at least between the very early, on one hand, and mid-tolate LM IIIC phases, on the other. The excavations

LATE MINOAN IIIC–GEOMETRIC PERIODS: RETREAT INTO REFUGE SETTLEMENTS

in the Kavousi and Monastiraki regions (Map 1) have proved that the earliest LM IIIC settlements (dated by the occasional presence of LM IIIB pottery; Phase I at Kavousi Kastro) in the northern part of the Isthmus were very few and probably restricted to the most defensible sites at Kavousi Kastro and Monastiraki Katalimata. Conversely, the other LM IIIC settlements, such as Vronda and Chalasmenos, were founded a few decades later (Phase II at Kavousi Kastro; Day and Snyder 2004, 69; Tsipopoulou 2004a, 103). This means that the population of the Ierapetra Isthmus at the beginning of LM IIIC was not larger than it was in LM IIIB, and that the population growth took place later, during the middle and late phases of the LM IIIC period. The above remarks are necessary to explain the character and scale of the changes that took place around 1200 B.C. Parts of two LM IIIC “retreat areas” were investigated by the Gournia Survey Project (Map 33). These are (1) the Monastiraki–Cha Gorge area, and (2) the Kato Chorio area. The first cluster consisted of two defensible settlements, Katalimata and Chalasmenos, which most probably belonged to one and the same community, but represented two different phases. This area may have offered security for the LM IIIC population living earlier in the central part of the Ierapetra Isthmus, namely, between the sheer cliffs of Papoura and Lamia, on the east, and the heights immediately west of Vasiliki, on the west, and the hills between Vasiliki and Episkopi, on the south. Late Minoan IIIB people from the northern part of the Isthmus moved, probably at the beginning of LM IIIC, to the rocky ridge of Kastro, the most defensible and strategic point above Kavousi, and to the inland valley of Avgo, behind Kastro (Haggis 2005, 81). The situation in the southern part of the Isthmus is more complicated. At least four LM IIIC defensible sites have been identified there, all located on and around the huge massif of Katalimata stretching between Hagios Ioannis and Vainia (Nowicki 2008, fig. 1B). The foundation of these settlements may have been related to the shift of the population from the area between the southern coast and Episkopi, at the beginning of the LM IIIC period, but the chronological sequence and relationships among all these sites remain debatable. Unlike the Kavousi and Monastiraki clusters, where several sites were excavated, the Katalimata group is known solely

71

from surveys and topographical studies (Nowicki 2004, 267–275). The Monastiraki cluster of LM IIIC sites (Map 33) must have been located around the mouth of the Cha Gorge, mostly for security reasons and for easy access to freshwater and arable land. The scatter of LM IIIC pottery (Maps 34, 35) around these sites indicates local land use. Thus, the early LM IIIC refugees, despite the abandonment of their old houses, exploited the same land and cultivated the same fields as their LM IIIB parents. The excavations at two sites, Katalimata and Chalasmenos, have shown that the shift to defensible places, so characteristic of the LM IIIC period, was a dynamic and complex process, consisting of several phases, when people responded quickly to changes in historical conditions. These two settlements lay at a distance of about 400 m from each other, on two different sides of the Cha Gorge. Katalimata was a barely accessible site located on small rocky terraces/shelves on the precipitous cliff of the northern side of the gorge, about 100 to 200 m above the bottom of the gorge. Access to the site is very difficult, and construction of houses in such an extreme place must have been caused by an extreme situation. Life for the inhabitants of Katalimata must have been very inconvenient, considering the effort needed for everyday communication and transport between the houses and the fields/pasturages. Such an extreme location, therefore, might have suggested that the site functioned exclusively as a refuge, and that the “proper village” existed at Chalasmenos (Dickinson 2006, 65), located more conveniently on a hill defended by steep slopes on three sides and by a high cliff rising above, on the fourth. Despite its hilltop location, Chalasmenos was much less defensible than Katalimata. The distance of about 20 to 30 minutes between Chalasmenos and Katalimata might have allowed people to evacuate quickly in the case of danger, but the goods stored in the Chalasmenos houses would have been lost. The excavations (Nowicki 2008) have proved, however, that Katalimata was a permanently inhabited settlement, founded earlier (at the beginning of the LM IIIC period) than Chalasmenos, and it was replaced by the latter probably after a few decades when LM IIIB vessels disappeared from the household pottery. This sequence of events fits very well with evidence yielded by the Kavousi

72

KRZYSZTOF NOWICKI

cluster and by other archaeological sites identified by surveys (Mook and Coulson 1997; Nowicki 2000, 229–235). Katalimata was contemporary with Phase I at Kavousi Kastro, which seems to be absent at less defensible Vronda and Chalasmenos. In the Kavousi cluster, the foundation of lower sites, such as Vronda (and Azoria?), did not result in the abandonment of its most defensible element, the Kastro, whereas in the Monastiraki case it did. The reason is obvious: Katalimata was an extreme site much less convenient for habitation than Kastro. Whether there was some overlapping between Katalimata and Chalasmenos, with the houses permanently inhabited in both settlements, is still the question to be solved when the complete evidence from Chalasmenos is published. Such overlapping was possible during Phase II of Kavousi Kastro, somewhere close to the middle of the 12th century B.C., since by that time Chalasmenos was already inhabited and Katalimata was still used (occasionally?). The problem of the chronological relationship between Katalimata and Chalasmenos is complicated further by the fact that no excavation has yet been undertaken in the upper settlement of Chalasmenos, which is located on the northern rocky spur of the hill, and has much more defensible characteristics than the excavated lower quarter of the settlement. Proper dating of this part of the site may be crucial for understanding changes in settlement locations during the LM IIIC period in the Monastiraki cluster. Monastiraki Katalimata was, in the first decades of the LM IIIC period, probably the only permanent settlement in the area within a radius of about 2 km from the mouth of the Cha Gorge. About 12 houses were constructed on natural terraces hanging on the Cha cliff (Nowicki 2008, 65), and these may indicate the number of families that lived here at the beginning of the 12th century B.C., that is, before the settlement of Chalasmenos was founded. These 12 or so families were probably of local origin, coming from the central part of the Isthmus, as far as Vasiliki and the hills north of Episkopi. Whether Katalimata also housed some of the people who had earlier lived on the northern coast, between Gournia and Pacheia Ammos, is difficult to say. The inhabitants of the Gournia Plain may have retreated south, up the Moulas Gorge to the much closer LM IIIC “refuge area” in the Asari

Valley, located about 30 to 45 minutes by foot south and west of Gournia (Hayden 2004a, fig. 2). A defensible settlement on the summit of Asari Kephala may have been founded already in the very beginning of the LM IIIC period (Nowicki 2000, 107). Relocation of the Gournia Valley inhabitants to Asari, at the beginning of the withdrawal from the coast, would be the most logical solution, if we accept that people tried to maintain links to their old and thus well-organized economic hinterland. Late Minoan IIIC sherds found near Gournia (Maps 34, 35) show that the land was still being used. The only reservation concerns the defensibility of Asari Kephala. The site is very well defended by steep slopes and cliffs from the route leading from the sea, up the Moulas Gorge, but is rather easily accessible from the inland valley of Asari (Map 1). The ridge of Kephala does not offer the same natural defensibility as Monastiraki Katalimata and Kavousi Kastro, but the settlement was well hidden from the coast and coastal plain. Evidence for Middle Minoan occupation on Asari Kephala indicates that the place may have played a similar role in the history of the Gournia region as Katalimata did for the Vasiliki–Monastiraki area and the hills above Kavousi for the Kavousi–Tholos plain: it must have been the first choice whenever people had to look for security. Reconstructing the changes in settlement location ca. 1200 B.C. around Kato Chorio and Episkopi, on the southern outskirts of the Gournia survey area, is more difficult. The survey here covered only a small part of the natural “refuge area” as offered by a huge massif of Katalimata, which constitutes the most southwestern part of the West Siteia Mountains. The landscape here is very rough, with a great number of rocky ridges and difficult communication within the region. Under peaceful historical conditions, Katalimata Mountain and the hills around it have been seasonally used by shepherds and farmers who grazed their animals, and cultivated grain and pulses on gentler slopes and in upland basins or lakkoi. Permanent settlements have been located on the Isthmus and coastal plains, and in the Hagios Ioannis Valley. When the situation changed, however, most of this lowland population looked for security on inaccessible rocky ridges above. The Gournia Survey has identified one of these sites: Kato Chorio Prophetes Elias, which is

LATE MINOAN IIIC–GEOMETRIC PERIODS: RETREAT INTO REFUGE SETTLEMENTS

the nearest defensible place for the inhabitants of the Kato Chorio–Episkopi area. Farther to the south, already beyond the Gournia survey borders, lay Vainia Stavromenos and Skouro (Map 1), two sites dominating the coastal part of the Isthmus and the south coast, east of Ierapetra. Surface pottery from Vainia Stavromenos shows that the settlement was occupied on several occasions between the Final Neolithic and the Venetian period, but the most substantial evidence represents the LM IIIC period (Nowicki 2000, 86–89). Since Stavromenos does not belong to the group of the most defensible sites, one may wonder if the settlement was able to survive in this rather exposed place (about 1 to 1.5 hours by foot from the sea) during the period when another group of people, only several kilometers to the north (Monastiraki Katalimata), was so desperate that they decided to build their houses on a precipitous cliff. Vainia Stavromenos was probably founded still in the LM IIIB period and was defended by a fortification wall, fragments of which could be seen on the western side of the hill (before the old path was bulldozed and converted into a dirt road and “a parking” serving the chapel above). Pottery dates mostly to early LM IIIC, but a number of fragments were of LM IIIB date. The settlement was abandoned in the LM IIIC period, most probably in its early phase. The site may have been founded, therefore, somewhat earlier than Phase I of Kavousi Kastro, and abandoned before Phase II or during this phase. In general, Vainia Stavromenos was probably contemporary with Palaikastro Kastri and Petras Kephala (Tsipopoulou 2008, xviii), and it may have represented a similar phenomenon of transitional LM IIIB/IIIC settlements, the security of which, apart from some natural defensibility, was reinforced by the military strength of the community, occasionally expressed by the construction of fortification walls (see, for example, Petras in East Crete, Koukounaries on Paros, and Moulas on Karpathos). This latter fact suggests that there were probably differences in the social organization of the groups responsible for the foundation of Monastiraki Katalimata and Vainia Stavromenos. Another defensible LM IIIC site is situated about 1 km east of Stavromenos on the rocky ridge of Skouro (about 400 m asl), rising from the southern slope of the Katalimata Massif (Nowicki 2008, fig. 1B:site 15). The ridge was fortified with two

73

lines of walls that enclosed an upper rocky citadel and a more extensive lower area. Pottery is not very abundant, but it allows dating the main phase of occupation to the transitional period of LM IIIB/IIIC. The site was also used in the Middle Minoan and Medieval periods, as were most of the similar defensible hills in this region. Some elements of the fortification may go back to the Protopalatial period. The close distance between Vainia Stavromenos and Skouro raises the question about the functional and chronological relationship between these two sites. No answer is possible, however, without more detailed investigations and excavations at both sites. Among the most important elements of the LM IIIC defensible settlement system in this region was Hagios Ioannis Katalimata, a large settlement (12,000 to 15,000 m2) situated on the long but narrow summit of Katalimata Mountain (Nowicki 2008, fig. 1B:site 17). The site extends over an area of ca. 300 x 30–50 m and is densely covered with architectural remains. On the basis of the probable number of houses, a population of at least 50 to 60 families can be estimated. The site must have been settled very early in LM IIIC, as is indicated by some of the surface pottery, although the occupation probably continued through most of this period. A few Protogeometric (and perhaps Geometric) sherds do not allow us to reconstruct a proper settlement, but may be related to post-abandonment activity, similar to that recorded at other Early Iron Age settlements. Probably by the middle-to-late LM IIIC period, and certainly before the Protogeometric, Hagios Ioannis Katalimata lost its importance in favor of Kato Chorio Prophetes Elias (150) and was abandoned. A secure interpretation of the role of Hagios Ioannis Katalimata in the regional settlement system is difficult, owing to the lack of excavated material from this and other sites in the same cluster. Hagios Ioannis Katalimata was probably founded together with Kavousi Kastro and Monastiraki Katalimata in early LM IIIC, but a little later than Vainia Stavromenos. Early LM IIIC pottery seems to show that Hagios Ioannis Katalimata and Vainia Stavromenos coexisted during the first decades of the LM IIIC period. Topographical differences between these sites, however, are considerable and require some explanation. If changes in security along the southern coast around 1200 B.C. were

74

KRZYSZTOF NOWICKI

very dramatic, then the sequence of settlement relocations may have been very fast—too fast, in fact, to be identified by the pottery. The evacuation of Vainia Stavromenos might be immediately followed by the foundation of Hagios Ioannis Katalimata, something that cannot be identified by archaeological evidence. There is, however, another possibility: that these two topographically different settlements coexisted but were inhabited by different groups of people. In such a scenario, Vainia Stavromenos would represent a large town community with elements of an old (LM IIIB) social organization and a local chieftain still maintaining his authority, and capable of building a substantial fortification wall, whereas the inhabitants of Hagios Ioannis Katalimata would represent loosely connected groups of farmers and shepherds from all around the massif, who concentrated on that inaccessible summit in a more spontaneous way. Kavousi Kastro, Monastiraki Katalimata, and Hagios Ioannis Katalimata, all founded in very early LM IIIC or even at the transition between LM IIIB and LM IIIC, prove that the beginning of the collapse of the LM IIIB settlement patterns was the most difficult time. Late Minoan IIIB habitation sites in the Ierapetra Isthmus, on the plain and both coasts, were abandoned within a short time, at most a few decades, and were replaced with extremely well-defended, new settlements. The comparison between the hypothetical population in late LM IIIB and early LM IIIC, as based on surface evidence, allows us to reconstruct LM IIIC defensible settlements as being built by local people. The situation, however, may have changed a few decades later when the defensible system was remodeled and better adjusted to environmental factors. Perhaps new elements and changes in social development of the local “refuge communities” led to the expansion of settlement and another relocation of sites around the mid 12th century B.C. It is probably also this phase that can be better identified in the area covered by the Gournia Survey Project, with LM IIIC evidence related to the settlements at Monastiraki Chalasmenos (67, 68, 70) and Kato Chorio Prophetes Elias (144). A group of sites about 1–1.5 km southwest of Chalasmenos (83, 84, 90, 97) may represent other elements of this process.

The foundation of a village at Chalasmenos marked the general expansion of LM IIIC settlement after a few decades of apparent demographic crisis. Some of the early LM IIIC “refuge settlements,” which were particularly inconvenient for habitation, were abandoned as permanent villages (Monastiraki Katalimata and Anatoli Elliniki Korphi), and perhaps used only occasionally as true “refuge places.” Whenever an early LM IIIC settlement offered characteristics acceptable for further occupation (Kavousi Kastro), however, it was inhabited. Close to the middle of the 12th century, together with Chalasmenos, other settlements were built in this region: Kavousi Vronda and probably Vasiliki Kephala. Kato Chorio Prophetes Elias (144) may have been first settled at that time or (if settled earlier) took a leading role (from Vainia Stavromenos and Hagios Ioannis Katalimata) as the most important settlement in the southern part of the Isthmus. A similar central position in the LM IIIC cluster, in the northern part of the Isthmus, may have been taken by Kavousi Azoria, although the Kastro stronghold continued to be inhabited. The foundation of Vasiliki Kephala may have been linked to the early LM IIIC community in the Asari Valley and thus, indirectly, to the LM IIIB inhabitants of Gournia as well. Despite Eliopoulos’s suggestion (2004, 83), it does not seem that Vasiliki Kephala was founded immediately after the final abandonment of Gournia, ca. 1200 B.C. The pottery (published to date) and topographical characteristics suggest, instead, the foundation during Phase II and not Phase I of Kavousi Kastro. The number, and in some cases the size, of settlements increased within a relatively short time around the middle of the 12th century, and there must have been some reasons for that. The new locations, although less defensible, did not reflect the LM IIIB pattern, which indicates that insecurity was still a problem. The mid 12th-century settlement expansion may have been stimulated by some of the following factors (or all of them): (1) the inhabitants of the very early defensive LM IIIC settlements were abnormally “squeezed” in their new houses (for the security reasons), and once the hardest time was over, the first reaction was to look for more habitation space, which could not always

LATE MINOAN IIIC–GEOMETRIC PERIODS: RETREAT INTO REFUGE SETTLEMENTS

be found within defensible borders of the early LM IIIC “refuge sites”; (2) natural population growth during the first decades of the 12th century caused population pressure on the restricted space in the early LM IIIC “refuge sites”; (3) the early LM IIIC settlement pattern may have consisted not only of defensible settlements, but also of numerous

75

refugee families scattered in upland areas such as the West Siteia Mountains; once the system started to recover, a part of this population might have opted for a return to village life; finally, (4) when the process of disintegration of the Mycenaean states progressed during the first decades of the 12th century, newcomers arrived.

Protogeometric–Geometric Periods The mid 12th-century B.C. settlement expansion was a short-lived phenomenon which terminated, after 100 to 150 years, with abandonment of the settlements at Kavousi Vronda, Monastiraki Chalasmenos, and Vasiliki Kephala at the end of the LM IIIC or the beginning of Protogeometric period. The settlement system changed around 1000 B.C. (Table 9; Map 36), and this is reflected by the “empty” landscape within the surveyed area, comparable to that of ca. 1200 B.C. The changes that can be seen in the Ierapetra Isthmus are very similar to those that took place in the Lasithi Mountains, and which resulted in the abandonment of many LM IIIC defensible settlements in favor of fewer sites that became regional centers. Settlement was concentrated on site 150 and the small outlying sites 84 and 97. In the Ierapetra Isthmus, the area between Kavousi and Kato Chorio was abandoned, and the population moved either to the south, to the acropolis of Kato Chorio Prophetes Elias, or to the Kavousi cluster, with two major settlements at Kavousi Kastro and Azoria. The growth and territorial expansion of those and other similar centers may have created a new political situation in which the internal tension became more important than the external threat, as it had been at the beginning of the 12th century. The consolidation of the Isthmus population from this point onward at two main centers or clusters (Kavousi and Kato Chorio Prophetes Elias) stimulated further social and political processes, leading eventually to the appearance of towns. The lack of apparent outer rural elements in the Protogeometric settlement pattern suggests growing tension between different units. The similar process affected not only the settlements on the edge of the Ierapetra Isthmus, but also upland sites

in the inner valleys in the West Siteia Mountains, and has also been recorded in the Lasithi Mountains. The abandoned LM IIIC settlements of Monastiraki Chalasmenos, Vasiliki Kephala, and Kavousi Vronda were probably occasionally revisited in the Protogeometric period, and the ruins may have been used for temporary habitation, burials (Chalasmenos?), and other activity (Eliopoulos 2004, 84; Rapp 2007, pl. 7:1). This phase is only poorly known at the main Protogeometric sites in the region because later (in particular, in the Geometric period) substantial rearrangements and rebuilding destroyed much of the evidence. The Geometric period shows the settlement pattern well defined (Table 10; Map 37), with the two main centers on their way to acquiring an urban character rather than continuing an old LM IIIC model of an extensive, but still rural settlement with a few public buildings and little differentiation in house plans and sizes. One of these centers was situated on the hill of Azoria, southeast of Kavousi, and the second on Prophetes Elias (150), above Kato Chorio. Azoria was still supplemented with the high acropolis of Kastro, and Prophetes Elias shared the topographical characteristics of Kavousi and Azoria together. A few off-site Geometric sherds (pieces of fine ware and cookware) were found some 500 m southeast of Gournia (8) and near the location of the Neopalatial site 60. An interesting phenomenon of this period is the reuse of some abandoned LM IIIC settlements like Chalasmenos and Vronda. At Chalasmenos, after a few centuries of abandonment, a building was constructed in the Geometric period on the ruins of earlier megara. According to Tsipopoulou (2004b), the Late Geometric building erected here did not mark a true reoccupation, but

76

KRZYSZTOF NOWICKI

was only periodically used for communal meals(?). If such an interpretation is confirmed, it would be tempting to see new activity at Chalasmenos as related to the improvement of the political situation, decrease of tension between competing centers, and increase of peaceful interaction between different communities. It is possible, however, to propose another explanation for the ceremonial/symbolic return to Chalasmenos. The hill rises immediately south of the Cha Gorge and a deep streambed, which was an obvious natural border between the Kavousi and Kato Chorio clusters. Did this restricted return to Chalasmenos express some territorial

claims of one of the groups? At Vronda, a series of Geometric cremation burials were built in ruined houses. Both cases indicate that the communities concentrated at and around Kavousi Azoria, and Kato Chorio Prophetes Elias, entered a phase during which old and abandoned settlements were given a symbolic (but also practical) function to the nucleated societies (Wallace 2003, 262; 2005, 264– 270; Haggis et al. 2004, 341; Tsipopoulou 2004b). By that time, the Ierapetra Isthmus and neighboring regions were probably already well divided into territorial units, which were on their way to the next stage of their social and political development.

10

Orientalizing–Classical Periods: Population Nucleation and Development of the Polis L. Vance Watrous

Orientalizing–Archaic Periods Settlement in the survey zone remained severely nucleated during the Orientalizing–Archaic periods, much as it had been in the earlier Iron Age (see Ch. 9). Only the town at Prophetes Elias (150), the fortified hill site 144 at its base, and the coastal site 2 near the Vrokastro survey area were inhabited (Table 11; Map 38). Judging from its size during the Classical period, the town at Prophetes Elias (150) probably grew during this period. The site produced cookware, mortars, relief pithoi, and drinking vessels (a krater, banded jugs, and black-glazed and monochrome cups). Few off-site finds were identified (besides two skyphoi from site 144). In the Kavousi area, the LM IIIC–Geometric settlements, such as Kavousi Kastro (Coulson et al. 1997), were dispersed in the mountains above and behind Kavousi village in clusters near a water source. Haggis (2005, 83–84, fig. 14) reconstructs local Early Iron Age settlement growth; settlement

began on Kastro by the end of LM IIIB and spread to Vronda and Azoria in LM IIIC. Vronda was deserted in LM IIIC, while Kastro continued to grow in the Protogeometric–Geometric periods. The local population grew significantly during the LM IIIC period. Protogeometric–Geometric sites were probably made up of separate single or extended families forming independent agricultural units (Haggis 2005, 83–84). During the 8th and 7th centuries B.C., many of these sites were abandoned for the hilltop settlement at Azoria (K71; Haggis et al. 2004). By the end of the 7th century this synoikism had created a town-sized polis at Azoria that possessed an agora, andreion, and a monumental civic building. These public structures suggest that by ca. 630 B.C., Azoria had become the center of a small polis (Haggis and Mook 2011). By the Archaic period, three major sites—Prophetes Elias (150), Azoria, and Istron—dominate the surveyed areas. At the end of the 6th century, Azoria was destroyed,

78

L. VANCE WATROUS

with a small reoccupation lasting perhaps 25 years (Haggis et al. 2004, 345). In the Vrokastro area, the population during the LM IIIC period (Hayden 2004a, fig. 20) was nucleated at some 13 sites. Based on site size and numbers, Hayden (2004a, 131, 151) estimates that the local population doubled in the LM IIIC period. Four settlements—at Vrokastro (Map 1), its fortified port at Elias to Nisi on a promontory east of Priniatikos Pyrgos, along the Istron Valley, and in the mountains—were village-sized (Hayden 2004a, fig. 21). Late Minoan IIIC settlements remained in contact with the sea. By the Protogeometric–Geometric period, settlement had nucleated at Vrokastro and Elias to Nisi, with a halo of smaller surrounding

sites (Hayden 2004a, 153). The coastal site and harbor at Elias to Nisi may have been seasonally occupied during this period (Hayden 2004a, 139). In the 7th century, the inhabitants of Vrokastro descended to the coast to form the polis of Istron, on and near the promontory of Nisi Pandeleimon (Hayden 2004a, 160, fig. 22). Except for a small site at Priniatikos Pyrgos (B. Hayden, pers. comm.) and settlement of Nisi Pandeleimon, the catchment surrounding Istron was only gradually resettled (Hayden 2004a, 171). At the same time, some 25 new sites, including the ancient settlement of Oleros (OL1A), were settled in the southern inland area of the Meseleri Valley, Schinavria Ridge, and the Prina area of the Vrokastro zone.

Classical Period Settlement changed very little from the previous period. In the Classical period (Table 12; Map 39), settlement continued to remain strictly nucleated at Prophetes Elias (150). During this period, the settlement reached its largest size. Generally speaking, our knowledge of the Mirabello area in the Classical period remains skeletal, since excavation and research on this period have been focused on central and western Crete (Erickson 2005). Late in the Classical period, the settlement pattern began to change. Two new sites were established on the north coast. A farm (2) appeared at the edge of the Vrokastro survey zone, and a hamlet-sized settlement (23) was established at Halepa, on the north coast. The foundation of a site at the harbor of Halepa (23) is a major departure from previous practice in our area and is probably a sign of the commercial expansion of the polis of Hierapytna to the north coast. An Attic red-figure amphora, dated to 450 B.C., found in Ierapetra (Davaras 1972, 647, pl. 602), may be evidence of this opening. By the 2nd century B.C., Halepa was complemented by a port facility at Mochlos (Vogeikoff-Brogan 2004). The mercantile function of Halepa is not in doubt, as it produced distinctive Hierapytnian amphorae as well as an imported non-Cretan amphora. At this time, Prophetes Elias (150), the main settlement in the survey zone, was almost 10 ha in size, making it the largest known Classical settlement in

the Gournia survey region. This Classical settlement produced roof tiles, utilitarian vases (e.g., cookware, mortars, basins, amphorae, pithoi, beehives, and spindle whorls), as well as fine tableware, such as black-glazed kraters, cups, and kantharoi. Outside of Prophetes Elias, roof tiles, pithoi, basins, cookware, cups, jugs, and a lamp were found at rural farming sites. Off-site finds recognized by the survey were minimal: a piece of fine ware and mortar found some 500 m west of Monastiraki. The latest vase identified from site 150 (before its reoccupation in the Roman era) is an Attic black-glazed kantharos of 425 B.C. The site therefore appears to have been abandoned by the end of the 5th century B.C. For this reason, it seems probable that the Prophetes Elias settlement, located 5 km north of the ancient city of Hierapytna, is to be identified with the polis of Larisa mentioned by Strabo (Geo. 9.440.19). Writing in the late 1st century B.C., Strabo says that Larisa had been physically absorbed into Hierapytna, and that the plain below it was called the Larisan Plain. This passage fits both the local archaeological evidence and local topography. Plate 3A shows a plain, the Episkopi Basin, below Prophetes Elias. The Kavousi area shared the same experience as the Gournia area during the Classical period: it was entirely depopulated. Azoria was destroyed at the end of the 6th century, with a small reoccupation lasting perhaps 25 years (Haggis et al. 2004, 345),

ORIENTALIZING–CLASSICAL PERIODS: POPULATION NUCLEATION AND DEVELOPMENT OF THE POLIS

and was only resettled to any extent in the Hellenistic period (by the 1st century B.C.; Haggis 2005, 86). Out of the reach of Hierapytna, settlement in the Vrokastro area was less constrained (Hayden 2004a, 107–198, fig. 23). At least 18 sites existed in the area during the Classical period. During the 5th century B.C., most of the settlement in the mountains around Vrokastro moved down to the coast at Istron. Istron became a sizable (25 ha) polis surrounded by farms and hamlets. North of Vrokastro, another polis existed at Olous, near Hagios Nikolaos, but little is known about it, as Platon’s excavations (Platon 1960) there remain unpublished. In ca. 180 B.C., Istron was destroyed, probably by Hierapytna. Only a few coastal sites survived after the destruction of Istron. In the southern area of the Vrokastro Survey, many sites were abandoned at the end of the Archaic period as part of the synoikism at the polis site of Oleros (Hayden 2004a, 182–192). During the

79

Archaic–Early Classical periods, Oleros was located on a fortified peak. Shortly thereafter, residents of the town moved down to the valley floor and the area was absorbed by Hierapytna, perhaps as early as the 4th century B.C. (Hayden 2004a, 100). Most of the larger sites in the survey zone, however, continued through the Classical and into the Hellenistic period. The extreme nucleation of Early Classical settlement in the northern Isthmus, in contrast to the dense settlement in the Vrokastro area, looks as if it were the result of local insecurity, a no man’s land, between the rival poleis of Hierapytna, Istron, Oleros, Olous, and Lyttos. Once Hieraptyna had eliminated the large site at Prophetes Elias (150), as well as Istron and Oleros, and gained control of the region, settlements began to be reestablished across the Isthmus, beginning in the Hellenistic period (see Ch. 11).

11

Hellenistic and Roman Periods: Expansion of the Isthmus in an International Era Natalia Vogeikoff-Brogan

Hellenistic Period The Gournia Survey has produced scant evidence for any significant occupation on the Isthmus Plain, the surrounding lowlands, and hill slopes during the Hellenistic period. Only three sites (5, 23, 77) preserve some evidence of continuity from the Classical period. The farm at Palaiobarda (5), west of Gournia on the south slope of a steep coastal hill, and the hamlet-sized settlement at Halepa (23) were established late in the Classical or early in the Hellenistic period. Site 77, a hamlet near Monastiraki, was short-lived, occupied ca. 350–200 B.C. The site at Prophetes Elias (150) at Kato Chorio, which shows continuous habitation from LM IIIC to the Classical period, declined in the 4th century B.C., but it was reoccupied in the Roman period. The Gournia Survey identified few sites with Hellenistic pottery, most of them being field sites and farm sites (32, 33, 81, 88, 89, 104, 114; see Table 13 and Map 40). There is no village-sized settlement on the Isthmus

until the Late Hellenistic period (139). Off-site pottery is also rare (Map 41). The conspicuous lack of any significant Hellenistic settlements on the Isthmus agrees with the results of the neighboring Kavousi Survey, which produced very little evidence for any important activity in the area before the 2nd century B.C., when port facilities were established at Mochlos. In the early 1st century B.C., Kavousi village (K24) and four other sites (K5, K40–K42) were settled. Tholos and Agriomandra (K1, K39) became ports (Haggis 2005, 86, fig. 16). The coast was dominated by the large (25 ha) polis center of Istron. To the south, Oleros, a second polis, remained independent until the 4th or 3rd centuries B.C., when it was absorbed by Hierapytna (Hayden 2004a, 185–187, 192). Puzzling at first, the thin distribution of sites during the Hellenistic period, both on the Isthmus and in the Kavousi area, finds some explanation when viewed

82

NATALIA VOGEIKOFF-BROGAN

in relationship with the political history of Hierapytna. In contrast, the Vrokastro area was settled with some 20 sites (Hayden 2004a, fig. 23). The literary and epigraphic testimonia bear evidence for the gradual growth of Hierapytna’s political strength and sovereignty from the 3rd century B.C. onward. A number of treaties initiated by Hierapytna attest to the expansion of the city as far as the Sanctuary of Zeus Diktaios to the west, and near Istron on the northern coast of the Bay of Mirabello (Chaniotis 1996). The southwestern boundaries of Hierapytna are less clear, but a series of isopolity treaties with Arkades, Priansos, and Viannos suggest that Hierapytna had expanded her territory to the lower elevations of the Lasithi Mountains by the early 2nd century B.C. (VogeikoffBrogan 2004, 214, with related bibliography). The expansion of Hierapytna to the north, however, remains poorly documented, even if widely assumed. It is largely based on a reference by Strabo (Geo. 10.4.475) to the synoikism between Larisa and Hierapytna, the first being absorbed by the second, probably in the 4th century B.C. A large number of sites near Hierapytna have been proposed as candidates for the possible location of Larisa (Stefanaki 2005, 26–27), with Prophetes Elias (150), near Kato Chorio, being the most likely. The decline of the site in the 4th century B.C. coincides with the rise of Hierapytna on the south coast (Erickson 2000, 312, 314; Watrous et al. 2000, 477). Although many scholars would accept Oleros, near modern Meseleroi, as the mother city of Hierapytna, Stefanaki has recently suggested that pre-Hellenistic Hierapytna was located at Prophetes Elias, while placing Larisa at the hill of Kephala near Vasiliki (Stefanaki 2005, 26–28), in the same line with other scholars who believe that Kephala Vasiliki could be identified with Minoan Hierapytna (Iliopoulos 1998, 311–312; Nowicki 2000, 106). All these hypotheses about the location of earlier Hierapytna, however, are based on the assumption that no pre-Hellenistic settlement existed below modern Ierapetra. This must remain uncertain, for the rescue excavations by the 24th Ephorate at Ierapetra rarely have gone below the Roman levels (M. Tsipopoulou, pers. comm.). The best evidence for the presence of Hierapytna in the northern part of the Isthmus comes from a funerary inscription of the late 2nd/early 1st century B.C., found near Vasiliki. H. and M. van Effenterre

have suggested that the inscription marked a mass grave connected with Hierapytna’s not necessarily successful attempts at northern expansion (van Effenterre and van Effenterre 1989). With the exception of the Vasiliki inscription, the lack of any other historical information concerning the area of the Isthmus in the Hellenistic period coincides with the absence of any major settlements in the triangle roughly defined by Lato, Hierapytna, and Praisos. In addition, Strabo’s description (Geo. 10.4.475) of the Isthmus confirms the scarcity of sites in the area during the Hellenistic period. According to Strabo, the Isthmus extended from Minoa of Lyttos to Hierapytna at the Libyan Sea, covering a distance of 60 stadia. Noticeably, Strabo does not refer to any other poleis within this distance. Strabo’s mention of Lyttos in connection with Minoa is puzzling (whether Minoa is placed near Istron, or at Gournia, or at Pacheia Ammos, following different suggestions), for it implies that Lyttos had a harbor at a considerable distance from her base, near a territory that the Latians and/or the Hierapytnians controlled (Chaniotis 1996, 346 n. 1717; Guizzi 2001, 313–314; for a recent review of the issue, see Stefanaki 2005, 22–25). Finally, the meager archaeological evidence from site 23, at the east end of Pacheia Ammos, does not support its identification with Strabo’s Minoa. The lack of any major Hellenistic sites in the northern half of the Isthmus could also possibly indicate that this area represented an “apple of discord” between Lato, Hierapytna, and possibly Lyttos. This would also explain the lack of any substantial settlements either at Pacheia Ammos or Tholos until Early Roman times. H. and M. van Effenterre are probably right in suggesting that there was a transverse demarcation at the Isthmus, separating the north from the south coast (van Effenterre and van Effenterre 1989). The desire on the part of Hierapytna to enlarge its territory probably reflects a combination of needs. The absence of any important establishments at the harbors of Pacheia Ammos and Tholos, and along the Isthmus during the Hellenistic period, suggests that trade might not have been the only reason for Hierapytna’s northern expansion. Although the Isthmus provided an alternative avenue for overland transshipment, avoiding the trip around the east coast of Crete, overland transportation of goods required a network of land facilities in the form of

HELLENISTIC AND ROMAN PERIODS: EXPANSION OF THE ISTHMUS IN AN INTERNATIONAL ERA

farmsteads and hamlets, which were absent from the Isthmus and the Kavousi pass until the Early Roman period. In his study of the Cretan isopolity treaties, Chaniotis has argued that Hierapytna’s large territorial expansion in the 2nd century B.C., through demonstrating aggressive (conquest) and nonaggressive (isopolity treaties) behavior, should be associated with its need for more pastureland, in order to cope with its growing population and other social problems (e.g., concentration of land in the hands of a few landowners; Chaniotis 1995, 72–77; Chaniotis, ed., 1999). Search for extra pastureland may have been a motivation, but it was not the prime reason for the northward expansion of Hierapytna. Strategically located at a harbor for the trade between Rhodes and Alexandria, the city of Hierapytna, like other Cretan cities, increased her revenues by collecting harborage fees and levies on non-Cretan products in transit (Viviers 1999; Vogeikoff-Brogan 2004, 217–219). The establishment of a coastal settlement at Mochlos further strengthens this case. I have argued elsewhere, on the basis of ceramic evidence, that Mochlos was settled by Hierapytnians with a primary interest in maritime profit, whether this was limited to the collection of harborage fees and levies, or possibly extended to the export of Cretan goods, most likely wine (Vogeikoff-Brogan 2004; Vogeikoff-Brogan et al. 2004). The context for Hierapytna’s interest was the important Rhodes–Alexandria trade in grain and

83

wine. Alexandria exported grain in return for Rhodian wine, as indicated by the large quantities of Rhodian stamped amphora handles found in Egypt (Fraser 1972, 162–169; Perlman 1999, 151– 153). The value of this trade can be gauged by the fact that Itanos (Map 5) was a Ptolemaic protectorate that oversaw and profited from the trade between Rhodes and Egypt. In an effort to be part of this trade, Hierapytna produced amphorae (presumably filled with local wine) that imitated Rhodian examples. A stamped amphora of the Rhodian type in Hierapytnian clay was found by the survey at site 75. Hierapytnian conflict with and eventual conquest of the poleis of East Crete, including Itanos, must have been fueled by a desire to control this major trade route and its goods. Recent fieldwork in East Crete, especially at the site of Trypetos (Map 5), has confirmed the local production of transport amphorae as early as the end of the 3rd/beginning of the 2nd century B.C., which could have been used for the intra-island trade of Cretan wine, without excluding off-island trade activities (Vogeikoff-Brogan and Apostolakou 2004; contra Chaniotis 2008, 89–90, without citing the evidence from Trypetos). The choice of Mochlos by Hieraptyna for its northern outlet, over the harbors at Pacheia Ammos and Tholos, is puzzling at first, for the site is difficult to access by land, but the harbors at Tholos and Mochlos are superior to any other place along the north coast of the Isthmus.

Roman Period The Romans finally conquered Crete in 67 B.C., after several unsuccessful attempts, beginning in 72 B.C. The Cretan coalition with King Mithridates of Pontus, the enemy of Rome, angered the Romans, who first tried to take the island unsuccessfully under M. Antonius Creticus. His defeat in 72–71 B.C. is a powerful testimony to the military strength of the Cretan cities during the early 1st century B.C. (De Souza 1998, 113). Revenge cannot have been the only reason for the persistent Roman efforts to conquer the island. According to De Souza, Rome was interested in annexing Crete not only for its strategic location in the eastern Mediterranean trade,

but also for its wealth. The high quality of the silver coinage of Cretan cities bears testimony to the prosperity of Crete (Stefanaki 2006). In addition, the high indemnity of 4,000 talents that the Romans imposed on Crete as a punishment for its support of Mithridates further strengthens the image of a prosperous island (De Souza 1998, 113), which most likely had amassed its wealth through the sea, with aggressive (piracy) or peaceful (mercenary service, harborage fees, possible exports) methods. The eventual conquest of Hierapytna by Metellus in 67 B.C. did not cause the city’s decline. Instead, to judge from the literary and epigraphic sources,

84

NATALIA VOGEIKOFF-BROGAN

Hierapytna entered a new period of prosperity, achieving a status close to that of Gortyn (Di Vita, ed., 1988), which was the capital of the new Roman province. Although today little archaeological evidence exsits to support the elevated status of Hierapytna in the Roman period, travelers like Christoforo Buondelmonti in 1415, Francisco Barozzi in 1577, Onorio Belli in 1590, and Admiral Spratt as late as the 19th century described a wealth of Roman buildings that were visible from the sea (Papadakis 1986, 36–54). Throughout her history, the prosperity of Hierapytna was largely connected with trade. In addition to her superb location at the center of the eastern Mediterranean trade routes, the Pax Romana offered Hierapytna the opportunity to finally exploit the advantages of the Isthmus corridor. Unlike mainland Greece, which showed a significant drop in the number of rural sites from the 2nd century B.C. to the 3rd century A.D., the Isthmus and the Kavousi regions (as well as the Mesara Plain) bear evidence for the establishment of a large number of rural sites during the Early Roman period (Alcock 1993, 72; Watrous and Hatzi-Vallianou 2004; Haggis 2005, 86–87). Most of the Roman sites at the Isthmus were established some time between the 1st century B.C. and the 1st century A.D. The increased number of sites on the Isthmus reflects the prosperity of this city during the Roman period. While there were fewer than 10 sites in the Hellenistic period, there are more than 20 sites that can be dated to the Early Roman period, of which 11 are village-sized (9, 33, 61, 73, 80, 86, 88, 91, 104, 117, 139; see Map 42 and Pls. 9B–10B). Offsite pottery density was high (Map 43). In the Gournia survey zone, a number of sites (11) reached village size (Table 14). Population density became relatively high in this period. Settlement— between Vasiliki, Monastiraki, and Episkopi—was concentrated in areas that had access to water. Site 9 received its water via an aqueduct (Pl. 9B); site 61 may have also. A few Hellenistic sites (23, 88, 114, 139) continued. Although the Hellenistic sites were located at the southern half of the Isthmus (with the exception of site 23), in the Early Roman period settlements expanded all the way to the Pacheia Ammos and the Tholos harbors. Two new sites (17, 21) appeared in the vicinity of Halepa (23), with the obvious goal to participate in north-coast trade. It is unclear whether they were

all functioning simultaneously—possibly not. The ceramic evidence from Halepa (23) does not go beyond A.D. 100, while the large Pacheia Ammos building (21) dates to the 2nd century A.D. The nature of the settlement at Alatzomouri (17), west of the village of Pacheia Ammos, is unclear, but the ceramic evidence suggests an occupation from Early Roman to the 3rd/4th century A.D. West of the Pacheia Ammos Bay, the site of Hagia Pelagia (9), south of Gournia, presents solid evidence for a larger, village-sized settlement. During the Roman period there is evidence for houses (walls), an aqueduct that is still visible today (Pl. 9B), considerable amounts of imported pottery, and a massive trapetum (oil press) and millstones. The construction of a port facility at Tholos led to the occupation of the Kambos Plain in the Kavousi region during the Roman period. Haggis (1996a, 2005) has interpreted this facility as primarily for the transshipment of goods from the north to the south coast and vice versa. The establishment of a village at Kamina (61), at the southern end of the Kambos Plain in the Early Roman period, should be viewed as another hub of the north–south coast commercial network, but a nearby massive oil press (Pl. 10A) points to local production for export. The significant number of transport amphorae found in the site (Map 44), as well as the imported pottery (Maps 45, 46), also testify to local participation in the production of goods for export. After Kamina (61), the next Early Roman village to the south is at the Church of Hagia Marina near Monastiraki (73). At a small distance (less than 500 m), another village (80) developed concurrently with that of Hagia Marina, taking advantage of its surrounding fertile environment. To the south, the next village-sized settlements (86, 88, 91, 104) are clustered near the middle of the Isthmus Plain. Sites 86 and 104 are situated on the plain, adjacent to a riverbed. Site 86 is a large production center of transport amphorae (the overall size of the site is unknown because it falls outside the boundaries of the Gournia Survey). This site may have used the clay beds south of Vasiliki village. Site 104 is situated on the north and south sides of the riverbed about 1 km south of 86. Sites 88 and 91 are located at the base of Mount Thriphti, near the Church of Hagios Georgios. Slightly larger, site 91 preserves house walls, heavy terrace walls, as well as architectural elements

HELLENISTIC AND ROMAN PERIODS: EXPANSION OF THE ISTHMUS IN AN INTERNATIONAL ERA

(a column, blocks, and a threshold). The location of these four sites on the Isthmus suggests that they may have developed as pairs (86 and 104; 88 and 91), to guard both sides of the north–south route and facilitate an important midway stop. Moving south, the survey recovered evidence for two Roman villages (117, 139) at a fair distance from each other. Site 117, on a flat spur on the slope of Thriphti just north of the Church and spring of Ephendis Christos, was built on arable land overlooking the north–south road. The next village (139) was more than 3 km away, west of Kato Chorio. In addition to the 10 villages, there is evidence for a number of hamlets or farmsteads developing in the orbit of these villages, such as sites 46 and 56 in the vicinity of the village at Kamina (61) or hamlets 113 and 114 near the village at Ephendis Christos (117). Almost all of the villages that were established on the Isthmus in the Early Roman period continued to be occupied throughout the Late Roman period, without any significant hiatus (Map 47). Furthermore, the Gournia Survey produced evidence for an increase in the number of sites during the Late Roman period, especially in the area around the middle of the Isthmus (41, 71, 97, 102, 103, 107, 125). The development of villages 41 and 71 may have followed the pattern of building sites in pairs, in order to control traffic on both sides of the north–south route. In the case of site 41, the red soils of the valley and the brown colluviums near the river must have played an additional role in the establishment of the site. In addition to an increase in villages and hamlets, the Late Roman period also witnessed the foundation of several farmsteads in the middle of the Isthmus (85, 87, 92, 97, 103, and 110). In sum, the settlement pattern on the Isthmus throughout the Roman period is shaped by the north–south route connecting Hierapytna with the harbors of Pacheia Ammos and Tholos. The density of Roman sites shows some significant exceptions, which become apparent when compared with the Protopalatial (Map 20) and Neopalatial (Map 25) settlement patterns. Roman settlements in areas along the coast are restricted to four sites; settlements in the plain north of Monastiraki are conspicuously fewer and spaced farther apart. One wonders if this is due to a lower population or to

85

landownership. The large unsettled area around site 21 could indicate an estate. Proximity to arable land and water sources is not unrelated to the choice of location, with a considerable number of sites developing along the streambeds. Settlements on the flat plain, particularly south of Monastiraki, are more numerous than at any other time period, although the preferred location continues to be slightly higher, on hill slopes of about 200 m altitude. The continuous and extensive occupation of the Isthmus and the Kavousi area suggests agricultural and pastoral exploitation as well as a role in the transshipment of goods.

Subsistence “Consumption in [Roman] Crete depended directly and heavily on what it could export,” wrote Harrison (1998, 132) a decade ago, based on Pliny’s long list of Cretan goods that reached Rome and Italy (Chaniotis 1999, 219–220, for a list of ancient sources mentioning Cretan herbs and medicinal plants). This list praised Crete’s wine, candied quinces, and lumber, as well as a number of medicinal plants (Harrison 1993, 116–117). Could the Roman desire for a certain number of Cretan products, however, be interpreted as evidence for agricultural intensification and agricultural specialization on the island during the Roman period? Most scholars would accept that the economy of Crete before the advent of the Romans was based on subsistence (Chaniotis 1999, with criticism by W.V. Harris in the same volume). The sociopolitical organization of the Cretan poleis, which evolved around the Doric syssitia or common meal, presupposed the holding of communal lands for cultivation and pasture within a state-controlled environment, leaving little room for entrepreneurship. Although this archaic socioeconomic structure could have continued its existence into the Hellenistic period, the descent of the Cretans to coastal sites in the 3rd century B.C., their involvement with mercenary service, as well as their exposure to maritime trade, must have influenced the socioeconomic structure, “giving way to a more typical form of city-state life” (Bintliff 1997, 23–24). It is hard to believe that the Cretans switched into a market economy without

86

NATALIA VOGEIKOFF-BROGAN

any previous exposure to or experience of it. The seeds must have been planted earlier in the Hellenistic period. For example, changes in the settlement pattern on the plain of the Mesara indicate social and political transformations already from the early Hellenistic period. The large rural expansion of Phaistos has been associated with a growing lower class that cultivated land intensively in order to respond to new market opportunities (Watrous, Hadzi-Vallianou, and Blitzer 2004, 448–449). One widely accepted rule of thumb among archaeologists and surveyors is that the more traces of human activity in the landscape, the more intensively the land was exploited (Alcock 1993, 82). Indeed, the establishment of new sites on the Isthmus and in the Kavousi region in the Early Roman period, with several of them on valley bottoms (21, 28, 42, 46, 61, 104 in the Isthmus region, and 1, 5, 37, 53 in the Kavousi region), could be interpreted as evidence for agricultural exploitation (Haggis 2005). The settlements on the Isthmus are probably connected to the growing prosperity of Hierapytna. These sites produced food for the asty, surplus for export, and facilitated the land transportation of goods from the north coast to the south and vice versa. Roman sites produced copious evidence of agricultural activity (Maps 43, 44): work basins, storage jars/pithoi, amphorae, and querns. Most sites produced multiple local amphorae, and several (21, 32, 56, 86, 150), without evidence of ceramic production, yielded many examples (7–11). These numbers suggest the production of wine and oil. The large storage facilities at Tholos, Sta Lenika, and possibly at Pacheia Ammos were probably built to offer storage for export goods and tax collections. The warehouse at Tholos is a long building (55.70 x 9.60 m), divided into four compartments inside, with 10 supporting buttresses on the exterior of its east side. The type of brickwork and tile-faced concrete supports a date in the 2nd century A.D. Haggis, following Boyd’s earlier hypothesis, has suggested that the Tholos warehouse served as “a contingency harbor and temporary storage facility” in the Alexandria–Rome corn route (Haggis 1996b, 199, 205). According to Haggis, there is evidence for another warehouse (two buildings, one with a vaulted ceiling) at Sta Lenika, on a hill overlooking the southern Kambos Plain (Haggis 2005, 125).

A large, rectangular structure (21) at the northern end of the Isthmus by Pacheia Ammos Bay was excavated by Boyd in 1903. This large building, hardly recognizable today, was probably constructed in the 2nd century A.D. It has been interpreted as a temporary storage facility for goods traveling overland to Hierapytna through the Isthmus (Sanders 1982, 140). Harrison (1993, 191) has suggested that it functioned as “a seaside way station for goods and people moving along the north coast.” One problem with these hypotheses, however, is that there is no good harbor near the site. Recently, in reevaluating the original plans of Boyd’s excavations, Kelly identified a plunge bath, a pool, and a praefurnium (furnace/heating room), all attached to the east side of the building and fed through a complicated system of an aqueduct and connecting cisterns. The aqueduct is still traceable all the way to Vasiliki, tapped to a source either at Episkopi or more likely from the Cha Gorge (Kelly 2006). The inclusion of a bath at the Pacheia Ammos building, the relatively few imports found there, and the lack of surrounding sites support its identification as the villa center of an estate. In the Vrokastro region, most sites produced insufficient evidence for Early Roman occupation, except for a few places that continued from Greek to Roman times (Hayden 2004b, 271–272). Writing about the Vrokastro region, Hayden (2004a, 189) suggested that “different areas of the island [had] individual histories.” In contrast to the Vrokastro area, the prosperity of the Isthmus seems to have stemmed from its active contribution to and involvement in the transshipment route across the Isthmus.

Viticulture I have argued extensively elsewhere about the possible involvement of Crete in the Mediterranean wine trade as early as the Hellenistic period (Vogeikoff-Brogan et al. 2004; Vogeikoff-Brogan and Apostolakou 2004). The publication of the transport amphorae from the sites of Trypetos (Siteia) and Mochlos, dating to the early 2nd and early 1st centuries, respectively, add archaeological substance to Polybius’s mention of Cretan wine (Hist. 6.11a.4), as well as to an inscription of the 3rd century B.C. that refers to several hectares of vineyards at Kydonia (Marangou-Lerat 1995, 11–13).

HELLENISTIC AND ROMAN PERIODS: EXPANSION OF THE ISTHMUS IN AN INTERNATIONAL ERA

In the area of the Isthmus, the best evidence for wine production and possible trade in the Hellenistic period comes from a small number of stamped amphora handles from the 2nd century B.C. that were found in Alexandria, Callatis, and Trypetos. Their stamps indicate Hierapytna as their source of origin (Marangou-Lerat 1995, 123–124; VogeikoffBrogan and Apostolakou 2004). In addition, the discovery of transport amphorae made of clay from the Myrtos Valley and their distribution in the southern and northern ends of the Isthmus (at Myrtos/Pyrgos and Mochlos) strongly suggest that they were products of Hierapytnian pottery workshops (VogeikoffBrogan, Nodarou, and Boileau 2008). The Gournia Survey has produced evidence that Hierapytna continued to contribute to the flourishing viticulture of Crete during the Roman period. Many survey sites yielded amphora fragments (Maps 45, 48). The presence of an amphora workshop in the middle of the Isthmus (86) should be associated with nearby vineyards and farmsteads (85, 87) and exploitation of agricultural resources (Marangou 1999, 270–271). Although largely destroyed by the modern asphalt road and cultivation, site 86 yielded wasters of amphorae (Type AC1) and other types of plain pottery, as well as overfired wall pins and slags. Isolated wasters from other sites (74, 103, 104) suggest that there were more workshops of transport amphorae on the Isthmus. Marangou-Lerat’s extensive survey of Crete has identified at least seven places producing amphorae of Type AC1 and its variants, dating to the beginning of their production in the second half of the 1st century A.D., without excluding that variant AC1 might have commenced earlier, in the first half of the century (Marangou-Lerat 1995, 74). The wide distribution of AC1 amphorae in the Mediterranean, including places like Rome and Ostia, matches well with the literary popularity of the sweet Cretan wine passum in the Roman Empire (Marangou-Lerat 1995, 15–29; Marangou 1999, 278).

Olive Cultivation The survey of the Isthmus has also produced evidence for olive cultivation (Map 44), with the discovery of pressing equipment in several sites (9, 46, 56 [Pl. 10A], 66 [Pl. 9C], 97, 117 [Pl. 10B]). In

87

addition, significant examples of pressing equipment dating to the Roman period have been collected through recent fieldwork in other parts of Crete (Hadzi-Vallianou 2004). Although the oil trade is difficult to trace, the large size of some of the presses (46, 66) suggests that oil production exceeded subsistence levels. In sum, six oil presses were found in the survey zone. Large-size pressing equipment is usually associated with increased manpower and labor, as well as with wealthy proprietors investing in agricultural intensification and specialization. Several Roman authors (Cato, Varro, and Columella) discuss the agricultural workloads and manning ratios of the large estates in Italy (Duncan-Jones 1974, 327–333; Thompson 2003). Large tracts of unsettled land around sites 21 and 61 may be the remains of estates. The large olive presses on the Isthmus may have served large or small-scale producers and produced oil for export.

Pastoralism Based on the abundance of ancient literary, epigraphic, and archaeological evidence for breeding animals on Dorian Crete, as well as the passage by Diodorus Siculus (5.65.2) that describes the breeding of cattle and sheep as the Cretan contribution to civilization, it is safe to assume that Cretans carried their great pastoral tradition into Roman times. The isopolity treaties of the Hellenistic period between Hierapytna and neighboring cities indicate specialized pastoralism in the area of East Crete, which developed as a response to the political/demographic crisis of the 3rd and 2nd centuries B.C. (Chaniotis 1999, 210–212). The fact that grazing rights were considered so important as to be the focus of several isopolity treaties indicates the economic significance of pastoralism for Hierapytna. It is, therefore, possible that the pastoral goods of Hierapytna were intended for more than subsistence, but to produce surplus for exchange (Forbes 1995; Nixon and Price 2001, 407). Several vows by shepherds or cattle owners to the Kouretes, from a passage in Vitruvius (De arch. 1.4.10) referring to intensive cattle breeding in the plain of Mesara, and an account from the sanctuary of Diktynna in West Crete with implications for organized shepherding and production of wool, constitute our sole evidence for specialized

88

NATALIA VOGEIKOFF-BROGAN

pastoralism in the Roman period (Chaniotis 1999, 210–212). The occupation of upland locations (such as Tzamachi, Schinavria Koriphi, and the Ephendis Christos cave) in the mountains to the west of the Isthmus throughout the Roman period, suggests that the population continued to practice pastoralism (Hayden 2004a, 200–201, 205). According to Hayden, the settlements located on Tzamachi and Schinavria communicated with the Isthmus through Asari, east of Sopata (Hayden 2004a, 200, and fig. 4 for routes and toponyms). The survey of the Ziros highlands, to the east of the Isthmus, has also shown extensive occupation of upland locations in the Early Roman period (Branigan 1998). At this time, the occupation of upper elevations should be viewed more as an increase in pastoral activity than a flight to the hills, as had occurred in the Early Iron Age (Hayden 2004a, 205). There is no concrete evidence, except for the occasional loomweights and spindle whorls found in the survey (9, 88), for pastoralism in the area of the Isthmus during the Roman period, although the political and administrative unity that followed the Roman conquest would have made transhumance of large flocks more attainable. The citizens of Hierapytna and her catchment could have taken their herds as far as the highlands surrounding Meseleri to the west and the Ziros Mountains in the east, the way their Hellenistic forefathers grazed their sheep on the highlands of Priansos and Praisos, but without being constrained by state boundaries. In addition to transhumance, recent work on ancient pastoralism (from the Archaic to the Roman period) has indicated that animal husbandry could have been integrated with farming activities (Forbes 1995).

Beekeeping The Gournia Survey yielded a small number of plain pottery sherds with interior scoring (at 61, 67, 71, 88), which have been identified as beehives (Maps 45, 48). The scarcity of the archaeological evidence may be attributed to the fact that the survey zone included only a small area of mountainous terrain. We know little about beekeeping and honey production on Crete in the historical periods. The Sphakia Survey in western Crete has produced by far the largest assemblages of ceramic beehives

known on Crete (Francis 2006). The discovery of a dedicated structure (the so-called Beehive Enclosure) of the Roman period at the site of Ergasteria in central Sphakia indicates systematic production of honey and wax. Roman authors like Pliny (HN 21.46) and Dioscorides (2.101) were aware of the superb quality of Cretan honey. Francis has associated Tarrha with the production of honey because of the proximity of the city to the Sphakia area and her coinage, which carried the bee. The bee appears on the coinage of several other Cretan cities in western Crete (Elyros, Hyrtakina, and Lissos). In eastern Crete, only Praisos (and possibly Hierapytna) struck her coins using the bee as a motif. The bee has been struck on the reverse of two bronze coins, which have been attributed to Hierapytna based on the abbreviation “I/E” that appears together with the bee. The bee also appears on the handle of a stamped Hierapytnian amphora found in Egypt, and on the handle of a possible Hierapytnian amphora found at Mochlos (Marangou-Lerat 1995, 123–124; VogeikoffBrogan 2004, 216; Stefanaki 2006, 311). Although the bee as a coin motif could be indicative of a city’s involvement in honey production, the bee should be viewed in the context of both sides of a coin. In the case of Praisos, the bee appears together with deities like Apollo or Demeter or Persephone, whose cults are connected with the bee and its products. Concerning Hierapytna, Stefanaki has suggested that the use of the bee on her coins and stamped amphorae could be associated with the conquest of Praisos by Hierapytna (ca. 145 B.C.); thus, the adoption of the bee could be seen as a political statement by Hierapytna (Stefanaki 2005, 177–182).

Fishing Not much is known about the role that fishing played in the subsistence of the Cretans during the Roman period, although recent archaeological work at Itanos and Eleutherna, as well as Hellenistic Kommos, have produced evidence for the fishing and consumption of inshore species (Mylona 2003, 2007). The discovery of a number of fish tanks on the eastern part of the island, both on the north and south coasts (Chersonissos, Mochlos, Siteia, Ferma Ierapetra, and Zakros), not far from the area of the Isthmus, could indicate a more systematic method

HELLENISTIC AND ROMAN PERIODS: EXPANSION OF THE ISTHMUS IN AN INTERNATIONAL ERA

of fish exploitation. Without proper excavation and contextual information, the purpose of the Cretan fish tanks and the time of their construction remain elusive. They have been dated to the Roman period, solely based on information from Latin authors like Columella and Varro, who describe breeding or conserving fish in “piscinae.” Although their appearance should be associated with a renewed interest in fish exploitation, most of the Cretan fish tanks, except for the Siteia ones, represent small-scale ventures (Mylona 2007, 92–94; contra Harrison [1993, 193–194; 1998, 132], who suggests that the size of some of these tanks could have serviced more than their local communities). The Siteia tanks are large and numerous, suggesting a more prosperous enterprise. The best evidence for the exploitation of marine resources in East Crete comes from an inscription of the 3rd century B.C. (IC 3.6.7), which refers to a decree regulating the relations between Praisos and Stalis, according to which coastal Stalis was allowed to keep a fraction of the income in purpledye and fish, derived from the port of Stalis, but also from “the islands” (Chaniotis 1996, 388–393, no. 64; Viviers 1999, 225). Most scholars place Stalis at modern Makrygialos, some kilometers east of Ierapetra (Perlman 1996, 257–258). Assuming that Hierapytna took over the former possessions of Praisos, it is almost certain that it also controlled the income from the two nearby islands (Leuke and Chryssi). The Stadiasmus Maris Magni informs us that Hierapytna had the advantage of exploiting the harbor of the islet of Chryse (Stadiasmus = Periplus 319–320; also Stefanaki 2005, 25–26 n. 62).

Exchange Whether or not the economy of Crete in the Hellenistic period was based solely on subsistence, or had already shifted to a market economy (as argued above; see sections on Subsistence and Viticulture), the conquest of Crete by the Romans in 67 B.C. placed the island within the orbit of the Roman economy. It is widely attested that Crete participated in Roman trade, facilitated the transshipment of goods, and exported some of its own goods, wine and herbs in particular.

89

Large amounts of imported pottery found on Roman sites in the survey area point to the commercial connections of the area with the outside world, specifically with southern Italy, Asia Minor/Eastern Aegean, Syria, and North Africa. John Hayes recognized the following Early Roman imports at our sites: 4 South Italian amphorae, 16 pieces of Italian Sigillata from Italy (Arezzo), 6 examples of Eastern Sigillata A from Syria, 5 Eastern Sigillata B dishes from Asia Minor (Tralles), 13 Chandarli Ware pots from Asia Minor (Pitane), 11 Eastern Aegean cooking pots, and 2 or 3 North African amphorae (Table 15; Map 46). During the Early Roman period, over half (22 of 30) of the Early Roman sites yielded imported pottery, making it clear that the region was closely connected to the outside world at this time. Imports are concentrated in villages (8 out of 11) and hamlets (6 out of 15), rather than farms (1 out of 10). Haggis (1996b, 204–205; 2005, 87) has suggested that the warehouses near Kavousi functioned primarily as storage facilities for the transshipment of goods across the Isthmus to Rome. What this hypothesis does not explain is why Hierapytna, once it finally controlled East Crete, would want to reject the Hellenistic sea route (via Itanos) for a land-based movement of goods (requiring offloading and on-loading) across the Isthmus. Within our survey zone, North African pottery reaches many (11 of 30) of the Early Roman sites. Their absence along the north coast, however, is notable. On the other hand, a string of sites, from 150 in the south to 56 in the north, where North African pottery was found, extends north–south along the eastern edge of the Isthmus. Italian pottery was also found in 13 of the 30 Early Roman sites; most is concentrated in the northern half of the survey zone. Half of the sites that produced African wares (6 of 13) also yielded Italian pottery. These two types of pottery could be the result of goods transshipment and/or local exchange. Imported pottery within the Isthmus may have been partly the result of the transshipment of goods, but the pattern of imported pottery across the Isthmus tells a more complicated story. The most common imported pottery found in the Isthmus came not from North Africa or Italy, but from the Aegean and Asia Minor. Fifteen sites (out of 30), evenly spread across the Isthmus, yielded pottery

90

NATALIA VOGEIKOFF-BROGAN

imported from sites in the Aegean and Asia Minor. Much of this pottery probably arrived at the Isthmus as part of direct exchange between local settlements and sites in the Aegean and Asia Minor. Many settlements in the agricultural heartland of the north Isthmus (as indicated by the distribution of off-site pottery; see Maps 43, 44) had access to imported pottery, also suggesting local production was connected with the foreign exchange that brought imported pottery. Foreign amphorae were found at 20 (out of 30) Early Roman sites (Map 45). These transport jars would have originally arrived in the Isthmus containing goods, but they may well have been saved for export of local produce, a common practice in Crete by the Roman period. Their distribution among hamlets (12), villages (7), and farms (1) in the same areas that exhibit the most intensive land use (i.e., the highest off-site pottery density; see Maps 43, 44) strengthens this connection. The presence of at least seven massive trapeta in our area points to the importance of oil production. Mattingly (1988) and Hitchner (1993) have described and interpreted the establishment of oil presses on small farms in North Africa during A.D. 100–300 as a similar sign of surplus oil production for export within the economy of the Roman Empire. Additionally, it cannot be a coincidence that all sites where a trapetum (or the accompanying millstones) were found also possessed imported pottery. The amounts and types of imported pottery in the Isthmus increase in the Late Roman period. Hayes identified 82 examples of Phocean Red Slip, 37 pieces of African Red Slip, 10 North African amphorae, 5 Cypriot Red Slip ware, 3 East Aegean/Asia Minor cookpots, 1 Syrian mortarium, 1 Egyptian lamp, and 1 Egyptian amphora (Maps 48, 49). Fine-ware and cookware imports rise from 72 in the Early Roman period to 113 in the Late Roman period. The number of imported amphorae increases from 18 (Early Roman) to 26 (Late Roman). In the Late Roman period, 19 sites produced imported fine wares or cookwares. Thirteen Late Roman sites yielded imported amphorae (as opposed to 10 Early Roman sites [Maps 45, 48]). The Late Roman distribution of imported pottery represents an expansion over the Early Roman pattern. Imports rose in all site categories: villages (10 out of 14), hamlets (8 out of 17), and farms (4 out of 10).

In the Late Roman period, foreign connections with the Eastern Mediterranean increase (Maps 48, 49). Connections with Asia Minor, the Levant, and North Africa grew closer. Imports from the Aegean and Asia Minor also grew (from 15 to 19) and were more densely spread across the north Isthmus. The distribution of these imports closely correlates with the incidence of off-site pottery (cf. Maps 43, 44), likely a sign of agricultural intensification. In the Late Roman period, Italian imports disappear, while North African imports, especially the number of amphorae (2 Early Roman, 10 Late Roman), increase. African imports generally occur in more sites (11 Early Roman and 15 Late Roman) and became more numerous (21 Early Roman and 36 Late Roman). Imported vases from the northern Aegean or the Black Sea were new in the Late Roman period. The discovery of a number of road inscriptions to the north of Hierapytna, referring to repairs of highways during the reign of Emperor Claudius, bears testimony to the development of a major road network for wheeled vehicles with a northeast–southwest direction, probably already established from the time of Augustus (Baldwin Bowsky 2006a, 565–567). Rome’s interest in developing the road network of Hierapytna should be associated with the prime position of the city in regard to Mediterranean trade routes and its access to north and south harbors through the Isthmus corridor. Cretan wine is widely mentioned and praised in the Roman sources (Chaniotis 1988, 72–73). Short texts incised or painted on Cretan wine amphorae provide information about Cretan participation in Roman trade, with Lyttian amphorae traveling to Pompeii (Chaniotis 2008, 90). Survey in the Isthmus produced evidence for production of Type AC1 amphorae at site 86. Pliny’s long list of Cretan medicinal herbs, as well as information from other sources (Chaniotis 1999, 219–220), suggests that the export of herbs from Crete became an important activity in Roman times, partly controlled by the emperor. Dioscorides, Pliny, and Galen refer to the import of Cretan herbs to Rome. Galen’s text, in particular, leaves little doubt about the existence of special herbal gardens on Crete administered by specialists in the service of the emperor (Kühn, 14, 9–10). Very little is known, however, about the

HELLENISTIC AND ROMAN PERIODS: EXPANSION OF THE ISTHMUS IN AN INTERNATIONAL ERA

location and administration of the imperial farms in Crete. The best evidence comes from a small number of inscriptions from Crete, dating from the time of Nero to the second half of the 2nd century A.D., which record the title procurators Augusti. Rouanet-Liesenfelt has suggested that the imperial property resulted after Octavian’s confiscation of the land that Antony handed to Cleopatra as a wedding gift in 37 B.C. It is very possible that this land was located in East Crete, around Itanos, the only part of Crete that had been exposed to strong Ptolemaic influence during the previous centuries (Rouanet-Liesenfelt 1992, 175–176). Rouanet-Liesenfelt proposed, however, to extend the limits of the imperial land as far as Lyttos to the west, based on a comment by Galen that the plain around Lyttos (Πεδιάδα) produced the best herbs in the Roman period, without excluding that other areas of Crete could have supplied the Roman emperor with herbs (Rouanet-Liesenfelt 1992, 184, 187–188). Galen’s description suggests that the Roman emperor imported large quantities of Cretan herbs to cover not only Rome’s needs but those of the provinces as well. The export of salt might have been another possible source of income for the inhabitants of Hierapytna in the Roman period. An inscription of the 2nd century A.D. at the Archaeological Collection of Ierapetra refers to the existence of a corps of special officers, the αλωροί, whose main task was to collect salt and safeguard the salt lakes/mines (Davaras 1980, 1–8). The construction of a special structure for the αλωροί of Hierapytna indicates that the extraction and exploitation of the salt was an organized activity in Roman Hierapytna. In addition to the shores of Hierapytna, the island of Chryse may have been amenable to salt extraction. The geomorphology of the island is very close to that of the island of Gaudos, to the south of Gortyn, which produced large quantities of salt and cedar in the Roman period (Davaras 1980, 3; also IC 4.184). Finally, the work of Paton and Schneider has showed convincingly that Crete played a significant role in the imperially protected marble trade of the 2nd and 3rd centuries A.D. (Paton and Schneider 1999). The abundant presence of imported marble on Roman Crete, both in public and private buildings, suggests that the local elites were rich and

91

sophisticated (Paton and Schneider 1999, 291). In the case of Hierapytna, Paton and Schneider have collected evidence indicating that the city made extensive use of marmor Proconnesium and Claudianum for its public buildings. Furthermore, the grand and luxurious villa at Makrygialos, a few kilometers to the east of Hierapytna, used large quantities of imported marble, even for its tiles (Paton and Schneider 1999, 288–289). Finally, Hierapytna is the only place on Crete, so far, that produced evidence for using imported marble earlier than the 2nd century A.D. Ian Sanders witnessed the excavation of a large stone structure at Viglia with numerous marble columns found together with pottery of Augustan date (Paton and Schneider 1999, 288 n. 37; also Davaras 1974, 50). If true, the association of imported marble with Augustan pottery constitutes further evidence that Hierapytna subscribed to the Roman culture and ideology as early as the 1st century A.D., as marble, especially colored, became a distinctive symbol of Roman architecture, culture, and atmosphere (Paton and Schneider 1999, 296). The only evidence for the use of marble within the Isthmus comes from site 91, which yielded several architectural members (a column, blocks, and a threshold). Harrison has argued that the “magnificence and elegance of Roman Crete were paid for privately from what wealth the island could plant and harvest” and that its consumption depended “directly and heavily on what it could export” (Harrison 1998). Harrison’s statement presupposes extensive land exploitation, for which we have evidence on the Isthmus, in the form of a dense settlement pattern, intensive off-site land use (Maps 43, 44), and the presence of imported pottery found at many sites (15 out of 30 in the Early Roman period; 24 out of 38 in the Late Roman period). The presence of Cretan amphorae found overseas, especially in Italy (Marangou-Lerat 1995), points to the same conclusion. Hierapytna was a major stop on the Alexandria–Ostia trade route. Our survey evidence indicates that this route had at least two branches: along the south coast of Crete, and overland across the Isthmus to points north and west. As we have discussed above (see the section concerning Subsistence), the economy of Hierapytna and the Isthmus relied on a combination of export and transshipment of goods.

92

NATALIA VOGEIKOFF-BROGAN

Social Organization Although we do not know exactly when Crete became a Roman province, scholars would agree that the process of Romanization did not start immediately after the conquest. In fact, the question one should ask is “when [did] Crete decide to accept its status as part of the Empire and to pursue what benefits could accrue?” (Harrison 1998, 129). The material Romanization of Hierapytna, as well as its highest prosperity (two theaters, one amphitheater, and even a rare naumachia), largely dates to the 2nd and 3rd centuries A.D. (Sanders 1982, 57–59, 139; Papadakis 1986). Baldwin Bowsky’s study of onomastics from Hierapytna has demonstrated that many Hierapytnians had developed a Graeco-Roman cultural identity (and enjoyed Roman civitas) by the 2nd century A.D., with the family of the Flavii Sulpiciani Doriones being the most Romanized group (Baldwin Bowsky 1994). There is evidence, however, suggesting that in Hierapytna’s case, the spread of Roman influence began shortly after her Roman conquest. Baldwin Bowsky has argued that the two inscriptions from Episkopi and Kato Chorio (IC 3.3.62–63) naming Augustus must be associated with the emperor’s interest in the reorganization of the road network within the Isthmus (Baldwin Bowsky 2006a). The onomastic evidence also supports the pre-Flavian recovery of Hierapytna. The epigraphic record preserves evidence showing that the city’s elite adopted the triumviral nomen “Antonius” and the early imperial nomina “Iulius” and “Claudius” (Baldwin Bowsky 2006a). To this, we could add the large public building with red-veined white columns excavated on the west side of Ierapetra in 1972 near the larger theater (Davaras 1974, 50), which Ian Sanders has dated to the late 1st century B.C./early 1st A.D., because of the presence of Arretine pottery in the excavation (Sanders 1982, 139 n. 5). Other evidence for early prosperity comes from the silver coinage that Hierapytna issued in the reigns of Tiberius (A.D. 14–37) and Caligula (A.D. 37–41), together with eight other Cretan cities (Stefanaki 2005, 392–397). According to Baldwin Bowsky (2006a, 571), “As the only city in eastern Crete to mint silver coinage, Hierapytna’s coins should have been the dominant currency of this part of the island.” All of the above evidence suggests that

Hierapytna had access to imperial patronage as early as Gortyn, for reasons that must be associated with her strategic location at the south end of the Isthmus and her role in the transhipment trade. In addition to imperial patronage, the city of Hierapytna also cultivated and benefited from local patronage. The study of the onomastics shows that Gortyn, Knossos, Lyttos, and Hierapytna shared a number of Roman family names that originated in all likelihood from Italian traders (Baldwin Bowsky 1994, 28–29; 2006b). Even more interesting is the fact that only Cretans from Gortyn and Hierapytna reached the Roman equestrian or senatorial status. Moreover, the Hierapytnian elite maintained strong ties with the rest of Roman Greece through offices in the Panhellenion (L. Flavius Sulpicianus Dorion and T. Flavius Xenion were Panhellenic archons from A.D. 161 to 169). To Baldwin Bowsky (1994, 32–36), Hierapytna “was one of the most Romanized cities of Crete, second only to Gortyn.” Old institutions like syssitia do not seem to survive long after the Roman conquest. A reference to dianomai (charities) in a Lyttian inscription (IC 1.18.11) dating to the end of the 2nd/beginning of the 3rd century A.D. should not be associated with syssitia but with charitable actions of the prosperous members of Cretan society. According to Paluchowski (2005), the Cretan poleis developed into “république[s] de notables” during the Roman period. This oligarchic system, very different from the egalitarian character of the previous centuries, is evident in the Lyttian inscriptions, where the names of the magistrates are repeated without following any rotating system (Paluchowski 2005, 427). For example, the governance of Lyttos, during Classical and Hellenistic times, was based on the annual rotation of the four phylai (tribes). More than half of the Lyttian magistrates are related, which indicates that the power was in the hands of a few powerful families, and not in the phylai. This “république de notables” is best illustrated in an inscription of the 2nd century A.D. from Hierapytna, recording a long list (more than 30) of individuals/members of the assembly, who ratified the erection of a number of statues (IC 3.3.7). The fact that there is no evidence for any social unrest on the island of Crete during the Roman period strongly suggests that the new political system replaced the traditional Cretan ways of social welfare (syssitia)

HELLENISTIC AND ROMAN PERIODS: EXPANSION OF THE ISTHMUS IN AN INTERNATIONAL ERA

with other forms of wealth distribution, like charities/dianomai (Paluchowski 2005, 438). In addition to the lost Roman buildings, the wealth of the city is also evident in its art (mostly sculpture that was removed from the city in modern times and is now in foreign museums [Papadakis 1986, 46–54]). Most of the sculpture was found at the site of Viglia, to the west of modern Ierapetra, where various travelers of the past centuries described elaborate Roman buildings, like the large theater (also recently identified as a bouleuterion; see Balty 1991, 549). The bronze statue of a youth (Herakleion Museum, no. 2677), the so-called Kore of Ierapetra (Archaeological Collection of Ierapetra, no. 833), a marble statue of Artemis with quiver and strap (Archaeological Collection of Ierapetra, no. 63), and the exquisite peplos-clad statue of “Demeter” are among the several statues that have been unearthed at Viglia by salvage excavation (Raftopoulou 1975; Apostolakou 1980; Kane 1985, pl. 51; Papadakis 1986, 84–85; Ridgway 2002, 124–125). It is very likely that most of them formed part of the sculptural decoration of the city’s large theater or bouleuterion. In addition to statues of deities, Balty and Thomas have proposed that the statues of L. Flavius Sulpicianus and his extended family, the most prominent Roman family of Hierapytna, also figured at the bouleuterion, together with statues of the imperial family from the Antonine period (Balty 1991, 549; Thomas 2007, 154). In fact, an impressive and colossal statue of Hadrian treading on a barbarian, now at the

93

Istanbul Archaeological Museum, was taken from Ierapetra to Constantinople in 1870 (Hannestad 1986, 200–201; Papadakis 1986, 51; Hallett 2005, 119). Despite the onomastic evidence, the material record for Roman influence during the early years of the Roman conquest remains slim in East Crete. Referring to the Vrokastro Survey, Harrison reports that the “opening decades of the new millennium are represented by a single Arretine sherd” (Harrison 1998, 131–132; Hayden 2004b, 272). The development of the Vrokastro area, however, was different from that of the Isthmus after the Roman conquest. The settlement pattern shows limited continuity from the Late Hellenistic to the Early Roman period, with only a small number of sites surviving the transition. With the exception of coastal Priniatikos Pyrgos, the rest of the settlements were inland and produced cooking and plain wares, suggesting farms and hamlets. Little imported pottery was recognized by the Vrokastro Survey. Priniatikos Pyrgos developed into an important harbor later, during the middle Roman period (150–425 A.D.), because of the increased exploitation of the fertile hinterland (Hayden 2004a, 211–212). By contrast, the Isthmus and the Kavousi regions developed rapidly, with an upsurge of settlement in the Early Roman period, following abandonment in the Greek period. Imports were widely distributed within the Isthmus during the Early and Late Roman periods.

Part III

Conclusions

12

Conclusions L. Vance Watrous

We conclude by considering one of the main focuses of this study, the development of social complexity in the Mirabello region that culminated in the formation of a Minoan polity centered at Gournia. The subject of Minoan state formation has been warmly debated since the publication of Renfrew’s book The Emergence of Civilisation in 1972. The main theories are listed below in rough chronological order of their appearance: 1) Diffusion: A. Evans (1921–1935) and Childe (1947) attributed increased social complexity in the Aegean and Europe to powerful influences from the Near East. 2) Managerial Chiefs: Renfrew (1972) posited that regional economic networks grew up in Early Minoan centers, such as Knossos, that required a managerial authority, a chief, who used his economic power to become a palatial ruler. 3) Neoevolution: Branigan (1988) and Warren (1987) viewed the Minoan palaces as the inevitable result of incremental population (Boserup 1981) and economic growth over time.

4) Social Storage: Halstead (1981) argued that Early Minoan chiefs used the storage of food and the lending and exchange of surplus as a basis of power that enabled them to become state rulers. 5) Peer-Polity Interaction: Renfrew and Cherry suggested that the regional communities of Central Crete developed common cultural features, such as palaces and peak sanctuaries, through the process of mutual emulation, competition, and the exchange of ideas and goods (Renfrew and Cherry, eds., 1986; Cherry 1986). 6) Wealth Finance: Manning (1994, 1996) hypothesized that elites transformed themselves into state rulers through the control and aggrandizement of foreign prestige goods. 7) Stimulus Diffusion: Watrous (Watrous, HadziVallianou, and Blitzer 2004) suggested that one member of the elite class convinced others to make him ruler through the use of a new religious ideology derived from Egypt.

98

L. VANCE WATROUS

What all these theories have in common is that they seek to explain Minoan state formation by focusing solely on the first Minoan palaces at Knossos, Phaistos, and Malia. But what the present study shows is that the developments leading up to the establishment of a Minoan state, centered at Gournia, differ substantially from those attendant to the first palaces. For example, the palace at Gournia was ushered in on a wave of settlement destructions and abandonments, but there is little or no evidence of regional warfare taking place at the time the first palaces were constructed. Nor is there any real basis for the tacit assumption among scholars that the three first palaces were all somehow the result of the same sequence of events. This disparity between theory and archaeological fact becomes more obvious if one turns to the subject of state formation in Early Iron Age Crete. In fact, the theories listed above hardly relate to Early Iron Age Crete at all: hence, the conclusion that the processes of state formation varied with time and place. Another complicating factor is that the cultural changes leading up to state formation were rarely

simple and were spread over a relatively long period. For the first palaces, each phase—EM II, EM III, MM IA, and MM IB—exhibits different developments that contributed to the establishment of the Minoan polities. Early Iron Age state formation is also the product of changes in the LM IIIC, Protogeometric (PG), and Geometric periods and 7th century B.C. Most of the above theories relate to only one phase of this development or one aspect of cultural change (see below), and hence, lack the power to offer an overall explanation of how a specific state actually arose. The processes leading to Minoan state formations were long and complicated; a fact that precludes any single theory of social complexity from being completely explanatory. For this reason, this study concludes by reviewing the chronological sequence of developments leading to the formation of a Minoan state at Gournia in the light of the processes suggested by the theories above.

Final Neolithic Final Neolithic settlement within the Mirabello region consisted of camps, farms, and hamlets situated on low slopes along the border of arable valleys in the immediate proximity of a water source. These sites, located some distance from one another, appear to have been small independent communities of no more than one to four families. Excavated Final Neolithic sites elsewhere in Crete suggest that these small communities lived off a combination of

hunting, gathering, farming, and seasonal pastoralism. Brian Hayden’s 1995 study of hunter-gatherer groups has shown that their subsistence strategy is based on sharing, interfamily exchange, cooperation, and diverse economic activities, and that their society is characterized by complexity and individual status, but it probably lacked a class structure or a developed social hierarchy.

Early Minoan I–II Early Minoan I–II is a period of dynamic social growth. Settlement hierarchy developed with some communities becoming village-sized. Specialfunction sites, for knapping and ore smelting, appeared. Settlement expanded across the countryside; many smaller farms and hamlets were founded near good arable land, suggesting the increased importance of agriculture. In the EM I–II period,

subsistence strategy expanded, diversified, and became more intensive, providing a basis for the sharp rise in population growth. The largest settlements (e.g., Priniatikos Pyrgos, Gournia, and Mochlos), however, grew up along the coast. Priniatikos Pyrgos, Gournia, and site 23, as well as Vasiliki (135), sit on the agriculturally productive colluvial marl slopes with no surrounding sites,

CONCLUSIONS

which mean they have large, rich catchments. This situation contrasts with the dense settlement pattern of small, closely spaced farms and hamlets farther inland (south of Priniatikos Pyrgos and between Monastiraki and Episkopi), some of which occupy marginal environments with poorer soil and less arable land and water. This difference between land-rich and land-poor settlements suggests the emergence of economic inequality. Here, the neoevolutionary argument (e.g., Warren 1987; Branigan 1988) seem, in part, to be pertinent: diversified agriculture led to settlement growth and increased population. Excavated EM II settlements, especially Mochlos, produced a wealth of imported material, such as Cycladic and Egyptian stone vases, a Syrian cylinder seal, silver and lead from Siphnos, and daggers alloyed with Near Eastern tin (Soles 2005). Copperbearing ore from the Cycladic island of Kythnos was crushed and smelted at Chrysokamino to make local copper (Betancourt 2006). Large amounts of obsidian imported from Melos reached the coastal sites, especially at Mochlos, Gournia (8), and Halepa (23). By the EM IIB period, architecture and tomb groups provide clear evidence of social stratification. The two large two-storied houses (Zois’s Red House) at Vasiliki, set next to an open paved court, dominate the surrounding smaller rubble houses on the site. These two households possessed singlestory ancillary structures, identified as “servants’ quarters or storerooms” by the excavator (Seager 1905, 210). Given their location immediately adjacent to the probable production center of Vasiliki Ware, it seems likely that the Red Houses controlled

99

this important export industry. High-quality redpolished Vasiliki Ware produced within the Isthmus was traded throughout most of Crete (Day 1997; Day, Wilson, and Kiriatzi 1997). At Mochlos, two or perhaps three tomb groups (I/II/III, IV/V/VI, and Lambda) stand out from the rest of the cemetery by their size and their rich assemblage of elite and imported items: gold jewelry, bronzes, and stone vases (Soles 1988). A tomb at Gournia also had gold and silver items (Soles 1992, 13–16) that may date to this period. It is at this point that the strategy of the managerial theory of Renfrew (1972) may be relevant. What we know of EM IIB Vasiliki and the production of Vasiliki Ware fits Renfrew’s description of a managerial authority. Finds that are associated with Vasiliki and Mochlos suggest the presence of attached and specialized craftsmen. Stone vases were manufactured at Mochlos and Pseira. The communities at Vasiliki, Mochlos, and perhaps Gournia seem to have been headed by two or more separate family groups that are distinguishable by their large landholdings and their privileged access to precious materials, international connections, and craft production. These families appear to have converted their agricultural wealth and association with crafts and trade into elite status. The archaeological facts here correlate with the scenario of wealth finance as suggested by Manning (1996). The elevation of these few families at local centers into chiefly positions of economic and social power may be linked to the subsequent outbreak of destructions evident at local sites toward the end of the EM IIB period.

Early Minoan III–Middle Minoan IA The end of the EM II period was characterized by the abandonment and destructions of sites by fire in our region (Phournou Koriphi, Myrtos/Pyrgos, Vasiliki, and Malia). Following the desertion of the Cyclades at the end of EC II, certain signs of Cycladic influence or presence appear along the north coast of Crete. The clearest example comes from the cemetery of Hagia Photia, where most of the local pottery, among other finds, is Cycladic in nature (Betancourt 2003; Davaras and Betancourt 2004). Cist graves, a Cycladic type, and pithos

burials, a western Anatolian type, appear locally at Sphoungaras, Pseira, and Pacheia Ammos, as well as at many other north coast sites. The local hallmark fabric for this period is Haggis and Mook’s Fabric XX/XXI, which is distinguished by many inclusions of angular crushed quartz. As Day and Wilson (2006) have pointed out in their discussion of a Cycladic presence in EM I Crete, this ceramic technique can be traced to the Cyclades. During EM III–MM IA, the number of sites within the Mirabello region decreased dramatically.

100

L. VANCE WATROUS

Overall, the settlement pattern indicates both continuity at centers and major discontinuities in rural areas. Concern with security seems to have been important: coastal sites in particular were abandoned, and new sites—e.g., Azoria—appeared in defensive locations. Massive perimeter walls guarded several of these sites, e.g., Vasiliki, and sites 131 and 144. A few centers, notably at Gournia and Mochlos, seem to have survived fairly intact amid a general population drop. Immigrants must have been assimilated into these centers, while some new rural sites appear to have been established by newcomers (see Ch. 5). Intersite distances between rural settlements increase sharply, suggesting relative isolation and perhaps hostility between site clusters. Seager’s excavations at Vasiliki give us some specific details of what happened to the site during the EM III–MM IA period. Seager (1905, 218) reported that after the EM IIB destruction of the site, the succeeding (his Period IV) inhabitants built their “hovels” over a small portion of the site. By MM IA, however, it regained its former size, included Seager’s Houses A and B, Zois’s Houses Gamma, Phi, and X, as well as a massive circuit wall and tower. The sequence of events at this time can be reconstructed as follows: initial destruction and abandonment of sites in EM IIB, Cycladic immigration in EM III, followed by a recovery beginning in MM IA. Trade connections with the Near East reappear in the MM IA period. Seals were made from ivory imported from Syria and/or Egypt (Krzyszkowska 1988). Industry shows a rise in craft specialization, visible in new types of stone vases, ivory seals, and bronze objects from tombs at Mochlos and Gournia. Some of the gold, ivory, jewelry, and stone vases from the Mochlos tombs date to this period

(Watrous 2005a). A number of Egyptianizing stone vases, jewelry, and seals in the tombs were probably locally made. New types of religious paraphernalia occur in some tombs. Society became markedly more hierarchical. Evidence of newcomers and a new settlement pattern consisting of site clusters point to a polarization of the population along ethnic lines. Simple individual burials began to be made at local cemeteries, often in marginal areas occupied by newcomers. At the same time, new types of elite burials appear at the centers. At Gournia, the cemetery has a single tomb (Tomb II) distinguished from the others by its large size and built altar (kernos) for public ceremony. Earlier skulls were purposely saved in Tomb I (and perhaps in Tomb II), suggesting a reverence for ancestors. At Myrtos/Pyrgos, a large rectangular house tomb with two attached ossuaries built on the top of the Pyrgos hill attracted elite male burials. The tomb was reached by a long paved processional road, flanked by another paved space, ending in a kernos. This arrangement is paralleled by the ramp and platform at Mochlos outside of Tombs IV/V/VI. These monumental displays of ancestor worship would seem to be local versions of the contemporary Egyptianizing Chrysolakkos tomb at Malia and Building 6 at the Phournoi cemetery, which show new public funerary ceremonies associated with a single elite social group within each of these communities. Egyptianizing funerary architecture and objects suggest that the local elite had also begun to assimilate foreign (as Helms 1988) ideas that helped to distinguish them from other sectors of society. This adaptation of foreign funerary ideas and customs is a form of stimulus diffusion (Watrous, Hadzi-Vallianou, and Blitzer 2004, 265–276).

Middle Minoan IB–II An enormous expansion of settlement takes place in the Mirabello region during the Protopalatial period. Overall, there is a three-fold increase in the total number of sites from the Late Prepalatial period. Settlement in the Gournia survey zone consisted of three main site clusters: around Gournia, around site 60, and north of Episkopi. In the Kavousi area, the

settlement pattern also consisted of a series of decentralized site clusters. The density of settlement and the diverse new environments exploited indicate a sharp intensification of land use. A number of new sites were established on marginal locations, occupied by a land-poor population. Each cluster, perhaps an extended kinship group, is separated from

CONCLUSIONS

the next by an expanse of unoccupied land, suggesting these groups were partly independent, if not competitive and hostile. Excavated settlements exhibit greater social hierarchy. At Gournia, the Protopalatial House He is appreciably larger than the other Protopalatial houses. At Myrtos/Pyrgos, the MM IB settlement

101

grew sharply, and an elite building was erected on the hilltop. As in the preceding period, simple pithos and jar burials and elite tombs point to social stratification. The poor burials within the region may have been for newcomers as well as commoners, and the elite tombs seem to distinguish old families as separate from the immigrant population.

Middle Minoan III–Late Minoan I The Protopalatial period ends with the destructions and abandonment of several sites in the Mirabello region. Pseira, Mochlos, Myrtos/Pyrgos, and perhaps Priniatikos Pyrgos suffer destruction by fire in MM II. At Vasiliki, House Gamma was violently burned down. The settlement on Pseira was deserted in MM IIB. This sequence of events seems tied to the MM II reoccupation of refuge settlements such as Katalimata (Nowicki 2001, 2008). Who caused the destructions at Pseira, Mochlos, and Vasiliki? We know that the eventual winner was Gournia; hence, it is tempting to suppose Gournia was a major aggressor. Following these events, a palatial structure was built at Gournia in the Neopalatial period. Excavation at Gournia in 2011 has shown the initial construction of the palace dates to MM IIIA. The large storage wing of this structure suggests that it had a redistributive function. The storage capacity of the palace also implies that personnel with specialized functions (e.g., craftsmen and craftswomen) received rations. Gournia became a major production center of pottery (Hawes et al. 1908, pls. VI–IX) and exported pottery to Akrotiri on Thera, Pseira, and perhaps to Mochlos (K. Bernard, pers. comm.). During the Neopalatial period, a peak sanctuary was established on Mount Ephendis Stavromenos, the highest mountaintop visible from Gournia. Gournia may have moved the population on site 1 to Pseira, following the destruction of the settlement at Pseira. Settlement became nucleated around the site of Gournia during this time, leaving it the only settlement on the north coast of the Isthmus. If viewed through kinship lines, the changes in the Mirabello early Neopalatial settlement pattern suggest the establishment of a new dendritic relationship in which the elite families at the major settlements in the region formed part of a large corporate group

(Blanton et al. 1996) whose head resided in the new palace at Gournia. How did Gournia manage to become a state center in the Mirabello region? Several scholars (Betancourt 1984; Soles 1992, 255–258; Soles and Davaras 1996, 180 n. 8) have constructed a continuous (uniformitarian) ceramic sequence for the local EM II–LM I period and on that basis have argued in the neoevolutionary vein for an unbroken cultural continuity between EM II and LM I for the Mirabello region, in which the Minoan state must have directly evolved from the local EM II centers of power (“chiefdoms”). If this were true, however, we could expect that Mochlos would have become the palatial center of the region in the Neopalatial period. But this did not happen. Instead, Gournia took control of the Mirabello region in the Neopalatial period. The destructions of Mirabello sites during the MM IIB–III period suggest that Gournia achieved palatial status through armed force. Evidence of a military presence certainly exists at Neopalatial Gournia, i.e., 2 swords, 14 daggers, 1 spearhead, as well as 3 double axes. The wider context for these events in the Mirabello region is significant. During the MM II–LM IA periods, both Knossos and Malia (Watrous 1996, 47–52; 2005b, 142–145, table 1) were forcibly expanding their territories in Central Crete. In this respect, Gournia seems to have been following the example of the other powers on the island. This form of emulation might be likened to peer-polity interaction (Renfrew and Cherry, eds., 1986; Cherry 1986), but the interaction in Central Crete and in the Mirabello area took place between groups of different size and social complexity— palatial states on the one hand and village-sized communities and their allies (“chiefdoms”) on the other. More importantly, this interaction does not

102

L. VANCE WATROUS

explain how Gournia actually became the center of a polity. The archaeological evidence from our area suggests that the reality of state formation may have been much grimmer than the theory of peerpolity interaction would suggest. State formation in the Mirabello region involved violence, warfare, and the movement of people. The events that unfolded in the MM II–LM I Mirabello region resemble “chiefly cycling” as defined by Wright (1984) and Flannery (1999). This cultural process involves long-term competition between chiefly families or factions with varying complexity (i.e., chiefdoms with two levels of hierarchy = “simple;” and chiefdoms with three levels of hierarchy = “complex”), who recurrently compete for resources, raid, expand territory, and fragment. This cyclical competition can occur over centuries with little or no increase in

sociopolitical complexity, until eventually, a state emerges (when one chiefdom successfully conquers its neighbors) or the result of a regional collapse. In such a process, similarly sized chiefdoms on the verge of state formation may have a dispersed population or groups separated by buffer zones between one another as in the MM I–II Mirabello region. Next, sudden and rapid nucleation of population occurs at one center, usually at the expense of its neighbors (who are defeated through raiding or warfare), always resulting in territorial expansion (Carneiro 1981). This is what seems to have happened in the Mirabello region during the MM II–III period on the eve of the appearance of a Minoan palace at Gournia. State formation in East Crete seems to have been very different from the process that brought about the first palaces at Knossos, Phaistos, and Malia.

Part IV

Appendices

Appendix A

Catalog of Sites L. Vance Watrous

Site entries consist of site number, (field number when applicable), local toponym if known, transect location, physical description, site size, (number of artifacts collected), artifacts by period, site date and type, and bibliography. Abbreviations include:

MM = Middle Minoan NP = Neopalatial PG = Protogeometric PP = Protopalatial

ARS = African Red Slip

RS = Red Slip

BG = Black Glaze

SM = Subminoan

BSJ = bridge-spouted jar

SS = straight-sided cup

EC = Early Cycladic

T = Turkish/Ottoman Period

EH = Early Helladic

V = Venetian Period

EM = Early Minoan ER = Early Roman FN = Final Neolithic H = Hellenistic LM = Late Minoan LPP = Late Prepalatial LR = Late Roman

1. (1) Metochi tou Nikolaou, Fields 26, 34 and 35 Located roughly 700 m northwest of Gournia at the west end of the valley, this large PP settlement is located at the base of an east-facing slope and looks across the valley to Gournia at the eastern end of the valley. Recent bulldozing has exposed the granodiorite bedrock and destroyed much of the

106

L. VANCE WATROUS

site. Site catchment was probably the western half (divided by the Gournia River) of the valley. Traces of ancient walls are extant among recent terraces on the slope. The site is partially hidden from the sea (and winter winds) by the ridge of Palaiobarda. Small quantities of MM IA sherds were recovered. PP size: 200 x 150 m. Artifacts (Figs. 11:B135, 12:B147, 14:B187, 16:B216, B217, 17:B231, 18:B237, B241–B243, B255, 19:B263, 20:B273, B275, B280, B285, 21:B300, B301, B305, B311, 22:B317, B318, B324, 23:B327, B329–B331, B334, B338, B340, B341, 24:B343, B345, B346, B348, B351, B352, B355, 25:B357, B361, B362, 26:B370, 27:B373–B375; Pls. 16:B60, B65, 18:B78, 20:B102, B106, B110, B111, 23:B134, B135, B144, 25:B163): fine ware: much dark ground, striped decoration, BSJ, amphorae, hole-mouth jar, large bowl, many SS and Vaphio cups (MM III), and carinated cups; large handmade bull figurine (est. length 20 cm) painted a red monochrome, conical cups (MM II). Coarse ware: body sherd with cross incised on exterior, phyllite and granodiorite wares, spherical and oval loomweight, thin and thick oval tripod legs, many pithoi, heavy basins, BSJs and jars; tub, tray, pithos lids, bathtub(?), 5 water channels, mudbrick, cooking ware. Eight obsidian blades, 1 obsidian chip, chert chunk, many stone hand axes (choppers); whetstone, ground stone. V–T sherds. Murex and topshell. Larnax fragments south of site. MM IA hamlet, MM IB–II village, MM III (LM IA?) presence, V–T presence. Possible cemetery to the south.

2. (2) Gournia Moon, Field 2 On a north-facing slope roughly 200 m north of site 1, this settlement is ca. 250 m south-southwest of a small bay (where the Gournia Moon camping ground is today). PP size: 20 x 60 m. Heavy terraces are based on a granodiorite and limestone bedrock. The site overlooks a small valley that may have been its agricultural base. Artifacts (Figs. 12:B156, B160, 13:B167, B168, 14:B184, 20:B272, 21:B310): EM: base; PP: lid, large jar, plume jars, carinated and SS cups, jug, hole-mouth jars, amphora, cookware; Iron Age, Archaic: pithos, cup; Classical: BG fragments. MM IB–II farm, Archaic–Classical farm.

3. (3) Panagia, Fields 44, 56 and 57 Located in the elevated (100 m) southwestern corner of the Gournia Valley, 700 m west-southwest from Gournia, the site has an area strewn with sherds measuring 150 (north-northwest to southsouthwest) x 60 m. The site extends northward from the Church of the Panagia (and a nearby well) to the north side of a riverbed. Immediately north of the church are the remains of a small Venetian(?) metochi and massively built terraces; old olive trees along the slopes below are of the same date. The EM site consists of two concentrations of pottery: 80 x 80 m on the north edge of the river and 40 x 50 m around the church. Gentle slopes and river (bed) are nearby. Artifacts (Figs. 3:B22–B33, 5:B59, 19:B269, 20:B276, B284): black- and brown-burnished FN wares; EM wares; several obsidian blades, red chert blade and flake, black chert chunk debris and flakes, whetstone, ground stone; Venetian sgraffito wares, T sherds. FN camp site, EM I–II hamlet, V–T presence.

4. (4) Gournia Moon II, Field 60 Situated on a steep slope at the western edge of the small bay of Palaiobarda, this site consists today of a 40 x 40 m concentration of sherds. In the Minoan period, there may have been a single structure here, although there are no signs of ancient retaining walls or terracing. Artifacts: EM: 2 possible sherds, 21 pieces of chert, and 2 of obsidian; PP: coarse and fine wares; NP: Vaphio cup, BSJs, jug, pithos. EM I–II knapping site, MM IB–LM I farm.

5. (8) Palaiobarda, Field 59 Located on the top (probably) and highest southern slope of a steep hill (elevation 150 m) on the coast, the site is missing its upper portion due to rock quarrying. The remaining part of the site measures 17 x 40 m. Together with nearby sites 2 and 4, it may have used the land in the small valley to the east. Artifacts (268 sherds; Fig. 48:D46): PP: cup, coarse and fine wares; NP: amphora, basin, lid (NP?); Late Classical–Hellenistic: BG jug, cup, cookpot. MM IB–LM I farm, Late Classical–H farm.

CATALOG OF SITES

6. (5) Panagia II, Fields 51–53 On the red soil slope above and 500 m west of Gournia, next to a streambed, this newly heavily plowed site produced several concentrations of fieldstones (house walls?), much NP fine-ware pottery and as much chipped stone. Unlike the later site, a dense EM scatter of chipped stone extends over a wide area, roughly 200 x 200 m, and does not seem to have much accompanying pottery. This EM knapping area is associated with site 3. NP size: 45 x 75 m. Artifacts (Fig. 18:B239): EM I–II: 1 EM I marble/calcite ware, 2 red chert blades, 73 red and black chert flakes (some worked), 45+ chunks (debris), 3 obsidian blades and 4 flakes; PP: conical, carinated cup, hole-mouth jar, basin, cookware; NP: BSJs, many rounded, globular, Vaphio and conical cups, large spouted tub (for wine, NP?), basin, bowl, cookware, one LM IB(?) cup; LM III(?): kalathos. EM I–II knapping site, MM IB–LM I hamlet.

7. (6) Palaiobarda, Field 72 This site is situated on the edge of a high cliff (next to an outcrop of granodiorite) above the sea, 300 m west of Boyd’s “Shore House” at Gournia. Several ancient walls are visible, including a rectangular structure (16 x 12 m) around which was found much pottery and chipped stone and a large quern. During EM II, the site measured 60 x 40 m; in PP, 75 x 40 m. The coastal location and large number of obsidian indicate this community was involved in trade. Artifacts (Figs. 5:B50, 8:B94, 22:B313; Pls. 12:B14, 15:B49): EM I–II fine (EM I handle) and coarse wares; PP coarse (thin oval tripod leg) and fine wares (one SS cup, potter’s mark?); ground stone hand tool, chipped stone, obsidian, two ground stone tools. EM I–II hamlet, MM IB–II hamlet with a single structure at its center.

8. (7, 19) Gournia, Fields 75–77, 87–100 The site is located on a low ridge at the eastern end of a coastal valley, a kilometer west of the village of Pacheia Ammos. Gournia is situated to take advantage of the adjacent river and the protection

107

from winter winds afforded by the ridge of Pera Alatzomouri. Its inhabitants probably farmed the valley to the west. The upper portion of the town was excavated by Harriet Boyd in 1901, 1903, and 1904. The Gournia upper town may have been first settled in the EM IIA period. Early Minoan IIA material comes from Tombs III, V, and VI on the north spur of the ridge: Tomb I: pit, two vases, one a Mesara EM IIA import (Soles 1992, 13, nos. G1–G14). Tomb III: EM IIA vases, bronze awl, and fragment of sheet gold. Rock shelters V and VI: several vases including a jug imported from Knossos (Soles 1992, 38) found under room 59 (House Dh) in the town. Early Minoan II pottery was also found on the north spur of the acropolis on the terraces on the west side. Some of the burials, found there when the foundations for the present cement phylakeion (guardhouse) were dug (M. Perakis, pers comm.), may also have dated to EM II. In any case, the wide distribution of EM II sherds argues for a villagesized settlement. A small cluster of EM II sherds was also found by the coast about 50 m westsouthwest of the Shore House, perhaps indicating a field house. Boyd found Late Prepalatial(?) pottery in an “early house” northwest of House Ea, in the nearby North Trench; the survey noted LPP pottery northwest of Boyd’s excavation within the fencedin area, along the river west of the excavated area and north of the asphalt road some 100 m from the coast, which indicates that the settlement had grown considerably by this time. Cemetery data corroborates this view: Tombs I, II, IV, VII, and VIII were built at the end of the LPP period. Boyd found a large wall near the North Trench that she interpreted as a fortification wall of the early town. Boyd discovered PP pottery in fills along the north and west edges of the town and scattered in sectors C and D. Houses Aa and Ek date to the PP period. The valley immediately north of the Gournia ridge apparently was unsettled; Boyd’s trenches there revealed no ancient traces (Hawes et al. 1908, 56), and survey results were similar. The survey found much PP pottery northwest of Boyd’s excavation and along the river to the west of

108

L. VANCE WATROUS

the upper town as far north as the asphalt road, on the hill across the river, and a large concentration (160 x 100 m) by the coast. Boyd’s trial excavation in this area revealed ancient house and terrace walls as well as a street; the associated pottery is PP and NP. Protopalatial burials (including a pithos burial) continue in the North Spur tombs. Pithos and larnax burials were made at Sphoungaras and along the base of the Alatzomouri slope over a stretch of 300 m. By this period, the town had grown substantially from its LPP size to include the north hill and southern saddle on Pera Alatzomouri. In the NP period, most of the upper town exposed by Boyd was constructed. The palace was built in MM III (Soles 1991, 26) or LM I (Hawes et al. 1908, 24), and remodeled, probably in LM IB. The Shore House (Pl. 8C; Fotou 1993, 98, pls. XXI, XLVIII), a shipshed complex with two galleries, dates to the NP period. During the MM IIIA period, the North Spur cemetery went out of use, apparently replaced by the Sphoungaras cemetery. The NP town seems to have been smaller than its PP predecessor. One spot on the east slope of the north hill of Pera Alatzomouri, two clusters along the east bank of the river, and one on the west bank produced NP sherds. Scatters of NP sherds (including pithoi and fine ware) on the southern part of the Alatzomouri ridge top are probably also burials. Most of the terrace walls and a fortification wall by the coast appear to be NP in date. Even though the NP town can be assumed to have been built after the MM II earthquake, it shows no signs of agglutinative/organic growth. Rather, except for possibly Houses Eb:11 and 12, the houses were clearly built in relation to the street plan, the regularity of which suggests it predates the houses and was planned. Thus, at least two of the streets enter the central court at its corners (northeast and southeast). The NP history of the town is not uniform. Boyd noted that certain houses (perhaps Bb, Bc, Cl, and Ce, and El) and sector D went out of use earlier (LM IA?) than the rest of the town, although this has recently been disputed (Soles 2002; Watrous and Heimroth 2011). This pattern of abandonment may be related to the Theran eruption. Other houses (Cf, Ck, Cm, and Cc) continued in use until the LM IB destruction.

After the LM IB destruction, portions of the town were reoccupied. Boyd (Hawes et al. 1908, 23) noted in LM III that “a few scattered dwellings on the old akropolis were in use.” Houses He, Eh, and perhaps Hf were built, plus a small structure over Ei. The small shrine on the top of the acropolis, north of House Fk, dates to this period. Burials were made in the LM I settlement (House Ej, room 58), on the hill on the western edge of the riverbank, and probably in the two empty chamber tombs (now destroyed) found on the southern top of Pera Alatzomouri (Pl. 8B). Four LM IIIA–IIIB sherds were found west of the upper town along the river, and Boyd (Hawes et al. 1908, 20) claims to have found a LM III settlement across the river on the west side, though no LM III traces were noted by the survey. The latest pottery from the site dates to LM IIIB (Hawes et al. 1908, pl. X), including an LM IIIB:2 bowl base found by the survey, suggesting the date of the final occupation of Gournia. EM II–LM IIIB village, EM II–MM III cemetery. Hawes et al. 1908, 20, 23–24, 56, pl. X; Soles 1991, 26; 1992, 38; 2002; Fotou 1993, 98, pls. XXI, XLVIII; Watrous and Heimroth 2011; M. Perakis, pers. comm.

9. (9) Hagia Pelagia, Fields 101, 102, 104, and 105 South of Gournia in the area between the river and the Church of Hagia Pelagia is the Roman successor to Gournia. The area is thick with evidence of habitation: tiles, pottery, and a massive trapetum, and a possible kiln separator. Although immediately adjacent to Minoan Gournia, the Roman site only slightly overlaps the earlier site, and consisted of houses (whose walls are preserved) built along the ridge. In the Venetian–Turkish period, the site featured the Church of Hagia Pelagia and a kalderimi connecting it to the Panagia metochi (3) to the west. Boyd (Hawes et al. 1908, 20) mentions an inscription with a man’s name, tombs, and an aqueduct leading southward up the riverbank. Today the aqueduct can be seen some 50 m south of the church on the slope immediately east of the road. Artifacts (Figs. 50:D92, 58:E47–E49, 60:E64– E66, 61:E75, E76, E78, 62:E79, E82, E83, 63:E87): ER: one Chandarli Sigillata (stamped) dish, Chandarli Ware, Italian Terra Sigillata, Eastern Sigillata

CATALOG OF SITES

A, cookware; LR: Hayes Type 3 rouletted bowls, 12 Phocean RS dishes, 1 ARS form 20, cookpots, 1 amphora; Roman tiles, bricks, amphorae, cooking ware, loomweight, beehive, spirally grooved (6th–7th c. A.D., Byzantine?) wares; V–T glazed wares and waster. Six millstones, olive press. Roman–V–T (Byzantine?) village. Hawes et al. 1908, 20.

10. (10) Pera Alatzomouri and Sphoungaras, Fields 143–145, 150–153, and 158–161 This site occupies the northern portion of the long ridge of Pera Alatzomouri as well as its eastern and western slopes. The ridge top consists of a hill at its northern end from which a long ridge extends south. A house wall excavated by Boyd is visible on the western slope. Walls are visible on the ridge top. The lack of domestic sherds along the base of the ridge slope, and pithos burials found there, seem to indicate this was a physically separate site from Gournia. Originally its catchment may have been the land to the east, including the valley and hill of Alatzomouri. Boyd noted house remains on the part of the north hilltop (Fotou 1993, plan A). Hall claimed that Shelter I at Sphoungaras was occupied during the FN period (Hall 1912, 46–48). It may have been used for burials for a settlement above. Early Minoan pottery was found on the northern hill of the ridge and also on the southern hill. In the EM IIA phase, burials were made in at least two rock shelters at Sphoungaras; burials continue at Sphoungaras (Deposits A and B) in EM IIB and in the EM III–MM IA period. The survey recovered Late Prepalatial (EM III–MM IA) pottery just south of the northern hill (Cooper 2002, 9). Middle Minoan IB–II pottery was found on the north hill and the entire southern ridge. Middle Minoan III–LM I pottery south of the north hill and on the southern ridge may be from burials. Pithos burials appear over a wide area in Sphoungaras in MM IA and continue along with larnakes into MM II. Early Minoan II sherds were found on the hilltop and western and eastern slopes (and base) of the northern hill, indicating a village-sized settlement in EM II. Late Prepalatial sherds over a wider area

109

of the hilltop and western and eastern slopes indicate, likewise, a village-sized settlement in LPP. Protopalatial sherds (LPP–PP wasters) cover the hilltop and east and west slopes, evidence of a larger village in PP. Very few MM III–LM I sherds were found south of the northern hill and along the southern ridge. Some sherds, including PP and NP pithoi and jars, a larnax, and a snake tube, found south of the northern hill, may come from a cemetery. Settlement on the site may have ceased after MM II. Excavations at Sphoungaras produced cist (EM I–MM I), pithos (MM I–LM I), and larnax (MM I– LM III) burials. Final Neolithic–LM I finds include at least five certain FN sherds (Betancourt 1999, 36). The sandy inclusions of Hall’s “Neolithic” pottery imply an EM I date. The Cycladic-looking pyxis (Hall 1912, fig. 22:B) may date to EM I. Early Minoan IIB Vasiliki Ware consisted of jugs, cups, plates, and bowls; other EM II pottery includes lamps, a mug, and four potter’s rellas or polishers (Hall 1912, fig. 27). Other EM II artifacts include a gold ring, soapstone dish (Deposit A), stone rings (as Warren 1972, pl. 79:d, dibble sticks?), and bronze tweezers. Early Minoan III–MM IA pottery consisted of a jug (Hall 1912, fig. 22:G), cups (Hall 1912, 50; Cooper 2002, fig. 21), and a Vasiliki imitation tumbler (collection of the University of Pennsylvania Museum of Archaeology and Anthropology, Philadelphia, PA). Two gold heart pendants, two serpentine bowls (one miniature), a green steatite bowl, a breccia bowl, and two ivory seals (Hall 1912, 53) are probably MM I (Higgins 1980, 64). Middle Minoan–LM I pottery was common. Triton shells and skeletal remains were also recovered. Artifacts (Figs. 11:B137, B138, B140, 12:B149, B154, B159, 13:B161, B162, B164, B165, B172– B176, 15:B194, 16:B210, 18:B253, 19:B257, B261, 20:B271, B274, B282, B288, 22:B316, 25:B359; Pls. 16:B64, 17:B66, 19:B85, B89, B90, B98– B101, 22:B123, 23:B133, B138, B139, B146, B147): EM: fine (EM IIA fine gray ware, EM II red-slipped and burnished) and coarse ware. Late Prepalatial: a White-on-Dark Ware bowl, cup, jugs, BSJ, basin, trickle pithos, hole-mouth jar, tumbler; much MM IB–II fine ware (SS, carinated globular cups, jugs, amphora, BSJ, lamp[?], handmade snake tube), and coarse ware (large basins, BSJ, pithoi,

110

L. VANCE WATROUS

pithos lid, burnished tray, thin oval tripod leg), and stone “dibble stick” or weight. A potter’s wheel (similar to Hawes 1908, pl. VIII:33) was found. Neopalatial (MM III–LM I): fine and coarse (pithos, ribbed tripod leg). Wasters (one an amphora) from the northern edge of the north hill, and the second terrace on the eastern slope may indicate the presence of a kiln. Stone hand tool and several ground stone pieces. One perforated Classical(?) slab. Possible FN settlement, EM I/II–MM IA village, MM IB–II village. FN–LM I burials at Sphoungaras, MM III–LM III burials on southern ridge. Hawes 1908, pl. VIII:33; Hall 1912, 46–48, 50, 53, figs. 22:B, 22:G, 27; Warren 1972, pl. 79:d; Higgins 1980, 64; Fotou 1993, plan A; Betancourt 1999, 36; Cooper 2002, 9, fig. 21.

11. (12) Selli, Fields 116, 117, 121, and 122 Immediately south (about 250 m) of Gournia there are two steep hills. Sherds are distributed on the east and west slopes of the eastern hill; there is only a little pottery visible (except in patches of red soil) on the plowed marl soil of the olive groves on the hilltop, but the presence of sherds around the edges of the hill probably indicates the hilltop was also settled. A wall and a possible chamber tomb were noted on the east face of the site. The size of the settlement—150 x 100 m (perhaps two to four houses)—is calculated on this assumption. The catchment for this settlement was probably the poor land on the hilly slopes to the south of the site. One built rectangular terraced platform (for a house?) and rock-cut cavities (tombs?) were noted on the east slope. Artifacts (446 sherds; Fig. 12:B157; Pls. 19:B83, 21:B121): Mostly PP–NP (especially NP); LPP: EM coarse hole-mouth jar, bowl; PP: pithoi, basins, jug, carinated cups, jar, cookware; NP: MM III conical cup, LM I rounded cup, cookpot. EM III–LM I hamlet.

12. (13) Selli II, Field 128 On the western hilltop immediately south of Gournia, a dense patch of PP–NP sherds, 40 x 30 m (farmstead?), was collected just south of the summit of the hill. Two stretches of rough north–south walls (LM IIIC?) of massive fieldstones (not cyclopean

style) run along the western edge of the ridge to the north and overlook the ravine to the west (Fields 128–130). There is a light scatter of PP and NP as well as Ottoman(?) sherds to the east of the walls. The area (ca. 200 m2) of the marl hilltop is available for cultivation. Artifacts (201 sherds; Pl. 19:B86): PP–NP: fine and coarse wares, LM I cup. MM IB–II farm, MM III–LM I field site, V–T presence.

13. (14) Selli, Field 201 Some 300 m south-southeast of site 12 is a small concentration of sherds (20 x 20 m) situated at the edge of the ravine separating the two hills of Selli. This field house may have taken advantage of the small area of flat land with red soils on the saddle. Artifacts (140 sherds): PP: cookware, basin (beehive?), pithos, amphora; NP: jug, conical cup, coarse ware. MM IB–II field site, MM III–LM IA field site.

14. (16) Fields 188 and 197 On the next ridge east of Pera Alatzomouri, this site is based on a level shoulder high on the ridge (elevation 90 m) some 800 m south of the coast (400 m south Alatzomouri/Phylakeion [17]). Concentrations of sherds and walls are most visible in the areas of deepest plowing (for morelias). A small cluster of EM sherds indicates that the site was established in EM IIB. Two separate concentrations of LPP sherds in the southwest (13 x 12 m) and northeast (40 x 60 m) portions of the site may represent three houses. During the PP period the settlement expanded: sherds are continuously distributed over an area of 170 x 100 m. Slight traces of MM III–LM IA sherds (12 examples) in the northern part of the site represent the final phase of occupation. Artifacts (Figs. 4:B39, B42, B43, 6:B61, B67– B70, 7:B76–B78, B80, B82–B84, 8:B98, B100, B103, B104, B106, 9:B108, B113, B115, B116, B118–B123, 15:B202, 16:B212–B214, 31:B455, B456; Pls. 13:B25, B30, 14:B36, B39, 15:B46, B47, B53, B54): PP: cookpots, brazier, cup; NP: giant ribbed tripod leg, conical cup. A coin from 1926. EM II–MM IA hamlet, MM IB–II village, MM III–LM I field site.

CATALOG OF SITES

15. (17) Pacheia Ammos II, Fields 206–208 This small (ca. 100 x 60 m) site is located at an elevation of 100 m on the gentle eastern slope of the Alatzomouri/Phylakeion ridge just west of Pacheia Ammos. Located on a small flat area (like the mandri [sheepfold] located there today), this NP (MM III probably) farmstead may have used the flat area (300 x 200 m) and slopes immediately around it. Artifacts: Cookware, brazier, conical and Vaphio cup. MM III(–LM I?) farm.

16. (15) Pacheia Ammos I, Fields 212–214 On a low (elevation 35 m) rise about 200 m from the valley bottom and immediately south of the INSTAP Study Center for East Crete, the site sits on a south-facing slope above a small ravine (to trap water?). At the southern edge of the site, two cists cut into the bedrock may be graves or olive slots. The site (ca. 120 x 50 m) may have consisted of several houses (plus a cemetery?) that could have used the gentle slope above (200 m2) or the valley to the east. Artifacts: PP: pithos, cookware, basins, larnax; NP: coarse wares. MM IB–LM I hamlet and possible cemetery.

17. (18) Alatzomouri/Phylakeion/Pacheia Ammos, Fields 165–168, 179, and 180 This site is on the southern portion of the coastal promontory of Alatzomouri, in a low saddle hidden from the sea by the hill summit to the north. The modern asphalt road is immediately to the south. The distribution of LPP sherds is roughly 125 x 250 m, allowing for slope wash; the actual size of the settlement may be more like 125 x 150 m. Protopalatial sherds occur over an area of about 150 x 300 m; actual settlement size may be 150 x 150 m. Traces of NP sherds (eight examples) were found in one small area some 70 m south of the cemetery next to the Church of Hagios Dimitrios. Two tiny clusters of sherds on the north face of the hill about 100 m north of the church probably represent burials. Roman sherds were found in an area 100 x 50 m near the shore just west of the modern harbor and houses. A dexameni (cistern), 4 x 5 m, is cut in the bedrock and plaster-lined, with a water channel leading down to the sea cliff. Under the

111

Church of Hagios Nikolaos, inside the village of Pacheia Ammos, there is said to be a large rectangular (Roman?) building. Recent excavations by the 24th Ephorate of Prehistoric and Classical Antiquities have found LPP burials, and a MM IIB industrial area with murex shells (Apostolakou 2008; T. Brogan, pers. comm.). Artifacts (Figs. 11:B141, 16:B209, 17:B219, B220, B222, B223, B234, B236, 18:B252, 19:B270, 21:B299, B306, 22:B315, B320, B321, B23:B328, B333, B335; Pls. 20:B103, B104, B107, 24:B153, B155): EM–LM I fine and coarse wares: EM: burnished coarse, EM II red-painted sherd, cookpot. PP: SS cups, lid, pithoi, BSJs, amphorae, basin, pithoid jar, cookpot. NP: Vaphio cup, thick oval tripod leg. Stone hand tools, used obsidian flake. Roman: H/ER amphora, dish (3rd–4th century A.D.). EM III–MM II village, Roman hamlet. MM IB– LM I burials north. Apostolakou 2008; T. Brogan, pers. comm.

18. Pacheia Ammos III, Under present village This site is the pithos (larnax?) burial excavated by Seager (1910) along the littoral of Pacheia Ammos. Residents there tell of other pithos burials found under buildings from the El Greco Bar to the Golden Beach Hotel, an area of 100 x 40(?) m. This cemetery (Pls. 20:B108, 21:B120) probably served site 17 in the PP period and site 20 in the NP period. MM I–LM I cemetery.

19. Pacheia Ammos V, East of Field 180 On the slope about 80 m northeast of Seager’s House is a chamber tomb excavated by Alexiou in 1951. The tomb dates to LM IIIB and produced three larnakes, a stirrup jar, a stone bowl, and a bronze scale pan and pyxis containing beads and two gold rings. One of the two caves (northeast of Seager’s House) and on the seashore (under the church) may be the “Neolithic Cave” (see Alexiou 1954, 1967) mentioned by Boyd (Hawes et al. 1908, 46). In 1963 Sakellarakis excavated a LM IIIA:2–IIIB tomb south of the Seager House that produced three larnakes, a stirrup jar, cups, a jug, beads, and a sard seal stone (Kanta 1980, 143–144). Early(?) and LM IIIA–IIIB tombs.

112

L. VANCE WATROUS

Hawes et al. 1908, 46; Alexiou 1954; 1967; Kanta 1980, 143–144.

20. (63) Pacheia Ammos IV, Field 707 (the Zeus Hotel) This site lies underneath the village of Pacheia Ammos, immediately south of Seager’s cemetery (18). Sherds can be seen today in an area of roughly 100 x 50 m. Only the southern and eastern boundaries are relatively certain; modern construction prohibits an estimate of its total size. It may be considerably larger. Pithoi, large basins (larnakes?), and an alabastron may suggest the site includes the southern edge of the LM I cemetery (18). Protopalatial and NP cookware, however, indicates a settlement associated with the cemetery (18). Artifacts (Fig. 18:B246–B249): PP: carinated cups, basins (several quite large), lid, hole-mouth jar, cookware, pithoid jar, pithos; NP: lids, cookware, pithoi, banded and burnished rounded cup (LM IA?), burnished alabastron decorated with horizontal floral band (LM IA?). MM IB–LM I hamlet.

21. (47) Ellinika/Pacheia Ammos, Fields 677 and 678 The site is located at the eastern edge of the village of Pacheia Ammos south of the asphalt road and 50 m south of the coast. Boyd excavated a large Roman structure here in 1903 (Boyd 1904; Sanders 1982, 140, fig. 50). Walls and tiles continue some 80 m to the west. The portion exposed today on the eastern side of the site has a vault that may have been part of the substructure of a platform. Pottery is distributed over an area of 120 x 70 m. Traces of a Roman aqueduct running from Vasiliki to this site are still visible. The aqueduct consists of pieces of cement channeling along the slope of the eastern edge of the river. This aqueduct appears to be part of a system that included Vasiliki and Episkopi. Sanders (1982, 140) suggests that this site may have been a villa or warehouse like that at Tholos. Kelly’s (2006) identification of a bath at the building indicates the structure was probably a villa. Artifacts (Figs. 57:E36, 61:E77, 62:E84): Roman tiles, amphorae, cookware, pithoi, large jar, lid for a jar; ER: 2 double-handled amphorae,

Campanian amphora, 2 Chandarli Ware dishes; LR: 5 ARS dishes, 9 Phocean RS dishes, 1 Cypriot RS dish, 11+ amphorae, 1 North African amphora. Roman hamlet. Boyd 1904; Sanders 1982, 140, fig. 50; Kelly 2006.

22. (50) Kale, Field 509 Situated 40 m from the coast and just east of the main crossing of the Pacheia Ammos–Kavousi and Ierapetra roads, this site has been much destroyed by road building, German bunkers, and later houses. A scatter of EM II (red-burnished) sherds over the north part of the site may be the remains of perhaps two farms that were part of site 23. The PP site measures 70 x 70 m. Construction for the rear court of a modern house on the site has revealed two cuttings for kilns. One kiln with many wasters dates to pre–World War II and was made by itinerant potters from Thrapsanos. The other had PP sherds in it. Recent pottery includes tsoukalia (cooking pots), pithoi, and stamnoi (water jars). Wasters, pottery, and burned material (earth, slag) occur over a wide area. Artifacts: EM I–II red-burnished pottery; PP: hole-mouth jars, cookware, vat slab, pithoid jar, pithoi, lid/tray, basins (one internally scored, a beehive?). EM II and MM IB–II hamlet. Recent kilns.

23. (48) Halepa, Field 545 The site is located on a small promontory at the east end of the Bay of Pacheia Ammos. A chapel of Hagia Eirene (completed in 2005) is said to be a replacement for an older Church of Hagia Photini. The site has been damaged by sea erosion, a German bunker, quarrying, a tar and creosote spill, and use as a garbage dump. Walls are visible from the sea in the north-facing scarp, and at the base of the scarp there are EM I sherds. Recent excavations by the Greek Archaeological Service on the tip of the promontory have revealed house walls and EM I–IIA pottery. The remains of an earlier structure (church?) and a German bunker can be seen north of the church. There are several freshwater springs in the area and a pond in the riverbed immediately to the east. Cement walls of 19th–20th century apothekes for carob are cut into EM levels at the

CATALOG OF SITES

southwest border of the site. The EM I site measures 150 m east–west; it stretches from the coast to the riverbed. The southern boundary of the site is obscured, but EM sherds are found along the coast for 200 m south of the promontory, so it likely measures around 200 x 150 m. This may be the largest EM settlement in the northern Isthmus of Ierapetra. Artifacts (Figs. 4:B34, B35, 5:B51–B53, B56; Pls. 12:B9, B10, B16, 43:D81): Many pieces of obsidian (collected) and some worked chert indicate foreign contacts; stone tool. EM: predominantly EM I “marble/calcite” wares; several with granodiorite inclusions, one EH or EC pithos. Late Classical–H pottery comes from the central and western part of the site, including Attic BG cup, amphorae (probably of Hierapytna fabric and an imported example), cookware, BG lamp, kalathos, bowl (100 B.C.–A.D. 100), lamp, cookware, bowl, H amphora. EM I/IIA village, Late Classical–Roman hamlet.

24. (49) Fields 507, 508, 542, and 543 This site is located on the top of a low rise 50 m southeast of site 22, so it may have been the cemetery for that site. The main road to Ierapetra runs through the middle of the site next to the riverbed. Protopalatial–NP sherd distribution is 30 x 150 m in extent. Artifacts (786 sherds; Figs. 6:B65, 7:B91; Pls. 13:B21, 19:B92): EM (from site 23?): three burnished sherds, chert flakes; PP: cookware, drain, larnax(?), hole-mouth jar, pithoi, pithoid jars, waster(?), large jugs or amphorae, conical and carinated cups, fine ribbed jug (MM II?); NP: lid with two holes to strap to container, hole-mouth jars, jug. Roman: 12 fine and cooking-ware sherds. Six grape-sized, waterworn pieces of pumice. Recent stamnoi and tsoukalia. EM I–II cemetery, MM IB–LM I hamlet, Roman presence.

25. (46) Pacheia Ammos VI, Field 709 Set on the eastern slope above the riverbed and 170 m from the coast, 300 m southeast from site 20, the site (Pl. 6A) consists of a cyclopean terrace, rectangular in plan, probably the retaining wall forming an MM II house platform, and three concentrations of sherds: (a) in and around the terrace, (b) 20 m

113

south of the terrace, and (c) 50 m south-southeast of the terrace. The MM II house is set up against/on a low slope some 30 m from the river, so that the owner could farm that 30-m strip, probably irrigating it. The land between the terrace and the river (40 m) is the best land available; it is gray, silty, and rich (annual overflow of the river). Today it is the preferred spot for perivoliae (gardens); where irrigated, corn, squash, and other vegetables are grown. Carob, broom, reeds, fennel, and olives grow there too. In contrast, the rocky red soil on the hill slope above (east) bears only shrubs (such as thyme). This is prime land and explains why there is a line of sites along the river stretching to Vasiliki. Artifacts (882 sherds; Figs. 31:B432, B435, B438, B440, B442, B443, B445, B451, B457, B463, 32:B464; Pl. 27:B181): Per concentrations noted above, (a) PP cookware, pithos, pithoid jar, amphora, SS and carinated cups; NP coarse ware; (b) PP cookware, hole-mouth jar, pithos and amphora, NP coarse ware; (c) PP: oval-mouth amphora, hole-mouth jar, cookware, pithos, carinated, SS, and conical cups, NP: cookware, jug, foliate band and spiral cups. Concentration (a) may have been a MM II–III farmhouse, (b) a MM II structure (house?) for storage and possibly cooking, and (c) a MM II–LM IA house. MM II–LM I farm.

26. (52) Fields 654, 655, 713, and 714 This site is situated on the slope above (east of) the river and on the eastern slope of the river valley, 120 m south of site 25. Plowing on the slope has revealed three dense concentrations of sherds and wall debris 3–9 m2 within a larger scatter of EM II–MM IA pottery (35 x 105 m). Cluster A dates to EM II; cluster B to MM IA; cluster C to MM IA. Sherd distribution for EM I–I is 30 x 70 m; for MM IA it is 35 x 105 m. The presence of EM II sherds in the unplowed, higher eastern sector of the site on the ridge may indicate the settlement spanned EM IIA–IB. The site is situated to take advantage of the river and riverine land below and the slopes/hilltop above for different kinds of agriculture. Soils are red, overlying a conglomerate bedrock. Newly bulldozed terraces have destroyed any possible traces of the site to the south. Artifacts (Figs. 6:B62–B64, B66, B71–B74, 7:B75, B79, B81, B85–B90, 8:B95–B97, B99,

114

L. VANCE WATROUS

B101, B102, B105, 9:B109–B112, B114, B117, 10:B130, 15:B199, 17:B229, B230; Pls. 13:B19, B20, B22–B24, B26, B27, B32, B33, 15:B41, B43, B45, B48, B50, B52): EM I marble/calcite ware; EM IIA: gray ware; EM IIB Vasiliki Ware, cook dish; much EM I–II: dishes, cookwares, one Hagios Onouphrios import; LPP: calcite inclusion wares, cookware, monochrome washy reddish-brown decoration, basins, lid, pithos, hole-mouth jar; PP: cookwares, pithos, hole-mouth jar, thin oval tripod leg, basin, monochrome jug. EM I–MM IA hamlet, MM IB–II farm.

27. (51) Palaimandria, Field 737 This tiny site (field house?) consists of a 20 x 20 m scatter of PP and NP sherds on a level stretch of red soil at the base of the north face of Palaimandria. The site is 120 m west of the river but near the low slopes next to the river. Artifacts (52 sherds): PP: cookware, basin, stone hand tool (chopper); NP coarse and cooking pots. MM II–LM I field site.

28. (55) Field 716 Some 100 m south of site 16, this site is on the top and slopes of a ridge on the eastern side of the river. The site was founded in EM II; small amounts (total 12 sherds) of EM II pottery are found on top and along the river in an area 70 x 70 m. Heavy overgrowth and bulldozing prevented precise spatial estimates (except for the Roman period). In the MM IA period, the site is inhabited and measures roughly 70 x 70 m. In the Roman period, the site reached its maximum size, 100 x 70 m. A possible wall was noted on the ridge top and on the south slope. The catchment could have included the river valley below and the flat red soils above, to the east. Artifacts: EM I–II: marble/calcite ware, burnished and inwardly thickened rim bowl (EM II), red-burnished pithos(?); LPP: basins, pithos, cookware, jug; PP: cook dishes, ribbed tripod leg, Roman tiles; ER: five Italian Terra Sigillata dishes, two Eastern Sigillata A sherds, one Eastern Sigillata B sherd, amphorae, cookwares; LR: cookwares. EM I–MM II hamlet, Roman hamlet.

29. (105) Field 719 Located about 250 m south of site 28 and 100 m above the eastern slope of the river on the red soils of the Isthmus floor, this tiny (15 x 15 m) mound site represents a single EM I–IIA farm. The site is situated to take advantage of the river valley, the marl slope above, and the red soils of the Isthmus. Artifacts (187 sherds): EM I–II: EM I horned handle, EM I wiped ware, cook dish, EM IIA fine gray ware, large bowl, welk shell, red wash sherds, handmade phyllite ware. No EM I–IIA marble/ calcite ware. Large amounts of stone tools: ground stone hand tools, querns; worked chert. Used as a mandri in the 19th–20th century. EM I–II field site.

30. (56) Field 633 This site is roughly 300 m south of site 28 on the northern and southern slopes of a ridge overlooking the river. The concentrations of sherds measure 30 x 26 m (on the north slope, these perhaps represent two houses) and 30 x 30 m (on the south slope); both slopes are now terraced and planted in olives. The catchment is similar to that of site 28; marl soil is found on the slopes. Three blocks of a possible wall were noted on the north slope. Artifacts (342 sherds): Mostly LPP. EM II: sandy ware, bowl, basin, pithos; LPP: many hole-mouth jars, cookware (thin oval tripod legs), dishes, pithoid jars, pithoi, basins, burnished bowl, three fine cups. No Protopalatial. EM I–IIB two field sites, EM III–MM IA farm.

31. (58) Field 691 Located some 400 m south-southwest of site 30 on the western side of the river, this site sits on the gentle slope about 70 m above the river. There are terraces for olives and a dirt road running through the middle of the site. Sherd distribution is 25 x 15 m. Obsidian and pumice were also found. This small farm site was occupied in EM I–II, MM IA, PP, and NP. Its principal catchment probably was the marl land between the site and the river. Artifacts (Fig. 18:B250, B251): EM: EM I marble/calcite ware; dish, cookware, jar, burnished red and brown wares, one obsidian and black chert

CATALOG OF SITES

piece, pieces of lemon-sized pumice; LPP: an “EM coarse ware” lekanis, cookware; PP: pithoi, tumbler, basin, hole-mouth jar, pithoid jar, cookware, pithoi; NP: cookware, Vaphio cups (MM III?). EM I–LM IA field site.

32. (61) Field 609 About 200 m southeast of site 31, the site is located on a low bluff directly overlooking the river to the west. Large stone piles are what remain of a PP–NP and Roman structure (farmstead?) that could take advantage of the nearby river and the flat red soils to the east. Site size: 55 x 25 m and 40 x 10 m in the Roman period. Artifacts (729 sherds; Fig. 61:E74): PP: cookware, pithoi, hole-mouth jar, pithoid jar, cup; NP: cookware, MM III conical cup, Vaphio cup, BSJ; Roman: seven amphorae, tiles, ER bowl, cookware, pithos. Three Hellenistic sherds: BG bowl and cup. MM IB–LM IA farm, H–ER field site.

33. (57) Hagios Ioannis, Vasiliki. Sta Logenna, Fields 589–592 and 604–606 Located on the northern end of the ridge in which the village of Vasiliki is situated, the site looks down on the confluence of two streams, one passing east and the other west of Vasiliki. The gentle marl slopes to the east may have supported the site, as they do the village of Vasiliki (with irrigated gardens, olives, grains, mulberry, carob, and grapes). A unique feature of the site is the large hill of gray clay (lepidochoma) directly to the west. Several old (20th century?) kilns and wasters, and the vast amount of pottery on the site, suggest it may have included a potter’s establishment. The clay here was mined recently to make bricks and to seal the village roofs. A similar outcropping south of Vasiliki (Selli) still has one of its two kilns partly preserved. The pottery is spread over an area of about 250 x 250 m east and west of the Church of the Archangel Michael. Seager excavated part of an Early Christian basilica near here (Seager 1905, 207; 1907, 113). The EM I–IIA site is located north and east of the church over an area of about 90 x 40 m. Late Prepalatial and PP pottery is found north, east, and west of the church, in an area 90 x 90 m. In the Roman period, the site expands to around 250 x 250 m.

115

Artifacts (1,258 sherds; Figs. 51:E2, 54:E17, 57:E37; Pl. 43:D93): EM: EM I marble/calcite wares; LPP: hole-mouth jar, small bowls and 1 lekanis; PP: pithos, hole-mouth jar, cookware. Five Hellenistic sherds: unguentarium (2nd–1st century B.C.), BG jug, bowl, cup, amphora. ER: cookwares, amphora, 2 Italian Terra Sigillata, 2 Eastern Terra Sigillata A, 2 Chandarli Ware dishes, 1 dish imported from Pergamon(?), tiles, 3 basins, 3 ARS dishes (1 cookware), jug, 1 North African amphora and 1 local. LR: 2 ARS plates, 12 Phocean RS dishes, 2 Cypriot RS bowls, amphorae, 2 basins, jar, North Syrian mortarium (cooking bowl), cookpot (Asia Minor shape). EM I–MM II hamlet, LH–LR field site and basilica. Seager 1905, 207; 1907, 113; Zois 1972, 118; 1978, 73; Sanders 1982, 140.

34. (27) Fields 408 and 409 This small site comprises two small concentrations (8 x 8 m and 2 x 2 m) of PP–NP sherds. Artifacts (70 sherds; Fig. 33:B495): PP: mediumsized jar, fine BSJ, pithos, cookware; NP: basin, cookware. MM IB–MM III/LM I field site.

35. (39) Field 536 On a low slope facing west-southwest, this patch (30 x 30 m) of NP sherds (along with large nearby piles of stones) may have been a farmstead. Artifacts (307 sherds; Fig. 33:B499): NP: cookware, pithoi, large jar, conical cup, spout, amphora, large BSJ. MM III–LM IA field site.

36–38. Vacat 39. (54) Fields 585, 586, 596, and 597 Located 200 m east of and above the river on the floor of the valley and roughly 200 m north of the site of Vasiliki, this site could have taken advantage of the colluvium along the river and the red soils adjacent to it. It consists of two concentrations of sherds (15 x 50 m and 10 x 35 m) separated from each other by 30 m.

116

L. VANCE WATROUS

Artifacts (182 sherds): North scatter: PP and NP cookware, MM II carinated cups. South scatter: PP: pithos, pithoid jars, cookware. The absence of fine ware and the types of vases in the southern scatter may indicate it was a cemetery. MM IB–MM III/LM I farm. MM III–LM I possible burials.

2 plain bowls, 18+ amphorae, 2 large plain plates (1 with pierced base for hanging), pithoi, 1 rouletted (3rd–5th century A.D.) plate. Quern. Roman: bricks, tiles. ER: 1 Italian Terra Sigillata, 1 Adriatic amphora, 1 ARS dish, 1 local amphora, cookware, and a basin; LR: 4 Phocean RS plates, 1 basin, local amphora, and 2 imported amphorae. LR hamlet with kiln.

40. (53) Fields 570 and 580 Located on the slope about 100 m west of the river, this site is at the edge of the red soils of the valley and the brown colluvium near the river. One stretch of large blocks in a terrace wall may mark the location of this tile-roofed Roman farmstead. Sherds are distributed over an area of 50 x 80 m. Artifacts (417 sherds): Roman: tiles, four (one large) amphorae, basin, cookware, many large pithoi, one BG (1st century A.D.?); ER: one Eastern Sigillata A plate, Campanian amphora, ER red plate; LR: local amphora. Two stone choppers. Roman hamlet.

41. (59) Panekklesia, Fields 555, 556, 562, and 563 This site is situated 400 m east of Vasiliki village on the west alluvial slope of the river. The apse of a small church (inner diameter 0.80 m, wall thickness 0.60 m) made of reused limestone columns, tiles, cement, and several blocks is preserved by the riverbank. Sherds are distributed in an area measuring 150 x 90 m. Artifacts (269 sherds): Tiles, brick, one Hayes Type 3 bowl; the rest LR–Byzantine. Roman–Byzantine village.

42. (60) Fields 557 and 561 On the eastern side of the river across from site 41, a scatter (100 x 40 m) of sherds marks the location of a Roman site of tile-roofed houses. Kiln remains (ceramic slag) were also found. An asphalt road runs through the middle of the site, which is larger than the scatter on the east side of road. Artifacts (499 sherds; Figs. 52:E6, 57:E42): Roman: tiles, bricks, pithos, cookware, giant basin,

43. (45) Field 504 Located 400 m southeast of site 24 at the western edge of the Cha riverbed, this small (30 x 10 m) field house or farmhouse(?) may have cultivated the small valley and surrounding slopes. Artifacts (34 sherds): PP: cookware, one vat/slab (wine press?), hole-mouth jar; NP: one tripod leg of a cookpot. MM IB/II–MM III/LM I field site.

44. (24) Evraika, Field 298 At the northern edge of the Isthmus Plain some 600 m east-northeast of site 43, this small site measured 35 x 30 m: it may have been a MM IA–II farm situated on a low hill using the surrounding red soils and water in the streambed at the base of the mountain Halepa. Artifacts (862 sherds; Figs. 13:B169, 14:B181, B182, 15:B192, B193, 16:B208, 21:B302, B303, B307, 26:B368, 27:B371; Pl. 18:B77, B79): LPP: jug, four bowls (lekanides), jar, three amphorae, lids, larnakes(?), pithoi, cookware (pots, dishes), seven hole-mouth jars, carinated cup; PP: ovalmouth amphorae, pithoi, pithoid jars, internally scored basin, fine and large BSJs, jug, three collared jars, carinated cup, Vaphio cups, basins, spouted basin, bowl (lekanis), conical cups, holemouth jars, plume jug. MM IA–II farm.

45. (—) Evraika, Field 228 Small shallow caves in the eroded northeastfacing slope of the gully at the base of Mount Halepa were used for MM IA–IB burials, probably for site 44. Excavated by M. Tsipopoulou (pers. comm.). MM IA–IB cemetery. M. Tsipopoulou, pers. comm.

CATALOG OF SITES

117

46. (23) Palaioekklesia, Field 244

48. (26) Ikonostasis, Field 403

On a low hill at the base of Mount Halepa, this site (Pl. 6B) consists of a Minoan structure and sherd scatter (50 x 90 m), and a Roman settlement of tile-roofed houses, a central building, trapetum, collecting basin, and pottery (50 x 120 m). The site overlooks the valley floor to the south that forms a natural corridor between the Minoan and Roman settlement at Tholos on the north coast and the Isthmus of Ierapetra to the south. The Minoan pottery extending 80 m to the east of the cyclopean structure implies a central building and outlying houses. A small basin constructed of cement with attached channel and Roman ground stones were used to press oil. The locale takes its name from a ruined church 500 m to the south. Artifacts (958 sherds): PP: cookware, SS cups, jars; NP: many cookpots, pithos; Roman: tiles, cookpot; ER: local amphora, basin, pithoi, ER red-painted dishes (Chandarli), one Italian amphora, one large pithos; LR: cookpots, East Aegean cookpot, two Phocean RS dishes, three African RS dishes. MM IB/II–LM I hamlet, ER hamlet.

About 100 m west-southwest of site 47, a concentration of sherds (30 x 40 m) and large plowed up fieldstones mark the location of a field house or farm building. Sixty-four clay lumps (mudbrick?) were recovered. A small cluster of pithos sherds is located some 27 m southeast of the site center. Artifacts (401 sherds; Figs. 31:B450, 32:B478): NP: large basins, pithoi, fruit-stand/bowl, jug, cookware, LM I everted rim cup, one pithos from North-Central Crete. Four stone tools: two of red chert; drill bit guide. MM III–LM IA farm/industrial site.

47. (25) Ikonostasis, Field 402 This small scatter of NP sherds (50 x 95 m) is located at the base of the north face of Thriphti (immediately south of the asphalt road) and about 800 m southwest of Kavousi village. Scatters of baked clay bits and the presence of two possible wasters on the site may be the remains of pottery kilns. No PP fabrics were found. A well immediately south of the site, drilled in 1971, produces water at 60 m below surface. Flat land for a farm is located immediately to the north. Artifacts (671 sherds; Figs. 33:B489, B497, B498, 34:B505): Giant base to pithos or larnax(?), fine jugs, basin, many cookpots (including pot lid, tray, and 20 legs), 2 larnax fragments(?), oval-mouth amphora, large BSJ, many pithoi, jar, vat/slab (wine press?), incised basin, Vaphio cup, small pithoid jar; 2 ground stones (one mano for crushing grain on a quern). MM III–LM IA hamlet, ceramic kilns, and possible wine press.

49. (28) Ikonostasis, Field 413 This small (10 x 12 m) PP–NP site sits on the slope of Thriphti about 200 m south of site 47. Some of the pottery, pithoi, larnakes(?), an incenseburner, suggests that part(?) of the site may have been a cemetery. Artifacts (298 sherds; Figs. 33:B487, B491, 59:E56, 60:E67): PP: bowl/basin, cup, large jar, cookware, thymiaterion (incense burner), ovalmouth amphorae, pithos; NP: pithos, cookware, hole-mouth jar, larnakes(?). MM IB–LM I field site and possible cemetery.

50. (31) Field 429 This is a concentration (25 x 40 m) of sherds on the south slope of Thriphti almost 100 m northeast of site 53. Massive terrace walls may have been built of material from a house. Artifacts (69 sherds; Fig. 33:B485): NP: cookpots, pithos, large jar. MM III–LM I farm.

51. (30) Field 440 This is a dense concentration (55 x 70 m) of PP and (mostly) NP pottery on the steep slope of Thriphti. Artifacts (384 sherds; Fig. 32:B480): NP: cookware, oval-mouth amphora, pithoid jars, pithoi, jugs, fine amphora, cup, MM III–LM IA conical cups, one possible drain, collared jar. MM IB–LM IA hamlet.

118

L. VANCE WATROUS

52. (29) Prinous, Field 439 The site consists of a scatter (40 x 30 m) of PP sherds and a short stretch of wall of large fieldstones oriented north–south, set high on the scree-covered slope of Thriphti. Artifacts (128 sherds): PP: jug, pithoi, basins, cookware, pithoid jar, carinated cup; NP (four sherds): cookware, basin. MM IB–II farm, MM III–LM I field site.

53. (32) Prinous, Field 429 The site consists of a concentration (20 x 20 m) of NP sherds about 60 m southwest of site 50. Artifacts (257 sherds; Figs. 30:B428, 32:B472, B477, 34:B507, B509, B510): NP: 11 clay lumps, cookware (pots, lid and plates), basins (for grapes?), pithoi, jars, jugs, MM III conical cups, cups. MM III–LM IA field site.

Artifacts (617 sherds, mostly Roman; Figs. 31:B431, 32:B473): NP: cookware, large basin, Vaphio cup; Roman: tiles, nine+ amphorae (one North African, 3rd–4th century A.D.), three ARS plates (3rd–4th century A.D.), jugs, bowl. MM III–LM IA field site, ER hamlet.

57. (35) Fields 515 and 532 Sherds were distributed over an area of 25 x 40 m, perhaps representing the cemetery for site 56 (ca. 80 m northeast) or site 58 (ca. 90 m to the southeast). A single concentration of larnax sherds and bones represents a burial. Artifacts (234 sherds; Fig. 26:B363): PP: basins, four (two oval) larnakes, pithos, jug, cups, cookware, tumbler, BSJ, clay lump. MM IB–II farm and cemetery.

58. (37) Field 531 and 532 54. (33) Field 511 The site is an area (80 x 45 m) of sherds around two partly preserved walls of large boulders that form a right angle. It was a small site in LPP, largest in MM III, situated on the low slope of Thriphti at the edge of the plain. Artifacts (233 sherds): MM IA: cookware, scored basin, red-painted basin; PP: pithos, basin, holemouth jar, oval-mouth amphorae, SS cups, larnakes, fine jug, bowl, cookware; NP: MM III conical and Vaphio cups, cookpot, cook dish, jar, jug, basin; three clay lumps, possibly from mudbrick. EM III–MM IA farm, MM IB–LM I hamlet.

55. (34) Field 513 This is a light scatter of sherds (25 x 10 m) at the edge of the plain 300 m south of the KTEO building (auto inspection station) and 175 m southsouthwest of site 54. Artifacts (52 sherds): PP: cookware, pithoid jar; NP: Vaphio cup, cookware. MM IB/II–MM III/LM I field site.

56. (36) Field 533 This is a small scatter (ca. 10 x 10 m) of NP sherds inside a larger (60 x 40 m) Roman site on the valley floor of the Isthmus.

This is a light scatter (50 x 50 m) of sherds, and a clay lump, on the terraced slope of the plain. Artifacts (115 sherds; Figs. 32:B470, B476, 33:B494, 34:B508): NP: cookware, pithoi, jar, pithoid jars, vat/slab or tray, pithoi, basin, ovalmouth amphorae, cup. MM III–LM IA farm.

59. (38) Plakoures, Field 530 This is a small scatter (40 x 35 m) of NP sherds on low slope, 200 m east of site 60. Artifacts (107 sherds; Fig. 34:B504): NP: cookware, pithos, fine jug(?), MM III oval-mouth amphora. MM III–LM I farm.

60. (44) Plakoures, Field 323 Set in the plain midway between Thriphti and Halepa, this site consists of a large concentration (170 x 200 m) of sherds and the remains of a cyclopean structure. Much of the site has been destroyed by bulldozing. A few blocks of an exterior wall (with a setback) are all that remain of what was reported to be a large cyclopean building. The settlement controls access between the Kavousi Plain and shore, and the Isthmus. A few

CATALOG OF SITES

MM IA sherds indicate that the site started at the end of the LPP period. Sites 46, 56, 58, 59, and 62 could all be interpreted as outlying farms or field houses belonging to this settlement in the PP and NP periods. There is no natural water source in the immediate area. Its catchment is level but dry (marginal agricultural) land. Collection of sherds at the site by sectors indicated that the principal settlement was founded in LPP times, and the settlement did not radically change size during the PP–NP period. Artifacts (1,975 sherds; Figs. 24:B349, 32:B481, B482): LPP: jar, cookpot, hole-mouth jar; PP: lid, several drains, many basins (4 scored), large jars, fine and coarse hole-mouth jar, pithoi, cookware, many pithoid and hole-mouth jars (several with granodiorite), jug, bowls (lekanides), carinated and SS cups, lamp(?), larnakes(?); NP: at least four drains, large basin, loomweight, pithoi, pithoid jar, cookware, two lamps or stands (LM IA), Vaphio and conical cups, fine jug; 52 clay lumps, several large (8.3 cm thick). MM IA village, MM IB–LM I village.

61. (62) Kamina, Fields 248, 249, and 258–264 The site is at the base of Thriphti on the edge of the plain. In the PP period, the distribution of sherds is about 90 x 50 m; in the NP it measures roughly 60 x 60 m. Pithoi sherds found outside the main concentration on the north end and east sides may be burials. The Roman-period settlement is a large one (330 x 160 m). The remains of two asbestos kilns (asbestokamina) and associated tiled buildings are relatively recent, as the toponym indicates. Artifacts (1,613 sherds): PP: cookware, large jar, basin, pithoid jar; NP: jug, pithos, cookware, Vaphio cup; Roman: tiles, 4 pithoi, 18+ amphorae; ER: BG sherd, cookware, ER bowl (Chandarli or Phocean), basins, amphorae, beehive; LR: amphora, basin, 4 Phocean RS dishes, 4 ARS dishes, tiles. MM IB–LM I hamlet, ER–LR village.

62. (41) Field 521 This site consists of four small concentrations of NP sherds within a 200 x 200 m area on the scree-covered foothill at the base of Thriphti. Rough walls, a mandri, and a well are more recent

119

features. Three of the concentrations can be dated to PP and NP. Artifacts (471 sherds): PP: cookware, hole-mouth jar; pithoi; NP: many pithoi, oval-mouth amphorae. MM IB/II–MM III/LM I field site.

63. (22) Kamina, Field 247 This is an area (23 x 15 m) of sherds some 300 m southwest of site 61. The separate patches of pithoi (one with bones identified by Maria Liston as a human adult tibia and humerus) and larnax fragments indicate this is a cemetery. A few Roman sherds. Artifacts (174 sherds; Figs. 23:B342, 24:B347, 62:E81): LPP: pithoid jars, larnax; PP: cookware, SS cup, jug, basins, larnakes. MM IA–II field site and cemetery.

64. (40) Fields 521 and 522(?) This is a dense concentration (55 x 60 m) at the base of the scree slope of Thriphti. The site (farmstead?) faces west-northwest onto the red-soil plain. Artifacts (369 sherds; Figs. 20:B279, 21:B304; Pl. 23:B148): PP: cookware, carinated cup, pithoi, conical cups, jar/BSJ; NP: cookware, basin. MM II–LM IA hamlet.

65. (21) Fields 283, 284, and 312 The site sits on an alluvial fan at the base of Thriphti and has three certain periods: LPP, PP, and NP. In the LPP period, the settlement is about 80 x 40 m, with perhaps two houses; in the PP period, the site is 90 x 80 m; the NP settlement (90 x 110 m) expands to the northeast and may include six houses. Few Roman sherds. Artifacts (Figs. 31:B441, 32:B479, 33:B490, B496, 61:E73, 62:E80, 63:E88–E90, E92, 64:E100): EM I–II: five sherds; LPP: hole-mouth jar, bowl, cup, fine monochrome jug, amphora; PP: basins, hole-mouth jar, rilled carinated jar, BSJ, cookware, cups (two carinated, SS, conical), ovalmouth amphora, larnax(?); NP: pithoi, cookware, cups (Vaphio, rounded, conical), drain(?), pithoid jar, basins (scored), fine spiral decorated jug (LM IA). Whetstone. EM I–II field site, MM IA–LM I hamlet.

120

L. VANCE WATROUS

66. (43) Paper Mill, Fields 395–397 Set at the base of the Thriphti slope, the site faces west (ca. 90 m to the Chalasmenos streambed), and the plain to the east. The site was much destroyed by bulldozing. Its southern half is partly covered by a terrace of fill. Protopalatial and NP sherds occur over an area of 110 x 40(?) m. The Roman site is smaller (70 x 40 m). A massive circular stone trapetum of the Roman period (Pl. 9C) was removed from the site to the Ierapetra Museum. Artifacts (813 sherds; Figs. 20:B286, B290, 22:B322, B325, 23:B337, 33:B486): PP: cookware, large jars, brazier(?), amphora, pithos lid, pithos, pithoid jar or larnax, basin, BSJ(?), cups; NP: cookware, conical cup, cups (LM IA), pithoi, pithoid jar, flowerpot, BSJ; Hellenistic (four): BG bowl, cup, pithos; Roman: tiles, cookware, ER Terra Sigillata, Italian Terra Sigillata plates (3rd–5th century A.D.), two Chandarli Ware dishes, basin; LR: three ARS dishes, four Phocean RS bowls. Clay lumps and chert. MM IB–LM IA hamlet, H–ER hamlet.

site faces out onto the gentle slope to the west. The PP sherd distribution is 40(?) x 40 m; NP: 75 x 40 m; LM IIIC: 100 x 40 m. A pithos found 60 m south of the site may represent a burial. In the LM IIIC period, this site must have been the outskirts of the upper Chalasmenos settlement. Artifacts (967 sherds; Fig. 32:B467; Pl. 27:B179): PP: cookware, jar, pithos, cups, large jar; NP: cookware, conical cup, large jar, Vaphio cup, one Marine Style(?) fragment; LM IIIC: cookware, pithoi, deep bowl; one LM IIIA cup. MM IB–II farm, MM III–LM IB hamlet, LM IIIA presence, LM IIIC hamlet and cemetery.

69. (64) Field 785 The limited (40 x 20 m) scatter of sherds is located on the lower slope of Thriphti (west and below Chalasmenos), next to a small ravine and streambed. Artifacts (120 sherds): PP: cookware, pithos, basin; NP: pithos. MM IB/II–MM III/LM I field site.

67. (42) Cha, Fields 392 and 393 The site (elevation 120 m) is under and around the Church of Hagios Pandelemion on a hill next to the Chalasmenos streambed. Early Minoan I–II sherds are concentrated in an area (30 x 30 m?) around the church, and LM IIIC sherds also occur within this area. The PP and NP site measures 100 x 45 m. The Ottoman-period settlement stretches down the slope to the north and west. Artifacts (Fig. 34:B518, B519): EM I–II burnished sherds, many red chert chunks and flakes, one LPP(?) vase; PP: coarse and cookware, pithoi, 4 pithoid jars, hole-mouth jar, scored basin (beehive?); NP: cookware, coarse basin, jar; LM IIIC: tripod legs, a deep bowl or krater, large krater, bowl, larnax; 19 chaff-tempered clay lumps, perhaps for industrial use (in LM IIIC?); Roman: cookpot, one stamped ER plate, 2 Hayes Type 3 dishes; V–T: sgrafitto, beehive, glazed wares. EM I–II field site, MM IB–LM I hamlet, LM IIIC field site, Roman farm, Ottoman presence.

68. (65) Chalasmenos, Field 782 At the base of the excavated site of Chalasmenos, next to (south of) the Chalasmenos streambed, the

70. (66) Chalasmenos, Field 799 Situated on the upper slopes (elevation 180 m) of Thriphti some 800 m southwest of the excavated site of Chalasmenos and near the streambed next to site 69, this site has a EM I–EM II scatter of 15 x 15 m, a PP scatter of 70 x 40 m, and an LM IIIC scatter of 70 x 70 m. Mudbrick fragments indicate a LM IIIC structure. Roman sherds are limited to an area of roughly 20 x 20 m. Artifacts (Figs. 61:E71, 63:E91): EM: dish, lug handle, worked chert debris; PP: cookware, jar, basin; NP: cookware, pithos, pithos lid; LM III: stirrup jar(?); LM IIIC (many sherds): pithoi, basins, hole-mouth jar, many kraters and pithoi, kylikes, skyphos, footed bowl, larnax; Late Roman: amphora, two ARS dishes, tile. Many post-Roman: tiles, pithos, green-glazed wares. Clay lumps (mudbrick?). EM I–II field site, MM IB–II hamlet, MM III– LM I field site, LM IIIC village, Roman field site, V–T presence.

CATALOG OF SITES

71. (67) Gonika, Fields 742–744, 751–753, 758, 759, 767, and 768 Located approximately 1 km below (northwest) the village and the spring of Monastiraki, this site is roughly 350 x 100 m in size. Artifacts: LM IIIC larnax(?); LR: tiles, bricks, cookware, amphorae, four Phocean RS plates, (one giant) pithoi, spirally grooved ware, beehive(?). LM IIIC possible cemetery, Roman village.

72. (68) Monastiraki, Under Village and Fields 792, 793, 795, 801, and 802 This site consists of a Roman mosaic discovered above a cistern and the Church of Hagios Stephanos. Artifacts (537 sherds; Fig. 52:E9): 3rd-century B.C. Hellenistic amphorae(?); Roman: bricks, tiles, cookware; ER: two amphorae made of volcanic fabric, cookware mug; LR: one ARS dish, amphorae, four Phocean RS dishes, one Cypriot dish, one North African amphora, basin, cookware. ER–LR/Early Byzantine hamlet. Schachermeyr 1938, 470; Sanders 1982, 140.

73. (69) 1994. Hagia Marina, Fields 836 and 837 The site is located below (west) of the church. About 20 m west-southwest of the Church of Hagia Marina are the remains (cavities, wasters, stamnoi, tsoukalia, brazier, pithoi, brownstone, and earth) of two pottery kilns built by Thrapsanos potters. The Roman sherds and walls are in an area of 100 x 150 m; the Ottoman site is immediately below the church and measures 45 x 140 m. Artifacts (1,147 sherds; Fig. 54:E24; Pl. 43:D90): Roman tiles, bricks, marble wall slab hypocaust plug(?); ER: one Italian Terra Sigillata, amphora; LR: two ARS dishes, basin, cookware, four Phocean RS bowls; LR/Early Byzantine: amphorae, cookware. ER–Early Byzantine village, Ottoman hamlet.

74. (70) Fields 807 and 808 Located some 300 m southeast of the VasilikiMonastiraki-Ierapetra crossroads, this site is in the middle of the Isthmus Plain and about 300 m east of a streambed. This Roman site measures 75 x 30 m.

121

Artifacts (424 sherds; Fig. 55:E29): Late Roman tiles, amphora, bricks, pithos, two ARS plates (3rd–4th century A.D.), overfired LR amphora, two North African amphorae. Roman hamlet.

75. (74) Field 872 On the south slope of a hillock high (elevation 195 m) in the foothills of Thriphti, this is a small (20 x 25 m) cluster of PP and NP sherds, probably a sign of a field house. Artifacts (52 sherds): PP: cookware, bowl, holemouth jar; NP: fine jar. MM IB/II–MM III/LM I field site.

76. (72) Fields 871 and 872 This site comprises two small clusters (30 x 40 m; 50 x 30 m) of sherds on a shoulder of the slope of Thriphti 250 m west of site 75. Limited good arable land suggests a pastoral function (at least in part). One or two Roman sherds. Artifacts (150 sherds; Fig. 64:E97): PP: cookware, bowl, hole-mouth jar; NP: coarse ware. MM IB/II–MM III/LM I hamlet.

77. (73) Fields 826 and 827 This site is located on a flat table of land on the rugged slope of Thriphti, with a cliff face along its western side, steep S-slopes on three sides, and a spring at its western base. Sherds, tiles, and walls occur over an area of 125 x 30 m. At least three separate series of house walls, constructed of large fieldstones, are preserved. There is little good arable land in the immediate vicinity. Many overfired sherds suggest local ceramic production. A few fine-ware shapes were found. The site was occupied from ca. 350–200 B.C. Artifacts (804 sherds; Fig. 49:D82, D86, D87, D90; Pl. 43:D85): Cooking ware: Fabric (2) (Hierapytna), 73 sherds, including 11 tiles, 20 pithoi, 3 amphorae, 3 jars. Fabric (7), 96 sherds, including 4 cups/kantharoi, 1 amphora, 2 dishes/ bowls, 3 jugs(?). Fabric (8), 63 sherds, including 8 amphorae, 1 basin, 1 jug, 13 large jars (amphorae?; 1 overfired, Coan 4th–3rd century B.C.), 1 pithos. Fabric (9), 3 sherds, including 1 coarse basin.

122

L. VANCE WATROUS

Mortars, 1 stamped amphora. One Attic(?) bowl. (See App. D for a description of the fabrics.) Hellenistic hamlet.

78. (77) Fields 822 and 823 The site is on a low slope of Thriphti about 150 m north of a spring and near (east of) arable land in the plain. The PP and NP site measures 110 x 75 m. One LM IIIA(?) bowl was found; mostly NP sherds were recovered. Artifacts (415 sherds; Fig. 34:B517): PP: cookware, basins, SS cup, large jars; NP: cookware, pithos, fine jar, spiral on dark ground jug (MM III– LM I), jugs, pithoid jar, pithoi, amphorae, large basin. Chert flakes and chunks, knapping debris. MM IB/II–LM I hamlet, LM IIIA presence.

79. (82) Fields 817 and 818 This large site (550 x 150 m) is in the middle of the Isthmus Plain at the narrowest part. There is good arable land to the north and west. Artifacts: Ottoman-period tiles, glazed and cooking ware, threshing-sledge flake, Turkish clay pipe. Late Venetian/Ottoman village.

80. (81) Fields 846 and 847 This is a Roman site with a pottery scatter of 125 x 135 m. Springs are located southeast of the site, and arable land to the west. Artifacts (1,163 sherds; Fig. 54:E22): Bricks, tiles, beehive; ER: one BG handle, amphora; LR: amphorae (3rd–4th century A.D.), cookpots; Roman–Byzantine: amphora; a column base. ER–Byzantine village.

81. (71) Fields 809, 810, and 845 This MM III(?)–LM IIIB site is in the middle of the Isthmus Plain, about 300 m east of a riverbed. Good arable land is available on all sides. The NP scatter measures 65 x 45 m; the LM III scatter 40 x 45 m. Artifacts (709 sherds; Figs. 13:B171, 32:B471, B475): PP: pithoi, cooking ware, amphora, pithoid jars; NP: cooking ware, amphora, conical cups, Vaphio cup, BSJ, pithos, pithoid jar; crucible(?);

cookpot from Mochlos (purple schist inclusions); a few Geometric–Archaic sherds. MM III(?)–LM I hamlet, LM IIIA–IIIB farm, LM IIIC–Geometric presence, Archaic–Hellenistic field site (small house with tiles).

82. (78) Field 866 This is a small (70 x 40 m) PP farmstead(?) located in the middle of the Isthmus Plain, about 500 m east of the riverbed. Artifacts (200 sherds): Pithoid jar, cookware, Vaphio cup. Quern, stone tool, flake. MM II–III farm.

83. (110) Fields 1036 and 1037 High in the foothills (elevation 170 m) of Thriphti on a flat stretch of land at the foot of site 84, the site comprises several sherd scatters: PP (100 x 105 m); NP (60 x 80 m); LM IIIC (65 x 70 m). Fragments of PP–NP pithoi, larnakes, and an incense burner or lamp suggest the site was a cemetery for site 84. On the other hand, the quern, cooking pots, and mudbrick point to a settlement. Late Minoan IIIC vases are mostly large shapes (e.g., pithoi). Artifacts (503 sherds): PP: cookware, pithos lid, pithoid jar, basin, lamp/thymiaterion; NP: cookware, pithoi, cups; LM IIIC: pithoi. Quern and clay lumps (mudbrick). MM IB–II village, MM III–LM I hamlet, LM IIIC hamlet and cemetery.

84. (111) Fields 1035 and 1036 This site is located on the steep slope of a natural amphitheater, at the base of which sits site 83. There is a spring at its upper edge. Steep local slopes provide limited arable land. Sherd distribution is as follows. EM I–II: 20 x 20 m; LPP: 70 x 65 m; PP: 255 x 75 m; NP: 115 x 105 m; LM IIIA– IIIB?: 40 x 15 m; LM IIIC–PG: 90 x 15 m; Late Venetian–Ottoman: 400 x 150 m. Late Prepalatial occurs in the north and center of the site. There are three areas (houses?) of PP at the northern end of the site: (1) on the ridge top (ca. 20 x 70 m with walls), (2) on the slope to the south, and (3) about 75 m to the south on a slope (40 x 45 m). Dense NP

CATALOG OF SITES

remains are concentrated in the central portion of the site. Late Minoan IIIA–IIIB and LM IIIC sherds are concentrated in the northern part of the site: one LM IIIC area (ca. 15 x 30 m; a house?) has a jar (burial?) below it. Artifacts (Pl. 57:E115): EM: red-painted EM II vases, chipped stone, stone tool; LPP: hole-mouth jar, amphora, lekanis, basin, thin oval tripod leg, pithoid jars; PP: SS cups, pithoi, pithos lid, clay drain, cookware, basin, pithoid jars; NP: pithos, cookware, beehive(?), Vaphio cups, pithoid jar, scored basin; LM IIIA–IIIB: jug, cookware; LM IIIC: skyphos, krater, jug, pithoid jars, cookpot; V–T: sgraffito, tile. EM I–II field site, EM III–MM IA hamlet, MM IB–LM I village, LM IIIA–IIIC field site, V–T village.

85. (76) Field 876 Located at the base of Thriphti about 300 m southeast of site 82, this small (ca. 40 x 40 m) site consists of coarse NP sherds. Arable land is present west of the site. Artifacts (210 sherds): NP: pithos, bowl, cookware, vat/slab (wine press), coarse feeding bottle; LR: one Phocean RS dish, amphora, tiles. Two chipped stone flakes. MM III–LM I farm with wine press, Roman tile-roofed farm building.

123

87. (84) Fields 890, 893 This is a small scatter (30 x 35 m) of sherds located 150 m southeast of site 86. Artifacts (102 sherds): tiles, cup(?). ER–4th c. A.D. farm.

88. (100) Hagios Georgios, Fields 901, 902, 904, and 905 About 150 m north of site 100, massive walls and a scatter of tiles and sherds occur in an area of 150 x 300 m. Artifacts (416 sherds; Fig. 53:E15): Hellenistic: tiles, (stamped) amphora handle, pithos; H/ER: two North African(?) amphorae, local amphora; LR: one Phocean RS dish, beehive; Roman: tiles, brick. Spindle whorl (site 93?) and beehive found to the south of the site. Hellenistic farm, Roman–Byzantine village.

89. (107) Fields 1036–1038 Situated on the slope (elevation 210 m) of Thriphti immediately north of site 90, this site is a large (200 x 150 m) scatter of sherds and tiles of the Ottoman period. Few Roman sherds. Artifacts (99 sherds; Fig. 59:E53): Tiles, pithoi, plates; one Hellenistic amphora. Hellenistic field site, Ottoman village.

90. (108) Hagios Georgios, Field 1038 86. (75, 83) Fields 888, 889, 894, and 895 This site is located in the middle of the Isthmus Plain adjacent to a riverbed. Since the site partly lies outside the boundary of our survey, its overall size is unknown. It was a large settlement in the ER–Byzantine period. The presence of kiln fragments, burned stone, and wasters indicate a local ceramic production center of amphorae. Water and arable land are locally available. Its central location was ideal for the export of these products. Artifacts (793 sherds; Figs. 51:E3, E4, 54:E19, E21, E23, 60:E62): ER: one basin; 3rd–4th c. A.D.: amphorae, waster; LR: three basins (overfired), bowl, jug, amphorae, cookware; Roman tiles, bricks, kiln fragments, wasters, cookware, many amphorae. ER village and amphora production center.

This is a large concentration of sherds located on a shoulder (elevation 221 m) on the Thriphti foothills. There is a little arable land below the site, and a spring above it. Above the Church of Hagios Georgios, the distribution of sherds is as follows: EM I–II, one medium-size cluster (ca. 30 x 30 m) just below the spring and another cluster (ca. 20 x 20 m) 30 m below (to the west); LPP, two small clusters (one to the south of the EM I–II area; one in the same area as the second EM I–II cluster); PP, 110 x 150 m; NP, 110 x 150 m; LM IIIC, traces at the southern edge and on the lower east slope. Artifacts: EM: marble/calcite ware (EM I), fine gray ware (EM IB–IIA), burnished sherds, redpainted sherds, red-burnished sherds, hole-mouth jar, chipped stone; LPP: lekanis, pithoid jars; PP:

124

L. VANCE WATROUS

jug, oval-mouth amphora, tumbler, carinated cup, cookware, pithoi, basin, pithoid jar, semi-globular cup, bowl (lekanis); NP: cookware, basins, Mochlos import, larnax(?), pithoi; LM IIIC: krater, pithos; Hellenistic: one pithos; V–T: tiles. EM I–MM IA farm, MM IB–LM I village, LM IIIC field site, V–T presence.

91. (93) Hagios Georgios, Fields 905, 906, 918, 921, and 922 Around and below (south of) the Church of Hagios Georgios, house walls, large terrace walls, and architectural parts (column, blocks, threshold) were found. Roman sherds were scattered over 250 x 220 m. Artifacts (276 sherds; Figs. 53:E12, 54:E20, 55:E26, 56:E34, E35, 57:E42): Tiles, brick, wall slab; ER: amphorae, cookware, jug/lagynos; LR: amphora, two wall pins, cookwares, amphora stand, five Phocean RS dishes; LR/Byzantine: two imported amphorae (one volcanic fabric), stopper. ER–LR village, Early Byzantine and Late Ottoman presence.

92. (117) Fields 1038 and 1039 The site consists of a small (25 x 35 m) scatter of NP and Roman sherds and tiles on a shoulder (elevation 254 m) on the slopes of Thriphti. Artifacts (134 sherds; Figs. 59:E57, E59, 64:E101): LR tiles, basin, amphorae. MM III–LM I field site, LR (3rd–4th century A.D.) field site.

93. (116) Fields 1098 and 1109 A scatter (25 x 20 m) of pebbles and worn sherds occurs on the bare peak of a bedrock outcrop with wild olives on the slope (elevation 260 m) of Thriphti. Artifacts (353 sherds; Fig. 31:B446): PP: coarse, one carinated cup; NP: many cups (some Vaphio), and juglets; three chipped stone pieces. MM II–LM I field site(?).

94. (115) Field 1108 On the slope (elevation 254 m) of Thriphti, a concentration (80 x 70 m) of NP sherds occurs. A little arable land is available to the north. Artifacts (217 sherds): NP: conical cup, Vaphio cups, MM III bowl, basin, jugs, cookware. Large quern. MM III–LM I hamlet.

95. (112) Field 1087 High (elevation 325 m) on the Thriphti slope, this site lies on a shoulder, with flat arable land to the east, and a spring. Sherds (45 x 55 m) and two separate sets of walls (one a platform[?], 10 x 18 m) were found directly to the west. PP farmstead? Artifacts (459 sherds; Fig. 14:B179): PP: cookware, pithos, basin, amphora, pithoid jar, holemouth jar, cup. Two chipped stone pieces (one obsidian). MM IB–II hamlet.

96. (120) Field 1099 This is a small scatter (35 x 50 m) of sherds on a steep slope some 100 m south of site 95. Artifacts (19 sherds): Pithoid jar, pithos. MM IB–II farm.

97. (114) Hagios Georgios, Fields 1105–1107 Situated in a stretch of flat land (ca. 400 x 200 m) on the slope (elevation 220 m) of Thriphti, the site consists of sherds distributed over an area of 130 x 210 m. Remains of a cyclopean building were found under a mandri at the east edge of the site. The sherd distribution was as follows: EM I–EM II: 60 x 80 m; LPP sherds around building and at western edge: ca. 80 x 30 m; PP: 130 x 210 m; NP: 130 x 210 m (same as PP except for northwest corner); Roman: 40 x 40 m. An asbestos kiln was also found. Artifacts (1,292 sherds; Figs. 56:E33, 60:E68, E69): EM: pithos, dark burnished, gray ware, cookware, marble/calcite ware, 26 chipped stone, stone

CATALOG OF SITES

tool; LPP: hole-mouth jar, bowl; PP: tumbler, SS cups, pithoi, pithoid jar, cookware, hole-mouth jars, amphora, larnax(?), lekanides; NP: pithos, pithoid jars, Vaphio cup, cookware, basins, amphorae and jug (MM II–III), BSJ, larnax?, Minoan animal figurine; 8 LM IIIC sherds (two kraters); LR: 1 local amphora, 2 North African amphorae, red-painted dish; tiles. EM I–MM IA hamlet, MM IB–LM I village, LM IIIC–PG farm, LR farm. MM IB–LM I possible cemetery.

98. (113) Hagios Georgios, Fields 1107 and 1108 On a slope 50 m north of site 97, pottery and chipped stone were scattered over an area 30 x 40 m. A PP burial was found at the western edge in a rock cleft, along with a patch (5 x 5 m) of sherds. The site is a FN farmstead; arable land is found to the west and east, and there is a spring at the Church of Hagios Georgios some 200 m to northwest. Artifacts (Figs. 1:B1–B13, 2:B14–B21; Pl. 11:B1–B8): FN: dark burnished ware, 2 incised sherds (chevrons, meander pattern), lug handles, 1 chalice? (EM I?), 1 jug (EM I Mesara import?); 100+ chipped stone, obsidian and chert, stone tool, 1 unfinished blossom bowl; PP: basket-impressed dish, SS cup, basin, oval larnax, leg bone (human?), cookware, thick oval tripod leg. FN farm, EM I field site. MM IB–II field site and cemetery.

99. (94) Field 919 About 150 m southeast of site 100, at the eastern edge of the Isthmus Plain, the site consists of a scatter of EM II (15 x 15 m), EM III–MM IA (15 x 15 m), and MM I–III sherds (15 x 25 m), and stone tools and two crucibles. Large amounts (130 pieces collected) of chipped stone were also found. Artifacts (315 sherds): LPP: bowl, jar; PP: cookware, basin pithos, carinated cups; NP: pithoid jar, conical cups, large cups. EM I–II knapping site, EM III–LM I field site.

125

100. (79) Hagios Georgios, Fields 892, 907, 908, 914, and 915 This site is located at the eastern edge of the plain below a conical rock outcropping, just west of site 91. Middle Minoan IB–LM IA sherds were scattered over 150 x 180 m. Chipped stone was densely distributed over 300 x 200 m. The site was a local source of chert; a stone drill guide was found 150 m to north, at site 88. Traces of LM IIIA–IIIB ceramics were reported in a 20 x 20 m area in the western part of site. A few Hellenistic–Roman sherds. Artifacts (2,643 sherds; Figs. 49:D84, 50:D94): PP: pithos, pithoid jar, cookware, cup, basins, beehive, BSJ, Vaphio cups (including an MM IIB Knossian import), SS cup, hole-mouth jars; NP: Vaphio cups, cookware, pithoi; huge vat/slab wine press; ground stone. MM IB/II–LM IA village, LM IIIA–IIIB field site.

101. (85) Field 897 This site consists of a scatter (30 x 50 m) of tiles, brick(?), and sherds in the plain 200 m southeast of site 86. Artifacts (103 sherds): Brick, tiles, ER pottery. ER field site.

102. (101) Fields 897, 898, and 909 This site is a concentration (210 x 90 m) of sherds and tiles in the plain just north of the riverbed, ca. 200 m north of site 104. Artifacts (165 sherds): Tiles, one ARS plate (3rd–5th century A.D.), one Phocean RS plate; V–T: tsoukalia, glazed wares. LR and V–T village.

103. (90) Fields 942 and 943 This is a small (70 x 25 m) scatter of LR–Early Byzantine sherds and tiles south of the riverbed across from site 102 and 50 m west of site 104.

126

L. VANCE WATROUS

Artifacts (503 sherds; Fig. 55:E31): Roman amphorae, basin or jar (overfired), cookpots, tiles, one Phocean RS plate (3rd–5th c. A.D.). LR–Early Byzantine farm.

104. (89, 91) Fields 910, 911, 943, 985, and 986 This site is located on the north and south sides of the riverbed, in the plain just north of the Kazarma hill. In the Roman period, the site (sherds, tiles, bricks, walls) extended to both sides of the river (220 x 320 m). A marble wall slab, BG, and other high-quality pottery from the southern half of the site may be the remains of a villa that began in the 1st century B.C. Protopalatial sherds are scattered (30 x 30 m) in the western half; NP in both the eastern (20 x 30 m) and western (20 x 20 m) halves; LM IIIA–IIIB sherds (20 x 20 m?) in the western half; Early Byzantine and Late Venetian–Ottoman sherds covered an area 80 x 130 m. Artifacts (1,404 sherds; Figs. 34:B511, B516, 52:E7, E8, 53:E11, E13, 54:E18, E25, 55:E27, E30, 56:E32, 58:E45, E50, 60:E63, 61:E70): PP: basins, pithoid jar, amphorae, BSJ, cookware; LM IIIA–IIIB: kylix stem; Hellenistic (three): BG bowl, plate, krater(?); Roman: many tiles, bricks, pithoi (one huge), large basins, amphorae (one huge), coarse mortars; ER: one ARS plate, waster, one North African amphora, four Terra Sigillata plates (two Italian), one Eastern Sigillata B, one Knidian(?) Relief ware, local lamp, amphora; LR: seven ARS plates, one Italian amphora, eight Phocean RS dishes, cup (Asia Minor), notched rim basin, cookware (one East Aegean dish), imported cookpot; LR/Early Byzantine: amphorae, jar; Venetian: sgraffito wares. Three stone tools (one obsidian flake), core, and chipped stone. One weight, marble wall slab, triangular perforated stone (olive-press weight?). MM IB–LM I field site, LM IIIA–IIIB field site, H field site, Roman/V–T village.

105. (88) Field 979 At the northern base of the hill of Kazarma, this site looks out to a large area of arable land and riverbed to the north. A small scatter of LPP–PP

sherds are distributed as follows: LPP, 25 x 40 m; PP, 30 x 110 m. Chipped stone (one obsidian piece) and a celt were found 150 m southwest of the PP site 104. Artifacts (100 sherds): EM I–II: marble/calcite ware; LPP: cup, jar, hole-mouth jar; PP: pithos, cookware, bowl. Forty pieces of chipped stone, one celt. EM I–II farm, MM IA–II hamlet.

106. (87) Kazarma, Field 960 At the eastern edge of the Kazarma hilltop, this site consists of an EM II sherd scatter (45 x 20 m). Possible walls were found. There is little obvious arable land nearby. Perhaps this was a seasonal site. Artifacts: EM: 12 gray ware (EM IIA) sherds, much coarse, unburnished pottery, cookpot; chipped stone. EM IIA farm.

107. (123) Fields 916, 1192, and 1193 Walls and Minoan pottery (45 x 155 m) were located just north of a riverbed and about 300 m east of site 104. Arable land is present to the west and south. Traces of LM IIIA pottery were found, as were Roman sherds (40 x 80 m?). Artifacts (455 sherds; Figs. 53:E16, 59:E54): PP: cookware, large jars, cups, basin, carinated cup, tumbler, amphora; NP: cookware, basin, pithoi, Vaphio cup; Roman: tiles, brick; cookware and beehive (3rd–4th century A.D.); LR: North African lamp, cookpot. MM IB/II–LM I hamlet, LM IIIA–IIIB field site, LR hamlet.

108. (96) Panagia, Field 991 Located just south of a riverbed at the base of Thriphti, facing an arable tract to the west, this site has scatters of PP sherds (80 x 20 m), MM II–III sherds (20 x 50 m), and NP sherds (20 x 30 m). Artifacts (375 sherds): PP: three pithoi, pithoid jar, cookware, collared jar; NP: cookware (one huge), conical cup, Vaphio cup, jug; four chipped stone pieces. MM IB/II–LM I farm.

CATALOG OF SITES

109. (118) Panagia, Field 1102 On the slope of Thriphti, this site has three small, ill-defined concentrations of EM I–II and PP–NP sherds in an area 40 x 15 m, on a shoulder overlooking (to the south of) a riverbed. Limited arable land is available nearby. The pottery appears worn and may not be settlement debris. Artifacts: EM: sherds, chipped stone; PP: large jar, basin, cookware; NP: jar, BSJ, cup(?). EM I–II seasonal farm, MM IB–LM I field site.

110. (92) Panagia, Fields 981 and 982 Situated in a valley east of Kazarma hill, the site consists of a scatter of Venetian–Turkish (85 x 60 m) and Roman sherds and tiles (ca. 50 x 30 m). Artifacts (308 sherds; Fig. 57:E43): Roman: 1 ARS dish, basin, cookpots, 30+ tiles, pithos. Roman (3rd–4th century A.D.) farm, V–T hamlet.

111. (99) Panagia Kera, Fields 992, 993, 1005, and 1006 Situated on the lower slope of Thriphti, below the Church of Panagia Kera (and a spring), a scatter of house walls, tiles, and sherds of the Late Venetian(?) and Ottoman period occur over an area of 90 x 120 m. Arable fields are located to the north. Artifacts (345 sherds): Tiles, hand mill, glazed wares, tsoukalia (cooking pots), stamnia (amphoras). V–T village.

112. (97) Panagia, Field East of Field 1005 High (elevation 200 m) on the steep slope of Thriphti near a streambed, there is a scatter of PP sherds (50 x 50 m) and NP sherds (25 x 20 m). Farmstead for arable land to south(?). Artifacts (196 sherds; Fig. 33:B492): PP: pithos, cookware, pithoid jars, hole-mouth jar, burnished lamp; NP: huge pithos, hole-mouth jar, cookware, Vaphio cups. MM IB–LM I hamlet.

113. (104) Panagia, Fields 1025–1026 On the slope (elevation 240 m) of Thriphti, on a level arable area, there are ER sherds and tiles (65 x

127

60 m), and Venetian(?) and Ottoman sherds around a possible lime kiln. Artifacts (113 sherds; Figs. 58:E44, 62:E85, 63:E93, E95): Roman: tiles; V–T: glazed wares. Roman hamlet, V–T presence.

114. (95) Panagia, Fields 996, 1002 and 1003 At the base of Thriphti, on a steep conical hill overlooking the valley and streambed south of Kazarma hill, there is the following distribution of sherds: EM I–II, 80 x 90 m; LPP, 60 x 80 m; PP, 100 x 110 m; NP, 55 x 100 m; ER, 45 x 60 m. The Roman farmstead(?) is occupied in the Hellenistic period, perhaps soon after 67 B.C. Artifacts (Figs. 10:B129, 47:D27, 49:D88): EM I–II: Hagios Onouphrios, bowls, jar, chipped stone, obsidian blade and flake, stone hand tools (chopper), two querns; LPP: cups, pithoid jar, cook dish, hole-mouth jar, ribbed jar; PP: basin, pithos, pithoid jar, cookware, plume jug, incised basin, conical cup, SS cup, carinated cup(?); NP: pithoi, pithoid jars, cookware, spiral decorated jar (LM I); H: BG dish or bowl, BG cup; ER (18 pieces): cup, pithos, two Eastern Sigillata B, one Chandarli Ware bowl, basin, amphora, tiles; V–T: tsoukalia, glazed wares. EM I–MM IA hamlet, MM IB–II village, MM III–LM I hamlet, H farm, R hamlet, V–T presence.

115. (102) Panagia, Field 1000 Situated on the plain northeast of the Papadiana hill near a streambed, there is a scatter (65 x 30 m) of 19th–early 20th-century sherds. Artifacts (114 sherds): Glazed wares, Thrapsanos pithos. 19th–20th century hamlet.

116. (103) Panagia, Field 1017 This site is on the steep, lower slope of Thriphti northeast of Papadiana. Limited arable land is available. Artifacts (58 sherds): Glazed wares. V–T presence.

128

L. VANCE WATROUS

117. (98) Aphendi Christos, Fields 1016, 1022, and 1023 On a flat spur (elevation 240 m) on the slope of Thriphti just north of the Church of Aphendi Christou Spring, there is a site with arable land and scatters of EM I–II sherds (45 x 30 m), ER–LR sherds (120 x 170 m), and Early Byzantine–Ottoman sherds (200 x 150 m). A Roman tiled farmstead, with heavy trapetum (Pl. 10B), was also found, as were later house walls and one Hellenistic(?) bowl. Artifacts (Figs. 51:E5, 58:E46): EM I–II: pottery, obsidian blade, chipped stone; Roman (408 sherds): BG plate (1st century A.D.?), cookpot, pithoi, basins; LR: Chandarli plate, amphorae, cookwares (two dishes and one cookpot from the eastern Aegean), three ARS dishes, four Phocean RS dishes, one Cypriot dish, one Egyptian RS, basin; LR/Early Byzantine: three imported amphorae; one marble wall slab, bricks, tile; Byzantine–Turkish pottery. EM I–II farm, Roman/T village.

118. (131) Selima, Field 1200 and 1202 High (elevation 380 m) on a shoulder on the Thriphti slope, below the springs, this site is an area of 80 x 60 m with sherds and (V–T) tiles. Six springs are nearby. The largest scatter is of LPP and 18th– 19th-century sherds; PP and NP finds are less numerous. The site is a LPP–NP cyclopean farmstead. Artifacts: EM I–II: bowl, chipped stone; LPP: large jars; PP: cookwares, hole-mouth jars; NP: pithos, cup, jar, cookware; ground stone, quern. EM I–II farm, MM IA–LM I hamlet, V–T presence.

119. (132) Selima, Fields 1201 About 200 m south of site 118, there are two small concentrations of sherds below Selima: EM I–II, ca. 20 x 20 m; Early Byzantine, 30 x 30 m (with tiles). Artifacts: EM I–II: sherds, 42 chipped stone; Byzantine sherds and tiles. EM I–II seasonal field site, Byzantine field site.

120. (122) Selima, Field 1200 High (elevation 540 m) on the Thriphti slope, near the main pass eastward in the southern Isthmus, the cyclopean structure (Pl. 6C) sits on a bluff overlooking the entire valley of Episkopi and the approach to the pass. There are two cyclopean rooms (4.95 x 5.0+ m and 2.5 x ? m), possibly a “fort”(?). MM I–NP pottery scatter: 25 x 25 m. Six springs are present just below the bluff; there is some arable land around and to the east. A late Venetian–Ottoman metochi consisting of perhaps six houses was found around the springs; it was used during the German World War II occupation by women from the Isthmus. Artifacts (51 sherds): PP: one handmade cup, jars, cookware; two large chunks of non-local granodiorite. MM IB–II farm, MM III–LM I field site, V–T presence.

121. (134) Papadiana, Fields 1093 and 1094 This site is based on and around a conical hill (elevation 260 m) on the lower slope of Thriphti about 500 m east of the village of Papadiana. A streambed is situated to the north, and arable land is 150 m to the west and 70 m below. Sherd distributions are as follows: EM I–II, west slope: 60 x 10 m and east slope: 40 x 20 m; NP, west slope: 30 x 10 m and south slope: 40 x 30 m. Traces of PP cookware were found. Artifacts (Fig. 10:B127; Pl. 14:B35): EM: marble/calcite ware (EM I), chipped stone, six obsidian flakes; NP: large jars, one granodiorite jar, at least seven pithoi, pithoid jar, larnax(?); clay lumps (mudbrick?); chaff-tempered clay fragment. EM I–II farm, MM III–LM I farm.

122. (145) Papadiana, Field 1094 This site is on a slope at the base of Thriphti, next to a streambed just north of the Church of Timios Stavros, with arable land to the west. The spring of Kephalovrysi is 150 m to the south. Roman sherds and tiles are concentrated in an area of 40 x 150 m.

CATALOG OF SITES

129

Artifacts (375 sherds; Figs. 57:E38–E41, 58:E51): ER tiles, cookpot, amphorae, basins, jugs, pilgrim flask?; LR/Byzantine: cookpot. ER–Byzantine hamlet.

Artifacts (329 sherds; Fig. 53:E14): Beehive, tile, bricks; LR (beginning 3rd–4th century A.D.): cookware, large jar, amphorae, two beehives. LR hamlet, V hamlet.

123. (121) Episkopi, Fields 1114–1116

126. (128) Epano Chorio. Village, Fields 1136, 1137, and 1141–1143

On a low hill in the middle of the Episkopi Valley, 200 m northeast of village, this site is cut in half by the modern road to Ierapetra. In the middle of the Isthmus (not visible from either coast), the site flanks the riverbed and arable land. The sherd distribution is as follows: EM I–II, 40 x 65 m; PP and NP, ca. 40 x 30 m; LM IIIA–IIIB, 125 x 65 m. The site grows dramatically in LM IIIA. This is the settlement that produced the LM IIIA–IIIB Episkopi tombs excavated by Seager, Xanthoudides, and Platon (Kanta 1980, 146–160). Artifacts: EM I–II: cookware; PP: pithos, spouted jar (for wine?), jar, cup (MM II–III); NP: cookware; LM III: pithoi, cookware, large jar, bowls, LM IIIA–IIIB ladle; V–T: glazed wares, Siphnian tsoukali, sgraffito. EM I–II hamlet, MM IB–LM I farm, LM IIIA– IIIB hamlet, V–T presence. Kanta 1980, 146–160.

124. (129, 130) Episkopi Village and Fields 1116–1139 This site consists of masses of pottery immediately south of the village of Episkopi. Artifacts: Tiles, glazed wares, beehive, Thrapsanos pithos, tsoukalia (cooking pots), kiln separator, Chanakali Ware, ribbed wares. Early Byzantine and V–T presence.

125. (135) Papadiana, Fields 1123 and 1124 Just northwest of the village of Papadiana, there are scatters of Roman (60 x 60 m) and Venetian sherds (105 x 60 m). Bricks and tiles belong to a Roman farmstead. Later sherds (105 x 60 m) belong to the village.

This site is a pottery concentration found outside the present-day village of Epano Chorio. Artifacts: tiles, glazed wares, beehive, brick. V–T presence, perhaps part of Epano Chorio village.

127. (153) Kato Chorio, Field 1140 At the northern edge of the village of Kato Chorio, this site is a collection of Roman bricks, tiles, and sherds (40 x 80 m). Arable land and a riverbed are nearby. Inscriptions collected by Guarducci (1942, 28–29, 57, 62) mention possible estate boundaries in the area. Venetian and Turkish pottery is debris from the village. Artifacts (70 sherds): tiles, brick; LR: cookwares, one Phocean RS dish, one imported amphora. R hamlet, V–T presence. Guarducci 1942, 28–29, 57, 62; Sanders 1982, 140.

128. (133) Kato Chorio. Village, Fields 1149 and 1150 Northwest and south of the village, there are walls and Venetian–Ottoman sherds over an area of 140 x 150 m. Artifacts (310 sherds): tiles, giant basin, stamnostatis (stone platform for placement of stamnos), red-painted bowl. V–T village.

129. (143) Church of Stavromenos, Field 1217 On the lower steep slope of Thriphti, an exposed scarp revealed MM IIB–III pottery and field

130

L. VANCE WATROUS

blocks possibly from walls. Kephalovrysi Spring is 50 m to the north; good arable land is located 100 m to the northwest. Artifacts (595 sherds; Figs. 27:B376, B377, B379–B382, 28:B384, B386–B399, 29:B400– B404, B406–B408, 30:B409, B412–B420, 32:B468; Pls. 26:B172, B176, B178, 27:B180): five white Late Cycladic sherds, six dark ground, bevel-based Vaphio cups, cookware, oval mouth amphorae, tortoise-shell ripple decoration, conical cups, BSJs, flower pot, large spouted jar, carinated cups, small MM III saucers, large SS cup, rounded dark ground cup with spirals in added white, jugs, basin. Pottery similar to Pyrgos III. MM III/LM IA farm.

130. (144) Cave near Stavromenos, Field 1216 This site (Pl. 3B) consists of FN sherds (50 x 50 m) from inside and in front of a cave (elevation 340 m), on a steep slope 200 m southeast of the Church of Stavromenos. The cave is above (and 150 m from) Kephalovrysi, the largest spring in the Isthmus. Arable land is located 200 m to the northwest. Artifacts (239 sherds; Fig. 5:B47; Pl. 14:B34): Dark burnished sherds: two pithoi; gray ware sherd (EM I?), chipped stone, BG fragments (one bowl) left as H/ER votives? FN hamlet, EM I/IIA field site, H/ER shrine.

131. (147) Epano Chorio, Field 1220 This site (Pls. 4C–5B) is based on a natural ledge (elevation 280 m) within a ravine on the slope of Thriphti, about 500 m southeast of Epano Chorio. The site has one natural point of access along a gorge that runs along the south edge of the site. A rectangular structure of cyclopean walls blocks the path at the entrance to the site. The structure leaves a narrow (ca. 5 m) ledge one must pass through between the structure and the edge of the ravine before entering the site. Fortification walls of large cyclopean blocks run along the north and south at the edge of the cliffs. The site has cliffs on three sides. Three sections of house walls are visible within the site, which is overgrown. Good arable land is situated 200 m below. A ravine may have provided

water; a spring is located 350 m to northeast. Sherds are distributed as follows: EM I–EM II, 40 x 70 m; LPP, 80 x 80 m; PP, ca. 40 x 40 m; a few 19th–20thcentury sherds and possibly a havani (mortar for grinding coffee beans). Artifacts (366 sherds): EM I–II: red-brown and dark burnished sherds, jug, chipped stone; LPP: hole-mouth jar, pithos, bowl (lekanis); PP: cookware, pithos, SS cups; V–T sherds. EM I–MM IA hamlet, MM IB–II farm, V–T presence.

132. (149) Fields 1251 and 1252 Situated in a small saddle behind a steep hill (elevation 275 m) at the base of Thriphti, this site (Pl. 4A) overlooks arable land 100 m below. Early Minoan I–II sherds are scattered over 20 x 25 m. Artifacts (30 sherds): EM I–II dark burnished and cookware. EM I–II field site.

133. (148) Epano Chorio, Field 1252 At the base of Thriphti (200 m southeast of Epano Chorio), the site is on a gentle colluvial marl slope 120 m from the riverbed. Sherds were exposed by plowing for an olive terrace about 1.5 m below the present surface. Early Minoan I–IIA sherds are scattered over 50 x 80 m. Artifacts (Figs. 5:B57, 6:B60): black, brown, and red-burnished wares, gray ware foot (chalice?), pithos, chipped stone. EM I–IIA hamlet.

134. Aphendi Stavromenos, Peak Sanctuary On the peak top of Aphendi Stavromenos is a small clearing, 5 m southeast of the geological column on the peak (elevation 1,476 m, the highest peak in East Crete), which produced Minoan pottery and (65+) pebbles. The site is located at the only spot where Mochlos is visible (just to the east of Mount Kapsas). All of the Mirabello Bay and Isthmus of Ierapetra is also visible from this spot. A cistern dug into the site next to the chapel and xenona (guest house) has destroyed much of the site.

CATALOG OF SITES

Artifacts: PP(?): Thin oval tripod legs, cookware, fine ware, barbotine ware(?); kylix(?); NP: phyllite ware; LM IIIC cookpot, one blue ware. MM IB–LM I peak sanctuary, LM IIIC presence.

135. Vasiliki, outside Survey Zone Established in EM IIA, the site had at least four houses, including Theta, Xsi, and P, which were associated with obsidian, knapping debris, ground stone tools, a bronze tweezer, and caprine bones. In EM IIB, the site was reorganized: Zois’s Red House was built. Two-storied, it possessed a paved courtyard to the west and north and a well. This new construction probably comprised two houses. See Seager 1905, 1907, 1908; Zois 1992, 2000, for a fuller description. Artifacts: Much obsidian, 2 broken bronze axes, a mold for an ax and a knife, a steatite seal, a large vase stamped with an EM II seal, a Koumasa-type figurine, amphorae, loomweights, many jugs and cups and pithoi. EM IIB (in two new houses built on top of the paved court): pithoi, 70 loomweights. Rubble houses around (one produced a carbonized olive stone) perhaps a 80 x 40 m hamlet or perhaps larger. EM IIA–LM I village. Seager 1905, 1907, 1908; Zois 1992, 2000.

131

Artifacts: EM I–II: dark burnished, red-painted, cook and marble/calcite ware, chipped stone (including a core); LPP: pithoid jars, pithoi, bowl; PP: pithoid jar, jugs, amphora, cookware, brazier(?), carinated cups, basin, amphora; NP: pithos, jar, semi-globular cup; Archaic–Classical: basin; V–T: glazed ware and tiles. Four ground stone pieces. FN–EM II field site, EM III–MM IA farm, MM IB–II hamlet, MM III–LM I farm, Archaic– Classical presence, V–T farm.

138. (146) Iero Bakania, Fields 1170 and 1171 About 300 m west of Kato Chorio on the Isthmus floor near (100 m east of) a riverbed, a large concentration (286 pieces) of chipped stone (90 x 300 m) was found in mostly brown alluvial soil. Patches of original red soil were present, though there was no sign of prehistoric habitation debris. Stone scatter may represent the halo (perhaps some washed) of a site that lies buried in the heavy colluvium to the east. EM knapping site.

139. (152) Fields 1165–1168

Some 250 m south of Epano Chorio, the site is on top of a steep spur at the base of Thriphti. It has similar topography and catchment to site 132. The EM I–II scatter covers 40 x 40 m. Artifacts (Fig. 10:B125): EM I–II chipped stone (and four obsidian pieces), dark burnished and red-painted pottery. EM I–II farm.

Located 600 m west of Kato Chorio in the plain near the riverbed, this site consists of Roman tiles, sherds and an ancient(?) wall over an area 150 x 120 m. Three Minoan sherds and a stone drill guide represent earlier activity. Artifacts (258 sherds): Tiles, brick, millstone; ER: amphorae (one Campanian), cookpot, cookware jug, ER red-painted plates, giant pithos, basin; LR: North African amphora (3rd–4th century A.D.), three Phocean RS dishes, two ARS plates, amphorae (one micaceous fabric). LH–Roman village, Early Byzantine presence.

137. (125) Kato Chorio, Field 1125

140. (124) Kato Chorio, Field 1180

At the eastern edge of the village, the site flanks a riverbed and good arable land. Sherds were distributed as follows: Neolithic, one sherd; EM I–II, ca. 15 x 15 m; LPP, ca. 40 x 30 m(?); PP, ca. 60 x 50 m; NP, ca. 40 x 40(?) m; V–T, 80 x 15 m. The site is below 154 and may be seasonally related to it. Walls are extant.

Located at the southwest edge of the village of Kato Chorio on the south side of the asphalt road, the site consists of PP–NP sherds (30 x 20 m), cyclopean walls, a rock cutting (water channel?), and stone tools. A football-sized chunk of granodiorite was also found. Much destroyed, the site is situated next to a riverbed and at the base of

136. (150) Epano Chorio, Field 1253

132

L. VANCE WATROUS

Thriphti where the plain of Ierapetra opens out to the south. This site is at the boundary between the Ierapetra Plain and the valley of Episkopi. Artifacts (17 sherds; Fig. 12:B151): PP: pithoid jar, cookware. MM IB–III field site.

141. (142) Kato Chorio, Field 1229 At the base of Thriphti, 250 m southeast of village, a concentration of sherds (55 x 20 m) was found near a riverbed and arable land. Artifacts (92 sherds): Thrapsanos pithos, glazed wares. Ottoman farm.

142. (136) Kato Chorio, Field 1234 On the slope of Thriphti southwest of the village and 60 m above good arable land, a scatter (30 x 40 m) of sherds was found on a level area (near ravine). Artifacts (170 sherds; Fig. 4:B36, B37, B40, B41): EM: EM I marble/calcite and dark burnished ware, EM II light on dark (two EM I) pithoi, jug, jar, bowl. EM I–II farm.

143. (137) Field 1233 On the slope 20 m above (southeast) site 142, a concentration (20 x 15 m) of PP and NP sherds were found. Two pithoid jars (burials?) were set on the southwest edge. Artifacts (190 sherds; Pl. 26:B171, B174, B175, B177): Huge pithoid jars, cookpots; NP: cookware, jar. MM IB/II–MM III/LM I field site.

144. (154) Kato Chorio, Field 1265 About 300 m southeast of Kato Chorio, the site (Pl. 8A) is located (100 m above the valley floor) on a ridge top overlooking Episkopi Valley. Surrounded on all sides by steep slopes (except for the narrow ridge top that forms the natural entrance to the site on the southwest), the site has a massive cyclopean (fortification?) wall and stretches of similar walling around the edge of the site across the ridge top on the southwest. A possible second entrance appears on the northeast, which is a narrow cutting in the

rock slope leading down to the riverbed below (as at Karphi). Cyclopean walling flanks the entrance on both sides; additionally, a lower wall to the west of the entrance may be the remains of a bastion. Little good arable land can be found nearby. A spring is located 200 m to the south (at site 149?). The site as defined by walls is 60 x 30 m; with sherd scatter (wash?) on slopes: ca. 90 x 40 m. A cave is located at the southeast corner of the site. Larnax and pithoi on slopes (burials?): EM I–II, 50 x 40 m; LPP, 50 x 100 m; PP, 50 x 100 m; LM III, 50 x 100 (traces of LM IIIC). Cyclopean wall may date to LPP. Artifacts (Figs. 11:B131, 12:B148): EM I–II, marble/calcite, red-burnished wiped ware, pithos, jar, burnished bowl, perforated pithos rim, obsidian arrowhead(?); LPP: cups, jar, basins, jug, pithoi; PP: jars, pithoi, pithoid jar, cookware, SS, conical, semiglobular cups, basin; LM III: deep bowl, larnax. Early Iron Age: two skyphoi. Stone bowl fragment. EM I–MM II hamlet, LM IIIA–IIIB hamlet, LM IIIC farm, Archaic field site. Possible LM IIIA–IIIB cemetery.

145. (155) Above Field 1265 Situated in a hidden valley behind (south of) site 144, a small concentration of house walls and sherds represents an 18th–20th-century hamlet. A Greek coin of 1869 was also found. A spring was located above what may be a sterna (cistern). There is little arable land nearby. Ottoman presence.

146–149. Vacat 150. (106) Prophetes Elias About 500 m south of Kato Chorio, this large site is on the top, east, and west sides of a steep conical hill 220 m above the valley floor (Pl. 3A). The nearest arable land is in the Episkopi Valley below. House walls are visible on the western slope. LM IIIC–PG, 150 x 340 m; Geometric, 240 x 380 m; Orientalizing–Archaic, at least 150 x 280 m; Classical, 265 x 415 m; Roman, 65 x 200 m; Ottoman (17th–19th century), 100 x 60 m. The settlement reaches its largest extent during the Classical period. The east slope is reoccupied during the ER–LR period and in the 17th–19th century.

CATALOG OF SITES

Artifacts (8,318 sherds; Figs. 4:B38, 12:B153, B158, 35–46:C1–C143, 47:D2, D4, D6–D9, D28, D30, D32–D34, D38, 48:D39, D44, D47, D49, D51, D52, D56, D57, D67, D79, 53:E10, 59:E60, E61; Pls. 28–37:C1–C99, 38:D1, D3–D5, D9–D15, 39:D16–D26, D29, 40:D31, D33–D38, D40–D42, 41:D43–D46, D48, D50, D51, 42:D59–D73, 43:D74–D80, D83, D89, D91): LM IIIC/PG: cookware, stoppers, kraters, lekanides, deep bowls, skyphoi (PG bell skyphoi), pendent semicircle cups, incised/impressed pithoi; Geometric: incised basin, cookware, blob cup, krater, imported (island?) cup, Late Geometric metope cup, short-necked cup, bowls, jug, Protogeometric B cups decorated with quirks, Middle Geometric krater, pyxis, brazier; Archaic: many heavy mortars (some Classical?), BG cups, banded jugs, cookware, krater, relief pithoi (rosette, guilloche stamped), monochrome cup, high-necked cup, skyphos; Classical: tiles, much BG, BG basin, BG cups, BG kantharos, BG kraters, BG lekanis, spindle whorl, one Attic import, BG skyphos, many pithoi (one stamped with a palmette), amphorae, mortars; Early Iron Age: several giant basins and pithoi; Roman: tiles, cookware;

133

ER: two Italian Terra Sigillata, one Eastern Sigillata B, amphorae (double-rolled handle import?); LR: one ARS plate, two Phocean RS dishes, cookpots, six amphorae (one import), five basins, two jars, one bowl, two jugs; Ottoman: tiles, sgraffito, foliateimpressed pithos, glazed wares, drip ware, pithoi, pilgrim flasks, stamnos; stone hand tool, hand mill, whetstone. LM IIIC–Classical town, Roman/V–T hamlet.

151. (151) Arkolounero, Field 1261 Set on a small spur in the middle of an upland valley at an elevation of 400 m, two walls enclose an area 15 x 15 m; the sherd scatter covers 40 x 75 m. Sherds include MM IA and PP. The site was established in MM IA. Artifacts (Figs. 14:B180, B186, 18:B244, B245, 19:B259, B260, 22:B312; Pl. 18:B72): LPP: holemouth jars, basins (lekanides), cookware, handmade cups, pithos; PP: cookware, amphora, carinated cup, Vaphio cup, pithoid jar, hole-mouth jar, dark ground cups; four pieces of chipped stone. MM IA–II hamlet.

Appendix B

Neolithic and Bronze Age Pottery Donald Haggis

This chapter presents an overview of the Neolithic and Bronze Age pottery that was recovered in the survey of the Gournia Valley and the north Isthmus of Ierapetra. The main purpose here is to describe the kinds of pottery found in the survey, the characteristics of Neolithic and Bronze Age ceramics in the region, and observable changes in forms and fabrics through time. Another aim is to summarize the criteria used to date specific potsherds and individual sites in the survey zone. The study of the pottery was conducted primarily to establish the periods of occupation on specific sites, while providing chronological and functional information on material collected in transects, that is, the “off-site” material. At the beginning of the survey, it was often difficult to define the exact date and form of many Bronze Age potsherds found scattered about fields during the initial stage of field walking. These potsherds were usually coarse wares, in poor condition because of various surface conditions and exposure to the elements, and they were collected irrespective of period or type along 100-m lines in 1-ha-square transects. Furthermore, because of their size and durability, fragments of

coarse-ware vessels were more prevalent and obvious on the site surface, and therefore more likely to be picked up by field walkers traversing a field than tiny fragments of small fine-ware vessels. This selection bias, along with the formation processes of the potsherd itself as an artifact, suggests that coarse wares constitute a majority of the material collected off-site in surface surveys in Greece. This was the case in the Gournia Project, and efforts were made to refine a coarse-ware ceramic chronology for the Bronze Age, with the purpose of dating and defining the origin and function of off-site material, as well as supplementing the information from the fine wares recovered in on-site samples. From the outset, we used a skeletal coarse-ware fabric typology that had been established for survey in the neighboring Kavousi region (Haggis and Mook 1993), but at every stage we attempted to reestablish the periodicity of types and wares by examining the stratified excavation assemblages and unstratified survey finds in the Mirabello Bay area. The nearby excavations at Mochlos provided a number of sources for comparison with material in the survey zone, particularly from levels dated to EM

136

DONALD HAGGIS

IIB, EM III–MM IA, MM IIB, MM III, LM IB, and LM III. Excavation in the Kavousi area at Vronda and Chrysokamino also produced stratified assemblages of Late Prepalatial date, and the Kastro offered an unbroken LM IIIC–Early Orientalizing sequence. Further afield, excavations at Kalo Chorio (EM I) and Myrtos Phournou Koriphi (EM IIA–IIB) were important sources of comparanda during both fieldwork and study phases of the project. Examination (and in some cases repeated reexamination) of this excavation material, stored in the INSTAP Study Center for East Crete, was not only useful in dating the fine-ware pottery derived from the survey, but in reaffirming and, in some cases, adjusting the chronological parameters of specific coarse-ware fabrics, ware groups, and shapes. It is rare to have so many excavations contiguous to and concurrent with a survey project, and perhaps also unusual to work in a region that overlaps two identifiable ceramic production areas (Day 1991, 1995, 1997). The Gournia survey area is located at the transition between two broad zones that are distinctive in their bedrock geology, each region evidently providing different source materials for clays and inclusions in the pottery originating in these areas. The rock and mineral inclusions, either naturally present in the clay or added as temper, are macroscopically observable, especially in coarseware pottery. Earlier surveys in the immediately neighboring regions of Vrokastro (Hayden, Moody, and Rackham 1992; Hayden 2004a, 2005) and Kavousi (Haggis and Mook 1993) had long recognized not only extraordinary variation in the coarse fabrics recovered on Bronze Age and Early Iron Age sites in those areas, but also periodic changes in these fabrics that could be correlated to changes in shapes and differences in the date of the associated fine wares. In our study of diachronic changes in coarse fabric types in Gournia, we were aware of two important aspects of production: (1) on the regional scale, pots were evidently produced in these two distinct geological zones of the north Isthmus and Mirabello coast—one in the area between Gournia and Priniatikos Pyrgos, containing granodiorite, and another on the eastern side of the Isthmus from Pacheia Ammos to Mochlos, which contained obvious sources of phyllite, quartz, and quartzite; (2) among the samples derived from the Gournia survey zone, it was clear that the distribution of

certain coarse-ware vessels that were made of fabrics derived from these two different sources varied through time. Furthermore, this diachronic variation was corroborated by the examination of pottery from stratified excavation deposits in both of these production zones, and by observable changes in the shapes, styles, and wares of the vessels themselves. In light of this extraordinary variation, the situation of the Gournia Project study area was advantageous for examining evidence of ceramic production, distribution, and consumption. This proved to be important in understanding relationships among sites, the function of sites, and the extent of hypothetical culture regions—those areas of similar or identical artifact types. Changes in the coarse-ware fabric types appeared along with changes in the ranges of shapes that were represented in each fabric type group that we defined, suggesting shifting patterns of diversification and standardization of production though time. With the exception of Gournia (8) and the Roman site 86, it has been difficult to locate specific workshops or kiln sites, which through excavation could tell us more about the organization and administration of production. Even so, the observable changes in fabrics and forms do suggest that modes and locales of production did indeed change, perhaps in response to economic and sociopolitical changes on a broader regional scale. The emphasis on coarse-ware fabrics also afforded us the opportunity to assess changes in the shapes and functions of a variety of utilitarian vessels, which are, of course, the most common survey finds but also frequently underemphasized, if not omitted, in most excavation reports. This neglect of coarse wares derived from excavation has hindered their potential use in survey, especially in reconstructing broad patterns of land use and economic organization in the landscape. The diverse range of storage, transport, and industrial functions represented by the great variety of coarse-ware types collected in “off-site” samples offered a means to explore different kinds of agricultural and industrial activities throughout the countryside (cf. Watrous and Blitzer 1999, 907). Although it is generally acknowledged that the shapes of some coarse-ware vessels do change through time, e.g., the common Minoan tripod cooking pot (and indeed the section of the cooking tripod leg; cf. Betancourt 1980; Watrous 1982, 68;

NEOLITHIC AND BRONZE AGE POTTERY

Haggis and Mook 1993, 269), archaeologists are still reluctant to try to develop form typologies of other coarse wares, such as storage, transport, cooking, and processing vessels. This assumes conservatism in the coarse-ware production tradition, predicating such a slow rate of change as to render a typology irrelevant to the dating of an excavation level or site. In the context of the objectives of excavation, coarse wares indeed do not seem to offer a very refined chronology, especially when compared with the information derived from fine tablewares. For survey, however, the hypothetical slower rate of change may still be accommodated by the coarser chronological resolution (broader chronological categories) usually offered by surface remains. That is to say, the changes observable in coarse-ware shapes and fabrics provided as much information about the periods of occupation at a given site as most surface conditions will normally allow. In

137

fact, the chronology suggested by the coarse ware fabric and shape typologies at Gournia was scarcely broader or less certain than that of the fine wares. For most sites, the coarse wares and fine wares independently suggested broad chronological ranges, or separate “phases” of occupation, such as FN, EM I–II, EM III–MM IA, MM IB–II, MM III–LM I, LM IIIA–IIIB, and LM IIIC. Even though individual, well-dated fine potsherds—such as “type fossils”—could be dated with precision (to, for example, EM IIB or MM IIB–IIIA), the definition of any single period by any single potsherd was rarely very helpful in actually narrowing the range of dates proposed for any site. Given that we lacked stratified excavation samples for each and every site, we found that the necessarily broader chronological divisions were effective deterrents to “reading” surface collections as if they were synchronous deposits or systemic assemblages.

Neolithic Neolithic pottery was found throughout the survey zone, but only a few concentrations were identified with a sufficiently substantial component to be called actual sites. The fabrics and shapes described below indicate a date in the Final Neolithic, but we are not certain that the sites are exactly contemporaneous. Sites 3 and 98 may serve here to represent the range of shapes and fabrics recovered across the survey zone. Furthermore, the potsherds from these two sites look very different from each other, reflecting very different geological environments and production traditions. The coarse-textured fabrics have inclusions that are derived from bedrock and clay sources that are found in the likely areas of ceramic production: phyllite, quartzite, quartz, and calcite of the eastern Isthmus and western Siteia Mountains, and granodiorite and quartz of the western zone of the Gournia Valley.

Red and Red-Black Mottled Coarse Wares The pottery from site 98 forms a homogeneous group with good typological links to Kastelli Phournis, Nerokourou, and Chrysokamino Kavousi

(Manteli 1992; Vagnetti 1996; Betancourt et al. 1999, 354–356). The most common and distinctive wares have a dull red (Fig. 1:B1) or reddish-black mottled (Fig. 1:B2–B11; Pl. 11:B1, B2, B5, B8) exterior surface, which is carefully smoothed or lightly burnished. The cores are usually evenly fired red or dark gray, with color differentiation apparent only at the exterior surface, which has a 1.0 to 2.0mm-thick red (2.5YR 5/6), black (2.5YR 4/0), or reddish-black mottled (2.5YR 5/8–4/8) layer or slip. The surfaces of FN sherds are generally worn, and the burnish of the exterior slip has been dulled, cracked, and in some cases, fractured, causing pieces to flake off the core itself. Some examples (Fig. 1:B12, B13; Pl. 11:B7) have a friable and distinctly grayish-brown or yellowish-brown core, suggesting variations in firing practices. Red-slipped varieties are covered with a thick, opaque red layer of clay (2.5YR 4/6; 10R 4/6) that appears to have been very evenly smoothed. Fragments of bowls are usually burnished red (Fig. 1:B5, B10) or black (Fig. 2:B14) on their interiors, though jars such as B15 have rough or unevenly smoothed interior surfaces (Fig. 2:B15). The coarse fabrics from site 98, located on the eastern edge of the survey zone in a geological zone

138

DONALD HAGGIS

of phyllite-quartzite, contain distinctive sub-angular quartz, quartzite, and phyllite inclusions (1.0 mm and smaller), microscopic quartz sand, and occasionally calcite particles. The look of the fabric is very gritty, and chaff voids are evident on the surfaces. The fabrics of small vessels, such as tiny strap handles (Fig. 1:B4), are densely packed with microscopic fine, sandlike particles. The larger particles, examined under a 10x hand lens, look like white and yellow quartz and calcite. Phyllite inclusions are red, green, and gray but are generally difficult to distinguish in the evenly colored cores of most sherds. They are more obvious in examples with eroded surfaces. Although elongated and laminated schist pieces do appear rarely, the visible phyllites are most commonly sub-rounded mudstones, fusing with the color and fabric of the core. The shapes are evidently round and concave bottom bowls and jars. No jugs could be identified with certainty. Bases were extremely rare, and no articulated or flat bases were recovered, which is unusual given their prevalence at contemporary FN sites, such as Nerokourou and Phourni. The most common handle type is a horned, perforated handle (Fig. 2:B14–B17) belonging to fairly thick-walled bowls and necked jars (cf. Vagnetti 1972–1973, 57, 76; Vagnetti, Christopoulou, and Tzedakis 1989, 36–37, fig. 21:80). The red-slipped handle in Figure 1:B1, for example, even in its fragmentary state, is 5.0 cm in height and has preserved on its inner face the black-polished interior surface of the bowl. Other handle types include large tubular lugs (Fig. 2:B18), wide strap handles, and miniature strap handles of very thin-walled jars and bowls (Figs. 1:B4, B5, 2:B19). Bowl rims are straight (Fig. 1:B6–B9) or inturning (Fig. 1:B10, B13), and one unusual example (Fig. 1:B11) is thickened, squared, and incised with a chevron pattern. Incised decoration is found on the exterior surfaces of bowl fragments. Rows of chevrons (Fig. 1:B2; Pl. 11:B5) and zigzags with horizontal lines (Fig. 1:B3) are features of carinated bowls, and comparable to examples from Nerokourou and Phaistos (Vagnetti, Christopoulou, and Tzedakis 1989, 48, 52–53, figs. 27:150, 29:183–185). The overall design principle is scarcely indicated by the few incised fragments from site 98, but the handling of the incision of the chevrons (Fig. 1:B2, B11; Pl. 11:B5) suggests an “open style” similar to examples from FN deposits at Phaistos (Manteli and Evely

1995, 10–11), though the zigzag fragment reflects aspects of the more “closed style” at Late Neolithic Knossos (cf. Furness 1953; Evans 1964).

Grayish-Brown Burnished Ware Two fine-ware sherds from site 98 are fragments of carinated bowls. One of them is a vertical handle (Fig. 2:B20; Pl. 11:B4) with two parallel vertical grooves incised symmetrically down the spine. The other (Fig. 2:B21; Pl. 11:B6) is a bowl rim with an impressed linear pattern—two sets of vertical and horizontal parallel lines, the outer set framing the inner one. The pinched rim is a feature of both LN and FN carinated bowls (Evans 1964, 195, fig. 38:22, 23, 29; Vagnetti, Christopoulou, and Tzedakis 1989, 43, fig. 25:134), but probably more common in the Late Neolithic. The sherds are light and smooth to the touch, the fabric is a porous light gray to grayish-brown clay with no visible inclusions, and the surfaces have a high polish.

Dark Grayish-Brown Burnished Wares A ware commonly found on the western side of the survey zone, and well represented at site 3 in the Gournia Valley, has dark gray to grayish-brown (7.5YR 4/1) to brown (10YR 5/2; 5YR 5/4) burnished surfaces, usually with an extremely high polish. The fabric is characterized by a dark gray color and dense, packed, equidimensional, and microscopic quartz and granodiorite inclusions. Calcium carbonate inclusions were not apparent, though samples of the fabric were tested with hydrochloric acid. The density and even distribution of the fine white, yellow, and black particles may indicate that these materials were added to the clay as temper. Although the fabric resembles some smaller examples of the red and reddishblack mottled wares, the inclusions here are denser and more evenly sorted than those of wares from site 98. Some brown burnished-ware jars have thickened collars (Fig. 3:B22–B24) or are simply inturned (Fig. 3:B25), though bowls exhibit characteristic pinched out (Fig. 3:B26) or sharply everted rims (Fig. 3:B27; cf. Vagnetti 1972–1973, 65) and high

NEOLITHIC AND BRONZE AGE POTTERY

burnish on the interior. The jar rim illustrated in Figure 1:B2 is black burnished on the interior. A dark gray bowl (Fig. 3:B28) has a high polish on the exterior, and the interior has a 1-cm-wide reserved band below the rim, suggesting pattern burnishing typical of FN–EM I across the island. The large bowl rim in Figure 3:B29 is made of the coarsetextured fabric similar to other grayish-brown burnished wares, but the burnishing is on the interior, while the exterior is heavily wiped or scraped. Both pattern burnishing and scraping are known in FN levels at Phaistos (Vagnetti 1972–1973) and Partira (Mortzos 1972), and at Knossos they are LN features, which continue into EM I (Hood 1990, 369). Very large bowl fragments were also found at site 3. The bowl handle (Fig. 3:B30) is a wide saddled strap that is lightly burnished brown to reddish brown (5YR 5/6) on the exterior and burnished black on the interior (7.5YR 4/3–5/4); and the bowl in Figure 3:B31 is of similar size and reddish brown

139

in color, also burnished on the interior, though its handle is not evidently saddled in the preserved portion (Vagnetti, Christopoulou, and Tzedakis 1989, 34, fig. 20:75). Two grayish-brown, burnished, horizontal elliptical handles present problems. One of them is only the start of the handle, rising above the bowl rim (Fig. 3:B32), so not much can be said about its form. The other example (Fig. 3:B33), however, is likely to be a wishbone type, common in Early Neolithic at Knossos. The presence of such handles at Nerokourou and in the FN levels at Phaistos (Vagnetti, Christopoulou, and Tzedakis 1989, 42, figs. 24–121) begs the question, implying a lingering use of the type down into the Final Neolithic, at least outside of Knossos (Vagnetti 1996, 32–33). While the burnished wares from site 3 appear to differ remarkably from the dull red and red-black mottled potsherds from site 98, there were simply an insufficient number of shapes recovered from either site to indicate a difference in date.

Early Minoan I–II The ceramic phases spanning EM I and II are well documented in East Crete (Map 5). Published stratified deposits of EM I at Kalo Chorio–Istron (Haggis 1996a) and Periods I and II (EM IIA and IIB) at Myrtos Phournou Koriphi (Warren 1972) provide the closest comparanda for material of these periods at Gournia. Although more distant, Knossos itself offers a fairly detailed picture of EM I (Hood 1990; Wilson and Day 2000), EM IIA (Wilson 1985; Wilson and Day 1994), and EM IIB (Momigliano and Wilson 1996; Wilson and Day 1999). Furthermore, recent work on ceramic regionalism in EM I–II eastern Crete (e.g., Day, Wilson, and Kiriatzi 1997; Whitelaw et al. 1997) has provided the framework for visualizing ceramic production and distribution in these periods in the Mirabello area. Rarely could single-phase or single-period Prepalatial sites be determined with certainty in the study area. The purpose of the present discussion is to explain the rationale for including the phases— EM I and EM IIA–IIB—in the site descriptions, by describing the fabrics and ware groups represented in survey samples and their correlates in excavation assemblages.

Dark-on-Light Painted Wares Dark-on-light wares were recovered in small quantities on a number of EM I–II sites, but surfaces are usually not preserved well enough to indicate the specific kind of painted decoration. The fabrics, however, are consistent with examples from published assemblages at Kalo Chorio (Haggis 1996a) and Myrtos Phournou Koriphi (Warren 1972). The dark-on-light sherds have a distinctive pinkish-buff core (5YR 6/6–6/8; 5YR 7/6–6/6) that is commonly salmon colored, a yellow or buff slip (10YR 8/6–7/6; 7.5YR 7/6), and in examples that have paint preserved, the decoration is red to reddish brown (5YR 5/4–5/6; 2.5YR 5/8) or dark brown (5YR 4/3). The fabrics are on the whole very hard and well fired, with evenly colored cores and evenly distributed sub-angular red and purple phyllite inclusions, mixed with rounded gray and white quartz and quartzite bits. Very rare are chalky white particles that protrude from the surface slip. Collared jar and jug fragments are the most common shapes in dark-on-light ware. The fabric of jar rims in Figures 4:B34, B35 is consistent with that of

140

DONALD HAGGIS

EM I wares from Kalo Chorio (Haggis 1996a, 668), and the shapes are not likely to be later than EM IIA (Blackman and Branigan 1982, 32; cf. Wilson 1985, 322–347, nos. P195, P327, P395). The jug wall fragment in Figure 4:B36, with its radiating vertical stripes in a thin washy orange slip applied directly to the pink-buff ground of the vessel, is identical to wares from Kalo Chorio (Haggis 1996a, 666, 667, 669, figs. 24:KT35, 26:KT56, KT57, 29:KT85). It is taken here to be EM I, though jug fragment Figure 4:B37 from site 142 is likely to be from an EM IIA south-coast import. They are very hard fabrics with a pink-buff core, evenly applied slip, and sub-rounded red, gray, and white inclusions and distinctly rounded gray and black pieces, which are probably basaltic and serpentiniferous rocks from the area west of the Myrtos Valley (Warren 1972, 94–95; Whitelaw et al. 1997, 268). Very fine EM IIA dark-on-light wares include Figure 4:B38, a small jar with a knob on the shoulder and the start of an outturned collar (Wilson 1985, 311, fig. 16:P120), and the bowl rim in Figure 4:B39 (Warren 1972, 155, fig. 39:P13, P17). The fabric of these pieces is a very fine buff (7.5YR 8/6–7/6). Where the slip is worn away, the fabric appears soft and porous to the touch. Large shallow bowls (Fig. 4:B40, B41) have slightly inturning rims, and the fabric suggests local production. The clay is a pinkish-buff color, and identifiable inclusions are very large angular quartz and granodiorite pieces. While only traces of paint are preserved on the sherd illustrated in Figure 4:B40, the other rim in Figure 4:B41 has an internal and external rim band, comparable to published examples from Knossos (cf. Wilson and Day 1999, 11, fig. 3:P47; Warren 1972, 155, fig. 39:P28). Also of local manufacture is Figure 4:B42, which has a similar fabric but a monochrome black burnished slip.

Fine Gray Ware Fine gray ware, a characteristic feature of EM IIA, was found in very fragmentary examples at a number of EM I–II sites. Even a tiny, otherwise undiagnostic sherd is very distinctive. The fabric is fine, and surfaces often are burnished to a soapy smooth polish. The core is fired evenly with no visible inclusions, and the slip rarely is differentiated.

The color is light gray (2.5YR 7/2; 5Y 7/1–7/2) to dark gray (5Y 6/1–5/1), and occasionally the slip is a mottled orangish-gray color. Several examples include goblets (Fig. 4:B43; cf. Haggis 1996a, 670, fig. 31:KT98), bowls (Figs. 4:B43, B44, 5:B45), a carinated bowl or pyxis (Fig. 5:B46), jars with a low collar (Fig. 5:B47; cf. Warren 1972, 101, no. P32), and molded bases common on spouted bowls from Myrtos in EM IIB (Warren 1972, 170, fig. 54:P247, P694). It is interesting that although dark gray or “Pyrgos” ware was found to be the most common EM I ware for open shapes in excavation contexts at Kalo Chorio (Haggis 1996a, 664–668; 1997), it was not identified with certainty in the Gournia survey samples.

Vasiliki Ware Vasiliki Ware, a veritable type fossil for EM IIB in eastern Crete, was found in surprisingly small amounts and very rarely in recognizably diagnostic shapes. Jugs (Fig. 5:B48) and goblets (Fig. 5:B49) are evidently the most common shapes (Warren 1972, 127–130, nos. P372, P419). While Vasiliki Ware has been long recognized as local to the north Isthmus area (Warren 1972; Betancourt et al. 1979), recent analyses of the pottery from Phournou Koriphi demonstrate the widespread distribution of wares originating in the northwest Ierapetra Isthmus and southeastern Bay of Mirabello (Whitelaw et al. 1997, 268).

Marble-Tempered Ware Marble/calcite-tempered ware, commonly recovered from coastal sites in the Gournia area, is perhaps the most distinctive of the EM I–IIA coarse wares. The fabric is the equivalent of fabric Type IX in the Kavousi area (Haggis and Mook 1993, 275), cookpot ware at Kalo Chorio (Haggis 1996a, 669), and fabric Type 8 at Myrtos (Warren 1972, 95). It has a reddish-brown to orange clay (5YR 4/6), a diffuse grayish-brown core (10YR 5/2–4/2), and both small (< 1.0 mm) and large (1–3 mm) white, opaque, angular inclusions. Some microscopic grayish-black particles are present in most samples. While surfaces are rarely preserved, they appear to have been covered by a thin slip, lightly burnished

NEOLITHIC AND BRONZE AGE POTTERY

or smoothed, usually permitting the white angular bits to show through. Horizontal lugs (Fig. 5:B50, B51; Pl. 12:B13, B14) belong to the walls of pedestaled bowls, and large bases (Fig. 5:B52), and handles (Fig. 5:B53; Pl. 12:B16) are pieces of jars and cooking pots.

Grog-Tempered Cooking Ware Grog-tempered cooking-ware fabric has a brown to reddish-brown core, and a reddish-brown or reddish-black mottled surface (2.5YR 4/4–5/6). The surfaces of open shapes, such as cheese pots (Fig. 5:B54; Pl. 14:B37) and trays (Fig. 5:B55), are burnished on both the interior and exterior, but the cores are very coarsely textured and crumbly. Inclusions are large white, gray, and yellow subangular pieces, and what appears to be mudstone or grog temper that is identical in color to the fabric matrix, making them difficult to discern without a 10x hand lens. The bottom of the tray in Figure 5:B55 is burned black on the exterior, confirming the “cooking-ware” attributed function.

Black-Burnished Ware Black-burnished ware is common on EM I–II sites (cf. Blackman and Branigan 1982, 29) and is evidently used for a wide variety of bowls, small basins, and jars. The core is dark gray or grayish brown (10YR 3/1–3/3), occasionally reddish brown (5YR 4/4), with densely packed microscopic white particles, probably quartz, black-and-white granodiorite, and gold biotite. The interiors of bowls are wiped, smoothed, or burnished, while the exteriors usually have a dull burnish or high polish (Fig. 5:B56; Pl. 12:B9). The fabric is identical to blackburnished ware from EM I contexts at Kalo Chorio (Haggis 1996a, 670). The steep-sided bowl rim in Figure 5:B57 is a distinctive EM IIA shape found in cooking pot and painted coarse wares at Knossos (Wilson 1985, 339, no. P322). The bowl rim in Figure 5:B58 has a smoothed exterior and burnished interior, and the carinated bowl fragment in Figure 5:B59 has a wiped interior. These latter pieces are possibly FN–EM I.

141

Red-Slipped Ware Red-slipped ware sherds were rare in the Gournia area. Examples are fragments of open shapes, probably bowls (Fig. 6:B60) and a collared jar (Fig. 6:B61). The clay is a fine light red to reddish-yellow (2.5YR 6/6; 5YR 6/6–5/6) fabric, and there are traces of burnished and washy slip (2.5YR 4/4–4/6). Although the ware group at Gournia is hardly large or consistent enough to make secure comparisons to stratified deposits, some links can be made to EM II assemblages. Collared jars in red-washed and red-burnished wares are found in EM IIA at Knossos (Wilson 1985, 319, no. P173). A number of bowls and jars from Myrtos (Warren 1972, 35, nos. P29, P30) have a red washy painted exterior similar to examples from Gournia.

Early Minoan Medium Coarse (EMMC) One coarse fabric, Early Minoan Medium Coarse (EMMC), was found consistently on EM II sites across the survey zone. The wide range of vessels produced in this ware includes bowls, steep-sided bowls, lekanides, spouted bowls, piriform jars, jugs, and small pithoi. The forms recovered in the Gournia area are consistent with published EM II examples from Myrtos Phournou Koriphi and Knossos. The core is gray to greenish gray or grayish brown (2.5Y 4/2–5/2) at the center, diffusing to orange or pinkish-buff (5YR 6/8–5/8) at the edges. The slip is usually buff colored (10YR 6/4). In some examples, the core is pink (5YR 6/6), pink buff (7.5YR 6/6), or tan (7.5YR 6/4), and frequently a dark red or gray (10YR 5/8; 7.5YR 5/0) washy slip was applied to the buff slip or grayish-brown ground of the vessel. The fabric is very gritty and sandy looking, with densely packed microscopic inclusions, generally smaller than 1.0 mm in size. Identifiable inclusions are quartz, feldspar, granodiorite, and biotite. Distinctly larger and more prominent (1–2 mm) white quartz, and white and black granodiorite pieces appear in larger vessels, such as the jars and pithoi. The presence of granodiorite in the EMMC fabric suggests local manufacture, but

142

DONALD HAGGIS

also a distribution considerably wider than the eastern Mirabello Bay. The ware is identical to fabric Types 3 and 5 from Myrtos Phournou Koriphi (Warren 1972, 94–95). Common shapes are collared jars (Fig. 6:B62; Pl. 13:B32), Warren’s piriform jars (Warren 1972, no. P579), and jugs (Fig. 6:B63, B64). Wide-mouth (Fig. 6:B65, B66; Pl. 13:B21) and collared (Fig. 6:B67) pithoi are easy to distinguish in survey samples and have good parallels at Myrtos (cf. Warren 1972, 159, 197–198, figs. 43:P74, 81:P620, P622, 82:P627, P628). Deep bowls (Fig. 6:B68, B69) are common EM II shapes recognizable in surface collections. One hole-mouth jar type with an articulated rim (Fig. 6:B70; Pl. 13:B27) is found at Knossos in EM IIA (cf. Wilson 1985, 338, 340, figs. 32:P320, 33:P323, P324), but not at Myrtos, perhaps suggesting an early date in EM II for the type. Large deep bowls (Fig. 6:B71–B73) are thickened on the rim interior and exterior, sometimes with an emphatic T shape. The bowls generally have very thin walls, sometimes less than 5.0 mm thick. Wide-mouthed pithoi and collared varieties (Fig. 6:B72), in fragmentary form, are difficult to distinguish from large, deep bowls since the fabric and surface treatment are essentially the same. The rim in Figure 6:B74 is probably a pithos, given the stance of the profile. Bases with distinctive grooves (Fig. 7:B75–B77; Pl. 13:B30, B31) are easily recognizable as fragments of spouted bowls (cf. Warren 1972, nos. P251, P252) or small deep open bowls common in assemblages at Myrtos. Plain bowl bases (Fig. 7:B78, B79) were also found in EMMC fabric. They are burnished on the interior and have a dull gray washy slip on the exterior. Both bowl and jug bases (Fig. 7:B80) have a slight and irregular concavity on the bottom, while some jar and jug bases (Fig. 7:B81, B82) have a distinctive shallow groove or smoothing mark at the base, creating a beveled edge. Shallow open bowls (Fig. 7:B83, B84; Pl. 14:B36, B39) were rare on EM II sites. The consistency in clay fabric, the wide variety of forms, and the widespread regional distribution of EMMC suggest a highly standardized and perhaps centralized production of coarse-ware vessels at some locale in the area between Gournia and Kalo Chorio–Istron on the Mirabello coast. This standardization could be evidence of controlled or administered production of pottery and a highly organized distribution system in complex exchange

networks along the Mirabello coast and across the Isthmus of Ierapetra. This EM II ceramic system is evidence of an integrated sociopolitical structure that operated on a regional scale and perhaps was based at a site such as Priniatikos Pyrgos, Gournia, Alatzomouri, or perhaps even Vasiliki.

Early Minoan Cooking Ware (EMC) Early Minoan I–II cooking ware is orange (5YR 5/8–6/8), red (2.5YR 4/8), reddish brown (5YR 4/6), or brown (7.5YR 5/4) in color with a dark grayish-brown core (2.5Y 4/2–3/2). The fabric is extremely coarse textured with densely distributed sandy inclusions that have a sub-angular to subrounded appearance. They are uniformly white, gray, yellow, and black particles, macroscopically identifiable as granodiorite, quartz, and calcite. This fabric is equivalent to Type II in the Kavousi region (Haggis and Mook 1993, 277) and reddish-brown ware from EM I Kalo Chorio (Haggis 1996a, 668). The most common shapes are large dishes or “baking plates” (Warren 1972). The most distinctive variety is a wide shallow bowl with a thick rounded rim and, below the rim on the exterior, an upward (Fig. 7:B85–B89; Pl. 15:B41, B48, B52) or downward (Fig. 7:B90, B91; Pl. 15:B43) projecting ridge. The bowl interior is smoothed or lightly burnished over the edge of the rim on the exterior down to the ridge. Below the ridge, the surface is very rough or haphazardly smoothed. Exact parallels for this common EM II shape are found in deposits dated to Period II at Myrtos (Warren 1972, 161–162). Another but less common form has parallels at Kalo Chorio (Haggis 1996a, 667, fig. 27:KT64), Myrtos (Warren 1972, 207, fig. 91:P107) and Knossos (Wilson 1985, 338–339), and it has a straight or upturned rim with an exterior ridge that forms a carinated profile (Fig. 8:B92, B93; Pl. 15:B47). Other dish types have a simple rounded and thickened rim (Fig. 8:B94–B102; Pl. 15:B45, B49, B50) comparable to standard shapes at Myrtos (Warren 1972, 163, fig. 47:P116) and Knossos (Wilson 1985, 337–339). Examples with sharply inturning rims (Fig. 8:B101, B103, B104; Pl. 15:B50) are clearly fragments of the simple thickened type that have been thinned and folded inward to form an internal handle (cf. Warren 1972, 111, no. P108).

NEOLITHIC AND BRONZE AGE POTTERY

Deep bowls (Figs. 8:B105, B106, 9:B107) and open jars belong to cooking utensils (cf. Wilson and Day 1999, 20, fig. 14:P212–P214). Internally thickened and collared jars (Fig. 9:B108) are found alongside open hole-mouth types (Fig. 9:B109– B115). Both varieties have good parallels at Myrtos (Warren 1972, 177, fig. 61:P333). Another jar type has a thickened interior rim (Fig. 9:B116, B117), a very distinctive profile with few East Cretan (Warren 1972, 176, fig. 60:P327) or Knossian (Wilson and Day 1999, 30, fig. 14:P220) parallels. Some of the “jar” or “bowl” rims surely belong to tripod cooking pots, the legs of which are commonly found at EM II sites. They have carefully tapered sides and regularly elliptical sections. Some have a low vertical rib (Fig. 9:B118–B120; Pl. 15:B54, B55) and others a folded-out bottom edge (Fig. 9:B121, B122; Pl. 15:B53). Both ribbed and outturned varieties are found in contemporary deposits at Phournou Koriphi, as are types with thick oval sections, such as Figure 9:B123 (Pl. 15:B56; cf. Warren 1972, 179, fig. 63:6).

Early Minoan Pithoi (Pithos Ware)

143

The fabric is essentially a very coarse version of the dark-on-light ware. Both core and slip are usually light red (2.5YR 6/6–5/6), with variations appearing orange, pink, and salmon colored. Sub-rounded and sub-angular phyllite, quartz, and calcite inclusions are both large (1–2 mm) and obtrusive in the core and protruding from the surface. In some examples, the inclusions are rounded sand particles, purple and gray mudstones being the most common. The fabric and surface treatment of EM pithoi are best known from EM I examples at Kalo Chorio (Haggis 1996a, 670). Collared pithoi (Fig. 10:B124; Pl. 13:B33) with rope bands are known at EM II Myrtos (Warren 1972, 195, fig. 79:P599, P606), while the distinctive modeling of the surface with parallel horizontal ribs (Fig. 10:B125–B128; Pl. 14:B34, B35) is the custom in EM I at Kalo Chorio (Haggis 1996a, 670) and Knossos (Wilson and Day 2000, 50, nos. P270, P271). In addition to the ribbing, scoring (Fig. 10:B129) and finger-impressed bands (Fig. 10:B130) are hints of the variety of modeled decorative designs on EM pithoi. Deep scoring is a characteristic of large open jars of EM I date at Knossos (Wilson and Day 2000, 50, nos. P265–P267).

Early Minoan I–II pithoi are distinguishable from later Minoan types in both form and fabric.

Early Minoan III–Middle Minoan IA (Late Prepalatial) It was a challenging task to identify EM III and MM IA phases of occupation in the survey zone, as neither period is adequately represented in wellpublished contexts in East Crete. Problems abound in defining this material in excavation contexts, and in disaggregating EM III–MM IA phases on sites with earlier (EM II) and later (Protopalatial) occupation (cf. Andreou 1978, 164–172; Cadogan 1986, 158; Momigliano 1991; Manning 1994, 231; Watrous 1994, 718–720). The term used here, “Late Prepalatial,” avoids the problems and any polemic, but it is also very useful in its inclusion of material that we think is later than EM IIB and yet earlier than MM IB–IIA. Our interest was in discerning regional patterns that might help us to interpret the nature of culture change in the transition from Prepalatial to Protopalatial.

Late Prepalatial Medium Coarse (LPMC) This fabric is a local Mirabello product ranging from fine—with few or no visible inclusions—to medium coarse, with visible, closely spaced, but microscopic granodiorite and biotite particles, and widely spaced quartz and phyllite pieces. The coarser examples are equivalent to fabric Types XX and XXI in the Kavousi region (Haggis and Mook 1993). In both fine and coarse examples, the core is evenly fired tan (10YR 6/4, light yellowish brown), pinkish buff (10YR 5/6–4/6), or reddish yellow (5YR 6/6). Rarely is it reduced to a dark grayish brown (10YR 4/2) or dark reddish brown (5YR 3/3). The most common varieties have a pinkishtan core (5YR 6/6–7.5YR 7/6, reddish yellow) or

144

DONALD HAGGIS

pinkish-brown core (5YR 5/4, reddish brown). Some of the coarser versions resemble Early Minoan Medium Coarse ware in texture, which is usually very gritty, like a fine sand with a diverse array of sizes and colors of the particles. Most abundant are white and black-and-white granodiorites and larger (> 1.0 mm) angular quartz and phyllite inclusions. Consistently present in all examples are tiny gold biotite particles that give the sherd cores and surfaces a glittery look in the sun. There is extreme variation among LPMC fabrics and surface treatments, suggesting a trend toward destandardization of ceramic production at the end of EM IIB, and for our typological purposes, defying strict categorization in functional or formal groups clearly representing fine, semi-fine, medium coarse, and coarse fabrics. Although no two sherds in this group look exactly alike macroscopically, in all examples there are visible inclusions, tending toward a medium-coarse attribution. The broad category itself was called “medium coarse” as a matter of convenience during site recording, given the extreme variation in fabric texture, inclusion density, and the diversity of fabrics for similar shapes. The group has been separated into finer versions and coarser versions that correlate generally to vase types. On the finer end of the highly variable fabric scale are cups, deep bowls, saucers, spouted bowls, small jugs, and jars. On the coarser end are large jars, bridge-spouted jars, pithoi, pithoid jars, large deep bowls, lekanides, and larnakes.

Fine Version of LPMC Among fine versions of LPMC are a variety of monochrome (Fig. 11:B131) and banded (Fig. 11:B132–B134; Pl. 16:B57–B60) conical cups with rim bands on the interiors. The slip is a dull washy dark gray (10YR 5/1–5/2). These cups are handmade, with parallels found at Knossos in the Upper East Well and Kouloures Groups (Andreou 1978, figs. 1, 2), and more recently among the EM III material from excavations on the south front of the palace (Momigliano and Wilson 1996, 50–51, figs. 29, 30). Typical MM IA angular cups (Fig. 11:B135; Pl. 16:B60) are rare in East Crete, but they are known from Hagia Photia Kouphota (cf. Tsipopoulou 1988, 36, fig. 4) and the Kouloures Group at Knossos. Round cups with a rim band, probably one-handled varieties in White-on-Dark

ware (Fig. 11:B136), are diagnostic for EM III and MM IA (cf. Momigliano 1991, 223, fig. 24; Betancourt et al. 1999, 355, fig. 6:10). The Whiteon-Dark ware conical bowl (Fig. 11:B137) has the characteristic dark gray slip (5YR 4/1) and internal rim band (cf. Fig. 11:B136). Such bowls, usually decorated with pendent hatched triangles in a thick yellowish-white paint (10YR 8/4–7/4), may be spouted, plain, or have a frying-pan handle (Betancourt 1984, 2, fig. 1–1). Although plain, dark gray monochrome and White-on-Dark wares are most prevalent, buffburnished ware was also apparent. The open jar rim in Figure 11:B138 has the characteristic dark gray wash on the interior, but the exterior has a thick buff-burnished slip (2.5Y 7/4). The glossy pale yellow and tan of the exterior presents a striking contrast to the rough black interior. Buff burnishing is common on conical (Fig. 11:B139, B140) and round (Fig. 11:B141) cups. Closed shapes, probably jugs and small jars (Figs. 11:B142–B145, 12:B146– B149; Pl. 16:B61–B65), are usually coated with a monochrome dark gray slip on the exterior, but some are burnished yellowish brown, tan, or buff. Other White-on-Dark ware forms include jugs (Fig. 12:B150; Pl. 17:B67) and saucers (Fig. 12:B151; Pl. 17:B70). Spouted jars, or “North Trench” style teapots, were found with thickened (Fig. 12:B152) and squared (Fig. 12:B153; Pl. 18:B81) rims, in the dark gray slip characteristic of White-on-Dark ware (Betancourt and Silverman 1991, fig. 3:347, 348). The bridge-spouted jar rim in Figure 12:B154 has the carination similar to jars in the Upper East Well Group at Knossos and in the North Trench at Gournia (Andreou 1978, figs. 1:14, 15, 7:4), but the slip in this example is a lustrous red color (10R 5/8) with a high polished surface. Collared jars or amphorae (Fig. 12:B155, B156) are generally rare in the LPMC fabric, but the shape is known from both the Kouloures and North Trench Groups (Andreou 1978, figs. 4, 7). Other fine wares include lids (Fig. 12:B157), shallow bowls (Fig. 12:B158, B159), and deep bowls (Fig. 12:B160). The shallow bowl in Figure 12:B159, with an everted and downturned rim, has a black washy slip on the exterior and a burnished interior. The shape is common at Knossos in MM IA (cf. Momigliano 1991, 158, fig. 2).

NEOLITHIC AND BRONZE AGE POTTERY

Coarse Version of LPMC The extraordinary diversity of wares produced in the LPMC fabric is best exemplified in the variety of coarse jars, lekanides, bowls, and pithoi recovered in the survey samples. Jar rims are the most common sherds in the LPMC fabric, representing both bridge-spouted and more rarely, hole-mouth types. The rims are rounded (Fig. 13:B161–B165; Pl. 19:B85, B87, B88, B96, B99), inturning (Fig. 13:B166, B167), internally thickened (Fig. 13:B168–B173; Pl. 19:B84, B86, B95, B101), and thickened and squared (Fig. 13:B174–B176). Internally thickened varieties are the most common (Fig. 14:B177–B179; Pl. 19:B83), while square types are rarely seen until MM IB. The basic forms of the EM III–MM IA bridge-spouted jar are represented in the catalog of East Cretan White-on-Dark ware from Gournia (Betancourt and Silverman 1991, figs. 2–3:327–346). Dark-on-light painted wares are the most common ware group for these jars. They have a buff core and slip (2.5Y 8/4; 7.5YR 8/4) and a reddishbrown rim band, usually best preserved on the interior (Figs. 13:B161, B170, B172, 14:B180, B181; Pl. 19:B84, B91, B95, B99). Some examples show evidence of banding (Fig. 14:B182; Pl. 19:B92), blobs or disks (Fig. 13:B164; Pl. 19:B96), and burnished surfaces (Fig. 14:B183; Pl. 19:B97). Given the state of preservation of these survey finds, it is generally difficult to determine whether the monochrome pieces (Fig. 13:B165, B168, B169, B173, B176; Pl. 19:B85, B86, B90, B101) are not banded or even white-on-dark. Another common surface treatment for jars is burnishing. The brown or buff exterior surface of the vessel might be plain burnished (Fig. 14:B184; Pl. 19:B89), or pattern burnished with horizontal lines (Fig. 13:B162; Pl. 19:B88). Some examples have a gray or black slip (10YR 5/2) with a highly polished surface (Fig. 13:B173; Pl. 19:B101). Actual examples of Whiteon-Dark ware (Fig. 14:B179, B185) are rare among the jar fragments. Pithoi, large jars, and lekanides are similar to their EM IIB predecessors. Pithoi with an everted and thickened rim (Fig. 14:B186–B188; Pl. 18:B72, B78), and collarless varieties (Fig. 14:B189, B190; Pl. 18:B73), have a distinctive rib or ridge below the rim on the exterior, which is characteristic of EM IIB pithoi and amphorae from Myrtos (Warren

145

1972, 159, 197, figs. 43:P74–P76, 81:P620). Collared forms (Fig. 15:B191; Pl. 18:B74)—essentially an EM IIB type—are found alongside collarless types (Fig. 14:B189; Pl. 18:B73). The most common type has a thickened and outturned rim (Figs. 14:B186–B188, 15:B192–B196; Pl. 18:B72, B77–B79), the main pithoid jar shape in Period IIa–b deposits at Myrtos/Pyrgos. Painted decoration includes trickle (Fig. 15:B192; Pl. 18:B77), bands and disks (Fig. 15:B191; Pl. 18:B74), incision (Fig. 15:B197), and monochrome slip (Figs. 14:B190, 15:B198, B199). Lekanides (Fig. 15:B198–B200) were sometimes difficult to distinguish from open or collared pithoi (e.g., Figs. 14:B188, 15:B193, B194, B201, B202; Pls. 18:B79, 20:B109). In form, Figures 15:B198 and B203 (Pl. 18:B80) look very much like EM IIB large deep bowls, but their fabrics suggest a EM III–MM IA date. A vertical handle (Fig. 16:B204) probably belongs to a large painted jar or small pithos. The stripes are bright red and are painted on a thick, creamy buff slip. Small lekanides and large bowls show exactly the same variety of surface treatments and decorative styles as the bridge-spouted jars, perhaps suggesting workshops specializing in specific wares. Hammer rims on deep bowls and internally thickened rims on shallow bowls (Fig. 16:B205–B207; Pls. 20:B115, 21:B119) exhibit links to EM IIB contexts at both Myrtos and Knossos (cf. Warren 1972, 169–170, 172, 174, figs. 53:P217, 54:P252, 56:P278, 58:P291; Wilson and Day 1999, 10–11, figs. 2, 3). Their presence in the Kouloures Group (Andreou 1978, fig. 3:19) indicates the longevity of these forms into EM III. Deep bowls with thickened rims (Fig. 16:B208–B210; Pls. 17:B66, 21:B120) and flaring shallow bowls (Fig. 16:B211–B215; Pl. 21:B116, B117) have EM III parallels at Chrysokamino (Betancourt et al. 1999, 357, fig. 7:13–15), Knossos (Andreou 1978, figs. 1:12, 3:5, 8, 16–18; Momigliano and Wilson 1996, 50–51, figs. 29:177, 30:189–191), and Gournia (Betancourt and Silverman 1991, fig. 1:306, 307). These shallow bowls are the formal predecessors of the wide flaring bowl or saucer of MM IB–IIA date. Lekanides with thick ledge or offset rims (Figs. 16:B216, B217, 17:B218–B223; Pl. 20:B102–B107, B110, B113) are generally more common than those with simple rounded or thickened rims (Fig. 17:B224–B227).

146

DONALD HAGGIS

This Late Prepalatial phase is clearly transitional, with shallow and deep bowls presenting formal links to earlier EM IIB forms; the saucer, round cup, and angular cup are clearly predecessors of later MM IB–IIA forms. The period is marked in the Gournia area, however, by the extraordinary diversity of forms and variation of wares within the LPMC group. The hole-mouth or bridge-spouted jar is really a new shape in EM III, and its popularity and standardization are observable in the White-onDark ware examples from the North Trench deposit at Gournia. The shallow bowl, deep bowl, pithoi, and lekanides have strong connections to EM IIB forms. In spite of the similarities, and the developmental dependence on the EM II repertoire, the Late Prepalatial phase represents something completely new—countless variations in shapes and fabrics suggesting a real departure from the uniformity and standardization of the EM II tradition. Compare, for example, the EM II sherds in Figures 6:B71–B74 and 17:B228, and Late Prepalatial sherds in Figures 14:B186, B189, B190, 15:B191, B192, 16:B204– B217, 17:B218–B227, B229–B236, 18:B237–B240 and Plates 17:B66, 18:B72, B74, B77, 20:B102, B103, B105–B108, B110–B113, B115, and 21:B116, B117, B119–B121. Deep bowls, shallow bowls, and lekanides were produced in a variety of wares, mirroring the diversity of the jars. Monochrome wares usually have a dark gray or brown washy slip on the interior (Fig. 16:B206), or both the interior and exterior (Figs. 16:B215, 17:B234, 18; Pl. 21:B116), and some examples have a red (10R 5/4–4/4) burnished slip (Fig. 16:B209; Pl. 21:B120). Dark-onlight painted wares have a characteristic band on the top of the rim and on the interior (Figs. 16:B205, B207, B208, B217, 17:B219, B220, B223, B232, B235, 18:B239; Pls. 20:B104, B106, B107, B110, B112, B115, 21:B119), usually painted in a thick dark brown (7.5YR 6/3–6/4) or red (2.5YR 5/6) slip. An unusual piece (Fig. 17:B236; Pl. 20:B108) has the rim band, but below the band on the exterior is a thick white (10YR 8/2, very pale brown) burnished slip. Burnishing is common on both plain and darkon-light wares. Some bowls with a painted rim band have a thick, buff-burnished (10YR 7/3) slip on the interior below the band (Figs. 16:B207, B217, 17:B223, B232; Pl. 20:B107, B110, B111, B115).

Others are buff-burnished on both the interior and exterior (Fig. 16:B205; Pl. 21:B119). Others still are burnished on the interior while the exterior is smoothed or wiped (Fig. 17:B218, B220, B233; Pls. 20:B105, B106, 21:B121). Rare are plain wiped or smoothed surfaces (Figs. 16:B216, 17:B221, B222; Pl. 20:B102, B103, B113). On bowls, the contrast between the dull washy exterior slip or plain surface (Fig. 17:B218, B232; Pl. 20:B105, B111), and the fine buff-burnished interior, presents a striking departure from the simple painted wares of EM IIB. Coarse versions of White-on-Dark wares were recovered, but like the fine-ware examples, they are very rare. Large deep bowls (Fig. 16:B210; Pl. 17:B66) are generally medium coarse in texture, similar to the fabric of White-on-Dark conical bowls (Fig. 11:B137), shallow bowls (Fig. 12:B151; Pl. 17:B70), and jars (Fig. 14:B179). The variations of shape and diversity of wares that characterize EM III–MM IA pottery are echoed in the LPMC fabric itself, with its wide range of subtle variations. The fabric designation actually consists of variations on a theme, rather than a specific type. Cores vary from reddish orange to red, and pinkish-buff to tan. The density, sizes, shapes, and frequency of visible rock inclusions vary as well. Although granodiorite and biotite particles are present in all examples, the volume, distribution, sizes, and density vary inconsistently; quartz and phyllites are equally variable. This lack of standardization in both wares and fabrics is a marked change from the consistent handling of medium coarse, cooking wares, and pithoi of EM I–II. The local Mirabello medium-coarse fabrics, EMMC and LPMC, present the most obvious contrasts—an evident shift from extreme standardization, if not centralization of production in EM II (cf. Day, Wilson, and Kiriatzi 1997, 286–288), to increased variability and diversity in EM III–MM IA. This change might be evidence of decentralization of production and increased competition on a local level, or perhaps even of changes in patterns of distribution and consumption on a regional scale. Do these changes in pottery production reflect broader economic changes in the Late Prepalatial transition? Are they related to new patterns of agricultural production and exchange and a new social or political organization?

NEOLITHIC AND BRONZE AGE POTTERY

Early Minoan (III) Cooking Ware (EMC) Late Prepalatial cooking ware has a very distinctive dark gray (5YR 4/1–3/1) or very dark red (2.5YR 3/2–2.5/2) core that is evenly fired through the wall of the vessel. Most examples look black or very dark red brown, completely reduced from the center of the core to the surface of the sherd. The surface, where preserved, is a brown, gray, or yellowish-red (5YR 5/6) slip that is smoothed or lightly burnished. Sherds in Late Prepalatial EMC fabric are dense and heavy; broken edges are sharp and very hard to the touch. Inclusions are densely spaced, well-sorted, equidimensional white particles—quartz and granodiorite—that are microscopic to about 1.0 mm in size. The texture is very gritty or sandy, and similar to LPMC. The fabric resembles EM II cooking ware, but the even distribution and regular sorting of the inclusions in the Late Prepalatial version suggest both improved technology and systematic preparation of the quartz and granodiorite temper. Large, thick-walled trays, possibly tripod forms, are introduced in the EMC fabric (Fig. 18:B242–B245; Pl. 22:B126–B128). The usual rim type is thickened and drawn slightly into the bowl (Fig. 18:B243, B245; Pl. 22:B127, B128). Shallow cooking dishes or baking plates were also found, but their simple thickened and rounded profiles (Fig. 18:B246–B249; Pl. 22:B125) are much less elaborate than their EM II predecessors, looking more like shallow bowls than dishes (cf. Betancourt et al. 1999, 357, fig. 7:17). Wide shallow bowls are represented in EMC fabric (Fig. 18:B250, B251), but there is no indication that these vessels were ever used for cooking. Similarly, jar rims from Figures 18:B252–B255 may belong to cooking pots, but the fabric type itself is not an indication of function. Pithoid jars (Fig.

147

19:B256–B258; Pl. 22:B123, B124), for example, are identical to LPMC varieties (cf. Fig. 14:B187, B193; Pl. 18:B78, B79), emphasizing the wide variety of forms produced in the EMC ware that overlap with other ware and fabric groups. Everted jar rims from Figures 19:B259 and 19:B260 are similar to EM IIB cooking pot rims from Myrtos (Warren 1972, 204, fig. 88:P681), although tripod legs have extremely thin, flattened oval sections (Fig. 19:B261, B262), a clear change from the thicker EM II varieties. Decorated legs have fingerimpressed vertical bands (Fig. 19:B263; Pl. 22:B131) and single or double plain vertical ribs (Fig. 19:B264; Pl. 22:B130). The presence of shallow bowls, deep bowls, pithoi, and jars in the EMC group is an indication of the unusual diversity of non-cooking forms produced in the traditional Prepalatial cooking-ware fabric. This crossover phenomenon—typically fineor medium-coarse ware shapes manufactured in a cooking ware fabric—is well in keeping with the extraordinary variation and diversity in the Late Prepalatial ceramic production tradition in general. That is to say, even within the Mirabello-area production zone, between Gournia and Kalo Chorio, there appears to be little standardization of fabrics within specific shape or function categories, and very little consistency of fabrics within ware groups. There is even a remarkable diversity of wares. The same shape and fabric might have a burnished surface, pattern burnishing, white-on-dark, dark-onlight, and monochrome styles, or even combinations of these treatments. This variation and apparently dynamic experimentation with fabric and ware groups are possibly results of decentralized ceramic production after EM IIB, involving multiple centers manufacturing a variety of similar, but not identical pots. Such diverse and potentially competing traditions may be related to a new regional social and political structure in the Late Prepalatial period.

Middle Minoan IB–II (Protopalatial) Sherds of Protopalatial date were easily recognizable, although it was not possible to distinguish separate MM IB and MM IIA phases on survey sites. Such phasing is, however, apparent in stratified

excavation assemblages at Petras, Palaikastro, Mochlos, and Myrtos/Pyrgos, which provides useful comparanda for dating the material in the Gournia Survey.

148

DONALD HAGGIS

Fine Wares Fine cups, saucers, bridge-spouted jars, and jugs were the most frequently recognized fine-ware shapes. The clay is a fine pink to pinkish-buff, and the surface treatment is a dark monochrome slip, that is, a washy dark brown (Fig. 19:B265–B268; Pl. 23:B137, B141), red (Fig. 19:B269, B270; Pl. 23:B136), or a metallic black (Fig. 20:B271, B272; Pl. 23:B138, B139). Carinated cups are common indicators of Protopalatial date. Tall-rimmed black monochrome cups in Figures 19:B269 and 20:B273, B274 (Pl. 23:B134, B136, B140) are MM IB–IIA, and the ribbed fragment in Figure 20:B275 (Pl. 23:B135) is no earlier than MM IIA. The short-rimmed example in Figure 20:B276 (Pl. 23:B133) is MM IB. The base fragment in Figure 19:B265, with its straight lower wall and parallel striations in the bottom, is MM IB. The application of the slip over the bottom edge, thus leaving an irregular reserved disk on the underside, is a technique common at Pyrgos in Period II. A variety of flat bases (Fig. 19:B266; Pl. 23:B137) belong to carinated or rounded cups and cannot be dated any more closely than MM IB–IIA, although some bases do have the reserved disk on the underside, suggesting a MM IB date. Tumbler bases (Figs. 19:B268, 20:B277–B279; Pl. 23:B141–B143, B148) are both wheelmade (Fig. 20:B277–B279; Pl. 23:B142, B143, B148) and handmade (Fig. 19:B268; Pl. 23:B141), and plain slipped and monochrome painted. Tumblers are most prevalent in MM IB–IIA, and very small, thin-walled versions continue into MM IIB in eastern Crete. Large conical goblets are either plain (Fig. 20:B280; Pl. 23:B144) or have monochrome red or black slip on the interior (Fig. 20:B281–B284; Pl. 23:B145, B147). The monochrome interior slip is characteristic of Myrtos/Pyrgos, Periods II and III (MM I–II). Examples with parallel striations on the base (Fig. 20:B283), or irregular profiles suggesting handmade production, are taken to be MM IB. Straight-sided or Vaphio cups are not common in East Crete until MM IIA (Andreou 1978, 108), where they appear in Andreou’s Vasiliki House A–Zakros Group and the Malia Town Group. At Petras, in the Siteia Valley, they are altogether rare in MM IB, not becoming common until MM IIB

(Haggis 2007). Straight-sided cups with a flaring wall (Fig. 20:B285, B286) are MM IB–IIA at Knossos (MacGillivray 1998, 69–71), and concave walls (Fig. 20:B287, B288) and beveled bases (Fig. 20:B289) are MM IIB. Deep, round cups (Fig. 20:B290) were also difficult to identify in surface collections. Middle Minoan I–II cup rims are not, however, very diagnostic in the fragmentary form in which they are found in survey. Their fine, pinkish-buff fabric and washy brown slip are often better indications of date than the actual extant form. Rims in Figure 20:B291 and B292 belong to large conical goblets, probably MM IB in date, and Figure 21:B293 is a tumbler rim. Figure 21:B294 could be a straightsided or carinated cup, and Figure 21:B295 is a rare MM IB deep, round cup. Flaring bowls or saucers with sharply everted rims (Fig. 21:B296) are common in Pyrgos Period III (MM II). Unusual finds among the fine wares include a dark-on-light style shallow bowl or lid (Fig. 21:B297) and the rim of a Chamaizi pot (Fig. 21:B298; Pl. 24:B149), probably imported from Malia. Fine bridge-spouted jars have rounded (Fig. 21:B299, B300; Pl. 24:B153), internally thickened (Fig. 21:B301, B302), squared (Fig. 21:B303), and triangular rims (Fig. 21:B304; Pl. 24:B151). The most common MM IB–IIA types are internally thickened and squared. Although carinated jars (Fig. 21:B305) begin at Knossos in MM IB, ribbed and collared varieties (Fig. 21:B299; Pl. 24:B153) are probably not earlier than MM IIA. They are most common in Quartier Mu and the Town Group at Malia, and Pyrgos, Period III (cf. Andreou 1978, fig. 20:1). The jar rims in Figures 21:B300, B301, and B303 can probably be dated to MM IB, based on similarities to Pyrgos Period II and Petras Lakkos types. Bases of jars and jugs have plain, flaring, or offset (Fig. 21:B306; Pl. 24:B155) and beveled (Fig. 21:B307) bases. Most of them are monochrome (Fig. 21:B307, B308, B310; Pl. 24:B152, B154), although dark-on-light styles are also popular in East Crete (cf. Betancourt 1977; 1985, 85–89). The jug base illustrated in Figure 21:B306 (Pl. 24:B155) has sets of radiating lines or floral sprays, and Figure 21:B311 shows a plume jug of MM IB–IIA date (Betancourt and Silverman 1991, fig. 6:381–386).

NEOLITHIC AND BRONZE AGE POTTERY

Protopalatial Medium Coarse (PPMC) Protopalatial Medium Coarse (PPMC) is the most distinctive fabric in the eastern Mirabello area, and the most common find on Protopalatial sites. The fabric ranges in texture from semi-fine to coarse, and the color is consistently pinkish buff (2.5YR 6/6, light red) or brownish yellow (10YR 6/6). The core is usually a diffuse grayish pink or orange (2.5YR 5/2; 5YR 6/6), and the slip is buff. The inclusions stand out in the core and, in many examples, protrude from the surface slip of the vessel. They are large, black-and-white granodiorite, angular or sub-angular, equidimensional in size and shape (0.25–3.5 mm), and evenly distributed throughout the core. Microscopic biotite (gold mica) and white quartz particles are also present, but they are much less frequent or obtrusive than the granodiorite. This PPMC fabric is used for a number of transport, storage, and processing vessels, including jars, amphorae, lekanides, pithoi, and larnakes. The consistent angularity of the inclusions, their regular size, and even distribution in the paste suggest that they were processed and had temper added. While the same basic elements characterize the LPMC fabrics of EM III–MM IA, these Protopalatial wares show none of the variation, experimentation, and irregularities in the handling of the clays and inclusions, and inconsistency in firing and finishing that characterizes the earlier LPMC group. Furthermore, there is a marked decrease in the range of shapes and much less variation within the shape categories themselves. The regularity and consistency in the PPMC fabric reflect a tendency toward standardization, if not centralized control of production of coarse wares, in the Mirabello area by MM IB. Among the jars, buff-burnished collared shapes (Fig. 22:B312, B313) are not common, but they do have good parallels at Myrtos/Pyrgos and Petras in MM IB–IIA. The jar (Fig. 22:B313) with the sharply offset rim and a finger-impressed band on the shoulder is a shape known at Myrtos and Malia, where it is called a “baggy jug.” Conventional bridge-spouted and hole-mouth jars are usually black wash (Fig. 22:B314–B318), plain, or dark-on-light (Fig. 22:B319). Some rim types have an internal ridge for a lid (Fig. 22:B314);

149

others have an external rib (Fig. 22:B317). Internal rim bands are common (Fig. 22:B315–B317), and the rims themselves are internally thickened, like their EM III–MM IA predecessors, but more angular or squared in profile. Most large jars are roughly painted in a washy brown, black, or red slip. Oval-mouth amphorae begin to be manufactured in extraordinary numbers in MM IB (Fig. 22:B320–B326). Most are produced in dark-onlight ware with a host of simple blob, disk, and linear motives. The well-preserved necks of these vessels demonstrate the characteristically irregularly painted rim (Fig. 22:B320, B321, B323, B325) and shoulder (Fig. 22:B322, B323) bands. Lekanides or basins were apparently produced exclusively in the PPMC fabric. The usual types are straight-walled (Fig. 23:B327), slightly flaring (Fig. 23:B328–B332), and only rarely incurving (Fig. 23:B333, B334). The rims may be squared, thickened, and ledged (Fig. 23:B327, B328, B331–B333). Thickened and flaring rounded rims were also found (Fig. 23:B329, B330, B335). Most lekanides are plain, buff-slipped, but some examples have notched bands (Fig. 23:B328) or other incised decoration. Dark-on-light ware is not common, but disks and hatched bands (Fig. 23:B327, B336) do appear. Pithoid and hole-mouth jars are the most prevalent coarse closed shapes in PPMC. The shapes are known at both Petras and Myrtos/Pyrgos in MM I–II. Collared varieties (Fig. 23:B337) were less frequently recovered than inturning and thickened rims (Figs. 23:B338–B342, 24:B343–B346; Pl. 25:B157, B158, B160, B161). Some profiles are very straight, with everted (Fig. 24:B344) or rounded rims (Fig. 24:B346; Pl. 25:B161). Protopalatial pithoi are easy to recognize and distinguish from EM and LM I varieties. They have very thick and elaborately molded torus bases (Fig. 24:B347, B348), and their rims are thickened, rounded, and sharply folded outward, back over the shoulder of the vase, forming a distinctive collar (Fig. 24:B349–B355; Pl. 25:B159, B162–B165). Some examples have an inner ledge (Fig. 24:B352, B355; Pl. 25:B163) for a lid, and one rim is plainly squared (Fig. 24:B353; Pl. 25:B165). Traces of paint were found on all pithos fragments. In most cases, trickle decoration is the preferred treatment (Fig. 25:B356–B359), and interiors can be scored.

150

DONALD HAGGIS

Larnakes have three basic rim types, a range confirmed by excavations in the nearby Protopalatial cemetery in Kalo Chorio (Haggis 1996a, 651–652, 655–656, figs. 11–12:HT22–HT25, HT30). The most easily recognizable type is very straight, with a thickened rim that has a wide shallow or deep groove on the top for securing a lid (Fig. 25:B360–B362). The second type is very straight and plain, with a squared or slightly rounded rim profile. A thickened and squared variety is also known from Kalo Chorio (Haggis 1996a, 655, fig. 11:HT23, HT24), but it was frequently difficult to distinguish between this type and large lekanides in fragmentary form. In larger sherds, the straight or slightly inturning stance of the piece, a thinning of the vessel wall midway between the base and rim, and an irregular diameter proved to be the best indicators of the larnax form. One larnax base (Fig. 26:B363) has a distinctive beveled edge or shallow finger groove where the wall meets the bottom. Most larnax fragments have traces of paint on the exterior, probably originally bands, disks, spirals, and trickle decoration, and a roughly applied or wiped slip on the interior. Lid fragments belong either to pithoi or larnakes (Fig. 26:B364). These lids are fairly thin, less than 2.0 mm thick, and the most common types have a projecting lip (Fig. 26:B365), a beveled edge (Fig. 26:B366, B367), or a simple rounded profile (Fig. 26:B368). Unusual pieces produced in PPMC, indicating the wide variety of uses of the fabric type, include

a zoomorphic rhyton fragment (Fig. 26:B369) and a loomweight (Fig. 26:B370).

Protopalatial Cooking Ware (PPC) This PPC fabric is similar to the Prepalatial cooking wares (EMC), but like the Protopalatial Medium Coarse, it is more consistent in its morphology and shapes than its predecessors. It is uniformly reddish-brown in color (5YR 4/4), and almost all examples have a distinctive and diffuse greenish-gray core (2.5Y 4/2). As in the PPMC, the dominant inclusion is granodiorite, consisting of very visible, angular, and equidimensional blackand-white particles. They are large in size (1–3 mm) and evenly distributed. Microscopic sandy white bits and gold biotite particles are also apparent. The pots in PPC are usually not as well fired or in as good condition as the Late Prepalatial cooking wares (EM III–MM IA EMC). The shapes in PPC include cooking dishes and trays (Fig. 27:B371), as well as tripod legs. The legs are less frequently decorated in MM I–II (Fig. 27:B372) than in Prepalatial, and the leg section is commonly a thin oval (Fig. 27:B373, B374), Figure 27:B375 being the typical type for the Gournia area in MM I–II.

Middle Minoan IIB–III The end of the Protopalatial period and the transition between Protopalatial and Neopalatial is marked in excavation contexts by phases of destruction and rebuilding. In the Gournia survey area, it was frequently difficult, on the one hand, to disaggregate MM IIB–IIIA material on sites with a substantial Protopalatial component, and on the other hand, to distinguish an easily separable MM III phase on Neopalatial sites. This section is meant to provide a description of wares and forms that are no earlier than MM IIB and no later than MM III—that is, before the full florescence of the Neopalatial period.

Fine and Medium-Coarse Wares The straight-sided or Vaphio cup, with a distinctive beveled bottom edge, is a MM IIB feature that continues into MM III (Fig. 27:B376–B378; Pl. 26:B172, B176; cf. Betancourt 1990, fig. 37:789; Sakellarakis and Sapouna-Sakellaraki 1997, 418, figs. 383–385; MacGillivray 1998, 70–71). The cups are most frequently monochrome dark brown to dark gray, and some with dark-on-light banding (Fig. 27:B379) are as late as MM IIIB. Regular horizontal ribbing (Fig. 27:B376; Pl. 26:B172) is a characteristic that was introduced during MM

NEOLITHIC AND BRONZE AGE POTTERY

IIB (Catling, Catling, and Smyth 1979, 26, fig. 18:95–98; MacGillivray 1998, 70–71, Types 11, 12), and the distinctly concave profile (Fig. 27:B377, B378, B380; Pl. 26:B175, B176) is a noticeable development of the MM IB–IIA varieties that have cylindrical or straight walls. Tall flaring conical cups (Figs. 27:B381, B382, 28:B383–B387) are a regular feature in MM IIB contexts, as are short-rimmed carinated cups (Fig. 28:B388, B389; Pl. 26:B167; cf. Betancourt 1990, fig. 53:1326). For these cups, dark gray monochrome is the rule. Bell cups with a rim band (Fig. 28:B390, B391) make their first appearance in MM III (Betancourt 1990, fig. 34:723). The crudely made monochrome carinated cup in Figure 28:B392 (Pl. 26:B169) dates to either MM IIB or IIIA. Semi-globular, hemispherical, or rounded cups with a pulled-out rim are rare in East Crete until MM IIB, not really becoming frequent in assemblages until MM III–LM I. Figures 28:B393 and B394 (Pl. 26:B166) are MM III. The cups are small in size, fairly thick-walled, and slightly rounded rather than distinctly globular, with a gently outturned, rather than offset, rim. The cup in Figure 28:B394 (Pl. 26:B166) has a white-ondark spiral, which is a decorative feature in MM III (cf. Catling, Catling, and Smyth 1979, 26, fig. 18:91, 92; Betancourt 1990, fig. 60:1710, 1721; Sakellarakis and Sapouna-Sakellaraki 1997, 417, fig. 383). The white-on-dark finish is a typically East Cretan conservative element, a carryover from the Protopalatial tradition. Saucers (Fig. 28:B395–B397; Pl. 26:B170, B177) with a distinctive ledge rim begin to appear in MM IIB, and types with a rib located just below the rim on the exterior (Fig. 28:B396; Pl. 26:B170), although present at Myrtos/Pyrgos during Period III and Quartier Mu, are more common in MM III. Tortoise-shell ripple (Figs. 28:B398, B399, 29:B400; Pl. 26:B173, B174) and “feather-wave” (Fig. 29:B401) are new decorative features in MM III, typically applied to ledge-rim bowls, saucers, and small closed shapes (cf. Catling, Catling, and Smyth 1979, 26, fig. 18:100; Betancourt 1990, fig. 63:1829; Sakellarakis and Sapouna-Sakellaraki 1997, 417, 419, figs. 383, 386). Bridge-spouted jars (Fig. 29:B402–B405) and oval-mouth amphorae (Fig. 29:B406, B407) are

151

difficult to distinguish from their MM I–II predecessors, except by fabric. Ribbed and carinated jar rims (Fig. 29:B402, B403) are present in MM IIB contexts at Myrtos and Malia (cf. Stürmer 1993, 129, 159, figs. 3:14, 12:128), but these examples have a fine tan-buff fabric rather than the customary pinkish-buff medium-coarse fabric of MM I–II. Absent as well are the formerly ubiquitous granodiorite inclusions, characterizing the typical PPMC jar fabric. The amphora in Figure 29:B406 is a coarse phyllite-tempered fabric, and Figure 29:B407 is fine buff. The change in fabrics from MM I–II to MM III is a good indication of a break with the Protopalatial tradition, as well as a change in production areas. While the MM III pottery is linked formally to the Protopalatial tradition, in fabric and technique, the pottery looks forward to the Neopalatial tradition of the eastern north Isthmus, which is dominated by phyllitetempered fabrics for coarse wares, and a soft, fine, porous fabric for fine wares, almost to the exclusion of granodiorite temper.

Coarse Wares Two basic coarse-ware fabrics are introduced in MM III in the Gournia area. One fabric, which we call “Gournia A,” is akin to Protopalatial Cooking Ware (PPC), in its gritty texture and use of granodiorite and quartz inclusions. Gournia A is different because it also contains phyllite inclusions, entirely absent in the earlier PPC. The Gournia A fabric is dark reddish brown to very dark brown in color, with a sandy texture and generally friable core and surface. Visible inclusions are subrounded to sub-angular white quartz and gray, purple, and red phyllites that are small in size and very densely distributed. Granodiorite does appear consistently, but is rare compared with the frequency of the other constituents. The phyllite and quartz particles are 1.0 mm and smaller, thus giving the fabric its characteristic sandy appearance. The visible phyllites are generally sub-rounded, only occasionally larger than 1.0 mm, and look like mudstones or shale. Some calcite bits dissolved when hydrochloric acid was applied, and spalling voids are common on both the sherd surfaces and cores.

152

DONALD HAGGIS

Vessels made of the Gournia A fabric include fruit-stands (Fig. 29:B408), bowls (Fig. 30:B409), cooking pots (Fig. 30:B410), and tripod trays with legs (Fig. 30:B411–B414). The legs are tall, slightly arched, and tapered, and have a distinctly squared section. Similar tripod tray legs are found in MM III–LM IA deposits at Petras. The other common coarse-ware fabric type in MM III is called Neopalatial Phyllite (NPP). As the name indicates, it is a phyllite-tempered ware that continues in use throughout the Neopalatial period. This fabric was identified as the principal coarseware type in MM III–LM I in the neighboring Kavousi region, where it has been called fabric Type I (Haggis and Mook 1993). The fabric probably originated in the eastern part of the isthmus and Mirabello coastal areas, drawing on the phyllitequartzite rocks of the western Siteia Mountains. It is the most prevalent coarse fabric recovered in Neopalatial levels in excavations at Mochlos, Pseira, and Kavousi.

The NPP fabric has an orange-red or reddishbrown color with a diffuse grayish-brown or brown core and a resilient thick surface slip. The phyllite inclusions are very large and prominent (2–5 mm in size); they are foliated and angular in appearance, and their even and regular distribution through the core suggests that they were added as temper. Tiny quartz and carbonate rocks are rare, but present in most samples. The phyllites are gray, purple, red, and occasionally yellowish green in color. Sherds are heavy, hard, and solid to the touch, appearing much better fired than the Gournia A fabric. The shapes in NPP mirror those of Gournia A, including fruit-stands or tripod stands with widesplaying rims (Fig. 30:B415, B416), jars or cooking pots (Fig. 30:B417–B419), tripod cooking pots with thick, oval-section legs (Fig. 30:B420, B421), and tripod trays with squared and tapered sections (Fig. 30:B422, B423). The full range of shapes in NPP is discussed below in the context of the Neopalatial pottery.

Middle Minoan III–Late Minoan I (Neopalatial) Fine Ware Fine ware of Neopalatial date was easily identifiable, and there is remarkably little variation in composition or fabric. A semi-fine to mediumcoarse version of the NPP fabric is commonly used for conical (Fig. 30:B424, B425) and round (Fig. 30:B426) cups. The other fabric for conical cups (Fig. 30:B427) is a medium-coarse buff with microscopic and sparse phyllite inclusions. Most Neopalatial fine wares, however, are very fine to semi-fine buff or pink buff, with traces of a dark brown slip or wash. Sometimes a fine burnished slip is preserved in traces, or well enough to suggest the original lustrous surface of the pot. The sherds have an even buff core with good surfaces; they are light to the touch and evidently well fired. The smallest piece of Neopalatial fine ware is easy to distinguish from the Protopalatial version: the hard, dense, and heavy pinkish-buff Protopalatial sherd, with its metallic monochrome red or black finish, is replaced by LM I with a lighter, more porous clay, and a tan to buff burnished surface that is detectable even in the tiniest fragment.

Cups are essentially three types: conical, straightsided, and round (or hemispherical/semi-globular). Conical cups with an inturning or internally thickened rim (Fig. 30:B424–B426) are most likely earlier in date—MM III or early LM IA—although the thick, straight-walled versions (Figs. 30:B427, B428, 31:429) are typical of LM I (Betancourt 1990, 42; Watrous 1992, 111). Rounded conical cups (Fig. 31:B430) were probably introduced well into LM I (cf. Watrous 1992, 112). Straight-sided cups, now more commonly called “Vaphio” cups (Fig. 31:B431–B435), were slow to take hold in East Crete, but they are a standard type in LM I. Bases of conical cups (Fig. 31:B436–B443) are distinguishable from round (or semi-globular) cups (Fig. 31:B444–B446) by their uniform base diameter and straight walls. Round cup bases usually have an offset or pinched-out edge (cf. Fig. 31:B445) and flaring walls (Fig. 31:B444, B446). Semi-globular cups exhibit a round profile with a slightly everted rim (Fig. 31:B447) or a sharply pulled-out rim (Fig. 31:B448–B451; Pl. 26:B168). Bell cups (Fig. 31:B452) were rarely encountered on sites in the Gournia area (cf. Watrous 1992, 113).

NEOLITHIC AND BRONZE AGE POTTERY

Saucers (Fig. 31:B453, B454) were not very common on Neopalatial sites, possibly indicating a tapering off of the popularity of the form. The ledge rim of Figure 31:B454 is a fine Neopalatial version of the usual MM IIB saucer (cf. Fig. 28:B396, B397; Pl. 26:B170); the latter is on the whole larger in size and coarser in fabric. Offset or torus bases belong to bridge-spouted jars (Fig. 31:B455, B456), the rims of which are either straight (Fig. 31:B457, B458), articulated (Fig. 31:B459, B460), or outturned (Fig. 31:B461; cf. Watrous 1992, 112–114). Decoration on Neopalatial sherds is normally not well preserved, but typical examples include foliate bands (Fig. 31:B462, B463; Pl. 27:B182, B185) and running spirals (Fig. 32:B464–B466; Pl. 27:B183, B184, B186); fragments of large jars have lilies (Fig. 32:B467; Pl. 27:B179) and simple banding (Fig. 32:B468; Pl. 27:B180).

Coarse Ware The phyllite-tempered fabric, called Neopalatial Phyllite, already described above, was introduced into the Gournia area in MM III. By LM I, it had become the standard fabric for the full range of cooking, storage, transport, and processing vessels. This tendency toward standardization of coarse wares is perhaps a continuation of a trend begun in the Protopalatial period, with the sharp division between the reddish-brown cooking wares (PPC) and the pinkish-buff medium-coarse wares (PPMC) for storage and transport containers. That is to say, already in MM I–II there is extreme uniformity and standardization within the PPC and PPMC groups, suggesting administered or centralized production in the coastal zone of Gournia–Kalo Chorio in the eastern Mirabello region. By LM I, there is not only increased uniformity and standardization in the exclusive use of phyllite-quartz tempered pottery, but there has also been a west-to-east shift in the production areas manufacturing the wares that appear on sites in the survey—from the granodioritecentered production in the Kalo Chorio–Istron area in the west to the phyllite-quartzite zone around Mochlos and Kavousi in the east. The question remains as to the economic and political significance of this change in production locales, the evidently increased concentration of manufacturing,

153

and the peculiar dominance of the Neopalatial Phyllite fabric. The coarse-ware shapes, along with the change in fabric, represent a clear break with the Protopalatial tradition, and are, in their own right, good indicators of chronology. Amphora necks are thick, and rims are straight and squared in profile (Fig. 32:B469–B471; cf. Catling, Catling, and Smyth 1979, 38, 49, figs. 26:151–153, 35:232). Fewer amphorae were collected on LM I sites than on Protopalatial or MM III sites. Lekanides, basins, and tubs (Fig. 32:B472–B475) have a squared or ledge rim, and less common types have a ridge just below the rim on the exterior (Fig. 32:B476, B477; cf. Sackett and Popham 1970, 227–228, figs. 17:NP 89, 18:NP 97). The rims of pithoi are very straight, articulated only by a slight outturned, squared (Fig. 32:B478– B480), or rounded lip (Fig. 32:B481, B482; cf. Sackett and Popham 1970, 226, 229, figs. 16:NP 83, 19:NP 121). The other Neopalatial rim types have a distinctive triangular (Fig. 33:B483, B484) or rounded (Fig. 33:B485) section, similar to pithoi from Knossos and Palaikastro (Sackett and Popham 1970, 226, fig. 16:NP 68; Catling, Catling, and Smyth 1979, 44, fig. 30:212–214). These pithoi are evidently new forms in MM III–LM I and differ remarkably from their Protopalatial predecessors (cf. Fig. 24:B349–B355), which have low flaring collars and sharply everted rims that curl back over the shoulder of the vase. In contrast, the LM I pithoi from the Gournia area have a tall, straight neck or high collar, and a square, round, or triangular rim, features probably introduced in MM III. Cooking pots, like the pithoi and lekanides, are produced exclusively in the Neopalatial Phyllite fabric. The walls of cooking pots are generally straight or slightly rounded, and the rims are round (Fig. 33:B486–B490), articulated (Fig. 33:B491– B497), or pulled-out (Fig. 33:B498–B500). The LM I types fit neatly into Betancourt’s (1980) “Type B” category from Kommos, although the Gournia examples seem to be more globular in profile than the Kommos pots. Cooking pot legs have a thick oval section and tapered bottom edges. The most common form of tripod leg in the Neopalatial period is plano-convex in section (Figs. 33:B501, B502, 34:503), with a flat interior surface and a wide arching exterior. This tripod leg

154

DONALD HAGGIS

is a veritable type fossil for LM I in Gournia and the north Isthmus areas. Trays show little variation from earlier forms. Both round (Fig. 34:B504) and square (Fig. 34:B505) rimmed varieties continue essentially unchanged from the Protopalatial

period. Dishes (Fig. 34:B506–B510) are fairly consistent in size and shape, and like the trays, they do not differ substantially from their MM I–II predecessors, except in fabric.

Late Minoan II–IIIB Late Minoan II material was not recognized in the Gournia area survey. This is not surprising, since this short stylistic phase is rarely identified by surveys. Furthermore, only the slightest traces of LM IIIA–IIIB material were recovered from across the Gournia Valley and north Isthmus. This difficulty in finding diagnostic LM II and LM IIIA–IIIB pottery is evidently also the case in nearby surveys in Vrokastro and Kavousi. A fragmentary kylix stem (Fig. 34:B511), two deep bowl fragments (Fig. 34:B512, B513), and ledge rim and pulled-rim bowls (Fig. 34:B514, B515) represent the total diagnostic fine ware of LM IIIA–IIIB date collected during the survey. None of it is sufficiently preserved to suggest anything but a late LM IIIA or IIIB date. Such meager remains are sometimes accompanied by large, heavy, and thick tripod legs with a distinctly round section (Fig. 34:B516–B520). The tendency toward

round-section legs in LM III has been documented at Kommos (Betancourt 1980). By LM IIIC, the type is standard (cf. Cadogan 1967, 259). The LM III coarse fabric is similar to NPP in its use of phyllite inclusions, but by the Postpalatial period, the phyllite-quartz fabric is much less well made. The slip is usually badly preserved, if extant at all, and the fabric itself is very gritty, with large and obtrusive quartz and calcite inclusions. The phyllite inclusions are of various irregular sizes, and the fabric is sometimes friable; spalling and chaff voids are not uncommon. While LM IIIA–IIIB pottery is poorly represented in the survey zone, the LM IIIC period is well known from a number of excavations on the edges of the Gournia survey zone. Late Minoan IIIC pottery was recovered from two sites, and this material is presented in Appendix C, which begins discussion with the Early Iron Age.

Appendix C

Late Minoan IIIC–Orientalizing Pottery Donald Haggis

One site in the Gournia area—site 150, located on the Prophetes Elias hill, above and southeast of the Isthmus watershed at Episkopi—produced a substantial sample of Early Iron Age, Geometric, Orientalizing, Archaic, and Classical pottery. This appendix presents the wares and forms spanning these periods from this single site. Although LM IIIC pottery is properly a development of indigenous and Mycenaean elements already present in LM IIIA–IIIB, the period is included in the present appendix because the onset of the LM IIIC ceramic phase on Crete coincides with a significant historical transition involving changes in settlement patterns, economy, land use, and social structure that inaugurate the Early Iron Age. That is to say, LM IIIC is closely related ceramically to the Bronze Age, but historically to the Early Iron Age– Archaic tradition. Excavations at a number of contemporary sites on the fringes of the Gournia survey zone have provided ample comparative material to date the pre-classical pottery from site 150. Kavousi Vronda and Kastro, Monastiraki Katalimata and Chalasmenos, and Vasiliki Kephala were all occupied during LM IIIC.

Kavousi Kastro, demonstrating continuous occupation into the 7th century B.C. (Coulson et al. 1997), provides the best published Dark Age sequence in East Crete. The implications of drawing on a single site to represent features of a number of ceramic phases are both methodological and historical. The pottery from site 150 is presented here as a continuation of the overview of ceramic phases begun in Appendix B; in many ways, however, the assemblage from a single-site surface collection is more akin to that recovered from excavation than from survey. But while we are lacking the broad regional perspective derived from a large number of sites situated across the survey zone, this presentation does provide an opportunity to consider the special function and importance of the physical location of this site and the implications of the context for the history of the region. Furthermore, the size of site 150, about 10 ha, and the surprising lack of other comparable LM IIIC–Classical sites in the survey zone, should indicate not only the regional importance of the site, but that the process of nucleation was an important part of political consolidation and urbanization in this

156

DONALD HAGGIS

area of eastern Crete, even as early as the Early Iron Age. The site is located at a visible and defensible point midway along the Isthmus corridor, suggesting a physical connection to the lowland route and a confluence of potential effects on the ceramic assemblage. Generally speaking, the Early Iron Age pottery from Prophetes Elias has strong

affinities with the Kavousi and northeast Isthmus. Granodiorite-tempered wares probably were imported from east of Episkopi. The coarse wares from Prophetes Elias are phyllite- and quartztempered, probably locally manufactured, although a number of LM IIIC pithos and lekanis fabrics may be products of the kiln at Vronda.

Late Minoan IIIC Fine Wares The LM IIIC fine wares are very distinctive, and their quality overall is very high. They typically have a hard pink (7.5YR 7/6, reddish yellow), light reddish-brown (5YR 7/6–6/4), or reddish-yellowish (5YR 6/6) core, usually with a fine buff-burnished slip (10YR 8/4–7/4) varying from very pale brown to pale yellow (2.5Y 8/4). The fabric itself is well fired and even-colored throughout the core. Sherds are hard and heavy to the touch, even in examples with eroded surfaces. Only occasionally is the core misfired to a pinkish gray (10YR 7/3, very pale brown). The consistency, similarity, and high quality of LM IIIC wares across the north Isthmus suggest a remarkable degree of regional interaction and exchange, and perhaps administered production of fine tablewares. Deep bowls and kraters are the most common open shapes, kylikes and cups being extremely rare. Deep bowls (Fig. 35:C1–C7; Pl. 28:C5) have offset or torus bases, concave bottoms, and sometimes a ring base, articulated by a low squared foot (Fig. 35:C1–C6; Pl. 28:C5). A late feature in LM IIIB that is prevalent in LM IIIC is the reserved disk at the bottom of the otherwise monochrome interior of the bowl (Fig. 35:C1–C7; Pl. 28:C5). The walls of deep bowls are fairly straight, with slightly everted, pulled-out, or thickened rims (Fig. 35:C8–C11; Pl. 28:C2). Kraters (Fig. 35:C12, 36:C13), cups (Fig. 36:C14), and kylikes (Fig. 36:C15) are very rare, compared with bowls, a disparity observed also in stratified excavation deposits at Kavousi Kastro (Mook and Coulson 1997, 354–357). The beveled base of Figure 36:C16 belongs to a late LM IIIC or Early Protogeometric bowl or krater. Closed shapes, probably jugs or amphorae (Fig. 36:C17–C21; Pl. 28:C8–C10), make up a large

portion of the pottery recovered from site 150. Although most fragments are not diagnostic, some forms have an articulated base with a concave underside (Fig. 36:C22), while the usual type (Fig. 36:C17, C18, C20; Pl. 28:C8, C9) has a slightly depressed bottom with a shallow groove running around the bottom edge (cf. Mook and Coulson 1997, 354). The latter form is typical of LM IIIC at Kavousi Kastro. Simple banding is the usual decoration on bowls (Fig. 35:C6, C11), kylikes (Fig. 36:C15), and jugs/ amphorae (Fig. 36:C17; Pl. 28:C8), in instances where paint is preserved. Interior surfaces of open shapes are monochrome, but reserved disks (Fig. 35:C1, C6, C7; Pl. 28:C5) and reserved rim bands (Fig. 35:C8, C10; Pl. 28:C2) are characteristics of LM IIIC deep bowls (Mook and Coulson 1997, 345); Mook and Coulson have argued that the reserved rim band identifies the onset of LM IIIC. Complex decoration was observed in a few pieces. The main forms of decoration include arches, spirals, and diagonal banding. The painted slip was applied to the buff-burnished surface. The most common slip is a thick red (2.5YR 4/6–5/6) matte slip probably originally very lustrous (Fig. 35:C1). In some examples, the slip is dark red or a dark reddish brown (5YR 3/2–3/1).

Coarse Wares Late Minoan IIIC coarse wares are easily identified because their fabrics are very distinctive (Haggis and Mook 1993, 275–276). Two phyllite-tempered wares were identified as belonging to LM IIIC, based on comparisons with material from Kavousi Vronda and Kastro.

LATE MINOAN IIIC–ORIENTALIZING POTTERY

The most common type is the equivalent of Types X and XI at Kavousi. It has a buff or yellow slip (7.5YR 7/6; 10YR 8/6) with an orange (5YR 6/6, reddish yellow) or pink (10R 5/6) core. Large (2–4 mm) and abundant red and gray phyllites and small (0.3–2.0 mm) quartz and quartzite bits are found densely distributed throughout the core; the inclusions even protrude from the thick buff slip. Basins or lekanides (Fig. 37:C23–C26; Pl. 29:C11, C13, C14), kraters (Fig. 37:C27), jugs (Fig. 37:C28), jars (Fig. 37:C29), and pithoi (Fig. 37:C30, C31; Pl. 29:C15) are found in this fabric. Basins have either a straight wall (Fig. 37:C23, C25; Pl. 29:C11, C13) or flaring profile with an offset square or round rim (Fig. 37:C24, C26; Pl. 29:C14; Mook and Coulson 1997, 355). Pithoi have straight necks and a rounded, hooked rim (Fig. 37:C31). Surface decoration on LM IIIC pithoi consists of hatched wave patterns, incised bands, chevrons, and chevron bands (Mook and Coulson 1997; Nowicki 2000). Another fabric type used primarily for pithoi (Fig. 37:C32) is pink to pinkish orange in color (2.5YR 6/6, light red; 5YR 6/6, reddish yellow), with a thick buff slip and very large, densely packed

157

red and purple phyllite inclusions. The fabric, called Type XIII at Kavousi, is a common pithos ware (Haggis and Mook 1993, 276) at LM IIIC sites in the northern Isthmus. Cooking wares are made of a very coarse phylliteand quartz-tempered fabric, identical to the cooking wares of LM IIIA–IIIB. Tripod legs have a round section, a finger impression at the base (Fig. 37:C33; Pl. 30:C17), and a deep vertical groove down the center of the exterior surface (Fig. 37:C34; Pl. 30:C19; cf. Mook and Coulson 1997, 349–350, 362, figs. 17:39, 37:159, 160). A rare piece is the tripod leg in Figure 37:C35 (Pl. 30:C18), which is made in a granodiorite-tempered fabric from the western side of the Isthmus. The sherd has buff to greenish-gray clay, a pinkish-brown core, and abundant black-andwhite granodiorite inclusions. Although this fabric has been observed at a number of LM IIIC settlements, including the Kavousi and Monastiraki sites, it has been found far west of the Kalo Chorio production zone at the site of Kastello in the Kritsa Valley. It is usually reserved for basins and pithoi (cf. Haggis and Mook 1993, 276 [Type XV]).

Protogeometric Fine Ware Protogeometric pottery on Crete is best known from stratified habitation deposits from excavations on the Royal Road at Knossos (Coldstream 1972), and in East Crete, from the Kastro at Kavousi (Coulson et al. 1997). Protogeometric fine wares from site 150 are either pink-buff ware or tan-buff ware. Pink-buff ware (Figs. 38:C36–C48, 39:C49, C50; Pl. 31:C20–C24, C26, C27), typically used for skyphoi, has a fine, even pink core (7.5YR 7/4–7/6, reddish yellow), a pale brown slip (10YR 8/4), and washy grayishbrown paint (10YR5/1–5/2). Some surface slips (Fig. 38:C37–C39; Pl. 31:C21, C24, C27) are a thick and lustrous red to reddish brown (2.5YR 5/8; 5YR 6/6). Examples of tannish-buff ware (Figs. 38:C42, 39:C51, C52; Pl. 31:C23, C25) have a very pale brown core and slip (10YR 7/4–8/3) and the same washy slip as the pink-buff ware. The wares are similar enough to suggest that these color differences

are because of variations in firing practices. Both wares are very hard and heavy to the touch, and surfaces retain a fine burnish. Sherds break in very sharp, almost concoidal, fractured edges and have a very distinctive feel. Skyphoi are equipped with a low flaring foot (Figs. 38:C36–C38, C42, 39:C51, C52; Pl. 31:C23–C27) or taut conical base (Fig. 38:C39, C40; Pl. 31:C21, C22), the largest being over 9.0 cm in diameter. Wheel ridging and regular horizontal grooves (Figs. 38:C36, C41, 39:C51; Pl. 31:C20, C25, C26) are apparent, and examples are either plain or monochrome dark gray or red. Figure 38:C39 and C46 (Pl. 31:C21) are fragments of small dipped bell skyphoi, which at Knossos are often dated to Subminoan rather than Early Protogeometric (cf. Coldstream 1972, 69–70, fig. 2:A18). An argument in favor of an early date might be the reserved disk in the bottom of the bowl (Fig. 38:C39; Pl. 31:C21), a carry-over from the LM IIIC deep bowls. The irregular application

158

DONALD HAGGIS

of dark slip on the base illustrated in Figure 38:C36 (Pl. 31:C26) could also be an indication of dipping. Skyphos walls are distinctly rounded, compared with the LM IIIC deep bowls, and most have thin walls and delicate S-shaped profiles terminating in an outturned rim (Fig. 38:C43–C46; cf. Coldstream 1972, 70, fig. 3:A18, C35; Coulson et al. 1997, 368, fig. 33:1–3). Small, fairly thick-walled skyphoi or cup rims (Figs. 38:C47, C48, 39:C49, C50) were also recovered (cf. Coldstream 1972, fig 3:A26, C38). The reserved band on the interior of Figure 38:C48 may also link the skyphos form to its LM IIIC predecessors. Kraters are semi-fine, medium coarse, or coarse versions of the pinkish-buff ware. There is considerable variety, including types with outturned (Fig. 39:C53, C54) and ledge rims (Fig. 39:C55–C57; Pl. 32:C28, C29, C31; Coldstream 1972, 75). In examples that have a sufficient portion of the body preserved, there is a characteristic ridge (Fig. 39:C53, C55, C56; Pl. 32:C28, C31) or carination (Fig. 39:C58; Pl. 32:C30) below the rim. The straight-walled variety with a squared ledge rim and ridge is clearly a carry-over from the typical LM IIIC basin or lekanis (cf. Seiradaki 1960, 8; Mook and Coulson 1997, 355). Fragments of hydriai are generally difficult to date, and in some cases difficult to distinguish from jugs and amphorae. The sharply flaring rim of Figure 39:C59 is a fine pinkish-buff example with a monochrome red exterior and a red interior rim band. The shape is common in PG contexts at Knossos (Coldstream 1972, 68) and at Kastro Kavousi (Coulson et al. 1997, 368, fig. 33:6). Decoration on closed PG shapes consists of banding near the base (Fig. 39:C60) and concentric

circles that were drawn with a compass (Fig. 39:C61, C62) on the upper body or shoulder of the vessel. Although these sherds are perhaps too fragmentary to indicate a specific date, based on decoration alone, the pinkish-buff fabric is a good indication of a PG or Sub-PG date.

Coarse Ware The PG coarse wares are very similar to their LM IIIC predecessors, but on the whole, the fabrics are harder and finer. Frequently there is little difference between the fabric of a semi-fine krater and a medium-coarse pithos. The jar and pithos fabrics are simply coarser versions of the pinkishbuff fabric, perhaps a finer and harder version of the LM IIIC basin and pithos wares. The core is an even pink or orange color, the slip is an even buffyellow or tan, and sherds are very hard and heavy to the touch. Purple and red phyllite inclusions are small, sub-rounded, and densely distributed. Pithoi have a distinct low collar and squared rim, similar to a LM IIIC–SM example from Karphi (Seiradaki 1960, 4, fig. 1). Jar bases (Fig. 40:C63, C64) regularly have one or more carefully tooled grooves, directly above the base. Cooking wares are essentially the same fabric as in LM IIIC, a reddish-brown phyllite- and quartztempered ware, but the PG varieties are better fired and have a well-preserved slip. Tripod cooking pots, as in LM IIIC, have a thumb impression at the base of the bowl (Fig. 40:C65; Pl. 32:C35) and a long, deep, vertical groove incised down the length of the leg (Fig. 40:C66; Pl. 32:C34).

Geometric The pottery described in this section spans the broad period that is contemporary with phases of Protogeometic B (PGB) and Early, Middle, and Late Geometric (EG, MG, LG) represented in cemetery and town deposits at Knossos. Stratified occupation levels on the Kastro have produced three definable stratigraphic and ceramic phases covering this period: Sub-Protogeometric, Geometric (G), and Late Geometric, the latter corresponding to LG

at Knossos and Athens. Because of the difficulties in associating East Cretan Geometric pottery with the Knossian and Attic sequences, and because the Sub-PG and Geometric pottery from the Kastro has not yet been fully published, we have chosen the inclusive rubric “Geometric” for this period, commenting wherever necessary or possible on the date of specific sherds.

LATE MINOAN IIIC–ORIENTALIZING POTTERY

Fine Wares Geometric fine wares have a pink core (5YR 7/4; 7.5YR 7/3), occasionally looking very red (2.5YR 6/6) or pinkish buff (7.5YR 6/6–5/6, reddish yellow). The usual slip is dark gray to metallic black (2.5Y 4/1–3/1). Forms with pedestals, such as cups and skyphoi, are very rare on Crete after the Protogeometric period. A ribbed conical foot of a skyphos (Fig. 40:C67; Pl. 33:C50) probably dates to Sub-PG, representing the earliest Geometric phase at site 150. Middle Geometric–LG deep cups usually have flat bases and wide, rounded profiles (Figs. 40:C68–C70, 41:C71, C72; cf. Coldstream 1972, 83, fig. 8; 1992, pl. 54:GD21, GD22). Sharply concave bottoms (Fig. 41:C73) and bases with a depression or groove forming a ridge on the underside (Fig. 41:C74, C75; Pl. 33:C42, C51) probably belong to kotylai or skyphoi (Coldstream 1972, 82, fig. 7). Cups and skyphoi are regularly monochrome dark gray or black on the interior, although the exterior can be plain buff, monochrome, or banded. One LG cup or skyphos (Fig. 41:C76; Pl. 33:C37) has a reserved band on the interior and a row of S-spirals on the exterior (cf. Coulson et al. 1997, 327, fig. 8:2). Late Geometric deep cup rims were the easiest Geometric features to identify (Figs. 41:C77–C85, 42:C86). They have a characteristic collar with an everted rim, forming a sharp transition from the shoulder of the round body. Given the fragmentary state of the survey finds, it is possible that some thickened and straight rims (e.g., Fig. 41:C85) belong to skyphoi. Late Geometric cup bases are usually very thin, and some have slightly concave bottoms (Fig. 42:C87; Pl. 34:C61); offset, beveled, and torus bases belong to skyphoi or small kraters (Fig. 42:C88–C91; Pls. 34:C65, C67, 35:C71). Closed shapes, such as jugs and hydriai (Fig. 42:C92; Pl. 33:C49), have torus bases; the thick collared forms illustrated in Figure 42:C93, C94 also belong to closed vessels. The thick-walled and sharply offset rim (Fig. 42:C95) probably belongs to an early (Sub-PG–G) skyphos, as does the wall fragment (Fig. 42:C96), given their similarities to EG and MG types from Knossos (Coldstream 1972, 96, fig. 15:G101). More unusual is a cup with a straight wall and thickened rim (Fig. 42:C97), and another with a carination high up on the body leading to a hooked

159

rim (Fig. 42:C98). Both types are rare in comparison with the traditional deep, round cups, but both have LG parallels at Knossos (Coldstream 1960, 169, fig. 14:109, 110). Kraters have offset rims (Fig. 42:C99), straight collars (Fig. 43:C100–C102), or ledge rims (Fig. 43:C103; cf. Coldstream 1972, 95, fig. 14). The small krater in Figure 42:C99 anticipates a common Orientalizing type (Coldstream 1973a, 38, fig. 1:H35). A variety of sandy buff (Fig. 43:C104, C105; Pl. 34:C56, C60) and gritty micaceous reddish-brown (Fig. 43:C106–C108; Pl. 33:C38, C39, C41) bowls are difficult to date and are presented here tentatively as LG. The bowls are usually buff, brown (10R 5/6–5/8), or orange (5YR 7/8–6/8) in color, and the fabric is sandy or gritty medium coarse in texture. The sandy buff varieties are very straightwalled, with a simple round rim and a series of regular horizontal grooves on the exterior. The coarse micaceous variety has an articulated rim, and either single (Fig. 43:C107, C108; Pl. 33:C38, C39) or multiple (Fig. 43:C106; Pl. 33:C41) grooves. While this basic bowl form begins in LG (Coldstream 1960, 169, fig. 14), it is present and probably more common in Orientalizing and Archaic. Similar to the bowls are collared jars and cooking pots (Fig. 43:C109; Pl. 33:C45) with a series of regularly tooled grooves on the neck and usually a set of grooves set low on the shoulder of the vessel (Coldstream 1973b, 86, fig. 9:F27, G132; 1992, pl. 55:GD32, GE25). The collar fragment in Figure 43:C109 (Pl. 33:C45) is a very gritty, coarse, orangish-red clay with obvious phyllite and quartz inclusions. The small estimated diameter—only 9.0 cm—suggests that the vessel was a small cooking pot or a large jug. Similar to the bowls discussed above (Fig. 43:C104–C108; Pls. 33:C38, C39, C41, 34:C56, C60), these forms are common in LG but continue throughout the 7th century. Fragments of closed vessels produced the widest array of painted decoration. The fragment with pendent concentric semicircles (Fig. 43:C110; Pl. 33:C44), probably Sub-PG in date, has parallels at Knossos in Late PG and PGB deposits (Coldstream 1992, pl. 52:GB1), as does the diamond/lozenge or checkerboard pattern in Figure 43:C111 (Pl. 33:C46). Rows of neatly drawn concentric circles (Fig. 44:C112; Pl. 34:C59) and the hatched wavy band or circle (Fig. 44:C113; Pl. 34:C58)

160

DONALD HAGGIS

are decorative features most common in LG (Coldstream 1968, pl. 57; Coulson et al. 1997, 379, fig. 38:5, 7). The cross-hatched triangle within a panel (Fig. 44:C114; Pl. 33:C48), and the concentric circles with the hourglass center, both have earlier parallels but are most likely fragments of large LG kraters or jars.

Coarse Ware Basins have either wide, flaring bowls (Fig. 44:C115, C116) or straight sides with a ridge below

the rim (Fig. 44:C117; Pl. 36:C84) in the style of the LM IIIC–PG lekanides and kraters (Coldstream 1972, 68, fig. 9:F32; G137). The fabric of these basins is a very hard, semi-fine to medium-coarse pinkish-buff, with sparse but visible phyllite and quartz inclusions. In the same fabric is the collared pithos rim of Figure 44:C118 (Pl. 36:C83) and the jar base of Figure 44:C119 (Pl. 36:C86). The squared cooking pot leg (Fig. 44:C120; Pl. 36:C85), in a sandy pinkish-brown fabric, indicates the longevity of the tripod shape in Crete (cf. Coldstream 1972, 68, fig. 9:F27). The thick, square section of the leg is a style beginning in Sub-PG or G.

Orientalizing Fine Ware Site 150 produced an array of 7th-century forms. Fabrics are very hard, well-fired, with even buff (7.5YR 6/4) to pink (7.5YR 7/4) cores, often appearing a light tan in color (10YR 6/4, light yellowish brown). Surfaces have a burnished slip, and although painted decoration can vary from red to black, or brown to bluish gray, the most distinctive and common slip color is reddish brown or purple (5YR 5/3; 5YR 6/6–5/6). Fine cups (Fig. 44:C121, C122; Pl. 35:C79) have an articulated torus base, slightly concave on the bottom and sharply offset and beveled (Fig. 44:C121; Pl. 35:C79). Jug bases are similarly concave and beveled (Fig. 44:C123; Pl. 35:C80), rounded (Fig. 45:C124; Pl. 35:C82), or flat with a groove on the bottom (Fig. 45:C125; Pl. 35:C81; cf. Coldstream 1973b, 40, fig. 3:H58, H60; Coldstream and Sackett 1988, 52–53). Cup walls are very thin and gently rounded with a tall outturned rim (Fig. 45:C126–C129; Pl. 35:C68–C70, C74; cf. Coldstream 1973b, 40, fig. 3:H54, H55; Coulson et al. 1997, 386, fig. 41:2). Very fragmentary jug bases (Figs. 44:C123, 45:C124, C125; Pl. 35:C80–C82) are decorated with horizontal bands (cf. Coldstream and Sackett 1988, 52–53), and their shoulders have the characteristic pendent languette and tongue patterns (Fig. 45:C130–C132; Pl. 35:C75–C77). The lekanis retains the squared rim and ridge (Fig. 45:C133) of the earlier LM IIIC–G basins and kraters, but new, more sinuous fine-ware

forms are introduced in the Orientalizing period. Figure 45:C134 has a curving wall and rounded rim (cf. Coldstream and Sackett 1988, 57–58, fig. 9:44), a form that continues into the 6th century (Coldstream 1973b, 53, fig. 8:L34–L36). Hydria or amphora rims are sharply articulated by a triangular-shaped overhanging lip (Fig. 45:C135; cf. Coldstream 1973a, 38, fig. 1:H11, H24–H26), and jug rims may be straight, trefoil, or downturning (Fig. 45:C136).

Coarse Wares Coarse basins (Fig. 45:C137) and trays (Fig. 45:C138) have a pink core, pinkish-buff or tan slip, and very dense angular granodiorite inclusions, suggesting a place of manufacture in the area of Kalo Chorio–ancient Oleros and Istron. Granodioritetempered fabrics are reintroduced into the east Isthmus area in LM IIIC, following a prolonged hiatus, beginning in the Neopalatial period. They are not common, however, in the post–Bronze Age until the Orientalizing period, when production and exchange of a variety of coarse-ware vessel types resumed on a large scale. Most interesting is the small “rosette relief pithos” (Fig. 46:C139; Pl. 37:C87). A large number of sherds of these pithoi were recovered at site 150. The rim is beveled, and its inward sloping and tapered outer edge creates a 4.5-cm-wide register that contains a row of stamped rosettes, a typical

LATE MINOAN IIIC–ORIENTALIZING POTTERY

framing element and rim decoration on 7th-century relief pithoi (cf. Schäfer 1957; Woodard 1982). Although the preponderance of relief pithos fragments from site 150 are made in the local reddishbrown phyllite-tempered fabric, the stamped rosette pithoi contain granodiorite and may have been a source near Episkopi. The vessels would have been transported through the Meseleri Valley and then southeastward down to the Isthmus of Ierapetra. This evidence of specialized and centralized production, and the widespread distribution of a small pithos type, indicates the economic ties between site 150 and the Archaic-period Mirabello communities. The location of the Prophetes Elias site is a controlling point, not only along the north–south Isthmus corridor connecting Hierapytna directly with the north coast at modern Pacheia Ammos, but also the site lies at a crucial juncture between the Isthmus route and the mountain passes into the Meseleri, Kalo Chorio, and Xerapotamos valleys (cf. Hayden, Moody, and Rackham 1992, 319–320). The site is based at the center of the most efficient and widely

161

used pre-modern passage between the south coast and the north coastal regions of Kalo Chorio and Meseleri, a route that would have been extremely important in antiquity, linking the territories first of Oleros, and then of Lato in the west and Hierapytna in the east. The appearance of granodioritetempered pottery at site 106—in the form of these 7th-century relief pithoi—strongly suggests the efficacy of this mountain passage and the extreme economic importance of the Prophetes Elias site. The locally made relief pithoi from site 150 show a diverse array of stamped, incised, and molded decoration. The large vertical handle fragment, Fig. 46:C140, is oval in section, tapering to a narrow equid-like spine on the top and outer edge; it is decorated with an incised foliate band flanked by parallel lines. Common decorative motifs include volute and lotus patterns (Fig. 46:C141; Pl. 37:C95), guilloche bands (Fig. 46:C142; Pl. 37:C93), dart and lozenge chains (Fig. 46:C143; Pl. 37:C99), as well as simple stamped circles and crosshatched bands.

Appendix D

Archaic–Hellenistic Pottery L. Vance Watrous

Archaic In this appendix, clay color names do not always match those referenced by the Munsell numbers. Ceramic fabrics of the Archaic period fall into three main types: 1) Fine, soft buff (5YR 8/4) to light brown (2.5YR 7/6) to pinkish buff (2.5YR 8/3). Clay contains a few specks of sand and mica; some examples have rounded quartz and chert inclusions. Used for fine-ware shapes including cups, kraters, amphorae, and bowls. 2) Hard, orange (2.5YR 5/8) to pale orange (10YR 8/3); large granules to very coarse inclusions of rounded and angular quartz, chert, and limestone. This fabric is similar to Classical Fabric 2 (see below) and is used for pithoi. 3) Hard, coarse, reddish orange (10R 5/8) to light red (2.5YR 6/8); purple phyllite, angular, mostly granule-sized and smaller inclusions; used for medium to large vases, that is, mortars and

coarse bowls. This fabric is similar to Classical Fabric 4. Almost all of the pottery recognized from this period comes from Prophetes Elias (150). Cups are the most commonly recognized fine-ware shape. High-necked cup bases have a low concave (Pl. 38:D1), splayed (Fig. 47:D2), or torus base (Fig. 47:D4; Pl. 38:D3, D4), similar to examples from Knossos (Coldstream 1973a, 56, fig. 10). Early 6thcentury vases are monochrome gray inside and out, though later pieces are glazed a thick matte black. One high-necked cup (Pl. 38:D5), painted a washy brown, illustrates the sharp inset juncture of neck and body (Coldstream 1973a, 56, fig. 10:L55). A Cycladic type (Boardman and Hayes 1966, 68, fig. 33:821) has a high disk foot; a similar example is seen in Figure 47:D4 (Pl. 38:D4). The cup in Plate 38:D1 has a low disk foot painted in grayish-dark glaze. A skyphos base (Fig. 47:D6) dates to the late 7th/early 6th century (Callaghan 1978, fig. 10). One

164

L. VANCE WATROUS

cup has a base similar to the foot of an Attic cup of 575–550 B.C. (Boardman and Hayes 1966, 130, fig. 62:1326, see also 71, fig. 29:2319, a plain bowl, ca. 590–560 B.C.; Ducrey and Picard 1969, 811, fig. 22:380). Fine basins have horizontal roll or strap handles (Coldstream 1973a, 52, fig. 7:L32) and are painted inside in brown chocolate or light brown (some are also glazed on the exterior). Basins often have a ridge below the rim (Fig. 47:D7–D9; Pl. 38:D9). Kraters have a disk foot (Pl. 38:D3, D10–D13, D15) and may be banded (Pl. 38:D14). The most common identified coarse shape is the domestic bowl or mortar. An outwardly thickened rim bears an impressed band of single or double

wheat stocks (Pl. 39:D16–D20) or chevrons (Pl. 39:D18). Some examples (Pl. 39:D18) are plain and smoothed inside, and others (Pl. 39:D16, D17, D20) are painted monochrome inside and out. Examples from Lato are similar (Ducrey and Picard 1969, 810, 811, 813, figs. 21–23). Mortar bases are flat (Pl. 39:D21) or have a low disk foot (Pl. 39:D22). The base may repeat the decorative pattern on the rim (Pl. 39:D23–D25) as if made to be displayed while hanging or leaning on the wall. One Archaic tray (Pl. 39:D26) was recognized. Pithoi have an overhanging rim (Fig. 47:D27) and may be stamped (Fig. 47:D28). A beehive (Pl. 39:D29) has a perforated base.

Classical The Classical fabrics from our area can be categorized as follows: 1) Hard, dark red (10R 4/8) to black; coarse; quartz; white, yellowish, and dark rounded and angular inclusions (sand); granular to coarse sand (1–2 mm); smoothed surface; frequently thin-walled; used for cooking ware. 2) Usually hard, orange (2.5YR 5/8) to pale (10YR 8/3); large granule-size to very coarse inclusions of rounded and angular quartz, chert, and limestone. Used for coarse, heavy/thick shapes, such as pithoi, tiles, and amphorae. The color of the tiles varies from greenish white to orange and light red. 3) Hard, coarse, reddish orange (10R 5/8) to light red (2.5YR 6/8); purple phyllite, angular, mostly granule-size and smaller inclusions: used for medium to large vases. 4) Soft, fine, white (2.5Y 8/2) to pale green (5Y 8/3); porous; angular and rounded quartz and chert coarse sand inclusions. Used for amphorae and bowls. 5) Usually soft, fine buff (10YR 8/3) to pale red (10R 6/8); a few specks of fine sand, calcium carbonate, and mica; some examples have medium sand, rounded quartz, and chert inclusions; some examples have a creamy slip on the exterior. Used for amphorae, cups, kantharoi,

and bowls. Some of these soft, fine, welllevigated examples may have come from the Hierapytna area. 6) Hard, coarse, brownish red (10R 5/4); quartz inclusions, very coarse to medium coarse and angular. Two certain examples: a bowl and a Black Glaze (BG) handle. 7) Hard, coarse, brown (2.5YR 5/3); granule-size to coarse sand inclusions of angular quartz and granodiorite. One coarse bowl. Classical shapes were recognized primarily at Prophetes Elias (150). Roof tiles are made in Fabric 2. Pithoi, made in Fabric 2 (e.g., Fig. 47:D30), are white and light red in color. A second pithos fabric type (Pl. 40:D31) has large angular red phyllite inclusions; it is hard, coarse, and light (2.5YR 6/8) to dark red (10R 5/8). Pithos decoration seems limited to thick bands and plain horizontally ribbed bands. Rims may be of an overhanging type. Basins and coarse bowls/mortars are made of the second pithos fabric above, or Fabric 7 (with some granodiorite and dark mica inclusions). A mortar (Fig. 47:D32), made in Fabric 4, has a monochrome exterior. Two basins (Fig. 47:D9, D33; Pls. 38:D9, 40:D33) have an articulated rim. Two coarse bowls/basins have a light brown fabric with many granodiorite inclusions. Interiors are smoothed (Fig. 47:D34; Pl. 40:D34) or have a metallic BG glaze inside (Pl. 40:D35) or all over (Pl. 40:D36), or are reddish on

ARCHAIC–HELLENISTIC POTTERY

the exterior. Decoration consists of a rim band. Amphorae, in Fabric 2 (Fig. 47:D38; Pl. 40:D37, D38), are reddish or white in color. A jug, in Fabric 5, has a raised torus foot with a beveled profile. Kraters are in two types, made either in a coarse or semi-fine clay. Their interiors are polished, painted matte BG, and have overhanging or rolled rims (Coldstream 1973b, 59, fig. 12:78). One fine krater (Fig. 48:D39) is in fine deep red (Attic?) clay; the vessel in Plate 40:D40 may be Knossian. A column krater has clay that is brown/buff, fine, and hard, with tiny dark and white inclusions. A local krater is made of fine brown clay, with much granodiorite; it has a polished surface, a brown glaze both inside and out. Another krater (Pl. 40:D41) has a fine, light red, micaceous clay, painted dark matte BG inside and out. A krater (Pl. 40:D42), in a fine pinkish-buff clay, is painted a dull BG inside and out, with an exterior floral pattern. The high-necked cup (Pl. 41:D43) is covered with a dull brown to black BG. Skyphos or cup bases (Fig. 48:D44 [dated ca. 450 B.C.]; Pl. 41:D44, D45, and Fig. 48:D46, D47; Pl. 41:D46) have a fine pinkish-buff clay and a glossy BG exterior; Pl. 41:D48 and Figure 48:D49 have a torus base; Plate 41:D50 has an articulated underside. Plate 41:D50 is early Classical (Coldstream 1973b, 24, fig. 14:B17/18). For the Classical cup in Figure 48:D44, compare Boardman and Hayes 1966, 57, fig. 23:2209 (550–525 B.C.) from Rhodes. The foot of a cup/bowl (not illustrated) resembles Sparkes and Talcott 1970, fig. 8:757 (ca. 375 B.C.),

165

a one-handled cup, and fig. 8:828 (375–350 B.C.), a bowl. A cup or kantharos base (Pl. 41:D51) dates to ca. 425 B.C. (cf. Sparkes and Talcott 1970: a kantharos [molded foot], fig. 7; Coldstream 1973a, 56, fig. 10:157 [500–480 B.C.]). Bowls may be in a fine pinkish buff (ca. 400 B.C.) and may have a BG interior. Bowls may have a pulled-out rim (Fig. 48:D52) and a torus or disk foot (Fig. 48:D44). Amphorae have a rolled (Pl. 41:D53), outturned (Fig. 48:D56, D57; Pl. 41:D53–D56), or thickened rim (Fig. 48:D56, D57; Pl. 40:D36); handles can be ribbed or of Coan type. An amphora (Pl. 41:D58) in Hierapytnian clay has a thickened rim. Lids (Pl. 42:D59–D62) are fashioned from coarse bases. Basins (Fig. 47:D9, D34; Pls. 38:D9, 40:D33–D36, 42:D63, D64) have pulled-out and articulated rims. Mortars (Pl. 42:D65) are heavier versions of these basins. Pithoi (Fig. 48:D67; Pls. 40:D31, 42:D66, D67) have simple relief bands and a flat base (Pl. 42:D68). A lampstand (Pl. 43:D91) is painted in a monochrome BG. Beehives, made in Fabric 2, have horizontal striations inside. Spindle whorls (Pl. 42:D69–D72) are biconical. Roof tiles come in different shapes: some are high-flanged (Pls. 42:D73, 43:D74); another type has a curved profile (Pl. 43:D75–D77); others have an irregular, thickened edge (Pl. 43:D78); and others an internal flange (Pl. 43:D79). Attic imports include the kantharoi/cup (Pls. 41:D51, 43:D79) and a skyphos or cup base of ca. 450 B.C. (Pl. 41:D45).

Hellenistic From the small local sample available within our survey zone, it appears that Hellenistic pottery fabrics changed little from the Classical period. Hierapytna produced amphorae of high quality, made in a fine version of Classical Fabric 5. This clay is usually hard, fine, slightly silty, and buff colored (2.5YR 8/4 to 2.5YR 7/6), with a few small brown inclusions. John Hayes dated two amphora handles (Pl. 43:D80) to the 3rd–2nd century B.C. and to the 4th–3rd century B.C., respectively. Amphora toes (Fig. 49:D82; Pl. 43:D81) are of the same fabric. Only two characteristic examples of Hellenistic amphorae come from the Isthmus of

Ierapetra. The first one has a conical toe that terminates with a flat bottom (Fig. 49:D82, from site 23). The outer surface is wet smoothed. Its fabric is hard fired, of light brown color (7.5YR 8/4–7/4), with a few white and black grits. Local production of Hellenistic amphorae has been identified in Gortyn, Keratokambos, and Hierapytna (Rendini 1988a, 263; Empereur, Kritzas, and Marangou 1991, 503–506, 519, 522; Empereur and Marangou 1992, 639–642, fig. 9). The amphora illustrated in Plate 43:D83, possibly an import, is cream-slipped. A Hierapytnian amphora (from site 86) of this fabric imitates Rhodian

166

L. VANCE WATROUS

stamped amphorae, but uses a blank stamp (Fig. 49:D84). Cooking-ware fabric is soft, orange, with many dark and rounded quartz inclusions, and bits of mica. Hellenistic dishes or bowls were often identified. Dish rims (Fig. 49:D86) have Hellenistic parallels (Coldstream 1973b, 30, fig. 15:F13, F14; cf. Pl. 43:D85). A small dish (Fig. 49:D87) with a relatively thin ribbed wall comes from site 73. Dishes of corresponding shape with foot-rings are known from Cyprus and are dated to the mid 2nd to early 1st century B.C. or later (Hayes 1991, 27, fig. XV:5). Two bowls, from sites 77 and 114, with plain, vertical to slightly incurved rims (e.g., Fig. 49:D88), are dated to the 3rd–2nd century B.C. On their outer and inner surfaces they preserve traces of black slip. Our examples belong to the earlier series, with a low foot, thin wall, and shallow body that is curved to the rim (Coldstream 1973b, 30, fig.15:H66, F14; Hayes 1991, 26, fig. XIV:9, 4, 5). The cup in Plate 43:D90, with its concave base, resembles a Knossian example of Augustan date (Sackett 1992, pl. 131:58). A bowl base (Pl. 43:D89) has Hellenistic

parallels (Coldstream 1973b, 30, fig. 15:F15). The flaring rim of another bowl (Pl. 43:D91) resembles Coldstream 1973b, 30, fig. 15:K22. A basin with a flaring rim comes from site 9 (Fig. 41:D92). Two thin ridges are formed on the outside, at the junction with the body. On the inner surface there are traces of brown to black slip. An unguentarium (Pl. 43:D93), with traces of BG inside, is a type that is widespread in the Aegean from the 2nd century B.C. (Hayes 1991, 68–72, pls. 14–16). Our example is more probably a transitional type from Hellenistic to Roman times (Robinson 1959, nos. F48, F49; Hayes 1971, 257, fig. 8:5). Two BG kantharoi (Fig. 48:D51; Pls. 41:D51, 43:D79) have parallels at Knossos dated to the 3rd–2nd century B.C. (Coldstream 1973b, 34, fig. 16:H88, H89; Callaghan 1992, 119, pl. 96:H28, no. 7). A pithos has a sharply projecting/overhanging rim with a thin, incised wavy line on its upper flat surface. On the outside, under the rim, there is a cylindrical band in relief with hollow/concave notches (Fig. 50:D94).

Appendix E

Roman Pottery John Hayes and Angeliki Kossyva

Plain Wares Roman plain wares are mostly local products. They include open and closed shapes for table and kitchen use, storage, and transportation. The main fabric is a relatively soft, orange, brown, buff, or cream color. The clay usually has limestone or quartz inclusions and some fine mica. The basin (Pl. 44:E1–E12) is the most common open shape. Like Knossos-type examples, it has a flat base, straight or sloping sides, and a flat or slightly raised rim (Hayes 1983, 108–109, figs. 15–16; Sackett 1992, 176, pls. 175, 179). The rim, with a diameter mainly between 30 and 40 cm, is plain or decorated with one (Fig. 51:E2) or more grooves (Fig. 51:E3, E4). The flat rim of an example of the 6th–7th century bears incised, wavy lines instead of grooves (Fig. 51:E5; cf. Hayes 1983, fig.15:182, 184; Sackett 1992, pl. 188:19). Basins are most common in the sites of the 2nd–3rd century. A feature of late date is a rising rim (Fig. 52:E6; Sackett 1992, 176). A basin with notches on the edge of the rim was found in site 89 (Fig. 52:E7). Analogous vessels at Knossos are considered not for

everyday use but as possible adaptations of a metalware type from the western part of the Roman Empire (Hayes 1983, 111, fig. 16:192–196; Sackett 1992, 176, pl. 195:82, 85). This type of decoration is also known at Gortyn in the 5th century (Rendini 1988b, 240–241, fig. 202:223). Two coarse basins represent the transition from Late Roman to early Byzantine times. One has a triangular rim and ridges on the upper part of the wall (Fig. 52:E8). The other has a relief band with incised drops under a sharp raised rim (Fig. 52:E9). A mortarium (Pl. 45:E17) was recovered; it is a thick basin with hard-fired fabric and a rough surface. This mortarium probably comes from Syria. Analogous examples are widespread in the eastern Mediterranean; they are dated from the 3rd–4th century onward (Hayes 1967, pl. 73; Riley 1979, 292, 299; Blakely, Brinkmann, and Vitaliano 1992). Bowls are hemispherical (Fig. 53:E10) or carinated, with a raised or flat rim, respectively (Sackett 1992, 176). Site 91 revealed a possible example of a conical lid with ridges (Fig. 53:E11; cf. Hayes 1983,

168

JOHN HAYES AND ANGELIKI KOSSYVA

fig. 12:154; Sackett 1992, 175, pl. 149:109). A reelshaped stand, possibly of an amphora, from site 93 (Fig. 53:E12), preserves traces of greenish-brown slip (Pl. 46:E29). This type of stand is widely known in Crete (Markoulaki, Empereur, and Marangou 1989, 561, fig. 8; Empereur, Kritzas, and Marangou 1991, 493, 495, 507, 510, 516; Empereur and Marangou 1992, 634, 638, 644; Sackett 1992, 217, pl. 162:45). At site 91, a possible fragment of a pithoid jar, with a plain vertical rim and a straight, grooved and ridged neck (Fig. 53:E13), was found. The hardfired fabric is yellowish brown (7.5YR 8/4–7/4), with a little lime and fine mica. The jar is probably a Cretan product of the 6th–7th century (Sackett 1992, 176, pl. 195:102–104). A bucket found has a raised rim and vertical wall. Its outer surface and rim are covered with greenish-brown (2.5Y 6/3–5/3) slip (Sackett 1992, 176, pl. 184:12). Beehives (Pl. 47:E30–E34) feature horizontal, thin grooves on the inner surface (Fig. 53:E14). Less coarse examples may be slipped brown. Rims are horizontal (Fig. 53:E15) or raised. Under the rim of the second type is a sharp ridge and relief band with pie-crust decoration (Fig. 53:E16; Hayes 1983, 110, fig. 14:177; Sackett 1992, 176–177; see also the discussion about beehives by Jones, Graham, and Sackett [1973, 397–452, pls. 75–86]). Jugs are relatively thin-walled with an overhanging, collar-like rim (Fig. 54:E17) and a base with a central button (Fig. 54:E18; cf. Hayes 1983, 109, fig. 11). Some have fabric that is analogous with the Late Hellenistic–Early Roman jugs from the Ierapetra area. A trefoil-rim jug with a ribbed neck (Fig. 54:E19), slipped brown, is possibly a local product, dated to the 4th–5th century. A twisted handle (Fig. 54:E20) from site 93 perhaps belongs to a jug or lagynos (Sackett 1992, pl. 125:10). Amphorae are the most common Roman vase shape found in the Isthmus area. They include Cretan (Pl. 46:E21–E29) and imported types. Cretan types of the 1st–3rd century are identified as Types 1 and 2 (Fig. 54:E21; in Hayes’s classification: Hayes 1983, 141, 143, figs. 20, 21) or AC3 and AC1, respectively (in Empereur’s classification: Markoulaki, Empereur, and Marangou 1989, 554– 566, 570–574; Empereur, Kritzas, and Marangou 1991, 520–522; Empereur and Marangou 1992, 645–648, fig. 9). Double-handled examples (Fig.

54:E22) of Type 39 (in the Peacock-Williams classification: Peacock and Williams 1986, 173–174) or AC2 (in Empereur’s classification: Markoulaki, Empereur, and Marangou 1989, 566–569; Empereur, Kritzas, and Marangou 1991, 489–490, 493, 499–500, 510, 521; Empereur and Marangou 1992, 638, fig. 5:d–e) have been recognized. The most common type of local amphora seems to have been Type AC1 (Type 2 in Knossos), with cylindrical neck, high convex collar-like rim, and short splayed handles, round in section. They date to the 2nd–3rd century (Fig. 54:E23), but also continue into the 4th–5th century (Fig. 54:E24; Hayes 1983, 140; Markoulaki, Empereur, and Marangou 1989, 578–579). This type is widespread in Crete. It was exported to Italy (Hayes 1983, 140, 143; Markoulaki, Empereur, and Marangou 1989, 578–579; Empereur, Kritzas, and Marangou 1991, 489–490, 493, 499–500, 510, 521) and imitated in mainland Greece during the 2nd century (examples have been found in Isthmia and the Athenian Agora; see Robinson 1959, 43, pl. 23; Gebhard 1973, 103, 132, fig. 53). Amphorae were imported from the volcanic region of South Italy during the Early Roman period. Type 10 (in the Peacock-Williams classification: Peacock and Williams 1986, 105–106), with the characteristic carination on the shoulder (Fig. 54:E25) and the double handles (Fig. 55:E26, E27), has been identified. Analogous amphorae have also been found at Knossos (Hayes 1983, 149, Type 20) and Gortyn (Rendini 1990, 236). Handles with a central groove, oval in section (Fig. 55:E28), possibly a north Italian import, date to the same period. Similar examples have been found at Knossos (Hayes 1983, 145, Type 7, fig. 21:33, 35) and Gortyn (Rendini 1990, 236). The rest of the imported amphorae date to the 3rd century and later. They seem to come from other centers in the eastern Mediterranean. Amphora fragments from North Africa (Fig. 55:E29) are characterized by a sandy red fabric, fired gray at surfaces (Hayes 1983, 151, 153, fig. 24:72–77; Peacock and Williams 1986, 153–170). Some have white surfaces because the clay was mixed with salted water (Peacock and Williams 1986, 45). African amphorae are also known from Knossos (Hayes 1983, 151, 153, fig. 24:72–77;

ROMAN POTTERY

Sackett 1992, 178, pl. 199:150) and Gortyn (Rendini 1990, 236–238). Amphora Type 43 (in the Peacock-Williams classification: Peacock and Williams 1986, 182–184) is well represented in the Ierapetra Isthmus. It has a globular body, with horizontal grooves on the upper part (Fig. 55:E30), a short conical neck (Fig. 55:E31), a tall everted rim, and bowed handles (Fig. 56:E32). The fabric includes lime and muscovite mica. The origin of the type is not exactly known; it is generally localized in the regions of the North Aegean and Black Sea. The type is dated to the 4th–6th/7th centuries. Straight grooves (Fig. 56:E33) are considered features of an early date; undulating ones (Fig. 56:E34) are later. A fragment of that type of amphora in Cretan fabric was found at site 91. Because of the undulating grooves, it is dated in the 6th–7th century. Local imitations of such amphorae are also known from Gortyn in the same period (Rendini 1990, 239, pl. 42a). Fragments of amphora Type 44 (in the PeacockWilliams classification: Peacock and Williams 1986, 185–187) were found at sites 91 and 104. They are covered with ridges from the base to the shoulder, which become widely spaced in the middle of the body. They have stumpy handles (Fig. 56:E35) and a thickened rim slightly concave on its outer face. The type is considered to originate in

169

Cyprus and is dated in the 5th to mid 6th century. Analogous examples are known from Gortyn (Rendini 1990, 239, pl. 42b). Sherds from amphora Type 45 (in the Peacock-Williams classification: Peacock and Williams 1986, 188–190) reveal a long, slender neck, with one or two short strap handles, a tall rounded shoulder, and a tapering, hollow, or solid foot. Most of the vessel is covered with ribbing. The fabric includes large amounts of mica. The area of western Asia Minor is considered as the probable source of the type. The version with one handle is dated from the late 1st century; the version with two handles appears in the late 4th century. Both continued until the late 6th century. These types have been also found at Knossos (Hayes 1983, 158, Type 49, fig. 26:105–107) and Gortyn (Rendini 1990, 239). Fragments of amphora Type 47 (in the PeacockWilliams classification: Peacock and Williams 1986, 193–195) from sites 47 and 57 are dated in the 3rd–4th century. They have a tapering body, terminating in a tubular, hollowed base, a tall conical neck (Fig. 57:E36), and a narrow rim with a prominent flange. Handles are thick, broad, and arched above rim level (Fig. 57:E37). Analogous examples have been identified in Knossos (Hayes 1983, 155, Type 37, fig. 25:86).

Cooking Wares Cretan products (Pls. 48:E35–E47, 49:E48–E60) have relatively thick walls and a very sandy fabric of reddish-brown color (Hayes 1983, 106; Sackett 1992, 172). Imports (Pl. 50:E61–E67) probably come from the eastern Aegean where the trade of cooking wares had been developed from earlier periods. Kilns of such wares have been excavated in Pergamon and Phocea (Ízyigit 1990, 1991). These wares are characterized by thin walls and a hard fabric of pinkish-red or brownish-red color with a fired gray surface (Hayes 1983, 106; Sackett 1992, 169–170, 172). Globular cooking pots are common. The earlier examples, at the transition from Late Hellenistic to Early Roman times, are Cretan products with various rim types: plain vertical (Fig. 57:E38), stubby

upright (Fig. 57:E39; cf. Sackett 1992, fig. 5:3:a), rim with hooked lip (Fig. 57:E40, from site 145), or flat with strap handle (Fig. 57:E41). Cretan and imported cooking pots of types known from Knossos are dated in the 2nd–3rd century. Some have a short, downturned rim, triangular in section (Fig. 57:E42); others are broad, flat, and sloping (Fig. 57:E43), or outcurved, thickened, and slightly concave inside (Fig. 58:E44; they correspond with Types 1, 2, and 4, respectively, at Knossos: Hayes 1983, 105–106, figs. 5–7). The type with the broad, flat, sloping rim seems be the most common one in the Ierapetra Isthmus. One example of a cooking pot has a rim (Fig. 58:E45) with a pronounced flange on the outside (Sackett 1992, 172, fig. 5:2:a). Imports from the eastern Aegean continue

170

JOHN HAYES AND ANGELIKI KOSSYVA

in the Late Roman period, such as the cooking pot with small ledge on top of its rim (Fig. 58:E46; cf. Hayes 1983, 126, fig. 8:94, 98). Contemporary Cretan vessels have a thickened, sharply outcurved rim (Fig. 58:E47–E49; such types are known at Knossos from the 4th century onward: Hayes 1983, fig. 8:92, 93) or a flat rim thickened on the outside (Fig. 58:E50). A local cooking pot with a sharply outcurved rim and a handle arched above the rim (Fig. 58:E51) represents the transition from Late Roman to Byzantine times; it dates to the 7th century or later. Casseroles correspond with those at Knossos of the 2nd–3rd century; they are carinated (Fig. 59:E52) with a downturned rim, triangular in section (Fig. 59:E53; cf. Type 1 at Knossos: Hayes 1983, 106, fig. 7:80), or more often with a broad, flat, sloping rim (Fig. 59:E54; cf. Type 2 at Knossos: Hayes 1983, 106, fig. 7:81–89). A Late Roman casserole has a sloping rim with a ledge on the edge and a groove inside (Fig. 59:E55). It imitates Late Roman cooking pots from Asia Minor, known also from Knossos and Gortyn (Rendini 1988c, 256, fig. 211:260). Our example differs in that it lacks the step under the groove of

the rim. The fabric includes quartz and particles of mica, indicating that it is an import from a nonlocal Cretan production center. Two thickened, rounded rims (Fig. 59:E56) represent the fryingpan type of the 2nd–3rd century, known from Knossos (Hayes 1983, 107, fig. 9:99–102; Sackett 1992, 173). Their micaceous fabric suggests that they are imports. A cooking dish (Fig. 59:E57) with a sloping wall and flat rim is probably an import. This type is also known at Knossos from the 2nd century onward (Hayes 1983, 107, fig. 9:103; Sackett 1992, 173). Mugs with raised flat bases (Fig. 59:E58) date to the 2nd–3rd century. Their fabric suggests they are probably imports. This type of mug is very well known in the Aegean (cf. Type 1 in Knossos: Hayes 1983, 107, fig. 6:72). Mediumsized trefoil-mouthed jugs (Fig. 59:E59) with ribbed bodies are mainly Cretan products, dated in the 2nd–3rd century or later, as at Knossos (Hayes 1983, 105, fig. 6:76, 77; Sackett 1992, 173–174). Two squared (Fig. 59:E60) and a knobbed rim (Fig. 59:E61) probably belong to basins and are dated in the 3rd–4th century or later.

Lamps Two small fragments of Roman lamps were identified (Pl. 51:E71, E72). A fragment of the Cretan type of lamp was found at site 104. The upper part of the body is usually decorated with ivy leaves and relief buttons in combination with various linear motifs. The shape of these lamps is connected with Late Hellenistic–Early Roman lamps from Ephesus. They are also known from areas in Central and East Crete (from Knossos, Gortyn, and Hagios Nikolaos) and are dated in the

1st century (Mercando 1974, 235, pl. XXXIV; Apostolakou 1987; Catling and Catling 1992, 265–273, pls. 250, 251). A North African lamp is Type XXXI (in Broneer’s classification: Broneer 1930, 118–119, pls. XXI, XXII) and is dated in the late 5th–6th century. Relief decoration with floral motifs, animals, human protomes, and crosses or linear ornamental motifs often embellishes the discus and rim (Brants 1913, 65–67, pl. VIII; Hayes 1972, 310–314, pl. XXI:a).

Wall Pins and Tile Roman-period buildings in the Ierapetra Isthmus have produced terracotta architectonic elements— specifically, wall pins and tiles (Pl. 52:E73–E78). Wall pins (Fig. 60:E62) are common finds on Crete—e.g., at Chania, Lappa, Eleutherna, Knossos,

Gortyn, Tsoutsouros, and Dermatos (Empereur, Kritzas, and Marangou 1991, 495, figs. 18:f, g, 48:a–d). Analogous examples also come from Cyprus (Empereur, Kritzas, and Marangou 1991, 495 n. 24) and Pergamon (De Luca 1984, 55–57, pl.

ROMAN POTTERY

21). They are generally considered anti-earthquake devices on vaulted roofs (Hayes 1983, 103) or supports for the air insulation in the villa hypocausts (Empereur, Kritzas, and Marangou 1991, 497; Empereur and Marangou 1992, 645; Sackett 1992, 256). Dated to the 2nd–4th century (Hayes 1983,

171

102–103, 132; De Luca 1984, 55–57, pl. 21; Sackett 1992, 256), they are local products (Empereur, Kritzas, and Marangou 1991, 495, 510; Empereur and Marangou 1992, 644–645). Local tiles, of the Corinthian type (Empereur and Marangou 1992, 644–645, fig. 11:b–d), are flat.

Fine Wares Examples (16) of Italian Terra Sigillata (Fig. 60:E63) date to the Early Roman period (Pl. 53:E79–E89). Bowls and dishes come from the Arezzo workshop (Fig. 60:E64–E66). Their pinkish fabric and the dark red to brown slip date mainly to the Augustan period (Ettlinger et al. 1990, 4–6, 52–117; Sackett 1992, 153–155). The Eastern Sigillata A ware (six examples) was probably manufactured in a Syrian center. Its fabric is pale yellow to orange (10YR 8/4 to 7.5YR 6/6–8/6) with dark red to reddish-brown slip. Chronologically it dates from the mid 2nd century B.C. to the late 2nd century A.D., with the main production in the 1st century B.C. to the 1st century A.D. (Hayes 1985, 9–13, pls. I–X). Form 4 (in the classification of Hayes: Hayes 1985, 15–16, pl. I:9–12) is popular locally. This dish, with a plain rim curved upward and a low foot, dates to the late 1st century B.C. to the 1st century A.D. Dishes of the same form have been found at Knossos (Sackett 1992, 150–151). Eastern Sigillata B ware found locally comes from Tralles in Asia Minor. Its fabric is orangish red with fine silver mica and red slip. This ware includes two series: the early B1 and the late B2; the latter has more plain shapes with thicker walls than the B1 series. From the B2 series, dated A.D. 70–150, Forms 60 (Fig. 60:E68) and 80 (Fig. 60:E69; classification of Hayes: Hayes 1985, 64, 69–70, pls. XIV:5–8, XV:15) were recognized. Form 60 is a dish with a sloping wall, incurved rim, and flat base, dated ca. A.D. 50–150. Similar vases come from Knossos (Hayes 1983, 104, fig. 4:32; Sackett 1992, 156–157, pl. 172:2, 4). Form 80 is a shallow bowl with a flattened rim and low foot, dated A.D. 80–150. Parallels are known at Knossos (Coldstream 1973b, 48, fig. 17; Sackett 1992, 157, pl. 167:19). A Knidian Gray Ware bowl has a thin, hard gray fabric and is dated 150 B.C.–A.D. 50 (Kenrick 1985,

58–63, 509, fig. 10; Hayes 1991, 12, fig. 5:12, 13). It has wide distribution in the eastern (see also a late example in Hagios Nikolaos: Davaras 1985, 194, pl. 54e) and central Mediterranean. A mold-made rim (Fig. 61:E70), without parallel ribbings, probably belongs to a vessel of Knidian Relief Ware. Its fabric is brownish red (10R 6/6–5/6) with slip of the same shade. This ware has wide distribution in the Mediterranean and is mainly dated in the 2nd century (Kenrick 1985, 327–337, pls. XX–XXII; Sackett 1992, 160, pls. 186, 192). Chandarli Ware is locally represented by 13 examples. A product of Chandarli (ancient Pitane) in the area of Pergamon, this ware dates to the 1st–3rd century. Its fabric is distinguished by large flakes of biotite mica. There are two versions: the earlier series (1st century) has orange to red clay (2.5YR 6/2) and orangish-red gloss thicker on the inside, which usually does not cover the underside of the base. The later series (2nd–3rd century) has reddish-brown to purplish clay (10R 5/8) with gloss of the same shade. The early series is represented by the base of a hemispherical bowl with a flange and low foot (Fig. 61:E71), dated in the 1st–2nd century (Form 3 in the Loeschcke’s classification: Hayes 1985, 76, pl. XVII:5–7). The late series produced many examples, such as: six dishes with plain or thickened rims, flaring walls, and low bases (Fig. 61:E72; cf. Hayes’s Form 2: Hayes 1972, 319–321; 1985, 77–78, pl. XVIII:2); one hemispherical bowl with a flange and low, heavy foot (Fig. 61:E73; cf. Hayes’s Form 3: Hayes 1972, 320–321; 1985, 78, pl. XVIII:3); and one dish with a curved wall and triangular tapering foot (Fig. 61:E74; cf. Hayes’s Form 4: Hayes 1972, 320–322; 1985, 78, pl. XVIII:4).All these forms are known at Knossos (Coldstream 1973b, 47, fig. 17; Hayes 1983, 118, figs. 2, 3; Sackett 1992, 158–159).

172

JOHN HAYES AND ANGELIKI KOSSYVA

A stamped base of a dish (Fig. 61:E75; Pl. 53:E87) bears a portrait of a man and an illegible inscription, probably made with a local coin of Augustan times. The dish may be an example of Cypriot Sigillata (Hayes 1985, 79–91) or Chandarli Ware. It has red to brown fabric (2.5YR 8/4–7/4) with fine lime and gold mica, the latter pointing to Chandarli Ware, which also features stamped vases in Augustan times (Hayes 1985, 72). Dishes and bowls of Phocean Red Slip Ware or Late Roman C (Pl. 54:E90–E99) became quite popular (84 examples) in the 4th century. This ware is a product of Asia Minor, possibly the Pergamon area. Its hard-fired and fine-grained fabric is red, with fine red slip that is mixed with the clay. The main decoration is rouletting and stamps (Hayes 1972, 323–370). Forms 3 (35 examples) and 10 (29 examples) occur in the Gournia survey area, though there is scant evidence of Forms 1 (2 examples), 2 (2 examples), or 8 (only 1 example; the forms follow Hayes’s classification: Hayes 1972). The Form 3 dish/bowl has a vertical rim incorporating a flange. It has a flaring wall and a low or false foot (Figs. 61:E76–E78, 62:E79). This is the most typical form of the ware and it is divided into eight types based on differences within the main shape. It is dated to the second half of the 5th–first half of the 6th century (Hayes 1972, 329–338). Among the Isthmus types, B–F are the most common. Form 10 is the successor of Hayes’s Form 3. It is a dish/bowl with a knobbed or flattened rim, a sloping wall, and a low foot (Fig. 62:E80, E81). It is divided into three types. Types A and B have been found; they are dated in the late 6th–early 7th century (Hayes 1972, 343–346). A Form 1 dish with a vertical, curved rim and foot of varying heights (Fig. 62:E82) is dated to the late 4th to mid 5th century (Hayes 1972, 325–327). A Form 2 dish with a broad flaring rim, flattened on top, and a shallow curved wall (Fig. 62:E83) has a flat base or low foot. It is dated in the late 4th to mid 5th century (Hayes 1972, 327–329). Form 8 is a bowl with a flange, carinated body, and small foot (Fig. 62:E84). It is not a common form and is dated in the second half of the 5th century (Hayes 1972, 340, 342). Vases of Phocean Red Slip Ware have also been found in early and late groups of Knossos (Sackett 1992, 161). A vase (Fig. 62:E85) has been identified as an imitation of Form 3 or 10. Its fabric is pinkish

(2.5YR 7/6–6/6) and is full of fine mica, which suggests it originated in a production center on the Asia Minor coast. Our examples of Phocean Red Slip Ware are decorated with rouletting, common in the 5th and early 6th centuries (Hayes 1972, 324). Stamped decoration is characteristic of the 5th to mid 6th century (Hayes 1972, 346). A fragmentary stamped floor with an elaborate floral motif (Fig. 62:E86; Pls. 54:E99, 55:E100) is classified, according to decorative motif, in Group II (ca. A.D. 440– 490) or III (A.D. 470–580), following Hayes’s classification. These groups are common for Forms 3 and 10 (Hayes 1972, 346–347, 349, 351–357). Light-Colored ware, associated with Phocean Red Slip Ware, is a product of a certain center in the eastern Aegean, likely in the Knidos area, and is dated to the mid 5th to early 7th century. It has fine clay of light orange or light brown color, with grits of gold mica. The thin slip is the same color but slightly darker. This ware includes dishes and small bowls, undecorated or with rouletting, grooving, stamps, and in the sgraffito technique (Hayes 1972, 408–410). Our area produced a knobbed rim of a dish and a foot fragment. Other examples are known from Itanos (Hayes 1972, 408). The survey area produced some 60 examples of African Red Slip Ware, the most widespread category of Late Roman pottery in the Mediterranean. The ware is made from granular clay of orange-red to brick-red color. Its main impurity is lime. It also has fine quartz, some black particles, and specks of silver mica. The slip has the same color as the clay. Decoration is stamped, incised, or modeled in relief. Often, however, the vessels are only decorated with grooves and rouletting. This ware is a product of Tynesia and is dated from the late 1st to the 7th century (Hayes 1972, 12–299). The survey area yielded vase forms (Pl. 56:E101–E111) dated in the 3rd–7th centuries A.D. (the forms are given according to Hayes’s classification: Hayes 1972). Dish (Fig. 63:E87) with plain rim, straight wall, and beveled foot, dated ca. 230/240–400 (Form 50: Hayes 1972, 69–73). Dish with flat base, curved flaring wall, and flat rim, dated ca. 325–400 (Form 57: Hayes 1972, 91–93).

ROMAN POTTERY

Dish (Fig. 63:E88) with flat base, curved wall, and short flat rim, dated ca. 290/300–375 (Form 58: Hayes 1972, 92–96). Dish corresponding with Form 58 but more shallow and with a broad flat rim, dated ca. 320–420 (Form 59: Hayes 1972, 96–100). Dish with flat base and vertical or slightly incurved rim, dated ca. 325–450 (Form 61: Hayes 1972, 100–107). Dish (Fig. 63:E89) with a plain rim, vertical wall, and tiny beveled foot, and is dated in the last quarter of the 4th century (Form 63.1 or 80A: Hayes 1972, 109–110). Form 80A is a shallow bowl with flat base, straight flaring wall, and plain rim, marked off by two grooves, and is dated in the mid to late 5th century (Hayes 1972, 124, 127–128). Bowl (Fig. 63:E90) with a flat base, curved wall, and two-part flaring rim, dated ca. 360–470 (Form 67: Hayes 1972, 112–116). Plate (Fig. 63:E91) with a knobbed rim, shallow sloping floor, and angular foot, dated in the 6th century (a variant of Forms 90/105: Hayes 1972, 138–140, 164, 166, 169). Dish/bowl (Fig. 63:E92) with a hooked or rolled rim, a sloping wall, broad shallow floor, and a large foot, dated ca. 500 to the third quarter of the 6th century (Form 103: Hayes 1972, 157–160). Dish/bowl (Fig. 63:E93) with a heavy knobbed rim, sloping or curved floor,and a foot of varying height, dated ca. 530–625 (Form 104: Hayes 1972, 160–166). Plate (Fig. 63:E94, E95) with a heavy knobbed rim, shallow sloping floor, and a rounded foot, dated ca. 580/600–660 (Form 105: Hayes 1972, 166–169). Two of the four local examples have different fabrics, suggesting they are imitations. Thickened rim (Fig. 63:E96) of a cooking-ware casserole or lid, dated in the 2nd or 3rd century (Hayes 1972, 17–18, 44–48, 201–209). Of the above forms, only those of the 3rd to 4th century (Forms 50, 58, 59, 61) are represented at

173

Knossos (Hayes 1983, 104, 118, fig. 4; Sackett 1992, 160–161, pls. 190, 219). Six examples of Cypriot Red Slip Ware (Pl. 64:E112–E117) were identified. Probably Cypriot in origin, this ware is dated from the late 4th century to ca. A.D. 700. The clay is very fine, with a few impurities of lime; in color it is almost yellow, or various shades of orange, brown, and red to maroon. The slip is a shade darker than the clay. The main decoration of the ware is coarse rouletting. Very often, the vessel rims have grooves, and some of the dishes have stamped floors (Hayes 1972, 371–386). According to Hayes’s classification (Hayes 1972), the following forms were recognized: Dish (Fig. 64:E97) with a plain, thickened rim or beveled on the outside, a sloping wall,and low foot, dated to the late fourth to third quarter of the 5th century (Form 1: Hayes 1972, 372–374). Dish with a knobbed, grooved rim, steep wall bearing rouletting, and a low foot, dated to the late 5th to mid 6th century (Form 2: Hayes 1972, 373–376). Dish/bowl (Fig. 64:E98) with straight, flaring sides and a flat rim terminating in a slightly upward hook, probably dated in the 6th century (Form 8: Hayes 1972, 378–379). Dish (Fig. 64:E99) with a broad, flat base, flaring wall, and thickened vertical or incurved rim. It has a low ledge-foot or groove around the edge of the base. It is dated ca. 550 to the end of the 7th century (Form 9: Hayes 1972, 378–382). Basin (Fig. 64:E100) with a vertical or incurved rim and a flat base grooved close to the edge. It is probably dated in the second half of the 7th century (Form 10: Hayes 1972, 380, 382–383). Dishes of Forms 1 and 2 have been also found in Ierapetra, suggesting the distribution route of the ware in the area. The latest fine ware found in the Gournia survey area is Egyptian Red Slip Ware. One example, perhaps an imitation, is a thickened incurved rim of a dish or bowl with eroded red slip (Fig.

174

JOHN HAYES AND ANGELIKI KOSSYVA

64:E101). The clay is pinkish to orange in color (7.5YR 8/3–8/4), with lime, quartz, dark grits, and

little fine mica, probably dated in the 7th century (Hayes 1972, 387–401).

Local Production Site 86 produced wasters of amphora Type AC1, wall pins, and slag possibly from the construction of the kiln or kilns (Pl. 58:E118–E123; Davaras 1973b). The asphalt road to Ierapetra and the continuous cultivation have removed any trace of the kiln(s). The local fabric is pinkish cream (2.5YR 7/8–6/8) with quartz, limestone, mica, and iron grit. Some sherds are covered with a greenish-brown slip. This site manufactured amphorae of Type AC1, stands, wall pins, basins, and perhaps buckets, bowls, and jugs. Production can be dated to the 2nd–4th centuries because AC1 amphorae and wall pins do not occur in contexts later than the 4th century (Hayes 1983; Markoulaki, Empereur, and Marangou 1989, 556; Empereur and Marangou

1992, 638). Similar items have also been found in other Cretan production centers of the same period (Empereur, Kritzas, and Marangou 1991, 493, 495, 499, 507, 510). This same fabric occurs in plain wares from other sites on the Ierapetra Isthmus (in particular, at sites 33, 40, 46, 73, 88, 91, 104, 110; see Fig. 51:E1) and from Hagios Nikolaos (from excavations opposite the O.T.E. telephone exchange building). Site 86 was an important production center of Roman pottery in the Isthmus area and beyond. Wasters and the overfired sherds (a handle of an amphora from site 74, a ring foot from site 104, and a rim of a basin or jar from site 103 [Pl. 59:E124–E126]) indicate the existence of other production sources.

References

Abbreviations follow the conventions of the American Journal of Archaeology 111.1 (2007), pp. 14–34.

Alcock, S.E. 1993. Graecia Capta: The Landscapes of Roman Greece, Cambridge.

EM III in Eastern Crete,” Aegean Archaeology 9, pp. 35–48.

Alexiou, St. 1954. “Υστερομινωικός τάφος Παχύαμμου,” CretChron 8, pp. 399–412.

Aruz, J. 1984. “The Silver Cylinder Seal from Mochlos,” Kadmos 23, pp. 186–188.

———. 1967. “Αρχαιότητες και Μνημεία Κρήτης,” ArchDelt 19 [1964], p. 441.

Ault, B. 1994. “Koprones and Oil Presses: Domestic Installations Related to Agricultural Productivity and Processing at Classical Halieis,” in Structures Rurales et Sociétés Antiques, P. Doukellis and L. Mendoni, eds., Paris, pp. 197–206.

Andreou, S. 1978. Pottery Groups of the Old Palace Period, Ph.D. diss., University of Cincinnati. Apostolakou, S. 1980. “Η Κόρη της Ιεράπετρας,” ArchDelt 35, pp. 31–36. ———. 2008. “A Workshop for Dyeing Wool at Pefka near Pacheia Ammos,” Kentro: The Newsletter of the INSTAP Study Center for East Crete 11, pp. 1–2. Apostolakou, V. 1987. “Λύχνοι ε Κρητικού Τύπου,” in ΕιλαΠινη: Tόμος τιμητικὸς γιὰ τὸν καθηγητὴ νικόλαο Πλάτωνα, L. Kastrinaki, G. Orphanou, and N. Giannadakis, eds., Herakleion, pp. 35–44. Apostolakou, V., P. P. Betancourt, and T.M. Brogan. 2007–2008. “The Alatzomouri Rock Shelter: Defining

Baldwin Bowsky, M.W. 1994. “Cretan Connections: The Transformation of Hierapytna,” Cretan Studies 4, pp. 1–44. ———. 2006a. “Highways and Byways of Roman Hierapytna (Crete): Four New Claudian Road Inscriptions,” ASAtene 84, pp. 551–580. ———. 2006b. “Roman Patronage, Greek Identity: How Lyttos Finally Won the War against Knossos,” in Πεπραγμένα Θ' Διεθνούς Κρητολογικού Συνεδρίου A' (5), Herakleion, pp. 407–419.

176

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

Balty, J.-C. 1991. Curia ordinis. Recherches d’architecture et d’urbanisme antiques sur les curies provincials du monde romain, Brussels. Barnard, K., and T. Brogan. 2003. Mochlos IB: Period III. Neopalatial Settlement on the Coast: The Artisans’ Quarter and the Farmhouse at Chalinomouri. The Neopalatial Pottery (Prehistory Monographs 8), Philadelphia. Bennet, J. 1985. “The Structure of the Linear B Administration at Knossos,” AJA 89, pp. 231–249. Betancourt, P.P. 1977. “Some Chronological Problems in the Middle Minoan Dark on Light Pottery of Eastern Crete,” AJA 81, pp. 341–353.

Betancourt, P.P., T.K. Gaisser, E. Koss, R.F. Lyon, R.F. Matson, S. Montgomery, G.H. Myer, and C.P. Swann. 1979. Vasilike Ware: An Early Bronze Age Pottery Style in Crete, Göteborg. Betancourt, P.P., and E. Hadjidaki. 2009. “Excavation of the Minoan Shipwreck near Pseira, Crete,” in Archaeological Institute of America 110th Annual Meeting Abstracts, January 8–11, 2009, Philadelphia, Pennsylvania, vol. 32, Boston, pp. 8–9. Betancourt, P.P., and R. Hope Simpson. 1992. “The Agricultural System of Bronze Age Pseira,” Cretan Studies 3, pp. 47–54.

———. 1980. Cooking Vessels from Minoan Kommos: A Preliminary Report, Los Angeles.

Betancourt, P.P., V. Karageorghis, R. Laffineur, and W.-D. Niemeier, eds. 1999. MELETEMATA: Studies in Aegean Archaeology Presented to Malcolm H. Wiener as He Enters His 65th Year (Aegaeum 20), 3 vols., Liège.

———. 1984. East Cretan White-On-Dark Ware: Studies on a Handmade Pottery of Early to Middle Minoan Periods (University Museum Monograph 51), Philadelphia.

Betancourt, P.P., J.D. Muhly, W.R. Farrand, C. Stearns, L. Onyshkevycha, W.B. Hafford, and D. Evely. 1999. “Research and Excavation at Chrysokamino, Crete, 1995–1998,” Hesperia 68, pp. 343–370.

———. 1985. The History of Minoan Pottery, Princeton.

Betancourt, P.P., and J. Silverman. 1991. The Cretan Collection in the University Museum, University of Pennsylvania II: Pottery from Gournia (University Museum Monograph 72), Philadelphia.

———. 1990. Kommos II: The Final Neolithic through Middle Minoan III Pottery, Princeton. ———. 1999. “What is Minoan? FN/EM I in the Gulf of Mirabello Region,” in Betancourt, Karageorghis, Laffineur, and Niemeier, eds., 1999, pp. 33–40.

Betts, J. 1967. “New Light on Minoan Bureaucracy,” Kadmos 6, pp. 15–40.

———. 2003. “The Impact of Cycladic Settlers on Early Minoan Crete,” Mediterranean Archaeology and Archaeometry 3, pp. 3–12.

Bintliff, J. 1997. “Regional Survey, Demography, and the Rise of Complex Societies in the Ancient Aegean: Core-Periphery, Neo-Malthusian, and Other Interpretive Models,” JFA 24, pp. 1–38.

———. 2004. “Pseira and Knossos: The Transformation of an East Cretan Seaport,” in Day, Mook, and Muhly, eds., 2004, pp. 22–28. ———. 2006. The Chrysokamino Metallurgy Workshop and Its Territory (Hesperia Suppl. 36), Princeton. ———. 2008. “Aphrodite’s Kephali,” Kentro: The Newsletter of the INSTAP Study Center for East Crete 11, pp. 12–13. Betancourt, P.P., and E.S. Banou. 1991. “Pseira and Minoan Sea Trade,” in THALASSA: L’Egée préhistorique et la mer. Actes de la 3e Recontre égéenne internationale de l’Université de Liège, Station de recherches sous-marines et océanographiques, Calvi, Corse, 23–25 avril 1990 (Aegaeum 7), R. Laffineur and L. Basch, eds., Liège, pp. 107–110. Betancourt, P.P., and C. Davaras. 2000. “Pseira,” in Muhly and Sikla, eds., 2000, pp. 38–57. Betancourt, P.P., C. Davaras, and R. Hope Simpson, eds. 2005. Pseira IX: The Archaeological Survey of Pseira Island. The Intensive Surface Survey (Prehistory Monographs 12), Philadelphia.

Bintliff, J., and A. Snodgrass. 1988. “Off-Site Pottery Distributions: A Regional and Inter-Regional Perspective,” CurrAnthr 29, pp. 506–513. Blackman, D., and K. Branigan. 1982. “Excavation of an Early Minoan Tholos Tomb from Ayia Kyriaki,” BSA 77, pp. 20–39. Blakely, J., R. Brinkmann, and C.J. Vitaliano. 1992. “Roman Mortaria and Basins from a Sequence at Caesarea: Fabrics and Sources,” in Caesarea Papers: Straton’s Tower, Herod’s Harbour, and Roman and Byzantine Caesarea (JRA Suppl. 5), Ann Arbor, pp. 194–213. Blanton, R., G. Feinman, S. Kowalewski, and P. Perigrine. 1996. “A Dual-Processual Theory for the Evolution of Mesoamerican Civilization,” CurrAnthr 37, pp. 1–14. Blitzer, H. 2004. “Agriculture and Subsistence in the Late Ottoman and Post-Ottoman Mesara,” in The Plain of Phaistos: Cycles of Social Complexity in the Mesara Region of Crete, L.V. Watrous, D. Hadzi-Vallianou, and H. Blitzer, eds., Los Angeles, pp. 111–220.

REFERENCES

Boardman, J., and J. Hayes. 1966. Excavations at Tocra, 1963–1965: The Archaic Deposits I (BSA Suppl. 4), London. Boserup, E. 1981. Population and Technological Change: A Study of Long-Term Trends, Chicago. Boyd, H.A. 1904. “Gournia: Report of the American Exploration Society’s Excavations at Gournia, Crete, 1901–1903,” in Transactions of the Department of Archaeology, FreeMuseum of Science and Art, University of Pennsylvania 1, i, Philadelphia, pp. 7–44. ———. 1904–1905. “Gournia: Report of the American Exploration Society’s Excavations at Gournia, Crete in 1904,” in Transactions of the Department of Archaeology, Free Museum of Science and Art, University of Pennsylvania 1, iii, Philadelphia, pp. 177–190. Branigan, K. 1988. “Some Observations on State Formation in Crete,” in French and Wardle, eds., 1988, pp. 63–72. ———. 1991. “Mochlos—An Early Gateway Community?” in THALASSA: L’Egée Préhistorique et la mer. Actes de la 3e Recontre égéenne internationale de l’Université de Liège, Station de recherches sousmarines et océanographiques, Calvi, Corse, 23–25 avril 1990 (Aegaeum 7), R. Laffineur and L. Basch, eds., Liège, pp. 97–107. ———. 1998. “Prehistoric and Early Historic Settlement in the Ziros Region, Eastern Crete,” BSA 93, pp. 74–90. Brants, J. 1913. Antieke terra-cotta lampen uit het Rijksmuseum van oudheben te Leiden, Leiden. Brogan, T.M., R.A.K. Smith, and J.S. Soles. 2002. “Mycenaeans at Mochlos? Exploring Culture and Identity in the Late Minoan IB to IIIA1 Transition,” Aegean Archaeology 6, pp. 89–118. Broneer, O. 1930. Terracotta Lamps (Corinth 4 [2]), Cambridge, MA. Brumfiel, E. 1976. “Regional Growth in the Eastern Valley of Mexico: A Test of the Population Pressure Hypothesis,’” in The Early Mesoamerican Village, K. Flannery, ed., New York, pp. 234–250. Brumfiel, E., and J. Fox. 1994. Factional Competition and Political Development in the New World, Cambridge. Cadogan, G. 1967. “Late Minoan IIIC Pottery from the Kephala Tholos Tomb near Knossos,” BSA 62, pp. 257–265. ———. 1977–1978. “Pyrgos, Crete, 1970–1977,” AR 24, pp. 70–84. ———. 1986. “Why Was Crete Different?” in The End of the Early Bronze Age in the Aegean, G. Cadogan, ed., Leiden, pp. 154–165.

177

———. 1992. “Myrtos/Pyrgos,” in Myers, Myers, and Cadogan, eds., 1992, pp. 202–209. ———. 2006. “A Power House of the Dead: The Functions and Long Life of the Myrtos-Pyrgos House Tomb,” in Archaeological Institute of America 108th Annual Meeting Abstracts, January 4–7, 2007, San Diego, Californa, vol. 30, Boston, p. 62. Callaghan, P.J. 1978. “KRS 1976: Excavations at a Shrine of Glaukos, Knossos,” BSA 73, pp. 1–30. ———. 1992. “Archaic to Hellenistic Pottery,” in Sackett, ed., 1992, pp. 89–136. Carneiro, R.L. 1981. “The Chiefdom: Precursor of the State,” in Transition to Statehood in the New World, G. Jones and R. Kautz, eds., Cambridge, pp. 37–79. Catling, H.W., and E.A. Catling. 1992. “The Lamps,” in Sackett, ed., 1992, pp. 257–322. Catling, H.W., E.A. Catling, and D. Smyth. 1979. “Knossos 1975: Middle Minoan III and LM I Houses by the Acropolis,” BSA 74, pp. 1–80. Chalikias, K., 2011. Settlement Patterns and Socioeconomic Change in the Ierapetra Region from the Bronze Age to the Ottoman Period: Chryssi Island— A Case Study, Ph.D. diss., Ruprecht Karl University of Heidelberg. Chaniotis, A. 1988. “Vinum creticum excellens: Zum Weinhandel Kretas,” Münstersche Beiträge zur antiken Handelsgeschichte 7, pp. 62–89. ———. 1995. “Problems of ‘Pastoralism’ and ‘Transhumance’ in Classical and Hellenistic Crete,” Orbis Terrarum 1, pp. 39–89. ———. 1996. Die Verträge zwischen kretischen Poleis in der hellenistischer Zeit (Heidelberger althistorische Beiträge und epigraphische Studien 24), Stuttgart. ———. 1999. “Milking the Mountains: Economic Activities on the Cretan Uplands in the Classical and Hellenistic Period,” in Chaniotis, ed., 1999, pp. 181– 220. ———. 2008. “What Difference Did Rome Make? The Cretans and the Roman Empire,” in The Province Strikes Back: Imperial Dynamics in the Eastern Mediterranean (Papers and Monographs of the Finish Institute at Athens 13), B. Forsén and G. Salmeri, eds., Helsinki, pp. 83–105. Chaniotis, A., ed. 1999. From Minoan Farmers to Roman Traders: Sidelights on the Economy of Ancient Crete, Stuttgart. Cherry, J.F. 1986. “Polities and Palaces: Some Problems in Minoan State Formation,” in Peer-Polity Interaction and Socio-Political Change, C. Renfrew and J.F. Cherry, eds., Cambridge, pp. 19–45.

178

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

Childe, V.G. 1947. History, London. Coldstream, J.N. 1960. “A Geometric Well at Knossos,” BSA 55, pp. 159–171.

Davaras, C., and J.S. Soles. 1995. “A New Oriental Cylinder Seal from Mochlos,” ArchEph 134, pp. 261–267.

———. 1968. Greek Geometric Pottery, London.

———. 2000. “Mochlos,” in Muhly and Sikla, eds., 2000, pp. 22–37.

———. 1972. “Knossos 1951–1961: Protogeometric and Geometric Pottery from the Town,” BSA 67, pp. 63–98.

Davis, E. 1979. “The Silver Kantharos from Gournia,” TUAS 4, pp. 34–45.

———. 1973a. “Knossos 1951–1961: Orientalizing and Archaic Pottery from the Town,” BSA 68, pp. 33–63. ———. 1973b. Knossos: The Sanctuary of Demeter (BSA Suppl. 8), Oxford. ———. 1992. “Early Hellenic Pottery,” in Sackett, ed., 1992, pp. 67–88. Coldstream, J.N., and H. Sackett. 1988. “Knossos: Two Deposits of Orientalizing Pottery,” BSA 73, pp. 45–60. Coleman, J.E. 1992. “Greece, the Aegean and Cyprus,” in Chronologies in Old World Archaeology I–II, 3rd ed., R.W. Ehrich, ed., Chicago, pp. 247–279, 203–221.

Davis, J.L. 1992. “The Islands of the Aegean,” AJA 96, pp. 699–756. ———. 2001. “Review of Aegean Prehistory I: The Islands of the Aegean,” in Aegean Prehistory: A Review, T. Cullen, ed., Boston, pp. 19–94. Day, L.P., N. Klein, and L. Turner. 2009. Kavousi IIA: The Late Minoan IIIC Settlement at Vronda. The Buildings on the Summit (Prehistory Monographs 26), Philadelphia. Day, L.P., M.S. Mook, and J. Muhly, eds. 2004. Crete beyond the Palaces. Proceedings of the Crete 2000 Conference (Prehistory Monographs 10), Philadelphia.

Cooper, J. 2002. The Cemetery of Sphoungaras: New Evidence for Social Hierarchy in Minoan Crete, M.A. thesis, State University of New York.

Day, L.P., and L.M. Snyder. 2004. “The ‘Big House’ at Vronda and the ‘Great House’ at Karphi: Evidence for Social Structure in LM IIIC Crete,” in Day, Mook, and Muhly, eds., 2004, pp. 63–79.

Coulson, W.D.E., D.C. Haggis, M.S. Mook, and J. Tobin. 1997. “Excavations on the Kastro at Kavousi: An Architectural Overview,” Hesperia 66, pp. 315–390.

Day, P.M. 1991. A Petrographic Approach to the Study of Pottery in Neopalatial East Crete, Ph.D. diss., University of Cambridge.

Davaras, C. 1972. “Αρχαιότητες και Μνημεία Ανατολικής Κρήτης,” ArchDelt 27, pp. 645–654.

———. 1995. “Pottery Production and Consumption in the Siteia Bay Area during the New Palace Period,” in Achladia: Scavi e recherché della Missione Greco-Italiana in Crete Orientale (1991– 1993), M. Tsipopoulou and L. Vagnetti, eds., Rome, pp. 148–175.

———. 1973a. “A New Hieroglyphic Seal from Mochlos,” Kadmos 12, pp. 109–113. ———. 1973b. “Κεραμεική κάμινος εις Ιστρώνα Ανατολικής Κρήτης,” ArchDelt 28, pp. 110–115. ———. 1974. “Αρχαιότητες και Μνημεία Ανατολικής Κρήτης,” Amaltheia 5, pp. 40–62. ———. 1975. “Early Minoan Jewelry from Mochlos,” BSA 70, pp. 101–114. ———. 1980. “Κρητικές Επιγραφές III,” ArchEph 1980, pp. 1–42. ———. 1985. “Ρωμαϊκό Νεκροταφείο Αγίου Νικολάου,” ArchEph 1987, pp. 130–212. Davaras, C., and P.P. Betancourt. 2000. “Pseira,” in Muhly and Sikla, eds., 2000, pp. 38–57.

———. 1997. “Ceramic Exchange between Towns and Outlying Settlements in Neopalatial East Crete,” in The Function of the Minoan Villa. Proceedings of the Eighth International Symposium at the Swedish Institute at Athens, 6–8 June, 1992, R. Hägg, ed., Stockholm, pp. 219–228. Day, P.M., and D.E. Wilson. 2006. “Reconstructing the Kampos Group: Technology, Mobility, and Identity in the Late Early Bronze I Aegean,” in Archaeological Institute of America 108th Annual Meeting Abstracts, January 4–7, 2007, San Diego, California, vol. 30, Boston, pp. 157–158.

———. 2004. The Hagia Photia Cemetery I: The Tomb Groups and Architecture (Prehistory Monographs 14), Philadelphia.

Day, P.M., D.E. Wilson, and E. Kiriatzi. 1997. “Reassessing Specialization in Prepalatial Cretan Ceramic Production,” in Laffineur and Betancourt, eds., 1997, II, pp. 275–289.

Davaras, C., P.P. Betancourt, and W. Farrand. 1992. “Pseira,” in Myers, Myers, and Cadogan, eds., 1992, pp. 276–281.

De Luca, G. 1984. Das Asklepieion: Via Tecta und Hallenstraße. Die Funde (Altertümer von Pergamon 4 [4]), Berlin.

REFERENCES

Demargne, P. 1945. Mallia: Exploration des Necropolis (1921–1933) (ÉtCrét 7), Paris. De Souza, P. 1998. “Late Hellenistic Crete and the Roman Conquest,” in Post-Minoan Crete. Proceedings of the First Colloquium on Post-Minoan Crete Held by the British School at Athens and the Institute of Archaeology, University College London, 10–11 November 1995 (BSA Studies 2), W.G. Cavanagh and M. Curtis, eds., Athens, pp. 112–116. Dickinson, O.T.P.K. 2006. The Aegean from Bronze Age to Iron Age: Continuity and Change between the Twelfth and Eighth Centuries B.C., London and New York. Dietz, S. 1971. “Aegean and Near-Eastern Metal Daggers in Early and Middle Bronze Age Greece,” ActaArch 42, pp. 1–22. Di Vita, A., ed. 1988. Gortina I (Monografie della Scuola Archeologica di Atene e delle Missioni Italiane in Oriente 3), Rome. Driessen, J. 1989–1990. “The Proliferation of Minoan Palatial Architectural Style: (I) Crete,” ActaArchLov 28–29, pp. 3–23.

179

Erickson, B. 2000. Late Archaic and Classical Crete: Island Pottery Styles in an Age of Historical Transition, ca. 600–400 B.C., Ph.D. diss., University of Texas at Austin. ———. 2005. “Archaeology of Empire. Athens and Crete in the Fifth Century B.C.,” AJA 109, pp. 619–663. Ettlinger, E., B. Hedinger, B. Hoffmann, P. Kendrick, G. Pucci, K. Roth-Rubi, G. Scheider, S. von Schnurbein, C. Well, and S. Zabehlicky-Scheffenegger. 1990. Conspectus formarum terrae sigillatae italico modo confectae, Bonn. Evans, A. 1921–1935. The Palace of Minos at Knossos I–IV, London. Evans, J. 1964. “Excavations in the Neolithic Settlement of Knossos, 1957–1960, Part I,” BSA 59, pp. 132–179. Flannery, K. 1972. “The Cultural Evolution of Civilizations,” Annual Review of Ecology and Systematics 3, pp. 399–426. ———. 1999. “Process and Agency in Early State Formation,” CAJ 9, pp. 3–21.

Driessen, J., and C.F. Macdonald. 1997. The Troubled Island: Minoan Crete before and after the Santorini Eruption (Aegaeum 17), Liège.

Floyd, C.R. 1998. Pseira III: The Plateia Building (University Museum Monograph 102), Philadelphia.

Driessen, J., J.A. MacGillivray, and H. Sackett, eds. 2003. Excavations at Palaikastro in East Crete: 1902–2002. A Centenary Exhibit, Athens.

Forbes, H. 1995. “The Identification of Pastoralist Sites within the Context of Estate-based Agriculture in Ancient Greece: Beyond the ‘Transhumance versus Agro-pastoralism Debate,’” BSA 90, pp. 325–338.

Ducrey, P., and O. Picard. 1969. “Recherches a Lato,” BCH 93, pp. 792–829. Duncan-Jones, R. 1974. The Economy of the Roman Empire, Cambridge. Earle, T. 1991. “The Evolution of Chiefdoms,” in Chiefdoms: Power, Economy and Ideology, T. Earle, ed., Cambridge, pp. 1–15.

Fotou, V. 1993. New Light on Gournia: Unknown Documents of the Excavation at Gournia and Other Sites on the Isthmus of Ierapetra by Harriet Ann Boyd (Aegaeum 9), Liège. Francis, J. 2006. “Beehives and Beekeeping in GraecoRoman Sphakia,” in Πεπραγμένα Θ' Διεθνούς Κρητολογικού Συνεδρίου A' (5), Heraklion, pp. 379–389.

Ehrenreich, R.M., C.L. Crumley, and J.E. Levy. 1995. Heterarchy and the Analysis of Complex Societies, Arlington.

Fraser, P.M. 1972. Ptolemaic Alexandria, Oxford.

Eliopoulos, T. 2004. “Gournia, Vronda, Kavousi, Kephala Vasilikis: A Triad of Interrelated Shrines of the Expiring Minoan Age on the Isthmus of Ierapetra,” in Day, Mook, and Muhly, eds., 2004, pp. 81–90.

Fried, M.H. 1967. The Evolution of Political Society: An Essay in Political Anthropology, New York.

Empereur, J.-Y., Ch. Kritzas, and A. Marangou. 1991. “Recherches sur les amphoras Crétoises II: Les centres de fabrication d’amphores en Créte centrale,” BCH 115, pp. 481–523. Empereur, J.-Y., and A. Marangou. 1992. “Recherches sur les amphores Crétoises (III),” BCH 116, pp. 633–648.

French, E.B., and K.A. Wardle, eds. 1988. Problems in Greek Prehistory, Bristol.

Furness, A. 1953. “The Neolithic Pottery of Knossos,” BSA 48, pp. 94–134. Gebhard, E.R. 1973. The Theater at Isthmia, Chicago. Gesell, G.C., L.P. Day, and W.D.E. Coulson. 2000. “Kavousi,” in Muhly and Sikla, eds., 2000, pp. 76–87. Gkiasta, M. 2008. A Historiography of Archaeological Surveys on Crete, Leiden.

180

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

Goodison, L. 2004. “From Tholos Tomb to Throne Room: Some Considerations of Dawn Light and Directionality in Minoan Buildings,” in Knossos: Palace, City, State (BSA Studies 12), G. Cadogan, E. Hatzaki, and A. Vasilakis, eds., Athens, pp. 339–350. Guarducci, M. 1942. Inscriptiones Creticae III: Tituli Cretae Orientalis, Rome. Guizzi, F. 2001. Hierapytna. Storia di una polis cretese dalla fondazione alla conquista romana, Rome. Hadjidaki, E., and P. Betancourt. 2006. “Το Πρώτο Μινωικό Ναυάγιο,” Κρητικό Πανόραμα 16, pp. 20–29. Hadzi-Vallianou, D. 2004. “Ελαιοκαλλιέργεια και ελαιοπαραγωγή στην Κρήτη κατά τη μινωική και έως την πρωτοβυζαντινή εποχή,” in Ελαιοσοδεία: Μελέτες για τον πολιτισμό της ελιάς. Παράρτημα της Επετηρίδας του Κέντρου Ερεύνης της Ελληνικής λαογραφίας της ακαδημίας αθηνών, 29–30 (1999–2003), A. Polymerou-Kamelaki, ed., Athens, pp. 65–130. Haggis, D.C. 1996a. “Excavations at Kalo Khorio, East Crete,” AJA 100, pp. 645–681. ———. 1996b. “The Port of Tholos in Eastern Crete and the Role of a Roman Horreum along the Egyptian ‘Corn Route,’” OJA 15, pp. 183–209. ———. 1997. “The Typology of the Early Minoan I Chalice and the Cultural Implications of Form and Style in Early Bronze Age Ceramics,” in Laffineur and Betancourt, eds., 1997, pp. 291–299. ———. 2001. “A Dark Age Settlement System in East Crete, and a Reassessment of the Definition of Refuge Settlements,” in Defensive Settlements of the Aegean and the Eastern Mediterranean after c. 1200 B.C. Proceedings of an International Workshop Held at Trinity College, Dublin, 7–9 May, 1999, V. Karageorghis and C.E. Morris, eds., Nicosia, pp. 41–57. ———. 2005. Kavousi I: The Archaeological Survey of the Kavousi Region (Prehistory Monographs 16), Philadelphia. ———. 2007. “Stylistic Diversity and Diacritical Feasting at Protopalatial Petras: A Preliminary Analysis of the Lakkos Deposit,” AJA 111, pp. 715–775. Haggis, D.C., and M.S. Mook. 1993. “The Kavousi Coarse Wares: A Bronze Age Chronology for Survey in the Mirabello Area, East Crete,” AJA 97, pp. 265–293. ———. 2011. “The Early Iron Age–Archaic Transition at Azoria in Eastern Crete: The Evidence from Recent Excavations at Azoria, Eastern Crete,” in The

‘Dark Ages’ Revisited: An International Symposium in Memory of William D.E. Coulson, A. MazarakisAinian, ed., Volos, pp. 515–528. Haggis, D.C., M.S. Mook, C.M. Scarry, L.M. Snyder, R.D. Fitzsimons, E. Stephanakis, and W.C. West. 2007. “Excavations at Azoria in 2003 and 2004, Part 1: The Archaic Civic Complex,” Hesperia 76, pp. 243–321. Haggis, D., M.S. Mook, C.M. Scarry, L.M. Snyder, and W.C. West. 2004. “Excavations at Azoria, 2002,” Hesperia 73 (3), pp. 339–400. Hall, E. 1904. “Early Painted Pottery from Gournia, Crete,” in Transactions of the Department of Archaeology, Free Museum of Science and Art University of Pennsylvania 1, pp. 191–206. ———. 1912. Excavations in Eastern Crete: Sphoungaras (University of Pennsylvania Museum Anthropological Publications 3 [2]), Philadelphia. ———. 1914. Excavations in Eastern Crete: Vrokastro (University of Pennsylvania Museum Anthropological Publications 3 [3]), Philadelphia. Hallager, E., and B.P. Hallager. 1995. “The Knossian Bull—Political Propaganda in Neo-Palatial Crete?” in POLITIEA: Society and State in the Aegean Bronze Age. Proceedings of the 5th International Aegean Conference, University of Heidelberg, Archäologisches Institut, 10–13 April 1994 (Aegaeum 12 [1]), R. Laffineur and W.-D. Niemeier, eds., Liège, pp. 547–560. Hallett, C.H. 2005. The Roman Nude: Heroic Portrait Statuary 200 B.C.–A.D. 300, Oxford and New York. Halstead, P. 1980. “Counting Sheep in Neolithic and Bronze Age Greece,” in Pattern of the Past: Studies in Honour of David Clarke, I. Hodder, G. Isaac, and N. Hammond, eds., Cambridge, pp. 307–339. ———. 1981. “From Determinism to Uncertainty: Social Storage and the Rise of the Minoan Palace,” in Economic Archaeology: Towards an Integration of Ecological and Social Approaches (BAR-IS 96), A. Sheridan and G. Bailey, eds., Oxford, pp. 187–214. Hannestad, N. 1986. Roman Art and Policy, Aarhus. Hansen, J.M. 1988. “Agriculture in the Prehistoric Aegean: Data versus Speculation,” AJA 92, pp. 39–52. Harrison, G.W.M. 1993. The Romans and Crete, Amsterdam. ———. 1998. “Crete the Ordinary,” in Post-Minoan Crete. Proceedings of the First Colloquium on PostMinoan Crete Held by the British School at Athens and the Institute of Archaeology, University College

REFERENCES

181

London, 10–11 November 1995 (BSA Studies 2), W.G. Cavanagh, and M. Curtis, eds., Athens, pp. 129–134.

Mediterraneo (tardo ellenismo e primo Impero), Rome, pp. 1–96.

Hawes, H. Boyd, B.E. Williams, R.B. Seager, and E.H. Hall. 1908. Gournia, Vasiliki and Other Prehistoric Sites on the Isthmus of Hierapetra, Crete. Excavations of the Wells-Houston-Cramp Expeditions, 1901, 1903, 1904, Philadelphia.

———. 1991. The Hellenistic and Roman Pottery (Paphos 3), Nicosia.

Hayden, B.J. 2000. “Vrokastro,” in Muhly and Sikla, eds., 2000, pp. 136–151. ———. 2003a. “Final Neolithic–Early Minoan I/IIA Settlement in the Vrokastro Area, Eastern Crete,” AJA 107, pp. 363–412. ———. 2003b. Reports on the Vrokastro Area, Eastern Crete I: Catalogue of Pottery from the Bronze and Early Iron Age Settlement of Vrokastro (University Museum Monograph 113), Philadelphia. ———. 2004a. Reports on the Vrokastro Area, Eastern Crete II: The Settlement History of the Vrokastro Area and Related Studies (University Museum Monograph 119), Philadelphia. ———. 2004b. “The Roman Period Settlement in the Vrokastro Area, Eastern Crete: Settlement Pattern, Sites, and Subsistence,” in Creta romana e protobizantina. Atti del congresso internazionale (Iraklion, 23–30 settembre 2000), II, M. Livadiotti and I. Simiakaki, eds., Padua, pp. 269–290. ———. 2005. Reports on the Vrokastro Area, Eastern Crete III: The Vrokastro Regional Survey Project. Sites and Pottery (University Museum Monograph 123), Philadelphia. Hayden, B.J., J. Moody, and O. Rackham. 1992. “The Vrokastro Survey Project, 1986–1989: Research Design and Preliminary Results,” Hesperia 61, pp. 293–353. ———. 2004b. “Introduction,” in Hayden 2004a, pp. 2–35. Hayden, B. 1995. “Pathways to Power: Principles for Creating Socioeconomic Inequalities,” in Foundations of Social Inequality, T. Price and G. Feinman, eds., New York, pp. 15–86. Hayes, J.W. 1967. “North Syrian Mortaria,” Hesperia 36, pp. 337–347. ———. 1971. “Four Early Roman Groups from Knossos,” BSA 66, pp. 249–275. ———. 1972. Late Roman Pottery, London. ———. 1983. “The Villa Dionysos Excavations, Knossos: The Pottery,” BSA 78, pp. 97–169. ———. 1985. “Sigillate orientali,” in Enciclopedia dell'arte antica classica e orientale: Atlante delle forme ceramice II. Ceramica fine romana nel bacino

Helms, M.W. 1988. Ulysses’ Sails: An Ethnographic Odyssey of Power, Knowledge, and Geographical Distance, Princeton. Higgins, R.A. 1980. Greek and Roman Jewelry, Berkeley. Hitchner, B. 1993. “Olive Production and the Roman Economy: The Case for Intensive Growth in the Roman Empire,” in La Production du vin et de l'huile en la Méditerranée (BCH Suppl. 26), M.-C. Amouretti and J.-P. Brun, eds., Athens, pp. 499–503. Hood, S. 1990. “Autochthons or Settlers? Evidence for Immigration at the Beginning of the Early Bronze Age in Crete,” in Πεπραγμένα ΣΤ' Διεθνούς Κρητολογικού Συνεδρίου A' (1), Chania, pp. 367–375. Iliopoulos, T. 1998. “A Preliminary Report on the Discovery of a Temple Complex of the Dark Ages at Kephala Vasilikis,” in Proceedings of the International Symposium: Eastern Mediterranean. Cyprus–Dodecanese–Crete, 16th–6th Centuries B.C., V. Karageorghis and N. Stampolidis, eds., Athens, pp. 301–313. Ízyigit, Í. 1990. “1989 yili Phokaia Kazi Çalişmalari,” Kazi Sonuçlari Toplantisi 12, pp. 127–153. ———. 1991. “1990 yili Phokaia Kazi Çalişmalari,” Kazi Sonuçlari Toplantisi 13, pp. 99–122. Jones, J.E., A.J. Graham, and L.H. Sackett. 1973. “An Attic Country House below the Cave of Pan at Vari,” BSA 68, pp. 357–452. Kalpaxis, Th., K. Athanassas, I. Bassiakos, T. Brennan, B. Hayden, E. Nodarou, K. Pavlopoulos, and A. Sarris. 2006. “Preliminary Results of the Istron, Mirabello, Geophysical and Geoarchaeological Project, 2002–2004,” BSA 101, pp. 135–181. Kane, S. 1985. “The Kore Who Looks after the Grain,” AJA 89, pp. 455–463. Kanta, A. 1980. The Late Minoan III Period in Crete: A Survey of Sites, Pottery and Their Distribution (SIMA 58), Göteborg. Karageorghis, V., and C.E. Morris, eds. 2001. Defensive Settlements of the Aegean and the Eastern Mediterranean after c. 1200 B.C., Nicosia. Kelly, A. 2006. “Distributions of Cretan Aqueducts; A Window onto Romanization,” in Πεπραγμένα Θ' Διεθνούς Κρητολογικού Συνεδρίου A' (5), Herakleion, pp. 391–405.

182

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

Kenrick, P.M. 1985. Excavations at Sidi Khrebish, Benghazi (Berenice) III.1: The Fine Pottery (LibAnt Suppl. 5 [2]), J.A. Lloyd, ed., Tripoli.

Manteli, K., and D. Evely. 1995. “The Neolithic Levels from the Throne Room System, Knossos,” BSA 90, pp. 1–16.

Knappett, C. 1999. “Assessing a Polity in Protopalatial Crete: The Malia-Lasithi State,” AJA 103, pp. 615– 640.

Marangou, A. 1999. “Wine in the Cretan Economy,” in Chaniotis, ed., 1999, pp. 269–278.

Koh, A. 2006. “A Late Minoan I Perfumed Oil Workshop at Mochlos, Crete,” in Archaeological Institute of America 108th Annual Meeting Abstracts, January 4–7, 2007, San Diego, Californa, vol. 30, Boston, pp. 30. Kosse, K. 1990. “Group Size and Societal Complexity: Thresholds in the Long-Term Memory,” JAnthArch 9, pp. 275–303. Kriga, D. 2010. “Which Markets Did the Theran Merchants Prefer for Exchanging Commodities? Imported Late Bronze Age Pottery from Akrotiri,” in Archaeological Institute of America 111th Annual Meeting Abstracts, 6–9, 2010, Anaheim, California, vol. 33, Boston, pp. 20–21. Krzyszkowska, O. 1988. “Ivory in the Aegean Bronze Age,” BSA 83, pp. 209–234. Laffineur, R., and P.P. Betancourt, eds. 1997. TEXNH: Craftsmen, Craftswomen and Craftsmanship in the Aegean Bronze Age. Proceedings of the 6th International Aegean Conference, Philadelphia, Temple University, 18–21 April 1996 (Aegaeum 16), Liège. La Rosa, V. 1997. “La ‘Villa Royale’ de Haghia Triada,” in The Function of the Minoan Villa. Proceedings of the Eighth International Symposium at the Swedish Institute at Athens, 6–8 June, 1992, R. Hägg, ed., Stockholm, pp. 78–89. Lekson, S. 1984. “Largest Settlement Size and the Interpretation of Socio-Political Complexity at Chaco Canyon, New Mexico,” paper given at the 49th Annual Meetings of the Society for American Archaeology, Santa Fe, NM. MacGillivray, J.A. 1998. Knossos: Pottery Groups of the Old Palace Period, London. Manning, S.W. 1994. “Emergence of Divergence: Development and Decline on Bronze Age Crete and the Cyclades,” in Development and Decline in the Mediterranean Bronze Age, C. Mathers and S. Stoddart, eds., Sheffield, pp. 221–270. ———. 1996. The Absolute Chronology of the Aegean Early Bronze Age, Sheffield. ———. 1999. A Test of Time, Oxford. Manteli, K. 1992. “The Neolithic Well at Kastelli Phournis in Eastern Crete,” BSA 87, pp. 103–120.

Marangou-Lerat, A. 1995. Le vin et les amphores de Crète de l’époque classique à l’époque impériale (ÉtCrét 30), Athens. Marinatos, S. 1974. Excavations at Thera VI, Athens. Markoulaki, St., J.-Y. Empereur, and A. Marangou. 1989. “Recherches sur les centres de fabrication d’amphores de Créte Occidentale,” BCH 113, pp. 551–580. Mattingly, D. 1988. “Megalithic Madness and Measurement, or How Many Olives Could an Olive Press Press?” OJA 7, pp. 177–195. Mavriyannaki, K. 1984. “Μινωικὴ μεταλλική μήτρα από την Βασιλική της Ιεράπετρας,” Κρητολογία 9, pp, 140–176. McGuire, R. 1983. “Breaking Down Cultural Complexity: Inequality and Heterogeneity,” in Advances in Archaeological Method and Theory 6, M. Schiffer, ed., New York, pp. 153–222. Mercando, L. 1974. “Lucerne romane del museo di Iraklion,” Antichità Cretesi: Studi in onore di Doro Levi II (Cronache di archeologia 14), Rome, pp. 235–239. Momigliano, N. 1991. “MMIA Pottery from Evans’ Excavations at Knossos: A Reassessment,” BSA 86, pp. 149–272. Momigliano, N., and D. Wilson. 1996. “Knossos 1993: Excavations outside the South Front of the Palace,” BSA 91, pp. 1–57. Mook, M.S., and W.D.E. Coulson. 1997. “Late Minoan IIIC Pottery from the Kastro at Kavousi,” in Late Minoan III Pottery: Chronology and Terminology (Monographs of the Danish Institute at Athens 1), B. Hallager and E. Hallager, eds., Aarhus, pp. 337–370. Mortzos, C. 1972. “Πάτιρα: Μια πρώιμος μινωικὴ κεραμεική ομάς,” Επετηρίς Κέντρου Επιστημονικών Ερευνών 3, pp. 386–421. Muhly, J., and E. Sikla, eds. 2000. One Hundred Years of American Archaeological Work on Crete, Athens. Myers, J.W., E.E. Myers, and G. Cadogan, eds. 1992. The Aerial Atlas of Ancient Crete, Berkeley. Mylona, D. 2003. “Fishing in Late Antiquity: The Case of Itanos, Crete,” in Zooarchaeology in Greece: Recent Advances (BSA Studies 9), E. Kotjabopoulou, Y. Hamilakis, P. Halstead, C. Gamble, and P. Elefanti, eds., Athens, pp. 103–110.

REFERENCES

———. 2007. Fish-Eating in Greece from the Fifth Century B.C. to the Seventh Century A.D.: A Story of Impoverished Fisherman or Luxurious Fish Banquets? Ph.D. diss., University of Southampton. Myrivilis, S. 1959. The Mermaid Madonna, New York. Narroll, R. 1962. “Floor Area and Settlement Population,” AmerAnt 27, pp. 587–589. Nikolakopoulou, I. 2006. “A Reappraisal of the Middle Bronze Age in the Cyclades in the Light of the New Evidence from Akrotiri, Thera,” in Archaeological Institute of America 108th Annual Meeting Abstracts, January 4–7, 2007, San Diego, Californa, vol. 30, Boston, p. 158. Nixon, L., and S. Price. 2001. “The Diachronic Analysis of Pastoralism through Comparative Variables,” BSA 96, pp. 395–424. Nowicki, K. 1999. “Final Neolithic Refugees or Early Bronze Age Newcomers? The Problem of Defensible Sites in Crete in the Late Fourth Millennium B.C.,” in Betancourt et al., eds., 1999, II, pp. 575–581. ———. 2000. Defensible Sites in Crete c. 1200–800 B.C. (LM IIIB/IIIC through Early Geometric) (Aegaeum 21), Liège. ———. 2001. “A Middle Minoan II Deposit at the Refuge Site of Monastiraki Katalimata (East Crete),” Aegean Archaeology 5, pp. 27–45. ———. 2004. “South of Kavousi, East of Mochlos: The West Siteia Mountains at the End of the Bronze Age,” in Day, Mook, and Muhly, eds., 2004, pp. 265–280.

183

Perlman, P. 1996. “Πόλις Υπήκοος: The Dependent Polis and Crete,” in Introduction to an Inventory of “Poleis”: Symposium August 23–26, 1995 (Acts of the Copenhagen Polis Centre 3), M.H. Hansen, ed., Copenhagen, pp. 233–287. ———. 1999. “Kretes aei Leistai? The Marginalization of Crete in Greek Thought and the Role of Piracy in the Outbreak of the First Cretan War,” in Hellenistic Rhodes: Politics, Culture, and Society (Studies in Hellenistic Civilization IX), V. Gabrielsen, T.-E. Pedersen, L. Hannestad, P. Bilde, and J. Zahle, eds., Aarhus, pp. 132–166. Pini, I. 2000. “Eleven Early Scarabs,” in Κρήτη– αίγυπτοϚ: Πολιτισμικοί δεσμοί τριών χιλιετιών, A. Karetsou, ed., Athens, pp. 86–93. Pini, I., ed. 1975. Kleinere griechische Sammlungen (CMS V), 2 vols., Berlin. Platon, L. 1999. “Ανυπόγραφα ‘έργα τέχης’ στα χέρια ιδιωτών κατά τη νεοανακτορική περίοδο στην Κρήτη,” in Eliten in der Bronzeit (Monographien des Römisch-Germanisches Zentralmuseum, Forschungsinstitut für Vor- und Frühgeschichte 1), Mainz, pp. 37–50. Platon, N. 1960. “Έρευναι της Αρχαιολογικής Εταιρείας εις Ανατολικήν και Κεντρικήν Κρήτην,” ArchDelt 16, pp. 258–262. Platon, N., ed. 1969. Iraklion, Archäologishes Museum: Die Siegel der Vorpalastzeit (CMS II, 1), Mainz. Platon, N., I. Pini, and G. Salies, eds. 1977. Iraklion, Archaologisches Museum. Die Siegel der Altpalastzeit (CMS II, 2), Berlin.

———. 2008. Monastiraki Katalimata: Excavation of a Cretan Refuge Site, 1993–2000 (Prehistory Monographs 24), Philadelphia.

Pullen, D. 1992. “Ox and Plow in the Early Bronze Age Aegean,” AJA 96, pp. 45–54.

Paluchowski, A. 2005. “Le régime des cités crétoises sous le Haut Empire,” Klio 87, pp. 421–444.

Raftopoulou, E.G. 1975. L’Enfant d’Hiérapétra (TravMém 20), Paris.

Papadakis, N. 1986. Ierapetra: Bride of the Libyan Sea, Ierapetra.

Rapp, D. 2007. “Building Megara for Dummies: The Conception and Construction of Architectural Forms at Late Minoan IIIC Halasmenos (Monasteraki, Ierapetra, Crete),” in Krinoi kai Limenai: Studies in Honor of Joseph and Maria Shaw (Prehistory Monographs 22), P.P. Betancourt, M.C. Nelson, and E.H. Williams, eds., Philadelphia, pp. 61–66.

Papaiannopoulou, A. 1991. The Influence of Middle Minoan Pottery on the Cyclades, Göteborg. Paton, S., and R.M. Schneider. 1999. “Imperial Splendour in the Province: Imported Marble on Roman Crete,” in Chaniotis, ed., 1999, pp. 279–304. Peacock, D.P.S., and D.F. Williams. 1986. Amphorae and the Roman Economy, London and New York. Peatfield, A. 1990. “Minoan Peak Sanctuaries: History and Society,” OpAth 18, pp. 117–131. Peebles, C.S., and S.M. Kus. 1977. “Some Archaeological Correlates of Ranked Societies,” AmerAnt 42, pp. 421–448.

Rehak, P., and J. Younger. 1998. “A Review of Aegean Prehistory VII: Neopalatial, Final Palatial, and Postpalatial Crete,” AJA 102, pp. 91–173. Rendini, P. 1988a. “Anfore,” in Di Vita, ed., 1988, pp. 263–277. ———. 1988b. “Ceramica acroma,” in Di Vita, ed., 1988, pp. 229–251.

184

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

———. 1988c. “Ceramica da cucina,” in Di Vita, ed., 1988, pp. 253–261. ———. 1990. “Anfore di produzione locale e d’importazione a Gortina,” in Πεπραγμένα ΣΤ' Διεθνούς Κρητολογικού Συνεδρίου A' (2), Chania, pp. 233–240. Renfrew, C. 1972. The Emergence of Civilisation: The Cyclades and the Aegean in the Third Millennium B.C., London. Renfrew, C., and J. Cherry, eds. 1986. Peer-Polity Interaction and Socio-Political Change, Cambridge. Rethemiotakis, G. 1999. “Social Rank and Political Power: The Evidence from the Minoan Palace at Galatas,” in Eliten in der Bronzezeit (Monographien des Römisch-Germanisches Zentralmuseum, Forschungsinstitut für Vor- und Frühgeschichte 43), Mainz, pp. 9–26. Ridgway, B.S. 2002. Hellenistic Sculpture III: The Styles of ca. 100–31 B.C., Madison.

Schultz, M. 2007. Social Organization in the Mirabello Region of Crete: From Farm to Palace, Ph.D. diss., University at Buffalo. Schwartzott, A. 1997. An Investigation of the Social Organization at Gournia Based on Object Typologies, M.A. thesis, State University of New York. Seager, R.B. 1905. “Excavations at Vasiliki, 1904,” in Transactions of the Department of Archaeology, Free Museum of Science and Art, University of Pennsylvania 1, iii, Philadelphia, pp. 207–221. ———. 1907. “Report of Excavations at Vasiliki, Crete, in 1906,” in Transactions of the Department of Archaeology, Free Museum of Science and Art, University of Pennsylvania 2, ii, pp. 111–132. ———. 1908. “Excavations at Vasiliki,” in Gournia, Vasiliki, and Other Prehistoric Sites on the Isthmus of Hierapetra, Crete: Excavations of the Wells-HoustonCramp Expeditions, 1901, 1903, 1904, H.A.B. Hawes, B.E. Williams, R.B. Seager, and E.H. Hall, eds., Philadelphia, pp. 49–50.

Riley, J. 1979. Excavations at Sidi Khrebish Benghazi (Berenice) II: Coarse Pottery (LibAnt Suppl. 5 [2]), J.A. Lloyd, ed., Tripoli, pp. 91–467.

———. 1909. “Excavations on the Island of Mochlos, Crete, in 1908,” AJA 13, pp. 272–303.

Robinson, H.S. 1959. Pottery of the Roman Period: Chronology (Agora 5), Princeton.

———. 1910. Excavations on the Island of Pseira, Crete (The University of Pennsylvania, The Museum Anthropological Publications III[i]), Philadelphia.

Rouanet-Liesenfelt, A.-M. 1992. “Les plantes medicinales de Crète a l’époque romaine,” Cretan Studies 3, pp. 175–190.

———. 1912. Explorations on the Island of Mochlos, Boston and New York.

Sackett, L.H. 1992. “Roman Pottery,” in Sackett, ed., 1992, pp. 148–480.

———. 1916. The Cemetery of Pachyammos, Crete (The University of Pennsylvania, The Museum Anthropological Publications VII[i]), Philadelphia.

Sackett, L.H., ed. 1992. Knossos: From Greek City to Roman Colony. Excavations at the Unexplored Mansion II (BSA Suppl. 21), London.

Seiradaki, M. 1960. “Pottery from Karphi,” BSA 55, pp. 1–37.

Sackett, L.H., and M.R. Popham. 1970. “Excavations at Palaikastro VII,” BSA 65, pp. 203–242. Sakellarakis, J. 1970. “Το Μινωικόν Ορεινόν Ιερόν του Θύλακα,” AAA 3, pp. 252–259. Sakellarakis, Y., and E. Sapouna-Sakellaraki. 1997. αρχάνες: Μια νέα ματιά στη μινωική Κρήτη, Athens. Sanders, I.F. 1982. Roman Crete, Warminster. Schachermeyr, F. 1938. “Vorberichtäber eine Expedition nach Ostkreta,” AA 1938, pp. 470. Schäfer, J. 1957. Studien zu den griechischen Reliefpithoi des 8–6 Jahrhunderts v. Chr. Aus Kreta, Rhodes, Tenos, und Boiotien, Stuttgart. Schultke, K.-H., and A. Zois. 1993. Aerodromio Anatolikis Kritis, Athens.

Shaw, J. 1986. “Excavations at Kommos (Crete) during 1984–1985,” Hesperia 55, pp. 240–269. ———. 1990. “North American Archaeological Work in Crete, 1880–1990,” Expedition 32, pp. 6–14. Sherratt, A. 1981. “Plough and Pastoralism: Aspects of the Secondary Products Revolution,” in Pattern of the Past: Studies in Honour of David Clarke, I. Hodder, G. Isaac, and N. Hammond, eds., Cambridge, pp. 261–305. Smith, R.A.K. 2005. “Minoans, Myceneans, and Mokhlos: The Formation of Regional Identity in Late Minoan III Crete,” in Ariadne’s Threads: Connections between Crete and the Greek Mainland in the Late Minoan III (LM IIIA2 to LM IIIC), A.L. D’Agata and J.A. Moody, eds., Athens, pp. 185–204.

REFERENCES

185

———. 2009. Mochlos IIB: Period IV. The Mycenaean Settlement and Cemetery: The Pottery (Prehistory Monographs 27), Philadelphia.

Stefanaki, V. 2005. Le monnayage d’Hiérapytna (Crète orientale) de la fin de l’époque classique a l’époque impériale, Ph.D. diss., Université Paris IV Sorbonne.

Soles, J.S. 1979. “The Early Gournia Town,” AJA 83, pp. 149–167.

———. 2006. “Η οικονομική ανάπτυξη της Ιεράπυτνας,” in Πεπραγμένα Θ' Διεθνούς Κρητολογικού Συνεδρίου A' (5), Herakleion, pp. 303–318.

———. 1988. “Social Ranking in Prepalatial Cemeteries,” in French and Wardle, eds., 1988, pp. 49–61. ———. 1991. “The Gournia Palace,” AJA 95, pp. 17–78.

Steponaitis, V. 1981. “Settlement Hierarchies and Political Complexity in Nonmarket Societies: The Formative Period of the Valley of Mexico,” American Anthropologist 83, pp. 320–363.

———. 1992. Prepalatial Cemeteries at Mochlos and Gournia and the House Tombs of Bronze Age Crete (Hesperia Suppl. 24), Princeton.

Stos-Gale, Z.A. 1985. “A Puzzle of the Bronze Age Silver-Lead on Cyprus,” in Proceedings of the Fifth Cypriote Congress, Nicosia, pp. 67–72.

———. 2002. “A Central Court at Gournia,” in Monuments of Minos: Rethinking the Minoan Palaces. Proceedings of the International Workshop “Crete of the Hundred Palaces?”, Université Catholique de Louvain-la-Neuve, 14–15 December 2001 (Aegaeum 23), J. Driessen, I. Schoep, and R. Laffineur, eds., Liège, pp. 123–142.

Stos-Gale, Z.A., and C.F. Macdonald. 1991. “Sources of Metals and Trade in the Bronze Age Aegean,” in Bronze Age Trade in the Mediterranean. Papers Presented at the Conference Held at Rewley House, Oxford, in December 1989, N.H. Gale, ed., Stockholm, pp. 249–288.

———. 2003. Mochlos IA: Period III. Neopalatial Settlement on the Coast: The Artisans’ Quarter and the Farmhouse at Chalimouri. The Sites (Prehistory Monographs 7), Philadelphia.

Stürmer, V. 1993. “La céramique de Chrysolakkos: Catalogue et réexamen,” BCH 117, pp. 123–187. Tainter, J. 1988. The Collapse of Complex Societies, Cambridge.

———. 2005. “From Ugarit to Mochlos – Remnants of an Ancient Voyage,” in EMPORIA: Aegeans in the Central and Eastern Mediterranean. Proceedings of the 10th International Aegean Conference, Italian School of Archaeology in Athens, 14–18 April 2004 (Aegaeum 25), R. Laffineur and E. Greco, eds., Liège and Austin, pp. 429–448.

Taramelli, A. 1897. “Cretan Expedition VIII: The Prehistoric Grotto at Miamu,” AJA 1, pp. 287–312.

———. 2009. “2009 Greek-American Excavation at Mochlos,” Kentro: The Newsletter of the INSTAP Study Center for East Crete 12, pp. 9–11. Soles, J.S., and C. Davaras. 1992. “Excavations at Mochlos, 1989,” Hesperia 61, pp. 413–445.

Tomkins, P. 2007. “Neolithic: Strata IX–VIII, VII–VIB, VIA–V, IV, IIIB, IIIA, IIB, IIA, and IC Groups,” in Knossos Pottery Handbook: Neolithic and Bronze Age (Minoan) (BSA Studies 14), N. Momigliano, ed., London, pp. 9–49.

———. 1994. “Excavations at Mochlos, 1990–1991,” Hesperia 63, pp. 391–436.

Tsipopoulou, M. 1988. “Αγία Φωτιά Σητείας: Το νέο εύρημα,” in French and Wardle, eds., 1988, pp. 31–48.

———. 1996. “Excavations at Mochlos, 1992–1993,” Hesperia 65, pp. 175–230. ———. 2000. “Mochlos,” in Muhly and Sikla, eds., 2000, pp. 22–37. Soles, J.S., and C. Davaras, eds. 2008. Mochlos IIA: Period IV. The Mycenaean Settlement and Cemetery: The Sites (Prehistory Monographs 23), Philadelphia. Sparkes, B., and L. Talcott. 1970. Black and Plain Pottery of the 6th, 5th, and 4th Centuries B.C. (Agora 12), Princeton. Stein, G. 1999. Rethinking World Systems, Tucson.

Thomas, E. 2007. Monumentality and the Roman Empire: Architecture in the Antonine Age, Oxford. Thompson, F.H. 2003. The Archaeology of Greek and Roman Slavery, London.

———. 2001. “A New Late Minoan IIIC Shrine at Halasmenos, East Crete,” in POTNIA: Deities and Religion in the Aegean Bronze Age. Proceedings of the 8th International Aegean Conference, Göteborg University, 12–15 April 2000 (Aegaeum 22), R. Laff ineur and R. Hägg, eds., Liège, pp. 99–101. ———. 2004a. “Halasmenos, Destroyed but Not Invisible: New Insight on the LM IIIC Period in the Isthmus of Ierapetra. First Presentation of the Pottery from the 1992–1997 Campaigns,” in Day, Mook, and Muhly, eds., 2004, pp. 103–123. ———. 2004b. “Μια περίπτωσι πρώιμων συμποσίων ή απλώς ηρωολατρείας: Γεωμετρική ανακατάληψη

186

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

στο Χαλασμένο Ιεράπετρας,” in Το αιγαίο στην Πρώιμη Εποχή του Σιδήρου, N. Stampolidis and A. Giannikouri, eds., Athens, pp. 127–141.

Επιστημονική Συνάντηση για την Ελληνιστική Κεραμική, Βόλος, 17–23 απριλίου 2000, Athens, pp. 327–334.

———. 2005. “‘Mycenoans’ at the Isthmus of Ierapetra,” in Ariadne’s Threads: Connections between Crete and the Greek Mainland in the Late Minoan III (LM IIIA2 to LM IIIC), A.L. D’Agata and J.A. Moody, eds., Athens, pp. 303–334.

Vogeikoff-Brogan N., E. Nodarou, and M.C. Boileau. 2008. “New Evidence for Wine Production in East Crete in the Hellenistic Period: An Integrated Approach of Typology and Thin Section Petrography,” in Proceedings of the 4th Symposium of the Hellenic Society for Archaeometry, Held in Athens 2003, Oxford, pp. 327–334.

———. 2008. “Foreword,” in Monastiraki Katalimata: Excavation of a Cretan Refuge Site, 1993–2000 (Prehistory Monographs 24), K. Nowicki, Philadelphia, pp. xvii–xxi. Tsipopoulou, M., and W.D.E. Coulson. 1994. “Preliminary Investigations at Halasmenos, East Crete, 1992–93,” Aegean Archaeology 1, pp. 65–97. Tsipopoulou, M., L. Vagnetti, and M. Liston. 2003. “New Evidence for the Dark Ages in the Isthmus of Hierapetra: An Unplundered Tholos Tomb at Vasiliki,” SMEA 45, pp. 85–124. Tzedakis, Y., St. Chryssoulaki, S. Voutsaki, and Y. Venierei. 1989. “Les Routes Minoennes: Rapport Préliminaire Defense de la Circulation ou Circulation de la Defense?” BCH 113, pp. 44–78. Vagnetti, L. 1972–1973. L'insediamento neolitico di Festos (ASAtene 50/51), pp. 7–138. ———. 1996. “The Final Neolithic: Crete Enters the Wider World,” Cretan Studies 5, pp. 29–39. Vagnetti, L., A. Christopoulou, and Y. Tzedakis. 1989. “Saggi negli strati neolitici,” in Scavi a Nerokourou, Kydonias. Ricerche greco italiane in Creta Occidentale I (Incunabula Graeca 91), Y. Tzedakis, A. Saccopni, and L. Vagnetti, eds., Rome, pp. 11–97. van Effenterre, H., and M. van Effenterre. 1989. “Un obituaire Crétois?” Ariadne 5, pp. 99–107. Viviers, D. 1999. “Economy and Territorial Dynamics in Crete from the Archaic to the Hellenistic Period,” in Chaniotis, ed., 1999, pp. 221–233. Vogeikoff-Brogan, N. 2004. “The Late Hellenistic Period in East Crete,” in Day, Mook, and Muhly, eds., 2004, pp. 213–220. Vogeikoff-Brogan, N., and S. Apostolakou. 2004. “New Evidence for Wine Production in East Crete in the Hellenistic Period,” in Transport Amphorae and Trade in the Eastern Mediterranean. Acts of the International Colloquium at the Danish Institute at Athens, 26–29 September 2002, J. Eiring and J. Lund, eds., Aarhus, pp. 417–427. Vogeikoff-Brogan, N., J. Eiring, M.-C. Boileau, and I.K. Whitbread. 2004. “Transport Amphoras and Wine Trade in East Crete in the Late Hellenistic Period. Evidence from Mochlos and Pyrgos Myrtos,” in ΣΤ'

Wallace, S., 2003. “The Perpetuated Past: Re-use or Continuity in Material Culture and the Structuring of Identity in Early Iron Age Crete,” BSA 98, pp. 251–277. ———. 2005. “Last Chance to See? Karfi (Crete) in the Twenty-First Century: Presentation of New Architectural Data and Their Analysis in the Current Context of Research,” BSA 100, pp. 215–274. Warren, P.M. 1965. “The First Minoan Stone Vases and Early Minoan Chronology,” CretChron 1965, pp. 7–43. ———. 1969. Minoan Stone Vases, London. ———. 1972. Myrtos: An Early Bronze Age Settlement in Crete (BSA Suppl. 7), London. ———. 1987. “The Genesis of the Minoan Palace,” in The Function of the Minoan Palaces, R. Hägg and N. Marinatos, eds., Stockholm, pp. 47–56. Warren, P.M., and V. Hankey. 1989. Aegean Bronze Age Chronology, Bristol. Watrous, L.V. 1982. Lasithi: A History of Settlement on a Highland Plain in Crete (Hesperia Suppl. 18), Princeton. ———. 1992. Kommos III: The Late Bronze Age Pottery, Princeton. ———. 1994. “Review of Aegean Prehistory III: Crete from Earliest Prehistory through the Protopalatial Period,” AJA 98, pp. 695–753. ———. 1996. The Cave Sanctuary of Zeus at Psychro: A Study of Extra-urban Sanctuaries in Minoan and Early Iron Age Crete (Aegaeum 15), Liège. ———. 2000. “Gournia,” in Muhly and Sikla, eds., 2000, pp. 126–135. ———. 2001a. “The Isthmus of Ierapetra in East Crete and the Dark Age Refuge Settlement of Profitis Elias: Diaspora or Local Population Change?,” in Defensive Settlements of the Aegean and the Eastern Mediterranean after c. 1200 B.C., V. Karageorghis and C.E. Morris, eds., Nicosia, pp. 85–94. ———. 2001b. “Review of Aegean Prehistory III: Crete from Earliest Prehistory through the Protopalatial

REFERENCES

Period,” in Aegean Prehistory: A Review, T. Cullen, ed., Boston, pp. 157–223. ———. 2005a. “Cretan International Relations during the Middle Minoan IA Period and the Chronology of Seager’s Finds from the Mochlos Tombs,” in EMPORIA: Aegeans in the Central and Eastern Mediterranean. Proceedings of the 10th International Aegean Conference, Italian School of Archaeology in Athens, 14–18 April 2004 (Aegaeum 25), P. P. Betancourt and R. Laffineur, eds., Liège, pp. 107–115. ———. 2005b. “New Pottery from the Psychro Cave and Its Implications for Minoan Crete,” BSA 99, pp. 129–147. ———. 2007. “How Was the Palace at Gournia Established? Minoan State Formation in the Mirabello Region of East Crete,” in Archaeological Institute of America 109th Annual Meeting Abstracts, January 3–6, 2008, Chacago, Illinois, vol. 31, Chicago, p. 101.

187

Archaeology of Houses and Households in Ancient Crete (Hesperia Suppl. 44), K. Glowacki and N. Vogeikoff-Brogan, eds., Princeton, pp. 199–212. Weingarten, J. 1986. “The Sealing Structures of Minoan Crete: MM II Phaistos to the Destruction of the Palace at Knossos,” OJA 5, pp. 279–298. ———. 1990. “Three Upheavals in Minoan Sealing Administration: Evidence for Radical Change,” in Aegean Seals, Sealings, and Administration Proceedings of the NEH-Dickson Conference of the Program in Aegean Scripts and Prehistory of the Department of Classics, University of Texas at Austin, January 11–13, 1989 (Aegaeum 5), T. Palaima, ed., Liège, pp. 105–120. Whitelaw, T., P. Day, E. Kiriatzi, V. Kilikoglou, and D. Wilson. 1997. “Ceramic Traditions at EM IIB: Myrtos, Fournou Korifi,” in Laffineur and Betancourt, eds., 1997, II, pp. 265–274.

———. 2008. “Secondary State Formation on Crete: The Mirabello Region during the Bronze Age,” paper presented at the INSTAP Study Center for East Crete, Pacheia Ammos, Greece, June 2008.

Whittaker, H. 2002. “Minoan Board Games: The Function and Meaning of Stones with Depressions (So-called Kernoi) from Bronze Age Crete,” Aegean Archaeology 6, pp. 73–88.

———. 2010. “The Coastal Town at Gournia: Fortification Walls and Towers, Street, Dam and Agricultural Terraces,” in Archaeological Institute of America 111th Annual Meeting Abstracts, 6–9, 2010, Anaheim, California, vol. 33, Boston, p. 113.

Wiener, M.H. 2007. “Neopalatial Knossos: Rule and Role,” in Krinoi kai Limenes: Studies in Honor of Joseph and Maria Shaw (Prehistory Monographs 22), P.P. Betancourt, M.C. Nelson, and E.H. Williams, eds., Philadelphia, pp. 231–242.

———. 2012. “The Harbor Complex of the Minoan Town at Gournia,” AJA 116, pp. 521–541.

Wilson, D. 1985. “Pottery and Architecture from the EM IIA West Court House,” BSA 80, pp. 281–364.

Watrous, L.V., and H. Blitzer. 1995. “The Gournia Survey Project: A Preliminary Report on the 1992– 1994 Field Seasons,” AJA 99, p. 313.

Wilson, D., and P. Day. 1994. “Ceramic Regionalism in Prepalatial Central Crete: The Mesara Imports from EM I to EM IIA Knossos,” BSA 89, pp. 1–87.

———. 1999. “The Region of Gournia in the Neopalatial Period,” in Betancourt et al., eds., 1999, pp. 905–909.

———. 1999. “EM IIB Ware Groups at Knossos: The 1907–1908 South Front Tests,” BSA 94, pp. 1–62.

Watrous, L.V., H. Blitzer, D.C. Haggis, and E. Zangger. 2000. “Economy and Society in the Gournia Region of Crete. A Preliminary Report on the 1992–1994 Seasons of the Gournia Project,” in Πεπραγμένα η' Διεθνούς Κρητολογικού Συνεδρίου A' (3), Herakleion, pp. 471–483. Watrous, L.V., and D. Hadzi-Vallianou. 2004. “Mesara Romana (150 B.C.–A.D. 400),” in Watrous, HadziVallianou, and Blitzer, eds., 2004, pp. 351–358. Watrous, L.V., D. Hadzi-Vallianou, and H. Blitzer. 2004. The Plain of Phaistos: Cycles of Social Complexity in the Mesara Region of Crete (Monumenta Archaeologica 23), Los Angeles. Watrous, L.V., and A. Heimroth. 2011. “Household Industries of Late Minoan IB Gournia and the Socioeconomic Status of the Town,” in ΣΤΕγα: The

———. 2000. “EM I Chronology and Social Practice: Pottery from the Early Palace Tests at Knossos,” BSA 95, pp. 21–63. Woodard, M.E. 1982. The Principal Figural Reliefs of the Cretan Relief Pithoi of the Seventh Century B.C., M.A. thesis, University of North Carolina. Wright, H.T. 1977. “Recent Research on the Origin of the State,” Annual Review of Anthropology 6, pp. 379–397. ———. 1984. “Prestate Political Formations,” in On the Evolution of Complex Societies: Essays in Honor of Harry Hoijer 1982, W. Sanders, H. Wright, and R.M. Adams, eds., Malibu, pp. 41–77. Xanthoudides, S. 1920–1921. “Chronicles,” ArchDelt 6, pp. 157–162. Zois, A.A. 1972. “Βασιλική Ιεράπετρας,” Ergon, pp. 113–118.

188

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

———. 1978. Βασιλική Ι, pp. 67–73, Athens. ———. 1979. “Ανασκαφή εις Βασιλικήν Ιεράπετρας,” Prakt 134 (1981), pp. 323–330.

———. 1992. “Vasiliki,” in Myers, Myers, and Cadogan, eds., 1992, pp. 276–281. ———. 2000. “Vasiliki,” in Muhly and Sikla, eds., 2000, pp. 118–122.

Index

agriculture, 9–11, 20, 25, 26, 27, 30, 43, 46, 52, 57, 85, 86, 87 Alexandria (Egypt), 87, 91 Alexiou, S., 4 Anatolia/Asia Minor, 10, 38, 39, 47, 49, 56, 83, 86–89, 91, 99, 115, 126, 168, 170, 171, 172 Apostolakou, S., xxi aqueduct, xvi, 11, 12, 84, 86, 108, 112 Asari, 11, 12, 13, 70, 72 74, 88 Azoria, 19, 25, 36, 38, 43, 72, 74, 75–77, 78, 100

bees, 20, 43, 45, 52, 88 Betancourt, P., xx, xxii, 5, 18, 20, 26, 38, 46, 47, 61, 99, 101 Blitzer, H., xx, xxii, 8, 57 Brogan, T., xx, 38, 111

Cadogan, G., xx, xxi, 5, 45 Cha Gorge, 5, 11, 12, 42, 49, 120 Chalasmenos, 5, 69, 71, 72, 74, 75 Cherry, J., 97, 101 chert, 19, 21, 23, 27 chiefdom. See social organization Chrysokamino, 5, 20, 25, 27, 28, 36, 38, 47, 99

climate. See environment conflict, 31, 40, 49, 101, 102 craft specialization, 26–28, 29, 31, 99, 100 crafts. See economy Cyclades, 10, 19, 20, 22, 25, 26, 27, 28, 38, 39, 47, 59, 99 Cyprus, 10, 39, 61, 166, 169, 170, 172

Davaras, C., xix, xxi, 4, 5, 28, 29, 37, 45, 46, 60, 66, 78, 91, 171, 174 defensive sites, 19, 20, 24, 26, 34, 45, 50, 70, 71, 72, 73, 74, 75, 100

economy, 20, 28, 39, 46, 48, 62, 85, 89, 91 Egypt, 10, 28, 39, 40, 99 environment, xix, 5, 8, 9–13, 26, 35, 44, 46, 48, 53, 74, 84, 86, 99, 100, 136 Episkopi, 4, 5, 9, 10–13, 20, 24, 35, 42, 49, 56, 66, 67, 78, 86, 92, 112 Episkopi, LM III site, 66, 129 exchange. See trade

fishing, 10, 12, 20, 88, 89 Flannery, K., 29, 102

190

AN ARCHAEOLOGICAL SURVEY OF THE GOURNIA LANDSCAPE

geology, 9, 11, 12, 136, 137 geomorphology. See geology Gournia site, 3, 22, 34, 35, 40, 41, 45, 47, 57, 59, 60, 66, 101, 107, 108 granodiorite, 20, 43, 136, 140–144, 146, 147, 149–153, 156, 160, 161, 164, 165 ground stone, 23, 27, 42

Hall, E., 3, 4, 18, 22, 37, 38, 59, 61, 109 Haggis, D., xxi, 3–7, 19, 89 Hankey, V., 37 Hayden, B., xxi, 4, 55 Hawes Boyd, H., 3, 11, 19, 22, 58, 59, 86 Herapytna, 10, 55, 78, 81, 82, 83, 84, 87, 88, 89, 91, 92

ideology, 29, 39, 40, 47, 60, 100 isopolity, 82, 83, 87 Ierapetra, 5, 10, 12, 49, 56, 78, 82, 91, 93 Istanbul, 93 Isthmus of Ierapetra, xix, 5, 10, 11, 12, 55, 70

Kamares Ware, 38, 47, 48 Katalimata, 17, 19, 46, 50, 69, 70, 72 Kavousi Kastro site, 3, 4, 71, 72, 74, 75, 77, 155 Kavousi survey, 5, 11, 12, 18, 24, 35, 43, 52, 55, 66, 77–79, 81, 84, 93 knapping site, 21, 22, 24, 26 Konida isle, 10

land use, intensified, 35, 46, 52, 86 Levant, 10, 27, 28, 30, 39, 47, 61, 99, 100 Linear B, 67 loomweight, 20, 23, 28, 42, 52, 57, 88, 106, 109, 119, 131, 150

Malia, 10, 31, 39, 40, 47, 100 metallurgy, 20, 26, 28 metochi, 42, 106, 108, 128 Mirabello Bay, xix, xx, 4, 7, 9, 10, 130 Mochlos, 4, 26, 27, 28, 31, 36, 37, 39, 40, 45, 46, 47, 49, 60, 62, 66, 67, 83, 89, 99 Myrivilis, S., 49 Myrtos (Phournou Koriphi), 5, 27, 28, 29 Myrtos/Pyrgos, 5, 37, 38, 39, 40, 45, 47, 100

Nowcki, K., 5, 17, 19, 20, 50, 56, 69–76, 82

obsidian, 20, 21, 22, 23, 24, 25, 28, 60

Pacheia Ammos, 4, 5, 7, 12, 42 Palaikastro, 10, 31, 38, 67, 73, 147 pastoralism, 10, 13,20, 27, 74, 87 peak sanctuary, 55–57, 59, 65, 97, 101, 130, 131 peer polity interaction, 97, 101 Perakis, G., 3 Phaneromeni Monastery, 11 Platon, N., 4, 66, 79, 129 plow, 6, 7, 27 population, 10, 21, 24, 25, 26, 30, 33, 41, 43, 44, 51, 52, 55, 65, 66, 69, 70, 77, 78, 81, 84, 85, 100 polis, 5, 10, 56, 77–79, 81 pottery production, 19, 23, 28, 47, 99, 136–174 Priniatikos Pyrgos, 5, 25, 26, 28, 31, 36, 39, 54, 61 Profetes Elias site, 7, 55, 74, 75, 76, 77, 81, 82, 132, 133, 155–158, 163–165 Pseira, 4, 10, 28, 37, 45, 49, 52, 61, 62 Pseira shipwreck, 47 pumice, 22, 113, 115

rainfall, 11, 25, 59 Renfrew, C., 97, 99, 101 romanization, 92, 182

Sakellarakis, J., 4, 56, 111, 150 Seager, R., 3, 12 social complexity. See social organization social organization, 20, 29, 30, 31, 40, 48, 62, 63, 92, 97–102 social storage, 97 soils, 6, 9, 11, 12, 13, 19, 22, 26, 44, 45, 53, 85, 99 Soles, J., 3, 4, 30, 31, 36, 39, 41, 45, 46–48, 57–60, 62, 70, 99, 101 Sphoungaras, 4, 19, 22, 28, 39, 40, 41, 57, 109, 110 springs, 12, 13, 18, 23, 24, 25, 43 state. See polity state formation, 97–102 stone vases, 25, 27, 28, 30, 31, 39, 40, 46–48, 60, 62, 63, 99, 100 Strabo, 55 56, 78, 82 surplus, 27, 30, 55, 56, 86, 87, 90, 97 synoikism, 77, 79, 82 syssitia, 85, 92

Thera, 10, 38, 55, 59, 60, 61–63, 101, 108 Thryphti, 11, 12, 13, 18, 23, 34

INDEX

trade, 20, 27, 28, 30, 31, 39, 47, 61, 66, 83, 84, 86, 89, 91, 99, 100 Tsipopoulou, M., xxii, 4–6, 67, 71, 73, 75, 76, 82, 116

Warren, P., 5, 22, 38, 97, 99, 139 wealth finance, 97, 99

Vasiliki site, 4, 5, 18, 19, 23, 26, 28, 30, 31, 37, 40, 45, 49, 61, 99, 100 Vasiliki Kephala, 74, 75, 82 Vasiliki Ware, 22, 28, 31, 99, 140

191

Vrokastro, 4, 78 Vrokastro survey, 4, 11, 18, 19, 25, 26, 36, 44, 54, 55, 66, 77, 82, 86, 93 Vronda, 3, 19, 25, 71, 74, 75, 76, 77, 79

Xanthoudides, S., 4, 66, 129, 188

Zakros, 10, 28, 62, 63, 88 Zangger, E., 10–12, 58, 59 Zois, A., 4, 5, 23, 37, 45, 61, 99, 131

Tables

TABLES 1 AND 2 Function

Size

3

Field site

Yes

10

Possible field site and burial

Yes

98

Farm

0.09 ha

Yes

130

Hamlet (with cave)

0.25 ha

Yes

137

Field site

Yes

Katalimata

Farm

Yes

Kephala

Farm

Yes

3

New Number of Sites

Site

2

1

0 Field site (